Immunomodulatory activity of nivolumab in previously treated and untreated metastatic renal cell carcinoma (mRCC): Biomarker-based results from a randomized clinical trial.

5012 Background: Nivolumab, a fully human IgG4 programmed death-1 (PD-1) inhibitor antibody, has shown encouraging activity in mRCC. This trial assessed the immunomodulatory and clinical activity, and safety of nivolumab in patients (pts) with mRCC. Methods: Ninety-one pts received nivolumab IV Q3W: pretreated pts (1–3 prior therapies; ≥1 anti-angiogenic agent) received 0.3 (n=22), 2 (n=22), or 10 mg/kg (n=23); 24 treatment-naïve pts received 10 mg/kg. Fresh biopsies and serum were obtained at baseline (BL) and nivolumab cycle 2 day 8 (C2D8; biopsy) and cycle 4 day 1 (C4D1; serum). Primary objective was to assess the immunomodulatory activity of nivolumab on serum chemokines (CXCL9, CXCL10) and tumor T cell infiltrates from BL to post treatment. Secondary/exploratory objectives included safety and tolerability, antitumor activity (ORR; RECIST 1.1), BL and treatment-induced changes in PD-1 ligand (PD-L1) expression (Dako immunohistochemistry assay; PD-L1 positivity: >5% tumor membrane staining at any intensity) and association of clinical activity with BL PD-L1 expression. Results: Mean increase from BL to C4D1 was 191% for CXCL9 and 90% for CXCL10. T cell infiltrates increased by a median of 70% (CD3+; range 53–220%) and 88% (CD8+; 61–257%) from BL to C2D8. Of 56 evaluable fresh pretreatment biopsies, 18 (32%) were PD-L1+. ORR was 22% (4/18) for PD-L1+ pts vs 8% (3/38) for PD-L1–. In 5/27 (19%) matched biopsy pairs PD-L1 expression increased >5% by C2D8. For evaluable pts ORR was 16% (14/90); 16% in previously treated pts, 13% in untreated pts. Median duration of response was 15 months; 6 (43%) responses are ongoing. 14/91 (15%) pts had grade 3–4 related AEs, most commonly colitis and elevated AST (n=2 each), diarrhea and pneumonitis (n=1 each), all grade 3. Conclusions: In this prospective biomarker-based study, nivolumab showed clinical activity and manageable safety in pts with previously treated and untreated mRCC. Responses were numerically higher in PD-L1+ pts but also seen in PD-L1– pts. Changes in biomarkers were consistent with PD-1 inhibition and provided evidence of immunomodulatory effects in serum and in the tumor microenvironment. Clinical trial information: NCT01358721.