Skip to main content
Premium Trial:

Request an Annual Quote

MedSeq Study Yields Preliminary Results on Costs of Adding WGS to Patient Care


NEW YORK (GenomeWeb) – Investigators involved in the MedSeq Project have found that at least in the short term, incorporating whole-genome sequencing data in the care of patients did not significantly increase healthcare costs compared to patients who didn't receive such genomic testing.

The results, although preliminary, serve as a jumping-off point for a broader discussion on the costs and utility of whole-genome sequencing in mainstream care, according to Robert Green, MedSeq's principal investigator, who presented on the project at the Precision Medicine World Conference last month. "These are tough questions," he said in an interview. "But these are the questions that payors are asking us. They want to know the value and they want to know the cost."

The study investigators this week completed the data collection phase of the project. Green, who is also a geneticist at Brigham and Women's Hospital, noted that the preliminary results of this pilot project suggest the need for an even larger randomized study, and that his team is seeking funding for a MedSeq 2.

The randomized study, which received a four-year $9.6 million grant from the National Human Genome Research Institute in 2011 and began enrolling patients the following year, is evaluating the medical, behavioral, and economic impact of adding whole-genome sequencing to everyday patient care.  

In one part of the study, MedSeq enrolled 100 healthy patients between the ages of 40 and 65, who are seen by 10 primary care doctors and are randomized to receive either standard care informed by family history or care informed by family history and whole-genome sequencing. In a second part, 100 patients with hypertrophic or dilated cardiomyopathy seen by 10 cardiologists are randomized to receive standard care informed by family history and traditional cardiomyopathy genetic testing, or all of this plus WGS testing.

The early MedSeq findings show that 92 percent of participants who had their genomes sequenced had recessive carrier traits and that 21 percent harbored a previously unknown risk variant for hereditary cancers, heart disease, and other rare conditions. Green previously discussed these top line findings at conferences, and has talked about how doctors handled genomic information in the care of their patients, and what impact the results had on healthcare costs.

Six months after receiving the WGS report, the genetic testing group had no statistically significant cost increase from ordering additional testing or consultations compared to the control group, Green and his colleagues found. Depending on how researchers calculate the downstream costs incurred in the first six months — without including the cost of sequencing and interpretation — the difference between the two groups is anywhere between $40 and $200, according to Green. He acknowledged, though, that the costs for the WGS group are higher compared to controls when sequencing and interpretation costs are added in.

Although this is a small study, the cost findings are "remarkable," Green said, when one considers the amount of information doctors received. Within MedSeq, physicians and patient participants learn all unanticipated findings across 5,000 disease-associated genes, when testing reveals variants that are classified as pathogenic or likely pathogenic.

"We're not just talking ACMG-56," Green said, referencing the American College of Medical Genetics and Genomics' recommendation that labs report incidental findings in a list of 56 disease-linked genes after exome or whole-genome sequencing, even if testing was ordered for a different indication. The recommendation has been controversial, partly because of the lack of evidence on the benefits, harms, and costs of providing such information.

This is the evidence that researchers are trying to gather through MedSeq, and the randomized-controlled design was key to investigating what can be discovered with WGS, and whether that knowledge adds costs compared to the standard of care. 

Too often when studying the impact of precision medicine, researchers tend to not conduct randomized control trials. Given the excitement around personalized medicine, researchers have noted the difficulty of enrolling for studies with a control arm, where a proportion of patients would get a placebo or a standard option with lots of toxicities, like chemotherapy.

For example, the Lung-MAP study, which was originally designed to be an umbrella trial in which patients in biomarker-based sub-studies would be randomized to either targeted treatments or standard chemotherapy, had to be redesigned due to lackluster accrual. As co-principal investigator Roy Herbst of Yale Cancer Center previously said in an interview, while Lung-MAP was being designed, immunotherapy became available and "it's hard to tell someone that you're doing a precision medicine program, we found this new genetic mutation, and we're going to randomize you to this chemotherapy, which has a 5 percent chance of a response and a 100 percent chance of side effects."

Green noted that doing a study like MedSeq was an "enormous" undertaking, involving training doctors to read and understand the test report after their patients have been evaluated for more than 5,000 disease-associated genes, and following up on physicians' recommendations to evaluate their "appropriateness" with a panel of genetics specialists. Physician/patient interactions were even recorded to "make sure doctors didn't recommend anything inappropriate but also to hear how they expressed the information," he added.

There are limitations to MedSeq, too. For example, the study only involves 200 individuals and there's only been six months of follow up. 

Still, the preliminary data offers an opportunity to start a much-needed conversation about the utility of broadly implementing precision medicine, in Green's view. It's a complicated discussion to have because on the one hand, for payors, before incorporating genomics in mainstream care, its benefits, harms, and costs must be vetted through rigorous, randomized controlled studies. On the other hand, the very nature of genomic information also confounds the current thinking about utility and costs.

Green provided a hypothetical example to illustrate the difficulty: "When you give an entire family access to this kind of information, they say, 'Oh, my son is 16 but in 10 years, he's going to be glad to know he's carrying these three recessive carrier traits,' and then, when he's 26, [he and his partner] use the information to do in vitro fertilization and preimplantation genetic diagnosis." This added expense wouldn't be there without having done that initial genomic test years ago, he said.

At the PMW Conference last month, Green noted that although there were only modest cost increases within the pilot effort, cost considerations are quite different on a population level. "When you multiply it by millions, you're still costing the healthcare system quite a bit," he said at the conference.

This is perhaps why within the Precision Medicine Initiative, a project aiming to collect a variety of data on 1 million Americans, whole-genome sequencing is a goal, but a long term one. According to Eric Dishman, who is heading recruitment efforts, the plan is to first scale up genotyping on arrays in 2018, and transition to whole-genome sequencing 18 to 20 months later, or once it becomes cheap enough to test a million people.

Green's team, meanwhile, is also looking to study the impact of WGS in a larger cohort. "That's one of the reasons we need a MedSeq 2," he said. In an expanded project, researchers would like to enroll more African Americans and Hispanics, enroll at least 1,000 people, and follow participants for a longer time.

MedSeq was split into looking at the impact of WGS data in a healthy population and those with cardiomyopathy. Green envisions that MedSeq 2 will enroll from all comers and include doctors from different medical specialties. "Imagine a world where sequencing is ubiquitous," he said. "How do we demonstrate the clinical utility and cost of that sequencing, if it is able to be used by virtually any provider?"