Trodelvy misses overall-survival goal in Phase 3 bladder-cancer trial; Gilead withdrew the indication

Gilead Sciences’ antibody-drug conjugate Trodelvy (sacituzumab govitecan) did not meet its primary goal of significantly prolonging overall survival in TROPiCS-04, a Phase 3 trial in heavily pretreated metastatic or locally advanced unresectable urothelial cancer, according to results posted to the federal trial registry.

The open-label study randomized 711 patients whose disease had progressed after both platinum-containing chemotherapy and a PD-1/PD-L1 checkpoint inhibitor, comparing Trodelvy against physician’s choice of single-agent chemotherapy — paclitaxel, docetaxel, or vinflunine. Per the trial’s arm description, sacituzumab govitecan was given at 10 mg/kg intravenously on Day 1 and Day 8 of each 21-day cycle.

A narrow miss on survival

Median overall survival was 10.3 months with Trodelvy (95% CI 9.1–11.8) versus 9.0 months for chemotherapy (95% CI 7.5–9.7). The hazard ratio was 0.86 (95% CI 0.73–1.02), with a p-value of 0.0870 — short of the threshold for statistical significance. Because the confidence interval crosses 1.0, the trial cannot rule out that the two arms produce equivalent survival.

A 1.3-month numerical survival edge that the data cannot statistically distinguish from chance — set against more deaths from adverse events in the Trodelvy arm.

Several secondary endpoints favored Trodelvy numerically. Investigator-assessed progression-free survival reached 4.0 months versus 2.9 months with chemotherapy (HR 0.81; 95% CI 0.68–0.95; p=0.0117), but progression-free survival assessed by blinded independent central review was not significant (median 4.2 vs 3.6 months; HR 0.86; 95% CI 0.72–1.03; p=0.1095).

Response figures depended on who did the reading. The objective response rate by blinded independent central review was 22.5% with Trodelvy versus 13.8% with chemotherapy (p=0.0022), and the clinical benefit rate by that same review was 29.9% versus 20.5% (p=0.0034). Investigator-assessed clinical benefit rate was higher still, at 31.5% versus 21.9% (p=0.0032). But because the primary endpoint was not met, the trial’s hierarchical testing renders these secondary results descriptive rather than confirmatory; they did not establish a net clinical benefit.

An excess of treatment-related deaths

The lack of a survival benefit was compounded by a safety signal. Gilead’s October 2024 update and the data presented at ESMO Asia 2024 reported a higher number of deaths from adverse events in the Trodelvy arm than in the chemotherapy arm. These excess deaths occurred primarily early in treatment and were driven by neutropenic complications, including infection. Grade 5 (fatal) treatment-emergent adverse events were reported in about 7% of Trodelvy-treated patients versus about 2% of those on chemotherapy.

The result lands in a difficult treatment setting. Patients entering TROPiCS-04 had already exhausted the two pillars of advanced urothelial cancer therapy, and the chemotherapy comparator arm itself reflects limited remaining options. Trodelvy had held U.S. accelerated approval in this indication: the FDA granted it on April 13, 2021, based on the response rate seen in the single-arm Phase 2 TROPHY-U-01 trial (confirmed objective response rate 27.7%, 95% CI 19.6–36.9, in 112 patients), with continued approval contingent on a confirmatory study. TROPiCS-04 served as that confirmatory trial.

The basis for withdrawal was an unfavorable benefit-risk balance — the absence of a survival benefit combined with the excess of toxicity-related deaths — rather than the efficacy miss in isolation. Gilead moved in October 2024 to voluntarily withdraw the urothelial-cancer indication, in consultation with the FDA, and the FDA announced the final withdrawal of the approval on November 22, 2024.

The trial is listed as completed, with primary completion in July 2025. The registry posting does not address further regulatory steps, and the detailed safety profile and any prespecified subgroup analyses await peer-reviewed publication. For now, the conclusion is unambiguous: on the endpoint that mattered most, the difference between Trodelvy and chemotherapy was not statistically significant, and the drug carried an excess of treatment-related deaths.