Researchers from the Mayo Clinic presented data at the annual meeting of the American Association for Cancer Research this week from their work using exome sequencing to identify gene mutations that could help gauge a patient's risk of developing peripheral neuropathy as a side effect of chemotherapy.
Starting with a subset of genes known to be implicated in inherited forms of neuropathy and then expanding to examine larger portions of the exome, the team identified two promising predictive targets — the genes ARHGEF10 and PRX — in a study of 73 cases and 43 controls and also bookmarked a number of good candidates for future study.
Andreas Beutler, a lead author of the study the group presented at the AACR meeting yesterday, told PGx Reporter this week that the Mayo researchers are planning to follow up on this initial effort to hopefully identify a number of additional markers associated with increased risk for chemotherapy-induced neuropathy. Once these are well characterized, clinicians with exome sequencing data available for their cancer patients could potentially use that information to make decisions about whether and how to use chemotherapy for patients with elevated risk.
"You might say it's not prime time right now to come up with individual prediction, but in the future we will have whole exomes for all these cancer patients, so let's get ready to use that information when it is available. That's what we are doing here," Beutler said.
"I do not envision that this would become a fee-for-service lab test," he explained. "Much more likely in a few years we will have CLIA-certified genomes or exomes costing a couple hundred bucks, so this will be something my clinical colleagues will be able to just look up hopefully with sufficient computerized support."
Peripheral neuropathy can be mild, but can also cause lingering and debilitating pain for months, or even indefinitely in some patients. Ideally, having a solid panel of risk genes for the side effect could help clinicians make difficult decisions about when and how to treat with different chemotherapeutic drugs.
"We are quite a bit away from using this to make clinical decisions," Beutler said. "But a realistic scenario might be for a woman with completely resected early stage breast cancer with low-to-moderate risk of recurrence — with more than 50 percent likelihood of never hearing from that cancer again. Should she take chemo? And if so, which one and how much? When you can identify patients with higher risk of a significant side effect, they may want to stay away from it. Or maybe you would choose a regimen with downsides different than the ones your patient is particularly susceptible to."
Beutler said the Mayo group has focused its work thus far on the drug paclitaxel, because the researchers had a paclitaxel-treated cohort available to them with very well-characterized phenotypes in respect to when and how quickly patients developed peripheral neuropathy symptoms. This allowed the group to conduct a well-powered study with relatively few subjects.
"Unfortunately, we are not yet the TCGA of symptom control genetics. When we started in the middle of last year, we had a budget for 120 exons but not for 1,200, so we focused on a strategy with sophisticated phenotyping and statistical modeling."
In the study, Beutler and his colleagues started with the hypothesis that genes that have been implicated in hereditary neuropathy might also play a role in the peripheral neuropathy caused by chemotherapy.
"There are many case reports of patients diagnosed with a very mild form of hereditary peripheral neuropathy in retrospect after developing catastrophic neuropathy after receiving often just one dose of chemotherapy," Beutler said. "It turns out these patients come from families with very, very mild [neuropathy] symptoms, but they [appeared to be] asymptomatic."
"This suggested to us and others in the field that the same genes might be involved in the hereditary and acquired neuropathy," he said. "So, we formulated a primary endpoint looking at a subset of established hereditary neuropathy genes."
The researchers sequenced about 20,000 genes from blood samples of the study participants, looking for an association between any of these genes and chemotherapy-induced neuropathy.
Overall, the group identified two hereditary neuropathy genes as new CIPN candidate genes. One, ARHGEF10, was "highly significant" based on single-gene testing with three single nucleotide variants predictive of the side effect. The researchers validated this finding in an independent 75-patient group from another paclitaxel trial, which confirmed the finding.
The team also found that another gene, PRX, was disrupted by a single mutation in eight cases and zero controls, a secondary finding Beutler said is "still exploratory, but … very promising."
Though investigation into genomic determinants of specific drug side effects has lagged behind pharmacogenomic efforts to define drug response, the Mayo group is not alone in pursuing this type of work.
A company called Inform Genomics reported last year that it was developing its own SNP-based test, called onPART, to gauge chemo side-effect risk for a range of different side effects and drugs including peripheral neuropathy, but it has not published a detailed description of its findings (PGx 10/10/2012).
According to the Mayo team, theirs is the first study to profile the exome broadly for gene mutations linked to risk for chemotherapy side effects, an approach they believe has great potential for elucidating such complex pharmacogenomic traits.