Aims: Anti-PD-1/PD-L1 can face the issue of tumor resistance in advanced non-small cell lung cancer (aNSCLC), mostly as primary resistance to the treatment. Still, disease progression (PD) also occurs after prior durable clinical benefit (DCB). Comparison of both situations and tools to evaluate residual benefit of PD-1/PD-L1 blockade are needed to understand and manage resistance. Methods: We reviewed aNSCLCs with anti-PD-1/PD-L1 monotherapy, and disease progression per RECIST 1.1 (PD) in our center. Primary objective was comparison of post-progression survival (PPS) in primary resistance versus PD after DCB. Secondary objectives were to characterize patterns of PD after DCB; assess the feasibility and relevance of Tumor Growth Rate (TGR) in PD after DCB. Results: 148 patients were included. Median PPS were 5.2months (95% CI: 2.6-6.5) and 21.3months (95% CI: 18.5-36.3) (p<0.0001) for primary resistance and PD after DCB respectively. Multivariable hazard ratio for death in PD after DCB versus primary resistance was 0.14 (95% CI: 0.07-0.30). 76.7% of PD after DCB occurred in ≤3 lesions; 72.1% occurred only in preexisting lesions. TGR suggested a persistent benefit of PD-1/PD-L1 blockade in at least 44.2% of cases. Conclusions: PD after DCB was an independent factor of longer post-progression survival, with specific patterns of progression. TGR is a promising tool to assess residual benefit of immunotherapy.

The development of immune checkpoint inhibitors (ICIs) constitutes a major therapeutic advance in the treatment of a number of malignancies [...]