Sasanlimab plus bacillus Calmette-Guérin in BCG-naive, high-risk non–muscle-invasive bladder cancer (NMIBC): Exploratory biomarker analysis of the phase 3 CREST trial.

Abstract

806 Background: The primary analysis of the phase 3 CREST trial (NCT04165317), evaluating sasanlimab, a PD-1 inhibitor, combined with BCG induction and maintenance (I+M) showed statistically significant and clinically meaningful improvement in event-free survival (EFS) vs BCG-I+M alone in patients (pts) with BCG-naive, high risk, non-muscle invasive bladder cancer (NMIBC). A trend toward improved EFS was noted in pts with high PD-L1 expression treated with sasanlimab and BCG-I+M. We present additional exploratory analyses examining associations between tumor microenvironment (TME)-related features and response to sasanlimab plus BCG-I+M. Methods: Pre-treatment tumor biopsies were assessed for CD8 + T-cell infiltration by immunohistochemistry (IHC) (N = 609), tumor mutational burden (TMB) by whole exome sequencing (WES) (N = 543), and gene expression by whole transcriptome sequencing (WTS) (N = 534). Exploratory analyses were performed correlating these features with EFS using Cox proportional hazards models. Results: Neither baseline tumor-infiltrating CD8⁺ T cells nor TMB was associated with improved EFS with sasanlimab in combination with BCG I+M versus BCG I+M alone. This observation, coupled with our previously reported PD-L1 data (Powles, ASCO, 2025), suggested that single biomarkers may not adequately reflect the complexity of the TME and we therefore focused on transcriptomic analyses. Transcriptomic profiling revealed that inflammatory and immune signatures (effector and inhibitory signatures) were associated with inferior EFS in the BCG monotherapy arm, but not in the sasanlimab plus BCG-I+M arm. Consistent with these findings, bladder cancer transcriptomic subtypes known to be heavily immune infiltrated, such as UROMOL class 2b and TCGA luminal infiltrated, were associated with poor EFS with BCG-I+M but demonstrated improved EFS with sasanlimab+BCG-I+M (HR = 0.41 [95% CI, 0.23-0.75], P < 0.01; HR = 0.46 [95% CI, 0.25-0.86], P = 0.01 respectively). Importantly, such immunobiological features were present in subsets of tumors across NMIBC stages, underscoring that stage alone does not reflect the biological heterogeneity captured by transcriptomic profiling. Conclusions: These findings demonstrate that pre-treatment NMIBC TMEs characterized by high immune infiltration are associated with poor outcomes with BCG monotherapy, potentially due to BCG promoting further upregulation of inhibitory molecules and adaptive immune resistance. The addition of sasanlimab to BCG-I-M may help overcome this resistance, offering a promising therapeutic strategy for pts with immune-infiltrated high-risk NMIBC. This combination has the potential to address a critical unmet need in the management of high-risk NMIBC, particularly within molecular subgroups associated with poor outcomes with BCG monotherapy. Clinical trial information: NCT04165317 .