Real-world evidence demonstrates PurIST’s ability to guide first-line treatment decisions and improve outcomes for patients with advanced pancreatic ductal adenocarcinoma (PDAC)
Tempus AI, Inc. (NASDAQ: TEM), a technology company leading the adoption of AI to advance precision medicine, today announced the publication of a study in JCO Precision Oncology validating the clinical utility of the company’s PurIST® algorithmic diagnostic. The study provides the largest real-world evidence to date supporting the integration of PurIST into routine clinical care for patients with advanced PDAC, with the aim of informing first-line chemotherapy selection and improving patient outcomes.
Pancreatic cancer remains one of the most lethal malignancies, with limited therapeutic options and a five-year survival rate of just 12%. For patients with advanced, unresectable PDAC, the two most common first-line chemotherapy regimens, FOLFIRINOX (FFX) and gemcitabine plus nab-paclitaxel (GnP), have shown variable efficacy, and clinicians have lacked robust biomarkers to guide optimal therapy selection. To address this challenge, Tempus collaborated with GeneCentric to develop and deploy PurIST, a clinically validated, RNA-based algorithm test that classifies PDAC tumors as either “classical” or “basal” subtypes.
The Tempus-led study analyzed a real-world cohort of 931 patients with advanced PDAC, using the Tempus xR RNA sequencing platform to assign PurIST subtypes. Patients were treated with either first-line FFX or GnP, and clinical outcomes were assessed according to PurIST classification. The study’s findings establish PurIST as both a prognostic and predictive biomarker, enabling clinicians to personalize first-line therapy for advanced PDAC patients and maximize the likelihood of improved survival.
- Prognostic Value: Among patients treated with FFX (N=536), those with the classical subtype had a significantly longer median overall survival (OS) of 11.8 months, compared to 7.0 months for basal subtype patients (Hazard Ratio [HR]=1.86; p<0.001).
- Predictive Value: In patients with the classical subtype and good performance status (ECOG 0 or 1, N=311), treatment with FFX was associated with a 33% relative risk reduction in death compared to GnP (HR=0.67; p<0.009). No comparable benefit was observed in basal subtype patients.
“Pancreatic cancer is a challenging disease, and the ability to match patients to the most effective first-line therapy is critical. Prior research has suggested that certain molecular subtypes are associated with distinct prognoses and responses to therapy, but until now, large-scale real-world validation in a clinical setting has been lacking,” said Ezra Cohen, MD, Chief Medical Officer, Oncology. “Our study demonstrates that PurIST subtyping enables a biomarker-driven approach to therapy selection, addressing a major unmet need in the management of pancreatic cancer.”
“PurIST is a step forward for pancreatic cancer molecular testing, giving healthcare providers a new tool to better inform personalized treatment options for this devastating disease,” said Michael Milburn, PhD, President and CEO of GeneCentric and co-author on the publication. “GeneCentric’s collaboration with Tempus allowed us to validate the algorithm leveraging the power of Tempus’ multimodal data to rapidly deliver a much-needed test for patients with PDAC.”
