United States Court of Appeals
for the Federal Circuit
______________________
ROCHE MOLECULAR SYSTEMS, INC.,
Plaintiff-Appellant
v.
CEPHEID,
Defendant-Appellee
______________________
2017-1690
______________________
Appeal from the United States District Court for the
Northern District of California in No. 3:14-cv-03228-EDL,
Magistrate Judge Elizabeth D. Laporte.
______________________
Decided: October 9, 2018
______________________
STEPHEN S. RABINOWITZ, Hughes Hubbard & Reed
LLP, New York, NY, argued for plaintiff-appellant. Also
represented by JAMES W. DABNEY, MITCHELL EPNER,
PATRICE POLYXENE JEAN, DAVID E. LANSKY, LYNN M.
RUSSO.
ERIK R. PUKNYS, Finnegan, Henderson, Farabow,
Garrett & Dunner, LLP, Palo Alto, CA, argued for de-
fendant-appellee. Also represented by MICHAEL PAUL
BARKER; GRANT L. KIM, WESLEY ELLSWORTH OVERSON,
Morrison & Foerster LLP, San Francisco, CA.
______________________
2 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
Before O’MALLEY, REYNA, and HUGHES, Circuit Judges.
Opinion for the court filed by Circuit Judge REYNA.
Concurring opinion filed by Circuit Judge O’MALLEY.
REYNA, Circuit Judge.
Appellant Roche Molecular Systems, Inc. (“Roche”)
owns U.S. Patent No. 5,643,723 (“the ’723 patent”), titled
“Detection of a Genetic Locus Encoding Resistance to
Rifampin in Microbacterial Cultures and in Clinical
Specimens.” The United States District Court for the
Northern District of California found that the asserted
claims of the ’723 patent are directed to patent-ineligible
subject matter and are therefore invalid under 35 U.S.C.
§ 101. Roche appeals from a grant of summary judgment
of invalidity. We affirm.
I. THE ’723 PATENT
The ’723 patent is directed to methods for detecting
the pathogenic bacterium Mycobacterium tuberculosis
(“M. tuberculosis” or “MTB”). ’723 patent col. 2 ll. 50–54.
MTB infection is a major cause of tuberculosis. Id. col. 1
ll. 13–30. In 1994, before the priority date of the ’723
patent, the general method of MTB detection in a tubercu-
losis patient was known as sputum examination by the
acid-fast bacilli smear. For this test, a biological sample
taken from a patient is subjected to cell culture in a
process that can take three to eight weeks. Id. col. 2 ll. 9–
11. This test has limitations: it can identify the presence
of bacterial cells in a biological sample, but cannot identi-
fy the cells as MTB. There is a need to know whether the
MTB from a patient is resistant to antibiotics. The
standard of care for MTB treatment at the time involved a
regimen of antibiotics, with rifampin being a first-line
anti-tuberculosis drug. Id. col. 1 ll. 31–33. Tuberculosis
outbreaks, however, still resulted because of delays in
diagnosis and reporting of rifampin-resistant tuberculosis
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 3
due to the inability to rapidly identify MTB strains that
are resistant to rifampin and put a patient on an appro-
priate alternative therapy. Id. col. 1 ll. 61–65.
Prior to the ’723 patent, scientists in the field had
been working on diagnostic tests for faster detection of
MTB, particularly rifampin-resistant MTB strains. Id.
col. 2 ll. 18–46. It was speculated that “[g]enotypic detec-
tion of multi-drug resistant MTB [strains] directly from
clinical specimens is theoretically the fastest and most
direct step toward determining effective therapy for
patients.” Id. col. 2 ll. 39–42. It was known in the art
that rifampin has a unique site of action on a particular
gene that encodes the β subunit of bacterial RNA poly-
merase (“the rpoB gene”). Id. col. 1 ll. 31–42. The rpoB
gene is present in MTB and other bacterial species, and
its deoxyribonucleic acid (“DNA”) sequences were known
to be highly conserved, with little variation from one
bacterial species to another. In 1994, single site muta-
tions in the rpoB gene that confer rifampin resistance in
some bacteria, such as Escherichia coli (“E. coli”), were
well characterized, making rpoB a prime candidate for
studying rifampin resistance in MTB. Id. col. 1 ll. 42–52.
The inventors of the ’723 patent—scientists from
Roche and the Mayo Foundation for Medical Education
and Research (“Mayo”)—sequenced the rpoB gene from
various bacteria species, including MTB, obtained from a
commercial vendor. 1 Id. col. 8 ll. 1–3 and col. 8 l. 15–col. 9
l. 20. After comparing rpoB DNA sequences across differ-
ent species, the inventors discovered that the rpoB gene
in MTB contains eleven “position-specific ‘signature
nucleotides’” (i.e., naturally occurring single nucleotide
1 Mayo later assigned its rights in the ’723 patent to
Roche. See Roche Molecular Sys., Inc. v. Cepheid, No. 14-
CV-03228-EDL, 2017 WL 6311568, at *1 (N.D. Cal. Jan.
17, 2017).
4 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
mutations) that are only present in MTB but not in other
bacteria. Id. col. 2 l. 60–col. 3 l. 2. In other words, these
naturally occurring signature nucleotides are like finger-
prints of MTB: if an investigator detects one of the eleven
signature nucleotides from a biological sample, she knows
the sample contains MTB, and vice versa. These signa-
ture nucleotides, therefore, could be used to identify MTB
using genetic testing, which is both faster and more
accurate than the traditional MTB detection methods. Id.
col. 2 ll. 9–31.
Based on these eleven MTB-specific signature nucleo-
tides, the Roche inventors devised a diagnostic test that
could (1) identify whether or not a biological sample
contains MTB, and (2) if MTB is present, predict whether
that MTB is a strain that is resistant to rifampin treat-
ment. The diagnostic test of the ’723 patent involves
subjecting DNA extracted from a biological sample taken
from a patient (e.g., a tissue or fluid sample) to amplifica-
tion by polymerase chain reaction (“PCR”) using a short,
single-stranded nucleotide sequence (a “primer”) that can
hybridize (i.e., bind) to at least one of the eleven position-
specific signature nucleotides in the MTB rpoB gene.
PCR is a method of amplifying DNA exponentially.
See Roche, 2017 WL 6311568, at *2. In PCR, a pair of
primers effectively “flanks,” or marks the start and finish
of, the DNA segment—e.g., the rpoB gene or a portion of
it—to be copied. Strands of DNA are then replicated
between the primer pair by a DNA polymerase. This
process is repeated until a sufficient number of copies of
the desired DNA segment are generated. These copies,
known as “amplification product,” make it possible to
detect whether a specific type of DNA is present. Id. It is
undisputed that by the time of the invention in 1994, PCR
had become a well-understood, routine, and conventional
technique. Id.
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 5
After PCR is performed, the presence of DNA amplifi-
cation product in sufficient copies from the reaction
indicates that MTB is present in the biological sample.
The absence of DNA amplification product (i.e., below the
detection limit using standard assays) indicates that MTB
is absent from the biological sample. The amplified rpoB
DNA segment from the PCR can, in turn, be tested for the
presence of known genetic mutations associated with
rifampin resistance. Thus, the ’723 patent represents an
improvement over the traditional sputum examination
method for detecting MTB, as its genetics-based diagnos-
tic method is faster and more accurate.
The ’723 patent provides two types of claims: (1) com-
position-of-matter claims for the primers used in the PCR,
which could hybridize to the rpoB gene of MTB at a site
that includes at least one of the eleven signature nucleo-
tides (“the primer claims”); and (2) process claims for
methods for detecting MTB that include amplifying target
sequences by PCR and detecting amplification products,
which, if present, indicate the presence of MTB (“the
method claims”).
Claims 1–13 are the method claims. Claim 1, the sole
independent method claim, recites:
1. A method for detecting Mycobacterium tubercu-
losis in a biological sample suspected of containing
M. tuberculosis comprising:
(a) subjecting DNA from the biological
sample to polymerase chain reaction
[PCR] using a plurality of primers under
reaction conditions sufficient to simplify a
portion of a M. tuberculosis rpoB [gene] to
produce an amplification product, wherein
the plurality of primers comprises at least
one primer that hybridizes under hybrid-
izing conditions to the amplified portion
of the [gene] at a site comprising at least
6 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
one position-specific M. tuberculosis sig-
nature nucleotide selected, with reference
to FIG. 3 (SEQ ID NO: 1), from the group
consisting
a G at nucleotide position 2312,
a T at nucleotide position 2313,
an A at nucleotide position 2373,
a G at nucleotide position 2374,
an A at nucleotide position 2378,
a G at nucleotide position 2408,
a T at nucleotide position 2409,
an A at nucleotide position 2426,
a G at nucleotide position 2441,
an A at nucleotide position 2456, and
a T at nucleotide position 2465; and
(b) detecting the presence or absence of an
amplification product, wherein the pres-
ence of an amplification product is indica-
tive of the presence of M. tuberculosis in
the biological sample and wherein the ab-
sence of the amplification product is indic-
ative of the absence of M. tuberculosis in
the biological sample.
’723 patent col. 25 l. 57–col. 27 l. 6. 2 Dependent claims 2–
13 add various limitations to claim 1 concerning PCR,
PCR analysis, and primer preparation details.
2 Adenine (“A”), Thymine (“T”), and Guanine (“G”),
together with Cytosine (“C”), constitute the four nucleo-
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 7
Claims 17–20 are the primer claims. Independent
claim 17 is representative and recites:
17. A primer having 14–50 nucleotides that hy-
bridizes under hybridizing conditions to an M. tu-
berculosis rpoB [gene] at a site comprising at least
one position-specific M. tuberculosis signature nu-
cleotide selected, with reference to FIG. 3 (SEQ ID
NO: 1), from the group consisting of [the same 11
nucleotides at the positions disclosed in claim 1].
Id. col. 28 ll. 14–31. Dependent claims 18–20 each add
further limitations. 3 Id. col. 28 ll. 32–46. Dependent
claim 20, for example, discloses full DNA sequences of
certain primers. Id. col. 28 ll. 44–46.
II. DISTRICT COURT PROCEEDING
Appellee Cepheid makes and sells “Xpert® MTB/RIF
Assay,” an assay that can detect MTB in a biological
sample and can identify rifampin-resistant MTB. Roche
brought a patent infringement case against Cepheid
asserting that Cepheid’s product infringed the ’723 pa-
tent. Roche, 2017 WL 6311568, at *1. Cepheid filed a
motion for summary judgment, arguing that all of the
asserted claims claim patent-ineligible subject matter
under 35 U.S.C. § 101. Id. at *9.
The district court granted Cepheid’s motion. Id. at
*19. The district court found that the primer claims of the
’723 patent, “which have genetic sequences identical to
those found in nature, are indistinguishable from those
held to be directed to nonpatentable subject matter” and
tide building blocks of DNA. See Roche, 2017 WL
6311568, at *2.
3 The remaining claims of the ’723 patent, claims
14–16 and 21–23, are also primer claims, but they are not
asserted in this litigation.
8 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
are thus invalid. Id. at *14. The district court also found
that the method claims are invalid because they are
directed to “nonpatentable laws of nature or natural
phenomena” and “the use of newly developed, non-
patentable primers to bind to newly identified naturally
occurring signature nucleotides . . . using the well-known,
routine process of PCR in a conventional way does not
transform the claimed methods into” patent-eligible
subject matter. Id. at *16–17.
Roche timely appealed. We have jurisdiction under 28
U.S.C. § 1295(a)(1).
III. DISCUSSION
We review the grant of summary judgment under the
law of the regional circuit, in this case the Ninth Circuit.
Charles Mach. Works, Inc. v. Vermeer Mfg. Co., 723 F.3d
1376, 1378 (Fed. Cir. 2013). The Ninth Circuit reviews
the grant or denial of summary judgment de novo. Leever
v. Carson City, 360 F.3d 1014, 1017 (9th Cir. 2004). We
also review de novo the question of whether a claim is
invalid under 35 U.S.C. § 101. Voter Verified, Inc. v.
Election Sys. & Software LLC, 887 F.3d 1376, 1384 (Fed.
Cir. 2018).
The only issues on appeal are whether the aforemen-
tioned primer claims and the method claims of the ’723
patent are patent-ineligible within the meaning of § 101.
Section 101 provides that “[w]hoever invents or discovers
any new and useful process, machine, manufacture, or
composition of matter, or any new and useful improve-
ment thereof, may obtain a patent therefor, subject to the
conditions and requirements of this title.” 35 U.S.C.
§ 101. There are certain exceptions to this provision: laws
of nature, natural phenomena, and abstract ideas are not
patent-eligible subject matter. Alice Corp. v. CLS Bank
Int’l, 134 S. Ct. 2347, 2354 (2014) (collecting cases).
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 9
The Supreme Court has established a two-step
framework for distinguishing patents that claim laws of
nature, natural phenomena, and abstract ideas from
those that claim patent-eligible applications of those
concepts. Id. at 2355 (citing Mayo Collaborative Servs. v.
Prometheus Labs., Inc., 566 U.S. 66, 77–79 (2012)).
Under the Alice/Mayo two-step framework, we first
“determine whether the claims at issue are directed to one
of those patent-ineligible concepts.” Id. (emphasis added);
see also Enfish, LLC v. Microsoft Corp., 822 F.3d 1327,
1335 (Fed. Cir. 2016). “[T]he ‘directed to’ inquiry applies
a stage-one filter to claims, considered in light of the
specification, based on whether ‘their character as a whole
is directed to excluded subject matter.’” Enfish, 822 F.3d
at 1335 (quoting Internet Patents Corp. v. Active Network,
Inc., 790 F.3d 1343, 1346 (Fed. Cir. 2015)). At step one,
“it is not enough to merely identify a patent-ineligible
concept underlying the claim; we must determine whether
that patent-ineligible concept is what the claim is ‘di-
rected to.’” Rapid Litig. Mgmt. Ltd. v. CellzDirect, Inc.,
827 F.3d 1042, 1050 (Fed. Cir. 2016).
If a claim is directed to one of those patent-ineligible
concepts, we move to step two of the Alice/Mayo inquiry
to “examine the elements of the claim to determine
whether it contains an ‘inventive concept’ sufficient to
‘transform’ the claimed abstract idea into a patent-eligible
application.” Alice, 134 S. Ct. at 2357 (quoting Mayo, 566
U.S. at 72–73, 78). At step two, there must be a further
inventive concept to take the claim into the realm of
patent-eligibility. Id. For claims that encompass natural
phenomena, the method steps are the “additional features
that must be new and useful.” Ariosa Diagnostics, Inc. v.
Sequenom, Inc., 788 F.3d 1371, 1377 (Fed. Cir. 2015); see
also Genetic Techs. Ltd. v. Merial L.L.C., 818 F.3d 1369,
1376 (Fed. Cir. 2016), cert. denied, 137 S. Ct. 242 (2016).
Following the Alice/Mayo framework, we address the
primer claims and the method claims in turn.
10 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
A. The Primer Claims
Roche argues that at Alice/Mayo step one, the primer
claims (claims 17–20) are patent-eligible because they are
directed to artificial, man-made primers that are different
from naturally occurring DNA. Roche, 2017 WL 6311568,
at *10. Specifically, Roche argues that the claimed pri-
mers have both a 3-prime end and a 3-prime hydroxyl
group, while the naturally occurring bacterial MTB DNA
contains neither of these. Id. at *11. The district court
rejected Roche’s arguments and found that “the primer
claims in this case, which have genetic sequences identi-
cal to those found in nature, are indistinguishable from
those held to be directed to nonpatentable subject matter
in In Re BRCA1.” Id. at *14 (citing In re BRCA1- &
BRCA2-Based Hereditary Cancer Test Patent Litig., 774
F.3d 755, 760 (Fed. Cir. 2014) (“BRCA1”)). We agree.
BRCA1 forecloses Roche’s arguments. There, this
court examined the subject matter eligibility of similar
primer claims and held that those primers “are not dis-
tinguishable from the isolated DNA found patent-
ineligible in Myriad” and thus are not patent-eligible.
BRCA1, 774 F.3d at 760 (citing Ass’n for Molecular Pa-
thology v. Myriad Genetics, Inc., 569 U.S. 576, 591 (2013)).
It is well established that primers are short, single-
stranded nucleic acid molecules that bind to their com-
plementary nucleotide sequence. 4 Id. at 761; see Roche,
2017 WL 6311568, at *10. As this court found in BRCA1,
“[p]rimers necessarily contain the identical sequence of
4 Complementarity refers to the inherent property
of nucleotides that causes them to pair with each other:
Adenine pairs with Thymine, and Cytosine pairs with
Guanine. J.A. 1129–30. This property allows investiga-
tors who know the sequence of a single strand of DNA to
accurately predict the sequence of the opposite DNA
strand. Id.
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 11
the [nucleotide] sequence directly opposite to the [DNA]
strand to which they are designed to bind. They are
structurally identical to the ends of DNA strands found in
nature.” 774 F.3d at 760. The court further held:
A DNA structure with a function similar to that
found in nature can only be patent eligible as a
composition of matter if it has a unique structure,
different from anything found in nature. Myriad,
133 S. Ct. at 2116–17. Primers do not have such a
different structure and are patent ineligible.
Id. at 761 (citation omitted).
It is undisputed that the primers before us have the
identical nucleotide sequences as naturally occurring
DNA, just like the primers found subject matter ineligible
in BRCA1. Appellant Br. 23 (admitting that “the claimed
primers include DNAs with the same nucleotide sequence
as portions of the MTB rpoB gene”). Nothing in the ’723
patent suggests that the Roche inventors introduced any
mutations that would have made the primers’ nucleotide
sequences different from those found in nature. Thus,
Roche’s primers are indistinguishable from their corre-
sponding nucleotide sequences on the naturally occurring
MTB rpoB gene.
Roche argues that the claimed primers are nonethe-
less patent-eligible because they “are chemically and
structurally distinct from any nucleic acid that occurs in
nature or that can be isolated from naturally occurring
DNA.” Appellant Br. 23. According to Roche, its claimed
primers have a 3-prime end and a 3-prime hydroxyl
group, which are absent in naturally occurring DNA.
This distinction is unavailing. The same argument was
raised in the opening brief in BRCA1. Roche, 2017 WL
6311568, at *14 (citing Appellant Br. at 11–12, 65–66,
BRCA1 774 F.3d 775 (Nos. 14–1361, –1366), 2014 WL
1668324). This court rejected this argument, holding that
the primers at issue were patent-ineligible subject matter.
12 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
BRCA1, 774 F.3d at 761. As we held in BRCA1, “it makes
no difference that the identified gene sequences are
synthetically replicated.” Id. at 760. It was undisputed
that the primers in BRCA1 contain 3-prime ends and 3-
prime hydroxyl groups, exactly as Roche’s primers in this
case. Roche, 2017 WL 6311568, at *14. Thus, except for
the nucleotide sequences, the primers before us are not
chemically or structurally different from the primers that
we held patent-ineligible in BRCA1.
Roche also contends that BRCA1 is distinguishable
because, as a bacterium, MTB has “a circular chromo-
some, which has neither a 3-prime end nor a 3-prime
hydroxyl [group],” while “[t]he primers at issue in BRCA1
were derived from human DNA, in which each chromo-
some occurs as a linear molecule.” Appellant Br. 23, 31.
Roche’s emphasis on the chromosome is misplaced. The
shape of MTB’s chromosomes is not relevant to the in-
quiry on the subject matter eligibility of the claimed
primers. As this court determined in BRCA1, the subject
matter eligibility inquiry of primer claims hinges on
comparing a claimed primer to its corresponding DNA
segment on the chromosome—not the whole chromosome.
774 F.3d at 760–61 (emphasizing the appropriate compar-
ison being between the primers and “the relevant portion
of the naturally occurring sequence” (emphasis added)).
Neither the claims nor the written description of the ’723
patent contain any reference to the circular nature of
MTB chromosomes. Indeed, Roche’s expert admitted that
“whether a chromosome is linear or circular makes no
difference in designing a primer.” Roche, 2017 WL
6311568, at *12. Therefore, at Alice/Mayo step one, we
find that the asserted primers are indistinguishable from
naturally occurring DNA and that the primer claims are
directed to a natural phenomenon.
Roche next argues that its primers can be distin-
guished from the patent-ineligible primers of BRCA1
because they can hybridize to only one of eleven position-
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 13
specific signature nucleotides on the MTB rpoB gene.
Appellant Br. 24. This is an Alice/Mayo step two argu-
ment: Roche is arguing that the specificity of its primers
to the eleven signature nucleotides would “transform” the
claimed naturally occurring phenomenon into patent-
eligible subject matter. But Roche’s emphasis on hybrid-
izing to particular DNA sequences is unavailing. A
primer that is otherwise patent-ineligible does not gain
subject matter eligibility simply because it can selectively
hybridize to a certain position of naturally occurring
DNA, because a primer having an identical nucleotide
sequence to naturally occurring DNA without further
chemical modification is a natural phenomenon. See
BRCA1, 774 F.3d at 760. Here, the primers before us
have no further chemical modification. 5
The eleven position-specific signature nucleotides on
the MTB rpoB gene that Roche’s primers are designed to
hybridize to are naturally occurring; the Roche inventors
identified these eleven positions after sequencing MTB
DNA. ’723 patent col. 12 ll. 1–23. In other words, Roche
identified these pre-existing position-specific signature
nucleotides; it did not create them. There is no doubt that
Roche’s discovery of these signature nucleotides on the
MTB rpoB gene and the designing of corresponding pri-
mers are valuable contributions to science and medicine,
5 We do not address the subject matter eligibility of
primers that have been altered—e.g., investigator-
induced mutation(s) such that their nucleotide sequences
are not found in nature, or primers which are chemically
modified or labeled by investigators such that they cannot
be isolated directly from naturally occurring DNA. See
Myriad, 569 U.S. at 596 (“Scientific alteration of the
genetic code presents a different inquiry, and we express
no opinion about the application of § 101 to such endeav-
ors.”).
14 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
allowing for faster detection of MTB in a biological sample
and testing for rifampin resistance. However,
“[g]roundbreaking, innovative, or even brilliant discovery
does not by itself satisfy the § 101 inquiry.” Myriad, 569
U.S. at 591; Ariosa, 788 F.3d at 1379. The primers are
not patent-eligible because they can be found in nature,
not because they are not valuable scientific discoveries.
We hold that the primers before us are indistinguish-
able from their corresponding nucleotide sequences on the
naturally occurring DNA, and that the primer claims,
therefore, are patent-ineligible within the meaning of
§ 101. We next address the asserted method claims.
B. The Method Claims
At Alice/Mayo step one, the plain language of the as-
serted method claims, viewed in light of the written
description, demonstrates that they are directed to natu-
rally occurring phenomena. The method claims disclose a
diagnostic test based on the observation that the presence
of the eleven position-specific signature nucleotides of the
naturally occurring MTB rpoB gene indicates the pres-
ence of MTB in a biological sample. Claim 1, the sole
independent method claim, provides “[a] method for
detecting Mycobacterium tuberculosis in a biological
sample” and contains two steps: (1) PCR amplification of
DNA (“the amplification step”), and (2) determination of
the presence of MTB based on the presence or absence of
PCR amplification product (“the detecting step”). ’723
patent col. 25 l. 57–col. 27 l. 6. The amplification step
subjects DNA from a biological sample—naturally occur-
ring matter—to amplification by PCR using primers that
are designed to hybridize to at least one of the eleven
naturally occurring position-specific signature nucleotides
on the MTB rpoB gene. Id. at col. 25 l. 60–col. 26 l. 67.
The primers, as discussed above, are indistinguishable
from their corresponding naturally occurring segments on
DNA. The detecting step of claim 1 is a mental determi-
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 15
nation step: if a PCR amplification product is detected,
MTB is present in the biological sample, and vice versa.
Id. at col. 27 ll. 1–6.
Claim 1 establishes that the method claims are di-
rected to a relationship between the eleven naturally
occurring position-specific signature nucleotides and the
presence of MTB in a sample. In other words, the method
claims assert that if an investigator detects a signature
nucleotide from a sample, she knows the sample contains
MTB. This relationship between the signature nucleo-
tides and MTB is a phenomenon that exists in nature
apart from any human action, meaning the method claims
are directed to a natural phenomenon, which itself is
ineligible for patenting.
The written description supports the conclusion that
the method claims of the ’723 patent are directed to a
patent-ineligible natural phenomenon. In the Summary
of the Invention, the patent states:
This invention involves a comparative analysis of
the rpoB sequences in MTB, other mycobacteria
and related . . . bacteria . . . demonstrating the
heretofore undiscovered presence of a set of MTB-
specific position-specific “signature nucleotides”
that permits unequivocal identification of MTB
....
’723 patent col. 2 ll. 60–65 (emphasis added). The lan-
guage makes clear what the inventors’ discovery entails:
the revelation of a previously undiscovered natural phe-
nomenon. Id. The patent further states that “[i]t was
found upon inspection of the sequence alignment that
there were eleven sites . . . at which the nucleotide ob-
served for Mycobacterium tuberculosis (MTB) differed
from all or most related organisms.” Id. col. 12 ll. 18–23.
Because these signature nucleotides are naturally occur-
ring, we conclude that the method claims, as informed by
16 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
the written description, are directed to a patent-ineligible
natural phenomenon. We turn to Alice/Mayo step two.
We hold that the method claims do not contain an in-
ventive concept that transforms the eleven position-
specific signature nucleotides of the MTB rpoB gene into
patent-eligible subject matter. PCR amplification of DNA
is the main action step of the method claims. ’723 patent
col. 25 l. 60–col. 26 l. 67. The district court found, and
Roche does not challenge, that “the background tech-
niques of PCR amplification and detection were ‘routine’
when the patent application was filed in 1994.” Roche,
2017 WL 6311568, at *16 (emphasis added); Appellant Br.
45 (“In the case at bar, the technique of PCR was well
known in the prior art.”). Indeed, the ’723 patent itself
makes clear that “[t]he methods of the present invention
use standard PCR techniques.” ’723 patent col. 3 ll. 65–66
(emphasis added). Neither the claims nor the written
description of the ’723 patent disclose any “new and
useful” improvement to PCR protocols or DNA amplifica-
tion techniques in general. The detecting step of claim 1
is similarly devoid of an inventive concept because it
involves a simple mental determination of the presence of
MTB based on the presence or absence of a PCR amplifi-
cation product.
Roche asserts that the method claims constitute more
than a patent-ineligible natural phenomenon. Roche
argues that at the time of the invention, it was “not
routine or conventional to use PCR (or any other genetic
test) to detect the presence of MTB in a biological sample”
and “unprecedented to perform PCR using the type of
primer specified in claims 1 through 13.” Roche, 2017 WL
6311568, at *15; Appellant Br. 42.
While it may be true that Roche inventors were the
first to use PCR to detect MTB in a biological sample,
being the first to discover a previously unknown naturally
occurring phenomenon or a law of nature alone is not
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 17
enough to confer patent eligibility. Many groundbreak-
ing, innovative, and brilliant discoveries have been held
patent-ineligible. See e.g., Mayo, 566 U.S. at 73–77 (dis-
covery of natural correlation between level of certain
metabolites and drug dosage, resulting in claimed method
of optimizing treatment using standard techniques to
administer the drug and then check if the metabolite level
indicated the need for a dosage change); Genetic Techs.,
818 F.3d at 1374 (discovery of natural correlation between
non-coding regions of DNA and the presence of an allele
in the coding region, resulting in claimed method of
detecting alleles using standard PCR to amplify and
detect); Ariosa, 788 F.3d at 1373 (discovery of natural
phenomenon that pregnant women’s blood contains
cffDNA, resulting in claimed methods using standard
techniques to amplify and detect cffDNA in maternal
blood). The Supreme Court in “Mayo made clear that
transformation into a patent-eligible application requires
more than simply stat[ing] the law of nature while adding
the words ‘apply it.’” Ariosa, 788 F.3d at 1377 (internal
quotation marks omitted) (quoting Mayo, 566 U.S. at 72).
Alice/Mayo step two’s requirement of “additional features
that must be new and useful” is simply not met in this
case because the asserted method claims recite standard
PCR methods applied to a naturally occurring phenome-
non; there is no additional innovation. See id.
Roche argues that to use its primers to detect MTB “is
no less an inventive act than to make a specific artificial
drug that is effective to treat an MTB infection.” Appel-
lant Br. 26 (emphasis omitted). We disagree. It is a well-
established law of nature that “complementary nucleotide
sequences bind to each other.” BRCA1, 774 F.3d at 761.
Roche’s method claims exploit the same law of nature—
the primer binds to its complementary nucleotide se-
quence on the MTB rpoB gene. This court’s holding in
BRCA1 applies to this case with equal force: the primers
“do not perform a significantly new function. Rather,
18 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
[they are] used to form the first step in a [PCR] chain
reaction—a function that is performed because the primer
maintains the exact same nucleotide sequence as the
relevant portion of the naturally occurring sequence.” Id.
Thus, unlike a method of treating a disease with a new
drug, Roche’s method claims do not involve “a significant-
ly new function” for the primers. 6 Id.
This case is distinguishable from CellzDirect, where
this court vacated a district court’s decision that the
method claims at issue were ineligible for patenting. 827
F.3d at 1052. This court held that while the claims were
based on the discovery of a natural phenomenon (the
ability of certain liver cells, or hepatocytes, to survive
multiple freeze cycles), they were “directed to a new and
useful laboratory technique for preserving hepatocytes”—
namely, freezing and thawing hepatocytes twice even
though the prior art taught away from this process. Id. at
1046–51 (emphasis added) (distinguishing cases that did
not invent a new and useful method based on the discov-
ery of a natural phenomenon). This court held that “[t]his
type of constructive process, carried out by an artisan to
achieve ‘a new and useful end,’ is precisely the type of
6 DNA or RNA can sometimes be used as a drug.
For example, technologies such as RNA interference
(“RNAi”) and small inhibitory RNAs (“siRNAs”) use RNA
to silence the expression of individual genes. See, e.g.,
Univ. of Utah v. Max-Planck-Gesellschaft Zur Forderung
Der Wissenschaften E.V., 734 F.3d 1315, 1318 (Fed. Cir.
2013) (an attorneys’ fees case not involving subject matter
eligibility of the claims); Ex parte Reich, No. 2013-004817,
2016 WL 750325, at *1 (P.T.A.B. Feb. 24, 2016); Ex parte
Khvorova, No. 2012-010359, 2015 WL 4267897, at *1
(P.T.A.B. July 10, 2015). We express no opinion on sub-
ject matter eligibility of method claims that exploit DNA
or RNA for drug-like new applications.
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 19
claim that is eligible for patenting. . . . [The inventors]
employed their natural discovery to create a new and
improved way of preserving hepatocyte cells for later use.”
Id. at 1048 (emphasis added) (quoting Alice, 134 S.Ct. at
2354). Unlike the method claims of the ’723 patent, the
invention in CellzDirect went beyond applying a known
laboratory technique to a newly discovered natural phe-
nomenon, and instead created an entirely new laboratory
technique that “is not simply an observation or detection”
based on the natural phenomenon. Id. (emphasis added).
In contrast, the ’723 patent claims a method of detection
based on a natural phenomenon and employs only con-
ventional, well-known laboratory techniques, which are
the opposite of those at issue in CellzDirect. 7
7 Similarly, this case is also distinguishable from
Vanda Pharmaceuticals Inc. v. West-Ward Pharmaceuti-
cals International Ltd., where this court found that the
claimed methods of treatment are subject matter eligible
because they claim “‘a new way of using an existing drug’
that is safer for patients” with schizophrenia at Al-
ice/Mayo step one. 887 F.3d 1117, 1135 (Fed. Cir. 2018)
(quoting Mayo, 566 U.S. at 87). But see id. at 1142 (Prost,
C.J., dissenting) (stating that the treatment claims are
“no more than an optimization of an existing treatment of
schizophrenia, just as the claims in Mayo”). Vanda “un-
derscore[s] the distinction between method of treatment
claims and those in Mayo,” i.e., claims “directed to a
diagnostic method.” Id. at 1134–35 (majority opinion). In
contrast, Roche’s method claims, consisting of a standard
PCR amplification step and a mental determination step,
are not directed to a method of treatment. Every time an
investigator practices Roche’s claimed invention—
detecting the presence or absence of the eleven signature
nucleotides of MTB rpoB gene in a sample—she is simply
rediscovering a preexisting natural phenomenon. Unlike
20 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
Roche attempts to distinguish its invention from the
patent-ineligible method claims in BRCA1. In BRCA1,
this court invalidated claims 7 and 8, which were directed
to methods of diagnosing genetic mutations of the BRCA1
gene in patients, as subject matter ineligible. 8 774 F.3d
at 763–65. Relevant here, BRCA1 distinguished the
subject matter ineligible method claims 7 and 8 from the
potential subject matter eligibility of a different method
claim, claim 21, which was not asserted and was thus not
at issue; in dicta, the court pointed out that claim 21
claims a method of detecting alterations in which
the alterations being detected are expressly iden-
tified in the specification by tables 11 and 12.
These tables expressly identify ten predisposing
mutations of the BRCA1 gene sequence discovered
by the patentees. Thus, the detection in claim 21
is limited to the particular mutations the inven-
tors discovered: detecting ten specific mutations
from the wild-type, identified as “[p]redisposing
[m]utations,” for the specific purpose of identify-
ing increased susceptibility to specific cancers.
Claims 7 and 8 are significantly broader and more
abstract, as they claim all comparisons between
the patient’s BRCA genes and the wild-type BRCA
genes.
Id. at 765 (footnote and citations omitted).
in Vanda, Roche does not claim a method of treatment
based on an underlying natural phenomenon, but the
natural phenomenon itself. Id. at 1135.
8 Briefly, claim 7 requires “1) hybridizing a BRCA
gene probe and 2) detecting the presence of a hybridiza-
tion product. Similarly, claim 8 requires 1) amplification
of the BRCA1 gene and 2) sequencing of the amplified
nucleic acids.” BRCA1, 774 F.3d at 764.
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 21
Roche argues that its method claims are analogous to
claim 21, and distinguishable from claims 7 and 8 in
BRCA1, because they require primers hybridizing to one
of eleven signature nucleotides expressly identified in the
claims, and thus would not preempt or limit use of other
DNAs for detecting MTB. Appellant Br. 34, 50; see ’723
patent col. 26 ll. 58–68. To be clear, BRCA1 did not find
claim 21 subject matter eligible; in response to the par-
ties’ arguments, this court emphasized that “we express
no view” on whether claim 21 is subject matter eligible,
and simply noted that “claim 21 is qualitatively different
from” claims 7 and 8. 774 F.3d at 765. Roche is mistaken
that limiting the scope of an otherwise ineligible method
claim would necessarily confer subject matter eligibility.
Roche’s attempt to limit the breadth of the method claims
by showing alternative uses of MTB DNA outside of the
scope of the claims “does not change the conclusion that
the claims are directed to patent ineligible subject mat-
ter.” See Ariosa, 788 F.3d at 1379. “While preemption
may signal patent ineligible subject matter, the absence of
complete preemption does not demonstrate patent eligibil-
ity.” Id. Thus, the method claims before us cannot gain
subject matter eligibility solely because they are limited
to specific signature nucleotides.
Therefore, we hold that the asserted method claims of
the ’723 patent are patent-ineligible because they are
directed to a natural phenomenon and lack any inventive
concept that transforms them into patent-eligible subject
matter.
IV. CONCLUSION
For the foregoing reasons, we affirm the district
court’s summary judgment ruling.
AFFIRMED
COSTS
No costs.
United States Court of Appeals
for the Federal Circuit
______________________
ROCHE MOLECULAR SYSTEMS, INC.,
Plaintiff-Appellant
v.
CEPHEID,
Defendant-Appellee
______________________
2017-1690
______________________
Appeal from the United States District Court for the
Northern District of California in No. 3:14-cv-03228-EDL,
Magistrate Judge Elizabeth D. Laporte.
______________________
O’MALLEY, Circuit Judge, concurring.
I agree with the majority that our decision in In re
BRCA1- & BRCA2-Based Hereditary Cancer Test Patent
Litigation, 774 F.3d 755, 758 (Fed. Cir. 2014) (“BRCA1”)
compels the conclusion that the primer and method
claims of U.S. Patent No. 5,643,723 (“the ’723 patent”) are
not eligible for patent protection.
I write separately to express my belief that we should
revisit our holding in BRCA1 at least with respect to the
primer claims. Specifically, I believe that our holding
there was unduly broad for two reasons: (1) the question
raised in BRCA1 was narrower than our holding in that
case; and, (2) our interpretation of the nature and func-
2 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
tion of DNA primers lacked the benefit of certain argu-
ments and evidence that the patent owner presents in
this case.
First, the question before us in BRCA1 was not
whether the primer claims were patent-ineligible, but,
rather, whether the district court abused its discretion
when it denied the patent owner’s motion for a prelimi-
nary injunction. 774 F.3d at 757. After filing an in-
fringement suit in district court, the patent owner in
BRCA1 moved for a preliminary injunction arguing that it
was likely to succeed on the merits of its infringement
claim. In re BRCA1-, BRCA2-Based Hereditary Cancer
Test Patent Litig., 3 F. Supp. 3d 1213, 1256 (D. Utah
2014). The district court denied the motion, in part,
because it found that the accused infringer had “raised a
substantial question concerning whether [the patent
owner]’s [p]rimer [c]laims are drawn to patent ineligible
subject matter.” Id. at 1263. The district court made no
findings regarding whether the primer claims were indeed
patent ineligible. Id. It correctly acknowledged that, “[a]t
this early stage, . . . the court ‘does not resolve the validity
question,’ but instead assesses ‘the persuasiveness of the
challenger’s evidence, recognizing that it is doing so
without all the evidence that may come out at trial.’” Id.
at 1257 (quoting Titan Tire Corp. v. Case New Holland,
Inc., 566 F.3d 1372, 1377 (Fed. Cir. 2009)).
The patent owner appealed. We affirmed by expressly
concluding that the primer claims were directed to patent-
ineligible subject matter. 774 F.3d at 757 (“Because we
hold that these claims are directed to ineligible subject
matter under 35 U.S.C. § 101, we affirm and remand.”);
id. at 761 (“Primers . . . are patent ineligible”). But that
was not the question before us; it was only whether the
district court abused its discretion when it found that the
accused infringer raised a substantial question regarding
invalidity under § 101. Appellants’ Br., BRCA1, 2014 WL
1668324, Nos. 14-1361, -1366 (Fed. Cir. Apr. 18, 2014), at
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 3
*2 (listing in its “Statement of Issues” the question of
“[w]hether the district court erred in finding that Ambry
raised a substantial question that Myriad’s pair of primer
claims are not directed to patent eligible subject mat-
ter . . . .”).
This procedural context in BRCA1 is important. We
have routinely recognized that the question of whether an
accused infringer has raised a substantial question of
invalidity in the context of a motion for a preliminary
injunction—such as the question before the district court
in BRCA1—presents a different type of inquiry than the
question of whether an asserted claim is invalid—such as
the question that was before the district court on sum-
mary judgment in this case. Indeed, “[w]hile the eviden-
tiary burdens at the preliminary injunction stage track
the burdens at trial, importantly the ultimate question
before the trial court is different” because “[i]nstead of the
alleged infringer having to persuade the trial court that
the patent is invalid, at [the preliminary injunction]
stage[,] it is the patentee, the movant, who must persuade
the court that, despite the challenge presented to validity,
the patentee nevertheless is likely to succeed at trial on
the validity issue.” Titan Tire, 566 F.3d at 1377. Signifi-
cantly, “the trial court ‘does not resolve the validity ques-
tion, but rather must . . . make an assessment of the
persuasiveness of the challenger’s evidence, recognizing
that it is doing so without all of the evidence that may
come out at trial.” Id. (emphasis added) (citations omit-
ted).
This brings me to my second point. Because of the
procedural posture in BRCA1, it is hardly surprising that
we did not have the benefit of certain arguments and
evidence that the patent owner presents in this case when
we decided BRCA1. As noted, in BRCA1, we considered
whether claims directed to the following were patent-
eligible under § 101:
4 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
A pair of single-stranded DNA primers for deter-
mination of a nucleotide sequence of a BRCA1
gene by a polymerase chain reaction, the sequence
of said primers being derived from human chro-
mosome 17q, wherein the use of said primers in a
polymerase chain reaction results in the synthesis
of DNA having all or part of the sequence of the
BRCA1 gene.
774 F.3d at 759 (quoting U.S. Patent No. 5,747,282 col.
155, ll. 23–29). In concluding that these claims were not
eligible for patent protection, we relied primarily on the
Supreme Court’s decision in Ass’n for Molecular Pathology
v. Myriad Genetics, Inc., 569 U.S. 576 (2013), which
involved other claims related to the BRCA1 gene. See
BRCA1, 774 F.3d at 759–61.
In Myriad, the Supreme Court separately considered
the eligibility of claims directed to “isolated DNA” having
a specific genetic sequence, on the one hand, and those
directed to complementary DNA (“cDNA”)—synthetically
created DNA “which contains the same protein-coding
information found in a segment of natural DNA but omits
portions within the DNA segment that do not code for
proteins”—on the other. 569 U.S. at 580. The Court
began its analysis by observing that the patent owner in
Myriad “did not create or alter any of the genetic infor-
mation encoded in the BRCA1 and BRCA2 genes,” nor did
it “create or alter the genetic structure of DNA.” 569 U.S.
at 590. “Instead, [the patent owner]’s principal contribu-
tion was uncovering the precise location and genetic
sequence of the BRCA1 and BRCA2 genes within chromo-
somes 17 and 13.” Id. The Court concluded that, in light
of this “principal contribution,” the isolated DNA claims
were not eligible for patent protection, while the cDNA
claims, which do not occur in nature, were. Id. at 593–94.
The Court distinguished the isolated DNA claims
from the “modified bacterium” claims held patent-eligible
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 5
in Diamond v. Chakrabarty, 447 U.S. 303 (1980). Id. at
590–91. In the Court’s view, Chakrabarty’s claims were
eligible because his bacterium was “new ‘with markedly
different characteristics from any found in nature.’” Id.
(quoting Chakrabarty, 447 U.S. at 310). The patent
owner in Myriad, in contrast, “did not create anything”
when it isolated DNA having a particular sequence—
though “it found an important and useful gene, . . . sepa-
rating that gene from its surrounding genetic material is
not an act of invention.” Id. at 591. Critically, the Court
recognized that claims are not “saved by the fact that
isolating DNA from the human genome severs chemical
bonds and thereby creates a nonnaturally occurring
molecule”: the “claims are simply not expressed in terms
of chemical composition, nor do they rely in any way on
the chemical changes that result from the isolation of a
particular section of DNA.” Id. at 593. It went on to
explain that “[i]f the patents depended upon the creation
of a unique molecule, then a would-be infringer could
arguably avoid at least Myriad’s patent claims on entire
genes . . . by isolating a DNA sequence that included both
the BRCA1 or BRCA2 gene and one additional nucleotide
pair.” Id.
The Court reached a different conclusion with respect
to the patent owner’s cDNA claims. The Court began by
stating that the “creation of a cDNA sequence from
mRNA results in an exons-only molecule that is not
naturally occurring.” Id. at 594. It noted that, although
viruses can incorporate cDNA into the human genome in
rare instances, “[t]he possibility that an unusual and rare
phenomenon might randomly create a molecule similar to
one created synthetically through human ingenuity does
not render a composition of matter nonpatentable.” Id. at
594 n.8 (emphasis omitted). The Court rejected the
accused infringers’ argument that cDNA is not patent
eligible because “[t]he nucleotide sequence of cDNA is
dictated by nature, not by the lab technician,” writing
6 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
that “the lab technician unquestionably creates something
new when cDNA is made.” Id. at 595. This is because,
while “cDNA retains the naturally occurring exons of
DNA, . . . it is distinct from the DNA from which it was
derived” because the intron sequences are removed. Id.
Thus, the Court concluded that “cDNA is not a ‘product of
nature’ and is patent eligible under § 101, except insofar
as very short series of DNA may have no intervening
introns to remove when creating cDNA.” Id. “In that
situation, a short strand of cDNA may be indistinguisha-
ble from natural DNA.” Id.
In BRCA1, we discussed Myriad’s teachings, culmi-
nating in a summary of that case by citing the Supreme
Court’s observation that, “[t]o the extent that the exon-
only sequence does not exist in nature, the lab technician
‘unquestionably creates something new when cDNA is
made.’” 774 F.3d at 760 (quoting Myriad, 569 U.S. at
595). But, in the very next sentence, we concluded that
“[t]he primers before us are not distinguishable from the
isolated DNA found patent-ineligible in Myriad and are
not similar to the cDNA found to be patent-eligible.” Id.
We arrived at this conclusion based on two facts that we
perceived as entirely resolving the question of whether
primers are structurally identical to that which exists in
nature. We found that “[p]rimers necessarily contain the
identical sequence of the BRCA sequence directly opposite
to the strand to which they are designed to bind,” and
that “[t]hey are structurally identical to the ends of DNA
strands found in nature.” Id.
But it is not clear from the BRCA1 opinion or record
why we reached this conclusion. The lack of record evi-
dence underlying BRCA1’s conclusion on this point is
important in light of the record in this case. Specifically,
BRCA1 concludes that primers have “identical sequences”
to the natural DNA strands directly opposite the strands
to which they bind, but, as the record in this case reveals,
a finding that the two have identical sequences does
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 7
entirely resolve the question of whether they are struc-
turally identical because structure is not defined solely by
nucleotide sequence. 1 Nor is it clear how primers “are
structurally identical to the ends of DNA strands found in
nature.” Id. As I explain below, the additional facts in
this record, viewed in the light most favorable to Roche,
give rise to genuine issues of material fact 2 regarding
whether the claimed primers have a “unique structure,
different from anything found in nature,” and therefore,
challenge the conclusion that this entire class of mole-
cules is ineligible under § 101.
Roche developed a record in this case demonstrating
the ways in which the claimed primers differ structurally
from anything that occurs in nature. Roche first submit-
ted evidence supporting a finding that the claimed pri-
mers differ from primers that naturally occur in the
bacteria of the M. tuberculosis complex (“MTB”). Roche’s
expert explained that, unlike the claimed primers, the
naturally occurring MTB primers are never single-
stranded. J.A. 1892 at ¶ 88. Roche’s expert also ex-
plained that the naturally occurring MTB primers are
comprised of RNA whereas the claimed primers are
comprised of DNA. J.A. 1892 at ¶ 89. This means that
1 Notably, there can be no dispute that primers,
though complementary, are structurally different in
sequence from the strands to which they hybridize.
Indeed, a primer comprising a nucleotide sequence of
ATCG is complementary to, but unquestionably different
from, a natural DNA strand comprising a sequence of
TAGC.
2 See Berkheimer v. HP Inc., 890 F.3d 1369, 1370
(Fed. Cir. 2018) (Moore, J., concurring in the denial of
rehearing en banc) (“While the ultimate question of
patent eligibility is one of law, it is not surprising that it
may contain underlying issues of fact.”).
8 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
naturally occurring MTB primers differ chemically from
the claimed primers “both (i) in the type of sugar they
contain (ribose [in MTB primers] v. deoxyribose [in the
claimed primers]) and (ii) the sets of bases that they use
(A, C, G, U [in the MTB primers] v. A, C, G, T [in the
claimed primers]).” J.A. 1892 at ¶ 89. Finally, Roche’s
expert explained that “[n]aturally occurring MTB primers
are 3–10 nucleotides long, and thus differ structurally
from the claimed primers, which are at least 14 nucleo-
tides long.” J.A. 1892 at ¶ 90. This record evidence
supports a finding that the claimed primers differ from
naturally occurring MTB primers.
Roche also explained that the claimed primers differ
structurally from the native MTB rpoB gene. Roche
explained that because “the DNA of MTB occurs in the
form a circular chromosome,” the native MTB rpoB gene
lacks a 3-prime end with a 3-prime hydroxyl (-OH) group.
J.A. 1891 at ¶ 85. In contrast, the parties agree that the
claimed primers necessarily have hydroxyl groups at their
3-prime end, and that skilled artisans would recognize as
much. Roche’s expert explained that “[a] free hydroxyl
group at the 3’ end of the primer or extension product is
essential for DNA replication, because it provides a free
end to which the next nucleotide can be attached.” J.A.
1881 ¶ 40 (emphasis added). Indeed, “[a] DNA that lacks
a free hydroxyl group at the 3’ end cannot support replica-
tion and thus cannot serve as a primer.” J.A. 1881 ¶ 40;
see also J.A. 1884–85 ¶ 48 (“If a DNA does not have a free
hydroxyl at the 3’-end, it cannot be a primer.”).
Moreover, because the DNA of MTB found in nature
occurs in the form of a circular chromosome, and therefore
lacks any sort of end, J.A. 1876 at ¶ 26 (“If the DNA is in
the form of a closed circle, there are no . . . 3-prime ends,
nor a 3-prime hydroxyl group.”), the claimed primers
cannot be “structurally identical to the ends of DNA
strands found in nature,” as we concluded in BRCA1, 774
F.3d at 760. And, even if the DNA in MTB occurred in
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 9
linear strands, Roche’s expert testified that, in linear
strands of native DNA, only the last nucleotide in the
entire strand typically has a hydroxyl group at its 3-prime
end. Thus, based on the above evidence, Roche’s expert
concluded that, while a portion of the native MTB rpoB
gene has the same nucleotide sequence as the claimed
primers, the two differ chemically vis-à-vis the presence of
a 3-prime end with a 3-prime hydroxyl group at a
nonnaturally occurring location. See J.A. 1892 at ¶ 86
(“Thus, there is a chemical difference between the primer
and the longer DNA strand of which the primer has, in
part, the same sequence.”).
Said another way, although it is undisputed that all
the claimed primers here have nucleotide sequences that
are identical to segments of the naturally occurring rpoB
gene found in MTB, a genuine factual dispute exists as to
whether they have a materially different structure than
any DNA molecules typically found in nature. Cf. Myriad,
569 U.S. at 595 (deeming relevant whether short strands
of cDNA are “indistinguishable from natural DNA”);
Chakrabarty, 447 U.S. at 310 (holding claims patent-
eligible where “the patentee has produced a new bacte-
rium with markedly different characteristics from any
found in nature and one having the potential for signifi-
cant utility”).
Not only does the record demonstrate that the claimed
primers could be structurally different from that which
exists in nature, but the claims here also appear to be
distinguishable from the molecule that would result from
isolating the sequence of the strand directly opposite the
strand to which the claimed primers hybridize. This is
critical because, in Myriad, the Court explained that
claims directed to isolated DNA sequences “are not saved
by the fact that isolating DNA from the human genome
severs the chemical bonds that bind gene molecules
together.” 569 U.S. at 592. But, unlike the claims in
Myriad, which were neither “expressed in terms of chemi-
10 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
cal composition, nor” reliant “in any way on the chemical
changes that result from the isolation of a particular
section of DNA,” 569 U.S. at 592, the primer claims here,
by virtue of being directed to the well-known, structural
term “primer” and by virtue of including “14-50 nucleo-
tides that hybridize[] under hybridizing conditions” to at
least one signature nucleotide, ’723 patent, col. 28, ll. 14–
31, are expressed in terms of chemical composition and
are reliant on the presence of a 3-prime end and a 3-prime
hydroxyl group at a nonnaturally occurring location.
Therefore, Roche’s evidence regarding the presence of a 3-
prime end with a 3-prime hydroxyl group, coupled with
the claim language, support a finding that the claims here
are distinguishable from the DNA claims in Myriad.
These structural differences between the claimed
primers and that which exists in nature are not my only
concerns with our conclusion in BRCA1, however: we also
held that primers “do not perform a significantly new
function” than does naturally occurring DNA. 774 F.3d at
761. In reaching this conclusion, we rejected the patent-
ee’s contention that DNA, when part of the naturally
occurring genetic sequence, “stores the biological infor-
mation used in the development and functioning of all
known living organisms,” but when isolated as a primer,
the DNA fragment “prime[s], i.e., . . . serve[s] as a starting
material for a DNA polymerization process.” Id. We
disagreed, writing that:
[i]n fact, the naturally occurring genetic sequences
at issue here do not perform a significantly new
function. Rather, the naturally occurring material
is used to form the first step in a chain reaction—
a function that is performed because the primer
maintains the exact same nucleotide sequence as
the relevant portion of the naturally occurring se-
quence. One of the primary functions of DNA’s
structure in nature is that complementary nucleo-
tide sequences bind to each other. It is this same
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 11
function that is exploited here—the primer binds
to its complementary nucleotide sequence. Thus,
just as in nature, primers utilize the innate ability
of DNA to bind to itself.
Id. at 760–61 (emphasis added). We then concluded that
Myriad does not “confer[] patent eligibility on composition
of matter claims directed to naturally occurring DNA
strands under such circumstances.” Id. at 761. Thus, “[a]
DNA structure with a function similar to that found in
nature can only be patent eligible as a composition of
matter if it has a unique structure, different from any-
thing found in nature.” Id. (citations omitted). We,
therefore, held that “[p]rimers do not have such a differ-
ent structure and are patent ineligible.” Id.
Here, not only is there at least a genuine issue of ma-
terial fact as to whether the claimed primers have a
different structure from anything found in nature, the
record also suggests that the claimed primers may have a
different function from that of native DNA. Particularly,
record evidence shows that, unlike native DNA, which
merely stores genetic information and serves as a tem-
plate for replication, the claimed primers can selectively
hybridize, or bind, to specific nucleotides of a target
gene—here, the “signature nucleotides” of the MTB rpoB
gene. This function is reliant on the presence of the free
3-prime hydroxyl group at a nonnaturally occurring
location and allows scientists, among other things, to
amplify and detect the MTB rpoB gene using techniques
such as polymerase chain reaction (“PCR”). The fact that
claimed primers, once synthesized, “utilize the innate
ability of DNA to bind to itself,” id. at 761, to achieve this
selective hybridization should not render them wholesale
patent-ineligible. In Myriad, the Supreme Court ex-
plained that, although “[t]he nucleotide sequence of cDNA
is dictated by nature, not by the lab technician,” this is
irrelevant for § 101 purposes because “the lab technician
unquestionably creates something new when cDNA is
12 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
made.” 569 U.S. at 595. Similarly, this record contains
evidence that the lab technician creates something new
when the claimed primers are made, even though, once
made, the primers “utilize the innate ability of DNA to
bind to itself.” BRCA1, 774 F.3d at 760–61.
Cepheid argues on appeal that the patent owner in
BRCA1 pointed to the same differences that Roche points
to here. Specifically, Cepheid contends that the patent
owner in BRCA1 raised the same 3-prime hydroxyl group
argument that Roche makes here, and that we squarely
rejected that argument. But, that is not an accurate
characterization of the record in BRCA1. There, the
patent owner merely alluded to the structural distinction
between primers and native DNA when explaining the
functional differences between the two. Appellants’ Br.,
In re BRCA1, 2014 WL 1668324, at *50. The only refer-
ence the patent owner made to the free 3-prime hydroxyl
group was in its statement of facts in its opening brief
where it stated that, “[t]here are no short, single strands
of DNA with a free 3’-OH group in nature that can serve
as primers.” Id. at *8. Notably, when attempting to
establish the existence of structural differences between
native DNA and primers, the patent owner in BRCA1 did
not reference the presence of a hydroxyl group at a
nonnaturally occurring location and focused instead on
the fact that primers are designed and synthetically
created in a lab. Id. at *48–50. Therefore, perhaps be-
cause of the procedural posture in which the issue was
developed, the patent owner in BRCA1 did not develop a
record demonstrating that primers differ structurally
from native DNA based on the presence of a hydroxyl
group at a nonnaturally occurring location. Nor did we
expressly address this hydroxyl group “argument” in
BRCA1 or include language indicating that we had con-
sidered, but rejected, any such argument.
The patent owner in BRCA1 also never argued that
its claimed primers were structurally distinct from natu-
ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID 13
rally occurring primers. While the patent owner in
BRCA1 pointed out that “[t]here are no short, single
strands of DNA with a free 3’-OH group in nature that
can serve as primers” and that “[i]n natural DNA replica-
tion, RNA primers are used as the starting material,” it
never used these facts to demonstrate that the primers at
issue were structurally distinct from anything that exists
in nature. Appellants’ Br., BRCA1, 2014 WL 1668324, at
*8. Simply, the patent owner in BRCA1 did not make the
specific arguments Roche makes here.
These points were also not raised or addressed in
Myriad. While the patent owner there argued that native
DNA differs from isolated DNA because “[n]ative DNA
cannot be used as a molecular tool, such as a probe or a
primer,” it did not explain that isolated DNA differs
structurally from native DNA vis-à-vis a 3-prime end and
a 3-prime hydroxyl group at a nonnaturally occurring
location. Myriad’s Mem. of Law in Support of Mot. for
Summary J. and in Opp. to Pl’s Mot. for Summary J.,
Myriad Genetics v. Assoc. for Molecular Pathology, No.
1:09-cv-04515-RWS (S.D.N.Y. Dec. 23, 2009), ECF No.
153, 8–9. Significantly, and as explained above, the
claims at issue in Myriad were not reliant on the presence
of this 3-prime hydroxyl group at a nonnaturally occur-
ring location. Accordingly, Myriad could not have proper-
ly raised, and the Supreme Court could not have
considered, the particular points that Roche now raises.
See Myriad, 569 U.S. at 593 (“Myriad’s claims are simply
not expressed in terms of chemical composition, nor do
they rely in any way on the chemical changes that result
from the isolation of a particular section of DNA.”).
Therefore, unlike the appellants in Myriad and in
BRCA1, here, Roche submitted evidence of record that, at
the very least, raises genuine issues of material fact as to
whether there exists anything in nature that both has the
structure and performs the function of the claimed pri-
mers. For these reasons, while I agree with the majority
14 ROCHE MOLECULAR SYSTEMS, INC. v. CEPHEID
that the broad language of our holding in BRCA1 compels
the conclusion that the primer claims in this case are
ineligible under 35 U.S.C. § 101, I believe that holding
exceeded the confines of the issue raised on appeal and
was the result of an underdeveloped record in that case. I
believe, accordingly, that we should revisit our conclusion
in BRCA1 en banc.