EVOKE-03 / KEYNOTE-D46 at a Glance
Design
Global, open-label, randomized phase III: sacituzumab govitecan 10 mg/kg (days 1, 8 q3w) plus pembrolizumab vs pembrolizumab monotherapy, first line in metastatic NSCLC with PD-L1 TPS ≥50% and no EGFR/ALK/ROS1 alterations; ~620 patients; dual primary endpoints of BICR PFS and OS. EVOKE-03 (Gilead's name) and KEYNOTE-D46 (Merck's) are the same study. (Merck/Gilead press release; Targeted Oncology)
Efficacy
Final PFS analysis: 11.8 vs 7.7 months, HR 0.81 (95% CI 0.66–1.00 — the upper bound exactly 1.00) — a numerical difference that did not meet the threshold for statistical significance; interim OS analysis HR 1.07 (95% CI 0.85–1.35), median OS 21.5 vs 22.8 months numerically FAVORING pembrolizumab alone, with futility indicated. (WCLC 2026 presented slides via verbatim physician captures, incl. Liu on this page)
Safety
Grade ≥3 treatment-related AEs 55.7% vs 16.5% — substantially more toxicity for the combination against no significant efficacy gain. (WCLC 2026 presentation via verbatim KOL capture)
Status
Discontinued June 8, 2026 on the external data monitoring committee's recommendation after the prespecified final PFS and interim OS analyses. Trodelvy remains approved in its existing indications; this first-line NSCLC combination is not moving forward. (Merck/Gilead press release)
KOL pulse
Amol Akhade: “Evoke 03 . Important negative study.” Tom John: “A negative study with another ADC+Pembro as first line treatment. More tox with ADC.”
KOLs Discussing EVOKE-03 / KEYNOTE-D46
Key Slides & Data
What Physicians Said
🌟 Truly honored that our work has been selected for presentation at the Presidential Symposium 1 of the World Lung Cancer Conference #WCLC26 in Seoul: ⏰ Sunday, 13 September 08:00-10:30 KST 🗒️ Presidential Symposium 1 📌 Plenary, Hall D2, 3F Primary analysis of the global, randomized phase III EVOKE-03/KEYNOTE D46 clinical trial on the combination of Sacituzumab Govitecan plus pembro vs pembro monotherapy in patients with advanced #NSCLC and high #PDL1 expression. Thank you @IASLC for selecting a negative (by press release) phase III trial for the Presidential Symposium, those data are important to be shared and communicated in detail, because they provide food for thought and can be hypothesis generating for future trials. Looking forward to the discussion that will follow, especially in he context of the Optitrop Lung 05 results. @IASLC @myESMO @lcsmchat @OncoAlert @OncBrothers @OncLive @PeerView @oncodaily @VJOncology @MedscapeOnc @LungCancerEu @YoungLungCancer @lcrf_org @LUNGevity @HeSMO_X @fairlifelcc @HenryDunantGR #IASLC #ThoracicMalignancy #WCLC26 https://t.co/pvnFKqNzXc
Dr. @g_mountzios at #WCLC26 presents results of EVOKE-03 / KEYNOTE D46: phase III study of sacituzumab govitecan (Trop2 ADC) with pembro vs pembro alone in NSCLC with PD-L1 ≥50%. Median f/u 14.7m. https://t.co/U1jhvkkDXB
EVOKE-03/KEYNOTE D46; Sac-govitecan trop2ADC; P3RCT; high PDL1 (>50%) randomised to Pembro vs SG+ Pembro; n620; 4 month longer mPFS but HR not stat significant ; OS flat ; Interesting as suggests chemoIO might not beat IO alone in high PDL1 within a prospective RCT #WCLC26 https://t.co/v6BUMaBEWK
EVOKE-03: More responses — but no survival gain? Phase III | 620 patients with untreated metastatic NSCLC, PD-L1 ≥50% SG + pembrolizumab vs pembrolizumab alone ORR: 56.6% vs 43.7% → more tumors responded PFS: 11.8 vs 7.7 months | HR 0.81 → numerical improvement, not statistically significant per prespecified boundary OS: 21.5 vs 22.8 months | HR 1.07 → no survival benefit Grade ≥3 TRAEs: 55.7% vs 16.5% → substantially more toxicity Non-squamous and East Asian subgroups showed interesting signals — hypothesis-generating, not practice-changing Take-home: More responses ≠ longer survival. Pembrolizumab remains a standard option in PD-L1-high, driver-negative NSCLC. Reference: Mountzios G et al. EVOKE-03/KEYNOTE-D46, PL02.06 | WCLC 2026 #MVOnco #WCLC2026 #EVOKE03 #NSCLC #LungCancer #Immunotherapy #TROP2
EVOKE-03/KEYNOTE-D46 at #WCLC26 In 1L metastatic NSCLC with PD-L1 TPS ≥50%, adding sacituzumab govitecan to pembrolizumab did not meet the primary PFS endpoint despite numerically longer PFS (11.8 vs 7.7 mo; HR 0.81). No OS benefit was observed at the interim analysis (HR 1.07). @OncoAlert @ManuelDomine @weoncologists @OpenMedKate @g_mountzios
🫁 EVOKE-03: adding a Trop-2 ADC to pembrolizumab did not improve outcomes in PD-L1-high NSCLC. PFS: 11.8 vs 7.7 months, HR 0.81 OS: 21.5 vs 22.8 months ORR: 56.6% vs 43.7% Grade ≥3 TRAEs: 55.7% vs 16.5%. A negative phase III study despite higher response rates with the combination. #CánCare #NSCLC #LungCancer #ThoracicOncology #ADC #Immunotherapy #WCLC26 @IASLC
As would be expected, much greater toxicity with sacituzumab govitecan + pembro arm vs pembro alone. With no PFS/OS signal and worse toxicity, this strategy fails in an unselected population. Other ADC trials in different PD-L1 subsets and exploring biomarkers pending. #WCLC26 https://t.co/jpwoUE29EB
Evoke 03 Sacituzumab Gov+ Pembro no benefit in OS. In 2026 it's time to integrate other biomarkers different than high PD-L1 expression in clinical trials to test escalation strategies. PFS data in PD-L1 >90% curve will overlap? #WCLC2026 https://t.co/JecL0Y8IUy
EVOKE-03 Ph3, Saci-Gov+Pembro v Pembro 1L NSCLC PD-L1 >50% SG+Pembro: ❌🔼mPFS 11 v 7m, HR 0.81 (not sig) ❌No OS benefit ❓benefit in Asian & non-sq popn ❌🔼 Gr3+ TRAE 73 v 45% 🤔 Early Ph2 activity lost in Ph3 Post-hoc Asia data = limited value End of SG in NSCLC IMO #WCLC26 https://t.co/MW60DL1p0m
EVOKE-03/KND46 negative readout for a Sac-GT. Take: Dual primary endpoint, neither met. PFS numerically favored SG + pembro but wasn't statistically significant. Interim OS: 21.5 vs 22.8 mo, HR 1.07. ORR (55.6% vs 43.7%) and DCR (83.3% vs 73.1%) both favored the combo, but no PFS/OS. I think SG is a first-generation ADC with inferior efficacy and higher toxicity. Response rate gains without OS or durable PFS benefit in settings where pembrolizumab is good and safe as discussed by @Dr_R_Kurzrock @oncology_bg Interesting split by geography in Asia… #NSCLC #LCSM #WCLC26
EVOKE-03 did not meet the primary endpoint. PFS 11.8 vs 7.7m, HR 0.81 but did not meet threshold for statistical significance. Consistent across subgroups. No difference in OS: 21.5 vs 22.8m, HR 1.07. #WCLC26 https://t.co/j3v6jr8N7H
@g_mountzios presenting EVOKE-03. A negative study with another ADC+Pembro as first line treatment. More tox with ADC. The most interesting slide for me is the result based on region. Asian patients do better. Will this be the same for other studies too? #WCLC26 #LCSM https://t.co/wNtaXlRTBB
🆙#WCLC26 #LCSM Plenary Session 🔥EVOKE-03/KEYNOTE D46: Primary Results from Phase 3: Sacituzumab Govitecan + Pembrolizumab in PD-L1 TPS ≥50% Metastatic NSCLC 🎙️ @g_mountzios 🎯PFS HR 0.81 (95% CI 0.66-1.00) 🎯OS HR 1.07 (95% 0.85-1.35) 🔢PL02.06 ☑️NCT05609968 🔗 https://t.co/JvkpEBLlSo @OncoAlert @Larvol @IASLC
Evoke 03 . Important negative study. SG plus Pembrolizumab in 1st line NSCLC with PDL1 above 50 % @IASLC @StephenVLiu @DrRiyazShah @RManochakian @FordePatrick @PatelOncology https://t.co/PDuPiGhxX3
EVOKE-03: SG + Pembro did not improve clinical outcomes over Pembro alone but with an increased rate of known side effects in patients with NSCLC and high PD-L1 expression (>50%) #WCLC26 #LCSM @IASLC https://t.co/zDPQEJIo1p
Dr. Giannis Mountzios @g_mountzios presented at @IASLC #WCLC26 the EVOKE-03 Phase 3 study investigating sacituzumab govitecan (SG) + pembrolizumab vs pembrolizumab alone in PD-L1 >50% NSCLC. Unfortunately, while there was an improvement in PFS, ORR, and DCR, no significant improvement in OS was observed. Unclear what the rates of oncogenic drivers and co-alterations here were amongst the patients included in this study. Difficult to improve outcomes with adjunct therapies in PD-L1 high NSCLC. More research is needed for additional therapeutic targets in this patient population to improve outcomes and prevent recurrence/resistance. @LUNGevity @LauraAlderMD @lungoncdoc @BrunaPellini @RManochakian @StephenVLiu @FordePatrick @FSkoulidis
About the EVOKE-03 / KEYNOTE-D46 Trial
EVOKE-03 / KEYNOTE-D46 asked whether adding a TROP2-directed ADC to pembrolizumab could beat pembrolizumab alone in the population where checkpoint monotherapy is already most active — PD-L1 TPS ≥50% first-line metastatic NSCLC. The answer was no: a numerical PFS edge that missed significance, an OS hazard ratio above 1 at the interim with futility indicated, and nearly 3.4× the severe treatment-related toxicity.
Physicians at WCLC 2026 read it as a class lesson rather than a molecule failure alone: several noted the mounting difficulty of improving on pembrolizumab monotherapy in PD-L1-high disease with ADC add-ons, with subgroup signals (non-squamous, East Asian patients) called hypothesis-generating at most.
Trial Methodology & Results
Study Design
Global, open-label, randomized 1:1 phase III; sacituzumab govitecan + pembrolizumab vs pembrolizumab; dual primary endpoints BICR PFS (RECIST v1.1) and OS. (Targeted Oncology; Merck/Gilead PR)
Population
Previously untreated metastatic NSCLC, PD-L1 TPS ≥50%, no sensitizing EGFR/ALK/ROS1 alterations; ~620 patients worldwide. (Merck/Gilead PR)
Outcome
Final PFS not significant (11.8 vs 7.7 mo, HR 0.81); interim OS HR 1.07 with futility; Grade ≥3 TRAEs 55.7% vs 16.5%. (WCLC 2026 presentation via verbatim KOL posts)
Status
Discontinued June 8, 2026 per eDMC recommendation; final data presented at WCLC 2026. (Merck/Gilead PR)
Final efficacy — endpoints not met
PFS was 11.8 versus 7.7 months, HR 0.81 with a 95% CI of 0.66–1.00 — the upper bound landing exactly on 1.00 — numerically longer with the combination but below the threshold for statistical significance under the trial's dual-primary design. The interim OS analysis showed HR 1.07 (95% CI 0.85–1.35) with median OS 21.5 versus 22.8 months — numerically favoring pembrolizumab alone — and futility for eventual OS superiority. (WCLC 2026 presented slides, via the verbatim physician captures on this page incl. Liu)
PFS HR 0.81 (NS) · OS HR 1.07Source: WCLC 2026 presentation (verbatim KOL captures) ↗
Toxicity — the deciding margin
Grade ≥3 treatment-related adverse events occurred in 55.7% with the combination versus 16.5% with pembrolizumab alone — the toxicity cost that, against non-significant efficacy, ended the program in this setting. (WCLC 2026 presented slides via verbatim physician capture on the KOL Pulse conference page)
Gr≥3 TRAE 55.7% vs 16.5%Source: Merck/Gilead June 8, 2026 update · WCLC 2026 data ↗
WCLC 2026: Live KOL Reaction
Physician posts about this trial captured live during the IASLC 2026 World Conference on Lung Cancer (Seoul, September 12–15). Quotes are verbatim; each card links to the original post. Last updated 2026-09-15.
🆙#WCLC26 #LCSM Plenary Session 🔥EVOKE-03/KEYNOTE D46: Primary Results from Phase 3: Sacituzumab Govitecan + Pembrolizumab in PD-L1 TPS ≥50% Metastatic NSCLC 🎙️ @g_mountzios 🔢PL02.06 ☑️NCT05609968 🔗 https://t.co/icbDk1Zjnh @OncoAlert @Larvol @IASLC https://t.co/pMAIqOHdqF https://t.co/4ShK6WS6fB
1/2 🚨 Several #WCLC26 Presidential Symposium trials already have topline press releases 👀 🙌 ARTEMIS-008 | Risvutatug Rezetecan • Phase 3 in relapsed SCLC • Met the primary endpoint of OS vs. topotecan • Statistically significant and clinically meaningful OS benefit! 🔗 https://t.co/WavE0BuSbb 🤦♂️ EVOKE-03 / KEYNOTE-D46 | Sacituzumab Govitecan + pembrolizumab • 1L PD-L1 TPS ≥50% metastatic NSCLC • PFS improvement did not reach statistical significance • Study was discontinued after an eDMC review 🔗 https://t.co/t0DcvrKJmz
We learn from every trial Be it positive or not . Excellent discussion on Evoke 03. @IASLC @GlopesMd @StephenVLiu @RManochakian @OncBrothers #wclc2026 https://t.co/dUnY6Qdtet
5/6 EVOKE-03 / KEYNOTE-D46 Sacituzumab govitecan + pembrolizumab vs pembrolizumab alone in untreated metastatic NSCLC with PD-L1 ≥50%. The early data looked promising. But phase III tells a different story: ❌ PFS numerically favored the combination but did not reach statistical significance, and the study was discontinued. At WCLC I want to understand why — and what the OS, subgroup and safety data teach us.
EVOKE-03: More response ≠ longer survival SG + pembrolizumab vs pembrolizumab in PD-L1 ≥50% metastatic NSCLC. More responses: ORR 56.6% vs 43.7%. PFS numerically longer: 11.8 vs 7.7 months | HR 0.81. But no OS benefit: 21.5 vs 22.8 months | HR 1.07. Much more Grade ≥3 toxicity: 55.7% vs 16.5%. Take-home: More tumors shrank, but patients did not live longer. Mountzios G et al. EVOKE-03/KEYNOTE-D46 | WCLC 2026 #MVOnco #WCLC2026 #EVOKE03 #NSCLC #LungCancer
The fact that Gilead Sciences’ #sacituzumabgovitecan (#SG)failed as a first-line treatment for non-small cell lung cancer was not new. Yet newly revealed global phase 3 data could reignite the question around translating Chinese data to the West. Evoke-03 trial https://t.co/Dg3KNyHfqc
📝EVOKE-03 ✓PD-L1高発現では👑PBR単剤にADC追加効果なし ↔️EV-302(mUC)👏 👀両試験でのNeutropenia発現率の違い(=宿主免疫への負の影響)がICIの効果に直結? 🤔類似試験(TroFuse-007, TROPION-Lung08)の行く末は? 🤔PD-L1低発現/陰性ではどうか? #WCLC26 https://t.co/dKyvWferwj
@Latinamd @IASLC @GlopesMd Honestly I'm skeptical on EVOKE-03 after how the earlier sacituzumab + pembro data read out. PD-L1 >=50% enrichment may not be enough. My money's on the B7-H3 ADCs instead.
🌟 Truly honored that our work has been selected for presentation at the Presidential Symposium 1 of the World Lung Cancer Conference #WCLC26 in Seoul: ⏰ Sunday, 13 September 08:00-10:30 KST 🗒️ Presidential Symposium 1 📌 Plenary, Hall D2, 3F Primary analysis of the global, randomized phase III EVOKE-03/KEYNOTE D46 clinical trial on the combination of Sacituzumab Govitecan plus pembro vs pembro monotherapy in patients with advanced #NSCLC and high #PDL1 expression. Thank you @IASLC for selecting a negative (by press release) phase III trial for the Presidential Symposium, those data are important to be shared and communicated in detail, because they provide food for thought and can be hypothesis generating for future trials. Looking forward to the discussion that will follow, especially in he context of the Optitrop Lung 05 results. #ThoracicMalignancy #WCLC26 #IASLC #ThoracicMalignancy #WCLC26 https://t.co/pvnFKqNzXc
🆙#WCLC26 #LCSM Plenary Session 🔥EVOKE-03/KEYNOTE D46: Primary Results from Phase 3: Sacituzumab Govitecan + Pembrolizumab in PD-L1 TPS ≥50% Metastatic NSCLC 🎙️ @g_mountzios 🔢PL02.06 ☑️NCT05609968 🔗 https://t.co/icbDk1Zjnh @OncoAlert @Larvol @IASLC https://t.co/4ShK6WS6fB
🌟 Truly honored that our work has been selected for presentation at the Presidential Symposium 1 of the World Lung Cancer Conference #WCLC26 in Seoul on Sunday, September 13: Primary analysis of the global, randomized phase III EVOKE-03/KEYNOTE D46 clinical trial on the combination of Sacituzumab Govitecan plus pembro vs pembro monotherapy in patients with advanced #NSCLC and high #PDL1 expression. Thank you @IASLC for selecting a negative (by press release) phase III trial for the Presidential Symposium, those data are important to be shared and communicated in detail, because they provide food for thought and can be hypothesis generating for future trials. Looking forward to the discussion that will follow, especially in he context of the Optitrop Lung 05 results. #ThoracicMalignancy #WCLC26 ⏰ Sunday, 13 September 08:00-10:30 KST 🗒️ Presidential Symposium 1 📌 Plenary, Hall D2, 3F #IASLC #ThoracicMalignancy #WCLC26 https://t.co/pvnFKqNzXc
🌟 Truly honored that our work has been selected for presentation at the Presidential Symposium 1 of the World Lung Cancer Conference #WCLC26 in Seoul: Primary analysis of the global, randomized phase III EVOKE-03/KEYNOTE D46 clinical trial on the combination of Sacituzumab Govitecan plus pembro vs pembro monotherapy in patients with advanced #NSCLC and high #PDL1 expression. ⏰ Sunday, 13 September 08:00-10:30 KST 🗒️ Presidential Symposium 1 📌 Plenary, Hall D2, 3F Thank you @IASLC for selecting a negative (by press release) phase III trial for the Presidential Symposium, those data are important to be shared and communicated in detail, because they provide food for thought and can be hypothesis generating for future trials. Looking forward to the discussion that will follow, especially in he context of the Optitrop Lung 05 results. #ThoracicMalignancy #WCLC26 #IASLC #ThoracicMalignancy #WCLC26 https://t.co/pvnFKqNzXc
EVOKE-03 / KEYNOTE-D46 FAQ
Are EVOKE-03 and KEYNOTE-D46 the same trial?
Yes — one phase III study with two names: EVOKE-03 in Gilead's nomenclature and KEYNOTE-D46 in Merck's (NCT05609968). Merck sponsored and operationalized the trial; Gilead's Trodelvy (sacituzumab govitecan) was the investigational addition. Some conference trackers list the two names as separate trials; they are not.
What did the trial test?
Whether adding sacituzumab govitecan, a TROP2-directed antibody-drug conjugate, to pembrolizumab improves outcomes versus pembrolizumab alone as first-line treatment for metastatic NSCLC with PD-L1 TPS ≥50% and no EGFR/ALK/ROS1 alterations, in about 620 patients with dual primary endpoints of PFS and OS.
What were the results?
Negative on both primaries: final PFS 11.8 versus 7.7 months (HR 0.81, 95% CI 0.66–1.00) missed statistical significance, the interim OS analysis showed HR 1.07 (95% CI 0.85–1.35) with median OS 21.5 versus 22.8 months and futility, and Grade ≥3 treatment-related adverse events were 55.7% versus 16.5%. The trial was discontinued in June 2026 on the data monitoring committee's recommendation, and the full data were presented at WCLC 2026.
Does this affect Trodelvy's approvals?
No. Sacituzumab govitecan remains approved in its existing indications (including metastatic triple-negative and HR+/HER2- breast cancer); it was never approved for first-line NSCLC, and this combination is not advancing in that setting.
What do KOLs take from it?
That beating pembrolizumab monotherapy in PD-L1-high first-line NSCLC remains extraordinarily hard: the toxicity of adding an ADC payload was not paid back in efficacy, echoing other negative ADC-plus-IO first-line attempts discussed at the meeting. Subgroup signals were described as hypothesis-generating only.
Key KOL Sentiments
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
Compiled and reviewed by the KOL Pulse research team, led by Brian Shields, Founder, KOL Pulse. Last updated September 13, 2026. Every statistic carries its source label in place; every quote is verbatim from the physician's own post.