Porter, R. v. SmithKline Beecham Corp. ( 2017 )

J -A31039-16
NON-PRECEDENTIAL DECISION - SEE SUPERIOR COURT I.O.P. 65.37
ROBERT AND KATHERINE PORTER, IN THE SUPERIOR COURT OF
INDIVIDUALLY, AND AS PARENTS AND PENNSYLVANIA
NATURAL GUARDIANS OF ROBERT T.
"BO" PORTER, A MINOR
Appellants
v.
SMITHKLINE BEECHAM CORPORATION,
PFIZER, INC. AND WOLTERS KLUWER
HEALTH, INC.
No. 3516 EDA 2015
Appeal from the Order Entered October 8, 2015
in the Court of Common Pleas of Philadelphia County Civil Division
at No(s): September Term, 2007 No. 03275
BEFORE: BENDER, P.J.E., DUBOW, J., and FITZGERALD, J.*
MEMORANDUM BY FITZGERALD, J.: FILED MAY 08, 2017
Appellants, Robert and Katherine Porter, individually, and as parents
and natural guardians of Robert T. "Bo" Porter, a minor, appeal from the
order entered in the Philadelphia County Court of Common Pleas granting
the motion of Appellee, Pfizer, Inc., for summary judgment.' Appellants
* Former Justice specially assigned to the Superior Court.
' On September 29, 2014, the trial court granted GlaxoSmithKlein, LLC's
renewed motion to dismiss and ordered the claims against them be
dismissed with prejudice. See R.R. at 21a. Appellants had settled with
Wolters Kluwer Health, Inc. prior to trial. See id. at 96a. Therefore, this
appeal is properly before this Court. See Pa.R.A.P. 341(b)(1). For the
parties' convenience, we refer to the reproduced record where applicable.
J -A31039-16
contend the trial court erred in precluding the testimony of their expert
witness based upon Frye v. United States, 293 F. 1013 (D.C. Cir. 1923).
We affirm.
The trial court summarized the facts and procedural posture of this
case as follows:
On June 2012 [Appellants] filed an Amended
15,
Complaint against [Appellee] Pfizer alleging that the
ingestion of Zoloft[2] by [Appellant] Mrs. Porter during her
pregnancy caused Minor [Appellant] to be born with the
serious birth defect onnphalocele.[3] On August 14, 2015
[Appellee] filed Frye Motions seeking to preclude the
Expert Testimony of Dr. [Michael] Freedman [M.D., Ph.D.]
and [Robert M.] Cabrera[, Ph.D].[4] On August 26, 2015
[Appellants] filed a Response. A two day hearing was held
on September 16 and September 17, 2015. At that
hearing the court heard from Dr. Freeman and [Appellee's
expert, Dr. Stephen Edward Kimmel M.D.] and received
into evidence numerous documents including the written
report of Dr. Cabrera. On September 30, 2015 Appellee's
Motion was Granted as to Dr. Freeman and he was not
permitted to testify at trial. On October 5, 2015
[Appellee's] Motion was Granted as to Dr. Cabrera and he
2 Zoloft is a sertraline.
Sertraline is a medication used to treat depression,
obsessive -compulsive disorder, panic disorder, and post -
traumatic stress disorder. A brand name for sertraline is
Zoloft. Zoloft belongs to the class of antidepressants
known as selective serotonin reuptake inhibitors (SSRIs).
R.R. at 906a.
3 "An omphalocele is a midline abdominal wall birth defect. This defect
occurs when the intestines and potentially other visceral organs protrude
outside the body through the umbilical opening." R.R. at 959a.
4 See R.R. at 247a.
-2
J -A31039-16
was not permitted to testify at trial. On September
. . .
15, 2015[, Appellee] filed a Motion for Summary
Judgment. On October 8, 2015[, Appellant] filed a
Response to [Appellee's] Motion for Summary Judgment.
On October 8, 2015[, Appellee's] Motion for Summary
Judgment was Granted.
Trial Ct. Op., 2/10/16, at 1-2 (footnotes omitted). This timely appeal
followed. Appellants filed a Pa.R.A.P. 1925(b) statement of errors
complained of on appeal and the trial court filed a responsive opinion.
Appellants raise the following issue for our review: "Did the trial court
improperly preclude Dr. Cabrera from testifying on Frye grounds?"
Appellants' Brief at 3. Dr. Cabrera was offered as an expert "on general and
specific causation." Id. Appellants contend the trial court "overlooked the
general acceptance of Dr. Cabrera's methodological tools, and inserted
[itself] as an independent assessor of Dr. Cabrera's credibility and
persuasiveness." Id. at 22. Appellants argue:
Dr. Cabrera described the foundational principles of the
modern study of teratology as expressed by James Wilson,
the co-founder of the Teratology Society and founder of
the field. The principles include the so-called "dose
response" principle restated in the context of teratology:
"Manifestations of deviant development increase in
frequency and degree as dosage increased from the No
Observable Adverse Effect Level to a dose producing 100%
Lethality." The central point made by Dr. Cabrera is that
low doses "may exert no or very little toxicity or
teratogenicity, while higher doses are expected to increase
the incidence and severity of the observed malformations."
-3
J -A31039-16
Dr. Cabrera explained the importance of animal studies in
rabbits and rats as a vehicle for testing whether a
compound is a human teratogen.
Dr. Cabrera then discussed numerous animal studies that
affirmatively suggest that Zoloft is a teratogen that
causes birth defects.
At the same time, Dr. Cabrera identified shortcomings with
the studies that he found reduced their statistical power.
He ultimately concluded that "while the reported data
provided indications of the presence of cranial, kidney, and
heart defects due to [Zoloft] exposure, the studies lacked
any detailed pathology that might have allowed the
investigators to draw more reasoned conclusions."
For present purposes, the key point is that animal studies
represent a generally accepted tool for building an
assessment about the teratogenicity of a pharmaceutical
compound.
[H]e focuses on the 2007 Louik study[5] The Louik
. . . .
study was a peer reviewed, case -controlled study reflecting
a strong, statistically significant positive correlation
between first trimester Zoloft exposure and omphalocele.
The study found that maternal use of Zoloft during the first
trimester was associated with a nearly six times greater
risk of the infant developing an omphalocele.
5 Louik C., A. Lin, et al. (2007).
E. "First -trimester use of selective
serotonin-reuptake inhibitors and the risk of birth defects." N. Engl. J. Med.
2007, Vol. 356(26): 2675-2683. R.R. at 624a.
-4
J -A31039-16
Of course, Dr. Cabrera did not limit his inquiry to the
Louik study. . .Concededly, some of these studies did
.
not involve enough subjects to show a "statistically
significant" association between Zoloft taken during
pregnancy and omphalocele.
Dr. Cabrera also discussed studies by Reefhuis[61 and
Furum that, although they did not show a statistically
significant association between omphalocele and Zoloft in
particular, did show statistically significant associations
between omphalocele and other SSRIs exhibiting the same
mechanism of action as Zoloft.
Dr. Cabrera made careful and appropriate use of the
Bradford -Hill criteria[8] in further developing his causation
analysis.
6
Reefhuis J., S. M. Gilboa, et al. (2015). "The national birth defects
prevention study: A review of the methods." Birth Defects Res A Clin Mol
Teratol. R.R. at 630a.
Furu, K., H. Kieler, et al. (2015). "Selective serotonin reuptake inhibitors
and venlafaxine in early pregnancy and risk of birth defects: population
based cohort study and sibling design." BMJ 350: h 1798. R.R. at 628a.
8 "The Bradford -Hill criteria were developed as a mean[s] of interpreting an
established association based on a body of epidemiologic research for the
purpose of trying to judge whether the observed association reflects a causal
relation between an exposure and disease." Soldo v. Sandoz Pharms.
Corp., 244 F.Supp.2d 434, 514 (W.D.Pa. 2003). "While we recognize
federal district court cases are not binding on this court, Pennsylvania
appellate courts may utilize the analysis in those cases to the extent we find
them persuasive." Stephens v. Paris Cleaners, Inc., 885 A.2d 59, 68 (Pa.
Super. 2005) (citations omitted).
-5
J -A31039-16
It conceded that Dr. Cabrera's analysis has limitations
is
based on the limitations of the available scientific
evidence, which in turn is based on the impossibility of
directly conducting studies on pregnant women.
In other words, Dr. Cabrera explained, after finding a
general causal relationship between drug and disorder, the
researcher identifies and rules out other potential causes
to determine whether the drug caused the disorder in that
specific instance. That is the differential causation
analysis that Dr. Cabrera performed here.
Appellants' Brief at 26-27, 29-30, 33, 35, 38-39 (citations and footnote
omitted) (emphasis added).
Dr. Cabrera opined:
It my opinion, within a reasonable degree of scientific
is
certainty, that Zoloft (sertraline) is teratogen,[91 both in
animals and in humans, when ingested during pregnancy.
The teratogenicity of sertraline, has been amply
demonstrated in animal studies, as well as in a number of
human epidemiological and registry studies. There exists
a biologically plausible mechanism of teratogenic action
(MOA) based on the MOA of SSRIs. In general, it is my
opinion, within a reasonable degree of scientific certainty,
that alteration of serotonin signaling by sertraline, can
impact embryonic development resulting in several
different congenital malformations, involving various organ
and body systems, including, but not limited to abdominal
wall defects such as omphalocele. It is my opinion that
Kathy Porter's ingestion of Zoloft while pregnant caused Bo
Porter's omphalocele.
R.R. at 578a.
9 "Teratology is the study of abnormal embryonic development. The
founding principles of the modern study of teratogens were initially
articulated by Janes G. Wilson (Wilson 1973), co-founder of The Teratology
Society, in his monograph, 'Environment and birth Defects.' R.R. at 578a.
-6
J -A31039-16
Although the order currently before this court awarded summary
judgment, "an appeal of a final order subsumes challenges to previous
interlocutory decisions," such as preclusion of expert testimony. Betz v.
Pneumo Abex, 44 A.3d 27, 54 (Pa. 2012). "Generally, the appropriate
appellate standard of review is the one pertaining to the underlying ruling."
Id. Instantly, the trial court granted summary judgment after precluding
Appellant's expert testimony. "[The a]ppellant's issue[ ] on appeal,
therefore, challenge[s] the court's preclusion of [her] expert testimony."
Haney v. Pagnanelli, 830 A.2d 978, 980 (Pa. Super. 2003)
Our review is governed by the following principles:
[A]s to the standard of appellate review that applies to the
Frye issue, we have stated that the admission of expert
scientific testimony is an evidentiary matter for the trial
court's discretion and should not be disturbed on appeal
unless the trial court abuses its discretion. An abuse of
discretion may not be found merely because an appellate
court might have reached a different conclusion, but
requires a result of manifest unreasonableness, or
partiality, prejudice, bias, or ill -will, or such lack of support
so as to be clearly erroneous.
Grady v. Frito-Lay, Inc., 839 A.2d 1038, 1046 (Pa. 2003) (citations
omitted).
Rule 702 of the Pennsylvania Rules of Evidence governs the
admissibility of expert opinion. Rule 702 provides as follows:
A witness who is qualified as an expert by knowledge, skill,
experience, training, or education may testify in the form
of an opinion or otherwise if:
-7
J -A31039-16
(a) the expert's scientific, technical, or other specialized
knowledge is beyond that possessed by the average
layperson;
(b) the expert's scientific, technical, or other specialized
knowledge will help the trier of fact to understand the
evidence or to determine a fact in issue; and
(c) the expert's methodology is generally accepted in
the relevant field.
Pa.R.E. 702 (emphasis added).
"The Frye test . . . adopted in Pennsylvania in Commonwealth v.
Topa, [] 369 A.2d 1277 (Pa. 1977), is part of Rule 702." Grady, 839 A.2d
at 1043. "Frye only applies when a party seeks to introduce novel scientific
evidence." Trach v. Fellin, 817 A.2d 1102, 1109 (Pa. Super. 2003) (en
banc). In Betz, our Supreme Court opined that "a reasonably broad
meaning should be ascribed to the term novel." Id. 44 A.3d at 53.
However, "Frye only applies to determine if the relevant scientific
community has generally accepted the principles and methodology the
scientist employs, not the conclusions the scientist reaches, before the court
may allow the expert to testify." Trach, 817 A.2d at 1112. In Trach, this
Court noted that
the scientific method is a method of research in which a
problem is identified, relevant data are gathered, a
hypothesis is formulated from these data, and the
hypothesis is empirically tested. Within the meaning of
the definition of the scientific method, empirical means
provable or verifiable by experience or experiment. Key
aspects of the scientific method include the ability to test
or verify a scientific experiment by a parallel experiment or
-8
J -A31039-16
other standard of comparison (control) and to replicate the
experiment to expose or reduce error.
Id. at 1113 (citations and quotation marks omitted).
Our Pennsylvania Supreme Court in Grady "emphasize[d] that the
proponent of expert scientific evidence bears the burden of establishing all of
the elements for its admission under Pa.R.E. 702, which includes showing
that the Frye rule is satisfied." Grady, 839 A.2d at 1045. Pennsylvania law
does not permit "experts to evade a reasoned Frye inquiry merely by
making references to accepted methods in the abstract." Betz, 44 A.3d at
58.
In the case sub judice, the trial court opined:
[Appellants] seek[ ] to have [Dr. Cabrera] testify as to
general and specific causation of the birth defects of this
case. Within Dr. Cabera's forty-seven page report,[10n are
five pages devoted to his training, education, and
experience and twelve pages devoted to animal studies
concerning SSRIs, Zoloft, and birth defects. The most
recent animal study referenced is a seventeen year old
study from 1998. Animal studies can be instructive in
determining the teratogenicity of a pharmaceutical and
indeed in the absence of human studies may become the
basis for a valid extrapolated scientific opinion. However,
animal studies are of limited utility in determining
teratogenicity where a significant body of human exposure
studies exists in the published medical literature. Dr.
Cabrera does not acknowledge these limitations.
Dr. Cabrera's opinions rely on a limited number of the
peer review articles. Dr. Freeman relied on these same
studies and formed a comparable analysis. Dr. Cabrera's
10 See R.R. at 572a.
-9
J -A31039-16
analysis suffers from many of the same methodological
defects in Dr. Freeman's opinions. The methodological
defects identified as to Dr. Freeman's opinions are
incorporated herein.
Dr. Cabrera's report contains other methodology
failings. Dr. Cabrera finds that the studies show that
SSRIs significantly increase the risk of birth defects in
human studies and opines that SSRIs are teratogenic.
However, he does not specifically analyze SSRIs results
which exclude the pharmaceutical Paxil.E11] This is a fatal
methodological flaw because Paxil has been identified as
having significantly different effects from Zoloft and other
SSRIs. Paxil is a causal factor in birth defects. Dr.
Cabrera's opinion, as reflected in his report, does not
contain any adequate discussion of the differences
between Paxil and Zoloft with respect to causation of birth
defects.
Dr. Cabrera opines that the mechanism of action
resulting in birth defects is that an "alteration of serotonin
signaling by sertraline, can impact embryonic development
resulting in several different congenital malformations.ff[12]
This opinion is speculative and without scientific basis. Dr.
Cabrera presents no information as to the baseline
serotonin level in the developing fetus or the change
caused by Zoloft. Without this data there can be no valid
opinion as to whether the level of serotonin changes in
positive or negative manner or has any outcome
determinative effect at all. Likewise, Dr. Cabrera did not
perform the dose response analysis necessary to draw a
valid scientific conclusion that a medication causes a
specific biological mechanism.
For the reasons precluding the testimony of Dr.
Freeman and those herein Dr. Cabrera likewise was
11"In 2005, Mrs. Porter's Paxil was prescribed by her primary care physician
. . She was switched to Zoloft
. . by her obstetrician
. . . in . . .
approximately the 7th week of her pregnancy." R.R. at 948a.
12 See R.R. at 578a.
- 10 -
J -A31039-16
properly precluded from offering causation opinions in this
matter.
Trial Ct. Op. at 18-19 (footnotes omitted and emphasis added).
The trial court precluded Dr. Freeman from testifying on Frye grounds
and opined:
Dr. Freeman correctly concedes that the studies reveal
precious little specific data on Zoloft and omphalocele. Dr.
Freeman effectively solely relies on the Louik study as
evidence of causation between Zoloft and the birth defect.
The Louik study is the only study to report a statistically
significant association between Zoloft and omphalocele.
Reliance on this one study is questionable because of its
limitations. Louik's confidence interval which ranges
between 1.6 and 20.7 is exceptionally broad. Equally
significant is the lack of power concerning the omphalocele
results. The Louik study had only 3 exposed subjects who
developed omphalocele thus limiting its statistical power.
Studies that rely on a very small number of cases can
present a random statistically unstable clustering pattern
that may not replicate the reality of a larger population.
The Louik authors were unable to rule out confounding or
chance. The results have never been replicated concerning
omphalocele. Dr. Freeman's testimony does not explain,
or seemingly even consider these serious limitations.
[T]he Louik authors themselves expressed concern that
they cannot distinguish true associations from random
elevations of risk. The Louik authors were unable to rule
out the possibility of chance:
"The previously unreported associations we
identified warrant particularly cautious
interpretation. In the absence of preexisting
hypotheses and the presence of multiple
comparisons, distinguishing random variation from
true elevation in risk is difficult. Despite the large
size of our study overall, we had limited numbers to
evaluate associations between rare outcomes and
J -A31039-16
rare exposures. We included results based on small
numbers of exposed subjects in order to allow other
researchers to compare their observations with ours,
but we caution that these estimates should not be
interpreted as strong evidence of increased risks."[13]
Dr. Freeman ignores the issues specifically pointed out by
the authors.
In addition to proper analysis of the appropriate
literature, generally accepted methodology required that
the epidemiologist consider the problem of confounding by
indication. Women who are depressed and take SSRIs
13 The Louik study concluded:
Our findings do not show that there are significantly
increased risks of craniosynostosis, omphalocele, or heart
defects associated with SSRI use overall. They suggest
that individual SSRIs may confer increased risks for some
specific defects, but it should be recognized that the
specific defects implicated are rare and the absolute risks
are small.
R.R. at 915a. Dr. Cabrera relied upon the Louik study in his expert report.
He opined as follows:
Louik and co-workers published on SRRI exposure and
risk of birth defects in the New England Journal of Medicine
(NEJM) in 2007. . . .In regard to study limitations, the
authors report concerns for recall bias were minimized
through multilevel structured questionnaire design. Recall
bias will also not explain specific risk associated with
individual SSRIs. Selection bias was also reported as
unlikely, since mothers were invited to participate without
knowledge of exposure. In regard to confounding by
indication, the authors report, "the absence of significantly
increased risk of various birth defects associated with the
use of non-SSRI antidepressants suggests that depression
itself is unlikely to be the cause of the defects studied."
R.R. at 605a -606a.
- 12 -
J -A31039-16
have been more likely to smoke, be older, have less
education, have poor nutrition, use other drugs, and have
chronic diseases such a diabetes and hypertension, than
women who do not use SSRIs. These factors have been
linked to an increased risk of birth defects. Dr. Freeman
does not properly analyze confounding.
In his report Dr. Freeman cites Jinnene[z]-Solenn,[14]
Furu,[15] and Huybrechts as three studies that also found
14 Dr. Cabrera opined:
Two additional studies that specifically report on SSRIs
or sertraline exposure and risk of omphalocele merit
mentioning. The first is a Danish cohort study on the
teratogenicity of covering pregnancies from 1997-2009
[Jiminez-Solem, E., J. T. Andersen et al. (2012) "Exposure
to selective serotonin reuptake inhibitors and the risk of
congenital malformations: a nationwide cohort study."
BMJ Open 2(3); e001148]. This study reported association
with major congenital malformations and exposure to
sertraline, adjusted [ ] and association between congenital
malformations of the heart and exposure to sertraline,
adjusted [ ]. However, they indicate, "We found an
association for exposure to an SSRI in the first
trimester and craniosynostosis [ ] but not for
omphalocele or anencephaly."
R.R. at 608a (emphasis added), 629a.
15
Dr. Cabrera opined:
The latest study to report on omphalocele is a Nordic
population based cohort study (Denmark, Finland, Iceland,
Norway, and Sweden) conducted between the periods of
1996-2010 (Furu, Kieler et al. 2015). The study utilized
national health registry data from each of the participating
countries. It reported an odds ratio (OR) for SSRI
exposure and omphalocele risk [ ], indicative of a
significant increased risk for omphalocele amongst infants
- 13 -
J -A31039-16
statistically significant increased risk of congenital
malformation associated with SSRI exposure. All three of
these studies acknowledged the problem of confounding,
and discussed the problem in their analysis. Dr. Freeman
does not address the authors' conclusions about
confounding.
Generally accepted methodology considers statistically
significant replication of study results in different
populations because apparent associations may reflect
flaws in methodology. Dr. Freeman claims the Alwan[161
of mothers receiving SSRI treatment during pregnancy,
but no Zoloft-specific risk was reported.
There are three additional studies that examined SSRI
exposure and present general population rates of
omphalocele, but in each report the incidence is too
low generally and in exposed population to perform
statistical testing for risk of omphalocele. These
studies include reports from Danish [Pedersen, L. H., T. B.
Henriksen et al. 2009). "Selective serotonin reuptake
inhibitors in pregnancy and congenital malformations:
population based cohort study." BMJ.1 339:b3569.],
Canadian [Berard, A., J. P. Zhao, et al. (2015) "Sertraline
use during pregnancy and the risk of major
malformations." Am J Obstet Gynecol 212(6): 795 e791-
795 e712.] and European registries [Wemakor, A., K.
Casson et al. (2015). "Selective serotonin reuptake
inhibitor antidepressant use in first trimester pregnancy
and risk of specific congenital anomalies: a European
register -based study." Eur J Epidemiol.]
R.R. at 609a -610a (emphases added), 625a, 627a, 631a.
16 Alwan, S., J. Reefhuis, et al. (2007). "Use of selective serotonin-reuptake
inhibitors in pregnancy and the risk of birth defects." New England Journal
of Medicine 356(26): 2684-2692. R.R. at 621a. The Alwan study concluded
that
[m]aternal use of SSRIs during early pregnancy was not
associated with significantly increased risks of congenital
heart defects or of most other categories of birth defects.
- 14 -
J -A31039-16
and ReefhuisEln studies demonstrate replication. However,
the population Alwan studied is only a subset of the
Reefhuis population and therefore they are effectively the
same. More significantly neither Reefhuis nor Alwan
reported statistically significant associations between
Zoloft and onnphalocele. . .[18]
.
Associations were observed between SSRI use and three
types of birth defects, but the absolute risks were small,
and these observations require confirmation by other
studies.
R.R. at 2014a. The authors identified the three types of birth defects as
anencephaly, craniosynostosis, and omphalocele. Id. at 2016a.
17 Reefhhuis et al. "Specific SSRI's and birth defects: bayesian analysis to
interpret new data in the context of previous reports." BMJ 2015; 350;
h3190. R.R. at 839a. The study acknowledged its limitations, inter alia, as
follows:
This analysis does not address whether the birth defect
associations we observed were caused by maternal SSRI
treatment, underlying maternal disease, or some other
factor. Since there was no specific question on depression
and we cannot identify all participants with untreated
depression, there is the possibility of confounding by
indication.
R.R. at 935a.
18
Dr. Cabrera opined:
In 2007, Alwan also published on SSRI exposure and
risk of birth defects in the NEJM. The study employed a
case -control design using data from the US National Birth
Defects Prevention Study (Alwan, Reefhuis et al. 2007).
The study utilized maternal reporting of SSRI usage
(fluoxetine, sertraline, and paroxetine). The study
reported an odds ratio for SSRI exposure and anencephaly
risk [ ], indicative of a significant increased risk for neural
tube defects amongst infants of mothers receiving SSRI
treatment during pregnancy, and a Zoloft-specific
increased risk [ ]. For those birth defects previously
- 15 -
J -A31039-16
Dr. Freeman agrees that he must, and claims he has,
applied the Bradford -Hill Criteria to support his opinion.
However, the starting procedure of any Bradford -Hill
analysis is "an association between two variables" that is
"perfectly clear-cut and beyond what we would care to
attribute to the play of chance." Dr. Freeman testified that
generally accepted methodology requires a determination,
first, that there's evidence of an association and, second,
whether chance, bias and confounding have been
accounted for, before application of the Bradford -Hill
criteria. Because no such association has been properly
demonstrated, the Bradford -Hill criteria could not have
been properly applied.E19]
associated with SSRI use (anencephaly, craniosynostosis,
and omphalocele) there was also an increased risk [ ]
associated with reported maternal sertraline usage. There
was a significant increased risk [ ] with any SSRI usage
and omphalocele [ ] and a non -significant increased risk [ ]
with sertraline exposure and omphalocele [ ]. Limitations
reported by the authors in the study included the inability
to separate the effect of maternal SSRI use from that of
the underlying depression, under -reporting of other SSRI
usage, potential for recall bias, and selection bias towards
the null hypothesis.
Reefhuis et al. utilized the data from 10 centers in the
United States, as part of the National Birth Defects
Prevention Study, to test SSRI exposure and risk of birth
defects (Reefhuis, Gilboa et al. 2015). The study
specifically reported an odds ratio [ ] for sertraline
exposure and omphalocele risk [ ] indicative of a non-
significant increased risk for omphalocele amongst
infants of mothers receiving SSRI treatment during
pregnancy. Limitations reported by the authors in the
study included the inability to separate the effect of
maternal SSRI use from that of the underlying depression,
self -reporting, multiple testing, and small numbers for
individual defects.
R.R. at 606a -607a (emphasis added).
19
Dr. Cabrera stated:
- 16 -
J -A31039-16
Dr. Freeman opines specifically that [Appellant] Bo
Porter's giant omphalocele birth defect was caused by
exposure to Zoloft in utero.
The temporal relationship between the exposure and
disease is also a factor which must be considered in
assessing specific causation. For an exposure to be the
cause of a disease the exposure must have occurred prior
to the disease. Dr. Freeman fails to address the temporal
failure of exposure between Mrs. Porter's use of Zoloft and
minor plaintiff's giant omphalocele. A Giant Omphalocele
is the result of an incomplete folding of the abdominal wall
during the third to fifth weeks of pregnancy. During the
third to fifth weeks of her pregnancy Mrs. Porter was
taking Peroxetine (a generic version of the known
teratogene Paxil). Mrs. Porter did not being taking Zoloft
until her seventh week of pregnancy. While Dr. Freeman
concedes this timing failure is an issue, he does not form
any opinion of his own but instead claims to defer to other
experts offering opinions which have not been revealed
and therefore necessarily not subject to cross-
examination.
A review of the literature indicates that Zoloft is
associated with teratogenicity. Association, however, does
not alone prove causation and further analysis is
necessary. Sir Bradford -Hill proposed a set of criteria to
determine association between exposure and outcome.
These criteria are accepted in the scientific community as a
viable method of determining the potential existence of
causality.
R.R. at 614a (emphasis added).
- 17 -
J -A31039-16
Clinical differential diagnosis is a generally accepted
methodology. Dr. Freeman is not a clinician and does not
profess to perform a clinical differential diagnosis of cause.
Dr. Freeman fails to properly rule out genetic causes.
Freeman does not discuss and fails to rule out
Dr.
maternal risk factors such as age, obesity, cigarette
smoking, alcohol, maternal stress, and family history. Dr.
Freeman fails to exclude Paxil (Peroxetine) as a risk factor.
Dr. Freeman's failure to analyze, discuss, and exclude
other possible causes departs from generally accepted
methodology.
Trial Ct. Op. at 8-10, 12, 13-15, 17 (footnotes in original omitted).
Stephen Edward Kimmel, M.D., Appellee's expert in epidemiology and
a pharmacoepidemiology, testified at the Frye hearing, inter alia, as follows:
[Appellee's counsel]: Did Dr. Freeman employ generally
accepted methodology?
A: No.
Q: Did you identify a number of specific ways in which you
believe that Dr. Freeman deviated from a generally
accepted methodology?
A: Yes, I did.
Q: And is this-does
this slide summarize the
methodological flaws you identified with respect to Dr.
Freeman's methodology?
A: Yes, it does.
Q: Would you please briefly describe what those flaws are?
- 18 -
J -A31039-16
A: Sure. So there are four groupings listed there. The
first ignoring chance, confounding and bias as a possible
is
cause of false associations, essentially not applying any
epidemiological principles to reviewing the results and
ignoring the lack of replications . . .
Q: I'd just ask you to slow down a little bit, if you don't
mind?
A: And ignoring the lack of replication. The second is
improperly grouping all SSRIs as a class to draw
conclusions about Zoloft. The third is improperly drawing
conclusions about Zoloft and omphalocele based on
findings from other unrelated congenital defects. And the
forth is incorrectly reporting several findings from the
literature.
Q: You're familiar with the Louik study?
A: Yes.
Q: Louik study
The has a finding with respect to
omphalocele, correct?
A: Correct.
Q: So when it says three exposed subjects, what does that
mean?
A: That's only three Zoloft-three women used Zoloft had
omphaloceles. So there's only three Zoloft-exposed
women.
Q: In the Louik group, did they conduct multiple
comparisons?
A: They did.
Q: How many comparisons did they do?
- 19 -
J -A31039-16
A: Just from the paper, 42 initial, 66 additional that they
reported in the paper, so over a hundred comparisons.
Q: Did the authors recognize the limitations of doing a
hundred plus comparisons?
A: Yes, they do. The quote here is that they state in the
presence of multiple comparisons distinguishing random
variation from two elevations in risk is difficult. Meaning,
we can't tell whether these are true findings or false
positives.
Q: So assuming the authors of the Louik study were aware
of the concept of chance, right?
A: Yes.
Q: Is what they did improper by conducting a hundred and
six comparisons?
A: No, it's not improper.
Q: Why would authors like the Louik authors do so many
comparisons understanding the potential limitations that
arise from concepts like chance?
A: Because they were doing it to generate hypothesis, not
to answer or address hypothesis. As they state, in the
absence of preexisting hypothesis. So it was to generate
hypothesis that could then be tested later.
The Court: Could any scientific conclusion about
omphalocele be made on the basis of the Louik paper?
The Witness: No.
The Court: Next question.
[Appellee's counsel]: And have subsequent studies looked
at the issue of Zoloft and omphalocele?
A: Yes.
- 20 -
J -A31039-16
The Court: Can any scientific conclusion be drawn from
about omphalocele from the Louik study and any other
studies subsequent that examined it?
The Witness: Yes, when you put it all together you can.
[Appellee's counsel]: . Dr. Freeman lists four studies
. .
with respect to Zoloft and omphalocele, correct?
A: Correct.
Q: Does Alwan replicate the Louik finding with respect to
omphalocele?
A: No, it does not.
Q: Why not?
A: Because the odds ratio of 1.5 . . . is consistent with the
play of chance.
Q: Andwould it be fair and generally accepted and
accurate to say that Alwan shows that a woman taking
Zoloft had a 50 percent increased risks of having a child
with omphalocele?
A: No.
The Court: Why is that improper to conclude?
The Witness: Because this is based on the results of this
population -based study, they did not demonstrate a
relationship overall between Zoloft and omphalocele that
could be attributed to a real association. The relationship
there is attributable to chance . . . .
[Appellee's counsel]: Dr. Freeman also cites this Jiminez-
Solem finding as at least having a relationship between
Zoloft and omphalocele, correct?
A: He does in his report.
- 21 -
J -A31039-16
what he cited here with respect to
Q: Is the odds ratio in
Jiminez-Solem and the confidence interval, are those
correct?
A: No, they're not.
Q: You've had a chance to take a look at the Jiminez-
Solem paper, right?
A: Yes, I have.
Q: But there is actually data that's reported or information
reported in Jiminez-Solem about abdominal wall defects,
right?
A: There was.
Q: And abdominal wall defects would be something that
would, a category that would encompass omphalocele?
A: Yes, that's correct.
Q: And what were the findings with respect to abdominal
wall defects?
A: So in this study there were no omphalocele defects
among the women who took Zoloft.
Q: . . . And then Reefhuis also looked at Zoloft and
omphalocele, right?
A: Yes.
Q: Do you remember who sponsored the Reefhuis study?
A: That was a study sponsored by the Centers For Disease
Control.
Q: . . . Carol Louik is an author on this paper, right?
- 22 -
J -A31039-16
A: Correct.
Q: That's the same Carol Louik who actually wrote the
Louik paper?
A: Correct.
Q: Dr. Freeman cites this CDC study as supporting his
findings in his paper, right?
A: He does.
Q: And what do the authors say about their own data?
A: So the authors say this is just a quote from their
paper-and I just want to clarify, they were doing this very
specifically because they were trying to answer the
question of does-can Zoloft be associated with
omphalocele because of the prior literature. And what
they found and what they said was it is reassuring that
none of the five previously reported associations between
Zoloft and birth defects, which includes omphaloceles were
confirmed in this analysis.
Q: Now, did Reefhuis and . . . her colleagues employ a
generally accepted methodology?
A: Yes, they did.
Q: Peer reviewed?
A: Yes, it is.
Q: Did Dr. Freeman use a generally accepted methodology
in evaluating these papers that he relied on?
- 23 -
J -A31039-16
A: No, he did not.
Q: Why not?
A: Soseveral reasons. First of all, there was no
application of epidemiologic methods to determine the
issues that we talked about before, chance, bias and
confounding. There was no critical appraisal of these
results. And there was-he did not account for or
acknowledge the play of chance in these findings.
Q: Based on all the data, both the data on Dr. Freeman's
chart and the other data that you evaluated, is there a
consistent association between Zoloft and omphalocele?
A: No, there's none.
Q: Has the CDC also indicated whether SSRIs can be
treated as a class with respect to birth defects generally
and omphalocele specifically?
A: In the Reefhuis paper, yes, they did.
Q: What did the Reefhuis authors of the CDC paper say
specifically about the class effect?
A: So they said, this is a quote, this analysis confirms the
need to assess the association between specific SSRIs and
specific birth defects rather than combining an entire drug
class.
Q: So in viewing all the data regarding SSRIs and class
effect, is generally accepted methodology to treat
it a
SSRIs as class with respect to causation when assessing
birth defects, and in particular, when assessing something
like omphalocele?
A: No, it's not.
- 24 -
J -A31039-16
Q: Did Dr. Freeman do that?
A: He did.
* * *
Q: Now, let's move to your third criticism of Dr. Freeman's
methodology. And this deals again with drawing improper
conclusions about omphalocele based on findings of
unrelated birth defects.
Now, in his report Dr. Freeman grouped together various
types of malformations into a single outcome, right?
A: Yes.
Q: you explain
Can whether such grouping raises
methodological issues?
A: It does.
Q: Can you explain why and how?
A: So, if you find a relationship between a drug and one
type of defect, those data can't be used to say that that
drug causes a different defect. . . .
Q: Do the authors of the CDC study also address generally
accepted methods for evaluating specific birth defects?
A: Yes, they do.
Q: This is from the Reefhuis paper again?
A: Yes, it is.
Q: What do they say?
A: They say the analysis confirms the need to assess
associations between specific birth defects rather than
combining heterogeneous groups of birth defects.
- 25 -
J -A31039-16
Q: Is it consistent with what you believe is generally
accepted in the field of epidemiology?
A: Yes.
Q: Is it consistent with what Dr. Freeman did?
A: No.
R.R. at 5039a, 5041a -5044a, 5046a -5047a.
Dr. Cabrera was deposed and testified, inter alia, as follows:
Q: . .The same type of birth defects that are at issue in
.
this litigation can and do occur in children whose mothers
have not taken Zoloft or any other SSRI during pregnancy,
correct?
A: That is correct.
Q: Am I correct that there's no scientifically validated test,
procedure or protocol that you could point me to in the
medical literature, in a textbook that provides a method to
determine whether or not a specific birth defect in an
individual has been caused or contributed to by Zoloft; is
that right?
A: My understanding is that's something that a medical
doctor would do for individual cases. As far as individual
cases goes, I'm not aware, but I don't practice at-you
know, at a case level for individuals, like a medical doctor.
Q: And my question is directed to your knowledge. You're
not aware, then, of any specific test that would identify
whether or not a specific birth defect in an individual was
caused or contributed to by Zoloft?
A: We're actually developing tests for that, but they're not
available right now.
R.R. at 3670a -3671a.
- 26 -
J -A31039-16
In sum, Dr. Cabrera's opinions rely upon peer review articles. See
R.R. at 605a -610a. Dr. Cabrera relied upon the Louik study. See R.R. at
605a -606a. However, the Louik authors found that their analysis did not
confirm an association between the use of SSRIs and omphalocele. Id. at
919a. Dr. Cabrera refers to the Alwan and Reefhuis studies. Id. at 606a -
607a. The Alwan study concluded that confirmation by other studies was
required. Id. at 2014a. The Reefhuis study stated that its "analysis does
not address whether the birth defect associations we observed were caused
by maternal SSRI treatment." Id. at 935a. Dr. Cabrera concedes that the
Jiminez-Solem, Andersen study did not find an association for exposure to
an SSRI and Omphalocele. See id. at 608a. Furthermore, he acknowledged
that in the Furu, Kieler study, there was no Zoloft specific risk reported with
mothers receiving SSRI treatment during pregnancy. See id. at 609a -610a.
He refers to three other studies, viz., Pedersen, Henriksen, Berard, Zhao,
and Wemakor, Cassib, which concluded that "the incidence is too low
generally and in exposed population to perform statistical testing for risk of
omphalocele" assocation with SSRI exposure. See id. at 610. In his
deposition, Dr. Cabrera conceded he was not aware of any tests that were
available to determine whether Zoloft contributed to any birth defects. See
id. at 3670a -3671a.
It was Appellant's burden to establish all of the elements of Rule 702
for the admission of Dr. Cabrera's expert report. See Pa.R.E. 702. Rule 702
- 27 -
J -A31039-16
includes the Frye test. See Grady, 839 A.2d at 1043. The trial court found
Dr. Cabrera's report contained methodological defects, for the reasons
precluding the expert testimony of Dr. Freeman. See Betz, 44 A.3d at 58.
We agree that Appellants failed to prove the methodology Dr. Cabrerra
employed was generally accepted in the relevant scientific community. See
Trach, 817 A.2d at 1112. Accordingly, we discern no abuse of discretion by
the trial court in precluding Dr. Cabrera from testifying on Frye grounds.
See Grady, 839 A.2d at 1046.
Order affirmed.
Judgment Entered.
J seph D. Seletyn,
Prothonotary
Date: 5/8/2017
- 28 -