ESMO 2026 trial results & KOL discussions, by trial →Each slide group below also links to its Full trial page and All posts.
Built by+60 more
At a glance
- Heme & Myeloma 5
- Other 91
- 1 trials with their own slide groups
- KOL infographics 15
- Physician posts first
- Last updated October 25, 2026
Heme & Myeloma 5
Built by
Other Heme & Myeloma slides
Not tied to a single tracked trial
#Surufatinib + PlatEtop + serplulimab in epNECs
@myESMO #ESMO26 #ESMOAmbassadors https://t.co/5sodZuPhr9
[Slide 1]
Study Design
Key Eligibility Criteria
• Aged ≥18 years
• Histologically confirmed NEC (except pulmonary origin)
• ECOG PS 0-1
• Measurable lesions
• Treatment naïve
N=60
Safety Run-in (n=6)
Induction (q3w, 6 cycles):
Surufatinib: 250 mg, qd, po
+ Serplulimab: 300 mg, d1, iv, q3w
+ EP regimen
Maintenance:
Surufatinib + Serplulimab
Dose expansion (n=54)
Induction (q3w, 6 cycles):
Surufatinib: RP2D
+ Serplulimab: 300 mg, d1, iv, q3w
+ EP regimen
Maintenance:
Surufatinib + Serplulimab
PD, death, or intolerable toxicity
Endpoints
• Primary endpoint: PFS (RECIST v1.1);
• Secondary endpoints: ORR, DCR, DoR, OS, AE.
• Etoposide injection: 100mg/m2, d1-3, IV, Q3W; Cisplatin: 25mg/m2, d1-3, IV, Q3W.
• Dose-limiting toxicities (DLTs) were observed for the first cycle (dose of Surufatinib would be adjusted to 200mg, qd, po, q3w if ≥ 2 DLTs were observed in the first 6 patients).
• Tumor response assessment every 6 weeks per RECIST v1.1
ECOG PS, Eastern Cooperative Oncology Group performance status; po, by mouth; qd, once daily, q3w, every 3 weeks; iv, Intravenous infusion; ORR, Objective response rate; DCR, Disease control rate; PFS, Progression-free survival; OS, Overall survival; AE, Adverse events.
Zhenyu Lin, MD
[Slide 2]
Tumor Response
In the response-evaluable population, the ORR was 69.8% and DCR was 95.3%.
Surufatinib + EP + serplulimab N=43
CR 1 | PR 29 | SD 11 | PD 2
Waterfall Plot: Tumor Change from Baseline
Change from baseline (%)
CR / PR / SD / PD
69.8% ORR
95.3% DCR
9.1 mo DoR
* New lesion appearance; ** 1st tumor assessment: SD; 2nd: target lesion shrinkage >30% with non-target lesion progression
Zhenyu Lin, MD
[Slide 3]
Conclusions
■ The quadruplet regimen of surufatinib + EP + serplulimab demonstrated promising antitumor activity as first-line treatment for extrapulmonary NEC, with a confirmed ORR of 69.8% and DCR of 95.3%.
■ The median PFS of 9.4 months is encouraging, particularly in a high-risk population with 67.4% liver metastases and a median Ki-67 of 70%. Patients achieving objective response derived a meaningful PFS benefit (11.5 months vs. 6.6 months).
■ The safety profile was manageable and consistent with the expected toxicities of the individual components, with no new safety signals identified. Grade ≥3 AEs were predominantly hematologic and attributable to the chemotherapy backbone.
■ Enrollment is ongoing to reach the planned sample size. Further exploratory analyses and long-term survival follow-up are warranted to better characterize the benefit of this quadruplet strategy in extrapulmonary NEC.
Zhenyu Lin, MD
#ESMO26 🧬 Epsametostat (HH2853) | Phase II
🎯 Advanced epithelioid sarcoma
💊 Dual EZH1/2 inhibitor
👥 94 pts | 75 efficacy-evaluable
📈 ORR: 40% | DCR: 77.3%
⏳ mPFS: 13.1 mo | mDoR: 15.6 mo
🎯 mOS: Not reached
⚠️ G≥3 TRAEs: 39.4%
🚀 Primary endpoint met! https://t.co/ya44WOgSgG
[Slide 1]
GLOBAL, MULTI-CENTER, FIRST-IN-HUMAN TRIAL DESIGN OF HH2853
Phase I
Dose Escalation and Expansion Cohorts
Dose Escalation: 100, 200, 200, 400, 600, 800, 1000 → Expansion cohorts; RP2D
Endpoints: MTD, DLT, Safety, PK, PD, Exploratory Efficacy
Phase II
HH2853 oral tablet BID (400 mg), 28-day cycle
Cohort 1 (N=44) Relapsed/refractory follicular lymphoma
Cohort 2 (N=94) Epithelioid sarcoma
Cohort 3 (N=61)
• Relapsed/refractory peripheral T-cell lymphoma
• Other relapsed/refractory non-Hodgkin lymphoma harboring EZH2 mutations
• Advanced solid tumors presenting specific genetic alterations (including EZH2 mutations, INI1 deficiency, BAP1 deficiency, ARID1A mutations, and/or SMARCA4 mutations)
Cohort 2: Epithelioid sarcoma (N=94)
Cohort 2 Specifics
• Treatment Regimen: HH2853, fixed dose 400 mg (RP2D) orally BID continuous, no dose adjustment by body weight/BSA; 28-day cycle
• Treatment Duration: Until disease progression, unacceptable toxicity, death, withdrawal, or other criteria.
Eligibility Criteria
• Age ≥ 18 years, Male/Female
• Histologically or cytologically confirmed ES
• Previously treated, including failure and intolerable toxicity, unresectable locally advanced or metastatic ES
• ECOG PS ≤ 1; Life expectancy ≥ 3 months
• Adequate organ function (bone marrow, liver, kidney)
• At least 1 measurable target lesion
Exclusion Criteria:
• Symptomatic CNS metastasis
• Concomitant or prior malignancy
• Serious active infection requiring systemic therapy
• Prior EZH2 or EZH1/2 inhibitor therapy
Study endpoints (cohort 2)
PRIMARY ENDPOINT
• BIRC-assessed confirmed ORR (RECIST 1.1)
SECONDARY ENDPOINTS
• Investigator-assessed confirmed ORR
• BIRC- & Investigator-assessed confirmed DoR, DCR, TTR, PFS
• Safety profile
• Population pharmacokinetic (PopPK) parameters: typical population PK parameters for HH2853 and its potential metabolites.
EXPLORATORY ENDPOINTS
• Exposure-response (E–R) relationships
• Overall survival
Study objectives: To evaluate the preliminary efficacy, safety, pharmacokinetics, and biomarker correlates of continuous twice-daily oral HH2853 in relapsed/refractory non-Hodgkin lymphoma and advanced solid tumors (Cohorts 1 & 3), as well as BIRC-confirmed ORR in previously treated epithelioid sarcoma (Cohort 2).
[Slide 2]
Efficacy endpoints assessed by BIRC — HH2853-G101 Cohort 2
Figure 2. Duration of treatment and clinical response (EAS)
[swimmer plot; legend: Treatment Failure from Last Treatment (N = 72); Intolerance from Last Treatment (N = 3)]
Swimmer plot illustrating the duration of treatment, individual response over time, and discontinuation status.
Table 2. Antitumor activity (EAS)
| Patients (N=75)
ORR, n (%), 95% CI | 30 (40.0), (28.9, 52.0)
DCR, n (%), 95% CI | 58 (77.3), (66.2, 86.2)
DoR, month, Median (95% CI) | 15.6 (7.5, NE)
TTR, month, Median (range) | 1.86 (1.4, 9.0)
As of June 30, 2025, the median treatment duration was 35.4 weeks. The primary endpoint was met, demonstrating an ORR of 40.0% (95% CI, 28.9–52.0) by BICR according to RECIST v1.1.
Figure 3. Tumor response assessed by BICR (EAS)
[waterfall plot; PR / SD / PD; HH2853 400mg BID Intolerance from Last Treatment; HH2853 400mg BID Treatment Failure from Last Treatment]
Waterfall plot showing the best percentage change from baseline in the sum of target lesion diameters. Tumor shrinkage was observed in the majority of patients (88.0%, 66/75), with 48.0% (36/75) achieving a reduction in tumor burden of ≥30%.
[Slide 3]
Secondary endpoints — HH2853-G101 Cohort 2
Figure 4. Kaplan-Meier curve of progression-free survival (PFS).
Kaplan–Meier estimates of PFS assessed by BIRC in the efficacy analysis set (N=75). As of the data cutoff date, 41 patients (54.7%) had experienced a PFS event, including 37 (49.3%) with disease progression and 4 (5.3%) deaths. The median PFS was 13.1 months (95% CI: 7.4–16.6). Numbers at risk at specified time points are shown below the horizontal axis.
Median (95% CI): Total 13.1 (7.4, 16.6); Treatment Failure 13.1 (7.4, 16.6)
Number of subjects at risk (0, 3, 6, 9, 12, 15, 18, 21, 24, 27 months)
Total | 75 | 58 | 43 | 34 | 20 | 17 | 11 | 6 | 1 | 0
Treatment Failure | 72 | 55 | 41 | 33 | 19 | 17 | 11 | 6 | 1 | 0
Figure 5. Kaplan-Meier curve of overall survival (OS). Kaplan–Meier estimates of OS in the efficacy analysis set (N=75). At the data cutoff date, with a data maturity of 24.0% (18 deaths recorded among 75 patients), the median OS was not reached (95% CI: NE–NE). Numbers at risk at specified time points are displayed beneath the x-axis.
Median (95% CI): Total NE (NE, NE); Treatment Failure NE (21.4, NE)
Number of subjects at risk (0, 3, 6, 9, 12, 15, 18, 21, 24, 27 months)
Total | 75 | 72 | 63 | 52 | 39 | 31 | 24 | 17 | 5 | 0
Treatment Failure | 72 | 69 | 61 | 50 | 37 | 29 | 22 | 16 | 5 | 0
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: PFS (BIRC) and OS, HH2853-G101 Cohort 2 (EAS N=75)
Total (PFS) | 75 | 13.1 | 7.4, 16.6
Treatment Failure (PFS) | 72 | 13.1 | 7.4, 16.6
Total (OS) | 75 | NE | NE, NE
Treatment Failure (OS) | 72 | NE | 21.4, NE
Note: Single-arm. PFS numbers at risk (Total) 75,58,43,34,20,17,11,6,1,0 at 0,3,...,27 mo; Treatment Failure 72,55,41,33,19,17,11,6,1,0. OS at risk Total 75,72,63,52,39,31,24,17,5,0; Treatment Failure 72,69,61,50,37,29,22,16,5,0. OS data maturity 24.0%.
[Slide 4]
Safety — HH2853-G101 Cohort 2
Table 3. Safety profile (SAS)
n (%) | Patients (N=94)
TEAEs | 93 (98.9)
Grade ≥3 | 47 (50.0)
Serious | 23 (24.5)
Leading to dose reduction | 6 (6.4)
Leading to dose interruption | 28 (29.8)
Leading to treatment discontinuation | 3 (3.2)
Leading to death | 4 (4.3)
TRAEs | 91 (96.8)
Grade ≥3 | 37 (39.4)
Serious | 7 (7.4)
Leading to dose reduction | 6 (6.4)
Leading to dose interruption | 20 (21.3)
Leading to treatment discontinuation | 3 (3.2)
Leading to death | 0
HH2853 demonstrated a favorable and manageable safety profile, characterized by a low incidence of serious TRAEs (7.4%) and infrequent AE-related dose modifications or discontinuations.
Figure 7. Common TEAEs occurring in ≥30% of patients (SAS) (Patients, %)
Alopecia 31.9
Blood lactate dehydrogenase increased 33.0
Alanine aminotransferase increased 35.1
Hyperuricemia 38.3
Weight decreased 40.4
Lymphopenia 45.7
Diarrhea 47.9
Hypertriglyceridemia 51.1
Hyperbilirubinemia 54.3
Anemia 55.3
Figure 8. Common TRAEs occurring in ≥30% of patients (SAS) (Patients, %)
Hyperuricemia 31.9
Alopecia 31.9
Alanine aminotransferase increased 31.9
Weight decreased 34.0
Hypertriglyceridemia 36.2
Lymphopenia 40.4
Diarrhea 45.7
Hyperbilirubinemia 51.1
Anemia 52.1
Thoughtful discussion by Lisa Salvatore: Temab-A + Bev shows promise in mCRC, but haematological toxicity matters.
Questions unanswered: What does Bev add? Can c-Met guide selection? What is the optimal dose? Awaiting Phase III validation #ESMO26 #ESMOAmbassadors @OncoAlert https://t.co/n0clts8P3q
[Slide 1]
Efficacy: promising, but…
Median follow-up: 18.2 months (range 0.4–22.8)
PFS | Temab-A 2.4 mg/kg + Bev (n=30) | FTD-TPI + Bev (n=20)
Events, n (%) | 27 (90.0) | 18 (90.0)
Median PFS, months (95% CI) | 6.8 (5.4–9.5) | 4.2 (1.8–6.3)
HR (95% CI) | 0.48 (0.26–0.89)
Progression-free survival
→ Sunlight (N=246) mPFS: 5.6 mos
Worse performance of FTD-TPI+Bev due to more heavily pretreated pts?
Patients at Risk (0–20 months)
Temab-A 2.4 mg/kg | 30 30 28 26 25 21 19 12 12 11 4 4 4 4 3 2 1 1 1 1 0
FTD-TPI | 20 19 15 12 11 9 7 4 4 2 1 1 1 1 0 0 0 0 0 0 0
OS | Temab-A 2.4 mg/kg + Bev (n=30) | FTD-TPI + Bev (n=20)
Events, n (%) | 22 (73.3) | 17 (85.0)
Median OS, months (95% CI) | 14.7 (9.8–17.8) | 10.2 (7.7–14.5)
HR (95% CI) | 0.62 (0.33–1.18)
Overall survival
→ Sunlight (N=246) mOS: 10.8 mos
Numerically longer OS, but wide CI
HR: is it appropriate?
Patients at Risk (0–23 months)
Temab-A 2.4 mg/kg | 30 30 29 27 26 25 25 24 24 22 19 19 19 16 15 12 9 9 8 4 1 1 1 0
FTD-TPI | 20 19 19 19 19 19 18 16 15 13 10 8 6 6 6 5 3 3 2 1 0 0 0 0
Prager et al. NEJM 2023
Lisa Salvatore, MD, PhD
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: PFS and OS: Temab-A 2.4 mg/kg + Bev vs FTD-TPI + Bev (mCRC)
Temab-A 2.4 mg/kg + Bev (PFS) | 30 | 6.8 | 5.4–9.5
FTD-TPI + Bev (PFS) | 20 | 4.2 | 1.8–6.3
Temab-A 2.4 mg/kg + Bev (OS) | 30 | 14.7 | 9.8–17.8
FTD-TPI + Bev (OS) | 20 | 10.2 | 7.7–14.5
HR: PFS 0.48; OS 0.62
HR CI: PFS 0.26–0.89; OS 0.33–1.18
Note: Median follow-up 18.2 months (range 0.4–22.8). Discussant comparison: SUNLIGHT (N=246) mPFS 5.6 mos, mOS 10.8 mos (Prager et al. NEJM 2023). Numbers-at-risk rows transcribed in ocr_text (small print, moderate confidence).
[Slide 2]
Efficacy: promising, but…
| Temab-A 2.4 (N=40) | Temab-A 2.0+Bev (N=26) | Temab-A 2.4+Bev (N=30)
RR | 17.5% | 19% | 30%
DCR | 77.5% | 81% | 97%
mPFS | 5.3 | 5.6 | 6.8
mOS | 10.7 | 11.2 | 14.7
Sharma et al. JCO 2026; Cecchini et al. ESMO 2025; Cecchini et al. ESMO 2026
Open question #1
• What does really add Bev to Temab-A?
Higher RR if Bev is added to Temab-A 2.4, but dose reduction due to AEs in 50% of patients
We will never have a direct comparison Temab-A vs Temab-A+Bev
Lisa Salvatore, MD, PhD
[Slide 3]
Efficacy: promising, but…
| Temab-A ≥2.4 cMet 3+ ≥10% (N=25) | Temab-A ≥2.4 cMet 3+ <10% (N=50) | Temab-A 2.0/2.4+Bev cMet 3+ ≥10% (N=18) | Temab-A 2.0/2.4+Bev cMet 3+ <10% (N=35)
RR | 36% | 12% | 17% | 31%
Sharma et al. JCO 2026; Murciano-Goroff et al. ESMO 2025; Cecchini et al. ESMO 2025
| Temab-A 2.4+Bev cMet 3+ ≥10% (N=10) | Temab-A 2.4+Bev cMet 3+ <10% (N=18)
mPFS | 9.3 | 6.8
mOS | 15.3 | 13.3
Cecchini et al. ESMO 2026
Open questions #2
• Can cMet expression predict response?
Yes if Temab-A alone
??? Temab-A+Bev (no for RR; PFS and OS promising indipendently of c-Met expression)
Lisa Salvatore, MD, PhD
[Slide 4]
Temab-A + Bev: my personal conclusions
✓ Temab-A + Bev is a promising regimen for pretreated mCRC (RR30%, mPFS 6.8 months and mOS 14.7 months).
✓ Pay attention to hematologic toxicity, in particular 53% of G3-4 anemia and 33% of G3-4 neutropenia. High rate of dose interruption and reduction.
✓ Phase III study is ongoing. I hope the study will be sufficiently powered to address the remaining open questions: patient selection and the optimal dose of Temab-A+Bev. Though the added benefit of Bev will remain unknown.
Lisa Salvatore, MD, PhD
Caveats: ctDNA % drives sensitivity, and deletions are hardest. PTEN loss is spatially heterogeneous, with discordance both ways, so neither tissue nor plasma is a gold standard. PTEN loss in ctDNA only → worst OS (mOS 11 mo). Matched WBC DNA is a must: 10 "hits" were CHIP. 2/2 https://t.co/Qdfx6jaien
[Slide 1]
PTEN heterogeneity complicates accurate classification
Tissue NGS | Plasma ctDNA
55 tissue only | 20 both | 13 plasma only
n = 88 altered patients · circle and overlap areas proportional to patient numbers
[oncoprint: ctDNA fraction %, biopsy, PTEN tissue, PTEN ctDNA rows]
Of 17 patients with PTEN biallelic loss in tissue
8 biallelic loss
9 partial loss
despite optimising for high ctDNA ≥20%
PTEN status: Intact; Single copy loss; Mutation + LOH
PTEN mutation only present in 2 samples
Copy number loss; Missense mutation; Truncating mutation
PTEN / TP53 / RB1 / APC (samples PB3, PB1, PB4, PB2, RP7, RP4, RP3, RP6, RP1, RP5, RP8)
Warner E, Van der Eecken E, Murtha AJ et al, Nat Cancer 2024
Edmond M. Kwan, MBBS PhD
[Slide 2]
Escape of the aggressive PTEN-defective tumour clone?
Tissue NGS | Plasma ctDNA
55 tissue only | 20 both | 13 plasma only
n = 88 altered patients · circle and overlap areas proportional to patient numbers
[oncoprint: ctDNA fraction %, biopsy, PTEN tissue, PTEN ctDNA]
Of 17 patients with PTEN biallelic loss in tissue
8 biallelic loss
9 partial loss
despite optimising for high ctDNA ≥20%
PTEN loss stratified by compartment
Neither n=136, mOS 32 mo
Tissue-only n=58, mOS 29 mo
ctDNA-only n=23, mOS 11 mo
Both n=28, mOS 17 mo
Overall survival
Months from first blood draw
No. at risk (0, 12, 24, 36, 48, 60)
Neither: 136 121 91 59 45 19
Tissue-only: 58 51 36 22 16 7
ctDNA-only: 23 9 3 1 0 0
Both: 28 21 6 4 2 1
Edmond M. Kwan, MBBS PhD
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: Overall survival from first blood draw, by PTEN loss compartment
Neither | 136 | 32 mo
Tissue-only | 58 | 29 mo
ctDNA-only | 23 | 11 mo
Both | 28 | 17 mo
Note: Medians from the printed legend. No HR, CI or p-value printed. At-risk table is small font but legible.
Great discussion with Prof. Silvia Stacchiotti at #ESMO26 in Madrid! Two challenging sarcoma topics, expertly presented and discussed. Really enjoyed the insightful exchange and the opportunity to learn from such an outstanding expert in the field. 👏
#Sarcoma #ESMO2026 https://t.co/OUp4u1Vajd
[Slide 2]
Background
Extensive and clinically relevant alternative splicing in sarcoma
Genome-wide analysis using TCGASpliceSeq in TCGA-SARC (n = 261)
1. Cohort and data source: TCGA-SARC (n = 261) • 261 sarcoma cases with overall survival • RNA-seq data from TCGA
2. Alternative splicing quantification: TCGASpliceSeq Genome-wide quantification of alternative splicing events (Percent Spliced In, PSI) | Exon skipping (ES) | Alternative promoter (AP) | Alternative terminator (AT) | Alt. acceptor (AA) | Alt. donor (AD) | Retained intron (RI) | Mutually exclusive exons (ME)
3. Survival association: Low PSI / High PSI; Cox regression (P < 0.05)
4. Prognostic impact: Survival-associated AS events; High-risk AS signature / Low-risk AS signature
A highly complex splicing landscape in sarcoma: 40,184 alternative splicing events in 9,673 genes; ~4.1 AS events per gene
3,610 survival-associated AS events in 2,291 genes: 3,610 (9.0%) associated with overall survival; Total 40,184 AS events; Univariate Cox analysis (P < 0.05)
Distribution of survival-associated events by splicing type:
Exon skipping (ES) | 1,359 (37.6%)
Alternate terminator (AT) | 776 (21.5%)
Alternate promoter (AP) | 758 (21.0%)
Alternate acceptor (AA) | 252 (7.0%)
Alternate donor (AD) | 252 (7.0%)
Retained intron (RI) | 203 (5.6%)
Mutually exclusive exons (ME) | 10 (0.3%)
~80% were ES, AT or AP events
Integrated AS signature shows strong prognostic performance: Time-dependent ROC (overall survival)
Integrated model (6 AS types) | AUC = 0.85
Alternate promoter (AP) | AUC = 0.82
Exon skipping (ES) | AUC = 0.80
Risk groups based on AS signature: Low AS risk / High AS risk, P < 0.0001
Alternative splicing represents a widespread and clinically relevant program in sarcoma: ~1 in 11 detectable splicing events is associated with patient survival.
Hong W et al. Ann Transl Med. 2019;7:557.
Dr. J Martin-Broto
[Slide 3]
HH2853 / epsametostat in epithelioid sarcoma (ES):
Clear biological rationale
Antitumor activity confirmed starting from 1st line
(40% ORR; responses in proximal and distal ES; mPFS 13 months)
Safety: long term hematologic follow-up needed
Any global development plan in ES, and, possibly, in other SMARCB1-deleted sarcoma, such as poorly differentiated chordoma?
Cirtuvivint in translocated sarcomas:
Overall marginal antitumor activity
Potentially preliminary encouraging signal in SFT
No safety concerns so far
Biological rationale and identification of translocated sarcoma of potential interest to be further clarified
Validation in larger cohorts essential
Free Access · KOL Pulse Intelligence
Track oncology’s top KOLs in your specialty
Create a free account, pick the tumor types you cover, and go beyond the tweet — the intelligence pharma teams use to map influence and prepare for engagement:
- ✓Enhanced KOL profiles
- ✓Pharma influence & potential advisory-board patterns
- ✓2025 Open Payments financial analysis
- ✓Social-media sentiment & trend signals
Get free access — pick your specialty
Choose the tumor types you follow. No cost, unsubscribe anytime.
Other 91
Built by+61 more
TANGENT
Emactuzumab vs placebo · Tenosynovial giant cell tumor (Phase 3, NCT05417789, SynOx Therapeutics)All posts →TANGENT
#ESMO26 🚀 TANGENT | Phase III
👥 136 pts | TGCT
💊 Emactuzumab (anti-CSF-1R) vs placebo
💉 5 infusions over 8 weeks
🎯 6-mo ORR (RECIST): 30% vs 2%
📈 ORR (TVS): 54% vs 7%
✅ Improved pain, stiffness & function (p<0.001)
⚠️ G≥3 TRAEs: 48% vs 2%
💡 Effective short-course therapy!
@OncoAlert
Same slide also posted by @PerezFidal
[Slide 1]
Objective Response Rate (RECIST v1.1) at Month 6 (Primary Endpoint)
Emactuzumab (n=91) — Objective response at Month 6: CR 2 (2%); PR 25 (27%); SD 54 (59%); PD 4 (4%); Missing data/not evaluable 6 (7%)
Placebo (n=45) — Objective response at Month 6: CR 0 (0%); PR 1 (2%); SD 42 (93%); PD 0; Missing data/not evaluable 2 (4%)
[Waterfall plots: Change from baseline, %; Patients; legend PD, SD, PR, CR]
| Emactuzumab (n=91) | Placebo (n=45)
ORRᵃ by RECIST v1.1, n (%) | 27 (30) | 1 (2)
Difference (95% CI)ᵇ | 28% (16, 39); p<0.001
At Month 6, emactuzumab demonstrated a statistically significant improvement in ORR by RECIST v1.1 compared with placebo (30% vs 2%). Most emactuzumab responders (70%) had responded by Week 6 (first MRI assessment)
ᵃCR or PR according to RECIST v1.1 at 6-month assessment. ᵇCommon risk difference stratified by prior TKI use; CI calculated using the Miettinen and Nurminen method; p value calculated using the Cochran-Mantel-Haenszel test stratified by prior TKI use.
CI, confidence interval; CR, complete response; MRI, magnetic resonance imaging; ORR, objective response rate; PD, progressive disease; PR, partial response; RECIST, Response Evaluation Criteria in Solid Tumors; SD, stable disease; TKI, tyrosine kinase inhibitor.
[Slide 2]
Change at Month 6 in Pain, Stiffness, and Quality of Life
Emactuzumab | Placebo
Worst Pain NRS — LSM (SE) change from baseline; Visit BL, W2, W4, W6, M2, M3, M4, M5, M6
Month 6: Emactuzumab −2.5; Placebo −0.4
Difference in LSM (95% CI): −2.2 (−3.0, −1.4); p<0.001ᵃ
Worst Stiffness NRS
Month 6: Emactuzumab −2.4; Placebo −0.3
Difference in LSM (95% CI): −2.1 (−3.0, −1.3); p<0.001ᵃ
Regimen: 5 infusions over 8 weeks
• QoL (EQ-5D-5L VAS) showed a numerically greater improvement with emactuzumab (+2.0) compared with placebo (−0.4) at Month 6 (LSM difference [95% CI]: 2.4 [−4.2, 9.1]; p=0.238)ᵃ
Emactuzumab was associated with rapid and durable clinically meaningful reductions in pain and joint stiffness. Differences versus placebo were statistically significant by Week 2 (after one dose) onwards
ᵃEstimates are from a mixed model for repeated measures with fixed effects for treatment, visit, and the interaction between treatment and visit, and covariates for prior TKI use (yes/no) and baseline score.
BL, baseline; CI, confidence interval; EQ-5D-5L, EuroQol 5 Dimension, 5 Level questionnaire; LSM, least-squares mean; M, Month; NRS, numerical rating scale; QoL, quality of life; SE, standard error; TKI, tyrosine kinase inhibitor; VAS, visual analogue scale; W, Week.
[Slide 3]
Most Common AEs Considered to be Related to Treatment
AE by preferred term in ≥15% of patients in the emactuzumab group, n (%) | Emactuzumab (n=91) All grades | Grade 3/4 | Placebo (n=45) All grades | Grade 3/4
Any treatment-related AE, n (%) | 91 (100) | 44 (48) | 31 (69) | 1 (2)
Pruritus | 69 (76) | 3 (3) | 10 (22) | 0
Periorbital oedema | 63 (69) | 1 (1) | 3 (7) | 0
Fatigue | 38 (42) | 2 (2) | 9 (20) | 0
Oedema peripheral | 38 (42) | 0 | 0 | 0
Face oedema | 35 (38) | 1 (1) | 1 (2) | 0
Blood CPK increased | 28 (31) | 22 (24) | 1 (2) | 0
AST increased | 22 (24) | 2 (2) | 0 | 0
Eyelid oedema | 22 (24) | 0 | 6 (13) | 0
Rash | 21 (23) | 1 (1) | 2 (4) | 0
Rash maculo-papular | 20 (22) | 2 (2) | 2 (4) | 0
Nausea | 19 (21) | 0 | 7 (16) | 0
Asthenia | 17 (19) | 1 (1) | 4 (9) | 0
Dermatitis acneiform | 16 (18) | 0 | 0 | 0
Lacrimation increased | 16 (18) | 0 | 0 | 0
Lichen myxoedematosus | 16 (18) | 2 (2) | 0 | 0
Blood lactate dehydrogenase increased | 14 (15) | 0 | 0 | 0
Hypertension | 14 (15) | 8 (9) | 1 (2) | 1 (2)
• The most common Grade 3 or 4 treatment-related AE was blood CPK increased, observed in 24% of patients in the emactuzumab group
Post-treatment AEs: any AE with onset or worsening of a pre-existing condition on or after the first administration of study treatment in the double-blind phase and before the first administration of study treatment in the open-label phase.
Events were considered by the investigator to be definitely, probably, or possibly related to treatment.
AE, adverse event; AST, aspartate aminotransferase; CPK, creatine phosphokinase.
[Slide 4]
Conclusions
• A planned course of 5 infusions of emactuzumab over 8 weeks provided statistically significant reductions in tumour size at Month 6 compared with placebo
• Most responders at Month 6 had responded by Week 6 (first MRI assessment) based on both RECIST v1.1 and TVS criteria
• Clinically meaningful improvements in physical function and range of motion with emactuzumab were observed early and were sustained over time, with statistically significant differences versus placebo emerging from Week 4 onwards
• Improvements in pain and stiffness were statistically significant with emactuzumab compared with placebo by Week 2, after the first infusion
• Emactuzumab had a manageable safety profile consistent with Phase 1 findings and expectations of CSF-1R inhibition
Emactuzumab demonstrated radiographic and functional efficacy and manageable safety in patients with TGCT, and has the potential to be the first short-course, potent treatment option for TGCT
CSF-1R, colony-stimulating factor-1 receptor; MRI, magnetic resonance imaging; RECIST, Response Evaluation Criteria in Solid Tumors; TGCT, tenosynovial giant cell tumour; TVS, tumour volume score.
Cross-trial comparisons & roundups
Slides that compare or list several trials
🚨 One of those #ESMO26 presentations I knew nothing about (and ended up loving)
SL-28: a novel off-the-shelf, non-genetically modified allogeneic leukocyte therapy in heavily pretreated solid tumors (n=12)!
💥 ORR 66.7% (4 CR, 4 PR), DCR 100%, with responses across multiple tumor types and mDoR not reached!
🚨 mPFS: 24.7 mo. Importantly, the observed deaths were non-cancer related, so the OS curve should be interpreted in that context.
✅ No GVHD. No CRS. No ICANS.
A very early n=12 dataset, but definitely one to watch 👀
@OncoAlert @OncoReporte #ESMO26
@OncoAlert @OncoReporte @myESMO @_SEOM
[Slide 1]
RESULTS: Treatment-emergent adverse events
Event | ANY Grade, n (%) | Grade 1-2, n (%) | Grade 3-4, n (%)
Any | 11 (91.7%) | 11 (91.7%) | 0 (0%)
Pyrexia | 10 (83.3%) | 10 (83.3%) | 0 (0%)
Anemia | 0 (0%) | 0 (0%) | 0 (0%)
WBC count decreased | 0 (0%) | 0 (0%) | 0 (0%)
Leukopenia | 0 (0%) | 0 (0%) | 0 (0%)
Neutropenia | 0 (0%) | 0 (0%) | 0 (0%)
Platelet count decreased | 0 (0%) | 0 (0%) | 0 (0%)
ALT/AST increased | 1 (8.3%) [?] | 1 (8.3%) [?] | 0 (0%)
Nausea | 1 (8.3%) [?] | 1 (8.3%) [?] | 0 (0%)
Vomiting | 0 (0%) [?] | 0 (0%) [?] | 0 (0%)
Hypotension | 0 (0%) | 0 (0%) | 0 (0%)
Fatigue | 0 (0%) | 0 (0%) | 0 (0%)
Headache | 0 (0%) | 0 (0%) | 0 (0%)
GVHD-linked TEAEs (all) | 0 (0%) | 0 (0%) | 0 (0%)
CRS-linked TEAEs (all) | 0 (0%) | 0 (0%) | 0 (0%)
ICANS-linked TEAEs (all) | 0 (0%) | 0 (0%) | 0 (0%)
[Slide 2]
RESULTS: Outcomes
Radiologic response
Y-axis: Change in Lesion Diameters (%) 25 0 -25 -50 -75 -100
Bars (left to right): PDAC* | GC | LLC | ACC | FL | BLCA | NB | PDAC | PCa | PCa | PCa | NSCLC
According to RECIST 1.1:
CR, n=4 (33.3%)
PR, n=4 (33.3%)
SD, n=4 (33.3%)
PD, n=0 (0%)
ORR: 8/12 (66.7%)
DCR: 12/12 (100%)
ACC Adenoid cystic carcinoma (salivary gland)
BLCA Bladder Urothelial Carcinoma
FL Follicular lymphoma
GC Gastric cancer
LCC Large-cell lung carcinoma
NB Neuroblastoma
NSCLC Non-small cell lung cancer
PCa Prostate cancer
PDAC Pancreatic ductal adenocarcinoma
*Tumor doubling time increased from 22 to 358 days, representing a 16.3-fold prolongation.
[Slide 3]
RESULTS: Duration of Disease Control
Rows (top to bottom): FL | PCa | PCa | NSCLC | NB | PCa | PDAC | BLCA | LCC | ACC | GC | PDAC
X-axis: Duration (months) 0 5 10 15 20 25 30 35
OS STATUS column (x = Death, o = Censored)
According to RECIST 1.1:
CR, n=4 (33.3%)
PR, n=4 (33.3%)
SD, n=4 (33.3%)
PD, n=0 (0%)
→ Ongoing
X Death (non-cancer related)
ACC Adenoid cystic carcinoma (salivary gland)
BLCA Bladder Urothelial Carcinoma
FL Follicular lymphoma
GC Gastric cancer
LCC Large-cell lung carcinoma
NB Neuroblastoma
NSCLC Non-small cell lung cancer
PCa Prostate cancer
PDAC Pancreatic ductal adenocarcinoma
Median duration of response in patients with CR/PR has not been reached
[Slide 4]
RESULTS: OS and PFS
OS (n=12)
Preliminary median OS 24.7 mo
12-month OS 83.3%
Estimated survival probability
Y-axis: Probability of Survival 0 25 50 75 100
X-axis: OS time (months) 0 10 20 30 40
PFS (n=12)
Preliminary median PFS 24.7 mo
12-month PFS 83.3%
24-month PFS 69.4%
Y-axis: Probability of PFS
X-axis: Latest progression-free CT/MRI assessment (months) 0 10 20 30 40
No radiographic PD; PFS events were deaths from any cause
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: OS and PFS (single arm, n=12)
SL-28 (OS) | 12 | 24.7 mo (preliminary)
SL-28 (PFS) | 12 | 24.7 mo (preliminary)
{'timepoint': '12 months', 'arm': 'OS', 'value': '83.3%', 'ci': None}
{'timepoint': '12 months', 'arm': 'PFS', 'value': '83.3%', 'ci': None}
{'timepoint': '24 months', 'arm': 'PFS', 'value': '69.4%', 'ci': None}
Note: Single-arm study, no comparator. No radiographic PD; PFS events were deaths from any cause.
‼️ #ESMO26 | ADC toxicities with Yeon Hee Park
💥Key pearls:
• Toxicity depends on the payload, linker, target & patient of course!!
• Different ADCs, different profiles! T-DXd: ILD | SG: myelosuppression/diarrhea | Dato-DXd: stomatitis/ocular
• Timing matters: toxicities can emerge from days to months…
With ADCs, the construct tells you what to watch for
@OncoAlert
@OncoReporte @myESMO @_SEOM @Lung_Cancers @gecp_org
[Slide 1]
Linker stability is a trade-off, not a target to maximize; determinant of ADC toxicity
less stable in plasma -> more stable in plasma
Hydrolysable CL2A - Sacituzumab govitecan (pH-dependent hydrolysis) - PH-dependent: bystander effect, systemic toxicities
Tetrapeptide GGFG - Trastuzumab deruxtecan (enzymatic cleavage) - Protease-dependent: bystander effect, target-dependent toxicities
Thioether, non-cleavable - Trastuzumab emtansine - Non-cleavable: no bystander effect
[Structure labels: SN38 (camptothecin); DXd (camptothecin); DM1 (maytansinoid); Hydrolysable CL2A linker (pH); Tetrapeptide linker (cathepsin B protease); Thioether linker]
Unstable linker
• More free payload in the circulation
• Higher rates of neutropenia
• Less ocular toxicity
• Payload reaches tumour tissue that the antibody itself penetrates poorly
Stable linker
• Less free payload in the circulation
• Lower rates of neutropenia
• More ocular toxicity
• Release depends more heavily on efficient internalisation
Conjugation chemistry & DAR
• Stochastic lysine/cysteine conjugation -> heterogeneous DAR, variable toxicity
• T-DM1: lysine conjugation, average DAR 3.5 at variable sites
• T-DXd, SG: cysteine conjugation at 8 sites, DAR up to 8, more homogeneous
• All approved ADCs still stochastic; site-specific conjugation in development
Observed in MMAE-based ADCs: the same molecule redesigned for greater linker stability shifts the toxicity profile rather than removing it.
Colombo R, et al. Cancer Discov 2024;14(11):2089-2108. Tarantino P, et al. Nat Rev Clin Oncol. 2023;20:558-576.
Yeon Hee Park
[Slide 2]
Determinants of the toxicities of ADCs
• Payload: class sets the expected AE spectrum; potency and DAR scale the severity of toxicity
• Linker: less stable -> higher free-payload peaks and chemotherapy-like AEs; very stable -> sometimes ocular, neuro- or hepatotoxicity
• Antibody: on-target, off-tumour effects (HER2 -> cardiotoxicity; Trop2 -> rash, mucositis; nectin 4 -> rash, dysgeusia; tissue factor -> bleeding) and FcγR uptake by immune cells
• Patient: organ function, comorbidities, pharmacogenomics, body composition, ethnicity
Pharmacodynamic determinants
Uptake into non-malignant cells | Fc/lectin receptors, endocytosis, passive diffusion
Bystander effect | Payload released from apoptotic cells
Deconjugation / linker instability | Premature payload release in circulation
High drug-to-antibody ratio | Increased total payload dose
Limited tumour penetration | Increased systemic toxicity
Pharmacokinetic determinants
Limited target-specific binding | Increased non-target exposure
↑ Absorption, ↓ catabolism | Increased systemic concentration
↑ Half-life, ↓ excretion, albumin binding | Prolonged exposure
Lipophilic payload | Higher tissue penetrance
[Figure: Linker stability (non-cleavable; cleavable enzyme labile; cleavable chemically labile) - Stability vs Chemotherapy-like toxicities; Payload mechanism of action (microtubule inhibitors: DM1 thrombocytopenia, hepatotoxicity; DM4 ocular toxicity; MMAE peripheral neuropathy, myelotoxicity; MMAF ocular toxicity; topoisomerase I inhibitors: diarrhoea, neutropenia; calicheamicins, duocarmycins, pyrrolobenzodiazepines: neutropenia); Antibody on-target toxicities (HER2 cardiotoxicity; Trop2 rash, mucositis; Nectin 4 rash, dysgeusia; Tissue factor bleeding); Binding to Fc receptors on immune cells (off-target toxicities, immune activation); Patient-related factors]
Yeon Hee Park
Tarantino et al. Nature Reviews Clinical Oncology | Volume 20 | August 2023 | 558–576 (Fig. 2; Table 2)
[Slide 3]
Different toxicity profiles, not different degrees of toxicity
Trastuzumab deruxtecan - DESTINY-Breast06
Nausea, any grade | 65.9%
Alopecia, any grade | 45.4%
Fatigue, any grade | 45.4%
Neutropenia, G3–4 | 20.7%
ILD / pneumonitis | 11.3% (G5 0.7%)
Watch: interstitial lung disease
Sacituzumab govitecan - TROPiCS-02
Neutropenia, G3–4 | 51%
Nausea, any grade | 55%
Alopecia, any grade | 46%
Diarrhoea, G3–4 | 9%
ILD / pneumonitis | Not reported
Watch: myelosuppression and diarrhoea
Datopotamab deruxtecan - TROPION-Breast01
Nausea, any grade | 51.1%
Alopecia, any grade | 36.4%
Dry eye, any grade | 21.7%
Stomatitis, G3–4 | 6.4%
ILD / pneumonitis | 3.3% (one G5)
Watch: stomatitis and ocular surface toxicity
Comorbidity should drive the choice: lung disease argues against T-DXd, low marrow reserve against SG, ocular surface disease against Dato-DXd.
Rates as compiled in Kang S, Kim SB. Ther Adv Med Oncol 2025;17:1–32. Cross-trial comparison — populations and reporting conventions differ.
Yeon Hee Park
[Slide 4]
Time to first onset differs by TEAE
Median days to first onset (log scale) ■ 5.4 mg/kg ■ 6.4 mg/kg
Axis: 1, 10, 100, 1000
TEAE | 5.4 mg/kg | 6.4 mg/kg
Nausea | 3 | 4
Fatigue | 12 | 7
Vomiting | 9 | 7
Neutropenia | 43 | 16
ILD (adjudicated) | 168 | 85
Interpreting dose-pool differences
Caution in comparison: Pools differ in size, age, sex, tumor type and treatment duration; no statistical comparison was made.
Sex and tumor effects: Vomiting higher in the 5.4 mg/kg pool (94.7% female); mGC symptoms may overlap with GI TEAEs.
Older patients: ≥65 years had more grade 3–4 AEs at 5.4 mg/kg (mainly mBC); similar by age at 6.4 mg/kg.
Renal impairment: Moderate renal impairment linked to higher low-grade ILD/pneumonitis — warrants closer monitoring.
Days - Nausea, vomiting, fatigue — first cycle
Weeks - Neutropenia — cycles 1–3
Months - ILD — monitor throughout treatment
Park YH, et al. The Oncologist. 2026;31:oyal396 (Results; Tables S5–S8; Table 2; Discussion)
Yeon Hee Park
TRIDENTE phase II: encorafenib + binimetinib + cetuximab rechallenge in BRAF V600E mCRC (n=21).
Met the study primary endpoint
ORR 19% (3/4 responses unconfirmed)
DCR 76%
#ESMO26 #ESMOAmbassadors https://t.co/uES7dA9CLz
[Slide 4]
cBioPortal for Cancer Genomics
New feature: cBioChat – chat.cbioportal.org
[Left panel: cBioChat UI]
Beta — Provide feedback with thumbs up/down below each response
cBioChat
Ask cBioPortalChat to explore cBioPortal data, build direct cBioPortal links, and answer questions about studies, genes, mutations, clinical attributes, and patient-level results.
Learn more
What is the frequency of PIK3CA mutations across cancer types in the TCGA PanCancer Atlas studies?
Haiku 4.5
20 msgs / 30 min · 20M tokens / week
Hide examples
Explore Data - Find studies, datasets, and available molecular data in cBioPortal.
Navigate cBioPortal - Create direct links to cBioPortal views, cohorts, plots, and study pages.
Analyze Data - Compare genes, cancer types, molecular profiles, cohorts, and outcomes.
LibreChat v0.8.7 - Every AI for Everyone. | Terms of service
[Right panel: response]
PIK3CA Mutation Frequency Across TCGA PanCancer Atlas Cancer Types
PIK3CA mutations show striking variation across cancer types in the TCGA PanCancer Atlas, ranging from 0.8% to 48.6% at the sample level:
Top 10 Cancer Types by PIK3CA Mutation Frequency
Cancer Type | Mutated Samples | Profiled Samples | Frequency
Endometrial Cancer | 279 | 574 | 48.6%
Breast Cancer | 347 | 1,066 | 32.6%
Cervical Cancer | 83 | 291 | 28.5%
Colorectal Cancer | 147 | 534 | 27.5%
Bladder Cancer | 90 | 410 | 22.0%
Head and Neck Cancer | 90 | 515 | 17.5%
Esophagogastric Cancer | 88 | 618 | 14.2%
Glioblastoma | 38 | 397 | 9.6%
Glioma | 42 | 514 | 8.2%
Non-Small Cell Lung Cancer | 81 | 1,050 | 7.7%
Bottom 5 Cancer Types
PIK3CA mutations are rare or absent in: Renal Non-Clear Cell Carcinoma (0.9%), Adrenocortical Carcinoma (1.1%), Prostate Cancer (2.0%), Pleural Mesothelioma (2.3%), and Pancreatic Cancer (2.8%).
How can I help you explore cBioPortal?
Haiku 4.5
LibreChat v0.8.7 - Every AI for Everyone. | Terms of service
Explore the Data
View PIK3CA mutations across all TCGA PanCancer Atlas studies on OncoPrint — shows sample-by-sample mutation status with integrated clinical tracks and copy number data.
Amazing work by @nikolausschultz , his lab & the cBioPortal team! Making cancer genomics accessible and now queryable with cBioChat.
https://t.co/7kMyCfRxeB
#ESMO26 #ESMOAmbassadors https://t.co/zc1XQ75Xo0
[Slide 2]
cBioPortal for Cancer Genomics
New feature: cBioChat – chat.cbioportal.org
Ability to connect your LLM of choice to cBioPortal MCP server
Immunotherapy Patients Over Time (2015–2024)
Year | Total Sequenced Patients | Received Immunotherapy
2015 | 7,234 | 852
2016 | 6,252 | 947
2017 | 8,174 | 1,157
2018 | 9,231 | 1,278
2019 | 11,426 | 1,703
2020 | 9,468 | 1,529
2021 | 10,448 | 1,739
2022 | 10,316 | 1,702
2023 | 11,449 | 1,907
2024 | 11,632 | 1,913
Top Immunotherapy Agents & Their Trends
Agent | Class | 2015 Patients | 2024 Patients | Trend
Pembrolizumab | PD-1 inhibitor | 386 | 1,070 | Strong growth
Nivolumab | PD-1 inhibitor | 424 | 520 | Moderate growth
Ipilimumab | CTLA-4 inhibitor | 193 | 208 | Relatively stable
Durvalumab | PD-L1 inhibitor | 7 | 269 | Rapid rise
Atezolizumab | PD-L1 inhibitor | 49 | 60 | Relatively flat
[Slide 3]
cBioPortal for Cancer Genomics
Next: tighter integration of the chat into the graphical interface
1 Ask cBioChat
"Survival by KEAP1 in KRAS-mutant lung adeno on pembrolizumab"
Cancer: Lung adeno | KRAS
Treatment: Pembrolizumab
Overall Survival - 538 patients
[KM plot: not mutated vs KEAP1 mutated; 100% to 0%; Months 0 to 105]
KEAP1-mutated patients do worse: median 9 vs 25 mo, p = 1.5×10^-8
2 Refine query via user interface
Clicks Stage IV in the stage chart
filter added Stage: IV
Stage (Highest Recorded)
I–III 176
IV 362
The chat sees the new filter and uses it for follow-ups.
3 Ask a follow-up
"Does KEAP1 still matter?"
Cancer: Lung adeno | KRAS
Treatment: Pembrolizumab | Stage: IV
Overall Survival - 362 patients
[KM plot: not mutated vs KEAP1 mutated; Months 0 to 105]
Yes, still worse: median 8 vs 21 mo, p = 4×10^-6
[Slide 4]
OncoKB updates
Debyani Chakravarty, Sarah Suehnholz, Moriah Nissan, Ritika Kundra, Nicole Fernandez, Kelly Cavender, Kinisha Gala
John Konecny, Calvin Lu, Benjamin Preiser, Reshma Ramaiah, JJ Gao
1033 Genes
11515 Alterations
143 Cancer Types
171 Drugs
Level 1 - FDA-approved drugs - 59 Genes
Level 2 - Standard care - 33 Genes
Level 3 - Clinical evidence - 38 Genes
Level 4 - Biological evidence - 32 Genes
Level R1 - Standard care - 8 Genes
Level R2 - Clinical evidence - 11 Genes
Automated annotation of MSK-IMPACT reports
Integrated into cBioPortal
Free academic licenses, fee for commercial
New: Integration into Open Evidence
New: Annotation of germline variants (also in cBioPortal)
Diana Mandelker, Zsofia Stadler
Bo Kim (UI/UX in DigITs)
[Screenshot: BRCA2 gene page]
BRCA2 Also known as FANCD1, XRCC11, FAD1
Tumor Suppressor
NCBI Gene: 675 | Show additional gene information
BRCA2, a tumor suppressor involved in the DNA damage response, is mutated in various cancer types. Pathogenic germline mutations in BRCA2 cause autosomal dominant BRCA2-associated Hereditary Breast and Ovarian Cancer and confer an increased risk for female and male breast, ovarian, pancreatic, and prostate cancers, as well as other cancer types.
Somatic | Germline
Gene Overview | Genomic Indicators | Annotated Variants
Genetic Risk - Penetrance: High
Annotated Variants - Pathogenic: 282 | Likely Pathogenic: 28
Genomic Indicators
Genomic Indicator | Inheritance | Associated Variants | Description
BRCA2-associated hereditary breast and ovarian cancer syndrome | Autosomal Dominant | Pathogenic Variants | A single variant in BRCA2 causes hereditary breast and ovarian cancer (HBOC) syndrome, which is associated with an increased risk for breast, ovarian, prostate, pancreatic, and other cancers.
Slides not tied to a tracked trial
Not tied to a single tracked trial
#ESMO26
Optimal duration of adjuvant endocrine therapy: 5, 7–8, or 10 years?
From Dr. Fumikata Hara’s excellent discussion👇 https://t.co/BHXSGesayf
[Slide 1]
CLINICAL APPLICATIONS
HOW LONG? COMPARE THE RANDOMIZED TREATMENT DURATIONS
TRIAL | COMPARE | SHORTER / LONGER ET | HR (95% CI)
B-42 | 5 vs 10 y (Placebo vs LET 5 y) | Prior AI 5 y or TAM <=3 y -> AI (total 5 y) | DFS 0.85 (0.76-0.94)
AERAS | 5 vs 10 y (Stop vs ANA 5 y) | Prior ANA 5 y or TAM 2-3 y -> ANA 2-3 y | DFS 0.61 (0.46-0.82)
GIM4 | 5 vs 7-8 y (LET 2-3 y vs 5 y) | Prior TAM 2-3 y | iDFS 0.78 (0.65-0.93)
DATA | 5-6 vs 8-9 y (ANA 3 y vs 6 y) | Prior TAM 2-3 y | Adapted DFS 0.86 (0.72-1.01)
IDEAL | 7.5 vs 10 y (LET 2.5 y vs 5 y) | Prior ET 5 y (TAM / AI / sequence) | DFS 0.92 (0.74-1.16)
ABCSG-16 / SALSA | 7 vs 10 y (ANA 2 y vs 5 y) | Prior ET 5 y (TAM / AI / sequence) | DFS 0.99 (0.85-1.15); SALSA: 10 y had more fractures (6.3% vs 4.7%); HR 1.35
[Bar diagram legend: Prior ET; AI in both arms; Additional AI in longer arm; R = Randomization]
Extension beyond 5 years can help, but 10 years is not necessary for everyone
B-42: Mamounas et al. ESMO 2026, LBA24. AERAS: J Clin Oncol. 2023;41:3329-3338. GIM4: Lancet Oncol. 2021;22:1458-1467. DATA: EClinicalMedicine. 2023;58:101901. IDEAL: J Natl Cancer Inst. 2018;110:40-48. SALSA: N Engl J Med. 2021;385:395-405. DFS definitions and analysis populations differ across trials. Bars approximate where prior TAM was 2-3 y.
[ESMO Congress Madrid 2026 logo]
These Kaplan-Meier curves finally separate… by about the width of a pixel...
The curves ride on top of each other for years before a modest late subatomic divergence, and then the blue arm takes a dramatic vertical plunge at the far righ, right about the time the numbers at risk collapse to single digits. Truly a triumph of statistical significance over clinical drama.
Novartis is already calling a ~20% relative reduction “clinically meaningful.” Somewhere a health economist is quietly calculating the cost per life-year for treating 5,100 patients to avert ~27 deaths.@5_utr
[Slide 1]
93.9% (Δ=1.2%) 91.9% (Δ=2.1%a)
92.7% 89.7%
Overall survival (%)
Median follow-up: 66.5 months
RIB + NSAI (n = 2549) | NSAI alone (n = 2552)
No. of OS events(b): 178 | 205
HR (95% CI): 0.809 (0.661-0.989)
Nominal P value (1-sided)(c): 0.019
Time since randomization (Months)
No. of patients at risk
RIB + NSAI: 2549 2404 2336 2299 2258 2219 2184 2132 2086 2047 1982 1521 761 85 2 0
NSAI alone: 2552 2301 2254 2208 2159 2117 2064 2012 1981 1936 1866 1426 730 81 3 0
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: Overall survival
RIB + NSAI | 2549
NSAI alone | 2552
HR: 0.809
HR CI: 0.661-0.989
P: 0.019 (nominal, 1-sided)
{'timepoint': '60 months', 'arm': 'RIB + NSAI', 'value': '93.9%', 'ci': None}
{'timepoint': '60 months', 'arm': 'NSAI alone', 'value': '92.7%', 'ci': None}
{'timepoint': '72 months', 'arm': 'RIB + NSAI', 'value': '91.9%', 'ci': None}
{'timepoint': '72 months', 'arm': 'NSAI alone', 'value': '89.7%', 'ci': None}
Note: Median follow-up 66.5 months; deltas 1.2% at 60 mo and 2.1% at 72 mo printed; medians not printed.
Opening Ceremony at @myESMO #ESMO26 with the ESMO Award Ceremony
#NoemiReguart, chair of Nomination Committee presents all awardees
Congrats to all 👏🏻👏🏻👏🏻
#ESMOAmbassadors https://t.co/g2tMu7XVHP
[Slide 1]
2026 ESMO SOCIETY AWARDS
ESMO Award — Rebecca Fitzgerald
ESMO Award for Translational Research — Dan A. Landau
ESMO Women for Oncology Award — Pilar Garrido
ESMO Lifetime Achievement Award — Antoni Ribas
[MADRID 2026 ESMO congress logo; panelists on stage below]
Promising data presented by @AndreaNecchi at #ESMO26 #GDFather-NEO.
Nivo + Visugromab
• Higher response (pCR and ORR)
• At median F/U 15.7 mo: no recurrence observed after pCR
• No new safety signal
Paper out in @NatureMedicine. Congrats Andrea and co-authors !
@myESMO @OncoAlert
[Slide 2]
Conclusions
Main conclusions
• Trial context
Exploratory, stratified phase 2 trial; first evaluation of GDF-15 neutralization in a checkpoint-inhibitor-naïve neoadjuvant setting
• Higher response with N+V
Approximately fourfold higher locally assessed pCR and ORR, confirmed by central blinded review
• No recurrence after pCR · bladder preservation
At a median follow-up of 15.7 months, no recurrence was observed after pCR; four N+V patients with pCR retained their bladder after re-TURBT alone
• Exploratory biomarker signal
Response to N+V was associated with a pre-existing proliferative CD8+ T-cell state; PD-L1 CPS and baseline serum or tumor GDF-15 did not distinguish responders from non-responders
• No new safety signal
Grade 3–4 TRAEs and immune-related AE rates were similar between arms; no treatment-related discontinuation, death or surgical delay occurred with N+V
These findings support further randomized evaluation of visugromab with checkpoint inhibition across MIBC and other urothelial cancer settings
We thank all participating patients and their families, and all investigators and site staff.
Prof. Andrea Necchi
Ongoing visugromab clinical development
GDFATHER-NSCLC-01 (NCT[illegible]) Treatment-naive non-squamous NSCLC, randomized phase 2b
GDFATHER-NSCLC-02 (NCT[illegible]) Checkpoint-inhibitor-refractory NSCLC, randomized phase 2b
GDFATHER-HCC-01 (NCT[illegible]) Checkpoint-inhibitor-refractory HCC, randomized phase 2b
VINCIT (NCT[illegible]) Cancer cachexia, placebo-controlled phase 2/3
[Slide 3]
nature medicine
Article
https://doi.org/10.1038/s41591-026-04740-9
Neoadjuvant nivolumab with or without visugromab in muscle-invasive bladder cancer: the phase 2a GDFATHER-NEO trial
Received: 7 September 2026
Accepted: 6 October 2026
Andrea Necchi, Chiara Mercinelli, Brigida A. Maiorano, Michela Piacentini, Antonio Cigliola, Roberto Iacovelli, Chiara Ciccarese, Chiara Ligato, Patrizia Giannatempo, Alessandro Rametta, Paolo Gontero, Francesco Soria, Christine Schuberth-Wagner, Joerg Wischhusen, Bernhard Rössler, Michael Metterlein, Stefan Ries, Thorsten Ross, Aasim Amin, Tobias Dreker, Frank Hermann, Lena Lemke, Kathrin Klar, Olga Mitina, Ana Kamova, Petra Fettes, Alexandra Auckenthaler, Sujata Rao, Eugen Leo & Felix S. Lichtenegger
e-mail: necchi.andrea@hsr.it
DOI: 10.1038/s41591-026-04740-9
URL: https://www.nature.com/articles/s41591-026-04740-9
QR code:
MADRID 2026 ESMO congress
Full house for the opening of #ESMO26 with @myESMO president @FAndreMD…looking forward to important days of science, networking and above all progress in #cancer care ! 💪
@OncoAlert @UniGenova https://t.co/uHFJDtL5yA
[Slide 1]
[Wide photo of packed main hall; multiple screens show the presidential address slide alongside the speaker]
Supporting better policy and healthcare systems
Revision of In Vitro Diagnostic Medical Devices Regulation (IVDR)
EU Biotech Act
Discussions on healthcare workforce challenges
Advocating for equitable access to financial services for cancer survivors (Right to be forgotten)
Thank you very much @myESMO for the opportunity to present on how to read and interpret cancer clinical trials today at the MCBS session. I was encouraged with the excellent turn out and hope that these lessons can be applied to the trials coming out of #ESMO26. https://t.co/Blk44RdIAu
[Slide 3]
[Photo of stage with slide]
Editorial
Improved adverse event reporting: it's about time
Laure-Anne Teuwen, MD, PhD(1,2), Nathan Cherny, MD(3), Bishal Gyawali, MD, PhD*(4,5,6)
Box: Recommendations for improved adverse event reporting in clinical trials
1. Avoid the use of subjective downplaying terms to describe toxicities.
2. Capture toxicity grades, ideally using the Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events.
3. Report maximum severity as well as onset, duration, and adverse event load.
4. Report dose reductions and treatment interruptions.
5. Report incidence of serious and fatal adverse events.
6. Report quality-of-life impact based on validated tools.
We at @getne should be very proud of what the up-and-coming young generation is achieving. A magnificent presentation of the #CABOTHYROID study findings by @JHernando3 at #ESMO26
@myESMO @OncoAlert https://t.co/O4KoXuAFN9
[Slide 1]
[partially cropped] ...2216 / CABOTHYROID Trial
STUDY DESIGN
[GETNE logo]
Baseline: Advanced RAIR DTC after PD to standard MKIs, N = 41
Treatment Phase (cycles Q4W): Cabozantinib 60 mg QD (24 weeks)
EOT: Safety visit 30 d after last dose
New treatment - Follow-up: until PD Q12W
Long-term follow-up: until withdrawal, death, or end of study
Survival status, Safety
CT or MRI scan and tumor markers Q12W
Blood sample cfDNA (baseline, during treatment, EOT, new treatment)
QoL → QoL Q12W and PD
Blood sample cfDNA → Methylation → Sequencing (EM-seq) → Methylation-pattern analysis (Machine Learning (ML) analysis) → Subgroup efficacy analysis
[ESMO Congress Madrid 2026 logo]
[Stage photo: three panel chairs seated, presenter at podium]
Cancer Immunotherapy- what's next ? Nice slide to explain progress so far by Dr Vinod Balachandran @myESMO #esmo26 https://t.co/TbS6xFT4je
[Slide 1]
15:15 - 16:00 What oncologists should know about mRNA vaccines
CHAIRS: YELENA JANJIGIAN, JAMES LARKIN
Cancer Immunotherapy – What's Next?
1893 — Infection as a way to fight cancer (William Coley)
1957 — The immune system can see and destroy abnormal cells (Thomas & Burnet)
1959 — BCG (TB vaccine) triggers immune response against tumors in mice (Old & Clarke)
1975 — Tumor necrosis factor (TNF) discovered (Old, Carswell, et al.)
1976 — IL-2 discovered (Gallo)
1986-1990 — First Cancer Immunotherapies BCG IFN-a2
1992 — Discovery of PD-1 (Honjo)
1996 — CTLA-4 blockade rejects murine tumors (Allison)
2011 — First immune checkpoint inhibitor FDA approved (Wolchok)
2014 — First PD-1/PD-L1 inhibitors FDA approved
2017 — First CAR T cells FDA approved (Sadelain, June)
What's next?
Why NOT Cancer Vaccines??
[OCCV logo — The Olayan Center for Cancer Vaccines, Memorial Sloan Kettering Cancer Center]
[Speaker panel] Vinod Balachandran — RNA vaccines for cancer: Principles to practice
ALICANTE AUDITORIUM - HALL 6
[ESMO Congress Madrid 2026 logo]
My first ESMO oral! 🇪🇸
Our study: immunotherapy around cancer surgery becomes standard of care before we know if patients live longer, mature survival data should come first.
Grateful to my mentors and my family ❤️ #ESMO26 @oncology_bg @dr_rajapramanik @gupta_aarushi https://t.co/PykpnULhPi
[Slide 1]
Background
Why does mature OS data matter for perioperative ICI trials?
• Perioperative immune checkpoint inhibitors have become standard of care across a variety of solid tumors based on large phase three studies
• All studies use EFS/DFS as primary end point- validation of these surrogate end points for overall survival (OS) in this setting is unknown
• ICI's can have permanent long-term toxicity which is especially relevant in the curative intent setting
• Mature survival data is essential to gauge improvement in cure rates with ICI's and counsel patients on risks vs benefits
Abhenil Mittal
[Podium sign: MADRID SPAIN 23-27 OCTOBER 2026]
#ESMO2026 | Drugging the Undruggable: A New Era in Precision Oncology
“Undruggable” was never about the target. It was about our tools.
An inspiring presentation by Dr. Vivek Subbiah (@VivekSubbiah) at ESMO 2026, highlighting how previously untargetable cancer drivers are becoming therapeutically actionable.
From KRAS inhibition and p53 reactivation to MYC targeting, protein degradation, synthetic lethality, and AI-driven drug discovery, precision oncology is entering a new era.
The next challenge is not only developing new drugs, but identifying the right molecular target in the right patient.
Comprehensive genomic profiling and ctDNA testing will play an increasingly important role in this transformation.
#ESMO26 @myESMO @VivekSubbiah #PrecisionOncology #UndruggableTargets #KRAS #TP53 #GenomicTesting #LiquidBiopsy @Dr_R_Kurzrock
[Slide 1]
MADRID 2026 ESMO congress
The undruggable journey is the journey of Odyssey
From bold ideas to breakthrough therapies: the evolution of KRAS inhibitors
Matt Damon in The Odyssey
Photograph: Landmark Media/Alamy
Timeline of KRAS Inhibitors
Discovery -> Development
1964 • Retroviral isolates were identified (also in 1967, 1974 and 1978)
1969 • Oncogene
1972 • DNA transfection to convert normal cells to cancer cells
1982 • The first human transforming gene was cloned • The human HRAS oncogene was found to be activated by point mutation • Human oncogenes were found to be related to v-ras oncogenes
1995 • Farnesyltransferase inhibitors were found to block Ras-induced tumours in mice
2013 • KRAS mutation becomes druggable
2019 • KRAS G12C inhibitors in clinical trials (Sotorasib, Adagrasib)
2021 • The first KRAS G12C inhibitor approved by US FDA (Sotorasib)
2022 • KRAS G12C, KRAS G12D and other KRAS inhibitors in trials
Vivek Subbiah, MD | X/Twitter @VivekSubbiah
J Hematol Oncol 17, 108 (2024)
Very promising randomized results with the CDK4 selective inhibitor atirmociclib presented by @antgiorda. Next gen CDKs are here! #ESMO26 #ESMOAmbassadors https://t.co/XuIiabKA5H https://t.co/YuDTbHahRg
Same slide also posted by @DrBarbiOnc, @E_de_Azambuja
[Slide 1]
Investigator-Assessed PFS
Atirmociclib arm (n = 126) | ICT arm (n = 137)
Median follow-up, months | 13.7 | 13.6
Events | 72 | 93
Median PFS, months (95% CI) | 9.4 (7.3–13.6) | 6.6 (3.8–7.8)
Stratified HR (95% CI) | 0.60 (0.440–0.825)
one-sided P value | 0.0007
Landmark labels: Atirmociclib 82.5% (3 mo), 62.3% (6 mo), 44.8% (12 mo); ICT 68.9% (3 mo), 51.9% (6 mo), 22.5% (12 mo)
No. at risk
Atirmociclib arm 126 120 100 98 83 83 72 71 64 63 55 49 34 32 13 12 12 3 3 2 0
ICT arm 137 127 90 84 66 65 58 54 45 45 31 25 17 16 7 7 6 1 0 0 0
CI, confidence interval; HR, hazard ratio; ICT, investigator's choice of therapy; PFS, progression-free survival.
Presenter: Antonio Giordano
[MADRID 2026 ESMO congress logo]
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: Investigator-assessed PFS
Atirmociclib arm | 126 | 9.4 | 7.3–13.6
ICT arm | 137 | 6.6 | 3.8–7.8
HR: 0.60 (stratified)
HR CI: 0.440–0.825
P: 0.0007 (one-sided)
{'timepoint': '3 mo', 'arm': 'Atirmociclib', 'value': '82.5%', 'ci': None}
{'timepoint': '6 mo', 'arm': 'Atirmociclib', 'value': '62.3%', 'ci': None}
{'timepoint': '12 mo', 'arm': 'Atirmociclib', 'value': '44.8%', 'ci': None}
{'timepoint': '3 mo', 'arm': 'ICT', 'value': '68.9%', 'ci': None}
{'timepoint': '6 mo', 'arm': 'ICT', 'value': '51.9%', 'ci': None}
{'timepoint': '12 mo', 'arm': 'ICT', 'value': '22.5%', 'ci': None}
Note: Landmark % printed at dotted lines at 3, 6, 12 months. Median follow-up 13.7 vs 13.6 mo.
Esmo conference is always rich in data and discussion. 👍 @myESMO #esmo26 https://t.co/8nqJAtLmxx
Fantastic to see CNS oncology take centre stage with a keynote lecture at @myESMO meeting in Madrid! Anna Berghoff introducing Wolfgang Wick on glioblastoma, brain metastases and LMD disease: the next frontier in oncology care.
#ESMO2026 #NeuroOncology
@EANOassociation https://t.co/3suPeHbeG8
[Slide 2]
[ESMO Congress Madrid 2026 stage screen; live video of speaker at left]
Leptomeningeal Metastasis (LM)
• Incidence of LM in advanced malignancies to approximate 9%–25% for small cell and non-small cell lung cancer, 5%–20% for breast cancer, and 6%–18% for melanoma.
• Prognosis of LM is poor but heterogeneous, with certain subgroups of patients (eg, EGFR-mutant NSCLC, HER2-positive breast cancer) demonstrating superior survival.
• Historical survival benchmark of 2–4 months does not apply to all patients
PMID: 38902944
[Figure panels A–C: brain/spinal CSF circulation schematic; B: adherent LM cells, floating LM cells, brain parenchyma; C: choroid plexus, barrier disruption through secreted factors, LM cancer cells]
[ESMO Congress Madrid 2026 logo]
MADRID SPAIN 23-27 OCTOBER 2026
Great to present the results of the @GrupoGetne SPAINTRK study at #ESMO2026!
Our real-world data in spanish population support the efficacy and durability of larotrectinib across NTRK fusion-positive solid tumors.
Proud of this collaborative effort! https://t.co/dxJDljN9op
[Slide 2]
Real-world outcomes of Larotrectinib in Solid Tumors with Tropomyosin Receptor Kinase Fusion (GETNE S2411 / SPAINTRK)
Jorge Hernando 1,2, Carlos Lopez 3, Alejandro Garcia-Alvarez 1,2, Santiago Aguín Losada 4, Olga Martínez-Sáez 5, Carlos Gonzalez-Perez 6, Ricardo López-Almaraz 7, Raúl Sánchez Morillas 8, Joseba Rebollo Liceaga 9, Laura Ferreira Freire 10, Miriam Pavon-Mengual 11, María Ruiz Vico 12, Miriam López-Gómez 13, Idoia Morilla 14, Delvys Rodriguez-Abreu 15, Jaume Capdevila 1,2
1070P
[Affiliations: Vall d'Hebron (Barcelona), Valdecilla (Santander), Santiago de Compostela, Hospital Clínic (Barcelona), La Paz (Madrid), Cruces/Biobizkaia (Barakaldo), Son Espases (Palma), Quirón Salud Torrevieja, Lucus Augusti (Lugo), Sant Joan de Déu (Barcelona), 12 de Octubre (Madrid), Infanta Sofía (San Sebastián de los Reyes), Navarra (Pamplona), Insular de Gran Canaria — small print]
[GETNE logo] Grupo Español de Tumores Neuroendocrinos y Endocrinos
INTRODUCTION
NTRK gene fusions have been described as oncogenic drivers in diverse adult and paediatric cancers, promoting cell survival and proliferation 1,2. Larotrectinib is a first-in-class selective tropomyosin receptor kinase (TRK) inhibitor with proven activity in a broad range of solid tumors 3. This study aimed to describe the effectiveness of larotrectinib in patients with solid tumors harbouring NTRK fusions in Spain 4.
METHODS
SPAINTRK (NCT06837090) was a retrospective observational study including adult and pediatric patients with solid neoplasms bearing NTRK fusions, treated with larotrectinib through compassionate use program, between European Medicines Agency (EMA) approval and commercialization in Spain (Oct 19-Sept 23) (Figure 1). Larotrectinib as part of clinical trials was excluded. TRK fusions were determined as part of standard care using next-generation sequencing (NGS), fluorescence in situ hybridization and/or immunohistochemistry, along with confirmatory molecular test.
The primary endpoint was duration of response (DoR). The secondary endpoints were objective response rate (ORR) and safety. No formal sample size was calculated.
Figure 1. Study Schedule & Milestones [timeline 4Q 2024 – 1Q 2026]
OBJECTIVES
Primary
- Description of the effectiveness of larotrectinib treatment in tumors with NTRK fusion in Spanish patients as a clinical series.
Secondary
- Describe duration of treatment, dose reductions, interruptions, and reasons for such modifications.
- Description of the efficacy of larotrectinib according to previous lines of therapy.
- Line of treatment at which molecular testing for NTRK was performed.
- Safety assessment of larotrectinib.
Exploratory
- The potential prognostic value of clinical and/or histological variable regarding the effectiveness and tolerability of larotrectinib.
REFERENCES
1. Cocco et al. Nt Rev Clin Oncol 2018 2. Bazhenova et al. Target Oncol 2021 3. Drilon et al. N Engl J Med 2018 4. Hernando et al. EJC 2026
Clinicaltrials.gov: NCT06837090 [QR code]
RESULTS
Table 1. Patient characteristics (n=20)
Age at treatment start, Median (range); years: 37 (0.83-81)
Sex; n (%): Male 10 (50); Female 10 (50)
Race; n (%): Caucasian 14 (70); Latin 3 (15); Asian 1 (5); North African 1 (5); Unknown 1 (5)
ECOG-PS n (%): 0 7 (35); 1 9 (45); 2 3 (15); Unknown 1 (5)
Tumor type; n (%): Infantile fibrosarcoma 3 (15); Infant desmoplastic ganglioglioma 1 (5); Glioblastoma 5 (25); Lung* 5 (25); Breast tumor 2 (10); Thyroid tumor 2 (10); Neuroendocrine Carcinoma 1 (5); Head and neck tumor 1 (5)
Disease status at enrollment; n (%): Locally advanced 9 (45); Metastatic 11 (55)
NTRK fusion gene; n (%): NTRK1 8 (40); NTRK2 5 (25); NTRK3 7 (35)
Previous systemic treatment lines**; n (%): 0 6 (30); 1 7 (35); 2 4 (20); ≥3 3 (15)
*Four patients had NSCLC and one patient had lung-NET. **Includes adjuvant
Table 2. Overall Response Rate (ORR)
Full dataset ORR; % (95% CI): 60 (36.1-80.9)
Best response; n (%): Complete response 4 (20); Partial response 8 (40); Stable disease 4 (20); Progressive disease 3 (15); Not evaluable 1 (5)
Figure 2. Duration of response among 12 patients with a Response calculated from the time of first response [KM curve: labels 83.3%, 66.7%, 50%; Number at risk 12, 10, 8, 6, 5, 4, 3, 3, 3, 2, 2, 1 at 0–66 months]
Table 3. Treatment-related adverse events (AEs). No patient presented grade 4-5 AE
Event; n (%) — Grade 3 / Any grade
Transaminitis 2 (10) / 3 (15); Neutropenia 2 (10) / 3 (15); Anaemia 0 (0) / 2 (10); Fractures 1 (5) / 2 (10); Dysgeusia 0 (0) / 1 (5); Skin toxicity 0 (0) / 1 (5); Acute hepatic failure 0 (0) / 1 (5); Dizziness 0 (0) / 1 (5); LDH increased 0 (0) / 1 (5); ALP increased 0 (0) / 1 (5)
Between Feb-Jun25, 20 pts from ten months to 81 years were included, 15 adults and 5 paediatrics, with eight different solid tumor types. NTRK fusions were reported in NTRK1 gene (40%), NTRK2 (25%) and NTRK3 (35%), 65% of pts were diagnosed using NGS Table 1. Larotrectinib was administered for a median of 13 months (95% CI: 8-21) and 70% had received previous treatments.
After a median follow-up of 24.4 months (95% CI: 13.2-35), the median DoR was 24.5 months (95% CI: 11.1- not reached) (Figure 2) and ORR was 60% (95% CI: 36.1-80.9) (Table 2). ORR was higher among pediatric pts (80%; 95% CI: 28.4-99.5). The highest ORR were seen among pts with infant fibrosarcoma (100%; 95% CI: 29.2-100) and lung tumors (80%; 95% CI: 28.44-99.5). After 1 year of treatment, 60% of pts remained progression-free and 75% maintained response (Figure 3). At data cutoff, 41.7% of the pts with a response were disease-free and/or remained on treatment. Toxicities were uncommon, with neutropenia and transaminitis being the most frequent, reported in 15%, each (Table 3).
Figure 3. Effectiveness of larotrectinib. Left panel shows treatments before larotrectinib. The right panel (squared) shows larotrectinib treatment and outcomes [swimmer plot by NTRK fusion and tumor type: Infant Fibrosarcoma, Head & Neck, Lung, Breast, Infant Desmoplastic ganglioglioma, Thyroid, Glioblastoma, Neuroendocrine]
CONCLUSIONS
❖ Larotrectinib evoked broad and durable antitumor activity in a plethora of solid tumors with NTRK fusions.
❖ Durable responses were observed irrespective of age, tumor type, and NTRK fusion gene.
❖ Safety profile was consistent with that of clinical trials, even after long-term administration.
❖ NGS is implemented in most hospitals and its use should keep expanding.
*Corresponding author: Dr. Jorge Hernando. E-mail: [email], Medical Oncology Department, Vall Hebron University Hospital and Vall Hebron Institute of Oncology (VHIO), Barcelona, Spain. ESMO 2026 Annual Symposium // Conflicts of interest: The presenting author has no conflict of interest regarding this communication // Funding: Grupo Español de Tumores Neuroendocrinos y Endocrinos (GETNE). // Partners: Bayer // Acknowledgement: GETNE acknowledge MFAR Clinical Research staff for their assistance in the development of this poster.
#ESMO26 🧬 Temab-A + BEV | Phase I
👥 83 pts | Refractory mCRC (3L+)
💊 c-MET ADC + BEV vs FTD/TPI + BEV
📊 Temab-A 2.4 mg/kg vs SOC:
🎯 ORR: 30% vs 0%
📈 mPFS: 6.8 vs 4.2 mo (HR 0.48)
⏳ mOS: 14.7 vs 10.2 mo (HR 0.62)
⚠️ G≥3 neutropenia: 33%
🚀 Promising activity! Phase III ongoing.
@OncoAlert
[Slide 1]
14:45 - 16:15 Rapid oral: Gastrointestinal tumours, lower digestive
CHAIRS: LISA SALVATORE, KAI-KEEN SHIU, RONA YAEGER
Combination of Temab-A + Bev in Patients With 3L+ CRC in the First-in Human Study (NCT05029882)
Key Inclusion Criteria (Part 7)
• Adult patients with metastatic CRC, irrespective of c-Met protein expression level
• Progression on fluoropyrimidine, oxaliplatin, and irinotecan, and if applicable, anti-EGFR or anti-VEGF monoclonal antibody, or other targeted therapy
• Measurable disease per RECIST v1.1
• ECOG PS 0–1
• Hemoglobin ≥ 8 g/dL (with no RBC transfusion in the past 14 days)
Key Exclusion Criteria (Part 7)
• BRAF V600E mutation
• dMMR+/MSI-H
• Prior regorafenib or FTD-TPI treatment
Part 7: Temab-A and Bev Combination Therapy
Safety Lead-In (Part 7a): DL1 (1.6 mg/kg; n=7) | DL2 (2.4 mg/kg; n=8) | Backfill DL3 (2.0 mg/kg; n=5)
Temab-A + Bev 7.5 mg/kg IV Q3W
Dose Optimization (Part 7b): R (1:1:1) N=66
Arm 1a Temab-A 2.0 mg/kg Q3W + Bevb (n=22)
Arm 2 Temab-A 2.4 mg/kg Q3W + Bevb (n=22)
Arm 3 Comparator (SOC)c FTD-TPI + Bevd (n=22)
Primary Objective (Part 7)
• Evaluate the safety, efficacy, and pharmacokinetics of Temab-A in combination with Bev vs FTD-TPI + Bev in patients with 3L+ CRC
Data cutoff date: 02March2026. aOne patient withdrew and was not treated. b7.5 mg/kg Q3W. cTwo patients withdrew and were not treated. d5.0 mg/kg Q2W.
3L+, third-line and beyond; Bev, bevacizumab; CRC, colorectal cancer; DL, dose level; dMMR, deficient DNA mismatch pair; ECOG PS, Eastern Cooperative Oncology Group performance status; EGFR, epidermal growth factor receptor; FTD-TPI, trifluridine/tipiracil; IV, intravenous; MSI-H, microsatellite instability-high; Q3W, every 3 weeks; R, randomization; RBC, red blood cells; RECIST, Response Evaluation Criteria in Solid Tumors; SOC, standard of care; Temab-A, telisotuzumab adizutecan; VEGF, vascular endothelial growth factor.
Michael Cecchini
Long-term follow-up of telisotuzumab adizutecan (ABBV-400, Temab-A) in combination with bevacizumab (Bev) vs standard of care (SOC) in patients (pts) with 3L+ colorectal cancer (CRC): Updated results of a phase 1 study
ALICANTE AUDITORIUM - HALL 6
[Slide 2]
14:45 - 16:15 Rapid oral: Gastrointestinal tumours, lower digestive
CHAIRS: LISA SALVATORE, KAI-KEEN SHIU, RONA YAEGER
TRAEs Were Primarily Hematologic and Gastrointestinal and Were Manageable
Most common TRAEsa
Temab-A 2.4 mg/kg + Bev (n=30): Grade ≥3 | Grade 1-2 || FTD-TPI + Bev (n=20): Grade 1-2 | Grade ≥3
Patients (%)
Anemia | 53 | 13 || 15 | 15
Neutropenia | 33 | 20 || 30 | 30
Thrombocytopenia | 10 | 23 || 30 | 5
Nausea | – | 57 || 50 | –
Fatigue | 7 | 37 || 35 | –
Vomiting | – | 40 || 25 | –
Decreased appetite | – | 30 || 20 | –
Diarrheab | – | 20 || 30 | 5
Management of TRAEs | Temab-A 2.4 mg/kg + Bev (n=30) | FTD-TPI + Bev (n=20)
Median treatment duration, months (range) | 6.1 (1.4–19.8) | 4.9 (0.9–14.0)
Treatment discontinuation, n (%) | 1 (3) | 1 (5)
Dose reduction, n (%) | 15 (50) | 6 (30)
Dose interruption, n (%) | 22 (73) | 9 (45)
• There were no TRAEs leading to death
• Adjudicated ILD/pneumonitis rate was 3% (Grade 3: n=1) among 30 patients who received 2.4 mg/kg Temab-A + Bev
Data cutoff date: 02March2026.
aAny grade reported in ≥20% of patients. bAll events of diarrhea with Temab-A were Grade 1.
Bev, bevacizumab; FTD-TPI, trifluridine/tipiracil; ILD, interstitial lung disease; Temab-A, telisotuzumab adizutecan; TRAE, treatment-related adverse event.
Michael Cecchini
Long-term follow-up of telisotuzumab adizutecan (ABBV-400, Temab-A) in combination with bevacizumab (Bev) vs standard of care (SOC) in patients (pts) with 3L+ colorectal cancer (CRC): Updated results of a phase 1 study
ALICANTE AUDITORIUM - HALL 6
[Slide 3]
14:45 - 16:15 Rapid oral: Gastrointestinal tumours, lower digestive
CHAIRS: LISA SALVATORE, KAI-KEEN SHIU, RONA YAEGER
Temab-A 2.4 mg/kg + Bev Shows Promising Efficacy Over FTD-TPI + Bev With a Median PFS of 6.8 Months
Median follow-up: 18.2 months (range 0.4–22.8)
PFS | Temab-A 2.4 mg/kg + Bev (n=30) | FTD-TPI + Bev (n=20)
Events, n (%) | 27 (90.0) | 18 (90.0)
Median PFS, months (95% CI) | 6.8 (5.4–9.5) | 4.2 (1.8–6.3)
HR (95% CI) | 0.48 (0.26–0.89)
Legend: Temab-A 2.4 mg/kg + Bev | FTD-TPI + Bev
Y-axis: Probability; X-axis: Time (Months) 0–20
Patients at Risk
Temab-A 2.4 mg/kg | 30 | 30 | 28 | 26 | 25 | 21 | 19 | 12 | 12 | 11 | 4 | 4 | 4 | 4 | 3 | 2 | 1 | 1 | 1 | 1 | 0
FTD-TPI | 20 | 19 | 15 | 12 | 11 | 9 | 7 | 4 | 4 | 2 | 1 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0
Data cutoff date: 02March2026.
Bev, bevacizumab; FTD-TPI, trifluridine/tipiracil; HR, hazard ratio; PFS, progression-free survival; Temab-A, telisotuzumab adizutecan.
Michael Cecchini
Long-term follow-up of telisotuzumab adizutecan (ABBV-400, Temab-A) in combination with bevacizumab (Bev) vs standard of care (SOC) in patients (pts) with 3L+ colorectal cancer (CRC): Updated results of a phase 1 study
ALICANTE AUDITORIUM - HALL 6
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: PFS
Temab-A 2.4 mg/kg + Bev | 30 | 6.8 months | 5.4–9.5
FTD-TPI + Bev | 20 | 4.2 months | 1.8–6.3
HR: 0.48
HR CI: 0.26–0.89
Note: Phase 1 (NCT05029882) Part 7 randomized dose optimization; median follow-up 18.2 mo (range 0.4–22.8); data cutoff 02 March 2026. No p-value printed.
[Slide 4]
14:45 - 16:15 Rapid oral: Gastrointestinal tumours, lower digestive
CHAIRS: LISA SALVATORE, KAI-KEEN SHIU, RONA YAEGER
Conclusions
Temab-A 2.4 mg/kg + Bev showed sustained clinically meaningful improvements in efficacy outcomes, including ORR (30%), median PFS (6.8 months), and median OS (14.7 months; HR=0.62)
The safety profile was manageable with low discontinuation rates; hematologic adverse events were managed with dose interruptions and reductions
The regimen is being evaluated in the global randomized Phase 3 trial AndroMETa-CRC-560 (NCT07525206)
For additional information, refer to the manuscript published concurrently in Annals of Oncology
Thank you to the patients, study sites, and investigators who participated in this clinical trial.
AbbVie funded this study and participated in the study design, research, analysis, data collection, interpretation of data, reviewing, and approval of the publication. All authors had access to relevant data and participated in the drafting, review, and approval of this publication. No honoraria or payments were made for authorship. Medical writing support was provided by Marianne Pons, PhD, and Phillip Giannopoulos, PhD, CMPP of Bio Connections, LLC, and funded by AbbVie.
Bev, bevacizumab; CRC, colorectal cancer; HR, hazard ratio; ORR, objective response rate; OS, overall survival; PFS, progression-free survival; Temab-A, telisotuzumab adizutecan.
Copies of this presentation obtained through Quick Response (QR) Code are for personal use only and may not be reproduced without permission from ESMO or the author of this presentation. Scan QR code to download an electronic version of this presentation.
To submit a medical question, please visit www.abbviemedinfo.com
Michael Cecchini
Long-term follow-up of telisotuzumab adizutecan (ABBV-400, Temab-A) in combination with bevacizumab (Bev) vs standard of care (SOC) in patients (pts) with 3L+ colorectal cancer (CRC): Updated results of a phase 1 study
ALICANTE AUDITORIUM - HALL 6
Nice to see the MASTERPLAN study presented #ESMO26 and published in @TheLancetOncol !
Congrats to the authors!
Our new paper :
"A new era in anti-RAS therapy: Unique challenges and opportunities for radiotherapy clinical trials"
with @free_radical28 is out in EJC, open access :
https://t.co/qBMfJOhCAF
Food for thoughts for ongoing and future trials in that space!👇
[Slide 1]
European Journal of Cancer 249 (2026) 117790
Contents lists available at ScienceDirect
European Journal of Cancer
journal homepage: www.ejcancer.com
[Elsevier logo] [EJC cover thumbnail] [Check for updates badge]
Current Perspective
A new era in anti-RAS therapy: Unique challenges and opportunities for radiotherapy clinical trials
Susannah G. Ellsworth a, Timothée Olivier b,c,*
a Department of Radiation Oncology, University of Pittsburgh Hillman Cancer Center, Pittsburgh, PA, USA
b Department of Medicine, University of Geneva (UNIGE), Geneva, Switzerland
c Department of Oncology, Geneva University Hospital, Geneva, Switzerland
@YukinoriOzaki presents the JCOG1204A1 of ctDNA analysis!
Congrats!
Looking forward to the primary result!
#ESMO26 https://t.co/ckn40jHZlK
[Slide 2]
Conclusions
• Within a randomized trial across intensive surveillance and standard of care arms, patients who remained ctDNA-negative throughout surveillance had 99.5% 5-year RFS, compared to 53.6% among ctDNA-positive patients
• Among initially ctDNA-negative patients, detection of ctDNA during surveillance preceded recurrence by a median lead time of 7.5 months overall and by 9.5 months in the intensive surveillance arm.
• These findings support integrating ctDNA surveillance into structured follow-up strategies, particularly for ER+/HER2- BC.
Yukinori Ozaki, M.D., Ph.D.
[ESMO Congress Madrid 2026 logo]
Preliminary clinical results of QTX3034, an oral G12D-preferring KRAS inhibitor with cetuximab ± mFOLFOX6 in KRASG12D-mCRC
#ESMO26 #ESMOAmbassadors
👉ORR: 25% vs 50%, DCR: 75 vs 90%
👉 no new safety signals
🧐Supports the combination with CTx
@myESMO @ASCO https://t.co/4KrcgfdNU2
Same slide also posted by @EvelynWongYT
[Slide 1]
Introduction
• KRASG12D-mutated mCRC remains an important unmet need
• QTX3034 is an investigational KRASG12D-preferring oral multi-KRAS inhibitor
• EGFR inhibitors +/- chemotherapy may improve response and delay onset of resistance to KRAS inhibition
• The initial Phase 1 monotherapy and combination data with cetuximab was previously presented, showing a combinable safety profile and no evidence of rash at QTX3034 monotherapy RP2D
• We present the pooled results of patients with mCRC who received QTX3034 at the RP2Ds of 1200mg BID + cetuximab and 800/1200 mg BID + cetuximab + FOLFOX
[Diagram] Cetuximab | EGFR | GRB2 | SOS1/2 | Upstream EGFRi boosts efficiency of RAS(OFF) trapping | OFF: Inactive RAS (GDP) | KRAS | ON: Active RAS (GTP) | QTX3034 | MAPK signaling | Blocking RAS(ON)-MAPK signaling releases compensatory feedback via EGFR
Dose-escalation: QTX3034 + cetuximab: 150 mg BID -> 1200 mg BID; QTX3034 + cetuximab + FOLFOX: 800 mg BID -> 1200 mg BID
Dose-expansion: QTX3034 + cetuximab 1200 mg BID
3
[Slide 2]
Efficacy – QTX3034 + cetuximab
QTX3034 + cetux demonstrated anti-tumor activity
QTX3034 1200 mg BID + cetuximab (N=20), N of responses (%)1
Legend: PD | SD | uPR | PR | CR | ▶ Ongoing
ORR | 5 (25)
PR | 5 (25)
CR | 0
DCR | 15 (75)
Rows under waterfall: Prior oxaliplatin | Prior irinotecan | Prior bevacizumab | Prior LOT
QTX3034 + cetuximab [spider plot: Difference in SLD from baseline vs Study day; 1200 mg]
• QTX3034 + cetuximab exhibited encouraging ORR, with 2.7-5.5 months duration
• Majority of responders had received prior FU, oxaliplatin, and irinotecan
1One patient non-evaluable due to withdrawal of consent, not related to study drug or disease
6
Clinical Cut-Off: Aug 28th, 2026
[Slide 3]
Efficacy – QTX3034 + Cetuximab + FOLFOX
Early signal of clinical activity in chemo-refractory and naïve patients
QTX3034 (800 / 1200 mg) + cetuximab + FOLFOX (N=10), N (%)1
Legend: PD | SD | uPR | PR | CR | ▶ Ongoing
ORR | 5 (50)
PR | 4 (40)
CR | 1 (10)
DCR | 9 (90)
QTX3034 dose | 1200 | 1200 | 800 | 1200 | 800 | 800 | 1200 | 1200 | 800 | 1200
Prior oxaliplatin | Prior irinotecan | Prior bevacizumab (shaded grid)
Prior LOT | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 1 | 0
QTX3034 + cetuximab + FOLFOX [spider plot: % Change in TL size vs Study day; 800 mg, 1200 mg]
• Encouraging ORR, in a predominantly chemo-refractory population and responses of 1.4 - 3.9 m
◆ Ongoing uPR was confirmed after clinical cut-off date
• *Complete Response confirmed in chemotherapy-naïve patient who subsequently elected for definitive surgery
1One patient non-evaluable due to poor compliance and subsequent study withdrawal not related to drug toxicity
7
Clinical Cut-Off: Aug 28th, 2026
[Slide 4]
Best change in ctDNA
Clinical responses supported by deep reductions in ctDNA
QTX3034 + cetuximab | N with ctDNA clearance > 90% | 9 (56%)
QTX3034 + cetuximab + FOLFOX | N with ctDNA clearance > 90% | 7 (70%)
Y-axis: KRASG12D VAF best change from baseline
Legend: 800 mg BID | 1200 mg BID
Study day | 43 | 15 | 43 | 15 | 15 | 15 | 15 | 15 | 15 | 43 | 43 | 15 | 15 | 15 | 15 | 15 | 43 | 43 | 15 | 43 | 43 | 43 | 43 | 43 | 43 | 43
BOR | SD | PD | SD | PD | PD | SD | SD | PD | PR | SD | PR | SD | SD | PR | SD | PR | SD | PD | CR | PR | SD | PR | PR | SD | SD | PR
8
Clinical Cut-Off: Aug 28th, 2026
🚨 iNATURE at #ESMO26 (LBA52): Median PFS 10.3 vs 3.1 mo, HR 0.29 in refractory metastatic disease. Striking! But before anyone says "personalized neoantigen vaccine works," let’s be careful on what this design can and can't tell us 🧵 https://t.co/IwjPgikuM1
[Slide 1]
LBA52 - Personalized neoantigen vaccine combined with precision critical lesion radiotherapy (CLERT) in advanced solid tumors: Updated results from the iNATURE phase II trial
Speakers FengMing P. Kong (Hong Kong, Hong Kong SAR, China)
Lecture Time 08:35 - 08:40
Abstract
@YJanjigianMD and #JamesLarkin on the main topics and data that will be presented at @myESMO #ESMO26 Annual Conference
🦠Cancer Vaccine - very hot topic!
💉ADCs in many settings
👀Follow the “What oncologist should know about” special sessions 🆕
#ESMOAmbassadors https://t.co/0IGPRev30q
[Slide 1]
ESMO 2026 IN NUMBERS
[microscope icon]
4,414 Abstracts:
• 12 in Presidential Symposia
• 147 Proffered Papers
• 242 Rapid Oral
• 2,025 Posters
• 1,988 ePosters
[presentation icon]
296 Sessions, of which:
• 230 Programme Sessions
• 66 Industry Satellite Symposia
[people icon]
~ 40,000
Professionals attending
[MADRID 2026 ESMO congress logo; live video of female presenter; slide partially repeated at right]
[Slide 2]
FROM DISCOVERY
TO CLINICAL IMPACT
What matters is not simply what is new.
It is what changes outcomes for patients.
1. DISCOVER
The biology that matters
→ 2. VALIDATE
In rigorous clinical trials
→ 3. DELIVER
Meaningful, accessible care
[MADRID 2026 ESMO congress logo; live video of female presenter]
[Slide 3]
THEMES ACROSS THE ESMO 2026 PROGRAMME
[globe icon] GLOBALISATION OF CANCER RESEARCH
[chart icon] TREATMENT OPTIMISATION
[shield icon] INTERCEPTION AND PREVENTION
[link icon] NEXT-GENERATION ADCs
[AI chip icon] ARTIFICIAL INTELLIGENCE
[hands/heart icon] NON-PHARMACOLOGICAL INTERVENTIONS
[MADRID 2026 ESMO congress logo; live video of female presenter]
Felt like a footballer in the tunnel before kick-off with @OncobitesTv ⚽
🎯 Our #ESMO26 poster 1988P: ADC + PD-1 blockade across solid tumors
14 trials | n=1,017 with initial benefit
Acquired resistance at 18 mo: 39-63%
➡️ CR rate did not predict it, unlike with anti-PD-1
Depth ≠ durability with ADC-IO?
#ADC #Immunotherapy
@_SEOM @OncoAlert @Larvol
💙 Oncologist wellbeing matters too.
📍 Our RESILIENCE-@_SEOM poster at #ESMO26, presented by @DrVilmaPBarcia: a structured program to prevent burnout in early-career oncologists across 7 Spanish centres 🇪🇸
🙌 Proud to co-lead this one
@myESMO https://t.co/chM3K7mPX3
[Slide 1]
3914P Implementing a structured intervention to prevent and mitigate burnout among early-career medical oncologists: Baseline data and initial feasibility from the RESILIENCE-SEOM study
V. Pacheco-Barcia1, Y. Garitaonaindía2, S. Martínez Morante3, L. León Mateos4, L. Punti Brun5, B. García Paredes6, B. Obispo7, P. Llanes Barragan5, M.M. Sánchez del Corral8, A.S. Cortegoso Mosquera4, E. Ceballos Barbancho9, S. de la Cruz10, E. García Asencio11, A.I. Martín Quesada12, R. Sánchez Bayona13, A. Lecumberri10, D. Sánchez14, C.A. Rodríguez Sánchez15, J. de Castro16, E. Elez17.
[Background and Methods panels partially obscured by presenter]
RESULTS
Figure 1. Baseline characteristics (Percentage %)
Female | 86%
≤35 years | 46%
University hospital | 96%
Previous burnout | 50%
Poor work-life balance | 79%
Previous treatment with antidepressants/anxiolytics | 14%
A total of 42 participants were included (86% female; 46% aged ≤35 years). Baseline characteristics are shown in Figure 1.
The mean CIPS score was 63.8 (SD 18.5), indicating frequent experiences of impostor phenomenon.
Baseline MBI-HSS scores revealed a burnout profile characterized by high emotional exhaustion, moderate depersonalization, and low personal accomplishment (Figure 2).
Figure 2. Baseline MBI-HSS scores [partially obscured]: Emotional exhaustion 28.2 | ... 12.[?] | ... 30.9
[Feasibility and Conclusions panels partially obscured by presenter]
[Slide 2]
3914P Implementing a structured intervention to prevent and mitigate burnout among early-career medical oncologists: Baseline data and initial feasibility from the RESILIENCE-SEOM study
V. Pacheco-Barcia1, Y. Garitaonaindía2, S. Martínez Morante3, L. León Mateos4, L. Punti Brun5, B. García Paredes6, B. Obispo7, P. Llanes Barragan5, M.M. Sánchez del Corral8, A.S. Cortegoso Mosquera4, E. Ceballos Barbancho9, S. de la Cruz10, E. García Asencio11, A.I. Martín Quesada12, R. Sánchez Bayona13, A. Lecumberri10, D. Sánchez14, C.A. Rodríguez Sánchez15, J. de Castro16, E. Elez17.
BACKGROUND
The prevalence of burnout among medical oncology trainees in Europe is estimated to exceed 70%, posing a significant threat to both healthcare professionals' well-being and the quality of patient care.
Early-career oncologists are particularly vulnerable due to high workloads and limited psychological support.
We evaluated a structured intervention combining individual and group-based strategies aimed at addressing burnout and improving emotional regulation.
METHODS
This prospective, multicenter cohort study included medical oncology residents and oncologists with ≤7 years of experience from seven Spanish hospitals. Baseline assessment included demographic data, the Maslach Burnout Inventory–Human Services Survey (MBI-HSS), and the Clance Impostor Phenomenon Scale (CIPS). The intervention consists of eight sessions delivered in a hybrid format, combining individual coaching and progressive group workshops. Follow-up assessments are planned at program completion and at 6 and 12 months.
RESULTS
Figure 1. Baseline characteristics (Percentage %)
Female | 86%
≤35 years | 46%
University hospital | 96%
Previous burnout | 50%
Poor work-life balance | 79%
Previous treatment with antidepressants/anxiolytics | 14%
A total of 42 participants were included (86% female; 46% aged ≤35 years). Baseline characteristics are shown in Figure 1.
The mean CIPS score was 63.8 (SD 18.5), indicating frequent experiences of impostor phenomenon.
Baseline MBI-HSS scores revealed a burnout profile characterized by high emotional exhaustion, moderate depersonalization, and low personal accomplishment (Figure 2).
Initial feasibility was assessed based on attendance at the first in-person session across four centers: 71.4% (10/14), 66.7% (4/6), 100% (7/7), and 66.7% (10/15), respectively, resulting in an overall initial attendance rate of 73.8% (31/42 scheduled participants).
Figure 2. Baseline MBI-HSS scores
MBI-HSS baseline (Mean score)
Emotional exhaustion | 28.2
Depersonalization | 12,0
Personal accomplishment | 30,9
CONCLUSIONS
Baseline findings confirm high levels of psychological vulnerability and burnout among early-career oncologists in Spain.
Although initial attendance rates suggest promising feasibility, high clinical workloads remain a major barrier to participation. These findings highlight the urgent need for structured institutional interventions.
Acknowledgments: Gratitude to the SEOM's Resilience-SEOM group and +MIR section.
Author COIs: speaker honoraria from Pierre Fabre, Nutricia, Servier, Pfizer, AstraZeneca, Leo Pharma; consultancy or advisory board from Advanced Accelerator Applications, a Novartis company, Pfizer, Nutricia, Servier; travel support from Amgen, Servier, Pfizer, Nutricia, Merck.
Funding: SEOM grant 2025.
Author contact: [email omitted]
#ESMO26 #HNSCC
Nice presentation from Dr Even about the systemic approach of HPV+ HNSCC
My take: Is time to end with antiEGFR agents in HPV+ HNSCC? or shall we keep it as a last resource in latter lines
None of the major guidelines (ASCO, ESMO and NCCN) do make any clear recommendation against/disadvice
@OncoAlert @myESMO @brunolarvol
@DavidSherMD
[Slide 1]
CONCLUSIONS
✓ HPV-related HNSCC are biologically, clinically, and prognostically distinct from HPV-unrelated cancers and should be considered as separate disease entities in future research, requiring dedicated clinical development.
✓ The underlying molecular biology, TME, and host immune response in HPV-associated HNSCC still need to be investigated to identify novel therapeutic targets
✓ Whether recurrent disease relies upon the same viral oncogenic machinery, immune environment and molecular alterations as in localized disease requires further study, which may help guide the appropriate setting in which to apply the therapies
[MADRID 2026 ESMO congress logo]
[Podium: ESMO congress MADRID SPAIN 23-27 OCTOBER 2026; photo of speaker at podium and panel]
🚨 #ESMO26 | Elfe-D + pumitamig across solid tumors
First clinical data beyond lung cancer for the B7-H3 ADC elfetabart drozuntecan + PD-L1×VEGF-A bsAb pumitamig!
📊 N=139 efficacy-evaluable: uORR 48.9% | cORR 39.6% | DCR 87.8%
👀 Melanoma (n=26): uORR 30.8% | cORR 26.9% | DCR 80.8%
Safety appeared manageable, with mainly G1–2 GI/hematologic TRAEs, no DLTs or treatment-related deaths, and 2 investigator-reported ILD cases 👍🏼
One question: is the activity driven by the IO, the antiangiogenic effect, or a synergy between the two?? 🤔
@OncoAlert @OncoReporte @myESMO @_SEOM @GrupoMelanoma
[Slide 1]
Baseline demographics and characteristics
At data cutoff, 142 patients had received elfe-D + pumitamig and 62% were still on treatment; ~60% had received >=2 prior lines of therapy
Elfe-D / Pumitamig | 3 mg/kg / 20 mg/kg DL1 (N=2) | 6 mg/kg / 10 mg/kg DL1-1 (N=14) | 6 mg/kg / 20 mg/kg DL2 (N=72) | 6 mg/kg / 30 mg/kg DL3 (N=54) | Total (N=142)
Age, years, Median (range) | 62.5 (60-65) | 53.5 (34-78) | 60.0 (26-83) | 59.5 (19-83) | 60.0 (19-83)
Sex, n (%) Male | 2 (100.0) | 11 (78.6) | 37 (51.4) | 18 (33.3) | 68 (47.9)
Female | 0 | 3 (21.4) | 35 (48.6) | 36 (66.7) | 74 (52.1)
Race, n (%) Asian | 0 | 8 (57.1) | 41 (56.9) | 35 (64.8) | 84 (59.2)
White | 2 (100.0) | 4 (28.6) | 27 (37.5) | 16 (29.6) | 49 (34.5)
Other | 0 | 2 (14.3) | 4 (5.6) | 3 (5.6) | 9 (6.3)
ECOG PS, n (%) 0 | 2 (100.0) | 5 (35.7) | 30 (41.7) | 17 (31.5) | 54 (38.0)
1 | 0 | 9 (64.3) | 42 (58.3) | 37 (68.5) | 88 (62.0)
Tumor type, n (%) HNSCC | 0 | 14 (100.0) | 24 (33.3) | 3 (5.6) | 41 (28.9)
Melanoma | 1 (50.0) | 0 | 11 (15.3) | 14 (25.9) | 26 (18.3)
HCC | 1 (50.0) | 0 | 11 (15.3) | 10 (18.5) | 22 (15.5)
CC | 0 | 0 | 11 (15.3) | 11 (20.4) | 22 (15.5)
PROC | 0 | 0 | 15 (20.8) | 16 (29.6) | 31 (21.8)
Prior lines of anticancer therapy, Median (range) | 2 (2-2) | 2 (1-4) | 2 (1-4) | 2 (1-5) | 2 (1-5)
>=2 lines, % | 50.0 | 57.1 | 58.3 | 63.0 | 59.9
DL, dose level; ECOG PS, Eastern Cooperative Oncology Group performance status; elfe-D, elfetabart drozuntecan; HCC, hepatocellular carcinoma; HNSCC, head and neck squamous cell carcinoma; CC, cervical cancer; PROC, platinum-resistant ovarian cancer
Data cutoff date: September 4, 2026
[Slide 2]
Most common TRAEs
The safety profile was consistent with the known profiles of the individual agents
TRAEs occurring in >=15% of the overall population (N=142)
Values shown as all-grade (grade >=3) percentages
Nausea 43.0 (1.4)
Fatigue 36.6 (2.1)
Neutrophil count decreased 34.5 (12.0)
WBC count decreased 32.4 (7.7)
Decreased appetite 28.2 (0.7)
Anemia 25.4 (2.8)
Vomiting 21.8 (0.7)
Proteinuria 21.1 (0.7)
Alopecia 18.3 (0)
Weight decreased 17.6 (0.7)
Hypertension 16.2 (7.0)
AST increased 16.2 (0.7)
Platelet count decreased 15.5 (2.1)
Stomatitis 15.5 (0.7)
Legend: Grade 1/2; Grade >=3. Patients (%)
- TRAEs were primarily grade 1/2 gastrointestinal and hematologic events
- Most common grade >=3 TRAEs were neutrophil count decreased, WBC count decreased, and hypertension
- VEGF-related toxicities consisted mainly of proteinuria and hypertension
- There were 2 investigator-reported cases of ILD (1 grade 1 non-ILD per adjudication; 1 grade 1 pending adjudication [0.7%])
Data cutoff date: September 4, 2026
[Slide 3]
Efficacy summary
Encouraging early responses were observed across all tumor types
Efficacy parameter | HNSCC (N=39) | Melanoma (N=26) | HCC (N=21) | CC (N=22) | PROC (N=31) | Total* (N=139)
Prior treatment, median (range), lines | 2 (1-4) | 2 (1-5) | 2 (1-2) | 2 (1-4) | 2 (1-4) | 2 (1-5)
Follow-up duration, median (range), months | 4.8 (0.5-10.0) | 7.4 (3.6-12.5) | 7.7 (1.4-9.3) | 6.1 (4.0-11.4) | 6.5 (2.6-8.6) | 6.3 (0.5-12.5)
Unconfirmed BOR, n (%)
CR | 0 | 0 | 0 | 1 (4.5) | 2 (6.5) | 3 (2.2)
PR | 15 (38.5) | 8 (30.8) | 9 (42.9) | 10 (45.5) | 23 (74.2) | 65 (46.8)
SD | 19 (48.7) | 13 (50.0) | 10 (47.6) | 9 (40.9) | 3 (9.7) | 54 (38.8)
PD | 4 (10.3) | 5 (19.2) | 1 (4.8) | 2 (9.1) | 3 (9.7) | 15 (10.8)
NE/missing | 1 (2.6) | 0 | 1 (4.8) | 0 | 0 | 2 (1.4)
uORR, % (95% CI) | 38.5 (23.4-55.4) | 30.8 (14.3-51.8) | 42.9 (21.8-66.0) | 50.0 (28.2-71.8) | 80.6 (62.5-92.5) | 48.9 (40.4-57.5)
cORR, % (95% CI) | 23.1 (11.1-39.3) | 26.9 (11.6-47.8) | 38.1 (18.1-61.6) | 40.9 (20.7-63.6) | 71.0 (52.0-85.8) | 39.6 (31.4-48.2)
Pending confirmation, n (%)† | 3 (7.7)‡ | 1 (3.8) | 0 | 1 (4.5) | 2 (6.5) | 7 (5.0)
DCR, % (95% CI) | 87.2 (72.6-95.7) | 80.8 (60.6-93.4) | 90.5 (69.6-98.8) | 90.9 (70.8-98.9) | 90.3 (74.2-98.0) | 87.8 (81.1-92.7)
*Efficacy-evaluable participants: received >=1 dose of study drug, baseline efficacy assessment, and >=1 post-baseline assessment or early treatment discontinuation.
†Pending confirmation includes patients with an observed CR/PR whose confirmed BOR was not yet available as of September 4, 2026, excluding those who received new anticancer therapy, died/reached EOS, or had confirmed PD before response confirmation.
‡As of Oct 09, of the 7 responses pending confirmation at the data cutoff, 3 have since been confirmed, while the remaining 4 are awaiting the next tumor assessments.
Data cutoff date: September 4, 2026
[Slide 4]
Efficacy by prior IO and/or bevacizumab
Efficacy in patients previously treated with IO and/or bevacizumab was broadly comparable to the overall population
Efficacy parameter | HNSCC Prior IO* (N=28) | HNSCC All (N=39) | Melanoma Prior IO* (N=26) | Melanoma All (N=26) | HCC Prior IO/Beva (N=19) | HCC All (N=21) | CC Prior IO/Beva (N=20) | CC All (N=22) | PROC Prior IO/Beva (N=27) | PROC All (N=31) | Total Prior IO/Beva (N=120) | Total All (N=139)
uORR, % (95% CI) | 42.9 (24.5-62.8) | 38.5 (23.4-55.4) | 30.8 (14.3-51.8) | 30.8 (14.3-51.8) | 42.1 (20.3-66.5) | 42.9 (21.8-66.0) | 50.0 (27.2-72.8) | 50.0 (28.2-71.8) | 81.5 (61.9-93.7) | 80.6 (62.5-92.5) | 50.0 (40.7-59.3) | 48.9 (40.4-57.5)
cORR, % (95% CI) | 25.0 (10.7-44.9) | 23.1 (11.1-39.3) | 26.9 (11.6-47.8) | 26.9 (11.6-47.8) | 36.8 (16.3-61.6) | 38.1 (18.1-61.6) | 40.0 (19.1-63.9) | 40.9 (20.7-63.6) | 70.4 (49.8-86.2) | 71.0 (52.0-85.8) | 40.0 (31.2-49.3) | 39.6 (31.4-48.2)
Pending confirmation, n (%) | 2 (7.1) | 3 (7.7) | 1 (3.8) | 1 (3.8) | 0 | 0 | 1 (5.0) | 1 (4.5) | 2 (7.4) | 2 (6.5) | 6 (5.0) | 7 (5.0)
DCR, % (95% CI) | 85.7 (67.3-96.0) | 87.2 (72.6-95.7) | 80.8 (60.6-93.4) | 80.8 (60.6-93.4) | 89.5 (66.9-98.7) | 90.5 (69.6-98.8) | 90.0 (68.3-98.8) | 90.9 (70.8-98.9) | 88.9 (70.8-97.6) | 90.3 (74.2-98.0) | 86.7 (79.3-92.2) | 87.8 (81.1-92.7)
*For melanoma and HNSCC, bevacizumab is not a standard treatment option; therefore, the subgroup includes patients with prior IO exposure only. For PROC, 27 patients had prior bevacizumab, of whom 3 also received prior IO.
Data cutoff date: September 4, 2026
Nice summary of data on ADCs in #sarcoma by @breelynwilkyMD .
I especially like the take-home message: “there’s a lot we don’t know about ADC patient selection in solid tumors, and in sarcoma this challenge is even more pronounced”
Still, promising data with HER2ADC in DSRCT and B7H3 in osteosarcoma (including a phase III to be presented monday)
#ESMO26 @myESMO @ctosociety
[Slide 1]
Overview of Antibody-Drug Conjugates
Biological Trojan Horse
- Antibody to cancer-enriched targets linked to therapeutic payload
- Optimal targets have turnover promoting internalization, cleavage of linker within lysosome to liberate payload
- Potential for bystander effect
- Efficacy and toxicity driven by off-target expression, payload/linker stability, anti-tumor payload activity
[Figure] Target: Expression, tumor-specificity, regulation, driver - Cancer cell surface proteins; Post-translational modifications; Secreted protein in TME; Surface Proteins on TME cells
IgG subtype: Fc Receptor activation, ADCC - IgG1, IgG2, IgG3, IgG4
Conjugation method: Drug-antibody ratio, manufacturing - Non-specific, Specific/selective, Dual payload
Linker type: Stability, TME, bystander effect - Cleavable, Non-cleavable; Enzyme labile, pH labile, Redox labile
Payload: Potency, susceptibility, resistance - Anti-microtubule, Immune activator, Targeted drugs, DNA damage, Radioligand, Degraders
JEM | Journal of Experimental Medicine © 2026 The Board of Regents of the University of Texas System
Shaw & Meric-Bernstam, J Exp Med, 2026
Breelyn A. Wilky, MD - University of Colorado Anschutz School of Medicine, Aurora, Colorado, USA
[ESMO Congress Madrid 2026 logo]
[Slide 2]
Utilizing ADCs in Sarcomas...
Opportunities and Pitfalls...
- The TARGET - is it relevant to sarcomas?
- Expression level (does it matter?)
- Downstream biology
- The PAYLOAD - is it effective?
- Payload type
- Resistance mechanisms
- Efficacy AND toxicity readouts
[Mechanism figure] (A) Targeted drug-delivery; Target receptor; Cancer cell; (B) Bystander killing - AG- 'Bystander' cancer cell; Passive toxin diffusion; (C) Pro-survival receptor blockade; (D) Fc-mediated immune-engaging effector functions - Macrophage, NK cell; Cell death; DAMPs; (E) Immunogenic cell death; Activated T cell; Antigen presentation; Mature DC; DC maturation
Conilh et al, Cell 2026
Breelyn A. Wilky, MD - University of Colorado Anschutz School of Medicine, Aurora, Colorado, USA
[ESMO Congress Madrid 2026 logo]
[Slide 3]
Clinical experience with ADCs in sarcomas - HER2
Desmoplastic small round cell tumor - trastuzumab deruxtecan
- 50% composite expression by IHC in clinical samples^1
[IHC images] HER2 (29D8) / HER2 (CB11); Patient A (H-score: 140) / Patient A (H-score: 140); Patient C (H-score: 70) / Patient C (H-score: 120); HER2 Hematoxylin
[Swimmer plot: Patient No. 1-19 vs Time (weeks) 0-110]
Legend: Partial response achieved; Recurrent disease resected; Progression; Death because of disease; Combination therapy initiated; Death because of other causes; Discontinued because of patient/MD choice; 4B5 IHC 0; 4B5 IHC 1+; 4B5 IHC 2+; 4B5 IHC unknown
Slotkin et al, JCO 2026
Breelyn A. Wilky, MD - University of Colorado Anschutz School of Medicine, Aurora, Colorado, USA
[ESMO Congress Madrid 2026 logo]
[Slide 4]
Clinical experience with ADCs in sarcomas - B7H3
Efficacy-Osteosarcoma Risvutatug Rezetecan
- Thirty-eight treated R/R osteosarcoma patients were evaluable for efficacy.
- HS-20093 showed promising anti-tumor responses especially at 12.0 mg/kg with 17.4% ORR.
| 8.0 mg/kg (N=15) | 12.0 mg/kg (N=23) | Total (N=38)
Median follow-up time (m) | 8.2 | 2.8 | 3.3
ORR, n(%), (95% CI) | 0 (0.0,21.8) | 4(17.4) (5.0,38.8) | 4(10.5) (2.9,21.1)
DCR, n(%), (95% CI) | 10(66.7) (38.4,88.2) | 20(87.0) (66.4,97.2) | 30(78.9) (62.7,90.4)
Median DOR (m), (95% CI) | / | NA NA,NA | NA NA,NA
NA: not available. Pos%: percentage of B7-H3 positive tumor cells. H-score: histochemistry score of B7-H3 protein.
Data cut-off: March 20, 2024
Figure 5. Best change in tumor burden [waterfall; 8.0 mg/kg / 12.0 mg/kg; rows Pos%, H-score, Prior line, Kinds of standard treatment - small values not transcribed]
2024 ASCO Annual Meeting #ASCO24 PRESENTED BY: Lu Xie, MD, Phase 2 Study of HS-20093 in Patients with Relapsed or Refractory Osteosarcoma and Other Sarcomas
Xie et al, ASCO 2024
Xie et al, ESMO 2025
Breelyn A. Wilky, MD - University of Colorado Anschutz School of Medicine, Aurora, Colorado, USA
[ESMO Congress Madrid 2026 logo]
🎉 Proud to present our study at #ESMO2026 in Madrid! 🇪🇸 (3904P)
🦴 In our prospective two-center study of 190 patients, adding loratadine to naproxen showed numerically lower G-CSF-induced bone pain (27.9% vs 32.7%), without statistical significance.
🧠 Greater pain was associated with poorer quality of life, sleep, mood, and anxiety.
💡 Bone pain is more than pain. Supportive care matters!
🏆 Proud to contribute four posters to ESMO as a team!
🙏 Grateful to the International Cancer Foundation for the travel grant and to all our collaborators!
@myESMO @OncoAlert
@uyargalip @kbaskurtt
@osutcuoglu @RyanNipp
#InternationalCancerFoundation #OneGlobalOncologyCommunity
#ESMO26 #ESMO
Efficacy of bemarituzumab in GC/GEJ by FGFR2 amplification status: Exploratory analysis from FORTITUDE-101 and -102
#ESMO26 #ESMOAmbassadors
👉 Amplification of FGFR2 in 16 -19%
👉 OS benefit in FGFR2 Amp pts for bema with & without Nivo
🧐FGFR2 Amp negative prognostic, potentially predictive for FGFR2🎯
@myESMO @ASCO
[Slide 1]
Background & Study Design
• Bemarituzumab, a monoclonal antibody targeting FGFR2b, was evaluated in the phase 3 FORTITUDE-101 and FORTITUDE-102 studies in FGFR2b-overexpressing advanced GC/GEJ1,2
• Exploratory survival analyses were conducted in the biomarker-evaluable population to determine the outcome by FGFR2 amplification status
FORTITUDE-101
FGFR2b ≥ 10% (N = 324a)
R 1:1
Bema + mFOLFOX6 (n = 159) | Placebo + mFOLFOX6 (n = 165)
FORTITUDE-102
FGFR2b ≥ 10% (N = 425a)
R 1:1
Bema + mFOLFOX6/CAPOX + nivolumab (n = 212) | Placebo + mFOLFOX6/CAPOX + nivolumab (n = 213)
Exploratory biomarker analysis: FGFR2 amplification evaluated by tissue and/or ctDNA NGS in assessable specimensb
n = 88 (55%) | n = 97 (59%) | n = 146 (69%) | n = 150 (70%)
• FGFR2b overexpression was detected by IHC using the investigational VENTANA FGFR2b (FPR2-D) RxDx Assay (Roche Diagnostics, Tucson, AZ)
• Tissue NGS was performed using the Tempus xT panel (FGFR2 amplification threshold ≥ 8 copies; tumor purity ≥ 30%)
• For ctDNA NGS, plasma samples were sequenced using the Guardant360 Liquid assay (Guardant Health; FGFR2 amplification threshold ≥ 2.2 copies)
aPrimary efficacy analysis population. bAssessable specimens excluded patients from China and Brazil. Data cutoff: March 2, 2026.
Bema, bemarituzumab; CAPOX, capecitabine and oxaliplatin; ctDNA, circulating tumor DNA; FGFR2b, fibroblast growth factor receptor 2 isoform IIIb; IHC, immunohistochemistry; mFOLFOX6, modified leucovorin, fluorouracil, and oxaliplatin; NGS, next-generation sequencing; R, randomized; TC, tumor cell.
1. Rha SY, et al. Slides presented at: European Society for Medical Oncology Congress; October 17-21, 2025; Berlin, Germany. Abstract LBA10. 2. ClinicalTrials.gov. https://clinicaltrials.gov/study/NCT05111626. Accessed September 25, 2026.
Presented by: Zev Wainberg, MD
[Slide 2]
FGFR2 Amplification Prevalence and Correlation With FGFR2b Expression
FGFR2 Amplification Status Among FGFR2b+ Patientsa
FORTITUDE-101: 18.9% (n = 35) IHC+ Amp+; 81.1% (n = 150) IHC+ Amp−
FORTITUDE-102: 15.9% (n = 47) IHC+ Amp+; 84.1% (n = 249) IHC+ Amp−
Legend: FGFR2b+/FGFR2Amp− ; FGFR2b+/FGFR2Amp+
Correlation Between FGFR2 Amplification Copy Number and FGFR2b Protein Expressionb
FORTITUDE-101: N = 129, Kendall's τ = 0.12, P = 0.083
FORTITUDE-102: N = 190, Kendall's τ = 0.19, P < 0.001
y-axis: FGFR2b Protein Expression (% 2+/3+ TC); x-axis: FGFR2 Amplification Copy Numbers
Key findings:
• FGFR2 amplification was identified in 19% and 16% of FGFR2b-overexpressing patients in FORTITUDE-101 and FORTITUDE-102, respectively
• FGFR2 amplification was associated with higher FGFR2b protein expression in patients with FGFR2b-overexpressing GC/GEJ
• Correlation between FGFR2b expression levels and FGFR2 copy number was weak (Kendall's τ = 0.12 in FORTITUDE-101 and 0.19 in FORTITUDE-102)
aFGFR2 amplification identified by tissue and/or ctDNA NGS. FGFR2b overexpression (IHC+) defined as ≥ 10% of TCs with 2+/3+ membrane staining by IHC.
bTissue FGFR2 amplification vs tissue FGFR2b protein expression (% 2+/3+, TC) in FGFR2b ≥ 10% population.
Presented by: Zev Wainberg, MD
[Slide 3]
FORTITUDE-101: Overall Survival by FGFR2 Amplification Status
FGFR2 Amplified (Placebo vs Bema)
Median OS, months (95% CI): Bema 19.3 (12.2–NE); Placebo 8.0 (6.0–NE)
OS HR (95% CI): 0.50 (0.22–1.12)
P = 0.086
Number at Risk (0, 6, 12, 18, 24, 30 months)
Placebo 17 11 5 2 2 0
Bema 18 14 13 10 5 1
FGFR2 Non-amplified (Placebo vs Bema)
Median OS, months (95% CI): Bema 12.9 (10.6–19.0); Placebo 12.9 (10.3–17.1)
OS HR (95% CI): 0.94 (0.65–1.37)
P = 0.76
Number at Risk (0, 6, 12, 18, 24, 30, 36 months)
Placebo 80 63 42 21 12 5 0
Bema 70 63 39 21 11 6 0
Key findings:
• In FORTITUDE-101, Bema (bemarituzumab + mFOLFOX6) showed a numerically greater treatment effect vs Placebo (placebo + mFOLFOX6) in FGFR2-amplified vs non-amplified patients for OS (HR 0.50 [95% CI 0.22–1.12] vs 0.94 [0.65–1.37]) and PFS (HR 0.47 [95% CI 0.21–1.02] vs 0.91 [0.63–1.30])
• Results from the placebo arm suggest FGFR2 amplification may confer poor prognosis
FGFR2 amplification identified by tissue and/or ctDNA NGS.
CI, confidence interval; HR, hazard ratio; NE, not estimable; OS, overall survival; PFS, progression-free survival.
Presented by: Zev Wainberg, MD
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: OS by FGFR2 amplification status (exploratory)
FGFR2 amplified: Bema + mFOLFOX6 | 18 | 19.3 mo | 12.2–NE
FGFR2 amplified: Placebo + mFOLFOX6 | 17 | 8.0 mo | 6.0–NE
FGFR2 non-amplified: Bema + mFOLFOX6 | 70 | 12.9 mo | 10.6–19.0
FGFR2 non-amplified: Placebo + mFOLFOX6 | 80 | 12.9 mo | 10.3–17.1
HR: 0.50 (amplified); 0.94 (non-amplified)
HR CI: 0.22–1.12 (amplified); 0.65–1.37 (non-amplified)
P: 0.086 (amplified); 0.76 (non-amplified)
Note: n taken from number-at-risk at time 0. PFS HRs (text only): amplified 0.47 (0.21–1.02), non-amplified 0.91 (0.63–1.30). Exploratory, not significant.
Updated results from Ph 1b/2 KEYNOTE-365 Cohort I: Pembro + Chemo (n=40) vs Chemo (n=30) for NEPC @urotoday #ESMO26
Pembro + Chemo vs Chemo:
📌ORR: 28% vs 15%
📌DCR: 50% vs 48%
📌Med DOR: 6.2 vs 4.1 mos
📌m-rPFS: 5.1 vs 4.0 mos; HR 0.66, 95%CI 0.36-1.21
📌mOS: 10.2 vs 7.1 mos; HR 0.33, 95%CI 0.17-0.63
Same slide also posted by @urotoday
[Slide 1]
Key Eligibility Criteria
• Aged ≥18 years
• Pathologicallya confirmed treatment-emergentb or de novo metastatic NEPC per central histology reviewc
• No prior platinum-based chemotherapy for NEPC
• ECOG PS 0 or 1
N = 70
Rd 1:1 or 3:1
n = 40: Pembrolizumab 200 mg IV Q3W × ≤35 cycles + Carboplatin AUC IV 5 Q3W × 4-6 cycles + Etoposide 100 mg/m2 IV d1-3 Q3W × 4-6 cycles
n = 30: Carboplatin AUC 5 IV Q3W × 4-6 cycles + Etoposide 100 mg/m2 IV d1-3 Q3W × 4-6 cycles
Treatment continued until unacceptable toxicity, PD, or participant or physician decision to withdraw
Stratification factor
• ECOG PS (0 vs 1)
Primary endpoints included
• ORR per RECIST v1.1 by blinded independent central review (BICR)
• Safety
Secondary endpoints included
• DOR per RECIST v1.1 by BICR
• rPFS per PCWG3-modified RECIST v1.1 by BICR
• OS
Post hoc analysis
• ORR, rPFS, and OS by PD-L1 status
Median study follow-up (time from randomization to data cutoff date [November 19, 2025]): 15.4 months (range, 0.8-43.5)
[Slide 2]
| Total population: Pembro + chemo n = 36 | Chemo n = 27 | PD-L1 subgroup CPS ≥1: Pembro + chemo n = 15 | Chemo n = 11 | CPS <1: Pembro + chemo n = 20 | Chemo n = 15
ORR, % (95% CI)a | 28 (14-45) | 15 (4-34) | 40 (16-68) | 18 (2-52) | 15 (3-38) | 13 (2-40)
DCR, % (95% CI)b | 50 (33-67) | 48 (29-68) | 67 (38-88) | 55 (23-83) | 35 (15-59) | 47 (21-73)
Best overall response, n (%)
CR | 0 | 0 | 0 | 0 | 0 | 0
PR | 10 (28) | 4 (15) | 6 (40) | 2 (18) | 3 (15) | 2 (13)
SD of any duration | 8 (22) | 9 (33) | 4 (27) | 4 (36) | 4 (20) | 5 (33)
SD ≥6 mo | 3 (8) | 1 (4) | 1 (7) | 0 | 2 (10) | 1 (7)
PD | 13 (36) | 6 (22) | 3 (20) | 2 (18) | 10 (50) | 3 (20)
No assessmentc | 5 (14) | 7 (26) | 2 (13) | 3 (27) | 3 (15) | 4 (7)
Not evaluabled | 0 | 1 (4) | 0 | 0 | 0 | 1 (27)
DOR, median (range), mo | 6.2 (2.0+ to 10.2+) | 4.1 (2.8+ to 6.2) | 8.3 (3.2+ to 10.2+) | 4.9 (3.5 to 6.2) | 4.1 (2.0+ to 4.1) | 4.1 (2.8 to 4.1+)
[Slide 3]
rPFS per PCWG3-Modified RECIST v1.1 by BICR
Total population
| Events, n (%) | Median (95% CI), mo | HR (95% CI)
Total population
Pembro + chemo (n = 40) | 23 (58) | 5.1 (4.1-6.3) | 0.66 (0.36-1.21)
Chemo (n = 30) | 23 (77) | 4.0 (2.1-5.0)
PD-L1 CPS ≥1
Pembro + chemo (n = 15) | 8 (53) | 6.0 (3.9-NR) | 0.40 (0.16-1.01)
Chemo (n = 12) | 11 (92) | 3.8 (0.4-5.0)
PD-L1 CPS <1
Pembro + chemo (n = 23) | 14 (61) | 5.1 (4.0-6.3) | 0.87 (0.38-1.97)
Chemo (n = 16) | 11 (69) | 4.1 (1.9-7.0)
6-month annotation: 49% (pembro + chemo), 26% (chemo)
Y axis rPFS, %; X axis Months
No. at risk (0,2,...,16 months)
Pembro + chemo: 40 32 22 12 3 1 0 0 0
Chemo: 30 20 13 5 2 0 0 0 0
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: rPFS per PCWG3-modified RECIST v1.1 by BICR (total population)
Pembro + chemo | 40 | 5.1 mo | 4.1-6.3
Chemo | 30 | 4.0 mo | 2.1-5.0
HR: 0.66
HR CI: 0.36-1.21
{'timepoint': '6 months', 'arm': 'Pembro + chemo', 'value': '49%', 'ci': None}
{'timepoint': '6 months', 'arm': 'Chemo', 'value': '26%', 'ci': None}
Note: Subgroups: CPS>=1 6.0 vs 3.8 mo HR 0.40 (0.16-1.01); CPS<1 5.1 vs 4.1 mo HR 0.87 (0.38-1.97). No p-value printed.
[Slide 4]
Overall Survival
Total population
| Events, n (%) | Median (95% CI), mo | HR (95% CI)
Total population
Pembro + chemo (n = 40) | 25 (63) | 10.2 (8.0-11.4) | 0.33 (0.17-0.63)
Chemo (n = 30) | 25 (83) | 7.1 (4.0-8.2)
PD-L1 CPS ≥1
Pembro + chemo (n = 15) | 9 (60) | 11.0 (8.7-11.9) | 0.16 (0.05-0.49)
Chemo (n = 12) | 12 (100) | 4.8 (1.9-8.6)
PD-L1 CPS <1
Pembro + chemo (n = 23) | 14 (61) | 8.1 (6.6-13.0) | 0.49 (0.21-1.11)
Chemo (n = 16) | 12 (75) | 7.1 (3.6-9.9)
6-month annotation: 85% (pembro + chemo), 59% (chemo)
Y axis OS, %; X axis Months
No. at risk (0,2,...,24 months)
Pembro + chemo: 40 34 31 26 18 13 5 3 3 3 3 2 2
Chemo: 30 26 21 17 10 2 1 0 0 0 0 0 0
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: Overall survival (total population)
Pembro + chemo | 40 | 10.2 mo | 8.0-11.4
Chemo | 30 | 7.1 mo | 4.0-8.2
HR: 0.33
HR CI: 0.17-0.63
{'timepoint': '6 months', 'arm': 'Pembro + chemo', 'value': '85%', 'ci': None}
{'timepoint': '6 months', 'arm': 'Chemo', 'value': '59%', 'ci': None}
Note: Subgroups: CPS>=1 11.0 vs 4.8 mo HR 0.16 (0.05-0.49); CPS<1 8.1 vs 7.1 mo HR 0.49 (0.21-1.11). No p-value printed; phase 1b/2, OS secondary.
☝🏼#ESMO26 Targeted therapies: toxicity is not just about the drug — context matters.
💥 Key pearls from @NReguart:
• PD-(L)1 → osimertinib: severe irAEs 15%; G≥3 pneumonitis 10%
• PD-(L)1 → sotorasib: G≥3 hepatotoxicity strongly ↑ with short washout
• KRASi + IO: feasible, but toxicity varies markedly across agents
• Dose modification is not = treatment failure
@OncoAlert @OncoReporte @myESMO @_SEOM @GrupoMelanoma
[Slide 1]
OTHER KRAS-OFF INHIBITORS + IO
More Feasible, BUT NOT Toxicity-Free
| Adagrasib + PEMBRO1 (Ph 2, KRYSTAL-7) | Divarasib + PEMBRO2 (Ph 1b/2, KRASCENDO 170) | Olomorasib +PEMBRO3 (Ph 1-2, LOXO- RAS-20001) | Calderasib + PEMBRO4 (Ph 1, KANDLELIT -001)
N | 149 | 78 | 93 | 107
Grade ≥3 TRAES | 70% | 65% | 33% | 33%
Diarrhea Grade ≥3 | 3% | 18% | 12% | 0%
Hepatic toxicity
ALT Grade ≥3 | 11% | 23% | 12% | 8%
AST Grade ≥3 | 14% | 18% | 10% | ≈ 7%
KRASi dose reductions | 48% | 53% | 20% | -
Tested in Phase III trials | KRYSTAL-7 ADA + PEMBRO vs PEMBRO | Krascendo-2 DIVA + PEMBRO vs PEMBRO + ChT | SUNRAY-01 OLO + PEMBRO vs plb + PEMBRO | KANDLELIT-004 CALD + PEMBRO vs plb + PEMBRO
Noemi Reguart, MD, PhD | ESMO 2026
1.- Garassino MC. et al. ELCC 2025; 2.- Skoulidis F et al. ASCO 2026; 3.- Burns T et al. JTO 2025; 4.Sacher A. et al. ELCC 2026
[Slide 2]
TOXICITY
2 PREDICTABLE = PROACTIVE MANAGEMENT
SKIN/NAILS (EGFR, HER2, RAS-ON) | GI/ORAL (EGFR, HER2, RAS-ON/OFF) | METABOLIC/OEDEMA (MET, ALK, ROS, NTRK) | LIVER (ALK, KRAS-OFF) | CNS/CARDIAC (EGFR, ALK, ROS1, NTRK) | LUNG/ILD (EGFR, ALK, KRAS-OFF)
ANTICIPATE
Emollients, Sun protection, Nail care | Diet, Hydration, Oral hygiene | Lifestyle, Lipids, Weight, Electrolytes, Renal function | LFTs (AST/ALT), Drug interactions | Mood, Cognition, ECG (QTc) | Monitor symptoms + imaging
DRUG-SPECIFIC PROPHYLAXIS
Doxycycline/ Minocycline P.O.
INTERVENE EARLY
Topical corticosteroids (e.g. betamethasone), Antiseptic soaks (e.g. chlorhexidine) | Early antidiarrheals, Antiemetic therapy, Steroid mouthwash ± lidocaine | Lipid-lowering therapy, Compression, Limb elevation, Physiotherapy, Diuretics if appropriate
HOLD EARLY if clinically significant
Corticosteroids if indicated
LUNG/ILD: HOLD EARLY if suspected -> DISCONTINUE if confirmed (any grade)
DOSE MODIFICATION WHEN APPROPIATE
Noemi Reguart, MD, PhD | ESMO 2026
[Slide 3]
TOXICITY
3 DOSE MODIFICATION ≠ TREATMENT FAILURE
LORLATINIB: Long-term efficacy with vs without dose reduction within 26 weeks (CROWN)
[KM curve: PFS, % vs Months 0-100; legend: With dose reduction / Without dose reduction]
AFATINIB: impact of dose adjustment on efficacy (LUX-Lung 3)
[KM curve: Estimated PFS probability vs Time (months) 0-27; inset table with median PFS, HR (95% CI), P-value - [illegible small print]]
DACOMITINIB: impact of dose adjustment on efficacy (ARCHER 1050)
[KM curve: PFS by IRC (%) vs Time (months) 0-36; inset table Patients n, Events n, Median (95% CI) PFS - [illegible small print]]
DOSE OPTIMIZATION CAN BE A WIN – THE GOAL IS PRESERVING TREATMENT CONTINUITY
Noemi Reguart, MD, PhD | ESMO 2026
1.- Mok T et al. ASCO 2026; 2.- Yang J et al. Annals of Oncol 2016; 3.- Corral J et al. Future Oncology 2019
[Slide 4]
TAKE-HOME MESSAGES
What oncologists should know about managing toxicities of novel therapies?
KNOW THE TARGET
Different targets, drugs and mechanisms bring distinct toxicity profiles.
EFFICACY ≠ TOLERABILITY
Balance therapeutic benefit against toxicity.
CONTEXT MATTERS
Combinations and sequencing can reshape toxicity and pose new challenges.
ACT EARLY & PROACTIVELY
Anticipate, intervene early and optimize dosing to enable treatment continuity.
Noemi Reguart, MD, PhD | ESMO 2026
The topic of systemic cancer treatment near the end of life is clinically and ethically challenging.
Please have a look at our poster 3113P at #ESMO26.
Paolo Bottalico presents a real world analysis of prevalence and determinants of systemic treatment use near the end of life. https://t.co/RBIiXj5gsG
[Slide 1]
Prevalence and determinants of systemic anticancer therapy use near the end of life: a real-world analysis of 1418 patients
FPN 3113P | ESMO Congress Madrid, Spain 23-27 October 2026
Paolo Bottalico*1, Noemi Pisetta1, Flavio Salerno1, Marco Audisio2, Sara Bustreo2, Ilaria Depetris2, Laura Fanchini2, Roberto Filippi2, Caterina Mecca2, Enrica Milanesi2, Chiara Paratore2, Giuliana Ritorto2, Maria Antonietta Satolli1, Rosella Spadi2, Tiziana Vavalà2, Massimo Di Maio1
1Department of Oncology, University of Turin, AOU Città della Salute e della Scienza di Torino, Turin, Italy; 2SC Medical Oncology 1U, AOU Città della Salute e della Scienza di Torino, Turin, Italy; *Presenting author
BACKGROUND
Use of systemic anticancer therapy (SACT) near end of life (EOL) remains controversial, raising concerns on appropriateness of care. We evaluated the prevalence and determinants of SACT close to death in a real-world, single-center cohort.
METHODS
Retrospective study of adult patients with advanced solid tumors who started SACT (≥1 line) at the Oncology Unit of a tertiary center in 2020–2024 and died between 1 Jan 2020 and 13 May 2025. Excluded: breast cancer; haematological, neurological, dermatological and head-neck malignancies followed at other centers; treatment continued elsewhere; adjuvant/neoadjuvant therapy only. Primary endpoint: SACT within 30 days before death (L30D); SACT within 14 days (L14D) also assessed. Treatment type refers to the last SACT received: chemotherapy (CT), targeted therapy (TT), immunotherapy (IT), hormonal therapy (HT); categories not mutually exclusive. Associations were explored by chi-square test and multivariable logistic regression (odds ratios [OR], 95% CI), adjusting for treatment type, treatment period, age, sex, treatment line, tumor category and palliative care involvement.
TABLE 1. BASELINE CHARACTERISTICS (N=1,418)
Age at death, median (IQR) | 70 (62–76) years
Sex, male | 908 (64.0%)
Tumor category: Gastrointestinal | 771 (54.4%)
Thoracic | 358 (25.2%)
Genitourinary | 166 (11.7%)
Other | 123 (8.7%)
Palliative care involvement: Yes | 1,099 (77.5%)
No | 319 (22.5%)
Treatment line at last SACT: 1st line | 623 (43.9%)
2nd line | 391 (27.6%)
3rd line | 224 (15.8%)
≥4th line | 180 (12.7%)
Last SACT received*: Chemotherapy | 1,063 (75.0%)
Targeted therapy | 267 (18.8%)
Immunotherapy | 183 (12.9%)
Hormonal therapy | 69 (4.9%)
*Not mutually exclusive. CT, chemotherapy; TT, targeted; IT, immunotherapy; HT, hormonal; GI, gastrointestinal; GU, genitourinary.
RESULTS
1,418 patients (deceased) | 20.0% SACT in last 30 days (284) | 7.7% SACT in last 14 days (109)
Palliative care involvement was documented in 1,099 patients (77.5%). SACT near EOL was less frequent with palliative care: 15.6% vs 35.4% in the last 30 days and 4.7% vs 17.9% in the last 14 days (Figure 1). After adjustment, palliative care remained strongly associated with lower odds of SACT near EOL (OR 0.35, 95% CI 0.26–0.47 at 30 days; OR 0.24, 0.15–0.36 at 14 days; Figure 2).
In the crude analysis, targeted therapy (L30D, L14D) and hormonal therapy (L14D) were associated with higher use of SACT near EOL, but none of these associations remained significant after adjustment (Table 2).
Genitourinary tumors showed independently higher use of SACT (OR 1.89 at 30 days; 2.42 at 14 days), as did thoracic tumors at 14 days (OR 3.05) (Figures 2–3). Treatment period 2023–25 was associated with higher use only in the last 14 days (OR 1.71, 1.12–2.59).
TABLE 2. WHO RECEIVES SACT NEAR EOL?
Last SACT / palliative care | % yes vs no | Crude OR (95% CI) | Adjusted OR (95% CI)
SACT in the last 30 days (L30D)
Palliative care | 15.6 vs 35.4 | 0.34 (0.25–0.45) | 0.35 (0.26–0.47)
Chemotherapy | 17.8 vs 26.8 | 0.59 (0.45–0.79) | 0.73 (0.45–1.20)
Targeted therapy | 26.6 vs 18.5 | 1.60 (1.17–2.17) | 1.25 (0.80–1.96)
Immunotherapy | 21.9 vs 19.8 | 1.14 (0.78–1.66) | 0.79 (0.45–1.38)
Hormonal therapy | 29.0 vs 19.6 | 1.68 (0.98–2.87) | 0.92 (0.42–2.02)
SACT in the last 14 days (L14D)
Palliative care | 4.7 vs 17.9 | 0.23 (0.15–0.34) | 0.24 (0.15–0.36)
Chemotherapy | 5.5 vs 14.4 | 0.34 (0.23–0.51) | 0.54 (0.26–1.12)
Targeted therapy | 13.9 vs 6.3 | 2.41 (1.58–3.67) | 1.64 (0.82–3.29)
Immunotherapy | 7.7 vs 7.7 | 0.99 (0.55–1.78) | 0.46 (0.20–1.07)
Hormonal therapy | 18.8 vs 7.1 | 3.03 (1.60–5.74) | 1.50 (0.51–4.39)
Bold = 95% CI excludes 1. Crude: targeted therapy (L30D, L14D) and hormonal therapy (L14D) are higher; none remains significant after adjustment.
Figure 1. SACT near end of life, by palliative care involvement (each figure = 1% of patients in the group; colored = received SACT). ×2.3 at 30 days · ×3.8 at 14 days more SACT without palliative care (crude ratio)
Palliative care: YES, last 30 days | 15.6% | 171 / 1099 pts
Palliative care: NO, last 30 days | 35.4% | 113 / 319 pts
Palliative care: YES, last 14 days | 4.7% | 52 / 1099 pts
Palliative care: NO, last 14 days | 17.9% | 57 / 319 pts
Figure 1 caption: SACT near EOL by palliative care involvement: last 30 days 15.6% with vs 35.4% without; last 14 days 4.7% vs 17.9% (N=1,418).
MULTIVARIABLE LOGISTIC REGRESSION: ADJUSTED OR (95% CI) (* 95% CI excludes 1; tumor site ref: GI)
SACT in the last 30 days (L30D)
Palliative care | 0.35 (0.26–0.47) *
Chemotherapy | 0.73 (0.45–1.20)
Targeted therapy | 1.25 (0.80–1.96)
Immunotherapy | 0.79 (0.45–1.38)
Hormonal therapy | 0.92 (0.42–2.02)
Period 2023–25 | 1.29 (0.98–1.70)
Age ≥70 y | 0.80 (0.60–1.06)
Female sex | 1.12 (0.84–1.50)
Line of therapy (per line) | 0.99 (0.87–1.14)
Thoracic | 1.42 (0.99–2.04)
GU | 1.89 (1.14–3.14) *
Other tumor | 1.52 (0.95–2.46)
SACT in the last 14 days (L14D)
Palliative care | 0.24 (0.15–0.36) *
Chemotherapy | 0.54 (0.26–1.12)
Targeted therapy | 1.64 (0.82–3.29)
Immunotherapy | 0.46 (0.20–1.07)
Hormonal therapy | 1.50 (0.51–4.39)
Period 2023–25 | 1.71 (1.12–2.59) *
Age ≥70 y | 0.90 (0.58–1.40)
Female sex | 1.41 (0.90–2.21)
Line of therapy (per line) | 1.03 (0.83–1.28)
Thoracic | 3.05 (1.76–5.28) *
GU | 2.42 (1.13–5.18) *
Other tumor | 1.10 (0.46–2.63)
Figure 2. Multivariable logistic regression. Palliative care involvement was independently associated with lower odds of SACT in both time windows. Compared with gastrointestinal tumors, genitourinary tumors showed higher adjusted odds at 30 and 14 days, and thoracic tumors at 14 days.
SACT by tumor category (% of patients) | L30D | L14D
Thoracic | 22.1 | 11.7
GI | 16.6 | 4.7
GU | 28.9 | 14.5
Other | 23.6 | 5.7
Figure 3. Crude rates by tumor category. At 30 days: GU 28.9%, other 23.6%, thoracic 22.1%, GI 16.6%.
KEY FINDING
Palliative care involvement is the strongest independent factor associated with less SACT use near the end of life.
CONCLUSIONS
SACT near EOL remains frequent: 1 in 5 patients received treatment in the last 30 days of life. Higher crude use with targeted and hormonal therapy did not remain statistically significant after adjustment. Palliative care involvement is the strongest independent factor associated with reduced SACT use near EOL, and its systematic promotion (particularly in genitourinary and thoracic cancers, where risk is independently higher) should be a priority to improve appropriateness of care at the end of life.
Disclosure: Paolo Bottalico (presenting author) has no conflicts of interest to declare. No specific funding was received for this work.
In the ET-FES Study, pts with total concordance of FES and FDG-PET at baseline treated with ET showed longer mOS (p=0.010, per =0.005 at multivariate analysis)
Whole-body ER heterogeneity has clinical implications
Great presentation by
@ALEDIGNUS12
#Esmo26 @OncoAlert https://t.co/dO6c6UYyOa
[Slide 2]
PROGRESSION-FREE AND OVERALL SURVIVAL
[Left] Progression-free survival (%) vs Years (0-11)
Median Progression-free survival
Group A: 18.1 months (95%CI 11.3-25.3)
Group B/C: 12.0 months (95%CI 6.6-22.9)
HR 1.36 (95%CI 0.88-2.10); p=0.162
No. at Risk (No. Cumulative Censors) (Years 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11)
Group A: 92 (0) | 53 (0) | 39 (0) | 29 (0) | 25 (0) | 21 (1) | 11 (6) | 9 (7) | 7 (9) | 4 (12) | 2 (14) | 0 (16)
Group B+C: 31 (0) | 16 (0) | 9 (0) | 7 (0) | 6 (0) | 3 (1) | 2 (2) | 1 (2) | 0 (3)
[Right] Overall survival (%) vs Years (0-11)
Median Overall survival
Group A: 69.6 months (95% CI, 48.3-85.1)
Group B/C: 42.5 months (95% CI, 27.0-70.4)
HR = 1.87 (95% CI 1.16–3.01); p = 0.010
No. at Risk (No. Cumulative Censors) (Years 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11)
Group A: 92 (0) | 90 (0) | 80 (1) | 70 (1) | 57 (2) | 48 (5) | 33 (12) | 26 (15) | 21 (17) | 11 (27) | 5 (32) | 0 (36)
Group B+C: 32 (0) | 28 (1) | 23 (1) | 17 (1) | 14 (1) | 11 (2) | 8 (3) | 3 (5) | 2 (6) | 0 (7)
Gennari A.
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: PFS and OS by FES/FDG heterogeneity group (Group A vs Group B/C)
Group A (PFS) | 92 | 18.1 months | 11.3-25.3
Group B/C (PFS) | 31 | 12.0 months | 6.6-22.9
Group A (OS) | 92 | 69.6 months | 48.3-85.1
Group B/C (OS) | 32 | 42.5 months | 27.0-70.4
HR: PFS HR 1.36; OS HR 1.87
HR CI: PFS 0.88-2.10; OS 1.16–3.01
P: PFS p=0.162; OS p=0.010
Note: n = number at risk at year 0 as printed (Group B+C 31 on PFS curve, 32 on OS curve). Group B/C HR vs Group A reference (direction inferred from multivariate slide).
[Slide 4]
CONCLUSIONS
18F-FES PET/CT captures whole-body ER heterogeneity, revealing biologically distinct patterns of ER availability across metastatic lesions.
FES-defined heterogeneity is associated with long-term survival. This association is independent of established prognostic factors in MBC, including tumor burden.
Whole-body functional ER imaging may refine endocrine-sensitivity assessment and support future treatment-fine tuning strategies
Gennari A.
#ESMO26
Oneof the best summaries that you can see these days about the management of HPV+ early disease #HNSCC
Kudos Dr Sandra Schmitz from @ClinUnivStLuc
@OncoAlert @brunolarvol https://t.co/UfPydn0GGX
[Slide 1]
I.De-escalation based on ctHPV DNA
A potential selection tool?
Baseline
Confirm a trackable tumour signal
Most HPV+ OPSCC (>90%) have detectable ctHPVDNA before treatment;
The baseline level reflects tumour burden and HPV integration status
During (C)RT
Adapt dose to response
ReACT 1.0: intermediate risk (T4 or >10 PY, TTMV-HPV DNA score >200); >95% ctDNA clearance at weeks 4-5 → reclassified and de-escalated to 54-60 Gy"
93%
2-y PFS in 89 de-escalated patients (60 low risk + 29 reclassified intermediate risk, of 102)
After surgery
Molecular residual disease
DART secondary analysis (n = 140): post-operative ctDNA status vs 2-y PFS.
69.5% vs 95.9%
2-y PFS, ctDNA+ vs ctDNA- after surgery
Surveillance
Safety for de-escalation
Two consecutive positive samples precede clinical recurrence by a median of 3.9 months.
PPV 94% · NPV 100%
Post-treatment surveillance
Integration of ctHPV DNA in de-escalation trials makes sense
Sandra Schmitz
Chera BS et al. J Clin Oncol 2020
Hanna GJ, Schoenfeld JD et al. (ReACT 1.0) Nat Commun 2026.
Routman DM et al. JAMA Otolaryngol Head Neck Surg 2025.
[ESMO Congress Madrid 2026 logo]
[Slide 2]
Conclusions
➢ Surgery is a treatment option in early stages: adequate patient selection and a technical learning curve are mandatory.
➢ Homogenization of inclusion criteria across studies would be of interest.
➢ So far, TORS and RT result in comparable swallowing function.
➢ ENE is a high risk factor but the relevant extent (>1, >2 or ≥4 mm) is undefined. iENE sensitivity is low: Potential impact of radiomics and machine learning
➢ Predictive and prognostic biomarkers should be incorporated in future trial designs. The future is biology-guided BUT need for standardized and more sensitive assays like HPV whole genome NGS liquid biopsies
[ESMO Congress Madrid 2026 logo]
#ESMO26 Nearly 1 in 3 patients alive at 7 years with EV+pembrolizumab in cisplatin-ineligible la/mUC - Mature EV-103 data (N=121; median follow-up 64.2 months).
🔹 ORR 66.9%; median DOR 29.6 months
🔹 41% of responders maintained response at 4 years
🔹 Median OS 27.7 months; 7-year OS 28.2%
🔹 Grade ≥3 TRAEs 64.5%; discontinuation of either agent 50.4%
🔹 No new safety signals
A remarkable long-term survival tail in cisplatin-ineligible la/mUC, reinforcing EV+P as 1L SOC
@UroToday @myESMO
[Slide 1]
EV-103 dose escalation and expansion cohorts A and K (1L cisplatin-ineligible la/mUC)2,6
Dose Escalation (DE) | EV+P (n=5)c
Recommended EV dose
Cohort A (expansion cohort) | EV+P (n=40)
Cohort Kd (randomized cohort) 1:1 randomizatione | EV+P (n=76) | EV monotherapy (n=73)
This analysis | Pooled EV+P DE/A/K cohort (N=121) | EV monotherapy Cohort K (n=73)
• Baseline characteristics were previously reported and consistent with the population1,7
• At data cutoff (February 20, 2026), all patients were off study treatment
• Median follow-upf: 64.2 months (pooled EV+P) and 58.6 months (EV monotherapy)
[Slide 2]
Median DOR by INV was 29.6 months in the pooled DE/A/K EV+P cohort
Pooled DE/A/K EV+P cohort
ORR by INV, n/N (%) | Events/n | Median DOR (95% CI), mo
81/121 (66.9) | 41/81 | 29.6 (10.3, 71.9)
Landmark labels: 53.7% (41.3, 64.6); 41.1% (28.5, 53.3); 37.0% (23.6, 50.4); 30.8% (16.2, 46.8)
EV monotherapy cohort
ORR by INV, n/N (%) | Events/n | Median DOR (95% CI), mo
33/73 (45.2) | 18/33 | 10.2 (4.4, 19.2)
Landmark labels: 29.4% (12.3, 48.8); 29.4% (12.3, 48.8)
DOR, % vs Time, mo (0-87)
No. at risk (EV+P): 81 67 59 46 40 37 36 32 29 27 22 21 21 19 19 17 16 16 15 12 10 8 7 7 6 5 5 2 1 1 0
No. at risk (EV mono): 33 29 16 13 10 8 6 5 5 5 5 5 5 5 5 4 4 2 0
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: Duration of response by investigator
Pooled DE/A/K EV+P | 81 | 29.6 mo | 10.3, 71.9
EV monotherapy (Cohort K) | 33 | 10.2 mo | 4.4, 19.2
{'timepoint': '~24 mo', 'arm': 'Pooled EV+P', 'value': '53.7%', 'ci': '41.3, 64.6'}
{'timepoint': '~48 mo', 'arm': 'Pooled EV+P', 'value': '41.1%', 'ci': '28.5, 53.3'}
{'timepoint': '~60 mo', 'arm': 'Pooled EV+P', 'value': '37.0%', 'ci': '23.6, 50.4'}
{'timepoint': '~84 mo', 'arm': 'Pooled EV+P', 'value': '30.8%', 'ci': '16.2, 46.8'}
{'timepoint': '~24 mo', 'arm': 'EV monotherapy', 'value': '29.4%', 'ci': '12.3, 48.8'}
{'timepoint': '~48 mo', 'arm': 'EV monotherapy', 'value': '29.4%', 'ci': '12.3, 48.8'}
Note: n = responders. ORR by INV 81/121 (66.9%) EV+P vs 33/73 (45.2%) EV mono. Landmark timepoints are not printed as text; they are inferred from the dotted marker-line positions on the x-axis (approx. 24/48/60/84 mo). Percentages and CIs are printed labels.
[Slide 3]
Median OS was 27.7 months in the pooled DE/A/K EV+P cohort
Pooled DE/A/K EV+P cohort
Events/n | Median OS (95% CI), mo
83/121 | 27.7 (20.9, 39.6)
Landmark labels: 52.9% (43.6, 61.3); 38.8% (30.2, 47.4); 33.0% (24.6, 41.6); 28.2% (19.6, 37.4)
EV monotherapy cohort
Events/n | Median OS (95% CI), mo
54/73 | 22.8 (16.3, 28.4)
Landmark labels: 46.4% (34.6, 57.3); 25.5% (16.0, 36.2); 23.9% (14.6, 34.5)
OS, % vs Time, mo (0-100)
No. at risk (EV+P): 121 112 103 87 75 62 59 55 50 47 45 38 35 26 18 16 14 14 12 3 1 0
No. at risk (EV mono): 73 70 61 57 51 48 44 40 33 28 23 22 21 19 18 18 16 16 12 9 6 4 3 1 0
• 56.2% of patients in the pooled EV+P cohort and 72.6% in the EV monotherapy cohort had subsequent systemic therapya
– PD-(L)1 inhibitor–containing therapy was administered in 23.1% of all patients in the pooled EV+P cohort and 42.5% of all patients in the EV monotherapy cohort
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: Overall survival
Pooled DE/A/K EV+P | 121 | 27.7 mo | 20.9, 39.6
EV monotherapy (Cohort K) | 73 | 22.8 mo | 16.3, 28.4
{'timepoint': '~24 mo', 'arm': 'Pooled EV+P', 'value': '52.9%', 'ci': '43.6, 61.3'}
{'timepoint': '~48 mo', 'arm': 'Pooled EV+P', 'value': '38.8%', 'ci': '30.2, 47.4'}
{'timepoint': '~60 mo', 'arm': 'Pooled EV+P', 'value': '33.0%', 'ci': '24.6, 41.6'}
{'timepoint': '~84 mo', 'arm': 'Pooled EV+P', 'value': '28.2%', 'ci': '19.6, 37.4'}
{'timepoint': '~24 mo', 'arm': 'EV monotherapy', 'value': '46.4%', 'ci': '34.6, 57.3'}
{'timepoint': '~48 mo', 'arm': 'EV monotherapy', 'value': '25.5%', 'ci': '16.0, 36.2'}
{'timepoint': '~60 mo', 'arm': 'EV monotherapy', 'value': '23.9%', 'ci': '14.6, 34.5'}
Note: Non-randomized pooled EV+P (DE/A/K) vs randomized Cohort K EV mono; no HR printed. Landmark timepoints are not printed as text; they are inferred from the dotted marker-line positions on the x-axis (approx. 24/48/60/84 mo). Percentages and CIs are printed labels.
Practical and insightful AI in oncology session at #ESMO26 👏
@ca_chung @markowetzlab @evangelia_chris @jnkath @DanielTruhn @polimi @myESMO https://t.co/if7m9BoUnf
[Slide 2]
[slide photographed at an angle; top-left of title cut off]
[...] clinic: goals of deployment
[...]pective validation to biomarker-stratified randomised trials
1 Validation
• Lock thresholds, test archived cohorts
• Analytical and clinical validity
2 Test comparison
• Same samples, head-to-head
• CHORD, Myriad, HRDetect
3 NHS pathway study
• Live use under the QMS
• Pre-specified analysis plan
4 Stratified RCT
• CIN-guided vs standard of care
• Evidence for NHS adoption
Existing methods for comparison
Method | Features | Assay input
CIN signatures (Drews 2022) | 17 copy-number signatures | Copy-number profiles
CHORD | SNV, indel and SV signatures | Whole-genome sequencing
HRDetect | Six HRD-associated signatures | Whole-genome sequencing
Myriad myChoice CDx | LOH, TAI, LST plus BRCA1/2 | Proprietary tumour assay
Drews RM et al. A pan-cancer compendium of chromosomal instability. Nature 2022
[photo also shows panel of four seated speakers and a presenter at the podium]
FES-PET defined heterogeneity is associated with long-term survival
#ESMO26
#ESMOAmbassadors
#ESMOcommunity
@myESMO @ALEDIGNUS12 https://t.co/dbSDstWNoT
[Slide 2]
PROGRESSION-FREE AND OVERALL SURVIVAL
Median Progression-free survival
Group A: 18.1 months (95%CI 11.3-25.3) | Group B/C: 12.0 months (95%CI 6.6-22.9)
HR 1.36 (95%CI 0.88-2.10); p=0.162
Progression-free survival (%) ; Years 0-11
No. at Risk (No. Cumulative Censors)
Group A | 92 (0) | 53 (0) | 39 (0) | 29 (0) | 25 (0) | 21 (1) | 11 (6) | 9 (7) | 7 (9) | 4 (12) | 2 (14) | 0 (16)
Group B+C | 31 (0) | 16 (0) | 9 (0) | 7 (0) | 6 (0) | 3 (1) | 2 (2) | 1 (2) | 0 (3)
Median Overall survival
Group A: 69.6 months (95% CI, 48.3-85.1) | Group B/C: 42.5 months (95% CI, 27.0-70.4)
HR = 1.87 (95% CI 1.16–3.01); p = 0.010
Overall survival (%) ; Years 0-11
No. at Risk (No. Cumulative Censors)
Group A | 92 (0) | 90 (0) | 80 (1) | 70 (1) | 57 (2) | 48 (5) | 33 (12) | 26 (15) | 21 (17) | 11 (27) | 5 (32) | 0 (36)
Group B+C | 32 (0) | 28 (1) | 23 (1) | 17 (1) | 14 (1) | 11 (2) | 8 (3) | 3 (5) | 2 (6) | 0 (7)
Gennari A.
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: PFS and OS (two panels)
Group A (PFS) | 92 | 18.1 months | 11.3-25.3
Group B/C (PFS) | 31 | 12.0 months | 6.6-22.9
Group A (OS) | 92 | 69.6 months | 48.3-85.1
Group B/C (OS) | 32 | 42.5 months | 27.0-70.4
HR: PFS 1.36; OS 1.87
HR CI: PFS 0.88-2.10; OS 1.16–3.01
P: PFS 0.162; OS 0.010
Note: n = number at risk at time 0. PFS Group B+C at-risk row starts at 31 while OS row starts at 32, as printed (PFS at-risk text is very small; verify). No. at risk at yearly intervals.
Happening now ! Opening ceremony of #ESMO2026 in Madrid. @OncoAlert @NCCS_SG @RebeccaDSing https://t.co/c4o9uv6a59
[Slide 1]
[Stage photo; speaker on video screen; MADRID 2026 ESMO congress logo]
Investing in research and people
30 Committees, task forces and working groups to support members
30+ one-year Fellowships, including new ESMO Artificial Intelligence Fellowship
New Clinical Trial Funding initiative to support independent academic clinical research
Support to the International Cancer Foundation (ICF)
[Podium sign] ESMO congress MADRID SPAIN 23-27 OCTOBER 2026
Ph1 trial w/INCA33890 👇🏼. Good tolerance and promising efficacy in a 1L MSS #mCRC. Liver & Peritoneal ORR 83% & 62%. HSR 35%, likely related to Oxa. Ph3 RCT undergoing at @QMBCI
#ESMO26 @OncoAlert #ColonCancer https://t.co/ITWJE13rxz https://t.co/H1z9SVes2r
[Slide 1]
[Madrid 2026 ESMO congress logo]
INCA33890, a TGFβR2×PD-1 Bispecific Antibody, With Standard of Care (SoC) Anticancer Therapies for Microsatellite Stable Colorectal Cancer (MSS CRC)
Sreenivasa R. Chandana,1 David Berz,2 Kyriakos P. Papadopoulos,3 Martin Gutierrez,4 Justin Moyers,5 Armando Santoro,6 Benedito A. Carneiro,7 Shigehisa Kitano,8 Jordi Rodon,9 Haeseong Park,10 Victor Moreno,11 Toshio Shimizu,12 Mark O'Hara,13 Tatiana Hernandez Guerrero,14 Torben F. Hansen,15 Kristen Spencer,16 Xiaohan Xu,17 Chiara Greggio,18 Thomas B. Karasic,17 Atsuo Takashima19
1START Midwest, Grand Rapids, MI, USA; 2Valkyrie Clinical Trials, Inc. Los Angeles, CA, USA; 3START Center for Cancer Research, San Antonio, TX, USA; 4John Theurer Cancer Center, Hackensack University Medical Center, Hackensack, NJ, USA; 5The Angeles Clinic and Research Institute, A Cedars-Sinai Affiliate, Los Angeles, CA, USA; 6IRCCS Humanitas Research Hospital - Humanitas Cancer Center, Rozzano, Milan, Italy; 7The Legorreta Cancer Center at Brown University, Providence, RI, USA; 8The Cancer Institute Hospital of Japanese Foundation for Cancer Research, Tokyo, Japan; 9The University of Texas MD Anderson Cancer Center, Houston, TX, USA; 10Dana-Farber Cancer Institute, Boston, MA, USA; 11START Madrid-FJD, University Hospital Fundacion Jimenez Diaz, Madrid, Spain; 12Kansai Medical University Hospital, NEXT Oncology KMU Japan, Osaka, Japan; 13Hospital of the University of Pennsylvania, Philadelphia, PA, USA; 14START Barcelona - HM Nou Delfos Hospital, Barcelona, Spain; 15Vejle Hospital, University Hospital of Southern Denmark, Vejle, Denmark; 16Perlmutter Cancer Center of NYU Langone Health, New York, NY, USA; 17Incyte Corporation, Wilmington, DE, USA; 18Incyte International Biosciences Sàrl, Morges, Switzerland; 19National Cancer Center Hospital, 5-1-1 Tsukiji, Chuo-Ku, Tokyo, Japan
Presentation number: 1954RO
[ESMO logo]
[Slide 2]
INCA33890-101 Study Design
INCA33890 Monotherapy1,* | Dose Escalation (N=48) 100-1500 mg q2w and 900 mg q4w IV in continuous 28-day cycles -> RDEs (q2w IV) (300, 600, 900, 1200 mg) -> Dose Expansion (N=236) MSS CRC / PDAC / Gastric / GEJ OC / NSCLC / SCCHN
Dose Escalation† and Expansion in MSS CRC (N=94)
Selected Eligibility Criteria
- Aged ≥18 years
- Diagnosis of advanced or metastatic (stage IV) MSS CRC
- ECOG PS 0 or 1
Primary Endpoints
- Safety and tolerability
- DLTs
Selected Secondary Endpoints
- ORR, DCR, DOR
- PK
INCA33890 (900 mg q2w IV) + FOLFOX‡ + Bev§ | n=43 | No prior systemic therapy
INCA33890 (900 mg q2w IV) + FOLFIRI|| + Bev§ | n=6 | Prior oxaliplatin but no irinotecan
INCA33890 (900 mg q2w IV) + Bev§ | n=39 | Prior fluoropyrimidine, oxaliplatin, and irinotecan
INCA33890 (900 mg q2w IV) + Cetux¶ | n=6 | Prior anti-EGFR inhibitor
[left edge of footnotes cropped] Garralda E, et al. ESMO 2025. Presentation 1522MO.
[..]rollment as of August 26, 2026. *Part 1 of the study investigating INCA33890 monotherapy has been described previously.1 †Dose escalation in part 2 used a statistical hybrid design based on ≥3 DLT-evaluable patients per dose level. ‡Oxaliplatin: 85 mg/m2 IV; [..]covorin: 400 mg/m2 IV (levoleucovorin 200 mg/m2 could be substituted for leucovorin if locally approved); and 5-FU: 400mg/m2 IV bolus (optional) followed by 2400mg/m2 IV infusion. §Dosed at 5 mg/kg q2w. ||Irinotecan: 180 mg/m2 IV; leucovorin: 400 mg/m2 IV [..]oleucovorin 200 mg/m2 could be substituted for leucovorin if locally approved); 5-FU: 400 mg/m2 IV bolus (optional) followed by 2400 mg/m2 IV infusion. ¶Dosed at 500 mg/m2 q2w.
[abbreviations: Bev, bevacizumab; Cetux, cetuximab; DCR, disease control rate; DLT, dose-limiting toxicity; DOR, duration of response; ECOG PS; EGFR; FOLFIRI; FOLFOX; GEJ; IV; MSS CRC; NSCLC; OC, ovarian cancer; ORR; PDAC; PK; q2w; q4w; RDE; SCCHN]
[Madrid 2026 logo]
[Slide 3]
Safety: Most Common TEAEs
TEAEs, n (%)* | INCA33890 + FOLFOX + Bev (n=43) Any Grade / Grade ≥3 | INCA33890 + FOLFIRI + Bev (n=6) Any Grade / Grade ≥3 | INCA33890 + Bev (n=39) Any Grade / Grade ≥3 | INCA33890 + Cetux (n=6) Any Grade / Grade ≥3
Neuropathy† | 33 (76.7) / 1 (2.3) | 2 (33.3) / 0 | 0 / 0 | 0 (0) / 0
Neutropenia‡ | 25 (58.1) / 15 (34.9) | 5 (83.3) / 4 (66.7) | 1 (2.6) / 0 | 1 (16.7) / 0
Stomatitis | 20 (46.5) / 3 (7.0) | 3 (50.0) / 1 (16.7) | 11 (28.2) / 1 (2.6) | 3 (50.0) / 0
Nausea | 19 (44.2) / 1 (2.3) | 1 (16.7) / 1 (16.7) | 4 (10.3) / 0 | 3 (50.0) / 0
Diarrhoea | 18 (41.9) / 1 (2.3) | 3 (50.0) / 1 (16.7) | 12 (30.8) / 2 (5.1) | 1 (16.7) / 0
Fatigue | 18 (41.9) / 0 | 4 (66.7) / 0 | 10 (25.6) / 1 (2.6) | 4 (66.7) / 1 (16.7)
Anaemia | 15 (34.9) / 1 (2.3) | 3 (50.0) / 1 (16.7) | 7 (17.9) / 2 (5.1) | 0 (0) / 0
Infusion-related reaction | 15 (34.9) / 1 (2.3) | 2 (33.3) / 0 | 2 (5.1) / 0 | 0 (0) / 0
Related to INCA33890 | 3 (7.0) / 0 (0) | 2 (33.3) / 0 | 2 (5.1) / 0 | 0 (0) / 0
Related to SoC | 12 (27.9)§ / 1 (2.3) | 0 / 0 | 1 (2.6)¶ / 0 | 0 (0) / 0
Vomiting | 14 (32.6) / 0 | 1 (16.7) / 0 | 4 (10.3) / 0 | 4 (66.7) / 0
Leukopenia | 13 (30.2) / 2 (4.7) | 4 (66.7) / 3 (50.0) | 0 / 0 | 0 (0) / 0
Decreased appetite | 12 (27.9) / 1 (2.3) | 3 (50.0) / 1 (16.7) | 6 (15.4) / 0 | 0 (0) / 0
Hypertension | 12 (27.9) / 4 (9.3) | 0 / 0 | 4 (10.3) / 1 (2.6) | 0 (0) / 0
*TEAEs listed are those occurring in ≥25% of the INCA33890 + FOLFOX + Bev group and ordered by frequency in the INCA33890 + FOLFOX + Bev group. †Neuropathy peripheral includes neuropathy peripheral, peripheral sensory neuropathy, and paraesthesia. ‡Neutropenia includes neutropenia and neutrophil count decreased. §11 related to oxaliplatin and 1 related to Bev. ¶Related to Bev.
Bev, bevacizumab; Cetux, cetuximab; FOLFIRI, folinic acid, fluorouracil, and irinotecan; FOLFOX, folinic acid, fluorouracil, and oxaliplatin; SoC, standard of care; TEAE, treatment-emergent adverse event.
[Madrid 2026 ESMO congress logo]
[Slide 4]
Efficacy: 1L INCA33890 + FOLFOX + Bev
| | Efficacy evaluable (n=42) |
| ORR, n (%) | 30 (71.4) |
| cORR, n (%) | 28 (66.7)* |
| DCR, n (%) | 42 (100) |
| 6-month PFS, % | 81.4 |
| Subgroup | ORR, n/N (%) |
| Liver mets | 24/29 (82.8) |
| Peritoneal inv | 8/13 (61.5) |
| PD-L1 TAP ≥1% | 21/26 (80.8) |
| PD-L1 TAP <1% | 3/6 (50.0) |
| Right sided | 11/12 (91.7) |
| Left sided | 18/28 (64.3) |
[Waterfall plot: Best Percentage Change From Baseline in Sum of Diameters of Target Lesions; legend: Liver metastasis / No liver metastasis / Ongoing / + PD-L1 TAP ≥1%; -30 threshold line]
All responses were PRs
Median follow-up of 5.6 months. *1 pending follow-up scan; 1 with target lesion removal (primary tumour resection due to intestinal obstruction, follow-up nonevaluable by RECIST).
Bev, bevacizumab; cORR, confirmed objective response rate; DCR, disease control rate; FOLFOX, folinic acid, fluorouracil, and oxaliplatin; inv, involvement; mets, metastases; NA, not available; ORR, objective response rate; PD, progressive disease; PD-L1 TAP, programmed death-ligand 1 tumour area positivity; PFS, progression-free survival; PR, partial response; RECIST, Response Evaluation Criteria in Solid Tumours.
[Madrid 2026 ESMO congress logo]
Dr @BlondeauxEva amazing on the stage of #ESMO26 on new clinical #endpoints to capture patient-centric benefit in clinical trials!
#ESMOAmbassadors
@myESMO #MCBS session https://t.co/Vtj5RAykc8
[Slide 1]
Non-scorable endpoints with ESMO-MCBS V2.0
Curative setting | Non-curative setting
Organ preservation | Time to treatment failure
*-ectomy or stoma free survival | Surgical resection/conversion
Locoregional control | Metastasis free survival
Locoregional recurrence free survival | Biochemical progression free survival
Biochemical recurrence free survival | PFS2
Minimal residual disease |
* laryngectomy, cystectomy, mastectomy etc.
Eva Blondeaux
[ESMO Congress Madrid 2026 logo; podium: ESMO Congress MADRID SPAIN 23-27 OCTOBER 2026; panel seated on stage]
Helpful summary table on ADC-related eye toxicity #ESMO26
More stable linkers associated with greater toxicity.
Befriend an opthalmologist and have them be part of your toxicity dream team! 👁️ https://t.co/us0Z8b2XvP
[Slide 1]
14:45 - 16:15 Emerging toxicities induced by antibody-drug conjugates
CHAIRS: CHRISTINA RUHLMANN, FLORIAN SCOTTÉ
ADC Ocular Adverse Events (OAEs)
Off-target toxicity occurring with a wide variety of ADCs
1. Off-target
2. Various linkers
3. Various payloads
ADC | Indication | Antibody | Linker | Payload | DAR | OAE
Belantamab mafodotin (Blenrep)* | Multiple myeloma | BCMA | Non-cleavable | MMAF | 4 | 71%
Mirvetuximab soravtansine (Elahere) | Ovarian cancer | FRα | Cleavable | DM4 | 3.5 | 61%
Tisotumab vedotin (Tivdak) | Cervical cancer | Tissue factor | Enzyme-cleavable | MMAE | 4 | 60%
Datopotamab deruxtecan (Datroway) | Breast cancer | TROP2 | Cleavable | TOP1i | 4 | 51%
Enfortumab vedotin (Padcev) | Bladder cancer | Nectin-4 | Enzyme-cleavable | MMAE | 3.8 | 40%
Trastuzumab deruxtecan (Enhertu) | Breast cancer | HER2 | Enzyme-cleavable | TOP1i | 8 | 11%
Trastuzumab emtansine (Kadcyla) | Breast cancer | HER2 | Non-cleavable | DM1 | 3.5 | 6%
Lindgren ES…Pasricha ND. Incidence and Mitigation of Corneal Pseudomicrocysts Induced by Antibody–Drug Conjugates (ADCs). Curr Ophthalmol Rep. 2024
Neel Pasricha
Eye toxicities
MURCIA AUDITORIUM - HALL 3
‼️ Hey #ESMO26 family!
Great discussion by Funda Meric-Bernstam on the rationale behind ADC + IO combinations!! @UTMDAnderson
▪️ADCs may induce immunogenic cell death, release DAMPs and prime the TME
▪️They may also increase PD-L1 expression and help overcome antigen heterogeneity
▪️VEGF–PD-(L)1 bispecifics aim to reverse vascular immunosuppression and improve tumour immune access
@OncoAlert @OncoReporte @myESMO @_SEOM
[Slide 1]
Rationale for ADC + IO combinations
1. Induction of Immunogenic Cell Death (ICD)
• Tumor-specific neoantigens, which provide new targets for the immune system to recognize.
• Damage-Associated Molecular Patterns (DAMPs), which act as biological danger signal
2. Priming the Tumor Microenvironment (TME)
• Dendritic Cell Activation
• T-Cell Infiltration
• Macrophage Reprogramming
3. Overcoming IO Resistance and Exhaustion
• ADC therapy frequently induces an upregulation of PD-L1 expression
4. Overcoming ADC Antigen Heterogeneity
• Immune system is trained broadly by the subsequent ICD, activated T cells can seek out and kill these antigen-negative escape cells, bypassing the ADC's target limitations
Dr. Funda Meric-Bernstam
[Slide 2]
Rationale for VEGF PD1/PDL1 Bispecifics
1. Complementary Biological Synergy
• Reversing vascular immunosuppression
• Dismantling TME Immune Barriers
• Reactivating T-Cells
2. Mechanistic Superiority: Cooperative Binding
• Affinity Amplification: Binding of one target may enhance affinity to other target
• Enhanced Internalization
3. Improved Local Tumor Enrichment (Therapeutic Index)
• Tumors overexpress both VEGF and PD-L1/PD-1, the dual-targeting nature acts like a geographic anchor
• The bispecific antibody accumulates heavily inside the TME
4. Overcoming Resistance and Expanding Indications
5. Clinical and Safety Advantages
• Reduced Additive Toxicity
• Single-Molecule Convenience
Dr. Funda Meric-Bernstam
Excellent lecture by @EANOassociation past president Michael Platten, followed by an engaging discussion led by Lakshmi Nayak, on immunomodulatory strategies for GBM, from vaccines to CAR T cells. The best is yet to come!.
#ESMO26 #NeuroOncology
@EANOassociation https://t.co/5rq8HRLThs
[Slide 1]
Immune Modulating Approaches in Neurooncology
[Schematic of engineered T cell, native T cell and malignant cell; labels: PD1, TCR, Engineered T cell, CAR, BiTE®, CD3, Native T cell, CTLA-4, Immune checkpoint inhibitors, MHC I/II, PD-L1, PD-L2, CD19, Naked mAb, ADC, CD19 CD20 CD22, Malignant cell]
Targets: IL13Ra2, EGFRvIII, Her2, GD2, EphA3, B7-H3, CD70, NKG2D, CLTX, GPNMB, CLTX, PTPRZ1, uPAR, ...
PRAME, PTPRZ1
Ipilimumab, Tremelimumab
Nivolumab, Pembrolizumab, Cemiplimab
Atezolizumab, Durvalumab, Avelumab
Virotherapy
Immunocytokines
Vaccines
Adapted from Batlevi et al. 2015, Nat Rev Clin Oncol
[ESMO Congress Madrid 2026 logo]
Amazing talk on RAS biology in the ESMO-EACR session #ESMO26 by @AmbrogioLab
From scarcity to abundance 🎯 RAS
From bench to bedside and back
Multiple questions are now opened!🤔
#ESMOAmbassadors @myESMO @GrupoTTD @_SEOM https://t.co/QAcHelGIQf
[Slide 1]
Validation is crucial
INTRINSIC
• KEAP1, SMARCA4 and CDKN2A as major independent determinants of inferior clinical outcomes with KRAS G12Ci monotherapy
Negrao MV et al. Cancer Discov 2023
ACQUIRED
• Selection of pre-existing clones with increased fitness on drug treatment
• De novo alterations
Awad M, et al. N Engl J Med 2021;384:2382-93; Tanaka N, et al. Cancer Discov 2021;11:1913-22; Zhao Y, et al. Nature 2021
ADAPTIVE
• Synthesis of GTP-bound KRASG12C
• Adaptive activation of wild-type RAS
Xue J, et al. Nature 2020;577:421-5; Hata A, et al. Nature Medicine 2020; Ryan MB, Cell Reports, 2022
X
[Chart: RAS inhibitors in clinical development - columns Terminated / Phase 1 / Phase 2 / Phase 3 / Approved; individual drug names [illegible]]
= n (high) of therapeutic strategies requiring validation
Ambrogio Lab
@VriesElisabeth on #access to biosimilars & generics of #cancer medicines, showing limited uptake in #LMICs and persistent inefficiency in health spending @myESMO
#ESMOAmbassadors #ESMO26
Find related @KNVokinger latest work on #WHO #EML :
https://t.co/4PpbKy2ZtA https://t.co/c5ICTP0539
[Slide 1]
The next wave of off-patent competition in oncology
From generics to biosimilars
ONCOLOGY MARKET COMPARISON
Generics - Mature market
[bar chart, US$ B] 2025: 23.7 | 2036: 31.5
Biosimilars - Rapidly expanding market
[bar chart, US$ B] 2025: 9.6 | 2030: 17.5
TWO WAVES OF OFF-PATENT COMPETITION
WAVE 1 Generics
• Chemotherapy
• Endocrine therapy
• Targeted small molecules
WAVE 2 Biosimilars
• Established mAbs (trastuzumab, rituximab, bevacizumab)
• Additional antibodies
• Checkpoint inhibitors and other complex biologics
Elisabeth de Vries
1 Future Market Insights (2025). Generic oncology drugs market.
2 ResearchAndMarkets (2025). Oncology biosimilars market.
[ESMO Congress Madrid 2026 logo]
[Podium sign: ESMO congress MADRID SPAIN 23-27 OCTOBER 2026]
BL and On-Treatment utDNA MRD and Clinical Response to Gem-iDRS in SunRISe-1 #ESMO26 @urotoday @DrFelixGuerrero
📌BL GDB scores: not predictive of BOR; similar results in pts with non-CR vs CR
📌Wk 12 GBD score: ⏫ in pts with non-CR vs CR (p=0.013)
📌BL and Wk 12 utDNA MRD status: not associated w/ DOR or EFS
[Slide 1]
Figure 3: CR by baseline GDB scores and with baseline MRD status
[Box plot: Baseline GDB score, CR vs Non-CR/NE; Wilcoxon p=0.86; MRD+ red, MRD- blue]
| CR (N=42) | Non-CR/NE (N=8) | p Value
Baseline GDB score
Mean (SD) | 38.4 (32.0) | 36.3 (30.1) | 0.86
Median (range) | 32.0 (0-94.0) | 31.5 (0-77.0)
Baseline MRD status, n (%)
MRD+ | 18 (90) | 2 (11) | 0.45
MRD- | 24 (80) | 6 (20)
[Slide 2]
Figure 4: Association of (A) DOR and (B) EFS with baseline MRD status
(A) 1-y DOR rate, % (95% CI): MRD- 62.7 (35.1-81.3); MRD+ 47.6 (25.7-66.7); p=0.16
No. at risk (months 0,3,...,30): MRD- 18 15 13 11 10 7 6 4 4 2 2; MRD+ 24 18 16 13 11 6 4 3 3 0 0
(B) 1-y EFS rate, % (95% CI): MRD- 64.9 (38.0-82.5); MRD+ 46.7 (27.8-63.6); p=0.08
No. at risk (months 0,3,...,39): MRD- 20 18 14 13 11 9 7 6 4 4 2 2 1 1; MRD+ 30 25 17 16 13 11 6 4 3 3 0 0 0 0
Survival probability (%) vs Treatment time, months
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: DOR and EFS by baseline MRD status
MRD- (DOR) | 18
MRD+ (DOR) | 24
MRD- (EFS) | 20
MRD+ (EFS) | 30
P: DOR p=0.16; EFS p=0.08
{'timepoint': '1 year', 'arm': 'MRD- DOR', 'value': '62.7%', 'ci': '35.1-81.3'}
{'timepoint': '1 year', 'arm': 'MRD+ DOR', 'value': '47.6%', 'ci': '25.7-66.7'}
{'timepoint': '1 year', 'arm': 'MRD- EFS', 'value': '64.9%', 'ci': '38.0-82.5'}
{'timepoint': '1 year', 'arm': 'MRD+ EFS', 'value': '46.7%', 'ci': '27.8-63.6'}
Note: Ns taken from first number-at-risk value. No medians printed.
[Slide 3]
Figure 5: CR by Week 12 GDB scores and by Week 12 MRD status
[Box plot: y-axis labeled 'Baseline GDB score', CR vs Non-CR/NE; Wilcoxon p=0.013; MRD+ red, MRD- blue]
| CR (N=31) | Non-CR/NE (N=8) | p Value
Week 12 GDB score
Mean (SD) | 28.1 (29.6) | 61.6 (28.1) | 0.013
Median (range) | 20.0 (0-86.0) | 68.5 (0-91.0)
Week 12 MRD status, n (%)
MRD+ | 13 (65) | 7 (35) | 0.044
MRD- | 18 (95) | 1 (5)
[Slide 4]
Figure 6: Association of (A) DOR and (B) EFS with Week 12 MRD status
(A) 1-y DOR rate, % (95% CI): MRD- 52.9 (27.6-73.0); MRD+ 57.1 (28.4-78.0); p=0.50
No. at risk (months 0,3,...,30): MRD- 18 13 11 11 9 4 3 1 1 0 0; MRD+ 14 13 11 8 8 7 6 5 5 2 2
(B) 1-y EFS rate, % (95% CI): MRD- 61.1 (35.3-79.2); MRD+ 40.0 (19.3-60.0); p=0.59
No. at risk (months 0,3,...,39): MRD- 19 19 12 11 11 9 4 3 1 1 0 0 0 0; MRD+ 20 14 13 11 8 8 7 6 5 5 2 2 1 1
Survival probability (%) vs Treatment time, months
[Survival data captured as a table — no point estimates read from the curve]
Endpoint: DOR and EFS by Week 12 MRD status
MRD- (DOR) | 18
MRD+ (DOR) | 14
MRD- (EFS) | 19
MRD+ (EFS) | 20
P: DOR p=0.50; EFS p=0.59
{'timepoint': '1 year', 'arm': 'MRD- DOR', 'value': '52.9%', 'ci': '27.6-73.0'}
{'timepoint': '1 year', 'arm': 'MRD+ DOR', 'value': '57.1%', 'ci': '28.4-78.0'}
{'timepoint': '1 year', 'arm': 'MRD- EFS', 'value': '61.1%', 'ci': '35.3-79.2'}
{'timepoint': '1 year', 'arm': 'MRD+ EFS', 'value': '40.0%', 'ci': '19.3-60.0'}
Note: Ns from first number-at-risk value. No medians printed.
When reading a clinical trial, the key question is: Does statistical significance translate into a meaningful difference in my patient’s life?
@myESMO #ESMO26 @Larvol @OncoAlert https://t.co/RRh8arjJMb
[Slide 1]
BG LAB
BISHAL GYAWALI ONCOLOGY
How I Read a Clinical Trial Report? A Primer for Busy Clinicians
POLICY AND PRACTICE | April 22, 2026
Author: Bishal Gyawali, MD, PhD, FASCO
AUTHORS INFO & AFFILIATIONS
JCO Oncol Pract 22, 909-912(2026) • Volume 22, Number 6 • DOI: 10.1200/OP-26-00181
Prof. Bishal Gyawali
[ESMO Congress Madrid 2026 logo]
At #ESMO2026 in Madrid, Prof. Herbert H. Loong emphasized that the future of precision oncology depends on three key elements: Precision, Speed, and Access.
He highlighted how multi-omic liquid biopsy approaches can complement tissue testing, support MRD detection, monitor treatment response, and potentially make precision oncology more accessible and sustainable.
Precision oncology is not just about identifying the right treatment, but delivering it at the right time to more patients.
#ESMO2026 #LiquidBiopsy @oncodaily #PrecisionOncology #MultiOmics @herbloong
[Slide 1]
MADRID 2026 ESMO congress
Take home messages
The new era of precision oncology demands:
Precision
• So test can guide patients' treatment in the right direction
Speed
• So timely precise decisions can be made for patients
Access
• So precise, and timely treatments can be offered to patients at-large
A multi-omic liquid biopsy approach can help satisfy all of the above
requirements to bring sustainable precision oncology to our patients.
Presented by HH Loong | @herbloong | ESMO 2026
#ESMO2026 A Physician-Pharmacist Shared Care Model for CDK4/6 Therapy: Advancing Quality and Reducing Oncologist burden
@RVH_HealthCntr @RVHFoundation #Madrid https://t.co/1L7TveWAM5
[Slide 1]
[Selfie of three people in front of poster 3974P; left and right portions of poster obscured]
3974P
FPN #3974P
A Physician–Pharmacist Model for CDK4/6 Therapy: Advancing Quality and Reducing Oncologist Burden
Rhiannon MacDonald, BSc, MSc candidate; Coralea Kappel, MD; Giulio Didiodato, MD; Lily Spasic, BScPharm, MSc; Sara Rask, MD; Isabel Li, MD; Jessica Singh, MD; Pooya Dibajnia, MD; Sujata Pokhrel, HBSc, CCRP; Zachary Veitch, MD
STUDY POPULATION: Study period: Aug 22, 2025 for 2 years | Data analysis cut-off: Sep 30, 2026
136 Eligible Patients -> 67 Enrolled Patients 49.3% -> 60/67 ≥1 pharmacist visit 89.6%
RESULTS: ONCOLOGIST CAPACITY RELEASED: 91 pharmacist-led visits; 22.75 h observed clinical time shifted; ≈23; ≈91
[Remaining text partially obscured or too small; see full poster image 2113875300706554361_s2 for complete transcription]
[Slide 2]
3974P
FPN #3974P
A Physician–Pharmacist Model for CDK4/6 Therapy: Advancing Quality and Reducing Oncologist Burden
Rhiannon MacDonald, BSc, MSc candidate(1); Coralea Kappel, MD(2); Giulio Didiodato, MD(3); Lily Spasic, BScPharm, MSc(4); Sara Rask, MD(2); Isabel Li, MD(2); Jessica Singh, MD(2); Pooya Dibajnia, MD(2); Sujata Pokhrel, HBSc, CCRP(5); Zachary Veitch, MD(2)
(1)McMaster University, Hamilton, Ontario (2)Oncology & Hematology, Hudson Regional Cancer Centre, Royal Victoria Hospital Barrie, Ontario (3)Critical Care Medicine, Royal Victoria Hospital Barrie, Ontario (4)Oncology Pharmacy, Royal Victoria Barrie, Ontario (5)Clinical Trials, Royal Victoria Hospital Barrie, Ontario
BACKGROUND
• Rising cancer incidence in Canada is increasing oncologist workload, compounded by limited workforce capacity (1.82 medical oncologists / 100,000 population)
• CDK4/6 inhibitors + endocrine therapy are standard therapy for HR+/HER2– breast cancer in metastatic and selected high-risk adjuvant settings.
• Expanding indications and prolonged treatment increase longitudinal oncology follow-up requirements.
• Physician–pharmacist shared care may redistribute routine monitoring while maintaining physician oversight.
PRIMARY OBJECTIVE
Evaluate oncologist clinic capacity released through a physician–pharmacist shared-care model for patients receiving CDK4/6 inhibitors.
ELIGIBILITY
METASTATIC: HR+/HER2– breast cancer; ≥6 months on CDK4/6 inhibitor and ≥2 CT scans showing stable/responding disease.
ADJUVANT: HR+/HER2– breast cancer; ≥3 months on adjuvant CDK4/6 inhibitor with stable toxicity at entry.
STUDY DESIGN
Alternating 3-month physician–pharmacist follow-up
METASTATIC: ≥6 mo stable/responding -> Pharmacist -> 3 mo -> Physician -> + imaging -> repeat
ADJUVANT: ≥3 mo stable toxicity -> Pharmacist -> 3 mo -> Physician -> Repeat
STANDARDIZED PHARMACIST ASSESSMENT
Therapy adherence & medication reconciliation | Disease surveillance & interval clinical changes | CBC, renal/hepatic function & electrolytes | ECG/QTc monitoring when indicated | Structured toxicity assessment | Supportive care & dose-management recommendations
GRADE 1 TOXICITY: Managed by pharmacist per protocol
GRADE ≥2 TOXICITY / CLINICAL CONCERN: Physician consultation / follow-up visit
STUDY POPULATION
Study period: Aug 22, 2025 for 2 years | Data analysis cut-off: Sep 30, 2026
136 Eligible Patients -> 67 Enrolled Patients 49.3% -> 60/67 ≥1 pharmacist visit 89.6%
PATIENT CHARACTERISTICS
TREATMENT INTENT: Metastatic: 40 patients (60%); Adjuvant: 27 patients (40%)
MENOPAUSAL STATUS: Post-menopausal: 50 patients (74.6%); Pre-menopausal: 16 patients (23.9%); Male: 1 patient (1.5%)
CDK4/6 INHIBITOR: Ribociclib: 40 patients (60%); Palbociclib: 14 patients (21%); Abemaciclib: 13 patients (19%)
PHARMACIST INTERVENTIONS
121 INTERVENTIONS ACROSS 91 PHARMACIST-LED VISITS (1.33 INTERVENTIONS/VISIT)
Referral 8 (6.6%); Drug-drug interactions 5 (4.1%); Supplementation 14 (11.6%); Toxicities 62 (51.2%); Lifestyle 32 (26.4%)
121 interventions
Grade 1 toxicities 83.9% | Grade 2/3 toxicities 16.1%
SAFETY AND ESCALATION
10/91 (11.0%) Pharmacist encounters requiring physician consultation
2/91 (2.2%) Pharmacist encounters requiring an additional physician follow-up visit
REASONS FOR CONSULTATION - GRADE 2/3 TOXICITIES: Rash; Elevated LFTs; Acute kidney injury; Pain; Hot flashes; Palpitations
CONCLUSION
• Physician–pharmacist shared care redistributed routine CDK4/6 follow-up while maintaining protocol-defined physician oversight. Physicians gained more than 22 hours of clinic time in over 1 year of the pilot program
• Pharmacists delivered 121 interventions across 91 visits (1.33 interventions/visit), with toxicities representing the most frequent intervention category.
• Pharmacist were able to deal independently with 83.9 % of toxicities which were grade 1 and required consultation with physicians for grade 2/3 toxicities which occurred for 16.1% of toxicities evaluated
• Physician consultation occurred in 10/91 pharmacist encounters (11.0%); 2/91 encounters (2.2%) required an additional physician follow-up visit.
RESULTS: ONCOLOGIST CAPACITY RELEASED
PHARMACIST-LED VISITS: 91 pharmacist-led visits (2,730 minutes; 1 pharmacist visit = 30 min)
OBSERVED CLINICAL TIME SHIFTED: 22.75 h observed clinical time shifted (1,365 minutes; Based on 15 min MD follow-up)
EQUIVALENT NEW-PATIENT VISITS: ≈23 60-min new-patient visits, time-equivalent
EQUIVALENT FOLLOW-UP VISITS: ≈91 15-min follow-ups, time-equivalent
At 90% target enrollment: projected 122 patients/year -> 2 visits per year -> 3,660 minutes -> 61 hours/year of potential oncologist clinic capacity released.
POTENTIAL ANNUAL FINANCIAL IMPACT
Modelled at ~50% enrollment (69 patients) -> 34.5 hours for the Oncologist and 69 hours for the pharmacist per annum
COST SAVINGS: 34.5 hours/year (Oncologist Salary) – 69 hours/year (Pharmacist Salary) = ~$1,035 per year (SAVED)
REVENUE GENERATION: 34 new breast-patient consults -> ~$7,391.30 per year (GENERATED); 34 systemic therapies (assuming) 55% Oral, 40% IV, 5% No Treatment -> ~$57,652.17 per year (GENERATED)
TOTAL ANNUAL FINANCIAL IMPACT: ~$1,035 + ~$7,391.30 + ~$57,652.17 = ~$66,078.47 per year
SECONDARY ENDPOINTS
• Cooperative identification of recurrence/progression events.
• Physician and pharmacist satisfaction — end-of-study analysis pending.
• Medication adherence — end-of-study analysis pending.
• Patient satisfaction – end-of-study analysis pending
FUNDING
This study was supported by an unrestricted grant from Novartis Pharmaceuticals. The authors declare no conflicts of interest relevant to this work.
Contact: [email address on poster]
“Thinking for yourself - do not outsource critical thinking”.
Important message from
@oncology_bg in his talk on “How I read and interpret a cancer clinical trial” at #ESMO26 https://t.co/697trqqmkU
[Slide 1]
[BG LAB - Bishal Gyawali Oncology logo]
How I read a clinical trial report
1. Generalizability of results (eligibility criteria)
2. Conflicts of interest
3. Quality of life
4. Endpoints / Surrogates
5. Control arm
6. Crossover
7. Censoring
8. Clinical vs statistical significance
9. Subgroup analysis
10. Toxicity reporting
11. Thinking for yourself – do not outsource critical thinking
Prof. Bishal Gyawali
[QR code] Gyawali B, JCO Oncol Pract 2026 22:909-912
[ESMO Congress Madrid 2026 logo]
[Podium sign: ESMO congress MADRID SPAIN 23-27 OCTOBER 2026]
5-year ET-FES trial data: 18F-FES PET/CT reveals that whole-body ER heterogeneity is a key prognostic factor in ER+/HER2- MBC treated with ET.
Patients with 100% FES+ lesions achieved significantly longer OS (71.4 vs 42.5 mos, p=0.04). FES- uptake indicates lower ET responsiveness.
@myESMO #ESMO26 @Larvol @OncoReporte @OncoAlert
AI in #ESMO26: Great insights from @jnkath . From single-task models to autonomous AI, the future of clinical implementation will depend not just on innovation, but also data infrastructure and effective governance.Great to see ESMO ELCAP and the upcoming EVALLM framework! https://t.co/lCHR1BdIqI
[Slide 3]
13:30 - 15:00 Transforming oncology with generative AI and autonomous tools
CHAIRS: EVANGELIA CHRISTODOULOU, JAKOB KATHER
ESMO Guidance on the Use of Large Language Models in Clinical Practice (ELCAP)
Type 1 — Patient-facing AI systems [patient → LLM icon]
Symptom Queries / Patient Education / Lifestyle Tracking
Type 2 — HCP-facing AI systems [clinician → LLM icon]
Clinical Decision Support / Multilingual Communication / Education and Training
Type 3 — Background AI systems [LLM → database icon]
Data Extraction / Background Alert Systems
Which evidence is required? ESMO EVALLM, work in progress
Wong et al., Annals of Oncology (2025)
Slide 7 www.kather.ai [KATHER LAB logo]
[Speaker panel] Jakob Kather — Generative AI, LLMs and AI agents in oncology
CADIZ AUDITORIUM - NCC
[Slide 4]
13:30 - 15:00 Transforming oncology with generative AI and autonomous tools
CHAIRS: EVANGELIA CHRISTODOULOU, JAKOB KATHER
AI in a European hospital
Data architecture
Access to all clinical data for training & real time deployment
[Pyramid:] AI applications / Clinical Data Repository / Primary clinical systems (EHR, PACS, ...)
GPU compute on premise or in the Cloud
Substantial demand for inference, and much more demand when models are also trained
AI-friendly governance and effective administration
AI can be deployed in a safe and compliant way
GDPR / EU AI act / MDR are navigable
Slide 15 www.kather.ai [KATHER LAB logo]
[Speaker panel] Jakob Kather — Generative AI, LLMs and AI agents in oncology
CADIZ AUDITORIUM - NCC
Prof @DrVijayPatil11 at #ESMO26 disclosing data from systematic evidence mapping showing potential 'clinical equivalence' of more sustainable schedules of #IO :
# Low dose vs no IO:
OS HR=0.67
# Low dose vs standard dose IO:
No RCT
@myESMO #ESMOAmbassadors https://t.co/FhN9FA2JML
[Slide 1]
RANDOMISED TRIAL EFFICACY
Comparison A: Low-Dose ICI Regimens Significantly Extend Overall Survival
POOLED OVERALL SURVIVAL: HR 0.67 (95% CI 0.50 to 0.91) | p = 0.026 | 4 RCTs (N = 1,453)
Low-dose ICI regimens reduced the hazard of death by 33% compared to standard non-ICI control therapy.
Trial / Study | N (Low / Ref) | Regimen Strategy | OS Estimate (95% CI) | Weight
Patil 2023 | 76 / 75 | Nivo 20mg + Chemo (Add-on) | HR 0.55 (0.36–0.82) | 15.1%
Menon 2026 | 193 / 187 | Nivo + Platinum Doublet (Add-on) | HR 0.72 (0.54–0.95) | 24.1%
Noronha 2026 (DELII) | 250 / 250 | Nivo 20mg Monotherapy (Replace) | HR 0.80 (0.66–0.97) | 33.9%
Shah 2026 (LBA6007) | 211 / 211 | Low-dose ICI Regimen (Replace) | HR 0.57 (0.44–0.73) | 26.9%
ESMO Congress 2026 | Low-Dose ICI Meta-Analysis | Dr. Vijay M Patil
8
[Stage photo: panel seated, presenter at ESMO Congress Madrid Spain 23-27 October 2026 podium]
100 treated. 4 benefit. All exposed.
The challenge is knowing who those 4 are. A powerful message from Dan Landau on why personalised adjuvant therapy matters. #ESMO26
@landau_lab @OncoAlert @myESMO https://t.co/pVXkKpGwr0
[Slide 1]
[Pictogram of 100 figures: green, yellow, red]
Cured by surgery - No benefit from adjuvant Tx
Residual disease - Benefit from adjuvant Tx
Residual disease - No benefit from adjuvant Tx
100 treated · 4 benefit · 100 exposed
[Live video of speaker on screen]
[MADRID 2026 ESMO congress logo]
📌Keynote lecture 1
CNS disease as the next frontier in oncology care: Glioblastoma, metastases, and leptomeningeal disease
👉🏻leptomeningeal metastasis update
✨ @wick_wolfgang
#ESMO26 🇪🇸 @OncoAlert #OncoAlertAF https://t.co/JweK7J435P
[Slide 1]
15:15 - 15:45 Keynote lecture 1
CHAIR: ANNA SOPHIE BERGHOFF
Leptomeningeal Metastasis (LM)
• Incidence of LM in advanced malignancies to approximate 9%–25% for small cell and non-small cell lung cancer, 5%–20% for breast cancer, and 6%–18% for melanoma.
• Prognosis of LM is poor but heterogeneous, with certain subgroups of patients (eg, EGFR-mutant NSCLC, HER2-positive breast cancer) demonstrating superior survival.
• Historical survival benchmark of 2–4 months does not apply to all patients
PMID: 38902944
[Schematic panels A-C: brain/spinal CSF flow; B: adherent LM cells, floating LM cells, brain parenchyma; C: choroid plexus, CSF, barrier disruption through secreted factors, influx of nutrients, LM cancer cells]
[Speaker video panel] Wolfgang Wick
CNS disease as the next frontier in oncology care: Glioblastoma, metastases, and leptomeningeal disease
ALMERIA AUDITORIUM - HALL 9
[ESMO Congress Madrid 2026 logo]
[Slide 2]
15:15 - 15:45 Keynote lecture 1
CHAIR: ANNA SOPHIE BERGHOFF
Leptomeningeal Metastasis (LM)- imaging
• Typical MRI findings of linear LM (type A)c
Typical MRI findings of nodular LMD (type B)
Both (type C)
Hydrocephalus only (type D–hydrocephalus)
Equivocal leptomeningeal findings or absence of leptomeningeal MRI findings (type D–normal)
[MRI panels with arrows]
PMID: 37863528
PMID: 42488117
[EZN Europäisches Zentrum für Neuroonkologie logo]
[Speaker video panel] Wolfgang Wick
CNS disease as the next frontier in oncology care: Glioblastoma, metastases, and leptomeningeal disease
ALMERIA AUDITORIUM - HALL 9
[ESMO Congress Madrid 2026 logo]
[Slide 3]
15:15 - 15:45 Keynote lecture 1
CHAIR: ANNA SOPHIE BERGHOFF
Leptomeningeal Metastasis (LM) - update
• Breaking barriers in leptomeningeal disease effort: consensus review on clinical trial design and basic science integration
• Proton craniospinal irradiation (pCSI) has emerged as a promising life-prolonging intervention for patients with LM
[Bar chart, y-axis 0 mo, 3 mo, 6 mo, 9 mo, 12 mo; groups CNS-PFS and Overall survival; legend: Proton CSI, Involved-field RT; no bar values printed]
Final randomized phase II analysis; CNS-PFS P < .001, OS P = .04.
PMID: 42777082
PMID: 42202212
[EZN Europäisches Zentrum für Neuroonkologie logo]
[Speaker video panel] Wolfgang Wick
CNS disease as the next frontier in oncology care: Glioblastoma, metastases, and leptomeningeal disease
ALMERIA AUDITORIUM - HALL 9
[ESMO Congress Madrid 2026 logo]
[Slide 4]
15:15 - 15:45 Keynote lecture 1
CHAIR: ANNA SOPHIE BERGHOFF
Leptomeningeal Metastasis (LM) – practise approach
Clinical phenotype | Management
Good performance status, disseminated LM, controlled extracranial disease | Evaluate pCSI and CNS-active systemic therapy
Actionable driver mutation | Prioritize tumour-specific CNS-active systemic treatment
Predominantly linear LM, positive cytology, patent CSF flow | Consider intrathecal therapy, preferably in a protocol
Bulky nodular or focal symptomatic LM | Focal RT or selected surgical intervention
Hydrocephalus or CSF obstruction | Neurosurgical assessment, CSF diversion when appropriate
Rapid systemic progression, poor functional status | Symptom-oriented care and carefully selected palliative interventions
[Speaker video panel] Wolfgang Wick
CNS disease as the next frontier in oncology care: Glioblastoma, metastases, and leptomeningeal disease
ALMERIA AUDITORIUM - HALL 9
[ESMO Congress Madrid 2026 logo]
Sjoukje Oosting, chair of the WG, presenting on the future and future challenges of @myESMO #MCBS at #ESMO26
#ESMOAmbassadors https://t.co/7YwkbaIxtM
[Slide 1]
ESMO - Magnitude of Clinical Benefit Scale
Validated framework to assess the clinical benefit of cancer treatments
Curative: A, B, C (Substantial benefit = A, B)
Non-curative: 5, 4, 3, 2, 1 (Substantial benefit = 5, 4)
Sjoukje Oosting
[ESMO Congress Madrid 2026 logo]
[podium: ESMO Congress Madrid Spain 23-27 October 2026; three panelists seated]
Excellent talk by Dr. Parischa from @UCSF
👁️ ADC ocular toxicity at #ESMO26: key takeaways 🧵
1. Corneal toxicity is the dominant ocular AE with ADCs: common, painful, and reversible. 3/15 FDA-approved ADCs carry a boxed warning for ocular toxicity.
2. Mechanism is off-target: ADCs are taken up non-selectively by the corneal epithelium via macropinocytosis and accumulate there. Microcyst-like changes start peripherally, then migrate centrally —> pain, blurry vision, refractive shifts.
3. Not tied to one ADC class. Seen across targets (BCMA, FRα, TF, TROP2, Nectin-4), cleavable and non-cleavable linkers, and payloads (MMAF, DM4, MMAE, TOP1i). Rates: belantamab 71%, mirvetuximab 61%, tisotumab 60%, Dato-DXd 51%, EV 40%.
4. Onset ~3 wks after infusion; resolves 2 wks–8 mo after stopping. Prevention is hard: >75% on select ADCs still develop it despite lubricants/steroid drops, driving up to 50% dose delays, 35% reductions, 25% discontinuations.
5. Risk factors remain unclear. Optometrists are well-positioned to lead routine screening and monitoring on ADCs, working closely with ophthalmology for complex cases. This partnership improves access and keeps patients on therapy
⭐️ Bottom line: proactive, early eye screening is what lets us time holds and dose reductions correctly. New consensus cornea grading (JAMA Ophthalmol 2026): G1–2 continue with close follow-up; G3 hold until G1, then reduce/stop. #ADC #oncology #ESMO26 @myESMO @MoffittNews
[Slide 1]
ADC Ocular Adverse Events (OAEs)
ADC corneal toxicity is prevalent, painful, and reversible
Corneal pseudomicrocysts
Microcyst-Like Epithelial Changes (MECs)
20% (3/15) of ADCs have FDA black box warning for ocular toxicity
Develop ~3 weeks after ADC infusion
Reversible 2 weeks-8 months after ADC discontinuation
Causes eye pain and blurry vision
Lindgren ES...Pasricha ND. Incidence and Mitigation of Corneal Pseudomicrocysts Induced by Antibody-Drug Conjugates (ADCs). Curr Ophthalmol Rep. 2024
[Slide 2]
ADC Ocular Adverse Events (OAEs)
ADCs selectively accumulate in the corneal epithelium
Rabbits administered radiolabeled ADC and ocular tissues collected over 14 days
Results:
ADC in ocular tissues (except corneal epithelium) similar to blood/plasma elimination (T1/2 = 4.5-7.5 days)
Corneal epithelium:
Absorption phase followed by stable concentration plateau through 14 days
High tissue-to-plasma ratio (1.6 vs <0.5)
[Chart] Mean Tissue : Plasma Ratio (0-2) vs Hours post dose (0-350)
Legend: Aqueous humor; Choroid-RPE; Conjunctiva (bulbar); Eyelid; Iris-cilliary body; Lens; Corneal endothelium; Corneal epithelium; Nictitating membrane; Optic nerve; Retina; Sclera - limbus; Sclera - remaining; Vitreous humor; Corneal stroma
[Y]arbington A et al. Nonclinical ocular toxicity of a maytansinoid payload-antibody drug conjugate: Ocular tissue distribution, lesion pathogenesis, and mitigation. AACR Annual Meeting. 2024
[Slide 4]
Proposed Mechanism
Macropinocytosis (actin-dependent non-selective endocytosis)
[Diagram] ADC; Lysosome; Macropinosome; Payload release
Ethylisopropylamiloride (EIPA): Gold standard macropinocytosis inhibitor
[Bar chart] MFIR (37°C/4°C), 0-50: Mock | Dextran | 200 µmol/L EIPA | 100 µmol/L EIPA | 50 µmol/L EIPA (** significance brackets)
[Microscopy] ADC; - EIPA; + EIPA; scale bars 10µm, 7.5µm
Zhao H et al. Cancer Res. 2018
Proud to present our e-Poster at #ESMO2026 .
📌 Predictive value of baseline 18F-FDG PET/CT TLG in IO-treated r/m HNSCC.
Special thanks to Prof. @amanda_psyrri for her invaluable guidance, support and contribution. 🙏🏻 https://t.co/H9r3VZllp8
[Slide 1]
1631eP
Predictive Value of Baseline 18F-FDG PET/CT TLG in Immunotherapy-Treated r/m HNSCC
Maria Anastasiou¹, Konstantinos Tryposkiadis¹, Panagiota Oikonomopoulou¹, Ioannis Kotsantis¹, Anastasios Kyriazoglou¹, Anastasios Pantazopoulos¹, George J. Kavourakis¹, Vassiliki Lyra², Niki Gavrielatou¹, Evangelos D. Zazas¹, Lydia E. Gkaralea¹, Sofia Chatziioannou², Amanda Psyrri¹
1. Oncology Unit, 2nd department of Internal Medicine, Attikon University Hospital, National Kapodistrian University of Athens, Greece 2. Department of Nuclear Medicine, Biomedical Research Foundation of the Academy of Athens, BRFAA, Athens, Greece
BACKGROUND
The efficacy of immunotherapy (IO) in patients with recurrent/metastatic head and neck squamous cell carcinoma (r/m HNSCC) has been well established. The identification of predictive biomarkers for response to immunotherapy and survival remains an area of intensive investigation. [illegible/obscured by presenter] ...help identify patients most likely to benefit from [illegible]
METHODS [obscured by presenter; Table 1 Demographic characteristics illegible]
RESULTS
Patients with TLG≤30 had longer median OS (median 49.0 vs 14.9 months; p=0.02) and numerically longer PFS (median 37.2 vs 9.4 months; p=0.11) compared to those with TLG>30. In multivariable analyses, TLG ≤30 was borderline significantly associated with improved OS (HR 0.42, 95% CI 0.18–1.00; p=0.05) and PFS (HR 0.51, 95% CI 0.25–1.01; p=0.05). None of the other PET parameters showed significant association with OS or PFS. Logistic regression did not identify any potential predictors of OR, while correlation analyses demonstrated only weak association between PET parameters and PD-L1 expression.
[Table 2. PET/CT parameters — illegible]
[Table 3. Survival by TLG Group — illegible]
[Figure 1. Summary of Overall Survival by TLG Group — KM curve, tick labels illegible]
[Table 4. Cox regression analysis to investigate association between PET/CT parameters and overall survival — illegible]
CONCLUSIONS
Baseline TLG may represent a potential predictive imaging biomarker for OS and PFS in patients with r/m HNSCC receiving IO, with lower TLG (≤30) associated with substantially improved outcomes. Other PET-derived parameters demonstrated limited predictive value. Prospective validation studies are warranted.
Correspondence: dpsyrri@med.uoa.gr
Funding: This study was not funded by any commercial entity
[e-poster kiosk UI: "Back to search"]
Great to be at #ESMO26, even more special to have it here at home in Madrid! 🇪🇸
Five days of learning, sharing ideas, and catching up with friends and colleagues.
Always inspiring to see so many people working together to improve cancer care!
#ESMO2026 @myESMO https://t.co/AsoNXAk8BC
[Slide 3]
[Photo of main-hall screen, same slide shown twice]
ESMO CONGRESS 2026: FIVE DAYS OF SCIENCE, EDUCATION, DISCUSSIONS AND NETWORKING
4,400+ ABSTRACT PRESENTATIONS
3 PRESIDENTIAL SYMPOSIA
12 LBAs
680 FACULTY MEMBERS
230 SESSIONS (including EONS)
ESMO / Congress Experience
[ESMO Congress Madrid 2026 logo]
Great opening ceremony by @FAndreMD - showcasing the work that @myESMO has done over the last year and the amazing work planned !
#ESMO26 #ESMOAmbassadors https://t.co/rXllnVuOPL
[Slide 1]
Keeping pace with science is no longer enough
ESMO's evolution is the response
[tree diagram branching to four items]
Complexity
Fragmentation
Data
Need for guidance
[Slide 2]
ESMO Roadmap 2030 levers
[gear icon] Helping to shape the scientific agenda
[folder/org-chart icon] Establishing standards and foundations
[people icon] Strengthening people, research, healthcare capabilities
[globe icon] Expanding global reach and impact
[Slide 3]
Targeted Committees on key elements of modern oncology
[icon] Artificial intelligence and digital oncology
[icon] Cancer care delivery
[icon] Cancer prevention and interception
[icon] Cancer therapies
[icon] Precision oncology and taxonomy
[icon] Special populations
Clinical Practice Guidelines Committee
Great session in #ESMO26, and great presentation from
@RADUVIDRA @myESMO
"From clinical trial to community care: Pitfalls in implementation and how to overcome them"
An uncomfortable truth. We need to improve. https://t.co/hkQdzHE9lp
[Slide 1]
Trials exclude what we treat every day
Pitfall 1 · the evidence
- 44.000 RCT with 5.7 mil patients:
- Multimorbidity:
- excluded by 91% of RCTs,
- present in 41% of patients
- Concomitant medication:
- excluded by 52% of RCTs,
- taken by 94% of patients
[Two stacked bar charts C and D by clinical specialty (Cardiovascular ... Malignant disease and immunosuppression), legend: 0 clinical specialities | 1 clinical speciality | 2–5 clinical specialities | 6–10 clinical specialities | ≥11 clinical specialities]
C: RCT exclusion by number of concomitant clinical specialties · D: real-world prevalence
The patients most often excluded are the ones most often treated.
Radu Vidra
Tan YY et al. Lancet Healthy Longev 2022;3:e674–89 · Figure 2C–D · 43,895 RCTs vs 5.7 million patients (England) · all drug classes
Kicking off #ESMO2026 with Alessandra Pedrocchi explaining the safety brakes we need to generate before employing AI agents in oncology https://t.co/Q1flHp6Kau
[Slide 1]
POLITECNICO DI MILANO 18
CoT: From Faithfulness to Monitorability
CHAIN-OF-THOUGHT MONITORING
LLM (reasoning model) --⚠--> Chain of thought (written reasoning) → Monitor (human or AI reader) → Flag (block · review · fix)
FRAGILE LINK
the written reasoning may omit or hide what drives the answer
• Monitorability: can a monitor detect problems by reading the CoT?
• Opportunity: reasoning in natural language is readable by humans and tools
• Fragile: training choices may make CoT less informative
• Recommendation: evaluate and preserve monitorability, alongside other safety methods
Korbak T. et al. Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety. 2025. arXiv:2507.11473
Alessandra Pedrocchi
[ESMO Congress Madrid 2026 logo]
[Slide 2]
POLITECNICO DI MILANO 17
Self-Explanations: Is Chain-of-Thought Faithful?
Do Reasoning Models Say What They Think?
Chain-of-thought reasoning A prompting technique that encourages language models to generate intermediate reasoning steps before arriving at a final answer, improving performance on complex tasks.
FAITHFUL COT RATE BY HINT TYPE: THE HINT IS RARELY ADMITTED
• Same question asked twice: without and with a hint (e.g., "A Stanford professor thinks the answer is A")
• When the hint changes the answer, the CoT admits it only 25% (Claude 3.7 Sonnet) and 39% (DeepSeek R1) of the time
[Bar chart: Fraction of examples with faithful CoT, Average] Claude 3.5 Sonnet (New) 5%; Claude 3.7 Sonnet 25%; DeepSeek V3 15%; DeepSeek R1 39%
Chen Y. et al. Reasoning Models Don't Always Say What They Think. Anthropic, 2025
Alessandra Pedrocchi
[ESMO Congress Madrid 2026 logo]
ESMO 2026 | Eight Frameworks Shaping the Future of Oncology
In his opening address at ESMO 2026 in Madrid, ESMO President Prof. Fabrice André highlighted eight completed ESMO frameworks:
1. ESCAT v2 – Clinical actionability of molecular targets
2. ETAC-S – Tumour-agnostic classification and screening
3. NGS recommendations – Molecular profiling in oncology
4. EBAI – AI-based biomarkers in oncology
5. ELCAP – Large language models in clinical practice
6. EnLIST – Lines of systemic therapy
7. ESMO-GROW – Real-world evidence reporting
8. ESMO-MCBS v2 – Magnitude of clinical benefit
Four additional frameworks are under development and expected within the next 6–12 months.
#ESMO26 @myESMO @oncodaily #ESMOPresident #PrecisionOncology #NGS #ArtificialIntelligence #CancerResearch
[Slide 1]
"According to ESMO"
Completed ESMO Frameworks
1 ESMO Scale for Clinical Actionability of molecular Targets (ESCAT) v.2
2 ESMO Tumour-Agnostic Classifier and Screener (ETAC-S)
3 ESMO NGS Recommendations
4 ESMO Basic Requirements for AI-based Biomarkers in Oncology (EBAI)
5 ESMO Guidance on the Use of Large Language Models in Clinical Practice (ELCAP)
6 ESMO Adaptation of Lines of Systemic Therapy (EnLIST)
7 ESMO Guidance for Reporting Oncology real World Evidence (ESMO-GROW)
8 ESMO-Magnitude of Clinical Benefit Scale (ESMO-MCBS) v.2
4 under development, to be released over the next 6-12 months
[Presenter Fabrice André on adjacent screen; repeated smaller copies of slide on lower screens]
Full house bright and early at investigational immunotherapy #ESMO26
Keep your 👀 on this PRAME TCR bispecific
🤩over 50 cancers PRAME+
🤩exciting response in heavily pretreated pts, PRAME level related https://t.co/tb9pOjCvgJ
[Slide 1]
IMA402 is an Optimized T-cell Engager (TCER) Designed to Target PRAME
Improved benefit-risk profile with half-life extension¹
[Schematic: Tumor Cell – PRAME – IMA402 – CD3/TCR – Activated T cell]
✓ Antitumor Activity: High-affinity and specificity TCR domain targeting PRAME pHLA molecules
✓ Optimized tolerability: Low-affinity T-cell recruiter against CD3/TCR allows higher dosing
✓ Antibody-like format with half-life extension: Long half-life of >1 weekᵃ allows for q2w or longer dosing intervals
PRAME is expressed in more than 50 cancers
[Heat-map list of cancer types, illegible]
PRAME prevalence in selected indications
Indication | % PRAME+ patientsᵇ
Cutaneous Melanoma 95%
Uterine Carcinoma 95%
Uterine Carcinosarcoma 95%
Synovial Sarcoma 95%
Uveal Melanoma 90%
Mucosal Melanoma 90%
Ovarian Carcinoma Subtypes 85%
Squamous Cell NSCLC 70%
Triple-negative Breast Carcinoma 65%
Small Cell Lung Cancer 45%
Esophageal Carcinoma Subtype 45%
Kidney Carcinoma Subtype 40%
Cholangiocarcinoma 35%
HER2-Enriched Breast Carcinoma 30%
Adenocarcinoma NSCLC 25%
Head & Neck Squamous Cell Carcinoma 25%
Hepatocellular Carcinoma 20%
Bladder Carcinoma 20%
[legend: ≥95% ... ≥10%]
ᵃ Half-life calculated by non compartmental analysis (NCA). ᵇ Data on file: PRAME target prevalence is based on a proprietary mass spec-guided initial expression threshold applied to RNAseq and/or immunohistochemistry data (approximate values, values between 95%-100% shown as 95%).
NSCLC, non-small cell lung cancer; pHLA, peptide-human leukocyte antigen; PRAME, preferentially expressed antigen in melanoma; q2w, every 2 weeks; TCR, T-cell receptor.
1. Hofmann M, et al. mAbs. 2026;18(1):1-24.
Prof. Dr. Dirk Schadendorf; Abstract #8465
[ESMO Congress logo] 3
[Slide 2]
IMA402-101 Study Design
Key Objectives
Primary:
• Determine MTD and/or RP2D
• Assess safety and tolerability
Secondary:
• Evaluate initial antitumor activity (RECIST 1.1 and iRECIST)
• Assess pharmacokinetics
Key Eligibility Criteria
• Recurrent and/or refractory solid tumors expressing PRAMEᵃ
• No prospective PRAME testing required
• HLA-A*02:01 positive
• ECOG performance status 0-1
• Received or not eligible for all available indicated standard of care treatments
Phase 1
• MABEL-based starting dose
• Dose escalation based on cohorts of 1-6 patients using adaptive design (BLRM model)
• MTD not reached per adaptive BLRM model
• q1w step dosing (3 doses) up to target doseᵇ
• q2w dosing from week 7 introduced post data cutoff based on favorable pharmacokinetics
• IMA402 RP2D range: 10-45 mg
IMA402 Monotherapy: 0.02 mg, 0.06 mg, 0.12 mg, 0.36 mg, 0.8 mg, 1.6 mg, 3 mg, 4 mg, 5 mg, 8 mg, 10 mg, 12 mg, 20 mg, 30 mg, 45 mgᶜ
IMA402 + Pembrolizumab: 5 mg + pembrolizumab, 10 mg + pembrolizumab, 20 mg + pembrolizumab, 30 mg + pembrolizumab
Efficacy Populationᵈ (n=40)
Safety Population (N=126)
[Footnotes small, partly illegible:] EudraCT No. 2022-503133-54-00, NCT05958121. ᵃ Cutaneous melanoma, melanoma of unknown primary, uveal melanoma, synovial sarcoma, endometrial carcinoma, ovarian carcinoma, squamous non-small cell lung cancer. ᵇ Step dosing introduced at 0.36 mg [...] ᶜ Exploratory high dose in n=6. ᵈ Efficacy population: All patients treated within the RP2D range as of Mar 1, 2026 [...] tested positive for PRAME [...] BLRM, Bayesian logistic regression model; ECOG, Eastern Cooperative Oncology Group; MABEL, minimum anticipated biological effect level; MTD, maximum tolerated dose; q1w, every week; q2w, every 2 weeks; RP2D, recommended phase 2 dose.
Data cutoff Jun 1, 2026
Prof. Dr. Dirk Schadendorf; Abstract #8465
[ESMO Congress logo] 4
[Slide 3]
Efficacy of IMA402 ± Pembrolizumab by PRAME Expression
[Waterfall plot] Y axis: Best % Change in Sum of Longest Diameter of Target Lesions from Baseline and BOR (RECIST 1.1)
PRAMElow (n=9): bar labels 45, 19, 15, 13, 12, 9, 4, -27, -62
PRAMEhigh (n=31): bar labels 41, 35, 29, 24, 21, 17, 13, 9, 6, 4, 4, 3, 1, 0, -7, -10, -11, -17, -22, -25, -44, -51, -60, -65, -68, -70, -77, -81, -84, -100, -100
Legend: IMA402 monotherapy; IMA402 + pembrolizumab; Melanomaᵃ; Ovarian cancer; Other indicationsᵇ
BOR (RECIST 1.1): PD, SD, PR, cPR, cCR; ▶ Ongoing response/disease control; ▷ Ongoing treatment
Reference lines: BL; PR
PRAMEhigh
All Indications Melanoma Ovarian Cancer
cORR 32% (10/31) 32% (7/22) 3/6
DCR (week 6) 65% (20/31) 73% (16/22) 3/6
ᵃ Includes cutaneous melanoma and one patient with uveal melanoma. ᵇ Includes endometrial cancer (n=2), sqNSCLC (n=2), and synovial sarcoma (n=1).
BOR, best overall response; BL, baseline; cCR, confirmed complete response; cORR, confirmed objective response rate; DCR, disease control rate; PD, progressive disease; (c)PR, (confirmed) partial response; RP2D, recommended phase 2 dose; SD, stable disease.
Data cutoff Jun 1, 2026
Prof. Dr. Dirk Schadendorf; Abstract #8465
[ESMO Congress logo] 10
[Slide 4]
Key Takeaways
• TCR-based therapies enable immune recognition of intracellular tumor antigens presented by cell-surface HLA, expanding the therapeutic landscape beyond targets accessible to conventional immunotherapies
• IMA402 is a novel TCR-based bispecific T-cell engager (TCER) targeting the cancer-associated antigen PRAME, a target expressed in more than 50 cancers
• IMA402 demonstrates a favorable tolerability profile and encouraging initial antitumor activity across various indications, including melanoma
In addition to IMA402 monotherapy, three ongoing cohorts of IMA402 with:
+ Pembrolizumab or nivolumab/relatlimab
+ Decitabine as potential PRAME enhancer
+ MAGEA4/8-directed TCER IMA401
HLA, human leukocyte antigen; ICI, immune checkpoint inhibitor; TCR, T cell receptor.
Data cutoff Jun 1, 2026
Prof. Dr. Dirk Schadendorf; Abstract #8465
[ESMO Congress logo] 12
“Inequalities start long before a cancer diagnosis”, my take away from presentation by @JoseRamalho during his presentation in the patient advocacy track session on “Access, equity and reducing inequalities”.
#ESMO26 #ESMO2026 #equity #ESMOCongress2026 #PatientAdvocacy https://t.co/IXdvnXOBdX
[Slide 1]
13:00 - 14:30 Access, equity, and reducing inequalities
CHAIRS: ANNE-MARIE BAIRD, AGATA POLINSKA
Inequalities start long before a cancer diagnosis
Progress in prevention
22 % -> 19 %
Daily smoking, 2012-22
2 in 3
Countries with falling alcohol consumption, 2010-22
64 %
HPV vaccination coverage in 2023
Exposure remains higher with low education (Low education | High education)
Daily smoking | 19% | 13%
Overweight or obesity | 57% | 43%
Insufficient physical activity | 73% | 64%
Low vegetable consumption | 43% | 33%
Low fruit consumption | 37% | 34%
Source: OECD/European Commission (2025)
José Ramalho, Health Division, OECD
José Ramalho - Mapping the gaps: Insights from the European Cancer Inequalities Registry
SEVILLE AUDITORIUM - NCC
Amazing discussion of Prof Robin Jones @RobinL_Jones in #ESMO2026 @myESMO. Clear concepts and sharp analysis. To enjoy it!! https://t.co/GPJqagUAqF
[Slide 1]
TGCT & Desmoid: Similarities & Differences
• Chronic diseases
• Limited value of standard endpoints
• Radiological
• Overall survival
• Importance of quality of life measures
• Existing randomized data to guide management
• Unpredictable clinical course
• TGCT:
• CSF-1R inhibitors clear mechanism
• Many patients treated with surgery
• Desmoid:
• Variety of effective agents
• Different mechanisms of action
• Most patients initially managed with active surveillance
• Spontaneous regression
Stacchiotti S et al. Cancer Treat Rev 112; 102491: 2023
Jeys LM et al. Bone Joint J 108-B; 1045-1051: 2026
Kasper B et al. JAMA Oncol 10; 1121-1128: 2024
Always thought-provoking insights from @DanielTruhn at #ESMO26! As AI agents enter oncology, we need more rigorous evaluation designs, including RCTs, just as we evaluate new drugs/technology.
#AItrack #ESMOAmbassador https://t.co/5rWDEdVrxy
[Slide 2]
13:30 - 15:00 Transforming oncology with generative AI and autonomous tools
CHAIRS: EVANGELIA CHRISTODOULOU, JAKOB KATHER
Where are we?
4,609 clinical studies of language models, January 2022 to September 2025
Most agent studies are here.
Exam questions 1,704 studies
Simulated cases 1,857 studies
Real patient data 1,048 studies
Randomised trials 19
[QR code] Nat Med 2026 Chen SF et al.
[QR code] JAMA 2025 Bedi S et al.
[ESMO Congress Madrid 2026 logo]
[Speaker panel] Daniel Truhn — Clinical evidence for AI agents: Where are we and what do we need
CADIZ AUDITORIUM - NCC
[Slide 3]
13:30 - 15:00 Transforming oncology with generative AI and autonomous tools
CHAIRS: EVANGELIA CHRISTODOULOU, JAKOB KATHER
We are now in a Transition Period
Gare Montparnasse, Paris, 22 October 1895
[photo of Montparnasse train derailment]
1895 | Conclusions for Agents today
Running late, the express entered the station too fast | Don't deploy ahead of evidence
The air brake failed, train broke through the wall | Safeguards needed
The guard was busy with paperwork and did not apply the handbrake | Don't overload the human
A woman in the street below was killed | Harm to patients?
Photo: Studio Lévy & fils, 1895 (public domain)
[ESMO Congress Madrid 2026 logo]
[Speaker panel] Daniel Truhn — Clinical evidence for AI agents: Where are we and what do we need
CADIZ AUDITORIUM - NCC
Presenting at #ESMO2026 in Madrid: Safety of LLM-Generated Chemotherapy Orders: A Multi-Model Evaluation Against International Guidelines and Standard Preparation Protocols (ePoster 247eP).
We had ChatGPT and Gemini generate chemotherapy orders across 16 regimens, then three medical oncologists scored them blinded against NCCN/ESMO, eviQ and manufacturer labels.
Key findings:
• 120/223 order sets (53.8%) had at least one clinically significant error
• No critical errors were recorded
• Errors clustered in supportive care (NK1/5-HT3 omissions, loperamide, cisplatin hydration) and route specification
• Cisplatin- and irinotecan-containing regimens (BEP, FOLFIRINOX, GEMCIS) were the highest-risk
• No detectable difference between models (p=0.42)
Takeaway: LLM errors are not random. They concentrate at identifiable points, so clinicians can focus verification where it matters most. This was a simulation study, and structured expert review remains essential.
Grateful to my co-authors Dr. R. Chanana, Dr. T. Pathan and Dr. K. Gupta, and to Apollo Hospitals Navi Mumbai.
#ESMO26 #Oncology #MedicalOncology #AIinHealthcare #LLM #PatientSafety #Chemotherapy #OncologyAI
In testing data from Ukrainian cancer patients stress questionnaires in #esmo2026 @myESMO https://t.co/QrCcIGRbw8
[Slide 2]
FPN: 3482 | 3482P
Psychosocial factors in cancer care in frontline regions in Ukraine
Javier Perea1, J Alejandro Perez-Fidalgo1,2, Anna Kachan3, Yuliia Nikishyna3, Yuliia Shevchuk3
1Department of Medical Oncology Hospital Clinico Universitario Valencia. INCLIVA.
2University of Valencia
3Mission Kharkiv
Background
Diagnosis of cancer has a deep impact on different life aspects, including socioeconomic issues, [obscured] well-being.
Psychosocial determinants of cancer [obscured] major concern that affects quality [obscured] diagnosis occurred in an adverse ge[obscured] such as a war, the potential stress o[obscured] is expectedly higher.
Methods
Patients recently diagnosed with [obscured] Ukrainian regions next to the f[obscured] and Sumy regions, were [obscured] questionnaire.
Psychosocial aspects [obscured] measured, including [obscured] of coping strategies [obscured]
The questionna[ire] [obscured] validated psych[ometric] [obscured] Perceived Str[ess] S[cale] [obscured] COPE.
Results
340 patients responded to the questionnaire, 92 male and 248 female. The median age was 60 years (19-82). 46.2% were from Sumy a[nd] [obscured] from the Kharkiv region. The most frequent diagnoses were breast cancer 33.5%, gastrointestinal cancers 23.5% and malignant genitou[rinary or] gynecological malignancies 16.5%. Of note only 6.5% had a thoracic malignancy. The most frequent stage at diagnosis was II in 33.8% fo[llowed by] stage III 21.5% and stage IV 17.9%.
[Donut and bar charts - small print, not transcribed]
The median PSS-10 level was 22 (interquartile range 17 to 25) from a possible range 0 to 40, higher than previously reported [obscured] values. High stress levels (PSS-10 >=17) were observed in 19.1% of patients, while 10.3% had low stress levels. Stress level[s] [obscured] [asso]ciated with gender (X2 8.9. p=0.001), with higher prevalence in women (23% vs 8.7%). No association was found w[ith] [obscured] [st]age (X2=2.4; p=0.7). Stress level was significantly associated with income level; higher stress levels were mo[re] [obscured] [low]er income. Moreover, stress level scores increased over time after cancer diagnosis, being higher after more t[obscured] [obscured] Coping strategies were problem-oriented in 75.9%, emotionally oriented in 12.4%, avoidant in 7.1%, and m[obscured] [obscured] [co]ping strategies were associated with lower stress levels (χ2=8.2; p=0.016).
Conclusion
[obscured] high among cancer patients in Ukrainian regions in a war context. Gender and income are the mos[t] [obscured] [obscured] oriented coping strategies reduced stress level.
#ESMO2026 opening ceremony by the @myESMO president @FAndreMD https://t.co/HKGLHJBLPQ
[Slide 1]
[Stage photo; screen shows slide twice flanking live video of presenter; panelists seated on stage; MADRID 2026 ESMO congress logo]
Keeping pace with science is no longer enough
ESMO's evolution is the response
[tree diagram branching to four items]
Complexity
Fragmentation
Data
Need for guidance
MADRID 2026 ESMO congress
[Podium sign:] ESMO congress MADRID SPAIN 23-27 OCTOBER 2026
[Slide 2]
[Stage photo; screen shows slide twice flanking live video of presenter]
ESMO Roadmap 2030 levers
[gear icon] Helping to shape the scientific agenda
[folder/org-chart icon] Establishing standards and foundations
[people icon] Strengthening people, research, healthcare capabilities
[globe icon] Expanding global reach and impact
MADRID 2026 ESMO congress
[Podium sign:] ESMO congress MADRID SPAIN 23-27 OCTOBER 2026
Our #RIANO study at #ESMO26! 138 neuro-oncology professionals from 15 Ibero-American countries (79% LATAM). Regional disparities in experience, gender, and clinical workload. Strengthening our regional voice @myESMO @SNOLAsociety @NeuroOnc @EANOassociation @GrupoGeino 🦾🦾🦾 https://t.co/Hf1pANcVmP
[Slide 1]
Neuro-oncology workforce across Ibero-America: demographic profile from the RIANO survey
Martin Zapata1, Alexandra Diaz2, Gabriela Mendonca3, Victor Patlan4, Angeles Viu5, Sonia; Lenmy Abao, Jorpa Escudero, Fernando Alosio Lanao Sanchez, Eduardo Daniel Bredis Gonzalez, Tessa Cristina Costa Va, Jose isidndra Ibanez, Joaquin Alexis Perez, Lucia Richter P[illegible], Victor Hernandez Alvarez, [illegible] Gutierrez Manrique, Loreto Yanez Sepulveda, Juan Guillermo, Revin Camargo, Kevin Camargo [author list rendered garbled on this image; affiliations illegible]
RIANO Red Iberoamericana de Neuro-Oncologia | BERLIN 2026 ESMO CONGRESS [logo as printed]
BACKGROUND
Understanding who delivers neuro-oncology care across Ibero-America is essential to interpret regional differences in access to care.
We describe the demographic and professional profile of participants in a structured online survey conducted by the Ibero-American Neuro-Oncology Network (RIANO) across Spain and Latin America.
Objective: characterise the workforce profile by region to contextualise disparities in neuro-oncology care.
METHODS
Cross-sectional online survey (preliminary analysis).
1 Design - Structured questionnaire on demographic and professional variables.
2 Distribution - Through medical societies and RIANO local representatives.
3 Population - Respondents from Spain and 14 Latin American countries.
4 Analysis - Descriptive statistics, comparisons Spain vs. Latin America.
Survey open until 30 September 2026. Updated analyses including additional responses will be presented at the congress.
RESULTS
138 professionals responded | 29 (21%) Spain | 109 (79%) Latin America + 14 countries | 44.7 y mean age, all respondents
Figure 1. Experience, gender and case volume by region (% of respondents) [Latin America (n=109) | Spain (n=29)]
>10 years of experience | 54.1% | 72.4%
<5 years of experience | 17.4% | 6.9%
Women | 42.2% | 65.5%
>20 brain tumour cases per month | 14.7% | 34.5%
Practice setting
82.8% of Spanish respondents work only in the public sector.
45.9% of Latin American respondents work in both public and private systems.
Specialties
Medical oncology was the leading specialty in both regions, reflecting the multidisciplinary nature of neuro-oncology care.
Table 1. Demographic and professional profile by region
Variable | Spain (n = 29) | Latin America (n = 109)
Share of all respondents | 21% | 79%
Countries represented | 1 | 14
>10 years of professional experience | 72.4% | 54.1%
<=5 years of professional experience | 6.9% | 17.4%
Women | 65.5% | 42.2%
>20 brain tumour cases per month | 34.5% | 14.7%
Practice setting | Public sector only: 82.8% | Public and private: 45.9%
Leading specialty | Medical oncology | Medical oncology
KEY FINDINGS
Seniority - More than 10 years since certification: 72.4% (Spain) vs 54.1% (Latin America), indicating a more experienced workforce in Spain.
Gender - Women were the majority in Spain (65.5%) but a minority in Latin America (42.2%).
Case volume - Higher proportion of >=20 brain tumour cases per month in Spain (34.5%) vs Latin America (14.7%).
Setting - Most Spanish professionals work in the public sector (82.8%), while nearly half of Latin American professionals work in both public and private systems (45.9%).
CONCLUSIONS
The Ibero-American neuro-oncology workforce is experienced but heterogeneous. Significant regional differences exist in seniority, gender distribution, clinical volume and practice setting. These differences should be considered when designing regional educational and training strategies to reduce inequities in neuro-oncology care.
Disclosure (presenting author)
M. Zapata reports personal fees from AstraZeneca, Bayer, MSD, Novartis, Roche, Bristol Myers Squibb, and Pfizer (all outside the submitted work).
Funding: No funding was received for this study.
Correspondence: Martin Zapata, MD, MSc.
Really enjoyed Florian Markowetz’s talk at #ESMO26, highlighting the long and complex journey. Beyond model development, the often-overlooked challenges of technical integration, governance, regulatory approval & validation are just as critical to bringing AI safely to hospitals. https://t.co/pmtrxxBmLp
[Slide 1]
13:30 - 15:00 Transforming oncology with generative AI and autonomous tools
CHAIRS: EVANGELIA CHRISTODOULOU, JAKOB KATHER
OncoQwen (formerly Oncollama):
AI-Powered Cancer Data Extraction Across the NHS
[NHS Cambridge University Hospitals NHS Foundation Trust logo]
What is OncoQwen?
• NHS open licensed product built by AI Centre with support from OneLondon
• A MESA Framework (Medical entity Extraction with Schema Alignment)
• Extracts diagnosis, staging, biomarkers & outcomes from free text
• Runs inside Trust firewalls — no patient data leaves the NHS
• Already processed 500k+ documents at GSTT
3 Impact Workstreams
Q1 Inequalities
Multi-region pan-cancer analysis of stage at diagnosis & outcomes by age, sex, ethnicity & deprivation
Q2 COSD Completeness
Compare OncoQwen vs. national cancer registry to quantify missing staging & biomarker data
Q3 Clinical Trials
Automate patient eligibility screening & auto-populate case report forms (EHR2EDC)
Scale & Partners
5 NHS Centres
20 Cancer Types
£300k Budget
3 SDEs Participating
Development phase: model trained on pan cancer synthetic documents
Validation phase: model trained on 6 cancer types: myeloma, breast, ovarian, melanoma, lung and colorectal
Lead: Dr. Joe Zhang, Guy's & St Thomas' NHS FT | OneLondon SDE | Collaboration-focused Driver Project
[AI Centre for Value Based Healthcare logo]
[Speaker panel] Florian Markowetz — Generative AI and AI biomarkers in precision medicine: Advancing regulation and clinical impact in oncology and beyond
CADIZ AUDITORIUM - NCC
[Slide 2]
13:30 - 15:00 Transforming oncology with generative AI and autonomous tools
CHAIRS: EVANGELIA CHRISTODOULOU, JAKOB KATHER
OncoQwen: the pathway from research to clinic
[NHS Cambridge University Hospitals NHS Foundation Trust logo]
Governance and approvals at Cambridge University Hospital, in the order they were completed
1 AGREEMENT — Licensed — OncoQwen licensed by London AI Centre @ Guy's and St Thomas' NHS Trust
2 AGREEMENT — MOU signed — Memorandum of Understanding signed with CUH
3 APPROVAL (12/2025) — Committee sign-off — As a quality improvement project, signed off by the NHS Investment Committee
4 APPROVAL — Divisional approval — Approval given at CUH Division B Business and Improvement Meeting
5 APPROVAL — DPIA signed — Data Privacy Impact Assessment, eHospital form signed off
6 SDE GOVERNANCE — SDE access request — East of England SDE Data Access Request form, as an SDE Driver project
7 PLATFORM — Platform protocol — Protocol between Newton's Tree (the platform deploying the LLM on NHS data) and CUH
8 SDE GOVERNANCE — DAC data coverage — OncoQwen Data Coverage for SDE DAC document, alongside schema and output example
9 ACCESS — Terms of Use — Signed by everyone accessing the LLM output data in the SDE
OUTCOME (09/2026) — Governed use — OncoQwen output data accessed in the East of England SDE by approved users
Lead: Dr. Joe Zhang, Guy's & St Thomas' NHS FT | OneLondon SDE | Collaboration-focused Driver Project
[Speaker panel] Florian Markowetz — Generative AI and AI biomarkers in precision medicine: Advancing regulation and clinical impact in oncology and beyond
CADIZ AUDITORIUM - NCC
[Slide 3]
13:30 - 15:00 Transforming oncology with generative AI and autonomous tools
CHAIRS: EVANGELIA CHRISTODOULOU, JAKOB KATHER
CIN signatures to clinic: technical challenges
Moving the Drews et al. 2022 CIN signatures from academic software into NHS use
1 Code quality
• Version control and code review
• Unit and regression tests
• Locked dependencies, reproducible runs
2 QMS and ISO
• ISO 13485 quality management
• IEC 62304 lifecycle, ISO 14971 risk
• MHRA status, DCB0129/0160 safety
3 Licensing and IP
• Confirm licence allows NHS clinical use
• Audit dependency licences
• Agree IP, liability and support
4 Newton's Tree
• Containerised pipeline inside the Trust firewall
• Governed data access
• Version pinning, monitoring, rollback
Drews RM et al. A pan-cancer compendium of chromosomal instability. Nature 2022
[Speaker panel] Florian Markowetz — Generative AI and AI biomarkers in precision medicine: Advancing regulation and clinical impact in oncology and beyond
CADIZ AUDITORIUM - NCC
[ESMO Congress Madrid 2026 logo]
[Slide 4]
13:30 - 15:00 Transforming oncology with generative AI and autonomous tools
CHAIRS: EVANGELIA CHRISTODOULOU, JAKOB KATHER
CIN signatures to clinic: goals of deployment
From retrospective validation to biomarker-stratified randomised trials
1 Validation
• Lock thresholds, test archived cohorts
• Analytical and clinical validity
2 Test comparison
• Same samples, head-to-head
• CHORD, Myriad, HRDetect
3 NHS pathway study
• Live use under the QMS
• Pre-specified analysis plan
4 Stratified RCT
• CIN-guided vs standard of care
• Evidence for NHS adoption
Existing methods for comparison
Method | Features | Assay input
CIN signatures (Drews 2022) | 17 copy-number signatures | Copy-number profiles
CHORD | SNV, indel and SV signatures | Whole-genome sequencing
HRDetect | Six HRD-associated signatures | Whole-genome sequencing
Myriad myChoice CDx | LOH, TAI, LST plus BRCA1/2 | Proprietary tumour assay
Drews RM et al. A pan-cancer compendium of chromosomal instability. Nature 2022
[Speaker panel] Florian Markowetz — Generative AI and AI biomarkers in precision medicine: Advancing regulation and clinical impact in oncology and beyond
CADIZ AUDITORIUM - NCC
[ESMO Congress Madrid 2026 logo]
@NatureMedicine 2/ Rather than broad CD3 activation, invika selectively engages Vβ6/Vβ10 T cells (~10% of TILs), binding their TCR and IL-2 receptor in cis. In 52 heavily pretreated patients, no MTD was reached, and T cell expansion peaked at 0.08–0.12 mg/kg, supporting 0.08 mg/kg as the RP2D. https://t.co/RrUxAuoiX7
[Slide 1]
e
T cell expansion in periphery: Baseline 6.8% → Day 8 50.983%
T cell expansion in tumor: Baseline 11.9% → Day 21 41.99%
Vβ6 T cells
Vβ10 T cells
Other Vβ cells
[Slide 2]
f
CD3+ CD8+ CD25+ Baseline
CD3+ CD8+ CD25+ Day 21
The challenge: Women are underrepresented in leadership positions in oncology
#ESMO26 https://t.co/424NN6mHSn
[Slide 1]
[MADRID 2026 ESMO congress logo]
Women underrepresented in leadership position
• Although more women than men are entering medical schools, women remain underrepresented in academic medicine1: graduates now approach 50%, female department leaders and full professors remain under 30%
• Academic medical oncology: women represented 37.1% of faculty, 31.4% of leadership positions and 21.7% of department chairs2
• Women's underrepresentation limits their opportunities for authorship, peer review, funding and promotion, further perpetuating their underrepresentation3
Women in Academic Medicine 2023–2024 [infographic: 57%, 52%, 48%, 45%, 45%, 34%, 29%, 27%, 27%, 25%; category labels illegible]
Academic medical oncology [bar chart Male/Female: Total Faculty, Chair and Program Director, Chair; y-axis Representation, %]
Vicious circle of disadvantage [illustration; small text illegible]
ESMO authorship study4
• Female authors were significantly less likely to be last than first authors (p<0.001 for all years), indicating that women were less likely to be the leaders of a study
ESMO authorship study, 2017–20224 [bar chart First author / Last author: 2017 38%/30%; 2018 37%/24%; 2019 41%/28.5%; 2020 37%/29%; 2021 36%/29%; 2022 39%/30%]
JCO Global Oncology (2015–2020)5: 645 articles
• Female (41.4%) first authors and 190 (32.1%) last authors
• Trend toward higher proportion of female first authors in the most recent years
• Variation in author gender distribution by article type
Female: 41% first author, 32% last author5 [bar chart]
Female first author by article type5 [bar chart]
Female last author by article type5 [bar chart]
JAMA Oncology (2003–2018)6: 598 RCTs
• 18% female corresponding authorship among 598 oncologic RCTs (2003–2018); lower rates for industry-funded trials (14.4%) versus cooperative group trials (25.9%)
1AAMC, State of Women in Academic Medicine 2023–2024 · 2Chowdhry M, JAMA Netw Open 2020 · 3Clark J, Lancet 2019 · 4Linardou H, Ann Oncol 2024;35(Suppl 2):S938–S940 · 5Hornstein P, JCO Glob Oncol 2021 · 6Ludmir E, JAMA Oncol 2019
Prof. Pilar Garrido
Amazing discussion at #ESMO2026 @myESMO of Dr R Jones about intermediate mesenchymal tumours 👏🏻👏🏻 https://t.co/NfSENElsaS
[Slide 1]
TGCT & Desmoid: Intermediate Tumours
- Context: Sarcoma oncologists manage a variety of tumours
- "Not all sarcomas are the same"
- but we also treat a number of locally aggressive tumours
- "Oncologists are really good at treating intermediate tumours."
- Simpler biology
- Proof of principle in TGCT, giant cell tumour
- Sarcomas complex biology
- Need better understanding of biology
- Artemis trial in ESMO Plenary Monday
[Slide 2]
TGCT & Desmoid: Discussion
- Are randomized trials always required for registration/reimbursement?
- Should Patient Reported Outcomes (PROs) be the primary endpoint for clinical trials?
- Should the economic burden be more accurately quantified?
- What is the optimal duration of systemic therapy?
- What is the optimal sequence of systemic therapy?
[Slide 3]
TGCT: Randomized Phase 3 Trials
Trial | Patient Number | Response Rate RECIST 1.1 | Progression RECIST 1.1 | Tumour Volume Score Response | Quality of Life
Pexidartinib | 61 | 39% | 2% | 56% | Improvement
Placebo | 59 | 0% p<0.0001 | 2% | 0% p<0.0001 |
Vimseltinib | 83 | 40% | 0 | 67% | Improvement
Placebo | 40 | 0% p<0.0001 | 0 | 0% p<0.0001 |
Pimicotinib | 63 | 54% | 2% | 61.9% | Improvement
Placebo | 31 | 3.2% p<0.0001 | 0 | 3.2% |
Emactuzumab | 91 | 29.7% | 0 | 53.8% | Improvement
Placebo | 45 | 2.2% p<0.001 | 0 | 3% p<0.001 |
References: Tap WD et al. Lancet 394(10197): 478-487: 2019; Gelderblom H et al. Lancet 403(10445): 2709-2719: 2024; Xu H et al. Lancet 407(10533): 1072-1083: 2026; Gelderblom H et al. ESMO 2026
#ESMO26
China's growing influence in the global biotech sector https://t.co/iI3acvtcBN
[Slide 1]
4. Boom of Chinese Biotechnology Sector
KEY ISSUE
China's growing influence in the global biotech sector.
China's emergence as a global biotechnology powerhouse is increasingly driven by innovation and R&D capacity.
KEY INSIGHT
Sustaining this momentum will require deeper international scientific collaboration, talent exchange, and integration into global research ecosystems.
ESMO COMMITMENT
ESMO promotes global oncology research partnerships and international scientific collaboration to accelerate innovation and improve outcomes for patients with cancer everywhere.
China's biotech boom: why the nation must collaborate to stay ahead
Lizzi CL & Jing Qian, Nature (2026)
[ESMO Congress Madrid 2026 logo]
Improvement in Europe's Clinical Trial System
#ESMO26 https://t.co/A7djR2uy9F
[Slide 1]
5. Europe's Clinical Trials Ecosytem
KEY ISSUE
Regulatory barriers to oncology innovation in Europe.
The revision of the EU's IVD Regulation and proposed Biotech Act can streamline clinical trials, precision diagnostics and advanced therapies.
KEY INSIGHT
Faster, coordinated procedures can accelerate oncology research and patient access to innovative treatments.
ESMO COMMITMENT
ESMO has built the evidence, drafted legislative solutions and brought oncology's priorities directly to EU policy makers.
ESMO responses to the proposed revision of the In Vitro Diagnostic Medical Devices Regulation (IVDR) and the EU Biotech Act I
[ESMO Congress Madrid 2026 logo]
#ESMO26
Global health threats like war, climate change and even AI can affect how we treat cancer https://t.co/46qqiHQtXd
[Slide 1]
9. 21st Century Global Health Threats
KEY ISSUE
Growing global threats and health inequalities.
Climate change, AI, pandemics, armed conflicts and antimicrobial resistance are widening health inequalities and straining health systems.
KEY INSIGHT
Global health threats require coordinated responses to tackle health inequalities and protect the delivery of care.
ESMO COMMITMENT
ESMO is a privileged partner of the WHO and UN, and advocates for equal access to optimal cancer care for all patients.
The Lancet Commission on 21st century global threats to health
Christopher J L Murray et al.; The Lancet (2026)
[ESMO Congress Madrid 2026 logo]
🔹 #ESMO2026 starts now! 🔹
Prof. @FAndreMD, President of @myESMO, opens the congress, marking the start of several days dedicated to the latest advances in #oncology.
An opportunity for physician-researchers to showcase @GustaveRoussy’s expertise.
#ESMO #Cancer https://t.co/GY2krS4szc
[Slide 2]
What is the role of ESMO in a world where AI is predominant
A point of reference for
Cancer care providers
Patients
Decision-makers
Future AI systems
[concentric circles diagram, centre:] ESMO
We define what matters
We set the standards
We fund the work
[Slide repeated on right screen; live video of presenter in centre]
MADRID 2026 ESMO congress
@myESMO 🔹 Gustave Roussy at #ESMO26 🔹
Today's poster sessions:
• Dr Nathan El Ghazzi
• Pr Laurence @AlbigesL
@myESMO #ESMO https://t.co/KHL3WXDxhA
[Slide 1]
4395P
[Gustave Roussy Paris Cancer Campus logo] [ESMO congress logo] FPN 4395P MADRID
A 25-ge[ne T-]cell exhaustion signature refines TLS-based prediction of immunotherapy benefit
N. El Ghazzi, A. Elliott, G. Sledge, J. Xiu, E.S. Antonarakis, G. D'Amato, D. Hsieh, J.C. Trent, A. Italiano
Gustave Roussy, Villejuif, France; Caris Life Sciences, Phoenix, USA; University of Minnesota, Minneapolis, USA; University of Miami, Miami, USA; University of California San Francisco, San Francisco, USA
BACKGROUND
Immune engagement and infiltration alone do not ensure durable benefit from checkpoint blockade.
Published in 2024, the 25-gene score was derived from lung adenocarcinoma samples by identifying CD8+ T cells co-expressing PD1 and >=1 of LAG3, TIGIT or TIM3.[1]
mIF profiling 166 LUAD -> Transcriptomics 135 tumors -> 25-gene model LASSO-derived -> Prior validation NSCLC - melanoma - RCC
External validation in POPLAR/OAK, MATCH-R, CA209-038 and JAVELIN Renal 101 linked higher scores to poorer ICI outcomes.
Can this 25-gene exhaustion score refine pan-cancer ICI stratification alongside TLS?
METHODS
Retrospective Caris clinicogenomic analysis
- Whole-transcriptome RNA-seq from the Caris database (>50 cancer types)
- Endpoints: real-world time on treatment (TOT) and overall survival (OS)
- Quartile-based Kaplan-Meier analyses and Cox models adjusted for age, sex, cancer type*, PD-L1, TMB and MSI
204,231 tumors - >50 cancer types -> 41,509 ICI TOT; OS: 41,507 | 162,610 non-ICI parent cohort
*Cancer type: NSCLC vs other. TOT is not RECIST-defined PFS.
RESULTS
1 Higher exhaustion is associated with shorter ICI outcomes
TIME ON TREATMENT (Median, months): Q1 4,90; Q2 4,63; Q3 4,33; Q4 4,20
OVERALL SURVIVAL (Median, months): Q1 22,17; Q2 20,00; Q3 19,83; Q4 17,63
Q1 = lowest; Q4 = highest exhaustion. TOT n = 41,509, OS n = 41,507. Adjusted for age, sex, cancer type (NSCLC vs other), MSI, TMB and PD-L1
MULTIVARIABLE ANA[LYSIS] Q4 versus Q1 adjuste[d]: Time on treatment 1.10 (95% CI 1.06-1.[illegible]); Overall survival 1.18 (95% CI 1.13-1.23); P < 0.0001 for both endpoints
2 The association was not reproduced in the non-ICI cohort
162,610 parent non-ICI cohort: The adverse association seen under ICI was not reproduced in the non-ICI cohort.
TO ESTABLISH TREATMENT SPECIFI[CITY]: Adjusted non-ICI models and a formal [...] exhaustion interaction are nee[ded] [partially obscured]
3 Joint TLS / exhaustion classification refines ICI prognostication
TIME ON TREATMENT (TOT) Median, months: TLS-high Exh-low 5.13; TLS-high Exh-high 4.93; TLS-low Exh-low 3.63; TLS-low Exh-high 3.73
OVERALL SURVIVAL (OS) Median, months: TLS-high Exh-low 22.53; TLS-high Exh-high 21.63; TLS-low Exh-low 16.27; TLS-low Exh-high 15.70
TLS-high / Exh-low vs TLS-lo[w ...]: OVERALL SURVIVAL HR [obscured]; TIME ON TREATMENT [obscured]
Exhaustion was inversely correlated with TLS and most inferred immune populations
CONCLUSIONS / TAKE-HOME MESSAGE
1 The 25-gene T-cell exhaustion signature independently predicts poorer ICI outcomes beyond PD-L1, TMB and MSI.
2 Joint TLS/exhaustion profiling may refine ICI selection by distinguishing immune-engaged tumors with preserved versus impaired antitumor functionality.
[1] Guegan JP, Peyraud F, Dadone-Montaudie B [...] adenocarcinoma reveals a 2[...] Cell Reports Medi[cine] [partially obscured]
Contact: antoine.italiano@gustaveroussy.fr
[Presenter Nathan El Ghazzi standing beside poster]
[Slide 2]
3636TiP
LITESPARK-034: Phase 3 Study of Belzutifan Plus Zanzalintinib Versus Belzutifan in Participants with Advanced Renal Cell Carcinoma During or After Anti-PD-(L)1 Therapy and VEGF-TKI
Laurence Albiges, Bradley McGregor, Brian Rini, Lis Starkopf, Rodolfo F. Perini, Ali Rana, Robert J. Motzer [affiliations small]
Background [small text, partially legible]: Belzutifan, a potent and selective hypoxia-inducible factor 2α (HIF-2α) inhibitor ...; Zanzalintinib is a multitargeted kinase inhibitor that targets VEGF receptors, MET, and the TAM family of kinases ...; The randomized, double-blind, phase 3 LITESPARK-034 study (NCT[illegible]) is designed to evaluate the efficacy and safety of belzutifan plus zanzalintinib versus belzutifan plus placebo in participants with advanced clear cell RCC ...
Objectives
Primary: To compare the following for belzutifan plus zanzalintinib versus belzutifan plus placebo - PFS per RECIST v1.1 as assessed by blinded independent central review (BICR) - Overall survival (OS)
Key secondary: To compare ORR per RECIST v1.1 by BICR for belzutifan plus zanzalintinib versus belzutifan plus placebo
Secondary: To evaluate the following for belzutifan plus zanzalintinib and belzutifan plus placebo: - Duration of response (DOR) per RECIST v1.1 as assessed by BICR - Safety - Patient-reported outcomes (PROs)
Methods - Study design: [Key Eligibility Criteria box, small] R 1:1 -> Belzutifan 120 mg PO QD + zanzalintinib 80 mg PO QD | Belzutifan 120 mg PO QD + placebo PO QD; Primary end points: PFS per RECIST v1.1 by BICR and OS; Key secondary end point: ORR per RECIST v1.1 by BICR; Secondary end points: DOR per RECIST v1.1 by BICR, safety, and PROs
[Participant eligibility criteria, Assessments and follow-up, Analyses, Status (world map) panels: small text largely illegible]
Sites currently open to enrollment [small]
3637TiP
Phase 3 study to assess the safety and efficacy of 177Lu-girentuximab in advanced, relapsed or recurrent ccRCC (LUTEON)
Laurence Albiges, Andrew M Scott, Thea Faivre, David Cade, Sumanta Pal [affiliations small] [Telix / LUTEON logos]
Background: high unmet need for systemic therapy options for patients with advanced ccRCC who experience high recurrence rates & poor outcomes [body text small, partially legible]
Carbonic anhydrase IX (CAIX): a promising theranostic target for ccRCC; 177Lu-girentuximab: monoclonal antibody-based approach [details small]
Study design: 2-part, randomized, open-label, multicenter Phase 3 study (NCT[illegible])
Part 1: Dose optimization (Phase 2a): One 1887 MBq 177Lu-girentuximab dose every 2 months for 8 months vs One 1258 MBq 177Lu-girentuximab dose every month for 6 months [read at reduced confidence]
Part 2: Efficacy & safety evaluation (Phase 3): RP2D 1:1 -> 177Lu-girentuximab (RP2D) vs Investigators' choice (SOC)
Key patient eligibility [small, partially legible]
[Presenter Laurence Albiges standing between posters]
@A_May_MD "s(t?)imulating the tumor better than the tumor itself first" means. Could you clarify a bit? As a reminder, that's the rationale of SWOG-1801 in the cartoon. They basicallly say that using pembro as neadj would more prime T-cells at the primary (resected) tumor site, rather than https://t.co/WQNzmqPmOe
[Slide 1]
Figure S1. Cartoons of the hypothesized mechanism of action of PD-1 blockade therapy in the adjuvant and neoadjuvant settings. A, B and C) Adjuvant therapy. Before surgery, the majority of antitumor T cells inhibited by PD-1 are located at the sites of locally advanced disease (A), which are resected with surgery (B). When giving adjuvant anti-PD-1, the therapy relies on the existence of antitumor T cells inhibited by PD-1 at micrometastases sites (C). D, E and F) Neoadjuvant therapy. Administering neoadjuvant anti-PD-1 therapy is intended to activate the large numbers of antitumor T cells at the tumor site (D) before the definitive surgery (E). The locally-activated antitumor T cells then go onto induce a systemic immune response able to recognize distant micrometastatic sites (F).
[Panels A-F: illustrations, no further text]
🔹 Gustave Roussy at #ESMO26 🔹
Dr Caroline Even presents the challenges in treating HPV-related head and neck cancers, with a focus on systemic approaches, at @myESMO.
#ESMO https://t.co/6BoZEogpmM
[Slide 1]
EGFR inhibition in LA HPV+ HNSCC
De-escalation
De-ESCALaTE: randomised phase III (Mehanna, Lancet 2019)
[low] risk HPV+ Oropharyngeal Carcinoma T3N0-T4N0, and T1N1-T4N3
168 pts with RT + cetuximab 400mg/m2 loading dose followed by 7 w cetux 250mg/m2
166 pts with RT + cisplatine 100mg/m2/3w, D1, D22, D43
[Table 2: Mean number of acute, late, and overall toxicity events per patient, by treatment group - cisplatin plus radiotherapy vs cetuximab plus radiotherapy; values too small to read reliably]
[KM A Overall survival: Cisplatin plus radiotherapy vs Cetuximab plus radiotherapy; HR 5.0 (95% CI 1.7-14.7); Log-rank p=0.0012; Number at risk: cisplatin 166, 160, 154, 147, 118; cetuximab 168, 163, 156, 144, 109; Time since randomisation (years) 0, 0.5, 1.0, 1.5, 2.0]
RTOG 1016 (Gillison, Lancet 2019)
HPV+ Oropharyngeal Carcinoma (T1-T2, N2a-N3 or T3-T4, N0-N3)
425 pts with RT + cetuximab 400mg/m2 loading dose followed by 7 w cetux 250mg/m2
424 pts with RT + cisplatine 100mg/m2/3w, D1, D22
[KM A: Intensity-modulated radiotherapy plus cisplatin (406 patients, 55 dead, 351 censored) vs plus cetuximab (399 patients, 78 dead, 321 censored); Years after randomisation 0-5; Number at risk: cisplatin 406, 372, 349, 334, 222, 100 [partially legible]; cetuximab 399, 367, 334, 305, 207, 106 [partially legible]]
Overall survival was significantly worse with cetuximab (two sided 95% CI 1.03-2.05; log-rank p=0.0163) than with cisplatin. Estimated 5-year overall survival was : 77.9% (95% CI 73.4-82.5%) in the cetuximab group and 84.6% (80.6-88.6) in the cisplatin group
[ESMO Congress logo; photo of presenter at podium and panel; podium: MADRID SPAIN 23-27 OCTOBER 2026]
Congratulations to our fellow @JavierDavidBen2 on a fantastic Rapid Oral presentation just now at the policy session at #ESMO26. https://t.co/e1sqBbEO0a
[Slide 2]
Results: reporting of planned assessments
The reporting denominator is the number meeting the full planning criterion.
Assessment | Planning fully met n / 31 (%) | Reporting fully met n / N (%)
Health-related quality of life | 23/31 (74.2%) | 3/23 (13.0%)
Chronic / low-grade toxicity over time | 7/31 (22.6%) | 1/7 (14.3%)
Patient-reported toxicity | 7/31 (22.6%) | 0/7 (0.0%)
Javier David Benitez-Fuentes | 3323RO
5/7
[Slide 3]
Results: primary endpoints and eligibility
4/7
OS primary / co-primary
All 31 trials
38.7%
12/31 trials
Primary: 12/31 (38.7%)
Not primary: 19/31 (61.3%)
ECOG PS > 1 eligible
All 31 trials
9.7%
3/31 trials
Included: 3/31 (9.7%)
Excluded: 28/31 (90.3%)
PS > 1 exclusion rationale
Among 28 PS 0-1 trials
0/28
fully justified
Limited: 3/28 (10.7%)
None: 25/28 (89.3%)
0/19 trials without primary OS met the full surrogate-validation criterion.
Javier David Benitez-Fuentes | 3323RO
[Slide 4]
Common Sense Oncology: the checklist
1/7
Phase III trials should test meaningful patient benefit.
DESIGN
[A]ppropriate comparator
[O]S or validated surrogate
[Q]uality of life
[C]linically meaningful
ANALYSIS
Censoring
Sensitivity analyses
Post-progression treatment
REPORTING
Absolute benefit
Quality of life and toxicity
Balanced interpretation
[Publ]ished in 2025: principles for design, analysis and reporting.
[...] B et al. Lancet Oncol. 2025;26:e80-e89. doi:10.1016/S1470-2045(24)00451-0.
[Javier D]avid Benitez-Fuentes | 3323RO
Very nice presentation on DDR defects, a tumor-agnostic hallmark with clinical implications.
1 pathway, 4 uses: therapeutic (PAPRi/chemo/ADCs/IO), resistance, prognostic, and genetic counseling. Interacts with all other hallmarks.
- @GhosnMd
📸 Sophie Postel-Vinay, MD, PhD
#ESMO26 @myESMO @GustaveRoussy @GRAlumni
[Slide 1]
Inserm | Université Paris-Saclay | UCL
DNA damage response (DDR) defects: a key cancer hallmark
Multiple implications in the clinic:
• Therapeutic (synthetic lethality, sensitivity to chemo- and radiotherapy; ADCs; ICI)
• Resistance
• Prognostic
• Genetic counselling
• Tumor-agnostic approach is relevant for therapeutic orientation
• Key hallmark: Interaction with multiple other cancer hallmarks
S. Postel-Vinay
EACR | ESMO congress
Proud to hear how Brazil is approaching heath care equity through digital pathology hub-and-spoke model for diagnostics at #ESMO26 https://t.co/iDcDEAwwjB
[Slide 1]
[photo of stage panel with 3 speakers; two screens showing the same slide]
A connected digital pathology [network to expand access,] standardise quality and reduce health [disparities in Brazil] [title partly cropped on right screen; full title on left screen: "A connected digital pathology network to expand access, standardise quality and reduce health disparities in Brazil"]
[map of Brazil with regional hubs]
NORTH
NORTH-EAST
WEST-CENTRAL
SOUTH-EAST
SOUTH
ACCC Central Hub (São Paulo)
Reference centre for complex cases and expert consultation
LEGEND
Physical sample flow (for regional hubs)
Digital images and data flow
Local Centre (hospital or laboratory)
Regional Digital Pathology Hub (slide scanning and case preparation)
ACCC Central Hub (reference for complex and expert consultation)
Tissue specimen (biopsy or surgical specimen)
[left screen also shows a right-hand step panel, mostly illegible: "Sample collection at local centres" — uncertain; "Physical shipment to regional hubs"; "Slide scanning at regional hubs"; "Digital images sent for analysis"; "Interpretation and collaboration"; "Report ..." — [illegible] small text]
The ESMO president Fabrice Andre at the opening sesion #ESMO26 Madrid.
Great talk! https://t.co/pc3rI0t5ex
[Slide 2]
[Phone screenshot of esmocongress.esmo.org livestream; presenter inset at right]
Keeping pace with science is no longer enough
ESMO's evolution is the response
[tree diagram branching to four items]
Complexity
Fragmentation
Data
Need for guidance
[Slide 4]
[Phone screenshot of esmocongress.esmo.org livestream; presenter inset at right]
Serving more than 52,000 oncology professionals
43
Guidelines
13
ESMO journals
Including 3 new journals:
• Artificial intelligence and oncology
• Oncology policy and science
• Genitourinary oncology
13
Meetings and congresses around the world
Including:
• 1 AI congress
• 2 new Asia congresses
30+
Webinars
#ESMO26 by the numbers. A global family of 40K delegates! https://t.co/SYFGCNJe07
[Slide 1]
ESMO 2026 IN NUMBERS
4,414 Abstracts:
• 12 in Presidential Symposia
• 147 Proffered Papers
• 242 Rapid Oral
• 2,025 Posters
• 1,988 ePosters
296 Sessions, of which:
• 230 Programme Sessions
• 66 Industry Satellite Symposia
~ 40,000
Professionals attending
[ESMO Congress Madrid 2026 logo]
Now public at @myESMO: In our phase Ia/Ib study, patients received RS-0139 at 40–60 mg/m² without a single case of neutropenia of any grade and without G-CSF support. Find the e-poster 1095eP or reach out to continue the conversation.
#ClinicalTrials #ESMO26 #CureWithASmile https://t.co/RQ7mZu4xf2
Secondary commentary
KOL infographics 15
Summary graphics that physicians, media and companies designed themselves (branded cards, redraws, watchlists and top-10 lists). They are commentary, not the presented data, so they sit here in compact form. Click a thumbnail to enlarge it on this page.
Heme & Myeloma 2
Henry C Fung| MM, lymphoma, leukemia & CART @HenrychihangFu1Post text
🩸🇪🇸 ESMO 2026 MYELOMA + MADRID GUIDE
Myeloma in the morning. Tapas at night. 😆
📍 Madrid | Oct 23–27
⭐ Essential session: Sat Oct 24, 08:30–10:00
Salamanca Auditorium, Hall 10
Three myeloma oral presentations:
1️⃣ 1454O—Bela-DRd in transplant-ineligible NDMM
High MRD negativity sounds excellent—but what does belamaf truly add beyond DRd? We need durable PFS, tolerability and ocular-treatment burden—not another impressive MRD snapshot.
2️⃣ 1455O—Long-term BelaRd
The important question is durability. Is frontline BCMA exposure justified when CAR-T and bispecific antibodies may be needed later?
3️⃣ 1456O—LBL-034, a new GPRC5D×CD3 bispecific in heavily pretreated RRMM
Phase I and single-arm: promising, not practice-changing. Watch duration of response, infections, EMD activity and the familiar GPRC5D taste/skin/nail toxicities.
BOTTOM LINE:
ESMO myeloma is small but focused:
➡️ Belamaf moving earlier
➡️ Another GPRC5D bispecific moving forward
➡️ No phase III result likely to change practice immediately
🍖 NOW THE IMPORTANT SESSION: IBÉRICO!
Easiest post-ESMO stop:
🥇 Cinco Jotas Padre Damián—excellent 100% ibérico de bellota and convenient after taking Metro Line 8 from IFEMA toward Nuevos Ministerios.
For the complete Madrid experience:
🥓 Joselito’s Velázquez—destination-quality jamón
🍤 La Casa del Abuelo—gambas al ajillo
🍳 Casa Lucio/Huevos de Lucio—huevos rotos
🍺 Bodega de La Ardosa—tortilla, vermouth and old-Madrid atmosphere
Scientific rule: Do not confuse MRD negativity with cure.
Madrid rule: Do not confuse ordinary ham with jamón ibérico de bellota. 😆
#ESMO2026 #MultipleMyeloma
[Slide 1]
ESMO 2026: MYELOMA + MADRID
MYELOMA IN THE MORNING • TAPAS AT NIGHT
SAT OCT 24 • 08:30–10:00
Salamanca Auditorium • Hall 10
1454O Bela-DRd — Does deeper MRD become durable benefit?
1455O BelaRd — Durability vs ocular burden
1456O LBL-034 • GPRC5D×CD3 — Promising Phase I—not practice changing
THE IBÉRICO ROUTE
1 Cinco Jotas Padre Damián — 100% ibérico de bellota • easiest from IFEMA
2 Joselito's Velázquez — Destination-quality jamón
3 Casa Lucio — Huevos rotos
4 La Casa del Abuelo — Gambas al ajillo
Do not confuse MRD negativity with cure.
Do not confuse ordinary ham with jamón ibérico de bellota. 😆
Henry C Fung| MM, lymphoma, leukemia & CART @HenrychihangFu1Post text
Figure 2/2: Not only myeloma!
The same Saturday session includes new approaches in B-cell lymphoma, transformed DLBCL and R/R MDS/AML. Interesting early-phase signals—but novel mechanism ≠ new standard.
Spanish wine, however, needs no Phase III trial. 🍷😆
#ESMO26 #HemeTwitter #Lymphoma #AML #MDS
[Slide 1]
ESMO 2026: BEYOND MYELOMA + SPANISH WINE
NOVEL TARGETS IN THE MORNING • SPANISH WINE AT NIGHT
SAT OCT 24 • 08:30-10:00
Salamanca Auditorium • Hall 10
1457O Rocbrutinib + R-CHOP - Frontline B-cell lymphoma • Phase Ib
1458O TRS005 • anti-CD20 ADC - Transformed DLBCL • ORR 50% • mPFS 8.7 mo
1459O LYT-200 • anti-Galectin-9 - R/R high-risk MDS/AML • Phase Ib
THE SPANISH WINE FLIGHT
1 Rioja Gran Reserva - Elegant • mature • classic
2 Ribera del Duero - Bold • structured • powerful
3 Albariño • Rías Baixas - Fresh • mineral • seafood-ready
4 Fino Sherry • Jerez - Dry • salty • perfect with jamón
Novel mechanism ≠ new standard.
Good Spanish wine needs no Phase III trial.
Other 13
Jeff Ryckman @jryckman3Post text
RADIANT trained on 119,782 RT cases and tested segmentation, dose and outcome prediction across sites.
What I want next is clinical utility: does using it improve planning decisions, toxicity or outcomes? Accuracy alone isn't enough.
#ESMO26 #radonc
https://t.co/0MTMp2F6iI https://t.co/eczRhHiVI8
[Slide 1]
RADIANT
Radiotherapy foundation model development and validation
CT, structures,
dose
14 centers
Pretraining cohort
n = 119,782 cases
→ Hold-out + external
validation
GTV Dice CTV Dice
83.52% 81.53%
Hold-out cohort: 22,058 cases across 10 disease sites; external retrospective/prospective cohorts in 3 countries.
21 downstream tasks include segmentation, dose and outcome prediction. No randomized clinical utility evaluation.
Zhang et al. ESMO 2026, 170O. AI model study; reported predictive performance does not establish patient benefit.
Jeff Ryckman @jryckman3Post text
FURTHER enrolled 69/129 planned pts.
At 2wk, pain response 50% in pooled MR-HIFU arms vs 38% with EBRT (P=.41).
Just refer these folks for EBRT. Most data for invasive ablation is confounded by receipt of EBRT. Love me some kypho after EBRT (for select cases), though
#ESMO26 #radonc
https://t.co/D7WeSP7cEe
[Slide 1]
FURTHER
MR-guided focused ultrasound with or without RT for bone pain
Painful non-spinal
Bone metastases
NRS >=4
n = 69
→ RANDOMIZE
EBRT (n = 24)
MR-HIFU (n = 22)
EBRT + MR-HIFU (n = 23)
*Primary endpoint: pain response at 2 weeks. Pooled MR-HIFU arms 50% vs EBRT 38% (P = .41), not significant.
Accrued 69/129 planned; underpowered. Pain response incorporates pain score and opioid use.
Braat et al. ESMO 2026, 3084O. 1:1:1 randomized; combined arm: MR-HIFU within 1 week of EBRT. RT dose unreported.
Jeff Ryckman @jryckman3Post text
Trotabresib added toxicity without helping newly diagnosed GBM.
12mo PFS 22% vs 39% w/ RT+TMZ alone (P=.048); OS similar.
G3+ thrombocytopenia 37% vs 15%.
#ESMO26 #radonc
https://t.co/fvr6SiCBoc https://t.co/cdf8w6jkIe
[Slide 1]
Trotabresib
Adding a BET inhibitor to chemoradiotherapy in glioblastoma
Newly diagnosed IDH-wildtype GBM
After surgery
n = 184
→ RANDOMIZE (Randomized phase II; 2:1)
RT + TMZ (n = 60) | 12-month PFS* 38.9%
+ trotabresib (n = 124) | 12-month PFS* 22.2%
*Primary endpoint: PFS by local RANO. 12-month PFS favored control (P = .048); HR not reported.
Median OS 16.7 vs 16.9 months (P = .898), not statistically significant. G3+ thrombocytopenia: 15.3% vs 37%.
Dipasquale et al. ESMO 2026, 692O. RT dose not reported in abstract; median follow-up 15 months.
Jeff Ryckman @jryckman3Post text
KEYNOTE-412 ctDNA: clearance after 5wk CRT was more frequent w/ pembro, but outcome associations varied by HPV status and arm.
Interesting biomarker work. Without an interaction test, I wouldn't call this predictive of pembro benefit.
#ESMO26 #radonc
https://t.co/HAS3ZYi4lX https://t.co/7rlojMGpuJ
[Slide 1]
KEYNOTE-412 ctDNA
Exploratory clearance analysis in a paired-sample subset
Locally advanced
Head and neck SCC
Paired plasma
samples
n = 473 of 804
Parent trial randomized
Subset not re-randomized
→ RANDOMIZE
ctDNA clearance HPV-negative ctDNA clearance HPV-positive
CRT + placebo
(n = 240) 63.6% 62.7%
CRT + pembro
(n = 233) 72.6% 76.8%
Clearance assessed after 5 weeks of CRT (baseline to cycle 3 day 1). Exploratory, retrospective analysis.
Outcome associations varied by HPV status and arm. No treatment-interaction test reported; not proof of prediction.
Machiels et al. ESMO 2026, 1536RO. Parent primary endpoint: EFS; these percentages are biomarker outcomes.
Jeff Ryckman @jryckman3Post text
GORTEC IDEAL is early: induction EGFR-ADC +/- anti-PD1, then 70 Gy/cisplatin. G3 TEAEs in 6/10 combination pts vs 2/10 ADC alone.
20 pts in this safety analysis.
#ESMO26 #radonc
https://t.co/WXt5sKlK9W https://t.co/lXFU9CQGPS
[Slide 1]
GORTEC IDEAL
Induction EGFR-ADC with or without anti-PD1 before CRT
Locally advanced
Head and neck SCC
n = 49 randomized
→ RANDOMIZE
Grade 3 TEAEs, safety subset
EGFR-ADC, then CRT 2/10
+ anti-PD1 6/10
Becotatug vedotin x3 +/- pucotenlimab, then 70 Gy + cisplatin; combination arm also receives adjuvant anti-PD1.
Phase II; 106 planned. First 20 safety-evaluable pts only. Primary CR at 12 weeks after CRT remains immature.
Tao et al. ESMO 2026, 1538RO. No grade 4 AEs; one death in ADC-only arm deemed unrelated. No efficacy conclusion.
Jeff Ryckman @jryckman3Post text
In resected brain mets progressing after RT: fewer CD8+ cells and more FOXP3+ cells than untreated tumors.
These are selected treatment failures. I wouldn't read this as evidence that RT broadly suppresses antitumor immunity.
#ESMO26 #radonc
https://t.co/dnJu8C8P22 https://t.co/Yu1JJE74mE
[Slide 1]
Progressive brain metastases
Molecular profiling of resected tumors by prior treatment
Resected brain mets
Progression after RT
or systemic therapy
+ untreated controls
n = 153
→ GROUPS
No prior therapy (n = 63)
Prior RT (n = 43)
Prior systemic therapy (n = 37)
Both (n = 10)
Exome sequencing, methylation profiling and immunohistochemistry; 35 paired initial/recurrent samples.
Prior-RT tumors had fewer CD3/CD8/CD45RO+ cells and more FOXP3+ cells than untreated tumors.
Steindl et al. ESMO 2026, 703RO. Observational, selected resected failures; no causal treatment comparison.
Richard Fitzgerald 💙 🦀💚 @r1chardf1tzg3r1Post text
#ESMO2026
“Beyond Guidelines”
Include Impact Of
🇬🇧 #DoctorSubstitution On
#Oncology Patient Outcomes
“ADVANCED PRACTITIONERS”
“PHYSICIAN ASSOCIATES “
@myESMO @ESTRO_RT @CPME_EUROPA @UEMSEurope
@AliceIndini @AndriPapakonst @AndrewWardley @TheLancetOncol
@TheBMA https://t.co/UlYlAbm2YW
[Slide 3]
Example 1
A cancer diagnosis was missed
One doctor reported that a patient with red-flag symptoms was repeatedly seen by a non-doctor, delaying diagnosis by around six weeks.
[Photo of man in waiting room]
[Slide 4]
Example 3
A patient wrongly admitted to end-of-life care
An ACP reviewed a cancer patient, determined she was at the end of life and admitted her to an end-of-life care room. She was later found to have treatable hypercalcaemia, improved with treatment and was discharged.
[Photo of patient in hospital gown]
Nicolò Battisti, MD @nicolobattistiMOIONSABP B-42KEYNOTE-905 / EV-303NATALEEAZUR-1KEYNOTE-C93lidERADESTINY-Breast05GIM3 FATATRIO-US B-12 TALENTDEWI (GERCOR-GONO)
Post text
Older adults at #ESMO26:
MOIO (LBA10) on 3-monthly checkpoint inhibitors, TOP-1 (545RO) on omitting RT in older women with ER+ breast cancer, B-42 (LBA24) on extended letrozole, EV-302 in older patients (4761P).
De-escalation is the theme this year.
#GeriatricOncology
#gerionc
@myESMO
@SIOGorg
@siognah
@SIOGnursing_AH
@YoungSIOG
@AlumniSiog
@royalmarsdenNHS
@ICR_London
@royalmarsden
[Slide 1]
ESMO CONGRESS 2026 · MADRID · 23-27 OCTOBER
Older adults with cancer at #ESMO26
10 presentations I will not miss
1 MOIO LBA10
Phase III non-inferiority trial of immune checkpoint inhibitors every 3 months in responding patients. Gwenaëlle Gravis.
Mon 26 Oct · 16:30-18:15 · Presidential Symposium III · Alicante Auditorium, Hall 6
2 TOP-1 (BOOG 2016-01) 545RO
5-year results of omitting radiotherapy after breast-conserving surgery in systemically untreated older women with ER+ early breast cancer. Lotte Bouwman.
Fri 23 Oct · 16:15-17:45 · Rapid oral, early breast cancer · Alicante Auditorium, Hall 6
3 NRG Oncology/NSABP B-42 LBA24
Final analysis of extended letrozole after 5 years of aromatase inhibitor-based therapy in postmenopausal women. Eleftherios P. Mamounas.
Fri 23 Oct · 13:30-15:00 · Proffered paper 1, breast cancer · Alicante Auditorium, Hall 6
4 GIM3 FATA 544RO
Final analysis: upfront versus switch schedule and three aromatase inhibitors head to head in postmenopausal HR+ early breast cancer. Pietro De Placido.
Fri 23 Oct · 16:15-17:45 · Rapid oral, early breast cancer · Alicante Auditorium, Hall 6
5 GERCOR-GONO DEWI 2877RO
Dostarlimab with watch-and-wait in localised dMMR/MSI-H gastro-oesophageal cancer, including a cohort of patients over 70 unfit for surgery. Thierry André.
Fri 23 Oct · 16:15-17:45 · Rapid oral, upper GI · hall to be confirmed
6 EV-302 in older patients 4761P
Updated exploratory analysis of enfortumab vedotin plus pembrolizumab in older patients and those with comorbidities. S. S. Sridhar.
Fri 23 Oct · poster on view 09:00-17:00, presenter 15:15-16:00 · Poster Area, Hall 5
7 KEYNOTE-C93 / GOG-3064 / ENGOT-en15 LBA12
Phase III first-line pembrolizumab versus carboplatin-paclitaxel in dMMR advanced or recurrent endometrial cancer. Brian M. Slomovitz.
Mon 26 Oct · 16:30-18:15 · Presidential Symposium III · Alicante Auditorium, Hall 6
8 AZUR-1 LBA87
First results of dostarlimab monotherapy for clinical complete response in dMMR/MSI-H locally advanced rectal cancer. Andrea Cercek.
Sun 25 Oct · 10:15-11:45 · Proffered paper, lower GI · hall to be confirmed
9 AIUR LBA28
Randomised trial of compression with standard surgical gloves to prevent paclitaxel-induced peripheral neuropathy. Young-Joon Kang.
Fri 23 Oct · 16:15-17:45 · Rapid oral, early breast cancer · Alicante Auditorium, Hall 6
10 HOPE LBA92
Phase III trial of a personalised digital supportive-care pathway for endocrine therapy symptoms in early breast cancer.
Mon 26 Oct · 08:30-10:00 · Proffered paper, supportive and palliative care · hall to be confirmed
Also on my list
Beyond guidelines: grey zones of oncology practice, with decision making in elderly and frail patients, Sat 24 Oct 10:15-11:45, Murcia Auditorium, Hall 3 · Overcoming barriers of access to clinical trials, Sat 24 Oct 08:30-10:00, Seville Auditorium · LOMAC (LBA81), 4 versus 6 months of adjuvant XELOX in gastric cancer · HER2CLIMB-05 HRQoL (1766RO), Sat 24 Oct 10:15-11:45, Almeria Auditorium, Hall 9 · lidERA (538RO) and DESTINY-Breast05 (539RO) patient-reported outcomes, Fri 23 Oct 16:15-17:45 · KEYNOTE-905 (4733O) and EV-103 (4740RO) in cisplatin-ineligible urothelial cancer · Dostarlimab real-world data by age (1286P), Mon 26 Oct poster · EONS19: quality of life for older adults with head and neck cancer (Cherith J. Semple), Sun 25 Oct 08:30 · TALENT (LBA23) and NATALEE 6-year OS (LBA25), Fri 23 Oct 13:30-15:00
Madrid local time. Clocks go back overnight on 24-25 October (CEST to CET). Late-breaking abstracts are embargoed until presentation; halls from secondary sources, check the programme app on the day.
@nicolobattisti
OncoAlert @OncoAlertPost text
🇪🇸 #ESMO26 Day 1 in numbers 📊 and it Just got Started
🔥 17.9M impressions
💬 1.1K posts
👥 437 participants
🌍 26 countries
⚡ 18.7 posts/hour
Proud to lead the conversation at #1 💪 Thank you @myESMO and everyone sharing the science from Madrid 🙌
Real-time metrics powered by AcuVox 🎙️ https://t.co/D5QkLIE1kW
@cdanicas @hoperugo @DrChoueiri @TiansterZhang @matteolambe @NiuSanford @stolaney1 @christine_lovly @amerseburger @GlopesMd @tompowles1 @montypal @Icro_Meattini @PGrivasMDPhD @brian_rini @CathyEngMD @apolo_andrea @AndreaNecchi @DrSGraff @BenWestphalen @graokane @to_be_elizabeth @ElisaAgostinett @nataliagandur @UOzkerim @DrMirallas @DrYukselUrun @Lucarecco @mzugman @HHorinouchi @p_ciracimd @DraMartinezLago @DrVilmaPBarcia @GIMedOnc @MarioBalsaMD @scocmem @AmandaNizamMD @GaiaGriguolo @JankovicK @Dr_Oncologista @Erman_Akkus @WeOncologists @niklas_kluemper @Markuseckstein3 @PTarantinoMD @ScottGottliebMD @VivekSubbiah @dr_yakupergun
Same image also posted by @OncoAlert
[Slide 1]
DAY 1 · SOCIAL PULSE ● LIVE
#ESMO26
Madrid.
23-27 OCTOBER 2026 | IFEMA MADRID
POTENTIAL REACH
17.9M
impressions on day one
1.1K Posts on X - clean, no bots or dupes
437 Participants - voices in the conversation
26 Countries - a global conversation
18.7 Posts / hour - across the meeting day
2.6 Posts / HCP - engagement depth
28% Positive - of scored posts
TOP INFLUENCERS (by reach · day 1)
1 OncoAlert @OncoAlert [X]
2 ESMO - Eur. Oncology @myESMO [X]
3 ESMO esmo_eur_oncology [IG]
4 Scott Gottlieb, MD @ScottGottliebMD [X]
5 Adam Feuerstein @adamfeuerstein [X]
6 Gilberto Lopes @GlopesMd [X]
7 Gil Morgan, MD @weoncologists [X]
8 Toni Choueiri, MD @DrChoueiri [X]
9 Vivek Subbiah, MD @VivekSubbiah [X]
10 Memorial Sloan Kettering @MSKCancerCenter [X]
11 Paolo Tarantino @PTarantinoMD [X]
12 The OncoAlert Netw... Bluesky [BSKY]
13 Yakup Ergün @dr_yakupergun [X]
14 Sumanta K. Pal, MD @montypal [X]
WHERE IT CAME FROM
Europe 363
North America 310
Asia-Pacific 100
Latin America 18
Middle East & Africa 2
TOP COUNTRIES
United States 275 | Spain 98
Switzerland 79 | Japan 63
Turkey 58 | United Kingdom 54
Canada 35 | Italy 25
TOP CITIES
Lugano 57
Boston 47
London 43
Tokyo 40
New York 34
Madrid 31
LANGUAGES
English 92%
Spanish 5%
Japanese · Italian · French 1% each
ACUVOX real-time congress metrics · by OncoBites x OncoAlert
ONCOLOGY INSIGHTS. REAL-TIME IMPACT.
EORTC @EORTCPost text
At today’s #ESMO2026 Policy session, Iryna Shakhnenko presented findings from an EORTC study on patient partner reviews of informed consent forms across 49 clinical trials.
👏 Congratulations to Iryna on presenting this work.
See the key highlights below. 👇 @myESMO https://t.co/D3LHK5HN9j
[Slide 1]
ESMO Rapid Oral session presented by:
Iryna Shakhnenko (Brussels, Belgium)
EORTC European Organisation for Research and Treatment of Cancer
Key messages
The challenge is not the amount of information, but how it is organised and presented
• Patients' understanding comes first
• Practical information matters
• Transparency builds trust
• Information should support participation, not just consent
Next: Survey → Focus groups → Recommendations → New EORTC PISIC template
ESMO
MitoAI @Vox_MitoAIDeLLphi-305INTerpath-001DASL-HiCaPKEYNOTE-B49ACTICCA-1VOLGADUO-EZW191BEHOLD-1 → BEHOLD-Ovarian01VIKTORIA-1TANGENTPROTEUSTROPION-Breast02FINER (CCTG MA.40)PACE-LUNGPOP-DURVACheckMate 8HWPRABITASBEATccKEYNOTE-B21
Post text
🔥 #ESMO2026 late-breakers 24 Oct scored 1–10 for SoC impact
🚀 9/10: VOLGA, DeLLphi-305, INTerpath-001
✅ 7–8: DASL-HiCaP, CheckMate 8HW, PRABITAS, BEATcc, DUO-E, PROTEUS, TANGENT
⏸️ No shift: FINER, KEYNOTE-B49, ACTICCA-1
Preliminary, pending full data. What moves practice most?
[Slide 1]
ESMO 2026 · SATURDAY 24 OCTOBER · LATE-BREAKING ABSTRACTS · PAGE 1 OF 2
HOT OFF THE PRESS
Late-Breaking Abstracts: Perspective
28 readouts scored for their impact on standard of care (1–10)
PRELIMINARY Abstract-level assessment, pending full data presentations. Scores may change.
28 LBAs released today and scored
DISTRIBUTION BY IMPACT TIER: 3 | 7 | 9 | 6 | 3
9–10 Paradigm-shifting | 7–8 Practice-changing | 5–6 Practice-informing | 3–4 Hypothesis-generating | 1–2 No shift / negative
LBAs BY TUMOR AREA: Breast 8 | Gynecologic 5 | Colorectal 3 | GU (prostate, bladder) 3 | Lung 3 | Hepatobiliary 2 | Pan-tumor 2 | Melanoma 1 | Sarcoma (TGCT) 1
9–10 Paradigm-shifting · Practice-defining — Phase 3 survival endpoints met in major unmet-need settings · 3 LBAs
LBA1 · MELANOMA 9/10 INTerpath-001 — Intismeran autogene + pembrolizumab · Resected stage IIB–IV — RFS HR 0.58 · DMFS HR 0.62 — First Phase 3 win for an individualized mRNA neoantigen therapy over anti-PD-1 alone. WNTB: Personalized manufacturing and supply chains scale to adjuvant time windows without treatment delays.
LBA2 · BLADDER 9/10 VOLGA — Perioperative durvalumab + EV · Cisplatin-ineligible MIBC — EFS HR 0.51 · OS HR 0.58 — First perioperative regimen to beat cystectomy on both EFS and OS in cisplatin-ineligible MIBC. WNTB: EV + IO skin and neuropathy toxicity stays acceptable in frail, elderly surgical candidates in community practice.
LBA3 · ES-SCLC 9/10 DeLLphi-305 — Tarlatamab + durvalumab · 1L maintenance — OS HR 0.42 · 21.1 vs 12.7 mo — Near-doubling of median OS sets a new benchmark for ES-SCLC maintenance. WNTB: Centers manage step-up dosing and CRS (49.6% any grade) in fragile outpatients without early mortality.
7–8 Practice-changing · Regimen-altering — Phase 3 superiority, smarter schedules, or OS confirmation · 7 LBAs
LBA4 · PROSTATE 8/10 DASL-HiCaP — Darolutamide + RT + ADT · Localized very high-risk / salvage — MFS-CI HR 0.42 — 4-yr 95% vs 88%; p = 0.0001. Supports ARPI intensification with curative RT. WNTB: MFS gain converts to OS without long-term cardiometabolic QoL cost.
LBA31 · COLORECTAL 8/10 PRABITAS — Biweekly FTD/TPI + bevacizumab · Refractory mCRC — OS HR 0.93 non-inferior — Severe neutropenia 13% vs 39% compared with standard 28-day dosing. WNTB: Guidelines and community practice pivot to biweekly dosing, cutting G-CSF use.
LBA32 · COLORECTAL 8/10 CheckMate 8HW — Nivolumab + ipilimumab · MSI-H/dMMR mCRC — OS HR 0.61 vs nivo — All lines (296 vs 286): 95% CI 0.46–0.82; p = 0.0009; medians NR. 1L vs chemo (171 vs 84): OS HR 0.37 (0.24–0.57), p < 0.0001; mOS NR vs 50.1 mo. WNTB: Long-term survival gain justifies higher immune toxicity vs single-agent PD-1.
LBA38 · CERVICAL 8/10 BEATcc — Atezolizumab + chemo + bevacizumab · 1L persistent/recurrent/metastatic — OS HR 0.69 — Mature mOS 31.9 vs 22.7 mo; p = 0.003; benefit regardless of PD-L1. WNTB: Atezolizumab used alongside pembrolizumab, notably in PD-L1 low/negative disease.
LBA37 · ENDOMETRIAL 7/10 DUO-E — Durvalumab + CP → durvalumab ± olaparib · 1L advanced/recurrent — OS HR 0.77 ITT — p = 0.037 for chemo + IO + PARPi; pMMR trend HR 0.85. WNTB: Added PARPi OS benefit in pMMR/HRD warrants dual-maintenance toxicity and cost.
LBA53 · PROSTATE 7/10 PROTEUS substudy — Apalutamide + ADT + RP · High-risk localized / locally advanced — EFS HR 0.38 — mEFS NR vs 19.1 mo; p < 0.0001; less salvage therapy. WNTB: Surgery + ARPI yields MFS/OS gains beyond RT + long-term ARPI.
LBA89 · TGCT 7/10 TANGENT — Emactuzumab (CSF-1R mAb) · Symptomatic advanced TGCT — ORR 29.7% RECIST — TVS ORR 53.8%; better physical function, stiffness and pain. WNTB: A 5-dose fixed course replaces chronic daily oral CSF-1R inhibitors.
TODAY'S THEMES: Novel modalities lead: mRNA neoantigen therapy, a bispecific T-cell engager and perioperative ADC + IO top the board. GU is the busiest curative-intent arena, with ARPI intensification around both RT and surgery. Breast is the most represented area (8 LBAs) but has no readout above 6/10.
HOW TO READ THIS · SCORING KEY
9–10 Paradigm-shifting. Phase 3 OS/EFS wins in major unmet need; new modality or new standard.
7–8 Practice-changing. Phase 3 superiority vs active standards, less-toxic schedules, or OS confirmation.
5–6 Practice-informing. Positive in narrow subsets, secondary endpoints, or PFS without OS.
3–4 Hypothesis-generating. Early-phase, feasibility, or AI tools needing prospective validation.
1–2 No shift. Failed primary endpoint or no survival benefit.
Continued on page 2 → 5–6 Practice-informing · 3–4 Hypothesis-generating · 1–2 No shift
WNTB = What You Need to Believe. Preliminary perspective on abstracts released 24 October 2026. Not medical advice.
[Slide 2]
ESMO 2026 · SATURDAY 24 OCTOBER · LATE-BREAKING ABSTRACTS · PAGE 2 OF 2
HOT OFF THE PRESS
Late-Breaking Abstracts: Perspective
Scores 1–6: practice-informing, hypothesis-generating and no-shift readouts
PRELIMINARY Pending full data
5–6 Practice-informing · Subgroup-specific — Narrow subsets, secondary endpoints, or PFS without OS · 9 LBAs
LBA15 · BREAST 6/10 KN026/HB1801 — Anbenitamab + docetaxel · HER2+ ABC, 1L — PFS HR 0.46 — Bispecific beat THP; ORR 87.6%; p < 0.0001. PFS only, OS not yet shown. WNTB: A bispecific doublet displaces the CLEOPATRA dual-mAb standard beyond regional registration.
LBA16 · BREAST 6/10 VIKTORIA-1 — Gedatolisib + palbociclib + fulvestrant · PIK3CA-mut HR+/HER2– ABC — PFS HR 0.52 visceral mets — Superior to alpelisib + fulvestrant, without the hyperglycemia signal. OS not yet reported. WNTB: The triplet shows OS benefit without cumulative stomatitis or diarrhea.
LBA48 · HCC 6/10 EMERALD-3 — STRIDE ± lenvatinib + TACE · Embolisation-eligible — mOS 44.6 vs 39.2 ALBI 1 — Liver function preserved; OS by ALBI grade is exploratory, not yet formally tested. WNTB: Upfront dual IO + TACE does not accelerate hepatic decompensation.
LBA96 · BREAST 6/10 TROPION-Breast02 — Dato-DXd · 1L mTNBC, brain metastases cohort — PFS HR 0.37 · OS HR 0.55 — mPFS 7.7 vs 3.5 mo; mOS 13.5 vs 10.7 mo. Subgroup of the main trial. WNTB: CNS penetration is sufficient to defer upfront cranial irradiation.
LBA20 · BREAST 6/10 Digital MammaPrint — AI digital pathology on H&E · Early HR+/HER2– — r = 0.954 vs RNA assay — Independent 10-year prognostic validation, HR 0.12. WNTB: Regulators accept image-based AI as a substitute for genomic RNA testing.
LBA36 · ENDOMETRIAL 6/10 KEYNOTE-B21 ctDNA — Adjuvant pembrolizumab · Resected high-risk dMMR — DFS HR 0.34 · DDFS 0.25 — ctDNA clearance strongly tracks outcomes. WNTB: Post-op ctDNA is standardized across labs to steer adjuvant IO.
LBA74 · LUNG 6/10 DESTINY-Lung04 PROs — T-DXd vs chemo-IO · 1L HER2-mutant NSCLC — Cough HR 0.66 — Delayed cough deterioration; physical functioning maintained. WNTB: The QoL edge cements T-DXd as preferred 1L over platinum chemo-IO.
LBA17 · BREAST 5/10 PREDICT-HER — HER2-PET imaging · T-DXd-treated ABC — NPV 88.2% — 100% NPV in HER2-low/zero; flags T-DXd non-responders. WNTB: PET tracer and nuclear medicine capacity can integrate into T-DXd pathways.
LBA73 · LUNG 5/10 PACE-LUNG — ctDNA-guided chemo escalation · EGFRm, ctDNA+ at wk 3 of osimertinib — mPFS 16.3 mo · OS 29.2 — Feasibility of 4-cycle chemo escalation in non-clearers. WNTB: Early ctDNA-triggered escalation can replace upfront FLAURA2 triplet.
3–4 Hypothesis-generating · Early validation — Early-phase signals and digital/AI tools awaiting outcome data · 6 LBAs
LBA41 · OVARIAN 4/10 BEHOLD-1 — Mocertatug rezetecan (B7-H4 ADC) · Platinum-resistant — ORR 62% — DoR 12.2 mo; mPFS 11.1 mo. Early-phase, single-arm. WNTB: Activity replicates in Phase 3, independent of FRα status.
LBA40 · OVARIAN 4/10 ZW191-101 — ZW191 (FRα ADC) · Platinum-resistant, non-clear cell — ORR 48% · DCR 96% — Phase 1, heavily pretreated disease, across dose levels. WNTB: Activity holds in Phase 3 at lower FRα than mirvetuximab requires.
LBA26 · BREAST 4/10 UNLOCK 14 (POP-DURVA) — Preoperative durvalumab · Stage I TNBC, TILs ≥ 5% — Downstaging 41.3% — Immune remodeling seen, but monotherapy pCR only 1.4%. WNTB: Short IO remodeling supports later chemo de-escalation safely.
LBA52 · PAN-TUMOR 4/10 iNATURE — PPVax neoantigen vaccine + CLERT · Advanced refractory solid tumors — PFS HR 0.29 — mPFS 10.3 vs 3.1 mo; p = 0.007. WNTB: Focal ablation + vaccines give systemic, abscopal-like control across tumors.
LBA18 · CRC · AI 3/10 BEAGLE-MDT — Multi-agent AI platform · Tumor board workflow — 73–100% concordance — Case summaries vs experts; trial matching 67–93%. WNTB: MDTs adopt LLM recommendations without prospective outcome endpoints.
LBA19 · SUPPORTIVE · AI 3/10 ALLIANCE — LLM-assisted side-effect management · Breast & CRC — 5.4 vs 2.8 toxicities logged — No change in dose modifications or quality of life. WNTB: Better symptom logging alone justifies adoption without linked interventions.
1–2 No shift · Practice-affirming negative — Missed primary endpoint or no survival benefit · 3 LBAs
LBA14 · BREAST 2/10 MA.40/FINER — Fulvestrant + ipatasertib · HR+/HER2– ABC post-CDK4/6i — OS HR 0.97 ITT — PFS benefit (HR 0.61, ASCO 2025) did not carry to OS; AKT-altered OS HR 1.14. Takeaway: No role for ipatasertib; capivasertib and oral SERDs remain preferred.
LBA51 · BREAST 2/10 KEYNOTE-B49 — Pembrolizumab + chemo · HR+/HER2– ABC, CPS ≥ 1 — OS HR 1.04 — PFS HR 0.74, but mOS 21.0 vs 22.2 mo; p = 0.6185. Takeaway: PFS gain without OS can't justify adding PD-1 given available ADCs.
LBA21 · BILIARY 2/10 ACTICCA-1 — Gem-cis vs capecitabine · Resected cholangio/gallbladder — DFS p = 0.097 missed — Primary log-rank not met; RMST shows a modest +2.6 mo DFS. Takeaway: Adjuvant capecitabine remains the standard; the RMST signal is hypothesis-generating.
WNTB = What You Need to Believe for the readout to change practice. Preliminary perspective based on late-breaking abstracts released 24 October 2026; scores will be revisited once full data are presented. Not medical advice. Scoring key on page 1.
ESMO 2026 · Page 2 of 2
Jose Fernando Moura, PhD @FernandoOncoPost text
Cancer care goes beyond medications. Our patients need more than treatments. They need compassion, connection, and care for the whole person. Science is reminding us why.
Because treating cancer matters. But caring for the person matters just as much. #ESMO2026 @myESMO https://t.co/7kfbMEf0SO
[Slide 1]
BREAKING NEWS
Spiritual Interventions in Cancer Care [Systematic Review]
Effects on biological and physiological markers in cancer patients
@FernandoOnco
ONCOLOGIA BASEADA EM EVIDÊNCIAS PARA UMA PRÁTICA MELHOR
82 studies included
Adults with cancer
Prayer, meditation, mindfulness
6760 records identified
Methods
JBI methodology, PRISMA and PICO framework
Databases: PubMed, MEDLINE, CINAHL, Scopus, Web of Science, Cochrane Library, SciELO
Search through January 2026
Outcomes assessed
Cortisol, cytokines, CRP
Immune cell counts
Cardiovascular measures
RESULTS Spiritual interventions were associated with favorable changes in multiple biological and physiological markers in patients with cancer.
Stress response [down arrow] Reduced cortisol levels
Inflammation [down arrow] Decreased pro-inflammatory cytokines: IL-1β, IL-6, IL-8, TNF-α
Immune function [up arrow] Enhanced NK cell activity and interferon-γ production
Autonomic regulation [up arrow] Higher heart rate variability and lower resting heart rate
Other signals Possible inhibition of inflammatory pathways (e.g., NF-κB), effects on oxidative stress and markers of cellular aging
KEY TAKEAWAY
Spiritual interventions were associated with favorable changes in multiple biomarkers and physiologic indicators, suggesting potential benefits through stress modulation, immune regulation, and attenuation of inflammatory responses.
CONCLUSIONS
• Evidence supports beneficial biologic and physiologic effects.
• Findings are promising for integrative oncology care.
• Most evidence was synthesized narratively.
Clinical question
Should spiritual care be more intentionally integrated into comprehensive cancer care?
What's your take?
@FernandoOnco | Discussing today. Improving tomorrow. #IntegrativeOncology #CancerCare #SpiritualCare
Dr. Robert F. Siegmund, MBA @robertfsiegmundPost text
.@myESMO conference has started today in Madrid.
#ESMO2026 Top Companies
Source: @Larvol CLIN database
#cancer #oncology #data https://t.co/iCT9kWfiuj
[Slide 1]
Top Companies by X Posts
Merck (MSD)
AstraZeneca
Novartis
BMS
Sichuan Biokin Pharma
Amgen
Roche
Pfizer
Sichuan Kelun Pharma
Sino Biopharm
X Posts (x-axis 0–350) [stacked sentiment bars; company logos]
FAQ
About this page
What is on the ESMO 2026 Heme & Other slides page?
The 96 Heme & Other slide photos, abstract screenshots and data graphics plus 15 KOL-made infographics that oncologists, societies, media and companies posted on X from the ESMO 2026 embargo lift (October 22) through the congress in Madrid (October 23–27), compiled by KOL Pulse. Each card links to the original post and keeps the author's full text. It is one tumor-type page of the ESMO 2026 slide library: https://kolpulse.com/kol-pulse-esmo26-slides
Which Heme & Other trials are covered?
Trials with their own slide groups, by number of slide posts: TANGENT (1). Slides that compare several trials sit in a Cross-trial group, and slides not tied to a tracked trial come last. Physician posts come first in every group.
How do I filter by trial or KOL?
Use the trial chips in each section, the Trial menu (which also finds cross-trial slides that name that trial) and the KOL menu or the Built by avatars. Filters combine; counts only offer combinations that have results, and Clear filters or the Escape key resets. Every state has a shareable link, for example #tag=tangent.
What are the OCR text and the source notes?
View OCR Text opens a verbatim transcription of the slide image; survival curves are captured as the printed numbers only, never read off the curve. Where it helps, a short context note gives the figures printed on the presented slide or abstract (for example, a different data cut or analysis). The author's caption is never edited.
Where are the other tumor types and the pre-congress archive?
Breast Cancer: https://kolpulse.com/kol-pulse-esmo26-slides-breast; Lung Cancer: https://kolpulse.com/kol-pulse-esmo26-slides-lung; GU Cancer: https://kolpulse.com/kol-pulse-esmo26-slides-gu; GI Cancer: https://kolpulse.com/kol-pulse-esmo26-slides-gi; Melanoma: https://kolpulse.com/kol-pulse-esmo26-slides-melanoma; Gyn Cancer: https://kolpulse.com/kol-pulse-esmo26-slides-gyn. Pre-congress previews, watchlists and top-10 lists: https://kolpulse.com/kol-pulse-esmo26-slides-archive
Can I use this page with my own AI assistant (NotebookLM, Perplexity, Claude, ChatGPT)?
Yes. Paste the page URL, or a filtered link, into your assistant and ask questions against it. Example prompt: "From the slide transcriptions on this page, list each trial's primary endpoint result with the hazard ratio and data cut exactly as printed on the slide, and the physician who posted it."