Ex Parte Reiter et al

Patent Trial and Appeal Board

Aug 3, 2017

13163701 (P.T.A.B. Aug. 3, 2017)

United States Patent and Trademark Office
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O.Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
13/163,701 06/19/2011 Yoram REITER 51341 1316
67801 7590 08/07/2017
MARTIN D. MOYNIHAN d/b/a PRTSI, INC.
P.O. BOX 16446
ARLINGTON, VA 22215
EXAMINER
DIBRINO, MARIANNE
ART UNIT PAPER NUMBER
1644
NOTIFICATION DATE DELIVERY MODE
08/07/2017 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
usptomail@ipatent.co.il
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte YORAM REITER and GALIT DENKBERG1
Appeal 2015-004942
Application 13/163,701
Technology Center 1600
Before TAWEN CHANG, RYAN H. FLAX, and TIMOTHY G. MAJORS,
Administrative Patent Judges.
MAJORS, Administrative Patent Judge.
DECISION ON APPEAL
This is an appeal under 35 U.S.C. § 134 involving claims to a soluble
antibody with a certain specificity and binding affinity. The Examiner
rejected the claims for failure to satisfy the written description requirement,
as anticipated, and for obviousness. We have jurisdiction under 35 U.S.C.
§ 6(b).
We AFFIRM.
1 Appellants identify the Real Party in Interest as Technion Research &
Development Foundation Limited. (App. Br. 2.)
Appeal 2015-004942
Application 13/163,701
STATEMENT OF THE CASE
According to the Specification, the “invention relates to an antibody
having a T-cell receptor specificity and higher affinity.” (Spec. 1:15—16.)
As background, the Specification explains that “expression of specific
peptides in complex with major histocompatibility complex (MHC) class I
molecules on cells was shown to be associated with cancer and autoimmune
disorders [] and viral infections.” {Id. at 1:20—22.) The Specification further
explains that, “[bjecause the specificity of the immune response is regulated
and dictated by these class I MHC-peptide complexes, it should be possible
to use these very specific and unique molecular cell-surface markers as
targets to eliminate [] cancer cells, while sparing normal cells.” {Id. at 1:32—
2:2.) The Specification describes “[o]ne promising approach is to generate
recombinant antibodies that will bind the MHC-peptide complex expressed
on the cancer cell[’]s surface with the same specificity as the TCR [(T-cell
antigen receptor)].” {Id. at 2:7—9.) Such antibodies could then be armed
with a cytotoxic moiety (e.g., drug, toxin, etc.). (Id. at 2:9-10.)
Claims 1, 3—7, and 15—17 are on appeal. Claims 1 and 17 are
illustrative and reproduced below:
1. An isolated soluble antibody specifically bindable with a
binding affinity below 10 nanomolar to a soluble human major
histocompatibility complex (MHC) class I protein complexed
with a HLA-restricted tumor antigen, wherein said antibody does
not bind said soluble human MHC class I in the absence of said
HLA-restricted tumor antigen, wherein said antibody does not
bind said HLA-restricted tumor antigen in the absence of said
soluble human MHC class I, and wherein said soluble antibody
binds with an antigen-specific, MHC-restricted specificity
peptide unloaded tumor cells naturally presenting human MHC
class I complexed with a HLA-restricted tumor antigen.
2
Appeal 2015-004942
Application 13/163,701
17. The isolated antibody of claim 1, wherein a monovalent
or dimer configuration of said soluble antibody binds said
complex when naturally presented on a cell.
(App. Br. 24—25 (Claims App.).) Appellants elected a claim group drawn to
a molecule comprising an antibody specifically bindable with an HLA class
I/antigen, and elected as a species an antibody bindable with the HLA-
A2/gl00 peptide G9-209M (SEQ ID NO:2). (See Election dated Jan. 12,
2012.)
The claims stand rejected as follows:
I. Claim 17 under 35 U.S.C. § 112, first paragraph (pre-AIA), for
failure to satisfy the written description requirement.
II. Claim 17 under 35 U.S.C. § 102(b) as anticipated by Verma.2
III. Claims 1, 3—7, and 15—17 under 35 U.S.C. § 103(a) over
Andersen,3 Kawakami,4 Chames,5 Rhode,6 Rhode ’445,7 and
Reiter.8
2 Verma et al., TCR Mimic Monoclonal Antibody Targets a Specific
Peptide/HLA Class I Complex and Significantly Impedes Tumor Growth In
Vivo Using Breast Cancer Models, 184 J. Immunol. 2156—65 (2010).
3 Andersen et al., WO 97/02342 Al, published Jan. 23, 1997.
4 Kawakami et al., WO 95/29193 A2, published Nov. 2, 1995.
5 Chames et al., Direct selection of a human antibody fragment directed
against the tumor T-cell epitope HLA-A1—MAGE-A1 from a nonimmunized
phage-Fab library, 97(14) PNAS 7969—74 (2000).
6 Rhode et al., Single-Chain MHC Class IIMolecules Induce T Cell
Activation and Apoptosis, 157 J. Immunol. 4885—91 (1996) (“Rhode”).
7 Rhode et al., US 6,232,445 Bl, issued May 15, 2001 (“Rhode ’445”).
8 Reiter, WO 01/72768 A2, published Oct. 4, 2001.
3
Appeal 2015-004942
Application 13/163,701
When Appellants filed their Reply Brief, they also requested an oral
hearing before the Board. The Board scheduled a hearing for July 6, 2017,
but Appellants waived attendance by communication dated June 14, 2017.
I - WRITTEN DESCRIPTION
The Examiner rejected claim 17 for including new matter, and thus
failing to comply with the written description requirement.9 (Ans. 3.)
According to the Examiner, “[t]he amendatory [claim] material [is] not
supported by the disclosure as originally filed.” (Id.) More specifically, the
Examiner finds the phrase “wherein a monovalent or dimer configuration of
said soluble antibody binds said complex when naturally presented on a cell”
is unsupported because “[t]he originally filed disclosure is in the context of
two species, scFv or Fab, rather than a generic antibody.” (Id.)
Appellants argue “[m]onomeric and dimeric configurations of
antibodies are well known in the art” and that “configurations and methods
of preparing [the] same are described in great detail in page 16 line 23
through page 18 line 15” of the Specification. (App. Br. 7 (also citing a
“text book manual for generating these antibodies” that is referenced in the
Specification).)
Appellants’ argument is unpersuasive. As the Examiner points out,
claim 17 recites a “monovalent or dimer configuration” not a “monomeric or
dimer configuration.” (Ans. 15—16.) The Specification describes a
“monovalent antigen-binding fragment of an antibody molecule” (Spec.
9 A rejection of claims 1, 3—7, 15, and 16, also under 35 U.S.C. § 112, first
paragraph, for lack of written description per new matter, was withdrawn by
the Examiner. (Ans. 14.)
4
Appeal 2015-004942
Application 13/163,701
16:27—28 (emphasis added)), but no intact monovalent antibody is described.
And the Examiner explains “an intact antibody has two antigen binding sites
and is necessarily divalent rather than monovalent.” (Ans. 16.)
On these points, Appellants do not appear to dispute, much less
persuasively rebut, the Examiner’s findings. (See App. Br. 7—9; Reply Br.
2.) But neither does the Examiner appear to dispute that monomer and
dimer antibody configurations are well known in the art, and that skilled
persons know how to make them. Capon v. Eshhar, 418 F.3d 1349, 1357
(Fed. Cir. 2005) (“The descriptive text needed to meet these requirements
varies with the nature and scope of the invention at issue, and with the
scientific and technologic knowledge already in existence.”) So, both
Appellants and the Examiner suggest an amendment to claim 17 —
replacing the word “monovalent” with “monomer” — as a way to potentially
overcome the rejection. (Compare Ans. 16 with Reply Br. 2.) The Board
does not as a general matter, however, decide the patentability of
hypothetically amended claims. We review the rejection of the claims
before us. And, on the present record, we agree with the Examiner that
claim 17 does not comply with the written description requirement.
II-ANTICIPATION
The Examiner rejected claim 17 as anticipated by Verma.10 (Ans. 3—
4; Final Act. 3—4.) The decisive issue here is whether claim 17 is entitled to
the putative priority date of a parent patent application (Feb. 13, 2002),
10 A rejection of claims 1,5, and 6, also under 35 U.S.C. § 102 for
anticipation by Verma, was withdrawn by the Examiner. (Ans. 15.)
5
Appeal 2015-004942
Application 13/163,701
which predates the Verma’s publication in 2010. (App. Br. 10.) Because
claim 17 includes new matter as explained above in Section I, we conclude
that Verma is prior art and claim 17 is not entitled to a priority date in 2002.
Cf. Noelle v. Lederman, 355 F.3d 1343, 1348 (Fed. Cir. 2004). We
otherwise agree with and adopt the Examiner’s findings concerning Verma’s
disclosures and affirm the rejection of claim 17 as anticipated.
In Appellants’ Reply Brief, they argue “Verma would not establish
anticipation . . . even if published earlier, because Verma does not disclose
or address the elected species in this case.” (Reply Br. 2—3.) Appellants
waived this argument by omitting it from the Appeal Brief, and waiting until
their reply to raise it. 37 C.F.R. § 41.41(b)(1) & (2); See Ex parte Borden,
93 USPQ2d 1473, 1477 (BPAI 2010) (informative) (“Properly interpreted,
the Rules do not require the Board to take up a belated argument that has not
been addressed by the Examiner, absent a showing of good cause”); Ex parte
Nakashima, 93 USPQ2d 1834, 1837 (BPAI 2010) (informative) (explaining
that arguments and evidence not timely presented in the principal brief will
not be considered when filed in a reply brief, absent a showing of good
cause explaining why the argument could not have been presented in the
principal brief). We thus decline to consider the argument in this appeal.11
11 Appellants are correct that, when a species has been elected, the Board
ordinarily limits its analysis of the claims to the patentability of that species.
(Reply Br. 3.) See Ex parte Ohsaka, 2 USPQ2d 1460, 1461 (BPAI 1987).
But, Appellants’ argument is untimely. Because it was not raised in the
Appeal Brief, the Examiner had no opportunity to respond to it in the
Answer, including by withdrawing the rejection if appropriate.
6
Appeal 2015-004942
Application 13/163,701
III - OBVIOUSNESS
Issue
Has the Examiner established by a preponderance of the evidence that
claims 1, 3—7, and 15—17 would have been obvious over Andersen,
Kawakami, Chames, Rhode, Rhode ’445, and Reiter?
Findings of Fact (FF)
The Examiner’s findings and statement of the rejection are provided at
pages 4—14 of the Examiner’s Answer. (See also Ans. 16—32; Final Act. 4—
8.) We provide the following findings for emphasis and easier reference.
FF 1. Andersen teaches an “invention [that] relates to a method of
producing an antibody or an antibody fragment specifically recognizing a
peptide-MHC complex.” (Andersen Abstract; see also id. at 1:10-13 (“The
present invention relates to recombinant antibodies and fragments of
antibodies with the antigen-specific, MHC-restricted specificity of T
cells.”).) Andersen further teaches “a pharmaceutical composition
comprising antibodies or antibody fragments according to the invention for
the prevention or treatment of infectious and autoimmune diseases, and
cancer.” {Id. at Abstract; see also id. at 4:15—5:5.)
FF 2. Andersen teaches that “a peptide-MHC complex” is “a peptide
which is capable of binding to a particular MHC molecule thereby
establishing a specific peptide MHC complex.” {Id. at 12:23—25.) Andersen
teaches “it is possible to produce numerous different antibodies by . . .
making appropriate variations with respect to the particular peptide and
MHC molecule used to generate or select the antibody in question.” {Id. at
12:26—29.) Andersen further teaches “‘specifically recognizing a peptide-
7
Appeal 2015-004942
Application 13/163,701
MHC complex’ means the antibody is capable of binding to the peptide -
MHC complex with an equilibrium dissociation constant, KD in the range of
lO'7 to lO'10 M.” (Id. at 12:30-33.)
FF 3. Andersen teaches “[t]he human major histocompatibility
complex (MHC), also designated the human leucocyte antigen (HLA)
system” includes two classes of alleles, “HLA class I alleles and HLA class
II alleles.” (Id. at 10:32-11:3.)
FF 4. Andersen, in examples, describes the generation and selection
of particular peptide-specific MHC restricted antibodies. (See, e.g., id. at
36-43.) Andersen teaches “[cjomplexes between purified mouse MHC class
I, Kk, and the Kk-restricted Influenza virus derived peptides, Haemaglutinin
(Ha255-262) were generated.” (Id. at 36:4—7.) The Examiner summarizes the
generation of the antibodies in Andersen’s examples as follows:
[T]he exemplified antibodies were produced by immunizing
non-transgenic mice with recombinant MHC/peptide complexes
(made from affinity-purified Kk MHC class I heavy chain
extracted from detergent solubilized cells and human p2m light
chain) of which about 80% were loaded with one desired peptide,
and a large phage display antibody library (i.e., about (107) was
produced from RNA of mice after a prime and boost
immunization with the complexes spaced two weeks apart.
(Ans. 6.) The Examiner finds Andersen “exemplifies producing several
anti-murine MHC class I/peptide antibodies in the range of 51 nM to 60
nM.” (Id.: see also Andersen 43:10—11 (“the resulting KD values were very
similar: 53, 56, 60 and 51 nM”).)
FF 5. Although Andersen exemplifies non-transgenic mouse (i.e.,
murine) MHC/peptide complexes, the Examiner finds Andersen also teaches
8
Appeal 2015-004942
Application 13/163,701
that “mice transgenic for the said HLA or MHC class I molecules with . . .
said antigenic peptide/HLA or peptide/MHC molecules” may be used for
generating antibodies by phage display technology. (Ans. 5,7.)
FF 6. Andersen teaches that, for isolating antibody specificity, “the
selection power of the phage display technology ... is particularly helpful.”
{Id. at 2:29—32.) Andersen teaches “increasing somatic hypermutations
result[s] in affinities in the nanomolar range” and that “phage display
antibody fragments isolated from immunization based libraries in general
have thousand fold higher affinities than those isolated from naive non-
immunized libraries.” {Id. at 3:4—9.) Andersen suggests results may be
improved by further enriching the MHC molecules. {Id. at 13:25—14:12;
Ans. 6.) For example, Andersen teaches “it may be advantageous to further
enrich the MHC molecules e.g. up to about 90%, about 95% or even about
98% or about 99%[] i.e. substantially 100% specific peptide-MHC loading.”
{Id. at 14:2-5.)
FF 7. Kawakami relates to melanoma antigens in humans and the
treatment of human cancers. (Kawakami Abstract and 1:10—15.) Kawakami
teaches the G9209 peptide is a modified immunogenic peptide from the
gplOO melanoma antigen, which binds to HLA-A2. (Kawakami 103:22—
104:22, 107 (Table 16), and 110 (Table 19).) Kawakami teaches generating
antibodies reactive with the peptide/antigen. {Id. at 4:34—5:1 and 5:28—33.)
FF 8. Chames teaches “a human antibody directed against a peptide
encoded by gene melanoma-associated antigen (MAGE)-Al and presented
by HLA-A1 molecules” is obtained by “a large phage Fab antibody
repertoire on a recombinant version of the complex HLA-A1-MAGE-A1
9
Appeal 2015-004942
Application 13/163,701
produced by in vitro refolding.” (Chames Abstract.) Chames teaches
“results indicate that nonimmunized phage Fab libraries are a source of
antibodies with a T cell antigen receptor-like specificity” and that “[t]he
human anti-HLA-Al-MAGE-Al antibody . . . may prove very useful for
monitoring the cell surface expression of these complexes, and eventually,
as a targeting reagent for the specific immunotherapy of HLA-A1 patients
bearing a MAGE-A1-positive tumor.” (Id.)
FF 9. Chames teaches that, in designing the peptide/antigen complex
“[t]he conformation of the antigen has to be as ‘natural’ as possible.”
(Chames 7973, left col.) Chames teaches that,
several methods to produce a recombinant version of the
complex needed for [antibody] selection [were tested], including
secretion of a single-chain peptide—HLA molecule in E. coli
periplasm and expression in Drosophila cells . . . , but only in
vitro refolding from inclusion bodies produced in E. coli yielded
enough correctly folded protein.
(Id.)
FF 10. Chames teaches isolating antibodies binding to the complex,
“with one clone (G8) showing the capability to bind in a peptide-specific
manner.” (Id.) Chames discloses the “Fab-G8 has an affinity of 250 nM for
the complex HLA-A1-MAGE-A1” and that “[t]his rather low affinity was
not expected” given the size of the phage library. (Id. at 7973, right col.)
Chames teaches the affinity of 250 nM “is most likely too weak for in vivo
targeting purposes” and “[t]he next step is thus to mature the affinity of this
antibody without losing its fine specificity.” (Id.) Through “affinity
maturation of G8 by directed randomization of complementarity determining
region H3 and reselection” Chames discloses “a gain of up to 18-fold in
10
Appeal 2015-004942
Application 13/163,701
affinity without loss of peptide specificity has already been achieved.” (Id.)
The Examiner thus finds Chames teaches that modifying the antibody by
conventional affinity maturation techniques resulted in Fab with increased
affinity to 14 nM (250 nM/18). (Ans. 8—9.)
FF 11. Rhode teaches “MHC class II/peptide complexes displayed on
the surface of APCs play a pivotal role in initiating specific T cell
responses.” (Rhode Abstract.) Rhode teaches “components of this
heterotrimeric complex can be genetically linked into a single polypeptide
chain” and “findings reported here open up the possibility of producing large
amounts of stable sc [(single-chain)] class II/peptide fusion molecules for
structural characterization and immunotherapeutic applications.” (Id.)
FF 12. Rhode further teaches:
Production of multimeric proteins can be improved by
genetically linking their subunits into sc fusion molecules. This
strategy directs balanced production of functional domains and
may facilitate their association and proper folding. Numerous
reports have described the design and generation of sc Ab and
TCR molecules in which a flexible polypeptide linker joins the
two variable region subunits (20, 24). This approach has also
been applied to MHC class I molecules (21, 22). Connecting the
class I heavy chain to /^m allows the production of sc molecules
that have proven useful in the study of peptide binding and T cell
activation (21, 22, 42 44).
(Id. at 4889, right col.)
FF 13. Rhode ’445 teaches
a linked single-chain [MHC] complex can provide a number of
advantages. In particular, in reducing the complex to a single
molecule, yields and stability of the molecules are typically
enhanced. That can be especially important for soluble
molecules which may not be produced efficiently in active form.
11
Appeal 2015-004942
Application 13/163,701
. . . [T]he MHC complexes of the present invention include sc-
MHC molecules that can comprise a variety of class I (H-2 or
HLA) or class II. . . MHC molecules.
(Rhode ’445 6:67—7:12; see also id. at 3:3—8 and 3:36—37 (“The term ‘MHC
complexes of the invention’ or related term is used herein to denote the sc-
MHC class I and class II molecules . . . Rhode ’445 further teaches the
MHC complexes can be used to immunize mammals and to generate
antibodies. (See, e.g., id. at 4:1—23.)
FF 14. Reiter teaches generating a functional mammalian single­
chain MHC class I complex in eukaryotic or prokaryotic expression systems.
(Reiter Abstract.) Reiter teaches nucleic acid constructs encoding a
“functional human P-microglobulin, being translationally fused upstream of
a second polynucleotide encoding a functional human MHC class I heavy
chain” and constructs encoding “an antigenic peptide, the antigenic peptide
being capable of binding a human MHC class I complex.” (Id. at 4:20-28;
see also id. at 6:23—25 (“a recombinant polypeptide comprising an amino
acid sequence including a functional human P-2 microglobulin directly or
indirectly covalently linked to a functional human MHC class I heavy
chain.”)
FF 15. Reiter teaches “[expression of the scMHC gene in E. Coli
BL21 cells was very efficient and recombinant protein accumulated as
insoluble intracellular inclusion bodies.” (Id. at 32:18—20.) Reiter teaches
“[tjhree different peptides were used for the refolding of the scMHC
molecules. Two peptides (G9-209-2M and G9-280-2V) are tumor
associated antigens derived from the melanoma common antigen gplOO . . . .
These peptides were previously shown to be HLA-A2 restricted.” (Id. at
12
Appeal 2015-004942
Application 13/163,701
32:29—33:4.) Reiter further teaches “scMHC-peptide complexes can be
produced in vitro by refolding of E. Coli inclusion bodies in the presence of
antigenic peptides. The refolding process is efficient and a homogenous
population of complexes can be obtained with high yields and purity.” {Id.
at 34:3—7; see also id. at 41:17—30 (teaching co-expression of the scMHC
complex and peptide).)
FF 16. Reiter teaches “[rjecombinant scMHC-complexes can also be
used to generate specific antibodies by phage display technology” and that
“[s]uch antibodies with TCR-like specificity can be a valuable tool for
studying antigen presentation by tumor cells as well as to develop novel
targeting agents for immunotherapy.” {Id. at 40:27 41:2.)
Analysis
Claim 1
Appellants argue the patentability of claims 1, 3, 4, 6, 7, and 15—17 as
a group. We select claim 1 as representative. 37 C.F.R. § 41.37(c)(l)(iv).
The Examiner provides detailed findings and explanations in support
of the rejection of the claims for obviousness. {See Ans. 4—14; see also id. at
16—31 (Examiner’s Response to Appellants’ arguments).) We provide an
overview, rather than repeating all of the findings and explanations here.
The Examiner finds that Andersen teaches preparing antibodies to
MHC class I or II peptide complexes, and that the antibodies specifically
bind such complexes with an affinity in the range of 10'7 to 10'10 M. (Ans.
6.) The Examiner finds Andersen’s range (which spans 0.1 nM to 100 nM)
includes the range of “below 10 nanomolar” recited in claim 1. {Id. at 5—7.)
The Examiner acknowledges that in Andersen’s working examples the
13
Appeal 2015-004942
Application 13/163,701
affinity differed from what is claimed — 51 nM versus less than 10 nM. (Id.
at 6—7.) According to the Examiner, however, Andersen teaches the method
may be improved by known techniques (e.g., higher peptide loading of the
MHC, increasing somatic hypermutations, using immunized rather than
naive libraries, and using transgenic mice). (See, e.g., id. at 7 and 19.) The
Examiner thus reasons the skilled artisan would, by such techniques, have
been able to increase affinity to the claimed level.
The Examiner finds Andersen does not teach the HLA molecule is
HLA-A2 or the peptide of Appellants’ elected species (gplOO G9-209), and
that Andersen does not explicitly teach generating an antibody with the
claimed affinity (below 10 nM) and functionality (binding peptide unloaded
tumor cells naturally presenting the complex). (Ans. 7.) So the Examiner
turns to the other cited references. (Id. at 7.)
The Examiner cites Kawakami as teaching “the G9-209 peptide is a
modified immunogenic peptide from gplOO protein that binds to HLA-A2,
[and] that this peptide is used in medicaments for . . . treatment or prevention
of melanoma, and [generating] antibodies against the peptide.” (Id. at 8.)
The Examiner finds Chames teaches a high affinity TCR-like
antibody specific to HLA-A1/MAGE-A1 tumor peptide, and an affinity
matured version of the antibody. (Id.) According to the Examiner, Chames
“exemplifies] use of recombinantly produced HLA-Al/p2m molecules (as
separate chains) with peptide in the antigen binding groove, refolded from
inclusion bodies produced in E. coli in the presence of the MAGE-A1 tumor
antigen peptide.” (Id.) The Examiner finds Chames used a “non-immunized
phage display library” to select a Fab (G8) that possessed an affinity of 250
14
Appeal 2015-004942
Application 13/163,701
nM, and that Chames “engineered the Fab-G8 antibody to gain an 18-fold
increase to affinity to 14 nM” while retaining specificity for the MHC
complex. {Id. at 8—9.) The Examiner finds Chames teaches using transgenic
mice would help reduce pan-MHC-reactive antibodies. {Id. at 9.)
The Examiner finds Rhode and Rhode ’445 teach production of
single-chain MHC molecules loaded/complexed with an antigenic peptide
and advantages of doing so. {Id. at 9—10.) For example, the Examiner finds
Rhode teaches such single-chain constructs provide for balanced production
of the functional domains and may facilitate proper folding. {Id. at 9.) The
Examiner finds Rhode ’445 teaches producing single-chain MHC complexes
enhances yield and stability of the complexes. {Id. at 11.) According to the
Examiner, Rhode and Rhode ’445 teach that these techniques can be used
for MHC class I complexes. {Id. at 9—11.)
And the Examiner cites Reiter as teaching “making a single chain
class I MHC molecule comprising p2m fused to a functional MHC class I
heavy chain such as HLA-A2” and fusing a peptide/antigen of interest to the
construct. {Id. at 11.) The Examiner finds Reiter teaches such single-chain
peptide complexes “can be used to generate specific antibodies by phage
display technology,” and that such antibodies provide a tool to detect antigen
presentation on tumor cells and to develop immunotherapeutic targeting
agents. {Id. at 11—12.)
The Examiner concludes it would have been obvious to make the
antibodies of Andersen specific for human class I MHC molecule HLA-A2
in combination with the antigenic peptide from a tumor antigen, such as G9-
209 taught by Kawakami. (Id. at 12.) According to the Examiner, it would
15
Appeal 2015-004942
Application 13/163,701
have been obvious to immunize mice transgenic for the HLA-A2 with the
HLA-A2/G9-209 complex that is efficiently loaded with the peptide, stable,
and properly folded, such as with the single-chain complexes suggested in
Rhode, Rhode ’445, and Reiter, and to make a phage display library and
screen it to identify antibodies, such as taught in Andersen. (Id.) The
Examiner reasons the skilled person would have had a motivation to
combine the art in this way “to make a high affinity HLA/peptide-specific
antibody to monitor the expression of HLA-peptide complexes on tumor
cells or peptide loaded APCs, or to treat or monitor cancer.” (Id. at 13.) To
the extent affinity enhancement would have been required, the Examiner
reasons the skilled person would have predictably used techniques (e.g., an
immunization library, HLA transgenic mice, etc.) taught in Andersen and
Chames to do so and produce an antibody with the affinity of claim 1. (Id.)
As to the further functionality recited in claim 1 ’s wherein clause (i.e.,
related to the antibody binding tumor cells naturally presenting the MHC
complex), the Examiner finds the combined art teaches an antibody that is
substantially the same as (and produced by the same methodologies as) the
claimed antibody. (Id. at 14; see also id. at 21 (“the art references cited in
the instant rejection teach the same methodologies used by Applicant to
produce the G1 antibody disclosed in the specification on page 40”).)12 The
12 The Examiner specifically points to Reiter’s teachings for preparing the
single-chain MHC-peptide complex. (Ans. 21.) And Appellants’
Specification repeatedly cites to Reiter (see Spec. 26, 32, 35, 43) in
explaining the single-chain MHC-peptide complex that was used to generate
antibodies with the claimed affinity/functionality. (Final Act. 10
16
Appeal 2015-004942
Application 13/163,701
Examiner thus puts on Appellants the burden to present evidence of an
unobvious distinction over the prior art. (Id.) In other words, the Examiner
is requiring evidence from Appellants to show an antibody designed
according to the prior art’s teachings does not have the claimed
functionality. (Id. (citing In re Best, 562 F.2d 1252 (CCPA 1977).)
Unless otherwise noted, we agree with the Examiner’s findings,
reasoning, and conclusion of obviousness. On the present record, we are
unpersuaded of reversible error in the Examiner’s rejection of claim 1 as
obvious. We address below the Appellants’ arguments.
Appellants, in a section of their brief titled “Introduction to
Arguments,” describe the invention and some of the prior art (cited and
uncited), and assert that the invention was the first to provide a high-affinity
soluble antibody responsive to a long-felt need. (App. Br. 11—12.) It is
unclear if Appellants regard the “introduction” as a distinct argument, or
merely context for the arguments that begin in Section VII(B), at page 14, of
Appellants’ brief. Nevertheless, insofar as Appellants allude in the
“introduction” to a long-felt need that the invention purports to solve (a
recognized secondary consideration of nonobviousness), we address
Appellants’ assertions.
To establish a long-felt and unmet need, the following elements must
be proven: First, the need must have been persistent and recognized by
skilled artisans. In re Gershon, 372 F.2d 535, 538 (CCPA 1967). Second,
the long-felt need must not have been satisfied by another before Appellants’
(“Applicant has referenced [Reiter] on page 32 at lines 14-16 of the instant
specification, and this reference is also cited in the instant rejection”).)
17
Appeal 2015-004942
Application 13/163,701
invention. See Newell Companies, Inc. v. Kenney Mfg. Co., 864 F.2d 757,
768 (Fed. Cir. 1988) (“[OJnce another supplied the key element, there was
no long-felt need or, indeed, a problem to be solved”). And third,
Appellants’ invention must satisfy the need. In re Cavanagh, 436 F.2d 491,
496 (CCPA1971).
We are not persuaded on this record that Appellants have shown a
long-felt and unmet need that is satisfied by the claimed invention.
Appellants do provide some evidence with Tamminen13 that others were
working to generate MHC-restricted antibodies that recognize viral-infected
cells with CTL-like specificity at least as early as 1987. And Andersen and
Chames evidence that similar work took place in the late 1990s and early
2000s. None of these references, however, appears to set the threshold of
relevance (clinical, diagnostic, or otherwise) at an antibody with a binding
affinity below 10 nM as claimed.
We are also unpersuaded that Andersen and Chames failed to meet a
need for generating MHC-restricted antibodies with CTL-like specificity and
affinity. Chames, for example, discloses that “Fab-G8 already has an
affinity 5- to 500-fold higher than found for TCRs . . . [and] this work
demonstrates that very large . . . nonimmunized phage libraries can be used
to rapidly select antibodies of exquisite TCR-like specificity.” (Chames
7974.) Designing highly selective antibodies to an MHC-peptide complex
13 Tamminen et al., Searching for MHC-restricted anti-viral antibodies:
antibodies recognizing the nucleoprotein of influenza virus dominate the
serological response of C57BL/6 mice to syngeneic influenza-infected cells,
17 Eur. J. Immunol. 999-1006 (1987).
18
Appeal 2015-004942
Application 13/163,701
was an objective of Andersen too, and Appellants fail to persuade us
Andersen did not meet that objective in actually generating antibodies with
affinities of between roughly 50 to 60 nM. (FF 1—4.) Appellants critique
Andersen because it exemplified a mouse MHC-peptide complex. (App. Br.
11.) But Appellants fail to persuade us that it would have been nonobvious
to extend Andersen’s phage display method to the functional human MHC
class I complex described in Reiter, for example, which specifically teaches
its complexes can be used to generate antibodies through phage display
technology. (FF 14—16.)
And finally, even if the “need” was defined more narrowly as a
portion of Appellants’ “introduction” suggests, Appellants fail to provide
persuasive evidence the invention recited in claim 1 satisfies such a need.
Appellants assert “[tjhis antibody is the first clinically relevant T cell
receptor-like antibody which can be used in the treatment of cancer.” (App.
Br. 11 (emphasis omitted).) Appellants provide no adequate support for this
broad assertion. If Appellants were the first to design antibodies capable of
treating cancer, surely evidence of this success (e.g., publications showing
industry praise or recognition among the scientific community that a
persistent need had been met) would have been documented over the fifteen
years since Appellants’ patent application was filed, and thus available for
submission by Appellants now.
Turning to Section VII(B) of Appellants’ brief, Appellants argue the
prior art does not teach all the elements of claim 1. (App. Br. 14—15.)
Appellants assert “Andersen suggests antibodies to peptide human MHC
complexes of affinities 10'7 — 10'10 M but does not describe a single
19
Appeal 2015-004942
Application 13/163,701
embodiment for the generation of such antibodies.” (Id. at 15.) Instead,
according to Appellants, Andersen “achieve[d] an antibody having an
affinity of 56 nM . . . towards the murine complex as opposed to the human
complex claimed.” (Id.) Thus, Appellants contend Andersen is “a non­
enabling disclosure for an antibody which binds human pMHC complexes,
as claimed.” (Id.; see also Reply Br. 5—6.)
We are unpersuaded. “[A] prior art printed publication cited by an
examiner is presumptively enabling barring any showing to the contrary by a
patent applicant.” In re Antor Media Corp., 689 F.3d 1282, 1288 (Fed. Cir.
2012). Here, as Appellants acknowledge, Andersen suggests generating
antibodies to human MHC complexes of affinities between 0.1 —100 nM,
thus encompassing the claimed range of “below 10 nanomolar.” (FF 2.)
Although Andersen exemplifies mouse MHC complexes and generated
antibodies above the claimed range, prior art is not limited to its preferred
embodiments or working examples. Merck & Co., Inc. v. Biocraft Labs.,
Inc., 874 F.2d 804, 807 (Fed. Cir. 1989). Also, the other art (e.g., Reiter)
suggests generating antibodies to human MHC complexes. (FF 14—16; Ans.
17—19, 21.) And Andersen (and other cited art) teach techniques to enhance
the affinity of antibodies (e.g., somatic hypermutations) and these techniques
would have been reasonably expected to produce antibodies with the
affinities claimed absent persuasive evidence to the contrary from
Appellants. (Ans. 17—18; FF 5—6; 10.)
Appellants note that Chames, which did use human complexes, failed
to generate antibodies with the claimed affinity and functionality. But, as
the Examiner points out, Chames used a non-immunized phage display
20
Appeal 2015-004942
Application 13/163,701
library, whereas Andersen teaches that “the affinity of the antibody may be
improved in general a thousand fold by use of an immunized library.” (Ans.
30—31; FF 6, 8.) We are thus unpersuaded that Chames shows Andersen is
non-enabled, particularly where Andersen discloses a range that includes
affinities of, e.g., 10 nM, 1 nM, and 0.1 nM, suggesting preparation of
antibodies with these affinities was within the reach of skilled persons using
known techniques.
Appellants argue the Examiner erred in asserting that the combined
references inherently teach an antibody with the recited specificity and
affinity. (App. Br. 16.) According to Appellants, arguments for inherency
“while proper in anticipation rejections, are improper in obviousness
rejections.” {Id. (citing Ex parte Vigano et al., (Appeal No. 2010-007666)
(non-precedential).)
Appellants’ argument is unpersuasive and does not accurately reflect
the law. Rather, where the Examiner provides a reasoned basis to conclude
that the prior art would possess a functional characteristic (e.g., the claimed
affmity/specificity), the Examiner may appropriately require that an
applicant provide proof to the contrary. As In re Best confirms, that is true
whether the rejection is based on anticipation or obviousness:
Where ... the claimed and prior art products are identical or
substantially identical ... the PTO can require an applicant to
prove that the prior art products do not necessarily or inherently
possess the characteristics of his claimed product. . . . Whether
the rejection is based on “inherency” under 35 U.S.C. § 102, on
“prima facie obviousness” under 35 U.S.C. § 103, jointly or
alternatively,[] the burden of proof is the same, and its fairness is
21
Appeal 2015-004942
Application 13/163,701
evidenced by the PTO’s inability to manufacture products or to
obtain and compare prior art products.
In re Best, 562 F.2d 1252, 1255 (CCPA 1977) (internal citations and
footnote omitted).
It is, however, also true that caution should be observed in extending
an inherency rationale to the obviousness inquiry. See Par Pharma., Inc. v.
Twi Pharms., Inc., 773 F.3d 1186, 1195—96 (Fed. Cir. 2014) (holding that a
“high standard” must be met “to rely on inherency to establish the existence
of a claim limitation in the prior art in an obviousness analysis.”) As stated
in Par Pharma, “the limitation at issue necessarily must be present, or the
natural result of the combination of elements explicitly disclosed by the prior
art.” {Id. at 1196.) Here, we are persuaded the natural result of the prior art
combination — especially the phage display technology taught in Andersen
combined with the single-chain human MHC construct described in Reiter
— would produce an antibody with the claimed affinity and specificity.14
14 We note the Applicant-Initiated Interview Summary (“Summary”) mailed
on April 2, 2015. In the “Summary,” Appellants indicated “they are willing
to file a Katz declaration with respect to reference WO 01/72768 (Reiter)
that is cited in the 103(a) rejection of record among other things for a
particular configuration of single chain class I MHC that is used to make
antibodies with TCR-like specificity.” (Summary.) Appellants further
discussed potential claim amendments and “asserted that the said particular
configuration of MHC is critical for making the claimed antibody.” {Id.)
Because the case was at the appeals stage at the time of the interview, no
such declaration or amendment was submitted. We do not decide in this
appeal whether removing Reiter from the prior-art combination or amending
the claims to recite the allegedly “critical” MHC configuration would result
in patentable claims. That said, to guide prosecution, should it continue, we
found Reiter’s contribution to the combination of art highly persuasive on
the present record. If Appellants remove Reiter and amend the claims to add
22
Appeal 2015-004942
Application 13/163,701
Indeed, Appellants’ Specification lends support in repeatedly citing Reiter
for describing the particular complex used for generating antibodies, via a
phage display library, with an affinity of 5 nM. (See Spec. 32-40.)
Appellants point out that “‘[ojbviousness cannot be predicated on
what is unknown.’” (App. Br. 16 (quoting In re Rijckaert, 9 F.3d 1531,
1534 (Fed. Cir. 1993)).) But we are not persuaded this rule applies here. At
least Reiter discloses a single-chain human MHC complex with the elected
peptide (much like, if not the same as, that described in Appellants’
Specification) and teaches such scMHC-peptide complexes can be used to
generate specific antibodies by phage display technology. (FF 14—16.)
Andersen teaches phage display technology and suggests affinities between
0.1 nM to 100 nM are known and obtainable; and affinity of an antibody is a
results-effective variable. In re Boesch, 617 F.2d 272, 276 (CCPA 1980)
(“[Discovery of an optimum value of a result effective variable in a known
process is ordinarily within the skill of the art”); In re Peterson, 315 F.3d
1325, 1330 (Fed. Cir. 2003) (“The normal desire of scientists or artisans to
improve upon what is already generally known provides the motivation to
determine where in a disclosed set of percentage ranges is the optimum
combination of percentages.”). We thus disagree that the Examiner’s
rejection relies on matters wholly “unknown” to skilled persons.
specific limitations as suggested, we encourage the Examiner to consider all
the evidence anew. This includes, for example, evaluating the disclosures in
Chames expressing challenges and difficulties with designing MHC
complexes and specific antibodies and the extent to which these disclosures
undercut (or not) the ordinarily skilled person’s reasonable expectation of
success. (See, e.g., Chames 7969-70, 7973.)
23
Appeal 2015-004942
Application 13/163,701
Appellants next argue the prior art cannot be combined because
Chames teaches away from the use of single-chain MHC. (App. Br. 16—17.)
Appellants contend “generation of the claimed antibodies was achieved
through a novel method of antibody production which makes use of a
specific configuration of a single chain MHC (in which the P2m is
covalently linked to the N-terminus of the MHC heavy chain) in vitro folded
with a peptide.” {Id. at 17.)
We remain unpersuaded. Claim 1 recites an antibody specifically
bindable with an MHC complex, not a method. So too, claim 1 does not
specifically recite the structural elements of a “single chain MHC” with the
particular linkages that Appellants emphasize. Accordingly, Appellants
appear to be arguing that Chames — in exemplifying non-straight chain
MHC complexes — teaches away from unclaimed elements. This fails to
overcome the rejection.
Even assuming Chames did teach away from single-chain MHC, other
cited art teaches the opposite. Indeed, Rhode, Rhode ’445, and Reiter
describe the benefits of single-chain MHC complexes, including stability,
proper folding, and high yields and purity. (FF 11—16.) Medichem, S.A. v.
Rolabo, S.L., 437 F.3d 1157, 1165 (Fed. Cir. 2006) (“Where the prior art
contains apparently conflicting teachings (i.e., where some references teach
the combination and others teaching away from it) each reference must be
considered for its power to suggest solutions to an artisan of ordinary skill
. . . considering] the degree to which one reference might accurately
discredit another”) (internal quotation marks and citation omitted). When
we consider Chames (suggesting properly folded single-chain MHC
24
Appeal 2015-004942
Application 13/163,701
complexes were produced in low yield) against these other references, the
preponderance of the evidence favors the Examiner. That the references
may, to some degree, contradict each other does not show that the claims are
nonobvious, contrary to Appellants’ contentions. (Reply Br. 11—12.)
Appellants argue no single art on file teaches antibodies of affinities
below 10 nM. (App. Br. 18.) Appellants contend there is no evidence that
the antibody fragment of Chames could be affinity matured to less than 10
nM, and that Chames does not teach any protocol of how affinity maturation
obtained Fab-G8 with an affinity of 14 nM. (Id.) And, according to
Appellants, the maturation strategies described in a later Chames
publication15 were not straightforward or routine. (Id. at 18—19; see also
Reply Br. 7—9.)
Appellants’ argument is unpersuasive. As explained above, on this
record, the Examiner’s invocation of In re Best is reasonable. Appellants,
for example, provided no persuasive evidence that combining Reiter’s
single-chain complexes with the phage display technology described in
Andersen would result in antibodies with affinities outside the claimed
range. Neither did Appellants provide persuasive evidence that using other
known techniques to enhance antibody affinity (e.g., peptide loading,
somatic hypermutations) would not result in antibodies with affinities below
10 nM. (FF2, 5-6.)
15 Chames et al., TCR-Like Human Antibodies Expressed on Human CTLs
Mediate Antibody Affinity-Dependent Cytolytic Activity, 169 J. Immunol.
1110-18 (2002) (“Chames ’02”).
25
Appeal 2015-004942
Application 13/163,701
It is true that Chames affinity maturation did not produce an antibody
with an affinity below 10 nM. But Chames was able to achieve an 18x
affinity enhancement by directed randomization of the complementarity
determining region H3 and reselection. (FF 10.) And Chames used a less-
selective non-immunized phage library as discussed above. (FF 8.) As to
Appellants’ contentions that Chames’ affinity maturation was not known or
straightforward, the Examiner finds that Andersen teaches site-directed
mutagenesis in the CDR-region is a conventional technique. (Ans. 20, 27.)
Appellants provide argument but insufficient persuasive evidence to the
contrary. See In re Geisler, 116 F.3d 1465, 1470 (Fed. Cir. 1997)
(“[Attorney argument [is] not the kind of factual evidence that is required to
rebut a prima facie case of obviousness”).16
Appellants argue the claimed invention cannot be considered a result
of a finite number of identified, predictable solutions. (App. Br. 19—21.)
Appellants contend the claimed molecular assemblies are complex, include
multiple sub-components and expression systems, and yield hundreds of
options. (Id. at 20.) Without a specific pointer in the art, Appellants
contend the Examiner’s analysis is based on impermissible hindsight. (Id.)
We remain unpersuaded for reasons already explained. The
combination of Andersen with the other cited references (Reiter especially)
appears to teach and suggest antibodies with essentially the same structural
16 We note that Chames ’02 does state that the synergistic improvement
observed “is somewhat surprising because mutations selected in parallel are
usually difficult to combine in a single protein.” (Chames ’02 1115; see also
Reply Br. 8.) If prosecution of the claims continues (as noted supra n. 14),
the Examiner should further address these teachings in Chames ’02.
26
Appeal 2015-004942
Application 13/163,701
makeup prepared by essentially the same method. Reiter expressly teaches
that its MHC complexes can be used to generate antibodies and Andersen is
not limited to the particular MHC complexes in its working examples. (FF
1—6, 14—16.) So, hindsight gleaned only from Appellants’ disclosure is not
required to combine the art and arrive at Appellants’ invention.
Claim 5
Claim 5 depends from claim 1 and adds “wherein the isolated
antibody is detectable on cells displaying said complex by FACS analysis.”
(App. Br. 24.) The Examiner finds claim 5 recites language that does not
carry patentable weight per se, and that the claims read on the active or
essential ingredients of the composition. (Ans. 13.) The Examiner finds
“the art antibody appears to be the same antibody with regard to this said
limitation since it possesses the recited affinity.” (Id. at 29.) In other words,
the Examiner finds the antibody according to the prior-art combination
would reasonably be expected to have the functionality recited in claim 5.
Appellants argue the Examiner erred because “reciting the assay . . .
inherently defines the threshold of complex detection” and thus “the claim
provides a meaningful limitation.” (App. Br. 21.)
On this record, the Examiner has the better position. An antibody
composition prepared based on the prior-art combination discussed above
with respect to claim 1 would, absent persuasive evidence to the contrary,
possess the functionality recited in claim 5.
We address Appellants’ Reply Brief further. Appellants devote a
substantial portion of the Reply Brief to arguments and evidence that could
and should (to be presented for consideration) have been raised in the
27
Appeal 2015-004942
Application 13/163,701
principal brief. (Reply Br. 3—13.) For example, Appellants repeatedly refer
to declarations {id. at 3—8, 12)17 and assert “the Examiner has not addressed
these [declarations] in her Answer” {id. at 3). Because Appellants neither
mentioned nor cited any declarations in the Appeal Brief, it is not surprising
the Examiner did not address them. We decline to consider the declarations
as part of our analysis on appeal.18 37 C.F.R. § 41.41(b)(1) & (2); see Ex
parte Nakashima, 93 USPQ2d 1834, 1837 (BPAI 2010) (informative).
Appellants also newly argue that Rhode and Rhode ’445 focus on
MHC class II while the claims in this case are limited to antibodies to MHC
class I. (Reply Br. 9—11.) And Appellants contend the Examiner’s assertion
of “five additional references cited in Rhode” to show class I MHC
constructs “is a back-door way of introducing new grounds of rejection.”
{Id. at 10.) Appellants did not try to distinguish Rhode or Rhode ’445 on
this basis in their opening brief and, if Appellants thought the Examiner had
introduced new grounds, Appellants could have requested prosecution be
17 Declaration of Charles DeLisi dated July 17, 2007; Declaration of
Vincenzo Cerundolo dated July 16, 2007.
18 We do, however, note the declarations were signed in 2007, years before
the present application was filed. Both declarations state that the declarants
“have read the Office Action with respect to the above-identified patent
application [the header identifies the present application (No. 13/163,701)]”
and then point to various claims, e.g., “claims 141-149, 151-155 . . . .” {See
DeLisi Deck 2; Cerundolo Deck 2.) We assume the declarations were
originally prepared for submission in a related patent application because
those are not the claims on appeal, and given the date the declarations were
signed, it is doubtful the declarants read the Office Action and rejection on
appeal. So the relevance of the declarations is questionable in any event.
28
Appeal 2015-004942
Application 13/163,701
reopened. 37 C.F.R. § 41.39(b)(1).19 We are unpersuaded in any event
because Rhode and Rhode ’445 suggest the attributes of single-chain
constructs extend to class I and II MHCs. (FF 11—13; see also Andersen
11:26—27 (“HLA class II molecules are structurally highly related to class I
molecules”).)
Appellants’ argument in the Reply Brief that Chames teaches away
from immunization and the use of mice is more new argument. (Reply Br.
11—12.) Appellants have not shown good cause for why this could not have
been argued in the opening brief. We thus decline to consider it.
Finally, Appellants contend Reiter “is directed to single chain MHC-
antigen complexes, but there is no disclosure of the use of these complexes
to make antibodies as claimed or motivation to combine with the other
references.” (Reply Br. 13.) Appellants did not earlier make this argument
but, in any case, Appellants are mistaken. Reiter teaches that
“[rjecombinant scMHC-complexes can also be used to generate specific
antibodies by phage display technology” and that such antibodies can be
used for targeted immunotherapies. (FF 16.) Reiter thus provides express
motivation to combine with other art of record, such as Andersen.
19 While we agree with Appellants that the Examiner’s citation could have
been more explicit, the Examiner did identify “four Kourilsky group articles
that also use P2m fused to heavy chain of MHC class I” (Final Act. 10) and
“reference 22 [of Rhode]” (Ans. 22). Thus, the Examiner was identifying
references 21 and 42-44 (Kourilsky articles), and reference 22 in Rhode,
which is discernible in light of the portion of Rhode cited by the Examiner.
29
Appeal 2015-004942
Application 13/163,701
Conclusion of Law
The preponderance of the evidence supports the Examiner’s
conclusion that claims 1 and 5 would have been obvious over Andersen,
Kawakami, Chames, Rhode, Rhode ’445, and Reiter.
Claims 3, 4, 6, 7, and 15—17 have not been argued separately and fall
with claim 1. 37 C.F.R. § 41.37(c)(l)(iv).
SUMMARY
We affirm the rejection of claim 17 for lack of written description and
for anticipation. We affirm the rejection of claims 1, 3—7, and 15—17 for
obviousness.
TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a).
AFFIRMED
30