Court Zaps The Cardinal’s PCR HIV Patents

Posted on | September 9, 2008 | No Comments

2008-1509, 1510 Leland Stanford Junior University v. Roche
ND/CA 05-04158
Judge Marilyn Hall Patel

Stanford appeals from the summary judgment opinion of Judge Marilyn Hall Patel finding that its 5,968,730; 6,503,705 and 7,129,041 patents invalid as obvious under § 103.  The patents involve correlating measurements of HIV nucleic acids obtained via a polymerase chain reaction (PCR) assay with determining whether or not a particular therapy is effective.

The earliest patent was filed in 1995 and the key prior art reference is a 1991 article in the Journal of Infectious Diseases–an article written by some of the named inventors on the patents.

The court analyzed each of the approx. 23 claims asserted under Graham, and found each obvious in light of the JID article.  The article clearly disclosed 1) extraction of the HIV RNA from serum or plasma (“extraction”); 2) copying the single-stranded HIV RNA into a double-stranded DNA molecule (“reverse transcription”); 3) using PCR to make millions of copies of the DNA using primers (“amplification”); 4) detecting the amplified DNA using a DNA probe (“detection”); and 5) generation of a standard curve used to calculate the amount of virus in a patient’s blood using a cRNA standard (“quantitation”).

The article did, however, differ from the claims.  Roche asserted that the JID article differed from the asserted patents only to the extent that the patent claims teach the use of this assay to determine therapy effectiveness using results of the assay performed on patients receiving drug therapy.  Stanford countered, arguing that:

at the time of invention, the understanding and teachings of those skilled in the art contradicted and taught away from the idea that HIV RNA could be detected and quantified as an effective marker of HIV infection and therapy effectiveness. Stanford contends that the prevailing theories taught that monitoring disease progression and drug efficacy would effectively result from observing the stable, proviral HIV DNA—not HIV RNA. Minimally, Stanford contends, these contradictory theories create a genuine issue of material fact sufficient to preclude summary judgment.

The court reviewed the prior art research cited by Stanford and concluded that "the only thing that is clear from the library of references submitted by the parties is that HIV research–particularly quantification using PCR and identification of surrogate markers–was in a state of uncertainty."

This uncertainty, however, did not preclude summary judgment.  According to the court, Stanford’s choice in using HIV RNA, while perhaps not obvious, misses the point.  The claims are not directed at Stanford’s selection process, but rather at a method of quantifying HIV RNA and using the results to evaluate drug efficacy.  This, the court found, was obvious in light of the JID article.  Given the JID article, it was obvious to use the process described to evaluate a drug treatment regiment:

Here, however, the choice to research HIV RNA as a potential marker was merely a preliminary step which led, ultimately, to the claimed methods in the instant action. The choice of HIV RNA as a surrogate marker is not what was patented. Consequently, this choice is not determinative in the obviousness analysis.

Common sense, along with coherence under the penumbra of patent law, dictate this result. It is clear that if Stanford had patented the five-step method of quantifying HIV RNA that is specifically disclosed in the JID article, it would have been anticipated by the prior art. No analysis regarding the state of the art at the time would have been necessary. However, if any changes—even obvious ones—are made to the process contemplated by the JID article, the analysis shifts to one of obviousness, not anticipation. Consequently, under Graham and KSR, this court is required to analyze the scope and content of the prior art, an analysis wholly missing under the anticipation rubric. This prompts the question: can a purported inventor add an obvious limitation to a piece of prior art in order to secure a method patent and then claim that because his choice of selecting the particular piece of prior art was non-obvious, the patent cannot be invalid as obvious? This cannot be so. Unless the selection process is somehow incorporated into the patent claims, the claims must be viewed independently of the selection process.

Thus, the court distinguished Takeda Chem. Indus., Ltd. v. Alphaparhm Pty, Ltd., 492 F.3d 1350 (Fed. Cir. 2007), where the asserted patents claimed a genus and particular pharmaceutical chemical compounds–the Stanford claims at issue were method claims rather Takeda’s product claims.

More reading:

Summary Judgment

Comments