By Andrew Williams --
Earlier this week, we noted that parties wishing to challenge IPR Final Written Decisions from the PTAB on appeal to the Federal Circuit would face an uphill challenge. However, even if that challenge is extremely difficult, it is not necessarily futile. This was demonstrated by the Federal Circuit on June 16, 2015 when it remanded part of a PTAB decision in the Microsoft Corporation v. Proxyconn, Inc. case. The Court held that even though it does not constitute error to use the "Broadest Reasonable Interpretation" (or BRI) standard for claim construction, it is error to adopt a construction that is "unreasonable," which the Board did for two of the construed claim terms. Otherwise, the Court did not find error with the Board's validity determination with respect to the claims not impacted by the improperly construed claims. The Federal Circuit also addressed, for the first time, the standards implemented by the Board for amending claims during IPR proceedings. Specifically, the Court found that the Board's Idle Free decision included a permissible interpretation of the PTO regulations at least with respect to the burden placed on the patentee to demonstrate patentability of any claim amended in view of the prior art of record. The Court did, however, stop short of indicating that it is proper to require the patentee to address patentability over the entirety of the prior art.
At the risk of oversimplifying the technology at issue, the patent in this case, U.S. Patent No. 6,757,717, related to decreasing traffic over the internet. Specifically, it sought to address the problem of sending files between computers multiple times when either the original computer, or an intermediary computer, already contained a copy of the file. Basically, the computer wishing to access a particular file (the receiver/computer) sends a request to a sender/computer that has a copy of the file. Instead of immediately sending the file, the sender/computer calculates a "digital digest," or code (our term), that can be used to uniquely identify the file. This code is used by the receiver/computer to determine if it already has a copy of the file on its memory. Only if the answer is negative does the sender/computer transmit the whole file over the internet. The patent also envisioned an intermediary "gateway" computer which would use the code to determine whether a "caching" computer had a copy of the file that could be sent to the receiver/computer. This is demonstrated in Figure 11 from the '717 patent, which was reproduced in the opinion:
Microsoft filed two IPR petitions related to the '717 patent, each challenging different claims. The Board joined the two proceedings, and determined that claims 1, 3, 6, 7, 9-12, 14, 22, and 23 were unpatentable as anticipated, and claims 1, 3, and 10 were additionally unpatentable as obvious. However, claim 24 survived. Microsoft appealed the determination that claim 24 was found to be patentable, and Proxyconn cross-appealed almost everything else. Chief Judge Prost wrote the decision for a unanimous panel that included Judge Lourie and District Judge Gilstrap from the Eastern District of Texas sitting by designation.
Was the Use of the BRI Standard Revesable Error?
Proxyconn first challenged the Board's use of the BRI standard. While the appeal was pending, the In re Cuozzo case was decided. The Federal Circuit in this case said it was bound by the Cuozzo decision to approve of the use of the BRI standard.
However, the Court pointed out that even though the Board may construe the claims broadly, they may not do so unreasonably. Therefore, for example, if the PTAB adopts a construction that is contrary to arguments made or positions taken during the original prosecution, it could be reversible error. Moreover, "[a] construction that is 'unreasonably broad' and which does not 'reasonably reflect the plain language and disclosure' will not pass muster." In other words, the construction cannot be divorced from the specification or the record.
Was the Construction of the Claim Terms Really "Reasonable"?
Proxyconn next challenged the claim constructions as unreasonable. The first related to the phrase "gateway . . . connected to said packet-switched network in such a way that network packets send between at least two other computers." The Board adopted the construction proffered by Microsoft that the "two other computers" were not limited to just the sender/computer and the receiver/computer, but could encompass the gateway and caching computers. As a result, it found claims 6, 7, and 9 anticipated by the "DRP" reference.
The Federal Circuit did not find this construction to be reasonable. A plain reading of the impacted claims indicated that the "two other computers" were meant to be alternatives to the gateway and caching computers. Indeed, the term "other" denoted distinction between the "two other computers" and the gateway and caching computers separately identified in the claims. Therefore, a construction that included them would be nonsensical. Furthermore, the specification consistently used this term to distinguish the two computers from the gateway and caching computers. Therefore, the Board's construction was determined to be unreasonably broad.
Similarly, Proxyconn challenged the construction of the terms "sender/computer" and "receiver/computer." Specifically, the Board construed these claim terms in such a way as to encompass the intermediary gateway and caching computers. But, with similar logic as used above, the Federal Circuit found that such constructions did not reasonably reflect the language and disclosure of the '717 patent.
Because the Board's determination of patentability was necessarily flawed with regard to the claims that contained any of these claim terms, the Federal Circuit vacated the findings of unpatentability of these claims and remanded for proceedings consistent with the opinion. Because Microsoft filed a petition for panel rehearing at the end of July, the case has not yet made its way back to the PTAB. It will be interesting to see how they handle the remand, and how fast they handle it, once it returns. Without the "one-year-to-final-decision" time constraints, the PTAB may not feel compelled to deal with the remand in an expedited fashion. We will, of course, monitor the case and report any interesting developments.
Did the Board Properly Deny Proxyconn's Motion to Amend
As another interesting aspect of this case, the Federal Circuit for the first time addressed one the largest criticisms levied against the PTAB for these new post-grant proceeding challenges to issued patents -- the ability (or lack thereof) to amend the claims. Proxyconn had moved to replace claims 1 and 3 with claims 35 and 36, respectively. Microsoft alleged that the substitute claims were not patentable because they were anticipated by the prior art reference DRP, which was of record, but was not a reference cited when trial was instituted for these two claims. The Board denied the motion for two reasons. First, it found that Proxyconn had not met its burden of establishing a prima facie case of patentability. Essentially, among other defects, it had failed to provide constructions of the newly added claim terms, failed to address the manner in which the claims were patentable over the art, and failed to identify the closest prior art. Second, it rejected Proxyconn's argument that the Board was precluded from relying on the DRP reference.
At center stage in this appeal was the "informative" decision Idle Free, which was decided by a six-member panel. As such, even though it is not necessarily binding on other panels, it has been cited on numerous occasions in the denial of motions to amend. Basically, the Idle Free decision clarifies that the burden is on the patent owner to show "a patentable distinction over the prior art of record and also prior art known to the patent owner." Interestingly, because the DRP reference was of record in the case, the Federal Circuit made clear that it was not deciding whether every requirement in Idle Free constitutes permissible interpretations of the PTO's regulations. As indicated in footnote 4, "[t]he Idle Free decision is not itself before us . . . ." Therefore, it remains a question whether a patentee is properly required to provide a distinction over the all of the prior art (at least of which it is aware).
But, with regard to the prior art of record, the Court found the Idle Free requirement to be reasonable. First, it found that establishing "rules" within adjudicated cases was permissible. While it might be advantageous to establish such rules with traditional notice and comment rule-making, the decision ultimately rests with the agency's discretion. As such, the use of adjudication over rulemaking was not an abuse of discretion. Second, the Court found that the Board acted reasonably when it interpreted its rules to require a patentee to show patentability over the art of record, because nothing in the statute or regulations prohibits it. Moreover, it is consistent with the very nature of these IPR proceedings. If the motion is granted, the claims will be added to an issued patent without further examination. Therefore, to limit any challenge to only the art cited at institution could prevent the Board from denying the motion despite being in possession of art that undermines patentability. Finally, the Court rejected Proxyconn's argument that such an interpretation was "fundamentally unfair." Proxyconn was clearly aware of the DRP reference because it was a part of the entire proceedings. It also had an opportunity to respond to Microsoft's opposition which cited the DRP reference, although it chose not to. As such, the Court determined that this was not a case in which Proxyconn was taken by surprise and was not given the opportunity to respond.
Microsoft Corporation v. Proxyconn, Inc. (Fed. Cir. 2015)
Panel: Chief Judge Prost, Circuit Judge Lourie, and District Judge Gilstrap (sitting by designation)
Opinion by Chief Judge Prost
So if I've got a biotech patent in which hundreds of references were cited, I've got an effectively insurmountable burden if I want to amend my claims during IPR (or, presumably, PGR), unless I have an infinite budget and an army of low-cost scientists to review every reference. Ouch.
Based on this decision, it seems that if I'm monitoring a competitor's application and I see that not much is being cited, I should try to get as much art on the record as possible, even if only of marginal relevance, because when I go to challenge that application after it's issued, in an IPR or PGR, using good references instead of marginal ones, the patentee will need to relate to that marginal art if the patentee tries to amend its claims during post-grant proceedings.
If I've overlooked something, I'm sure this blog's readership will let me know.
Posted by: Dan Feigelson | August 06, 2015 at 04:17 AM