Analysis of gene expression levels and their impact on survival in 31 cancer-types patients identifies novel prognostic markers and suggests unexplored immunotherapy treatment options in a wide range of malignancies

Abstract

Abstract Background Immunotherapy has dramatically improved cancer treatment by inhibiting or activating specific cell receptors, thus unleashing the host anti-tumor response. However, the engagement of the three main immune checkpoints so far identified, CTLA4, PD-1 and PD-L1, is effective in a fraction of patients, therefore novel targets must be identified and tested. Methods We focused our attention on the following nine highly relevant immune checkpoint (ICR) receptors: CTLA4, PD1, PD-L1, LAG3, TIM3, OX40, GITR, 4-1BB and TIGIT. All of them are targets of existing drugs currently under clinical scrutiny in several malignancies. Their expression levels were evaluated in patient tissues of 31 different cancer types vs. proper controls, in a total of 15,038 individuals. This analysis was carried out by interrogating public databases available on GEPIA2 portal and UALCAN portal. By the Principal Component Analysis (PCA) their ability to effectively discriminate patients form controls was then investigated. Expression of the nine ICRs was also related to overall survival in 31 cancer types and expressed as Hazard Ratio, on the GEPIA2 portal and validated, for melanoma patients, in patients-datasets available on PROGgene V2 portal. Results Significant differential expression was observed for each ICR molecule in many cancer types. A 7-molecules profile was found to specifically discriminate melanoma patients from controls, while two different 6-molecules profiles discriminate pancreatic cancer patients and Testicular Germ Cell Tumors from matched controls. Highly significant survival improvement was found to be related to the expression levels of all nine ICRs in a wide spectrum of malignancies. For melanoma analysis, the relation with survival observed in TCGA datasets was validated in independent GSE melanoma datasets. Conclusion Analysis the nine ICR molecules demonstrates that their expression patterns may be considered as markers of disease and strong survival predictors in a variety of malignancies frequently associated to poor prognosis. Thus, the present findings are strongly advocating that exploratory clinical trials are worth to be performed, using available drugs, targeting these molecules.