federal circuitCross-posted at WLF’s Forbes.com contributor page

Last week, pharmaceutical products and services company Novo Nordisk petitioned the U.S. Court of Appeals for the Federal Circuit to rehear en banc a June 18 Federal Circuit panel ruling that invalidated a Novo patent (Novo Nordisk A/S v. Caraco Pharm. Labs; 2-1 with Judge Newman in dissent). Because the panel misapplied Patent Act § 103‘s standards for obviousness to Novo’s “combination drug,” the Federal Circuit should agree to rehear the case en banc and reverse the three-judge panel.

Background. In 1994, a Novo scientist conceived of an unorthodox way to improve a drug-based therapy for Type II diabetes. The scientist combined metformin, a compound that targets the liver’s sensitivity to insulin, with an insulin stimulator compound called repaglinide. The goal was to extend the amount of time that metformin worked in a patient’s body. The use of repaglinide was controversial for two reasons: 1) combinations of metformin and other insulin stimulators from a “genus” different from that of repaglinide (such as sulfonylurea) offered mixed results; and 2) research showed that repaglinide was eliminated from the body within one hour of use.

Hence, the synergy of the 1994 metformin-repaglinide combination was astonishing. The combination was eight times more effective than administering metformin alone. In other words, it led to an 800% increase in reducing diabetes patients’ fasting plasma glucose (FPG) over an eight-hour period. To this day, there is no scientific explanation for this synergistic effect.

Infringement Litigation. Novo Nordisk obtained patent protection on the combination. Caraco filed an Abbreviated New Drug Application for repaglinide and Novo sued for infringement. Caraco stipulated to infringement, but argued that because the combination was obvious in light of the prior art, the court should invalidate Novo’s patent. The Eastern District of Michigan held the patent invalid because the combination was “obvious to try.” The Federal Circuit affirmed, holding that the FPG reduction was “expected and explainable in light of the state of the art as of the critical date.” The appellate panel granted no deference to the patent office examiner’s finding that the eight-fold effectiveness increase was unexpected.

An Obvious Need for Rehearing. As Novo’s rehearing petition explains, the majority embraced the district court’s three-step rationale for why the metformin-repaglinide combination was obvious: 1) the prior art evident at the relevant time was the combination of metformin and a different insulin stimulator—sulfonylurea; 2) that combination provided some beneficial results for Type II diabetes; and 3) repaglinide was also an insulin stimulator which purportedly had an effect similar to sulfonylurea.

The court’s reasoning is both factually and legally flawed. Factually, both the trial and appellate courts failed to distinguish the scientific properties of the relevant compounds at issue and ignored a key piece of prior art. While sulfonylurea and repaglinide may both be “insulin stimulators,” their chemical properties are not similar. Thus, the metformin-sulfonylurea combination could not have been predictive of the metformin-repaglinide combination.

Legally, the majority rubber-stamped the trial court’s post hoc explanation for why the inventor thought the metformin-repaglinide combination was obvious to try. In dissent, Judge Newman forcefully argued that “obviousness may not be established using hindsight or in view of the teachings or suggestions of the inventor.” What matters is whether the outcome of combining the two drugs was predictable to the person of ordinary skill. That is an objective test, not a subjective test. That objectivity prevents courts from, as Judge Newman wrote, “us[ing] the inventors’ exceptional intellect against them.”

Also, when determining the evidence of unexpected results in the “obviousness” inquiry for combination products, legal precedent dictates that courts should consider how, at the relevant time, each distinct part of the combination would function. A study available before patent application showed that repaglinide alone disappeared from the body less than an hour after ingestion. The district and Federal Circuit opinions disregarded that key piece of prior art, which contradicted anyone’s thinking that the metformin-repaglinide combination was “obvious.”

diabetesIt is critically important for the Federal Circuit to get this case right. As Judge Newman stated in her dissent, “[metformin-repaglinide] is now apparently a treatment of choice for patients whose Type II diabetes had previously been untreated.” Novo Nordisk defied detractors and invested millions in testing and developing this combination. An invalidated patent would mean no return on that risk-taking.

Failure to correct the Novo Nordisk panel’s errors would have powerful ripple effects beyond this case. Combination drug treatment holds a great deal of promise, with research currently underway to treat devastating ailments like cancer and osteoporosis. But bringing those treatments to patients is staggeringly expensive and fraught with risk already. An excellent amicus brief the Pharmaceutical Research and Manufacturers of America filed supporting Novo’s en banc rehearing petition illuminates these costs and risks.

Adding legal uncertainty to the risk-reward calculus for combination drugs might be the last straw for many companies, to the lasting detriment of American health care.