Outcomes of long-term responders to immunotherapy in the treatment of advanced non-small cell lung cancer (NSCLC). A retrospective regional study.

BACKGROUND: For advanced lung cancer patients, an increasing number of long responders to Immune Checkpoint Inhibitors (ICI) is observed. Thus, the optimal treatment duration remains unclear whereas indefinite treatment is associated to immune-related adverse events (irAEs) and financial burden. OBJECTIVE: To describe medical evaluations and therapeutic decisions at the two-year mark for advanced NSCLC patients on ICI. METHODS: This retrospective cohort study included advanced NSCLC patients from two French academic hospitals who received ICI (first-line or subsequent, ± chemotherapy) for at least two years. Primary outcome was overall survival (OS), secondary outcomes were progression-free survival (PFS), medical exams performed at two years and the resulting therapeutic decisions. RESULTS: Among 191 patients included in the statistical analysis, OS and PFS did not differ between discontinuation and continuation groups. Median OS was not reached in the "stop group" versus 62.6 months in the continuation group (HR 0.75, p=0.42). Median PFS was of 60.7 months in the "stop group" vs. not reached in the continuation group (HR 0.63, p=0.14). Evaluations at 24 months included: whole-body computed tomography (WBCT) alone (43.5%), WBCT + PET (24.5%), WBCT + cerebral MRI (13%), WBCT + cerebral MRI + PET (11.5%), and cerebral MRI + PET (7.5%). PET scans were more frequent post-2020 (OR 6.23, p<0.01), associated with local treatments (OR 3.41, p<0.02), and with improved survival (OR 0.26, p<0.01). ICI were stopped for 99 patients (49.5%), with 4.5% receiving local treatments, and continued for 101 patients (50.5%), with 7% receiving local treatments. Reasons for continuation included oligoprogression (10%), persistent hypermetabolism (7.1%), early progression (6.1%), patient request (10.1%) and medical decision (58.6%). Discontinuation correlated with irAEs (p<0.05) and disease response (p<0.01). CONCLUSION: This study provides reassuring data supporting treatment cessation for patients with controlled disease after two years of ICI. However, it is subject to potential selection biases because of its retrospective nature.