In Re: Testosterone Replacement Therapy Products Liability Litigation

Filing 1896

CASE MANAGEMENT ORDER NO. 47 (Ruling on AbbVie's motion for summaryjudgment on failure-to-warn claims - dkt. 1745), signed by the Honorable Matthew F. Kennelly on 5/8/2017. (mk)

IN THE UNITED STATES DISTRICT COURT
FOR THE NORTHERN DISTRICT OF ILLINOIS
EASTERN DIVISION

In re Testosterone Replacement Therapy
Products Liability Litigation Coordinated
Pretrial Proceedings
(This document applies to all cases)

)
)
)
)
)

Case No. 14 C 1748
MDL No. 2545

CASE MANAGEMENT ORDER NO. 47
(Ruling on AbbVie's motion for summary
judgment on failure-to-warn claims, dkt. 1745)
MATTHEW F. KENNELLY, District Judge:
Plaintiffs in this multidistrict litigation (MDL) proceeding allege that they have
suffered either arterial cardiovascular injuries or injuries related to blood clots in the
veins (venous thromboembolisms, or VTEs) as a result of taking prescription
testosterone replacement therapy (TRT) drugs. Defendants AbbVie Inc., Abbott
Laboratories, AbbVie Products LLC, and Unimed Pharmaceuticals, LLC (collectively,
AbbVie) manufacture AndroGel, one of the TRT drugs at issue in this litigation. Seven
plaintiffs have been selected to proceed with bellwether trials beginning June 2017.
Each bellwether plaintiff—Edward Cribbs, Cecile Frost, Jeffrey Konrad, Jesse
Mitchell, Arthur Myers, Robert Nolte, and Robert Rowley—alleges that AndroGel
caused them to suffer either a cardiovascular or VTE injury. AbbVie has filed three
motions for summary judgment, each one addressing different claims or issues. In this
opinion, the Court addresses AbbVie's motion for summary judgment on (1) claims
based on allegations that AbbVie failed to adequately warn of the dangers of AndroGel;
(2) unjust enrichment claims; and (3) punitive damages. For the reasons stated below,

the Court grants AbbVie's motion for summary judgment on plaintiff Rowley's claims for
strict liability based on design defect and for breach of implied warranty as well as on all
of the plaintiffs' negligent design defect claims but otherwise denies the motion for
summary judgment.
Background
The Court assumes familiarity with the background as set out in its separatelyissued opinion on AbbVie's motion for summary judgment on the issue of causation.
The Court therefore discusses only those details uniquely relevant to plaintiffs' failure to
warn claims.
A.

Initial approval of AndroGel
In April 1996, AbbVie submitted an investigational new drug application (IND) to

the FDA for AndroGel in accordance with the FDA's procedures for approving a new
prescription drug. The IND proposed clinical trials to demonstrate the safety and
efficacy of AndroGel, a testosterone replacement drug, in treating men with
hypogonadism. AbbVie then performed controlled studies of the effects of AndroGel
and submitted the data to the FDA. In April 1999, AbbVie submitted a new drug
application (NDA) for AndroGel, in which AbbVie was required to demonstrate the
drug's efficacy and proposed labeling. In response, the FDA requested that AbbVie
provide supplements containing raw data, medical data, and other clinical trial
information.
In February 2000, the FDA's medical officer recommended approval of AndroGel
1%. See Defs.' Statement of Undisputed Material Facts (SUDMF), Ex. 19 (Application
Review for AndroGel 1%). The medical officer concluded that AndroGel 1% was safe

2

for use and that the clinical trials produced no deaths and no serious adverse events
that were determined to be directly related to the use of AndroGel. Id. at 5. The
medical officer also noted that AndroGel was associated with polycythemia—an
elevated volume of red blood cells—in a few clinical trial participants. Id. The reviewer
indicated, however, that "[c]lear labeling instructions should serve to minimize these
risks by recommending active medical surveillance" such as the monitoring of "periodic
hematology and chemistry laboratories." Id. The medical officer also observed that the
mean concentration of estradiol—a metabolite of testosterone and a form of estrogen—
of the clinical trial participants remained within normal range on all observation days. Id.
at 17. The officer also noted that some patients had estradiol concentrations above the
upper limit of normal, with a greater number of these patients in the group receiving the
higher dose of AndroGel. Id.
Later that same month, the FDA approved AndroGel 1% as safe and effective.
The approved label indicated that "[h]emoglobin and hematocrit levels should be
checked periodically (to detect polycythemia) in patients on long-term androgen
therapy." Defs.' SUDMF, Ex. 85 (AndroGel Advertising / Promotional Materials) at E850008.
B.

Monitoring and developments related to cardiovascular risk
AbbVie says that both it and the FDA continued to monitor available safety

information and data for potential risks from AndroGel use, including cardiovascular risk.
Specifically, AbbVie monitored safety information to look for evidence of a "safety
signal," which the FDA defines as "a concern about an excess of adverse events
compared to what would be expected to be associated with a product's use." Defs.'

3

SUDMF, Ex. 1 (FDA Guidance on Pharmacovigilance Practices) at 4. AbbVie
submitted to the FDA annual periodic safety update reports (PSURs), which provided
information about adverse events from patient reports, ongoing clinical trials, and
medical literature. For the period of February 2000 through February 2005, AbbVie
reported thirty-five cardiac events out of 463,200 patient-years of exposure. Defs.'
SUDMF, Ex. 33 (2000-05 PSUR) at 2, 14, 16. AbbVie concluded that this incidence of
cardiac disorders did not exceed the general frequency in the overall population—and
therefore did not constitute a signal—but that it would continue to monitor cardiac risk.
Id. at 67.
In September 2005, AbbVie provided the FDA with an in-depth analysis of safety
information, which it refers to as the 2005 White Paper. See Defs.' SUDMF, Ex. 43
(2005 White Paper). AbbVie concluded that no studies reported a link between high
endogenous testosterone and increased risk of coronary artery disease. Id. at 61. The
company determined that any possible unfavorable effects of testosterone on the
cardiovascular system are less pronounced at the prescribed doses of AndroGel and
thus that the drug is unlikely to alter the overall risk of cardiovascular events. Id. at 78.
AbbVie concluded that no changes in AndroGel's prescribing information were required
"until the potential relationship between TRT and the cardiovascular system is better
defined." Id. at 104. AbbVie also noted that "[t]here is agreement in the literature that
additional research is necessary." Id. In both 2007 and 2010, AbbVie conducted
additional signal evaluation reports for AndroGel and concluded that there was no
evidence of a signal for cardiovascular events. Defs.' SUDMF, Ex 39 (2007 Signal
Evaluation) at 7; Defs.' SUDMF, Ex. 40 (2010 Cardiovascular Signal Evaluation) at 12.

4

In 2010, the FDA received notice that a study evaluating the cardiovascular
safety of TRT drugs had been discontinued. See Defs.' SUDMF, Ex. 61 (FDA Memo
May 2010) at 1. In this study, a group of authors evaluated the use of testosterone gel
by elderly men. During testing, the study's data safety and monitoring board
recommended that the researchers discontinue the study "because of an increase in
cardiovascular events observed among participants treated with testosterone compared
to placebo." Id. In June of that same year, the authors published the results of this trial
and concluded that daily application of a testosterone gel was associated with a greater
frequency of adverse events, including cardiovascular events. Defs.' SUDMF, Ex. 60
(Basaria Article) at 115. The authors also noted that due to limitations in the study, the
results may have been due to chance alone. Id. at 118.
In response to the discontinuation of this study, the FDA undertook a qualitative
review of TRT and the risk of cardiovascular disease in adult men. A reviewer for the
FDA's division of epidemiology evaluated two meta-analyses and one qualitative review
on the topic and determined that the studies highlight some trends "that could represent
signals of unknown effects of TRT" and that such trends "include a possible increased
risk of cardiovascular events . . . in men aged 65 years or older." FDA Memo May 2010
at 1. The FDA nevertheless concluded that the results of these studies "do not support
an association between TRT and increased risk of cardiovascular events in men." Id.
The FDA also noted that the analyses fell short of providing a definitive answer and that
larger safety trials designed to evaluate cardiovascular outcomes were needed to
provide a more conclusive answer. Id. The FDA also evaluated the Basaria article and
determined that it had several limitations "that precluded a definitive assessment of the

5

role of testosterone therapy in the cardiovascular events noted in the study." Defs.'
SUDMF, Ex. 4 (Resp. to Citizen Petition) at 6.
Later in 2010, the FDA conducted another analysis of cardiovascular risk. This
time, the FDA evaluated all of the studies discussed within the three articles that the
FDA reviewed in May. Defs.' SUDMF, Ex. 63 (FDA Memo Dec. 2010) at 1. The FDA
concluded that "one cannot make the conclusion based on these studies that
testosterone therapy increases the risk of cardiovascular disease." Id. The memo also
identified the need for long-term studies focused on evaluating cardiovascular risk. Id.
The FDA noted that the label then in use for testosterone products (including AndroGel)
included a warning on risk of edema in patients with preexisting cardiac, renal, or
hepatic disease. Id. at 2. It went on to state that "the risk of myocardial infarction and
other cardiovascular disease remains uncertain." Id.
Following these evaluations, the FDA finally ruled on a new drug application that
AbbVie had first submitted in February 2009 for a new version of AndroGel: AndroGel
1.62%. In reviewing the new drug application for AndroGel 1.62%, the FDA considered
adverse event reports for AndroGel 1% and specifically the possibility of an increased
risk of cardiovascular injury. Defs.' SUDMF, Ex. 68 (Application Review for AndroGel
1.62%) at 107. The FDA pointed to its own December 2010 determination that
available scientific studies "do not support an association between TRT and an
increased risk of cardiovascular events in men." Id. at 108. The FDA did not require
the label for AndroGel 1.62% to include a warning regarding an increased risk of
cardiovascular events.
Between April 2013 and February 2014, three additional articles were published

6

that are relevant to this litigation. The first was a meta-analysis known as the Xu article
which evaluated the existing literature on whether there is an association between use
of testosterone and cardiovascular risk. The other two publications—the Vigen article
and the Finkle article—were independent studies that evaluated the same association.
In February 2014, the group Public Citizen submitted a citizen petition to the FDA
asking it to issue a warning that TRT increases risk of cardiovascular injury. See
generally Defs.' SUDMF, Ex. 74 (Citizen Petition). In doing so, Public Citizen relied on
the articles by Basaria, Xu, Vigen, and Finkle to argue that a causal association existed
between TRT and cardiovascular injury. The citizen petition asked the FDA to (1) add a
"black box" warning to the labels for all drugs containing testosterone highlighting the
increased risks of cardiovascular dangers; (2) send "Dear Doctor" letters to warn
physicians of the possibility of cardiovascular effects; and (3) require the medication
guide for all testosterone products to be updated to reflect this risk. Id. at 1.
The FDA denied the citizen petition's requests in July 2014. It concluded that
there was "insufficient evidence of a causal link between testosterone therapy and
adverse cardiovascular outcomes to support the [requested] regulatory actions." Resp.
to Citizen Petition at 5. The FDA again addressed the Basaria study and stated that its
limitations prevented a conclusion that testosterone caused the cardiovascular effects.
Id. at 6. The FDA evaluated the Xu, Vigen, and Finkle articles and concluded that they
similarly did not support the conclusion that TRT drugs increase the risk of
cardiovascular events. Id. at 7–14. The agency denied the requested labeling changes
but indicated that the studies warranted "further exploration of a possible safety signal
regarding testosterone and cardiovascular risk." Id. at 16. The FDA stated that its

7

evaluation of the possible safety signal was ongoing. Specifically, it was waiting for the
results from a trial investigating the effects of testosterone in men over sixty-five, and it
planned to present the question of a potential association between testosterone and
cardiovascular risk to an advisory committee in the fall. Id.
In September 2014, the FDA convened an advisory committee to consider "the
potential for adverse cardiovascular outcomes" associated with use of TRT. Defs.'
SUDMF, Ex. 75 (Minutes Summary of September 2014 Advisory Committee Meeting) at
2. The advisory committee agreed that available studies about cardiovascular risk were
limited in scope and quality but also agreed that a weak signal of cardiovascular risk
had emerged from the results of recent large epidemiologic studies. Id. at 7. The
committee decided that the potential signal for cardiovascular risk should be added to
the labeling for TRT drugs. Id.
In May 2015, the FDA required AbbVie to add the following warning to the
AndroGel label:
5.5 Cardiovascular Risk
Long term clinical safety trials have not been conducted to assess the
cardiovascular outcomes of testosterone replacement therapy in men. To
date, epidemiologic studies and randomized controlled trials have been
inconclusive for determining the risk of major adverse cardiovascular
events (MACE), such as non-fatal myocardial infarction, non-fatal stroke,
and cardiovascular death, with the use of testosterone compared to nonuse. Some studies, but not all, have reported an increased risk of MACE
in association with the use of testosterone replacement therapy in men.
Patients should be informed of this possible risk when deciding whether to
use or to continue to use AndroGel 1.62%.
Defs.' SUDMF, Ex. 77 (May 2015 AndroGel 1.62% Label) at 5.5.
C.

Monitoring and developments related to VTE risk
AbbVie also monitored the relationship between TRT and VTE events and
8

provided the FDA with PSURs that included reports on adverse clotting events.
Between 2002 and 2004, the PSURs reported two serious cases of pulmonary
embolism and one major case of deep vein thrombosis (DVT). A PSUR that evaluated
safety data from 2000 through 2005 reported four serious thromboembolic events (both
DVT and pulmonary embolism) that were deemed to be possibly related to AndroGel.
In August 2005, AbbVie revised the "Adverse Reactions" section of the AndroGel label
to note that "[t]wo patients reported serious events considered possibly related to
treatment: deep vein thrombosis (DVT) and prostate disorder requiring a transurethral
resection of the prostate (TURP)." Defs.' SUDMF, Ex. 45 (Aug. 2005 AndroGel 1%
Label) at 11.
In AbbVie's 2005 White Paper, the company addressed AndroGel's hematologic
effects in addition to the cardiovascular risks. See 2005 White Paper 45–52. The
report acknowledged that testosterone can stimulate red blood cell production, which
can lead to polycythemia. Id. at 45. For this reason, AbbVie recommended that the
label continue to recommend monitoring of hemoglobin and hematocrit. Id. AbbVie
also noted that elevated hematocrit can lead to thromboembolic events. Id. at 47. The
report evaluated adverse event reports from the company's database from February
2000 through February 2005 and found 45 instances of polycythemia and increased
hematocrit out of 463,200 patient-years of observation. Id. at 50. AbbVie also found
four thromboembolic events—two cases of DVT and two cases of pulmonary
embolism—all of which were considered serious. Id. Elevated hematocrit was
observed in only one of these cases. Id. AbbVie concluded that "[t]he underlying
medical conditions of the four patients" were the most plausible explanations for the

9

VTE events. Id. The company stated that it is unlikely that the events were caused by
testosterone gel. Id.
The report also analyzed adverse event reports from the FDA database on
multiple testosterone drugs, including AndroGel. Id. at 50–51. AbbVie discovered
eighty-two cases of hematologic adverse events, including polycythemia, increased
hematocrit, and thrombosis-related events. Id. at 51. Twenty of these cases identified
a testosterone product as the primary or secondary suspect medication; eight
specifically identified AndroGel. Id. Of the eight associated with AndroGel, there were
two instances of DVT, one instance of increased hematocrit, and three instances of
pulmonary embolism. Id.
In March 2007, AbbVie again considered the possibility of a connection between
TRT and VTE events. See Defs.' SUDMF, Ex. 39 (2007 VTE Signal Evaluation).
AbbVie noted two reports of venous thrombus activity between February 2007 and
February 2008. Id. at 1–2. The company ultimately concluded that a review of scientific
literature did not support a signal of VTE risk associated with the use of TRT drugs. Id.
at 7.
In December 2007, the FDA approved a new label for AndroGel. The company
updated the label to indicate: "Increases in hematocrit, reflective of increases in blood
cell mass, may require lowering or discontinuation of testosterone. Increase in blood
cell mass may increase the risk for a thromboembolic event." Defs.' SUDMF, Ex. 83
(Dec. 2007 AndroGel 1% Label) at 5.9. In September 2009, the FDA again approved a
new label. This time, AbbVie added a statement to the medication guide for patients
which listed as a possible side effect "[b]lood clots in the legs." Defs.' SUDMF, Ex. 47

10

(Medication Guide) at ABBVIE-FST17796240.
In March 2010, AbbVie received an adverse event report of DVT with pulmonary
embolism and subsequently conducted a signal evaluation. See Defs.' SUDMF, Ex. 84
Part 3 (2010–11 PSUR) at E84-0866. AbbVie concluded "that the potential signal of
VTE was not supported by the data available to date." Id. at E84-0880. The evaluation
further stated that the then-current warning—which advised patients to monitor
hematocrit for polycythemia—adequately reflected available scientific knowledge. Id.
As previously discussed, in April 2011 the FDA approved AndroGel 1.62%. The
FDA's medical review of the product noted that in clinical studies recipients of AndroGel
1.62% did demonstrate an overall increase in mean hematocrit. Application Review for
AndroGel 1.62% at 157. There were no thromboembolic events reported in these
patients. Id. The FDA concluded that the language in the proposed label, which
mirrored that of AndroGel 1%, adequately addressed the possibility of increased
hematocrit. Id. But the FDA also moved the discussion of possible increased
hematocrit from the "Laboratory Testing" section under warnings into a separate section
under warnings titled "Polycythemia." Compare Dec. 2007 AndroGel Label at 5.9 and
Defs.' SUDMF, Ex. 86 (April 2011 AndroGel 1.62% Label) at 5.3. Section 5.3 of the
April 2011 label states:
5.3 Polycythemia
Increases in hematocrit, reflective of increases in red blood cell mass, may
require lowering or discontinuation of testosterone. Check hematocrit
prior to initiating treatment. It would also be appropriate to re-evaluate the
hematocrit 3 to 6 months after starting treatment, and then annually. If
hematocrit becomes elevated, stop therapy until hematocrit decreases to
an acceptable concentration. An increase in red blood cell mass may
increase the risk of thromboembolic events.
April 2011 AndroGel 1.62% Label at 5.3.
11

In March 2014, the FDA notified AbbVie that it had "become aware of new safety
information related to the serious risk of venous thromboembolic events associated with
testosterone use." Defs.' SUDMF, Ex. 87 (2014 Safety Labeling Change Notification) at
ABBVIE-FST06145227. The FDA requested that AbbVie add additional safety
information to the AndroGel labeling. Id. Specifically, the FDA asked AbbVie to add to
the warnings and precautions section under the highlights heading the following
statement: "Testosterone use may increase the risk of venous thromboembolism
(VTE), including deep vein thrombosis (DVT) and pulmonary embolism (PE). Consider
VTE in patients with signs or symptoms consistent with DVT or PE." Id. The FDA also
required AbbVie to add to the full prescribing information the following warning and
precaution:
5.4 Venous Thromboembolism
There have been postmarketing reports of venous thromboembolism
(VTE) in men taking testosterone, including AndroGel 1%. Testosterone
use may increase the risk of VTE, including deep vein thrombosis (DVT)
and acute pulmonary embolism (PE). Consider VTE in patients who
report symptoms of pain, swelling, warmth and redness in the leg (DVT) or
acute shortness of breath (PE). In such patients, discontinue treatment
with AndroGel 1% and evaluate patients for possible VTE [see
Postmarketing Experience (6.2)].
Id. at ABBVIE-FST06145228. Finally, the FDA required AbbVie to update the warning
in the medication guide regarding blood clots in the legs to read: "Blood clots in the
legs or lungs. Clots in your leg can include leg pain, swelling or redness. Clots in your
lungs can include difficulty breathing or chest pain." Id.
In response, AbbVie analyzed available VTE safety data from February 2000
through March 2014 and submitted to the FDA another white paper summarizing its
findings. See Defs.' SUDMF, Ex 89 (2014 White Paper). The report noted that
12

although polycythemia may lead to an increased incidence of vascular events,
"randomized testosterone trials have reported very low frequency of thromboembolic
events with testosterone therapy." Id. at 17. The report also addressed the possibility
that TRT leads to VTE events not through polycythemia but through an increase in
estradiol. See id. The report indicated that "the modest increases in estradiol that
accompanies [sic] most forms of TRT (e.g., transdermal) are unlikely to account for any
increased propensity for thrombotic events." Id.
The FDA evaluated this "rebuttal response" and modified its label change
requirements "to better reflect the level of uncertainty of drug causality." Defs.' SUDMF,
Ex. 90 (2010 E-mail) at 2. Specifically, the FDA told AbbVie it was not required to add
to the warnings and precautions section the statement indicating that testosterone use
may increase the risk of VTE and that patients with certain symptoms should be
evaluated for VTE. Id. at 3. The FDA also removed from its requirement for the full
prescribing information the statements indicating that testosterone use may increase the
risk of VTE, including DVT and pulmonary embolism. Id. AbbVie added the modified
warnings required by the FDA to the AndroGel label in June 2014.
D.

The present suits
The seven bellwether plaintiffs addressed in this motion have each filed suit

against AbbVie, alleging that they suffered either a cardiovascular or VTE injury after
taking AndroGel. This opinion addresses only the plaintiffs' claims based on AbbVie's
alleged failure to warn. All of the bellwether plaintiffs allege that AbbVie is liable for both
negligence and negligent misrepresentation based on its failure to warn consumers of
AndroGel about the increased risk of cardiovascular and VTE injury. Five of the

13

plaintiffs—Frost, Konrad, Mitchell, Nolte, and Rowley—also allege that AbbVie is strictly
liable for its failure to warn and for manufacturing AndroGel with a defective design. All
of the bellwether plaintiffs allege that AbbVie has breached an implied warranty by
failing to disclose defects in AndroGel. Six of the plaintiffs—Frost, Konrad, Mitchell,
Myers, Nolte, and Rowley—allege that AbbVie has also breached express warranties
and engaged in fraud. Myers and Rowley allege that AbbVie's failure to warn
constitutes a violation of state consumer protection laws. And Mitchell and Nolte allege
that AbbVie is per se negligent because it violated the Food, Drug, and Cosmetic Act by
failing to provide adequate warnings. In addition, all seven bellwether plaintiffs seek
punitive damages based on these failure to warn claims. Cribbs also brings a claim for
unjust enrichment and seeks to collect from AbbVie the money he paid for AndroGel.
Discussion
AbbVie has moved for summary judgment on the bellwether plaintiffs' claims
based on failure to warn, as well as their request for punitive damages and Cribbs's
unjust enrichment claim. AbbVie argues first that the failure to warn claims are
preempted by federal law. Alternatively, AbbVie contends that the plaintiffs' failure to
warn claims fail under state law because (1) there is no evidence that AbbVie knew or
should have known that AndroGel increased risk of cardiovascular or VTE injury; and
(2) the warnings that AbbVie provided were adequate to warn of the known or knowable
risks associated with AndroGel. AbbVie further contends that four of the plaintiffs—
Cribbs, Konrad, Myers, and Rowley—have not provided evidence that AbbVie's alleged
failure to provide adequate warnings was the proximate cause of their injuries. AbbVie
also argues that Myers's failure to warn claims are barred by the statute of limitations.

14

In addition, AbbVie argues that it is entitled to immunity from the plaintiffs' strict liability
claims based on design defect and their claims based on breach of an implied warranty.
AbbVie also argues that the plaintiffs' claims based on negligent design defect are
preempted by federal law. Finally, AbbVie contends that the plaintiffs are not entitled to
punitive damages and that Cribbs's claim for unjust enrichment fails because he did not
directly bestow a benefit on AbbVie.
When considering a motion for summary judgment, a court construes the facts in
the light most favorable to the non-moving party. Suarez v. W.M. Barr & Co., Inc., 842
F.3d 513, 517 (7th Cir. 2016). Summary judgment is appropriate where there is no
genuine dispute regarding any material fact and the moving party is entitled to judgment
as a matter of law. Schaefer v. Universal Scaffolding & Equip., LLC, 839 F.3d 559, 604
(7th Cir. 2016). A dispute is genuine if the evidence is such that a reasonable jury could
return a verdict for the non-moving party. Whiting v. Wexford Health Sources, Inc., 839
F.3d 658, 661 (7th Cir. 2016).
A.

Preemption
AbbVie first argues that the plaintiffs' claims based on the company's alleged

failure to warn are preempted by federal law. Federal preemption occurs when a state
law is invalidated because it conflicts with federal law. Mason v. SmithKline Beecham
Corp., 596 F.3d 387, 390 (7th Cir. 2010). Courts have identified three forms of
preemption: (1) express preemption, where Congress declares its intent to preempt
state law; (2) implied preemption, where the structure and purpose of the federal law
evinces an intent by Congress to preempt state law; and (3) conflict preemption, which
occurs when it is impossible to comply with both federal and state law. Id. AbbVie

15

argues that the failure to warn claims fall into the third category, because it is impossible
to comply with both federal labeling requirements from the FDA and state-law duties
that underlie failure to warn claims. See Defs.' Mot. for Summ. J. on Pls.' Failure-ToWarn Claims and Mem. of Law in Support (Defs.' Mot. for Summ. J.) at 55.
The Supreme Court has held that conflict preemption exists for state-law claims
based on a manufacturer's alleged failure to warn consumers only when there is clear
evidence that the FDA would have rejected the proposed change to the drug's label.
Wyeth v. Levine, 555 U.S. 555, 571–72 (2009); Mason, 596 F.3d at 391. In doing so,
the Court relied on the FDA's procedures regarding labeling changes. The Court noted
that typically a manufacturer may only change a drug label after receiving approval from
the FDA. Wyeth, 555 U.S. at 568. FDA regulations qualify this restriction, however,
and indicate that a manufacturer may "add or strengthen a contraindication, warning,
precaution, or adverse reaction" without waiting for FDA approval. Id. (citing 21 CFR §§
314.70(c)(6)(iii)(A), (C)). Thus manufacturers are not limited to the warnings approved
by the FDA but may unilaterally add additional warnings—and thereby comply with
state-law duties under failure to warn—without violating federal requirements. Wyeth,
555 U.S. at 569–71. Conflict preemption therefore exists in this context only where the
manufacturer can provide "clear evidence" that the FDA would have rejected attempts
by the manufacturer to make the change to the label that plaintiffs contend was omitted.
Id. at 571–72; Mason, 596 F.3d at 391. The Court concludes that AbbVie has failed to
present clear evidence that the FDA would have rejected efforts to add warnings about
either cardiovascular or VTE risk to the AndroGel warning labels.
AbbVie first argues that the FDA considered any potential risks when it initially

16

approved AndroGel in 2000, because it concluded that the drug was safe and did not
require a warning about either cardiovascular or VTE risk. Defs.' Mot. for Summ. J. at
59. But the Seventh Circuit stated in Mason that the FDA's initial approval of a drug
"does not provide much, if any, evidence that the FDA would have rejected the warning
the plaintiffs say should have been in place." Mason, 596 F.3d at 391. The court
emphasized the fact that the manufacturer, and not the FDA, retains primary
responsibility for a drug's label. Id. In short, the fact that the FDA accepted AbbVie's
assessment of the risks of AndroGel and approved the label that AbbVie proposed
sheds little light on whether the FDA would have rejected any attempts by AbbVie to
provide additional warnings.
AbbVie next argues that the FDA had been evaluating and approving TRT drugs
since the early 1970s and never required the warnings that plaintiffs now allege were
missing. Defs.' Mot. for Summ. J. at 60. AbbVie contends that the FDA repeatedly
determined that TRT products did not present the types of risk that plaintiffs now claim.
But again, the Seventh Circuit in Mason determined that the administrative history of
"different drugs made by different manufacturers" is not useful in determining whether
the clear evidence standard has been met. Mason, 596 F.3d at 395. To put it another
way, the FDA's failure to require warnings on the labels of different drugs does not
answer whether they would have rejected an attempt by AbbVie to add the warning to
its own labels.
AbbVie also points to the fact that over the years it provided the FDA with
multiple analyses of potential safety signals and adverse event reports, including annual
PSURs and comprehensive white papers. Defs.' Mot. for Summ. J. at 61. AbbVie

17

contends that in each of these instances, it analyzed the possibility of cardiovascular or
VTE risk, and yet the FDA never required additional warnings in response to these
analyses. Further, AbbVie notes, the FDA approved numerous AndroGel labels
between 2000 and 2009 without requiring the labeling changes that plaintiffs now
endorse. This, AbbVie argues, constitutes clear evidence that the FDA would have
rejected any attempt by AbbVie to add such warnings themselves. Id. But in each of
the reports in question, AbbVie concluded that there was no evidence of an association
between AndroGel and either cardiovascular or VTE risk, and it consistently
recommended that no change be made to AndroGel's labeling. The fact that the FDA
accepted AbbVie's recommendation to maintain the current warnings does not
constitute clear evidence that it would have rejected an additional warning. Nor does
the fact that the FDA repeatedly approved labels that did not contain the warning. See
Mason, 596 F.3d at 395 (finding that "the FDA's inaction, as in its failure to mandate a
warning about the risk," does not support granting summary judgment on preemption).
The Court concludes that the plaintiffs' failure to warn claims are not preempted.
AbbVie next points to the fact that the FDA evaluated the possibility of an
association between TRT drugs and either cardiovascular or VTE risk and consistently
concluded that there was insufficient evidence to draw a causal connection. AbbVie
points to the following determinations by the FDA:
•

the FDA's conclusion in 2010 that the risk of cardiovascular disease
"remains uncertain" following the publication of the Basaria article, FDA
Memo Dec. 2010 at 2;

•

the FDA's conclusion in 2013 based on a safety review of AndroGel 1.62%
that "known safety issues associated with [AndroGel] are adequately
addressed in the current labeling," Defs.' SUDMF, Ex. 70 (2013 AndroGel
1.62% Postmarketing Safety Summary) at 4.
18

•

the FDA's conclusion in 2014 following further study that there was
insufficient evidence of a causal link between testosterone therapy and
cardiovascular injuries, Resp. to Citizen Petition at 5.

But the fact that the FDA was not "affirmatively convinced of a causal link between the
drug and the adverse event" would not necessarily preclude AbbVie from adding the
warning on its own. See In re Fosamax (Alendronate Sodium) Prods. Liab. Litig., 852
F.3d 268, 298 (3d Cir. 2017). As in Fosamax, the FDA recognized the possibility of a
link with both cardiovascular and VTE risk, even though it ultimately concluded that the
available literature did not support an association. See FDA Memo May 2010 at 1;
Minutes Summary of September 2014 Advisory Committee Meeting at 7. And as
AbbVie points out, the FDA approved three labeling changes between August 2005 and
September 2009 that added additional information relating to the possibility of VTE
injury, though not going so far as to state a causal relationship. See Defs.' Mot. for
Summ. J. at 62–63. This evidence indicates a reasonable possibility "that the FDA
would still have determined that 'reasonable evidence' of a link existed—or more
precisely, that the possibility of rejection was less than highly probable." Fosamax, 852
F.3d at 298. AbbVie has therefore failed to meet the "clear evidence" standard.
Finally, AbbVie points to two instances in which, it argues, the FDA rejected the
specific warnings that plaintiffs now argue were missing. In March 2014, the FDA asked
AbbVie to add the following statement to AndroGel's full prescribing information:
"Testosterone use may increase the risk of VTE, including deep vein thrombosis (DVT)
and acute pulmonary embolism (PE)." 2014 Safety Labeling Change Notification at 1–
2. AbbVie then conducted its own review of VTE safety information and concluded that
it was unlikely that TRT drugs led to increased risk of VTE, through either polycythemia
or other mechanisms. See 2014 White Paper. The FDA responded by deleting its
19

earlier statement from its warning requirement. 2010 Email at 2–3. AbbVie argues that
this amounts to a clear rejection by the FDA of the warning language that plaintiffs rely
on for their claims. But as AbbVie points out, this was a rejection of the FDA's own
proposal, and not a rejection of an attempt by AbbVie to add the language. This
evidence does not demonstrate that, had AbbVie amended the label to include the
language in the FDA's initial proposal instead of contesting the language, the FDA
would have rejected this modification.
The other instance that AbbVie points to is the FDA's rejection of the 2014 citizen
petition. The citizen petition asked the FDA to require a black box warning, Dear Doctor
letters, and an update to the medication guide to warn consumers of cardiovascular risk
associated with testosterone products. Citizen Petition at 1–2. The FDA ultimately
denied all three requests and indicated that the current evidence presented at most a
"possible safety signal." Resp. to Citizen Petition at 16. AbbVie relies on two circuit
court cases in arguing that this meets the clear evidence standard. In the first, the
Seventh Circuit indicated that the FDA's refusal to require a reference to a particular
injury on the label of an over-the-counter drug "when it had been asked to do so in the
submission to which the agency was responding" was clear evidence that the FDA
would not approve a similar warning from the manufacturer. Robinson v. McNeil
Consumer Healthcare, 615 F.3d 861, 873 (7th Cir. 2010). The Tenth Circuit reached a
similar conclusion last week in Cerveny v. Aventis, Inc., No. 16-4050, 2017 WL 1573309
(10th Cir. May 2, 2017), finding the FDA's rejection of "a citizen petition containing
arguments virtually identical" to plaintiffs' to be clear evidence supporting preemption.
Id. at *9.

20

First, the Court notes that the Seventh Circuit in Robinson was not addressing an
argument regarding preemption. Rather, it was evaluating the plaintiff's claim that the
district court erred in refusing to permit her to amend her complaint to add a claim
based on breach of implied warranty. Because the issue of preemption was not raised
by either party, the court's statement regarding what constitutes clear evidence is dicta.
More importantly, the facts of this case distinguish it from both Robinson and
Cerveny and therefore support the conclusion that AbbVie has failed to meet the
standard for preemption. In Robinson, the FDA declined to include the proposed
warning after it determined that "the addition would confuse rather than enlighten." Id.
at 870. The FDA specifically stated that "the overall benefit versus risk profile for
ibuprofen products remains very favorable when they are used according to the labeling
instructions." Id. But in AbbVie's case, the FDA did not indicate that the warning
requested in the citizen petition would confuse or otherwise discourage consumers for
whom the drug was beneficial from taking TRT drugs. In fact, the FDA agreed that the
evidence might demonstrate a "possible safety signal." Resp. to Citizen Petition at 16.
The FDA's rejection of a citizen petition in Cerveny is somewhat closer to the
rejection upon which AbbVie relies here. In considering the citizen petition at issue in
that case, the FDA stated that "the scientific literature did not justify ordering changes to
the labeling that warn of such risks" and "there was insufficient evidence to support [the
other] requests." Cerveny, 2017 WL 1573309, at *7. The Tenth Circuit found that the
FDA, in its rejection of the petition, "concluded that the warnings were unjustified for
risks in taking [the drug] prior to pregnancy" and therefore that the claims were
preempted. Id. at *10. In the present case, the FDA also concluded that there was

21

insufficient evidence to support the warnings requested by the citizen petition. But, the
FDA also indicated uncertainty, and a less-than-definitive determination, by stating that
it "believes that the publication of these studies warrants further exploration of a
possible safety signal regarding testosterone and cardiovascular risk" and that the
FDA's "evaluation remains ongoing." Resp. to Citizen Petition at 16. The FDA noted
that it was continuing to assess "this potential safety signal" and was waiting for the
results of a trial evaluating the effects of testosterone treatment in elderly men. Id.
Finally, the FDA stated that it intended to present the question of a potential association
between testosterone and cardiovascular events to an advisory committee. Id. This
evidence indicates that there may have been other reasons behind the FDA's denial of
the citizen petition—namely, that it wanted to complete its own evaluation—that caused
it to reject the citizen petition and that it would not have rejected a similar proposal from
the manufacturer itself. See Mason, 596 F.3d 387, 394–95 (finding that rejection of a
citizen petition did not constitute clear evidence where the FDA indicated that more
research was necessary). Although the court in Cerveny rejected plaintiffs' argument
that the FDA would have treated a request by the manufacturer differently than one in a
citizen petition, Cerveny, 2017 WL 1573309 at *8–9, the FDA's statements in this case
indicate that, had the manufacturer come forward with a request to make further
warnings, the FDA may have permitted the changes. And irrespective of the rationale
for the FDA's determination, its rejection of the warnings is not "clear-cut," Mason, 596
F.3d at 394, because the agency itself indicated the possibility of future action. 1

1

The Court also notes that in May 2015, the FDA required AbbVie to add additional
warnings about cardiovascular risk to the AndroGel label, based on part on the findings
of the advisory committee referenced in the agency's response to the citizen petition.
22

Further, the citizen petition in this case requested that the labeling changes apply
to "all testosterone-containing drugs presently on the market," not only to AndroGel.
Citizen Petition at 1. This change would therefore have swept much more broadly than
an attempt by AbbVie to change only the AndroGel label. This is significantly different
from Cerveny, where, as far as the Court can tell, the citizen petition requested changes
specifically to the label of the drug taken by one of the plaintiffs. See Cerveny, 2017 WL
1573309 at *7. Cerveny itself provides the importance of this distinction: "the FDA
views overwarnings as problematic because they can render the warnings useless and
discourage use of beneficial medications." Id. at *8. This rationale may explain not only
why the FDA may be reluctant to add too many warnings to the label of a particular
product, but also why the agency may be reluctant to add a particular warning to the
label of a number of products that may be varied in design and use. Thus the FDA's
rejection in Cerveny of the citizen petition's proposed warning for the precise drug at
issue—i.e., a warning that was narrow and focused—constituted clear evidence that it
would have rejected a similar attempt by the manufacturer. But the FDA's rejection in
this case of the citizen petition's effort to modify the labels of "all testosterone-containing
drugs" does not constitute clear evidence that the agency would have rejected an
attempt by AbbVie to modify the warning label of only its own drug.
In sum, because there are other explanations for the FDA's rejection of the
citizen petition, this rejection does not constitute clear evidence that the FDA would
have rejected an attempt by AbbVie to add the warnings that plaintiffs contend were
Though the significance of this point may be subject to question, see Mason, 596 F.3d
at 395, "[t]o the extent these subsequent events have any sway . . . they clearly cut
towards making it less likely that the FDA would have rejected plaintiffs' proposed
warning." Id. at 395.
23

wrongfully omitted.
Finally, AbbVie argues that plaintiffs cannot defeat preemption by arguing that
the company committed fraud on the FDA. Defs.' Mot. for Summ. J. at 80. Plaintiffs
concede that their failure to warn claims do not rely on allegations that AbbVie
committed fraud on the FDA or withheld information from the agency. Pls.' Resp. in
Opp'n to Mot. of AbbVie Defs.' for Summ. J. on Pls.' Failure To Warn Claims (Pls.'
Resp.) at 23. Therefore this argument is moot.
For all of these reasons, the Court finds that AbbVie has failed to present clear
evidence that the FDA would have rejected an effort by AbbVie to add the requested
labeling changes to the AndroGel label. The Court therefore concludes that plaintiffs'
failure to warn claims are not preempted.
B.

Failure to warn
Failure to warn claims are a product of state law, and the seven bellwether

plaintiffs reside in six different states. The law underlying failure to warn claims,
however, is largely similar across these states. The Court will make note of any
significant differences that impact its analysis.
Under a failure to warn claim, a product may be deemed defective if it is
unreasonably dangerous to place the product in the hands of a user without a suitable
warning and the product is supplied and no warning is given. See, e.g., Saller v. Crown
Cork & Seal Co., Inc., 115 Cal. Rptr. 3d 151, 166, 187 Cal. App. 4th 1220, 1238 (2010).
Plaintiffs must show that the defendant "did not adequately warn of a particular risk that
was known or knowable in light of the generally recognized and prevailing best scientific
and medical knowledge," which is typically measured by the information available at the

24

time of the manufacture and distribution. Id., 187 Cal. App. 4th at 1238; see also
McKenzie v. A.W. Chesterson Co., 277 Or. App. 728, 742, 373 P.3d 150, 158 (2016);
Flax v. DaimlerChrysler Corp., 272 S.W.3d 521, 541 (Tenn. 2008); House v. Armour of
Am., Inc., 929 P.2d 340, 343 (Utah 1996). In the prescription drug context,
manufacturers have a continuing duty to warn about the risks associated with their
products, even those discovered past the point of initial sale. See Payne v. Novartis
Pharm. Corp., 767 F.3d 526, 530 (6th Cir. 2014); Singleton v. Eli Lilly, Co., No. 1:10-CV02019-AWI-SKO, 2011 WL 2621067, at *3 (E.D. Cal. June 29, 2011). This duty to warn
also requires the manufacturer to "exercise ordinary and reasonable care in testing a
product for potential danger." Rodriguez v. Stryker Corp., 680 F.3d 568, 574 (6th Cir.
2012). Thus a manufacturer's failure to investigate risks of potential harm can create
liability under failure to warn, but not for "hidden risks that neither it nor the medical
community had a reasonable basis to suspect." Id.
Further, all of the states whose law is at issue in the AbbVie bellwether trial
cases apply the learned intermediary doctrine, under which the manufacturer's duty to
warn extends only to prescribing physicians and not to consumers of the drug itself.
See Walton v. Bayer Corp., 643 F.3d 944, 999–1000 (7th Cir. 2011) (applying Illinois
law). The physician is a "learned intermediary" "who, equipped with the knowledge
imparted to him by the drug's manufacturer, determines, weighing benefit against risk,
the drug's suitability for a particular patient." Id. at 1000. Because the physician
determines whether to prescribe the drug, manufacturers are required to provide only
warnings sufficient to make the physicians aware of the attendant risks of a particular
medication.

25

Finally, plaintiffs bringing failure to warn claims must also demonstrate that the
manufacturer's failure to provide adequate warnings was the proximate cause of their
injuries.
In moving for summary judgment, AbbVie argues that (1) it did not know, nor
should it reasonably have known, that AndroGel caused increased risk of
cardiovascular or VTE injury prior to the plaintiffs' injuries; (2) the warnings on the
AndroGel label were adequate to warn physicians of the known risks; and (3) four of the
plaintiffs have not provided evidence to support proximate causation.
1.

Knew or should have known

A drug manufacturer is deemed to be an expert in its field and is under a
continuing duty to be aware of scientific developments relating to its product.
Grundberg v. Upjohn Co., 813 P.2d 89, 98 (Utah 1991). Manufacturers are therefore
responsible not only for "actual knowledge gained from research and adverse reaction
reports, but also for constructive knowledge as measured by scientific literature and
other available means of communication." Id. Manufacturers are required to notify the
medical profession of any additional side effects discovered from the use of their
products. Id.
Because a prescription drug manufacturer has an ongoing duty to inform the
medical profession about the risks associated with its product even after the time of
distribution, the Court must consider all information available to AbbVie at any time prior
to the plaintiffs' injuries, the idea being that had AbbVie issued a warning even after the
plaintiffs had initially been prescribed AndroGel, they might have ceased taking the drug
and thereby avoided their injuries. At times, the relevant scientific literature differs for

26

cardiovascular and VTE risks, and therefore the Court considers them separately.
a.

Information on cardiovascular risk

Four bellwether plaintiffs allegedly suffered a cardiovascular injury after taking
AndroGel:
•

Cribbs (myocardial infarction on May 25, 2012)

•

Frost (myocardial infarction on February 21, 2013)

•

Konrad (myocardial infarction on July 9, 2010)

•

Mitchell (myocardial infarction on November 18, 2012)

The Court considers the scientific literature and other medical information available to
AbbVie prior to these dates. Neither party argues that there was any scientific
information made available between the first cardiovascular injury (July 2010) and the
last (February 2013) that might differentiate the Court's determination as between these
four plaintiffs. See Defs.' Mot. for Summ J. at 88–90.
The Court concludes that a reasonable jury could find that AbbVie knew or
reasonably should have known, before the injuries suffered by these four plaintiffs, that
TRT drugs increase the risk of cardiovascular injury. In support of their argument,
plaintiffs first offer two reports by experts who describe the mechanisms by which
testosterone can lead to cardiovascular injuries. In the first report, Dr. Perry Halushka
discusses a particular mechanism involving thromboxane A2 receptors. He explains
how thromboxane A2 plays a central role in platelet aggregation, and he opines that a
"robust body of medical and scientific literature" demonstrates that platelet factors play
a critical role in acute thrombotic events. Pls.' Resp. to Defs.' Statement of Material
Facts (SOMF), Ex. 2 (Halushka Report) at 8. In the second report, Dr. Burt Gerstman
discusses the thromboxane A2 mechanism, as well as other possible biological
27

mechanisms demonstrating a relationship between testosterone and cardiovascular
events. Pls.' Resp. to Defs.' SOMF, Ex. 3 (Gerstman Report) at 91–92. In discussing
these mechanisms, Dr. Gerstman relies primarily on studies that existed prior to 2010,
when the first bellwether plaintiff suffered a cardiovascular injury. He concludes that
"several lines of biological evidence support the epidemiologic finding of an increased
risk of heart attack and stroke with testosterone supplementation." Id. at 92.
Plaintiffs also provide the expert report of Dr. Hossein Ardehali. Dr. Ardehali
opines that "[e]xogenously administered testosterone therapy in men increases the risk
of a major cardiovascular event." Pls.' Resp. to Defs.' SOMF, Ex. 9 (Ardehali Report) at
7. Dr. Ardehali details the various mechanisms by which use of testosterone can
increase the risk of cardiovascular injury, including through an effect on (1)
thromboxane A2; (2) estradiol levels; (c) erythropoiesis; and (d) nitric oxide. Id. at 67.
He also indicates that the scientific community began looking into the effect that
testosterone has on cardiovascular risk as early as 1990. Id. at 66. Dr. Ardehali cites
numerous studies and reports as evidence of the existence of a causal relationship,
many of which were published prior to plaintiffs' injuries. See id. at 103–14.
Dr. Ardehali also reviewed numerous documents specific to AndroGel and
determined that by 2007 there was reasonable evidence of a causal association
between AndroGel therapy and cardiovascular injury. See id. at Appendix C. In
drawing this conclusion, he relied primarily on AbbVie's signal evaluation reports, the
PSURs AbbVie submitted to the FDA, and the FDA's own database on adverse event
reports. See id., Appendix C at 1. Dr. Ardehali concluded that the adverse event
reports from individuals who had used AndroGel and experienced a cardiovascular

28

injury supported the existence of a causal association. A reasonable jury could infer
from this evidence that AbbVie reasonably should have known that AndroGel increased
cardiovascular risk.
AbbVie argues that Dr. Ardehali's "long-after-the-fact litigation work" does not
give rise to a genuine dispute of fact, because he admitted that he does not have
training in pharmacovigilance and did not use FDA guidelines when analyzing the
adverse event reports. Defs.' Mot. for Summ. J. at 19–20. AbbVie points to no case
law, however, indicating that whether a manufacturer reasonably should have known of
a particular risk hinges on whether the FDA's guidelines have been met. The fact that
the FDA considered the same evidence and concluded that no safety signal existed
may support AbbVie's argument, but it does not prevent Dr. Ardehali's opinion from
giving rise to a genuine dispute on this particular issue. Dr. Ardehali reviews both the
scientific literature and the adverse event reports that would have been available to
AbbVie prior to plaintiffs' injuries to conclude that an expert in this field—as AbbVie is
deemed to be—reasonably should have known that use of TRT drugs is associated with
an increased risk of cardiovascular injury.
Plaintiffs also offer the Basaria article, which was published before any of the four
plaintiffs at issue suffered their cardiovascular injury, and concluded that use of
testosterone gel is associated with an increased risk of adverse cardiovascular events.
Basaria Article at 115. AbbVie argues that this study should be discounted due to the
fact that the authors acknowledged that the results might be due to chance alone.
Defs.' Mot. for Summ. J. at 88. AbbVie argues that plaintiffs cannot as a matter of law
rely on an article "where the authors of a study themselves are unwilling to suggest a

29

causal link." Id. But although the authors did concede that the results may have been
due to chance, they also stated in the conclusion of their study that "[i]n this population
of older men with limitations in mobility and a high prevalence of chronic disease, the
application of a testosterone gel was associated with an increased risk of cardiovascular
adverse events." Basaria Article at 109. Even if this conclusion cannot be extrapolated
to other populations, when considered in conjunction with the other evidence offered by
plaintiffs it is sufficient to permit a reasonable jury to conclude that AbbVie reasonably
should have known of a causal association.
Finally, the Court concludes that there is also ample evidence from which a
reasonable jury could conclude that AbbVie had a duty to perform additional tests in
order to better evaluate the effect of AndroGel on cardiovascular risk. One of the
plaintiffs' experts, Dr. Peggy Pence, opines that the lack of safety data regarding the
use of testosterone in elderly men was well-known as early as 2002. Pls.' Resp. to
Defs.' SOMF, Ex. 4 (Pence Report) at 71. Dr. Pence discusses a number of articles
published prior to 2010 that specifically address the need for further study to determine
the effects of TRT drugs on cardiovascular risk. AbbVie itself recognized in 2005 that
"[t]here is agreement in the literature that additional research is necessary." 2005 White
Paper at 104. Further, in 2006 Canada's regulatory body, Health Canada, provided
AbbVie with an assessment of regulatory issues surrounding testosterone therapy. See
Pls.' Resp. to Defs.' SOMF, Ex. 11 (Health Canada Analysis). The analysis indicated
that testosterone is known to increase hematocrit and that an above-normal hematocrit
level "is associated with increased risk for vascular disease," including myocardial
infarction, stroke, and thrombosis. Id. at 12. The analysis also stated that there is no

30

long-term clinical data available to determine the risk / benefit ratio of TRT in elderly
men and that one is needed particularly to determine the incidence of "coronary
events." Id. at 14. AbbVie asks the Court to discount this analysis because Health
Canada was addressing "a different label, under a different regulatory regime, in a
different country." Defs.' Reply at 8–9. But the analysis is not being used to dispute the
adequacy of AbbVie's FDA-approved labels. Plaintiffs offer the analysis to show only
that the scientific community agreed that more testing was required to determine
whether drugs such as AndroGel increased the risk of cardiovascular injury.
In sum, a reasonable jury could infer that AbbVie reasonably should have known
that there was a possible association between TRT drugs and increased cardiovascular
risk and that it needed to perform additional safety testing. The Court therefore denies
AbbVie's motion for summary judgment on claims brought by plaintiffs who experienced
cardiovascular injuries to the extent they are based on whether AbbVie knew or should
have known of an increased risk.
b.

Information on VTE risk

Three bellwether plaintiffs allegedly suffered a VTE injury after taking AndroGel:
•

Myers (pulmonary embolism on February 6, 2008)

•

Nolte (pulmonary embolism on November 1, 2012)

•

Rowley (DVT on April 27, 2013)

The Court considers the scientific literature and other medical information available to
AbbVie prior to these dates. AbbVie divides its argument between the evidence
available prior to 2009 (which encompasses Myers), the evidence available prior to
2014 (which encompasses Nolte and Rowley), and the evidence available after 2014
(which does not encompass any of the bellwether plaintiffs). Defs.' Mot. for Summ J. at
31

91–95. Because the Court concludes that a genuine dispute exists regarding whether
AbbVie knew or reasonably should have known that a risk of VTE injury existed prior to
Myers's injury in 2008, it is unnecessary to consider the remaining time periods.
Plaintiffs have provided evidence from which a reasonable jury could conclude
that AbbVie should have known by February 2008 that TRT drugs could lead to an
increased risk of a VTE injury. Plaintiffs offer the expert report of Dr. Henry Rinder, who
opines that exogenous testosterone increases the risk of VTE injuries. Pls.' Resp. to
Defs.' SOMF, Ex. 8 (Rinder Report) at 1. Dr. Rinder describes a number of biologically
plausible mechanisms by which testosterone can increase the risk of VTE injury, one of
which is an increase in hematocrit. Id. at 1–2, 10–13. AbbVie acknowledges that as
early as February 2000, the medical literature supported a need to monitor hematocrit in
patients taking TRT. Defs.' Mot. for Summ. J. at 91. Further, when discussing other
mechanisms, Dr. Rinder cites literature that was available prior to 2008. Rinder Report
at 11–13. Dr. Rinder's opinion is supported by that of Dr. Gerstman, who similarly
discussed mechanisms that can lead to thromboembolic events, including through an
increase in estradiol. In doing so, Dr. Gerstman relied on literature that was available
prior to 2008.
Dr. Rinder also reviewed adverse event reports for AndroGel and opines in his
report that by 2004, "there was evidence of heightened risk of VTE in patients exposed
to AndroGel." Id. at 18, Appendix A. AbbVie appears to criticize Dr. Rinder for the
same reasons as it challenges Dr. Ardehali—namely, that Dr. Rinder is not an expert in
pharmacovigilance and did not apply the FDA's signal detection guidance in his
analysis. Defs.' Mot. for Summ. J. at 42–43. But as discussed earlier, the fact that Dr.

32

Rinder did not apply the FDA's standards for signal detection does not mean that his
opinion is irrelevant in determining whether AbbVie reasonably should have known of
an association between TRT drugs and increased risk of VTE injury. Further, Dr. Pence
incorporated Dr. Rinder's analysis in her own report in concluding that by 2004
reasonable evidence of a causal association existed. Pence Report at 81. Dr. Pence
agreed that the adverse event reports between 1992 and 2004 constitute "particularly
compelling evidence" that an association existed. Id. at 81–82.
In sum, a reasonable jury could conclude that AbbVie reasonably should have
known from the scientific literature and AndroGel's adverse event reports about an
association between TRT drugs and increased risk of VTE injury. The Court therefore
denies AbbVie's motion for summary judgment on claims brought by plaintiffs who
experienced VTE injuries to the extent they are based on whether AbbVie knew or
should have known of an increased risk.
2.

Adequacy of warnings

AbbVie next argues that the warnings on the AndroGel label that existed prior to
each of the plaintiffs' injuries were adequate as a matter of law to warn of the risks of
cardiovascular and VTE injury. "Although the adequacy of warnings concerning drugs
is generally a question of fact, it can become a question of law where the warning is
accurate, clear and unambiguous." Thom v. Bristol-Myers Squibb Co., 353 F.3d 848,
853 (10th Cir. 2003) (internal quotation marks omitted). Warnings are generally
adequate when they contain "a full and complete disclosure of the potential adverse
reactions to the drug." Pittman v. Upjohn Co., 890 S.W.2d 425, 429 (Tenn. 1994).

33

a.

Cardiovascular risk

The Court finds that a genuine dispute exists regarding whether AbbVie
adequately warned physicians of all known or reasonably knowable cardiovascular risks
associated with AndroGel. In arguing that its labels were adequate, AbbVie points first
to three statements added to the label between 2000 and 2007. The first indicates that
"[e]dema with or without congestive heart failure may be a serious complication in
patients with preexisting cardiac, renal or hepatic disease." Defs.' SUDMF, Ex. 27
(2000 AndroGel Label) at ABBVIE-FST00355128. A reasonable jury could conclude,
however, that this warning is insufficient to inform consumers that regardless of
preexisting conditions AndroGel may increase the risk of cardiovascular injury. Further,
a genuine dispute exists regarding whether this statement adequately apprises
consumers of all possible types of cardiovascular injuries.
The second statement indicates that "[g]eriatric patients treated with androgens
may be at an increased risk for the development of prostatic hyperplasia and prostatic
carcinoma." Id. AbbVie argues that this statement is sufficient to inform physicians that
men over the age of 65 are at greater risk of serious side effects. But this warning does
not specifically identify either cardiovascular or VTE injury—it refers only to effects on
users' prostates. The third statement, added in December 2007, indicates that "there is
insufficient long-term safety data in geriatric patients to assess the potential risks of
cardiovascular disease." Dec. 2007 AndroGel Label at ABBVIE-FST14649560. But a
reasonable jury could conclude that this does not constitute a "full and complete
disclosure" of the risks reasonably knowable to AbbVie at this time, given that it does
not identify a relationship between AndroGel and cardiovascular injury.

34

AbbVie also points to three sets of clinical guidelines published between 2002
and 2010, each of which stated in some form that whether TRT is associated with
increased cardiovascular risk is uncertain. See Defs.' Mot. for Summ. J. at 99–100.
AbbVie fails to explain, however, how these guidelines are relevant in determining
whether the warnings that AbbVie provided on the AndroGel label were adequate.
AbbVie has not cited to any case law that requires the Court to look to warnings other
than those given by AbbVie in determining whether AbbVie has upheld its duty to warn
consumers of the knowable risks of its product. Further, these guidelines suffer from
the same potential deficiencies as those identified with the AndroGel label itself, given
that they indicate only that TRT's association with cardiovascular risk remains uncertain.
Finally, AbbVie argues that communications from the FDA in 2014 further
informed doctors about a possible increase in cardiovascular risk. But each of the
bellwether plaintiffs bringing claims based on cardiovascular risk suffered an injury prior
to 2014. Therefore any warnings AbbVie provided after their injuries are irrelevant in
determining whether these four plaintiffs received adequate warnings.
In light of the Court's conclusion that a reasonable jury could find that AbbVie had
reason to know that AndroGel increased the risk of cardiovascular injury, a reasonable
jury could also conclude that the warnings AbbVie gave indicating that the state of
research was uncertain failed to notify consumers of an association between the two.
AbbVie is therefore not entitled to summary judgment based on the adequacy of its
warnings with regard to the plaintiffs who suffered cardiovascular injuries.
b.

VTE risk

AbbVie next argues that it provided adequate warnings regarding the risk of VTE

35

injury associated with AndroGel. AbbVie points to the fact that from the initial approval
of AndroGel, the product's label instructed doctors to monitor hematocrit in order to
detect polycythemia. Further, in 2007 AbbVie added a statement that specifically
indicated that an increase in hematocrit could increase the risk of a thromboembolic
event. But plaintiffs argue—and a reasonable jury could conclude—that these warnings
did not adequately address the risk of VTE caused by a different mechanism and
otherwise independent of an increase in hematocrit. This is relevant because it appears
that both Myers and Rowley had normal levels of hematocrit at the time of their injuries.
Therefore a genuine dispute exists regarding whether these statements provided
physicians with adequate warning that AndroGel could lead to thrombotic events that
would not be indicated by an increase in hematocrit.
c.

Rebuttable presumption and overpromotion

AbbVie also argues that two of the states at issue here—Tennessee and Utah—
employ a rebuttable presumption that FDA-approved warnings are adequate and
therefore that the discussion above is irrelevant for plaintiffs Konrad and Rowley. Both
Utah and Tennessee appear to apply a presumption of non-defectiveness for products
manufactured in conformity with government standards. See Utah Code Ann. § 78B-6703 (West); Tenn. Code Ann. § 29-28-104 (West). But plaintiffs can rebut this
presumption by demonstrating that the product at issue was in fact unreasonably
dangerous. See Flax v. DaimlerChrysler Corp., 272 S.W.3d 521, 536 (Tenn. 2008);
Niemala v. Imperial Mfg., Inc., 2011 UT App. 333, ¶ 10, 263 P.3d 1191, 1196 (2011).
And plaintiffs have presented evidence that AndroGel did not have any benefits for
individuals who had "Low T" as opposed to classical hypogonadism. A reasonable jury

36

could conclude that AndroGel's risks outweighed its benefits and therefore that it was
unreasonably dangerous.
And even if the warnings the Court has discussed adequately addressed the risk
of cardiovascular and VTE injury, there is still reason to deny AbbVie's motion for
summary judgment. Many states apply the overpromotion theory, under which a
manufacturer can be liable for failure to warn even when the warnings were adequate if
it engaged in an advertising campaign that in effect negated the warnings. See Okuda
v. Wyeth, No. 1:04-CV-80 DN, 2012 WL 12337860, at *1 (D. Utah July 24, 2012)
(considering motions in limine) (indicating that the overpromotion theory comes into play
"if the warnings given were adequate and the prescribing physicians disregarded the
warnings in reliance on the promotional materials"); Salmon v. Parke, Davis & Co., 520
F.2d 1359, 1363 (4th Cir. 1975) (applying North Carolina law) (indicating that a jury
could infer that certain promotional materials provided to the physician were "a form of
overpromotion which nullified the effect of even a valid warning on the package");
Stevens v. Parke, Davis & Co., 9 Cal. 3d 51, 65, 507 P.2d 653, 661 (1975) ("[A]n
adequate warning to the profession may be . . . nullified by overpromotion of the
drug . . . which may have the effect of persuading the prescribing doctor to disregard the
warnings given.").
Plaintiffs contend that AbbVie engaged in an overpromotion campaign when it
exaggerated the benefits of AndroGel and promoted it "as a cure-all for numerous
afflictions for which it was never tested or approved." Pls.' Resp. at 44. They argue
essentially that this overpromotion skewed the prescribing physicians' risk / benefit
assessments and caused them to disregard the AndroGel warnings. Id. at 46–47. And

37

the Court has elsewhere determined, in its ruling on another of the three motions for
summary judgment brought by AbbVie, that a genuine dispute exists regarding whether
AbbVie engaged in such a campaign and whether the physicians for each of these
plaintiffs was impacted by it. Thus even for those plaintiffs for whom the warning
otherwise might have been adequate—such as plaintiffs from states with a rebuttable
presumption of adequacy or Nolte, who suffered a VTE injury with an accompanying
increase in hematocrit—a genuine dispute exists regarding whether AbbVie's conduct
amounted to overpromotion and caused the prescribing physicians to disregard the
otherwise adequate warnings. The Court therefore denies AbbVie's motion for
summary judgment on plaintiffs' failure to warn claims on the basis that the warnings
given were adequate.
3.

Proximate cause

The standard for proximate cause on a failure to warn claim varies somewhat
among different states. Generally speaking, a plaintiff must show that had the
manufacturer provided the warning at issue, the plaintiff's injury would have been
avoided. Some jurisdictions apply the heeding presumption, which indicates "that
where a warning is inadequate, the plaintiff is entitled to a rebuttable presumption that
an adequate warning would have been heeded if one had been given." Thom, 353 F.3d
at 855. AbbVie argues that four plaintiffs—Cribbs, Konrad, Myers, and Rowley—have
not provided evidence giving rise to a genuine dispute on this point because each of
their physicians testified that they still would have prescribed AndroGel even if the
desired warnings had been included. Defs.' Mot. for Summ. J. at 106.

38

a.

Cribbs

Under North Carolina law, a plaintiff in a failure to warn case must demonstrate
that the desired warning would have altered the physician's care and avoided the
plaintiff's injury. See Holley v. Burroughs Wellcome Co., 318 N.C. 352, 356, 348 S.E.2d
772, 774–75 (1986). AbbVie argues that Cribbs cannot meet this element because his
prescribing physician, Dr. Mark Ottelin, testified that he continued to prescribe AndroGel
for Cribbs even after receiving an alert from the FDA that it was investigating the risk of
stroke and heart attack in men taking testosterone products. Pls.' Resp. to Defs.'
SOMF, Ex. 16 (Ottelin Dep.) at 130:18–132:2. But Dr. Ottelin also testified that his
process in prescribing AndroGel to Cribbs would have been to discuss the drug and all
its potential complications and then permit Cribbs to choose whether to proceed. Id. at
87:8–16. Dr. Ottelin testified that he would not have discussed cardiovascular
complications with Cribbs when he initially prescribed AndroGel because that
information did not exist at that time. Id. at 87:24–88:2. And Cribbs testified that he
decided to stop taking AndroGel after he saw a commercial indicating that taking
AndroGel could lead to heart attacks. Pls.' Resp. to Defs.' SOMF, Ex. 17 (Cribbs Dep.)
at 156:2–8. A reasonable jury could infer from this evidence that, had AbbVie provided
Dr. Ottelin with the warning regarding cardiovascular risk, he would have informed
Cribbs of this risk, and Cribbs would have decided not to take AndroGel. AbbVie is not
entitled to summary judgment on Cribbs's failure to warn claims on the basis of lack of
proximate cause.
b.

Konrad

To show proximate cause under Tennessee law, the plaintiff must show "that a

39

warning would have altered the doctor's actions and that the change in the doctor's
actions would have averted the patient's injury." Payne, 767 F.3d at 531. Because
Tennessee law has not adopted the heeding presumption, id. at 532–33, Konrad must
present evidence from which a reasonable jury could conclude that his injury would
have been avoided if his doctor had received a different warning.
AbbVie argues that Konrad cannot do so because his doctor, Steven Overby,
testified that the primary effect of the new warnings was to reduce the number of
patients that requested testosterone replacement medication. Defs.' SUDMF, Ex. 111
(Overby Dep.) at 130:23–131:17. Dr. Overby further stated that "it is not necessarily
that [he is] fearful of using the prescription" and that Konrad had a clear screening
workup before he prescribed AndroGel. Id. at 131:18–24. Therefore, AbbVie argues,
Dr. Overby has not been discouraged from prescribing AndroGel and still would have
done so for Konrad, regardless of additional warnings.
But Dr. Overby also testified that his "prescription writing habits are basically run
by the patient request or the patient volume that demands a certain type of
prescription." Id. at 132:1–4. Further, he testified that, had the information in the 2015
label regarding the risk of heart attacks and stroke been available in 2010, he would
have discussed these risks with Konrad before writing a prescription. Id. at 133:14–22.
Dr. Overby stated he would also have included the later-added information regarding
DVT into his risk-benefit analysis when determining whether to prescribe AndroGel. Id.
at 133:23–134:4. And Konrad testified that it would have been important to him "to
know of any potential increased risk of heart attacks or strokes associated with
AndroGel use prior to making [his] decision to take the drug." Pls.' Resp. to Defs.'

40

SOMF, Ex. 19 (Konrad Dep.) at 226:6–9. The court in Payne indicated that causation
"ultimately rests with the patient's decision to take or reject the medication." Payne, 767
F.3d at 532. A reasonable jury could infer from the evidence cited by plaintiffs that Dr.
Overby would have informed Konrad about the associated cardiovascular risks and that
Konrad would have chosen not to take AndroGel. See Smith v. Pfizer Inc., 688 F.
Supp. 2d 735, 746 (M.D. Tenn. 2010) (denying summary judgment on the issue of
causation where plaintiff's physician would have warned him about the side effects at
issue and evidence showed that plaintiff was concerned with these particular side
effects). AbbVie is not entitled to summary judgment on Konrad's failure to warn claims
on the basis of lack of proximate cause.
c.

Myers

Under Arizona law, Myers must show that if AbbVie had issued a proper warning,
he would not have taken AndroGel. See D'Agnese v. Novartis Pharm. Corp., 952 F.
Supp. 2d 880, 890 (D. Ariz. 2013). Arizona courts apply the heeding presumption to
their determination of proximate cause. Id. AbbVie argues that there is no genuine
dispute regarding proximate cause because Dr. Michaela Tong, Myers's prescribing
physician, testified that she warned him of the risk of thromboembolic events and that
additional VTE warnings would not have affected her decision whether to prescribe
AndroGel. Defs.' SUDMF, Ex. 109 (Tong Dep.) at 109:2–16, 114:16–21. But Myers
testified during his deposition that he does not recall Dr. Tong providing any warning
about risks associated with AndroGel and that he would remember if this occurred. Pls.'
Resp. to Defs.' SOMF, Ex. 19 (Myers's Dep.) at 153:21–154:9. Myers also testified that
he would have trusted Dr. Tong to warn him of any adverse effects associated with

41

AndroGel. Id. at 156:3–10. Further, although Dr. Tong testified that she provided
Myers with the warning on possible VTE events that existed in the label in 2007, this
label only addressed VTE events associated with an increase in hematocrit. Dr. Tong
did not warn Myers of the risk of VTE injuries that could occur independently of an
increase in hematocrit. See Tong Dep. at 158:15–162:20. A reasonable jury could infer
from this evidence that Myers was not fully apprised of the risks associated with
AndroGel and that he would not have taken the drug if he had been fully informed.
Because AbbVie does not offer any evidence that Myers would still have taken the drug
even with these warnings, it is not entitled to summary judgment on Myers's failure to
warn claims on the basis of lack of proximate cause.
d.

Rowley

In a failure to warn claim brought under the law of Utah, a plaintiff "must show . . .
that had warnings been provided, the injured party would have altered his use of the
product or taken added precautions to avoid the injury." House, 929 P.2d at 346.
Courts in Utah apply the heeding presumption. See id. Proximate causation is
generally a question for the jury—"[o]nly in rare cases may a trial judge rule as a matter
of law on the issue of proximate cause." Steffensen v. Smith's Mgmt. Corp., 820 P.2d
482, 486 (Utah Ct. App. 1991).
AbbVie argues that Rowley has failed to present evidence to support proximate
cause because his prescribing medical professional, physician's assistant Teryl
Hunsaker, testified during his deposition that he was aware of the risk of blood clots and
that he took this risk into account when making his initial prescribing decision. Pls.'
Resp. to Defs.' SOMF, Ex. 23 (Hunsaker Dep.) at 64:12–23, 115:1–10. But Hunsaker

42

also testified that he was unaware of the risk of DVT at the time that he first prescribed
AndroGel to Rowley. Id. at 114:17–21. More importantly, when asked whether he
would have prescribed AndroGel to Rowley had he been advised that AndroGel led to
an increased risk of VTE independent of polycythemia, Hunsaker testified that he did
not know and that he would have discussed this risk with Rowley before prescribing the
medication. Id. at 112:15–113:3. This testimony is sufficient to give rise to a genuine
dispute regarding whether AbbVie's failure to provide this warning was the proximate
cause of Rowley's injury. AbbVie is not entitled to summary judgment on Rowley's
failure to warn claims on the basis of lack of proximate cause.
4.

Statute of limitations

AbbVie also contends that Myers's failure to warn claims are barred by the
statute of limitations. It argues that Myers's cause of action accrued in February 2008
when he suffered a pulmonary embolism. Because more than two years have elapsed
since this event, AbbVie contends, Myers is barred from bringing his claim.
The Arizona Supreme Court has indicated that the statute of limitations is not
automatically triggered each time a professional's services may have brought about an
adverse result. Walk v. Ring, 202 Ariz. 310, 314–15, 44 P.3d 990, 994–95 (2002) (en
banc) (considering a claim for medical malpractice). Instead, the statute begins to run
only when the plaintiff knows, or reasonably should know, of the wrong committed by
the defendant. Id. at 315, 44 P.3d at 995. The court stated that "it is not enough that a
plaintiff comprehends a 'what'; there must also be reason to connect the 'what' to a
particular 'who' in such a way that a reasonable person would be on notice to
investigate whether the injury might result from fault." Id. at 316, 44 P.3d at 996. Here,

43

Myers's "what" certainly occurred in February 2008 when he suffered a pulmonary
embolism. And AbbVie argues that Myers reasonably could have discovered the
existence of his claims against AbbVie in 2008 because the AndroGel label that existed
at the time included the statement that an increase in red blood cell mass may increase
the risk for a thromboembolic event. Defs.' Mot. for Summ. J. at 91–93. But the
statement on the label highlighted the risk of thromboembolic event only as related to an
increase in red blood cells. And Myers had a normal hematocrit (the ratio of red blood
cells to total volume of blood) at the time of his pulmonary embolism. The Court finds
that a reasonable jury could conclude that Myer did not have a reason to connect this
injury with AndroGel until much later—specifically, the point in time when he had reason
to believe that AndroGel could lead to VTE injury independent of an increase in
hematocrit. Because there is a genuine dispute regarding when Myers reasonably
should have discovered his claims against AbbVie, AbbVie is not entitled to summary
judgment on Myers's claims based on the statute of limitations.
D.

Other state law claims
1.

Strict liability—design defect

Plaintiffs Frost, Konrad, Mitchell, Nolte, and Rowley allege that AbbVie is strictly
liable for manufacturing AndroGel with a defective design. Plaintiffs concede that Utah
law precludes Rowley from bringing this claim. Pls.' Resp. at 64 n.28. The Court
therefore grants summary judgment in favor of AbbVie on Rowley's strict liability design
defect claim. AbbVie argues that three of the remaining plaintiffs live in states which
provide manufacturers with immunity from strict liability design defect claims when the
drug at issue has been approved by the FDA. Defs.' Mot. for Summ. J. at 108–109.

44

With regard to Nolte, AbbVie argues that he cannot meet the elements of a strict liability
design defect claim under Arizona law. Id. at 109–10.
AbbVie is correct that three of the states at issue here—California (Frost),
Tennessee (Konrad), and Oregon (Mitchell)—appear to have adopted comment (k) to
section 402A of the Restatement (Second) of Torts, which states that manufacturers of
products that are "unavoidably unsafe," such as prescription medications, are "not to be
held to strict liability for unfortunate consequences attending their use." Allen v. G.D.
Searle & Co., 708 F. Supp. 1142, 1149 (D. Or. 1989); see also Brown v. Superior Court,
44 Cal. 3d 1049, 1066–67, 751 P.2d 470, 481 (1988); Or. Rev. Stat. Ann. § 30.920(3);
Rodriguez v. Stryker Corp., No. 2:08–0124, 2011 WL 31462, at *6 (M.D. Tenn. Jan. 5,
2011). But comment (k) confers this immunity only "with the qualification that [the
drugs] are properly prepared and marketed, and proper warning is given." Allen, 708 F.
Supp. at 1149 (emphasis added) (citing Restatement (2d) of Torts § 402A, cmt. k
(1965)). Therefore numerous courts have held that a drug manufacturer receives
immunity from strict liability design defect claims only when there is no dispute
regarding whether the manufacturer provided adequate warnings. See Allen, 708 F.
Supp. at 1149; J.F. ex rel. Moore v. McKesson Corp., No. 1:13-CV-01699-LJO-JLT,
2014 WL 202737, at *7-8 (E.D. Cal. Jan. 17, 2014); Harwell v. Am. Med. Sys., Inc., 803
F. Supp. 1287, 1300 (M.D. Tenn. 1992). AbbVie is therefore not entitled to summary
judgment on the basis of comment (k), because a genuine dispute exists regarding
whether it provided adequate warnings.
AbbVie also argues that Nolte has failed to provide evidence to support the
elements of a strict liability design defect claim under Arizona law. AbbVie contends

45

that Arizona courts apply the standard from the Restatement (Third) of Torts, under
which a prescription drug is not considered to be defectively designed unless "a
reasonable healthcare provider would not prescribe the drug for any class of patients."
Defs.' Mot. for Summ. J. at 109 (internal quotation marks omitted). In support, AbbVie
points to two federal district court cases from Arizona that cite this standard when
considering design defect claims brought under Arizona law. See Harrison v.
Howmedica Osteonics Corp., No. CIV 06-0745 PHX RCB, 2008 WL 906585, at *21 (D.
Ariz. Mar. 31, 2008); Gebhardt v. Mentor Corp., 191 F.R.D. 180, 185 (D. Ariz. 1999).
But the court in Gebhardt itself acknowledged that "no Arizona case has formally
adopted the Restatement (Third) of Torts." Gebhardt, 191 F.R.D. at 185. And Arizona
courts considering strict liability design defect claims use the "risk / benefit analysis" to
determine whether it was reasonable for the manufacturer to put his product on the
market. Golonka v. Gen. Motors Corp., 204 Ariz. 575, 581–82, 65 P.3d 956, 962–63
(Ariz. App. Ct. 2003). This test considers whether the manufacturer would continue to
market the product in light of the product's potentially dangerous consequences.
Gomulka v. Yavapai Mach. and Auto Parts, Inc., 155 Ariz. 239, 243, 745 P.2d 986, 990
(Ariz. App. Ct. 1987). Thus Nolte's claim hinges on whether he can demonstrate that a
reasonable manufacturer in AbbVie's position would have continued to market AndroGel
in light of the attendant cardiovascular and VTE risks. Nolte argues that a reasonable
manufacturer would not have done so because the benefits that AbbVie claimed for
treating men with "Low T" did not actually exist and therefore could not have
outweighed the risks. The Court finds that this is sufficient to give rise to a genuine
dispute regarding whether Nolte can establish the elements of strict liability for a design

46

defect under Arizona law. Without a clear statement from the Arizona courts on
whether the state has adopted this "any class of patients" standard, the Court declines
to modify the risk / benefit analysis that Arizona courts have applied for years.
The Court therefore grants AbbVie's motion for summary judgment on the strict
liability design defect claim brought by Rowley but otherwise denies the motion for
summary judgment on these claims.
2.

Negligence—design defect

AbbVie also argues that the Court should grant it summary judgment on plaintiffs'
negligence claims to the extent that they are based on defective design. Defs.' Mot. for
Summ. J. at 110. AbbVie contends that any such claims are preempted by federal law,
relying primarily on Yates v. Ortho-McNeil-Janssen Pharmaceuticals, Inc., 808 F.3d 281
(6th Cir. 2015). Design defect claims typically require the plaintiff to demonstrate that
that there was a reasonably feasible alternative design. See, e.g., DeWitt v. Eveready
Battery Co., Inc., 144 N.C. App. 143, 154–55, 550 S.E.2d 511, 518–19 (2001). In
Yates, the court concluded that any negligence claim based on the argument that a
manufacturer should have used employed an alternative design for its drug after it was
approved by the FDA was preempted by federal law because FDA regulations prohibit a
manufacturer from making major changes to the formulation of the drug provided in the
approved application. Yates, 808 F.3d at 298. The court then indicated that any
argument based on an alternative pre-approval design was too attenuated because it
would require the court to speculate on whether the FDA would have approved the
alternative design and whether the plaintiffs would still have selected the drug in its
alternate form. Id. at 299.

47

Plaintiffs argue that none of this analysis applies to their claims for negligent
design defect because they "do not allege a safer alternative formulation or brand of
testosterone would have avoided their injuries." Pls.' Resp. at 65. Instead, plaintiffs
argue, they are asserting that there is no formulation of testosterone that would be
effective to treat individuals with age-related hypogonadism, which they refer to as an
"illusory" condition that does not require TRT. Id. But plaintiffs cite no case law which
supports their contention that they can bring a negligent design defect claim without
demonstrating the existence of a feasible alternative design. Plaintiffs' characterization
of these claims demonstrates that they are actually just a repackaging of their off-label
marketing claims—that AbbVie never should have promoted AndroGel for use by
individuals with age-related hypogonadism or "Low T." The Court grants summary
judgment in favor of AbbVie on these claims.
3.

Implied warranty

AbbVie argues in a footnote that manufacturers of prescription drugs receive the
same comment (k) immunity from liability on implied warranty claims as they receive on
strict liability design defect claims. Defs.' Mot. for Summ. J. at 109 n.57. This is true
under Utah law, which means that the Court must also grant summary judgment to
AbbVie on Rowley's claim for breach of implied warranty. See Straub v. Fisher &
Paykel Health Care, 990 P.2d 384, 389 n.1 (Utah 1999). As for the remaining plaintiffs,
even if AbbVie is correct, the Court determined that a genuine dispute exists regarding
whether AbbVie is entitled to comment (k) immunity due to its alleged failure to provide
adequate warnings. The Court therefore denies AbbVie's motion for summary judgment
on the remaining plaintiffs' implied warranty claims for the same reason.

48

4.

Unjust enrichment

AbbVie also argues that it is entitled to summary judgment on Cribbs's claim of
unjust enrichment. AbbVie first argues that Cribbs cannot show that he conferred a
benefit directly on AbbVie as required under North Carolina law. Defs.' Mot. for Summ.
J. at 112. But both appellate courts in North Carolina and federal courts applying North
Carolina law have held that a plaintiff can bring a claim for unjust enrichment even if he
has not directly conferred a benefit on the defendant. See New Prime, Inc. v. Harris
Transp. Co., No. COA12-271, 2012 WL 3192718, at *2–3 (N.C. App. Ct. Aug. 7, 2012);
Metric Constructors, Inc. v. Bank of Tokyo-Mitsubishi, Ltd., 72 F. App'x 916, 920–21 (4th
Cir. 2003); In re Processed Egg Products Antitrust Litig., 851 F. Supp. 2d 867, 932 (E.D.
Penn. 2012). AbbVie next argues that Cribbs cannot show that he did not receive the
benefit that he paid for, because he has admitted that he saw an overall improvement in
his well-being after taking AndroGel. Defs.' Mot. for Summ. J. at 112. But Cribbs also
alleges that he suffered a myocardial infarction due to taking AndroGel and thus that he
did not receive the safe treatment for which he paid. This is sufficient to give rise to a
genuine dispute regarding whether AbbVie was unjustly enriched. The Court therefore
denies AbbVie's motion for summary judgment on Cribbs's claim of unjust enrichment.
D.

Punitive damages
Finally, AbbVie argues that plaintiffs' requests for punitive damages fail as a

matter of law, either because the states in which the plaintiffs reside bar punitive
damages in these types of cases or because the plaintiffs have failed to provide clear
and convincing evidence that AbbVie acted fraudulently or maliciously. Defs.' Mot for
Summ. J. at 111–12. Plaintiffs argue that the law of Illinois—AbbVie's home state—

49

governs the question of punitive damages and that a genuine dispute exists regarding
whether plaintiffs can meet the standard under Illinois law.
1.

Choice of law

Plaintiffs contend that this Court should apply Illinois choice of law rules when
determining which state's law governs the availability and award of punitive damages.
Pls.' Resp. at 66–69. AbbVie does not argue otherwise, though it contends that under
Illinois choice of law rules, the punitive damages determination is governed by the law
of plaintiffs' home states. See Reply Mem. in Further Supp. of Defs.' Mot. for Summ. J.
on Pls.' Failure-To-Warn Claims (Defs.' Reply) at 14 & n.16.
The Seventh Circuit has indicated that in "foreign cases filed directly in a district
court as a part of ongoing multidistrict litigation," courts generally should treat the cases,
for purposes of applicable choice of law rules, as originating outside the district and thus
should apply the choice of law rules of the originating states. Dobbs v. DePuy
Orthopedics, Inc., 842 F.3d 1045, 1049 (7th Cir. 2016). But this rule applies only to
foreign cases that were directly filed in the district overseeing the MDL. As established
by the Court in Case Management Order No. 12, any actions filed in this MDL
proceeding after October 24, 2014 solely against AbbVie are treated as originally filed in
this district, if the plaintiff pleads in his complaint that venue is proper in this district.
See dkt. no. 440 (Case Mgmt. Order No. 12 Regarding the Filing of Actions in the
Northern District of Ill. or Directly in the MDL Proceedings) at II.B.ii. The bellwether
plaintiffs have brought their claims solely against AbbVie and have each alleged that
venue is proper in the Northern District of Illinois. Thus these cases in fact originate in
this district.

50

Further, the Seventh Circuit has also indicated that "[w]here the parties agree on
the law that governs a dispute, and there is at least a reasonable relation between the
dispute and the forum whose law has been selected by the parties, [courts] forego an
independent analysis of the choice-of-law issue and apply the parties' choice." Harter v.
Iowa Grain Co., 220 F.3d 544, 559 n.13 (7th Cir. 2000) (internal quotation marks
omitted). Neither party argues that the choice of law rules of a state other than Illinois
should apply to the selection of the law on punitive damages. And there is a reasonable
relation between the dispute and Illinois, because this Court sits in Illinois and AbbVie is
located in Illinois. The Court therefore applies Illinois choice of law rules when
determining what law governs plaintiffs' requests for punitive damages.
Illinois courts employ the "most significant contacts" test to resolve conflicts of
law. Miller v. Long-Airdox Co., 914 F.2d 976, 978 (7th Cir. 1990). Illinois law applies a
strong presumption that the law of the state where the injury occurred governs in a
personal injury case, which can be overcome only "by showing a more or greater
significant relationship to another state." Townsend v. Sears, Roebuck & Co., 227 Ill. 2d
147, 163–64, 879 N.E.2d 893, 903 (2007). A court considers where the injury occurred,
where the conduct that caused the injury occurred, the domiciles of the parties, and
where the parties' relationship is centered. Id. at 160, 879 N.E.2d at 901. The Court
also takes into consideration the relevant policies of the forum, the relevant policies of
the interested states, and the relative interests of those states in the determination of
the particular issue. Id. at 170, 879 N.E.2d at 906–07.
Here, the most significant contacts occurred both in plaintiffs' home states (where
the injuries occurred) and in AbbVie's home state (where the conduct that allegedly

51

caused the injury took place) and therefore do not favor one party over the other. In
considering the purpose of punitive damages, however, the Court concludes that Illinois
has the greatest interest in governing the determination of punitive damages. Punitive
damages serve a public goal of punishing the defendant for its wrongdoing and
protecting the public from future misconduct, either by the defendant or by others.
Ziarko v. Soo Line R. Co., 161 Ill. 2d 267, 276, 641 N.E.2d 402, 407 (1994). In the
Court's view, the state in which a defendant is domiciled therefore tends to have a
stronger policy interest in whether punitive damages are available than the state in
which the plaintiff's injury occurred. Thus Illinois' interest in regulating AbbVie's conduct
outweighs whatever interest plaintiffs' home states have in protecting non-resident
businesses against excessive liability.
AbbVie cites to Hammond v. System Transport, Inc., No. 11 CV 1295, 2012 WL
3234865 (C.D. Ill. Aug. 6, 2012), in which the court analyzed the same choice-of-law
authority—and this Court's decision in a prior case—and decided to apply the law of the
plaintiff's home state regarding the imposition of punitive damages. The Court
respectfully disagrees with the analysis in that case and found no Illinois case law that
suggests a different conclusion. The Court therefore applies Illinois law to the plaintiffs'
requests for punitive damages.
2.

Punitive damages standard

Under Illinois law, punitive damages are awarded "only where the defendant's
conduct is willful or outrageous due to evil motive or a reckless indifference to the rights
of others." Franz v. Calaco Dev. Corp., 352 Ill. App. 3d 1129, 1137, 818 N.E.2d 357,
366 (2004). The Court finds that a genuine dispute exists regarding whether AbbVie's

52

conduct was sufficiently willful or outrageous. Plaintiffs have provided evidence from
which a reasonable jury could conclude that AbbVie knew that AndroGel increased the
risk of cardiovascular and VTE injury and yet failed to provide adequate warnings.
Further, as the Court has concluded in a separate ruling, a reasonable jury could infer
that AbbVie promoted AndroGel for the treatment of a condition referred to as "Low T,"
knowing that AndroGel had no cognizable benefits for this class of patients. Taken
together, plaintiffs have provided sufficient evidence to give rise to a genuine dispute
regarding whether AbbVie advertised AndroGel to a class of patients who did not need
the drug, without warning of severe side effects. A reasonable jury could find this
conduct sufficiently willful or outrageous to support a claim for punitive damages. See
Proctor v. Davis, 291 Ill. App. 3d 265, 285–286, 682 N.E.2d 1203, 1216 (1997) (finding
sufficient evidence for a punitive damages award where defendant "not only knew of the
adverse effects of periocular use of [the product], but promoted and developed this offlabel use). The Court therefore denies AbbVie's motion for summary judgment on
plaintiffs' requests for punitive damages.
Conclusion
For the foregoing reasons, the Court grants AbbVie's motion for summary
judgment [dkt. no. 1745] on Rowley's strict liability design defect claim, Rowley's breach
of implied warranty claim, and all of the AbbVie bellwether plaintiffs' negligent design
defect claims but otherwise denies the motion.

________________________________
MATTHEW F. KENNELLY
United States District Judge
Date: May 8, 2017
53