Adding the antibody-drug conjugate (ADC) sacituzumab govitecan (Trodelvy) to pembrolizumab (Keytruda) did not improve survival outcomes in patients with metastatic non-small cell lung cancer (NSCLC), according to results from the EVOKE-03 phase III trial.
Median progression-free survival (PFS) numerically improved from 7.7 months with pembrolizumab monotherapy to 11.8 months with the combination regimen; however, the difference failed to meet prespecified criteria for statistical significance. An interim analysis of overall survival (OS), the second dual primary endpoint, showed a slight numerical advantage for the control arm (22.8 versus 21.5 months) in a study population selected for PD-L1 positivity. Prespecified subgroup analyses of patients from East Asia and China suggested a potential OS benefit for sacituzumab govitecan, but neither comparison achieved statistical significance.
Treatment-emergent adverse events (AEs), including serious, treatment-discontinuation, and fatal AEs, all occurred more frequently in the sacituzumab group, reported Giannis Mountzios, MD, PhD, of Henry Dunant Hospital Center in Athens, Greece, at the World Conference on Lung Cancer (WCLC) in Seoul, South Korea.
“At this primary analysis for PFS, the study did not meet its primary endpoint because the difference was not statistically significant,” said Mountzios. “Confirmed objective response rates and disease control rates were numerically higher with the combination than with pembrolizumab monotherapy, and median duration of response [DOR] was similar between the treatment groups. At the interim analysis for overall survival, a statistically significant difference for this dual primary endpoint was not met, and it is highly unlikely that it will occur in the future.”
He added that the safety profile of sacituzumab govitecan plus pembrolizumab was consistent with the known profiles of each agent, and no new or additional toxicities were observed with the combination.
The results reflected a familiar pattern of attrition from phase II to phase III trials, said WCLC discussant Ji-Youn Han, MD, PhD, of the National Cancer Center in Goyang, South Korea. Pembrolizumab had already established a high benchmark in the KEYNOTE-024 trial with an objective response rate (ORR) of 44.8%. In the phase II EVOKE-02 trial evaluating sacituzumab as an add-on, the 95% confidence intervals included 44.8%, which she described as a “recipe for overestimation.”
Although EVOKE-02 reported an ORR of 66.7%, those results came from a 30-patient study, Han noted. In EVOKE-03, ORR decreased to 55.6%, representing a 12% absolute difference over pembrolizumab alone—comparable to the incremental benefit previously observed with chemotherapy. The median DOR did not differ between treatment groups, suggesting that the sacituzumab payload “acts like chemotherapy on top of immuno-oncology, turning non-responders into short-term responders.”
“Pembrolizumab monotherapy remains a standard for PD-L1-high groups,” said Han. “Sacituzumab govitecan plus pembrolizumab should not be pursued in this setting outside of clinical trials. An ADC plus anti-PD-1 antibody provides an early chemotherapy-like effect without a survival tail. Future trials need overall survival as the primary endpoint, with deliverable biomarker selection and prespecified histology and geography stratification.”
EVOKE-03 was designed in recognition of the fact that more than half of patients with metastatic NSCLC do not respond to initial treatment with pembrolizumab monotherapy, said Mountzios. Sacituzumab is a TROP-2-directed ADC that showed preliminary encouraging activity in EVOKE-02 across tumor histologies and in patients with PD-L1 total positive score (TPS) ≥50%.
Patients eligible for EVOKE-03 had untreated stage NSCLC, PD-L1 TPS ≥50%, no untreated or unstable brain metastases, and no EGFR/ALK/ROS1 mutations. Investigators in the international trial randomized 620 patients to pembrolizumab alone or in combination with sacituzumab. PFS and OS served as dual primary endpoints, and secondary endpoints included ORR, DOR, patient-reported outcomes, and safety.
The primary analysis occurred after a median follow-up of 14.7 months. The 4.1-month improvement in PFS translated into a hazard ratio of 0.81 favoring the combination, with a P value of 0.0250. However, the protocol specified a P value of 0.007 for statistical significance, resulting in a negative trial.
A subgroup analysis suggested statistical superiority of the combination in patients younger than 65 (HR 0.65, 95% CI 0.47–0.91), non-white race/ethnicity (HR 0.69, 95% CI 0.51–0.98), patients from East Asia (HR 0.68, 95% CI 0.48–0.97), non-squamous histology (HR 0.70, 95% CI 0.54–0.90), no brain metastases (HR 0.77, 95% CI 0.62–0.95), and absence of liver metastases (HR 0.76, 95% CI 0.61–0.96).
The OS analysis showed no statistically significant advantages for the addition of sacituzumab. ORR favored the combination (55.6% versus 43.7%), as did disease control rate (83.3% versus 73.1%). Median DOR was 21.3 to 21.4 months in both groups.