Complement component C1q can predict patients’ response to cisplatin depending on tumour type: A step towards personalized medicine?

Background: C1q can exert complement-independent pro-tumorigenic functions [1], also by virtue of its binding to hyaluronic acid (HA). HA was previously demonstrated to influence tumor cells’ sensitivity to cisplatin [2]. Thus, we attempted to dissect the potential implication of HA-bound C1q in chemoresistance, evaluating its effects on drug sensitivity in both malignant pleural mesothelioma (MPM) and high-grade serous ovarian cancer (HGSOC). Methods: HA-C1q effect on in vitro response to chemotherapeutic agents was evaluated in MPM (H28, M14K and ZL34) and HGSOC (SKOV-3) cell lines. Its modulation of ABC transporters and detoxification enzymes was evaluated by qPCR, Western blot and functional assays. Isolated primary MPM (n=24) and HGSOC (n = 26) cells were also tested for cisplatin cytotoxicity. Patients’ clinical outcomes in terms of radiological response were correlated with in vitrocytotoxicity. Results: HA-C1q matrix significantly increased MPM cell mortality after cisplatin and PARP inhibitor treatments, restoring chemosensitivity also in a cisplatin-resistant MPM cell line (H28R). HA-C1q downregulated the expression and functionality of drug efflux transporters, comprising MDR1, and promoted cell proliferation. In primary MPM cells, we detected a strong correlation (R = 0.81, p = 0.002) between the percentage of in vitro cell mortality after cisplatin treatment on HA-C1q matrix and patients’ response to chemotherapy. Interestingly, no effect of HA-C1q matrix was observed in HGSOC cells. Conclusions: C1q can specifically increase MPM cell sensitivity to cisplatin treatment, due to drug transporters’ downregulation and increased cell proliferation. Moreover, we demonstrated that C1q-HA matrix is the optimal culture condition to mimic MPM microenvironment and to predict patients’ response to chemotherapy. HA-bound C1q exerted no effect on HGSOC cells, highlighting a tumor-type-dependent specificity. Our results confirm the contro- versial effect of C1q within the tumor microenvironment and allow to set up an easy testing method which may be applied in the future not only as a predictive tool for the individual patient but also as a translational model to test novel therapeutic targets. The prediction of patient response may allow to move a step towards the development of personalized medicine in different solid tumors.