The Marsico Pancreatic Cancer Initiative at the University of Colorado Anschutz is expanding into a broader multi-institutional research effort designed to give scientists a deeper view of pancreatic ductal adenocarcinoma (PDAC). RefinedScience has announced agreements that will provide access to more than 2,000 consented pancreatic cancer samples from academic and industry sources, supporting what the company expects to become the world’s largest multimodal single-cell PDAC dataset by the end of 2026.
The initiative is bringing together biological samples with clinical information collected over time, creating a research resource that can connect molecular changes inside tumors with treatment histories and patient outcomes. The approach is intended to address one of the major challenges in pancreatic cancer research: the disease can behave very differently between patients, making it difficult for conventional datasets to capture its full biological complexity.
For developers of pancreatic cancer therapeutics, the expanded dataset could provide a more detailed basis for studying why some patients respond to treatment while others do not. RefinedScience plans to process and release single-cell multi-omic profiles linked to matched clinical data through consented and institutional review board (IRB)-governed access frameworks. The company said the information will include treatment history and longitudinal patient outcomes.
Multi-Institutional Access Broadens the Research Base
The expansion is anchored by the Marsico Pancreatic Cancer Initiative at CU Anschutz, which was established following a $15 million philanthropic investment from Tom and Cydney Marsico. The initiative was created to apply artificial intelligence and advanced data science to pancreatic cancer research, with the broader objective of identifying biological drivers and supporting the development of more personalized treatment approaches.
RefinedScience already has contractual relationships with CU Anschutz and UCHealth. The company has now added a signed collaboration with the Brenden-Colson Center for Pancreatic Care at Oregon Health & Science University, giving the initiative access to additional patient samples.
The resulting resource is designed to combine several layers of information from the same patients. These include primary human tissue, single-cell sequencing, genomic profiling, spatial transcriptomics and longitudinal clinical data. Rather than examining molecular information separately from clinical history, researchers can study how biological characteristics correspond with treatment decisions and outcomes over time.
That connection could be particularly useful in PDAC research, where identifying meaningful differences between tumors remains a major scientific challenge. Researchers can use the dataset to examine disease mechanisms, investigate potential therapeutic targets and explore biological patterns associated with treatment response and resistance.
The initiative also has potential relevance for biomarker development. Linking molecular profiles with actual patient outcomes may help researchers investigate characteristics associated with response or non-response to specific therapies. Such evidence can be useful for pharmaceutical and diagnostic companies seeking to understand patient populations and potential applications for new products.
“By expanding this dataset across institutions, we are building a richer foundation to understand pancreatic cancer biology and inform drug and diagnostic development,” said Brett Atwood, CEO of RefinedScience.
The multi-institutional structure is also significant because access to well-characterized human biospecimens and associated clinical information can be a limiting factor in translational cancer research. By combining samples from multiple sources under established consent and governance processes, the initiative is seeking to create a broader research population without separating the biological material from the clinical context in which it was collected.
Linking Tumor Biology With Patient Outcomes
The initiative’s expansion comes as researchers and industry participants continue to search for better ways to understand pancreatic cancer progression and treatment resistance. Its dataset is intended to allow researchers to examine individual cells and their surrounding tissue environment while also considering what happened to the patient before and after treatment.
Tom Marsico said the philanthropic investment was intended to support the scientific and discovery process and help change the trajectory of pancreatic cancer. He added that seeing the initiative expand across institutions and putting the resulting data to use for patients was consistent with the original goal of the investment.
Kimberly Muller, vice chancellor for Innovation and Biotechnology at CU Anschutz, highlighted the importance of high-quality biospecimens and clinical data in advancing pancreatic cancer discoveries. She also described RefinedScience as a commercialization partner for single-cell and spatial profiling and welcomed the company’s collaboration with OHSU.
The timing of the announcement also places the initiative in the middle of discussions around pancreatic cancer research. The update comes ahead of scientific and partner meetings at the American Association for Cancer Research (AACR) Special Conference on Pancreatic Cancer, taking place September 25 to 28, 2026, in San Diego.
Among the scientific discussions is a plenary presentation on RAS inhibition by Rosalie “Rosie” Sears, Ph.D., a professor in the Department of Molecular and Medical Genetics at OHSU, co-director of the Brenden-Colson Center for Pancreatic Care and senior member of the OHSU Knight Cancer Institute.
RAS signaling has been an important area of investigation in pancreatic cancer, particularly because of its relationship with tumor development and treatment strategies. The presence of researchers from the participating institutions at the conference provides an opportunity for the initiative and its partners to engage with pharmaceutical, diagnostics and research organizations interested in accessing patient-linked single-cell data.
For RefinedScience, the expanded collaboration also represents a move toward making complex patient-derived datasets more useful to external researchers and developers. The company is inviting discussions with pharmaceutical and diagnostic organizations seeking single-cell information connected to clinical outcomes in PDAC.
As the dataset grows toward its 2026 target, its value will depend not only on the number of samples collected but also on the depth and quality of the information connected to each case. Bringing tissue-level biology, molecular profiling and longitudinal patient outcomes into the same research framework could give investigators a more complete basis for exploring pancreatic cancer and developing the next generation of diagnostics and pancreatic cancer therapeutics.
