The Phase III LAGOON trial has failed to confirm the second-line benefit of lurbinectedin (Zepzelca) in relapsed small cell lung cancer, a result that puts Jazz Pharmaceuticals plc (Nasdaq: JAZZ) under pressure to defend an accelerated approval that was always contingent on this outcome. The trial missed its primary endpoint of overall survival across all three arms, with lurbinectedin monotherapy producing a median OS of 8.7 months versus 10.7 months for the control arm — a hazard ratio of 1.190 that numerically favors the comparator. The combination arm of lurbinectedin plus irinotecan showed a median OS of 10.9 months, numerically similar to control, with an HR of 0.902.
Lurbinectedin is a selective inhibitor of oncogenic transcription that binds DNA and disrupts transcriptional processes on which small cell lung cancer cells are particularly dependent. The drug also modulates the tumor microenvironment by reducing tumor-associated macrophages and inflammatory signaling. These mechanisms contributed to the objective response rates observed in the Phase II study that supported the drug's accelerated approval in second-line SCLC in 2020.
Trial data
LAGOON is a Phase III, randomized, open-label, three-arm study enrolling 724 patients with SCLC who had progressed on prior platinum-based chemotherapy. Patients were randomized 1:1:1 to lurbinectedin 3.2 mg/m² monotherapy, lurbinectedin 2.0 mg/m² plus irinotecan 75 mg/m², or investigators' choice of topotecan or irinotecan.
A notable design feature was the inclusion of patients with a history of CNS involvement — a population excluded from the earlier Phase II study that supported accelerated approval. This broader enrollment appears to have materially affected the monotherapy result: patients with CNS metastases receiving lurbinectedin monotherapy had a median OS of 7.1 months versus 10.3 months for control (HR 1.791), while those without CNS involvement showed a median OS of 9.6 months versus 10.7 months for control (HR 1.106). The control arm also performed better than historical precedent, a pattern that can inflate hazard ratios against investigational arms. Tolerability data offered some distinction: Grade ≥3 treatment-related adverse events occurred in 35% of patients on lurbinectedin monotherapy versus 64.4% in the control arm, suggesting a cleaner safety profile even in the absence of a survival benefit.
