ADCs Returned 0.46 in Relapsed Disease and 1.15 in the Front Line at WCLC
Nine presentations across small cell and non-small cell lung cancer, and the antibody-drug conjugates left the meeting in two different directions. In relapsed small cell disease, two separate Phase 3 trials returned an overall survival hazard ratio of 0.46. In first-line non-small cell disease, two ADC trials produced response and progression benefits that did not reach survival at all.
B7-H3 went from unvalidated to class-defining in a single session
Two Phase 3 readouts in relapsed small cell lung cancer landed at the same meeting, both against the same target, both returning an identical overall survival hazard ratio.
| Trial | Population | Key results |
|---|---|---|
| ARTEMIS-008 risvutatug rezetecan | Relapsed SCLC after platinum | OS 18.5 vs 10.3 months, HR 0.46; PFS 7.2 vs 3.0 months; ORR 58.3% vs 12.6% |
| TAISHAN-302 tambotatug pelitecan | Relapsed SCLC after one prior platinum regimen | OS 13.3 vs 9.4 months, HR 0.46; PFS 7.4 vs 2.8 months, HR 0.29; ORR 59.1% vs 9.7%; Grade 3 or higher TRAEs 46.4% vs 74.7% |
| BNT324-01 pumitamig plus elfetabart drozutecan | Advanced SCLC, PD-L1 by VEGF-A bispecific with B7-H3 ADC | ORR 70.4%; DCR 93.0%; ORR 92.3% in first line, 77.3% in second, 52.4% in third or later; Grade 3 or higher TRAEs 23.3% |
| MAVERICK SWOG S1827 | SCLC without brain metastases after initial therapy | MRI surveillance reduced cognitive failure or death, HR 0.60; no apparent OS disadvantage |
Data as presented at WCLC 2026. Comparator arms are platinum-based or physician’s choice chemotherapy except BNT324-01, which is single-arm.
The TAISHAN-302 toxicity line deserves separate attention. Grade 3 or higher treatment-related adverse events ran at 46.4% on the ADC against 74.7% on the comparator. An ADC that is substantially less toxic than the chemotherapy it replaces is unusual, and in a relapsed population with poor performance status it changes who is eligible rather than only how well they respond.
MAVERICK sits apart from the ADC story and may outlast it. If MRI surveillance can replace routine prophylactic cranial irradiation without an overall survival penalty, a standard of care disappears and a category of long-term cognitive harm goes with it.
In first-line non-small cell disease, response did not convert
| Trial | Population | Key results |
|---|---|---|
| HARMONi-2 ivonescimab | First-line PD-L1-positive advanced NSCLC | OS 30.8 vs 22.6 months, HR 0.73; previously reported PFS 11.1 vs 5.8 months, HR 0.51 |
| REZILIENT 3 zipalertinib plus chemotherapy | First-line EGFR exon 20 insertion NSCLC | PFS 14.5 vs 8.5 months, HR 0.50; ORR 65.0% vs 40.3% |
| DESTINY-Lung04 trastuzumab deruxtecan | First-line HER2-mutant advanced NSCLC | PFS 14.3 vs 8.3 months, HR 0.63; ORR 70.0% vs 44.5%; median OS 29.3 vs 33.1 months, HR 1.15 |
| EVOKE-03 sacituzumab govitecan plus pembrolizumab | First-line PD-L1-high metastatic NSCLC | PFS 11.8 vs 7.7 months, HR 0.81, missed statistical threshold; OS 21.5 vs 22.8 months, HR 1.07; ORR 55.6% vs 43.7%; Grade 3 or higher TRAEs 55.7% vs 16.5% |
| Iza-Bren dose expansion | Previously treated EGFR-mutant NSCLC, EGFR by HER3 bispecific ADC | At 2.5 mg/kg: confirmed ORR 29.6%; median PFS 6.9 months; no treatment-related deaths; one TRAE-related discontinuation |
Data as presented at WCLC 2026. DESTINY-Lung04 and EVOKE-03 overall survival figures are numerically in favour of the comparator.
DESTINY-Lung04 is the result to read twice. Progression-free survival favoured the ADC at a hazard ratio of 0.63 and response nearly doubled at 70.0% against 44.5%. Median overall survival came in at 29.3 months against 33.1, a hazard ratio of 1.15.
Those are not contradictory findings. They are a warning about which endpoint carries a first-line decision. Moving an ADC ahead of chemo-immunotherapy improves the measured tumour response and has not yet improved how long patients live.
EVOKE-03 tells a harsher version. Progression-free survival improved numerically and missed its threshold, overall survival was numerically worse, and grade 3 or higher treatment-related adverse events ran at 55.7% against 16.5%. More than three times the severe toxicity for no survival gain is a clear result rather than an ambiguous one.
The pattern worth taking awaySet the four ADC readouts side by side and the dividing line is not the target, the payload or the company. It is the setting. The two trials that halved mortality ran in relapsed small cell disease as single agents against chemotherapy, in patients defined by prior treatment failure. The two that failed on survival ran in the first line, against or alongside chemo-immunotherapy, in populations selected broadly or by a single biomarker. An ADC beating chemotherapy in a relapsed population is a different proposition from an ADC beating a functioning first-line standard, and the meeting priced those two propositions very differently.
What to watch
Whether the two B7-H3 Phase 3 results survive regulatory scrutiny of their comparator arms. Both ran against chemotherapy in relapsed small cell disease, which is a low bar by design and an appropriate one, but the hazard ratio of 0.46 will be read against the control arm as much as against the drug.
Whether DESTINY-Lung04 reports mature overall survival differently. A hazard ratio of 1.15 on median OS at this analysis is not a final answer, and the gap between the PFS and OS signals is the number that decides whether an ADC displaces chemo-immunotherapy in HER2-mutant first line.
And whether unselected frontline TROP2 and immunotherapy combinations continue at all after EVOKE-03. Greater tumour shrinkage with triple the severe toxicity and no survival benefit is the kind of result that closes a development path rather than adjusting it.
The headline from this meeting will be that ADCs advanced again. The more useful reading is that they advanced in one setting and stalled in another, and that the same class of molecule produced an overall survival hazard ratio of 0.46 in relapsed disease and 1.15 in the front line within a single programme of presentations.