Physician discussion, October 23–26

Physician posts on X, October 23–26.

HARMONi-GI1 and CLARITY-Gastric01, both presented on October 25, moved to the top of the table on Days 3–4. INCB161734 (zecdarasib), an investigational KRAS G12D inhibitor, plus chemotherapy in KRAS G12D-mutant pancreatic cancer entered the table on October 26 (posts). Also discussed on October 26: LEAP-012 (posts) and HRS-4642 (posts).

Days 3–4: GI trials presented on October 25

All four were presented on October 25; physicians kept discussing them on October 26.

1. HARMONi-GI1: ivonescimab + gemcitabine-cisplatin vs durvalumab + gemcitabine-cisplatin in first-line biliary tract cancer

Double-blind Phase 3 (LBA8), N=682, enrolled in China: ivonescimab, a PD-1×VEGF bispecific antibody, plus gemcitabine-cisplatin vs durvalumab plus gemcitabine-cisplatin in untreated advanced biliary tract cancer, presented by Jian Zhou on October 25 in Presidential Symposium II, with James J. Harding as invited discussant. The primary endpoint, OS, was met at a prespecified interim analysis: median 14.9 vs 12.1 months, stratified HR 0.75 (95% CI 0.61–0.92), one-sided P=0.0031; 12-month OS 59.7% vs 50.7%. Key secondary endpoints, investigator-assessed: median PFS 8.4 vs 5.8 months, HR 0.61 (0.51–0.72); ORR 53.8% vs 32.5%. Grade 3 or 4 treatment-emergent adverse events occurred in 82.0% vs 72.3% of patients, and subsequent anti-tumor therapy was given to 31.8% vs 45.9%. (ESMO 2026 LBA8 slides, Jian Zhou; prespecified interim analysis, median follow-up 14.3 months; data cutoff not shown on the slides) The comparator is the approved first-line standard from TOPAZ-1. Ivonescimab is investigational: it is not approved for biliary tract cancer in any country and is not approved for any indication in the US or Europe. In the US, the ECOG-ACRIN Phase 2/3 trial EA2251 will compare ivonescimab plus chemotherapy with pembrolizumab or durvalumab plus chemotherapy. (KOL Pulse HARMONi-GI1 page)

Physicians called the survival gain against an active immunotherapy comparator meaningful, and noted the added toxicity.

The debate was the population. Some physicians asked for a global or Western trial before changing practice; Yakup Ergün argued that trials should be judged on their data, not their country of origin. On October 26, Angela Lamarca summarized the invited discussion.

Physicians also looked at subsequent therapy, meaning treatment given after the trial treatment, which was more common in the durvalumab arm on the presented slide (Jian Zhou, LBA8).

2. CLARITY-Gastric01: sonesitatug vedotin in CLDN18.2-positive gastroesophageal cancer, second line and later

Phase 3 (LBA7): sonesitatug vedotin (Sone-Ve), a CLDN18.2-directed antibody-drug conjugate, vs investigator’s choice in previously treated CLDN18.2-positive gastroesophageal cancer, presented by Rui-Hua Xu on October 25 in Presidential Symposium II, with Salah-Eddin Al-Batran as invited discussant. The trial had two dual primary endpoints. OS in patients treated in the third line or later (n=355) was met: median 10.0 vs 5.0 months, HR 0.47 (95% CI 0.36–0.62). PFS by blinded independent central review in all randomized patients, second line or later (n=532), was not met: median 4.2 vs 3.8 months, HR 0.77 (0.62–0.96), P=0.053193 on the slide, which did not reach the threshold for significance. OS in all randomized patients, the key secondary endpoint, was met through alpha recycling: median 10.6 vs 6.4 months, HR 0.58 (0.47–0.73). In the second-line subgroup, a descriptive analysis, the OS HR was 0.88 (0.60–1.30). Confirmed ORR by central review was 23.3% vs 13.8% in patients with measurable disease, and grade ≥3 treatment-related adverse events occurred in 36.4% vs 44.8%. (ESMO 2026 LBA7 slides and abstract, Rui-Hua Xu; planned interim analysis, data cutoff May 31, 2026; median follow-up 8.6 months) Sonesitatug vedotin is investigational and not approved by the FDA or any other regulator. The only FDA-approved CLDN18.2-directed agent is zolbetuximab, for first-line use with chemotherapy, and only 17 randomized patients had received zolbetuximab before. (KOL Pulse CLARITY-Gastric01 page)

Physicians welcomed the OS result in later lines, including in tumors with lower CLDN18.2 expression.

The debate was second line, where the OS benefit was less clear, and the CLDN18.2 definition, which differs from the zolbetuximab trials.

3. IROCAS: adjuvant mFOLFIRINOX vs mFOLFOX6 in high-risk stage III colon cancer

Phase 3 (LBA29) in resected T4 and/or N2 colon cancer, presented on October 25 by Jaafar Bennouna with Jeffrey Meyerhardt as invited discussant, and published concurrently in Annals of Oncology. Adding irinotecan to adjuvant oxaliplatin-based chemotherapy did not improve 3-year DFS, the primary endpoint, and OS did not differ significantly. Diarrhea was more frequent with mFOLFIRINOX. All drugs in both regimens are approved; IROCAS tested whether intensifying the adjuvant regimen helps.

Physicians agreed the trial answers the adjuvant irinotecan question.

The debate was what to try next in high-risk and ctDNA-positive stage III disease. On October 26, Nicholas Hornstein questioned the irinotecan-based escalation arm of CIRCULATE-US and argued for newer drugs instead.

4. AZUR-1: dostarlimab alone in dMMR/MSI-H locally advanced rectal cancer

Global, single-arm Phase 2, N=154: dostarlimab for 9 cycles (6 months) in untreated stage II/III dMMR/MSI-H rectal cancer, presented by Andrea Cercek on October 25 (LBA87) with David Tougeron as invited discussant, and published concurrently in Annals of Oncology. At this interim analysis, the primary endpoint, cCR12 (clinical complete response sustained for 12 months) by independent central review, was 47% (54 of 116); 38 more patients in clinical complete response were still being followed toward 12 months, and the presented slides describe the central-review analysis as 75% mature. Investigator-assessed cCR12, a secondary endpoint, was 68% (98 of 144). In an exploratory composite of investigator-assessed cCR12 or pathological complete response, the rate was 76% (110 of 144). 71% of patients (110 of 154) were managed without further treatment after dostarlimab alone, and grade ≥3 immune-mediated adverse events occurred in 3%. (ESMO 2026 LBA87 slides and abstract, Andrea Cercek, and Annals of Oncology; interim analysis, data cutoff Apr 20, 2026; median follow-up 26.3 months) The central-review and investigator figures use different evaluable populations; the presented slide shows that investigators recorded complete responses earlier than central review. FDA granted accelerated approval on October 19, 2026: Jemperli (dostarlimab) for adults with stage II/III dMMR/MSI-H locally advanced rectal cancer, as determined by an FDA-authorized test. (FDA approval notice, Oct 19, 2026, via KOL Pulse AZUR-1 page)

Physicians welcomed organ preservation with dostarlimab alone; Nicholas Hornstein called it an obvious new standard of care.

The main caveat raised was the gap between investigator and central-review results, along with the single-arm design and the length of follow-up.

Days 1–2: the earlier GI readouts

GI trials presented on October 23–24.

5. SANO: active surveillance vs esophagectomy after clinical complete response

Phase 3 stepped-wedge, cluster-randomized trial, N=309, after neoadjuvant chemoradiotherapy (41.4 Gy). 5-year OS 50% with active surveillance vs 54% with surgery, HR 1.17 (95% CI 0.77–1.77), not significantly different; the confidence interval is wide and equivalence was not formally shown. DFS HR 1.39 (0.94–2.06). 5-year distant metastases 46% vs 29%, P<.01. 85 of 198 patients (43%) under surveillance were spared esophagectomy; the slide labels this figure hypothetical. (ESMO 2026 LBA78 abstract and slides, Sanjiv Gangaram-Panday; median follow-up 76 months) SANO tests a treatment strategy, not a drug.

Physicians split on whether the trial supports offering surveillance. Points raised: the distant-metastasis rate, that a non-significant OS difference does not establish non-inferiority, and the 41.4 Gy dose.

6. HERIZON-GEA-01: zanidatamab ± tislelizumab in first-line HER2+ gastroesophageal adenocarcinoma

Phase 3, N=914: zanidatamab + chemotherapy, with or without tislelizumab, vs trastuzumab + chemotherapy. At the second interim OS analysis, zanidatamab + chemotherapy improved OS vs trastuzumab + chemotherapy: median 22.9 vs 19.6 months, HR 0.76 (95% CI 0.62–0.93), P=0.0088; 36-month OS 31.5% vs 22.9%. The triplet vs trastuzumab + chemotherapy: median OS 27.0 vs 19.6 months, HR 0.65 (0.52–0.80), a descriptive analysis because significance was already met at the first interim; 36-month OS 39.8%. Triplet vs doublet: HR 0.81 (0.65–1.00), P=0.0502, not statistically significant. (ESMO 2026 LBA79 slides, Kohei Shitara; second interim analysis, data cutoff Jul 15, 2026; median follow-up 35.3 months) The safety profile was consistent with the first interim analysis. (LBA79 safety slide) FDA approved on August 25, 2026: Ziihera + Tevimbra + chemotherapy for HER2 IHC 3+ or IHC 2+/ISH+ disease, and Ziihera + chemotherapy for IHC 3+. (KOL Pulse HERIZON-GEA-01 page)

The doublet’s OS result was not debated. The debate was whether to add tislelizumab, given the P=0.0502 comparison and added toxicity.

Doublet or triplet?

Nicholas Hornstein @GIMedOnc

GI medical oncologist, Northwell · Oct 23

HERIZON-GEA-01 second OS interim at #ESMO26, and I'm a triplet believer. In January I wondered if tislelizumab was necessary. With 35 months of follow-up, my answer is yes. 914 patients, 1L HER2+ GEA, randomized 1:1:1 to zani + chemo ± tislelizumab vs trastuzumab + chemo. Median OS
• Zani + chemo + tisle: 27.0 mo
• Zani + chemo: 22.9 mo
• Tras + chemo: 19.6 mo 3-year OS: 39.8% vs 31.5% vs 22.9% The doublet now has its own OS win over trastuzumab (HR 0.76, 95% CI 0.62–0.93, p=0.0088). The triplet vs trastuzumab got better with time (HR 0.72 at IA1, now 0.65). Triplet vs doublet: HR 0.81 (95% CI 0.65–1.00), p=0.0502. Yes, it technically misses. But 4 more months of median OS is clinically meaningful in metastatic gastric cancer. Patients also stayed on the triplet longer (~47 vs ~30 weeks), with no new safety signal. Takeaway: zani + chemo + tisle is my 1L default for HER2+ GEA. Bring on the bispecifics 🔥 @OncoAlert @TheGutOncLab @OncoAlert

View post on X ↗

Yakup Ergün @dr_yakupergun

medical oncologist · Oct 23

#ESMO26 | HERIZON-GEA-01 – Updated OS Her2 positive gastric/ge cancer At 35.3 months of follow-up, zanidatamab + CT now shows a significant OS benefit over trastuzumab + CT: 22.9 vs 19.6 months (HR 0.76; p=0.0088). With the addition of tislelizumab, median OS increased to 27 months, but the additional benefit was statistically borderline (HR 0.81; p=0.0502). Considering toxicity, we need to carefully analyze which patients should receive the triplet regimen.

View post on X ↗

Nurullah İlhan,MD @drnurullahilhan

medical oncologist, Istanbul · Oct 25

HERIZON-GEA-01 raises a treatment-selection question in 1L HER2+ advanced gastroesophageal adenocarcinoma. Zanidatamab + chemo improved OS vs trastuzumab + chemo: 22.9 vs 19.6 months; HR 0.76; P=0.0088. The tislelizumab triplet also sustained its OS benefit vs trastuzumab + chemo. But the direct comparison matters: Adding tislelizumab to zanidatamab + chemo gave OS of 27.0 vs 22.9 months; HR 0.81; P=0.0502 — not statistically significant. This does not establish equivalence between doublet and triplet. Who gains enough from adding immunotherapy to justify the extra treatment—and can biomarkers help us identify them? @KoheiShitara @YJanjigianMD @myESMO @OncoAlert #HERIZONGEA01 #ESMO26 #GICSM

View post on X ↗

7. COMPANION-002: tovecimig + paclitaxel in previously treated biliary tract cancer

Phase 2/3: tovecimig, a DLL4×VEGF-A bispecific antibody, plus paclitaxel vs paclitaxel alone after prior therapy for advanced biliary tract cancer, presented by Nilofer Azad. The trial met its primary endpoint of response rate and improved PFS. OS did not differ, and the presented slide describes the OS analysis as confounded by crossover from the paclitaxel arm. Tovecimig is investigational.

An investigator on the trial highlighted response and PFS. Others pointed to the toxicity, the crossover, and first-line testing.

Response and PFS vs OS and toxicity

ilyas sahin, MD @ilyassahinMD

GI medical oncologist, Mass General; COMPANION-002 investigator · Oct 24

Excited to see our COMPANION-002 trial published in JCO and presented at #ESMO26! Treatment options remain limited for advanced biliary tract cancer after first-line therapy, especially without targetable mutations. Tovecimig + paclitaxel vs paclitaxel alone: • Tumor response (ORR): 18% vs 5.3% • Time without progression (median PFS): 4.7 vs 2.6 months (HR 0.44) Proud to have contributed to this multicenter trial!

View post on X ↗

Arndt Vogel @ArndtVogel

liver cancer and precision oncology, Princess Margaret / MHH · Oct 24

A randomized trial of paclitaxel ± tovecimig in previously treated patients with advanced biliary tract cancer #ESMO26 #ESMOAmbassadors 👉 ORR: 18 vs 5.3% 👉 mPFS: 4.7 vs 2.6 mo 👉 mOS: 8.9 vs 9.4 mo 🧐Active drug, but no OS benefit in 2nd line, could be explored in 1L @myESMO @ASCO

View post on X ↗

Mario Balsa @MarioBalsaMD

medical oncologist, Hospital Clínic Barcelona · Oct 24

Signal in previously treated biliary tract cancer at #ESMO26: Tovecimig + paclitaxel improves ORR (17% vs 5%) and PFS (HR 0.44)…but with substantial toxicity, and OS remains uninterpretable given 54% crossover 🤷🏻‍♂️ @OncoAlert 4Sight is already dissecting the abstracts worth watching! 👀 @OncoReporte @_SEOM @myESMO

View post on X ↗

Grainne O'Kane @graokane

medical oncologist, pancreatic and hepatobiliary cancers · Oct 24

Paclitaxel +\-tovecimig: COMPANION-002 ph2/3 ➡️prev treated #BTC wo actionable alterations ➡️ORR 17.1 v5.3%; DCR 62 v 37% ➡️mPFS 4.7 v 2.6mths - benefits seen at PFS2 noting crossover ➡️OS not sig but nb crossover @myESMO #ESMO26 @OncoAlert https://t.co/eeR74Ehqed

View post on X ↗

8. G-HOPE-001: arcotatug tavatecan in CLDN18.2-positive gastric/GEJ cancer, third line and later

Phase 3, China and Japan: arcotatug tavatecan (IBI343), a CLDN18.2-directed antibody-drug conjugate, vs treatment of physician’s choice after at least two prior lines, presented by Kohei Shitara. PFS and OS both improved; the OS result was reported at the second interim analysis. Physicians quoted figures from two analyses, the abstract’s first interim and the updated second interim in the presentation, so the numbers differ between posts. Arcotatug tavatecan is not FDA approved.

Physicians called it a new option after two lines. Open questions: OS maturity, patients previously treated with zolbetuximab, and the lack of a Western cohort.

9. ACTICCA-1: adjuvant gemcitabine-cisplatin vs capecitabine in resected biliary tract cancer

Phase 3, presented by Henning Wege. The primary DFS endpoint was not met. OS was numerically longer with gemcitabine-cisplatin but not statistically significant, and toxicity was higher.

Physicians agreed it does not change the standard. They differed on whether gemcitabine-cisplatin is an option for selected patients.

10. RASolute 302: daraxonrasib by RAS genotype in previously treated pancreatic cancer

Phase 3, N=500. Primary results: median OS 13.2 vs 6.7 months with daraxonrasib vs chemotherapy, HR 0.40 (95% CI 0.30–0.53), overall population. (ASCO 2026 LBA5 / NEJM; data cutoff Feb 10, 2026) At ESMO 2026, Zev A. Wainberg presented subgroup analyses from the same data cutoff. OS HR by RAS genotype: G12D 0.49 (0.33–0.74), G12V 0.32 (0.19–0.54), G12R 0.61 (0.29–1.30), Q61 0.37 (0.09–1.44), no RAS mutation identified 0.43 (0.09–1.98). The G12R, Q61 and no-mutation groups were small (30 vs 31, 11 vs 11 and 9 vs 7 patients) and their confidence intervals cross 1. ORR by BICR in G12R was 7% vs 14%. (ESMO 2026 LBA82 slides; subgroup HRs from an unstratified Cox model; median follow-up 8.5 months) FDA approved on August 26, 2026 as Rasonque for metastatic pancreatic adenocarcinoma after at least one prior systemic therapy, or in patients who are not candidates for multiagent therapy. The label does not restrict use by RAS status. (KOL Pulse RASolute 302 page)

Physicians agreed on the overall benefit. The question was RAS G12R.

11. KEYNOTE-975: pembrolizumab with definitive chemoradiotherapy in esophageal cancer

Phase 3: pembrolizumab or placebo with definitive chemoradiotherapy in unresectable esophageal or GEJ cancer, presented by Manish Shah. EFS was not improved in PD-L1 CPS ≥10 tumors, and OS was not formally tested. Pembrolizumab is not approved with definitive chemoradiotherapy for esophageal cancer.

Physicians agreed the trial was negative. Discussion turned to radiation dose, timing of immunotherapy relative to radiation, and toxicity.

12. PRABITAS: biweekly vs conventional trifluridine-tipiracil + bevacizumab in refractory colorectal cancer

Pragmatic, non-inferiority Phase 3 trial from Japan, presented by Hiroya Taniguchi and published in The Lancet. The biweekly (14-day) schedule was compared with the conventional 28-day schedule. OS met non-inferiority, with less grade ≥3 neutropenia on the biweekly schedule. Trifluridine-tipiracil plus bevacizumab is FDA approved for previously treated metastatic colorectal cancer; PRABITAS tested the dosing schedule, not a new drug.

Physicians agreed on the schedule. The debate was why it took a randomized trial, after earlier real-world and Phase 2 data.

A new standard schedule, or overdue?

Nobuhiro Shibata MD, PhD @shibanob

medical oncologist, Kansai Medical University; PRABITAS co-author · Oct 24

#ESMO26 LBA31 PRABITAS Phase 3 Trial A pragmatic phase 3 non-inferiority trial comparing biweekly vs conventional FTD/TPI + bevacizumab in refractory mCRC. Efficacy:
• OS: 11.3 vs 10.3 mo (HR 0.93, 95% CI 0.80–1.08)
• PFS: 3.9 vs 3.9 mo (HR 0.93, 95% CI 0.81–1.06) Safety:
• Grade ≥3 neutropenia: 13% vs 39%
• Febrile neutropenia: 1% vs 5%
 Biweekly FTD/TPI + bevacizumab demonstrated non-inferior OS with reduced toxicity compared with the conventional schedule. Our institution, Kansai Medical University Hospital, participated in this study. I am pleased to have contributed as a co-author. Simultaneously published in The Lancet! https://t.co/O0yd2m1dqc #PRABITAS #mCRC #ColorectalCancer
@OncoAlert

View post on X ↗

Sharlene Gill, MD, MPH, MBA, FASCO @GillSharlene

GI medical oncologist, University of British Columbia · Oct 24

Shout out to the PRABITAS study team 🇯🇵 for providing Level 1 evidence to confirm earlier RWE (with a pragmatic design and no industry funding!) ➡️ biweekly TAS102 should be the new standard for SUNLIGHT ➡️Simultaneous publication in @TheLancet https://t.co/o6Vor5Ms9a ➡️Great discussion by @KaiKeenShiu #ESMO26 @myESMO

View post on X ↗

Nicholas Hornstein @GIMedOnc

GI medical oncologist, Northwell · Oct 25

PRABITAS: PIII biweekly vs conventional TAS-102 + bev in refractory mCRC 🇯🇵, presented at #ESMO26. Yay for confirmation. But everyone should have been doing this since the OG publications (and certainly after @CathyEngMD published). The SUNLIGHT schedule (TAS-102 days 1–5 and 8–12, bev days 1 and 15, every 28 days) works, but the neutropenia is savage at almost 50% G3-4 on SUNLIGHT. PRABITAS has the same dose but spreads it out a littttttle more: TAS-102 days 1–5 + 15-19 and bev every 14 days. 970 patients, dosing at physician discretion. The primary endpoint was OS with an NI margin of 1.25. OS 11.3 vs 10.3 mo, HR 0.93 (95% CI 0.80–1.08). Non-inferior, and the curves are the same. Massive tox reduction, consistent with prior data. This shouldn't surprise anyone. Cathy Eng's group published this schedule in 2025: 16% grade ≥3 neutropenia, no febrile neutropenia, no G-CSF. The Japanese phase 2 data go back to 2020. Do we really always need randomized data to change practice? Do patients deserve to wait for 5+ years? Asking for a friend. @OncoAlert @TheGutOncLab @Onco_Nexus

View post on X ↗

Dr. Cathy Eng @CathyEngMD

GI medical oncologist, Vanderbilt-Ingram · Oct 24

PRABITAS phase III trial pragmatic trial enrolled 972 pts in 17M evaluating the schedule of TAS-102 + bev (2 weeks on/2 weeks off) vs. (1 week on/1 week off) which validates non-inferiority and less dose reductions validating our RWD approach in the US. #colorectal #cancer #ESMO2026 @OncoAlert

View post on X ↗

Also new: telisotuzumab adizutecan

Telisotuzumab adizutecan + bevacizumab in refractory metastatic colorectal cancer

Long-term follow-up of a randomized Phase 1 comparison (791RO): telisotuzumab adizutecan (Temab-A, ABBV-400), a c-Met-directed antibody-drug conjugate, plus bevacizumab vs trifluridine-tipiracil plus bevacizumab in third-line or later disease. Michael Cecchini presented it on October 24 in the lower GI rapid oral session; Lisa Salvatore discussed it. The paper was published concurrently in Annals of Oncology. Efficacy favored Temab-A plus bevacizumab, with hematologic toxicity the main concern. A Phase 3 trial is ongoing, and Temab-A is investigational.

KOL Voices: GI cancer leaderboard

The top physicians on the ESMO 2026 KOL leaderboard, GI tab. One line from each, verbatim, with the trial it is about.

1. 1
   Nicholas Hornstein GI medical oncologist, Northwell · 51.6K impressions, 22 posts

   on PRABITAS → “Do we really always need randomized data to change practice?”
2. 2
   MV Chandrakanth medical oncologist, Narayana Health Kolkata · 48.7K impressions, 63 posts

   on SANO → “Surgery when needed — NOT surgery never!”
3. 3
   Nieves Martinez Lago MD PhD GI medical oncologist · 44.1K impressions, 38 posts

   on PRABITAS → “💡 Same survival, less toxicity!”
4. 4
   Jeff Ryckman radiation oncologist · 34.2K impressions, 32 posts

   on KEYNOTE-975 → “Another reason not to assume IO helps every CRT backbone.”
5. 5
   Arndt Vogel liver cancer and precision oncology, Princess Margaret / MHH · 31.8K impressions, 28 posts

   on COMPANION-002 → “Active drug, but no OS benefit in 2nd line, could be explored in 1L”
6. 6
   Anirban Maitra GI pathologist and pancreatic cancer researcher · 31.2K impressions, 5 posts

   on Daraxonrasib + vopimetostat (Tango) study “Reduced toxicity data is particularly important as a lower Daraxonrasib dose can be used.”
7. 7
   Angela Lamarca HPB and NET medical oncologist · 22.0K impressions, 42 posts

   on COMPANION-002 → “👉Development continues”
8. 8
   Daisuke Kotani, MD, Ph.D 小谷 大輔 GI medical oncologist, National Cancer Center Hospital East · 21.7K impressions, 16 posts

   on SANO → “Despite no significant OS or DFS differences, the wide CIs and significantly higher distant mets rate highlight important challenges for implementing active surveillance in clinical practice.”
9. 9
   Shankar Siva radiation oncologist, Peter MacCallum Cancer Centre · 20.2K impressions, 9 posts

   on MASTERPLAN → “Better efficacy without added harm.”
10. 10
    Erman Akkus medical oncologist · 17.4K impressions, 7 posts

    on HARMONi-GI1 → “mOS: 14.9 vs. 12.1 months; HR 0.75, 95% CI 0.61-0.92, P = 0.0031”

What’s yet to come: the last GI session, October 27

Times are Madrid time. ESMO 2026 ends on October 27.

FAQ

Which GI cancer trials did physicians discuss most at ESMO 2026?

From October 23 to 26, HARMONi-GI1 drew the most physician posts on X among GI trials, followed by CLARITY-Gastric01, HERIZON-GEA-01, SANO and G-HOPE-001.

What did HARMONi-GI1 show at ESMO 2026?

At a prespecified interim analysis (LBA8, median follow-up 14.3 months), ivonescimab plus gemcitabine-cisplatin improved OS, the primary endpoint, vs durvalumab plus gemcitabine-cisplatin in first-line advanced biliary tract cancer: 14.9 vs 12.1 months, stratified HR 0.75 (95% CI 0.61-0.92), one-sided P=0.0031. Grade 3 or 4 adverse events were 82.0% vs 72.3%. The trial enrolled patients in China only. Ivonescimab is investigational and not approved for biliary tract cancer in any country.

Did CLARITY-Gastric01 meet its primary endpoints?

One of its two dual primary endpoints. OS in patients treated in the third line or later was met: 10.0 vs 5.0 months, HR 0.47 (95% CI 0.36-0.62). PFS by central review in all randomized patients was not met: HR 0.77 (0.62-0.96), P=0.053193. OS in all randomized patients, the key secondary endpoint, was met: HR 0.58 (0.47-0.73). These are planned interim results (LBA7, data cutoff May 31, 2026). Sonesitatug vedotin is investigational.

What did HERIZON-GEA-01 show at ESMO 2026?

At the second interim analysis (ESMO 2026 LBA79, data cutoff July 15, 2026), zanidatamab plus chemotherapy improved OS vs trastuzumab plus chemotherapy: 22.9 vs 19.6 months, HR 0.76 (95% CI 0.62-0.93), P=0.0088. Adding tislelizumab vs zanidatamab plus chemotherapy: HR 0.81, P=0.0502, not statistically significant. Both regimens have been FDA approved since August 25, 2026, with different HER2 criteria.

Does SANO show active surveillance is as good as surgery?

SANO found no statistically significant difference in 5-year OS (50% vs 54%, HR 1.17, 95% CI 0.77-1.77), but distant metastases were higher with surveillance (46% vs 29%). Equivalence was not formally shown, and physicians on X disagreed on whether to offer surveillance.

What was new for RASolute 302 at ESMO 2026?

Subgroup analyses (LBA82) showed OS hazard ratios below 1 across RAS genotypes, with small subgroups such as G12R (HR 0.61, 95% CI 0.29-1.30) having wide confidence intervals. Daraxonrasib (Rasonque) has been FDA approved since August 26, 2026 without a RAS-status restriction.

What did AZUR-1 show at ESMO 2026?

At the interim analysis presented as LBA87 (data cutoff April 20, 2026; median follow-up 26.3 months), cCR12 by independent central review, the primary endpoint, was 47% (54/116), with 38 more patients in clinical complete response still being followed. Investigator-assessed cCR12 was 68% (98/144), and 71% of patients (110/154) were managed without further treatment after dostarlimab alone. Dostarlimab (Jemperli) received FDA accelerated approval for stage II/III dMMR/MSI-H locally advanced rectal cancer on October 19, 2026.

Which GI sessions are still ahead at ESMO 2026?

On October 27, the last day of the congress, Scientific Congress Highlights 1 includes highlights for lower GI cancers (Erika B. Ruiz-Garcia) and upper GI cancers (Florian Lordick).

Related ESMO 2026 pages

• ESMO 2026 GI slide library
• ESMO 2026 trials: physician posts by trial
• ESMO 2026 conference page: daily highlights and KOL leaderboard
• HERIZON-GEA-01 trial page
• SANO trial page
• RASolute 302 trial page
• AZUR-1 trial page
• HARMONi-GI1 trial page
• CLARITY-Gastric01 trial page