All roads lead to Rome? NetraMark’s AI platform to assess biomarker guided treatment from ROME trial

A new collaboration between a Toronto-based artificial intelligence company and an Italian research foundation could sharpen the case for genomically guided cancer treatment and potentially reshape how payers evaluate precision oncology coverage decisions.

NetraMark Holdings announced last late month that it has entered a strategic research collaboration with Fondazione per la Medicina Personalizzata (FMP) to analyze the ROME Phase II oncology trial dataset using its proprietary NetraAI platform. The stated goal is to identify clinically actionable patient subgroups and biomarker-linked treatment patterns that could inform future precision oncology strategies and clinical trial design.

The ROME trial itself represents one of the most rigorous tests to date of whether matching therapies to a tumor's genomic profile delivered better outcomes than standard approaches. Published in Nature Medicine in September 2025, the multicenter, randomized, open-label study enrolled 1,794 patients with advanced solid tumors across 40 Italian oncology centers between November 2020 and August 2023. Investigators used comprehensive genomic profiling on tissue and blood samples to detect actionable alterations, then convened a molecular tumor board (MTB), a multidisciplinary panel of oncologists, pathologists, geneticists, and bioinformaticians, to guide treatment selection. Of the screened population, 897 patients were evaluated by the MTB and 400 were ultimately randomized to receive either tailored treatment or standard of care.

Patients receiving genomically matched therapy achieved a higher overall response rate (17.5% vs. 10%) and longer median progression-free survival (3.5 vs. 2.8 months). At 12 months, 22% of patients in the tailored arm remained progression-free, compared with just 8.3% in the standard-of-care group. Overall survival was similar between arms, though investigators attributed that largely to a 52% crossover rate from standard care to tailored therapy. Importantly, subgroup analyses showed the strongest benefits among patients with Microsatellite Instability-high tumors, high tumor mutational burden, and BRAF or HER2 alterations.

For managed care stakeholders, the ROME data carry practical implications. The trial provided randomized evidence, not just single-arm or retrospective data, that a tumor-agnostic precision oncology strategy guided by molecular profiling and expert review can meaningfully improve clinical outcomes. That distinction matters as payers continue to grapple with coverage frameworks for next-generation sequencing, tumor-agnostic therapies, and MTB-driven treatment decisions.

The NetraMark collaboration adds another layer. The company's NetraAI platform is designed to uncover what it calls "model-derived subgroups”, concise, interpretable patient characterizations defined by small combinations of variables that may reveal disease structure and treatment-response dynamics not visible in conventional analyses. Applied to the ROME dataset, this approach could help identify which patient populations derive the greatest benefit from genomically guided care and which biomarker profiles most reliably predict treatment response, questions that sit at the center of payer coverage and utilization management discussions.

The collaboration is also expected to strengthen NetraMark's precision oncology analytics capabilities and support future trial design, biomarker development, and enrichment strategies for other sponsors. For FMP, led by ROME principal investigator Paolo Marchetti, M.D., Ph.D., the partnership offers a chance to extract additional clinical value from a dataset that already has produced landmark findings. In NetrMark news release, Marchetti said “We are pleased to collaborate with NetraMark to evaluate how advanced, explainable analytics can help identify patient subgroups most likely to benefit from tailored approaches and support the evolution of precision-oncology decision frameworks.”