In Cleveland Clinic Foundation v. True Health Diagnostic LLC, the Federal Circuit (CAFC) dealt another blow to the patent eligibility of diagnostic methods and the growing field of personalized medicine.
Cleveland Clinic obtained three patents (US Patent Nos. 7,223,552, 7,459,286 and 8,349,581) directed to the use of the activity and/or levels myeloperoxidase (MPO) or a MPO oxidation products (the foregoing referred to as simply MPO in this discussion) to characterize/assess/predict a risk of cardiovascular disease in a subject. At a high level, the claims required: 1) determining a level/activity of MPO in a subject; 2) comparing this value with a control value; 3) and determining risk of disease when the MPO level/activity is increased as compared to the control. The claims in the patents also describes various forms and features of cardiovascular disease, such as when a cardiovascular event might occur.
The CAFC used the Alice two-step framework and found the claims were directed to “multistep methods for observing the law of nature that MPO correlates to cardiovascular disease” under step one and no feature of the claims, either alone or in an ordered combination, was sufficient to transform the law of nature into a patent-eligible concept under step two (citing the use of prior art techniques to detect MPO and statistical methods to determine the level of MPO and the control value). There was nothing dramatically different in the court’s Alice analysis than in previous cases.
Is a Correlation a Law of Nature?
The courts have consistently refused to provide a concrete definition of what is a natural law or an abstract idea. The courts rely heavily on determinations in prior cases to guide the inquiry. It could be argued there is no “natural law” that a subject with a high level of MPO will have an increased risk of cardiovascular disease. In some cases the correlation will not prove true, but in some cases it will. In many subjects, a high level of MPO will be asymptomatic (perhaps due to other aspects of the subject’s physiology). Likewise, many subjects with low MPO levels will have cardiovascular disease due to factors unrelated to MPO levels (for example, poor lifestyle or genetic predisposition). Furthermore, the correlation was shown to vary in certain aspects depending on the population studied and the time frame with which a cardiovascular event might occur. Given these discrepancies, is a correlation that holds true for only a portion of a population or only under certain defined conditions properly defined as a natural law?
Characterizing the Improvement over the Prior Art
Let’s take a step back and look at the claims in the context of the prior art. The claimed methods allow an improvement in the technological field or medical care. The simple blood test allows for the identification of patients that may benefit from additional care without invasive testing at a low cost using a simple procedure. The CAFC recognized the difference in the claimed methods (one of which detected MPO in the blood but was not predictive of cardiovascular disease) and the benefits of the claimed methods. Viewed in this manner, the claims are similar to those in Diamond v. Diehr, where an abstract idea was held to be patent eligible when the abstract idea was used in a process designed to solve a technological problem in conventional industry practice.
The CAFC distinguished this case from Rapid Litigation Management, Ltd. v. CellzDirect, where claims directed to a method of cryopreserving cells were eligible under §101, as we discussed in a previous post. In CellzDirect, the basis for the claimed methods was the discovery of an inherent property of liver cells to undergo multiple freeze-thaw cycles while still maintaining viability. The methods claimed in CellzDirect, such as those here, did not invent new techniques for the cryopreservation process (prior art methods and reagents were used), but rather claimed additional steps in the cryopreservation process based on the discovered natural property of liver cells to provide a new and useful laboratory technique. The method claims in this case take the presumed natural law of MPO correlation to certain types of cardiovascular disease and add additional limitations (for example, time frames for the onset of disease) or steps (for example, comparison to a control value) to produce a new and useful laboratory technique.
One possible reason for the holding in the present case may be the lack of specific information on how an “elevated” MPO level is determined, how the control value is determined, or how much above the control value the MPO level in the subject needs to be in order to be at risk for the defined conditions. The provision of such a defined decision point would have allowed arguments that the claims were directed not to the presumed natural law, but to a specific value derived by the hand of man. As such, the claims would not “start and end with the naturally occurring phenomenon” as prohibited in Ariosa.
Such a decision point could be implemented using an algorithm or “rule set” that provides cut-off values and criteria for the evaluation of the subject’s risk based on MPO levels and possible other risk factors (the specification discusses several, such as troponin-T). Such a rule set could also specify comparisons among discrete populations. While the use of such rule sets obviously limits claim scope to some extent, defining the claimed relationships with greater specificity offers additional arguments in the eligibility battle. An additional benefit is this approach potentially allows the omission of an explicit comparison step (as any comparison is accomplished by the rule set).
The use of rule sets was favorably viewed by the CAFC in McRO v. Bandai, noted in a previous post, where the court allowed a rather generic rule set to provide for §101 eligibility in a method of computer animation. The CAFC noted that the rule set in McRO “transformed a traditionally subjective process performed by human artists into a mathematically automated process executed on computers.” Claims that define the MPO correlation to a baseline value in determining an outcome should be viewed as transforming the subjective process of diagnosis into a defined process carried out by the claimed methods. While McRO is directed to the computer arts, the mantra of the CAFC is that each technological field is treated the same in terms of patent eligibility analysis.
While the CAFC held the claims at issue here to be invalid under §101, I do not believe the door is shut for meaningful patent claims directed to diagnostic methods and their use in personalized medicine. While the prevailing judicial view may require the recitation of additional elements with more specificity, thoughtful drafting of the specification and claims provides an opportunity for claims of the type invalidated here to withstand the scrutiny of the §101 analysis.