Efficacy, safety, and biomarker analysis of neoadjuvant avelumab/axitinib in patients (pts) with localized renal cell carcinoma (RCC) who are at high risk of relapse after nephrectomy (NeoAvAx).

289 Background: Antibodies targeting PD-1/PD-L1 combined with vascular endothelial growth factor (VEGF) inhibitors are a first-line standard of care for metastatic RCC. Neoadjuvant use of these combinations may lead to downstaging and reduce the risk of recurrence. In addition, sequential tissue may allow identification of key immune biomarkers associated with outcome. Methods: Neoavax is a single arm phase II trial of 12 weeks neoadjuvant avelumab/axitinib prior to nephrectomy in 40 pts with high-risk non-metastatic clear-cell (cc) RCC (cT1b-4cN0-1M0, Grades 3-4). Primary endpoint is RECIST 1.1 partial response (PR) in the primary tumour (PT) in ≥25%. Secondary endpoints are disease-free survival (DFS), overall survival (OS) and safety. Biomarker analysis on sequential tissue is an exploratory endpoint. Expression of PD-L1 (SP263), CD8+, CD8-granzyme-B (CD8/GZMB)+, Foxp3+ cells, CD8/CD39+ and MHC-I were compared on pre-treatment biopsy and nephrectomy samples from 34 pts (NCT03341845). Results: Pts/tumour characteristics are shown in table. Twelve pts (30%) had a PR of the PT from a baseline mean diameter of 10.3 (range 5.6-16.4) cm. Median PT downsizing was 20 (0-43.5) % and median post-treatment vital tumour presence was 50 (1-100) %. At a median follow-up of 23.5 months, recurrence occurred in 13 (32%) pts at a median of 8 (2-23) months and 3 died of disease. Of the 12 pts with PT PR, 11 (92%) are disease-free. Median DFS and OS are not reached. Postoperative adverse events occurred in 8 pts (2 Clavien Dindo grade 3a). There were no treatment-related surgery delays and no PT progression. Post-treatment samples showed upregulation of PD-L1 expression (p <0.0001) and total CD8+ densities (p < 0.01) when compared to pre-treatment biopsies. Comparing samples of pts with PR vs no PR in the PT, no clear immune marker differences were observed. Post-treatment samples from pts that recurred were characterized by lower densities of total, intra-epithelial and stromal CD8+, intra-epithelial CD8+CD39+ (p<0.05) and total CD8+GZMB+ (p=0.1). Pre-treatment biopsies showed no clear differences. Conclusions: Neoadjuvant avelumab/axitinib for non-metastatic high-risk RCC leads to PR of the PT in 30% which is associated with DFS. Pts without recurrence had a significant increase in CD8+ densities compared to pts with recurrence suggesting expansion of a pre-existing immune response. Clinical trial information: NCT03341845. [Table: see text]