Personalized Medicine: A Patent Office Perspective

By Kevin E. Noonan --

On December 3rd, the biotechnology, chemical, and pharmaceuticals (BCP) technology groups at the U.S. Patent and Trademark Office held their most recent quarterly customer partnership meeting.  During one of six presentations, Kathleen Bragdon, a Quality Assurance Specialist from Technology Center 1600, presented "A Look at Personalized Medicine."  In prior years, this talk may have been anticipated to be a prospective look at a developing technology area, or represent the Office's attempt to open a dialog with applicants and their representatives.  But these have not been normal times, and the presentation contained a few unpleasant but not entirely unexpected surprises.

The majority of the presentation was a competent, but by no means comprehensive, review of personalized medicine (referencing scientific journal articles and Wikipedia) and pharmacogenetics.  Warfarin, and the use of genetic screening of single nucleotide polymorphisms (SNPs) to determine patients for whom warfarin treatment may be contraindicated, was used as a case study.  Ms. Bragdon also reviewed FDA approval of genetic testing of two human genes (VKORC1 and CYP2C9) having SNPs predictive for warfarin sensitivity.

The talk featured several examples of how these relationships between genotype and drug treatment could be claimed, including the following:

3. A method for determining whether a human subject having breast cancer will be effectively treated with "breast cancer drug X", said method comprising:
a) considering data in a database comprising genetic patient information about the ERBB2 gene at position 101 of SEQ ID NO:1; and
b) correlating the presence of a cytosine at position 101 of SEQ ID NO:1 with effective treatment of the human subject with "breast cancer drug X".

This is an invalid claim, according to Ms. Bragdon, for failing to qualify as statutory subject matter under the Bilski decision, because it is neither tied to a machine or apparatus nor performs a transformation (see ".  The following claim, on the other hand, would be statutory subject matter:

4. A method for treating a human subject having breast cancer, said method comprising:
a) obtaining a nucleic acid sample from said human subject;
b) subjecting the sample to PCR and identifying the nucleotide present at position 101 of SEQ ID NO:1; and
c) treating the human subject with "breast cancer drug X" when a cytosine is detected at position 101 of SEQ ID NO:1.

A comparison between these claims illustrates the Office's solution to how Bilski (or whatever form a future LabCorp decision may take regarding these kinds of claims) may be applied to biotechnology (or more specifically, molecular diagnostic) claims (see "The Relevance of In re Bilski to the Patentability of the Metabolite Claim").  By casting the claim as a method of treatment, and tying the step of identifying patients bearing a diagnostic SNP to a treatment step, the policy and philosophical problems presented by the LabCorp case are apparently solved.  While the identification step can still be interpreted to be either a "mental step" or a mere recitation of a natural principle, treatment methods are clearly statutory subject matter (in the U.S.).

One difficulty with this solution, however, is that it may frequently be the case that the actor who performs the diagnostic step of determining whether a patient bears a diagnostically-relevant SNP is a different actor than the one who performs the method of treatment step.  This dichotomy of action then directly implicates the principle enunciates in the Muniauction case, that a claim cannot be infringed by joint tortfeasors unless there is one actor exercising direction and control over the others (see "Adequate Method Claiming Requirements:  Muniauction, Inc v. Thomson Corp.").  This situation may frequently be the case in the practice of claims such as claim 4 in the presentation.  However, there may be instances when it is not the case.  In these cases, the Office's solution to the LabCorp problem will be unavailing.  The introduction of method of treatment claims in this regard is also problematical insofar as it implicates the infringement exemption embodied in 35 U.S.C. § 287(c) for physicians and other medical personnel.

The presentation also discusses the Office's interpretation of how the enablement requirement may be implicated in personalized medicine claims.  In this regards, Ms. Bragdon discussed the following claim:

5. A method for treating a human subject having breast cancer, said method comprising:
a) obtaining a nucleic acid sample from said human subject;
b) subjecting the sample to PCR and identifying the nucleotide present at position 101 of SEQ ID NO:1; and
c) treating the human subject with "breast cancer drug X" when a cytosine is detected at position 101 of SEQ ID NO:1.

The presentation supplies the (putative) disclosure from the specification supporting this claim, that SEQ ID NO:1 is a variant of the ERBB2 gene having an A (adenine) to C (cytosine) mutation at position 101 (A101>C).  This mutation (A101>C) is typically found in breast cancer patients, according to the hypothetical, and it correlates with a significantly better response to treatment with "breast cancer drug X" versus placebo (i.e., without the mutation, breast cancer drug X is ineffective).  Finally, the hypothetical specification does not distinguish among patient populations tested, but teaches that variability in treatment responses among patient populations may be an unpredictable (emphasis in original) factor in SNP correlation studies (presumably, this factor in the analysis could also be something within the skill of one having ordinary skill in the art).

The enablement issue arises under this hypothetical because post-filing data establishes that there is a patient population that does not show the claimed correlation between the mutation in the ERBB2 gene and responsiveness vel non to breast cancer drug X.  Ms. Bragdon did not assert that under these circumstances the claim would necessarily be invalid for lack of enablement, but suggested that "[t]he appropriateness of making any enablement rejection should be considered based on the foregoing facts."

While it is certainly the case that post-filing evidence can be used to show that a claim is not enabled, here it seems that the more appropriate question would be whether the specification provided sufficient guidance to enable the skilled worker to determine whether a particular patient (as opposed to a patient population) would bear a predictive SNP, and whether the practice of the claimed invention with a particular patient having a particular SNP to predict drug sensitivity would require undue experimentation.  Perhaps Ms. Bragdon was performing her analysis in view of the Federal Circuit's Monsanto Co. v. Syngenta Seed, Inc. case, where the Court found invalid claims reciting genetic transformation of plant cells where certain plant cells (monocots) could not be transformed at the time the application was filed, and the specification provided neither any teaching regarding how to transform monocot cells nor disclaimer or limitation to transformable dicot cells.  However, the Office appears to be suggesting that claims like claim 5 may be per se unpatentable in the face of after-filing evidence that not all patients fall within the scope of the claim.  This seems an unnecessary conclusion and extension of the requirements for enablement under 35 U.S.C. § 112, first paragraph.

A copy of Ms. Bragdon's presentation can be obtained here.

Additional BCP presentations will be discussed in future posts.  For information regarding other BCP posts, please see:

• "USPTO Implements New Program to Teach Examiners How to Read and Understand Case Law," December 4, 2008.

Comments

[Norman Siebrasse]

Thanks very much for this post. The presentation is something I would never find out about otherwise and your analysis of the problems with the USPTO approach is very helpful. My own view is that Bilski is a mess. They wanted to step back on business method patents, which is fine in itself, but the majority opinion is an incoherent hodge podge of doctrines that will have unintended consequences well outside the business methods area. (I can't blame the CAFC - the incoherence starts with the USSC computer trilogy.) From your post, it looks like we're already seeing the beginning.