This article has been corrected to note that the next-generation sequencing component of the trial is currently observational only.
NEW YORK (GenomeWeb) – A prospective observational trial of multiple myeloma patients is aiming to demonstrate how next-generation sequencing could guide therapy.
Jonathan Keats, head of the multiple myeloma research laboratory at the Translational Genomics Research Institute, reported on ongoing efforts of the Relating Clinical Outcomes in Multiple Myeloma to Personal Assessment of Genetic Profile, or CoMMpass, study at the Clinical Genome Conference in San Francisco earlier this month.
The study is a Multiple Myeloma Research Foundation-supported trial that kicked off in 2011 and aims to enroll 1,000 treatment-naive patients with multiple myeloma and follow them for five to 10 years. Eligible patients receive tumor/normal whole-genome sequencing, exome sequencing, and transcriptome sequencing.
In the first part of the trial, patients receive standardized care, and sequencing is done purely for observational purposes. Patients receive either standard therapy alone, in combination with a proteasome, or the proteasome alone.
The patients will be followed and clinical parameters collected every three months for at least five years. The group also plans to do whole-genome sequencing, exome sequencing, and transcriptome sequencing on 800 relapse samples. The goal is to then have a trial arm that uses the sequencing results to choose the next line of treatment and compare outcomes with a physician's choice arm.
In addition, genomic and clinical data is being made available for researchers through the MMRF Researcher Gateway. The MMRF is funding the study, while TGen is performing the bulk of the sequencing and GenoSpace, a Cambridge, Mass.-based bioinformatics company, designed the MMRF Researcher Gateway portal. Sagar Lonial, professor of hematology and oncology at Emory University, is the clinical lead for the study. Other collaborators include the Van Andel Research Institute, Spectrum Health, ACT Oncology, and more than 50 clinical sites in the US and Canada.
As of January 2014, 436 patients were enrolled and 196 had been sequenced. By the end of July, Keats anticipates that around 600 patients will be enrolled and 400 of those sequenced.
Researchers at TGen perform low-coverage, long-insert whole-genome sequencing to identify structural variations; high-coverage exome sequencing to identify point mutations; and transcriptome sequencing, which can identify gene fusions as well as confirm which point mutations identified from exome sequencing are actually being expressed.
Thus far, Keats said, the researchers have found that the majority of mutations are not expressed. Of the 28 genes that previous research has found to be mutated in over 3 percent of all multiple myeloma patients, RNA-seq performed in the CoMMpass trial to date has found that only 11 are clearly expressed, Keats said.
Layering the RNA-seq "adds a nice filter, particularly if you are going to treat the patient based on the molecular profile," he said.
One interesting finding from the long-insert whole-genome sequencing, Keats said, was the finding of diverse breakpoints to IgH. Chromosomal translocations involving IgH occur in approximately half of multiple myeloma patients. But surprisingly, the second most commonly identified translocation was between IgH and MYC. "We knew MYC was important in progression, but finding [that translocation] in about 15 percent of patients at diagnosis was new," he said.
Another common occurrence is what Keats described as "irrational marker positive expectations." For instance, he explained, in one patient example, the team identified a BRAF mutation that would indicate targeted therapy. However, when looking at the allele frequency, they found that only about 10 percent of tumor cells had that mutation. Simply reporting a BRAF testing result as positive or negative would not be sufficient to predict whether a targeted therapy would be effective.
"Even if the drug had 100 percent efficacy, would we see any difference in the patient?" Keats said. "If we start talking about giving people drugs based on being positive for something, we have to consider whether we expect a good result."
Knowing the mutation's frequency and how mutational frequency affects drug response will be important in order to make sure that potentially effective drugs "don't go down the trashcan," Keats said.