Document resource
Background Vilastobart is a tumor-activated, Fc-enhanced high-affinity anti-CTLA-4 designed to focus activity toward tumors while minimizing systemic exposure. We previously reported vilastobart combined with atezolizumab (Tecentriq®) had a 26% overall response rate (ORR) in heavily pretreated patients with MSS mCRC without liver metastases (NLM), and a favorable safety profile consistent with tumor-activated design. 1 High tumor mutational burden (TMB) is a biomarker predictive of response to immune checkpoint inhibitors (ICI) and is an approved tissue-based companion diagnostic for pembrolizumab in advanced solid tumors. In MSS mCRC, TMB from tumor tissue (tTMB) provided no predictive utility to ICI; however, a plasma-based assay showed high TMB improved survival with ICI (N=163), and pTMB was significantly higher than tTMB with a majority of patients with pTMB ≥10mut/Mb.2 We explored whether plasma TMB (pTMB) could serve as a predictive biomarker in NLM patients with MSS mCRC and enrich for response to treatment with vilastobart in combination with atezolizumab.Methods This Phase 2 study ( NCT04896697) evaluated vilastobart (100mg IV-Q6W) in combination with atezolizumab (1200mg IV-Q3W) in patients with MSS mCRC who had ≥1 prior chemotherapy regimen in the metastatic setting. Baseline pTMB was assessed retrospectively using the Guardant-Infinity Liquid Assay. We evaluated the relationship between pTMB status (pTMB-High[H]≥10mut/Mb; pTMB-Low[L]<10mut/Mb) and best overall response using 2-sided Fisher’s Exact Test.Results As of 12 May 2025, 44 heavily pretreated patients with MSS mCRC were treated with vilastobart in combination with atezolizumab, including 27 NLM. Twenty-four NLM patients had sufficient ctDNA to determine pTMB status. A TMB threshold of ≥10mut/Mb was selected as this captured all responders with known pTMB status. Overall, 63% (n=15) of TMB-evaluable patients were pTMB-H. In this pTMB-H population, the ORR was 40% (95% CI:16%, 68%), with 5 of 6 responses confirmed, while the ORR in TMB-L patients (n=9) was 0% ( figure 1, p=0.05). One additional confirmed responder was non-evaluable for pTMB status. Responses were deep and durable (figure 2). Consistent with the tumor-activated design of vilastobart, treatment was generally well-tolerated: overall (n=44), 11 patients (25%) required immunosuppression for immune-related adverse events; 3 (7%) experienced colitis; 2 (5%) discontinued treatment due to a related event.Conclusions Using pTMB status as a predictive biomarker, heavily pretreated patients with MSS mCRC who were pTMB-H had a 40% ORR following treatment with vilastobart in combination with atezolizumab. These encouraging data suggest pTMB is a promising biomarker approach that may enrich for response in a disease setting with a significant unmet need.Trial Registration NCT04896697References Davar D, Fakih M, Bekaii-Saab TS, DeVito NC, Knecht JG, Vandross AL, Perez CA, Bessudo A, Gupta A, Patel E, Fantini D, Paramasivan S, Crowe D, Duncan M, McConnell S, Luptakova K, Parikh AR. Phase 1/2 study of XTX101, a tumor-activated, Fc-enhanced anti-CTLA-4 monoclonal antibody, in combination with atezolizumab in patients with advanced solid tumors and in MSS CRC. Journal of Clinical Oncology 2025;43(4_suppl):206–206. https://ascopubs.org/doi/10.1200/JCO.2025.43.4_suppl.206Loree JM, et al. Plasma versus tissue tumor mutational burden as biomarkers of durvalumab plus tremelimumab response in patients with metastatic colorectal cancer in the CO.26 trial. Clin Cancer Res. 2024;30(15):3189–3199. https://pubmed.ncbi.nlm.nih.gov/38727700/Ethics Approval This study was approved by Advarra Institutional Review Board, approval number Pro00054400. All study participants gave informed consent before taking part in the study.Abstract 1315 Figure 1pTMB correlated with response following treatment with vilastobart in combination with atezolizumab: All NLM patients. (A) Waterfall plot showing pTMB status for all NLM patients who had target lesion measurements available (24/27). Note: Of 2 patients with 0% best change, one patient was TMB-Low, and the other was non-evaluable for TMB status. (B) Boxplot showing pTMB scores at baseline with respect to best overall response. TMB status was available for 24/27 NLM patients; 3 were non-evaluable for TMB status. pTMB=plasma tumor mutational burden; NLM=non liver metastases; PR=partial response.Abstract 1315 Figure 2Deep tumor reductions and durable responses observed in pTMB-H patients treated with vilastobart in combination with atezolizumab. Figures showing all pTMB-high (≥10 mut/Mb) NLM patients who had target lesion measurements available (12/15). pTMB=Plasma Tumor Mutation Burden; NLM=non liver metastases; PR=partial response