Unexpected Results Need to Match the Breadth of Claims

June 10, 2014

What This Means to You

  • Unexpected results should be commensurate in scope with claims.
  • Claims merely excluding prior art compounds are at risk of being found obvious.
  • An application should be drafted with multiple fall-back provisions and different scopes of subgenera.


In Allergan, Inc. v. Apotex Inc., the U.S. Court of Appeals for the Federal Circuit (Federal Circuit) reversed the district court’s findings, holding that the asserted claims were invalid for obviousness and vacating the district court’s grant of an injunction preventing generic versions of Latisse®.

Case Background

Allergan and Duke University own U.S. Patent Nos. 7,388,029 (“the ’029 patent”) and 7,351,404 (“the ’404 patent”), respectively, directed to a bimatoprost solution for treating eyelash hair loss. Allergan markets Latisse®, a bimatoprost solution, as a topical treatment to promote eyelash hair growth. When Apotex Inc. and a number of generic companies filed an Abbreviated New Drug Application (ANDA) with the Food and Drug Administration (FDA) to market a generic version of Latisse®, Allergan and Duke sued the generic companies for infringement.

Apotex argued that the ’029 patent and the ’404 patent are each anticipated by International Patent Application No. PCT/US98/02289 (“Johnstone”) and U.S. Patent No. 5,688,819 (“the ’819 patent”). The prosecution history indicated that to overcome a rejection based on novelty, the applicants had amended the claims to exclude Johnstone’s compounds. The district court found Johnstone’s specification too sparse and ambiguous to anticipate these ’029 patent claims.

Apotex further argued that the ’404 patent, directed to methods for stimulating hair growth by topical application of a bimatoprost solution, was inherently anticipated by Allergan’s ’819 patent, directed to methods of treating glaucoma using the same formulation. Evaluating the specification of the ’404 patent, expert testimony, and clinical trial statistics, the district court agreed with Allergan that the glaucoma treatment eyedrops can be properly applied to avoid transfer to skin around the eyes. Thus eyedrop application does not necessarily lead to topical application to promote eyelash growth.

Finally, Apotex argued that the ’029 patent is obvious over the combined teachings of the ’819 patent and Johnstone. The district court found no motivation to combine the references due to the unpredictable nature of hair loss art, pharmacological differences, and the minimized side effects of bimatoprost. As a result, the court concluded that the patents at issue are valid and issued an injunction against Apotex. They appealed the case to the Federal Circuit.

Decision Analysis

The Federal Circuit did not find any clear error in the district court’s ruling on novelty. It did point out that the district court erred in the obviousness analysis for failing to consider the full scope of the claims in the ’029 patent as well as the two cited references. According to the Federal Circuit, the proper analysis should focus on whether any compounds within the broad genus claimed by the ’029 patent were obvious at the time of the invention, not whether specific compounds like bimatoprost were rendered obvious.

The Federal Circuit further reasoned that given the breadth of the ’029 patent, Apotex only had the burden of showing that any compound that fell within the broad genus claimed by the patent was obvious at the time of the invention. The court found that Johnstone suggests the possibility of using a 17-phenyl PGF analog in hair growth and its mere disclosure of alternative preferences does not constitute teaching away from the broad swath of the compounds within the ’029 patent. The court also elaborated that the hair loss art may be unpredictable, but nonetheless, after Johnstone, the issue is whether using a compound from glaucoma treatment to grow hair is reasonably unpredictable or not. The court concluded the ’029 patent is obvious because Johnstone teaches this new utility for a finite number of already identified and isolated compounds characterized from other references. The court further cautioned that unexpected results and secondary considerations must be commensurate with the scope of the claims. Since the claims of the ’029 patent encompass two broadly defined genera, the improved pharmacological profile of bimatoprost is not sufficient to overcome the obviousness of the whole claim.


This case highlights the court’s heightened scrutiny on unexpected results in the pharmaceutical industry. Taken together with Bristol-Myers Squibb Co. v. Teva Pharms. USA, Inc., which was decided by the Federal Circuit in June 2014, courts evaluate unexpected results for not only differences in kind versus differences in degree, but also for the breadth of the claimed invention.

Since new medical indication patents or selection patents tend to rely on advantages of new uses, species, or formulations of known classes of compounds or compositions, factors to rebut obviousness should be taken into consideration during the early patent drafting stage. The claims and genera need to be crafted in different layers to match the scope of any unexpected results available. Although the prior art compounds may be excluded in a conditional clause, depending on the size and structural similarities of the compounds, reasonable claim breadth combined with unexpected results having differences in kind will likely prevail in a nonobviousness argument.