NEW YORK (GenomeWeb) – Researchers from Ohio State University's Comprehensive Cancer Center have launched a public database of somatic driver mutations.
Currently, the database, known as CanDL for Cancer Driver Log, includes 334 variants in 60 genes, all of which are associated with known sensitivity or resistance to approved and investigational therapies, are associated with clinical trials, or have been functionally characterized. The variants link up to 169 unique literature references.
The researchers described CanDL in a paper published this week in the Journal of Molecular Diagnostics.
Sameek Roychowdhury, assistant professor of internal medicine and pharmacology at OSU, told GenomeWeb that the team will continue to expand the database and that third parties can also contribute their own entries.
He added that entries can be searched by gene or the specific variant and can also be downloaded.
Currently, most of the variant entries are SNVs, Roychowdhury said, but later additions will include fusions, insertions, deletions, and other structural information, as well as additional genes.
CanDL also does not yet contain information on tumor suppressor genes, only oncogenes.
"Because mutations in tumor suppressors are typically deleterious, including frame shifts and truncations, the degree of functional characterization for each unique mutation is limited but can often instead be inferred," the authors wrote in the JMD paper.
Roychowdhury anticipated that the primary users would be other clinical labs running next-generation sequencing-based tests for cancer patients. "I think this will be a time saver and close the gap from mutation testing to figuring out what to do with the data," he said.
Roychowdhury added that the lab began developing the database when it launched a precision cancer medicine clinical trial for patients with metastatic disease. As part of that trial, the Ohio State team analyzed patients with an NGS panel in order to identify potential treatments. Roychowdhury said that, initially, researchers every week were trying to match the mutations with available therapies, which was a time consuming task. So, instead, they decided to create a database where they could deposit curated information.
Annotation has frequently been cited as a major barrier for the adoption of NGS tests. Even targeted panels generate a vast amount of information that physicians often do not know how to deal with.
In addition, because most clinical cancer labs now have the capability to develop their own NGS panels, companies such as Foundation Medicine and Molecular Health are looking to distinguish themselves through their annotation pipelines and physician portals that help match a patient's genomic profile with approved drugs, off-label access, clinical trials, or even other physicians treating patients with similar molecular profiles.
Other academic laboratories have also been working on their own pipelines and databases as they acquire more sequence data from patients. There are large databases like ClinVar, launched by Heidi Rehm's laboratory at Harvard's Laboratory for Molecular Medicine, which focuses on germline variants and their relationship to disease. In addition, a number of groups have focused on building up databases of BRCA variants following the US Supreme Court ruling, which has enabled labs other than Myriad to test for cancer predisposition mutations in those genes.
Roychowdhury said that CanDL is different from these databases in that it focuses on somatic mutations in cancer and focuses on variants that are linked to therapies, clinical trials, or preclinical work.
Researchers from Washington University's McDonnell Genome Institute have also been developing a number of variant databases, including their most recent one, Clinical Interpretations of Variants in Cancer (CiViC), which contains written summaries of the clinical relevance of mutations in sequenced tumor samples. Currently, 80 genes are described in the CiViC database.
CanDL is similar to CiViC, but Roychowdhury said the goal was to keep it as simple as possible — CanDL lists the mutation, ranks its level of evidence on a 1 to 4 scale, and then links to the supporting literature.
The CanDL database also currently contains information only on oncogenes, while CiViC also includes tumor suppressor genes like TP53 and PTEN.