When compounds in a class have divergent properties, positional isomer not obviousMay 18, 2009

In a decision Wednesday, the Federal Circuit affirmed a district court's determination that the asserted claims in a pharmaceutical patent were not proven obvious. As is common in pharmaceutical cases, the defendant filed an ANDA asserting the patent covering the compound and its use was invalid, and in the ensuing infringement suit admitted its proposed generic drug met the claim limitations. The district court specifically held the prior art did not suggest the closes prior art compound as a potential "lead compound," as the various members of the class of compounds had unpredictable properties rendering the claimed compound's beneficial properties unexpected, and secondary considerations of commercial success and long-felt need confirmed the defendant's failure to prove the claims obvious. The district court also rejected a separate argument of obviousness-type double patenting based on an expired patent also owned by the patentee.

The Federal Circuit affirmed. Under Takeda, the court observed obviousness for structurally similar chemical compounds depends on one of ordinary skill in the art selecting the asserted prior art compound as the "lead compound," or the starting point for further experimentation. Here, the court did not reach the question, as it determined that even if the closest prior art compound was selected, the defendant has not shown obviousness. There was simply no reason shown why one of ordinary skill in the art would modify even the closest compound to produce the claimed compound (a positional isomer) given the inherent unpredictability of the physical-chemical, biological, and therapeutic characteristics of the individual compounds in the class. The court also cited its recent Kubin decision, observing there was "no credible evidence that the structural modification was routine." Further, there was evidence of both commercial success ($2.7 billion in sales) and long-felt need. As a result, the court affirmed the district court's determination that the claims had not been proven obvious.

Similarly, the court affirmed the finding of no obviousness-type double patenting. Here, the prior art patent claimed intermittent dosing of compounds in the same class as the claimed compound (to compensate for side effects), whereas the asserted claims are directed to the compound itself (both isolated and in a pharmaceutical composition) and using the compound to treat osteoporosis. There was no clear and convincing evidence of overlap between the patents, so there was no double patenting.

Procter & Gamble (P&G) sells the osteoporosis drug Actonel®, the active ingredient of which is the compound risedronate. The structure of risedronate is shown below:


Risedronate is part of a class of compounds known as bisphosphonates, which generally have activity inhibiting bone resorption, making them useful to treat osteoporosis. Several other osteoporosis medications are of this class, including Fosamax® (alendronate) and Boniva® (ibandronate). Even with this similarity, the compounds in this class have many differences, and their physical-chemical, biological, and therapeutic characteristics are generally unpredictable.

P&G holds a patent covering risedronate, as well as pharmaceutical compositions containing risedronate and methods of use to treat osteoporosis. Teva Pharmaceuticals sought to market a generic version, and filed an ANDA asserting P&G's patent was invalid. P&G sued, and Teva counterclaimed that the asserted claims were obvious. Specifically, Teva asserted that a structurally similar compound disclosed in a prior P&G patent, 2-pyr EHDP, would be identified as a lead compound and rendered risedronate obvious. 2-pyr EHDP is a positional isomer of risedronate, with the only difference being the position of the hydroxy-ethane-diphosphonate group (located on the #2 carbon in 2-pyr EHDP and the #3 carbon in risedronate).

After a bench trial, the district court concluded one of ordinary skill in the art would not have used 2-pyr EHDP as the lead compound, but that even if it was, one of ordinary skill would have had no reason to make risedronate based on the prior art as there would have been no expectation of success. Further, the court held there was sufficient evidence of unexpected results and secondary inidicia of nonobviousness that even had Teva made out a prima facie case of obviousness, it would have been rebutted by the secondary indicia. The district court also rejected Teva's double patenting argument. Teva appealed.

The Federal Circuit affirmed. The court declined to consider whether one of ordinary skill in the art would have selected 2-pyr EHDP as a lead compound, because even if that was the case, the court held the claims not to be proven obvious. The court observed its statement in Takeda that sometimes "[a] known compound may suggest its homolog, analog, or isomer because such compounds often have similar properties and therefore chemists of ordinary skill would ordinarily contemplate making them to try to obtain compounds with improved properties." Even in such cases, however, "a showing that the prior art would have suggested making the specific molecular modifications necessary to achieve the claimed invention" is necessary to show obviousness.

Here, the various bisphosphonates did not generally share similar properties: instead, the compounds in the class were quite divergent in their various properties, and were "generally unpredictable." Further, P&G synthesized another close positional isomer, 4-pyr EHDP, but it was not active in inhibiting bone resorption even though it was a closely related compound. Thus, one of ordinary skill in the art would not have expected success with risedronate. Instead, the only motivation was to "vary all parameters or try each of numerous possible choices" (as described by the court in Kubin). As such, there was no prima facie case of obviousness made.

Further, the court held there was sufficient evidence of unexpected results to rebut any prima facie case that did exist. While this seems like a bit of bootstrapping, given that the unexpected nature of the class of compounds was one of the bases for the determination no prima facie case had been made, this further confirmed the court's conclusion. Similarly, secondary indicia of nonobviousness, namely evidence of over $2.7 billion in sales and solving a long-felt need for osteoporosis treatment sealed the court decision. Interestingly, the court observed that the relevant time for determination of a long-felt need was, as with the rest of the considerations of obviousness, the patent's priority date. This mattered here because while Actonel was not the first osteoporosis drug on the market, the need was still unmet as of the priority date. This leads to the possible conclusion that multiple patents with different treatments may benefit from a finding of long-felt need, particularly in the pharmaceutical industry where FDA approval often means the drug does not reach market until much after not only the priority date, but also the issue of the patent.

Finally, the court easily rejected Teva's double patenting argument. The court held there was no overlap between the prior art patent (which claimed intermittent dosing of other bisphosphonates to mitigate side effects) with the asserted claims, which covered risedronate and its use to treat osteoporosis.

To read the full decision in Procter & Gamble Co. v. Teva Pharms. USA, Inc., click here.

← Return to Filewrapper

Stay in Touch

Receive the latest news and updates from us and our attorneys.

Sign Up