Reasonable Expectation of Success to the Rescue

A prima facie case of obviousness requires that where there is a reason to modify or combine prior art to achieve the claimed invention, there must also a reasonable expectation of success.  It is not that common, however, for the Federal Circuit to address this prong of obviousness in much detail.  Interestingly though, in the last several months, lack of a reasonable expectation of success was a major factor in the reversal of two obviousness rulings by the Federal Circuit.  One originated from the Patent Trial and Appeals Board (“Board”) (In Re Stepan Company, Aug. 25, 2017) (“In re Stepan”) and the other from a district court Genzyme Corporation v. Dr. Reddy’s Lab, Dec. 18, 2017, (“Genzyme”).

In In re Stepan, the Court emphasized that reasonable expectation of success requires a motivation to do more than simply vary all parameters or try all possible choices until success is achieved.  In Genzyme, the Court explained that a hypothesis presented in passing in a reference, without more, is not enough for an ordinarily skilled person to have a reasonable expectation of success.  While In re Stepan has relevance for patent applicants facing conclusory obviousness rejections, Genzyme is cautionary for patent challengers banking too heavily on isolated, unsupported statements as a basis for obviousness.

In re Stepan Co.

In re Stepan deals with an appeal by Stepan Company from a decision of the Board affirming examiner’s rejection of all claims of U.S. Patent Application No. 12/456,567 (“’567 application”).  The sole independent claim in the ’567 application is directed to an ultra-high load, aqueous glyphosate salt containing concentrate that includes a surfactant system.  Surfactants make glyphosates more effective as herbicides by improving their adherence to leaves.  The claimed concentrate has a cloud point above at least 70 ºC or no cloud point when the concentrate is heated to its boiling point.  The surfactant system contains three components.

As the opinion explains, a cloud point is the temperature at which a solution becomes cloudy due to the surfactants becoming insoluble and separating into layers.  One can avoid cloudiness by choosing a surfactant having a cloud point higher than the solution’s temperature or by cooling the solution before adding the surfactant.  The ’567 application explains that because glyphosate salt is made at about 75 ºC, it is advantageous to formulate glyphosate with a surfactant system exhibiting a high cloud point so as to avoid having to wait for the glyphosate salt reaction product to cool down.

The vast majority of claims in the ’567 application were rejected for obviousness over a prior art reference disclosing highly-loaded glyphosate compositions containing surfactants with a cloud point of at least 50 ºC and ideally 60 ºC.  The examiner found that the reference teaches preferred surfactants having two of the three required components, the third component being optional, and further that “any combination” of surfactants could be used in the glyphosate compositions.  Id. at 4.  As to cloud point, the examiner stated that while the reference did not teach a cloud point above 70 ºC, achieving this cloud point would have been a matter of “optimizing the formulation” because of the teaching in the reference that the ideal cloud point should be above 60 ºC.  Id.

The Board adopted the examiner’s findings.  The Federal Circuit disagreed.  It found that the Board had failed to articulate why an ordinarily skilled person would have had a reasonable expectation of success to formulate the claimed surfactant system with a cloud point above at least 70 ºC.  Quoting from an earlier opinion, the Court said, “[T]o have a reasonable expectation of success, one must be motivated to do more than merely to vary all parameters or try each of numerous possible choices until one possibly arrived at a successful result.”  Simply reciting the teaching from the reference that “any combination” of surfactants might be used and that a cloud point above 60 ºC was desired was not sufficient.

Lastly, the Court found that the Board indisputably erred when it disregarded as irrelevant some of the evidence presented by Stepan.  Id. at 8.  The evidence showed that none of the examples in the prior art reference included all three of the surfactant components, and that the closest examples, having two of the three components, failed to achieve a cloud point above at least 70 ºC.  Id.  This evidence was clearly relevant to expectation of success.  Not surprisingly, the Court instructed the Board to take the evidence into consideration on remand.

In a nutshell, the Board’s obviousness decision was not upheld because of its failure to make specific factual findings on the issue of expectation of success and also for not considering all relevant evidence on the issue.

Genzyme Corporation v. Dr. Reddy’s Lab

Turning now to Genzyme, the technology underlying the dispute in this case is related to obtaining hematopoietic stem cells from blood.  These stem cells reside in the bone marrow where they develop into mature blood cells, including white blood cells (WBCs).  They are found in the blood only in very small numbers but can be made to move from the bone marrow into the blood through a process called mobilization.  In the bone marrow, they are anchored at least in part through an interaction between CXCR-4, a receptor located on the stem cell surface, and SDF-1, a protein (chemokine) produced in the bone marrow.

Genzyme Corporation (“Genzyme”) developed a method for mobilizing and harvesting hematopoietic stem cells by sequentially administering two drug products, G-CSF and plerixafor.  The method is covered by claim 19 of U.S. Patent No. 7,897,590 (“’590 patent”).  Plerixafor is the active chemical ingredient in Genzyme’s Mozobil®.  Genzyme brought a Hatch-Waxman action against Dr. Reddy’s Laboratories (DRL) which was seeking approval to market a generic version of Mozobil®.  In its defense, DRL asserted that claim 19 was invalid for obviousness.  The district court disagreed, and DRL appealed.  A major issue on appeal was whether an ordinarily skilled person would have had a reasonable expectation of success in combining the asserted prior art references to arrive at Genzyme’s method.

DRL’s obviousness claim was based on a combination of a journal article by Hendrix et al. and U.S. Patent 5,824,304 (’304 patent), or alternatively over the combination of WO 00/45814 (“’814 application”) and the ’304 patent.  Hendrix is focused on the use of plerixafor for treating HIV.  It reports an increase in WBC counts in subjects to whom plerixafor was administered.  To explain the increase, it referred to known information that plerixafor can “completely inhibit” binding of SDF-1 to CXCR-4, and hypothesized that “binding of [plerixafor] to CXCR4 may inhibit the chemotactic effects of SDF-1, causing release of WBCs from the endothelium and/or stem cells from bone marrow.”  Genzyme at 5 (emphasis added).

The ’304 patent is related to the interaction between another protein-receptor pair: the protein VLA-4, found in bone marrow and the receptor VLA-4 receptor, found on stem cells.  The patent discloses a method of increasing stem cell numbers in blood by administering an agent that blocks the binding of VLA-4 to its receptor, thereby releasing stem cells from the bone marrow.  The patent also discloses administering G-CSF to stimulate production of stem cells in the bone marrow.  The combination of G-CSF and the blocking agent, leads to mobilization of stem cells.

The ’814 application, like Hendix, reveals the relationship between plerixafor and WBC elevation, not information about using plerixafor to mobilize stem cells.

DRL argued that claim 19 differed from the ’304 patent only in the use of plerixafor as the blocking agent, but asserted that such use would have been obvious due to the express suggestion in Hendrix that plerixafor could function as a blocking agent for releasing stem cells from the bone marrow.  Id. at 6.  The district court disagreed and noted that there was evidence of a history of failure in the field of mobilization of stem cells.

The Federal Circuit also disagreed.  It pointed to expert testimony that a typical stem cell has around one hundred different types of receptors on its surface and that there was no evidence that VLA-4 localizes stem cells in the marrow similarly to the CXCR-4/SDF-1 bond.  Id. at 7.  Thus, there could have been no expectation that CXCR-4 and VLA-4 would behave similarly with regard to mobilization.  Id.  That plerixafor may cause stem cell mobilization, the Court emphasized, was hypothesized in the Hendrix article in an isolated sentence without explanation, the rest of the seven pages article being focused on increasing WBC counts.

Further, the Court observed that SDF-1 was mentioned for attracting lymphocytes, not stem cells, and CXCR-4 was described as being expressed in different types of WBCs, not stem cells.  Id. at 7-8.  An ordinarily skilled artisan would therefore have recognized that presence of stem cells was never tested by the Hendrix authors.  Id. at 8.  Other relevant facts were: CXCR-4 antagonists were studied only in the HIV field, there was a history of failure with respect to the search for better stem cell mobilization agents, and no one had ever mobilized stem cells with any CXCR-4 antagonist.  Id.

As to the combination of the ’814 application and the ’304 patent, the Court agreed with the district court that obviousness would have required the erroneous assumption that an ordinarily skilled person would have known that WBCs were a proxy for stem cells.  Id. at 9.

The Court found that under these facts, a reasonable expectation of success did not exist.  Although non-precedential, Genzyme is a reminder that unpredictability in a field can seriously undermine expectation of success.  In such fields, taking hypotheses or conclusory statements at their face value to demonstrate obviousness is unlikely to be successful.