Neoadjuvant lenvatinib plus pembrolizumab for resectable clear-cell renal cell carcinoma (PELUR): A prospective phase 2 study.

4541 Background: Lenvatinib plus pembrolizumab prolongs overall survival (OS) and progression-free survival (PFS) in advanced clear-cell renal cell carcinoma (ccRCC), with significantly improved objective response rate (ORR) and survival time compared with other competitors. However, more than 80% ≥Grade 3 adverse events (AEs) were emerged during treatment. The efficacy and safety of neoadjuvant low-dose lenvatinib plus pembrolizumab (ldLP) in ccRCC at high-risk of progression has not been assessed. Methods: This was an open-label phase 2 clinical trial including patients with resectable high-risk ccRCC who received neoadjuvant ldLP every 21 days for 3 cycles. Tumor responses and safety were both primary end points. The secondary end points were PFS, patient-reported quality-of-life and immune biomarkers. RNA and DNA were isolated from pretreatment tumor tissue was subjected to RNA and next-generation sequencings. Single-cell RNA sequencing (scRNA-seq) was performed in both pretreatment and posttreatment specimens from 6 patients. Results: A total of 33 patients were enrolled, 23 received neoadjuvant therapy followed by nephrectomy were included in the intention-to-treat (ITT) analysis. All patients received neoadjuvant therapy follow the dose of protocol. There was only one grade 3 AE (hypertension) emerged during neoadjuvant therapy. The most common AEs of neoadjuvant treatment were hypertension, fatigue, rash and pruritus (n = 5, 21.7%). During adjuvant stage, three grade 3 AEs were reported, including one case of rash, ALT/AST increase and acute kidney injury. The most common AEs during adjuvant were rash, fatigue and pruritus (n = 6, 26.1%). The EORTC QLQ-C30 questionnaires showed significant improvements in all emotional function and symptom of appetite loss. Total score of FKSI-DRS was also significantly improved. Tumor and thrombus regression occurred in all patients after neoadjuvant therapy, with 11/23 (47.8%) of them got partial response. After a median follow-up of 22 months (15-35 months), five patients experienced disease progression and 2 patients had ccRCC-related died. We characterized ~11500 single cells from 6 patients (12 samples), which were categorized into partial response (PR; n = 3) and stable disease (SD; n = 3). Our analysis revealed that the ARPP21 + /IGLL1 + B cell subcluster (AI + B cells) demonstrated the most substantial cellular perturbation within SD group. Furthermore, we observed AI + B cells experienced a significant reduction in PR group following treatment. The AI + B cells were predicted to interact with DCs to contribute to a poor therapy response. Conclusions: Our data preliminarily demonstrated safety and efficacy of neoadjuvant ldLP in ccRCC at high-risk of progression. Results also highlight the importance of AI + B cells in effective responses to ldLP and suggests potential strategies to overcome immunotherapy resistance. Clinical trial information: NCT05485896 .