Skip to main content
Premium Trial:

Request an Annual Quote

Genetic Testing Needs an FDA Fix

Premium

Recently, Gail Javitt, law and policy director of the Genetics & Public Policy Center at Johns Hopkins University, testified before the US Food and Drug Administration on the topic of genetic diagnostics. Genome Technology’s Meredith Salisbury caught up with Javitt to find out more.

Genome Technology: What is the Genetics & Public Policy Center, and what does it mean to be the law and policy director there?

Gail Javitt: We were founded in 2002 with a grant from the Pew Charitable Trusts to look at policy issues arising out of genome sciences — our first initiative focused on reproductive technologies. Our grant was renewed about a year and a half ago to focus on the quality of genetic testing, because one of the issues that emerged from our analysis of reproductive technologies was that there were concerns about how genetic testing was regulated currently. As law and policy director, I have been overseeing the research and analysis around the regulation of genetic technologies.

GT: In your testimony, you called the regulatory framework for genetic tests “incoherent and inadequate.” Why is that?

Javitt: By incoherent, I mean that there are arbitrary regulatory distinctions that are not based on levels of risk of a genetic test. Currently for the vast majority of genetic tests, there is not a regulatory body that reviews the quality of the test — by which I mean the analytic and clinical validity of the test — before it is marketed.

The system is set up so that laboratories register [under the CLIA statute] and they have to go through inspection and they are subject to quality control requirements: record keeping and similar kinds of requirements. It’s sort of like what the Board of Health does when they inspect a kitchen, although obviously with more complexity. There’s no pre-market requirement for any specific test — but the lab as part of being a lab has to have data on file demonstrating analytical validity.

For tests that are considered high complexity, meaning they’re more complicated to perform or interpret, there are additional requirements that [the Centers for Medicare and Medicaid Services] has imposed under CLIA. In particular what we have focused on is proficiency testing. That is essentially a pop quiz to the laboratory, or the laboratory can give itself, to show that laboratories can reliably get the right answer.

The mechanism by which the agency implements that requirement is through what’s called specialty areas — there are specialty areas for a wide variety of laboratory tests but there has been no specialty area developed for most genetic tests. In part that’s due to the fact that when the first group of specialty areas was set up back in the early ’90s, genetic testing was in its infancy; however, that does not serve as an excuse to never do it. As a result there is no specific requirement that genetic testing laboratories perform proficiency testing in order to comply with CLIA.

We believe that the enforcement of CLIA has been inadequate with respect to genetic testing laboratories. I should add as a caveat that many laboratories go beyond the CLIA minimum by enrolling in certification by external bodies, but that is not a requirement.

GT: You’ve also spoken about clinical validity issues.

Javitt: Not only does the laboratory know what it’s doing and can it get the right answer — that’s a very important piece of the puzzle, but it’s not the only piece. The other piece of the puzzle is, does this test provide value to the physician and the patients? Does it actually correlate to the patient’s current or future condition? Just because you can find a variant reliably over time doesn’t mean that variant has any clinical significance; that’s called clinical validity.

For the most part, with very few exceptions, there is no pre-market review of  laboratory tests to determine whether there’s sufficient evidence of clinical validity. To put that in context, why that is, is that most genetic tests are offered as services of the laboratory. They receive a patient’s specimen and they develop a test using reagents and report a result. Historically, while FDA has stated that it falls within their jurisdiction, they have chosen not to regulate it.

The only exception is laboratories that instead of making the test from scratch can instead purchase test kits — that’s prepackaged reagents and instructions from a manufacturer. Those prepackaged kits are regulated as medical devices by FDA.  There have been a handful of kits that have gone through this regulatory process. A real life example of how this dichotomy between laboratory-developed tests and test kits plays out is there’s an approved test for cystic fibrosis that tests mutations associated with that disease — [and] there are many other laboratory-developed tests for cystic fibrosis that have not gone through that process. So by “incoherent” I mean there are two different systems with vastly different regulatory requirements with the same end product, which is a result of a test to a patient.

GT: But FDA is testing some regulatory processes on IVDMIAs, right? How does that work?

Javitt: While they have traditionally been hands-off, FDA has said, ‘We want to carve off one type of laboratory-developed test known as the IVDMIA for more scrutiny — to bring it into our tent.’ FDA’s rationale, as stated in their draft guidance, is that because the test employs an algorithm — it takes data from the test and uses an analytical algorithm to come up with a result to treat or not treat, or treat in a certain way — and because what goes on in the algorithm is not transparent to the end user, i.e., the clinician, FDA believes they require more scrutiny. Because it’s not possible to know how the company or laboratory came up with that answer — they’re using a proprietary methodology to get there — FDA feels that because there is no transparency they need to give it more scrutiny. I do think also that the fact that it is giving a treat/don’t treat or you use it to prescribe a certain medication is a significant part, so it really comes very close to the core mission of the agency.

IVDMIAs have been singled out by FDA for regulation when the vast majority of laboratory tests, including genetic tests, are not currently regulated by FDA. What we would like to see is a system where there is transparency in information regarding all genetic tests. That said, not all genetic tests are going to require the same amount of scrutiny and the same amount of data — but at a bare minimum there should be information about all tests out there and who is doing them and what is the information on which the test is based.

GT: You talked about the incentive companies have to actually not participate in the regulatory process. What should be done about that?

Javitt: What I meant by that was by allowing two paths, one which requires pre-market review and one which does not, you are creating a disincentive to go through pre-market review. There should be one path. Now that path may be a graduated path; just because you’re all on the same path doesn’t mean you have to march exactly the same way. FDA’s medical device regulation system is already tiered based on risk, so that’s not a new concept. Right now tests for the same indication are subject to vastly different levels of oversight based solely on the source of the ingredients the laboratory chooses to perform the test.

GT: What steps would you like to see FDA take to accomplish that?

Javitt: We are very glad that they are focusing on the whole issue of laboratory-developed tests. There needs to be more thought and more discussion with all of the interested parties about how you design a rational system that gets at the good that you want to achieve without overdoing it. I don’t think that there’s one answer to how they get to that endpoint.

The Scan

Review of Approval Process

Stat News reports the Department for Health and Human Services' Office of the Inspector General is to investigate FDA's approval of Biogen's Alzheimer's disease drug.

Not Quite Right

A new analysis has found hundreds of studies with incorrect nucleotide sequences reported in their methods, according to Nature News.

CRISPR and mRNA Together

Time magazine reports on the use of mRNA to deliver CRISPR machinery.

Nature Papers Present Smartphone Platform for DNA Diagnosis of Malaria, Mouse Lines for Epigenomic Editing

In Nature this week: a low-cost tool to detect infectious diseases like malaria, and more.