The Northern District of California has just given us a great example of the use of Daubert in connection with summary judgment to get rid of cases that never should have been filed. You know the type. These are the cases where the plaintiffs plead a good game in connection with some prescription drug, but then cannot come up with reliable expert opinions that the drug actually could have caused their alleged injuries, which should lead to judgment in favor of the drug manufacturer.
That is what happened in Wendell v. Johnson & Johnson, No. C 09-4124 CW, 2014 U.S. Dist. LEXIS 89576 (N.D. Cal. June 30 2014), although it took far too long for the case to reach that result. Wendell was a wrongful death case where the decedent started treatment for inflammatory bowel disease at age 12, including a six-mercaptopurine (“6MP”) drug called Purinethol, initially on its own and later in combination with two different anti-tumor necrosis factor (“anti-TNF”) drugs. After eight years of drug therapy for his IBD, this unfortunate young man developed a rare and aggressive form of cancer known as hepatosplenic T-cell lymphoma (“HSTCL”), which led to his rapid and untimely demise.
Two years later – in July 2009 – his parents sued the manufacturers of the drugs he used, eventually alleging in a Fourth Amended Complaint in April 2011 that he had developed HSTCL as a result of taking his combination of drugs and that the drug manufacturers failed to warn adequately about the risks of taking 6MP drugs with anti-TNF drugs.
Over the next three years, all but one defendant got out of the case (including the defendant in the case name), either through summary judgment or settlement, leaving only one manufacturer of Purinethol to fight the battle resulting in the order that caught our attention.
And a good fight it was. The manufacturer argued first that the plaintiffs had no evidence sufficient to show that Purinethol, either alone or in combination with anti-TNF drugs, caused the decedent’s cancer. The district court agreed and held that the opinions given by the plaintiffs’ two causation experts were not based on sufficiently reliable scientific evidence and thus were not admissible under Rule 702 and Daubert.
Each of the court’s reasons is helpful. First, the district court noted that neither doctor had ever conducted any independent research or published any studies on the relationship between 6MP and anti-TNF drugs and the development of the HSTCL disease. Moreover,
Although both conclusorily stated during their depositions that their opinions on this subject were based on a reasonable degree of medical certainty, they also conceded that their opinions would not satisfy the standards required for publication in peer-reviewed medical journals. . . . The fact that both of Plaintiffs’ causation experts are reluctant to publish their opinions – and appear to have developed their opinions specifically for the purposed of this litigation – casts doubt on the reliability of their methodologies.
Id. at **12-13. Let’s unpack this statement, which we like a lot. To start, the court rejected the plaintiffs’ experts’ use of magic words – i.e., that they held their opinions to a “reasonable degree of medical certainty.” We have seen these words in many versions – reasonable degree of scientific certainty, reasonable degree of economic certainty, and so on – but merely saying something does not make it so in federal court, or anywhere else. The district court also carefully defined the issue on which the experts needed to opine – the relationship of these particular drugs to this kind of cancer. In other words, the court held the plaintiffs to the causation issue they actually had to prove at trial.
We particularly like the district court’s treatment of the doctors’ admitted reluctance to publish their opinions and their concessions that the opinion would not withstand peer review. We see these kinds of litigation-driven opinions with some regularity, and we are disappointed when courts admit the opinions and allow litigation to lead science, which should not occur. See, for example, a Drug and Device Law Blog favorite case, Rosen v. Ciba-Geigy Corp., 78 F.3d 316, 319 (7th Cir. 1996) (“Law lags science; it does not lead it.”). This district court saw it our way.
Second, the district court observed that the experts failed to identify any studies showing a causal link between HSTCL and the drugs at issue, and the court’s reasoning here is again very helpful. The plaintiffs admitted that no such studies existed, but argued that such studies are impossible because HSTCL is “exceedingly rare.” Hmm, could this be the plaintiffs attempting to shift the burden to the defendant to disprove causation? We think so, and the district court correctly rejected the plaintiffs’ explanation:
The difficulty of conducting these studies . . . does not relieve Plaintiffs of their burden to present evidence of causation. Indeed, the need for such evidence is especially important here in light of the fact that more than seventy percent of observed HSTCL cases are idiopathic, meaning they have no known cause. [Citing Plaintiffs’ experts]. Given this high rate of idiopathic cases, Plaintiffs cannot reasonably eliminate other causes of [the decedent’s] HSTCL without some reliable evidence of a positive link between the drugs at issue and the disease.
Id. at **14-15. Cases are more challenging when the disease at issue is most likely idiopathic, but this district recognized correctly that such a case is challenging especially for plaintiffs, who bear the burden of proving causation at all times. We think the lack of reliable scientific evidence should lead to the exclusion of expert opinions in all cases, but we cannot disagree with the district court’s view that the lack of relevant epidemiology is particularly damning when most cases have no known cause. The studies and case reports that these plaintiffs actually cited were not on point: None purported to show that the “specific combination of drugs . . . actually causes HSTCL,” and none demonstrated any statistically significant results or accounted for plausible alternative causes of HSTCL. Id. at *17.
The experts’ opinions therefore did not hold up, which was alone sufficient to grant summary judgment. The district court, however, held also that the plaintiffs’ claims failed because the prescribing physician did not recall reading the drug labeling and that it was not his “regular practice” to look at drug labeling. Id. at *20. Although the district court couched this holding in terms of a “duty to warn,” we call this a failure of warnings causation. Because the prescribing physician did not read the warnings, any purported inadequacy in the warnings could not have contributed to his prescribing decision and thus could not have caused the decedent’s disease. Moreover, the district court properly placed the burden again on the plaintiffs: “Plaintiffs have not identified any evidence to contradict this testimony or otherwise suggest that Dr. Rich actually relied on the [drug] warning label. Accordingly, they cannot prevail on their negligence claim under a failure-to-warn theory.” Id.
We will file this order away for future use and will track it on appeal, if there is one. After five years of litigation (and apparently a few settlements), these plaintiffs may have had enough, and if they fold their tent and go home, we will not blame them. Their case in end was a waste of time and eventually came to the correct result for the last defendant standing.