Evaluation of the Prognostic Power of Complement System in Malignant Pleural Mesothelioma

Malignant pleural mesothelioma (MPM), a rare tumour arising from the mesothelial cells of the pleura, is characterized by poor prognosis and limited therapeutic options. Over recent years, accumulating evidence has shed light on the controversial role of the complement system (C) in different tumour types and contexts, suggesting novel perspectives for therapeutic targeting in selected subgroups of patients. Here, we aimed to determine the prognostic power of C components in MPM by an innovative tool, the newly-defined “Complement Score”, as an evolution of the classical “Immunoscore”. This system may permit to stratify MPM patients by combining bioinformatics, immunohistochemical and survival analysis. Upon bioinformatics analysis, the protein expression of potentially prognostic C components (C1s, C1INH, C1q, CFB, and CFI) was evaluated both on tissue microarrays (n = 88) and whole tissue sections (n = 17) of MPM patients. Univariate analyses were performed to correlate the expression of each C component with MPM histotype, TILs (CD4+, CD8+), Ki-67, PD-L1, and overall survival. Interestingly, survival analysis showed that C1qHIGH (χ2=6.01; p=0.01) and C1INHHIGH (χ2=5.13; p=0.02) patients displayed significantly increased survival. The evidence of a favorable prognostic value of C1q in MPM was apparently inconsistent with our previous findings reporting a tumor-promoting role of C1q. Thus, we attempted to dissect the potential implication of C1q, in concert with HA, in the response to chemotherapeutic treatments. We demonstrated that HA-bound C1q could increase MPM cell mortality after cisplatin treatment, due to increased tumour cell proliferation and downregulation of several drug efflux transporters, including MDR1. Moreover, we detected a strong correlation between patients’ response to chemotherapy and the percentage of in vitro cell mortality after cisplatin treatment in MPM primary cells seeded on HA+C1q matrix (R=0.81, p=0.002). The effect of HA-bound C1q was investigated also in high-grade serous ovarian cancer to determine whether it could be tumor-type specific. Our results confirmed the controversial effect of C1q within the tumour microenvironment. Moreover, we demonstrated that HA-C1q matrix is the optimal culture condition to mimic MPM microenvironment and to predict patients’ response to chemotherapy. The prediction of patient response may allow to move a step towards personalized medicine in several solid tumours.