The fact that Gilead Sciences’ Trodelvy failed as a first-line treatment for non-small cell lung cancer was not new. Yet newly revealed global phase 3 data could reignite the question around translating Chinese data to the West.
According to results from the Evoke-03 trial presented at the 2026 World Conference on Lung Cancer, the combination of Trodelvy and Keytruda demonstrated a numerically longer progression-free survival versus Keytruda alone, reducing the risk of progression or death by 19%, which did not meet statistically significance. The Merck-sponsored study, also coded Keynote-D46, was conducted in patients with previously untreated, PD-L1-high NSCLC.
When it comes to overall survival, the Trodelvy-Keytruda combo performed even worse than Keytruda monotherapy, with a negative trend hinting at a 7% increased risk of death.
At an interim analysis conducted when 47% of patients in the 620-subject trial had passed away, the median OS length for the Trodelvy regimen was shorter, at 21.5 months, versus 22.8 months for Keytruda.
The trial therefore failed on both its dual primary endpoints, dealing another blow to Gilead’s plan to establish Trodelvy as a cornerstone of its solid tumor franchise. Following announcement of the surprise setback in June, Gilead in the second quarter recorded an impairment worth $1.75 billion related to the TROP2 antibody-drug conjugate for NSCLC.
Despite the Evoke-03 flop, a Gilead spokesperson told Fierce that the company “remains confident in the potential of Trodelvy across various solid tumors, including in phase 3 studies in early breast cancer, endometrial cancer and small cell lung cancer.” The first-in-class TROP2 ADC boasts approvals in certain triple-negative breast cancer and HR-positive, HER2-negative breast cancer settings.
At this point, Evoke-03 looked like little more than another lamentable phase 3 failure of a once-promising asset, which formed the centerpiece of Gilead’s $21 billion acquisition of Immunomedics. But in a small twist, the study’s presenter, Giannis Mountzios, M.D., of the Henry Dunant Hospital Center in Greece, provided a post-hoc analysis at WCLC showing better OS results in patients enrolled in East Asia and China.
Among 230 patients (37%) in East Asia, the Trodelvy-Keytruda regimen led to a 27% OS improvement over Keytruda alone. The combo performed even better, pulling off a 35% OS advantage, in Chinese patients.
Among 107 Chinese patients, median OS was not reached for the Trodelvy-Keytruda arm, with the lower bound of the 95% confidence interval reaching 24.2 months; whereas the median OS for Keytruda landed at 27.9 months, with 15.8 months as the lower bound of 95% CI.
In a statement to Fierce, Gilead’s spokesperson noted that all subgroup analyses of Evoke-03 are exploratory in nature as the trial missed its dual primary endpoints.
But the efficacy gap could again reopen questions about the translatability of China data to a global population. It could also affect perception of Merck’s own TROP2 ADC program, sac-TMT, partnered with China’s Kelun-Biotech.
The New Jersey pharma has launched nearly 20 global phase 3 trials for sac-TMT, emboldened by multiple positive phase 3 readouts from the drug’s China-only studies. One of them, OptiTROP-Lung05 in first-line, PD-L1-positive NSCLC, excited industry watchers during ASCO 2026 by showing a massive 65% PFS advantage over Keytruda monotherapy and a strong—though immature—45% trend toward an OS improvement.
Then in July, Kelun announced that its phase 3 OptiTROP-Lung06 study for sac-TMT and Keytruda in first-line PD-L1-negative nonsquamous NSCLC also met its PFS endpoint, alongside a favorable OS trend.
Sac-TMT recently chalked up its first global phase 3 win, as the TroFuse-005 trial in previously treated endometrial cancer met both PFS and OS endpoints.
During an interview with Fierce on the sidelines of ASCO 2026, Marjorie Green, M.D., Merck’s head of oncology clinical development, flagged that different clinical practices and available treatment options could impact patients’ OS outcomes from the same drug in different geographies. By comparison, other endpoints such as tumor response rate and PFS are not impacted by subsequent therapies and can therefore serve as “reasonable surrogates” for understanding an ADC’s potential based on Chinese data.
In response to a similar China data question during a press briefing about WCLC data, GSK’s oncology R&D head, Hesham Abdullah, M.D., pointed to consistent tumor response data observed for its B7-H4 ADC, mocertatug rezetecan (mo-rez), in both its partner Hansoh Pharma’s Chinese trials and GSK’s own development program.
“I think the one thing that we probably always have to take into account is the current standard-of-care availability of therapies,” he said.