Illumina, Inc. v. The Trustees of Columbia University in the City of New York

1. Case Identification

• Case #: IPR2018-00797
• Patent #: 9,868,985
• Filed: March 16, 2018
• Petitioner(s): Illumina, Inc.
• Patent Owner(s): The Trustees of Columbia University in the City of New York
• Challenged Claims: 1-2

2. Patent Overview

• Title: Method for Sequencing a Nucleic Acid
• Brief Description: The ’985 patent describes a method for sequencing a nucleic acid by detecting the identity of a specific nucleotide analogue as it is incorporated into a growing DNA strand. The method relies on using nucleotide analogues that have a removable 3'-OH capping group and a detectable fluorescent tag attached to the base via a cleavable linker.

3. Grounds for Unpatentability

Ground 1: Obviousness of Claim 1 over Tsien in view of Prober

• Prior Art Relied Upon: Tsien (WO 91/06678) and Prober (a 1987 article in Science).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Tsien taught the fundamental sequencing-by-synthesis (SBS) method, including using nucleotide analogues with a 3'-OH blocking group and a base-attached fluorescent label. Specifically, Tsien disclosed a 3'-O-allyl group as a "small," chemically cleavable blocking group (the "R" group) and the use of a cleavable linker (the "Y" group) to attach a fluorescent tag. Prober taught the known and advantageous method of attaching such labels to the 5-position of pyrimidine bases (cytosine and thymine), which minimizes interference with hybridization. Petitioner contended that the claimed pyrimidine structures (C and D) are merely a pictorial representation of nucleotides previously found unpatentable in prose form in related IPRs against the Patent Owner.
  • Motivation to Combine: A person of ordinary skill in the art (POSA) reading Tsien would be motivated to improve its method using Prober’s well-known base-labeling approach. Tsien itself disclosed 5-substituted pyrimidines and cited Prober for showing enzymatic incorporation, demonstrating a clear link between the references. The known advantages of 5-position labeling—placing the reporter group in the major groove of the DNA duplex and stabilizing the structure—provided strong motivation for the combination.
  • Expectation of Success: The Board and Federal Circuit had previously affirmed that a POSA would have a reasonable expectation of success in combining these references to make 3'-capped nucleotide analogues with a cleavably linked base label. The synthetic steps for creating the claimed analogues were well within the ordinary skill in the art, and the ’985 patent itself acknowledged that the necessary chemistry was "well-established."

Ground 2: Obviousness of Claim 2 over Tsien in view of Prober and Pallas

• Prior Art Relied Upon: Tsien (WO 91/06678), Prober (a 1987 article in Science), and Pallas (WO 98/53300).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground challenged claim 2, which adds the limitation of "simultaneously sequencing a plurality of different nucleic acids." Pallas disclosed a system for the simultaneous analysis of nucleotide sequences from a population of polynucleotides anchored to microparticles in a planar array. Crucially, Pallas expressly suggested that its system could be used with a "base-by-base" DNA sequencing methodology and specifically identified Tsien as a suitable method.
  • Motivation to Combine: The motivation was explicit, as Pallas directly directed a POSA to use Tsien's SBS methodology in its parallel sequencing apparatus. Furthermore, Tsien's own disclosure of using "four separate parallel reaction zones" would have informed a POSA that its methods were compatible with parallel applications.
  • Expectation of Success: A POSA would have had a high expectation of success, given Pallas's express teaching that Tsien's base-by-base method "permits the stepwise identification of a sequence of nucleotides" suitable for its system.

Ground 3: Obviousness of Claims 1-2 over Dower in view of Prober and Metzker

• Prior Art Relied Upon: Dower (Patent 5,547,839), Prober (a 1987 article in Science), and Metzker (a 1994 article in Nucleic Acids Research).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground presented an alternative obviousness rationale. Dower disclosed an SBS method using nucleotide analogues with a removable 3'-blocking group and a fluorescent label, and it repeatedly cited Prober for methods of labeling nucleotides. Prober, in turn, disclosed attaching labels to the 5-position of pyrimidines. Metzker disclosed and demonstrated the successful polymerase incorporation of specific 3'-ether capping groups, including the 3'-O-allyl group, which is small, removable, and not a ketone, methoxy, or ester group, thus meeting all functional requirements for the "R" group in the ’985 patent.
  • Motivation to Combine: A POSA would combine these references to create an optimized SBS system. Dower’s express and repeated citations to Prober would have motivated a POSA to use Prober's 5-position pyrimidine labeling in Dower's method. To implement Dower’s requirement for a "small" blocking group, a POSA would have turned to Metzker, which demonstrated that the 3'-O-allyl group was both small enough for polymerase incorporation and efficiently removable.
  • Expectation of Success: A POSA would have reasonably expected success in combining these elements, as each component's functionality and compatibility was established in the art. The combination would have been a predictable implementation of known, advantageous techniques to achieve the claimed method.

4. Key Claim Construction Positions

• The Petitioner argued that the term "or" in Claim 1, which recites four nucleotide analogues (A), (B), (C), or (D), means the items are alternatives. Therefore, the claim is unpatentable if the use of any single one of the analogues is found to be obvious. The petition's primary focus was on the pyrimidine analogues (C) and (D).

5. Key Technical Contentions (Beyond Claim Construction)

• The Petitioner argued that a pharmaceutical "lead compound" analysis was not required, as the Board had previously rejected this approach for related patents owned by Columbia. However, Petitioner also argued that even if such an analysis were applied, the claims would still be obvious. Tsien’s disclosure of 3'-O-allyl dNTPs provided a clear "lead compound" that a POSA would have been motivated to modify by adding a 5-position label (from Prober) and a cleavable linker (taught by Tsien itself).

6. Arguments Regarding Discretionary Denial

• The Petitioner argued that discretionary denial under §325(d) would be improper. It asserted that the Patent Owner obtained the ’985 patent by re-litigating issues already decided against it by the Board and Federal Circuit in previous IPRs, specifically the inventiveness of a "small" 3'-O-capping group. Petitioner contended that these improper prosecution arguments prevented the Examiner from appreciating the full scope of the prior art. The petition also presented significant new evidence not before the Examiner, including expert testimony from Dr. Romesberg, which further warranted institution.

7. Relief Requested

• Petitioner requested institution of an inter partes review and cancellation of claims 1-2 of the ’985 patent as unpatentable.