Cheryl Koehn, as Mother and Next Friend of Vanessia Koehn v. Secretary of Health and Human Services

In the United States Court of Federal Claims
OFFICE OF SPECIAL MASTERS

**********************
CHERYL KOEHN, *
as mother and next friend of * No. 11-355V
VANESSIA KOEHN, * Special Master Christian J. Moran
*
Petitioner, * Filed: May 30, 2013
*
v. *
* Entitlement, HPV vaccine (Gardasil),
SECRETARY OF HEALTH * systemic juvenile idiopathic arthritis
AND HUMAN SERVICES, * (sJIA)
*
Respondent. *
**********************

P. Leigh O’Dell, Beasley, Allen, et al., Montgomery, AL, for petitioner;
Darryl R. Wishard, United States Dep’t of Justice, Washington, DC, for
respondent.

PUBLISHED DECISION DENYING COMPENSATION1

Cheryl Koehn alleges that two doses of the human papillomavirus (“HPV”)
vaccine given to her daughter, Vanessia, caused her to suffer from systemic
juvenile idiopathic arthritis (“sJIA”).2 Ms. Koehn seeks compensation from the

1
The E-Government Act of 2002, Pub. L. No. 107-347, 116 Stat. 2899,
2913 (Dec. 17, 2002), requires that the Court post this decision on its website.
Pursuant to Vaccine Rule 18(b), the parties have 14 days to file a motion proposing
redaction of medical information or other information described in 42 U.S.C.
§ 300aa–12(d)(4). Any redactions ordered by the special master will appear in the
document posted on the website.
2
The Secretary recognized the HPV vaccine as a vaccine covered in the
Vaccine Program on April 20, 2007. National Vaccine Injury Compensation
(. . . continued)
National Childhood Vaccine Injury Compensation Program. 42 U.S.C. § 300aa-10
et seq. (2012). To establish that she is entitled to compensation, Ms. Koehn must
fulfill the three-pronged test set forth in Althen v. Sec’y of Health & Human
Servs., 418 F.2d 1274, 1278 (Fed. Cir. 2005).

Ms. Koehn relies primarily upon the opinion of Michael McCabe, Ph.D. Dr.
McCabe presents an innovative theory involving cytokines to explain how an HPV
vaccine can cause sJIA. The Secretary, however, undermined the persuasive value
of Dr. McCabe’s hypothesis by presenting a contrary opinion from Carlos Rose,
M.D., a board-certified rheumatologist. As detailed in section IV below, Dr.
McCabe’s theory has not been tested, has not been the subject of peer-review, is
not generally accepted in the relevant medical community, and is inconsistent with
epidemiological studies.

The flaws in Ms. Koehn’s evidence extend from the first prong of Althen to
the remaining two prongs. Ms. Koehn has not established that the onset of her
sJIA occurred in a temporal interval that Dr. McCabe’s theory would predict. See
section V. Additionally, Ms. Koehn’s case lacks a “logical sequence of events”
that connects her disease to the HPV vaccination as required by the second prong
of Althen. See section VI.

Consequently, Ms. Koehn has not established that she is entitled to
compensation. A full discussion follows.

I. Procedural History

Ms. Koehn filed her petition on June 6, 2011, and medical records on June
14, 2011. These medical records are summarized in section II.C, below. Ms.
Koehn filed a report from Dr. McCabe on August 24, 2011. Exhibit 9. Due to
concerns about the adequacy of the disclosure regarding Althen prong one, Ms.

Program: Addition of Meningococcal and Human Papillomavirus (HPV) Vaccines
to the Vaccine Injury Table, 72 Fed. Reg. 19937. Although many petitioners have
claimed that the HPV vaccine harmed them, this may be the first instance in which
a claim has reached a special master for resolution. (Other HPV vaccine cases
have been resolved when petitioners acknowledged that they were not likely to
prevail or when the parties reached a settlement.)

2
Koehn filed a supplemental report from Dr. McCabe on October 3, 2011. Exhibit
27. As discussed more extensively below, in sections II.D.1.b and c, Dr. McCabe
opined that the HPV vaccine caused Vanessia’s sJIA. Ms. Koehn also filed the
articles on which Dr. McCabe relied.

After Ms. Koehn made these submissions, the Secretary evaluated the
evidence. The Secretary recommended that compensation be denied because Ms.
Koehn had not satisfied any of the three elements set forth in Althen. In addition
to identifying perceived flaws in Dr. McCabe’s opinion, the Secretary also relied
upon an opinion presented by Dr. Rose. Resp’t Rep’t, filed Nov. 14, 2011. The
gist of Dr. Rose’s opinion is that there is not adequate evidence to support the
theory that the HPV vaccine can cause sJIA. See sections II.D.2.b and c, below.

The parties did not succeed in resolving the case through a settlement. Thus,
the case was set for a hearing. In advance of the hearing, the parties filed briefs
and additional medical literature. Dr. McCabe and Dr. Rose testified at a hearing
held on June 21, 2012. Following the hearing, the parties submitted additional
articles and briefs.

Ms. Koehn’s claim that the HPV vaccine caused Vanessia’s sJIA is ready for
adjudication. The foundational elements—the HPV vaccine and sJIA—are
discussed first. The following sections review Vanessia’s medical history as well
as the qualifications, reports, and testimony of the experts. After a short recitation
of the legal standards, this decision separately analyzes Ms. Koehn’s evidence for
each of the Althen prongs. Section VII provides the conclusion.

II. Background

To provide context to Vanessia’s medical history and the opinions of the
parties’ experts on the issue of vaccine causation, found below in sections II.C and
D, respectively, it is helpful first to review some preliminary information
concerning the vaccine Vanessia received and the condition from which she
suffers. Thus, sections II.A and B provide a brief overview of human
papillomavirus, HPV vaccine, and JIA.

3
 A. Human Papillomaviruses and Human Papillomavirus Vaccines

1. Human Papillomaviruses

There are more than 130 different types of human papillomaviruses. These
viruses tend to be found in cutaneous or mucosal epithelial surfaces. Some strains
of human papillomavirus are relatively benign, causing warts. Other strains, in
particular HPV 16 and HPV 18, cause cervical cancer. Exhibit 16 (Margaret
Stanley, Immunobiology of HPV and HPV vaccines, 109 Gynecologic Oncology
S15 (2008)) at S15-16.

Because of the cells that it infects, a human papillomavirus “is practically
invisible to the host who remains ignorant of the pathogen for long periods of
time.” A human papillomavirus does not cause cytolysis,3 necrosis,4 or
inflammation. Without exposure to the host’s immune system, “there is little or no
release into the local milieu of pro-inflammatory cytokines.” Id. at S16. This is
part of the virus’s strategy for survival.

Given enough time, “most [human papillomavirus] infections resolve.” But,
approximately 10-20 percent of infected individuals develop persistent infections.
One reason appears to be that humans produce relatively few antibodies in
response to the human papillomavirus. Id. at S17.

2. Vaccines against Human Papillomaviruses

Developing an effective vaccine against human papillomaviruses was
challenging, in part, because of the need to generate a robust response from the
person’s immune system. Exhibit 16 (Stanley) at S17-18. Researchers eventually
succeeded in creating an HPV vaccine that induces “high concentrations of
neutralizing antibodies.” Id. at S18. An HPV vaccine can cause the host to
produce more antibodies than the human papillomavirus because, in part, the
vaccine is given intramuscularly, close to the lymph nodes. This delivery system

3
Cytolysis is the “destruction of a cell by rupture of the cell membrane with
loss of cytoplasm.” Dorland’s Illustrated Medical Dictionary 466 (32nd ed. 2012).
4
Necrosis is “the sum of morphological changes indicative of cell death.”
Dorland’s at 1235.

4
“circumvent[s] the immune avoidance strategies of the viral intraepithelial
infectious cycle.” Id.

a) HPV Vaccine Composition

Another advance in the creation of vaccines against the human
papillomavirus was the reproduction of a portion of the virus known as the L1
protein. The resulting virus-like particle (VLP) stimulates the immune system to
produce antibodies and the antibodies confer immunity to the particular strand of
the human papillomavirus. Exhibit 17 (Margaret Stanley, HPV- immune response
to infection and vaccination, 5 Infectious Agents & Cancer 19 (2010)) at 2-3.
There are two different vaccines against human papillomavirus. One, known as
Cervarix, contains the L1 VLP for two strands, 16 and 18. The other, known as
Gardasil, contains the L1 VLP for four strands, 6, 11, 16, and 18. Id. at 3. In
addition to the difference in strands, Cervarix and Gardasil contain different
adjuvants.5 Cervarix uses an adjuvant known as AS04, which is comprised of a
lipid and an aluminum salt. Exhibit E (Thomas Verstraeten et al., Analysis of
adverse events of potential autoimmune aetiology in a large integrated safety
database of AS04 adjuvanted vaccines, 26 Vaccine 6630 (2008)) at 6631. On the
other hand, Gardasil uses amorphous aluminum hydroxyphosphate sulfate to
increase antibody production. Transcript (“Tr.”) 154; Physician’s Desk Reference
at 1828 (66th ed. 2012).

b) HPV Vaccine Effectiveness

Experiments on HPV vaccines have shown that “the peak geometric mean
antibody concentrations achieved are at least two [logarithmic] higher than those
after natural seroconversion”6 and for “the majority of vaccinated subjects, serum
antibody levels remain at concentrations greater than those found in natural
infection.” Exhibit 16 at S18. One article commented that “[i]t is fairly
uncommon that a vaccine will produce an immune response greater than that

5
An adjuvant is a stimulator of a more robust immune response. See
Dorland’s at 32
6
Seroconversion is “the change of a patient’s serologic test from negative to
positive, indicating the development of antibodies in response to infection or
immunization.” Dorland’s at 1698.

5
achieved by natural infection.” Exhibit 21 (Villa et al., Immunologic responses
following administration of a vaccine targeting human papillomavirus Types 6, 11,
16, and 18, 24 Vaccine 5571 (2006)) at 5581.7

The amount of time to generate the antibody response was discussed in
several articles. For example, one article reported that “vaccination induced a
marked immune response, beginning approximately 1 month after the initial dose,
peaking at approximately month 7, and thereafter declining to a stable plateau for
2.5 years after the last vaccine dose.” Exhibit 25 (Ian Frazer, Correlating
immunity with protection for HPV infection, 11 Int’l J. Infectious Diseases S10
(2007)) at S13; see also exhibit 22 (Elmar Joura et al., HPV antibody levels and
clinical efficacy following administration of a prophylactic quadrivalent HPV
vaccine, 26 Vaccine 6844 (2008)) at 6849; exhibit 28 (Alfonso García-Piñeres et
al., Cytokine and Chemokine Profiles following Vaccination with Human
Papillomavirus Type 16 L1 Virus-Like Particles, 14 Clinical & Vaccine
Immunology 984 (2007)) at 986; exhibit 29 at 331.

In addition to looking at the production of antibodies in response to an HPV
vaccine, researchers have also investigated the cytokine response. E.g., exhibit 28
(García-Piñeres), exhibit 32 (Rebecca T. Emeny et al., Priming of Human
Papillomavirus Type 11-Specific Humoral and Cellular Immune Responses in
College-Aged Women with a Virus-Like Particle Vaccine, 76 J. Virology 7832
(2002)), exhibit 30 (Thomas G. Evans et al., A Phase 1 Study of a Recombinant
Viruslike Particle Vaccine against Human Papillomavirus type 11 in Healthy Adult
Volunteers, 183 J. Infectious Diseases 1485 (2001)). At the hearing, relatively
little attention was paid to the García-Piñeres article, Emeny article, or the Evans
article because Dr. McCabe and Dr. Rose primarily discussed an article by Pinto.
See, e.g., Tr. 119-20.

7
Other studies that also reported that the vaccine produces a stronger
antibody response include exhibit 29 (Ligia A. Pinto et al., Cellular Immune
Responses to Human Papillomavirus (HPV)—16 L1 in Health Volunteers
Immunized with Recombinant HPV-16 L1 Virus-Like Particles, 188 J. Infectious
Diseases 327 (2003)) at 336; exhibit 35 (Purnima Bhat et al., Regulation of
immune responses to HPV infection and during HPV-directed immunotherapy,
239 Immunological Revs. 85 (2011)) at 87; and exhibit 18 (Luciano Mariani &
Aldo Venuti, HPV vaccine: an overview of immune response, clinical protection,
and new approaches for the future, 8 J. Translational Med. 105 (2010)) at 107.

6
 Dr. McCabe bases much of his opinion on the Pinto article, which, according
to Dr. McCabe, is “an important paper in vaccinology, the study of vaccines.” Tr.
100. The Pinto study is “a technical tour de force.” Tr. 100, 103. Therefore, due
to its complexity and its significance, the Pinto article is reviewed in detail.

When this study was conducted, vaccines against human papillomavirus
were being developed. Dr. Pinto and colleagues designed an experiment “to better
characterize the innate and acquired immune system cytokine response elicited by
L1 VLP vaccination.” Exhibit 26 (Ligia A. Pinto et al., HPV-16 L1 VLP vaccine
elicits a broad-spectrum of cytokine responses in whole blood, 23 Vaccine 3555
(2005)) at 3556. The vaccination referenced in the Pinto article contained one
protein present in Gardasil. Tr. 100.

Twenty-four women participated in the study. Blood specimens were
collected before the initial injection, known as month zero. Twenty women, then,
received a 50 µg dose of vaccination without adjuvant and four women received a
placebo of sterile saline solution. One month later, all women were given a second
injection of the same substance (either the vaccination or a placebo). At month
two, more blood was drawn. At six months, the women received a third injection.
At seven months, more blood was drawn. Exhibit 26 at 3556.

The researchers determined the level of cytokines for each of the three blood
samples after different types of stimulation. Exhibit 26 at 3557. This process was
done “in vitro,” id. at 3562, meaning in glass, like a test tube. Dorland’s at 956.
The blood from women who received the vaccine and women who received the
placebo was evaluated in the context of four substances. In the first, the blood was
not stimulated at all. The researchers refer to this as the “media.” In the second,
the blood was stimulated with 10 µg of the virus-like particle present in the
vaccine. In the third, the blood was stimulated with 1.0 µg of the virus-like
particle. In the fourth, the blood was stimulated with a control known as PHA.
Exhibit 26 at 3557, § 3.1; see also Tr. 292-93. The stimulation was for “24 [hours]
in the absence or presence of L1 VLP or PHA.” Exhibit 26 at 3559 (caption to
figure 1).

As discussed below in section IV.B.3, the researchers obtained different
results depending upon whether there was any stimulation. For cells in the
media—meaning no stimulation—the cytokine levels stayed relatively similar
from month zero to month two to month seven. “As shown in Fig. 1D,
spontaneous secretion of cytokines in the absence of any stimuli (media control)
7
did not show any significant increases following vaccination.” Exhibit 26 at 3560.
For blood that was stimulated either with 10 µg or 1.0 µg of the virus-like particle,
cytokines increased. “Stimulation of cells from vaccine recipients with L1 VLP
(10 µg/ml) induced significant increases in the median levels of inflammatory . . .
cytokines.” Id. at 3557-59. “Similar patterns of cytokine production to the ones
seen in response to L1 VLP at 10 µg/ml were observed when L1 VLP was tested at
1.0 µg/ml.” Id. at 3559.

c) HPV Vaccine Safety

Dr. McCabe and Dr. Rose each referenced one epidemiological study that
investigated the safety of an HPV vaccine.8 One was an article by Chun Chao.
Exhibit 34 (Chao et al., Surveillance of autoimmune conditions following routine
use of quadrivalent human papillomavirus vaccine, 271 J. Intern. Med. 193
(2012)). The other was an article by Thomas Verstraeten. Exhibit E.

(1) Chao Article

Dr. Chao and colleagues used a database to look at the medical history of
nearly 190,000 women. Their goal was to determine whether women who
received a dose of a quadrivalent human papillomavirus vaccine developed
autoimmune diseases within 180 days after the vaccination. Exhibit 34 at 193.
The researchers selected 180 days “to accommodate lag time for clinical work”
necessary for the treating doctor to arrive at the correct diagnosis. Id. at 201.

The article says that the researchers looked for cases of “juvenile rheumatoid
arthritis (JRA).” Id. at 194. The article explained how the researchers looked for
various diseases:

8
Dr. Rose also reproduced a portion of the package insert (also known as the
manufacturer’s label). See exhibit A at 3-4. The Secretary, however, did not
submit the package insert as an exhibit, did not ask any questions about the
package insert during direct examination of Dr. Rose, and did not cite it in her
post-hearing brief. Although Ms. Koehn asked some questions about the package
insert during the cross-examination of Dr. Rose, Tr. 252-60, the package insert
does not affect the outcome of this case.

8
 The method for case identification was designed to be highly sensitive
to capture any potential cases, to address potential undercoding or
miscoding in the early course of an autoimmune condition. To this
end, ICD-9 diagnosis codes, abnormal laboratory results or pharmacy
prescriptions possibly indicative of autoimmune conditions . . . were
captured.

Id. at 194-95. Information about the specific ICD-9 codes was contained in
Appendix A-C. Id. at 195. However, the copy of the Chao article that was filed as
exhibit 34 did not contain the appendices. See exhibit 34.

After the scope of the case ascertainment became an issue at the hearing, see
Tr. 248, Dr. Rose was permitted to file the relevant appendix and a report
commenting on the ICD-9 code. As a preliminary matter, Dr. Rose explained what
an ICD-9 code is:

The ICD-9 is a complex and evolving international coding system
utilized by patient care providers to identify the condition or
conditions suffered by their patients. The codes have a multiplicity of
uses including retrospective identification of cases for public health
projects (like the one in question), utilization of resources, quality
assurance and adequacy of charges for rendered services.

Exhibit I (Dr. Rose’s supplemental report, dated Nov. 1, 2012) at 2.9 Dr. Rose next
stated that under the ICD-9, the relevant code is 714.3, juvenile rheumatoid
arthritis. Id. The Chao researchers used this code. See exhibit H (reproduction of
Appendix A-1 from the Chao article). In addition, the Chao researchers also
searched for medications commonly prescribed for sJIA. Exhibit I at 3. Thus, Dr.
Rose concluded that “almost certainly all cases of JRA within the study population
would have been detected with the methodology utilized by the investigators.” Id.
at 4.

Given this understanding of what the researchers did, the results can be
stated. In the category of juvenile rheumatoid arthritis, the researchers found three

9
For more information about ICD-9 codes, see Fresco v. Sec’y of Health &
Human Servs., No. 06-469V, 2013 WL 364723, at *9 n.40 (Fed. Cl. Spec. Mstr.
Jan. 7, 2013).

9
cases arising after vaccination. Exhibit 34 at 197 (table 1, column E, line 6).
Among people who were not vaccinated, the researchers estimated that there were
43 cases. Id. at 199 (table 3, third column, line 6). The incidence rate ratios
(“IRR”) was 0.48 with a 95% confidence interval of 0.26-0.91. Id. (table 3,
columns 4-5, line 6). Dr. Rose explained that because the confidence interval was
below 1.0, there was “no increase in risk of developing new onset of JRA after
HPV vaccination.” Exhibit I at 3. Although Dr. Chao and colleagues did not make
a specific finding for juvenile rheumatoid arthritis, their overall conclusion was
similar. They stated that “this observational surveillance study offers some
assurance that amongst a large and likely generalizable female population, no
safety signal for autoimmune conditions was found following HPV4 vaccination in
routine clinical use.” Id. at 202.

(2) Verstraeten Article

The Verstraeten article collects several studies about the safety of vaccines
containing an adjuvant known as AS04. AS04 is the adjuvant in Cervarix, not the
adjuvant used in Gardasil. Exhibit E at 6631; Tr. 240 (Dr. Rose).

Dr. Verstraeten’s and colleagues’ goal was “to evaluate the safety of AS04
adjuvanted vaccines with regard to rates of AEs [adverse events] of potential
autoimmune aetiology.” Exhibit E at 6631. To address the problem that small
studies may not detect rare events, Dr. Verstraeten and colleagues collected “[a]ll
completed or ongoing controlled, randomised studies of ASO4 adjuvanted HPV-
16/18, HSV and HBV vaccines conducted by GSK Biologicals [GlaxoSmithKline,
the manufacturer of those vaccines] or collaborators,” with one exception. Id.
Forty-two studies were included. Id. at 6632 (table 1). More than 36,000 people
received a vaccine and more than 30,000 people served as controls. Id. In regard
to the number of people, Dr. Rose stated that the Verstraeten article was
“the closest that we can be to an epidemiological study” because it studied “about
60,000 individuals . . . [and] covered two years of followup.” Tr. 232. Dr.
McCabe did not address this article.

Using a database, Dr. Verstraeten and colleagues looked for adverse events
following the vaccination using terms in the Medical Dictionary for Regulatory
Activities. Exhibit E at 6631. One of the terms was “juvenile arthritis,” which,

10
according to Dr. Rose, encompasses sJIA. Id. at 6633 (table 2); Tr. 287.10 The
authors’ general conclusion was their study “did not show evidence of an overall
increase in relative risks for autoimmune disorders in participants receiving
vaccines containing AS04 compared with controls.” Exhibit E at 6633.

B. Juvenile Idiopathic Arthritis

1. Basic Information11

The term “juvenile idiopathic arthritis” encompasses several different
diseases. The form affecting Vanessia is known as sJIA.12 The diagnostic criteria

10
Another term was “rheumatoid arthritis,” an autoimmune disease that Dr.
Rose stated does not encompass sJIA. Exhibit E at 6634 (table 3); Tr. 287; see
also Tr. 186, 197-98. Among the vaccine recipients, there were 12 cases of
rheumatoid arthritis. Among the controls, there were nine cases. Exhibit E at 6634
(table 3). The relative risk was 1.17 and the 95 percent confidence interval ranged
from 0.47 to 2.86. Id. at 6635 (table 4). When asked about this article, Dr.
McCabe explained that a relative risk of greater than one means that the risk is
increased and a relative risk of less than one means that the risk is decreased. Tr.
188.
11
Dr. McCabe, who is not a medical doctor, testified that he learned more
about sJIA by reading articles about the disease in the course of preparing his
expert report. Tr. 164; see also Tr. 65 (discussing exhibit 12), 75-80 (discussing
exhibit 13). Dr. Rose, who is a pediatric rheumatologist with experience in
treating sJIA, generally did not challenge the accuracy of information provided
about the disease. Thus, the source of information about sJIA is the set of articles
filed as exhibits as well as the testimony.
12
Other names for this same entity include Still’s disease, systemic arthritis,
systemic-onset juvenile rheumatoid arthritis, and systemic-onset juvenile chronic
arthritis. Exhibit C (Fabrizio De Benedetti & Rayfel Schneider, Chapter 14:
Systemic Juvenile Idiopathic Arthritis, in Textbook of Pediatric Rheumatology
(“Textbook”) (James T. Cassidy et al. eds., 6th ed. 2011)) at 236. Exhibit G
contains a photocopy of the cover of this textbook, the book’s publication
information, and the first page (page 236) of chapter 14. For ease, all citations to
this textbook will be made to exhibit C.

11
include: arthritis and a quotidian fever13 for at least two weeks, plus a rash,
lymphadenopathy, enlargement of the liver or spleen, or serositis. Exhibit C at 236
(relying upon the criteria set by the International League of Associations for
Rheumatology).

The disease manifests in different parts of the body. Characteristically more
than one joint is affected. During active inflammation, a person often experiences
muscle pain, a fever and rash. The disease also causes problems in the person’s
spleen and lymph nodes. Less common features include problems in the heart,
liver and (more rarely) the central nervous system. Exhibit C at 238-41.

“The acute manifestations of sJIA are variable in duration and last from
weeks to months.” Id. at 246. While approximately 40 percent of patients nearly
completely recover after one course of the disease, more than half of the people
afflicted “have a persistent disease course.” Id. In the United States, less than 0.5
percent of people with sJIA die from it. Id. at 247.

Treatments for sJIA include “medications to minimize joint inflammation.”
Id. at 244. Prednisone is recommended.14 Other drugs that have some
effectiveness include anti-tumor necrosis factor,15 anti-interleukin 6 receptors, anti-
interleukin 1, methotrexate,16 intravenous immunoglobulin, cyclosporine-A, and
thalidomide. Exhibit C at 244-46.

13
A quotidian fever is one that “recurs every day.” Dorland’s at 693. The
fever in sJIA is also sometimes referred to as a “hectic fever,” which also means
recurring each day. Id. at 692.
14
Prednisone is a medication against inflammation and suppresses the
immune system. Dorland’s at 1509.
15
An example of a pharmaceutical that inactivates tumor necrosis factor is
etanercept. Dorland’s at 650. Enbrel is a trademarked name for etanercept. Id. at
612.
16
Methotrexate is a “folic acid inhibitor” used for many conditions,
including “severe rheumatoid and psoriatic arthritis.” Dorland’s at 1151.

12
 Studies from Europe suggest that sJIA has an annual incidence of between
0.3 and 0.8 per 100,000 children less than 16 years of age.17 Although the onset
peaks among children 1-5 years old, adolescents and adults can also develop the
disease. Males and females are affected equally. Id. at 236.

2. Causes

The term name of the disease—systemic idiopathic juvenile arthritis—
provides information about what is known about the cause of the disease.
According to a medical dictionary, “idiopathic” means “of unknown cause or
spontaneous origin.” Dorland’s at 912. “Idiopathic” does not mean that there is no
cause. While the cause or causes of sJIA have not been found, “there is substantial
evidence of a dysregulated innate immune response with consequent increased
production of inflammatory cytokines.” Exhibit G at 237.18

Cytokines are “nonantibody proteins released by one cell population . . . on
contact with specific antigen, which act as intercellular mediators, as in the
generation of an immune response.” Dorland’s at 466. Cytokines are “the ways
we tell one cell to the other what to do.” Tr. 279 (Dr. Rose). Cytokines are “very
ubiquitous” and the cytokine response is “almost . . . universal.” Id. After a
person encounters an antigen, the immune system responds with the production of
cytokines within hours. Tr. 281-82 (Dr. Rose), 295, 300 (Dr. McCabe).

17
The incidence rate refers to the number of new cases in a population over
a period of time. See Dorland’s 1595.
18
An autoinflammatory disease differs from an autoimmune disease. See
Dorland’s at 181 (defining autoimmune and autoinflammatory). Autoimmune
diseases, about which special masters often hear testimony, are caused by
autoantibodies and autoreactive T cells. However, in sJIA, autoantibodies and
autoreactive T cells are not involved. Thus, sJIA is not an autoimmune disease.
Exhibit 13 (Elizabeth D. Mellins et al., Pathogenesis of systemic juvenile
idiopathic arthritis: some answers, more questions, 7 Nature Revs. Rheumatology
416 (2011)) at 417-18.
While rheumatologists such as Dr. Mellins, distinguish autoimmune diseases
from autoinflammatory diseases, Dr. Chao and Dr. Verstraeten (two
epidemiologists) did not maintain this precision. Although both articles discuss
“autoimmune diseases,” that phrase is broad enough to include sJIA. See section
II.A.2.c.

13
 Human beings produce a finite number of types of cytokines, with perhaps
as many as 40 different cytokines being identified so far. Tr. 290; see also Tr. 172
(Dr. McCabe stating he “accept[s] to a certain extent that there is a commonality in
immune effector functions”). Depending on the context, cytokines have different
purposes. Some cytokines promote inflammation while other cytokines are anti-
inflammatory. See Tr. 78. Dr. McCabe stated that ordinarily, pro-inflammatory
cytokines can act on multiple tissues and can lead to (a) increased vascular
permeability, (b) fever, and (c) increased synovial inflammation. Tr. 77-78; see
also exhibit 13 (Mellins) at 418-21, reproduced as exhibit 38 (PowerPoint slides) at
5.

The specific pro-inflammatory cytokines that have been implicated in the
development of sJIA include interleukin (“IL”) 1, IL-6, IL-7, IL-8, IL-18,
macrophage inhibitory factor, and tumor necrosis factor. Exhibit C (Textbook) at
237; Tr. 66 (Dr. McCabe), 280 (Dr. Rose). “Many features of sJIA seem to be
explained by the known effects of innate proinflammatory cytokines, IL-1 and IL-6
in particular.” Exhibit 13 at 418.

How any of these cytokines contribute to sJIA is unknown.19 As one
textbook stated, “[t]he role of each one of these mediators is far from being
clarified.” Exhibit C at 237. At the hearing, Dr. McCabe recognized that the
medical community did not understand what the cytokines were doing at the
cellular level. Tr. 299.

Even accepting the proposition that pro-inflammatory cytokines contribute
to the course of sJIA, this observation does not identify the causes of the disease
because something must initiate the increase in cytokines. Hence, one of the
articles cited by Dr. McCabe asks “What are the initial triggers of sJIA?” Exhibit
13 (Mellins) at 423. As Dr. Rose explained, medical researchers are generating

19
The production of pro-inflammatory cytokines does not always result in
disease. In fact, as Dr. McCabe and Dr. Rose recognize in their expert reports, the
production of pro-inflammatory cytokines is a protective response that vaccines are
designed to elicit. See exhibit 9 (Dr. McCabe) at 3 and exhibit A (Dr. Rose) at 6;
see also exhibit 26 (Pinto). The production of these pro-inflammatory cytokines,
however, is associated with diseases, including diseases other than sJIA, such as
sarcoidosis and systemic lupus erythematosus. Tr. 279.

14
hypotheses to explain the development of pro-inflammatory cytokines. See Tr.
217.

Dr. McCabe identified three articles in which the authors stated that
infections or vaccinations could be the initial cause for sJIA. Tr. 136, 142-43,145-
46. One article stated, “in juvenile idiopathic arthritis (JIA) a temporal relationship
between disease onset, childhood vaccination, remissions and flares hint[s] at a
possible relation of JIA disease activity and vaccinations or infections.” Exhibit 15
(Arash Ronaghy et al., Vaccination leads to an aberrant FOXP3 T-cell response in
non-remitting juvenile idiopathic arthritis, 70 Ann. Rheum. Dis. 2037 (2011)) at
120 (footnote deleted without notation). Another article asserted that “[o]ne
scenario is that infectious agents that are typically encountered in childhood initiate
sJIA; no single environmental trigger has been identified, although this lack of an
obvious candidate could point to multiple common agents being capable of
initiating sJIA.” Exhibit 13 (Mellins) at 417. A third article stated “[i]n juvenile
idiopathic arthritis, infections and vaccinations have been suggested as two
candidate triggers.” Exhibit 12 (Berent Prakken et al., Juvenile idiopathic arthritis,
377 Lancet 2138 (2011)) at 2141. This article continued, “but neither has been
confirmed as a trigger because of a scarcity of proper controlled, prospective
studies.” Id.

In the context of discussing vaccination as a possible trigger, Prakken cited
two articles that were filed into the record. Exhibit 12 (Prakken) at 2141 nn. 46,
47. In both studies, the people given the vaccine already suffered from juvenile
idiopathic arthritis and the researchers were examining whether the patient’s
disease worsened after the vaccination. One study involved the mumps, measles,
and rubella (“MMR”) vaccine. In a retrospective analysis of 207 patients with
juvenile idiopathic arthritis (of which 17 had systemic arthritis), the researchers
found “no changes in disease activity, flare occurrence or medication use after the
MMR vaccination.” Exhibit 43 (Marloes W. Heijstek et al., Safety of measles,
mumps and rubella vaccination in juvenile idiopathic arthritis, 66 Ann. Rheum.
Dis. 1384 (2007)) at 1386. Thus, the researchers concluded that the “MMR
vaccination appears to be safe in JIA.” Id.

20
This article, as submitted, does not have the same pagination as originally
published in the Annals of Rheumatic Diseases.

15
 The other study used the meningococcal C vaccination. The total number of
participants was 234, including 34 with systemic arthritis. The researchers “did
not detect any worsening of disease activity within 6 months after MenC
vaccination.” Exhibit 47 (Evelien Zonneveld-Huijssoon et al., Safety and Efficacy
of Meningococcal C Vaccination in Juvenile Idiopathic Arthritis, 56 Arthritis &
Rheumatism 639 (2007)) at 644.

The parties did not submit any case reports linking the Gardasil vaccine and
JIA.

C. Vanessia’s Medical History before and after her sJIA Diagnosis21

Vanessia was born in February 1995. She was generally healthy for the first
12 years of her life. In February 2008, Vanessia saw her regular doctor, Dr. Elena
R. Regala for a routine check-up. Dr. Regala noted Vanessia’s history of asthma.
Exhibit 3 at 11. Dr. Regala’s office administered the first dose of the HPV vaccine
to Vanessia on this date. Id.; exhibit 2 at 3. Vanessia received the second dose of
the HPV vaccine on April 18, 2008. Exhibit 2 at 3; exhibit 3 at 8.

On approximately June 21, 2008, Vanessia developed a rash all over her
body. She reported to Dr. Regala on June 24, 2008, that she had this rash for “3
days.” Exhibit 3 at 8. Dr. Regala suspected an allergic reaction and prescribed
Benadryl and prednisone. Id. The rash disappeared in three days. Id. at 26 (notes
dated July 2, 2008).

Vanessia stopped taking the prednisone and, on June 27, 2008, she
developed pain in many places including her knees, thighs, and calves. Exhibit 4
at 14.22 Dr. Regala’s impression included juvenile rheumatoid arthritis. Exhibit 3
at 27.

On June 28, 2008, Vanessia was admitted to Marian Medical Center for
“high fever accompanied by severe joint pains of the knees and ankles,” which

21
The parties accept the accuracy of the medical records.
22
Given that Dr. Regala prescribed prednisone on June 24, 2008, and
Vanessia reported, on June 27, 2008, that she had stopped taking prednisone, it
appears that Vanessia actually took prednisone for fewer than three days.

16
started on June 25, 2008. Exhibit 3 at 26. While in the hospital, various laboratory
tests were done. Exhibit 3 at 12-21; exhibit 4 at 6. Dr. Frank Scott, a
rheumatologist, saw Vanessia. Dr. Scott’s impression was that she had “probable
Still’s disease (systemic onset juvenile arthritis).” Exhibit 4 at 11-12. Vanessia
was prescribed prednisone. By the day on which she was discharged (July 2,
2008), she had started to feel better, no longer had a fever, and was not suffering
from joint pains. However, she still had a rash. When she left the hospital, her
presumptive discharge diagnosis was JIA. Exhibit 4 at 6. At discharge, Dr. Regala
referred Vanessia to a pediatric rheumatologist. Exhibit 3 at 11.

On July 8, 2008, Vanessia saw Dr. Deborah McCurdy, a pediatric
rheumatologist at the University of California at Los Angeles Health System. Dr.
McCurdy recorded that Vanessia’s vaccinations were up-to-date, including a
second dose of the HPV vaccine. Dr. McCurdy also noted that Vanessia’s family
history included JIA. Exhibit 5 at 51. Dr. McCurdy stated that Vanessia’s
symptoms made “sJIA very likely.” Id. at 55. Dr. McCurdy continued the
prescriptions for prednisone and was waiting for the results of pending laboratory
studies to add methotrexate and Enbrel. Id. Dr. McCurdy sent a letter
summarizing her findings to Dr. Regala on July 8, 2008. Exhibit 5 at 20-26. Dr.
McCurdy’s letter mentioned that Vanessia had “just received the second of three
HPV vaccines.” Id. at 21.

Vanessia saw Dr. Regala again on August 19, 2008. Exhibit 3 at 6. Dr.
Regala knew that Vanessia was suffering from JIA from the previous
correspondence with Dr. McCurdy. See exhibit 5 at 20, 24 (Dr. McCurdy’s letter
to Dr. Regala dated July 8, 2008). Dr. Regala administered the third dose of the
HPV vaccine to Vanessia on August 19, 2008. Exhibit 3 at 6; see also exhibit 2
at 3.

On August 27, 2008, a physical therapist associated with a local public
health department, Sylvia Medinger, saw Vanessia in response to a referral from
Dr. McCurdy. In her history, Ms. Medinger recorded that Vanessia’s dose of
prednisone had ended on August 18, 2008. Vanessia was still receiving Enbrel.
On August 25, 2008, Vanessia had a “flare-up . . . with fever, rash and increase in
pain.” Ms. Medinger evaluated Vanessia and recommended that she have physical
therapy twice a week. Exhibit 8 at 48-50.

Vanessia returned to Dr. McCurdy on September 3, 2008. Vanessia
recounted that she was having some symptoms after stopping prednisone. Dr.
McCurdy recorded that Vanessia had “some improvement with Enbrel.” Vanessia
17
was also taking methotrexate. Dr. McCurdy examined Vanessia and found that she
had swollen knees and ankles. Dr. McCurdy’s impression was that she was
“improved but still [had evidence of] active [disease]” and was “better.” Exhibit 5
at 45-46.

Dr. McCurdy continued to care for Vanessia and follow-up appointments
were held in December 2008, 2009 (two appointments), and 2010. At these visits,
Vanessia, despite her JIA, was generally “doing well.” The doctors recommended
that she receive the influenza vaccine and H1N1 vaccine. Exhibit 5 at 32, 41, 44,
60.

Another follow-up appointment occurred on January 12, 2011, at UCLA.
This time, Vanessia saw Dr. Alice Hoftman, another pediatric rheumatologist. Dr.
Hoftman’s record states that Vanessia was “currently pursuing lawsuit against
Gardasil. [Received] Gardasil #2, 4/08. [Diagnosed] 7/08.” Exhibit 5 at 27.
During this visit, Dr. Hoftman apparently recommended that Vanessia receive the
flu vaccine. Despite having previously accepted the doctor recommendation that
Vanessia receive a flu vaccine in 2008-2010, Ms. Koehn refused at this visit. See
exhibit 5 at 28, 32 (H1N1 vaccine), 44, 60. Regarding her refusal, Dr. Hoftman
wrote: “[patient’s] mother refused flu vaccine this year. Discussed [with] mom
the importance of this vaccine. Mom hesitant [because] Gardasil. [Discussed
with] mom – no data but all vaccines and infections can trigger autoimmune
response.” Id. at 28.

D. Experts’ Qualifications, Reports and Testimony

1. Petitioner’s Expert, Michael J. McCabe, Ph.D.

a) Qualifications

Dr. McCabe earned a Ph.D. in microbiology and immunology from Albany
Medical College in 1991. He worked at the Karolinska Institute in Stockholm,
Sweden as a postdoctoral research associate from 1990 to 1992. In 1992, he joined
the faculty of Wayne State University as a research assistant professor at the
Institute of Chemical Toxicology. His research explored how chemicals, metals
and other contaminants from the environment affect the immune response. He also
held various other positions at Wayne State University until 2000. Exhibit 10
(curriculum vitae); Tr. 12-14, 50-53.

18
 From 2000 to 2009, Dr. McCabe worked, first as an assistant professor and
then as an associate professor, in the Department of Environmental Medicine at the
University of Rochester School of Medicine and Dentistry. Dr. McCabe’s duties
included research, a small amount of teaching, and administration. While
supervising approximately 25 scientists “working on lab-based and
epidemiological research projects,” Dr. McCabe’s research focused on
“mechanistic metal toxicology and immunotoxicology.” Exhibit 10; see also Tr.
15-17, 35-37 (detailing teaching responsibilities).

In 2009, Dr. McCabe started working at Robson Forensic, Inc. as an
associate. In that capacity, Dr. McCabe provides “reports and testimony toward
the resolution of . . . personal injury litigation of toxicology and human health
assessments involving environmental and occupational exposures to agents such as
metals.” Exhibit 10; see also Tr. 33-34.

Dr. McCabe has written about 40 articles that appear in peer-reviewed
publications and about 12 book chapters. Most, but not all, of Dr. McCabe’s
publications relate to the toxicity of metals. Tr. 15, 37-39.

Dr. McCabe has contributed to select committees exploring causation. For
example, Dr. McCabe participated on a National Academy of Science committee
exploring beryllium alloy exposure. He reviewed proposals about Gulf War
injuries for the Department of Defense. He was a co-author of a white paper about
the role of the environment in developing autoimmune diseases for the National
Institute of Environmental Health Sciences. Exhibit 10; Tr. 22-29.

In response to questions asked by the Secretary’s counsel during voir dire,
Dr. McCabe stated that he is not a medical doctor and does not treat patients. Tr.
33. He has not researched sJIA. Tr. 41. However, Dr. McCabe has been involved
in a small pilot study, examining how “lead-intoxicated girls” responded to
Gardasil. Tr. 42.

His current position at Robson Forensics, Inc. requires him to “review legal
cases, produce reports, and testify as needed.” Tr. 33. Dr. McCabe estimated that
activities related to litigation provide more than 95 percent of his income with most
of his work for plaintiffs. Tr. 33-34.

Ms. Koehn offered Dr. McCabe as an expert in the field of immunology to
which the Secretary did not interpose an objection. Tr. 31, 50. Dr. McCabe was
recognized as an expert in immunology. Tr. 53.
19
 b) Report23

Dr. McCabe’s report begins with a review of Vanessia’s medical history.
Dr. McCabe’s recitation is consistent with the information presented above.

Dr. McCabe describes “juvenile rheumatoid arthritis.”24 He emphasizes that
this disease is an autoinflammatory process “driven by dysregulation of the innate
immune system as evidenced by a role for pro-inflammatory cytokines (e.g. IL-6,
IL-1 and TNF-α).” Exhibit 27 at 2. He states, “[m]uch as the same with most
human autoimmune diseases, the cause of Juvenile Rheumatoid Arthritis is thought
to be multifactorial – with genetic susceptibility factors and environmental triggers
working together in complex ways to initiate and perpetuate adaptive and innate
immune activities resulting in tissue damage.” Id. at 2-3. “[T]he basis for the
argument for a causative role for these environmental triggers [referring to
infections and vaccinations] comes from mechanistic considerations.” Id. at 3.

Dr. McCabe also describes the HPV vaccine. Citing an article by Pinto, Dr.
McCabe asserts that “[i]n individuals immunized with [HPV vaccines], high levels
of both adaptive and innate immune cytokines are produced.” Id. at 3. “Notably,
many of these same vaccine-elicited cytokines are the pro-inflammatory cytokines
that have been implicated in the etiology of JRA.” Id. As made clear during the
hearing, this is the essence of Dr. McCabe’s theory: an HPV vaccine elicits a
certain cytokine pattern (particularly IL-6) and these cytokines cause sJIA. Tr.
123.

Dr. McCabe’s report also elaborates on the topic of the temporal interval that
is medically appropriate for causation. Dr. McCabe cites studies that showed that
within seven months of Gardasil vaccination, more than 99 percent of people have
seroconverted. Exhibit 27 at 4-6. This discussion implies that it was appropriate
to infer that development of a disease within seven months of a vaccination was
caused by the vaccination.

23
Dr. McCabe’s supplemental report encompasses his original report.
Therefore, citations will be only to the report dated October 1, 2011 (Exhibit 27).
24
Dr. Rose pointed out that “juvenile rheumatoid arthritis” is not the
currently preferred term. Exhibit A at 4-5.

20
 c) Testimony25

After presenting his qualifications, Dr. McCabe discussed Gardasil. Tr. 54-
55. He summarized Vanessia’s medical history, Tr. 55-61, and his synopsis is in
accord with the findings of fact set forth above. Dr. McCabe premised his opinion
on Vanessia’s diagnosis of sJIA. Tr. 60.26

Dr. McCabe’s next topic was explaining how Gardasil can cause sJIA. Dr.
McCabe began by explaining a prevailing theory of how sJIA originates. As
mentioned above in section II.B., sJIA is mediated by pro-inflammatory cytokines,
such as TNF, interleukin 1, interleukin 6, and interleukin 18. The role of these pro-
inflammatory cytokines leads to a classification of sJIA as an autoinflammatory
disease. Tr. 65-66. According to Dr. McCabe, when a person with a genetic
susceptibility encounters an environmental trigger, the person’s innate immune
system falls out of balance. The result of this imbalance, for some people, is sJIA.
Tr. 66-69, 92-93.

Dr. McCabe testified about the Bradford Hill criteria for causation.27 In Dr.
McCabe’s view, several of these criteria supported finding that Gardasil can cause
sJIA. Supporting criteria include the temporal sequence, the dose-response
relationship, and biological plausibility. Tr. 97-99. Another factor, experimental
evidence, was the springboard into a lengthy discussion about how human beings
respond to a vaccine against some types of human papillomavirus.

25
This section of the decision and the section on Dr. Rose’s testimony
summarize pertinent portions of their testimony without necessarily discussing
each page of the transcript. However, the entire transcript has been reviewed.
26
If Dr. McCabe had disagreed with the diagnosis from Vanessia’s treating
doctors, his testimony about an alternative diagnosis might have been problematic
because Dr. McCabe is not a medical doctor.
27
After the hearing, Ms. Koehn filed the article in which the Bradford Hill
criteria appear. Exhibit 48 (Sir Austin Bradford Hill, The Environment and
Disease: Association or Causation?, 7 Proc. of the Royal Society of Medicine 295
(1965)).

21
 Dr. McCabe spoke extensively about a 2005 article written by Dr. Pinto and
colleagues. Tr. 100-04; see Exhibit 26. Dr. McCabe interpreted this study as
showing that a vaccine against a particular strand of human papillomavirus caused
the production of various pro-inflammatory cytokines. Tr. 103-04, 110-11.

The discussion about the 2005 Pinto article flowed into testimony about a
more recent article in which Dr. Pinto appears as the senior author. Exhibit 28
(García-Piñeres). Again, the authors used a vaccine against one strand, type 16, of
the human papillomavirus. This study also showed that various cytokines
increased after the administration of a vaccine against human papillomavirus. Tr.
117-19.

Dr. McCabe summarized his opinion why Gardasil can cause sJIA. His
opinion is based, in part, upon “the scientific and medical literature that implicates
proinflammatory cytokines and inflammatory responses and innate immunity in the
pathogenesis of systemic juvenile arthritis.” His opinion is also based, in part,
upon the “scientific and medical literature that demonstrates that HPV vaccine is a
strong and potent immunogen that stimulates the production of these same
proinflammatory cytokines.” Tr. 123.

At this point, Dr. McCabe moved to explain why Gardasil caused Vanessia’s
sJIA. Dr. McCabe saw evidence that Vanessia was generating pro-inflammatory
cytokines in her clinical presentation, including a fever, rash and joint pain. Tr.
123. Dr. McCabe also maintained that when Vanessia was given medications
intended to reduce pro-inflammatory cytokines, such as Enbrel, methotrexate, and
prednisone, her disease improved. Tr. 124-25. Dr. McCabe also suggested that
Vanessia’s sJIA worsened after she received the third dose of Gardasil on August
19, 2008. Dr. McCabe, however, cautioned that when Vanessia received this dose,
she was on anti-inflammatory therapies. Thus, whether the third dose of Gardasil
caused Vanessia’s sJIA to flare was “not necessarily clear.” Tr. 126.

The next topic of Dr. McCabe’s direct testimony was the medically
appropriate interval between vaccination and the onset of symptoms. Dr. McCabe
stated that any adverse consequence of the vaccination is likely to arise in “the
time period that measurable changes in the immune response are known to be
elicited by the vaccine.” Tr. 128. Relying upon various studies, Dr. McCabe
stated, by reference, that the medically appropriate immune response range would
extend to approximately seven months after the vaccination. Tr. 127-29; see also
exhibit 25 at S13.

22
 Dr. McCabe’s final topic was to address a study by Chun Chao and others.
Exhibit 34. Despite involving approximately 189,000 people, Dr. McCabe
asserted that the size of the study was not sufficiently large to detect any increase
in the number of cases involving sJIA because sJIA is a rare disease. Tr. 133-34.
Therefore, Dr. McCabe agrees with Berent Prakken, the author of another article
on juvenile idiopathic arthritis, who recommended that “much larger studies . . .
will be needed to define the role of environmental triggers in JIA.” Tr. 136
(quoting exhibit 12 at 4).

For all these reasons, Dr. McCabe concluded, to a reasonable degree of
scientific certainty, that the first two doses of the Gardasil vaccine caused Vanessia
to develop sJIA. 28 Tr. 136-38.

On cross-examination, Dr. McCabe acknowledged that the Prakken article
states “‘Infections and vaccinations have been suggested as two candidate triggers,
but neither has been confirmed because of the scarcity of proper control
perspective studies.’” Tr. 140 (quoting exhibit 12 at 2141). The studies that
looked for a connection between vaccination and juvenile idiopathic arthritis
concerned the meningococcal vaccine and the MMR vaccine. Id.

Dr. McCabe stated that clinicians and basic researchers have been
investigating the causes of sJIA for a long time. But, they have not identified the
cause because it is a “multifactorial disease.” Tr. 143-44. In this regard, Dr.
McCabe stated that there is “no epidemiology that’s meaningful enough to inform
us” as to whether the HPV vaccine causes sJIA. Tr. 141-42. Dr. McCabe also
acknowledged that he had not located any case reports describing an association
between HPV vaccine and sJIA. Id. Dr. McCabe is not aware of anyone
conducting a case control study of whether HPV vaccine causes sJIA. Tr. 147.

Counsel for the Secretary probed Dr. McCabe’s reliance on medical articles.
For example, counsel noted the 2005 Pinto article does not mention any type of
arthritis, including sJIA, does not propose any theory to connect an HPV vaccine to
sJIA, and does not report that anyone who received the vaccination developed any

28
Ms. Koehn’s counsel asked Dr. McCabe if he held his opinions “to a
reasonable degree of scientific certainty,” and Dr. McCabe answered affirmatively.
Tr. 137. Dr. McCabe could have testified if he held his opinions only to a
reasonable degree of scientific probability.

23
symptoms. Tr. 147-48. For the last point, Dr. McCabe pointed out that because
the study did not report any consequence, it is impossible to know whether any test
subjects experienced any adverse events. Tr. 148-49. Government counsel and
Dr. McCabe reviewed similar limitations to other articles, including the García-
Piñares article. Tr. 149-50.

In regard to the Chao article, Dr. McCabe stated that the researchers
considered juvenile rheumatoid arthritis. Tr. 155 (citing exhibit 34 (Chao) at 197).
The abstract of this article states “No autoimmune safety signal was found in
women vaccinated with HPV-4.”29 Tr. 156 (quoting exhibit 34 at 193).

In Vanessia’s medical history, none of her treating doctors expressed any
opinion as to whether the HPV vaccine caused her to develop sJIA. Dr. McCabe
did not see any indication that Vanessia’s treating doctors were concerned about
giving her the third dose of Gardasil after she had developed sJIA. Tr. 156-57. Dr.
McCabe recognized that it appears that shortly before the third dose of Gardasil,
Vanessia had stopped taking prednisone but was improving on Enbrel. Tr. 158
(citing exhibit 5 at 45-46).

Dr. McCabe stated that hypothetically, if Vanessia had not received Gardasil
and still developed sJIA, then he would not know what caused her to develop the
disease. Tr. 160. His opinion that Gardasil caused Vanessia to develop sJIA is
based, in part, upon the timing and also upon the immunobiology of what is known
about sJIA. Tr. 161.

Dr. McCabe also answered questions the undersigned asked. Dr. McCabe
stated that after Ms. Koehn’s counsel first contacted him, there was a hypothesis
that Gardasil caused Vanessia’s sJIA. He investigated whether “there was a
tenable scientific argument” by conducting a scientific undertaking. He learned
about sJIA. Tr. 164. He also looked for data to support the proposition that
Gardasil causes an increase in pro-inflammatory cytokines. He also drew upon his
experience and background. Tr. 164-65.

Dr. McCabe testified that the limited number of cytokines does not detract
from his opinion that Gardasil causes a production of pro-inflammatory cytokines
and these cytokines caused Vanessia’s sJIA. He stated that although a tool box

29
Gardasil is the vaccine against four strands of the human papillomavirus.

24
may have many tools, it is likely that a hammer was used to pound a nail. Tr. 172-
73.

Dr. McCabe recognized that some of the Bradford Hill criteria do not
support a finding of causation. For example, the “strength of association” is not
supportive. In addition, the criteria of “analogy” would either be not supportive or
not relevant. The studies involving juvenile idiopathic arthritis and either the
MMR vaccine or the meningococcal vaccine do not link the disease with the
vaccine. If the MMR vaccine and/or the meningococcal vaccine were analogous to
Gardasil, then those studies would counter the hypothesis that Gardasil can cause
sJIA. However, Dr. McCabe suggested that the MMR vaccine and the
meningococcal vaccine differed from Gardasil. See Tr. 179-83.

Additionally, Dr. McCabe agreed that Gardasil is not the only cause of sJIA.
This fact is easily recognized because sJIA existed before Gardasil. Tr. 173-74,
190-91.

In regard to Vanessia’s case specifically, the undersigned asked how Dr.
McCabe could distinguish a case of sJIA caused by Gardasil from a case of sJIA
caused by something else. Dr. McCabe’s response was relatively weak. Although
not phrased in these terms, he essentially stated that among all the things to which
Vanessia was exposed, the only possible cause for sJIA was Gardasil. See Tr. 191-
94.

After a short redirect examination, Dr. McCabe confirmed that his opinion
remained unchanged. He stated that Gardasil caused Vanessia’s sJIA. Tr. 195-97.

2. Respondent’s Expert, Carlos Rose, M.D.

a) Qualifications

Dr. Rose graduated from Argentina’s University of Buenos Aires in 1977.
He passed his boards for rheumatology while still in Argentina in 1983. By 1987,
Dr. Rose was living in the United States, participating in an internship in pediatrics
at the Medical Center of Delaware. He had successive fellowships in pediatric
rheumatology, first at the Children’s Hospital of Philadelphia and then at Alfred I.
duPont Institute in Delaware. He has held a board certification in pediatrics with a
specialty in rheumatology since 1998. Dr. Rose estimated that there are 216
pediatric rheumatologists in the United States. Exhibit B (curriculum vitae) at 1-5;
Tr. 199-200.
25
 He has worked as a staff physician in pediatric rheumatology at the Alfred I.
duPont Institute since 1991. In 1994, he became chief of the rheumatology
division. He has taught pediatrics at the Jefferson Medical College of Thomas
Jefferson University since 1991, and he became a full professor at that school in
2002. Exhibit B at 6-7.

He has served on international committees and lectured to audiences in
foreign countries. Dr. Rose has written more than 70 peer-reviewed articles. He
also has written more than 25 book chapters or monographs. Some publications
relate to juvenile idiopathic arthritis, but not specifically to sJIA. Exhibit B at 13-
20; Tr. 202-03.

As part of voir dire, Ms. Koehn’s counsel elicited the following points about
Dr. Rose’s qualifications. He is not an immunologist and he has not researched the
role of pro-inflammatory cytokines in causing sJIA. He has not done any research
on any human papillomavirus vaccine, including Gardasil. Tr. 204.

Dr. Rose stated that he has worked for the Department of Health and Human
Services in the Vaccine Program for 21 years. Over that span, Dr. Rose estimated
that he has reviewed approximately 60 cases. He recommended compensation in
one case. Tr. 205-07.

The Secretary offered Dr. Rose as an expert in the field of pediatric
rheumatology. After Ms. Koehn did not object, he was recognized in that field.
Tr. 207.

b) Reports

In Dr. Rose’s first report, he begins with a summary of Vanessia’s medical
history. Dr. Rose agrees that she suffers from sJIA. Exhibit A at 1.

He states that sJIA is an “auto-inflammatory disease[] likely associated with
dis-regulation [sic] of cytokine networks likely IL-1 and IL-6 networks rather than
the adaptive immune system.” For this particular form of arthritis, Dr. Rose states
that he is not familiar with any infections being associated with sJIA, although
some infections have been associated with “transient self-limiting arthritis.” In
regard to the human papillomavirus, Dr. Rose states that that organism does not
produce any arthritis and “no syndrome even remotely reminiscent of sJIA is seen
in association with the infection.” Id. at 2.
26
 Dr. Rose rejects the idea that HPV vaccine can cause sJIA. Relying on a
study that integrated many clinical trials, Dr. Rose states there was “no statistically
significant difference in the event rates between vaccine and control groups.” Id.
at 3, 8 (citing exhibit E (Verstraeten)). Dr. Rose also reviewed the literature that
Dr. McCabe cited.

Dr. Rose’s first report concludes that “more likely than not Vanessia’s
disease emerged as the result of chance and it was not causally related to the
immunizations she received.” Id. at 7. Dr. Rose’s supplemental report makes a
similar point: “the temporal association between vaccine and disease onset is
coincidental.” Exhibit F at 1.

The supplemental report presents Dr. Rose’s opinion regarding Dr.
McCabe’s cytokine theory. Dr. Rose asserts that “[t]he cytokine response is
complex and cytokines that in certain circumstances are pro-inflammatory, in
others are anti-inflammatory, depending on the combination of signals, the tissue
in question and even perhaps the age of the individual.” He maintains that “[t]he
fact that similar cytokines are found in serum of sJIA patients and in vaccine
response is more a reflection of the somewhat limited and stereotypical
inflammatory response repertoire in mammals than a suggestion for a link
[between] vaccine [and] sJIA.” Id.

c) Testimony

After Dr. Rose was accepted as an expert in pediatric rheumatology, he
summarized the material that he reviewed in this case. He offered his opinion that
the Gardasil vaccinations were not related to Vanessia’s development of sJIA. Tr.
208.

Dr. Rose agreed that Vanessia suffers from sJIA. SJIA, as set forth above, is
not an autoimmune condition. It is an autoinflammatory condition. Dr. Rose
explained that the treatment for sJIA is different medications, including
corticosteroids (methotrexate). The purpose of some drugs is to inhibit cytokines
such as interleukin 1, interleukin 6, and TNF. The way Vanessia’s doctors treated
her was typical. Tr. 210-14.

Dr. Rose elaborated on cytokines and sJIA. He stated that interleukin 6 may
be a cause of sJIA. Even if it is not a cause, interleukin 6 influences the course of
the disease. For example, peaks in interleukin 6 preceding a rise in temperature
27
and a hectic fever are a defining characteristic of sJIA. Interleukin 1 has also been
linked causally to arthritis. Dr. Rose stated that his experience as a clinician who
has seen some (but not all) patients with sJIA improve after taking drugs that
control interleukin 1 and interleukin 6 informs his belief that these cytokines play a
role in the disease. Tr. 215-17.

Dr. Rose disputed the relevance of Dr. McCabe’s citation to the Prakken
(exhibit 12), Mellins (exhibit 13), and Rongahy (exhibit 15) articles. According to
Dr. Rose, although these articles mention that infections could cause juvenile
idiopathic arthritis, the articles were merely generating hypotheses that could be
tested. To Dr. Rose, the articles did not report any evidence supporting Dr.
McCabe’s theory. See Tr. 217-18. According to Dr. Rose, pediatric
rheumatologists are not discussing whether an HPV vaccine can cause sJIA.
Instead, pediatric rheumatologists are discussing how safe the vaccines are for
patients with sJIA. Tr. 219.

Dr. Rose testified that one way to look at the safety of vaccines is to give
vaccines to people who have a disease and to see what happens. For juvenile
idiopathic arthritis, there were studies involving the MMR vaccine and the
meningococcal vaccine. Those studies showed that the MMR vaccine and the
meningococcal vaccine did not affect people with juvenile idiopathic arthritis
adversely. Tr. 221-23.

Dr. Rose is not aware of any epidemiological data connecting the HPV
vaccine and sJIA. He also has not seen any case reports on this topic. Tr. 220.

Dr. Rose discussed some of the articles on which Dr. McCabe relied. For
the Pinto article (exhibit 26), Dr. Rose examined whether cytokines remained
elevated. Constancy in elevation was important to Dr. Rose because, as a clinician,
he sees patients with a pattern of continually elevated cytokines. When Dr. Rose
stops medications that inhibit the production of cytokines, the patients flare. But,
when Dr. Rose looked at the data presented in the Pinto article, he did not see
much difference in the amount of cytokines produced at zero months, two months,
and seven months. To Dr. Rose, the Pinto data is “very suggestive that the
response that this vaccine elicited in these normal people has not been sustained,”
and thus the vaccine-elicited cytokine response differs from the sustained pattern
of “upregulation” he sees in his patients with sJIA. Tr. 223-26.

According to Dr. Rose, the García-Piñares article from 2007 (exhibit 28) in
which Dr. Pinto appears as the senior author makes the same point. These
28
experiments showed that a vaccine can stimulate the production of cytokines when
administered. But, in Dr. Rose’s view, these experiments do not show how a
single incidence of cytokine production can cause a disease. See Tr. 227-28.

Dr. Rose also discussed the Verstraeten (exhibit E) article, which he cited in
his expert report. Exhibit A at 3, 8. Dr. Rose stated that “this is the closest that we
can get to an epidemiologic study. This is a study of about 60,000 individuals.”
Tr. 232. Dr. Rose stated that if the vaccine were “a significant trigger[,] I would
expect to see one or two cases of sJIA in the vaccinees.” Id. However, Verstraeten
and his colleagues found “no evidence of an overall increase in relative risks for
autoimmune disorders in participants receiving vaccines containing AS04
compared with controls.” Exhibit E at 6633.

In regard to Vanessia, Dr. Rose said that he did not see any evidence that her
treating doctors believed that the HPV vaccination caused her sJIA. Dr. Rose said
that according to the standard of practice, even after Vanessia was diagnosed as
having sJIA following the second dose of Gardasil, she still should receive the
third dose. Dr. Rose said that he recommends that his patients with sJIA receive
the HPV vaccine. For Vanessia, although she had a rash and increased joint pain
after the third dose of Gardasil, Dr. Rose stated that this flare was due to the
discontinuation of corticosteroids. Dr. Rose said that this pattern of stopping the
medication and worsening of the condition happened earlier. Tr. 232-34.

On cross-examination, Ms. Koehn’s counsel elicited the following testimony
from Dr. Rose. The Verstraeten article did not involve Gardasil and involved a
vaccine that had a different adjuvant from the adjuvant in Gardasil. There was
some question about whether the Verstraeten article was looking for cases of sJIA.
Tr. 240-44.

Dr. Rose stated that the incidence (new cases per year) of sJIA is between
0.3 and 0.8 per 100,000 people. Tr. 244. This fact was the starting point for a
discussion between counsel and Dr. Rose about how large a population sample
would be needed to power an epidemiological study involving sJIA. This colloquy
did not provide especially helpful testimony in that Dr. Rose said, “I really need a
calculator or somebody to help me to really calculate it. . . . So maybe you need
100,000. I don’t really know the answer. . . . At least you need 100,000.” Tr.
245-46.

Dr. Rose stated that in the Chao article, the researchers were looking for
cases of “juvenile rheumatoid arthritis.” Tr. 248 (discussing exhibit 34 at 194).
29
Dr. Rose maintained that, although the article was published in 2011, the
researchers were using an out-of-date term. Dr. Rose believed that the term
“juvenile rheumatoid arthritis” would capture cases of sJIA. Tr. 248-52.

Ms. Koehn’s counsel also inquired about the Gardasil package insert, which
Dr. Rose had cited in his expert report. Ms. Koehn’s counsel raised two issues.
First, whether the phase 3 or phase 4 clinical trials would have identified cases of
sJIA that followed the administration of Gardasil. Dr. Rose said that if there were
any cases, then they would have been reported. Second, whether the number of
subjects in the clinical trials would have detected any changes in the incidence of
sJIA. Dr. Rose stated that although there were about 10,000 vaccinees and a
similar number of controls, if the vaccine caused sJIA, there would be some cases
reported. Tr. 252-60.

d) Post-Hearing Report

Due to questions about the scope of the Chao research that arose during the
hearing, Dr. Rose was permitted to file a brief supplemental report. It explained the
process of identifying diseases in women who had received Gardasil. Exhibit F.

III. Standards for Adjudication

To receive compensation under the Program, Ms. Koehn must prove either:
(1) that Vanessia suffered a “Table Injury”—i.e., an injury listed on the Vaccine
Injury Table—corresponding to a vaccine that she received, or (2) that Vanessia
suffered an injury that was actually caused by Gardasil. See 42 U.S.C. §§ 300aa-
13(a)(1)(A) and 11(c)(1); Capizzano v. Sec’y of Health & Human Servs., 440
F.3d 1317, 1319-20 (Fed. Cir. 2006). Here, Ms. Koehn is not claiming an injury
listed on the Vaccine Table. Therefore, she must prove causation-in-fact.

When a petitioner proceeds on a causation-in-fact theory, a petitioner must
establish three elements. The petitioner=s

burden is to show by preponderant evidence that the vaccination
brought about [the] injury by providing: (1) a medical theory causally
connecting the vaccination and the injury; (2) a logical sequence of
cause and effect showing that the vaccination was the reason for the
injury; and (3) a showing of a proximate temporal relationship
between vaccination and injury.

30
Althen, 418 F.3d at 1278.

In this passage, Althen indicates that petitioner’s burden of proof is a
preponderance of the evidence. Accord 42 U.S.C. § 300aaB13(a)(1). The
preponderance of the evidence standard, in turn, has been interpreted to mean that
a fact is more likely than not. Moberly v. Sec’y of Health & Human Servs., 592
F.3d 1315, 1322 n.2 (Fed. Cir. 2010). Proof of medical certainty is not required.
Bunting v. Sec’y of Health & Human Servs., 931 F.2d 867, 873 (Fed. Cir. 1991).

Distinguishing between “preponderant evidence” and “medical certainty” is
important because a special master should not impose an evidentiary burden that
is too high. Andreu v. Sec’y of Health & Human Servs., 569 F.3d 1367, 1379-80
(Fed. Cir. 2009) (reversing judgment that petitioners were not entitled to
compensation); see also Lampe v. Sec’y of Health & Human Servs., 219 F.3d
1357 (2000); Hodges v. Sec’y of Health & Human Servs., 9 F.3d 958, 961 (Fed.
Cir. 1993) (disagreeing with dissenting judge’s contention that the special master
confused preponderance of the evidence with medical certainty). In this regard,
“close calls regarding causation are resolved in favor of injured claimants.”
Althen, 418 F.3d at 1280.

Ms. Koehn argues she has provided preponderant evidence to meet her
burden under Althen to prove Vanessia’s sJIA was caused in fact by her Gardasil
vaccinations. An evaluation of each prong follows.

IV. Prong One from Althen – Medical Theory

The starting point for analysis is the theory proposed by the expert that
“causally connect[s] the vaccination and the injury.” Althen, 418 F.2d at 1278.
This element of petitioner’s case is sometimes referred to as answering the “can
it” question. Pafford v. Sec’y of Health & Human Servs., No. 01-165V, 2004 WL
1717359, at *4, 9 (Fed. Cl. Spec. Mstr. July 16, 2004), mot. for review denied, 64
Fed. Cl. 19 (2005), aff’d, 451 F.3d 1352 (Fed. Cir. 2006).

To explain how Gardasil harmed Vanessia, Ms. Koehn presents a theory
dependent upon relatively complex medical knowledge. Special masters have
been instructed in how to evaluate this type of evidence. See section IV.A below.
The evidence is analyzed in section IV.B.

31
 A. Considerations of Scientific and Medical Evidence

As Congress authorized, the judges of the Court of Federal Claims have
collectively issued the Vaccine Rules. 42 U.S.C. § 300aaB12(d)(2). The Vaccine
Rules, in turn, provide that the special master “must consider all relevant and
reliable evidence governed by principles of fundamental fairness to both parties.”
Vaccine Rule 8(b)(1); see Cedillo v. Sec’y of Health & Human Servs., 617 F.3d
1328, 1339 (Fed. Cir. 2010) (stating “Vaccine Rule 8(b)(1) necessarily
contemplates an inquiry into the soundness of scientific evidence to be considered
by special masters”).

The reliability of expert testimony is a topic on which the Federal Circuit has
guided special masters. The leading case is Terran v. Sec’y of Health & Human
Servs., 195 F.3d 1302 (Fed. Cir. 1999). In Terran, the special master “examined”
the expert’s opinion “in the light of the four guideposts enumerated in Daubert,”
and “conclude[d] that petitioner’s theory of causation is not based on reliable
scientific evidence.” Terran v. Sec’y of Health & Human Servs., No. 95-451V,
1998 WL 55290, at *11 (Fed. Cl. Spec. Mstr. Jan. 23, 1998) (citing Daubert v.
Merrell Dow Pharma., Inc., 509 U.S. 579 (1993)). When Ms. Terran’s appeal
reached the Federal Circuit, she argued that “the Special Master improperly
applied the Daubert factors to the expert’s testimony.” The Federal Circuit
rejected this argument and indicated that the special master reasonably used
“Daubert’s questions as a tool or framework for conducting the inquiry into the
reliability of the evidence.” Terran, 195 F.3d at 1316.

After Terran, decisions from judges of the Court of Federal Claims have
consistently cited to the Daubert criteria as useful in assessing an opinion that a
vaccine can cause an injury. See, e.g., Snyder v. Sec’y of Health & Human
Servs., 88 Fed. Cl. 706, 742-45 (2009); Cedillo v. Sec’y of Health & Human
Servs., 89 Fed. Cl. 158, 181-82 (2009), aff’d, 617 F.3d at 1338; Bazan v. Sec’y of
Health & Human Servs., 70 Fed. Cl. 687, 699 n.12 (2006) (“A special master
assuredly should apply the factors enumerated in Daubert in addressing the
reliability of an expert witness’s testimony regarding causation.”), rev’d on other
grounds, 539 F.3d 1347 (Fed. Cir. 2008); Campbell v. Sec’y of Health & Human
Servs., 69 Fed. Cl. 775, 781 (2006); Piscopo v. Sec’y of Health & Human Servs.,
66 Fed. Cl. 49, 54 (2005).

The reliability of the expert’s theory is not presumed. A “special master is
entitled to require some indicia of reliability to support the assertion of the expert
witness.” Moberly, 592 F.3d at 1324 (citing Terran, 195 F.3d at 1316).
32
Furthermore, the reliability of an expert=s theory affects the persuasiveness of the
evidence. Special masters may “inquir[e] into the reliability of testimony from
expert witnesses. Weighing the persuasiveness of particular evidence often
requires a finder of fact to assess the reliability of testimony, including expert
testimony, and we have made clear that the special masters have that
responsibility in Vaccine Act cases.” Moberly, 529 F.3d at 1325 (citing Terran,
195 F.3d at 1316). 30

The mere proffer of a theory does not satisfy petitioners’ burden on this
prong. If the special master finds that the expert’s theory is supported by only an
“ipse dixit,” then the special master may reject this opinion. Snyder, 88 Fed. Cl. at
745 n.66 (2009) (quoting Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997)); see
also Cedillo, 617 F.3d at 1339 (also quoting Joiner, 522 U.S. at 146).

To avoid presenting just an unadorned statement from an expert, petitioners
typically present medical articles on which the expert relies. When the petitioner

30
In her post-hearing brief, Ms. Koehn consistently described Dr. McCabe’s
theory as “biologically plausible.” Pet’r Br., filed Sept. 21, 2012, at 8, 12 (citing
Doe 93 v. Sec’y of Health & Human Servs., 98 Fed. Cl. 553, 566-67 (2011)). The
Secretary argued that Ms. Koehn was using the wrong standard. Resp’t Br., filed
Nov. 19, 2012, at 5 (citing Pet’r Br. at 8). Nevertheless, Ms. Koehn continued to
assert that she has advanced a “biologically plausible theory of causation.” Pet’r
Reply Br., filed Dec. 4, 2012, at 4 (capitalization changed without notation).
As discussed in the text, Moberly establishes that the correct standard of
proof in evaluating a petitioner’s theory is the preponderance of the evidence.
Moberly, 592 F.3d at 1322. Although Ms. Koehn is accurate in citing Doe 93 in
support of a plausibility standard, another Court of Federal Claims opinion
respectfully disagreed with Doe 93. Caves v. Sec’y of Health & Human Servs.,
100 Fed. Cl. 119, 144 n.18 (2011). Rather than use a plausibility standard, Caves
used a preponderance of the evidence standard. Id. at 132 (stating “each of [the
Althen] requirements must be proven by a preponderance of the evidence”). When
the Federal Circuit reviewed Caves, it affirmed without opinion pursuant to
Federal Circuit Rule 36. 463 F. Appx. 932 (Fed. Cir. 2012). The Federal Circuit’s
Rule 36 adjudication indicates that “a judgment or decision has been entered
without an error of law.” Thus, the precedential authority supports the
preponderance of the evidence standard.

33
presents medical articles, the special master may evaluate those articles.31 Andreu,
569 F.3d at 1379-80 (“[T]he special master can consider [medical literature or
epidemiological evidence] in reaching an informed judgment as to whether a
particular vaccination likely caused a particular injury.”) (citing Daubert, 509 U.S.
at 593-97). The Secretary, too, may offer articles that contradict a petitioner’s
theory. Stone v. Sec’y of Health & Human Servs., 676 F.3d 1373, 1379-80, reh’g
en banc denied, 690 F.3d 1380 (Fed. Cir. 2012), cert. denied, --- S.Ct. ---, 2013
WL 328557 (2013); Bazan, 539 F.3d at 1353 (stating “[t]he government, like any
defendant, is permitted to offer evidence to demonstrate the inadequacy of the
petitioner’s evidence on a requisite element of the petitioner’s casein-chief [sic].”).

In evaluating expert testimony and scientific literature, special masters
should analyze scientific literature “not through the lens of the laboratorian, but
instead from the vantage point of the Vaccine Act=s preponderant evidence
standard.” Andreu, 569 F.3d at 1380. “In other words, a finding of causation in
the medical community may require a much higher level of certainty than that
required by the Vaccine Act to establish a prima facie case. The special master
must take these differences into account when reviewing the scientific evidence.”
Broekelschen v. Sec’y of Health & Human Servs., 89 Fed. Cl. 336, 343 (2009),
aff’d, 618 F.3d 1339 (Fed. Cir. 2010).

When an expert’s opinion is not supported, the special master may find
petitioner’s proof was inadequate. Althen, 418 F.3d at 1278 (“A persuasive
medical theory is demonstrated by ‘proof of a logical sequence of cause and
effect’ . . . supported by ‘reputable medical or scientific explanation[,]’ i.e.,
‘evidence in the form of scientific studies or expert medical testimony.’”) (quoting
Grant v. Sec’y of Health & Human Servs., 956 F.2d 1144, 1148 (Fed. Cir. 1992));
see also Shapiro v. Sec’y of Health & Human Servs., 105 Fed. Cl. 353, 360 n.5
(2012) (denying motion for review and stating “the Special Master merely
required that the theories be reliable and meet the preponderance of the evidence
standard. He found each of [petitioner’s expert’s] explanations lacking in this
regard, based upon major gaps and flaws in those theories, and instead was
persuaded by [respondent’s expert’s] contradicting testimony.”), aff’d mem., 2013
WL 1896173 (Fed. Cir. 2013).

31
The special master, however, may not require medical literature. Althen,
418 F.3d at 1280.

34
 These standards will be used to determine whether Ms. Koehn has met her
burden of proof for the first prong of Althen.

B. Evidence Related to Prong One of Althen

1. Overview

Dr. McCabe’s theory includes two distinct propositions: first, the
production of inflammatory cytokines can cause sJIA, and, second, Gardasil can
cause inflammatory cytokines. See Pet’r Br., filed Sept. 21, 2012, at 8-12
(organizing petitioner’s prong one evidence around these two points). As
previously summarized, Dr. McCabe relied primarily upon articles authored by
Prakken, Mellins, Ronaghy, and Pinto. Exhibits 12, 13, 15 and 26.

Dr. Rose questioned the reliability of using cytokines to link Gardasil and
sJIA. The formulation that: (1) Gardasil can cause an increase in particular
cytokines; (2) those cytokines can contribute to sJIA; and, therefore, (3) Gardasil
can be a significant factor in causing sJIA is oversimplified. The generation of
cytokines is “very ubiquitous and [is] almost a universal response.” Tr. 279.
Further, people produce a finite number of cytokines, with perhaps as many as 40
being identified so far. Tr. 290; see also Tr. 172 (Dr. McCabe stating he “accept[s]
to a certain extent that there is a commonality in the immune effector functions”).
Thus, to Dr. Rose, “similarities in cytokine patterns . . . do not mean much in terms
of causality.” Tr. 219.

Given this dispute between the experts, the special master’s responsibility is
“to assess the reliability of testimony, including expert testimony” Moberly, 592
F.3d at 1325. One accepted method for evaluating the persuasiveness of an
expert’s opinion is to conduct an analysis using Daubert. Id. at 1324, citing Terran
195 F.3d at 1316.

The Supreme Court listed several non-exclusive factors that trial
courts may consider in evaluating an expert’s opinion:

(1) whether a theory or technique can be (and has been) tested; (2)
whether the theory or technique has been subjected to peer review and
publication; (3) whether there is a known or potential rate of error and
whether there are standards for controlling the error; and, (4) whether
the theory or technique enjoys general acceptance within a relevant
scientific community.
35
Terran, 195 F.3d at 1316 n.2 (citing Daubert, 509 U.S. at 592-95). These factors
will be used here.

2. Whether a Theory or Technique can be (and has been)
Tested

One way to test the theory that Gardasil can cause sJIA is to administer
Gardasil to people and see how many people develop the disease. These
epidemiological studies are discussed separately below.

In lieu of that type of testing, scientists (including Dr. McCabe) look to
criteria listed by Sir Austin Bradford Hill. See exhibit 48; see also Tr. 97
(testimony from Dr. McCabe about the Bradford Hill criteria). Two criteria that
are potentially useful here are responses to other vaccines and animal models.

a) Other Vaccines

Dr. McCabe recognized that one way of inquiring was to “study
exacerbation in individuals who have already been diagnosed and have the
disease.” Tr. 134. Dr. McCabe also stated that considering studies with other
vaccines would be “a reasonable hypothesis . . . to consider with a few caveats.”
Tr. 181. Exploring how vaccinations affect people with a disease can inform the
assessment of whether the vaccinations cause the disease. Tr. 222; see also W.C.
v. Sec’y of Health & Human Servs., No. 07-456V, 2011 WL 4537877, at *14-15
(Fed. Cl. Spec. Mstr. Feb. 22, 2011) (considering three studies about flu
vaccination given to people with multiple sclerosis), mot. for review denied in
relevant part and granted in non-relevant part, 100 Fed. Cl. 440 (2011), aff’d 704
F.3d 1352 (Fed. Cir. 2013).

The record contains the results of two studies involving vaccines and
juvenile idiopathic arthritis. In both studies, the people given the vaccine already
suffered from juvenile idiopathic arthritis and the researchers were studying
whether the patient’s disease worsened after the vaccination. See generally,
exhibit 43 (Heijstek) and exhibit 47 (Zonneveld-Huijssoon). One study involved
MMR vaccine. In a retrospective analysis of 207 patients with juvenile idiopathic
arthritis (of which 17 had sJIA), the researchers found “no changes in disease
activity, flare occurrence or medication use after the MMR vaccination.” Thus, the

36
researchers concluded that the “MMR vaccination appears to be safe in JIA.”
Exhibit 43 (Heijstek) at 1386.32 The other study used the meningococcal C
vaccination. The total number of participants was 234, including 34 with sJIA.
Exhibit 47 (Zonneveld-Huijssoon) at 641. The researchers “did not detect any
worsening of disease activity within 6 months after MenC vaccination.” Id. at 644.

Analogizing between how other vaccines affect patients with juvenile
idiopathic arthritis and how Gardasil affects patients with sJIA, Dr. Rose stated
that “these two vaccines seem to have a wonderful record of safety in patients with
JIA.” Tr. 222. Dr. Rose added, that all vaccines, except live viral vaccines, are
recommended to those who have JIA. Id.

Dr. McCabe pointed out that Gardasil induces a stronger response from the
immune system than the natural infection. Tr. 181-82. Dr. McCabe also did not
know whether the MMR vaccine or the meningococcal C vaccination induced the
same type of cytokines as Gardasil induces. Tr. 183. Thus, the analogy between,
on the one hand, either MMR vaccine or the meningococcal vaccine, and, on the
other hand, Gardasil, is not perfect.

To the extent that some differences can be overlooked, the Heijstek and
Zonneveld-Huijssoon studies suggest that when researchers have explored whether
vaccinations affect juvenile idiopathic arthritis, they have not found that the
vaccine worsens the disease. Because they are studies, the Heijstek and
Zonneveld-Huijsoon findings are entitled to more weight than speculative passages
in other articles. For example, a group of researchers, including Dr. Zonneveld-
Huijsoon, stated “in juvenile idiopathic arthritis (JIA) a temporal relationship
between disease onset, childhood vaccination, remissions and flares, hint at a
possible relation of JIA disease activity and vaccinations or infections.” Exhibit 15
(Ronaghy) at 1. As the Secretary argued, this language (“hint”) is “equivocal.”
Resp’t Br. at 6.

b) Animal Models

Another way to test whether a substance causes a disease is to substitute
animals for people. If the animals develop the disease, then people might, too. See

32
Dr. Heijstek and colleagues added a caveat that the statistical power of
their study was limited and recommended a prospective trial. Id.

37
Tr. 176 (Dr. McCabe’s testimony that if he had unlimited funding to study the
causes of sJIA, he would “possibly look for some changes in animal models”).

Here, there are no animal models for sJIA. However, animal models do
exist for a related disease, macrophage activation syndrome. Dr. McCabe and Dr.
Rose agreed that macrophage activation syndrome is similar to, although not
exactly the same as, sJIA. Tr. 76 (Dr. McCabe), 219 (Dr. Rose), 285 (Dr. Rose),
297-98 (Dr. McCabe); see also exhibit C (Textbook) at 241 (discussing
macrophage activation syndrome within a chapter on sJIA). While Dr. Rose
suggested that a worsening of symptoms after injecting MAS-afflicted mice with
specific vaccines would give us clues about the effects of certain vaccines
compared to others, Tr. 219, 285, there was no evidence showing that this
experiment was conducted.33

Dr. Rose further indicated that, in his opinion, this experiment is unlikely to
be conducted. Dr. Rose explained that researchers are pursuing hypotheses around
sJIA that are more likely to produce advancements than the theory that Gardasil
can cause sJIA. Tr. 220-22.

Overall, the evidence relating to testing does not assist Ms. Koehn. From
her perspective, the most favorable interpretation is that this factor is neutral
(neither supporting nor discounting) because the most on-point testing has not been
done. Another interpretation is that this factor is against Ms. Koehn’s theory
because the testing that has been done with other vaccines and sJIA has refuted a
connection between those vaccines and sJIA.

3. Whether the Theory or Technique has been Subjected to
Peer Review and Publication

The theory that Gardasil can cause sJIA has not been subject to peer review
or publication. Dr. McCabe’s attempt to combine two ideas— (1) that pro-
inflammatory cytokines can cause sJIA and (2) that Gardasil can cause pro-
inflammatory cytokines—appears to be unprecedented. As the Secretary points
out, until Ms. Koehn’s case, there was not even one case report published in the

33
Neither party introduced any articles discussing the extent of experiments
on animals with macrophage activation syndrome.

38
medical journals showing even a temporal sequence in which a Gardasil
vaccination preceded sJIA. Resp’t Br. at 9.

The peer-reviewed article on which Dr. McCabe most heavily relied was
Pinto. Dr. McCabe saw Pinto as supporting his theory because Pinto demonstrates,
in some circumstances, increased levels of cytokines are present seven months
after vaccination. The specific part of the Pinto experiment on which Dr. McCabe
relied was when blood from a vaccinated person was stimulated with the virus-like
particle. See Tr. 104-12.

Dr. Rose opined that a different part of the experiment was more
meaningful. He stated that for purposes of evaluating a possible connection
between HPV vaccination and sJIA, the relevant portion is the media. To Dr.
Rose, this part of the experiment showed how the cells “are before and after
vaccination, how the cells behave when you leave them alone.” Tr. 224. When
Dr. Rose analyzed the data regarding the media, he saw that “for almost no
cytokine there’s a spontaneous release of cytokines that is different at time zero
compared to time two and time seven.” Tr. 225. The researchers came to the same
conclusion: “As shown in Fig. 1D, spontaneous secretion of cytokines in the
absence of any stimuli (media control) did not show any significant increases
following vaccination.” Exhibit 26 (Pinto) at 3560. In other words, each
successive administration of HPV vaccine did not produce any increase in
cytokines. Cytokine levels increased only when researchers reintroduced the agent
against which the vaccine was designed to protect.

Dr. Rose was less interested in the data showing the production of cytokines
after the blood cells were stimulated with more of the L1 virus-like particle. He
stated: “[o]f course, when you stimulate with an antigen you get more” cytokines
released. Tr. 265.

Despite contrary testimony from Dr. McCabe (see Tr. 293-96, 301-03), Dr.
Rose’s focus on the media column is logical. The blood in the media encountered
the L1 virus-like particle only in the context of the three doses of vaccination. This
pattern resembles what happened to Vanessia in the sense that no medical record
suggests that she was exposed to a living strand of the human papillomavirus. If
Vanessia encountered the human papillomavirus after the vaccination, the Pinto
article predicts that she would produce a robust immune response like the ones
reported for 10 µg and 1.0 µg of the virus-like particle.

39
 The Pinto experiment also undermined the cohesiveness of Dr. McCabe’s
theory, particularly in regard to timing both for onset of symptoms and duration of
symptoms. To review, after human beings are exposed to an antigen, they produce
cytokines immediately. Tr. 281-82 (Dr. Rose’s testimony that “from stimulus to
response is a question of hours”). But, in Dr. McCabe’s theory, the onset of
disease can occur as long as seven months after vaccination. Tr. 127-29 (citing
exhibit 25 (Frazer) at S13).

Dr. McCabe explained that the delay could be due to the time needed to
amplify the immune system’s response. Tr. 300-01; see also Tr. 295. However,
the media portion of the Pinto experiment contradicts Dr. McCabe’s speculation
about an amplification process. In Pinto, the cytokines increased only when the
blood was restimulated. When the blood was left alone, the cytokine level stayed
relatively constant. This lack of continued elevation in pro-inflammatory
cytokines was inconsistent with how sJIA persists. In Dr. Rose’s experience in
treating people with sJIA, those patients constantly need to receive medications to
prevent development of pro-inflammatory cytokines. When the medication stops,
the person has a flare in her (or his) disease. Tr. 224. Dr. McCabe, who is not a
medical doctor, agreed that “cytokine disregulation in sJIA isn’t a transient event.”
Tr. 305. But, when he was asked about why sJIA is a chronic disease, Dr. McCabe
did not provide a persuasive explanation. Tr. 305.

Overall, the evidence relating to peer review and publication does not assist
in finding that Dr. McCabe’s theory is probable. The peer-reviewed articles about
epidemiology are taken up separately.

4. Whether There is a Known or Potential Rate of Error and
Whether There are Standards for Controlling the Error

No evidence was introduced on this topic. An error rate for Dr. McCabe’s
theory cannot be calculated. Thus, this factor does not constitute affirmative or
negative evidence.

40
 5. Whether the Theory or Technique Enjoys General
Acceptance within a Relevant Scientific Community34

Except for the portion of the Prakken article discussed above, Ms. Koehn
has not presented any evidence that the relevant scientific community generally
accepts the theory that Gardasil can cause sJIA. See Pet’r Reply Br. at 14-15. The
Secretary has presented evidence (the opinion of Dr. Rose) that shows that the
theory is not generally accepted.

Dr. Rose is the head of pediatric rheumatology at the Alfred I. DuPont
Hospital for Children. Exhibit B (curriculum vitae); Tr. 200. He conducts
research on juvenile rheumatoid arthritis, although not on sJIA. Tr. 202. He
attends conferences held by associations of rheumatologists. Tr. 283-84. He
serves as an editor for Clinical Rheumatology. Exhibit B at 9. Given this
background, it seems likely that if rheumatologists were considering whether
Gardasil can cause sJIA, then Dr. Rose would have heard some discussion about
this theory. However, Dr. Rose testified that he did not recall hearing about this
idea. Tr. 284.

Furthermore, Dr. Rose stated that the general practice among
rheumatologists is to recommend vaccinations for their patients with sJIA. See
Tr. 219, 222. This practice reflects a belief that the benefits from vaccination
outweigh the potential harm from vaccination. Although, conceivably, at some
future time, rheumatologists will generally accept the theory that Gardasil can
cause sJIA, the evidence in this case is that they do not.

6. Additional Considerations

In defining how district court judges should determine whether expert
opinion is admissible, the Supreme Court has emphasized that the approach should
be “flexible.” Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 150 (1999)
(citing Daubert, 509 U.S. at 594). Thus, the analysis of whether the theory that

34
Citing Capizzano, 440 F.3d at 1325, Ms. Koehn argues that a petitioner is
not required to show a particular theory has general acceptance. Pet’r Br. at 15. It
is correct that special masters may not require general acceptance. However,
pursuant to Terran, special masters may consider whether a particular theory has
general acceptance as one factor in the overall analysis. 195 F.3d at 1316.

41
Gardasil can cause sJIA may consider more than just the four factors explicitly
listed in Daubert. Two other factors are the origins of the theory and
epidemiological studies.

a) Genesis of the Expert’s Theory

One consideration is why the expert came up with the opinion. On remand
from the Supreme Court, the Ninth Circuit stated:

One very significant fact to be considered is whether the experts are
proposing to testify about matters growing naturally and directly out
of research they have conducted independent of the litigation, or
whether they have developed their opinions expressly for purposes of
testifying. That an expert testifies for money does not necessarily cast
doubt on the reliability of his testimony, as few experts appear in
court merely as an eleemosynary gesture. But in determining whether
proposed expert testimony amounts to good science, we may not
ignore the fact that a scientist’s normal workplace is the lab or the
field, not the courtroom or the lawyer’s office.

Daubert v. Merrell Dow Pharm., Inc., 43 F.3d 1311, 1317 (9th Cir. 1995).

Here, Dr. McCabe developed his theory for the purpose of litigation. From
his initial consultation, he understood that petitioner was hypothesizing that
Gardasil caused Vanessia’s sJIA. From that starting point, Dr. McCabe
investigated whether “there was a tenable scientific argument” and produced his
report accordingly. Tr. 164. This factor, although not decisive, weighs against Dr.
McCabe’s theory.

b) Epidemiological Studies

The Federal Circuit has endorsed consideration of epidemiological studies as
one factor in the special master’s analysis. See W.C. v. Sec’y of Health & Human
Servs., 704 F.3d 1352, 1361 (Fed. Cir. 2013) (holding that the special master was
not arbitrary in denying compensation and summarizing epidemiological studies
cited by the special master); Lampe, 219 F.3d at 1365 (stating “[a]n
epidemiological study may be probative medical evidence relevant to a causation
determination”). The special master may not find against a petitioner solely
because the petitioner did not introduce supporting epidemiology. Capizzano, 440
F.3d at 1325.
42
 Ms. Koehn acknowledges that she has not presented any epidemiology.
Pet’r Br. at 15. The Secretary, on the other hand, relies upon the results of two
studies that looked for but did not find an increased incidence of disease after
vaccines against human papillomavirus. Resp’t Br. at 7-8 (citing exhibit 34 (Chao)
and exhibit E (Verstraeten)).

Ms. Koehn’s cross-examination of Dr. Rose brought out two weaknesses in
this reliance on the Verstraeten article. First, as mentioned above, Gardasil is not
the same as Cervarix. Tr. 240. Logically, it is possible that a component of
Gardasil that is not in Cervarix could cause unintended side effects that would not
be identified in studies about Cervarix. Second, the size of the Verstraeten study,
despite including more than 60,000 people, is still not large enough to discover an
increased risk of sJIA. This argument derives from the incidence of sJIA, which is
approximately 0.3 to 0.8 per 100,000 people in the United States. Tr. 244-45; but
see Tr. 133 (Dr. McCabe’s testimony that the incidence of sJIA is between 2 and
20 cases per 100,000 people). Given the frequency with which new cases develop,
Dr. Rose was reluctant to estimate the size of an adequately powered study,
although he speculated the size might be 100,000 people. Tr. 245-46.

In addition to the Verstraeten study, the other epidemiological study was
authored by Chun Chao. Ms. Koehn reasonably could not repeat the attacks used
against the Verstraeten article in response to the Chao article. Unlike the
population Verstraeten analyzed, the Chao study subjects received Gardasil.
Compare exhibit E (Verstraeten) at 6631 with exhibit 34 (Chao) at 193 and Tr. at
132 (Dr. McCabe noting that HPV-4, referred to in the Chao study, is Gardasil). In
addition, Chao looked at more than twice as many women. Compare exhibit 34
(Chao) at 193 (n = 189,629) with exhibit E (Verstraeten) at 6630 (n = 68,512).
Instead, Ms. Koehn called into question the supposition that Chao researchers
would have identified cases of sJIA. See Pet’r Reply Br. at 14.

Based on the population analyzed by Chao and her colleagues and the
incidence of sJIA, the study appears to be robust. According to the results,
however, “no cluster of disease onset in relation to vaccination timing, dose
sequence or age was found for any autoimmune condition.” Exhibit 34 at 193. In
other words, “[n]o autoimmune safety signal was found in women vaccinated with
HPV4.” Id. While an epidemiological study cannot prove that Gardasil does not
cause autoimmune diseases as an absolute proposition, the results suggest that
Gardasil causes an autoimmune disease extremely rarely, if it causes an
autoimmune disease at all.
43
 Taken together, the Verstraeten and the Chao articles are an additional (but
not decisive) reason for finding that Dr. McCabe’s theory that a vaccine against
human papillomavirus can cause sJIA to be unlikely. The same result would have
occurred even if the epidemiological studies were not part of the record.

C. Summary

The Supreme Court has recognized that a novel theory that is relatively
unexamined by the relevant scientific community may not be as persuasive as a
theory that has been thoroughly peer-reviewed. This is so because “submission to
the scrutiny of the scientific community . . . increases the likelihood that
substantive flaws in methodology will be detected.” Daubert, 509 U.S. at 593-94.
The Daubert Court added, however, that the lack of publication is a “relevant,
though not dispositive, consideration in assessing . . . scientific validity.” Id. at
594. Special masters, too, have recognized that a theory’s novelty is not
dispositive in determining its scientific validity. Cedillo v. Sec’y, Health &
Human Servs., No. 98-916V, 2009 WL 331968, at *111 (Fed. Cl. Spec. Mstr.
Feb. 12, 2009) (“At times novel theories can be persuasive.”), mot. for review
denied, 89 Fed. Cl. 158, aff’d, 617 F.3d 1328. Ultimately, however, it is
petitioner’s burden to support her theory with “sufficient supportive evidence to
justify the adoption of a proffered new theory.” Cedillo, 2009 WL 331968, at
*111.

With respect to the first prong of Althen, Ms. Koehn’s burden is to establish
that Gardasil can cause sJIA. Her proof does not need to be scientifically certain;
preponderant evidence suffices.

Here, the evidence does not weigh in Ms. Koehn’s favor. Dr. McCabe, a
Ph.D. immunologist, has pieced together a theory that, although not entirely
impossible, contains sufficient gaps to make it unpersuasive. See Joiner, 522 U.S.
at 146 (affirming exclusion of an expert’s report when the trial court “conclude[d]
that there [was] simply too great an analytic gap between the data and the opinion
proffered”). Consequently, Ms. Koehn has not met her burden of proof.

44
V. Prong Three from Althen – Timing35

Petitioners are required to establish a “showing of a proximate temporal
relationship between vaccination and injury.” Althen, 418 F.3d at 1278. The
Federal Circuit has elaborated that the third prong of the Althen test requires
“preponderant proof that the onset of symptoms occurred within a timeframe
which, given the medical understanding of the disorder’s etiology, it is medically
acceptable to infer causation.” Bazan, 539 F.3d at 1352. “Under this test,
petitioner [is] first required to establish the timeframe for which it is medically
acceptable to infer causation, that is, the timeframe in which symptoms would be
expected to arise if the [disease] was caused by the vaccination. Then, she [is]
obliged to show that the onset of her [disease] occurred during this causation
period.” Shapiro v. Sec’y of Health & Human Servs., 101 Fed. Cl. 532, 542
(2011), recons. denied after remand, 105 Fed. Cl. 353 (2012), aff’d mem., 2013
WL 1896173 (Fed. Cir. 2013).

These two aspects are separately considered, beginning with findings related
to when Vanessia’s sJIA began. After the relevant time for Vanessia is
established, the next section reviews whether her onset date falls within a
medically acceptable timeframe.

A. What happened to Vanessia

Vanessia received the first dose of Gardasil on February 18, 2008, and the
second dose on April 18, 2008. Exhibit 2 at 3. The experts agree that Vanessia’s
sJIA became manifest in late June 2008. See Tr. 129 (Dr. McCabe stating “the
disease emerged manifest in June of 2008”); Exhibit A (Dr. Rose’s report) at 1.36
As the Secretary points out, the interval between vaccination and onset is
approximately four months (using the date of the first dose) and approximately two
months (using the date of the second dose). See Resp’t Br. at 14.

35
Since the third prong of Althen ties directly to Dr. McCabe’s theory, this
decision discusses the third prong now. The second prong of Althen is discussed
in section VI.
36
Facts supporting the onset include: On June 21, 2008, Vanessia reported
she had a rash all over her body. Exhibit 3 at 8. While hospitalized, Vanessia was
diagnosed with systemic onset juvenile arthritis. Exhibit 4 at 11-12.

45
 For Ms. Koehn to prevail, she must establish that two months (or four
months) falls within the medically acceptable timeframe. Bazan, 539 F.3d at 1352;
Shapiro, 101 Fed. Cl. at 542.

B. Time Expected by Medical Science

The Court of Federal Claims has recognized that petitioners’ proof of the
medically acceptable time for an injury to appear after vaccination depends upon
the petitioners presenting, pursuant to Althen prong one, a “reputable theory as to
how the vaccination could cause the injury.” Langland v. Sec’y of Health &
Human Servs., 109 Fed. Cl. 421, 443 (2013). This linkage makes sense. If
medical science understands how an injury might occur, then there would be some
basis for understanding when the injury would occur. Conversely, if there is little
understanding about the cause of a disease, then it is difficult to say when the
disease should begin.37 See Veryzer v. Sec’y of Health & Human Servs., 100 Fed.
Cl. 344, 356 (2011) (“[T]he ‘etiology’ of the disorder determines the appropriate
temporal relationship.”), aff’d without opinion, 475 Fed. Appx. 765 (Fed. Cir.
2012). Moreover, analyzing the medically appropriate time from prong three in
terms of the medical theory from prong one is in accord with the observation that
evidence from one prong may overlap with another prong. Capizzano, 440 F.3d at
1326.

To Dr. McCabe, the “expected interval between vaccination . . . and the
onset of the autoinflammatory disease is predicted by the time period that
measurable changes in the immune response are known to be elicited by the
vaccine.” Tr. 128. As discussed above, people receiving Gardasil produce
antibodies against the human papillomavirus within seven months. See exhibit 25
(Frazer) at S13. Therefore, Dr. McCabe implies that an appropriate timeframe in
which an individual can first exhibit symptoms of sJIA caused by Gardasil extends
to up to seven months. See Pet’r Reply Br. at 16 (“[I]t follows that the interval

37
Dr. Rose expressed this idea when he stated “two months is as good as
two hours or as good as six months since we really don’t know what’s going on.”
Tr. 308.

46
during which sJIA could be said to have a temporal association with Gardasil is the
same 3-dose time frame, or within 7 months.”).38

Dr. Rose questioned why a theory involving cytokines could produce an
injury after a delay of several months. As discussed in reference to Althen prong
one, Dr. Rose stated, and Dr. McCabe agreed, that the immune system produces
cytokines very quickly after it encounters an antigen. Tr. 281-82 (Dr. Rose); Tr. at
295 (Dr. McCabe). Thus, Dr. Rose expected that if vaccine-triggered cytokines
contributed to the pathogenesis of sJIA, then symptoms of sJIA would “start[] right
away.” Tr. 282.

Dr. McCabe’s theory holds that cytokines that are produced in response to
the vaccination could lead to sJIA. He acknowledged that sJIA has genetic factors,
“meaning that certain susceptible members of the population likely exist and
develop this disease with or without environmental triggers.” Tr. 76. He then
added that cytokines, activated by the vaccine, act on multiple tissues causing fever
and the release of acute phase reactive proteins. Tr. 77-80. In his PowerPoint, he
also acknowledged increased vascular permeability and increased synovial
inflammation in response to cytokine activation. Exhibit 38 at slide 5 (reproducing
figure 1 from exhibit 13 (Mellins) at 419). When asked to explain how the
vaccine-stimulated cytokines cause the disease, Dr. McCabe referred to these
consequences. Tr. 299. He also expected that “cytokine-mediated interactions
between cells of the adaptive immune system and the innate immune system . . .
are somehow playing a role in the disease,” but more sophisticated information
was lacking. Tr. 299-300.

The lack of specificity in Dr. McCabe’s theory creates a gap in Ms. Koehn’s
case. The consequences of cytokine production that Dr. McCabe identifies, such
as fever, are usually apparent very quickly. The body’s rapid cytokine response
appears inconsistent with Dr. McCabe’s assertion that the onset of disease could
take many months.

Dr. McCabe attempted to answer this conundrum by opining that the onset
of sJIA could be delayed because “there’s an amplification process.” Tr. 301.
However, Dr. McCabe did not explain persuasively what he meant by that term.

38
As discussed below, Dr. McCabe did not directly discuss the interval in
Vanessia’s case, which is two months.

47
see also Tr. 295. And specifically, Dr. McCabe did not explain why the immune
system’s production of cytokines would be amplified for weeks and months
without a stimulant being present.

In sum, the record does not support a finding that the medically appropriate
interval for a cytokine-mediated theory would extend out to seven months as Dr.
McCabe proposed. Seven months might be appropriate for a different theory.39
And seven months might be appropriate for a cytokine-mediated theory if there
were some reliable evidence about how the cytokines start a lengthy process. But,
because cytokines exist for a short duration, a preponderance of evidence does not
support the finding that seven months is an appropriate medical interval.

More important for Ms. Koehn’s case is whether a preponderance of the
evidence establishes that two months is a medically appropriate interval because
Vanessia’s sJIA symptoms were recognized approximately two months after the
second dose of Gardasil. See section V.A above. There was no testimony from
either Dr. McCabe or Dr. Rose saying that two months is medically appropriate. In
the absence of evidence, it is difficult to find that Ms. Koehn has met her burden of
proof. Even two months is probably too long an interval for a cytokine-driven
reaction. See James v. Sec’y of Health & Human Servs., No. 09-284V, 2010 WL
4205699, at *6 (Fed. Cl. Spec. Mstr. Sept. 30, 2010) (summarizing testimony of
the petitioner’s expert that a child’s death 14 hours after vaccination was consistent
with release of cytokines); Doe/11 v. Sec’y of Health & Human Servs., No.
99-212V, 2008 WL 4899356, at *28-30 (Fed. Cl. Spec. Mstr. Oct. 29, 2008)
(discussing whether a cytokine storm can arise in four hours), mot. for review
denied, 87 Fed. Cl. 1 (2009), aff’d, 601 F.3d 1349 (Fed. Cir. 2010).40

39
For example, in other cases, petitioners have proposed that a vaccine
caused an autoimmune response involving either antibodies or T-cells. Ms. Koehn
has not proposed a theory involving antibodies or T-cells because neither appears
to be involved in the pathogenesis of sJIA. See Exhibit C (Textbook) at 236.
40
These cases are consulted because (a) petitioner did not introduce any
evidence about whether two months is a medically appropriate time and (b) special
masters may use their “accumulated expertise” in evaluating the cases. Lampe,
219 F.3d at 1362 (quoting Hodges v. Sec’y of Health & Human Servs., 9 F.3d 958,
961 (Fed. Cir. 1993)).

48
 Ultimately, a finding on Althen prong three is not needed. Even if Ms.
Koehn had established that there was a proper temporal sequence, timing does not
entitle her to compensation. Grant, 956 F.2d at 1148. She is also required to
establish a persuasive medical theory. Althen, 418 F.3d at 1278. As explained
above, she has not met the first element and the failure to meet this element means
that she cannot be compensated. See Hibbard v. Sec’y of Health & Human Servs.,
698 F.3d 1355, 1364-65 (Fed. Cir. 2012) (holding special master did not err in
resolving the case pursuant to prong two when respondent conceded that petitioner
met prong three).

VI. Prong Two from Althen – Logical Sequence of Cause and Effect

The remaining Althen prong is “a logical sequence of cause and effect
showing that the vaccination was the reason for the injury.” 418 F.3d at 1278.
Given the finding that Ms. Koehn has not established a persuasive theory to
explain how Gardasil can cause sJIA, as a matter of logic, she cannot show that
Gardasil did cause her sJIA. See Caves, 100 Fed. Cl. at 145. Nevertheless, the
evidence particularly relevant to this factor is discussed for the sake of
completeness.41

A. Factors to Consider in regard to Prong Two from Althen

While the first prong from Althen is sometimes shortened to “can it?,” the
second prong can be summarized as asking “did it?” See Pafford, 2004 WL
1717359, at *4-5, 9. Evidence relevant to this prong “tends to be evidence specific
for the petitioner.” Viscontini v. Sec’y of Health & Human Servs., No. 98-619V,
2011 WL 5842577, at *20 (Fed. Cl. Spec. Mstr. Oct. 21, 2011), mot. for review
denied, 103 Fed. Cl. 600 (2012). This focus on the petitioner particularly reflects
the separate inquiries into the question of general causation (Althen prong one) and
question of specific causation (Althen prong two). Veryzer, 100 Fed. Cl. at 353.

41
As part of her argument regarding Althen prong two, the Secretary argues
that the two (or four) month delay between vaccination and onset of symptoms
makes the logical sequence of events questionable. See Resp’t Br. at 10. Because
section V.B above discussed the timing issue, the analysis of Vanessia’s
chronology is not repeated here.

49
 According to the Federal Circuit, the petitioner might present preponderant
evidence on this prong by submitting evidence from treating doctors and/or
evidence demonstrating challenge / rechallenge. Capizzano, 440 F.3d at 1325-26.
This type of evidence focuses on the overriding issue in this case—whether
Gardasil was a substantial factor in causing Vanessia’s sJIA. See Shyface v. Sec’y
of Health & Human Servs., 165 F.3d 1344, 1352 (Fed. Cir. 1999). Although Ms.
Koehn argued that Vanessia’s presentation was consistent with pro-inflammatory
cytokines, see Pet’r Br. at 13-14 (citing Tr. 123-25), this showing merely
establishes that Vanessia suffered from sJIA. It does not show that the pro-
inflammatory cytokines resulted from Gardasil. To be entitled to compensation,
Ms. Koehn must present additional evidence. See Moberly, 592 F.3d at 1323
(holding that an appropriate temporal onset and “a simplistic elimination of other
potential causes of the injury” does not meet petitioner’s burden) (quoting Althen,
418 F.3d at 1278).

B. Evidence Related to Prong Two from Althen

1. Statements of Treating Doctors

To demonstrate the “logical sequence of cause and effect,” the Federal
Circuit has identified statements of treating doctors as probative evidence.
Capizzano, 440 F.3d at 1326. Their views, however, are not necessarily
“sacrosanct.” Snyder, 88 Fed. Cl. at 745 n.67.

Here, Ms. Koehn acknowledges that “Vanessia’s treating physicians did not
express any opinion as to whether Gardasil was a cause of her development of
sJIA.” Pet’r Br. at 14 (citing Tr. 157). Although this lack of connection from a
treating doctor tends to make her claim less likely, Ms. Koehn points to a statement
from one of Vanessia’s rheumatologists, Dr. Hoftman.

Dr. Hoftman worked within the University of California at Los Angeles
(UCLA) Health System. Exhibit 5 at 28. Vanessia had been seen at UCLA since
July 2008. Exhibit 3 at 11; exhibit 5 at 51. Dr. Hoftman saw her on January 12,
2011, as part of a periodic follow up. Exhibit 5 at 28. In the context of presenting
a plan until Vanessia’s next appointment in three months, Dr. Hoftman wrote “Pt
mother refused flu vaccine this year. Discussed [with] mom importance of this
vaccine, risk < benefit. Mom hesitant b/c Gardasil. [Discussed with] mom – no

50
data but all vaccines and infections can trigger autoimmune response.” Exhibit 5
at 28.42

Dr. Hoftman does not express any agreement with Ms. Koehn’s concern
about Gardasil. Dr. Hoftman actually appears to have recommended that Vanessia
receive the flu vaccination and Vanessia would have been vaccinated against the
flu at the January 12, 2011 appointment except that Ms. Koehn “refused flu
vaccine this year.” In other years after Vanessia was diagnosed with sJIA, doctors
had recommended, and Ms. Koehn had accepted their recommendation, that
Vanessia receive a flu vaccination. See exhibit 5 at 32, 44, 60.

2. Challenge / Rechallenge

The advice to receive a flu vaccination is not necessarily inconsistent with a
theory that Gardasil caused Vanessia’s sJIA because the flu vaccine is not the same
as Gardasil. The more relevant inquiry is whether the doctors recommended an
additional dose of Gardasil.

When patients encounter a putative causative agent a second time, they are
considered to be facing a “rechallenge.” See Capizzano, 440 F.3d at 1322 (stating
“[a] rechallenge event occurs when a patient who had an adverse reaction to a
vaccine suffers worsened symptoms after an additional injection of the vaccine”).
The challenge-rechallenge paradigm is relevant to determining whether petitioners
have demonstrated a “logical sequence of cause and effect.” Capizzano, 440 F.3d
at 1326.

When the vaccinee’s medical history supports challenge-rechallenge, special
masters have accepted this evidence as persuasive. Freeman v. Sec’y of Health &
Human Servs., No. 04-1528V, 2009 WL 5103594, at *12 (Fed. Cl. Spec. Mstr.
Dec. 9, 2012); Hall v. Sec’y of Health & Human Servs., No. 02-1052B, 2007 WL
312084, at *7 (Fed. Cl. Spec. Mstr. Oct. 4, 2007). On the other hand, petitioners
have sometimes fallen short of demonstrating that their case truly fits the

42
To the extent that Dr. Hoftman expressed an opinion that “all vaccines . . .
can trigger [an] autoimmune response,” Dr. Hoftman’s statement provides a
modicum of support for Ms. Koehn’s prong one argument. It does not weigh very
heavily in that regard because, as Dr. Hoftman states in that same sentence, there is
“no data” for the proposition.

51
challenge-rechallenge model. Doe 70 v. Sec’y of Health & Human Servs., 95 Fed.
Cl. 598, 608 (2010) (denying motion for review when the special master did not
arbitrarily find that “the facts of petitioner’s case did not fit the challenge-
rechallenge model”), aff’d sub nom., Rickett v. Sec’y of Health & Human Servs.,
468 Fed. Appx. 952 (Fed. Cir. 2011); Locane v. Sec’y of Health & Human Servs.,
No. 99-589V, 2011 WL 3855486, at *11 (Fed. Cl. Spec. Mstr. Feb. 17, 2011), mot.
for review denied, 99 Fed. Cl. 715, 732 (2011), aff’d, 685 F.3d 1375 (Fed. Cir.
2012).

Here, the parties draw different conclusions from how Vanessia fared after
her third vaccination. The relevant chronology shows:

Date Event Citation
7/1/08 Dr. Scott diagnoses Vanessia with probable sJIA while Exhibit 4 at
she was hospitalized 11-12.
7/2/08 Dr. Regala, Vanessia’s pediatrician, refers Vanessia to Exhibit 3 at
UCLA / Dr. McCurdy 11.
7/8/08 Vanessia’s first appointment with Dr. McCurdy. The Exhibit 5 at
history notes that Vanessia had received two doses of the 51-55.
HPV vaccine.
Dr. McCurdy continued the prescriptions for prednisone,
methotrexate, Enbrel.
8/19/08 Dr. Regala administers the third dose of Gardasil Exhibit 3 at
6; see also
exhibit 3 at
3-4.
8/25/08 During physical therapy, Vanessia had rash, chills, and Exhibit 8 at
joint pain. 48-50.
9/3/08 Vanessia returns to Dr. McCurdy. Her current Exhibit 5 at
medication was Enbrel. By history, Vanessia had “some 45-46.
improvement [with] Enbrel,” although she had “swollen
knees [and] ankles.” Also, by history, after Vanessia
“stop[ped] prednisone, [her] rash returned.”
The doctor’s plan included continuing Enbrel and
starting methotrexate. The doctor also ordered laboratory
tests and if there were an increase in “inflammatory
markers[,] may need prednisone.”

Relying upon Dr. Rose’s testimony, the Secretary interprets this sequence as
contrary to the challenge-rechallenge paradigm. According to the Secretary, “if the
52
HPV vaccine caused or substantially contributed to Vanessia’s sJIA, then it would
seem logical that a third dose of it on August 19, 2008 would have significantly
exacerbated her symptoms.”43 The Secretary argues that the third dose of Gardasil
did not make Vanessia worse because the rash was associated with stopping
prednisone, not with the administration of the vaccine. Resp’t Br. at 12.

Dr. McCabe’s response is to emphasize a medication that Vanessia was
taking—Enbrel. When Vanessia received the third dose of Gardasil, she was also
taking an “anti-inflammatory therapy.” Tr. 126. Enbrel is a confounding factor.
As Dr. McCabe explained: “If there were no changes, part of that would be I
would suspect or wonder and consider whether well, the reason that there’s no
change is because at the same time that a stimulus is given an inhibitor is present.”
Tr. 126.

In this context, Dr. McCabe stated that trying to determine whether the third
dose of Gardasil made Vanessia worse is “difficult to say” because there are “[t]oo
many variables.” Tr. 127. Any worsening could have been due to her stopping
prednisone. Her continued use of Enbrel could have prevented any worsening that
Gardasil would have caused absent the use of Enbrel.44 In addition, there is the
normal waxing and waning of sJIA.

The many confounding factors make reliance on Vanessia’s experience after
the third dose of Gardasil difficult in either respect. While it cannot be said that
the Secretary has proven the absence of rechallenge, Ms. Koehn has not met her
burden of proving that Vanessia’s case constitutes an example of rechallenge.45

43
Dr. McCabe indicated that on an abstract level, this logic is an appropriate
way to explore a cause and effect relationship. Tr. 125-26.
44
Enbrel appears to help Vanessia cope with her disease. See exhibit 8 at 43
(noting, on February 25, 2009, that her hand hurt after she missed one dose of
Enbrel).
45
In its most recent report addressing whether vaccines cause injuries, the
Institute of Medicine discussed the value of the rechallenge paradigm.

It is possible that one or more of the ‘challenges’ in an individual case
patient reporting is related to coincidental exposure; thus, the
committee looked for other information. . . . The value for the
(. . . continued)
53
 3. Relative Qualifications of Experts46

In weighing the persuasiveness of opinion testimony, special masters may
consider the relative expertise of the witness. Locane v. Sec’y of Health & Human
Servs., 685 F.3d at 1380 (stating “[t]he Special Master found Dr. Warner’s
testimony more persuasive than Dr. Bellanti’s because of their different
backgrounds and specialties and because the medical literature supports Dr.
Warner’s theory. . . . We find nothing arbitrary or capricious.”); Stone, 676 F.3d at
1382 (noting “[t]he special master found the respondent’s experts’ testimony on
that issue to be more reliable than Dr. Kinsbourne’s in view of their more extensive
and more recent experience”).

Dr. McCabe is not a medical doctor. Tr. 33. While Dr. McCabe’s lack of
training and experience as a medical doctor could decrease the value of his opinion
for any of the Althen prongs, see Resp’t Br. at 4-5 (discussing Dr. McCabe’s

committee of rechallenge cases is much greater for monophasic
conditions (events that typically happen only once, e.g., vasculitis)
than for relapsing-remitting conditions, such as multiple sclerosis or
rheumatoid arthritis.

Institute of Medicine, Adverse Effects of Vaccines: Evidence and Causality
(Kathleen Stratton et al., eds. 2012). Although reports from the Institute of
Medicine have informed decisions of special masters, see, e.g., Terran, 1998 WL
55290, at *10-12, mot. for review denied, 41 Fed. Cl. 330, 337 (1998), aff’d 195
F.3d 1302 (Fed. Cir. 1999), the decision in Ms. Koehn’s case does not depend
upon the views of the Institute of Medicine.
46
In addition to the relative qualifications of the experts, both sides suggest
that the other side’s expert may be biased. Neither of these arguments found their
targets because both Dr. McCabe and Dr. Rose appeared to offer sincerely held
opinions.
Nevertheless, Dr. McCabe derives more than 95 percent of his income from
participating in litigation. Tr. 34. In Ms. Koehn’s words, his “professional
activities revolve in large measure around participation in litigation.” Pet’r Reply
Br. at 3. This concentration leaves Dr. McCabe open to a challenge that he is
simply a professional witness.

54
credentials and background), the Secretary makes a particular argument for prong
two. The Secretary contends that he “is not qualified to independently provide
medical testimony and evidence on this issue.” 47 Resp’t Br. at 13.

Ms. Koehn replies that Dr. McCabe’s opinion should be given “substantial
weight.” Pet’r Reply Br. at 3. Ms. Koehn notes that Dr. McCabe earned a Ph.D in
microbiology and immunology. Id. While an assistant professor at Wayne State
University, he researched cytokines. Tr. 21-22. When he moved to the University
of Rochester School of Medicine and Dentistry, he led researchers who were
exploring how vaccines “modulate the immune response.” Tr. 20-21. Ms. Koehn
argues that Dr. McCabe’s specific training in immunology makes him “more
qualified” than Dr. Rose “to discuss the effects of vaccines on cell biology.” Pet’r
Reply Br. at 2-3.

Dr. McCabe is qualified to discuss immunologic principles and that
expertise naturally fits in the discussion of theory under prong one of Althen.
However, when those principles are applied to Vanessia specifically as part of the
prong-two analysis, Dr. McCabe’s inexperience with diagnosing diseases in human
beings becomes more problematic. Dr. McCabe does not have the experience of
Dr. Rose, who has diagnosed and treated 150-200 patients with sJIA. Tr. 278.
Thus, when it comes to evaluating their opinions, Dr. Rose’s opinion is given more
weight.

Dr. Rose’s opinion is that Gardasil did not cause Vanessia’s sJIA. To him,
Vanessia’s Gardasil vaccinations and her development of sJIA were “unrelated
events.” Tr. 208. This opinion is persuasive.

4. Summary

The Althen prong two analysis is necessary only if it is found (or assumed)
that the petitioner met the burden regarding Althen prong one. In the present case,
Ms. Koehn’s evidence on prong one was not persuasive. Hence, the foregoing

47
Despite this argument, the Secretary did not file a Daubert-type motion to
exclude his testimony. Such a motion to exclude testimony is relatively rare in the
Vaccine Program. Fresco, 2013 WL 364723, at *21; Garcia v. Sec’y of Health &
Human Servs., No. 05-720V, 2010 WL 2507793, at *2 (Fed. Cl. Spec. Mstr. May
19, 2010).

55
analysis about Vanessia’s case was undertaken for the sake of completeness and to
ensure that the entire record was considered.

The evidence about Vanessia does not persuasively show that she developed
sJIA because of Gardasil. Her treating doctors gave her the third dose of Gardasil
after she had been diagnosed with sJIA and the treating doctors continued to
recommend other vaccinations to her. These vaccinations did not clearly
exacerbate Vanessia’s sJIA as might be expected if the Gardasil vaccine were
causative.

As discussed in the context of Althen prong three, a sequence in which the
vaccination preceded the development of the disease does not establish causation.
In some cases, the disease followed the vaccination only as a matter of
coincidence. See Capizzano, 440 F.3d at 1327 (recognizing the possibility of
coincidence).

Dr. Verstraeten anticipated that coincidence and causation might be
confused. He wrote:

Bearing in mind the background incidence of autoimmune disorders
in adolescents and young adult population, it seems likely that, with
broader use of HPV vaccines or other vaccines targeting this age
group, autoimmune disorders will be reported in temporal association
with vaccine administration even in the absence of a causal
relationship.

Exhibit E (Verstraeten) at 6633. Vanessia’s case fits this description. Ms. Koehn
has accurately reported that Vanessia’s sJIA started after the vaccination but she
has not established the necessary “causal relationship.”

VII. Conclusion

Through the testimony of Dr. McCabe, Ms. Koehn has presented some
evidence on each of the Althen prongs. However, Dr. McCabe’s opinions are not
persuasive. Ms. Koehn has not established, under a more likely than not standard,
that the two doses of Gardasil caused Vanessia to suffer sJIA.

56
 She is not entitled to compensation. The Clerk’s Office is instructed to enter
judgment in accord with this decision unless a motion for review is filed.

IT IS SO ORDERED.

s/Christian J. Moran
Christian J. Moran
Special Master

57