Chris Anderson

Behind a given tumor type, many pathogenic mechanisms may lurk. Uncover them with NGS and customize therapies accordingly.

There is little doubt that the future of solid tumor cancer treatment will be increasingly precise and tailored to individual patients based on specific predictive biomarkers. Next-generation sequencing (NGS) technologies, which have become more refined in just the past few years, are a leading breakthrough that has allowed researchers and molecular pathologists alike to shift from querying single genes to panels that can simultaneously examine 40, 50, to even hundreds of genes in one run.

This is not to say that NGS will be the silver bullet of personalized cancer treatment, as there is a wide array of genetic tests available today to predict the risk of developing specific cancers and ones that also help target specific therapies to cancer subtypes. Rather, the power of these new sequencing technologies will be to unlock information about the variable nature of solid tumors to help provide clinicians with a better understanding of cancer’s mechanisms. Using this information, they can choose from the rapidly expanding roster of cancer therapeutics, specifically which one is most appropriate to best treat an individual patient based on the genetic profile of their cancer.

“Until recently, the majority of molecular diagnostic laboratories have been using standard sequencing and various different types of mutation detection technologies that are aimed at looking at one gene and one target at one time,” noted Rajyalakshmi Luthra, Ph.D., professor of hematopathology, and director of the molecular diagnostic library at the University of Texas MD Anderson Cancer Center. “With the advent of all the targeted therapies and the advances we made in our field, we now know that for each tumor type we may have to look for more than one marker at a time.”

NGS Informing Treatment

While NGS technology has made great strides since its introduction nearly a decade ago, it is still a relatively expensive technology to deploy, upgrade, and also on a per test basis. Granted, the cost of sequencing a panel of as many as a couple hundred genes now is roughly equivalent to the cost five years ago of the same analysis on a single gene, but that still equates to an average cost of anywhere from $2,500 to $5,000 per test. For this reason, deployment of NGS is still largely confined to large academic medical centers and in the laboratories of commercial diagnostics companies.

“It is still early days of having broad spread adoption of this technology,” said Robert Daber, Ph.D., director of research and development and director of cancer genomics at clinical diagnostic company BioReference Laboratories. “As the field matures, we feel there are a couple of things that need to be in place. First, it needs to cover actionable mutations, so these need to clinically relevant. And second, it needs to be cost effective and able to work with really minute specimens.”

At MD Anderson, the decision to begin using NGS technology three years ago was made based on the limitations its lab worked under previously, Dr. Luthra noted. “The goal is that we should be able to dissect the tumor at a molecular level and provide the clinicians with the markers that they can target to treat the patients,” she said. “The amount of tumor specimen we get is very limited. So you cannot do one test at a time as it was in the past. If you had to look at ten different markers then, you wouldn’t have enough material to look at all of them.”

Just as important as the markers that are clinically relevant are those discovered mutations that are currently not considered clinically relevant. “There is an element of discovery here, as well, where we are finding complex profiles that maybe are only seen in one percent or one-half percent of patients with a given tumor type,” Dr. Daber said. “So the question becomes are they going to respond like others with a similar profile or will they respond differently? There is a bit of knowledge gap since we don’t have outcome data yet.”

In some cases, while the data may not suggest a therapy, the genetic profile of the tumor may point the way toward a specific clinical trial underway that could be of benefit to the patient. This method of helping to better define appropriate patients for trials can speed enrollment, but also begin to fill the knowledge gap Dr. Daber mentioned by validating additional relevant genetic markers.

Current NGS Limitations

Because NGS technology is so new, tool companies are continually finding ways to tweak and improve methods and the corresponding data and informatics tools they offer. For Dr. Luthra, this has presented challenges and is also another reason why other health systems may not yet want to deploy NGS in their pathology labs.

“The vendors are constantly trying to improve their platforms both from a hardware and software standpoint. This is reasonable; you expect that with a new platform they will want to improve it and make frequent changes,” Dr. Luthra added. “However, at a clinical laboratory you cannot constantly go back and re-validate whatever it is that they have validated initially. It is easy for a cancer hospital like us to have enough samples with the various types of mutations to validate the platform. But when it is a smaller or medium-sized laboratory, they may not have collected enough samples to do that.”

Another limiting factor today, and one that needs to be addressed in the future, is the relative genetic literacy among oncologists. According to Bruce Korf, M.D., Ph.D., chair of the department of genetics, and director of the Heflin Center for Genomic Sciences at the University of Alabama at Birmingham, the glut of genetic information that will be used to inform treatment decisions will require additional education of clinicians, especially since a sizable cohort received their genetics training before the genomic age. Even among those who’ve received more recent training, who understand the role genomics can play, he thinks it is unlikely these doctors are ready to employ it in their practices.

“It’s not enough to say [physicians] should know this or that,” Dr. Korf said in a podcast prior to the Next Generation Dx Summit in August. “They should know about the indication for genetic testing or they should be able to understand the concept of the nine variants versus a pathological mutation. We need to get beyond that into the specifics of what do you actually expect them to do and what kind of skills do they need to carry that out.”

But even bringing physicians to this level won’t likely be enough to fully leverage the genetic information available to physicians. “The practicing physician is going to need to be backed up by medical geneticists and genetic counselors who can help them to deal with the day to day opportunities in genomic medicine and there’s a significant challenge facing us in terms of building that work force,” Dr. Korf concluded.

The Future of NGS in Cancer Care

As Dr. Daber noted, it is still the early days of deploying NGS technology to inform clinical decision making, but the data being generated daily by the academic medical centers and diagnostics manufacturers is the key to further unlocking value. “I think the technology allows us to look at so many genes, it has gotten us as far as we can get today,” he said. “What we really need is to start collating information around the experiences and knowledge of the significance of these markers. IT technologies can be leveraged, but we don’t need a faster computer and bigger storage. Right now we need more knowledge.”

Dr. Luthra agrees with this and added that as more markers and combinations of markers become validated via the expanding datasets and associated outcomes, it will provide an opportunity to make the technology accessible to a larger number of hospitals. “From this data we can really see if it is necessary to test fifty genes. Even though they are cancer related they may not show mutations for [a specific cancer],” she added. “If that is the case, then I believe vendors will come up with smaller, very targeted panels to be used by smaller hospitals for immediate care of the patient and then there would be option to use a more comprehensive panel if after the first test more information was needed.”

While these developments may be a few years down the road, Dr. Luthra is confident the trajectory of NGS is leading to this clinical reality. “We have tested more than 10,000 cases now using NGS. It seems like a lot of cases, but then you subgroup these and show who got secondary treatment and follow these for outcomes. It will take a couple of years, but we have big data efforts and clinical trials that will allow us to provide more information about these treatments, this will make [NGS] available to not just the major cancer centers, but medium throughput hospitals, as well.”

Chris Anderson ([email protected]) is the former chief editor of Drug Discovery News, which he helped launch in 2005.

This article was originally published in the October 22 issue of Clinical OMICs. For more content like this and details on how to get a free subscription to this digital publication, go to www.clinicalomics.com.

Previous articleInvestor’s Gift Yields $10M for the Jackson Laboratory
Next articleLexicon Licenses Some Ex-U.S. Telotristat Etiprate Rights to Ipsen for $145M+