The fate of sotorasib: a regulatory failure potentially harming patients

• Published in: The lancet oncology 2024-05, Vol.25 (5), p.549-552
• Main Authors: Ranganathan, Sruthi, Prasad, Vinay, Olivier, Timothée
• Format: Article
• Language: English
• Subjects: Antineoplastic Agents - adverse effects; Antineoplastic Agents - therapeutic use; Cancer therapies; Comorbidity; Drug Approval - legislation & jurisprudence; FDA approval; Humans; Oncology; Patients; Survival; United States; United States Food and Drug Administration
• ISSN: 1470-2045; 1474-5488
• DOI: 10.1016/S1470-2045(23)00616-2

[...]a protocol amendment in CodeBreaK 200 permitted crossover in a trial testing the fundamental efficacy of sotorasib. Since crossover was permitted and no survival benefit was seen, one could argue that patients in both groups benefited from sotorasib, regardless of whether they received the therapy early or late and leading to the non-significant overall survival benefit. [...]the numerical survival data are worse in the sotorasib group than in the control group, with 109 (64%) of 171 patients receiving sotorasib having died as of data cutoff date, compared with 95 (54%) of 174 patients who received docetaxel. [...]informative censoring might have occurred, meaning that patients censored because of loss to follow-up were different—in terms of characteristics and comorbidities—than those remaining in the study who were censored at data cutoff, and these differences varied by group. [...]the potential overestimation of the progression-free survival benefit of sotorasib over docetaxel was further supported by an interval-censoring analysis.