Predictive value of longitudinal systemic inflammatory markers for pathologic response to neoadjuvant PD-1 blockade in resectable non-small cell lung cancer.

Background: Identifying biomarkers to predict responses for neoadjuvant immunotherapy in resectable non-small cell lung cancer (NSCLC) is under intensive study. Considering the interplay between cancer, inflammation, and immunosuppression, we hypothesized that circulating and imaging inflammatory markers could serve as indicators of anti-tumor immune responses, and thus conducting an exploratory study to reveal the predictive value of combining longitudinal systemic inflammatory markers in stratifying pathologic response to neoadjuvant sintilimab.

Methods: We retrospectively reviewed 36 patients (29 male and seven female) with NSCLC (stage IA-IIIB) who underwent pre- and post-treatment peripheral blood tests and F-fluorodeoxyglucose positron emission tomography/computed tomography (F-FDG PET/CT) scans before and after two cycles of neoadjuvant sintilimab (registration number: ChiCTR-OIC-17013726). The neutrophil-to-lymphocyte ratio (NLR), immune-related adverse events (irAEs) on imaging, and lymphoid organ metabolism [spleen-to-liver ratio (SLR) and bone marrow-to-liver ratio (BLR)] were evaluated to examine their predictive value for the major pathologic response (MPR). Significant variables were used to classify patients into low, intermediate, and high inflammatory burden groups for stratifying pathologic regression and tumor-infiltrating immune cells abundance in the tumor microenvironment. Spearman's correlation analysis was performed to explore the correlation between systemic inflammatory markers, primary tumor metabolism, and tumor-infiltrating immune cells abundance at various time points.

Results: Of the 36 enrolled patients, 13 (36.1%) exhibited MPR. ΔNLR% was a significant negative predictor of MPR (P=0.047) and negatively correlated with pathologic regression (r=-0.34, P=0.045). Pre- and post-treatment SLRs were potential negative predictors of MPR (P=0.06; P=0.055) and negatively correlated with pathologic regression (r=-0.30, P=0.07; r=-0.31, P=0.06). The high inflammatory burden group (pre-treatment SLR >0.83 and ΔNLR% >-17%) had the lowest pathologic regression (P=0.01) and the highest infiltration abundance of pre-treatment CD68 macrophage (P=0.01-0.04). irAEs on imaging did not have significant effects on MPR and pathologic regression in overall and per-organ analyses.

Conclusions: The combination of pre-treatment SLR and ΔNLR% demonstrates predictive value in stratifying pathologic response to neoadjuvant immunotherapy in resectable NSCLC. The high inflammatory burden group had the lowest pathologic regression and the pre-treatment immunosuppressive microenvironment with macrophage enrichment.