Alicia Zhou, Chief Executive Officer of the Cancer Research Institute, shared a post on LinkedIn:
“For more than seven decades, the Cancer Research Institute (CRI) has helped propel immunotherapy from an idea on the fringe to a pillar of modern cancer care. Along the way, we’ve learned an essential truth: progress accelerates when we share what we know, and stalls when we don’t.
Today, I’m proud to share a milestone that reflects that same progress, in action. We have launched the CRI Discovery Engine, a first-of-its-kind, open, AI-ready database designed specifically to advance cancer immunotherapy. It’s more than a repository. It’s a shared foundation for understanding how immune cells respond to treatment over time, in space, and at cell-level resolution.
Why does this matter now? Because our field has reached an inflection point. We finally have the technologies to study immunotherapy and immune responses as dynamic, living systems — but our data practices haven’t kept pace. Too much critical information remains siloed, proprietary, or inaccessible. Fewer than half of high-impact cancer studies can be reproduced, and only about 1% are standardized in a way that allows other scientists to actually use them.
The CRI Discovery Engine is built to change that.
A Shared Answer to Shared Problems
At its core, the Discovery Engine systematically captures how both immune cells and cancer cells respond to immunotherapy interventions over time and in space. The data are generated using standardized, reproducible protocols, and optimized for AI and machine learning — so that researchers can ask better questions and get answers faster, without starting from scratch.
This effort is a true collaboration. We’re building the CRI Discovery Engine alongside Stanford University, the University of Pennsylvania Perelman School of Medicine, and Memorial Sloan Kettering Cancer Center, with technology support from 10x Genomics. The initiative is led by three extraordinary principal investigators: CRI STARs Andrea Schietinger, and Ansuman Satpathy, and CRI Scientific Advisory Council Associate Director E. John Wherry.
Dr. Wherry captured the heart of this collaborative effort when he said, ‘One of the biggest challenges in academic research is that we work in silos. There’s competition and proprietary knowledge that institutions feel they need to protect. But that approach slows everyone down.’
The CRI Discovery Engine represents a collective decision to move faster together — because patients can’t afford for us to do otherwise.
Designed For Reality, Not Just Success Stories
The initial phase of the CRI Discovery Engine focuses on melanoma and colorectal cancer: two diseases where immunotherapy has transformed outcomes for some patients, yet still falls short for many others.
Importantly, the dataset will include not only treatments that worked, but those that didn’t.
That choice is intentional. Negative and null results are essential for understanding resistance, refining hypotheses, and avoiding costly dead ends — but they’re rarely shared. By including these data, we aim to illuminate not just what works, but why it works, why it fails, and how to design what comes next.
As Dr. Satpathy put it, ‘Someday we’ll look back on this as a turning point for immunotherapy. By building a shared, high-resolution understanding of how the human immune system responds to interventions over time, we’re unlocking a new era of discovery — one that shows us why treatments work, why they fail, and how to design what comes next.’
Built For Reproducibility – Ready For AI
The CRI Discovery Engine is purpose-built to address the reproducibility crisis head-on. Standardized experimental design and consistent controls mean results can be reliably reproduced — no matter who runs the experiment or where the bench is.
It’s also designed for the future. Large, well-annotated, harmonized datasets are the fuel that AI and machine learning require. By making these data accessible in an AI-ready format, we’re making it possible for researchers to identify patterns, to generate insights, and to jumpstart years of scientific rigor.
The technology is ready. Now the data can be, too.
A Living Resource for a Global Community
We will seed the CRI Discovery Engine with our own research, and more scientists will be able to contribute over time. Together, we are creating a living resource that grows more valuable with every addition. The initial dataset will be made publicly available by the end of this year.
The challenges facing cancer research are real. Federal funding is under threat. Public trust in science is strained. This moment calls for urgency, collaboration, and courage.
Cancer doesn’t care about institutional egos, proprietary data, or who gets credit. Neither do we.
The CRI Discovery Engine is our commitment to a different path forward: one grounded in shared data, illuminated with shared purpose, and driven by the belief that when we build the right foundations, discovery always follows.
Behind every data point is a patient who wants more time. This is how we intend to give it to them.”
You can also read: CRI Discovery Engine Inaugurates New Chapter in Immunotherapy, Creating the First-of-its-Kind AI-Ready Database
