Boston Heart Diagnostics Corporationv.HEALTH DIAGNOSTIC LABORATORY, INC

Patent Trial and Appeal Board

Jul 28, 2015

12106070 (P.T.A.B. Jul. 28, 2015)

Trials@uspto.gov Paper 11
571-272-7822 Entered: July 28, 2015

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

BOSTON HEART DIAGNOSTICS CORP.,
Petitioner,

v.

HEALTH DIAGNOSTICS LABORATORY, INC.,
Patent Owner.
____________

Case IPR2015-00557
Patent 8,119,358 B2
____________

Before ERICA A. FRANKLIN, RAMA G. ELLURU, and
CHRISTOPHER G. PAULRAJ, Administrative Patent Judges.

ELLURU, Administrative Patent Judge.

DECISION
Denying Institution of Inter Partes Review
37 C.F.R. § 42.108

Case IPR2015-00557
Patent 8,119,358 B2

2
I. BACKGROUND
Boston Heart Diagnostics Corporation (“Petitioner”) filed a Petition
requesting inter partes review of claims 1–10 of U.S. Patent No. 8,119,358 (Ex.
1001; “the ’358 patent”). Paper 1 (“Pet.”). Health Diagnostics Laboratory, Inc.
(“Patent Owner”) filed a Preliminary Response. Paper 9 (“Prelim. Resp.”). We
have jurisdiction under 35 U.S.C. §§ 6(b) and 314.
Under 35 U.S.C. § 314(a), an inter partes review may be instituted only if
“the information presented in the petition . . . shows that there is a reasonable
likelihood that the petitioner would prevail with respect to at least 1 of the claims
challenged in the petition.” See also 37 C.F.R. § 42.108(c).
For the reasons given below, on this record, we are not persuaded that
Petitioner has established a reasonable likelihood of prevailing with respect to at
least one challenged claim of the ’358 patent. Accordingly, we deny the petition.
A. Related Proceedings
The parties represent that the ’358 patent is the subject of the following
district court case: Health Diagnostic Laboratory, Inc. v. Boston Heart
Diagnostics Corporation, Case No.3:14-cv-00796 (E.D. Va.). Pet. 2; Paper 3;
Paper 5, 3. Patent Owner also avers that Application Nos. 13/826,398 and
14/559,058, which stem from the same application that led to the ’358 Patent, are
currently before the Office. Paper 5, 3.
B. The ’358 Patent
The ’358 patent is directed to biomarkers that can be used to predict the risk
that an individual will develop diabetes or for identifying members of a population
at risk of developing diabetes, and methods of calculating such risks, advising
individuals of such risks, providing diagnostic test systems for calculating such
risks, and various other embodiments. Ex. 1001, Abstract, 1:24–28, 4:10–16.
Case IPR2015-00557
Patent 8,119,358 B2

3
Relevant to the challenged claims, the ’358 patent Specification discloses methods
for determining an individual’s risk of developing a diabetic condition. The
method comprises determining the levels of particular biomarkers in an individual
and inputting those levels into a model (e.g., a mathematical equation) to derive a
value that correlates to a risk of developing a diabetic condition in the individual.
Important to our analysis, the model is developed by analyzing the levels of those
same biomarkers in a study population.
Specifically, the ’358 patent Specification acknowledges biomarkers, such
as glucose and HBA1c,1 were used in the prior art for diabetes risk assessment.
Ex. 1001, 2:30–54. The Specification further states that “several single time point
biomarker measurements have been attempted for the use of risk assessment for
future Diabetes,” such as C-Reactive Protein (“CRP”) and Interleukin-6 (“IL-6”),
used alone and as an adjunct to the measurement of HBA1c. Id. at 2:54–59. Those
tests were not adopted, however, “for practical reasons relating to clinical
performance, specifically poor specificity and high false positive rates.” Id. at
2:60–62. The Specification states “the [present] invention provides novel panels of
biomarkers which can be measured and used to evaluate the risk that an individual
will develop Diabetes in the future.” Id. at 4:17–19. According to the
Specification,
the invention includes [a] method of evaluating risk for developing a
diabetic condition, the method comprising: obtaining biomarker
measurement data, wherein the biomarker measurement data is
representative of measurements of biomarkers in at least one
biological sample from an individual; and evaluating risk for
developing a diabetic condition based on an output from a model,

1 Our decision refers, as do the cited references, to hemoglobin A1c
interchangeably as “HBA1c,” “HBA1C,” or “HbA1c.”
Case IPR2015-00557
Patent 8,119,358 B2

4
wherein the model is executed based on an input of the biomarker
measurement data; wherein said biomarkers are defined.
Id. at 5:15–23.
With respect to “model,” the Specification expressly states that “[t]he terms
‘formula,’ ‘algorithm,’ and ‘model’ are used interchangeably for any mathematical
equation, algorithmic, analytical or programmed process, or statistical technique
that takes one or more continuous or categorical inputs (herein called ‘parameters’)
and calculates an output value, sometimes referred to as an ‘index’ or ‘index
value.’” Id. at 19:9–14; see also id. at 15:33–36, 79:5–80:49; Fig. 34 (“flow
diagram of an example method for developing a model which may be used to
evaluate a risk of a person, or group of people for developing a diabetic
condition”). The Specification further explains that “[i]n some variations, the step
of evaluating risk comprises computing an index value using the model based on
the biomarker measurement data, wherein the index value is correlated with risk of
developing a diabetic condition in the subject.” Id. at 5:34–38.
C. Illustrative Claim
Petitioner challenges claims 1–10 of the ’358 patent. Claims 1 and 2 are
independent, and claims 3–10 depend directly or indirectly from claim 2; claim 10
is a multiple dependent claim that can depend from claims 2, 3, 4, or 9. Claim 1 is
illustrative of the claimed subject matter and recites:
1. A method comprising:
(a) obtaining measurements of biomarkers from at least one biological
sample isolated from an individual, wherein said biomarkers
comprise:
(i) at least three biomarkers, where three of the biomarkers are
selected from the RDMARKER sets listed in FIG. 6A; or
(ii) at least four biomarkers selected from RD MARKERS; or
(iii) at least three biomarkers, where two biomarkers are selected from
ADIPOQ; CRP; GLUCOSE; GPT; HBAlC; HSPAlB; IGFBPl;
Case IPR2015-00557
Patent 8,119,358 B2

5
IGFBP2; INS, LEP; and TRIG; and one biomarker is selected from
the ALLDBRISKS, CPs, and TLRFs of Table 1, Table 2, and
Table 3; or
(iv) at least three biomarkers, where at least one biomarker is selected
from GLUCOSE and HBAlC; at least one biomarker is selected
from ADIPOQ, CRP, GPT, HSPAlB, IGFBPl, IGFBP2, INS, LEP,
and TRIG; and at least one biomarker is selected from the
ALLDBRISKS, CPs, and TLRFs of Table 1, Table 2, and Table 3;
or
(v) at least three biomarkers, where at least two biomarkers are
selected from the biomarkers within the group consisting of Core
Biomarkers I and Core Biomarkers II and at least a third biomarker
is selected from any of the biomarkers listed in Table 4; or
(vi) ADIPOQ, GLUCOSE, CRP and one biomarker selected from the
group consisting of HBAlC, IGFBPl, IGFBP2, Insulin, LEP and
TRIG;
(b) calculating an index value from the output of a model, wherein the
inputs to said model comprise said measurements, and further
wherein said model was developed by fitting data from a
longitudinal study of a selected population of individuals and said
fitted data comprises levels of said biomarkers and an end point in
said selected population of individuals, wherein said end point is
selected from risk for developing a diabetic condition, the
diagnosis of a diabetic condition, response to a Diabetes-
modulating drugs, a surrogate diabetes endpoint, glucose class, a
complication of a diabetic condition; and
(c) administering to said individual a Diabetes-modulating drug.
Ex. 1001, 102:23–67 (claim limitations relevant to our anlaysis are italicized). As
Petitioner explains (Pet. 7–8), independent claims 1 and 2 of the ’358 patent
include Markush groups of Markush groups. In particular, claims 1 and 2 each
recite six alternative sets of biomarkers ((i) – (vi)). Because the six
alternative sets of biomarkers are separated by “or,” they represent six alternate
elements in a Markush group. Further, each of the six alternative biomarker sets of
claims 1 and 2 recites a Markush group of biomarkers.

Case IPR2015-00557
Patent 8,119,358 B2

6
D. References Relied Upon
Petitioner relies upon the following references (Pet. 3–4):
WO 02/48715 A2, Paul M. Ridker & JoAnn E. Manson, Inflammatory Markers as
Tools in the Detection and Prevention of Diabetes Mellitus and as Tools to Aid in
the Selection of Agents to be Used for the Prevention and Treatment of Diabetes,
The Brigham and Women’s Hospital, Inc., published June 20, 2002. (“BWH
Application”) (Ex. 1003)

Helle Harder et al., The Effect of Liraglutide, a Long-Acting Glucagon-Like
Peptide 1 Derivative, on Glycemic Control, Body Composition, and 24-h Energy
Expenditure in Patients with Type 2 Diabetes, DIABETES CARE, vol. 27, no. 8, pp.
1915–1921 (2004), available online August 2004 (“Harder”) (Ex. 1004)

Jonathan Krakoff, MD et al., Inflammatory Markers, Adiponectin, and Risk of Type
2 Diabetes in the Pima Indian, DIABETES CARE, vol. 26, no. 6, pp. 1745–1751
(2003), available online June 2003 (“Krakoff”) (Ex. 1005)

Robert S. Sherwin, MD et al., Prevention or Delay of Type 2 Diabetes, DIABETES
CARE, vol. 27, Supplement 1, pp. S47–S54 (2004), available online January 2004
(“ADA position statement”) (Ex. 1006)

Joachim Spranger et al., Adiponectin and Protection Against Type 2 Diabetes
Mellitus, LANCET, vol. 361, p. 226–228 (2003), January 18, 2003 (“Spranger”) (Ex.
1007)

Petitioner did not provide a declaration in support of its Petition.
E. The Asserted Challenges
Petitioner argues that challenged claims 1–10 of the ’358 patent are
unpatentable based on the following grounds (Pet. 5):
References Basis Claim(s)
BWH Application § 102(b) 1–9
BWH Application and Harder § 103(a) 10
Krakoff and ADA position statement § 103(a) 1–8
Krakoff, ADA position statement, and Harder § 103(a) 9 and 10
Spranger and ADA position statement § 103(a) 1–8
Spranger, ADA position statement, and Harder § 103(a) 9 and 10
Case IPR2015-00557
Patent 8,119,358 B2

7
II. ANALYSIS
A. Claim Interpretation
In determining whether to institute a review, we construe the claim terms
that are necessary for an understanding of whether the prior art teachings anticipate
or render obvious the claimed invention. In an inter partes review, a claim in an
unexpired patent is given its broadest reasonable interpretation in light of the
specification of the patent in which it appears. 37 C.F.R. § 42.100(b); see also In
re Cuozzo Speed Techs., LLC, No. 2014-1301, __F.3d __, 2015 WL 4097949, at
*5–8 (Fed. Cir. July 8, 2015) (“Congress implicitly approved the broadest
reasonable interpretation standard in enacting the AIA,” and “the standard was
properly adopted by PTO regulation.”). Under the broadest reasonable
interpretation standard, claim terms are given their ordinary and customary
meaning, as would be understood by one of ordinary skill in the art in the context
of the entire disclosure. In re Translogic Tech., Inc., 504 F.3d 1249, 1257 (Fed.
Cir. 2007). For purposes of this decision, we construe the following claim terms.
“model”
Petitioner contends that as used in the ’358 patent, “model” is “any
mathematical technique or algorithm that is used to calculate an index value.” Pet.
13 (citing Ex. 1001, 19:9–25). Patent Owner did not propose a contruction for
“model.” Prelim. Resp. 16–18. We agree, in part, with Petitioner’s proposed
construction for this claim term.
The ’358 patent Specification expressly states that “[t]he terms ‘formula,’
‘algorithm,’ and ‘model’ are used interchangeably for any mathematical equation,
algorithmic, analytical or programmed process, or statistical technique that takes
one or more continuous or categorical inputs (herein called ‘parameters’) and
Case IPR2015-00557
Patent 8,119,358 B2

8
calculates an output value, sometimes referred to as an ‘index’ or ‘index value.’”
Ex. 1001, 19:9–14.
Accordingly, for purposes of this decision, we construe “model” to mean
“any mathematical equation, algorithmic, analytical or programmed process, or
statistical technique that takes one or more continuous or categorical inputs (herein
called ‘parameters’) and calculates an output value, sometimes referred to as an
‘index’ or ‘index value.’” Id.
“index value”
Petitioner contends that “an index value covers any calculated value that can
be used to enumerate a risk of developing diabetes.” Pet. 12 (citing Ex. 1001,
8:47–57). Patent Owner argues that “index value” should be construed as “an
output value derived from a model, wherein the output value is correlated with the
risk of developing a diabetic condition, e.g., a Diabetes risk score.” Prelim Resp.
17 (citing Ex. 1001, 5:34–38, 8:47–48, Examples 4–7).
We decline to adopt either party’s proposed construction. Although both
parties agree that the “index value” correlates to or enumerates a risk of developing
diabetes or a diabetic condition, Petitioner construes the term as “any calculated
value,” while Patent Owner construes it as “derived from an output model.” The
claim language, however, expressly requires that the “index value” is “from the
output of a model” (claims 1 and 2). Thus, the claim language specifies how the
“index value” is derived, i.e., from a “model,” which we have construed above. In
addition, the ’358 patent Specification states that “the step of evaluating risk
comprises computing an index value using the model based on the biomarker
measurement data, wherein the index value is correlated with risk of developing a
diabetic condition in the subject.” Ex. 1001, 5:34–38 (emphasis added).
Case IPR2015-00557
Patent 8,119,358 B2

9
Based on the claim language and the Specification, we construe “index
value” as “a value that correlates to the risk of developing a diabetic condition.”
B. BWH Application (Ex. 1003)
Petitioner contends that claims 1–9 are anticipated by the BWH application
(Pet. 19–36) and claim 10 is unpatentable over the combination of the BWH
application and Harder (Pet. 36–42). Claims 1 and 2 recite “calculating an index
value from the output of a model,” “wherein the inputs to said model comprise
[measurements of biomarkers . . . wherein said biomarkers comprise at least three
biomarkers or at least four biomarkers from the specified biomarkers listed in
categories (i)–(vi)],” and “wherein said model was developed by fitting data from a
longitudinal study of a selected population of individuals and said fitted data
comprises levels of said biomarkers.” Ex. 1001, 102:23–104:12 (emphases
added). As indicated above, we construe “model” as “any mathematical equation,
algorithmic, analytical or programmed process, or statistical technique that takes
one or more continuous or categorical inputs (herein called ‘parameters’) and
calculates an output value, sometimes referred to as an ‘index’ or ‘index value,”
and “index value” as “a value that correlates to the risk of developing a diabetic
condition.” Further, the claim language expressly requires that the claimed
“model” must be developed by “fitting data” comprising the measured levels of
certain recited biomarkers from a study population and the model must output an
an index value.
According to Petitioner, “the BWH application teaches that traditional
diabetes markers, such as GLUCOSE, INSULIN, HBA1C, and BMI should be
combined with biomarkers of inflammation, such as C-reactive protein (“CRP”),
and interluken-6 (‘IL-6[’]) to provide a more complete picture of diabetes risk.”
Case IPR2015-00557
Patent 8,119,358 B2

10
Pet. 19 (citing Ex. 1003, 2:21–24, 2:29–32, 5:3–21). Petitioner further states the
following:
[u]sing the measured biomarker levels and the diabetes outcomes for
the study participants, the BWH application developed a diabetes risk
model. Ex. 1003 at 25:23-26:14. Using the “Student’s t-test,” the
collected body of statistical data was evaluated, after which the
biomarker distributions were adjusted using a rank-sum test. Ex. 1003
at 25:23-26. The authors next correlated each biomarker range with a
diabetes risk score based upon longitudinal outcome, i.e.,
development of diabetes within the first four years of the study. Ex.
1003 at 25: 26-28. Conditional logistic regressions, adjusting for BMI,
were then used to assign a diabetes risk score to weighted
combinations of biomarkers. Ex. 1003 at 25:28-26:14 and Table 5
(page 38). Table 4 of the BWH application shows adjusted relative
risks of developing diabetes based upon baseline biomarker
measurements. Ex. 1003 at Table 4 (page 37).
Pet. 21. In addition, Petitioner’s claim chart for the BWH application refers to
“[a]n index value (a.k.a. diabetes risk score) [that] was developed by using
statistical analysis on biomarker measurements” obtained from a longitudinal
study. Pet. 31 (citing Ex. 1003, 6:26–28, 24:1–25, 25:23–26:14). Based on the
record before us, Petitioner has not persuaded us sufficiently that the BWH
application discloses the claimed “model” for the following reasons.
The BWH application discloses case subjects “providing blood specimens
who were free of reported diabetes at enrollment and subsequently developed
newly diagnosed diabetes during a four-year observation period.” Ex. 1003, 24:9–
25. Control subjects (individuals free of self-reported diabetes at the time the case
subject reported the event of developing confirmed diabetes) were matched to
study subjects. Id. at 24:24–31. The baseline plasma and blood samples of the
case subjects were thawed and assayed for IL-6, CRP, insulin, and HbA1C. Id. at
24:1–8, 24:32–25:1 (because of the prevalence of undiagnosed diabetes among the
Case IPR2015-00557
Patent 8,119,358 B2

11
sample population, the BWH study limited case subjects to individuals with certain
baseline HBA1c levels), 25:10–21. Statistical analysis was then performed on
certain biomarkers. Id. at 25:23–26:14. It is this statistical analysis that Petitioner
contends discloses “a diabetes risk model.” Pet. 21 (citing Ex. 1003, 25:23–26:14),
Pet. 31 (claim chart) (citing Ex. 1003, 6:26–28, 24:1–25, 25:23–26:14), Pet. 24
(citing Pet. 21). Petitioner further contends that “the collective BWH biomarker
set of CRP, GLUCOSE, INS, HbA1C, and BMI anticipates five of the [claimed]
six alternative biomarker sets, i.e., (i)–(v).” Pet. 23.
Petitioner does not explain adequately, however, how the BWH
application’s statistical analysis discloses a “model,” as we construe that term, i.e.,
“any mathematical equation, algorithmic, analytical or programmed process, or
statistical technique that takes one or more continuous or categorical inputs (herein
called ‘parameters’) and calculates an output value, sometimes referred to as an
‘index’ or ‘index value.” Petitioner also does not explain how the alleged model
“was developed by fitting data” comprising the relied upon set of biomarkers, i.e.,
“CRP, GLUCOSE, INS, HbA1C, and BMI.” See Pet. 23–24. In other words,
Petitioner does not explain adequately how the statistical model was developed by
using the collective levels of all the biomarkers relied upon as the claimed set of
biomarkers to output an “index value,” i.e., “a value that correlates to the risk of
developing a diabetic condition.” Id.
Petitioner contends that “[c]onditional logistic regressions, adjusting for
BMI, were then used to assign a diabetes risk score to weighted combinations of
biomarkers.” Pet. 21 (citing Ex. 1003, 25:28–26:14, Table 5) (emphasis added).
Petitioner further contends that “Table 4 of the BWH application shows adjusted
relative risks of developing diabetes based upon baseline biomarker
measurements.” Pet. 21 (citing Ex. 1003, Table 4). We note also the BWH
Case IPR2015-00557
Patent 8,119,358 B2

12
application states that the risk assessment was adjusted for baseline fasting insulin
in additional to clinical parameters. Ex. 1003, 27:8–10, Table 4. Petitioner’s
argument is deficient, however, because the record does not explain sufficiently the
BWH application’s statistical analysis, and specifically what “adjusted for” entails
and how adjusting for certain parameters (e.g., BMI and insulin) discloses “fitting”
a model with these parameters. For example, Patent Owner disputes Petitoner’s
assertions and argues that all the biomarkers relied upon by Petitioner were not
analyzed by BWH to assess diabetes risk because some factors were “[a]djusted
for.” Prelim. Resp. 22–23. Thus, Petitioner’s contentions do not persuade us,
particularly when those contentions stand alone, without support from a declaration
of an asserted expert in the field explaining how the BWH application discloses a
“model.”
“A reference anticipates a claim if it discloses the claimed invention ‘such
that a skilled artisan could take its teachings in combination with his own
knowledge of the particular art and be in possession of the invention.’” In re
Graves, 69 F.3d 1147, 1152 (Fed. Cir. 1995) (citation and emphasis omitted). To
establish anticipation, each and every element in a claim, arranged as recited in the
claim, must be found in a single prior art reference. Net MoneyIN, Inc. v. VeriSign,
Inc., 545 F.3d 1359, 1369 (Fed. Cir. 2008). Therefore, based on the current record,
Petitioner has not adequately persuaded us that the BWH application anticipates
the claimed “model,” recited in independent claims 1 and 2.
For the same reasons, we are not persuaded adequately that the BWH
application anticipates dependent claims 3–9. With respect to Petitioner’s
challenge to claim 10 as having been obvious over the combination over the BWH
application and Harder, Petitioner relies on the same disclosure of the BWH
application as teaching the claimed “model.” Pet. 36–38. Thus, for the same
Case IPR2015-00557
Patent 8,119,358 B2

13
reasons discussed above, Petitioner has not persuaded us that the combination of
the BWH application and Harder teaches all the limitations of claim 10.
Accordingly, we are not persuaded that Petitioner has established a
reasonable likelihood of prevailing in showing that claims 1–9 are anticipated by
the BWH application and that claim 10 is rendered obvious by the combination of
the BWH application and Harder.
C. Krakoff (Ex. 1005)
Petitioner contends that claims 1–8 are unpatentable over the combination of
Krakoff and the ADA position statement (Pet. 38–45) and claims 9 and 10 are
unpatentable over the combination of Krakoff, the ADA position statement, and
Harder (Pet. 45–46).
Petitioner contends that “Krakoff teaches collecting a biological sample
(blood) and screening for a biomarker set comprising ADIPOQ, GLUCOSE, CRP,
HBA1C, Insulin, vWF, and BMI.” Pet. 41 (citing Ex. 1005, 1746, col. 2, l. 19–col.
3, l. 31, Table 1, Table 4). Petitioner further contends that this biomarker set
“anticipates each of the six alternative biomarker sets [(i)–(vi)] in claims 1 and 2.”
Pet. 41. Petitioner further states the following:
After Krakoff accumulated biomarker measurements for the case and
control groups, Krakoff used statistical analysis to determine
correlations between biomarker values and a risk of developing
diabetes. Ex. 1005 at 1746, col. 3, lines 33-66. First, the accumulated
biomarker measurements were log-transformed to achieve normalized
distributions and then a paired t-test was used to compare baseline
biomarker measurements between case and control subjects. Ex. 1005
at 1746, col. 3, lines 35-42. The incidence rate ratios (IRR) for
developing diabetes based upon the various biomarker levels were
then calculated using a conditional logistic regression that accounted
for the BMI/sex/age matching between case and control subjects. Ex.
1005 at 1746, col. 3, lines 49-53. Krakoff reported that adiponectin
levels were strongly correlated with diabetes risk. Ex. 1005 at 1748,
Case IPR2015-00557
Patent 8,119,358 B2

14
col. 1, line 10-col. 2, line 9; FIG. 1. Furthermore, a multivariate
analysis of biomarkers, including ADIPOQ, CRP, vWF, BMI, GLU,
HbA1c, and INS showed correlations between those biomarkers and a
risk of future diabetes. Ex. 1005 at 1748, col. 3, line 17- 1749, col. 1,
line 5; FIG. 2.
Pet. 39; see also Pet. 41–42 (“Krakoff, like the BWH application, also performed
statistical analysis on biomarker distributions from a longitudinal study to
determine a risk factor for developing diabetes.”). Based on the record before us,
Petitioner has not persuaded us sufficiently that Krakoff teaches the claimed
“model” for the following reasons.
Krakoff conducted a longitudinal study in which members of a community
were screened every two years for plasma glucose levels, as well as other
parameters. Ex. 1005, 1745, col. 3, ll. 35–49. Based upon accumulated data, “case
subjects” were identified as having developed type 2 diabetes during the study. Id.
at 1746, col. 1, ll. 22–32, Table 1. Each case subject was matched against a
“control subject” with similar body-mass index (“BMI”) that did not develop
diabetes over the period of the study. Id. Using samples collected from the
subjects at the beginning of the study, Krakoff determined baseline levels of
biomarkers for case and control subjects. Id. at 1746, col. 2, l. 19–col. 3, l. 31,
Table 1, Table 4. The biomarkers measured included BMI, HbA1c, insulin,
glucose, adiponectin, CRP, and von Wilebrand factor (“vWF”). Id. Statistical
analysis was then performed on certain biomarkers. Id. at 1746, col. 3, ll. 31–55.
It is this statistical analysis that Petitioner contends teaches the claimed “model.”
Pet. 39 (citing Ex. 1005, 1746, col. 3, ll. 33–66), 41–42 (citing Ex. 1005, 1746, col.
3, ll. 35–53, Table).
Petitioner does not explain adequately, however, how Krakoff’s statistical
analysis discloses a “model,” as we construe that term, i.e., “any mathematical
Case IPR2015-00557
Patent 8,119,358 B2

15
equation, algorithmic, analytical or programmed process, or statistical technique
that takes one or more continuous or categorical inputs (herein called ‘parameters’)
and calculates an output value, sometimes referred to as an ‘index’ or ‘index
value.” Petitioner also does not explain how the alleged model “was developed by
fitting data” comprising the relied-upon set of biomarkers, i.e., “ADIPOQ,
GLUCOSE, CRP, HBA1C, Insulin, vWF, and BMI.” See Pet. 41–42. In other
words, Petitioner does not adequately explain how the statistical model was
developed by using the collective levels of all the biomarkers relied upon as the
claimed set of biomarkers to output an “index value,” i.e., “a value that correlates
to the risk of developing a diabetic condition.” Id.
Petitioner contends that “[t]he incidence rate ratios (IRR) for developing
diabetes based upon the various biomarker levels were then calculated using a
conditional logistic regression that accounted for the BMI/sex/age matching
between case and control subjects.” Id. at 39 (citing Ex. 1005, 1746, col. 3, ll. 49–
53). Petitioner further contends that “a multivariate analysis of biomarkers,
including ADIPOQ, CRP, vWF, BMI, GLU, HbA1c, and INS showed correlations
between those biomarkers and a risk of future diabetes.” Id. (citing Ex. 1005,
1748, col. 3, l. 17–1749, col. 1, l. 5, Fig. 2). Petitioner’s argument is deficieint,
however, because the record does not explain sufficiently Krakoff’s statistical
analysis, and specifically the details of calculating the incidence rate ratios for
developing diabetes and the multivariate analysis and whether Krakoff uses all the
biomarkers Petitioner relies upon for these calculations and analysis. For example,
Patent Owner disputes Petitoner’s assertions and argues that “Krakoff looked at
each marker independent[ly] of the others evaluated, not collectively, and adjusted
for other factors to ensure that they did not interfere with the results.” Prelim.
Resp. 34. Thus, Petitioner’s contentions do not persuade us, particularly when
Case IPR2015-00557
Patent 8,119,358 B2

16
those contentions stand alone, without support from a declaration of an asserted
expert in the field explaining how the BWH application discloses a “model.”
Therefore, based on the current record, Petitioner has not persuaded us
adequately that the combination of Krakoff and the ADA position statement
teaches the claimed “model,” recited in independent claims 1 and 2. Petitioner
relies on the same teachings in support of its argument that dependent claims 3–8
are unpatentable over the combination of Krakoff and the ADA position statement
and that claims 9 and 10 are unpatentable over the combination of Krakoff, the
ADA position statement, and Harder. Pet. 43–46. For the same reasons, we are
not adequately persuaded that dependent claims 3–10 are unpatentable over the
asserted combinations including Krakoff.
Accordingly, we are not persuaded that Petitioner has established a
reasonable likelihood of prevailing in showing claims 1–8 are unpatentable over
the combination of Krakoff and the ADA position statement and claims 9 and 10
are unpatentable over the combination over the combination of Krakoff, the ADA
position statement, and Harder.
D. Spranger (Ex. 1007)
Petitioner contends that claims 1–8 are unpatentable over the combination of
Spranger and the ADA positon statement (Pet. 47–52) and claims 9 and 10 are
unpatentable over the combination of Spranger, the ADA position statement, and
Harder (Pet. 52–54).
According to Petitioner, Spranger teaches the “biomarker set of BMI,
adiponectin (ADIPOQ), HbA1c, and GLUCOSE,” which Petitioner contends
“discloses alternative biomarker sets (iii) and (iv) of claims 1 and 2 of the ’358
patent.” Pet. 48–49. Petitioner further states the following:
Case IPR2015-00557
Patent 8,119,358 B2

17
After Spranger accumulated biomarker measurements for the diabetic
and control groups, Spranger correlated the risk of diabetes to
measured adiponectin levels using statistical methods. Ex. 1007 at
227, col. 1, line 40-col. 2, line 8. Spranger used both a binning and
continuous model, and found roughly the same correlation between
risk of diabetes and adiponectin. Id. Spranger additionally performed
a statistical analysis incorporating all of the clinical variables that
were measured, including BMI, adiponectin (ADIPOQ), HbA1c, and
GLUCOSE, and found the same trend, but with a higher p value. Ex.
1007 at 227, Figure and caption.
Pet. 47–48. Based on the record before us, Petitioner has not persuaded us
sufficiently that Spranger teaches the claimed “model” for the following reasons.
Spranger teaches a prospective study “to assess whether baseline
concentrations of adiponectin in plasma independently modify the risk of type 2
diabetes in apparently healthy individuals.” Ex. 1007, 226, col. 2, ll. 49–51. Case
subjects were initially free of diabetes and developed type 2 diabetes within two to
three years, and each of the diabetic case subjects was matched with non-diabetic
control subjects. Id. at 227, col. 1, ll. 12–33, 30–35. “The main baseline
characteristics of individuals with incident type 2 diabetes and controls are
summarised in [a] table.” Id. at 227, col. 1, ll. 36–38. According to Petitioner,
“[a]fter Spranger accumulated biomarker measurements for the diabetic and
control groups,” including measurements for “BMI, adiponectin (ADIPOQ),
HbA1c, GLUCOSE, and lipid levels,” “Spranger correlated the risk of diabetes to
measured adiponectin levels using statistical methods.” Pet. 47 (citing Ex. 1007,
227, col. 1, ll. 35–37, col. 1, l. 40–col. 2, l. 8, Table). It is this statistical analysis
that Petitioner contends teaches the claimed “model.” Pet. 47–48.
Petitioner does not explain adequately, however, how Spranger’s statistical
analysis teaches a “model,” as we construe that term, i.e., “any mathematical
equation, algorithmic, analytical or programmed process, or statistical technique
Case IPR2015-00557
Patent 8,119,358 B2

18
that takes one or more continuous or categorical inputs (herein called ‘parameters’)
and calculates an output value, sometimes referred to as an ‘index’ or ‘index
value.” Petitoner also does not explain how the alleged model “was developed by
fitting data” comprising the relied upon set of biomarkers, i.e., “BMI, adiponectin
(ADIPOQ), HbA1c, and GLUCOSE.” See Pet. 49. In other words, Petitioner does
not explain adequately how the statistical model was developed by using the
collective levels of all the biomarkers relied upon as the claimed set of biomarkers
to oupt put an “index value,” i.e., “a value that correlates to the risk of developing
a diabetic condition.” Id.
Petitioner contends that “Spranger used both a binning and continuous
model, and found roughly the same correlation between the risk of diabetes and
adiponectin.” Pet. 47–48 (citing Ex. 1007, 227, col. 1, l. 40–col. 2, l. 8). Petitioner
further contends that “Spranger additionally performed a statistical analysis
incorporating all of the clinical variables that were measured, including BMI,
adiponectin (ADIPOQ), HbA1c, and GLUCOSE, and found the same trend, but
with a higher p value.” Id. at 48 (citing Ex. 1007, 227 (figure and caption)).
Petitioner’s argument is deficient, however, because the record does not adequately
describe Spranger’s statistical analysis. For example, Patent Owner disputes
Petitoner’s assertions and argues that Spranger’s analysis “‘adjusted’” for certain
factors and “factored out data from other markers to confirm the independent
predictive value of adiponectin.” Prelim. Resp. 45–46 (citing (Ex. 1007, 227, Figs.
(a)-(c)), 48. Thus, Petitioner’s contentions do not persuade us, particularly when
those contentions stand alone, without support from a declaration of an asserted
expert in the field explaining how the BWH application discloses a “model.”
Therefore, based on the current record, Petitioner has not persuaded us
adequately that the combination of Spranger and the ADA position statement
Case IPR2015-00557
Patent 8,119,358 B2

19
teaches the claimed “model,” recited in independent claims 1 and 2. Petitioner
relies on the same teachings in support of its argument that dependent claims 3–8
are unpatentable over the combination of Spranger and the ADA position statement
and that claims 9 and 10 are unpatentable over the combination of Krakoff, the
ADA position statement, and Harder. Pet. 50–54. For the same reasons, we are
not persuaded adequately that dependent claims 3–10 are unpatentable over the
asserted combinations including Spranger.
Accordingly, we are not persuaded that Petitioner has established a
reasonable likelihood of prevailing in showing claims 1–8 are unpatentable over
the combination of Spranger and the ADA position statement and claims 9 and 10
are unpatentable over the combination over the combination of Spranger, the ADA
position statement, and Harder.

III. CONCLUSION
For the forgoing reasons, we are not persuaded, based on the Petition and the
accompanying evidence, that Petitioner has shown a reasonable likelihood that
Petitioner will prevail on any of its challenges to claims 1–10 of the ’358 patent.

IV. ORDER
In consideration of the foregoing, it is
ORDERED that the Petition is denied, and no inter partes review is
instituted.

Case IPR2015-00557
Patent 8,119,358 B2

20
PETITIONER:

Thomas C. Meyers
Brown Rudnick LLP
tmeyers@brownrudnick.com

Brian D. Bean
Brown Rudnick LLP
bbean@brownrudnick.com

PATENT OWNER:

Sanya Sukduang
Finnegan, Henderson, Farabow, Garrett & Dunner, LLP
Sanya.Sukduang@finnegan.com

Lillian M. Robinson
Finnegan, Henderson, Farabow, Garrett & Dunner, LLP
Lillian.Robinson@finnegan.com