Real-World Outcomes of Immunotherapy Discontinuation at Two Years in Advanced Non-Small Cell Lung Cancer: final results of the ISTOP study.

• Real-world outcomes after 24-month anti-PD-1/PD-L1 monotherapy remain unclear • No overall survival difference between discontinuation and continuation of anti-PD-1/PD-L1 monotherapy • Progression risk after discontinuation was higher in selected clinical subgroups • ICI rechallenge was feasible and effective in most relapsing patients • Late grade 3–4 immune-related events were uncommon after long-term ICI use Real-world outcomes after 24-month anti-PD-1/PD-L1 monotherapy remain unclear No overall survival difference between discontinuation and continuation of anti-PD-1/PD-L1 monotherapy Progression risk after discontinuation was higher in selected clinical subgroups ICI rechallenge was feasible and effective in most relapsing patients Late grade 3–4 immune-related events were uncommon after long-term ICI use The optimal duration of immune checkpoint inhibitors (ICIs) in advanced non-small cell lung cancer (NSCLC) remains undefined. While most trials adopt a 2-year treatment cap, evidence supporting this strategy in real-world practice is limited. The I-STOP study evaluated outcomes of patients with advanced NSCLC who achieved disease control after 24 months of ICI therapy and either discontinued or continued treatment. I-STOP is a multicenter, observational study including patients with advanced NSCLC treated with single-agent PD-1/PD-L1 inhibitors for ≥24 months in 12 Italian institutions. Patients with complete response, partial response, or stable disease at 24 months were categorized according to treatment discontinuation. Progression-free survival (PFS) and overall survival (OS) were compared using 24-month landmark analyses; prognostic factors were assessed through Cox models. Among 173 eligible patients (median follow-up 55.4 months), 28 (16.2%) discontinued ICI therapy at 24 months. Median OS was not reached in either cohort, with no difference between discontinuation and continuation groups (HR 1.22, 95% CI 0.50–2.95; p = 0.66). Median PFS was 35.6 months in the discontinuation group versus not reached in the continuation group (HR 2.16, 95% CI 1.04–4.48; p = 0.035). Higher progression risk after discontinuation was observed in patients with PD-L1 ≤ 50%, brain metastases, ECOG PS ≥ 1, or KRAS mutations. Rechallenge achieved a 62.5% response rate, and late grade 3–4 immune-related events were uncommon (3.5%). Discontinuing ICIs at 24 months appears to be a safe strategy for patients with durable disease control, without compromising survival outcomes. Nonetheless, some subgroups may benefit from continued treatment. Larger prospective studies are needed to identify which patients could gain from prolonged immunotherapy.