Gilead Sciences, Inc., Hoffmann-La Roche, Inc., F. Hoffmann-La Roche, Ltd., and Genentech, Inc., v.Natco Pharma Ltd.and Natco Pharma, Inc., 753 F.2d 1208 (Fed. Cir. 2014, cert denied 3/9/15)
About a year ago, the Supreme Court denied review of a Federal Circuit decision on the application of doctrine of obviousness-type double patenting to reject later filed, earlier issued application, allowing the lower court’s decision to stand. At the time, the decision did not seem to raise a stir in the bar, and indeed one of the attorneys who argued the case before the Federal Circuit recently said in a seminar that he believed that the decision did not change the law. Despite this, in the past few months a number of commentators have suggested that the case sounds a warning alarm for pharma companies and that sweeping action including a thorough portfolio review is needed in view of the decision. For that reason, we are taking a closer look at Gilead, and its possible ramifications.
*Spoiler alert* – we don’t believe that drastic action is warranted.
The orientation of the patents involved is as follows:
Each application claimed the inhibition of viruses by selective inhibition of certain enzymes, and for purposes of its opinion, the Court assumed that the ‘375 (earlier filed/later issued) application rendered obvious the (later filed/earlier issued) ‘483 application. This is a classic “Reversal of Order” situation as considered in In re Braat, 937 F.2d 589, 19 U.S.P.Q. 2d 1289 (Fed. Cir. 1991). Braat involved an improvement on construction of data surfaces of compact discs. After making his first invention, and filing for a patent, the inventor made a second invention that was an improvement on his first invention, and filed a patent on this improvement. At the risk of oversimplification here is a graphical representation of the data surface aspect:
Prior Art Disk:
First Application:
Second Application:
The first application was an improvement over the prior art by modifying the height of the wall dividing data channels, giving the software powering the reading laser a way to distinguish between adjacent channels and reduce mistracking. The second application involved a way to reduce mistracking by sloping the edges of the walls, thus directing the laser back into the track if it strayed – and that application also claimed the combination – modified wall edges plus modified height. This second application issued first, and was then used to reject the first application under the doctrine of obviousness-type double patenting. Application 2 clearly does render application 1 obvious, since it embodies the invention of application 1.
However, the Braat Court created an “exception” to the doctrine, holding that, in the situation where there is this reversal of order, and that reversal is not the “fault” of the applicant, a “two-way” test must be performed – not only should we ask whether the reference application (in this case, application 2) renders obvious the rejected application (application 1), but we must also ask if the earlier filed application (application 1) renders obvious the later filed one (application 2).
This situation is of considerable importance in pharma cases, especially in the arena of small molecules, where applicants are sometimes faced with a situation where they have (1) a generic invention that is arguably, but not clearly, patentable over the prior art and (2) a subgeneric (e.g. species) invention that is more clearly patentable over the prior art and that is arguably patentable over the generic invention. Even if the applicants disclose and claim the two inventions in one application, it frequently happens that the subgeneric claims are allowed first and the generic claims are rejected over the prior art. While applicants could keep the allowed subgeneric claims in the application and contest the rejection of the generic claims, doing so would delay the issuance of the subgeneric claims, perhaps by years, if they wind up taking an appeal from a final rejection of the generic claims. On the other hand, if applicants file a continuation application containing the rejected generic claims, cancelling the generic claims from the first application, they will receive an obviousness-type double patenting rejection of the generic claims over the then-patented subgeneric claim, unless they file a terminal disclaimer – even if the subgeneric claims are patentably distinct from the generic claims, because as in Braat the subgenus typically renders obvious that portion of the genus which it occupies. The two-way test of Braat would seem to resolve this problem, at least where the subgenus is patentably distinct from the genus, since the analysis would have to be done in that direction as well.
The two-way test, however, is an “exception” to the rule – and the limited scope of this exception became clear in In re Berg, 140 F.3d 1428, 46 U.S.P.Q. 2d 1226 (Fed. Cir. 1998). In Berg, the applicant filed two patents on the same day, each containing the genus invention and a patentably distinct subgenus – apparently assuming that they would first obtain a patent on the subgenus, and would save time by already having an application in condition for allowance (by cancelling the genus) and avoid delaying prosecution of the genus by having an already pending continuing application directed thereto. Applicants were apparently relying on the application of the two-way test, which would have enabled overcoming (or even precluded) a double patenting rejection of the genus over the subgenus. However, in denying applicability of the two-way test, the Court limited the “exception” to the situation where the reversal in order was not within the control of the applicant, i.e., defining “fault” in Braat to include situations where something the applicant did – or here, didn’t do, filing a single unitary application – contributed to the reversal of order.
Presumably, such “fault” would also include the situation where applicants took an early allowance of the subgenus, and filed a continuation to prosecute the genus, instead of maintaining all claims together in a single application. Accordingly, the not-uncommon course of action where applicant takes the allowable subject matter and then files a continuation to fight for broader scope would most always have double patenting issues where the earlier-filed, broader application would receive a double patenting rejection over the allowed/patented subgenus.
Of course, the Berg or Braat rejections could have been resolved by the filing of a terminal disclaimer. In the realm of 20 year patents, that wouldn’t seem to matter in the situation where the applications stem from a common ancestor and so have the same expiration date, and applicants are not concerned with maintaining common assignment. However, there are at least two situations where it does, in fact, matter. The first situation is the one that occurred in Gilead. Why applicants filed a second application – a later discovered utility, perhaps – is not clear, but as priority to the earlier application was not claimed, the second filed application had a 20-year term which ended later than that of the first-filed application. Despite the Braat-style reversal of order – the later filed application issued first — the Court implies, though never says outright, that the order of filing of the applications might have been the result of gamesmanship:
“Despite their similarities in content, however, the ‘375 and ‘483 patents are not part of the same family of patents and were not before the same patent examiner. Instead, Gilead crafteda separate “chain” of applications, having a later priority date than the ‘375 patent family. That separate chain resulted in the issuance of the ‘483 patent. Because the patents do not claim priority to any common application, they will expire at different times as governed by the provisions of the Uruguay Rounds Agreement Act.” (Emphasis added.)
“Crafted,” crafty, shifty, underhanded… the implication is that Gilead somehow did something wrong. In any event, the court focuses on the relative expiration dates of the patents, and indicates that their relative issue dates have no bearing on the applicability of a double patenting rejection. The “bedrock principle” of patent law, according to the Court, is that there can be no additional term for an invention that is obvious over an earlier expiring one, regardless of filing or issue dates.
According to various commentators, this has caused consternation at least in the pharma industry, with patentees said to be searching their portfolios for patents that might, under Gilead, need terminal disclaimers. One might wonder, should such patents have escaped the Patent Office in the first place? In the Gilead situation, the answer is, likely, “no.” Indeed, an attorney who argued the Gilead appeal stated in a recent seminar his belief that Gilead changed the law – not at all.
The two-way test is not applicable. The two-way test of Braat is applicable only where (according to Berg) the reversal of order was not within the control of the applicant. Clearly, here, applicant’s choice to file a later application negates applicability of the two-way test.
The USPTO’s internal procedure for removing provisional double patenting rejections in the earlier filed application is not applicable. According to MPEP §804 (I)B(1):
If a “provisional” nonstatutory double patenting rejection is the only rejection remaining in an application having the earliest effective U.S. filing date (including any benefit claimed under 35 U.S.C. 120, 121, 365(c), or 386(c) ) compared to the reference application(s), the examiner should withdraw the rejection in the application having the earliest effective U.S. filing date and permit that application to issue as a patent, thereby converting the “provisional” nonstatutory double patenting rejection in the other application(s) into a nonstatutory double patenting rejection when the application with the earliest U.S. effective filing date issues as a patent.
The rejected application here is the later filed application, and so the USPTO should not have removed any double patenting rejection therein, had one been made. In fact, the earlier filed application (where the rejection could have been removed) contained a terminal disclaimer over the later filed application, though presumably under the procedure of MPEP §804 above this would not have been necessary.
In fact, what happened in the later filed application in Gilead was that the USPTO failed during prosecution to make a double patenting rejection in the later filed application that would have required a terminal disclaimer over the earlier filing. Gilead does not, therefore, create any new law, though at first blush its emphasis on expiration rather than issue dates might appear to do so – indeed, in the dissent of Judge Rader, is argued to do so.
Thus, if this is the concern of the commentators, then it is a concern that the PTO did not issue a valid patent – something that presumably an aware counsel would have warned the client to take action against during prosecution (e.g., either address through argument or terminal disclaimer).
Yet, there is a second situation where Gilead justifiably causes concern, that situation where (unlike the facts in Gilead) the earlier filed application is entitled to Patent Term Adjustment (PTA). What would have happened in Gilead if the earlier filed application was entitled to sufficient PTA so that it expired later than the second filed application? Such is a common situation where the first filed, broader application often takes longer to reach first action than a continuation (where the examiner – likely the same examiner as in the parent – has already done searching). If that first application is limited to patentable species, it is likely those species are patentable because data is available, and likely that data is available because it was developed for what are believed to be important compounds. Thus, the application having PTA could be directed to compounds which become commercial.
If a double patenting rejection was the last remaining issue in the first filed case, under MPEP §804 as above, that rejection would have been withdrawn, and the application allowed to issue. Under Gilead, however, there is at least the possibility that this later expiration date should have required a terminal disclaimer. Gilead notes the above portion of the MPEP, characterizing it as requiring a terminal disclaimer “for the later of the two applications (which the hypothetical anticipates to have the later expiration date).” Accordingly, where there is PTA in the earlier of the applications that extends its expiry beyond the expiration of the later filed application, the Court’s reasoning suggests that perhaps a disclaimer is still required in the earlier – but later expiring – application.
Gilead argued that it is issue dates that control whether a double patenting rejection is appropriate. Had the court held so, indeed, that would have been new law. Instead, the Court focused on expiration dates, perhaps phrasing a little differently what was, in reality, the law all along.
On the one hand, while it is not unreasonable (at least, where a patent is being litigated) for patentees to be reviewing their portfolio for potential double patenting situations where a later filed , earlier issued patent with a longer term should have disclaimed an earlier filed, later issued patent – the Gilead situation – looking for the situation where the USPTO failed to do its job, one would hope that such situations would be rare. However, in the second situation, where the earlier filed application has PTA which makes it the later expiring application, the PTO has not fallen down on the job since PTA is awarded after prosecution is closed, and the examiner had no opportunity to issue a double patenting rejection. It is here that patentees’ concern should be focused, at least for patents which are important commercially. To the extent that Gilead created new law (or more accurately, used language which highlighted a situation that is as yet not litigated and thus uncertain), it is here that the PTO even operating flawlessly will not alert applicants to what might now be understood as a problem, and here that patentees may wish to consider – in patents that are commercial and/or likely to be litigated – whether action such as developing evidence of non-obviousness, or the filing of a terminal disclaimer, is needed to ensure withstanding a challenge