In inter partes review (IPR) proceedings, petitioners bear the burden of demonstrating that documents relied on as printed publications were publicly accessible before the effective filing date of a challenged patent. In Pfizer, Inc. v. Biogen, Inc., IPR2017-01166, Paper 9 at 10 (PTAB Nov. 13, 2017), the PTAB found that Pfizer failed to establish the public accessibility of a Rituxan® drug label relied on in a prior art challenge.

In its petition, Pfizer submitted prior art challenges based on the Rituxan® drug label and asserted that the label constituted a printed publication. The Board denied institution of the IPR. In the decision denying institution, the Board found that Pfizer’s evidence did not establish that “the Rituxan label was disseminated or otherwise made available to ordinarily skilled artisans exercising reasonable diligence prior to the critical date for the `172 patent.”

Current availability of a drug label on the FDA website does not demonstrate prior public accessibility

Pfizer argued that the label was publicly accessible because it was “a true and accurate copy of the original 1997 drug label” available from the FDA. Relying on the testimony of a retired academic librarian, Pfizer asserted that the Rituxan® label is “available today” from the FDA’s website, and that such availability “represents that it is the original approved label for RituxanTM as of November 26, 1997.” The Board, however, determined that the current availability of the label from the FDA’s website did not demonstrate that the label was publicly available as of the critical date:

[T]he fact that said label “is available today from the FDA’s website, which represents that it is the original approved label for Rituxan™ as of November 26, 1997” . . . is not pertinent to the question of whether the Rituxan Label was publicly accessible before August 11, 1998. In this regard, we note that the record is devoid of evidence concerning the availability of the Rituxan Label from the FDA website (or elsewhere) prior to the critical date of the ’172 patent.

[T]he FDA website’s present identification of the Rituxan Label as the originally approved label, without more, is insufficient to establish that such label was publicly accessible prior to August 11, 1998. For example, Petitioner neither asserts, nor offers evidence to suggest that the version of the Rituxan Label on the FDA website was in fact included as a package insert with Rituxan, or indicating when any Rituxan package insert was made available to interested artisans . . . . Absent additional context, we do not find the fact that the FDA website currently identifies the partially handwritten Rituxan Label as the first approved version of that label persuasive evidence, for purposes of this Decision, that the Rituxan Label was disseminated or otherwise available to reasonably diligent interested artisans before the critical date of the ’172 patent.

Evidence of public accessibility must be with respect to the “same document”

Pfizer also contended that “the well-known ‘Internet Archive’ service shows that the Rituxan® label was available on the website of Genentech, which markets Rituxan®, as of January 23, 1998.” The Board, however, noted that the label available on Genentech’s website was not the same document as the label that Pfizer asserted as a printed publication:

As to the printout of the January 23, 1998 version of the Rituxan webpage on the Genentech website obtained from the Internet Archive . . . , even accepting Petitioner’s contention that the webpage includes the same information as the Rituxan Label, the fact remains that the Rituxan Webpage and Rituxan Label are not the same document. Rather, “Rituximab” is partially written by hand on the Rituxan Label but not the Rituxan Webpage, and the two documents bear different layouts. Because the Rituxan Webpage and Rituxan Labels are materially different documents, and Petitioner offers no explanation as to why the purported public accessibility of the Rituxan Webpage establishes that the Rituxan Label was likewise accessible, we find Petitioner’s reliance on the Rituxan Webpage unpersuasive.

Given the fact that the label available from Genentech’s website was not the same document as the label asserted as a printed publication, the Board determined that the availability of the label from Genentech’s website did not demonstrate public accessibility of the label asserted as a printed publication. The Board also faulted the petitioner for failing “to establish that . . . the Genentech website . . . was well-known to the community interested in the subject matter of the reference, indexed, or that it included numerous related articles.”

Decisions on institution are based on the information presented in the petition

Concerning evidence of public accessibility, the dissenting opinion noted that the petitioner would have the opportunity at trial to respond to the patent owner’s contentions with regard to sufficiency of evidence with supplemental information under § 42.123(a) and/or evidentiary objections with supplemental evidence under 37 C.F.R. § 42.64(b)(2). The dissenting opinion stated that “Patent Owner is uniquely positioned to know whether or not those references were publicly disseminated prior to the critical date of August 11, 1998.” However, the majority criticized the dissent as “plac[ing] upon Patent Owner a burden to assist Petitioner in demonstrating that a cited reference is a prior art printed publication.” The majority noted that institution decisions are based on “the information presented in the petition,” not the information that a Petitioner might obtain during discovery.


The FDA website indicating the current availability of a drug label, by itself, is not sufficient to demonstrate that the drug label was actually publicly available as of the critical date. Petitioners must remember to thoroughly demonstrate that the specific document being relied upon was publicly available before the critical date. Other publicly available documents, even if they contain similar information, cannot be used to demonstrate the public availability of the specific document alleged to be a printed publication.