Primary results from the phase 3 EVOKE-03/KEYNOTE D46 trial, presented at the International Association for the Study of Lung Cancer 2026 World Conference on Lung Cancer, evaluated sacituzumab govitecan (SG) combined with pembrolizumab as first-line therapy for patients with PD-L1 high (TPS ≥50%) metastatic non-small cell lung cancer (mNSCLC) lacking EGFR, ALK, and ROS1 alterations.
The combination produced a numerically longer median progression-free survival compared with pembrolizumab monotherapy (11.8 months vs. 7.7 months) but did not meet the prespecified threshold for statistical significance (HR, 0.81; 95% CI, 0.66–1.00; nominal P=0.0252). Overall survival also did not reach statistical significance at this interim analysis, with median OS of 21.5 months in the combination arm versus 22.8 months with pembrolizumab alone (HR, 1.07; 95% CI, 0.85–1.35; P=0.7155).
Key finding
Adding sacituzumab govitecan to pembrolizumab extended median PFS by approximately four months over pembrolizumab alone in PD-L1 high mNSCLC, but the improvement did not meet the trial’s prespecified statistical significance threshold, and OS was not improved at the interim analysis.
Study Design
| Trial | EVOKE-03/KEYNOTE D46 (NCT05609968); open-label, phase 3 |
|---|---|
| Population | Adults with untreated mNSCLC, PD-L1 TPS ≥50%, and no EGFR, ALK, or ROS1 alterations (N=620, intent-to-treat) |
| Treatment | Sacituzumab govitecan 10 mg/kg IV (days 1 and 8) plus pembrolizumab 200 mg IV (day 1) every 21 days vs. pembrolizumab 200 mg IV (day 1) every 21 days alone |
| Primary endpoints | Progression-free survival by blinded independent central review (BICR) and overall survival (dual primary) |
| Key secondary endpoints | Objective response rate, duration of response, safety, and patient-reported outcomes (change from baseline and time to deterioration) |
Randomization was stratified by histology (squamous vs. nonsquamous), geographic region (East Asia vs. Western Europe/North America/Australia), and ECOG performance status (0 vs. 1). The data cutoff for this primary analysis was April 3, 2026, at a median follow-up of 14.7 months. The trial was preceded by the phase 2 EVOKE-02 study (NCT05186974), which had shown encouraging preliminary results for the combination across histologies, including in the PD-L1 high subgroup.
Results and Safety
Among 620 randomized participants (311 in the combination arm, 309 in the monotherapy arm), the overall population was predominantly male (71.5%), with a median age of 67 years; 69.0% had nonsquamous histology and 16.9% were never-smokers. The 12-month PFS rate was 48.3% with the combination versus 36.9% with pembrolizumab alone, and the 18-month PFS rate was 36.0% versus 29.9%, respectively. Confirmed objective response rates, assessed by BICR per RECIST v1.1, were 55.6% (95% CI, 49.9%–61.2%) with the combination and 43.7% (95% CI, 38.1%–49.4%) with monotherapy. The disease control rate was 83.3% versus 73.1%, and fewer patients experienced progressive disease as their best overall response (7.1% vs. 16.5%). Median duration of response was similar between arms: 21.4 months with the combination versus 21.3 months with pembrolizumab alone.
OS rates at 12 and 18 months were nearly identical between the two arms (68.1% vs. 68.5% at 12 months; 56.2% vs. 56.7% at 18 months), consistent with the lack of OS separation at this interim analysis. The investigators noted that OS data are not yet mature and that this represents an interim assessment.
Treatment-related adverse events (TRAEs) occurred in 93.2% of participants receiving the combination versus 65.4% receiving pembrolizumab alone. Grade ≥3 TRAEs were reported in 55.7% and 16.5% of participants, respectively. The most common TRAEs in the combination arm were anemia, alopecia, neutropenia, diarrhea, and nausea. The investigators reported no new or unexpected toxicities beyond the established individual profiles of each agent. Detailed discontinuation and dose-modification data were not disclosed in the available source material.
Clinical Context
Pembrolizumab monotherapy is the established standard of care for first-line PD-L1 high mNSCLC, yet more than half of patients in this setting do not respond to or ultimately progress on single-agent immunotherapy. EVOKE-03 was designed to test whether adding SG, a Trop-2-directed antibody-drug conjugate, could meaningfully improve outcomes in this biomarker-selected population. The failure to meet the PFS primary endpoint, and the absence of an OS benefit at this interim analysis, suggests the combination does not offer a clinically meaningful advantage over pembrolizumab alone for this indication, at least at the current follow-up.
The higher response rate with the combination did not translate into a statistically significant PFS benefit or any OS signal, raising questions about what, if any, role SG may play in the PD-L1 high first-line setting. Whether longer follow-up will reveal OS separation, or whether different patient subgroups might derive differential benefit, was not addressed in the data presented. These results add to the accumulating evidence that augmenting checkpoint inhibitor regimens with antibody-drug conjugates in biomarker-unselected or broadly selected NSCLC populations remains a challenging strategy.
Sources
- Mountzios G et al. “Primary Results from Phase 3 EVOKE-03/KEYNOTE D46: Sacituzumab Govitecan + Pembrolizumab in PD-L1 TPS ≥50% Metastatic NSCLC.” Presented at IASLC World Conference on Lung Cancer 2026. Seoul, South Korea. Abstract PL02.06. View source.
AI disclosure: This article was prepared with assistance from generative AI using the source material cited above. It was reviewed and edited by Brandon Twyford before publication.