The CD70-CD27 axis in cancer immunotherapy: Predictive biomarker and therapeutic target

Abstract The CD27-CD70 interaction is recognized as a positive costimulatory pathway for T cell priming and expansion. However, recent studies showed that chronic CD27-CD70 interaction in cancer can lead to apoptosis of T cells, rendering them dysfunctional. CD70 is expressed not only by hematological tumors but also by solid tumor cells. This expression is regulated by hypoxia-induced factor (HIF), Epstein-Barr virus (EBV) infection, and epithelial-mesenchymal transition. CD27 expression on intratumoral T cells identifies exhausted and dysfunctional T cells, as well as regulatory T cells with enhanced immunosuppressive activity. Given the preferential expression of CD70 on certain tumor cells, several therapeutic approaches including antibody-drug conjugate, anti-CD70 CAR T cells, and anti-CD70 monoclonal antibodies (mAb), have been investigated in various preclinical models and clinical trials. To date, the most significant clinical results are observed in hematological malignancies. However, no therapeutic tools specifically targeting the deleterious CD27-CD70 interaction have been developed. Most CD70-targeting mAb also deplete other CD70-expressing cells, such as activated T cells. Interestingly, chronic CD27-CD70 interaction results in the release of detectable soluble CD27 in patient plasma. The presence of high levels of soluble CD27 in plasma correlates with resistance to anti-PD-(L)1 in renal cancer, melanoma, and non-small cell lung cancer. Conversely, the absence of a predictive impact soluble CD27 in melanoma patients treated with the more toxic combination of anti-PD-1 and anti-CTLA-4 may justify therapeutic escalation with this regimen. Thus, the CD27-CD70 axis may serve as both a potential biomarker to guide the choice of immunotherapy and a novel clinical target.