Not surprisingly, the Fed. Cir., on December 2nd, denied Sequenom’s petition for rehearing en banc of the invalidation of certain of the claims of U.S. Pat. No. 6,258,540 which were directed to methods of detecting and using cffDNA. (Appeal no. 2014-1139, 2014-1144). (A copy of the decision can be found at the end of this post.) Judge Newman was the lone dissenter. Judge Lourie, joined by Judge Moore, and Judge Dyk wrote separate concurrences that read like dissents (“Mayo made us do it!”). The concurrences are not lengthy, but they bring into focus – finally – the basics of the s. 101 debate. A summary of their opinions might go like this: Disembodied abstract ideas and natural phenomena like mathematical formulae and drug metabolites in the blood –or cffDNA per se– are not patent-eligible. However, real-world applications of the phenomena, such as a new method of performing a specific prenatal diagnosis using amplified cffDNA, as recited in claim 21 of the ‘540 patent, should be found to meet the Mayo two-part test for patent-eligibility.
That is, if the claims are drafted narrowly enough, according to Lourie, Moore and Dyk. And drafted narrowly enough, such “application claims” are not abstract ideas, either. As Judge Lourie writes: “Moreover, the claims here are not abstract. There is nothing abstract about performing actual physical steps on a physical material.” This would appear to be an attempt to distinguish the ‘540 diagnostic claim 21 from diagnostic claim 2 of U.S. Patent No. 6,033,857, which Judges Lourie and Moore invalidated as an attempt to patent an abstract mental process in Myriad. In 2012, they wrote: “[Claim 2]…consist(s) of analyzing and comparing certain DNA sequences [and is] not patent-eligible subject matter on the ground that they claim only abstract mental processes.”
While Judges Lorie and Moore do not directly distinguish their earlier Myriad opinion, Judge Dyk does. He cites both the 2012 Myriad decision and “Myriad II” as improperly presenting claims that sought to capture “’all comparisons between the patient’s BRCA genes and the wild-type BRCA genes‘ to be overbroad and thus ineligible under s. 101, noting that ‘[t]he covered comparisons are not restricted by the purpose of the comparison or the alteration being detected.’” 774 F.3d 755, 763, 765 (Fed.. Cir. 2014). As an example of a “good” diagnostic claim, Judge Dyk points to claim 21 of Myriad’s U.S. Pat. No. 5,753, 441 that “covers a method of detecting any of several specific mutations in the BRCA1 gene, newly discovered by the patent applicant and shown to increase a person’s risk of developing particular cancers, using conventional methods.” See 774 F.3d at 765.
Still, Judge Dyk takes time to approve of the Mayo decision, which I have called “an old use of an old compound” and he calls “conventional applications of well-known laws of nature,” and criticizes diagnostic claim 21 of the ‘540 patent because it “still broadly encompasses any diagnosis of any disease, disorder, or condition….Such claims appear to be impermissible attempts to capture the entire natural phenomenon of cffDNA rather than any particular applications thereof developed and actually reduced to practice by the inventors.” In case you missed that last phrase, Judge Dyk would require both narrow claims and actual reduction to practice for both “diagnostic and therapeutic method patents” to “survive.” The commentators who predicted that method of medical treatment claims were going to get drawn into this quagmire were correct.
The Fed. Cir.’s “great dissenter” Judge Newman, would rely on the Supreme Court’s dicta in Mayo that “a new way of using an existing drug” is patent-eligible, and notes that “[i]n the case now before us, the claimed method was not previously known, nor the diagnostic knowledge and benefit implemented by the method.”
While this decision is not aimed to please the developers of companion diagnostics and personalized medicine, the battle lines have at least become somewhat more clearly drawn.