AACR Project GENIE Biopharma Collaborative Releases Colorectal Cancer Dataset

The American Association for Cancer Research (AACR) Project GENIE Biopharma Collaborative (BPC) has released a colorectal cancer dataset that includes genomic and clinical information that provides a fuller picture of a patient's journey, including tumor development and treatment responses. This will give researchers and clinicians more data to help understand how cancer patients’ genetics impact the course of their disease.

The AACR Project Genomics Evidence Neoplasia Information Exchange (GENIE) is a registry of real-world genomic and clinical cancer data assembled via data sharing among 18 international academic cancer centers. Every 6 months, new data are added to the registry.

To provide a rich resource to the oncology community, AACR Project GENIE launched the BPC in 2019 to add deep clinical annotation, including treatment data and tumor pathology, to a subset of approximately 20,000 de-identified patients from the main GENIE registry (Cancer Discov 2020;10:OF2). The colorectal cancer dataset is the second to be made publicly available through the BPC (see www.aacr.org/professionals/research/aacr-project-genie). A dataset for non–small cell lung cancer (NSCLC) was released in May. The BPC subcommittee determined the sequence in which the datasets would be released. Up next: breast cancer.

“Project GENIE has collected extensive genomic data from more than a hundred thousand patients, but we have lacked some of the very basic clinical data and outcomes information that we need to ask clinically relevant research questions. With the BPC, we home in on a smaller subset of the dataset to get some of those details,” says Kenneth Kehl, MD, MPH, of Dana-Farber Cancer Institute (DFCI) in Boston, MA, and co–principal investigator of the BPC.

Ten pharmaceutical companies—Amgen, AstraZeneca, Bayer HealthCare Pharmaceuticals, Boehringer Ingelheim, Bristol Myers Squibb, Genentech, Janssen Research & Development, Merck, Novartis, and Pfizer—are participating in this 5-year collaboration. As part of their commitment, the companies are collectively contributing $36 million and providing technical expertise.

“As an industry, we all have the need for reliable, quality real-world data that is truly reflective of the patient treatment pathway. So, I think that basic need, no matter what the research question is, drives us to contribute to how we can get that right,” says Karma Rabon-Stith, PhD, MBA, who leads the Global Oncology Real World Evidence team for Pfizer.

The colorectal cancer dataset includes 1,485 patients with colorectal cancer from DFCI; Memorial Sloan Kettering Cancer Center in New York, NY; and Vanderbilt-Ingram Cancer Center in Nashville, TN. As with the NSCLC dataset, the colorectal cancer dataset includes prior cancer treatments, tumor pathology, and clinical outcomes.

Kehl says these datasets will help answer questions that require retroactively exploring how biomarkers influence the effect of treatment. For example, the BPC collects CT, MRI, and PET-CT scan data from diagnosis through death or the patient's last follow-up, indicating whether the cancer was stable, responding to treatment, or progressing. Initiatives are also underway to determine molecular features that might correlate with sites of metastatic disease.

“In general, with some of the datasets that are not part of the BPC, you may have that data on the site of the biopsy, but not necessarily the longitudinal trajectory of metastatic disease sites for the patient,” explains Kehl.

The participating pharmaceutical collaborators are helping to spread the word among investigators about the opportunity to use the data and connecting them with the BPC where appropriate. In addition, they provide guidance on the development of current and future datasets. All BPC datasets are made public 1 year following initial release to the biopharma companies.

“The BPC is a unique opportunity for companies to come together in a collaborative spirit and engage in conversations about making a difference in the research to benefit patients. Collectively, we strive to curate the most reliable and informative real-world data about as many tumor types as possible and deliver it to the broader research community,” notes Michael Paskow, MPH, associate director of epidemiology for the Global Real World Evidence Generation–Oncology team at AstraZeneca.

The colorectal cancer and NSCLC datasets are part of the first phase of the BPC's upgrade of AACR Project GENIE's clinical datasets from common cancers—including bladder, breast, pancreatic, and prostate—already linked with the genomic profiles. Phase 2 will cover additional cancer types, and it will provide updates to and add additional patients to the current NSCLC and colorectal cancer cohorts.

“The overall goal is precision oncology, which has been the goal of Project GENIE all along. We're going to continue working to develop strategies to deliver the optimal treatment to the right patients with the right kind of tumor,” says Kehl. –Aaron Tallent