Abstract Background Immune checkpoint inhibitors (ICI) are approved for metastatic cervical cancer (mCC) after platinum failure, but only a minority of patients derive durable benefit. We aimed to compare long-term responders (LTR) and non–long-term responders (NLTR) to ICI and to identify clinical and molecular predictors of long-term response. Methods We retrospectively reviewed patients with mCC treated with ICI at Gustave Roussy. LTR were defined as patients achieving complete or partial response or stable disease lasting >12 months, while NLTR progressed within 12 months. Logistic regression analyses were performed to identify factors associated with LTR. Genomic analyses were available for a subset of patients using large-panel next-generation sequencing. Results Between January 2015 and September 2023, 106 patients received ICI; 20 (19%) were classified as LTR and 86 (81%) as NLTR. After a median follow-up of 42 months, median progression-free survival was not reached in LTR versus 3.5 months in NLTR. Eleven LTR maintained a sustained response after ICI discontinuation. In multivariate analysis, oligometastatic disease, lymph node–only metastases, and progression outside prior radiation fields were independently associated with LTR. Tumour mutational burden and mismatch repair status were not predictive. PIK3CA mutations were significantly more frequent in LTR, while STK11 alterations were enriched in NLTR. Conclusion Durable responses to ICI in mCC, although uncommon, are associated with prolonged disease control and may persist after treatment discontinuation. Specific clinical and molecular features are strongly associated with long-term benefit and may help refine patient selection for immunotherapy.
Barraud et al. (Tue,) studied this question.