AUTHOR=Mezi Silvia , Pomati Giulia , Fiscon Giulia , Amirhassankhani Sasan , Zizzari Ilaria Grazia , Napoletano Chiara , Rughetti Aurelia , Rossi Ernesto , Schinzari Giovanni , Tortora Giampaolo , Lanzetta Gaetano , D’Amati Giulia , Nuti Marianna , Santini Daniele , Botticelli Andrea TITLE=A network approach to define the predictive role of immune profile on tumor response and toxicity of anti PD-1 single agent immunotherapy in patients with solid tumors JOURNAL=Frontiers in Immunology VOLUME=14 YEAR=2023 URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2023.1199089 DOI=10.3389/fimmu.2023.1199089 ISSN=1664-3224 ABSTRACT=Background

The immune profile of each patient could be considered as a portrait of the fitness of his/her own immune system. The predictive role of the immune profile in immune-related toxicities (irAEs) development and tumour response to treatment was investigated.

Methods

A prospective, multicenter study evaluating, through a multiplex assay, the soluble immune profile at the baseline of 53 patients with advanced cancer, treated with immunotherapy as single agent was performed. Four connectivity heat maps and networks were obtained by calculating the Spearman correlation coefficients for each group: responder patients who developed cumulative toxicity (R-T), responders who did not develop cumulative toxicity (R-NT), non-responders who developed cumulative toxicity (NR-T), non-responders who did not develop cumulative toxicity (NR-NT).

Results

A statistically significant up-regulation of IL-17A, sCTLA4, sCD80, I-CAM-1, sP-Selectin and sEselectin in NR-T was detected. A clear loss of connectivity of most of the soluble immune checkpoints and cytokines characterized the immune profile of patients with toxicity, while an inversion of the correlation for ICAM-1 and sP-selectin was observed in NR-T. Four connectivity networks were built for each group. The highest number of connections characterized the NR-T.

Conclusions

A connectivity network of immune dysregulation was defined for each subgroup of patients, regardless of tumor type. In patients with the worst prognosis (NR-T) the peculiar connectivity model could facilitate their early and timely identification, as well as the design of a personalized treatment approach to improve outcomes or prevent irAEs.