Merck (NYSE: MRK) and Gilead Sciences (Nasdaq: GILD) announced the discontinuation of the Phase III KEYNOTE-D46/EVOKE-03 study after sacituzumab govitecan (Trodelvy) — Gilead's Trop-2-directed antibody-drug conjugate (ADC) — failed to produce statistically significant improvement in progression-free survival when added to pembrolizumab (Keytruda) in previously untreated metastatic non-small cell lung cancer (NSCLC) with high PD-L1 expression. The result closes off a meaningful expansion opportunity for sacituzumab govitecan beyond breast cancer and raises broader questions about whether Trop-2-targeted ADCs can add to checkpoint inhibitor monotherapy in a biomarker-selected lung cancer population.
The KEYNOTE-D46/EVOKE-03 study was a global, open-label, randomized Phase III trial enrolling approximately 620 patients with previously untreated metastatic NSCLC whose tumors expressed PD-L1 at a tumor proportion score of 50% or higher and lacked sensitizing EGFR, ALK, or ROS1 alterations — the same population for which pembrolizumab monotherapy is already an established standard of care. Patients were randomized 1:1 to receive sacituzumab govitecan (10 mg/kg intravenously on days 1 and 8 of a 21-day cycle) plus pembrolizumab (200 mg intravenously on day 1) or pembrolizumab alone. The trial carried dual primary endpoints of PFS assessed by blinded independent central review and overall survival.
Following a pre-specified final PFS analysis and interim OS analysis, the external Data Monitoring Committee recommended discontinuation. A numerical improvement in PFS was observed in the combination arm, but it did not reach statistical significance. Critically, the probability of achieving a statistically significant OS benefit at the planned final analysis was assessed as unlikely. No specific numerical values — median PFS, OS, hazard ratios, or p-values — were disclosed; full data are expected at a future medical meeting. The safety profile of the combination was consistent with the known profiles of each agent individually, and no new safety signals were identified, suggesting tolerability was not the limiting factor.