Foundation Medicine Inc v. Guardant Health Inc

1. Case Identification

• Case #: IPR2019-00634
• Patent #: 9,840,743
• Filed: February 1, 2019
• Petitioner(s): Foundation Medicine, Inc.
• Patent Owner(s): Guardant Health, Inc.
• Challenged Claims: 1-26

2. Patent Overview

• Title: Methods for Detecting Genetic Variations in Cell-Free DNA
• Brief Description: The ’743 patent discloses methods for detecting genetic variations, such as copy number variations (CNVs) and rare mutations, from cell-free DNA (cfDNA) in a bodily sample. The methods involve sequencing the cfDNA, filtering the sequence reads based on quality, mapping them to a reference genome, and performing quantitative analysis to identify the variations.

3. Grounds for Unpatentability

Ground 1: Anticipation of Claims 1-3 and 22-26 by Chiu

• Prior Art Relied Upon: Chiu et al. (2008), “Noninvasive prenatal diagnosis of fetal chromosomal aneuploidy by massively parallel genomic sequencing of DNA in maternal plasma,” Proc. Natl. Acad. Sci. USA, 105(51).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Chiu, which was not before the Examiner, discloses every limitation of claim 1 for detecting CNV. Chiu teaches a method for diagnosing fetal aneuploidy (a type of CNV) by sequencing cfDNA from maternal plasma. Chiu explicitly discloses sequencing extracellular polynucleotides, mapping them to a human reference genome, quantifying the reads per chromosome (predefined regions), and determining CNV. Petitioner contended that Chiu’s step of limiting its analysis to “unique sequences” that map to only one location inherently meets the claimed limitation of “filtering out reads that fail to meet a...mapping score threshold.” Finally, Chiu determines CNV by both normalizing the number of unique reads for each chromosome against the total number of reads and by processing this data against a control sample using a z-score analysis.
  • Key Aspects: Petitioner asserted that Chiu’s disclosure of filtering based on unique mappability directly anticipates the very limitation added during prosecution to overcome the Examiner’s initial rejection of the ’743 patent claims.

Ground 2: Claims 1-9, 20, and 21-26 are obvious over Kinde 2012 in view of Kinde 2011

• Prior Art Relied Upon: Kinde et al. (2012), “FAST-SeqS: A Simple and Efficient Method for the Detection of Aneuploidy by Massively Parallel Sequencing,” PLoS ONE, 7(7) (“Kinde 2012”); and Kinde et al. (2011), “Detection and quantification of rare mutations with massively parallel sequencing,” Proc. Natl. Acad. Sci. USA, 108(23) (“Kinde 2011”).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Kinde 2012 teaches a method for detecting fetal aneuploidy (a CNV) that meets most limitations of claim 1, including sequencing cfDNA, mapping reads to predefined regions of a reference genome, quantifying reads, and determining CNV by normalizing counts and comparing them to a control sample via z-score analysis. While Kinde 2012 teaches filtering for quality, Kinde 2011 explicitly teaches filtering out reads that fail to meet a specific quality score threshold (e.g., quality score ≥20), rendering this step obvious. Furthermore, Kinde 2011 teaches assigning unique identifiers (UIDs or "barcodes") to DNA molecules before amplification to create "unique sequence reads," and Kinde 2012 suggests that incorporating this method could increase its accuracy. Thus, a POSITA would find it obvious to apply the analytical steps of Kinde 2012 to the "unique sequence reads" generated by the method of Kinde 2011.
  • Motivation to Combine: A POSITA would combine the references because Kinde 2012, from the same author group as Kinde 2011, explicitly incorporates and suggests further integration of Kinde 2011’s techniques to improve accuracy by uniquely identifying template molecules. The combination would use a known technique (from Kinde 2011) to improve a similar, later-developed method (from Kinde 2012).
  • Expectation of Success: A POSITA would have a high expectation of success, as the two methods employ the same underlying sequencing and library preparation technologies, and the authors of Kinde 2012 already demonstrated success in combining aspects of the two methods.

Ground 3: Claims 10-13, 15-19, and 21 are obvious over Kinde 2011 in view of Forshew

• Prior Art Relied Upon: Kinde 2011 (a 2011 journal article) and Forshew et al. (2012), “Noninvasive Identification and Monitoring of Cancer Mutations by Targeted Deep Sequencing of Plasma DNA,” Sci. Transl. Med., 4(136) (“Forshew”).

• Core Argument for this Ground:

  • Prior Art Mapping: This ground targets the claims for detecting rare mutations. Petitioner argued that Forshew teaches a method for detecting rare cancer mutations in cfDNA that discloses most limitations of independent claim 10. Forshew teaches sequencing cfDNA, mapping reads, and for each mappable base position, calculating an allele frequency—the claimed ratio of variant reads to total reads. Forshew also teaches "processing" this ratio by comparing it to a reference sample (e.g., a patient's normal DNA or samples from earlier time points). Kinde 2011 addresses the known problem of sequencing errors by teaching the use of UIDs to uniquely identify original DNA molecules, allowing for error correction. It also explicitly teaches filtering reads based on a quality score threshold.
  • Motivation to Combine: A POSITA would combine Kinde 2011 with Forshew to improve the accuracy of rare mutation detection. Forshew acknowledges the challenge of avoiding false positives from PCR errors, and Kinde 2011 provides a well-known solution for minimizing such errors through its UID-based error correction method.
  • Expectation of Success: A POSITA would have a reasonable expectation of success because both methods utilize standard next-generation sequencing on cfDNA fragments. Combining the techniques would involve the predictable step of attaching Kinde 2011's barcoded adapters during the library preparation stage taught by Forshew.

• Additional Grounds: Petitioner asserted an additional obviousness challenge for claim 14 over the combination of Kinde 2012, Kinde 2011, and Forshew, relying on similar combination rationales.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that the Board should not exercise its discretion to deny institution under 35 U.S.C. §325(d). Although Kinde 2011 was cited on the face of the ’743 patent, it was one of over 400 references and the Examiner never relied on it or gave any indication of having appreciated its significance. More importantly, Petitioner’s primary references for its main grounds—Chiu, Kinde 2012, and Forshew—were never presented to or considered by the Examiner. Therefore, the petition raised substantial new arguments and evidence that were not previously before the USPTO.

5. Relief Requested

• Petitioner requests institution of inter partes review and cancellation of claims 1-26 of the ’743 patent as unpatentable.