By Kevin E. Noonan –

The Patent Trial and Appeal Board (like its predecessor, the Board of Patent Appeals and Interferences or BPAI) occasionally renders an opinion having the tendency to raise an eyebrow or two, which on occasion has led the Federal Circuit, like its predecessor the Court of Claims and Patent Appeals, to offer a correction (the exercise of such tendencies no doubt leading to the vigor with which the Patent and Trademark Office pursued correction of the Federal Circuit by the Supreme Court, in Dickenson v. Zurko).  The Board's recent decision in Interference No. 106,115 between the Broad Institute, Harvard University, and MIT (collectively, "Broad") as Senior Party against Junior Party the University of California/Berkeley, the University of Vienna, and Emmanuelle Charpentier (collectively, "CVC") (see "PTAB Grants Priority for Eukaryotic CRISPR to Broad in Interference No. 106,115") has raised more than a few eyebrows, and CVC's brief on appeal to the Federal Circuit seeks to convince the Court that once again correction is in order.

To recap, the Board was convinced by Broad's arguments that CVC's attempts to reduce eukaryotic CRISPR to practice were unavailing until after Broad's reduction to practice as evidenced by a manuscript submitted on October 5, 2012.  Operating on the legal principle that "priority of invention goes to the first party to reduce an invention to practice unless the other party can show that it was the first to conceive of the invention and that it exercised reasonable diligence in later reducing that invention to practice," Cooper v. Goldfarb, 154 F.3d 1321, 1327 (Fed. Cir. 1998), the Board was unconvinced that CVC's March 1, 2012 conception satisfied the requirements of "complete" conception.  Using much of the same argument (albeit for different purposes) as it had to prevail in Interference No. 106,048 (see "PTAB Decides CRISPR Interference in Favor of Broad Institute — Their Reasoning"), the Broad persuasively argued that the evidence of CVC's attempts to reduce eukaryotic CRISPR to practice showed sufficient uncertainty and failures for the Board to conclude that CVC did not satisfy the requirements for conception.  On this evidence the Board was unpersuaded that all that had been needed was the application of routine experimentation using the sgRNA detailed in CVC's March 1st priority statement.  Nor was the Board convinced that Broad derived the embodiments of eukaryotic CRISPR that they reduced to practice embodying sgRNA only after Dr. Marraffini obtained it from CVC and disclosed it to the Broad inventors (see "CVC Files Reply to Broad's Opposition to CVC's Priority Motion").

CVC begins its brief by asserting that its inventors, Professors Doudna and Charpentier invented CRISPR:

This case concerns who is entitled to patents for using that system to edit genes in eukaryotic (e.g., plant or animal) cells: Doudna and Charpentier, who invented the technology, announced it to the world, and received the Nobel Prize for it; or a scientist at the Broad Institute who took their design and then (along with many others) promptly reduced it to practice using routine methods.

On the contrary, they assert, the PTAB "awarded priority to Broad because it purportedly reduced the invention to practice first—even though the PTAB never identified any inventive contribution Broad made," and that this result was "backward" and arrived at through "multiple legal errors."  This last assertion is intentional and critically important, because if the Federal Circuit is convinced that such errors were legal and not factual in nature then the Board's decision will be entitled to no deference and CVC may avoid another affirmance (the result of Federal Circuit review of the Board's decision in Broad's favor in Interference No. 106,048; see "Regents of the University of California v. Broad Institute, Inc.").

The legal errors asserted by CVC include:  1) that the Board improperly did not use an objective standard to judge CVC's conception, but rather required evidence that CVC inventors knew CRISPR would work in eukaryotic cells; 2) that the Board awarded priority to Broad without there being anything inventive done by Broad that they didn't get from CVC; and 3) requiring that the written description of CVC's applications in suit were sufficient to "persuade skeptical artisans the invention would overcome various imagined hurdles to reduction to practice in eukaryotic cells" rather than merely allowing them to identify the invention, which was contrary to the Court's instruction that written description ""is not about whether the patentee has proven to the skilled reader that the invention works," citing Alcon Rsch. Ltd. v. Barr Labs., Inc., 745 F.3d 1180, 1190 (Fed. Cir. 2014).

These assertions are distilled into three Issues Presented for the Court:

1. Whether the PTAB legally erred by failing to apply an objective standard for conception, and/or impermissibly awarded priority without identifying any inventive contribution by the purported inventor.

2. Whether the PTAB's analysis is arbitrary and capricious.

3. Whether the PTAB applied an erroneous legal standard to conclude that CVC's first and second provisional applications lacked written description.

After a factual background, the brief reaches the legal arguments CVC relies upon to support its earlier assertions regarding the PTAB's decision.  These are generally aimed at supporting CVC's assertions that how to deliver CRISPR-Cas9 complexes was known in the art from earlier work with zinc-finger nucleases (ZFN) and transcription activator-like effector nucleases (TALENs).  Adapting the CRISPR-Cas9 system to cleave DNA (any DNA) was, according to the brief, a consequence of the CVC inventors recognizing that a combination of Cas9, crRNA, and tracrRNA was enough to effect cleavage, and their further ("crucial") recognition that the crRNA and tracrRNA could be combined into a single-guide, chimeric RNA (sgRNA), illustrated by this figure:

And this copy of the laboratory notebook page where sgRNA was first conceived by CVC:

In addition to citing contemporary statements from the CVC inventors regarding how their CRISPR-Cas9 system could be used, the brief also walks through the disclosures of CVC's first provisional application (P1, USSN 61/652,086, filed May, 2012) and mentions summarily the other two provisionals (P2, filed October 19, 2012 and P3, filed January 28, 2013, the latter being the only provisional to which the Board ruled CVC was entitled to priority), illustrating the disclosure that supported their priority entitlement arguments and specific disclosure related to CRISPR-Cas9 cleavage in eukaryotic cells.  (The brief also follows the consistent pattern of CVC mentioning the acclaim its inventors experienced upon their disclosure of CRISPR-Cas9 up to and including the Nobel Prize.)

These encomiums provided a basis for CVC to make its next argument, that the scientific community appreciated the significance of CVC's CRISPR-Cas9 and rapidly applied it to eukaryotic cells because it was evident that it could be so applied.  "The question 'was not whether the CRISPR-Cas9 system would work in eukaryotes'—skilled artisans 'all expected it would,'" they state, "but whether it would 'outcompete the existing genome-editing technologies, such as TALENs and ZFNs.'"

As for CVC's own efforts (the success, confidence in, and putative failures entailed were argued successfully by Broad as evidence of failure to achieve complete conception), the brief sets forth its planned activities (in some detail) to show actual reduction to practice during the summer and fall of 2012.  These plans included using expression vectors to introduce nucleic acids encoding Cas9 and sgRNA under control of transcription promoters by which the CRISPR-Cas9 components would be expressed in the cell, and microinjection techniques where the pre-formed CRISPR-Cas9 complex would be directly introduced into the cell nucleus.

Having shown CVC's own plans, the brief sets forth examples of many other laboratories showing CRISPR-Cas9 activity in eukaryotic cells.  One of these was the Broad itself, which CVC's explains with the preface "After Obtaining CVC's Chimera A Before Publication, Broad Promptly Reports Reduction to Practice," reciting the tale of Dr. Marraffini's "improper" disclosure of this aspect of CVC's invention to Broad inventors.  In this telling, however, CVC emphasizes Broad's putative record of failures, due to (according to CVC) that those inventors' "did not understand that mature tracrRNA was a necessary third component of the final DNA-cleavage complex" and thus they had "no idea that mature tracrRNA and crRNA could be linked to form sgRNA."  "All that changed on June 26, 2012," CVC asserts, which is the day Dr. Marraffini disclosed sgRNA to the Broad inventors by sending this e-mail containing the drawing (that at the time was under confidential review) showing the sgRNA structure (something CVC asserts Broad never disclosed until discovery in this interference):

(CVC also showed a comparison (substantial identity) between its sgRNA structure and the one disclosed in a scientific paper from the Broad group, Cong 2013, published in January 2013).

And as it had before the Board, CVC sets forth the results reported from four other groups during this period, with the following comparison of the methods, reagents, and techniques used:

At the same time CVC set forth its history of reduction to practice (although all but the most casual of readers would note that the first such reduction was achieved in October 2012, 4 months after CVC's disclosure of sgRNA in June).  One explanation for this delay recited obliquely in the brief was that "Doudna's laboratory was ill-equipped for human-cell experiments" and thus had to rely on someone not having ordinary skill in the art (a graduate student "from a neighboring lab") to perform the experiments that turned out to be crucial to Broad's ability to convince the Board of CVC's failures of achieving complete conception.  (In addition, this narrative casts in perhaps their best light the history of difficulties experienced by CVC in obtaining actual reduction to practice; for example, in describing the experiments that ultimately did show CRISPR-Cas9 mediated cleavage the brief notes that they employed "the same basic vectors Jinek designed by June: an sgRNA vector with a U6 promoter, and a codon-optimized Cas9 vector with a CMV promoter and an NLS.")  For the microinjection-based experiments, the brief chalks up their less than stellar results not as failures but the consequence of having fewer resources than other labs and the decision not to publish having to do more with no opportunity to publish in a "high impact" journal than a perception of failure by CVC's inventors and collaborators.

After setting forth its explication of the course of proceedings below (including the '048 Interference) CVC sets forth its legal arguments.  First, CVC argues that the Board erred by departing from the legal standard of conception, which they argue is objective — was conception complete enough for the skilled worker to put the invention into practice (wherein the success of other labs, including at the Broad, should have been sufficient evidence of complete conception that any purported difficulties the CVC inventors and collaborator encountered would not be dispositive)?  Second, CVC asserts that "[t]he PTAB erroneously rejected CVC's originality challenge" (that sgRNA was CVC's invention that Broad used in its practice of CRISPR in eukaryotic cells).  This error arose because:

The PTAB identified nothing [Broad inventor] Zhang added that CVC had not already conceived, much less anything in the count.  For good reason: Zhang obtained the blueprint for CVC's invention—including CVC's chimera A sequence and the use of the invention in eukaryotes—from CVC's unpublished manuscript.  Zhang cannot be the inventor when he added nothing but instead (improperly) obtained every limitation of the count from CVC.

Third, CVC asserts the Board's decision was contrary to the standards under the Administrative Procedures Act imposed on administrative agencies, because "[i]t failed to connect its conclusions to its findings, and repeatedly ignored relevant evidence."  Finally, CVC argues that the Board applied the wrong legal standard for written description (a question of fact) by imposing the requirement that CVC demonstrate with experimental evidence that CRISPR would work in eukaryotic cells (which requirement was dependent on the Board being convinced by Broad's arguments to the contrary and abetted by CVC's purported failure to so demonstrate, at least not satisfactorily).  In CVC's view, "[t]he inventor's obligation is to tell artisans what she invented, not convince skeptics the invention will work."

In support of their first argument CVC cites Cameron & Everett v. Brick, 1871 C.D. 89, 90 (Comm'r Pat.), Amgen, Inc. v. Chugai Pharm. Co., 927 F.2d 1200, 1206 (Fed. Cir. 1991), and Sewall v. Walters, 21 F.3d 411, 415 (Fed. Cir. 1994), for the proposition that the Board committed two legal errors.  First, conception must be "definite and permanent," which CVC maintains it was because it satisfied the standard that its CRISPR invention could be (and was) disclosed in its priority documents (specifically P1), illustrating "possession of an operative method of making it."  The evidence CVC relies upon in this part of its argument is the success of others skilled in the art, and the Board's error CVC asserts is in relying on its own difficulties and purported failures in reducing eukaryotic CRISPR to practice.  "Whether the inventor succeeds promptly is irrelevant when the evidence shows the invention is sufficiently firm, definite, and developed that skilled mechanics can do so—and the evidence [here] shows they did," the brief argues.  And the subjective requirement imposed by the Board, that the CVC inventors knew the invention would work is contrary to Federal Circuit precedent, which contains no requirement of either expectation not knowledge that a completely conceived invention would work, CVC citing Burroughs Wellcome Co. v. Barr Labs., Inc., 40 F.3d 1223, 1228 (Fed. Cir. 1994); and Univ. of Pittsburgh of Commonwealth Sys. of Higher Educ. v. Hedrick, 573 F.3d 1290, 1299 (Fed. Cir. 2009).  Their argument hies back to Mergenthaler v. Scudder, 11 App. D.C. 264, 276 (1897), for the well-worn aphorism of patent law that "[c]onception is the 'formation, in the mind of the inventor, of a definite and permanent idea of the complete and operative invention, as it is thereafter to be applied in practice,'" wherein "[t]he 'true date' of conception is 'the point where the work of the inventor ceases and the work of the mechanic begins.'"  Put succinctly in CVC's brief, conception is complete under this precedent "'when one of ordinary skill in the art could construct the apparatus without unduly extensive research or experimentation,'" citing Sewall and Burroughs (emphasis in brief).  This is precisely the case here, according to CVC, based on the success of others, including Broad, once the sgRNA component of the CRISPR-Cas9 complex had been disclosed by CVC.  And the fact that "much patience and mechanical skill, and perhaps a long series of experiments" might have been necessary to reduce the invention to practice doesn't negate conception, CVC argued, citing Barba v. Brizzolara, 104 F.2d 198, 202 (C.C.P.A. 1939); and Acromed Corp. v. Sofamor Danek Group, Inc., 253 F.3d 1371, 1380 (Fed. Cir. 2001).  Included in this litany is Dolbear v. American Bell Telephone Co., 126 U.S. 1 (1888) (the Bell patent case, reminding the Court that Alexander Graham Bell "could not himself construct a telephone that transmitted spoken words" (emphasis in brief)), as well as the Wright brothers who applied for their airplane patent almost a year before Kitty Hawk and the patent on using AZT to treat AIDS.

The reason for this wrong result by the Board, CVC argues, is that they asked the wrong question, focusing on whether the CVC inventors were able to reduce the invention to practice without experimentation.  Encouraged by Broad the Board was led to their conclusion by the "failure" of the CVC inventors, as individuals of greater than ordinary skill in the art, to indicate incomplete conception, ignoring the success by others of ordinary skill in the art.  "It is hard to imagine more powerful, objective, real-world evidence that conception was 'complete,'" they argue, "than the fact that so many actually did "construct the apparatus without unduly extensive research or experimentation," citing Sewall.  This willful ignorance of the evidence by the Board asserted by CVC included evidence of the response of the conference attendees where Doudna disclosed sgRNA, the response of Dr. Marraffini himself, and "significance of [Broad inventor] Zhang's alleged conception on June 26, 2012, the very day he learned of CVC's invention" (emphasis in brief), followed by a litany of all the ways Broad's reduction to practice mimicked CVC's conception.

As to CVC's purported failings, the brief argues the Board was "doubly wrong" first for relying on the necessity for experimentation per se to prove incomplete conception rather than whether these experiments required more than routine skill, citing Rey-Bellet v. Engelhardt, 493 F.2d 1380, 1387 (C.C.P.A. 1974).  And second, in relying on "how the inventors or their collaborators fared, or what they said along the way" rather than "whether the inventors' idea was sufficiently firm and definite that reduction to practice could be performed by "one skilled in the art without the exercise of invention," illustrated by several examples including Bell's telephone, Sewall and Acromed.  And CVC contends that the case law relied upon by the Board (Alpert v. Slatin, 305 F.2d 891 (C.C.P.A. 1962)) can be distinguished here because in that case the only evidence was of the inventor's failure to reduce to practice, whereas here several other laboratories were able to do so contemporaneously (which CVC argues the Board improperly characterized as nunc pro tunc conception).

Finally, in this portion of their argument, CVC addresses the subjective test for reduction to practice they allege the Board employed as being improper under City of Elizabeth v. Nicholson Pavement, 97 U.S. 126 (1877), Applegate v. Scherer, 332 F.2d 571, 573 (C.C.P.A. 1964); Dana-Farber Cancer Inst., Inc. v. Ono Pharm. Co., 964 F.3d 1365, 1372 (Fed. Cir. 2020), and in particular Burroughs, where the Court held the inventors did not need to know whether the invention worked to have complete conception.  CVC distinguished Hitzeman, relied upon by the Board, on the basis that in that case the Court merely held that for claims to a compound to be conceived the inventor must expect to be able to "produce" the composition (something CVC contends its inventors had done here, at least in vitro).

Next, CVC's brief turns to the issue of originality and the evidence and argument asserted before the Board that Broad had merely reduced to practice eukaryotic CRISPR using CVC's sgRNA after it was (improperly) disclosed to Broad's inventors by Dr. Marraffini (an argument that loses some equitable force because the disclosure was public).  According to CVC, the Board granted Broad priority without identifying any aspect of the invention original to their inventors.  CVC points out that the Board presumed Broad made an inventive contribution ("there must have been differences") (emphasis in brief) because the Broad inventors succeeded in reducing the invention to practice while CVC's inventors had not (what CVC terms the "ipse dixit" determination).  This conclusion, CVC maintains, was arrived at by the Board without considering much less determining "whether Zhang added anything inventive or was just a more efficient (or luckier) mechanic " (emphasis in brief).  The reason for this situation is clear to CVC: the Board did not make findings of anything inventive by the Broad inventors because they could not — "[Broad inventor] Zhang got the invention from CVC."  For CVC, the Board's decision "flouts the requirement that patents be awarded to an 'original inventor'—not a 'borrower or a copyist,'" citing 1 W. Robinson, The Law of Patents for Useful Inventions § 58 (1890) (as well as the U.S. Constitution) and Applegate.  Ironically, CVC argues that, had CVC hired Broad's inventor to reduce the invention to practice under these circumstances it would be beyond dispute that CVC's inventors would have prevailed.

The brief then sets forth CVC's argument that the Board violated the provisions of the APA for reasoned decision making and should be considered "arbitrary and capricious," "otherwise not in accordance with law," or "unsupported by substantial evidence," giving the Federal Circuit several ways to at least vacate the Board's decision for remand.  These arguments rely for their facts on what CVC set forth above:  for example, that the Board found in favor of the Broad inventors without any evidence of inventive contribution to CVC's conception (the ipse dixit, "there must have been differences" standard).  CVC contends that the APA requires that "[a]gency decisions must be grounded in evidence," citing Morall v. DEA, 412 F.3d 165, 178 (D.C. Cir. 2005), and evidence CVC contends was lacking for this conclusion (and in at least one instance involving use of the U6 promoter to make sgRNA CVC contends the Board committed a mistake of fact).  CVC cites the principle that "[c]ourts cannot 'uphold agency action if' the agency decision 'fails to consider "significant and viable and obvious alternatives,'" citing Dist. Hosp. Partners, L.P. v. Burwell, 786 F.3d 46, 59 (D.C. Cir. 2015), which "precisely describes the situation here" according to CVC.  Moreover, CVC asserts that the frequency of Broad's successes (0.75%) was sufficiently low that "many experiments may fail to demonstrate cleavage due to random chance" or "human error," "equipment quality," or "measurement failures" (or the fact that CVC's experiments were performed initially by a graduate student and CVC achieved success when a more experienced experimentalist took up the task, without and change in the protocol).  And some of those "failures" were attempts at CRISPR reactions that were outside the scope of the Count and thus should not have been applied to the Board's calculus.  The Board ignored "rafts" of evidence contrary to its conclusions, CVC states, including contemporaneous encomiums by scientists like Dr. Marraffini about sgRNA-based CRISPR; the timing of other laboratories successes relative to CVC's inventors' disclosure of sgRNA-based CRISPR; the persistence ("stability") of CVC's protocol, which did not change throughout its attempts to produce actual reduction to practice; and with regard to the subjective standard, that CVC's inventors never believed the protocols would not be successful despite the highly touted expressions of doubt asserted by Broad to support its incomplete conception narrative.  And CVC brings forth an argument it made during Final Hearing to the effect that the four months it took for CVC to reduce the invention to practice was not a significant delay (the Wright brothers took almost a year to achieve powered human flight, for example).  In some ways playing their trump card, CVC states that "[n]owhere did the PTAB explain why the inventors awarded the Nobel Prize for probably the most stunning advance of the century should be disqualified because their graduate students initially stumbled but ultimately succeeded in just four months."  CVC makes similar arguments regarding CVC's microinjection experiments and accuses the Board of conflating the requirements of these experiments with vector-based expression of CRISPR-Cas9 in eukaryotic cells.

Finally, the brief addresses the Board's decision to deny priority benefit to CVC's earlier P1 provisional application based on lack of written description, saying the Board applied the wrong standard once again.  In interferences, CVC states, all that is required is that a priority document describe one embodiment falling within the scope of the Count, citing Falkner v. Inglis, 448 F.3d 1357, 1362 (Fed. Cir. 2006).  Because the P1 provisional described a protocol for eukaryotic CRISPR that others (including Broad) followed to achieve the result and that protocol was what CVC's inventors ultimately themselves reduced to practice CVC's P1 application satisfied the standard.  The brief sets forth in detail (as CVC has done throughout the interference before the Board) what is disclosed in P1 and how the methods for practicing CRISPR in eukaryotic cells were conventional techniques used in earlier cleavage regimes like ZFN and TALENs.  The Board in error, CVC maintains, applied a "burden to convince," contrary to the Court's instructions in Alcon that:

Written description "is about whether the skilled reader of the patent disclosure can recognize that what was claimed corresponds to what was described."  . . .  Consequently, this Court has warned, it "is not about whether the patentee has proven to the skilled reader that the invention works, or how to make it work."

Written description does not require that the person of ordinary skill be persuaded that an invention works, nor does delay during the application of diligence negate the existence of an adequate written description, CVC asserts, citing BASF Plant Sci., LP v. Commonwealth Sci. & Indus. Rsch. Org., 28 F.4th 1247, 1267 (Fed. Cir. 2022) — again stating that the Board applied the wrong standard regarding what is required to provide an adequate written description under the statute.  What is required, CVC argues, is a description, not evidence that the described invention works, and here the successes of other labs, including Broad's, and the CVC inventors' eventual success establishes that CVC's P1 priority document contained the required description.  While possession is required proof is not, according to the brief, and the Board's imposition of the proof requirement was another error CVC lays at the Board's doorstep.  CVC cites the Board's differential treatment of P1 (not adequately described) and P3 (satisfies written description requirement) to illustrate the point, because the only difference between these two priority documents is that the P3 application included a working example of CRISPR in eukaryotic cells.  CVC closes this portion of the brief by reminding the Court that an inventor could never foreclose the possibility that an invention might not work, because: 

No inventor can anticipate and preempt with instructions every single litigation-inspired hypothetical problem that can be conjured.  Lawyers can always imagine 1,001 reasons an invention might not work.  Paid experts can invent still more.

"Possession" in this context, CVC asserts, means possession of the idea, the conception of the invention, nor "construction of a working example."  This is because, as CVC concludes its brief, the controlling principle is that "[t]he patent system does not punish inventors of breathtaking innovations by saddling them with the burden of convincing putative skeptics their invention will work."

By Kevin E. Noonan –

On September 28, 2022, the Patent Trial and Appeal Board denied all preliminary motions by Junior Party the University of California, the University of Vienna, and Emmanuelle Charpentier (collectively, "CVC") and Senior Party ToolGen in Interference No. 106,127.

On the same day, the Board suspended further proceedings in this interference (and in Interference No. 106,126 between Junior Party the Broad Institute, Harvard University and MIT (collectively, "Broad") and Senior Party ToolGen) "in the interest of not wasting resources" (see "PTAB Suspends ToolGen Interferences"), reasoning foreshadowed in the first portion of their decision on motions which notes that the Board's judgment in the '115 Interference is binding for the purposes of this interference even though the decision is on appeal.  But because the Federal Circuit has not overturned the Board's earlier judgment, the panel continues (again "in the interests of efficiency") in preparation for the now-suspended priority phase of this interference.

As a reminder, the parties had filed the following motions:

• For CVC, Substantive Motion No. 1 for priority benefit to U.S. Provisional Application No. 61/652,086, filed May 25, 2012 ("P1"), U.S. Provisional Application No. 61/716,256, filed October 19, 2012, ("P2"), and U.S. Provisional Application No. 61/757,640, filed January 28, 2013 ("Provisional 3"); Substantive Motion No. 2 to deny ToolGen benefit of priority to U.S. Provisional Application No. 16/717,324, filed October 23, 2012 ("P1"); Substantive Motion No. 3 asking the Patent Trial and Appeal Board to add claims in ToolGen's U.S. Patent No. 10,851,380 to this interference; and Contingent Responsive Motion No. 1 to be accorded benefit of priority to U.S. Patent Application No. 13/842,859, filed March 15, 2013, or in the alternative U.S. Patent Application No. 14/685,504, filed April 7, 2015, or U.S. Patent Application No. 15/138,604, filed April 26, 2016.

• ToolGen filed Substantive Motion No. 1 for priority to later-filed U.S. Provisional Application No. 61/837,481, filed June 20, 2013 ("P3" or "ToolGen P3"), or alternatively, International Application No. PCT/KR2013/009488, filed Oct. 23, 6 2013 ("PCT"), contingent on the Board granting CVC Substantive Motion No. 2 to deny ToolGen benefit of priority to U.S. Provisional Application No. 61/717,324, filed October 23, 2012 ("P1"), and Substantive Motion No. 2 to deny CVC priority to its U.S. Provisional Application No. 61/757,640, filed January 28, 2013 ("Provisional 3").

The Board's opinion does not address the parties' motions in order but starts with Junior Party CVC's Motion No. 1 to be accorded benefit to the P1 and/or P2 provisional applications.  The Board denied this motion based on its decision in Interference No. 106,115 on a related motion and the principle of issue preclusion based inter alia on the identify in scope of the interference count in this interference with the scope of the interference count in the '115 Interference, citing Lawlor v. National Screen Serv. Corp., 349 U.S. 322, 326 (1955).  Citing Parklane Hosiery Co. v.  Shore, 439 U.S. 322, 329 (1979), the Board asserts that ToolGen being CVCs opponent in this interference does not negate their application of the issue preclusion doctrine, which is based on "whether an issue of law or fact has been previously litigated," citing International Order of Job's Daughters v. Lindeburg & Co., 727 F.2d 1087, 1091 (Fed. Cir. 1984), and as it did in the '126 interference under similar circumstances, Mother's Restaurant, Inc. v. Mama's Pizza, Inc., 723 F.2d 1566, 1569 (Fed. Cir. 1983).  Because CVC's Motion No. 1 satisfies all the requirements for the Board to apply issue preclusion in this instance, citing A.B. Dick Co. v. Burroughs Corp., 713 F.2d 700, 702 (Fed. Cir. 1983), the Board denied this motion, with the opinion addressing and rejecting CVC's arguments to the contrary (e.g., that the determination of priority is distinct from determination of benefit and thus the principle is not applicable because the earlier decision was not necessary for the earlier judgment; the Board asserted that the issue before it here is the same as the issue before the Board in the '115 Interference and thus the principle was properly applied).

The opinion next addressed ToolGen Motion No. 2 to deny CVC benefit of its P3 provisional application.  ToolGen's argument as understood by the Board was that CVC did not show in P3 that its CRISPR-Cas9 system was "capable of cleaving or editing the target DNA molecule or modulating transcription of at least one gene encoded by the target DNA molecule" or that "meditate[s] double stranded cleavage at the [DNA] target sequence."  Citing its decision denying priority benefit to CVC's P1 and P2 provisional applications, the Board notes that it provided two alternatives regarding possession sufficient to support priority benefit: "experimental evidence of possession of an embodiment of Count 1" or in the alternative "instructions or discussion of conditions that would have been considered to hinder activity in eukaryotic cells."  ToolGen's arguments were unpersuasive for the Board because they focused only on the first alternative according to the opinion.  The question for the Board was whether a person of ordinary skill would have considered CVC's P3 provisional application to show possession of an embodiment of Count 1 under the second alternative basis.  Regarding this the Board performed a comparison of the P3 disclosure with what CVC's inventors (Jinek et al.) disclosed in a contemporaneous scientific publication.  This comparison was directed to Example 2 of the P3 application (which ToolGen specifically argued was insufficient, supported by expert testimony) and associated Figure 36E and 38B (the latter being particularly called out by ToolGen for having bands significantly differing in size or position from what was expected if cleavage had been successful).  While CVC countered with its own expert, what was ultimately persuasive to the Board (and fatal to ToolGen's motion) was this comparison between Figure 36E and Figure 38B in the P3 provisional application with figures in a paper published in eLife on January 29, 2013 ("the day after CVC P3 was filed" according to the opinion):

To the Board, "[t]he bands in Figure 36E of CVC P3 and Figure 1E of the e-Life manuscript appear to be at least similar" and "[a]s with Figure 36E of CVC P3 and Figure 1E of the e-Life manuscript, the bands of each figure [38B and 3] appear to be similar."  Further, the Board considered the opinion of the peer reviewers of the eLife manuscript, that:

This excellent paper demonstrates the capability of a method based on the bacterial CRISPR-Cas system to introduce targeted, mutagenic double-strand breaks into human chromosomal DNA.  This system could become a powerful alternative to protein-based targeting reagents, because it is based on simple Watson-Crick recognition and no protein design is required.  To put this into perspective, the first proof-of concept study with zinc finger nucleases (ZFNs) in human cells targeted plasmid, not chromosomal DNA [emphasis in opinion].

The Board concludes that "[t]hus, the reviewers, persons of ordinary skill in the art at the time CVC P3 was filed, considered the gels depicted in Figures 38B and 36E to contribute to evidence of a CRISPR-Cas9 system that successfully cleaved DNA within eukaryotic cells, noting 'the paper demonstrates' this and the study 'show[s]' it.  While agreeing with ToolGen that Ariad Pharmaceuticals, Inc. v. Eli Lilly & Co., 598 F.3d 1336, 19 1357–58 (Fed. Cir. 2010), requires an applicant to show possession of the claimed invention, the Board emphasizes that the possession test requires "an objective inquiry into the four corners of the specification from the perspective of a person of ordinary skill in the art" (emphasis in opinion).  The similarity of the disclosure in P3 and the eLife publication, and the publication reviewer's assessment of the science disclosed therein, was sufficient to convince the Board that the CVC inventors possessed the invention and were entitled to priority benefit to the P3 provisional application.  Accordingly the Board denied ToolGen's Motion No. 2.

The Board dismissed CVC's Responsive Motion No. 2 because it was contingent on the Board granting ToolGen's Motion No. 2 and in view of its denial the Board dismissed this motion as moot.

Turning to CVC Substantive Motion No. 2, to deny ToolGen priority benefit to its P1 provisional (in some ways the most significant motion, because grant of this motion would make CVC the Senior Party) the Board denied the motion because while ToolGen did not provide a constructive reduction to practice of an embodiment of a eukaryotic CRISPR Cas9 system encoded by a codon-optimized Cas9 protein-encoding nucleic acid (as recited in ToolGen's half of MacKelvey Count 1 as declared), the CVC half of the Count had no such requirement.  The legal basis for this decision, as set forth by the Board, was that "priority of the filing date of a prior application in an interference is granted with respect to counts not claims; all that is necessary for a party to be entitled to benefit of an earlier filed application for priority purposes is compliance with 35 U.S.C. § 112 with respect to at least one embodiment within the scope of 22 the count"  (emphasis in opinion), citing Falko-Gunter Falkner v. Inglis, 448 F.3d 1357, 1362 (Fed. Cir. 23 2006), and Hunt v. Treppschuh, 523 F.2d 1386, 1389 (C.C.P.A. 1975).  The Board rejected CVC's reliance on "judicial estoppel" (based on positions ToolGen had taken during ex parte prosecution of ToolGen's utility application involved in this interference) because the standard is whether ToolGen showed reduction to practice of an embodiment within the Count, and here the Count included a portion that did not require codon-optimized Cas9 protein-encoding nucleic acid.  As stated in the opinion, "[t]he purpose of an interference count is to provide the description of the interfering subject matter that sets the scope of admissible proofs on priority," citing 37 C.F.R. § 41.201 and Slip Track Sys., Inc. v. Metal-Lite, Inc., 304 F.3d 1256, 2 1263 (Fed. Cir. 2002).  As a practical matter, the Board recognized that "[i]f the parties were limited to their own 'half' of a count, the scope of admissible proofs could differ for each, making a determination of priority for a common invention problematic."  To CVC's accusations of unfairness, the Board stated that "CVC did not request authorization to argue that Count 1 should be changed to refine the boundaries of the parties' commonly invented subject matter[, n]or did CVC request authorization to argue that the parties' claims do not interfere."  Except for these arguments, the opinion notes, "CVC does not argue that the ToolGen P1 application fails to describe an embodiment within the scope of Count 1" and further rejected CVC's argument that ToolGen should be estopped from asserting a constructive reduction to practice.  Accordingly, the Board denied CVC's Motion No. 2, leaving the status of the parties undisturbed.

As it had for CVC's Responsive Motion No. 2, the Board dismissed as moot ToolGen's Motion No. 1 for priority benefit to applications later-filed than its P1 provisional application because it was contingent on grant of CVC's Substantive Motion No. 2 (which the Board denied).

The Board also denied CVC's Motion No. 3 to add claims of ToolGen Patent No. 10,851,380 to the interference.  The only distinction between these claims and the Count was the addition of two guanine residues ("GG") at the 5' end of the guide RNA, and CVC argued that this difference would be obvious in view of the Jinek reference.  The Board rejected many of ToolGen's arguments, particularly with regard to uncertainty in the art regarding eukaryotic CRISPR on the grounds that the Count recites eukaryotic CRISPR and the analysis here considers whatever is recited in the Count to be prior art.  Nevertheless, CVC's arguments attempting to rebut ToolGen's evidence of unexpected results (regarding decrease in off-target effects) were equally unavailing and the Board accordingly denied CVC's motion in view of that evidence.  And while ToolGen's evidence is somewhat thin ("a bare assertion that one of ordinary skill in the art would have found the results to be surprising and/or unexpected"), the Board finds that "CVC had an opportunity to put forth and cite contradictory evidence of how one of ordinary skill in the art would have understood the results reported in the '308 patent, including from the cross-examination of [ToolGen's expert witness]" which it did not do ("Rather, CVC presents arguments about the nature of the comparison presented in the '308 patent, its nexus to the claims, and whether ToolGen appropriately relied on these results.  As explained above, we are not persuaded by these arguments.")  On these grounds the Board denied CVC's Motion No. 3.

Finally, the Board dismissed as moot the parties' various motions to exclude or concluded it has no need to address parties' objections to admissibility in view of their decisions to deny the parties' motions.

The interference thus remains in the same posture between the parties as it was when declared, and the parties must await the outcome of CVC's appeal in the '115 Interference.  Unless the Supreme Court deigns to review the Federal Circuit's judgment (which is very unlikely regardless of the substance of that judgment) the suspension will not be lifted until sometime next spring, and a decision reversing the Board's awarding priority to Broad should result in the priority phase here ensuing.

By Kevin E. Noonan –

On September 28, 2022, the Patent Trial and Appeal Board denied all preliminary motions by Junior Party the Broad Institute, Harvard University, and MIT (collectively, "Broad") and Senior Party ToolGen in Interference No. 106,126.

On the same day, the Board suspended further proceedings in this interference (and in Interference No. 106,127 between Junior Party the University of California, the University of Vienna, and Emmanuelle Charpentier (collectively, "CVC") and Senior Party ToolGen) "in the interest of not wasting resources" (see "PTAB Suspends ToolGen Interferences"), reasoning foreshadowed in the first portion of its Decison on motions which notes that the Board's judgment in the '115 Interference is binding for the purposes of this interference even though the decision is on appeal.  But because the Federal Circuit has not overturned the Board's earlier judgment the panel continues (again "in the interests of efficiency") in preparation for the now-suspended priority phase of this interference.

As a reminder, the parties had filed the following motions:

• For Broad, Substantive Motion No. 1 to change the Count (a motion similar to the motion denied in Interference No. 106,115); Contingent Motion No. 2 to add U.S. Application Nos. 15/160,710 (having allowable claims 1, 40, and 41) and 15/430,260 (allowable claims 74, 94, and 95) should the Board grant Motion No. 1; and Substantive Motion No. 3 to de-designate Broad claims as not corresponding to either Count 1 or proposed Count 2.

• ToolGen had a single motion in this interference, Substantive Motion No.1 for priority to later-filed U.S. Provisional Appl. No. 61/837,481, filed June 20, 2013 ("P3" or "ToolGen P3"), or alternatively, International Appl. No. PCT/KR2013/009488, filed Oct. 23, 2013 ("PCT").

With regard to Junior Party's Motion No. 1 (to change the Count), the Board held that Broad had not satisfied its burden of proof at least because Broad's proposed Count No. 2 included changes other than merely broadening the count to include dual-guide RNA (dgRNA) as well as single-guide RNA (sgRNA), to which the count was limited in the interference as declared.  First, the Board agreed with ToolGen that Broad's proposed Count No. 2 added the words "or edited" to the portion of the Count that states "the guide sequence directs the Cas9 to the target sequence, whereby the DNA molecule is cleaved or edited in the eukaryotic cell" (emphasis in opinion).  Second, the Board agreed with ToolGen that Broad's proposed Count No. 2 deleted the phrase "expression of at least one gene product is altered" from Broad's portion of the MacKelvey count in the interference.  ToolGen argued that changes rendered proposed Count No. 2 "excessively broad" but while the Board did not indicate agreement with this characterization it held that Broad had not adequately explained why these changes were necessary as part of its burden in asserting its motion No. 1, the Board stating it would not make changes "for change's sake," citing Louis v. Okada, 59 U.S.P.Q.2d 1073, 1076 (BPAI 2001).

"Nevertheless," the Board stated, the decision addresses Broad's arguments in support of its Motion No. 1.  Despite basing its several arguments on the characteristic of Proposed Count 2 to encompass dgRNA as well as sgRNA, the Board stated that Broad had failed to show any ToolGen claims that are broader than sgRNA embodiments.  The Board disregarded Broad's arguments that ToolGen is pursuing dgRNA-reciting claims because such "ToolGen claims have not been allowed and are not involved in the current interference."  The Board recognized that "Broad's argument is that because its own claims and proofs are outside the scope of Count 1, the count is too narrow" (emphasis in opinion), and that "this is unjust because it puts Broad's generic RNA invention at risk, while excluding Broad's best proofs, which are based on dual-molecule RNA systems."  The Board turns on their head Broad's arguments that ToolGen would not be prejudiced because its claims do not recite dgRNA-containing eukaryotic CRISPR embodiments by saying that "Broad fails to explain why it would be just to put ToolGen's involved claims, limited to a single-molecule system, at risk with proofs to a dual-molecule system, unless both are the same patentable invention."  Ultimately the Board's decision is based on the Broad failing to show "why the single- and dual-molecule CRISPR-Cas9 systems should be included in one count."  And the Board was not persuaded by Broad's citation of authority regarding the permissibility for a party to broaden a count to include its "best proofs," first noting that none of this authority is binding and then distinguishing each case Broad relied upon, including Grose v. Plank, 15 USPQ2d 1338, 1342 (BPAI 1990); Kondo v. Martel, 220 USPQ 47, 49 (BPAI 1983); and Nelson v. Drabek, 212 U.S.P.Q. 98 (Comm'r 1979).  The upshot of this criticism is that Broad has not, in the Board's view, "cite[d] support, precedential or non-precedential, that a count should be broadened to include more than one patentable invention for the purpose of allowing a party to present its best proofs," citing the Board's precedential decision in Lee v. Mcintyre, 55 USPQ2d 1137, 1142 (BPAI 2000), that "there is no per se rule that the count ultimately must be as broad as the broadest patent claim corresponding to the count."  And because under the "presumption" rules the Board sees as governing the interference (that a claim corresponds to the count when it would be anticipated or obvious if the count is considered to be prior art) "a species count may anticipate a genus claim," and "designation of claim correspondence to the count thus does not equate to a presumption that the scope of the count is equal to the full scope of the designated claims.  In short, there is no presumption here that the species count is the same patentable invention as any genus claims designated as corresponding to the count."

In addition, the Board, asserts, Broad "fails to meet its burden in Motion 1 because it fails to present argument or direct us to evidence that the dual-molecule and single-molecule CRISPR-Cas9 systems are the same patentable invention" and alternatively "Broad fails to present argument or direct us to evidence that the dual-molecule and single-molecule CRISPR-Cas9 systems are separately patentable."  The Board also faulted Broad for failing to be thorough in its arguments regarding "common or separate patentability of the dual- and single-molecule CRISPR-Cas 9 systems," nor did Broad argue that either of the two systems are obvious over the other according to the opinion nor show evidence related to whether one species of CRISPR was obvious over the other.  These errors led to the Broad leaving it to the Board to make the determination, rather than bearing its burden to do so, in the Board's opinion, thus failing to satisfy the requirements of 37 C.F.R. § 41.208(b).

Under these circumstances, Broad failed to convince the Board to grant Motion No. 1, the opinion stating that their denial was "not because we are persuaded of any specific relationship between a single-molecule and a double-molecule RNA configuration, but because we are not persuaded of any such relationship" (emphasis in opinion).  The Board suggests that Broad had two choices: one, to show dual- and single-molecule CRISPR were the same patentable invention (which they did not do) or to move to add a separate count to the interference directed to dual-molecule CRISPR embodiments (which Broad chose not to do)(because, of course, Broad wanted the interference to have a single count encompassing both dual- and single-molecule CRISPR for which its assertion of proofs of earlier invention of dual-molecule CRISPR would garner priority to claims for both species).

The Board responded to Broad's unfairness arguments by stating that Broad failed to adduce evidence that Proposed Count 2 does not encompass more than one patentable invention.  In this regard, the opinion notes that Broad's Motion No. 3 asks the Board to designate claims not limited to single-molecule CRISPR as not corresponding to Count 1 (i.e., claims directed to dual-molecule CRISPR and so-called "generic" claims not limited to one or the other species of gRNA), again on putative fairness grounds. Here again the Board states that Broad did not present evidence or argument on obviousness and instead "argues for relief that would involve contradictory reasoning — in Motion 1 that the single- and dual molecule CRISPR-Cas9 systems are the same patentable invention and in Motion 3 that a 'non-limited' CRISPR-Cas9 system is separately patentable from a single-molecule CRISPR-Cas9 system — without directing us to evidence in either context."  For the Board, "fairness of the proceeding to both parties depends on whether the single-molecule and dual-molecule systems are the same patentable invention or are separately patentable" — and Broad has not convinced the Board of either proposition.

The Board specifically disagreed with Broad's contention that "the relevant question would be whether a generic RNA genus is directed to the same invention as the single-molecule RNA species" (emphasis in opinion), stating that the "relevant question" is the scope of Proposed Count 2 — something on which the Board accuses the Broad of not taking a position.

Accordingly, the Board held that Broad failed to meet its burden under 37 C.F.R. § 41.208(b) and denied the motion.

Turning to Broad Motion No. 2, a contingent motion designating certain claims of Application Nos. 15/160,710 and 15/430,260, the Board dismissed this motion as moot in view of its decision regarding Broad Motion No. 1.  And regarding Broad's contention that if the Board denied its Motion No. 1 it should designate claim 41 of the '710 application and claim 95 of the '260 application as corresponding to Count 1, the Board declined the invitation because Broad had not asked for leave to file a motion to have additional claims designated as corresponding to Count 1.

Returning to Broad Motion No. 3, the Board states that Broad was required in order to prevail to show that the claims it seeks to have designated as not corresponding to Count 1 be neither anticipated not obvious by the subject matter of the Count (under circumstances where features of the claimed CRISPR system, such as being capable of cleaving DNA and altering gene expression in eukaryotic cells, are presumed under 37 C.F.R. § 41.207(b)(2) to be taught in the art).  The Board opinion first finds estoppel arising from the Board's decision in the '115 interference regarding SaCas9, a CRISPR-Cas9 system with multiple nuclear localization signals, and claims not limited to an RNA configuration.  The Board in that interference decided claims reciting each of these limitations corresponded to the count, which included claim 18 of Broad patent No. 9,697,359 (as does the Count in this interference).  The Board in this opinion states that it "decline[s] to decide these issues again under the doctrine of issue preclusion," not due to any position ToolGen has taken but because "[a] party who has litigated an issue and lost should be bound by that decision and cannot demand that the issue be decided over again," citing  Mother's Restaurant, Inc. v. Mama's Pizza, Inc., 723 F.2d 1566, 1569 (Fed. Cir. 1983), and that the circumstances here satisfied the requirements of A.B. Dick Co. v. Burroughs Corp., 713 F.2d 700, 702 (Fed. Cir. 1983).  Nevertheless, the Board reviewed (and rejected) Broad's arguments regarding the patentable distinctness of claims to CRISPR encompassing generic RNA species, despite the fact that these claims satisfy the requirements of 37 C.F.R. § 41.207(b)(2) due to unfairness in considering this "a per se rule that must be rigidly applied."  This outcome was once again based on the Board's conclusion that Broad failed to argue or present facts supporting this distinction, because (referencing ToolGen's arguments) "even if Broad is correct that at least some of its claims are 'generic' as to RNA configuration, a count directed to species — single molecule RNA configuration — would anticipate these claims, if treated as prior art."  Relying on the presumption that under the declaration of interference these claims correspond to Count 1, the Board denied Broad Motion No. 3 because "Broad fails to argue or direct us to evidence to the contrary," nor did it support its "inequity" arguments with evidence that single- and dual-molecule RNA CRISPR species are the same invention (and thus that a finding based on earlier invention of single-molecule species would preclude separate priority to dual-molecule CRISPR species), and rejects Broad's citation of rulemaking involving 37 C.F.R. § 41.207(b)(2) based on a distinction between genus and species grounds because, inter alia, Broad did not seek leave to file a motion to add a genus count to the interference.

With regard to claims reciting SaCas9, the Board states that Broad had argued and failed to persuade regarding claims reciting this aspect in the '115 Interference.  Regardless of whether issue preclusion would properly prevent Broad from reasserting these arguments, the Board finds these claims to correspond to the Count on the same basis as it had in the '115 Interference and found unpersuasive evidence of commercial success, expert testimony, and citations to scientific literature adduced in support of Motion No. 3 in this interference.  The Board reached the same conclusion on similar grounds with regard to claims reciting two or more nuclear localization signals, finding Broad had not convinced them that such claims do not correspond to Count 1.  After a long disquisition of the parties' arguments regarding vector delivery of CRISPR components the Board concluded that "we are not persuaded that one of ordinary skill in the art would have considered Broad's involved claims limited to vector delivery to be non-obvious over the subject matter of Count 1 treated as prior art."  And finally, the Board states that Broad failed to rely on persuasive evidence that chimeric Cas9 species or fusion with protein or heterologous domains would not be obvious over the count considered to be prior art.  On these bases, the Board denied Broad's Motion No. 3.

With regard to ToolGen's Motion No. 1 the Board held this was moot because it was contingent on the Board granting Broad's Motion No.1, and dismissed ToolGen's miscellaneous motion to exclude evidence because it had denied Broad's Motion Nos. 1 and 3.

The interference thus remains in the same posture between the parties as it was when declared, and the parties must await the outcome of CVC's appeal in the '115 Interference.  Unless the Supreme Court deigns to review the Federal Circuit's judgment (which is very unlikely regardless of the substance of that judgment), the suspension will not be lifted until sometime next spring, if it is — a decision reversing the Board's awarding priority to Broad should dissolve this interference in ToolGen's favor, the ultimate question of priority thus being the province of the '127 Interference between CVC and ToolGen.

By Kevin E. Noonan –

On September 28, 2022, the Patent Trial and Appeals Board suspended proceedings in Interference No. 106,126, between junior Party the Broad Institute, Harvard University and MIT (collectively, "Broad") and Senior Party ToolGen, and in Interference No. 106,127, between Junior Party the University of California, the University of Vienna, and Emmanuelle Charpentier (collectively, "CVC") and Senior Party ToolGen.  On the same day, the Board had issued a Decision on Motions in each interference (to be discussed in another post), leaving only the priority phase to be contested between each pair of parties.

The basis for suspension was an acknowledgement that CVC and Broad had been involved in prior Interference No. 106,115 and that the decision in that interference was under consideration by the Federal Circuit on appeal.  Recognizing that "the count in [each of] the current interference[s] is similar in the count in the '115 Interference and some issues raised and decided during the priority phase in the '115 Interference are similar," a decision by the Federal Circuit could "potentially impact[] a decision on priority in [each of these] interference[s]" (the language in each Order being identical, it has been altered here accordingly).

The Order suspends proceedings in each intereference "in the interest of not wasting resources" and states that "a schedule will not be issued before the decisions in the '115 Interference are final and the Federal Circuit has issued a mandate."

Stay tuned.