Phase I/Ii Study Of Vorolanib Plus Nivolumab In Patients With Thoracic Malignancies: Immunotherapy (Io) Correlatives To Differentiate Responders From Nonresponders.

e21019 Background: VEGF inhibition is suggested to enhance innate T cell function, activate dendritic cells, block recruitment of regulatory T cells, and decrease myeloid-derived suppressor cells. Vorolanib is a tyrosine kinase inhibitor that is structurally similar to sunitinib but designed to improve safety without compromising efficacy. Response to checkpoint inhibitors in patients with thoracic tumors is limited, thus we designed this multi-institutional, phase I/II study (NCT03583086) to explore safety and efficacy of combining vorolanib and nivolumab. Here we present updated results in the NSCLC, thymic, and SCLC cohorts, including patients previously treated with single agent IO. Methods: The Phase I maximum tolerated dose was 200 mg vorolanib and 240 mg nivolumab q2w. Phase II is ongoing and uses a two-stage MinMax design to assess the objective response rate (RR) in SCLC, thymic carcinoma, and three NSCLC cohorts: 1) IO naïve; 2) primary refractory, defined as radiographic progression < 12 weeks of starting IO; and 3) acquired resistance, defined as achieving at least stable disease ≥12 weeks to IO followed by progression. Results: As of December 2019, 37 patients had enrolled, 38% female, 11% never smokers, median age 66. The disease control rate (DCR) across all cohorts is 65% (20/31 evaluable patients) and the RR is 13% (4/31). In the NSCLC cohort, the overall RR is 14% (3/21); the RR is 33% (2/6) in IO naïve patients, including 2 on therapy > 17 months. RR in the NSCLC patients with acquired resistance is 14% (1/7) with an 86% (6/7) DCR. No responses in the primary refractory cohort but the DCR is 50% (4/8). DCR in the SCLC and thymic cohorts is 40% (2/5) and 50% (2/4), respectively; 1 thymic patient had a partial response with a complete response in target lesions. Exploratory correlatives, including mass cytometry analysis and single cell gene expression, are being performed in order to understand changes in macrophages and T cell activation markers to differentiate the underlying biology of the tumor and the microenvironment between responders and nonresponders. Conclusions: The addition of vorolanib nearly doubled the RR in the IO naïve NSCLC patients compared to historical data with single agent nivolumab. Additionally, 67% of NSCLC patients with prior IO achieved at least stable disease, 1 patient with acquired resistance had a response. Correlatives that may help define biomarkers of response to therapy will be presented among different cohorts of patients. Clinical trial information: NCT03583086 .