Opinion
117852/04.
Decided on May 15, 2009.
Andrew G. Finkelstein, Finkelstein Partners, LLP, for Plaintiffs.
Beth L. Kaufman, Christopher Milito, Schoeman, Updike Kaufman, LLP, for Defendants.
Plaintiffs in this mass tort litigation seek damages for personal injuries allegedly sustained as a result of defendants' wrongful marketing of the prescription drug Neurontin (generically known as gabapentin) for off-label uses such as the treatment of bipolar disorder. Plaintiffs plead products liability causes of action alleging that Neurontin caused plaintiffs or their decedents to undergo suicide-related events, including suicidal ideation, suicide attempt, and completed suicide. Defendants Pfizer Inc., Warner-Lambert Company LLC, Parke-Davis, a division of Warner-Lambert Company, and Warner-Lambert Company (collectively "Pfizer") move to preclude the testimony of plaintiffs' experts on the issue of general causation — that is, the issue of whether Neurontin is capable of causing suicide-related injuries.
This court set the motion down for a Frye hearing which was held jointly with a Daubert hearing ordered by United States District Judge Patti B. Saris in the Neurontin federal Multi-District Litigation ("MDL"). The federal court has issued an exhaustive opinion denying defendants' motion to preclude. ( In Re Neurontin Marketing, Sales Practices, and Products Liability Litigation, US Dist Ct, D Mass, 04-10981-PBS [Saris, J.], May 5, 2009) ("federal opinion"). This court adopts the federal court's analysis of and findings on the reliability of plaintiffs' methodology and conclusions on the issue of general causation. The detailed analysis and findings will not be repeated here. Rather, this opinion will address the Frye standard applicable to determination of defendants' motion in the state cases, and this court's reasons for holding, as it now does, that defendants' motion to preclude should be denied under that standard.
Judge Saris graciously hosted the joint Frye/Daubert hearing and invited this court's full participation in the hearing. There are currently approximately 288 cases in the Neurontin state litigation and 244 cases in the Neurontin federal products liability MDL. The hearing was held jointly in recognition of the importance of coordinating related federal and state litigations in order to reduce costs and delays. ( See Manual for Complex Litigation [Fourth] §§ 20, 20.31.) In addition, with the knowledge and consent of the parties, the federal and state courts conferred on the issues raised in the Frye/Daubert hearing. Judge Saris afforded this court the opportunity to review the opinion before its issuance. Thus, this court was in the position to advise Judge Saris that it independently agreed with the findings and analysis of the federal opinion on the reliability of plaintiffs' methodology and conclusions, and that the state opinion would concur in such findings and analysis.
In both state and federal products liability actions involving exposure to an allegedly harmful substance, the plaintiff must prove both "general causation" — i.e., "that the toxin [or other chemical] is capable of causing the particular illness," and "specific causation" — i.e., that the exposure was sufficient to cause the particular plaintiff's illness. ( See Parker v Mobil Oil Corp. , 7 NY3d 434, 448, rearg denied 8 NY3d 828, citing McClain v Metabolife Intl., Inc., 401 F3d 1233, 1241 [11th Cir 2005].) It is well settled that "[t]he introduction of novel scientific evidence calls for a determination of its reliability." ( Parker, 7 NY3d at 446.) New York courts continue to adhere to the Frye standard in making this determination ( id. at 447 n 3), while federal courts apply the Daubert standard. Under both standards, the plaintiff has the burden of proving the admissibility of the expert evidence on general causation. ( See Fraser v 301-52 Townhouse Corp. , 57 AD3d 416 [1st Dept 2008].)
The Frye standard was first articulated in Frye v United States ( 293 F 1013 [DC Cir 1923]), a case involving a novel scientific technique (a lie detector test). The Court held that "while courts will go a long way in admitting expert testimony deduced from a well-recognized scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field in which it belongs." ( Id. at 1014.) The Court of Appeals has characterized this standard as requiring a determination of "whether the accepted techniques, when properly performed, generate results accepted as reliable within the scientific community generally." ( People v Wesley, 83 NY2d 417, 422 [reliability of then new technique — DNA evidence]. Accord Parker, 7 NY3d at 446.)
The Daubert standard was articulated in Daubert v Merrell Dow Pharmaceuticals, Inc. ( 509 US 579), a case involving an expert opinion on whether a pharmaceutical drug was capable of causing a certain injury. The Court there held that the federal Rules of Evidence superseded the Frye standard, and that the trial judge must perform a "gatekeeping role" of "ensuring that an expert's testimony both rests on a reliable foundation and is relevant to the task at hand." ( Id. at 589, 597.) Daubert also articulates four factors, none of which is determinative, for a trial court to consider in evaluating reliability: 1) whether the theory or technique can be or has been tested; 2) whether the theory or technique has been subjected to peer review and publication; 3) in the case of a technique, the known or potential rate of error; and 4) whether the theory has met with "general acceptance" in the relevant scientific community. ( See id. at 593-594.) Thus, Daubert holds that the general acceptance standard of Frye is no longer applicable in the federal courts but "can yet have a bearing on the inquiry" as to reliability. ( Id. at 594.)
Conversely, notwithstanding New York's adherence to the Frye standard, a Daubert-type analysis of the plaintiff's expert's methodology is relevant where the scientific issue is not whether a novel scientific technique should be held admissible, but whether the methodology employed by the plaintiff's expert leads to a reliable theory or opinion on causation. In Parker, the Court of Appeals clarified the applicability of the Frye standard to an expert's opinion on whether a toxin is capable of causing a particular disease (there, whether exposure to benzene in gasoline caused the plaintiff to develop a form of leukemia). Plaintiff offered the opinions of epidemiologists and toxicologists on causation. The Parker Court found that the question was "whether the methodologies employed by [plaintiff's] experts lead to a reliable result," and that "[t]here is no particular novel methodology at issue for which the Court needs to determine whether there is general acceptance. Thus, the inquiry here is more akin to whether there is an appropriate foundation for the experts' opinions, rather than whether the opinions are admissible under Frye." ( 7 NY3d at 447.) The Court also expressly noted that where the issue before the trial court is the reliability of the experts' methodology, federal cases evaluating reliability under Daubert are "instructive." ( Id. at 448 n 4.)
In People v Wesley ( 83 NY2d 417, supra), the Court had similarly explained: "While foundation concerns itself with the adequacy of the specific procedures used to generate the particular evidence to be admitted, the test pursuant to Frye [full citation omitted] poses the more elemental question of whether the accepted techniques, when properly performed, generate results accepted as reliable within the scientific community generally." ( Id. at 422.) "Once Frye has been satisfied, the question is whether the accepted techniques were employed by the experts in this case. The focus moves from the general reliability concerns of Frye to the specific reliability of the procedures followed to generate the evidence proffered and whether they establish a foundation for the reception of the evidence at trial." ( Id. at 429 [internal citation and quotation marks omitted].)
Cases considering the admissibility of a plaintiff's theory of causation, both before and after Parker, have adopted the formulation that the plaintiff's burden is to prove that the plaintiff's "expert's theory is generally accepted" in the relevant scientific community. ( Lara v New York City Health Hosps. Corp., 305 AD2d 106 [1st Dept 2003] [theory that precipitous delivery can cause infant cerebral palsy]; Marsh v Smyth , 12 AD3d 307 [1st Dept 2004] [theory that hyperabduction of arm was cause of nerve palsy]; Pauling v Orentreich Med. Gp. , 14 AD3d 357 [1st Dept 2005], lv denied 4 NY3d 710 [theory that facial injections of silicone can cause silicone toxicity]; Heckstall v Pincus ,19 AD3d 203 [1st Dept 2005] [theory that Bupropion, a smoking cessation aid, can cause arrhythmia]; Marso v Novak , 42 AD3d 377 [1st Dept 2007], lv denied ___ NY3d ___ [2009] [theory that bradycardia was a risk factor for stroke]; Leffler v Feld , 51 AD3d 410 [1st Dept 2008] [theory that Altace can cause pemphigus]; Fraser v 301-52 Townhouse Corp. , 57 AD3d 416, supra [theory that mold can cause respiratory diseases].)
After Parker, if the methodology is not novel and the issue is whether the methodology leads to a reliable theory of causation, the theory should arguably be scrutinized not under Frye for general acceptance, but under foundational principles for reliability. Significantly, however, even assuming that the Frye standard is applicable to a novel causation theory based on an established methodology, the courts have consistently held that the Frye requirement that the theory be generally accepted does not impose a rule "that a jury may hear only theories that are either conclusively established' by the scientific literature or unanimously supported by the scientific authorities." ( Fraser, 57 AD3d at 418 n 2 [internal brackets omitted].) Thus, "general acceptance does not necessarily mean that a majority of the scientists involved subscribe to the conclusion. Rather it means that those espousing the theory or opinion have followed generally accepted scientific principles and methodology in evaluating clinical data to reach their conclusions." ( Zito v Zabarsky , 28 AD3d 42, 44 [2d Dept 2006] [causation opinion on whether Zocor can cause an autoimmune disease], quoting Matter of Rezulin Litigation [Frances Beck] v Warner-Lambert Co., 2002 NY Slip Op 40431[U] * 6-7 [Sup Ct, NY County] [Freedman, J.] [causation opinion on whether Rezulin can cause cirrhosis of the liver]]. Accord Matter of Bextra Celebrex, 2008 NY Misc Lexis 720 [Sup Ct, NY County] [Kornreich, J.] [causation opinion on whether Celebrex increases risk of heart attacks or strokes at various doses].)
In a pre- Parker case in which the plaintiffs offered the expert opinions of epidemiologists and toxicologists to prove that exposure to toxins in a landfill caused cancers, this Department held that neither the deductions of the experts, nor their methodologies in reaching their conclusions, are "premised on the type of novel science' implicating the concerns articulated in Frye." ( See Nonnon v City of New York , 32 AD3d 91, 103 [1st Dept 2006] [opinion by Mazzarelli, J.], affd on other grounds 9 NY3d 825 [2007].) See also Marsh v Smyth , 12 AD3d 307, supra [Saxe, J., concurring].) Post- Parker cases have continued to hold, or left open the possibility, that novel opinions on causation should be scrutinized at a Frye hearing rather than in a foundational inquiry undertaken at trial. ( See Leffler, 51 AD3d at 410; Fraser, 57 AD3d at 418.)
As explained by the Court of Appeals, the Frye standard "emphasize[s] counting scientists' votes, rather than . . . verifying the soundness of a scientific conclusion." ( Parker, 7 NY3d at 447, quoting Wesley ( 83 NY2d at 439 [Kaye, C.J. concurring].) Thus, the purpose of a Frye hearing is not to decide which experts' conclusions are correct ( Marsh, 12 AD3d at 311 [Saxe, J., concurring]), but "to determine whether the experts' deductions are based on principles that are sufficiently established to have gained general acceptance as reliable." ( Id. at 308 [majority opinion].) Or, as originally stated in Frye and reaffirmed in Wesley, the court must determine whether the experts' deductions are "based upon a scientific principle or procedure which has been sufficiently established to have gained general acceptance in the particular field in which it belongs." ( Marso v Novak, 42 AD3d at 378, quoting Wesley, 83 NY2d at 423.) Further, as Parker cautions, the court must guard against the danger of allowing unreliable information or "junk science" to go before a jury, without imposing "an insurmountable standard that would effectively deprive toxic tort plaintiffs of their day in court." ( 7 NY3d at 447.)
Applying the Frye standard, the court holds that plaintiffs have met their burden of proving that Neurontin (gabapentin) has the capacity to cause suicide-related events. The parties' dispute on this motion focuses on the reliability of plaintiffs' evidence of a statistically significant association between gabapentin and suicide-related events, and of plaintiffs' biological plausibility theory of causation.
In support of their claim that a statistically significant association exists, plaintiffs rely on a study conducted by the Food and Drug Administration ("FDA") on whether the use of antiepileptic drugs led to an elevated risk of suicidality. On January 31, 2008, in response to this study, the FDA issued an Alert, entitled "Information for Healthcare Professionals — Suicidality and Antiepileptic Drugs," finding that patients receiving antiepileptic drugs, including gabapentin, had approximately twice the risk of suicidal behavior or ideation compared to patients receiving placebo. The FDA also issued a Statistical Review and Evaluation: AntiEpileptic Drugs and Suicidality, dated May 23, 2008 ("FDA study"), describing its methodology and analysis in evaluating the collected data on 11 antiepileptic drugs, and concluding that antiepileptic drugs are associated with increased risk of suicidality relative to placebo and that the effect appears consistent among the 11 drugs that were studied.
As discussed at length in the federal opinion, epidemiology is a branch of medicine that focuses on general causation — whether an agent is capable of causing a disease. (Federal Opinion at 10-15.) It is well recognized that "epidemiology itself is certainly not novel" and that epidemiological evidence is a "primary generally accepted methodology for demonstrating a causal relation between a chemical compound and a set of symptoms or a disease." ( Nonnon, 32 AD3d at 104 [internal quotation marks and citation omitted]. See generally Parker v Mobil Oil Corp. , 7 NY3d 434, supra.)
A useful discussion of epidemiology and epidemiologic studies is found in the Reference Guide on Epidemiology, contained in the Reference Manual on Scientific Evidence (Fed. Judicial Ctr., 2d ed 2000), a publication of the federal courts designed to assist judges in evaluating complex scientific evidence.
Here, defendants argue that the FDA study, a meta-analysis, was methodologically flawed because it was driven by two antiepileptic drugs (lamotrigine and topiramate), and that there is no independent evidence establishing a statistically significant association between gabapentin and suicidality. Plaintiffs' experts counter that the antiepileptic drugs were properly pooled, and that placebo-controlled studies on which the FDA study was based were designed for other purposes and were too small and contained too few patients at high risk for suicidality to establish a link between gabapentin alone and suicidality. ( See Federal Opinion at 40-50 [summarizing the parties' positions on the FDA study].) After a hearing in July 2008, at which defendants presented the same objections to the pooling of the 11 drugs, two FDA advisory committees, composed of eminent researchers and scientists, voted to affirm the findings of the FDA study. (Federal Opinion at 34-36.) The FDA subsequently issued an Updated Alert, dated December 16, 2008, requiring all manufacturers of antiepileptic drugs, including Neurontin, to include a warning in the drug label of the risk of suicidal thoughts or behavior.
This court concurs in the federal court's finding "that the FDA study is an epidemiological study that may be considered on the question of whether the Plaintiffs have produced evidence of an association between Neurontin and suicidality." (Federal Opinion at 40.) In so holding, the court recognizes that "standards promulgated by regulatory agencies as protective measures are inadequate to demonstrate legal causation." ( Parker, 7 NY3d at 450. See also McClain, 401 F3d at 1250 [same under Daubert].) Here, however, plaintiffs rely not on the FDA's Alert or subsequent decision to issue a warning, but on the underlying study performed by FDA scientists in support of those actions. Moreover, as the federal opinion states and other courts have observed: "Suicide presents researchers seeking to study it with both ethical and practical difficulties. As [suicide is] a rare event, studying it for purposes of causation requires a high number of participants. . . . Perhaps more important, researchers have ethical qualms over treating risky' patients with a placebo." ( Giles v Wyeth Inc., 500 F Supp 2d 1048, 1058 [SD Ill 2007]; Federal Opinion at 12.) It is undisputed that, whether for practical or ethical reasons, no large scale, randomized, placebo-controlled studies, designed specifically to test for an association between gabapentin and suicidality, have been performed. Under these circumstances, the absence of statistically significant information about gabapentin alone is not dispositive. ( See Federal Opinion at 50; Selig v Pfizer, Inc., 290 AD2d 319, 320 [1st Dept 2002] [holding that, in the absence of "any clinical data" to support causal link between drug (Viagra) and disease, plaintiffs could (but failed to) set forth "other scientific evidence" showing causal link].)
As to the reliability of the meta-analysis, while the parties' experts dispute whether the mechanism of action of gabapentin is GABAergic or otherwise comparable to that of the other antiepileptic drugs in the FDA study, all of the experts agree that the mechanism of action of gabapentin is not fully understood. Plaintiffs and defendants have both retained eminent experts whose opinions as to the effects of gabapentin are based on a large number of peer reviewed
in vitro and in vivo studies of animals and humans, the relevance of which the experts, however, dispute. ( See Federal Opinion at 53-54, 57, 61-71 [discussing studies].) As held above, the court's task on this motion is not to decide which of the experts' conclusions are correct but, rather, whether the conclusions have been reached in a methodologically reliable fashion. ( See supra at 6-7; Ruiz-Troche v Pepsi Cola of Puerto Rico Bottling Co., 161 F3d 77, 85 [1st Cir 1998] [same under Daubert].) The court finds that plaintiffs have submitted sufficiently reliable evidence that gabapentin is analogous to the other antiepileptic drugs in the FDA study, and that pooling and extrapolation are "scientifically sound and methodologically reliable" from a chemical and pharmacological standpoint. ( See Federal Opinion at 58 [and analysis at 53-58].)Finally, in order to demonstrate the causality of the association shown in the FDA study between gabapentin and suicidality, plaintiffs proffer a theory of biological plausibility. This theory is that gabapentin, whether by increasing GABA in the brain or by binding to the alpha-2-delta subunit of the calcium ion channel, decreases serotonin and other monoamine neurotransmitters; and, further, that the decrease in the monoamines prompts behavioral disturbances, depression, and suicidality. The theory that decreases in the monoamines can cause mood disturbances leading to suicide is well supported in the scientific literature and has gained widespread acceptance in the relevant field of biological psychiatry. ( See Federal Opinion at 72-78 [summarizing literature].) The parties thus focused at the hearing on the scientific evidence that gabapentin causes decreases in monoamine release.
It is well settled than an association demonstrated by an epidemiological study is not equivalent to causation and that an identified association must be evaluated by researchers to determine whether the association is causal. (Reference Manual on Scientific Evidence at 336, 374.)
As framed by the parties, their dispute over whether gabapentin can cause suicide-related events did not turn on "dose response relationship" or the level of exposure at which the drug produces illness — an issue that can be significant in products liability cases involving pharmaceutical drugs. ( See generally Parker v Mobil Oil Corp. , 7 NY3d 434, supra; Matter of Bextra Celebrex, 2008 NY Misc Lexis 720, supra.)
As indicated above, this evidence consisted largely of animal and human studies published in peer reviewed articles. The dispute between the parties' causation experts, while simplified here in lay terms, is a highly esoteric one about the relevance of the various studies to demonstrate causation. For example, defendants concede that increased GABA in the nuclei where serotonin originates reduces the rate of serotonin release, and they acknowledge that spectroscopy studies in humans, cited by plaintiffs, show that gabapentin increases whole brain GABA. However, they dispute the relevance of these studies because the studies do not distinguish between intra and extra-cellular GABA. ( See Federal Opinion at 61-62). Defendants also acknowledge that several in vitro studies cited by plaintiffs show that gabapentin inhibits or decreases monoamine release in stimulated rat brain slices treated with gabapentin. However, they dispute the extent of the decrease shown by the studies and criticize these studies on the basis, among others, that the degree of stimulation used in the studies would not occur in living beings. ( Id. at 62-64.) Defendants, for their part, cite studies with humans (the Ben-Menachem studies) which they claim show that gabapentin in fact increased the level of monoamines, or that there were no significant differences in the monoamine levels of patients before and after treatment with gabapentin. Plaintiffs counter with an explanation of the results of these studies based on the study models, and also disagree with defendants' interpretation of the results. ( Id. at 65-69.) Notably also, while defendants' expert offers cogent criticisms of plaintiffs' reliance on the in vitro studies, his report acknowledges that "[t]his type of test system has been used by numerous research teams for more than 30 years as a model to study various physiological and pharmacological aspects of neurotransmitter release" at monoamine-containing neurons. (Report of Charles Taylor, dated Dec. 20, 2007, at 18). Plaintiffs also point to acknowledgments in defendants' internal documents that "[g]abapentin has been known to reduce monoamine neurotransmitter release for many years." (Ps.' Sur-Reply, Ex. 5 [2001 Internal Project Operating Plan].) Defendants claim, however, that "additional studies" are needed to demonstrate the effect of gabapentin on the release of monoamine neurotransmitters in an intact organism. (Taylor Report at 20.)
The parties' principal biological causation experts were plaintiffs' expert Dr. Michael Trimble, a professor of behavioral neurology, whose publications include a textbook on Biological Psychiatry, and defendants' expert Dr. Charles Taylor, a neuroscientist who, for over 20 years until 2007, was the head preclinical pharmacologist on the gabapentin development team for defendants. Both experts submitted reports and testified at the hearing. Their impressive qualifications are further detailed in the federal opinion. ( Id. at 7 n 7, 9 n 11.)
While defendants assert that further study is needed of gabapentin's effects, neither Frye nor Daubert requires that an expert's causation opinion be based on conclusively established data. It is well recognized that "[t]rained experts commonly extrapolate from existing data." ( General Elec. Co. v Joiner, 522 US 136, 146.) Absent clinical data conclusively supporting a theory of medical causation, what is required is that the plaintiff "set forth other scientific evidence based on accepted principles showing such a causal link." ( Selig, 290 AD2d at 320.) Expert opinion must be supported by scientific data and must not be "connected to existing data only by the ipse dixit of the expert." ( Marsh, 12 AD3d at 312 [Saxe, J. concurring], citing Joiner, 522 US at 146.) However, "[i]t is sufficient if a synthesis of various studies or cases reasonably permits the conclusion reached by the plaintiff's expert." ( Zito, 28 AD3d at 44, quoting Marsh, 12 AD3d at 313.)
An expert opinion on causation will be excluded where it is unsupported by any scientific studies or medical literature or where the literature is plainly insufficient to support the opinion. ( See e.g. Fraser, 57 AD3d at 418 [precluding expert's theory that mold caused respiratory illness on the ground, among others, that " none of the medical literature in the record supports the stated position of plaintiffs' expert"]; Heckstall, 19 AD3d at 205 [precluding expert's opinion where plaintiff presented "no clinical or epidemiological data or peer reviews" linking the drug to the disease, and supported claim of causation solely with case reports]; Marsh, 12 AD3d at 313 [Saxe, J. concurring] [noting that many of the Court's Frye cases have precluded testimony based upon "a complete absence of literature or studies supporting the claim"]; Selig, 290 AD2d at 320 [precluding expert's opinion where the record lacked any clinical data showing a causal link between the drug and the disease, and any scientific literature supporting the expert's theory as to the mechanism of action of the drug.) Not surprisingly, the expert opinion will also be precluded if the opinion has been rejected by the relevant scientific community. ( See e.g. Marso, 42 AD3d at 378 [expert's causation opinion should be rejected where "there is a generally or widely held view in the scientific community rejecting such conclusions outright"]; Pauling, 14 AD3d at 358 [precluding expert's opinion where there was "no supporting medical literature whatsoever" for the opinion, and reputable scientific institutions "have issued statements that there is no evidence" of a relationship between the drug and the disease].)
That is not the case here. Rather, as discussed above and at length in the federal opinion, plaintiffs present an extensive body of both epidemiological and peer reviewed scientific literature that is sufficient to raise a triable issue of fact as to whether Neurontin causes suicide-related events. While the parties' experts vigorously disagree on the interpretation of the existing literature, their dispute over the validity of plaintiffs' theory "falls squarely within the range where experts might reasonably differ.'" (Federal Opinion at 72, citing Kumho Tire Co. v Carmichael, 426 US 137, 153.) The issue of general causation should therefore be submitted to the jury.
For the foregoing reasons, defendants' motion to preclude is denied.