Recently, Bexis attended the DRI drug and device committee spring conference. Among other things he heard a bang-up presentation on genomics and personalized (also known as “precision”) medicine from Paige Sensenbrenner. On that same day, co-blogger Steve Boranian alerted Bexis to a new defense argument in asbestos/mesothelioma cases that also utilizes genomics – certain mutations in a gene called “BAP1” – to identify persons at greater risk of idiopathic (that is, not related to asbestos) mesothelioma. Here’s a link to that article. A verifiable alternative cause could be a game-changer for asbestos litigation. The statement we quoted back in 2009, uttered by the first person ever to have his genome individually sequenced, that “individual genes are just not very informative,” appears in the process of being disproven by ongoing scientific events.
Both items, as informative as they were on scientific facts, were rather short on the law. That’s where we come in. We thought we’d take a look at what law exists concerning the intersection of pharmacogenomics, personalized medicine, and prescription medical product liability litigation. We’ve touched on these issues back in 2011, when we blogged about Mills v. Bristol-Myers Squibb Co., 2011 WL 4708850 (D. Ariz. Oct. 7, 2011), one of the first cases in which the plaintiff made allegations about pharmacogenomically-based risks. Back then we said:
The plaintiff is claiming that a drug is defective, not because of anything inherent in the drug itself, but solely because it is less effective (and therefore has a different risk-benefit profile) due to the plaintiff’s peculiar genetic makeup. Essentially, the allegations seek to impose a non-FDA-approved contraindication, using state law, based upon human genetic variability. With advances in computer technology making genetic testing exponentially cheaper and more detailed as times passes (see Moore’s law), more and more genetic variability in the efficacy of prescription drugs is bound to be discovered. Eventually – certainly within some of our lifetimes – we’ll be able to carry our entire individual genetic code around with us on a chip, should we so choose....
The complaint in Mills is a bare genetic susceptibility claim, frankly based on an allegation of “variant” genetic characteristics shared by only a minority of the population. In our view, unless and until – and only to the extent that – the FDA decides to assess drug approvals and contraindications on the basis of genetic subgrouping, this type of tort claim should not be recognized, because it is flatly contrary to the criteria by which the intended uses of drugs are currently determined. Claims such as in Mills, which are at loggerheads with FDA criteria for drug development, are precisely those with the most potential for making pharmaceutical manufacturers into “sitting ducks” for litigation, in this instance litigation based on extraneous genetic factors.
It may well be that the coming (and to some extent existing) revolution in genetically individualized medical therapy will require changes in how drugs are evaluated, labeled, etc., but this is a singularity-driven issue that needs to be addressed by the policy branches of our government, and not haphazardly in product liability litigation.
(Emphasis added).
FDA Regulation of Pharmacogenomic Information
We still feel the same way, but now with the caveat that the FDA is starting to lay the regulatory foundation for drug warnings in the coming age of personalized medicine. Way back in 2005, the FDA allowed voluntary addition of pharmacogenomic information in drug labeling:
The pharmacogenomic data and resulting test or tests may be intended to be included in the drug labeling to choose a dose and dose schedule, to identify patients at risk, or to identify patient responders. Inclusion of a pharmacogenomic test in the labeling would be contingent upon its performance characteristics.
FDA, Guidance for Industry, Pharmacogenomic Data Submissions, at 5 (March 2005) (available here). Under the heading “Clinical Pharmacology,” drug labeling can now contain a subsection specifically devoted to pharmacogenomics. As of last year, over 100 drugs contained such information, according to the FDA.
Since our post in 2011, the FDA has released several additional guidance documents in this field. This one, the most basic, simply defines the relevant terminology, particularly in the area of clinical investigations. This one addresses how to incorporate genetic variation into the design and implementation of clinical investigations. This one concerns inventing tools to identify biomarkers in the course of drug development. This one concerns how to report pharmacogenomic data to the Agency. The FDA’s online pharmacogenomics resources offer links to scads of other stuff. We don’t purport to be either scientists or regulatory lawyers, so now we’ll turn to the underlying purpose of this post, which is the legal precedent involving pharmacogenomics and product liability.
Pharmacogenomic Product Liability Claims
Pharmacogenomics is a double-edged sword. There will be some cases in which the results of genetic testing might assist plaintiffs (for example, in the absence of an adequate warning) and others in which the results of genetic testing will indisputably aid defendants – such as the mesothelioma alternative cause scenario mentioned above:
In the ideal plaintiff’s case, a person with an allele that made him or her specifically susceptible to the action of some toxin would be exposed to that toxin, which would cause a unique and detectable biochemical change, which in turn would be shown to cause an extremely high likelihood of contracting the plaintiff’s disease. The ideal defendant’s case might occur in several ways: similar biomarker evidence would point a finger at a purely genetic cause or at some other (perhaps voluntary or non-anthropogenic) exposure; or, a person exposed to a toxin known to cause the person’s disease in susceptible people might have a gene that completely neutralized the toxic effect and also might lack a biomarker that is uniformly found in people whose disease was caused by exposure.
Steve C. Gold, “The More We Know, the Less Intelligent We Are? − How Genomic Information Should, and Should Not, Change Toxic Tort Causation Doctrine,” 34 Harv. Envtl. L. Rev. 369, 392 (2010) (footnote omitted).
The plaintiffs’ side has sporadically argued that genetic markers should be warned about or, in some cases designed around, although we doubt the latter is even possible. So far many plaintiffs have had trouble coming up with factual support to back such allegations – and sometimes we’re not even sure why they’re making them. In the latter category we place Newman v. McNeil Consumer Healthcare, 2013 WL 9936293 (N.D. Ill. March 29, 2013), a case we blogged about a couple of years ago. Newman involved SJS/TEN, the autoimmune diseases (or different forms of the same disease) Stevens Johnson Syndrome and Toxic Epidural Necrosis. SJS/TEN is somewhat analogous in our sandbox to mesothelioma in asbestos cases, since plaintiffs often work backwards from the diagnosis to look for some drug company to sue. We mentioned in a recent post the wide variety of drugs that plaintiffs have claimed to cause these conditions. Perhaps to bolster the often weak SJS/TEN causation testimony, in Newman the plaintiff’s expert sought to discuss pharmacogenomics. However, the testimony about purported genetic predisposition to the disease was rejected as speculative, because no genetic link to SJS/TEN has yet been discovered:
Defendants are correct that [the expert] testimony on the subject would be speculative and irrelevant. First of all, Plaintiffs’ argument admits that the relevance and helpfulness of the information is conditioned on the discovery of a genetic link, which may not happen. Secondly, even if such a link were discovered, Plaintiffs fail to explain how it rebuts Defendants claim that SJS/TEN was unpredictable during the relevant time frame.
Id. at *8. As of the time of trial, the state of the art did not include a genetic marker for SJS/TEN. “That SJS/TEN may be more predictable in the future if a particular discovery is made says nothing about Defendants’ negligence.” Id. While the genetic testimony in Newman was offered by the plaintiff, should an SJS/TEN marker be discovered, we believe that overall such a development would help defendants far more than plaintiffs by exonerating most, and maybe all – if the marker is not drug related – of the plethora of drugs that plaintiffs have alleged as causative agents.
Then there’s the Mills case itself. Mills was similar to Newman in that the plaintiff once again made baseless allegations about possibly carrying a genetic marker. In Mills, the plaintiff claimed that, due to a variant gene (“CYP”), she could not metabolize the defendant’s drug as well as most other people. The presence of this adverse genetic marker purportedly supported a design defect theory:
Plaintiff alleges that the chemical structure of [the drug] is defective because it carries a higher risk of adverse events for patients who carry the genetic variant CYP, who are poor metabolizers of the drug. Plaintiff contends that [the drug] is the proximate cause of her injuries because, “[u]pon information and belief,” she is a CYP carrier.
Mills, 2011 WL 4708850, at *2. As in Newman, the court in Mills did not let the plaintiff get away with this subterfuge. A plaintiff’s own genetic markers, and genome generally, is something the plaintiff is uniquely in possession of. “Plaintiff’s genetic makeup is a fact solely within her control. Tests are available that can reveal whether plaintiff in fact possesses CYP.” Id. Where the plaintiff is asserting a claim based on pharmacogenomics, s/he is responsible for producing the evidence to prove it. Id.
Similar – and similarly unsupported – allegations of “genetic...
