SIM0237 Bladder Cancer Data and New PD-1/PD-L1 Biomarker Validations

Researchers presented and published multiple studies advancing PD-1/PD-L1-targeted immunotherapy, including promising phase 1/2 data for the anti-PD-L1/IL-15 variant fusion protein SIM0237 in BCG-unresponsive non-muscle invasive bladder cancer and new validation studies of biomarkers for treatment selection and response monitoring.

At the 2026 European Association of Urology (EAU) annual meeting, investigators reported results from the dose escalation and expansion parts of a phase 1/2 study of intravesical SIM0237 monotherapy in 49 patients with BCG-unresponsive high-risk non-muscle invasive bladder cancer (14 with carcinoma in situ [CIS] with or without Ta/T1, and 35 with papillary-only disease). With a median follow-up of 6.7 months, 8 of 10 CIS patients with at least one post-baseline tumor assessment achieved a best response of complete response, with a probability of duration of complete response at 12 months of 71.4%. In the 35 papillary-only patients, the 12-month disease-free survival rate was 65.8%. The treatment was safe and well tolerated: treatment-related adverse events occurred in 51.0% of patients, were mostly grade 1/2 and limited to the urinary system, with one grade 3+ treatment-related adverse event and no dose-limiting toxicities, immune-related adverse events, or adverse events leading to discontinuation. Systemic exposure of SIM0237 was undetectable in all 29 patients with serum samples analyzed. A phase 3 study is being planned.

In a separate study published in npj Precision Oncology, Adela reported clinical validation of its genome-wide methylated circulating tumor DNA (ctDNA) test for monitoring response to immunotherapy in advanced solid tumors. In banked samples from 64 patients with advanced head and neck, breast, ovarian, melanoma, or other solid tumors who received pembrolizumab, a decrease in methylated ctDNA from pre-treatment to before cycle 3 was associated with a significantly higher objective response (odds ratio [OR] 31.77) and clinical benefit (OR 15.55), as well as improved progression-free survival (hazard ratio [HR] 0.27) and overall survival (HR 0.49). The company plans to commercialize the test later this year.

Another validation study reported that a seven-gene predictive signature (CD86, CHI3L1, CXCL10, CXCL9, LAG3, NR4A1, and VCAM1) for the efficacy of immune checkpoint inhibitors in sarcomas stratified patients into significantly different progression-free survival groups. Using RNA sequencing data from pretreatment tumor samples from the SARC028 and GEIS-32 phase II trials, patients with scores of 0–4 had a median PFS of 49 days, versus 170 days for those with scores of 5–7 (HR 0.71, P = 0.007).

Investigators also presented research validating an AI-assisted PD-L1 scoring algorithm against original Blueprint study data in lung cancer. The AI platform achieved performance at least equivalent to manual assessment by expert pathologists and, in certain scenarios, exceeded it, with the most pronounced improvement seen in cases involving the SP142 assay. The study used consensus scores from 24 experienced pathologists as the reference standard.

A study published in Nature described INSPIRE, a machine learning approach that uses representation learning on real-world data to rank histology-based malignant indications for which a chosen mechanism of action is likely to be effective. When restricted to data before broad establishment of PD-1 inhibitors in the clinic, the method successfully prioritized 70% of subsequent approvals.