Caris Life Sciences (Nasdaq:CAI), the Nasdaq-listed Techbio company, has published a study describing an artificial intelligence (AI) tool designed to help doctors choose first-line treatment for pancreatic cancer.
The research, published in npj Precision Oncology, describes the development and validation of the Caris AI Insights model using outcomes from patients receiving FOLFIRINOX or gemcitabine plus nab-paclitaxel.
Patients classified by the model as standard molecular risk who received FOLFIRINOX, the regimen the tool predicted would benefit them most, survived for a median of 16 months, compared with 9.9 months for similar patients given gemcitabine plus nab-paclitaxel.
Around half of patients in the study received a different first-line therapy than the one the model would have recommended.
The signature is included in the Caris Molecular Tumor Board Report, built using whole exome and whole transcriptome sequencing, and is available at no extra cost to clinicians ordering the company's MI Cancer Seek test.
David Spetzler, president of Caris Life Sciences, said the approach aimed to move treatment decisions beyond trial and error by using tumor biology to identify which patients might benefit from more aggressive therapy and which could avoid unnecessary toxicity.
Caris received US Food and Drug Administration approval for MI Cancer Seek in November 2024, describing it as the first and only tissue-based assay combining whole exome and whole transcriptome sequencing with FDA-approved companion diagnostic indications for molecular profiling of solid tumors.