The formation of the Precision Cancer Medicine Initiative (PCMI) two years ago — a partnership between Duke Cancer Institute, the Department of Pathology, and Duke University Health System Clinical Laboratories — has made it possible to maximize the organization, availability, and interoperability of Duke cancer patients’ clinical and tumor genomic information for the benefit of cancer care and research into new treatments. It boasts an increasingly active weekly Molecular Tumor Board and a growing Molecular Registry of Tumors.
Last year Duke’s leadership in this field was recognized when Duke Cancer Institute was invited to join the elite international American Association for Cancer Research Project GENIE (Genomics, Evidence, Neoplasia, Information, Exchange) consortium, whose registry contains clinical-grade cancer genomic sequencing data from nearly 71,000 patients with 80 major cancer types — among the largest publicly available genomic datasets released to date. These data include age, sex, primary diagnosis, and type of cancer analyzed. There’s data from nearly 11,000 patients with lung cancer, greater than 9,700 patients with breast cancer, and nearly 7,000 patients with colorectal cancer.
Pathologist Shannon McCall, MD, leader of PCMI Research Initiatives, director of the Duke BioRepository & Precision Pathology Center (BRPC), and the Duke site principal investigator for Project GENIE, was involved in founding the Precision Cancer Medicine Initiative two years ago and encouraging DCI to support it.
“I was totally on fire to get this started,” she recalled. “We have so many big thinkers at Duke who say let’s think about data and what’s possible, but what we’ve done here with the Molecular Tumor Board and joining Project GENIE that I’m so proud of is that this group has really dug in … We’ve made a real difference.”
McCall gives enormous credit to Michael Datto, MD, PhD, medical director for DUHS Clinical Laboratories and director of the Duke University Health System Clinical Molecular Diagnostics Laboratory, and DUHS clinical informatics architect Christopher Hubbard, who designed, built and coded DCI’s in-house secure and searchable electronic Frameshift Molecular Registry of Tumors (or Frameshift MRT) that now stores clinical-grade cancer genomic sequencing data from more than 3,900 patients, across more than 40 cancers. (In general, metastatic lung cancer patients are the biggest beneficiaries of comprehensive molecular profiling tests. That’s because targeted therapies already exist for many of the most common molecular alterations seen in these patients. Metastatic breast cancer patients are the next largest group, followed by metastatic colorectal cancer patients. Even in cancer types where the possibility of a match is low, the tests may still be worth doing.)
“The GENIE thing didn’t happen by itself; we had to build the data infrastructure to make it happen,” said McCall, adding that DCI director of Bioinformatics, Kouros Owzar, PhD, and programmer Jeremy Gresham consulted with her and Department of Pathology programmer Michael Fox to ensure the Precision Cancer Medicine Initiative team was on the right track to upload Frameshift MRT data to AACR GENIE.
Duke Cancer Institute has so far contributed de-identified genomic sequencing data from 1,000 patients to the Project GENIE database.
As a member of the GENIE Consortium, DCI investigators benefit from early access to GENIE data. Data doesn’t become publicly available until six months after submission, but it’s immediately available to consortium members.
Jonathan Bell, MMCi, MEng, a data scientist and MD candidate at Duke, was hired at the end of August as bioinformatician for the Precision Cancer Medicine Initiative to help structure the clinical data in the MRT database and organize DCI’s annual data submissions to GENIE.
“We are thrilled that the Duke Cancer Institute is part of AACR Project GENIE,” said Shawn M. Sweeney, PhD, director, AACR Project GENIE Coordinating Center. “The Duke team has already made significant contributions to the project, in addition to data, which became publicly available in July 2019. We look forward to doing even more with Duke and the other 18 GENIE Participating Institutions."
As important as Frameshift MRT is for many DCI patients, it’s critical, McCall said, that researchers also have access to this wider GENIE data set, especially in rare and hard-to-treat cancers, so that DCI can help even more patients.
“When we spot a rare mutation in a patient’s tumor, we might ask, “Has that particular mutation been seen before?” said McCall. “We need the power of those 70,000 records to find out.”
On October 31, the AACR GENIE project announced a big expansion — a “BioPharma Collaborative” — that DCI, as a consortium member, is expected to join in year three of the five-year project.