Correlative serum biomarker analyses in the phase 2 trial of lenvatinib-plus-everolimus in patients with metastatic renal cell carcinoma.

BACKGROUND No biomarkers have been established to predict treatment efficacy in renal cell carcinoma (RCC). In an exploratory retrospective analysis of a Phase 2 study, we constructed composite biomarker scores (CBSs) to predict progression-free survival (PFS) and overall survival (OS) in patients with metastatic RCC randomised to receive lenvatinib-plus-everolimus. METHODS Of 40 biomarkers tested, the 5 most strongly associated with PFS (HGF, MIG, IL-18BP, IL-18, ANG-2) or OS (TIMP-1, M-CSF, IL-18BP, ANG-2, VEGF) were used to make a 5-factor PFS-CBS or OS-CBS, respectively. A 2-factor CBS was generated with biomarkers common to PFS-CBS and OS-CBS. Patients were divided into groups accordingly (5-factor-CBS high: 3-5, CBS-low: 0-2; 2-factor-CBS high: 1-2, CBS-low: 0). RESULTS PFS/OS with lenvatinib-plus-everolimus were significantly longer in the 5-factor CBS-high group versus the CBS-low group (P = 0.0022/P < 0.0001, respectively). In the CBS-high group, PFS/OS were significantly longer with lenvatinib-plus-everolimus versus everolimus (P < 0.001/P = 0.0079, respectively); PFS was also significantly longer with lenvatinib-plus-everolimus versus lenvatinib (P = 0.0046). The 5-factor-CBS had a predictive role in PFS and OS after multivariate analysis. Similar trends were observed with the 2-factor-CBS for PFS (i.e., lenvatinib-plus-everolimus versus everolimus). CONCLUSIONS The 5-factor CBS may identify patients with metastatic RCC who would benefit from lenvatinib-plus-everolimus versus everolimus; additional validation is required. CLINICAL TRIAL REGISTRATION The clinical trial registration number is NCT01136733.