When the Genomic Data Commons opened June 6, the $20 million portal that consolidates NCI’s datasets contained genomic information from 14,500 patients.
Before the end of the month, that number jumped to 32,500, as a result of a contribution from Foundation Medicine Inc., a molecular information company founded in 2011.
The GDC was announced in June by Vice President Joe Biden as part of the National Cancer Moonshot Initiative (The Cancer Letter, June 6).
“We’re adding 18,000 cases from Foundation,” said Louis Staudt, director of the NCI Center for Cancer Genomics. “Many important cancer genes, up to 287, have been sequenced by Foundation in these cases. We applaud their public spirit, and we are really glad that they anted up.”
FMI didn’t receive remuneration for the 18,000 de-identified cases. A conversation with Staudt appears here.
At the National Cancer Moonshot Summit, Biden recognized FMI’s contribution, saying that the GDC—as well as FMI’s dataset—is “critical for advancing the field of precision medicine and improving the care of cancer patients.” (The Cancer Letter, July 1.)
The FMI contribution is one of several bioinformatics initiatives Biden highlighted. The moonshot, which rapidly shifted away from the initial rhetoric of curing cancer, has focused largely on two concrete goals: streamlining review and oversight of oncology products at FDA, and inducing data hoarders to make systems interoperable.
This is not the first time FMI has released cancer data into the public domain. In February, the company made 1,300 pediatric cancer profiles publicly accessible.
“To our knowledge, that transfer increased the amount of publicly available genomic data for pediatric cancers by twofold or more,” said Michael Pellini, CEO of Foundation Medicine. “We’ve heard feedback from patients and researchers regarding the real value this dataset offers.”
Foundation Medicine has information on over 80,000 patients, gathered from its work with academic cancer centers, community facilities, and pharmaceutical companies.
“The majority of [cases] have come in through the United States, but certainly, there’s a growing percentage coming in from outside of the United States as well,” Pellini said to The Cancer Letter. “There’s also information that has been generated on many thousands of patients as part of clinical trials, but that information is not shared in the broader database. That information is still primarily walled off for our pharma partners.” A conversation with Pellini appears here.
FMI, based in Cambridge, Mass., was chosen to profile patients enrolling in Lung-MAP, a multi-arm clinical trial that assigns advanced squamous cell lung cancer patients to therapy based on molecular characteristics of their disease (The Cancer Letter, Feb. 11, 2013, Feb. 27, 2013, May 8, 2015).
“I really appreciate Foundation Medicine’s philanthropy, especially in pediatric cancer,” said Nancy Goodman, executive director and founder of Kids v Cancer. “They were the first genomic testing company to send their screening protocol to FDA for review as part of the Lung-MAP protocol.
“They pursued collaborations with many pediatric oncologists and released a lot of pediatric cancer profiles earlier this year,” said Goodman, who successfully advocated for the creation of priority review vouchers to incentivize development of drugs for rare pediatric diseases.
The GDC has logged significant use since its launch, said NCI’s Staudt.
“A lot of people are going to the site—many computer programmers would like to use our automated application programming interface that allows them to directly connect to the GDC,” Staudt said to The Cancer Letter. “We’ve been really busy. I think the busy next year or two will be to improve our visualization tools for the data.”
The FMI dataset is unique because it contains information beyond the “canonical hotspots” that usually draw the attention of cancer researchers, Staudt said.
“There are two or three other aspects that I think are special to the Foundation dataset,” Staudt said. “I would say, largely, almost exclusively, these would be patients who had a relapse of their disease after treatment.
“This is because the Foundation platform has been seen as a way to try to find a new therapeutic option when all others have been exhausted.”
Also, since the GDC consolidates genomic information from NCI’s existing repositories, the institute’s data tend to be skewed towards tumor samples that have not been influenced or modified via therapy (The Cancer Letter, April 29).
“Why that is important is that most of our sequencing in both The Cancer Genome Atlas and TARGET has been in pretreatment samples,” Staudt said. “So this dramatically expands our look at what relapsed cancer samples are genetically.
“We may find some samples in the Foundation sets that would be illuminating.”
As with other Database of Genotypes and Phenotypes (dbGaP) projects, researchers can submit a study proposal to access the FMI data and commit to publishing their findings.
“The data cannot be re-presented and sold for profit, but that’s not special to Foundation. That’s all dbGaP projects,” Staudt said. “This is a pure academic, philanthropic activity by Foundation.”
The GDC team is now looking at importing cancer genomic data from the National Center for Biotechnology Information, which contains at least 4,000 high-quality cases that have not been organized through a search portal like the GDC.
“Basically, NCBI, unlike the GDC, only enables data download, not data visualization and analysis,” Staudt said. “There are a surprising number of high quality cancer genomics projects there that, according to our conversations with NCBI, are underutilized.
“That’s free data, and I reiterate, I don’t think most cancer researchers are even looking at that data currently.
“I realized that these datasets represented a great opportunity to quickly expand the scope of the GDC. That is what we’re planning to do.”
Pellini: More Data to Come
Sharing data is central to Foundation Medicine’s mission, Pellini said.
“We were already deep into the process of determining how and where we would release a set of the data contained in FoundationCore,” Pellini said. “In parallel, about a year ago we learned the idea of the GDC—it wasn’t called GDC then— had surfaced in connection with the NCI.”
“We started a dialogue, and frankly in a relatively short time period both parties expressed a real interest in working together. We were prepared to go this route regardless, and the NCI and now Moonshot’s work with the GDC provides such a tremendous forum through which to make this type of data available to the broader community.”
“One reason FMI is the leading company in molecular information, because other companies and labs conduct profiling tests in a more limited fashion,” Pellini said.
“They’re doing a different type of standard molecular profiling than Foundation Medicine, a much more narrow approach,” Pellini said. “In terms of the ability to provide a comprehensive genomic profile for a patient who is battling a complex cancer, we are the leading company, and we have worked hard to set new standards in terms of how to profile a patient’s tumor in a comprehensive way delivering very high quality results time and time again.”
FMI will release additional data from time to time as FoundationCore, the company’s internal knowledgebase, continues to be built, Pellini said.
“We haven’t given any specific timelines, but I do think it’s reasonable to expect that as time goes on, and we continue to build FoundationCore, we will continue to make additional aspects of the FoundationCore publicly available,” Pellini said. While there are many aspects of our work that are proprietary, we see the broader data as one of the pieces to the puzzle to advance cancer care.
“What we have to do not only at Foundation Medicine, but also broadly across the oncology community is raise the level of awareness of the new therapeutic interventions that are available for patients with cancer today, and in order to appropriately select those therapies, a comprehensive understanding of the genomic basis of a patient’s tumor is vital.
“The great number of opportunities we have to put this information into the hands of researchers as well as clinicians, the greater the opportunity for us to really increase that sea level of awareness of this new era of molecular oncology.”