NEW YORK (GenomeWeb) – Personal Genome Diagnostics announced earlier this month that it is developing a non-invasive next-gen sequencing assay to help identify which cancer patients have the best chance at responding to treatment with immune checkpoint inhibitors.
The company received a Phase I contract from the National Institutes of Health at the end of 2016 to support development of the test, called MutatorDETECT, which infers a patient's overall tumor mutation load from analysis of circulating cell-free DNA using a targeted sequencing panel.
Mutational load, or burden, has begun to command significant attention from diagnostics developers as studies over the last decade have shown that patients whose tumors have a large number of mutations are also more likely to have a robust anti-cancer immune response, and tend to do better when treated with drugs that in some way precipitate this response.
The reason for this, evidence suggests, is that a greater number of mutations means a higher likelihood that a tumor has induced the development of neoantigens that allow the immune system to recognize and attack a cancer.
Immune checkpoint inhibitors have thus far been approved primarily for cancers like melanoma, where high mutational loads are common. But there is increasing evidence that subsets of patients with other types of cancer can also have highly mutated tumors that predispose them to response.
As a result, academic groups and companies have marked tumor mutational burden (TMB) as a target for diagnostics development, hoping to create tests that can pick out these responsive patients and better personalize treatment with this class of drugs.
John Simmons, PGDx's director of translational science and diagnostics, said in an interview that the announced NIH funding has helped support the company as it completed the initial informatic work to develop its assay and perform the first steps of analytic validation. Investigators are now clinically validating the test with a number of academic and pharma partners, he said.
According to Simmons, the test that PGDx has developed is different than some others that are being advanced, both because it is performed non-invasively — on circulating cell-free DNA —  and because it relies on a smaller sequencing panel.
Though he declined to detail the test's genomic targets, Simmons said that the assay covers about as many targets as current comprehensive liquid biopsy sequencing tests on the market.
"I can tell you that the size of the assay is in line with other plasma panels that are similar to our own," he said. "We tried to use that size as a benchmark."
In order to develop a test that could indicate the presence of a high number of mutations across that whole genome without having to sequence the whole genome, or even the hundreds of genes other tests require, Simmons said that the company utilized the wealth of whole-exome sequencing data that it has amassed in its operations so far, using an informatic strategy to identify a much smaller panel of regions that are highly predictive of overall TMB.
"We have been working in the immuno-oncology space … for many years," Simmons said. "We've had a lot of experience and we've done a lot of sequencing"
PGDx co-founder Luis Diaz, for example, ran the first prospective trial looking at immunotherapy response in patients with microsatellite instability, a deficiency of mismatch repair mechanisms that leaves tumors with hundreds to thousands of mutations in stretches of repetitive DNA called microsatellites.
"PGDx did the sequencing work for Luis's trial," Simmons said, among much other work in the area, "and we took that experience and [developed] a computational approach to identify regions throughout the genome that could be used in an informative way."
Importantly, Simmons said, the company's approach in identifying these regions was not restricted to genes, but include non-coding areas of the genome as well.
"We used thousands of whole exomes to narrow down to regions that we can show, when there are mutations there, it is highly correlated with mutational burden thresholds."
PGDx is not alone in its interest in tumor mutation load as a strategy to predict immunotherapy response. Foundation Medicine began sharing data from experiments last year in which it used its FoundationOne comprehensive targeted sequencing test to predict the overall mutational burden in a tumor genome.
Soon after, the company announced that this TMB analysis had been made available to clinical customers who order FoundationOne for their patients.
Foundation Medicine's test is tissue-based, and requires sequencing the full FoundationOne panel — more than 300 genes — to predict whether a patient's tumor exceeds the threshold for high TMB across its entire exome.
Foundation's President and Chief Operating Officer Steven Kafka said in an email this week that the company is seeing a positive response from oncologists to these new options, "especially among those who recognize the value of TMB as a clinical marker in helping to predict responses to FDA-approved checkpoint inhibitor cancer immunotherapies across multiple advanced cancers.
"Oncologists also value the ability to receive both comprehensive tumor genomics information and TMB and [microsatellite instability] scores from one sample," he added.
Foundation Medicine has recently expanded its testing portfolio to include liquid biopsy. It launched its cell-free DNA test, FoundationAct, last spring.
Kafka did not discuss whether the company might be working on a way to offer TMB inference in a non-invasive format.
Meanwhile, liquid biopsy firm Guardant Health has said that it is developing methodology for assessing patient TMB from blood samples.
Guardant's TMB analysis is part of a 500-gene panel it has developed for use in clinical trials and drug development.
In an email, Guardant CEO Helmy Eltoukhy said that the company's large panel, called Guardant Omni, is designed to have broad gene coverage to support tumor mutational burden for use with immunotherapy clinical trials, among its other applications.
"Some of our biopharma partners are already using Guardant360 for immuno-oncology applications today," he added, though he did not mention any plans by Guardant to develop a simpler TMB test for standard clinical use by oncologists outside of the context of a clinical trial.
According to PGDx's Simmons, while sequencing many hundreds of genes might be relatively cost-effective in tissue, as in Foundation Medicine's case, it's not very practical in the non-invasive setting, because analysis of circulating tumor DNA requires sequencing much more deeply than is required with tissue samples.
"At PGx we really have that broader strategy of bringing oncology NGS assays out to the marketplace in a decentralized fashion … working with drug companies to co-develop tests and bring them through the FDA," Simmons said.
Notably, the firm announced an IVD agreement with Illumina last year in which the  companies plan to develop two in vitro cancer diagnostics — one for tissue and another for plasma.
"With that charge in mind you are very cognizant of how that assay will be deployed in the final end-user space," Simmons added. "You want to be really cognizant of size of the assay, especially in plasma where you're sequencing at a much higher depth than with a tissue-based assay. Real estate is important. It's a key driver of the cost of that assay."
As it develops MutatorDETECT, PGDx is working with biopharmas to investigate checkpoint inhibitors, as well as with some academic collaborators interested in testing the assay in specific patient populations, Simmons said.