By Kevin E. Noonan —

On Monday, the U.S. District Court for the District of New Jersey (Judge Peter G. Sheridan, presiding) found after a bench trial on all issues that Bayer’s U.S. Patent 6,787,531 was invalid for obviousness.  In doing so, the Court’s decision illustrated anew the uncertainties that the U.S. Supreme Court’s KSR Int’l Co. v. Teleflex, Inc. decision has (re)introduced into patent law, and the folly of the Supreme Court’s simple-minded "common sense" approach to the issue.

Bayer’s patent claims a pharmaceutical composition of an oral contraceptive marketed as Yasmin® in the U.S., comprising a combination of drospirenone and ethinylestradiol.  Claim 1 of the ‘531 patent is representative of the claimed formulation:

1.  A pharmaceutical composition comprising from about 2 mg to about 4 mg of micronized drospirenone particles, about 0.01 mg to about 0.05 mg of 17-alpha-ethinylestradiol, and one or more pharmaceutically acceptable carriers, the composition being in an oral dose form exposed to the gastric environment upon dissolution, and the composition being effective for oral contraception in a human female.

This combination as an oral contraceptive was not new; the formulation was.  Bayer found that, despite the fact that drospirenone was known to be acid labile and to isomerize into an inactive form, the combination could be advantageously formulated using micronized drospirenone (which increased absorbsion of the drug but was expected to also increase the rate of isomerization) and by not coating the drospirenone with an enteric coating.  Since this was an oral formulation, an enteric coating would be expected in the art to protect the drug from isomerization in the acidic environment of the stomach.

The action was initiated pursuant to the provisions of the Hatch-Waxman Act in response to Barr having filed an Abbreviated New Drug Application (ANDA) with the Food and Drug Administration.  Barr’s position was that the use of micronization to improve drug absorbsion and enteric coatings to protect acid-labile drugs was "routinely considered" by drug formulators in making drugs for oral administration, and that Bayer’s claims were thus obvious.

The prior art considered by the Court included a common formulation text not particularly directed to oral contraceptives (Aulton, 1988, Pharmaceutics: The Science of Dosage Form Design) and a number of scientific journal articles relating to acid-lability of not drospirenone but a related compound, spirenone, both in vitro and in vivo.  Also considered were prior art patents for various combinations of oral contraceptives formulated for different administration schedules.

Bayer also introduced testimony relating to how the invention was made, specifically relating to the surprising findings that formulating the drug in ways that should have increased isomerization (micronizing the acid-labile drospirenone and dispensing with the enteric coating) in fact had the opposite effect.  According to this testimony, the reason for this paradoxical result is that the micronized drug was absorbed more rapidly than it isomerized in vivo, and enteric coatings made the effectiveness of the drug more variable; it was uncontested that the drug had to be 99% effective to be useful as a contraceptive.  This testimony and other evidence established that Bayer scientists had originally prepared formulations that were unmicronized and enterically-coated, and that they were successful only when they changed direction and prepared the formulation that was the subject of the ‘531 patent.

Judge Sheridan disregarded this evidence, on the grounds that it was not reasonable to him that Bayer’s scientists would have changed direction as abruptly as they purportedly did in their formulation work.  Moreover, the Court believed that the Aulton treatise and other scientific journal references, as well as Barr’s expert, taught that acid lability studies in vitro would never be sufficient to establish whether a drug was truly acid labile without complementary in vivo studies, and that had Bayer performed the in vivo studies they would have known that the absorbsion rate was sufficiently fast that enteric coatings were unnecessary and micronization was desirable.  In addition, the Court cited three studies (including both in vitro and in vivo arms) with the related drug spirenone that purported to show the absence of spirenone isomers in blood samples from experimental animals or human volunteers administered tableted spirenone.  In vitro studies on drospirenone, on the other hand, showed that 80% of the drug had isomerized in a 0.1N HCl solution at room temperature.  The Court found that spirenone and drospirenone were related "as close[ly] as fraternal twins" and thus imputed the behavior of spirenone to drospirenone.

The Court conducted its obviousness analysis based on these facts expressly using the rubrics of Graham v. John Deere, citing KSR as the impetus for this analysis.  In comparing the differences between the prior art and Bayer’s claimed invention, the Court found this critical factor that informed its analysis:

There is one other principle that is clearly established in the prior art which defeats both claims.  The prior art states the in vitro studies are unreliable unless they are correlated to in vivo testing (Aulton, McGilveray).  Unlike Nickisch, a person of ordinary skill in the art would conduct such testing.  Decisions about drug dose form are then made based upon the in vitro/in vivo testing.  That is, the decision whether to micronize and/or enteric coat comes second.  In this case, Bayer argues that the person of ordinary skill in the art would reverse this process ("put the cart before the horse").  This does not make sense.  Justice Kennedy in KSR observed that in analyzing an obviousness defense, the court must use its common sense.  KSR, 127 S. Ct. at 1732.  In this instance, Bayer’s alleged exception to the in vitro/in vivo correlation rule (that it does not apply to acid sensitive drugs) does not ring true.  A person of ordinary skill in the art must precisely know and verify the characteristics and chemical reactions of a drug in order to evaluate its therapeutic value in humans.  To follow McGinity’s alleged exception for acid-sensitive drugs can only lead to sketchy, imprecise formulation results and an increased risk of injury to users.

Judge Sheridan’s Opinion (emphasis added).  Using "its common sense," the Court found Bayer’s formulation claims obvious.

It is an everyday occurrence to note that "common sense" is anything but common, even among the most mundane of human activities.  Most would agree that pharmaceutical formulation is hardly a mundane activity, and is fraught with uncertainties exacerbated by the unpredictability of patient response to a particular formulation; indeed, a great deal of the regulatory apparatus for approving new drugs is based on establishing that the vast majority of patients will be helped and, more importantly, not be harmed by a formulation.  The evidence presented to Judge Sheridan established just this sort of unpredictable result, where an acid-labile drug is provided in a better formulation when it is treated in ways expected to increase isomerization and drug inactivation in the acidic environment of the stomach.

And yet Judge Sheridan, harkening to the admonitions of the Supreme Court that he should employ "common sense," disregards the extensive evidence of these uncertainties and decides that, to him, Bayer’s activities "put the cart before the horse."  Never mind the Supreme Court’s equal admonitions against the insidious effects of hindsight, or how "obvious" the result may be once it is found that the drug, surprisingly, is absorbed faster than it can isomerize.  No, Judge Sheridan is comfortable substituting his "common sense" interpretation of what is the cart and what is the horse, and the order they should go in, to decide that Bayer’s formulation would be obvious.

This is madness.  It cannot be the Supreme Court’s intention to unleash the generally uninformed "common sense" of the generalist judiciary to trump testimony by the individuals — the inventors — who actually perform the experiments and produce the inventions whose disclosure fulfills the Constitutional mandate to "promote the progress of the useful arts."  But that is precisely the effect that the KSR decision, and most of the rest of the Supreme Court’s obviousness jurisprudence will have (and is having) on U.S. patent law.  It is not enough that the Court mistakes the proper emphasis of the correct Constitutional analysis by putting the cart of restricting monopolies before the horse of promoting disclosure of new inventions.  In KSR and the rest of the Supreme Court’s "totality of the circumstances," "we know an obvious invention when we see one" jurisprudence, it substitutes judicial common sense for scientific common sense.  Folly, pure and simple.

Barr made two other allegations against the claims of the ‘531 patent.  First, that the clinical trials performed in the U.S. were a public use, because earlier clinical trials in Europe had established that the formulation was safe and effective and because the study participants were not bound by confidentiality to withhold from their personal physicians the identity of the drugs they were administered.  The Court found that Bayer’s proffered reason for performing clinical testing in the U.S. — the greater ethnic diversity of the U.S. versus the European population — constituted experimental use, not public use, and that Bayer had established the appropriate confidentiality conditions for the physicians and other healthcare and scientific workers involved with the clinical trials.  (It was also persuasive that there were ethical prohibitions against keeping the identity of the administered drugs from trial participants’ physicians.)  Second, the Court found that while the evidence of the European clinical trials was material to patentability on the public use question, Bayer declarants showed no intent to deceive the U.S. Patent and Trademark Office by only partially and obliquely referring to these trials (and the data obtained from them) during prosecution of the ‘531 patent.

Additional information regarding this case can be found at the Orange Book Blog.