NOT FOR PUBLICATION
UNITED STATES COURT OF APPEALS
FILED
FOR THE NINTH CIRCUIT
DEC 06 2017
MOLLY C. DWYER, CLERK
U.S. COURT OF APPEALS
In re: INCRETIN-BASED THERAPIES No. 15-56997
PRODUCTS LIABILITY LITIGATION,
______________________________ D.C. No.
3:13-md-02452-AJB-MDD
JEAN ADAMS, On Behalf of Herself and
All Other Similarly Situated Plaintiffs,
MEMORANDUM*
Plaintiff-Appellant,
v.
MERCK SHARP & DOHME CORP.,
FKA Merck & Co. Inc.; et al.,
Defendants-Appellees.
Appeal from the United States District Court
for the Southern District of California
Anthony J. Battaglia, District Judge, Presiding
Argued and Submitted October 3, 2017
Pasadena, California
Before: GRABER, MURGUIA, and CHRISTEN, Circuit Judges.
*
This disposition is not appropriate for publication and is not precedent
except as provided by Ninth Circuit Rule 36-3.
Plaintiff-Appellant Jean Adams and other plaintiffs in this consolidated,
multidistrict litigation appeal the district court’s determination that their state-law
claims were preempted under Wyeth v. Levine, 555 U.S. 555 (2009). We have
jurisdiction pursuant to 28 U.S.C. § 1291. We do not decide whether the
defendants met their burden under Levine’s “clear evidence” test because we hold
the district court misapplied Buckman Co. v. Plaintiffs’ Legal Committee, 531 U.S.
341 (2001), in two ways: first, the district court relied on Buckman to
impermissibly circumscribe discovery; and second, the district court relied on
Buckman to deem the plaintiffs’ newly discovered evidence “irrelevant” to the
court’s preemption analysis at the summary judgment stage. Either of these errors
would independently warrant reversal.
1. In Buckman, the plaintiffs were injured by the use of orthopedic bone
screws in their spines and claimed that the defendant, a consulting company that
assisted the screws’ manufacturer to secure regulatory approval, made fraudulent
representations to the Food and Drug Administration (FDA). Id. at 343. The
Supreme Court held the plaintiffs’ state-law claims were impliedly preempted by
2
federal law.1 Id. at 348. The Court ruled that the plaintiffs’ state-law “fraud-on-
the-FDA” claims “inevitably conflict[ed] with the FDA’s responsibility to police
fraud consistently with the Administration’s judgment and objectives,” id. at 350,
and reasoned, “were plaintiffs to maintain their fraud-on-the-agency claims . . . ,
they would not be relying on traditional state tort law which had predated the
federal enactments” in question, id. at 353; “[o]n the contrary, the existence of
these federal enactments is a critical element in their case.” Id.
Conversely, in Stengel v. Medtronic Inc., we held the plaintiffs’ state-law
failure-to-warn claim was not preempted. 704 F.3d 1224, 1233 (9th Cir. 2013) (en
banc), cert. denied, 134 S. Ct. 2839 (2014). We explained that because the
plaintiffs’ claim “rest[ed] on a state-law duty that parallel[ed] a federal-law duty”
and was “independent of the FDA’s pre-market approval process that was at issue
in Buckman,” Buckman did not control and the plaintiffs’ claims were not
1
The federal law at issue in Buckman, Stengel v. Medtronic Inc., 704 F.3d
1224 (9th Cir. 2013) (en banc), and McClellan v. I-Flow Corp., 776 F.3d 1035 (9th
Cir. 2015), was the Medical Device Amendments to the Food, Drug, and Cosmetic
Act (FDCA). We have previously suggested that the reasoning and policy of these
decisions also applies to drugs. See Gaeta v. Perrigo Pharm. Co., 630 F.3d 1225,
1234 (9th Cir. 2011) (distinguishing Buckman but giving no indication that
Buckman’s rule was inapplicable to preemption with respect to drug claims),
vacated on other grounds, 565 U.S. 973 (2011). We therefore assume without
deciding that Buckman would also preempt state-law fraud-on-the-FDA claims
concerning drugs.
3
preempted. Id.; see also McClellan v. I-Flow Corp., 776 F.3d 1035, 1040–41 (9th
Cir. 2015) (holding the plaintiff’s state-law failure-to-warn claims against the
manufacturer of an infusion pump were not fraud-on-the-FDA claims preempted
under Buckman because the claims “did not arise solely by virtue of” federal law,
and because there was “no suggestion that Congress intended to displace
traditional tort law by making all policing of medical labels and warnings the
exclusive province of the FDA”).
2. Here, the district court first relied on Buckman to circumscribe discovery.
Though it acknowledged that “this is not a case like Buckman that is predicated
upon a fraud-on-the-FDA basis,” the district court perceived that the plaintiffs
“invoke[d] allegations of misreporting and under-reporting as a justification for
additional discovery, and as pertinent to a preemption defense.” The court
concluded that the plaintiffs were asserting “fraud-on-the-FDA type allegations”
that were preempted by Buckman, and decided that the discovery they sought,
which included adverse event source documents and databases, was irrelevant to
whether federal law preempted the plaintiffs’ state-law failure-to-warn claims.
First, we disagree with the district court’s characterization of the plaintiffs’
state-law claims as “fraud-on-the FDA type allegations.” The plaintiffs asserted
common-law failure-to-warn claims arising from a state-law duty that paralleled an
4
FDCA-imposed duty, as was the case in Stengel and McClellan, where we found
the state-law claims not to be preempted. Stengel, 704 F.3d at 1233; see also
McClellan, 776 F.3d at 1040–41. The discovery the plaintiffs sought was directly
relevant to whether any causal connection existed between incretin use and
pancreatic cancer. The plaintiffs did argue that it would not be unduly burdensome
to produce the data they requested because the defendants were required to collect
and submit it to the FDA, but the duty the plaintiffs claim the defendants breached
was the parallel common law duty to warn, not a duty arising from the FDCA. As
pertinent to the defendants’ preemption affirmative defense, whether it would have
been possible for the defendants to comply with both their common law duty to
warn and the federally imposed reporting obligations is a separate issue that cannot
be resolved without knowing what information was available to the defendants.
See Levine, 555 U.S. at 573. Neither Buckman’s holding nor what the district court
termed the “policy underlying Buckman” can be read to preclude discovery of
evidence relevant to the plaintiffs’ state-law failure-to-warn claims. See Stengel,
704 F.3d at 1233.
The district court also ruled the request to compel production of the
defendants’ adverse event source documents and databases was unduly
burdensome. The plaintiffs sought the “source files” for each pancreatic cancer
5
event known to the defendants. Such files have been produced in pharmaceutical
litigation of this sort, it is undisputed that the defendants already maintained these
databases, and here, the volume of the requested data was limited. Defendants
established only that producing the files to the plaintiffs would require the
relatively modest task of redacting identifying information of patients and
reporters.2 The cost estimate provided to the district court included the cost of
producing the source files of both pancreatic cancer and pancreatitis. This figure
therefore overestimated the cost required to produce the pancreatic cancer files
alone. Thus, the defendants failed to show that complying with the plaintiffs’
discovery request would be unduly burdensome, and it was an abuse of discretion
to deny the plaintiffs’ motion to compel production for this reason. We reverse the
district court’s discovery orders that were premised on its misapplication of
Buckman, specifically: (1) the denial of the plaintiffs’ motion to compel production
of adverse event source documents and databases; and (2) the denial of the
plaintiffs’ motion to compel production of the defendants’ foreign regulatory files.
3. The district court also relied on Buckman to preclude its consideration of
“new safety information” the plaintiffs uncovered in the discovery they were
2
The number of pancreatic cancer adverse events (and the corresponding
number of source files the defendants would need to produce) appears to be
relatively low according to the FDA’s adverse event database.
6
allowed to conduct, including a signal assessment completed by Health Canada and
evidence from animal studies and clinical trials. The defendants argue the district
court considered and rejected the plaintiffs’ “new safety information” as non-
material. In its summary judgment order, the district court enumerated the “new
safety information” that the plaintiffs proffered, but also unambiguously stated that
it “maintain[ed] its position as set forth in previous orders regarding the relevance
of this data to the Court’s conflict preemption analysis,” reiterated its view that
“Buckman [was] implicated by Plaintiffs’ defense to the clear evidence standard,”
and concluded that the “new safety information” did not constitute “persuasive or
appropriate considerations in analyzing the clear evidence standard” (emphasis
added). These statements strongly suggest that the court deemed the new safety
information irrelevant at the summary judgment stage. Further, in its discussion of
the materiality of the “new safety information,” the district court stated, “it remains
unclear whether the FDA considered this information, and if it did not, whether
this data would have altered the FDA’s conclusion” (emphasis added).
Uncertainty about whether the FDA considered the “new safety information” and
whether it would have altered the FDA’s conclusion establishes that a disputed
issue of material fact should have prevented entry of summary judgment on the
defendants’ preemption claim. As the district court correctly noted, the parties’
7
experts disputed whether the “new safety information” would have been material
to the FDA’s analysis.
4. Finally, we hold the district court abused its discretion by partially
disqualifying the plaintiffs’ regulatory expert. The district court’s order correctly
observed that Dr. Fleming’s exposure to confidential information could not be
“entirely documented” by discoverable information such as meeting minutes, but
failed to account for the fact that this was the only evidence the defendants offered
to show they had disclosed to Dr. Fleming information relevant to the current
litigation. The defendants did not submit testimony or a declaration of anyone at
Novo Nordisk who could attest to whether or how any information provided to Dr.
Fleming was relevant to the current litigation, and Dr. Fleming averred that in
reaching his opinions and preparing his report for this litigation, he did not rely on
any information, confidential or otherwise, that he obtained from his consulting
relationship with Novo. Without more, the defendants did not meet their burden of
showing “specific and unambiguous” disclosures required to trigger
disqualification, Hewlett-Packard Co. v. EMC Corp., 330 F. Supp. 2d 1087, 1094
(N.D. Cal. 2004) (emphasis added). The district court abused its discretion by
partially disqualifying Dr. Fleming.
8
As an independent ground for disqualifying Dr. Fleming, the district court
concluded that he was a “competitor” under the court’s protective order. The
parties stipulated to the terms of the protective order, Dr. Fleming’s involvement
with Exsulin did not violate its express terms, and the defendants acknowledge that
the drug being developed by Dr. Fleming’s company was not a prescription
medication. The district court’s order concluded that “Plaintiffs’ argument
distinguishing between a manufacturer and a developer presumes a meaningful
difference in the context of the protective order,” but the record does not show that
the defendants were required to carry their burden of showing the distinction was
not meaningful. The district court abused its discretion by partially disqualifying
Dr. Fleming as a “competitor” under the stipulated terms of the protective order.
VACATED and REMANDED.
9