Prognostic and predictive biomarkers for immunotherapy in advanced renal cell carcinoma

The therapeutic algorithm of renal cell carcinoma has been revolutionized by the approval of immunotherapy agents by regulatory agencies. However, objective and durable responses are still not observed in a large number of patients, and prognostic and predictive biomarkers for immunotherapy response are urgently needed. Prognostic models used in clinical practice are based on clinical and laboratory factors (such as hypercalcaemia, neutrophil count or Karnofsky Performance Status), but, with progress in molecular biology and genome sequencing techniques, new renal cell carcinoma molecular features that might improve disease course and outcomes prediction have been highlighted. An implementation of current models is needed to improve the accuracy of prognosis in the immuno-oncology era. Moreover, several potential biomarkers are currently under evaluation, but effective markers to select patients who might benefit from immunotherapy and to guide therapeutic strategies are still far from validation. The approval of immunotherapy agents to treat patients with metastatic renal cell carcinoma revolutionized the therapeutic landscape for this disease, but a large number of patients still do not experience objective and durable response, and biomarkers to identify response to immunotherapy are urgently needed. In this Review, the authors describe strength and limitations of currently available prognostic and predictive biomarkers in metastatic renal cell carcinoma and discuss emerging promising biomarkers that deserve further investigation. Key points No prognostic and predictive biomarkers of response to immune checkpoint inhibitors (ICIs) are available in clinical practice. Current prognostic models for renal cell carcinoma (RCC) should be improved by integrating novel markers. Programmed cell death 1 ligand 1 (PDL1) expression level is a negative prognostic factor with an unclear predictive role for ICI response in patients with RCC. Tumour mutational burden and gene expression profile are promising predictive factors of response to ICIs that deserve to be further explored. Research is in progress to identify additional biomarkers, which, however, are still far from validation and approval.