Ariosa was a decision that essentially held that the novel discovery of a naturally-occurring phenomenon could not per se meet the Mayo/Alice requirement for an inventive concept, even though it was of “groundbreaking importance” in advancing medicine. The claims appealed in Ariosa were not directed to specific diagnostic conclusions, so I continued to cling to the possibility that the Fed. Cir., faced with a simple “If A, then B” diagnostic claim, might rule differently. Of course, the holding in Cleveland Clinic erased that hope.
The Fed. Cir. panel found that claims to diagnosing the presence of atherosclerotic cardiovascular disease (CAD) by measuring the blood level of an enzyme, myeloperoxidase (MPO) in a patient and comparing it to the level in a control subject diagnosed as not having the disease did not pass the Mayo/Alice test. The panel cited Arisoa as supporting its finding that this method “starts and ends with naturally occurring phenomena with no meaningful non-routine steps in between–the presence of MPO in a bodily sample is correlated to its relationship to [CAD]. The claims are therefore directed to a natural law”. Slip op. at 14.
The part of the decision that set off some legal alarm bells, at least for me, was the panel’s trivializing the techniques used to establish the controls and a relationships between MPO level in a patient vs. the control level in a patient population not having CAD. The panel wrote:
“[T]he claims instruct that MPO levels be detected or determined using any of these known techniques. The claims of the testing patent also contain a ‘comparing’ step where MPO levels are compared to statistically derived control or predetermined values. Here too, the Cleveland Clinic does not purport to derive new statistical methods to arrive at the predetermined or control levels of MPO that would indicate a patient’s risk of [CAD].” Slip op. at 16.
The panel goes on to describe the “known statistical models” by quoting eleven lines from the specification of one of the patents-in-suit. Apart from the fact that this decision supports the dreaded-but-expected “two inventions” rule by requiring that patentees have a second novel – if not unobvious – invention in their claims involving natural laws, the specification of the ‘260 patent (which I am working from) discloses an extensive amount of picking and choosing from various in vivo and environmental parameters to arrive at the control values of MPO and the significance of the differences between the control and patient populations MPO levels. See, e.g., all of Example 1 and Table I, which require 5 columns of text to explicate.
Shortly after the Mayo decision came down, one commentator argued that the recitation of undesired “side effects” in the claims is a reflection of human preference. Likewise, in this case, the presence or absence of CAD in the control populations, per se, was the result of value judgments made by clinicians, not absolute numbers spit out by a computer.
For example, one factor supporting the definition of CAD given in Ex. 1 requires at least 50% stenosis in one or more arteries, and that is just one of several factors defining CAD, including environmental factors like smoking and family history, and this set must be assembled before statistical analysis can begin.
It is the specification that informs the POSA how to practice the invention, not the claims, but you would not know that from the panel’s general description of how the method is carried out. It defies belief that the application of columns of selection and statistical analysis to the problem solved by the invention is not novel, or does not rise to the level of an inventive concept. Maybe Cleveland Clinic should file a divisional application and claim it.