Gilead Sciences is drawing back the curtain on the Trodelvy lung cancer data that previously caused a 10% slide in the company’s stock price.
The TROP2-directed antibody-drug conjugate (ADC) led to a 16% reduction in the risk of death compared with the chemotherapy docetaxel in non-small cell lung cancer patients who had failed on chemo and PD-1/L1 therapy. As Gilead revealed in January, the overall survival readout, from the phase 3 EVOKE-01 study, was not statistically significant.
The results are being shared at the American Society of Clinical Oncology annual meeting and published in the Journal of Clinical Oncology.
While the study missed its primary endpoint, Gilead is hanging onto a positive sign in a subgroup of patients, whom Bilal Piperdi, M.D., VP of oncology clinical development, said have the highest unmet medical need. However, convincing the FDA could be a stretch. And cross-trial comparisons with AstraZeneca and Daiichi Sankyo’s rival drug raise several important considerations.
In patients whose tumors didn’t respond to their last PD-1/L1 treatment, Trodelvy was associated with a 25% reduction in the risk of death. The ADC extended patients’ median life expectancy by 3.5 months to 11.8 months. In immunotherapy-responsive patients, Trodelvy performed numerically worse than chemo, with median overall survival fingures of 9.6 months and 10.6 months, respectively, resulting in a 9% negative trend against the Gilead drug.
The subgroup analysis was pre-specified in the trial design but was not powered for formal statistical testing. Immunotherapy-unresponsive patients made up about two-thirds of the trial population.
The composition of EVOKE-01’s population reflects the reality that about 40% to 50% of patients will respond to first-line PD-1/L1 treatments, Piperdi noted. Unresponsive patients typically don’t have good outcomes, he added, arguing that Trodelvy’s 3.5-month improvement in the group was clinically meaningful.
As to the negative trend seen in the PD-1-responsive group, the Gilead exec argued that chemotherapy had somehow outperformed its usual efficacy.
Holding the belief that Trodelvy could still fill a treatment gap, Gilead is assembling the data and plans to speak with the FDA and doctors “to see the path forward,” Piperdi said. He declined to answer whether Gilead would be open to running a separate clinical trial focused on the unresponsive patients.
Getting the FDA on board with a subgroup analysis will be a tall task, if not entirely impossible. Oncology officials from the agency have said multiple times that they would use subgroup analyses to limit the scope of an approval but not to salvage a failed clinical trial with an approval.
In a recent note to clients, Leerink Partners analysts said they do not see a path to approval for Trodelvy following the primary endpoint miss. Still, the team added that the detailed findings from EVOKE-01 could inform the broader TROP2 ADC class and Gilead’s development strategy.
Gilead’s setback comes as competition in the TROP2 ADC space heats up. Earlier this week, AstraZeneca and Daiichi Sankyo said a phase 3 trial for their rival TROP2 drug, datopotamab deruxtecan (Dato-DXd), in a similar previously treated NSCLC setting, also failed to meet statistical significance on overall survival.
Luckily for AZ and Daiichi, their TROPION-Lung01 trial uses a dual primary endpoint design. And the trial is still considered positive because it had previously hit statistical significance on progression-free survival (PFS). The FDA has accepted the companies’ application based on the PFS readout, albeit in a subgroup of patients with nonsquamous tumors, in which Dato-DXd showed a benefit.
But Trodelvy likely wouldn’t have gotten a positive readout even if PFS were EVOKE-01’s primary endpoint. In the study, Trodelvy only extended patients’ median PFS by 0.2 months compared with chemo, with an 8% reduction in the risk of tumor progression or death.
RELATED: AstraZeneca-Daiichi's Enhertu follow-up Dato-DXd unable to prove overall survival benefit in phase 3
Overall survival is the gold-standard endpoint from a regulatory perspective, particularly in second-line NSCLC, because it captures the overall efficacy-risk profile of a drug, Piperdi said. That’s why Gilead designed EVOKE-01 to show improvement in overall survival, with PFS being a secondary endpoint.
One area that Gilead’s Trodelvy appeared to have performed better than AZ/Daiichi’s Dato-DXd was in patients with squamous NSCLC. Previously, an interim analysis of TROPION-Lung01 linked Dato-DXd to a potential detriment to patient survival in squamous NSCLC, leading to an FDA filing solely in the nonsquamous group.
In EVOKE-01, Trodelvy showed roughly similar overall survival benefit between squamous and nonsquamous histologies, with death risk reductions at 17% and 13%, respectively.
Another subgroup analysis of interest divides patients by whether their tumors have actionable genomic alterations (AGAs) such as EGFR mutations. Previously, Dato-DXd showed a 62% PFS improvement in patients with AGA, whereas the benefit in non-AGA patients was smaller at just 16%.
Now in EVOKE-01, Trodelvy showed a similar pattern. In patients who had received previous therapy for AGA, Trodelvy cut the risk of death by 48%. The number was only 11% among those who had not.
The relatively muted benefits seen in the AGA-negative population—and the failed overall survival endpoints—across two agents in the same class don’t bode well for Dato-DXd’s ongoing FDA review. It raises the possibility that the agency might further limit a potential indication by AGA status.
The FDA is expected to deliver a verdict on the Dato-DXd NSCLC application in December. Meanwhile, Piperdi said Gilead will provide a regulatory update on Trodelvy’s NSCLC bid when appropriate.