In re: ONGLYZA (SAXAGLIPTIN) AND KOMBIGLYZE (SAXAGLIPTIN AND METFORMIN) PRODUCTS LIABILITY LITIGATION - MDL 2809.
No. 22-6078
UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT
February 13, 2024
24a0028p.06
RECOMMENDED FOR PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b)
LEATHA TAYLOR, individually and as administratrix of the estate of David Taylor, et al., Plaintiffs-Appellants, v. BRISTOL-MYERS SQUIBB COMPANY; ASTRAZENECA PHARMACEUTICALS LP; IPR PHARMACEUTICALS INC.; MCKESSON CORPORATION; ASTRAZENECA LP, Defendants-Appellees.
ORIGINATING CASES
| 18-cv-00053-57 | 18-cv-00127-32 | 18-cv-00250 | 18-cv-00320-21 | 18-cv-00578-79 |
| 18-cv-00062-63 | 18-cv-00134 | 18-cv-00252-53 | 18-cv-00325-26 | 18-cv-00596 |
| 18-cv-00065-67 | 18-cv-00139 | 18-cv-00256 | 18-cv-00328 | 18-cv-00598 |
| 18-cv-00070 | 18-cv-00145 | 18-cv-00258 | 18-cv-00330-31 | 18-cv-00634 |
| 18-cv-00072-74 | 18-cv-00148 | 18-cv-00260 | 18-cv-00333 | 18-cv-00648 |
| 18-cv-00076-78 | 18-cv-00150-51 | 18-cv-00266 | 18-cv-00336-39 | 19-cv-00023 |
| 18-cv-00080-82 | 18-cv-00154-55 | 18-cv-00268-71 | 18-cv-00341-45 | 19-cv-00025-26 |
| 18-cv-00085 | 18-cv-00162 | 18-cv-00279 | 18-cv-00347-49 | 19-cv-00055 |
| 18-cv-00087 | 18-cv-00167 | 18-cv-00281 | 18-cv-00354-56 | 19-cv-00068 |
| 18-cv-00089-91 | 18-cv-00170 | 18-cv-00285 | 18-cv-00372-75 | 19-cv-00070-71 |
| 18-cv-00094 | 18-cv-00180 | 18-cv-00287 | 18-cv-00394 | 19-cv-00076 |
| 18-cv-00098-01 | 18-cv-00217 | 18-cv-00297-98 | 18-cv-00448-49 | 19-cv-00078 |
| 18-cv-00104-05 | 18-cv-00219 | 18-cv-00300-02 | 18-cv-00496 | 19-cv-00089-92 |
| 18-cv-00107-08 | 18-cv-00223 | 18-cv-00308 | 18-cv-00505 | 19-cv-00095 |
| 18-cv-00110-11 | 18-cv-00233-34 | 18-cv-00312 | 18-cv-00562 | 19-cv-00112 |
| 18-cv-00116 | 18-cv-00237-39 | 18-cv-00314 | 18-cv-00568 | 19-cv-00122-23 |
| 18-cv-00118-24 | 18-cv-00247 | 18-cv-00316 | 18-cv-00571-72 | 19-cv-00125-30 |
| 19-cv-00134 | 19-cv-00218 | 19-cv-00398 | 20-cv-00124 | 20-cv-00340 |
| 19-cv-00138-45 | 19-cv-00221 | 19-cv-00401 | 20-cv-00141 | 20-cv-00383 |
| 19-cv-00147-56 | 19-cv-00223 | 19-cv-00403 | 20-cv-00143 | 20-cv-00405 |
| 19-cv-00161 | 19-cv-00227 | 19-cv-00422 | 20-cv-00153 | 20-cv-00419 |
| 19-cv-00166 | 19-cv-00230-32 | 19-cv-00439 | 20-cv-00206-08 | 20-cv-00438 |
| 19-cv-00169 | 19-cv-00252 | 19-cv-00461-63 | 20-cv-00236 | |
| 19-cv-00173 | 19-cv-00287 | 19-cv-00499 | 20-cv-00247 | |
| 19-cv-00190 | 19-cv-00350 | 20-cv-00007 | 20-cv-00281 | |
| 19-cv-00194 | 19-cv-00393 | 20-cv-00072-73 | 20-cv-00330 |
Argued: December 7, 2023
Decided and Filed: February 13, 2024
Before: SILER, NALBANDIAN, and DAVIS, Circuit Judges.
COUNSEL
ARGUED: Ashton Rose Smith, MOORE LAW GROUP, PLLC, Louisville, Kentucky, for Appellants. Paul W. Schmidt, COVINGTON & BURLING LLP, New York, New York, for Appellees. ON BRIEF: Ashton Rose Smith, Jennifer A. Moore, MOORE LAW GROUP, PLLC, Louisville, Kentucky, for Appellants. Paul W. Schmidt, COVINGTON & BURLING LLP, New York, New York, Carol Dan Browning, STITES & HARBISON PLLC, Louisville, Kentucky, Emily S. Ullman, COVINGTON & BURLING LLP, Washington, D.C., for Appellees.
OPINION
NALBANDIAN, Circuit Judge. In this multi-district litigation, plaintiffs claim that saxagliptin, a diabetes drug, caused their heart failure. During expert discovery, plaintiffs presented a single expert to show the drug can cause heart failure. After a Daubert hearing and expert motions, the district court excluded the expert, finding that methodological flaws rendered his testimony unreliable under
On appeal, plaintiffs challenge the district court‘s exclusion of their expert, its grant of summary judgment, and its refusal to give plaintiffs more time to find another expert witness. Because we conclude all three claims lack merit, we AFFIRM.
Defendants (Bristol-Myers Squibb, AstraZeneca, and McKesson) manufacture and sell FDA-approved type 2 diabetes drugs containing saxagliptin, a dipeptidyl-peptidase-4 (DPP-4) inhibitor.
In 2008, the FDA‘s Endocrinologic and Metabolic Drugs Advisory Committee recommended more clinical studies evaluating the link between diabetes drugs and cardiovascular risk.
SAVOR (“Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus“), a randomized-controlled trial (RCT), was commissioned in response. SAVOR “randomly assigned 16,492 patients with type 2 diabetes who had a history of, or were at risk for, cardiovascular events to receive saxagliptin or placebo and followed them for a median of 2.1 years.” Benjamin M. Scirica et al., Saxagliptin and Cardiovascular Outcomes in Patients with Type 2 Diabetes Mellitus, 369 New Eng. J. Med. 1317, 1317 (2013). SAVOR examined cardiovascular risk on a primary endpoint (“a composite of cardiovascular death, myocardial infarction, or ischemic stroke“), as well as a secondary endpoint (“a composite of cardiovascular death, myocardial infarction, stroke, hospitalization for unstable angina, coronary revascularization, or heart failure“). Id. SAVOR found no statistically significant difference between the groups for the primary endpoint or for five components of the secondary endpoint. Id. at 1322.
But the study did find a statistically significant difference between saxagliptin and a placebo on hospitalizations for heart failure (HHF). Id. at 1317, 1321–22. The study observed a statistically significant 27% increase in hospitalization for heart failure rates in patients administered saxagliptin compared to patients receiving a placebo (3.5% vs. 2.8%). Id. at 1317. But the study cautioned that the observed association “was unexpected and should be considered within the context of multiple testing that may have resulted in a false positive.” Id. at 1324. The study also noted that the HHF “finding merits further investigation and needs to be confirmed in other ongoing studies, and a class effect should not be presumed.” Id.
Saxagliptin‘s drug label was updated in response. The warning, directed at prescribing physicians, states:
In a cardiovascular outcomes trial enrolling participants with established ASCVD [atherosclerotic cardiovascular disease] or multiple risk factors for ASCVD (SAVOR trial), more patients randomized to [saxagliptin] (289/8280, 3.5%) were hospitalized for heart failure compared to patients randomized to placebo (228/8212, 2.8%). In a time-to-first-event analysis the risk of hospitalization for heart failure was higher in the [saxagliptin] group (estimated Hazard Ratio: 1.27, 95% CI: 1.07, 1.51).
R.749-2, Prescribing Information at 3, PageID 17379.
II.
SAVOR‘s HHF finding sparked this multi-district litigation. Plaintiffs sued defendants in federal courts across the country, asserting claims for strict product liability, negligence, failure to warn, breach of warranty of merchantability, and breach of express and implied warranties, all stemming from heart failure allegedly caused by saxagliptin. In February 2018, the Judicial Panel on Multidistrict Litigation consolidated 41 actions pending in 23 districts (as well as 43 potential tag-along actions in 26 districts), transferring them to the Eastern District of Kentucky as MDL 2809.1
infarction and/or cardiovascular injury.” R.179, Case Management Order No. 1 at 1, PageID 1049.
Plaintiffs presented a single general causation expert: Dr. Parag Goyal. Dr. Goyal, a cardiology professor at Weill Cornell Medicine was to testify that “it is more likely than not that saxagliptin is capable of causing heart failure.” R.633-10, Goyal Report at 14, PageID 12297.
Defendants moved to exclude Dr. Goyal, arguing that he: (1) unreliably based his causation finding on SAVOR alone while ignoring all subsequent human data, (2) unreliably used animal data, and (3) unreliably applied the Bradford Hill criteria.2
The district court conducted a Daubert hearing on Dr. Goyal‘s testimony on August 10, 2021, before granting defendants’ motion to exclude Dr. Goyal‘s testimony on January 5, 2022.3
After Dr. Goyal‘s exclusion, defendants moved for summary judgment, arguing that Dr. Goyal‘s exclusion meant that plaintiffs now lacked admissible evidence raising a triable issue of material fact on general causation. Plaintiffs opposed this motion, contending other evidence in the record was sufficient to create a disputed material fact as to general causation. Meanwhile, plaintiffs moved to modify the scheduling order to allow them another ninety days to identify a replacement general causation expert for Dr. Goyal.
On August 2, 2022, the district court denied plaintiffs’ motion for more time to identify a substitute general causation expert, concluding they could not show “good cause” to modify a scheduling order as required by
In the same order, the district court granted defendants’ motion for summary judgment, concluding that plaintiffs failed to produce admissible expert testimony that saxagliptin can cause heart failure.
Plaintiffs timely appealed, arguing the district court erred thrice: (1) by excluding Dr. Goyal, (2) by granting summary judgment, and (3) by refusing to grant plaintiffs more time to find a substitute general causation expert.
III.
We review the district court‘s decision to exclude Dr. Goyal‘s expert testimony
The party proffering the expert (here plaintiffs) bears the burden of showing by a preponderance of the evidence that the expert satisfies
A.
First, Dr. Goyal‘s reliance on SAVOR, to the exclusion of all other studies involving human data, was unreliable.
As Dr. Goyal conceded, no clinical study beyond SAVOR has found a statistically significant association between saxagliptin and heart failure. Instead, four post-SAVOR observational studies, collectively following 175,000 saxagliptin users, found no association. Dr. Goyal conceded that these observational studies were “reasonably designed” and that “the number of patients” was “a strength.” R.710, Tr. of Dr. Goyal‘s Daubert Hearing at 90–91, PageID 16078–79. Even so, Dr. Goyal‘s report failed to engage with these studies, dismissing them as “generally limited due to issues related to confounding.” R.633-10, Goyal Report at 9, PageID 12292. But Dr. Goyal identified no specific issues or confounders in the observational studies. Dr. Goyal‘s failure “to adequately account for contrary evidence is not reliable or scientifically sound.” In re Lipitor (Atorvastatin Calcium) Mktg., Sales Pracs. & Prods. Liab. Litig., 174 F. Supp. 3d 911, 932 (D.S.C. 2016).
Yet, despite admitting that basing causation on a single study is improper, Dr. Goyal still found causation based only on SAVOR. Indeed, Dr. Goyal stated that SAVOR‘s findings “should be interpreted as cause-and-effect unless there is compelling evidence to prove otherwise.” R.633-10, Goyal Report at 8, PageID 12291. But Dr. Goyal‘s conclusion that SAVOR shows causation in the absence of “compelling evidence” showing otherwise reverses the burden of proof. See Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997) (Exclusion is proper for “opinion evidence that is connected to existing data only by the ipse dixit of
So Dr. Goyal drew “unauthorized conclusions from limited data—conclusions the authors of the study d[id] not make,” betraying a “lack of scientific rigor.” McClain v. Metabolife Int‘l, Inc., 401 F.3d 1233, 1248 (11th Cir. 2005). By ignoring all other human studies, besides SAVOR, without an adequate explanation, Dr. Goyal failed to base his opinion on “sufficient facts or data.”
Plaintiffs raise two main arguments on appeal. Neither is persuasive. First, plaintiffs claim that the district court “erroneously concluded that a study—including an RCT—must be replicated to be reliable.” Appellant Br. at 21.5 Not so. The district court did not exclude Dr. Goyal based on a per se rule against studies that still haven‘t been replicated. Indeed, the district court clarified that Dr. Goyal‘s views on replication came into play because “there are multiple studies on the issue of whether saxagliptin can cause heart failure.” R.740, Experts Order at 31 n.3, PageID 17089. And the court concluded that Dr. Goyal acted unreliably by agreeing on the importance of replication but, without explanation, ignoring all human studies beyond SAVOR.
Second, plaintiffs claim that the district court “improperly concluded that Dr. Goyal could not make a causation finding based on an epidemiological study ... that does not itself make a causal determination.” Appellant Br. at 21 (citing In re Bair Hugger Forced Air Warming Devices Prods. Liab. Litig., 9 F.4th 768, 779 (8th Cir. 2021)).6 Again, not so. The district court did not fault Dr. Goyal for considering an epidemiological study. It faulted him for basing his opinion on SAVOR‘s finding without adequately explaining how he inferred a causal relationship from SAVOR, an epidemiological study that did not come to a conclusion about causation.
According to the district court, Dr. Goyal‘s sin was methodological inconsistency, not reliance on an un-replicated epidemiological study.
The district court did not abuse its discretion in concluding that Dr. Goyal‘s reliance
B.
Second, Dr. Goyal‘s use of animal data was unreliable because he has no expertise in interpreting animal studies.
Dr. Goyal conceded multiple times he is unqualified to analyze animal studies. He specifically conceded he was unqualified to conclude either that the animals in the studies he cited “actually had heart failure” or whether “saxagliptin is capable of causing heart failure in animals.” R.710, Hearing Tr. at 148, PageID 16136. Further, Dr. Goyal admitted to ignoring multiple peer-reviewed animal studies dispelling a causal link between saxagliptin and heart failure. And, since none of the studies diagnosed animals with heart failure, Dr. Goyal simply claimed that particular animals showed symptoms of heart failure, such as anemia or congestion. Since Dr. Goyal was admittedly unqualified to make these diagnoses, the district court did not err in finding Dr. Goyal‘s use of animal studies unreliable.
C.
Third, Dr. Goyal did not reliably apply Bradford Hill.
Courts must ensure both that expert opinions are “the product of reliable principles and methods,” and that these methods have been “reliably applied” to the “facts of the case.”
On the one hand, Dr. Goyal cherry-picked data to bolster his case. When considering the analogy factor, Dr. Goyal looked at thiazolidinedione (TZD) drugs (a different kind of second-line diabetes medication), rather than DPP-4 inhibitors (the class of drugs to which saxagliptin belongs). Dr. Goyal explained this unconventional choice simply by claiming that TZDs “provide[] an appropriate analogy” because they have been shown to “worsen heart failure.” R.633-10, Goyal Report at 14, PageID 12297. This cherry-picking “undermines principles of the scientific method and is a quintessential example of applying methodologies (valid or otherwise) in an unreliable fashion.” In re Lipitor (Atorvastatin Calcium) Mktg., Sales Pracs. & Prod. Liab. Litig. (No II) MDL 2502, 892 F.3d 624, 634 (4th Cir. 2018).
On the other hand, Dr. Goyal inconsistently applied several Bradford Hill factors. For instance, Dr. Goyal‘s report claimed that two of the factors (specificity and biological gradient) were satisfied, but he later testified that they were not. These changes in Dr. Goyal‘s Bradford Hill analysis—neither explained nor justified—cast doubt on the reliability of his testimony.
Plaintiffs don‘t defend Dr. Goyal‘s application of Bradford Hill per se. Instead, they contend that exclusion was improper because a “jury, not the trial judge, must evaluate and weigh conflicting
***
In sum, the district court had three good reasons to find Dr. Goyal‘s testimony unreliable: (1) improper reliance on SAVOR to the exclusion of all other human studies, (2) unqualified analysis of animal studies, and (3) cherry-picking and inconsistent consideration of the Bradford Hill factors. So we conclude that excluding Dr. Goyal was not an abuse of its discretion.
IV.
Plaintiffs claim that, even if Dr. Goyal were properly excluded, summary judgment was improper because “there is extensive other evidence from which a jury could properly determine that saxagliptin can cause heart failure.” Appellant Br. at 19.
We review a district court‘s grant of summary judgment de novo. Baggs v. Eagle-Pitcher Indus., Inc., 957 F.2d 268, 271 (6th Cir. 1992). Summary judgment is appropriate only where “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.”
As a preliminary point, plaintiffs claim that the district court erred by finding they cannot establish general causation without expert testimony. Instead, plaintiffs argue that “[j]uries are permitted to decide issues of causation without guidance from experts.” Appellant Br. at 40.
The district court concluded that, as an MDL, the issue was governed by “the substantive state law of the transferor state.” R.769, Summary Judgment Order at 14, PageID 18470; see, e.g., In re Ford Motor Co. Ignition Switch Prods. Liab. Litig., 194 F.R.D. 484, 490 (D.N.J. 2000).8 So the court reviewed the law of all fifty states, concluding that all states require “the
plaintiff in cases involving complex issues of medical causation to present expert testimony on the subject.” R.769, Summary Judgment Order at 14, PageID 18470 (citing R.746-2, Jurisdiction Survey of the Laws on Causation in Product Liability Actions).
The district court does not stand alone: other district courts have agreed that all jurisdictions require expert testimony to show general causation, “at least
While plaintiffs contest this requirement, they identify no state without a requirement of expert testimony to establish general causation in complex medical cases. Instead, plaintiffs resort to selectively quoting cases. Most important is a decision of this circuit, which plaintiffs claim affirmed a district court‘s finding that causation could be shown “based simply on the warning language in product inserts.” Appellant Br. at 40–41 (citing Meridia Prods. Liab. Litig. v. Abbott Lab‘ys, 447 F.3d 861, 866 (6th Cir. 2006)). But plaintiffs incorrectly characterize the district court opinion: it merely “assume[d] arguendo that no states’ laws erect” a requirement of expert testimony to establish general causation, because “[r]egardless, the Court ends up in the same place: summary judgment for Defendants.” In re Meridia Prods. Liab. Litig., 328 F. Supp. 2d 791, 802 (N.D. Ohio 2004). Plaintiffs’ string citation to four state-court cases is similarly inapposite and misleading.
So we conclude that the district court did not err in finding that all jurisdictions require expert testimony to show general causation in complex medical cases such as this MDL. Dr. Goyal‘s exclusion warranted the district court‘s grant of summary judgment to defendants.
V.
For their third and final issue on appeal, plaintiffs argue that the district court abused its discretion when it refused to grant them leave to identify a new general causation expert.
District courts may modify scheduling orders “only for good cause.”
We have specified that district courts should measure “good cause” mainly by reference to two factors. “[T]he moving party‘s diligence in attempting to meet the case management order‘s requirements” is the “primary measure,” but “possible prejudice to the party opposing the modification” is also relevant. Inge v. Rock Fin. Corp., 281 F.3d 613, 625 (6th Cir. 2002). Both factors cut against plaintiffs.
First, plaintiffs’ diligence in attempting to meet the scheduling order. Plaintiffs are right that they “were diligent and timely in identifying Dr. Goyal.” Appellant Br. at 57. But that misses the point. As the district court noted, “plaintiffs do not really request an extension of deadlines,” since expert discovery and motions
Second, prejudice to defendants. Plaintiffs claim that granting their motion would cause defendants “no prejudice” because of the “minimal amount of delay.” Appellant Br. at 58. But granting plaintiffs’ request would essentially restart expert discovery, requiring depositions, briefing, hearings, and motions on plaintiffs’ new expert. This would delay the MDL‘s resolution for years—just consider that plaintiffs requested three months to simply identify an expert. So granting the motion would likely impose significant costs on defendants in the form of substantial legal expenses and years of delay. See Com. Benefits Grp., Inc. v. McKesson Corp., 326 F. App‘x 369, 377–78 (6th Cir. 2009) (“[A]ny further delay in discovery would have resulted in additional time and expense incurred by both the parties and the court and would have unfairly prejudiced defendants.“).
We conclude the district court did not abuse its discretion by denying plaintiffs’ request for more time to identify a general causation expert to replace Dr. Goyal. The court properly refused to reward plaintiffs for their failure to identify a reliable general causation expert by imposing significant costs on defendants. To find otherwise would set a precedent that parties may drag out litigation by identifying “only one expert witness on a crucial issue who is later excluded.” R.769, Summary Judgment Order at 11, PageID 18467.
VI.
For these reasons, we AFFIRM.