On July 31st, the FDA announced its intention to regulate both laboratory developed tests (LDT”s) and in vitro diagnostic (IVD) companion devices, and it will soon be asking for public comment on the proposed regulations. Should genetic counselors be among the people commenting? Well yes, as the new rules are likely to affect genetics practice, since many of the tests that look at genetic susceptibility are LDT’s and could be subject to a premarket review by the FDA that will delay or deny the clinical availability of new tests, and a mandatory process of adverse result reporting. The impact will be felt most immediately in cancer settings, where genetic tests that look at tumor DNA for purposes of choosing targeted therapies or predicting prognosis are likely first candidates to draw FDA scrutiny, but eventually the new rules should affect a range of clinical specialties. At stake is finding the balance between too much regulation, wherein it becomes prohibitively difficult and expensive to introduce new tests that can help diagnose patients and personalize recommendations for screening and treatment, and too little regulation, wherein we suspect that our information on the accuracy and reliability of new tests is not adequately accurate or reliable (an ongoing issue, by the way, with non-invasive prenatal testing. See Katie Stoll’s post here and a study, new this week, suggesting that the dreaded false positive result may be more common than test makers have led us to believe).
A little background on the two closely related entities that are the focus of new regulations. LDT’s are what used to be called “home-brews”: tests that are used by a single lab and not marketed as a kit or a device. Somewhat by historical accident, LDT’s have come to exist in a regulatory grey area, effectively exempt from FDA oversight. The assumption behind this was that what went on in an individual lab affected only that lab’s patients and that no agency could track every one-off solution engineered by a mom-and-pop lab. As with everything else in 2014, the status quo has been disrupted by new technology – but in this instance the new technology isn’t the magic of Google or whole genome sequencing but overnight shipping. Yes, the world of genetic testing has been turned on its ear by the likes of UPS and Fed Ex.
In brief, now that the Pony Express has picked up its game, laboratories can test samples from all over the world in centralized locations with sophisticated and expensive testing capability that isn’t available back on the farm. At the same time, lab tests, including genetic tests and biomarkers like measures of gene expression, play an increasingly important role in making diagnoses and determining treatment. For this reason, the FDA has moved in its determined yet glacial manner to regulate a subset of tests that are considered high or medium risk – those tests which have the potential to alter medical care, and therefore have significant implications if the information they provide is incorrect. This risk-based approach is a measured step – it allows the FDA to continue to use discretion when tests are low risk or experimental or involve a rare disease for which there is no other test.
IVD companion diagnostics are tests developed to be used in conjunction with a drug or other therapy – tests that can be used to refine dosages or identify good candidates for a given therapy. Obviously pharmacogenetics is a subset of this broader category of companion testing. Again, the proposed regulatory framework would stratify the tests as high risk, moderate risk, low risk – requiring pre-market approval for higher risk tests, and allowing the agency to exercise “discretion” in low risk situations (discretion is FDA-speak for a wink and a nod). With regard to IVD diagnostics, the FDA intends not just that the tests on offer be confirmed as reliable, but is instituting the requirement that companion testing be included in the development of new therapies as a matter of course. In effect the government is mandating that all new therapies be individualized to the greatest extent possible: “When an appropriate scientific rationale supports such an approach, FDA encourages the joint development of therapeutic products and diagnostic devices that are essential for the safe and effective use of those therapeutic products.” The age of personalized medicine is upon us, and the FDA is ON IT.
If all this sounds familiar, it only means that you have been paying attention. Since 2010, the FDA has been asserting publically that it has both the intention and the authority to regulate LDT’s and IVD’s. Going back even further, the Genomics and Personalized Medicine Act of 2006, introduced by then Senator Barack Obama, emphasized the development of companion diagnostics, calling on the National Academy of Sciences to recommend incentives and requiring the Institute of Medicine to improve “oversight and regulation of genetic tests.” While the bill was never passed, it is not surprising to see a similar emphasis under the current administration.
So, genetic counselors, are we for or against the proposed regulations? Probably the answer to that question is — yes. Like the FDA, most people seem to be in favor of some middle option – regulating everything is virtually impossible and regulating nothing is an appealing libertarian fantasy, but in fact it would put counselors in the uncomfortable position of having to rely on figures supplied by the companies who manufacture the tests. Careful observers like the Genetics and Public Policy Center have been calling for increased oversight for genetic testing for years. Their 2006 summary of a genetic testing quality initiative sums it up this way:
“assessment of public attitudes shows that the public widely believes that the government regulates genetic tests to ensure their quality and, moreover, that the government should play this role. In fact, however, genetic tests are subject to very little governmental oversight when compared to other health care products. There is no formal approval procedure a laboratory has to go through before offering a new genetic test, and government requirements to ensure that genetic testing laboratories are getting the right answers to patients are minimal. Moreover, there is no government requirement that a test must be clinically valid – i.e., actually relate to a particular disease or risk of disease – in order to be sold.”
However, both the American Clinical Laboratory Association and the American Medical Association have reacted negatively to the proposed FDA regulatory strategy. The ACLA pushback comes as no surprise – few entities welcome idea of FDA regulation – and the organization has submitted a petition claiming that only CLIA and not the FDA had authority over LDT’s (the FDA rejects this). The more measured response of the AMA reflects the concerns of clinical care-givers, and may align with the attitude of many genetic counselors:
The draft FDA Framework for Oversight of Laboratory Developed Tests (LDTs) announced today, outlines a risk-based approach that raises a number of questions and concerns.
The FDA proposal adds an additional layer of regulatory requirements which may result in patients losing access to timely life-saving diagnostic services and hinder advancements in the practice of medicine.
The AMA is committed to ensuring that the proposal that is ultimately adopted by the FDA preserves rapid access to care and medical advancements.
What makes it difficult to respond to the FDA is that there is a lot of wiggle room left in the regulations as written. High and moderate risk tests will be required to report adverse results and apply for pre-market review according to separate timetables – but the FDA will not define those terms for up to 2 years after the regulations are finalized (Policy and Medicine has a useful chart if you are looking for specifics on timelines). In other words, the FDA has designed a system that gives them room to maneuver – and is asking for respondents to give feedback on the plan without knowing where the agency plans to draw the line. For example, breast cancer susceptibility panels probably aren’t low risk; they are medically actionable and complicated to interpret. Are they high risk or moderate risk? The somewhat hyperbolic letter from the FDA to 23andMe last fall* suggested that the agency believes the fallout from breast cancer risk prediction done badly might be unnecessary mastectomies. That sounds pretty high risk – but is that the perceived reality of counselors who work with these tests?
The rare disease exemption in the FDA plan means that whole exome or whole genome sequencing would not be affected, in those cases where the patient presents with an apparently genetic condition that has eluded diagnosis. WES for those with no apparent disease, who wish to use the information prophylactically? I have literally no idea what risk the FDA would assign to clinical versions of genome scanning. What about the genetic testing done for children with autism? These supplement rather than point to a diagnosis and would rarely change treatment but may have a big impact on the parents reproductive choices – is that consequence enough to bump a test from low risk to high risk?
I might sound like I am criticizing the FDA, but in fact I am sympathetic to the difficulties inherent in a modulated approach and appreciate that they are attempting to tread that knife’s edge. I do think it makes it difficult to provide feedback, and I would suggest that their policy be reopened for public comment at critical junctures, such as the point at which high, low and moderate risk categories are more carefully defined. Useful commentary now, I would suggest, will need to be far more granular than the FDA regulatory language itself. What tests do you feel work well for you and your patients? Are there tests in use or in the pipeline that concern you? Which ones? Why? Share your concerns here, and I will write up a response incorporating reader response when the draft regulations are posted for public comment.
*Note: don’t bother telling the FDA that you are concerned about direct-to-consumer testing, because the agency has already noted that this applies only to testing in a clinical context. No DTC testing will be exempt from review – a footnote to the FDA’s announcement that had DTC advocates screaming foul – for details see Jennifer Wagner’s irritated response at the Genomics Law Report.