NEW YORK (GenomeWeb) – In March, the US National Academy of Medicine published a report on biomarkers and how the US government should take up a role in ensuring confidence in testing for precision medicine.
"I was sort of stunned when this report came out that it wasn't all over the press," said Garret Fitzgerald, a committee member and professor at the University of Pennsylvania.
Titled, "Biomarker Tests for Moleculary Targeted Therapies: Key to Unlocking Precision Medicine," the report calls for standards in analytical validity and clinical utility, a national registry of biomarker-based tests, and cooperation from electronic health record software companies to support these tests, among others.
With funding from governmental organizations including the National Cancer Institute and the Centers for Disease Control and Prevention; non-profits including the American Society for Clinical Oncology and the College of American Pathologists; and for-profit entities like Novartis and Quest Diagnostics, the report's commissioners were just a small representation of the parties that might have an interest in how biomarker-based testing matures. Patients, public and private payers, and regulatory bodies added in make it hard to please everyone.
Fitzgerald is chair of the Department of Systems Pharmacology and Translational Therapeutics at Penn's Perelman School of Medicine. With experience developing biomarkers for the molecular clock in cardiac disease as well as with experience in the pharmaceutical industry with Genentech and Novartis, he was able to provide a unique perspective to the report committee.
"I was representative of the field beyond cancer," he told GenomeWeb.
He's also written about policy issues and published an article this week in Science Translational Medicine highlighting some of the report's findings and recommendations.
GenomeWeb spoke with Fitzgerald about the report and his article, digging into the "Wild West" nature of the precision medicine biomarker market and the political will that will be needed to make sure it works for as many people as possible. The following is an edited transcript of that interview.
GW: A lot has been said and written about biomarkers. What does this report add to the discussion?
What's been discussed before is the potential of biomarkers. We're at the early stages of precision medicine. We have a couple of examples like imatinib, where the identification of the appropriate biomarker does indeed identify people likely to respond. But we've loads of claims around that are unsubstantiated, loads of associations we have no idea how to interpret them.
I don't think what has been addressed so precisely is how to do that. Here we're talking about the application of tests that direct patient care. It's really a fundamental issue that has not been addressed — that there be transparent evidence around analytical validity. Having established that, the more challenging thing is to prove that that improves outcomes. Tell me where that's been shown or discussed in great detail.
The discussion about biomarkers has been as superficial and unregulated as the actual field is.
We're moving from an era where we've looked for large average effects in randomized trials, to a position where we want to be able to parse variability of drug response. That's a revolution in clinical research. In that context, the biomarker business has a fundamental role to play. We've got to know about analytical validity of what's out there, and if it does demonstrate clinical validity, the paymasters have to be convinced around that issue. That means resetting the incentive. To my knowledge, those sorts of issues haven't been addressed until now.
GW: So why now?
For example, with 23andme before the FDA curbed their enthusiasm, we know they were selling the idea that a few SNPs here and there and you're going to get Parkinson's. And then lo and behold, there was that study where they sent the same DNA to a bunch of these companies, they all got different answers as well as different advice.
GW: One concrete recommendation from the report is to develop labels for biomarker-based diagnostic tests. Why?
The labels just capture what we're trying to do. We're saying there should be an evidentiary framework that serves as a transparent basis for assessing the analytical and clinical utility, which is completely absent at the moment.
Without a common evidentiary framework for validity and utility, you don't know what you can rely on and you don't know how useful it is. The whole idea is to use quantitative or qualitative distinctions between people as a basis for making decisions about treatments they should have or not have. Without that, one of the fundamental pieces of infrastructure of the whole precision medicine idea is absent.
To do that requires the engagement of these very diverse stakeholders and it needs to be done urgently.
How do you arrange things so the maker of a biomarker is incentivized to provide the info that informs the label? We came up with a few suggestions. If labeling is used by reimbursement bodies, that is incentive right there [for the testing companies] to provide updated information. Perhaps the more challenging aspect is that the strength often becomes diluted by the accumulation of evidence, and that's where regulatory bodies have a role to play in terms of assessing evidence. We obviously can't mandate how this is approached. We suggest many bodies relevant to this feed up to the Department of Health and Human Services.
GW: Is there anything that can be done piecemeal? Or does this all have to be taken on at the same time.
I think it will be done piecemeal, to some degree, but the key issue is there needs to be an assumption of responsibility by, ideally, leadership in HHS, which means right at the top because many of the elements reside within it. Many of the elements reside within HHS to convene a representative and empowered body to begin to reduce this to practice. It's in everybody's interest, unless you're a fraud. If you're not a fraud, it's in everybody's interest.
GW: What's your hope for the next secretary of HHS?
My hope is this person would understand the importance of this and its fundamental relevance to the improvement of human health. Without this, we can't move away from old-style clinical research where we were looking for large average effects because that's all we could detect, and that didn't have any relevance at the individual level, except on the basis of probability, to something that really answered the question any patient asks, which is, "Will this help me and do it safely."
People are not too interested in whether there's a three-month average life extension. You can't begin to do that science, without having a credible, believable evidence base with respect to biomarkers.
GW: What are some other important takeaways from the report?
As precision medicine is deployed, it's important it does not become the domain of just the wealthy elite and the academic elite. We don't want this to be confined to the 1 percent in high-end academic medical centers.
One of the other things we recognized is that the existing regulatory framework, namely CLIA, is not positioned to deal with the explosion of sequencing capability, multi-omics integration, these sorts of things. Although there is a loose framework of US Food and Drug Administration approval, it's really not fit for purpose in terms of 21st century technology.
GW: What was it like working on this committee?
There was a real commitment. I was very impressed by the sense of commitment of the people, both at NAM and outside. In an environment where the experiences with outfits like 23andme and Theranos are really cautionary tales of how an unregulated environment can be misleading to patients, we thought that getting this right is a fundamental piece of the infrastructure in pursuing precision medicine. It's not the only piece, but If you don't have this bit right, so that it's believable, transparent, sustainable, it's very difficult to envisage how the goals of a more precision deployment of therapeutics to individuals, who are likely to benefit safely from them, is going to be achieved.
I've been on a bunch of IOM committees. I think the output of this one is the most important one I've been on. It's a challenge. The output is not, 'Here's the solution.' The output is, 'Here's the challenge of how to approach the solution.' The responsibility involves HHS, FDA, to some degree the Centers for Medicare & Medicaid Services, and the National Institutes of Health and so on to actually respond to this. They're the people with the resources to respond. They're the ones who commissioned the report.