BOSTON — Genialis said it will present new data at the American Association for Cancer Research Annual Meeting 2026 demonstrating that its artificial intelligence-driven model can better predict which breast cancer patients will benefit from the HER2-targeted therapy Enhertu.
The company’s Genialis Supermodel-powered RNA survival model was evaluated using real-world clinical data from Tempus, including 90 patients across HER2-positive, HER2-low, and HER2-ultra-low breast cancer groups. According to the company, the model showed statistically significant performance that outperformed standard HER2 diagnostic methods.
Antibody-drug conjugates, or ADCs, have become a rapidly expanding class of cancer therapies, but selecting the right patients remains a challenge. Current diagnostic approaches, such as immunohistochemistry and fluorescence in situ hybridization testing, focus on measuring HER2 expression and do not fully capture the biological factors that influence treatment response.
Genialis said its Supermodel, trained on large-scale RNA sequencing data, analyzes tumor gene expression and translates it into biological signals tied to treatment response. The model incorporates factors such as drug targeting, cellular internalization, and payload activity, which are not reflected in conventional testing.
In the study, patients identified by the model as likely to benefit from Enhertu had a median treatment duration of 345 days, compared with 245 days for those predicted to have less benefit, representing a 41 percent difference. The model also showed no predictive signal in control groups, indicating it is specific to treatment response rather than overall prognosis.
“Better biomarkers for HER2-directed ADCs are urgently needed as current diagnostics were designed to measure receptor expression, not the full biology of drug response,” said Mark Uhlik, Ph.D., chief scientific officer of Genialis. “Enhertu response involves multiple biological processes beyond HER2 expression, including internalization, payload activity, and the tumor’s damage response. These aren’t captured by IHC. The Supermodel is designed to represent those processes using comprehensive RNA data, which helps explain the signal we’re presenting at AACR.”
Company officials said the model’s predictive signals are tied to broader biological processes, including DNA damage response, tumor stress pathways, hypoxia, and drug payload activity, highlighting the limitations of receptor-based diagnostics alone.
With more than 1,000 ADC programs currently in development, Genialis said there is increasing demand for more advanced biomarker tools to guide drug development and patient selection. The company is applying its modeling approach across multiple ADC programs in collaboration with pharmaceutical and biotechnology partners.
“HER2 IHC tells you what a tumor expresses. The Supermodel tells you the biology present in a tumor and predicts what will happen when you treat it. That is a fundamentally different kind of information, and it is what drug developers need to make smarter decisions earlier in ADC development,” said Rafael Rosengarten, Ph.D., chief executive officer of Genialis. “This Enhertu predictor is another example of the Supermodel’s broad application across the ADC pipeline, and we are rapidly extending this work with our pharma and biotech partners.”
