Ex Parte Howley et al

Patent Trial and Appeal Board

Mar 10, 2015

11985510 (P.T.A.B. Mar. 10, 2015)

UNITED STATES PATENT AND TRADEMARK OFFICE
__________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
__________

Ex parte PAUL HOWLEY and SONJA LEYRER
_________

Appeal 2012-012206
Application 11/985,510
Technology Center 1600
__________

Before ERIC B. GRIMES, JEFFREY N. FREDMAN, and
CHRISTOPHER G. PAULRAJ, Administrative Patent Judges.

PAULRAJ, Administrative Patent Judge.

DECISION ON APPEAL
This is an appeal
1
under 35 U.S.C. § 134 involving claims to a
recombinant Modified Vaccinia Ankara (MVA) virus. The Examiner
rejected the claims for obviousness. We have jurisdiction under 35 U.S.C.
§ 6(b). We reverse.

STATEMENT OF THE CASE
Background
The invention at issue “relates to a recombinant poxvirus capable of
expressing two or more homologous foreign genes,” wherein “[s]aid genes

1
Appellants identify the Real Party in Interest as Bavarian Nordic A/S (see
App. Br. 2).
Appeal 2012-012206
Application 11/985,510

2
are heterologous to the viral genome” (Spec. 1 ll. 4–6). It is “an object of
the present invention to provide a stable, effective and reliable vaccine
against infectious diseases, which can be caused by more than one strain,
clade, variant, subtype or serotype of said infectious disease causing
microorganism.” (id. at 8 ll. 11–15). For example, the Specification
describes the desirability of “a multivalent vaccine that contains antigens
from all four Dengue virus serotypes” using such a recombinant virus (id. at
6 ll. 29–30). It was “expected that the insertion of homologous genes in a
single viral genome would result in homologous recombination and, thus, in
deletions of the inserted homologous genes” (id. at 9, ll. 20–23). The
Specification notes, however, that “when generating a recombinant poxvirus
comprising in its genome at least two foreign genes with a homology of at
least 60%, it was unexpectedly found that said homologous genes remain
stably inserted into the viral genome” (id. at 9 ll. 24–28).
The Claims
Claims 23–44 are under appeal. Independent claim 23 is
representative, and reads as follows:
23. A recombinant MVA comprising at least two foreign genes of
greater than 300 nucleotides which are 50-75% homologous in comparison
to each other, wherein each of the foreign genes is stably inserted into a
different site in the MVA genome.

The Rejection

The Examiner has rejected the claims under 35 U.S.C. § 103(a) as
being unpatentable over the combination of Cardosa,
2
Ball,
3
Pompon,
4

2
Cardosa et al., US 2004/0265324 A1, published Dec. 30, 2004 (hereinafter
“Cardosa”)
Appeal 2012-012206
Application 11/985,510

3
Huse,
5
Antoine,
6
and Genbank Accessions: M19197,
7
U88536,
8
and
M14931
9
(Ans. 4–6). The issue presented on appeal is whether there was a
reasonable expectation of success, at the time of Appellants’ claimed
invention, of being able to generate a recombinant MVA comprising two
foreign genes, which are 50-75% homologous to each other and inserted into
different sites in the MVA genome.
FINDINGS OF FACT
FF1. Cardosa teaches “a recombinant MVA . . . containing and
capable of expressing DNA sequences encoding antigens from all four
dengue virus serotypes (type 1, 2, 3, and 4)” (Cardosa, ¶ 17). Cardosa
further teaches:
When the DNA sequence encoding the dengue antigen is
integrated at a site in the viral DNA which is non-
essential for the life cycle of the virus, e.g. one of the
above mentioned deletions, the newly produced
recombinant MVA will be infectious, that is to say able
to infect foreign cells and it will express the integrated
DNA sequence. The recombinant MVA according to the
invention will be useful as extremely safe live vaccines

3
Ball, High-Frequency Homologous Recombination in Vaccinia Virus DNA,
61 J. VIROL. 1788–1795 (1987).
4
Pompon et al., US 5,635,369, issued June 3, 1997 (hereinafter “Pompon”).
5
Huse, US 6,893,845 B1, issued May 17, 2005.
6
Antoine et al., The Complete Genomic Sequence of the Modified Vaccinia
Ankara Strain: Comparison with Other Orthopoxviruses, 244 VIROL. 365–
396 (1998) (hereinafter “Antoine”).
7
GenBank Accession M19197, Dengue virus type 2 (S1 vaccine strain),
complete genome, 8/2/1993.
8
GenBank Accession U88536, Dengue virus type 1 clone 45AZ5, complete
genome, 9/19/1997.
9
GenBank Accession M14931, Dengue virus type 4 polyprotein precursor,
gene, complete cds, 2/15/2000.
Appeal 2012-012206
Application 11/985,510

4
for the treatment or prophylactics [sic, prophylaxis] of
infectious diseases.

(Id., ¶ 12). “Insertion of foreign genes into the MVA genome was
targeted precisely to the site of the naturally occurring deletion II in
the MVA genome” (id., ¶ 53).
FF2. Ball teaches:
A recombinant vaccinia virus genome was
constructed in which the viral thymidine kinase (tk) gene
was placed between direct repeats of a 1.5-kilobase-pair
DNA sequence of heterologous origin. . . . Under
nonselective conditions, however, the tk gene was
frequently excised by both inter- and intramolecular
recombination events because the repeated sequences
provided substantial targets for homologous DNA
recombination.

(Ball, Abstract). Ball further teaches that “[w]ithout stabilization,
such [repeat] sequence arrangements were so readily eliminated from
the viral DNA by homologous recombination that isolation of the
corresponding virus was impossible” (id., p. 1793).
FF3. Pompon teaches “a yeast strain in the chromosomes of which
several heterologous genes are stably integrated” (Pompon, col. 1 ll.
6–8). Pompon teaches that “[t]he main foreseeable genetic
instabilities which lead to the disappearance or modification of
heterologous activities introduced into yeast strains carrying one or
more foreign genes are . . . (a) inactivation by mutation of one of the
heterologous genes . . . (b) deletion by recombination of the
heterologous gene . . . (c) elimination of one of the heterologous
genes, or evolution of important characteristics of the strain . . . .” (id.,
Appeal 2012-012206
Application 11/985,510

5
col. 3 ll. 39–64). Pompon further teaches using “different 3’-flanking
sequences in order to avoid the risks of instability which might result
from homologous recombinations between plasmid and genome” (id.,
col. 18 ll. 8–12).
FF4. Huse teaches “prokatyotic cells [that] contain diverse
combinations of first and second DNA sequences encoding first and
second polypeptides which form a heteromeric receptor exhibiting
binding activity toward a preselected molecule” (Huse, Abstract).
Huse further teaches the construction of a vector wherein:
The pseudo-wild type gene encodes the identical amino
acid sequence as that of the wild type gene; however, the
nucleotide sequence has been altered so that only 63%
identity exists between this gene and the encoded wild
type gene VIII. Modification of the gene VIII nucleotide
sequence used for surface expression reduces the
possibility of homologous recombination with the wild
type gene VIII contained on the same vector.

(Id., col. 11 ll. 1–9).
FF5. “GenBank Accessions# M19197, U88536 and M14931 provide
the PrM genes from three different Dengue virus types, which share at
least a 71 % nucleic acid sequence identity” (Ans. 10). Appellants’
Specification indicates that “sequence comparison . . . between the
PrM genes of Dengue virus serotype 1, 2, 3 and 4 (PrM1-4) revealed a
sequence identity of 66.5-72.9%, i.e., a homology of appr. 65-75%”
(Spec. 9 ll. 4–7).
FF6. Antoine teaches the complete genomic sequence of the MVA
strain, and identifies “potential stable insertion sites for foreign genes”
(Antoine, p. 385).
Appeal 2012-012206
Application 11/985,510

6
FF7. Gritz10 teaches:
Recombinant vaccinia viruses that contain two divergent
env genes in tandem array have been constructed. In the
absence of selective pressure to maintain both genes,
recombination between conserved homologous regions in
these genes generated a wide range of progeny, each of
which expressed a novel variant polypeptide encoded by
the newly created hybrid env gene.

(Gritz, Abstract). Gritz teaches that the inserted “env genes differ by
14% in the nucleotide sequence” (86% homology), and that “under
nonselective conditions . . . nearly all of the viral genomes had
condensed to contain a single env gene” (id., pp. 5950, 5954).
FF8. The Specification states that “US Patent No. 5,338,683
discribes [sic] insertion of gp 13 and 14 of herpesvirus glycoprotein
genes into two different insertion sites of a single recombinant
poxvirus; however, both genes have a homology of 25.2% only”
(Spec. 11 ll. 1–4). The Specification also states that “US Patent No.
5,891,442 discloses a recombinant poxvirus [wherein genes] for the
polyprotein VP2, VP3 and VP4 of infectious bursal disease [were]
inserted into a single insertion site and have a homology of 41.9%-
50.3%” (id. at 11 ll. 10–14). The Specification further states that “US
Patent No. 6,217,882 describes a recombinant swinepox virus vector
containing pseudorabies antigens gp50 and gp63 with a homology of
52.7% inserted into the same insertion site” (id. at 11 ll. 15–18). “In
summary, it can be stated that according to the prior art homologous

10
Gritz et al., Generation of Hybrid Genes and Proteins by Vaccinia Virus-
Mediated Recombination: Application to Human Immunodeficiency Virus
Type 1 env, 64 J. VIROL. 5948–5957 (1990) (hereinafter “Gritz”).
Appeal 2012-012206
Application 11/985,510

7
genes or sequences having a homology of at least 50% are all inserted
into the same or a single insertion site within the viral genome” (id. at
11 ll. 19–22).

ANALYSIS
The Examiner finds that “Cardosa et al. teach the generation of a
recombinant MVA that encodes at least one Dengue virus gene from at least
one serotype in naturally occurring deletion sites within its genome,” and
“[e]xamples of these genes are PreM and NS-1 genes of the 4 serotypes of
Dengue” (Ans. 4). The Examiner acknowledges that “Cardosa et al. do not
teach the specific homology or nucleic acid length; or the specific intergenic
regions” (id. at 5). The Examiner also finds “Ball teaches that during
vaccinia virus replication, homologous recombination can result in the
deletion of inserted foreign nucleic acid sequences” (id.). The Examiner
relies upon the teachings in Pompon and Huse that, in order to avoid
deletion of the foreign gene through homologous recombination, one can
either modify “the 3'-flanking sequence on the plasmid thereby avoiding the
risks of instability associated with homologous recombination” or change
the sequence (i.e., codons) of one of the inserted nucleic acid sequences
(id.). The Examiner further relies upon Antoine’s teaching of the complete
genome of MVA, including information pertaining to open reading frames
and intergenic regions, and the complete genomes of Dengue virus types 1,
2, and 4 sharing at least 71% identity as set forth in the GenBank Accessions
U88536, M19197, and M14931, respectively (id. at 5–6).
Based on the forgoing teachings, the Examiner asserts that “[i]t would
have been obvious to one of ordinary skill in the art to modify the
Appeal 2012-012206
Application 11/985,510

8
composition taught by Cardosa et al. in order to encode at least two, or only
two or only three Dengue PrM antigens/genes stably inserted into different
sites of the MVA genome,” and that
There would have been a reasonable expectation of success,
given the knowledge that homologous recombination can affect
the stability of inserted foreign genes into vaccinia virus
genomes due to deletion of such genes, as taught by Ball, also
given the knowledge that the effects (e.g., deletion) of
homologous recombination can be overcome by modifying the
nucleic acid sequence of the inserted gene or changing the 3'
flanking region of the gene, as taught by Huse and Pompon et
al., also given the knowledge that the open reading frames and
therefore the intergenic sites of MVA are known, as taught by
Antoine et al., and also given the knowledge that the Dengue
serotypes 1[,] 2, and 4 with respect to their PrM genes share at
least 71% percentage of identify, as taught by GenBank
Accessions M19197, U88536 and M14931 in view of
Appendices A-C.

(Id. at 6).
Appellants argue that the Examiner and the cited references have
failed to establish that there would have been any expectation of success in
achieving the claimed invention in view of the known “issue of homologous
recombination between two homologous foreign sequences inserted into
different sites in an MVA genome, which was thought to be a problem prior
to Appellant’s invention” (App. Br. 8). For example, Appellants rely upon
Gritz’s teaching that even “reducing the homology between two repeated
sequences to 86% did not generate a stable [recombinant vaccinia] virus”
that contained both env gene sequences (id. at 13).
We determine that Appellants have the better position. Although
Cardosa generally teaches a recombinant MVA “containing and capable of
Appeal 2012-012206
Application 11/985,510

9
expressing DNA sequences encoding antigens from all four dengue virus
types” (FF1), there is no suggestion that the inserted foreign genes can have
a homology of 50–75% or that they can be stably inserted into different sites
within the MVA genome. The examples provided in Cardosa only
demonstrate insertion of a cDNA fragment from Dengue virus type 2 (see
Examples 5–8). Furthermore, “[i]nsertion of foreign genes into the MVA
genome was targeted precisely to the site of the naturally occurring deletion
II in the MVA genome” (FF1).
Moreover, in view of the known issue of homologous recombination
(FF2), the Examiner has not established that a skilled artisan would have had
a reasonable expectation of success in stably inserting two foreign genes
have 50–75% homology into different sites of the MVA genome. “A
showing of obviousness requires a motivation or suggestion to combine or
modify prior art references, coupled with a reasonable expectation of success
. . . .” Boehringer Ingelheim Vetmedica, Inc. v. Schering-Plough Corp., 320
F.3d 1339, 1354 (Fed. Cir. 2003). “[T]he reasonable expectation of success
must be founded in the prior art, not in the applicant’s disclosure.” In re
Vaeck, 947 F.2d 488, 493 (Fed. Cir. 1991). We recognize, but are not
persuaded by, the Examiner’s contention that “one of ordinary skill in the art
would expect as identity between two sequences decreases, the odds of
homologous recombination decline” (Ans. 11). Gritz demonstrates that
homologous recombination between genes inserted at different sites can
even occur with sequences having less than perfect homology (86%
homology) (FF7). Although Appellants’ Specification acknowledges that
the prior art taught the insertion of genes having a homology of only 25.2%
into different insertion sites of a single recombinant poxvirus (FF8), the
Appeal 2012-012206
Application 11/985,510

10
Examiner has not identified any basis on this record to conclude that the
skilled artisan would also have reasonably expected to achieve stable
insertion with two foreign genes having more than twice the level of
homology (50–75%) than previously recognized in the art.
We are also unpersuaded by the Examiner’s reliance on the teachings
of Huse and Pompon. Although these references indicate that modification
of the inserted gene sequences can reduce the possibility of homologous
recombination in prokaryotic cells and in yeast (FF3–4), the Examiner has
not identified any basis on this record to extrapolate those teachings to a
poxvirus (MVA) in eukaryotic cells. See Vaeck, 947 F.2d at 493 (finding
no prima facie obviousness where there was no equivalence established
between different host cell types (cyanobacteria vs. bacteria) for expression
of chimeric gene); cf. In re Droge, 695 F.3d 1334, 1338 (Fed. Cir. 2012)
(finding obviousness where prior art expressly taught that “‘[t]he method
according to the invention may be carried out in any type of cell host. Such
hosts can be, in particular, bacteria or eukaryotic cells (yeasts, animal cells,
plant cells), and the like.’”).
We conclude that the obviousness rejection is not supported by a
preponderance of the evidence of record. We accordingly reverse the
obviousness rejection.

REVERSED
cdc