Genialis and Inventia Life Sciences agreed to work together to improve how new treatments for pancreatic cancer are tested and matched to the patients most likely to benefit. The companies will combine Genialis’ AI-powered patient insights with Inventia’s RASTRUM 3D cell model platform to build human-relevant models of pancreatic ductal adenocarcinoma (PDAC) that correspond to the tumor biology of identifiable patient populations.

Initial results from this collaboration will be presented this week at the Pancreatic Cancer Action Network (PanCAN)’s 2026 Scientific Summit in San Diego.

Recent advances in targeted therapies for pancreatic cancer have brought renewed attention to PDAC and underscored the need for better strategies to identify which patients are most likely to benefit from new treatments. Inventia’s RASTRUM platform prints reproducible 3D pancreatic cancer models in which tumor and stromal components can be controlled and perturbed independently.

Genialis characterizes those models with the Genialis™ Supermodel—a large molecular model trained on a proprietary harmonized and curated transcriptomics dataset— and connects them to molecular states observed in the clinic. The collaboration also draws on data generated through Genialis’ long standing collaboration with PanCAN.

In the study being presented on Friday, September 25, Genialis identified PDAC patients with poor prognosis based on the composition and activity of the tumor microenvironment across 644 pancreatic cancer patient tumors taken from the PanCAN SPARK “Know Your Tumor” cohort.

Genialis then mapped those patient profiles to tumor-fibroblast co-culture models generated on RASTRUM, allowing researchers to investigate the biology associated with poor outcomes and identify potential therapeutic strategies. This gives drug developers a way to test new approaches in models that reflect the biology of the patients they are intended to treat.

“This collaboration closes an important gap in drug discovery by connecting what we learn from patients with what we can test in the lab,” said Cameron Ferris, PhD, CEO and co-founder of Inventia Life Science. “Genialis can identify clinically meaningful biological states in real tumors, while RASTRUM allows us to model and perturb those states at scale. Together, that creates a powerful way to generate and test therapeutic hypotheses that are grounded in the biology of the patients we ultimately want to treat.”

The design closes a loop that translational programs usually leave open: hypotheses are derived from patient data, tested experimentally in human-relevant systems, and returned to the clinical population to determine where they are most likely to apply, noted Ferris.

“Our pharma partners need to make critical decisions about which patients to enroll and which combinations to pursue, yet historically, whether a preclinical result would translate to patients often wasn’t clear until Phase II,” said Rafael Rosenbarten PhD, CEO and co-founder of Genialis. “Inventia gives us a model where we can test those hypotheses experimentally, while the Supermodel helps identify the patients most likely to benefit.”