# ROBERTS

> District Court, D. New Jersey · November 10, 2025

URL: https://www.frixlaw.com/law-library/cases/11201028

## Case

- **Full name:** Gaston Roberts, et al. v. Zhejiang Huahai Pharmaceutical Co., Ltd., et al.
- **Court:** District Court, D. New Jersey
- **Decided:** November 10, 2025
- **Opinion:** 100trialcourt
- **Cited by:** 0 later opinions in the Frix Law Library

## Citator (automated)

- No negative treatment found by the automated citator. That is not the same as a confirmation that the case is good law; read the citing cases.
- Full citator and citing cases: https://www.frixlaw.com/law-library/cases/11201028

## Opinion text

IN THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF NEW JERSEY
CAMDEN VICINAGE

IN RE: VALSARTAN,
LOSARTAN, AND
IRBESARTAN PRODUCTS
LIABILITY LITIGATION MDL No. 19-2875
(RMB/SAK)

THIS DOCUMENT RELATES TO: OPINION
Gaston Roberts, et al. v. Zhejiang Huahai
Pharmaceutical Co., Ltd., et al.,

Case No. 1:20-cv-00946 (RMB/SAK)

APPEARANCES

C. Brett Vaughn
Daniel Nigh
Ashleigh Raso
NIGH GOLDENBERG RASO & VAUGHN, PLLC
14 Ridge Square NW, 3rd Floor
Washington, DC 20016
Attorneys for Plaintiff Jan Roberts

Jessica Davidson
Allison M. Brown
KIRKLAND & ELLIS LLP
601 Lexington Avenue
New York, New York 10022
Nina R. Rose
Jordan M. Schwartz
KIRKLAND & ELLIS LLP
1301 Pennsylvania Avenue, N.W.
Washington, D.C. 20004
Attorneys for Defendants Zhejiang Huahai Pharmaceutical Co., Ltd., Solco Healthcare
U.S., LLC, and Prinston Pharmaceutical Inc.
RENÉE MARIE BUMB, Chief United States District Judge:

Now before the Court are a series of pre-trial motions in what was to be the first
bellwether trial in this multi-district litigation (“MDL”).1 Of particular import to both
the Court and the parties and, therefore, the focus of this Opinion is the Motion to
Exclude the Opinions of Dr. Fareeha Siddiqui [Docket No. 3068] filed by the
Defendants.
The Court prefaces this Opinion by highlighting the “special dangers to the
fact-finding process” inherent in scientific expert testimony. See United States v. Ford,

1 Plaintiff Jan Roberts (“Plaintiff”), as personal representative and spouse of
Mr. Gaston J. Roberts, Jr. (“Mr. Roberts”) has moved to exclude the testimony of
defense experts Dr. Andrew Thompson [Docket No. 3057], Dr. Victoria Chernyak
[Docket No. 3066], Dr. Nadim Mahmud [Docket No. 3067], and Dr. Gregory Diette
[Docket No. 3070]. The Court heard testimony from Dr. Thompson and, as set forth
on the record, had indicated its intention to reserve its decision. The Court also heard
testimony from Dr. Mahmud and, for the reasons set forth on the record, Plaintiff’s
motion to exclude his testimony was denied. Given the impact of the instant Opinion,
the motions to exclude the testimony of Drs. Diette and Chernyak are denied as moot.
Defendants Zhejiang Huahai Pharmaceuticals Co., Ltd., Prinston
Pharmaceutical Inc. d/b/a Solco Healthcare LLC, and Solco Healthcare U.S.
(collectively, the “Defendants” or “ZHP”) have moved to exclude the testimony of
Plaintiff’s experts Dr. John Russo [Docket No. 3058] and Dr. Fareeha Siddiqui
[Docket No. 3068], as well as for summary judgment [Docket No. 3061]. The parties
have resolved the issues as to Dr. Russo. Accordingly, the motion to exclude his
testimony is denied as moot. Defendants also moved to exclude the testimony of
Plaintiff’s expert Dr. William Sawyer [Docket No. 3060], which was referred by this
Court to the Special Master, Hon. Thomas I. Vanaskie (ret.). Judge Vanaskie denied
that motion [Docket No. 3167]. Defendants indicated their intent to appeal that denial
to this Court. In light of the instant Opinion, any appeal would be moot. But the
Court finds that Defendants have preserved their position should Dr. Sawyer’s
testimony be presented in another action in this MDL. [See 9/3/25 Tr. 87:7-88:4
(Docket No. 3170).]
481 F.3d 215, 220 n. 6 (3d Cir. 2007) (quoting Daubert v. Merrell Dow Pharms., Inc.,
43 F.3d 1311, 1321 (9th Cir. 1995)). “Expert evidence can be both powerful and quite
misleading because of the difficulty in evaluating it.” Daubert v. Merrell Dow Pharms.,

Inc., 509 U.S. 579, 595 (1993); see also United States v. Soler-Montalvo, 44 F.4th 1, 16 (1st
Cir. 2022) (“there is a particular worry with expert testimony that jurors may assign
more weight to it than it deserves. Expert testimony can carry with it an unwarranted
aura of special reliability and trustworthiness.”) (cleaned up). To safeguard against

these “special dangers,” district courts must “tread carefully when evaluating proffered
expert testimony.” Ford, 481 F.3d at 220 n. 6. Given these “special dangers,” this
Court, along with its sister courts, has been entrusted with an essential and rigorous
gatekeeping duty to ensure the reliability and relevance of an expert’s testimony prior
to a jury considering it. See Cohen v. Cohen, 125 F.4th 454, 460 (3d Cir. 2025).

Plaintiff implores the Court to allow the jury to sort through Dr. Siddiqui’s
testimony. This Court cannot do that – essentially taking the expert’s word for it –
without abandoning its gatekeeping responsibility. FED. R. EVID. 702, Advisory
Committee’s note to 2000 amendments. “Judicial gatekeeping is essential because just
as jurors may be unable, due to lack of specialized knowledge, to evaluate

meaningfully the reliability of scientific and other methods underlying expert opinion,
jurors may also lack the specialized knowledge to determine whether the conclusions
of an expert go beyond what the expert’s basis and methodology may reliably
support.” FED. R. EVID. 702, Advisory Committee’s note to 2023 amendments.
Having considered the parties’ submissions2 and having had the benefit of
extensive oral argument and a Daubert hearing, the Court now exercises its essential
gatekeeping function and resolves the motion before it. For the reasons set forth

herein, Defendants’ Motion to Exclude the Opinions of Dr. Siddiqui is GRANTED
in full.
I. PROCEDURAL HISTORY
This Court has had occasion to describe this litigation before:
This MDL as a whole is sprawling, encompassing dozens of classes of
plaintiffs, upstream and downstream defendants, and multiple theories
of liability covering both economic losses and personal injury. Valsartan
is the generic name of the now off-patent anti-hypertensive drug
Diovan®. It is also used in a combination heart failure drug called
Exforge®. At its core, the litigation involves the alleged contamination
with nitrosamines of certain Valsartan-containing drugs manufactured,
distributed, or sold by Defendants (the “VCDs”) from January 1, 2012
through the recalls of those drugs in the summer of 2018.3
In re Valsartan, Losartan, & Irbesartan Prods. Liab. Litig., 2025 WL 1024048, at *1
(D.N.J. Apr. 7, 2025).

2 Defendants filed a brief in support of their motion to exclude Dr. Siddiqui’s
expert testimony [Defs.’ Br. (Docket No. 3069)]. Plaintiff has opposed [Pl.’s Opp’n
(Docket No. 3097)] and Defendants have submitted a reply brief in further support of
the motion [Defs.’ Reply (Docket No. 3117)]. At the direction of the Court, the parties
filed supplemental submissions in support of their respective positions. [Defs.’ Supp.
Br. (Docket No. 3160); Pl.’s Supp. Br. (Docket No. 3161).]
3 The specific nitrosamine alleged to have been present in the VCDs consumed
by Mr. Roberts is N-Nitrosodimethylamine (“NDMA”), which has been classified as
probable genotoxic carcinogen by the U.S. Food and Drug Administration (“FDA”).
This Opinion concerns just one of over more than one thousand member cases
in the MDL, which was chosen by the parties as the first bellwether personal injury
trial. This action was filed by Mr. Roberts and his wife in January 2020, asserting

various claims under Alabama state law.4 In March 2020, Mr. Roberts passed away
and his wife was substituted as Plaintiff as the personal representative of her husband’s
estate. [Docket No. 6.] A jury trial was scheduled to begin in early September 2025.
In advance of trial, the parties submitted various pre-trial motions. The Court held
Daubert hearings and oral argument on the motions on August 26, 2025 and

September 3, 2025. The Court indicated to the parties on the record that, after
thoroughly considering the parties’ arguments and Dr. Siddiqui’s live testimony, it
intended to exclude the testimony of Dr. Siddiqui. [9/3/25 Tr. 8:16-9:2.] In response,
Plaintiff conceded that without Dr. Siddiqui’s testimony – Plaintiff’s only specific
medical causation evidence – summary judgment in favor of Defendants was

appropriate on all claims. [Id. 9:9-11:15.] As a result, the Court adjourned the trial.
II. FACTUAL BACKGROUND
The following facts are generally not in dispute, unless noted. Mr. Roberts
began taking Defendants’ NDMA-contaminated VCDs in September 2016. His last
prescription for the affected valsartan was filled in June 2018. On July 13, 2018,

4 Plaintiff intended to assert the following claims at trial: (i) manufacturing defect,
(ii) failure to warn, (iii) negligence, (iv) negligence per se, (v) common law fraud,
(vi) wrongful death, (vii) survival action, (viii) loss of consortium, and (ix) punitive
damages. Plaintiff voluntarily withdrew her breach of warranty and Alabama
Deceptive Trade Practices Act claims [Docket No. 3139].
Defendants voluntarily recalled their affected VCDs due to the detection of NDMA.
In August 2018, Mr. Roberts was diagnosed with hepatocellular carcinoma (“HCC”),
a common and aggressive form of liver cancer. Mr. Roberts unfortunately passed

away in March 2020 at the age of sixty-four.
The heart of the parties’ dispute is whether Plaintiff can establish specific
causation in light of Mr. Roberts’s other extensive and significant risk factors for
HCC.5 In other words, does Plaintiff have reliable, admissible evidence from which a
jury could find that Mr. Roberts’s cancer was caused by his exposure to NDMA in his

valsartan medication, as opposed to his numerous other risk factors? For the reasons
set forth herein, the Court finds that she does not.
A. MASH and MASLD
Mr. Roberts had a decades-long history of liver problems. In August 2009,

Mr. Roberts was referred by his physician for abnormal liver function tests. [Rose
Decl.6 Ex. 6 at GRobertsJr-CA-000661 (Docket No. 3072-5).] He reported struggling
with elevated “liver numbers” since his teenage years and that “[h]e has been told over

5 Defendants also contest Plaintiff’s ability to prove general causation. As the
adequacy of Plaintiff’s evidence establishing general causation is not at issue in the
motions presently before the Court, the Court does not address it.
6 “Rose Decl.” refers to the Declaration of Nina Rose in support of Defendants’
Motion to Exclude the Opinions of Dr. Fareeha Siddiqui [Docket No. 3071].
On occasion, the Court relies upon exhibits appended to the Declaration of Nina Rose
in support of Defendants’ Motion for Summary Judgment and Statement of Material
Facts Not in Dispute (“Summ. J. Rose Decl.”) [Docket No. 3064].
the years that he probably has a fatty liver, i.e. NASH.”7 [Id.] In September 2009,
after test results showing “abnormal mild transaminitis,”8 Mr. Roberts underwent an
ultrasound, which “demonstrate[d] fatty liver.” [Summ. J. Rose Decl. Ex. 27 at

GRobertsJr-CA-000654 (Docket No. 3065-26).] His treating gastroenterologist,
Dr. Christopher W. Ives, indicated that MASH was his primary impression of
Mr. Roberts’s condition, though he also noted that there was “no evidence of
decompensation at least biochemically and physical exam showed no evidence of
portal hypertension.” [Id.] Dr. Ives and Mr. Roberts discussed the possibility of a liver

biopsy, but Mr. Roberts was not interested, which his physician noted was
“understandable and probably not completely necessary at this time.” [Id.]
In March 2011, Mr. Roberts was seen again by Dr. Ives, who listed noted
nonspecific abnormal results of liver function study under Mr. Roberts’s past medical

history. [Rose Decl. Ex. 17 at GRobertsJr-ESMS-000017 (Docket No. 3072-16).] He
was not seen by Dr. Ives again until July 2015. [Id. at GRobertsJr-ESMS-000045.]

7 “NASH” refers to non-alcoholic steatohepatitis, which is now referred to as
metabolic dysfunction-associated steatohepatitis or “MASH.” The related condition
previously referred to as non-alcoholic fatty liver disease or “NAFLD” is now referred
to as metabolic dysfunction-associated steatotic liver disease or “MASLD.” The
nomenclature changed in 2023. [See Defs.’ SMF ¶ 38.] The parties and their experts
often utilize the current and outdated terminology interchangeably. The Court, for
clarity, will use “MASH” or “MASLD” as appropriate, unless quoting directly from
the record.
8 Transaminitis occurs when a patient has elevated levels of certain liver enzymes
in their blood, which indicates underlying liver stress or damage. [See Mahmud
Tr. 144:23-145:5 (Docket No. 3168).]
Dr. Ives noted both MASH and nonspecific abnormal results of liver function study
under Mr. Roberts’s past medical history. [Id.] Mr. Roberts visited Dr. Ives again in
early April 2016 because of constant abdominal pain that had been ongoing for about

a year. [Id. at GRobertsJr-ESMS-000006.] He saw Dr. Ives a few months later for
continued abdominal pain. [Id. at GRobertsJr-ESMS-000002.] Dr. Ives noted that his
“LFTs” (or liver function tests) were elevated and listed MASH and abnormal liver
function study as past medical diagnoses. [Id.] Dr. Ives’s own assessment included

“fatty liver.” [Id. at GRobertsJr-ESMS-000005.]
Mr. Roberts’s medical records from January 2018 indicate that his “past
medical history” included MASH and “[n]onspecific abnormal results of liver function
study.” [Id. at GRobertsJr-CA-000210, GRobertsJr-CA-000215.]
B. Cirrhosis

Mr. Roberts also had a history of liver fibrosis and cirrhosis. It is unclear from
his medical records when this first arose. His medical records from 2009 permitted
defense expert Dr. Mahmud to calculate his Fib-4 score, which was 1.99, representing
indeterminate liver fibrosis.9 [Mahmud Report at 21.] Laboratory results from

9 As Dr. Mahmud explained, “[t]he current best practice recommendations from
major society guidance are to calculate a blood-based fibrosis score called the Fib-4,
which is computed using a patient’s age, AST, ALT, and platelet count.” [Mahmud
Report at 18 (Docket No. 3067-3).] Fib-4 scores are an important “risk stratification
tool” routinely used “to identify patients who are highly likely to have cirrhosis.”
[Mahmud Tr. 147:10-23.] “In patients with MASLD, a Fib-4 <1.3 effectively rules
out advanced fibrosis, with a ≥90% negative predictive value. . . . By contrast, a Fib-4
>2.67 denotes a very high risk of advanced fibrosis. Specifically, patients with a Fib-4
>2.67 have a ~80% positive predictive value of having cirrhosis. In other words,
November 2015 showed that Mr. Roberts’s Fib-4 score had increased to 3.22, which
indicated advanced fibrosis or possible cirrhosis. [Id.]
In early April 2016, Mr. Roberts underwent an ultrasound scan of his abdomen.

The scan showed “hepatomegaly measuring 20 cm,” “fatty hepatic infiltrate,” and
“2 solid densities” in “the liver measuring 2.0 and 1.8 cm at their widest diameter.”
[Vaughn Cert. Ex. 17 at Infirmary_GRobertsJr00014 (Docket No. 3097-18).] A
follow-up CT scan later that month showed that the “[p]eripheral margin of
[Mr. Roberts’s] liver [was] somewhat lobulated” and that an “8 mm focal hypodensity

[was] seen in the right lobe.” [Rose Decl. Ex. 51 at GRobertsJR-TH-MD-000944
(Docket No. 3072-50).] The scan showed other cirrhosis symptoms, including
“enlargement of the caudate lobe, and diffuse heterogenous attenuation throughout
the liver parenchyma,” and “recanalization of the paraumbilical vein.” [Mele Report
at 2 (Docket No. 3072-38).] The radiologist reviewing the CT scan stated that the

“findings above may be evidence of liver cirrhosis.” [Rose Decl. Ex. 51 at
GRobertsJR-TH-MD-000944.] As to the hypodensity, the radiologist noted that it was
“too small to definitively characterize. Benign or malignant etiologies could have this
appearance.10 . . . Follow-up liver protocol CT 4-6 months from date of study would

among 100 patients with Fib-4 >2.67, about 80 of these patients will truly have
cirrhosis based on this test result alone. . . . Finally, patients with a Fib-4 between 1.3
and 2.67 are in an indeterminate range for advanced fibrosis/cirrhosis and should
undergo [further testing] as a next step to determine the likelihood of cirrhosis and
inform the need for hepatology referral.” [Mahmud Report at 18.]
10 Plaintiff’s expert radiologist, Dr. Christopher Mele, agreed that it was
impossible to characterize whether the hypodensities on the CT scan were malignant
be useful for further characterization.” [Id.] No such follow-up took place. At the
time of his cancer diagnosis in August 2018, described below, Mr. Roberts had
advanced decompensated cirrhosis. [Siddiqui Dep. 136:20-21, 267:10-12 (Docket

No. 3072-6).]
C. Obesity and Diabetes
Mr. Roberts’s medical records show that from 2008 through early 2018, he
consistently had a Body Mass Index (“BMI”) score indicating that he was obese or

morbidly obese. [Rose Decl. Ex. 17 at GRobertsJr-AMG-000028-30, GRobertsJr-CA-
000210, GRobertsJr-CA-000722, GRobertsJr-ESMS-000004, GRobertsJr-ESMS-
000047, GRobertsJr-ESMS-000053.] His medical records note that Mr. Roberts had
diabetes as early as December 2007, when his records showed his A1C at 6.9%. [Rose
Decl. Ex. 12 at GRobertsJr-CA-000729 (Docket No. 3072-11).] In July 2016, medical

records note that Mr. Roberts had Type 2 diabetes mellitus with hyperglycemia and
that his A1C levels were above 9.0%. [Rose Decl. Ex. 13 at GRobertsJr-AMG-000042-
44 (Docket No. 3072-12).]
D. HCC (Liver Cancer) Diagnosis

On July 17, 2018, Mr. Roberts presented to the emergency department at
Thomas Hospital in Fairhope, Alabama due to worsening abdominal pain in his left

or benign. [Mele Dep. 161:22-162:3 (Docket No. 3072-39).] Dr. Samuel Hooks,
Mr. Roberts’s treating gastroenterologist, could not exclude the possibility that the
lesion “evolved into” the HCC tumors identified in 2018. [Hooks Dep. 73:20-23
(Docket No. 3072-36).]
side and radiating to the back. [Summ. J. Rose Decl. Ex. 34 at GRobertsJR-TH-MD-
001026 (Docket No. 3065-33).] Mr. Roberts reported that he had been experiencing
this pain for two to three days, as well as nausea. [Id.] A CT scan of Mr. Roberts’s

abdomen and pelvis revealed an enlarged liver and spleen with “evidence of diffuse
hepatic parenchymal disease likely cirrhosis.” [Id. at GRobertsJr-TH-MD-001028.]
A right upper quadrant ultrasound revealed an enlarged liver, hepatic steatosis, and a
“vague hypoechoic area within the right lobe of the liver.” [Id.] A follow-up MRI was

recommended. [Id.] “Probable cirrhosis” and history of MASH were noted. [Id.]
The follow-up MRI showed multiple masses that were “highly suspicious for
multifocal HCC,” “especially in the setting of cirrhosis.” [Rose Decl. Ex. 24 at
GRobertsJr-PPR-000106 (Docket No. 3072-23); Rose Decl. Ex. 27 at GRobertsJr-
SouCC-000251 (Docket No. 3072-26).]

In August 2018, Mr. Roberts’s hepatologist, Dr. Jared White, noted that he had
“NASH cirrhosis and two hepatic lesions . . . concerning for HCC.” [Rose Decl.
Ex. 29 at GRobertsJr-PPR-000093 (Docket No. 3072-28).] Dr. White further noted
that Mr. Roberts had undergone an “extensive work up for cirrhosis” which “is
essentially negative.” [Id.] Dr. White noted that Mr. Roberts did not have a

“significant [history of] liver disease other than risk factors for NASH cirrhosis” and
that, while Mr. Roberts was “[w]ell compensated, [his liver] has cirrhotic
configuration.” [Id. at GRobertsJr-PPR-000097.]
Shortly thereafter, Mr. Roberts was diagnosed with HCC. [Rose Decl. Ex. 23
at GRobertsJr-UABHIM-MD-000002-6 (Docket No. 3072-22).] His treating radiation
oncologist described Mr. Roberts’s condition as HCC “arising from background of
NASH cirrhosis.” [Rose Decl. Ex. 5 at GRobertsJr-UABHIM-MD-000083 (Docket
No. 3072-4).] In December 2018, Mr. Roberts’s physician characterized his HCC as

an advanced “Stage IIIA” carcinoma and noted that Mr. Roberts had cirrhosis
secondary to MASH. [Rose Decl. Ex. 25 at GRobertsJr-SouCC-000235 (Docket
No. 3072-24).] Mr. Roberts continued to be treated at Thomas Hospital in Fairhope,
Alabama, from his admission to the emergency department in July 2018 until his death
on March 17, 2020. [Defs.’ SMF ¶¶ 60–61.]

III. THE LEGAL STANDARDS GOVERNING THE ADMISSIBILITY OF
EXPERT TESTIMONY
Federal Rule of Evidence 702 governs the admissibility of expert testimony,
permitting a witness “qualified as an expert by knowledge, skill, experience, training,
or education” to testify in the form of an opinion, provided that “the proponent
demonstrates to the court that it is more likely than not that:”
(a) 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;
(b) the testimony is based on sufficient facts or data;
(c) the testimony is the product of reliable principles and methods; and
(d) the expert’s opinion reflects a reliable application of the principles and
methods to the facts of the case.
FED. R. EVID. 702.11 The party that proffers the expert testimony bears the burden of
establishing its admissibility by a preponderance of the evidence. Daubert, 509 U.S. at
592 n.10 (citing Bourjaily v. United States, 483 U.S. 171, 175–76 (1987)).

Because Rule 702 “clearly contemplates some degree of regulation of the
subjects and theories” to which an expert may testify, the Supreme Court has stated:
[I]n order to qualify as “scientific knowledge,” an inference or assertion
must be derived by the scientific method. Proposed testimony must be
supported by appropriate validation—i.e., “good grounds” based on
what is known. In short, the requirement that an expert’s testimony
pertain to “scientific knowledge” establishes a standard of evidentiary
reliability.
Id. at 590; see also Oddi v. Ford Motor Co., 234 F.3d 136, 144–45 (3d Cir. 2000).
The Court’s Function as Gatekeeper
The Court must act as a “gatekeeper” to prevent expert testimony running afoul
of Rule 702 from ever reaching the jury. See Daubert, 509 U.S. at 596– 97; see also Cohen,
125 F.4th at 460 (“District courts are tasked with a ‘rigorous gatekeeping function’”)
(quoting Elcock v. Kmart Corp., 233 F.3d 734, 741 (3d Cir. 2000)). “This gatekeeping
function is necessarily ‘flexible,’ granting district courts ‘latitude in deciding how’ these
requirements are met.” Cohen, 125 F.4th at 460 (internal citations omitted). “But this
leeway ‘is not discretion to abandon the gatekeeping function’ or ‘perform the function

11 The Court applies the current version of Rule 702, which was amended effective
December 1, 2023. The amendment does not substantively alter Rule 702, but rather
“clarifies that the preponderance standard applies to the three reliability-based
requirements added in 2000—requirements that many courts have incorrectly
determined to be governed by the more permissive Rule 104(b) standard.” FED. R.
EVID. 702, advisory committee’s note to 2023 amendments.
inadequately.’” Id. (quoting Kumho Tire Co. v. Carmichael, 526 U.S. 137, 158–59 (1999)
(Scalia, J., concurring)).
Thus, the Court “must determine . . . whether the expert is proposing to testify

to (1) scientific knowledge that (2) will assist the trier of fact to understand or determine
a fact in issue.” Daubert, 509 U.S. at 592. “This entails a preliminary assessment of
whether the reasoning or methodology underlying the testimony is scientifically valid
and of whether that reasoning or methodology properly can be applied to the facts in
issue.” Id. at 592–93. The Third Circuit has described Rule 702 as embodying a

“trilogy of restrictions on expert testimony: [1] qualification, [2] reliability, and [3] fit.”
Calhoun v. Yamaha Motor Corp., U.S.A., 350 F.3d 316, 321 (3d Cir. 2003) (quoting
Schneider v. Fried, 320 F.3d 396, 405 (3d Cir. 2003)); see also Cohen, 125 F.4th at 460.
The Trilogy of Restrictions on Expert Testimony

First, the witness must be qualified to testify as an expert, which requires “that
the witness possess specialized expertise.” Calhoun, 350 F.3d at 321. This
requirement, however, has been interpreted liberally to encompass “a broad range of
knowledge, skills, and training.” In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 741
(3d Cir. 1994).

Second, the testimony must be reliable, which demands that “the expert’s
opinion must be based upon the ‘methods and procedures of science,’ rather than on
‘subjective belief or unsupported speculation.’” Calhoun, 350 F.3d at 321 (quoting
Paoli, 35 F.3d at 742). In other words, the Court must ensure that the evidence
adduced is scientifically valid. Daubert, 509 U.S. at 590. In determining the reliability
of expert testimony, the Court is guided by the following factors:
(1) whether a method consists of a testable hypothesis; (2) whether the
method has been subject to peer review; (3) the known or potential rate
of error; (4) the existence and maintenance of standards controlling the
technique’s operation; (5) whether the method is generally accepted;
(6) the relationship of the technique to methods which have been
established to be reliable; (7) the qualifications of the expert witness
testifying based on the methodology; and (8) the non-judicial uses to
which the method has been put.

Calhoun, 350 F.3d at 321 (quoting Paoli, 35 F.3d at 742 n.8). The Court is not restricted
to any “definitive checklist or test.” Daubert, 509 U.S. at 593. This inquiry is “a flexible
one” focusing “solely on the principles and methodology, not on the conclusions that
they generate.” Id. at 595.
While reliability does not require “correctness,” it does prohibit “too great a gap
between the data and the [expert’s] opinion proffered.” Oddi, 234 F.3d at 146
(quoting Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997)). Thus, the court “must
examine the expert’s conclusions in order to determine whether they could reliably
flow from the facts known to the expert and the methodology used.” Id. (quoting
Heller v. Shaw Indus., Inc., 167 F.3d 146, 153 (3d Cir. 1999)).
Third, the expert’s testimony must “fit” the case. Daubert, 509 U.S. at 592.
Otherwise known as the “helpfulness” standard, this restriction requires there to be “a
valid scientific connection to the pertinent inquiry as a precondition to admissibility.”
Id. at 591–92. The fit requirement “goes primarily to relevance.” Id. at 591. “[T]he
expert’s testimony must be relevant for the purposes of the case and must assist the
trier of fact.” Schneider, 320 F.3d at 404 (citations omitted); United States v. Zimmerman,
277 F.3d 426, 433 n.4 (3d Cir. 2002) (“It is well-established that an expert opinion
must be tailored to the specific facts of the case to have any value.”). While not a high

bar, the standard is “higher than bare relevance.” Ford, 481 F.3d at 220 n.6 (quoting
Paoli, 35 F.3d at 745).
IV. MOTION TO EXCLUDE DR. SIDDIQUI’S TESTIMONY
The Court now considers Defendants’ challenges to Dr. Siddiqui’s expert
testimony. The parties agree that the Court’s resolution of this motion is case

dispositive if the Court excludes Dr. Siddiqui’s testimony. Put plainly, Dr. Siddiqui’s
testimony is Plaintiff’s sole evidence of specific causation, and, without her testimony,
Plaintiff concedes that her case fails. Defendants contend that Dr. Siddiqui’s
testimony should be excluded because the differential diagnosis underlying her
opinion is fundamentally flawed in that she both failed to reliably “rule in”

Mr. Roberts’s exposure to NDMA in valsartan and “rule out” Mr. Roberts’s numerous
well-established independent risk factors for HCC.
A. Dr. Siddiqui’s Qualifications
Defendants do not dispute Dr. Siddiqui’s qualifications. She is a board-certified

hematologist and medical oncologist. She has over twenty years of clinical experience
in internal medicine, hematology, and oncology. After completing her residency in
internal medicine in 2003, she completed a three-year double fellowship in hematology
and oncology. Dr. Siddiqui spent about half of this fellowship studying the causation,
diagnosis, and treatment of solid cancers, such as HCC. She is currently an Associate
Physician Diplomate of Medicine at the University of California, San Diego. As part
of her practice, she diagnoses and treats patients with liver cancer daily. [Siddiqui
Report at 1–2 (Docket No. 3072).]

B. Dr. Siddiqui’s Methodology and Opinions
Plaintiff offers Dr. Siddiqui as a specific causation expert in this case.
Dr. Siddiqui testified that, in her medical opinion, Mr. Roberts’s HCC was caused by
his exposure to NDMA that was present in ZHP’s valsartan pills. Specifically, she

opined that Mr. Roberts’s exposure to NDMA was a substantial factor in causing his
HCC.12 She further opined that his myriad other risk factors, such as cirrhosis,
diabetes, obesity, and MASH/MASLD, were not substantial factors in causing his
cancer.
To reach this conclusion, Dr. Siddiqui performed a differential diagnosis, which

involves ruling in and ruling out possible causes and contributing factors of
Mr. Roberts’s HCC. In carrying out her differential diagnosis, Dr. Siddiqui first
identified risk factors associated with developing HCC. To do so, she reviewed
medical and scientific literature, medical databases, and guidance from organizations
like the American Cancer Society and the NIH regarding the risk factors for HCC.

[Siddiqui Tr. 25:3-11 (Docket No. 3168).] Although these organizations did not list

12 Dr. Siddiqui defined a “substantial factor” as “a factor that a reasonable person
would consider to have been a cause of the cancer. A substantial [factor] must be more
than a remote or trivial factor. A substantial factor does not have to be the only cause
of the harm.” [Siddiqui Report at 1.]
NDMA specifically as a risk factor, carcinogens in general were at times included.
Because NDMA is a carcinogen, Dr. Siddiqui included it as a risk factor to consider
in her differential diagnosis. [Id. 25:12-25.]

Dr. Siddiqui then determined which of these risk factors applied to Mr. Roberts
given his medical and social history. She was able to rule out several risk factors that
were inapplicable to Mr. Roberts.13 [Siddiqui Report at 29; Siddiqui Tr. 26:6-20.]
For example, viral hepatitis and chronic excessive alcohol consumption are risk factors
for HCC but could readily be ruled out because Mr. Roberts did not have viral hepatitis

and was not a heavy drinker. [Siddiqui Tr. 26:6-14.] She narrowed the relevant risk
factors to NDMA in valsartan, pre-existing cirrhosis, MASLD/MASH, obesity,
diabetes, and hyperlipidemia. [Siddiqui Report at 29–31.]
Finally, Dr. Siddiqui considered each of these applicable risk factors, ruling
each either in or out to determine which was a substantial factor in causing

Mr. Roberts’s cancer. As explained below, she ruled out Mr. Roberts’s diabetes,
MASLD/MASH, obesity, and cirrhosis. She ultimately opined that Mr. Roberts’s
exposure to NDMA in his valsartan was a substantial factor in causing his HCC.

13 These risk factors included viral hepatitis, chronic excessive alcohol
consumption, tobacco use, aflatoxins, anabolic steroids, gender, race/ethnicity, age,
iron overload, Glycogen Storage Disease, Alpha-1 Antitrypsin Deficiency,
hypercitrullinemia, Alagille Syndrome, and Acute Intermittent Porphyria.
i. Diabetes, MASLD/MASH, and Obesity Ruled Out
Dr. Siddiqui considered several of Mr. Roberts’s known risk factors “under the
umbrella of metabolic syndrome.” [Siddiqui Tr. 52:14-18; see also Siddiqui Report

at 29.] Dr. Siddiqui opined that “MASLD is a cluster of conditions – insulin
resistance, hypertension, hypertriglyceridemia, and abdominal obesity.” [Siddiqui
Report at 21.] She explained that this “metabolic syndrome” is an umbrella term that
includes “having obesity, having hyperlipidemia, even hypercholesteremia sometimes,
diabetes.” [Siddiqui Tr. 52:14-24.]

In her view, obesity is not a significant risk factor alone for liver cancer and is
only associated with an increased risk for developing liver cancer because it can result
in MASLD and cirrhosis. [Siddiqui Report at 22.] Dr. Siddiqui opined that “the
increased risk of liver cancer due to obesity would have been negligible, and again best
captured based on the degree of his liver cirrhosis.” [Id. at 30.] She applied the same

reasoning to Mr. Roberts’s hyperlipidemia, which she states increases the risk of liver
cancer by causing MASH, which then can cause cirrhosis. [Id.]
Dr. Siddiqui similarly recognized that “Type 2 diabetes is linked with an
increased risk of liver cancer” because it can contribute to MASLD. [Id. at 22–23.]
She also noted that Mr. Roberts was not diagnosed with diabetes until November 14,

2016 (after he began taking the contaminated valsartan) and that his diabetes was well
controlled. [Id. at 30.]
Turning to MASLD, Dr. Siddiqui recognized that it “has been increasingly
linked to liver cancer, especially in individuals with progression to [MASH].”
[Siddiqui Report at 21.] But she reasoned that the resulting increased risk of liver
cancer would be captured by the degree of his liver cirrhosis. [Id. at 30.] She explained
that the progression from MASLD/MASH to cirrhosis is “slow” and “stepwise,” as

is the progression from cirrhosis to HCC. [Siddiqui Tr. 53:2-6.] In her opinion,
Mr. Roberts’s progression from a questionable MASH diagnosis to eventual advanced
cirrhosis and cancer was inconsistent with this slow, stepwise progression.
Accordingly, Dr. Siddiqui ruled out metabolic syndrome as the cause of his HCC.
[Id. 54:1-9.]

ii. Cirrhosis Ruled Out
Dr. Siddiqui recognized that “[c]irrhosis is one of the most significant risk
factors for liver cancer,” and that the “increased risk of liver cancer associated with
cirrhosis is particularly pronounced in individuals who have . . . MASLD that
progressed to cirrhosis.” [Siddiqui Report at 21.] According to Dr. Siddiqui, “[f]rom

the time cirrhosis is first diagnosed, it takes approximately 7-10 years before increased
rates of HCC are observed.” [Id. at 21–22 (emphasis in original).]
Dr. Siddiqui testified that liver disease proceeds in a “stepwise progression” that
begins with a healthy liver, which over time – “over many, many years usually” –
develops inflammation, then develops areas of fibrosis and scar tissue, which

eventually results in cirrhosis. [Siddiqui Tr. 15:23-16:8.] “[A] small portion of people
with cirrhosis go on to develop hepatocellular or liver cancer. And a lot of them end
up developing end-stage liver disease.” [Id. 16:8-11.]
In Dr. Siddiqui’s view, “[a]ny increased risk of liver cancer that Mr. Roberts
was at would essentially be captured by the degree of cirrhosis he had prior to being
exposed to valsartan contaminated with NDMA.” [Siddiqui Report at 30.] In her

view, Mr. Roberts had only “incredibly mild, the earliest stages of cirrhosis prior to
being exposed to valsartan contaminated with NDMA.” [Id.] She would not have
expected him to receive an actual diagnosis of cirrhosis in 2016 because it was so
“mild.” [Id.] Based on her review of Mr. Roberts’s April 2016 CT scan, she believed
he would be “between inflammation and possible fibrosis at that time,” rather than

cirrhosis. [Siddiqui Tr. 16:13-18.] She rejected the radiologist’s reading that the
“findings above may be evidence of liver cirrhosis” because Mr. Roberts had no
physical symptoms, which she claimed are necessary to make a clinical diagnosis of
cirrhosis. [Id. 16:20-17:16; 19:2-3.]

Had Mr. Roberts developed HCC because of pre-existing cirrhosis that was
unrelated to any NDMA exposure, Dr. Siddiqui would not have expected it to develop
until August 2025 at the earliest, utilizing what she viewed as the “aggressive type time
frame of seven years from diagnosis in August 2018.” [Siddiqui Tr. 50:22-51:8.] And,
even if he had been diagnosed with cirrhosis in 2016, Dr. Siddiqui would not have

expected to see HCC develop until April 2023. [Id. 51:9-21.] Accordingly, because
she viewed Mr. Roberts’s timeline to be inconsistent with the typical progression from
cirrhosis to HCC, Dr. Siddiqui ruled out cirrhosis as the cause of his HCC.
[Id. 51:22-52:10.]
iii. NDMA Exposure from Valsartan Ruled In
Finally, Dr. Siddiqui determined that NDMA exposure in drugs was a risk
factor for HCC. She reasoned that NDMA is classified as probable human carcinogen

and that animal studies have consistently demonstrated NDMA-induced liver cancer
in rodents and a dose response. [Siddiqui Report at 25.] She also considered the
Hidajat Study, which showed dose response in rubber and dye factory workers who
were exposed to NDMA and developed liver cancer,14 as well as three epidemiological
studies – Pottegård,15 Gomm,16 and Mansouri17 – that attempted to measure the

association between exposure to NDMA in valsartan and risk of liver cancer.
The Pottegård study did not find an association between NDMA exposure in
valsartan and liver cancer. Dr. Siddiqui found this study to have limited, if any, value
due to its small sample size. [Siddiqui Tr. 32:2-33:3.]

14 Mira Hidajat, et al., Lifetime Exposure to Rubber Dust, Fumes, and N-nitrosamines
and Cancer Mortality in a Cohort of British Rubber Workers with 49 Years Follow-Up, 76 J.
Occupational & Env’tl Med. 250 (2019), https://doi.org/10.1136/oemed-2018-
105181 (“Hidajat”).
15 Anton Pottegård, et al., Use of N-nitrosodimethylamine (NDMA) contaminated
valsartan products and risk of cancer: Danish nationwide cohort study, 362 BMJ (2018),
https://pubmed ncbi nlm nih.gov/30209057/.
16 Willy Gomm et al., N-Nitrosodimethylamine-Contaminated Valsartan and the Risk
of Cancer, 118 Deutsches Arzteblatt Int’l 357, 360 (2021) (“Gomm”) [Rose Decl. Ex. 30
(Docket No. 3072-29)].
17 Imene Mansouri, et al., N-nitrosodimethylamine-Contaminated Valsartan and Risk
of Cancer: A Nationwide Study of 1.4 Million Valsartan Users, 11 J. of the Am. Heart Assoc.
(2022) (“Mansouri”) [Rose Decl. Ex. 31 (Docket No. 3072-30)].
Both the Gomm and Mansouri studies were much larger studies that found a
statistically significant increased risk of liver cancer in patients who ingested NDMA-
contaminated valsartan. Gomm reported a hazard ratio of 1.16, meaning a 16%

increased risk of developing liver cancer.18 No dose response was detected, however.
Gomm followed almost 800,000 patients in Germany and divided them into “ever
exposed” and “never exposed” groups. If a patient filled even one valsartan
prescription or ingested even one valsartan pill, that patient was included in the “ever
exposed” group. So, too, would a patient who used valsartan regularly for years.

[Siddiqui Tr. 33:7-20, 33:21-34:3.] The “ever exposed” group also included patients
who took valsartan manufactured by different manufacturers whose pills contained
vastly different amounts of NDMA. [Id. 34:4-9.] Comparing the two groups, the
Gomm study found a 16% higher risk of liver cancer in the “ever exposed” group.
Dr. Siddiqui, however, testified that the study results were “diluted” because the “ever

18 “Relative risk is commonly calculated by dividing the risk of developing a
disease observed in an exposed group by the risk observed in an unexposed, but
otherwise similar group.” Magistrini v. One Hour Martinizing Dry Cleaning, 180 F. Supp.
2d 584, 591 (D.N.J. 2002), aff’d, 68 F. App’x 356 (3d Cir. 2003). Relative risk may be
denoted in different manners, such as percents, hazard ratios, or odds ratios. Id. at 591
n.7. “If the risks of the unexposed and exposed are the same, then the relative risk
estimate (which mathematically is simply the former divided by the latter) is 1.0” or
100%. Id. at 591. “This value is also called the null value, and indicates that exposure
is not associated with the disease in that study. Thus, a relative risk [or hazard ratio]
of 1.0 means that the agent has no effect on the incidence of disease. Similarly, if the
relative risk estimate is 1.3, then risk appears to be 30% higher among the exposed
compared to the non-exposed. When the relative risk [or hazard ratio] reaches 2.0,
the risk has doubled, indicating that the risk is twice as high among the exposed group
as compared to the non-exposed group.” Id.
exposed” group included patients who had ingested far less NDMA than Mr. Roberts
had, given that he took valsartan daily for two years and that the valsartan he
consumed contained much higher amounts of NDMA than the valsartan

manufactured by other companies. Accordingly, she believed the true risk to someone
in Mr. Roberts’s shoes was much higher than observed in Gomm. [Id. 34:20-24.]
Mansouri reported a hazard ratio of 1.12, or a 12% increased risk. No dose
response was detected in Mansouri, either. Dr. Siddiqui had similar critiques of the
Mansouri study as she did with the Gomm study. Mansouri did not account for the

varying amounts of NDMA contamination in valsartan manufactured by different
companies. [Id. 40:15-41:7.] As a result, Dr. Siddiqui believed there was “a big
dilution effect” and that the true increased risk due to the amount of NDMA to which
Mr. Roberts was exposed is much greater than what was reported in Mansouri. [Id.

41:15-19.] She testified that Mr. Roberts would have had one of the highest exposures
compared subjects in that study and, therefore, would have had a higher risk of
developing cancer than the average study participant. [Id. 56:16-25.]
Dr. Siddiqui examined Mr. Roberts’s exposure to NDMA from the
contaminated valsartan pills he began taking in September 2016. At the time of the

July 2018 recalls, the FDA announced that the “reasonably safe” level of NDMA for
human ingestion was 96 nanograms. [July 27, 2018 FDA Press Release, available at
https://www.fda.gov/drugs/drug-safety-and-availability/fda-updates-and-press-
announcements-angiotensin-ii-receptor-blocker-arb-recalls-valsartan-losartan.] Yet
each valsartan pill taken by Mr. Roberts from September 2016 to September 2018
contained between 13,180 and 20,190 nanograms, far more than the FDA limit. He
took roughly 681 NDMA-contaminated valsartan pills prior to his cancer diagnosis.
[Siddiqui Report at 28–29.]

Dr. Siddiqui opined that the “aggressive behavior and rapidity of Mr. Roberts’
cancer progression was atypical and was most consistent with a carcinogenic
exposure.” [Id. at 31; see also Siddiqui Tr. 30:4-8 (“The timeline of two years wasn’t
making sense to me. And that’s why I believe that it was the NDMA that was
promoting that cancer and shortening the timeline and making it very, very fast.”).]

Typically, according to Dr. Siddiqui, it takes “many, many years, sometimes seven,
ten years or more” to progress from early liver disease to cirrhosis and “then cirrhosis
to HCC takes another sometimes seven to ten years or ten years or more.” [Siddiqui
Tr. 28:1-10.] Yet, in her view, he progressed from having “no obvious aggressive signs
of disease in 2016” to advanced cancer in 2018. [Id. 28:11-17.] Based on this “very

shortened latency period or time span,” she had “no doubt” that Mr. Roberts’s cancer
“was caused by his exposure to NDMA contaminated valsartan.” [Id. 28:16-17;
Siddiqui Report at 31.] She does not believe that there were any other substantial
factors in causing his liver cancer.

C. Dr. Siddiqui’s Testimony is Unreliable, Conclusion-Driven, and Speculative
There is no dispute that a differential diagnosis is a generally accepted method
of opining on specific causation. But “the differential diagnosis must be properly
performed in order to be reliable.” Feit v. Great W. Life & Annuity Ins. Co., 271 F. App’x
246, 254 (3d Cir. 2008). Here, argue Defendants, it was clearly not. Instead,
Defendants contend that Dr. Siddiqui’s opinion is “speculative, result-oriented, [and]
unscientific.” [Defs.’ Br. at 1.] The Court agrees. As explained in detail herein, the
Court finds that Dr. Siddiqui did not use, as required, a “scientifically valid

methodology to rule in and rule out the potential causes” of Mr. Roberts’s cancer.
See In re Zoloft (Sertralinehydrochloride) Prods. Liab. Litig., 176 F. Supp. 3d 483, 494–95
(E.D. Pa. 2016), aff’d sub nom. In re Zoloft (Sertraline Hydrochloride) Prods. Liab. Litig.,
858 F.3d 787 (3d Cir. 2017). Nor do her conclusions “reliably flow from the data and
methodology.” Feit, 271 F. App’x at 254. In short, she routinely ignored facts and

data that were inconvenient to her ultimate conclusion, demonstrating the
conclusion-driven nature of her testimony. Plaintiff accuses Defendants of merely
attacking Dr. Siddiqui’s conclusions, not her methodology. This accusation is
unfounded. While Defendants surely take issue with Dr. Siddiqui’s conclusions, their

Rule 702 challenges are limited to flaws in her methodology. While Rule 702
challenges must focus “solely on the [expert’s] principles and methodology, not on the
conclusions that they generate,” Daubert, 509 U.S. at 595, it does not follow that an
expert’s conclusions are entirely irrelevant to the analysis. Indeed, courts “must
examine the expert’s conclusions in order to determine whether they could reliably

flow from the facts known to the expert and the methodology used.” Oddi, 234 F.3d
at 146.
The Court takes this opportunity to reemphasize its critical gatekeeping
function. Plaintiff repeatedly urged this Court to permit the jury to hear Dr. Siddiqui’s
testimony and argued that any flaws with Dr. Siddiqui’s testimony should be
addressed, not by the Court on a Rule 702 motion, but through “vigorous
cross-examination.” [Pl.’s Supp. Br. at 1.] Alternatively, Plaintiff has suggested that
this Court reserve its decision until after trial so that it can rule on a “developed

record.” [9/3/25 Tr. 46:16-21.] This Court shall do no such thing. Given this Court’s
serious concerns with Dr. Siddiqui’s testimony, this would amount to a total
abdication of its gatekeeping responsibility. The entire purpose of Rule 702 is to keep
unreliable expert testimony from ever reaching a jury. And for good reason. How
could the Court ever “unring” the proverbial bell after the jury had listened to hours of

unreliable expert testimony? The prejudice to Defendants would be insurmountable.
The Court also rejects any insinuation on the part of Plaintiff that this Court is
usurping the role of the jury or improperly “nitpicking” Dr. Siddiqui’s opinions.
[See Pl.’s Supp. Br. at 1 (quoting FED. R. EVID. 702, Advisory Committee’s note to

2023 amendments).] This Court is required to regulate what expert testimony reaches
a jury precisely because of how persuasive and authoritative expert testimony can be.
See Daubert, 509 U.S. at 595; Ford, 481 F.3d at 220 n. 6. Of course, it is the jury’s job
to weigh the expert testimony presented to it. But first, this Court must ensure that the
expert testimony is based on sufficient facts and data and is the product of reliable

principles and methods.
i. Dr. Siddiqui Did Not Reliably Rule Out Cirrhosis
Cirrhosis is the primary and most potent risk factor for HCC. Eighty to ninety
percent of autopsied individuals with HCC were found to have cirrhosis.19

Peer-reviewed scientific studies have established that individuals with cirrhosis and
other underlying conditions, like MASLD, are up to 3,000% to 4,500% more likely to
develop HCC than those without cirrhosis.20 Dr. Siddiqui herself recognized that
cirrhosis is “one of the most significant risk factors” for liver cancer. [Siddiqui Report
at 21.] Mr. Roberts had – at the very least – the early stages of cirrhosis in April 2016.

[See Siddiqui Report at 30.] By the time of his cancer diagnosis, his cirrhosis was
aggressive and obvious. [Siddiqui Dep. 136:20-21, 267:10-12.] And his physicians
characterized his cancer as HCC arising out of or in the setting of “NASH cirrhosis.”
[Rose Decl. Ex. 5 at GRobertsJr-UABHIM-MD-000083.]
Notwithstanding all of this – the immensely increased risk of HCC associated

with cirrhosis and Mr. Roberts’s history of cirrhosis – Dr. Siddiqui determined that
cirrhosis was not a substantial factor in causing Mr. Roberts’s cancer. She reached this
conclusion for two primary, but critically flawed, reasons. First, she claimed that

19 Giovanna Fattovich et al., Hepatocellular carcinoma in cirrhosis: incidence and risk
factors, 127 Gastroenterology 35, 36 (2004).
20 Kazuo Tarao et al., Real impact of liver cirrhosis on the development of hepatocellular
carcinoma in various liver diseases—meta‐analytic assessment, 8 Cancer Med. 1054 (2019)
(“Tarao 2019”); Rongtao Lai et al., Incidence rates of hepatocellular carcinoma based on risk
stratification in steatotic liver disease for precision medicine: A real-world longitudinal
nationwide study, PLoS Med. (2024) (“Lai 2024”).
Mr. Roberts had only “mild” cirrhosis in April 2016, which resulted in only a
“very minor increased risk” of developing cancer. Second, she claimed that the
“timeline” during which Mr. Roberts developed cancer was inconsistent with the

typical progression of cirrhosis to HCC. The Court addresses each portion of her
opinion in turn.
a. “Mild” Cirrhosis
As a preliminary matter, regardless of whether Mr. Roberts was formally
diagnosed or whether Dr. Siddiqui would or would not have given him a diagnosis –

the prong on which she primarily hangs her hat – the overwhelming evidence in this
case points to the conclusion that Mr. Roberts suffered from some degree of cirrhosis
prior to ingesting any contaminated valsartan and prior to his cancer diagnosis. And
even Dr. Siddiqui agrees with this conclusion. Dr. Siddiqui’s expert report was clear:
“it is my opinion that Mr. Roberts had incredibly mild, the earliest stages of cirrhosis

prior to being exposed to valsartan contaminated with NDMA.” [Siddiqui Report
at 30.] That Dr. Siddiqui “would not have given him an actual diagnosis at this time”
is beside the point. [Id.]21 She reiterated that he had “very mild and undiagnosed
cirrhosis” prior to ingesting NDMA. [Id.] And, Dr. Mele, Plaintiff’s expert
radiologist, opined that the April 2016 CT scan was “consistent with cirrhosis.” [Mele

Report at 3; see also Mele Dep. 157:14-21.] Dr. Siddiqui agreed with his findings.
[Siddiqui Dep. 97:22-23.] The April 2016 scan itself showed findings that “may be

21 Dr. Siddiqui is a hematologist and oncologist. She is not a hepatologist.
evidence of liver cirrhosis.” [Rose Decl. Ex. 51 at GRobertsJR-TH-MD-000944.] The
liver appeared nodular and lobulated and, as even Plaintiff’s expert radiologist opined,
the CT scan showed recanalization of the umbilical vein, which is indicative of

cirrhosis. [Mele Report at 2; see also Mahmud Tr. 150:13-151:23.] Mr. Roberts’s Fib-4
scores, as calculated by Dr. Mahmud, likewise indicated that he may have had
cirrhosis for some time.22 [Mahmud Report at 21.]
At the Daubert hearing, Dr. Siddiqui attempted to downplay the implications of
the April 2016 CT scan, testifying that the radiologist reported that the “findings were

consistent with possible inflammation or nodular formation or fibrosis or cirrhosis.”
[Siddiqui Tr. 17:1-2.] But that is not what the radiologist reported. Nor did the
radiologist state, as Dr. Siddiqui claims, that the CT scan showed only “subtle
findings.” [Id. 17:8-10.] The report stated: “findings above may be evidence of liver

cirrhosis.” [Rose Decl. Ex. 51 at GRobertsJR-TH-MD-000944.] She then testified
that since Mr. Roberts had no symptoms, he could not have diagnosable cirrhosis,
which entirely ignores the possibility of compensated or asymptomatic cirrhosis.
[Siddiqui Tr. 18:3-7.]

22 Dr. Siddiqui did not consider the Fib-4 scores at all. Nor did she provide any
reasoning for ignoring the high Fib-4 score calculated by Dr. Mahmud. Plaintiff
attempts to excuse this by arguing that Fib-4 scores are not diagnostic. Defendants do
not dispute this. It is, however, an important “risk stratification tool” that
Dr. Mahmud testified is routinely used “as a best practice tool . . . to identify patients
who are highly likely to have cirrhosis.” [Mahmud Tr. 147:10-23.]
When pressed by the Court on her scientific basis for determining that
Mr. Roberts did not have cirrhosis in 2016, Dr. Siddiqui testified that “the liver looks
quite hydrogenous, as in it looks clean.” [Id. 20:13-19.] It is unclear to the Court what

exactly she meant by “hydrogenous.”23 But what is readily apparent is that the CT
scan did not show a “clean” liver. It showed a “lobulated” liver. [Rose Decl. Ex. 51
at GRobertsJR-TH-MD-000944.] As Dr. Mahmud explained, “[a] normal liver is
smooth, like smooth like a table. A liver that has cirrhosis, the scarring will cause
retractions along the surface and it looks bumpy or lobulated or nodular.” [Mahmud

Tr. 150:15-23.] To reiterate, Dr. Siddiqui hangs her opinion hat on the lack of a formal
cirrhosis diagnosis in 2016, while ignoring all the medical records suggesting it was
present. In any event, given the evidence before this Court of ample findings consistent
with cirrhosis, the lack of a formal diagnosis of cirrhosis until the time of his cancer
diagnosis is inconsequential. Her opinion is detached from the facts of the case and,

therefore, is both unreliable and unhelpful to a jury. There is simply “too great a gap
between the data” and the opinions she has preferred. Oddi, 234 F.3d at 146
(quoting Joiner, 522 U.S. at 146).
Dr. Siddiqui then opined that “Mr. Roberts would have had a very minor

increased risk of liver cancer because of the degree of cirrhosis he had in 2016.”

23 The Court suspects she meant “homogeneous.” As Dr. Mele explained, a
non-cirrhotic liver typically “has a very homogeneous attenuation.” [Mele Dep.
171:22-23.] By contrast, the CT scan of Mr. Roberts’s liver showed “heterogeneous
attenuation,” which “goes with individuals who have cirrhosis.” [Id. 172:3-6.]
[Siddiqui Report at 30.] But she made no attempt to quantify the increased risk.
See Magistrini, 180 F. Supp. 2d at 610 (excluding specific causation expert’s testimony
wher expert failed to undertake “a quantitative risk assessment” or assess the

“respective risks” of possible causes “using the same scientific method.”). Nor did she
provide any scientific or medical basis to support her opinion that “mild” cirrhosis
results in only a “very minor increased risk” of HCC. Instead, based on her ipse dixit
alone, she determined that she could ignore the “very minor” increased risk and rule
cirrhosis out as a cause of Mr. Roberts’s cancer. This is not a proper basis for expert

testimony. In re Zostavax (Zoster Vaccine Live) Prods. Liab. Litig., 579 F. Supp. 3d 675,
680 (E.D. Pa. 2021) (excluding expert testimony that “cite[d] no literature” or
“supporting scientific or medical studies” as “nothing more than an ipse dixit
pronouncement.”). “Nothing in either Daubert or the Federal Rules of Evidence

requires a district court to admit opinion evidence that is connected to existing data
only by the ipse dixit of the expert.” Joiner, 522 U.S. at 146; accord Player v. Motiva
Enters., LLC, 240 F. App’x 513, 520 (3d Cir. 2007) (“The District Court certainly had
the discretion to exclude opinion evidence that is connected to existing data only by
the ipse dixit [or say-so] of the expert.”) (cleaned up).

What’s more, “mild” cirrhosis is not a medical term and has no accepted
meaning in the medical or scientific literature. [Mahmud Report at 35; Mahmud
Tr. 126:1-19; see Siddiqui Tr. 17:17-25 (describing “mild” cirrhosis as a vague
“colloquial” term).] Dr. Siddiqui explained that she used “mild” in a colloquial sense
to convey early stage, asymptomatic or “compensated” (as opposed to symptomatic
or “decompensated”) cirrhosis. [Siddiqui Tr. 17:17-18:7; Siddiqui Dep. 129:5-13.]
But she provided no scientific support for her position that compensated cirrhosis does
not meaningfully increase the risk of developing liver cancer. Nor does she engage

with the extensive scientific literature cited by Defendants and Dr. Mahmud that
directly undermine her opinion, a point that cannot be emphasized enough.
For example, the relevant medical guidelines do not distinguish between
“compensated” and “decompensated” cirrhosis and recommend that all patients with
cirrhosis of any type immediately begin screening for HCC every six months.24 The

scientific literature likewise establishes that cirrhosis is associated with an extremely
high risk of HCC, regardless of whether it is compensated or decompensated.25
Defendants also cite to a 2022 study in Sweden that found significantly increased risks
of HCC for both compensated and decompensated cirrhosis patients.26 Roughly half
of the 15,000 study participants had compensated cirrhosis. In the patients with

24 Amit G. Singal et al., AASLD Practice Guidance on prevention, diagnosis, and
treatment of hepatocellular carcinoma, 78 Hepatology 1922 (2023) [Rose Decl. Ex. 42
(Docket No. 3072-41); EASL [European Association for the Study of the Liver] Clinical
Practice Guidelines on the management of hepatocellular carcinoma, 82 J. of Hepatology 315
(2025) [Rose Decl. Ex. 43 (Docket No. 3072-42)]; Yuri Cho et al., Overview of Asian
clinical practice guidelines for the management of hepatocellular carcinoma: An Asian
perspective comparison, 29 Clinical & Molecular Hepatology 252 (2023) [Rose Decl.
Ex. 44 (Docket No. 3072-43)].
25 See generally Tarao 2019; Lai 2024.
26 Bonnie Bengtsson et al., The risk of hepatocellular carcinoma in cirrhosis differs by
etiology, age and sex: A Swedish nationwide population-based cohort study, 10 United
European Gastroenterology J. 465 (2022) [Rose Decl. Ex. 45 (Docket No. 3072-44)].
compensated (or, as Dr. Siddiqui would put it, “mild”) cirrhosis, the study reported a
hazard ratio of 167, or an over 16,000% increased risk of HCC. Yet Dr. Siddiqui did
not address any of these studies.

Dr. Siddiqui’s failure to provide scientific support for her opinions and her
refusal to grapple with these contradictory studies renders her testimony unreliable and
“nothing more than an ipse dixit pronouncement.” Zostavax, 579 F. Supp. 3d at 680.
An expert cannot “ignore completely [a] vast body of contradictory evidence”
undermining her position, or “refuse to engage with contrary epidemiological studies

while offering no supporting studies in return.” Rimbert v. Eli Lilly & Co., 2009 WL
2208570, at *14 (D.N.M. July 21, 2009), aff’d, 647 F.3d 1247 (10th Cir. 2011); see also
McEwen v. Baltimore Washington Med. Ctr. Inc., 404 F. App’x 789, 791 (4th Cir. 2010)
(affirming exclusion of expert testimony where expert “failed to meaningfully account

for medical literature at odds with their testimony”).
Finally, in Dr. Siddiqui’s own words, Mr. Roberts had cirrhosis that was
“obvious,” “overwhelming,” and “aggressive” at the time of his cancer diagnosis.
[Siddiqui Dep. 92:9-15, 136:20-21, 244:17-19, 247:10-12, 267:10-12.] It is not
uncommon for cirrhosis and liver cancer to be diagnosed at the same time. [Mahmud

Tr. 128:2-11; Vaz 2022.27] But she provides no explanation for why this aggressive

27 “Vaz 2022” refers to Juan Vaz et al., Unrecognized liver cirrhosis is common
and associated with worse survival in hepatocellular carcinoma: A nationwide cohort
study of 3473 patients, 293 J. of Internal Med. 184 (2022) [Rose Decl. Ex. 46 (Docket
No. 3072-45)]. This study showed that 39% of HCC patients had unrecognized
decompensated cirrhosis was not the cause of Mr. Roberts’s cancer. At the Daubert
hearing, Dr. Siddiqui suggested that the NDMA exposure from valsartan likely caused
the cirrhosis that was discovered in August 2018. [Siddiqui Tr. 96:2-97:15.] But she

provides no scientific literature, record evidence, or analysis to support this opinion.
It is nothing more than speculative ipse dixit.28
In sum, Dr. Siddiqui ignored the immensely high increased risk of HCC
associated with cirrhosis based upon unreliable methods that ignored the facts of the
case and contradictory scientific literature. Her opinions – supported by nothing more

than her speculative ipse dixit – must be excluded.
b. Timeline
The true crux of Dr. Siddiqui’s opinion ruling out cirrhosis, however, is the
timeline during which Mr. Roberts developed HCC, which Dr. Siddiqui characterizes
as atypical. [See, e.g., Siddiqui Tr. 28:23-29:2 (“The only reason I ruled in NDMA and

[ruled out] cirrhosis not caused by NDMA is because . . . the studies show that
cirrhosis to HCC usually takes around seven to ten years.”) (emphasis added).] She is
persistent throughout her testimony that the cirrhosis may develop into HCC only

cirrhosis at the time of their cancer diagnosis. “These patients were more often male,
older, and had larger tumors” and “had worse median survival.” Id.
28 Plaintiff attempts to bolster Dr. Siddiqui’s testimony with a citation to a report
from the U.S. Department of Health and Human Services Agency for Toxic
Substances and Disease Registry. [Pl.’s Opp’n at 13–14.] But Dr. Siddiqui did not
rely on this report in forming her opinion and, as Defendant correctly notes, “Plaintiff
cannot unilaterally supplement [Dr. Siddiqui’s] opinion in Daubert briefing.” [Defs.’
Reply at 10 n.8 (citing In re Rezulin Prods. Liab. Litig., 369 F. Supp. 2d 398, 407
(S.D.N.Y. 2005).]
through a “slow stepwise progression” that takes seven to ten years. [See, e.g., id.
47:3-10, 50:3-11, 53:2-9.] Therefore, according to Dr. Siddiqui, the absolute earliest
Mr. Roberts’s cirrhosis could have developed into HCC is April 2023 (assuming he

had early-stage cirrhosis in April 2016). [Id. 51:9-21.]
Strictly on its face, this theory is appealing. How else can cirrhosis – with its
massive hazard ratio – be definitively ruled out? Upon this Court’s required scrutiny,
however, the opinion is just a house of cards. The timeline Dr. Siddiqui espouses is
backed by neither the scientific literature nor the record evidence, however. It is based

solely upon Dr. Siddiqui’s unsupported ipse dixit. Without this supposedly atypical
timeline, her opinion collapses. Dr. Siddiqui herself recognized this, conceding that
her entire opinion would need to be reworked if the timeline were different. [Id.
110:20-111:7.]

With respect to the timeline that Dr. Siddiqui embraces, there is no evidence
that Mr. Roberts’s cirrhosis began in April 2016. Because Mr. Roberts did not undergo
any imaging between 2009 and 2016, it is not possible to ascertain when it first arose.
Indeed, Plaintiff’s expert radiologist, Dr. Mele, recognized this. [Mele Dep. 172:22-
173:11.] And it defies common sense to assume that the indicia of cirrhosis evident in

the April 2016 CT scan arose for the first time on that day. In other words, there is no
scientific or factual basis for Dr. Siddiqui’s assumption (and that is what it is) that
Mr. Roberts did not have cirrhosis prior to the April 2016 CT scan. See Hoefling v. U.S.
Smokeless Tobacco Co., LLC, 576 F. Supp. 3d 262, 281 (E.D. Pa. 2021) (it is “improper
to treat [a well-established risk factor] as implausible simply because no conclusive test
was performed” demonstrating that plaintiff had it). What’s more, as the Court has
already explained, the medical records in this case strongly suggest that Mr. Roberts
may have had cirrhosis for many years prior.29 Yet Dr. Siddiqui did not even consider

this evidence in formulating her opinion.
The more fundamental issue with Dr. Siddiqui’s timeline opinion, however, is
that the timeline itself lacks a reliable scientific basis. Dr. Siddiqui’s seven-to-ten-year
timeline from cirrhosis to HCC derives exclusively from one inapposite study.30
[Siddiqui Tr. 70:18-22.] The Johnson 2024 study involved hepatitis C patients who

had received effective antiviral therapy. The study’s goal was to “assess the impact of
antiviral treatment on cirrhosis progression and HCC development.” [Johnson 2024
at 2; see also Mahmud Tr. 161:17-23.] Needless to say, Mr. Roberts did not have
hepatitis C nor was he cured of his risk factors for HCC. Dr. Siddiqui offers no
explanation for why the timeline set forth in a study about patients cured of hepatitis C

would be applicable to Mr. Roberts, a patient with several other known risk factors for
liver cancer, but not hepatitis C. At most, she speculates that this timeline is
“conservative” and that it likely would have taken even longer for MASH cirrhosis to
develop into HCC. [Siddiqui Tr. 50:6-21.] This is based on her “assumption” that

29 For example, per Dr. Mahmud’s calculations, Mr. Roberts’s Fib-4 score was
1.99 in September 2009, representing indeterminate fibrosis, and 3.22 in November
2015, representing advanced fibrosis and possible cirrhosis. [Mahmud Report at 21.]
30 Philip J. Johnson et. al, Progression of chronic liver disease to Hepatocellular (Liver)
carcinoma: implications for surveillance and management, BJC Reps. (2024) (“Johnson
2024”) [Rose Decl. Ex. 50 (Docket No. 3071-49)].
Mr. Roberts’s cirrhosis was less aggressive than the cirrhosis experienced by the cured
hepatitis C patients in the Johnson 2024 study. [Id. 72:18-73:16.] She admits this is
simply an assumption. The Court need not permit her testimony that is based on mere

“guesswork.” Zostavax, 579 F. Supp. 3d at 683 (“Reliable expert medical testimony,
not lay assumptions or guesswork, is required under Daubert.”).
What’s more, she overstates the Johnson 2024 study’s findings. Contrary to her
characterization, the study does not say that liver cancer can only develop seven to ten

years after cirrhosis arises. Instead, it found that the median time from cirrhosis to liver
cancer diagnosis in the specific patient population involved was seven years. [Johnson
2024 at 3.] That this is the median time, rather than the minimum time as suggested
by Dr. Siddiqui, means that there will necessarily be patients who develop liver cancer
sooner. Critically, she does not account for this possibility in categorically ruling out

cirrhosis as the cause of Mr. Roberts’s cancer.
Dr. Siddiqui cited no other scientific basis for the timeline.31 Nor did she
account for data and literature that contradicted her established position. She did not

31 At one point, Dr. Siddiqui testified that other studies confirm this timeline, but
at no point did she provide any other citations. [Siddiqui Tr. 111:8-17.] Incredulously,
Plaintiff’s counsel suggested at oral argument that Dr. Siddiqui did not even need a
study to back up her timeline opinion: “she could have no studies and talk about this
timeline from her own experience.” [9/3/25 Tr. 25:2-9.] Be that as it may, that was
not the basis for her opinion. She did not testify that in her experience, the progression
from cirrhosis to HCC in Mr. Roberts’s case was atypical. Rather, she expressly and
exclusively relied upon Johnson 2024 to develop her differential diagnosis. The Court
does not consider Dr. Siddiqui’s experience – proffered not by Dr. Siddiqui herself, but
rather after the fact by counsel – to be sufficient to rectify the flaws in her methodology.
account for the medical guidelines that recommend screening cirrhosis patients every
six months because the risk of developing liver cancer is so high. [Mahmud Tr. 124:20-
125:25, 159:11-19.] Nor did she account for the CDC guidance that indicates a four-

year latency period for solid cancers like HCC to develop.32 She also failed to consider
whether Mr. Roberts’s cancer progression was consistent with the medical literature
regarding Stage III cancer and survival rates. Mr. Roberts was diagnosed with
Stage III cancer in August 2018 and passed away in March 2020, approximately
eighteen months later. Defendants point out that this is consistent with the medical

literature establishing that the median survival time for patients with “large tumor
sizes” of “5 cm” or greater, like Mr. Roberts, is just 18.80 months.33 Unfortunately,
the five-year survival rate for patients with advanced HCC, like Mr. Roberts, is “less
than 20%.”34 Dr. Siddiqui’s refusal to engage with contradictory data while focusing
exclusively on data from the inapposite Johnson 2024 study is at odds with a

See Rezulin, 369 F. Supp. 2d at 407 (“The subject of [a Daubert] motion is the proposed
testimony of experts, not the theories of the lawyers.”).
32 WTC Health Guidance Document at 5,
https://www.cdc.gov/wtc/pdfs/policies/WTCHP-Minimum-Cancer-Latency-PP-
01062015-508.pdf.
33 Sangheun Lee et al., Subclassification of Barcelona Clinic Liver Cancer B and C
hepatocellular carcinoma: a cohort study of the multicenter registry database,
31 J. of Gastroenterology & Hepatology 842, 843 (2016) [Rose Decl. Ex. 34 (Docket
No. 3072-33)].
34 Evelyn Calderon-Martinez et al., Prognostic Scores and Survival Rates by Etiology
of Hepatocellular Carcinoma: A Review, 15 J. Clin. Med. Res. 200, 203 (2023) [Rose Decl.
Ex. 33 (Docket No. 3072-32)].
systematic differential diagnosis and follows no reliable methodology. See In re Zoloft
(Sertraline Hydrochloride) Prods. Liab. Litig., 858 F.3d 787, 797 (3d Cir. 2017) (“if an
expert applies certain techniques to a subset of the body of evidence and other

techniques to another subset without explanation, this raises an inference of unreliable
application of methodology.”).
The cornerstone of Dr. Siddiqui’s opinion – that HCC caused by cirrhosis arises
seven to ten years after cirrhosis diagnosis – lacks any reliable scientific foundation. It
is premised only upon her ipse dixit and misapplication of the scientific literature. Her

opinion that Mr. Roberts’s cancer could not have been caused by cirrhosis has no
reliable scientific basis and must be excluded.
ii. Dr. Siddiqui Did Not Reliably Rule Out Mr. Roberts’s Other Known
Risk Factors
Dr. Siddiqui’s treatment of Mr. Roberts’s other known risk factors fares no
better. Based on little else than her own say-so, she opined that his risk factors could
all be considered together and that their impact was limited to increasing Mr. Roberts’s

risk of developing cirrhosis. And since she had already ruled out cirrhosis as a cause,
the other risk factors fell by the wayside as well. But this approach is contrary to the
established scientific literature and not the product of a reliable methodology.
Dr. Siddiqui did not consider each of Mr. Roberts known risk factors
independently and instead lumped his numerous conditions – specifically,

MASLD/MASH, obesity, diabetes, and hyperlipidemia – under the umbrella of
metabolic syndrome. Dr. Siddiqui claimed that “the way we do it in basically
medicine now, is we put it all under [MASLD] . . . They put in hyperlipidemia,
diabetes, and obesity into a group of diseases when talking about liver cancer or liver
disease.” [Siddiqui Dep. 197:1-25.] But this approach is undermined by ample

scientific evidence that Dr. Siddiqui either ignores or misconstrues. Even the one
study Dr. Siddiqui cites as support for her approach does the exact opposite.35
As Defendants point out, Mantovani 2017 “emphasizes the ‘independent’ nature of
risk factors for HCC, including obesity, and notes that these risks are ‘additive’ in
individuals, like Mr. Roberts, with multiple risk factors for HCC.” [Defs.’ Reply at 15

n.10.]36 Accordingly, the Court finds that Dr. Siddiqui’s opinion employs an
unreliable method that lacks acceptance and support in the scientific literature.
See Magistrini, 180 F. Supp. 2d at 603 (excluding expert testimony as scientifically
unreliable where expert treated various cancers “together for etiological purposes”
without “sufficient scientific support for this theory”).

And, as explained below, the great weight of the scientific literature establishes
that each of Mr. Roberts’s conditions was an independent risk factor for HCC. Indeed,
as Dr. Siddiqui herself admitted, each independently carries a higher increased risk

35 Alessandro Mantovani & Giovanni Targher, Type 2 diabetes mellitus and risk of
hepatocellular carcinoma: spotlight on nonalcoholic fatty liver disease. 5 Ann. Transl. Med.
(2017) (“Mantovani 2017”) [Rose Decl. Ex. 53 (Docket No. 3072-52)].
36 This literature is consistent with the testimony of Defendants’ expert
hepatologist, Dr. Mahmud, who testified that each risk factor is studied independently
in the literature and that he did not understand Dr. Siddiqui’s approach because there
is “no justification for lumping them together . . . in the literature.” [Mahmud
Tr. 139:11-23.]
than NDMA exposure does. [See Siddiqui Tr. 103:16-104:4.]37 But she did not engage
with this literature, much less refute it. She made “no attempt to quantify the relative
contribution of [Mr. Roberts’s] risk factors,” rendering her opinion unreliable. Guinn

v. AstraZeneca Pharms. LP, 602 F.3d 1245, 1256 (11th Cir. 2010) (affirming exclusion
of expert). As part of its critical gatekeeping function, this Court must ensure that
Dr. Siddiqui “employs in the courtroom the same level of intellectual rigor that
characterizes the practice of an expert in the relevant field.” See Kumho, 526 U.S.

at 152. Based on the record before it, this Court simply cannot make that finding.
Dr. Siddiqui used an approach that is not supported by the medical and scientific
literature, rendering her testimony unreliable.
a. MASH/MASLD
Dr. Siddiqui appeared to equate MASLD/MASH and the umbrella metabolic
syndrome she described. But she distorted and misrepresented the record in her

attempt to rule out MASH or metabolic syndrome as the cause of Mr. Roberts’s HCC.
In her report, Dr. Siddiqui unmistakably opined that “[w]hile Mr. Roberts had NASH
with no evidence of decompensation, his resulting increased risk of liver would be
captured based on the degree of his liver cirrhosis.”38 [Siddiqui Report at 30 (emphasis

added).] She confirmed this belief at her deposition: “Yes, I do believe that

37 When pressed, Dr. Siddiqui testified that, although she cannot be sure, each of
these independent risk factors “may have” contributed to the cause of Mr. Roberts’s
cancer. [Siddiqui Tr. 109:20-23.]
38 Recall that NASH is now referred to as MASH. See supra n.7.
[Mr. Roberts] had NASH before his [cancer] diagnosis based on my review of his
medical records.” [Siddiqui Dep. 200:1-6.]
Yet in the Daubert briefing and at the Daubert hearing, Plaintiff and Dr. Siddiqui

changed tack. The line became that Mr. Roberts was never diagnosed with MASH
and so MASH could not have been the cause of his cancer. Indeed, Dr. Siddiqui
repeatedly and insistently testified as much. [See, e.g., Siddiqui Tr. 63:10-14, 84:11-13;
98:15-22.] She claimed, contradicting her report, that, at most, “there was a
questionable diagnosis of NASH,” but no “proper full diagnosis of NASH was [ever]

made.” [Id. 98:15-22.] And, according to Dr. Siddiqui, “[t]here were some notes that
said consistent with NASH. But there was no official diagnosis of NASH.”
[Id. 63:10-14.] Aside from unjustifiably elevating form over substance, this opinion is
contradicted by the record.

Mr. Roberts’s 2009 medical records state unequivocally that his treating
physician’s “impression” was “nonalcoholic steatohepatitis with no evidence of
decompensation.” [Summ. J. Rose Decl. Ex. 27 at GRobertsJr-CA-000654.] There is
nothing “questionable” about this noted diagnosis. “NASH” then continues to appear
throughout his medical records as part of his past medical history. [See, e.g., Rose Decl.

Ex. 17 at GRobertsJr-ESMS-000002, GRobertsJr-ESMS-000045, GRobertsJr-ESMS-
000051, GRobertsJR-CA-000210.] And Mr. Roberts’s cancer was frequently
described by his treating physicians as HCC in the setting of “NASH cirrhosis.”
[See, e.g., Rose Decl. Ex. 5 at GRobertsJr-UABHIM-MD-000083;39 Rose Decl. Ex. 29
at GRobertsJr-PPR-000096.] Her attempt to unilaterally discount this clear diagnosis
is not supported by the record and is based on pure conclusion-driven speculation.40

Dr. Siddiqui also improperly relied upon Dr. White’s note to support her
opinion that Mr. Roberts was never diagnosed with MASH. She claims that the note,
written in August 2018, states that his doctors “had done a full review in the past and
had not found liver disease” and that “he had a completely normal liver workup.”
[Siddiqui Tr. 84:7-14, 102:8.] Not so. She has cherrypicked one phrase from the note

and ignored the rest. While the note states that Mr. Roberts’s “[w]ork up for cirrhosis
is essentially negative,” it goes on to explain that the “[l]ocal workup is negative for

39 Mr. Roberts’s radiation oncologist described Mr. Roberts’s cancer as
“HCC arising out of NASH cirrhosis.” [Rose Decl. Ex. 5 at GRobertsJr-UABHIM-
MD-000085.] Dr. Siddiqui urges the Court to ignore this assessment because the
physician was a resident, rather than an attending, and because he “he has half an hour
to quickly make the history . . . he doesn’t know what [Mr. Roberts] was exposed to
. . . He is seeing a questionable history of NASH and he’s seeing cirrhosis.” [Siddiqui
Tr. 101:1-21.] But this is entirely speculative. Dr. Siddiqui admits as much. She never
spoke with the resident physician, and she had “no idea” what he was thinking when
he wrote his note. [Id. 102:3-6.]
40 In one instance, Dr. Siddiqui went so far as to testify that Mr. Roberts’s “fatty
liver was never diagnosed officially.” [Siddiqui Tr. 110:1-3.] But “fatty liver” appears
throughout Mr. Roberts’s medical records. [See generally Rose Decl. Ex. 17.] In fact,
Mr. Roberts had been consistently told by his doctors since he was a teenager that he
had fatty liver. [Rose Decl. Ex. 6 at GRobertsJr-CA-000661.] This is yet another
example of the lengths to which Dr. Siddiqui went to distort the record to serve
Plaintiff’s desired outcome. Such a results-oriented approach is unreliable and cannot
be presented to a jury.
viral source.”41 [Rose Decl. Ex. 29 at GRobertsJr-PPR-000093, GRobertsJr-PPR-
000096 (emphasis added).] It further states that Mr. Roberts had no “significant

[history of] liver disease other than risk factors for NASH cirrhosis.” [Id. at GRobertsJr-
PPR-000097 (emphasis added).] Dr. Siddiqui’s decision to rule out MASH as a cause
of Mr. Roberts’s cancer is additionally unreliable and disconnected from the record in
this case insofar as it cherry-picks and mischaracterizes portions of Dr. White’s note
to support or confirm her position. See Allen v. Am. Cap. Ltd., 287 F. Supp. 3d 763, 786

(D. Ariz. 2017) (excluding an expert’s testimony where expert “cherry-picked the data
for his analysis to produce a litigation-driven result.”).
Finally, Dr. Siddiqui simply assumed that – contrary to the scientific literature
– “[a]ny increased risk of liver cancer that Mr. Roberts was at [due to his metabolic
syndrome] would essentially be captured by the degree of cirrhosis he had prior to

being exposed to valsartan contaminated with NDMA.” [Siddiqui Report at 30.]
According to this assumption, she could then rule out MASLD/MASH as causes by
ruling out cirrhosis. For the reasons previously articulated, see supra Section IV.C.i.,

41 This is consistent with Dr. Mahmud’s interpretation of this record. He
explained that a “cirrhosis work up” involves a blood test “to identify potential causes
of cirrhosis, things like hepatitis B virus, hepatitis C virus, antibodies to look for
different types of autoimmune liver disease, certain types of testing to look for genetic
causes of liver disease.” [Mahmud Tr. 190:1-6.] “When all of that returns negative,
which means to say that we don’t identify any evidence of hepatitis B, hepatitis C,
autoimmune liver disease, et cetera, we refer to that as a negative cirrhosis workup.
There is no blood test for MASH or MASLD, so that’s never really kind of referred to
as being a positive element of this cirrhosis workup.” [Id. 190:7-12.]
the Court finds her exclusion of cirrhosis as a cause of Mr. Roberts’s cancer to be
unscientific, unreliable, and detached from the facts of the case.
Most problematic, however, is the fact that the scientific literature undermines

the validity of Dr. Siddiqui’s assumption. It is true that MASLD and MASH can lead
to HCC in part because they can cause cirrhosis. But the scientific literature also
establishes that MASLD and MASH are independent risk factors for HCC.
As Dr. Siddiqui herself recognizes, MASLD “has been increasingly linked to liver
cancer, especially in individuals with progression to [M]ASH” and “[a]pproximately

13% of patients diagnosed with HCC without a background of cirrhosis were noted to
have” MASLD. [Siddiqui Report at 21.]42 This is confirmed by studies establishing
that “[i]n non-cirrhotic subjects, those with [MASH] have a higher risk of HCC
compared to other []etiologies of liver disease.”43 Other published studies establish
hazard or odds ratios between 1.82 and 7.62.44 And, even putting MASH aside,

42 Perplexingly, despite her own expert agreeing that MASLD/MASH can be an
independent risk factor for HCC, Plaintiff apparently takes the position that it is not a
risk factor at all for HCC. [See Pl.’s Opp’n at 20 n.84.]
43 Jonathan G. Stine et al., Systematic review with meta-analysis: risk of hepatocellular
carcinoma in non-alcoholic steatohepatitis without cirrhosis compared to other liver diseases,
48 Ailment Pharmacol Ther. 696 (2018) (“Stine 2018”) [Rose Decl. Ex. 10 (Docket
No. 3072-9)].
44 See, e.g., Fasiha Kanwal, et al., Risk of Hepatocellular Cancer in Patients With
Non-Alcoholic Fatty Liver Disease, 155 Gastroenterology 6 (2018), available at
https://pubmed.ncbi.nlm.nih.gov/30144434/; Qingyan Kong et al., Impact of
Metabolic Dysfunction-Associated Fatty/Steatotic Liver Disease on Hepatocellular Carcinoma
Incidence and Long-Term Prognosis Post-Liver Resection: A Systematic Review and
studies establish that MASLD alone is associated with a hazard ratio of 1.80 – or 80%
increased risk of HCC – even in patients without cirrhosis.45 Notably, all of the
scientific literature establishes that MASLD and MASH carry an increased risk for

developing HCC far greater than that associated with NDMA exposure.
Dr. Siddiqui herself recognized this. [Siddiqui Tr. 102:12-24.] And yet, she
failed to engage with this scientific literature. Instead, she summarily excluded the
possibility of MASLD/MASH directly causing Mr. Roberts’s cancer for two
scientifically unreliable reasons. First, according to Dr. Siddiqui, the “metabolic

syndrome to HCC” pathway is “rare.” [Id. 53:12-15.] But Dr. Siddiqui made no
attempt to quantify how rare it is compared to what she calls the more common
pathway of metabolic syndrome to cirrhosis and then finally to HCC. Nor did she
explain why it is scientifically and medically appropriate to rule out a cause simply
because it is uncommon. As articulated, Dr. Siddiqui’s opinion amounts to nothing

more than speculative ipse dixit.
Second, Dr. Siddiqui rejected this pathway because Mr. Roberts “had cirrhosis
at some point,” so “the direct NASH to HCC” pathway “wasn’t making sense in [his]
case.” [Id. 53:16-25.] But the fact that Mr. Roberts had cirrhosis “at some point”

cannot eliminate the possibility that his MASLD/MASH contributed to both the

Meta-Analysis, Academic Radiology (2025) [Rose Decl. Ex. 59 (Docket No. 3072-58)];
Stine (2018).
45 Fahim Ebrahimi, et al., Familial coaggregation of MASLD with hepatocellular
carcinoma and adverse liver outcomes: Nationwide multigenerational cohort study, 79 J. of
Hepatology 1374 (2023) [Rose Decl. Ex. 11 (Docket No. 3072-10).]
development of cirrhosis and HCC. Dr. Siddiqui has presented no basis – other than
her own say-so – to assume that these pathways are mutually exclusive. The Court
need not take her word for it. See FED. R. EVID. 702, advisory committee’s note to

2000 amendments. Indeed, as Dr. Mahmud testified, the risk is “synergistic. . . . So if
you have multiple of these risk factors, that each independently increase the risk, there
is a kind of an additive effect where if you have all of these [risk] factors, your risk is
even higher.” [Mahmud Tr. 141:5-16.] Accordingly, the Court finds that
Dr. Siddiqui’s opinion that MASLD/MASH was not a cause of Mr. Roberts’s cancer

is speculative and based on nothing more than her ipse dixit. She did not employ a
reliable methodology in reaching her conclusion and so her opinion must be excluded.
b. Diabetes
Once again, Dr. Siddiqui ignored inconvenient medical records and scientific
literature to rule out diabetes as a cause of Mr. Roberts’s cancer, rendering her opinion

unreliable. It is undisputed that diabetes is an independent risk factor for HCC.
[Siddiqui Tr. 106:8-9.] Indeed, a 2017 study cited by both parties establishes that
Type II diabetes, which Mr. Roberts had, is “closely associated with increased risk of
HCC . . . independent of . . . obesity . . . and presence of cirrhosis and hepatic
steatosis.” [Mantovani 2017 at 3.] The scientific literature further establishes an 150%

increased risk associated with diabetes independent of other factors. [Id.]
Dr. Siddiqui, however, does not engage with these studies and wholly fails to explain
why diabetes can be reliably ruled out despite carrying a higher increased risk of liver
cancer than NDMA exposure. See Magistrini, 180 F. Supp. 2d at 610 (“to properly
assess the specific cause of an individual’s illness, a quantitative risk assessment must
be undertaken” for each risk factor); see also Kannankeril v. Terminix Int’l, Inc., 128 F.3d
802, 807 (3d Cir. 1997) (defining a proper differential diagnosis as “the determination

of which of two or more diseases with similar symptoms is the one from which the
patient is suffering, by a systematic comparison and contrasting of the clinical findings.”).
Rather than supporting her opinion with scientific data, Dr. Siddiqui purports
to rule out Mr. Roberts’s diabetes “as causing or even meaningfully increasing his risk

of liver cancer” for two reasons. [Siddiqui Report at 30.] First, according to
Dr. Siddiqui, “Mr. Roberts was not diagnosed with diabetes until November 14, 2016
– after [he] started taking NDMA contaminated valsartan.” [Id.] Second, Dr. Siddiqui
found his diabetes to be “well controlled.” [Id.] Both these reasons, however, are
belied by the record. Mr. Roberts’s December 2007 medical records note that he had

“DM,” or diabetes mellitus. [Rose Decl. Ex. 12 at GRobertsJr-CA-000729.]46 And in
July 2016 – months before ever taking a contaminated valsartan pill – Mr. Roberts’s
records show that he had “Type 2 diabetes mellitus with hyperglycemia,” and that his
hemoglobin A1C levels were 9.5%, representing uncontrolled glucose levels. [Rose
Decl. Ex. 13 at GRobertsJr-AMG-000042-44.] Dr. Siddiqui eventually conceded that

46 Plaintiff argues this document reported just possible diabetes, not an actual
diabetes diagnosis. [Pl.’s Opp’n at 18.] Either way, there is ample evidence of diabetes
in Mr. Roberts’s medical records well before he began taking the contaminated
valsartan, all which Dr. Siddiqui outright ignored. Indeed, Dr. Mahmud also noted
several other medical records that noted the diagnosis of “DM” or “diabetes mellitus”
in June 2007, December 2007, and May 2008. [Mahmud Report at 6.]
Mr. Roberts did have high hemoglobin A1C, “but he was trying to control it.”
[Siddiqui Tr. 110:1-2.]
Ultimately, her opinion regarding the link between Mr. Roberts’s diabetes and

his cancer is totally divorced from the facts of the case. Even assuming Mr. Roberts’s
diabetes was well-controlled – which the record evidence indicates it was not –
Dr. Siddiqui cites no scientific data to support her claim that well-controlled diabetes
does not increase the risk of developing liver cancer. This opinion relies solely on her
ipse dixit and is, therefore, patently unreliable and must be excluded.

c. Obesity
The scientific literature establishes that obesity is an independent risk factor for
HCC.47 Obesity has been associated with as high as a 286% increased risk for HCC,
even after adjusting for cirrhosis.48 Dr. Siddiqui does not dispute this increased risk
[Siddiqui Tr. 103:16-104:4], but she simply brushes it off.

47 See, e.g., Xuancheng Xie et al., Correlation analysis of metabolic characteristics and
the risk of metabolic-associated fatty liver disease-related hepatocellular carcinoma, Scientific
Reports (2022) (“Xie 2022”) [Rose Decl. Ex. 15 (Docket No. 3072-14)].
48 Xie 2022 at 6; see also Baek Gyu Jun et al., Impact of overweight and obesity on the
risk of hepatocellular carcinoma: a prospective cohort study in 14.3 million Koreans, 127 Brit.
J. of Cancer 109 (2022) [Rose Decl. Ex. 21 (Docket No. 3072-20)] (hazard ratio of 1.55
associated with obesity after adjusting for diabetes and hazard ratio of 1.60 for each
5 kg/m2 increase in BMI); Yi Chen et al., Excess body weight and the risk of primary liver
cancer: An updated meta-analysis of prospective studies, 48 Eur. J. of Cancer 2137 (2012)
[Rose Decl. Ex. 22 (Docket No. 3072-21)] (risk ratios of 1.69 associated with obesity
after adjusting for alcohol use, 1.89 associated with obesity after adjusting for hepatitis,
and 2.06 associated with obesity after adjusting for diabetes).
Despite Mr. Roberts’s longstanding history of both Class II and Class III obesity
[See generally Rose Decl. Ex. 17], Dr. Siddiqui chose not to address this literature at all.
Rather, she summarily concluded that “the increased risk of liver cancer due to obesity

would have been negligible, and again best captured based on the degree of his liver
cirrhosis.” [Siddiqui Report at 30.] She did not even attempt to quantify the increased
risk posed to Mr. Roberts due to his long-term significant history of obesity or compare
that increased risk to any increased risk caused by exposure to NDMA. Had she, she
would have been forced to grapple with the fact that the studies universally establish a

higher increased risk due to obesity than NDMA exposure. Dr. Siddiqui failed to
address relevant scientific literature, quantify and compare relative risk ratios for
different risk factors, or explain her reasoning for finding that any increased risk due
to obesity was “negligible.” For these reasons, the Court finds that Dr. Siddiqui’s
decision to rule out obesity was scientifically unreliable.

iii. Dr. Siddiqui Did Not Reliably Rule in NDMA Exposure from
Valsartan
Defendants contend that Dr. Siddiqui has no scientific basis for “ruling in”
exposure to NDMA-contaminated valsartan as a potential cause of Mr. Roberts’s
HCC. The Court agrees. Dr. Siddiqui had no reliable basis to conclude that the level
of NDMA to which Mr. Roberts was exposed could have, much less did cause, his
cancer. Instead, she inferred a causal relationship between NDMA exposure and liver
cancer from the limited human studies when the authors expressly disclaimed

causality. She speculated about the true increased risk of cancer and dose response
based upon her ipse dixit alone. And she opined on the timeline and temporal
proximity of Mr. Roberts’s exposure to NDMA and his cancer based on, in her own
words, “pure speculation.” [Siddiqui Tr. 88:9.] In short, her testimony was not based

on sufficient data and is not the product of reliable principles and methods.
Admittedly, there is a small body of scientific literature addressing the link
between NDMA exposure and liver cancer. Plaintiff argues that this is because
“NDMA is too toxic to ethically test on humans,” and that Dr. Siddiqui should not be
faulted for the paucity of on-point studies. [Pl.’s Supp. Br. at 1.] Surely, the Court

does not. But the lack of on-point studies does not excuse Dr. Siddiqui’s wholesale
unsupported speculation and ipse dixit assertions relating to the studies that do exist.
Perry v. Novartis Pharms. Corp., 564 F. Supp. 2d 452, 467–68 (E.D. Pa. 2008).
Dr. Siddiqui has no reliable basis rooted in the scientific literature for ruling in

NDMA exposure from valsartan as a possible cause of Mr. Roberts’s HCC. She
testified that her decision was based on several studies that purportedly showed a
causal relationship between NDMA exposure at the levels found in valsartan and liver
cancer. As noted above, the studies are few and far between. Dr. Siddiqui employed
no reliable methodology in assessing the findings of the few studies that exist at this

time. And none provides a reliable basis for her conclusions.
a. The Gomm and Mansouri Studies
Only two studies exist that have found a statistically significant association
between NDMA ingestion from a medication and cancer: Gomm and Mansouri.
Neither study, however, was able to establish a causal relationship.49
Mansouri qualified its findings, stating that “[m]ore research is needed to gain further
evidence and to understand more deeply the relationship between NDMA exposure

and the risks of liver cancer . . . and to establish clinically relevant causality.”
[Mansouri at 12.] Gomm put it more bluntly: “the present study can only state the
existence of a statistical association [between NDMA exposure and cancer]. Causality
cannot be inferred.” [Gomm at 360 (emphasis added).] Yet that is precisely what

Dr. Siddiqui did.
Not only did Dr. Siddiqui infer causality from these studies, but she also failed
to grapple with the very low risk levels reported in those studies – hazard ratios of just
1.16 and 1.12, respectively. Instead, she opined that any increased risk found in
Gomm and Mansouri was “diluted” and that the true risk was significantly higher.

[Siddiqui Tr. 33:7-34:24, 40:15-41:19.] The ZHP-manufactured valsartan taken by
Mr. Roberts contained far more NDMA than the valsartan manufactured by most
other companies. For example, a 320 mg valsartan pill manufactured by Defendants
contained 13.18 to 20.19 micrograms of NDMA. By contrast, a 320 mg valsartan pill
manufactured by Hetero Labs Ltd. contained just 0.33 to 0.44 micrograms of

49 Pottegård – the only other study that addressed this – found no association
between NDMA exposure in valsartan and liver cancer.
NDMA.50 According to Dr. Siddiqui, because the Gomm and Mansouri studies did
not distinguish between the type of pill taken by the study participants and necessarily
included participants who took pills with lower concentrations of NDMA, the results

must be diluted.
Intuitively, Dr. Siddiqui’s dilution theory makes sense. But expert testimony
cannot be based on intuition alone. It must be backed by data and scientific evidence.
Calhoun, 350 F.3d at 321 (“the expert’s opinion must be based on the ‘methods and
procedures of science’ rather than on ‘subjective belief or unsupported speculation.’”).

Under scrutiny, it is clear that this theory is based on nothing more than speculation.
Dr. Siddiqui admits as much. She concedes that the studies do not disclose what pills
were being taken by the study participants and that the participants “may have just
filled one prescription as compared to those people like Mr. Roberts who may have
filled multiple prescriptions.” [Siddiqui Tr. 80:17-81:4.] Underlying this speculative

theory, however, is an assumption that dose response has been established. In other
words, the more NDMA a person is exposed to, the more likely they are to develop
cancer. But this assumption is just as speculative. As Dr. Siddiqui herself recognizes,
“none of the valsartan epi studies detected a dose response.” [Siddiqui Report at 26.]
Putting that issue aside for the time being, the studies are missing the data that

is critical to support her theory: the types of valsartan pills ingested, the dose of

50 FDA Laboratory analysis of valsartan products, available at
https://www.fda.gov/drugs/drug-safety-and-availability/laboratory-analysis-
valsartan-products (“FDA Lab. Analysis”).
valsartan, the frequency at which the valsartan was taken, and the duration of
exposure. [See Siddiqui Tr. 81:20-83:7.] Also missing is any data on the market share
of the various manufacturers, which Dr. Siddiqui recognized is “really important.”

[Id. 82:8-83:7 (explaining that it is possible that Hetero dominates the French market,
but not considering the inverse that perhaps ZHP does, which would undermine her
assumptions).] Dr. Siddiqui assumes that the study participants overall were exposed
to less NDMA than Mr. Roberts. But that is purely speculation. As counsel put it and
Dr. Siddiqui could not deny, “it could go the other way.” [Id. 83:8-11.] Dr. Siddiqui

agreed that it is possible that the study participants could have been exposed to far
more NDMA than Mr. Roberts was either because of the type of pills they took or the
duration of exposure. [Id. 83:13-20.]
Dr. Siddiqui’s opinion that Gomm and Mansouri establish that NDMA

exposure in valsartan is a cause of cancer is simply not the product of reliable
methodology. Rather, this Court has no hesitancy in finding that it is patently
results-oriented and ignores what undermines her conclusion. In reaching this
conclusion, Dr. Siddiqui ignored other possible scenarios and overstated the scientific
data available to support her finding.51

51 This is not the only time Dr. Siddiqui has ignored scenarios or data that
undermined her conclusion. In its Laboratory Analysis, the FDA “estimated that if
8,000 people took the highest valsartan dose (320 mg) containing NDMA from the
recalled batches daily for four years, there may be one additional case of cancer over
the lifetimes of the 8,000 people.” [FDA Lab. Analysis.] Rather than consider this,
Dr. Siddiqui once again discounted the findings due to her dilution theory, despite
recognizing on cross-examination that the FDA was fully aware of the differing levels
b. The Hidajat Study
Next, Dr. Siddiqui relied upon Hidajat and various animal studies to establish
dose response. But she did not do so reliably. The Court first addresses her reliance

on Hidajat. Hidajat is “an occupational cohort study of 36,441 men who worked in a
rubber factory in the United Kingdom.” In re Zantac (Ranitidine) Prods. Liab. Litig.,
644 F. Supp. 3d 1075, 1214 (S.D. Fla. 2022). According to Dr. Siddiqui, this study
establishes dose response because the factory workers “who were exposed to higher
amounts of NDMA in the industry were more likely to die due to liver cancer than

workers who were exposed to lower amounts of NDMA in the factory.” [Siddiqui
Report at 26.] From this, she opines that “the human study also demonstrated a
dose-response.” [Id.] But her reliance on this study to establish dose response in the
context of NDMA in an ingested tablet is woefully misplaced and unreliable.
This Court is not the first to address an expert’s use of Hidajat to conclude that

an NDMA-contaminated medication can cause cancer. It has the benefit of a
thorough opinion from its sister court in the Southern District of Florida, which
concluded that “an expert opinion that relies upon the [Hidajat study] to conclude that
[a drug containing NDMA] can cause cancer utilizes an unreliable methodology” for
reasons that apply with equal force here. Zantac, 644 F. Supp. 3d at 1214–15.

“First, there are too many inherent uncertainties in [the Hidajat] occupational
study for such data to be reliably applied to the causation question in this MDL.” Id.

of contamination in different valsartan products when it issued its analysis. [Siddiqui
Tr. 91:8-92:22.]
at 1215. “Specific to the Hidajat occupational study, the number of assumptions and
estimations necessary to render this study helpful to a jury are staggering.” Id. The
Zantac court explained those “staggering” assumptions as follows:

workers continued to work in the factory after 1967; workers retired at
age 70; workers never ceased to work until they were 70; workers never
moved to another department with different NDMA-exposure levels; and
workers never left the rubber industry to be exposed to different
carcinogens in some other occupation. Additionally, an estimation was
necessary to compute the amount of NDMA each worker breathed in his
department, given that such data was never recorded. Finally, the study
authors had to try to adjust for confounding variables and for the
possibility that other carcinogens in the rubber factory could have caused
a worker's cancer.
Id. at 1215–16. The author of the study herself testified in the Zantac MDL “that her
study was not designed to assess NDMA exposure through an oral medication.”
Id. at 1216. “The route of exposure—skin absorption, inhalation, and ingestion—is
no small matter. Those routes affect the magnitude of ultimate exposures and because
they often affect health outcomes in different ways.” Id. (cleaned up) (quoting Reference
Manual on Scientific Evidence 71 (3d ed. 2011) at 518). Critically, in relying “on the
studies of others,” an expert “must not exceed the limitations the authors themselves
place on the study. That is, she must not draw overreaching conclusions.” Id. (cleaned
up) (internal citations omitted).
“For the Hidajat study to possess some level of relevance, an expert would have
to testify at trial that the inhaled and absorbed fumes from a 1967 rubber factory may
be reliably converted into an ingested dose of” contaminated valsartan. Id. “And of
course, for such a conversion to be part of a reliable methodology, it would also have
to be weighed in conjunction with assumptions about how long a worker worked,
where they worked, and what they were exposed to. Then further assumptions and
estimations would be necessary for comparison.” Id. at 1217. Additionally, an expert

would have to “account for the fact that . . . rubber fumes contain other chemicals,
including other established carcinogens” and determine whether “the levels of NDMA
in [valsartan] were comparable to the levels of NDMA that the subjects in” the Hidajat
study inhaled. In re Zantac (Ranitidine) Litig., 342 A.3d 1131, 1154 (Del. 2025).
Yet Dr. Siddiqui has made no attempt whatsoever – let alone a reliable attempt

– to extrapolate the Hidajat findings to the facts at bar. Instead, she concluded, without
explanation or reliable basis, that dose response exists between ingested NDMA in
valsartan and liver cancer because it was – shakily, at best – established in the Hidajat
study via completely distinct routes of exposure and at unknown levels. There is

“simply too great an analytical gap between the data and the opinion proffered.”
Joiner, 522 U.S. at 146. Accordingly, the Court further finds Dr. Siddiqui’s
methodology unreliable insofar as it relies upon the Hidajat study. See Zantac,
644 F. Supp. 3d at 1217.
c. Animal Studies
Finally, Dr. Siddiqui relied upon animal studies to establish a causal

relationship, dose response, and a significantly shortened latency period for cancer
caused by NDMA exposure. She testified that the animal studies show that animals
exposed to NDMA develop cancer “very rapidly,” and that the more NDMA the
animals were exposed to, the more likely they were to develop cancer. [Siddiqui
Tr. 38:4-8, 39:2-5, 85:22-25.] From this, she infers that dose response would hold true
in the context of humans at the levels exposed to in the contaminated valsartan. And,
despite being unable to opine on what amount of NDMA exposure was required to

cause Mr. Roberts’s cancer, she claims he could have developed cancer after “even a
few doses.” [Id. 88:1-3.]
Generally, courts are “reluctant to allow expert testimony based upon animal
studies to prove causation in humans unless there are good grounds to extrapolate data
and results from animals to humans.” In re Hum. Tissue Prods. Liab. Litig., 582 F. Supp.

2d 644, 670 (D.N.J. 2008) (citing Paoli, 35 F.3d at 742–43). And for good reason.
A rat is not a person. This should go without saying. Yet Dr. Siddiqui wholly “fail[ed]
to take into account critical differences between animal data and human experience—
including but not limited to extrapolations in dosing—render[ing] [her] methodology

scientifically invalid and unreliable.” Soldo v. Sandoz Pharms. Corp., 244 F. Supp. 2d
434, 547 (W.D. Pa. 2003); see also Joiner, 522 U.S. at 144–45.
“The use of animal studies to prove causation in human beings has ‘two
significant disadvantages,’ which ‘are almost always fraught with considerable, and
currently unresolvable, uncertainty.’ First, extrapolating from animals to humans is

difficult because ‘differences in absorption, metabolism, and other factors may result
in interspecies variation in responses.’” Soldo, 244 F. Supp. 2d at 466 (quoting Federal
Judicial Center, Reference Manual on Scientific Evidence, at 130 (1994)). Indeed, rats
typically have much shorter lifespans than humans, meaning that the timelines of
disease in rodents and humans differ considerably. See Zantac, 644 F. Supp. 3d
at 1279–84. But Dr. Siddiqui did not consider these differences or this so-called
“translation gap” in formulating her differential diagnosis. [See Mahmud Tr. 134:21-
136:20.]52 Plaintiff argues that Dr. Siddiqui relied on Dr. Panigrahy’s “translation

mechanism” as set forth in his general causation report. [9/3/25 Tr. 27:15-22.] But
the Court sees no reference whatsoever to Dr. Panigrahy’s – or any other – translation
mechanism in Dr. Siddiqui’s report or testimony. Instead, she testified that just a few
doses over the course of just weeks to months could cause cancer in humans, just as it
could in rodents. [Siddiqui Tr. 87:22-88:13.]

“A second difficulty is that ‘the high doses customarily used in animal studies
requires consideration of the dose-response relationship and whether a threshold no-effect
dose exists.” Soldo, 244 F. Supp. 2d at 466 (quoting Federal Judicial Center, Reference
Manual on Scientific Evidence, at 130 (1994)) (emphasis in original). Dr. Siddiqui

does not engage at all with the fact that the rats studied were injected with massive
amounts of NDMA – orders of magnitude higher than what Mr. Roberts was exposed
to. [See Chodosh Dep. 227:4-15 (Docket No. 3117-2) (extrapolating from animal study
inappropriate because Mr. Roberts’s cumulative exposure to NDMA was “several
thousand times lower” than the dose given to animals).] Instead, she simply assumes

52 Dr. Mahmud, Defendants’ specific causation expert, explained that
epidemiologists and clinicians are “always cautious [when] try[ing] to translate
findings from an animal study directly to humans because inherently humans are
different from mice.” [Mahmud Tr. 131:14-24.] He testified that any extrapolation
must account for this “translation gap” and not “just blindly extrapolate something
you see in an animal study and assume that the same thing will be observed in
humans.” [Id. 135:3-16.]
the existence of a dose-response relationship at any level of exposure in humans
because a dose response relationship at extremely high levels of exposure has been
established in rodents. [Siddiqui Tr. 85:22-25.] She does not explain the basis for this

extrapolation. She essentially just asks that the jury take her word for it. But the Court
need not accept expert opinions “connected to existing data only by the ipse dixit of the
expert.” Joiner, 522 U.S. at 146.53
When asked how much NDMA-contaminated valsartan was necessary to cause
Mr. Roberts’s cancer, Dr. Siddiqui was unable or unwilling to say. Her response was

simply: “I believe that the amount that he had caused his cancer.” [Siddiqui
Tr. 87:11-12.] The exchange that followed bears repeating here:
Q. What I’m trying to understand is when is the point, according
to your theory, that [Mr. Roberts] had consumed enough contaminated
valsartan such that he developed cancer? Does that make sense?
A. So my understanding based on my reading is this, that NDMA
is such a potent cancer-inducer that even a few doses, even a few, even
weeks to months can cause cancer.
Now, again, Your Honor, of course this is in animal studies, but if
we extrapolate the weeks to months, if we extrapolate it to a year, for
example, the fact that NDMA is such a strong cancer-inducer and
promoter – I don’t know at which point he got the cancer. I believe it

53 Dr. Siddiqui appears to suggest that because the FDA has stated that consuming
up to 0.096 micrograms of NDMA per day is considered reasonably safe for human
ingestion, any exposure above that is automatically capable of causing cancer. [See
Siddiqui Tr. 85:15-20, 86:3-14.] And Plaintiff argues that the fact that the FDA used
animal studies in formulating this cutoff somehow justifies Dr. Siddiqui’s reliance on
those studies. [9/3/25 Tr. 23:23-25.] Both miss the point. “[T]he FDA primarily
engages in a process of risk assessment rather than determination causation, and the
relevance to causation of evidence used to assess risk is not clear.” In re Diet Drugs,
2001 WL 454586, at *15 (E.D. Pa. Feb. 1, 2001).
was somewhere maybe in around 2018, but this is pure speculation. The
reason being that it was so aggressive that he started having symptoms.
So I would assume, my assumption is that it would have been a
fast-growing cancer, and he would have sought help as soon as it started
to grow.
[Id. 87:22-88:13 (emphasis added).] A break then ensued. Thereafter, the colloquy
continued:
Q. Dr. Siddiqui, when we left off we were looking at this slide that
you showed earlier this morning in talking about it’s true, is it not, that
you were unable to tell us at what point on this chart, in your view,
Mr. Roberts had consumed enough contaminated valsartan to cause his
HCC; true?
A. So what I was trying to say was that based on the data that I
reviewed, it can be -- NDMA can sometimes be as quick as weeks to months.
So it could be at any point. And then I was trying to -- I think I went into
this explanation of how with the carcinomas you can’t always tell when
they arise, so we assume, we make the assumption that they, depending
on the type of cancer, that they really started growing when the patient
first starts having symptoms. So I was just sort of talking about that.
Q. And so in your view, it could have been just a few weeks of
taking contaminated valsartan that Mr. Roberts had consumed enough
in your opinion to cause his HCC, correct?
A. My opinion is that although there are animal studies, Your
Honor, that talk about a few weeks to a few months, I think for Mr.
Roberts, it must have been -- again, this is pure speculation, Your Honor,
because we don’t know, we don’t know that, but I think what we do
know is it came up somewhere between 2016 and 2018, right, most likely,
I think. But, again, it’s pure speculation, I think it was somewhere
probably late 2017, maybe early 2018. And the reason I think that is
because he -- he went in when he started having symptoms. And he
started having symptoms when his cancer must have, you know, grown
to a point that it was causing him issues. And that’s why I’m speculating
that it may have been somewhere in late 2017, 2018. And he continued
to take the contaminated valsartan throughout that.
Q. And what that means, Dr. Siddiqui, just to round out this
questioning on this slide, is you can’t give us a particular dose of
contaminated valsartan that is necessary in order to induce HCC,
correct?
A. So there are no studies that I can think of that look at that.
[Id. 89:9-90:20 (emphasis added).] This rampant speculation cannot reliably support
Dr. Siddiqui’s decision to rule in NDMA exposure as the cause of Mr. Roberts’s
cancer.
Plaintiff argues that even if dose response cannot be reliably established,

Dr. Siddiqui’s testimony supports a temporal proximity theory of causation. But her
testimony that NDMA can cause significantly accelerated cancer development is also
based on speculative and unreliable extrapolation from animal studies. Dr. Siddiqui
testified that NDMA causes mutations that cause cancer “very rapidly, sometimes as
early as weeks to months in animals, of course, . . . because we don’t have human

studies.” [Id. 39:2-5.] But the lack of human studies does not excuse her failure to
account for the differences between rats and humans. She later testified that “the
literature and the data” shows that NDMA can cause cancer “in months,” without
even a caveat that this timeline comes from studies on rats, not humans. [Id. 47:11-13.]
And, as Dr. Siddiqui repeatedly stated, her opinion as to when Mr. Roberts developed

cancer and, therefore, the relevant latency period, is based on nothing more than pure
speculation. [Id. 87:22-88:13, 89:9-90:20.] She uses the word “extrapolate” but then
does not actually do any reliable extrapolation or translation from the animal studies
to humans. Any meager attempts to extrapolate a dose response relationship or
shortened latency period from the animal studies to humans are not based upon good
grounds or reliable methods, but rather speculation. Joiner, 522 U.S. at 146; McClain
v. Metabolife Int’l, Inc., 401 F.3d 1233, 1237 (11th Cir. 2005) (“Daubert requires the trial
court to act as a gatekeeper to [ensure] that speculative and unreliable opinions do not

reach the jury.”).
Dr. Siddiqui made no attempt to reliably extrapolate any of the data or results
in the animal studies to humans. She did not consider the drastically different life
cycles of rats and humans to reliably determine a latency period in humans, or the
massive difference in the doses to which the rats and Mr. Roberts were exposed to

reliably establish a dose-response relationship in humans. Soldo, 244 F. Supp. 2d
at 547 (expert testimony based on animal studies is unreliable where it does not “take
into account critical differences between animal data and human experience—
including but not limited to extrapolations in dosing.”). She made the questionable

and unsupported “assumption that chemicals behave similarly in different species.”
Daniels-Feasel v. Forest Pharms., Inc., 2021 WL 4037820, at *13 (S.D.N.Y. Sept. 3,
2021), aff’d sub nom. Daniels-Feasel v. Forest Pharms., Inc., 2023 WL 4837521 (2d Cir.
July 28, 2023). The analytical gap between the animal studies and Dr. Siddiqui’s
causation opinions is simply too great. See, e.g., Zantac, 644 F. Supp. 3d at 1280–84;

Oddi, 234 F.3d at 146 (reliability requirement prohibits “too great a gap between the
data and the [expert’s] proffered opinion.”). Accordingly, the Court finds that
Dr. Siddiqui’s causation, dose response relationship, and temporality opinions are
unreliable and must be excluded.54
As the Mansouri study noted, “[m]ore research is needed to gain further

evidence and to understand more deeply the relationship between NDMA exposure
and the risks of liver cancer.” [Mansouri at 12.] To be sure, limited studies exist at
this time. But, contrary to Plaintiff’s protestations, “the non-existence of good data
does not allow expert witnesses to speculate or base their conclusions on inadequate
supporting science.” Perry v. Novartis Pharms. Corp., 564 F. Supp. 2d 452, 467–68

(E.D. Pa. 2008). The Perry court put it well:
In cases where no adequate study shows the link between a substance
and a disease, expert testimony will generally be inadmissible, even if
there are hints in the data that some link might exist. This may mean
that early victims of toxic torts are left without redress because they are
unable to prove their cases with the scientific data that exists. While this
is a regrettable result in those individual cases, it is an unavoidable reality
of the structure of our legal system and is necessary to protect the interests
of defendants who might otherwise be subject to crippling verdicts on the
basis of slender scientific evidence. . . . As the Seventh Circuit has noted,

54 Even if Dr. Siddiqui had reliably ruled in NDMA exposure from valsartan as a
possible cause, the Court agrees with Defendants that her determination that this was
the only substantial factor in causing Mr. Roberts’s cancer is not the product of a
reliable application of established principles and methods. Her differential diagnosis
is fundamentally flawed in that it lacks any comparison or assessments of the relative
risks of HCC associated with each of Mr. Roberts’s known risk factors and his NDMA
exposure. See, e.g., Ervin v. Johnson & Johnson, Inc., 2006 WL 1529582, at *7 (S.D. Ind.
May 30, 2006), aff’d, 492 F.3d 901 (7th Cir. 2007) (excluding causation expert opinion
where expert ruled in one cause associated with a small increased risk while refusing
to rule in other causes associated with higher risks); In re Zimmer Nexgen Knee Implant
Prods. Liab. Litig., 218 F. Supp. 3d 700, 718 (N.D. Ill. 2016), aff’d sub nom. In re Zimmer,
NexGen Knee Implant Prods. Liab. Litig., 884 F.3d 746 (7th Cir. 2018) (expert’s
differential diagnosis was unreliable where it applied “inconsistent” standards “to
determine which potential causes are reasonable and should be ruled in.”).
“[t]he courtroom is not the place for scientific guesswork, even of the
inspired sort. Law lags science; it does not lead it.”
Id. at 468 (quoting Rosen v. Ciba-Geigy Corp., 78 F.3d 316, 319 (7th Cir. 1996)). “Based
on the data that exist today,” this Court easily concludes that Dr. Siddiqui’s opinions
regarding the causal and dose response relationships between NDMA exposure in
valsartan and liver cancer are “mere guesswork.” Id. at 469. “While such speculation

is appropriate in the laboratory where a hypothesis can be tested by experiment, it has
no place in the courtroom where no such testing is possible.” Id.

For these reasons, the Court finds that Dr. Siddiqui’s testimony is not based on
sufficient facts and data, is not the product of reliable principles and methods, and does

not reflect the reliable application of the principles and methods to the facts of the case.
Accordingly, her testimony must be excluded for failure to comply with the strictures
of Rule 702.
V. MOTION FOR SUMMARY JUDGMENT
A. The Legal Standards Governing Summary Judgment Motions

Summary judgment is granted where “the movant shows that there is no
genuine dispute as to any material fact and the movant is entitled to judgment as a
matter of law.” FED. R. CIV. P. 56(a). A fact is “material” if it “might affect the
outcome of the suit under the governing law.” Anderson v. Liberty Lobby, Inc., 477 U.S.
242, 248 (1986). A dispute of material fact is genuine “if the evidence is such that a

reasonable jury could return a verdict for the nonmoving party.” Id. In determining
whether a genuine dispute of material fact exists, “all evidence is viewed in the light
most favorable to the non-moving party and ‘all justifiable inferences are to be drawn
in his favor.’” Morgan v. Allison Crane & Rigging LLC, 114 F.4th 214, 220 (3d Cir. 2024)

(quoting Anderson, 477 U.S. at 255). A “mere scintilla of evidence,” however, does
not generate a genuine dispute of material fact. Anderson, 477 U.S. at 252.
The movant “always bears the initial responsibility of informing the district
court of the basis for its motion, and identifying those portions of ‘the pleadings,

depositions, answers to interrogatories, and admissions on file, together with the
affidavits, if any,’ which it believes demonstrate the absence of a genuine issue of
material fact.” Celotex Corp. v. Catrett, 477 U.S. 317, 323 (1986) (quoting FED. R. CIV.
P. 56(c)). Then, “when a properly supported motion for summary judgment [has been]
made, the adverse party ‘must set forth specific facts showing that there is a genuine

issue for trial.’” Anderson, 477 U.S. at 250 (citing FED. R. CIV. P. 56(e)).
The nonmovant’s burden is rigorous. The nonmovant “must point to concrete
evidence in the record”; mere allegations, conclusions, conjecture, and speculation
will not defeat summary judgment. Orsatti v. New Jersey State Police, 71 F.3d 480, 484
(3d Cir. 1995); accord Jackson v. Danberg, 594 F.3d 210, 227 (3d Cir. 2010) (citing

Acumed LLC v. Advanced Surgical Servs., Inc., 561 F.3d 199, 228 (3d Cir. 2009)
(“[S]peculation and conjecture may not defeat summary judgment.”)). If the
nonmoving party “‘fails to make a showing sufficient to establish the existence of an
element essential to [his] case, and on which [he] will bear the burden of proof at trial,’
then summary judgment is appropriate for the moving party.” SodexoMAGIC, LLC v.
Drexel Univ., 24 F.4th 183, 204 (3d Cir. 2022) (quoting Celotex, 477 U.S. at 322).

B. Summary Judgment is Appropriate Because Plaintiff Cannot Establish
Specific Causation
Under Alabama law, “[r]egardless of the cause of action asserted, whether a
defective product caused a plaintiff’s alleged injuries is ‘an essential element of all
product liability cases.’” Lowery v. Sanofi-Aventis LLC, 535 F. Supp. 3d 1157, 1175
(N.D. Ala. 2021).

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/cases/11201028. Public record. Not legal advice.