Pharmacogenetic Tests: US FDA Tries Off-Label Approach To Guide Consumers

US FDA’s latest effort to provide guidance for consumers on pharmacogenomics tests underscores the challenges the agency faces in the absence of a clear regulatory framework for lab-developed tests – and with no easy mechanism to update labels for off-patent drugs.

The US FDA's new pharmacogenomics table has opened the door ever so slightly on off-label test data.

The US Food & Drug Administration is trying a different approach to cataloguing pharmacogenetic test information in an effort to steer prescribers and consumers away from unproven associations and towards ones that the agency sees as providing useful information.

On 20 February, FDA announced a new catalogue of pharmacogenetic tests intended to provide “the agency's view of the state of the current science in pharmacogenetics.” The agency is publishing a table “that describes some of the gene-drug interactions for which we believe there is sufficient scientific evidence to support the described associations between certain genetic variants, or genetic variant-inferred phenotypes, and altered drug metabolism, and in certain cases, differential therapeutic effects, including differences in risks of adverse events.”

The agency’s announcement of the effort includes an explicit call for updated authority over lab-developed tests – but also (perhaps inadvertently) illustrates the case for enhanced resources to update old product labels by listing at least three drug-gene interactions in the new table that are not in the FDA-approved label.

“Many pharmacogenetic tests are being offered as laboratory developed tests (LDTs),” FDA said, adding that it “has, in an exercise of enforcement discretion, generally not enforced applicable regulatory requirements for LDTs, such as premarket review.”

However, “while the landscape of LDTs has changed, the agency's oversight framework has remained the same. The FDA is committed to continuing to work with Congress on a broader legislative solution to the oversight of in vitro clinical tests generally (including LDTs), which would modernize our regulation of these tests.”

The agency has worked with outside stakeholders and Congress for several years on model legislation to try to achieve that goal – so far without reaching consensus. (Also see "FDA And The Transition: Reading The LDT Leaves" - Pink Sheet, 24 Nov, 2016.)

“In the meantime, the FDA should not and cannot stand idly by when safety issues arise,” the agency’s statement continues. “Consistent with our mission to protect and promote public health, we believe it is important to take steps now to help ensure that claims being made for pharmacogenetic tests offered today are grounded in sound science to avoid inappropriate management of patients' medications – and to do so through approaches that both protect patients and advance the development of analytically and clinically validated pharmacogenetic tests.”

The agency frames the effort as following specifically on a 2018 warning about “FDA’s concerns with firms offering genetic tests making claims about how to use the genetic test results to manage medication treatment that are not supported by recommendations in the FDA-approved drug labeling or other scientific evidence.”

CDER/CDRH Collaboration

The new resource is a collaboration between the Center for Devices & Radiologic Health and the Center for Drug Evaluation & Research, but amounts to an effort by CDRH to repurpose information already published by CDER.

The CDRH-hosted table is formatted differently from CDER’s existing catalogue of pharmacogenomics markers included in drug labels. Both lists are organized alphabetically by drug (non-proprietary) name, and both include the identity of the genetic marker of interest. However, where CDER presents the drug class and then the affected sections of the approved product label, the CDRH table identifies the “affected subgroup” (i.e., “poor metabolizers” or a specific allele to identify) and then a description of the drug-gene interaction and brief statement of the implications (i.e., consider a lower dose, or results in higher risk of a given side effect).

The CDRH table also divides the list of markers into three groups:

1. Those “for which the data support therapeutic management recommendations”;

2. Those “for which the data indicate a potential impact on safety or response”; and

3. Those “for which the data demonstrate a potential impact on pharmacokinetic properties only.”

“The fact that FDA has included a particular gene-drug interaction in the table does not necessarily mean FDA advocates using a pharmacogenetic test before prescribing the corresponding medication, unless the test is a companion diagnostic,” the website notes.

More importantly, the list identifies by exclusion those tests that FDA does not deem reliable: “If no statements related to efficacy or toxicity are provided, the scientific evidence FDA reviewed was considered insufficient to support such associations.”

Three ‘Unapproved’ Claims Listed

CDRH’s list is also far shorter than CDER, with just over 100 entries (106) compared to nearly 400 on the CDER site. Some of that is formatting: CDRH mostly (but not completely) avoids listing the same drug twice, even when there is more than one pharmacogenetics market of interest. Primarily, however, it appears to reflect CDRH’s focus on tests other than companion diagnostics likely to be identified in the indication/contraindication section of the label.

CDRH’s list also includes three drugs that do not have labeling statements about the referenced test. All three – tacrolimus, simvastatin, and allopurinol – are off-patent products.

“The table posted today includes certain established gene-drug interactions that appear in FDA-approved drug labeling,” the agency says.

“Recognizing, however, that not all supported gene-drug interactions may be found in current FDA labeling, the table also includes some additional gene-drug interactions that are consistent with the current FDA labeling and for which there is sufficient scientific evidence based on published literature. This literature-based scientific evidence is often used in support of the recommendations found in professional guidelines used by clinicians.”

The three are:

• Tacrolimus (originally approved for organ transplant patients as Prograf) is listed in the first category of gene-drug associations where the data support therapeutic management recommendations. The specific gene cited is CYP3A5, and the agency says that “intermediate or normal metabolizers” may have “lower systemic concentrations” of the drug and therefore “lower probability of achieving target concentrations.” The agency advises providers to “measure drug concentrations and adjust dosage based on trough whole blood tacrolimus concentrations.” There is extensive literature published on the CYP3A5 metabolizer impact on tacrolimus. However, the FDA approved label does not mention the issue; it does include extensive advice about the risk of over-exposure (and side effects) when co-prescribed with CYP3A inhibitors and reduce availability when co-prescribed with CYP3A4 (as opposed to CYP3A5) inducers.

• Simvastatin (originally approved for LDL lowering as Zocor) is listed in the second group as having a “potential impact on safety or response.” The specific gene cited is SLC01B1, and the affected groups are those with 521 TC or 521 CC variants. The impact is higher systemic concentrations leading to increased risk of myopathy. The genetic marker is not referenced in the FDA-approved label.

• Allopurinol (originally approved for gout as Zyloprim) is also in the second category on the CDRH website, with the genetic marker cited as HLA-B. Those who are 58:01 allele positive have a higher risk of severe skin reactions, the agency says. Again, that information is not reflected in the FDA approved label.

“The table is not complete,” FDA stresses, “but we felt it is important to provide clinicians, patients and test developers with information now while we continue to review the scientific evidence, including the scientific evidence underlying various professional guidelines, such as those of the Clinical Pharmacogenetics Implementation Consortium. We anticipate updating this web-based resource periodically as the evidence evolves.”

To support that effort, FDA has opened a docket for public comment. The agency wants “all stakeholders to provide their feedback on specific pharmacogenetic associations that should or should not be included as the agency continues to update this table. Feedback should include supporting rationale and underlying evidence that supports the pharmacogenetic association.”