Bd. of Trs. of the Leland Stanford Junior Univ. v. Chinese Univ. of Hong Kong, 2015-2011

• Court: United States Courts of Appeals. United States Court of Appeals for the Federal Circuit
• Writing for the Court: O'Malley, Circuit Judge.
• Citation: 860 F.3d 1367
• Parties: The BOARD OF TRUSTEES OF the LELAND STANFORD JUNIOR UNIVERSITY, Plaintiff-Appellant v. The CHINESE UNIVERSITY OF HONG KONG, Defendant-Appellee
• Docket Number: 2015-2011
• Decision Date: 27 June 2017

860 F.3d 1367

The BOARD OF TRUSTEES OF the LELAND STANFORD JUNIOR UNIVERSITY, Plaintiff-Appellant
v.
The CHINESE UNIVERSITY OF HONG KONG, Defendant-Appellee

2015-2011

United States Court of Appeals, Federal Circuit.

Decided: June 27, 2017

Edward R. Reines , Weil, Gotshal & Manges LLP, Redwood Shores, CA, argued for plaintiff-appellant. Also represented by Derek C. Walter , Michele Gauger .

Charles E. Lipsey , Finnegan, Henderson, Farabow, Garrett & Dunner, LLP, Reston, VA, argued for defendant-appellee. Also represented by Steven O'Connor ; Jennifer Swan , Robert F. McCauley , Jeffrey Daniel Smyth , Lily Lim , Palo Alto, CA; Howard Warren Levine , Washington, DC.

Before O'Malley, Reyna, and Chen, Circuit Judges.

O'Malley, Circuit Judge.

The Board of Trustees of the Leland Stanford Junior University ("Stanford") appeals from orders of the Patent Trial and Appeal Board ("Board") in three interference proceedings between Stanford and the Chinese University of Hong Kong ("CUHK"). In all of these proceedings, the Board found that Stanford's claims were unpatentable for lack of written description. See Quake v. Lo , No. 105,920 (P.T.A.B. Apr. 7, 2014); Lo v. Quake , No. 105,923 (P.T.A.B. Apr. 7, 2014); Lo v. Quake , No. 105,924 (P.T.A.B. Apr. 7, 2014).¹ Because we conclude that the Board relied on improper evidence to support its key findings and did not cite to other substantial evidence to support its findings, we vacate the Board's interference decisions and remand for further proceedings.

860 F.3d 1370

I. BACKGROUND

A. Technology and Patents

This appeal concerns testing methods for fetal aneuploidies, conditions in which a fetus either has an abnormally high number of chromosomes (e.g., Down's syndrome, a result of trisomy 21 ) or an abnormally low number of chromosomes (e.g., Turner's syndrome, a result of a missing copy of an X chromosome). Prior to the methods developed by the inventors involved in this appeal, physicians typically diagnosed fetal aneuploidies using invasive amniocentesis or chorionic villus sampling procedures. Doctors used less invasive testing methods for identifying aneuploidies, such as ultrasonography and biochemical marker detection, but these methods had suboptimal diagnostic accuracy.

The two competing inventors in the underlying interferences on appeal—Stanford professor Stephen Quake and CUHK professor Dennis Lo—both developed methods for diagnosing aneuploidies using cell-free fetal DNA ("cff-DNA") from maternal blood samples. In 1997, Lo and a colleague discovered that cff-DNA circulates in maternal blood in small amounts. This discovery made possible new prenatal screening techniques for chromosomal and other abnormalities, but researchers developing techniques for assaying cff-DNA had to overcome interference from maternal DNA in the maternal blood sample.

In 2006, Quake developed a "digital analysis" method to detect small changes in the quantity of an aneuploid chromosome relative to the quantity of one or more normal chromosomes, without distinguishing between maternal and fetal DNA. '018 patent, col. 1, ll. 46–60; col. 2, ll. 7–9; col. 7, ll. 46–61. Quake describes "a method of differential detection of target sequences in a mixture of maternal and fetal genetic material."Id. col. 4, ll. 43–45. The '018 specification explains that the approach "involves the separation of the extracted genomic material into discrete units so that the detection of a target sequence (e.g., chromosome 21) may be simply quantified as binary (0, 1) or simple multiples, 2, 3, etc." Id. col. 1, ll. 49–52.

Quake's specification outlines the four steps in his method: (1) obtaining a maternal tissue sample, preferably blood; (2) distributing single DNA molecules from this sample to a number of discrete reaction samples; (3) "[d]etecting the presence of the target in the DNA in a large number of reaction samples"; and (4) performing "[q]uantitative analysis of the detection of the maternal and fetal target sequences." Id. col. 8, l. 35–col. 9, l. 6. The method requires a large number of samples, as only a small amount of cff-DNA is present in a maternal sample. The specification clarifies that the digital PCR technique, in which a known target DNA sequence in a reaction well is amplified by target-specific primers, is the preferred embodiment for amplifying and detecting target sequences.

The capabilities of second-generation massively parallel sequencing ("MPS") are useful for performing Quake's method, as this method can process large numbers of DNA samples simultaneously. Quake's specification discloses that second-generation MPS can be used for counting chromosomes through DNA sequencing using the Illumina sequencing platform. Id , col. 19, l. 59–col. 20, l. 3. MPS can be performed by "random" or "targeted" methods. In the random format, all DNA in a sample is linked to a leader sequence and amplified using a primer complementary to the leader. Appellee Br. 10. In the targeted format, the target sequence is specifically amplified, and then sequenced.

Quake claimed his method in an application filed on February 2, 2007; this application issued as

860 F.3d 1371

U.S. Patent No. 7,888,017 ("the '017 patent"). Quake filed continuation application no. 12/393,803 ("the '803 application") in February 2009. This continuation application issued as the '018 patent at issue in this appeal. The '017 and '018 patents share the same specification.

The original claims of Quake's '803 application explicitly recited methods that required the detection of "target sequences." For example, claim 1 of the '803 application read:

1. A method of differential detection of target sequences in a mixture of maternal and fetal genetic material, comprising the steps of:

a) obtaining maternal tissue containing both maternal and fetal genetic material;

b) distributing the genetic material into discrete samples, each sample containing on average not more than about one target sequence per sample;

c) measuring the presence of different target sequences in the discrete samples; and

d) analyzing a number of the discrete samples sufficient to obtain results distinguishing different target sequences .

J.A. 3253 (emphasis added).

In 2011, Quake cancelled all pending claims in the application which later issued as the '018 patent, and added new claims.² A representative later-added claim from the '018 patent states:

1. A method for determining presence or absence of fetal aneuploidy in a maternal tissue sample comprising fetal and maternal genomic DNA, wherein the method comprises:

a. obtaining a mixture of fetal and maternal genomic DNA from said maternal tissue sample;

b. conducting massively parallel DNA sequencing of DNA fragments randomly selected from the mixture of fetal and maternal genomic DNA of step a) to determine the sequence of said DNA fragments;

c. identifying chromosomes to which the sequences obtained in step b) belong;

d. using the data of step c) to compare an amount Stanford of at least one first chromosome in said mixture of maternal and fetal genomic DNA to an amount of at least one second chromosome in said mixture of maternal and fetal genomic DNA, wherein said at least one first chromosome is presumed to be euploid in the fetus, wherein said at least one second chromosome is suspected to be aneuploid in the fetus, thereby determining the presence or absence of said fetal aneuploidy.

'018 patent, col. 33, ll. 48–67 (emphasis added).

Lo's "random sequencing" method uses random MPS and does not require the detection of specific target sequences. The first step of Lo's method is to obtain a maternal blood sample, containing both maternal and cff-DNA. The researcher then sequences the mixed maternal and cff-DNA from the blood sample using random MPS. The sequence fragments obtained from random MPS are then aligned to a reference genome to determine a chromosome or chromosomal region of origin for each sequence. Once the chromosome fragments have been mapped to their respective chromosomes of origin, the researcher can compare the overall number of sequences mapped to each chromosome. A disproportionate number (e.g., greater frequency) of aligned sequences to

860 F.3d 1372

chromosome 21 reveals the presence of a Down's syndrometrisomy.

Lo filed provisional application no. 60/951,438 describing the "random sequencing" method on July 23, 2007, and subsequently filed application no. 12/178,181 on July 23, 2008. This application published in January 2009.

B. Interference History

CUHK claims that, in 2011, Quake realized that CUHK had claimed the "random sequencing" method. Quake then cancelled all pending claims in the application that later issued as the '018 patent, and added the claims listed above that, for the first time, explicitly cover random MPS methods. CUHK also claims that the '833 application copied claims from Lo. Stanford contends that random MPS is disclosed in the specification and supports these later-filed claims.

Both Quake and Lo requested interferences to determine who invented the random sequencing method, and when the method was invented. In early 2013, the PTO declared three interferences between Quake's patents and applications and Lo's...