Development of an efficacy and safety predictive tool for patients affected by advanced or metastatic non small cell lung cancer treated with immunotherapy or chemo-immunotherapy

Lung cancer remains a significant challenge in modern oncology, despite the progress made in therapeutic options for patients. Currently, immunotherapy, either alone or in combination with chemotherapy, plays a pivotal role in treating advanced or metastatic non-small cell lung cancer lacking specific oncogenic alterations. While PD-L1 serves as a biomarker for clinical decisions, there is still a dearth of validated predictive biomarkers. Up to 40% of patients derive little or no benefit from these advanced approaches. Therefore, we sought to investigate the potential role of lymphocyte subpopulations and circulating interleukins as predictive biomarkers in these patients. We conducted investigations on 57 patients designated for single-agent immunotherapy and 120 patients designated for immunotherapy in combination with chemotherapy. Blood samples were collected from these patients on two occasions: first, prior to commencing therapy, and second, after two to three cycles of treatment. Through multivariate Cox regression analysis, we observed that, in the immunotherapy group, patients with high T gamma delta cell counts, low levels of IL1, and low levels of IL6 exhibited the most favourable clinical outcomes (p < 0.001). Conversely, in the chemo-immunotherapy group, patients with low levels of IL1 and low T regulatory cells demonstrated the highest performance (p < 0.001). Early fluctuations in T regulatory cells correlated with clinical outcomes in the chemo-immunotherapy group (p = 0.002), and a similar trend was observed in the immunotherapy group, although it did not reach statistical significance (p = 0.051). In terms of safety, we observed that serum levels of IL1, IL2, and IL6 were associated with the development of severe (grade-3 or -4) toxicities in both groups. While acknowledging the limitation of a monocentric trial, our study presents a potential predictive tool for integration into routine clinical practice to assist in decision-making. Further external validation on larger populations is essential to substantiate our findings.