Background Blockade of PD-1 or PD-L1 (PD-1 blockade) with antibodies has revolutionized treatment options for patients with non-small cell lung cancer particularly with tumors that don’t present any targetable mutations. However, the benefit of PD-1 blockade is limited only to a subset of the patients. This study aims to identify biological markers of clinical response in a discovery and validation cohort of patients early on during treatment to distinguish responders from non-responders.

Methods Blood was drawn from patients with non-small cell lung cancer, at pre-treatment and at 4 cycles post-treatment. Immune cell subsets identified in the blood using a well characterized panel of immune markers that could distinguish CD4 and CD8 T cells based on proliferation, expression of immune checkpoint receptors, receptors of exhaustion and activation and cytokine secretion. Blood drawn from patients with non-small cell lung cancer before and after PD-1 blockade was also analyzed using a clinical flow cytometric assay for changes in the frequencies and concentration of immune cell populations particularly naive and memory CD4 and CD8 T cells. Further, next generation sequencing (NGS) was performed on the extracted DNA from archival tumor biopsies from the same patients as the flow cytometric analysis. The iRECIST criteria was used to evaluate the clinical responses with cut-off at 9 months.

Results In the responder patients a significant increase in the frequency of activated effector memory CD4 and CD8 T cells, and decreased frequency of central memory CD4 and CD8 T cells was observed in the blood post-treatment compared to pre-treatment. In a separate cohort of 29 patients, we could validate the initial findings and further report that CD8 T cells in non-responders expressed high levels of TNFα, TIGIT and PD-1 compared to non-responders at baseline. Our next-generation sequencing analysis of tumor biopsies identified the presence of pathogenic and likely-pathogenic mutations TP53, KRAS, KEAP1, NOTCH-1 and STK11 in the tumor of patients which correlated with response to PD-1 blockade. A multivariate analysis combining immune and genetic parameters could discriminate responders from non-responders for the discovery cohort of patients.

Conclusions We report that combined analyses of selected immune cell and genetic parameters could predict early clinical response to PD-1 blockade.