Ex Parte Bruhn et al

Board of Patent Appeals and Interferences

Mar 23, 2012

10744595 (B.P.A.I. Mar. 23, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
__________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
__________

Ex parte LAURAKAY BRUHN, ALICIA F. SCHEFFER,
MICHAEL T. BARRETT, DOUGLAS A. AMORESE, and
STEPHEN S. LADERMAN
__________

Appeal 2011-012798
Application 10/744,595
Technology Center 1600
__________

Before ERIC GRIMES, LORA M. GREEN, and
JACQUELINE WRIGHT BONILLA, Administrative Patent Judges.

GRIMES, Administrative Patent Judge.

DECISION ON APPEAL
This is an appeal under 35 U.S.C. § 134 involving claims to a method
for determining a change in gene copy number, which the Examiner has
rejected as obvious. We have jurisdiction under 35 U.S.C. § 6(b). We
reverse.
Appeal 2011-012798
Application 10/744,595

2
STATEMENT OF THE CASE
The Specification discloses that “many malignancies involve the gain
or loss of DNA sequences. . . . Comparative genome hybridization (CGH) is
one approach that has been employed to detect the presence and identify the
location of amplified or deleted sequences.” (Spec. 1:24-35.) The
Specification also states that “[i]n a recent variation of the above traditional
CGH approach, the chromosomes to which the labeled nucleic acids are
hybridized have been replaced with a collection of solid support bound
nucleic acids, e.g., an array of BAC (bacterial artificial chromosome) clones
or cDNAs” (id. at 2:9-12).
Claims 1, 3-11, 14-24, 26-28, 30-32, 50, 52, 53, 58-66, 70, 71, and 73
are on appeal. Claim 1, the only independent claim, is representative and
reads as follows:
1. A method for comparing the copy number of at least one nucleic
acid sequence in at least two genomic sources, said method comprising:
(a) preparing at least a first collection of nucleic acid molecules from
a first genomic source and a second collection of nucleic acid molecules
from a second genomic source, wherein said first and second genomic
sources have a complexity of 1 x 108 base pairs or more and said first and
second collections are of non-reduced complexity;
(b) contacting said first and second collections of nucleic acid
molecules with one or more pluralities of distinct oligonucleotide feature
elements bound to a surface of a solid support, wherein said one or more
pluralities of distinct oligonucleotide feature elements each consists
essentially of oligonucleotide feature nucleic acids that range in size from 20
nt to 200 nt in length and said contacting occurs under stringent assay
conditions;
(c) measuring the binding of the first and second collections of nucleic
acid molecules to said feature elements to produce data; and
Appeal 2011-012798
Application 10/744,595

3
(d) identifying a quantitative difference in the copy number of at least
one nucleic acid sequence in said at least two genomic sources using said
data.

The claims stand rejected under 35 U.S.C. § 103(a) as follows:
• Claims 1, 3-5, 7, 9-11, 14-16, 18-24, 26, 50, 58-60, 65, 66, 70, 71,
and 73 based on Winzeler1 and Pollack2 (Answer 6);
• Claims 6, 30, 61, 62, and 64 based on Winzeler, Pollack, and
Pinkel3 (Answer 15);
• Claims 8, 17, and 52 based on Winzeler, Pollack, and Garner4
(Answer 17);
• Claims 27 and 28 based on Winzeler, Pollack, and Wells5 (Answer
18);
• Claims 31 and 32 based on Winzeler, Pollack, Pinkel, and Li6
(Answer 19);
• Claim 53 based on Winzeler, Pollack, and Heyneker7 (Answer 21);
and
• Claim 63 based on Winzeler, Pollack, and Zoller8 (Answer 22).

1 Winzeler et al., Whole genome genetic-typing in yeast using high-density
oligonucleotide arrays, 118 PARASITOLOGY S73-S80 (1999).
2 Pollack et al., Genome-wide analysis of DNA copy-number changes using
cDNA microarrays, 23 NATURE GENETICS 41-46 (1999).
3 Pinkel et al., US 5,690,894, Nov. 25, 1997.
4 Garner et al., Patent Application Publication US 2003/0054388 A1, Mar.
20, 2003.
5 Wells et al., Telepathology: a diagnostic tool for the millennium?, 191 J.
PATHOL 1-7 (2000).
6 Li et al., Patent Application Publication US 2003/0082618 A1, May 1,
2003.
7 Heyneker, US 6,057,100, May 2, 2000.
Appeal 2011-012798
Application 10/744,595

4
All of the Examiner’s rejections rely on the combination of Winzeler
and Pollack. The Examiner finds that Winzeler teaches a method similar to
that of claim 1 (Answer 6-7) but using yeast genomic DNA (id. at 6), which
does not have the complexity required by claim 1 (id. at 13). The Examiner
finds that “Pollack teaches a sample with complexity of 1 x 10^8 base pairs
or more (p. 41, where the first and second collections were obtained from
normal and tumor DNA from human tumors and where the human genome
has the complexity required by the claim)” (id.).
The Examiner concludes that it would have been obvious “to have
applied the teachings of comparative genomic hybridization (CGH) using
reduced [sic, non-reduced] complexity samples and short oligonucleotide
feature elements to the analysis of human samples as taught by Pollack” (id.
at 14). The Examiner also concludes that a skilled worker would have had a
reasonable expectation of successfully using immobilized probes (“feature
elements” in the language of the claims) of 20-200 bp because Pollack uses
“feature elements that are as short as 0.5 kb or 500 bp in length” (id.).
Appellants argue that “the accepted wisdom at the time of filing was
that in order to use an oligonucleotide array to investigate a complex
genome such as the human genome, one is required to reduce the
complexity of the genome. . . . Reducing the complexity of the genome was
thought to increase the signal to noise ratio to a point where meaningful data
can be obtained.” (Appeal Br. 6.) Appellants cite several references as

8 Zoller et al., Comparative genomic in situ hybridization (cGISH) analysis
on plant chromosomes revealed by labeled Arabidopsis DNA, 9
CHROMOSOME RES. 357-375 (2001).
Appeal 2011-012798
Application 10/744,595

5
evidence that those skilled in the art recognized the need to reduce the
complexity of human genomic DNA when carrying out CGH procedures (id.
at 6-15). Appellants urge that “the use of non-reduced complexity human
genomic DNA in Winzeler’s oligonucleotide array method would be
contrary to the accepted wisdom as discussed above” (id. at 16) and that
“there is more than ample factual evidence supporting the Appellants[’]
contention that one of skill in the art would not modify Winzeler’s method in
the way proposed by the Examiner” (id. at 17).
We agree with Appellants that a person of ordinary skill in the art
would not have considered it obvious to apply Winzeler’s method using
human genomic DNA with non-reduced complexity instead of DNA having
a lower complexity, such as yeast disclosed in Winzeler. Appellants have
cited sixteen references as evidence that those skilled in the art expected that
reducing the complexity of human genomic DNA (i.e., so that complexity is
less than 108 base pairs) would be necessary to effectively carry out array-
based CGH with probes of the size recited in claim 1. Of these sixteen
references, however, nine are addressed to analysis of single-nucleotide
polymorphisms (SNPs).9 Although Appellants argue that SNP analysis and
CGH analysis are similar (Appeal Br. 15), they have not persuasively shown
that a skilled worker would have the same expectations for genomic DNA of

9 The SNP-related references are referred to as: Dong (2001), Kennedy
(2003), Syvanen (2001), US Patent Application publication no.
20040081996, Tsuchihashi and Drocopoli (2002), Jordan (2002), Matsuzaki
(2004), The ‘448 application, and Gunderson (2005). Full citations to these
references can be found in the Evidence Appendix to the Appeal Brief
(pages 30-31).
Appeal 2011-012798
Application 10/744,595

6
non-reduced complexity in a CGH analysis and in a SNP analysis. The
SNP-related references are therefore entitled to little weight on the issue
central to this appeal.
The remaining references cited by Appellants, however, address CGH
analysis, and we conclude that these references are sufficient to cast doubt
on a skilled worker’s expectation of successfully applying Winzeler’s
method using Pollack’s human genomic DNA of non-reduced complexity.
For example, Mir10 notes Pollack’s array-based method (Mir at 351, citing
Pollack as reference 81), but states that the “process of analyzing sequence
variation goes through a number of stages,” including “amplification of loci”
(id.). Mir states that “PCR is used to generate sufficient amounts of a target
for analysis. It also serves to reduce the complexity of the sample.” (Id. at
352.) In other words, Mir teaches reducing the complexity of a human
genomic DNA sample via PCR application.
Mir notes that amplification “is now the rate-limiting step in genome-
wide analyses of sequence variation” (id.) and suggests several possible
solutions (id. at 352-353), which do not include using non-amplified
genomic DNA. Finally, Mir states that “the need to amplify loci for analysis
of variants is a major bottleneck for large-scale analysis” (id. at 353) and
that “it would be useful to add measurement of repeated-gene copy numbers
to the other measures of genetic variation . . . but, at present, no efficient
high-throughput method is available for this” (id. at 354). Mir thus provides

10 Mir et al., Sequence Variation in Genes and Genomic DNA: Methods for
Large-Scale Analysis, 1 ANNU. REV. GENOMICS HUM. GENET. 329-360
(2000)
Appeal 2011-012798
Application 10/744,595

7
support for Appellants’ position that those skilled in the art expected that
reducing complexity of human genomic DNA, such as by PCR
amplification, was necessary to carry out array-based CGH.
Lucito (2000)11 states that, by using Pollack’s method, “[h]ighly
amplified regions of the genome, and perhaps some deletions, can be
detected” (Lucito (2000) at 1726). Lucito (2000) states that the
“disadvantages of this method are that the signal-to-noise ratio is poor
because of the complexity of the total human genome and the short length of
the arrayed probes” (id. at 1726-27) and that “the practical resolving power
of this method is likely to be poor” (id. at 1727). Lucito (2000) presents an
alternative method involving “hybridiz[ing] genomic representations of
tumor and normal DNA. Representations are reproducible samplings of
DNA populations in which the resulting DNA typically has a new format or
reduced complexity or both” (id.).
Lucito (2000) thus provides evidence that those skilled in the art
recognized that Pollack’s method suffered drawbacks because of the
combination of high-complexity genomic DNA and the short length of the
probes in its array. As a result, Lucito (2000) supports Appellants’ position
that those skilled in the art would not have considered it obvious to modify
Pollack’s method – using human genomic DNA of non-reduced complexity
and probes with a size in the range of 500-2000 nucleotides (Pollack 45,

11 Lucito et al., Detecting Gene Copy Number Fluctuations in Tumor Cells
by Microarray Analysis of Genomic Representations, 10 GENOME RES.
1726-1736 (2000)
Appeal 2011-012798
Application 10/744,595

8
last ¶) – by using an array of oligonucleotides having a size in the range of
20-200 nucleotides, as required by claim 1.
Appellants also cite several CGH-related references that that were
published after the effective filing date of the instant application, but we find
Mir and Lucito (2000) to be sufficient evidence to support Appellants’
position.
The Examiner argues that the “conventional wisdom in the art as
argued by Applicant is not sufficient to lead one of skill away from the
combination of references . . . because the cited references do not address
the role of oligonucleotide length on the use of non-reduced complexity
samples” (Answer 25). We disagree because, as discussed above, Lucito
(2000) attributes the poor signal-to-noise ratio of Pollack’s method to “the
complexity of the total human genome and the short length of the arrayed
probes” (Lucito (2000), at 1726-27 (emphasis added)). Lucito (2000) thus
provides evidence that those skilled in the art considered both sample
complexity and oligonucleotide length to be important factors in CGH
assays.
The Examiner also argues that the lower end of Pollack’s size range
(500 bases) is relatively close to the upper end of the claimed size range
(200 bp), and Appellants have not provided evidence of what size of
oligonucleotide would and would not be expected to function in a CGH
assay with genomic DNA of nonreduced complexity (Answer 25-26).
Again, we disagree with the Examiner’s reasoning. Pollack’s assay
used oligonucleotide probes with sizes ranging from 500 bp (0.5 kb) to 2 kb,
and Lucito (2000) provides evidence that the results obtained by Pollack
Appeal 2011-012798
Application 10/744,595

9
were poor. Thus, Pollack’s experiment, as viewed by those skilled in the art
at the time the present invention was made, would have led to an expectation
that carrying out the same process using even smaller oligonucleotide
probes, would produce results with an even worse signal-to-noise ratio. The
Examiner has not provided contrary evidence showing that Winzeler’s
experiment, using yeast DNA, would have been expected to provide similar
results if repeated using human genomic DNA with nonreduced complexity.
In summary, after considering the evidence cited by the Examiner and
the rebuttal evidence provided by Appellants, we conclude that the rejections
on appeal are not supported by a preponderance of evidence of record. See
In re Oetiker, 977 F.2d 1443, 1445 (Fed. Cir. 1992) (“After evidence or
argument is submitted by the applicant in response [to the Examiner’s
rejection], patentability is determined on the totality of the record, by a
preponderance of evidence with due consideration to persuasiveness of
argument.”).
SUMMARY
We reverse all of the rejections on appeal.

REVERSED

lp