A novel oncolytic vaccinia virus armed with IL-21 and 4-1BBL for intravesical treatment of NMIBC: Preclinical efficacy and first-in-human safety evaluation.

198

Background: Oncolytic viruses are emerging cancer immunotherapies. For non–muscle-invasive bladder cancer (NMIBC), intravesical delivery allows high local exposure with limited systemic toxicity. This study was conducted to evaluate anti-tumor efficacy, mechanism, and safety of KM1, a novel developed oncolytic vaccinia virus modified with tyrosine kinase deletion, L025 and A46R gene deletion, and armed with interleukin 21 and 4-1BBL in murine models and humans. Methods: In vitro assays assessed virus replication, cytotoxicity, and transgene expression in human and murine bladder cancer cell lines and organoids. In vivo studies assessed immune activation, toxicity, and antitumor efficacy in murine subcutaneous and orthotopic bladder tumor models. An investigator-initiated phase 1a trial enrolled NMIBC patients who had failed prior BCG therapy or declined further BCG. This multicenter, single-arm study used 3+3 dose-escalation with primary endpoints of safety and tolerability (AE and DLTs). Planned dose levels were 2×10^9 PFU (DL1), 5×10^9 PFU (DL2), and 1×10^10 PFU (DL3) per instillation. Following completion of DL1 and DL2, the protocol was amended to discontinue further dose escalation (DL3 was not opened) and to enroll an expansion cohort at DL1 (additional n = 3) to further characterize safety/tolerability and PK/PD at the putative recommended dose. The induction regimen consisted of bladder irrigation with sucrose monolaurate, followed by intravesical KM1 retained for 2 hours, given twice weekly for 3 weeks and then once weekly for 3 weeks. Subsequent maintenance therapy comprised three intravesical KM1 instillations every 3 months. Results: KM1 selectively replicated in human and murine bladder cancer cells, induced high IL-21 and 4-1BBL secretion. Supernatant from KM1-infected cells activated CD8⁺ T cells and enhanced tumor-killing in vitro. In vivo, KM1 showed better antitumor efficacy than BCG, increased CD8⁺ T and NK cell infiltration and function. Depletion of either population diminished efficacy. In the phase 1a study, 12 patients received intravesical KM1 (DL1, n = 9 [including n = 3 expansion]; DL2, n = 3) and were evaluable. Ten TEAEs were reported, predominantly grade 1–2; the most common were noninfective cystitis (5/12, 41.7%) and urinary tract pain (4/12, 33.3%). One DL1 patient experienced a DLT of transient, likely treatment-related grade 3 liver injury that resolved within 1 week without sequelae. No KM1 DNA was detected in saliva, blood, or feces at any assessed time point. At a median follow-up of 10.1 months, all patients remained recurrence-free by 6 months (CR 12/12; 100%). Conclusions: KM1 showed selective replication, oncolysis, and immune activation in preclinical models. Early clinical data suggest favorable safety and promising antitumor activity of intravesical KM1 in NMIBC. Clinical trial information: ChiCTR2400086709 .