Post-discontinuation survival in patients with advanced non-small cell lung cancer receiving immune checkpoint inhibitors: a pooled analysis of prospective cohort studies

The safety of discontinuing immune checkpoint inhibitors (ICIs) because of a durable response in patients with advanced NSCLC remains uncertain, and post-discontinuation survival outcomes based on the reason for cessation are not well defined. A pooled analysis was conducted using data from four prospective cohort studies involving 835 patients with advanced NSCLC who discontinued ICIs. Patients were categorized based on discontinuation reasons: durable response; immune-related adverse events (irAEs) (subcategorized by tumor response at discontinuation); non-irAE adverse events; disease progression; and other causes. Disease progression was the most common reason for ICI discontinuation (N = 528 [63.2%]), followed by irAEs (N = 187 [22.4%]) and tumor response (N = 23 [2.8%]). Regarding response status at ICI discontinuation due to irAEs, complete/partial response (CR/PR) was the most frequent (N = 85), followed by stable disease/not evaluable (SD/NE, N = 69) and disease progression (N = 33). After a median post-discontinuation follow-up of 15.8 months (interquartile range, 6.9-23.2), patients who discontinued because of a response had excellent outcomes, with no deaths and only three progression-free survival events. While post-discontinuation overall survival was comparable between the irAE-CR/PR and irAE-SD/NE groups, ICI therapy ≥12 months was associated with improved post-ICI discontinuation survival in the irAE-CR/PR group. Discontinuation of ICIs because of a durable tumor response is rare in real-world settings but represents a feasible strategy for patients with advanced NSCLC. Patients in the irAE-CR/PR group had favorable post-ICI discontinuation survival if they received ICI therapy lasting ≥12 months.