Board of Trustees of the Leland Stanford Junior University v. Roche Molecular Systems, Inc. (Fed. Cir. 2009)

    By Kevin E. Noonan --

The pitfalls attendant on collaborations between industry and academia are nicely illustrated in the Federal Circuit's decision last week in Board of Trustees of the Leland Stanford Junior University v. Roche Molecular Systems, Inc.

Stanford asserted three patents relating to methods for using polymerase chain reaction (PCR) detection of human immunodeficiency virus (HIV) RNA in blood samples for assessing the efficacy of antiretroviral drug treatment.  These patents were:

U.S. Patent No. 5,968,730 (Merigan, Katzenstein and Holodniy, inventors), issued on October 19, 1999:

1.  A method of evaluating the effectiveness of anti-HIV therapy of a patient comprising:
    (i) collecting a plasma sample from an HIV-infected patient who is being treated with an antiretroviral agent;
    (ii) amplifying the HIV-encoding nucleic acid in the plasma sample using HIV primers in about 30 cycles of PCR; and
    (iii) testing for the presence of HIV-encoding nucleic acid, in the product of the PCR;
in which the absence of detectable HIV-encoding nucleic acid correlates positively with the conclusion that the antiretroviral agent is therapeutically effective.

U.S. Patent No. 6,503,705 (Kozal, Merigan, Katzenstein and Holodniy, inventors), issued on January 7, 2003:

1.  A method of evaluating the effectiveness of anti-HIV therapy of an HIV-infected patient comprising:  a) collecting statistically significant data useful for determining whether or not a decline in plasma HIV RNA copy numbers exists after initiating treatment of an HIV-infected patient with an antiretroviral agent by: (i) collecting more than one plasma sample from the HIV-infected patient at time intervals sufficient to ascertain the existence of a statistically significant decline in plasma HIV RNA copy numbers; (ii) amplifying the HIV-encoding nucleic acid in the plasma samples using HIV primers via PCR for about 30 cycles; (iii) measuring HIV RNA copy numbers using the products of the PCR of step (ii); (iv) comparing the HIV RNA copy numbers in the plasma samples collected during the treatment; and b) evaluating whether a statistically significant decline in plasma HIV RNA copy numbers exists in evaluating the effectiveness of anti-HIV therapy of a patient.

And U.S. Patent No. 7,129,041 (Merigan, Katzenstein and Holodniy, inventors), issued on October 31, 2006:

1.  A method of evaluating the effectiveness of anti-HIV therapy of a patient comprising:  correlating the presence or absence of detectable HIV-encoding nucleic acid in a plasma sample of an HIV infected patient with an absolute CD4 count, wherein the presence or absence of said detectable HIV-encoding nucleic acid is determined by (i) collecting a plasma samples from an HIV-infected patient who is being treated with an antiretroviral agent; (ii) amplifying HIV-encoding nucleic acid that may be present in the plasma sample using HIV primers via PCR and; (iii) testing for the presence of HIV-encoding nucleic acid sequence in the product of the PCR.

All three patents were assigned to Stanford University.

The background facts recited in the Federal Circuit's opinion relate to co-development in the late 1980's and early 1990's of the application of PCR technology to HIV detection by Stanford and Cetus, a private company that developed and owned the PCR technology.  These collaborations were governed by a series of written agreements between the company and the University.

In 1988, Inventor Holodniy signed a "Copyright and Patent Assignment (CPA)" agreement as a condition for his employment as a post-doctoral researcher in Inventor Merigan's laboratory governing ownership of any inventions he developed.  However, in 1989 he began a series of "regular visits" to Cetus to learn the PCR technique.  The company imposed as a requirement that he sign a "Visitor's Confidentiality Agreement (VCA)" with Cetus that purported to assign his rights to any "ideas, inventions and improvements" made by him "as a consequence of" the work he performed at Cetus.

Complicating matters, Stanford and Cetus signed several "Materials Transfer Agreements" permitting Stanford to use PCR-related reagents and information supplied by Cetus during the performance of the work that culminated in the patents-in-suit.  In return, Cetus received licenses to the technology developed by the Stanford researchers using the materials supplied by Cetus.

The Stanford Technology Transfer Office and Roche (Cetus' successor-in-interest to agreements between Stanford and Cetus and the PCR technology) began negotiations regarding a license to the patents, but the University sued when Roche refused to agree to a license.  Stanford sued Roche for infringement based on Roche's sale of kits for PCR detection of HIV RNA for assessing the efficacy of antiretroviral therapy.  Roche raised its purported ownership rights as well as "shop rights" to the claimed invention as affirmative defenses, as a declaratory judgment counterclaim, and procedurally regarding Stanford's standing to bring the lawsuit.

The District Court ruled on summary judgment that Roche's assertion of ownership rights were properly viewed as a counterclaim subject to and barred by the California statute of limitations (which had expired before Roche raised the counterclaim).  The Court also ruled that Roche did not have a license to the technology because it failed to obtain consent from the patent holder, Stanford, and that Roche did not have shop rights in the invention.  The District Court granted Roche's summary judgment motion that the asserted claims of all three patents were invalid for obviousness.

The Federal Circuit issued a decision vacating the District Court's determinations in a decision by Judge Linn, joined by Judges Prost and Moore.  The Federal Circuit affirmed the District Court's determination that Roche's ownership claims were barred by the California statute of limitations.  But the CAFC vacated the remainder of the District Court's decision, based on its conclusion that Stanford did not have standing to sue.

The Federal Circuit first dismissed Stanford's objections to Roche's defenses and counterclaims on procedural grounds, finding critically that Roche had raised a question of standing that could properly be raised at any time during a litigation and that was not foreclosed by statutes of limitations.  Thus, the CAFC considered it error for the District Court to have refused to consider Roche's ownership counterclaims below.  Instead of returning the matter to the District Court, however, the Federal Circuit decided the question (in Roche's favor) in its opinion.

In its decision, the panel went right to the heart of the matter:  that Inventor Holodniy had made agreements to assign his rights to multiple parties.  Here, however, the Court made an important distinction based on the language of the CPA Inventor Holodniy signed with Stanford.  The agreement stated that Inventor Holodniy "agrees to assign" all inventions that he first "conceives or first actually reduces to practice" under the research grants to the University.  The Court held that this was not "an immediate transfer of expectant interests" (i.e., a present assignment of future inventions), but "a mere promise to assign rights in the future"; the Court's conclusions were bolstered by further portions of the CPA that required Inventor Holodniy to assign his rights "at an undetermined [future] time."  Thus, the panel held that Inventor Holodniy retained rights in his invention at the time he made the further agreement with Cetus; if the Court had ruled otherwise, then Inventor Holodniy would have transferred all his rights to the University prior to signing the agreement with Cetus, and would have had no rights to assign.

In contrast, the VCA Inventor Holodniy signed with Cetus contained the language required by the Federal Circuit to effect an immediate transfer of rights:  "I will assign and do hereby assign to CETUS, my right, title and interest in each of these ideas, inventions and improvements" (emphasis in original).  With the filing of patent applications naming Inventor Holodniy as an inventor, the Court said, equitable title conferred upon Cetus by Inventor Holodniy's VCA converted "by operation of law" to legal title in the "ideas, inventions and improvements" disclosed in the patent applications that issued ultimately as the patents-in-suit.  This negated Inventor Holodniy's subsequent assignment of his rights to these inventions to Stanford, because he no longer had any legal rights to assign (Stanford's rights to the patents presumably derive from the interests of the other named inventors).

The Federal Circuit also refuted Stanford's assertion that Holodniy's later assignment placed them in the position of a "bona fide purchaser for value" with regarding to Cetus' (now Roche's) interests in the patents.  The Court found that Stanford had at least constructive notice of Cetus' interests, if only as the result of the terms of several Materials Transfer Agreements between the University and the company.  In addition, the Court relied upon common-law agency principles that attribute knowledge to the employer possessed by the employee.  (Interestingly, the Court ignored any agency relationship between Inventor Holodniy and the University with regard to failure of the inventor to obtain Stanford's permission or consent to assign his rights to Cetus, since (according to the Court) "Holodniy signed away his individual rights as an inventor, not Stanford's, while performing work for Stanford after promising to assign his rights to the university" (emphasis in original).)

Finally, the Court also rejected Stanford's argument that Inventor Holodniy's assignment to Cetus was voided by the University's rights to federally-funded inventions under the Bayh-Dole Act (35 U.S.C. § 202(d)):

We are unconvinced of Stanford's interpretation of the Bayh-Dole Act in this case.  Stanford identifies no authorities or reasons why its election of title under Bayh-Dole had the power to void any prior, otherwise valid assignments of patent rights.  Stanford was entitled to claim whatever rights were still available after the Government declined to exercise its option, including the rights of co-inventors Merigan, Katzenstein, and Kozal.  However, Holodniy transferred his rights to Cetus more than six years before Stanford formally notified the Government of its election of title.  As previously noted, Stanford's invention rights policy "allow[ed] all rights to remain with the inventor if possible," . . . which supports the conclusion that Holodniy still possessed rights at the time he signed the VCA with Cetus.  Just as we explained that Bayh-Dole does not automatically void ab initio the inventors' rights in government-funded inventions, . . . we see no reason why the Act voids prior contractual transfers of rights.

(However, in footnote 1 the Court noted that "[w]e express no opinion as to whether Holodniy's execution of the VCA violated any provisions of the Bayh-Dole Act, or whether the Act provides the Government or Stanford some other legal recourse to recover Holodniy's rights.")

Because the Federal Circuit affirmed Roche's rights to the patents-in-suit, Stanford lost standing to sue under the principle that all co-owners must be joined in an infringement lawsuit.  Despite Roche's loss of its ability to assert an ownership interest in the patents due to the actions of the statute of limitations, its rights to the patents obtained through Inventor Holodniy's assignment agreement were not extinguished and deprived Stanford of standing to assert infringement against Roche.  And because Stanford lacked standing to bring the suit, the District Court lacked jurisdiction to invalidate the asserted patents claims for obviousness, which judgment the Federal Circuit also vacated.

While the outcome is not surprising and indeed is consistent with well-established patent law principles, wherein rights to an invention arise first in the inventor and can be freely alienated by the inventor (absent any agreement, such as an employment agreement, to the contrary), it is clear that this outcome is contrary in spirit to the intent of the Bayh-Dole regime.  It illustrates the necessity for universities to provide more than merely agreements to be signed by its researchers.  Rather, there is a critical need for information regarding the importance of understanding the interplay of university and corporate interests in inventions that its researchers may develop.  This is particularly true for junior researchers like Inventor Holodniy, who no doubt was more interested in gaining access to the reagents and techniques that would further his research interests than to any commercial applications of the technology.  These are not illegitimate concerns; indeed, it is preferable for university researchers to be focused on their primary goal, advancing knowledge through original research, than on the practical applications of that research.  However, in view of the different motivations and interests of academia and industry, it behooves technology transfer office personnel to have the capacity and authority to at least monitor activities of university research employees when they interact with or obtain technology owned by corporate sponsors or collaborators.  This remains a current problem, and one whose importance is demonstrated pointedly by Stanford's loss of standing (and licensing revenue) against Roche in this case.

Board of Trustees of the Leland Stanford Junior University v. Roche Molecular Systems, Inc. (Fed. Cir. 2009)
Panel: Circuit Judhes Linn, Prost, and Moore
Opinion by Circuit Judge Linn

Comments

[EG]

"it is clear that this outcome is contrary in spirit to the intent of the Bayh-Dole regime."

Kevin,

That's an understatement. I've talked to others in the know and they agree with me that this decision makes a mockery of the statutory scheme in Bayh-Dole:

First, this ruling misinterprets how title to “subject inventions” works under Bayh-Dole, and especially where title to the “subject invention” initially resides. Under Bayh-Dole, title to the “subject inventions” (i.e., those resulting from federally sponsored research which this Case appears to involve) initially resides with the organization (e.g., Stanford) which carries out the sponsored research. It’s only when the organization elects not to retain title that such title may then pass to the U.S., and only if the appropriate federal agency specifically requests that title pass to it. As the Federal Circuit correctly held in Central Admixture, title is potentially “voidable,” not void, if the organization doesn't choose to retain title, and the appropriate federal agency also fails to request title back.

Second, this ruling is inconsistent with 35 USC § 202(d). 35 USC § 202(d) says that if the university elects not to retain title, the university may request that the appropriate federal agency allow the inventor to retain rights in the “subject invention.” If 35 USC § 202(d) is to make sense, initial title to the “subject invention” must reside with the university under Bayh-Dole.

Third, this ruling appears to frustrate what is intended by 35 USC § 202(c)(7). 35 USC §202(c)(7) expressly prohibits a nonprofit organization, like Stanford, from assigning rights in federally sponsored research to third parties (like Cetus or Roche) without approval of the respective federal agency. That a university researcher (like Holodniy) can so easily defeat the prohibition of 35 USC § 202(c)(7) by signing a VCA that incidentally assign rights to “subject inventions” to third parties seems bizarre.

Fourth, this ruling cannot be so easily reconciled with the Federal Circuit's holding in Central Admixture. In essence, this ruling allows a "third party" (e.g., Roche) to challenge and “void” title to a “subject invention.” That is exactly what the Federal Circuit said in Central Admixture cannot be done. Instead, as was held Central Admixture, only the appropriate federal agency may “void” such title in appropriate circumstances under Bayh-Dole. (In footnote 1, while expressing “no opinion,” Judge Linn does appear to suggest that Bayh-Dole might allow Stanford to recover these rights if the Holodniy’s execution of the Cetus VCA violated Bayh-Dole, so this may not be the end of the story.)

Fifth, Judge Linn’s opinion views Stanford’s argument based on Bayh-Dole as requiring the “voiding” of the VCA assignment by Holodniy of the “subject invention” to Cetus. While Stanford could have phrased its Bayh-Dole argument more artfully in this regard, “voiding” of Holodniy’s assignment to Cetus under the VCA is not what is required by Bayh-Dole. Instead, Holodniy had no title to assign because, by operation of Bayh-Dole, title to the “subject invention” started and remained with Stanford. In fact, Judge Linn’s reliance upon the Townsend case is misplaced because the Eastern District of Tennessee in Townsend also got the issue of “where the title resides” wrong (i.e., in electing to retain title, the University of Pittsburgh was not “exercising its right to title” because it already had title by operation of Bayh-Dole).

Admittedly, Stanford might have avoided this result by phrasing the assignment in its researcher's CPA as a “present transfer of,” not a “promise to,” assign rights in future inventions to Stanford. But allowing such subtle phrasing in an assignment to determine whether or not a university retains title to “subject inventions” from federally-sponsored research, as Judge Linn’s opinion in Stanford University does, appears to frustrate how title to “subject inventions” was intended to work under Bayh-Dole, and especially who has title to those “subject inventions” in the first place. I suspect that the Stanford researcher didn't read the incidental assignment clause in the Cetus VCA, and certainly didn't understand what that assignment clause meant.

[Kevin E. Noonan]

Dear EG:

I strive for understatement in an overstated world.

But seriously, I think the issue is whether Inventor Holodniy had anything to assign to Cetus when he signed the VCA. I understand your position, that Bayh-Dole vests title in the institution for all work performed under a Federally-funded research program. I suspect the complicating factor is that Inventor Holdoniy did some of the work on Cetus premises using Cetus reagents and Cetus methods, which were important in achieving the claimed inventions. The status of that work then becomes the issue, and whether the Bayh-Dole umbrella extends that far. There are issues no matter how you come down on that question: if no, you get the result here. If yes, you provide a disincentive for collaborative work between university and corporate researchers, work that usually (albeit not here it appears) is necessary to commercialize inventions initially made at the university benchtop. This is in addition to the more general disincentive regarding Bayh-Dole "march-in" rights, which while they have never been exercised still exist (particularly with regard to inventions directed to public health matters such as AIDS).

I agree that the legal basis for the decision was hair-splitting by the court as to the meaning of the assignment terms in the CPA and the VCA (no surprise that the company got it right, is there?). The dismissal of the Bayh-Dole aspect is where Stanford needs to direct any efforts for rehearing or rehearing en banc, or legislative efforts to make it clear that inventors like Dr. Holodniy have no rights to assign for projects funded from the Federal government through the university.

Thanks for the comment.

[EG]

Kevin,

"Hair-splitting" isn't the half of it. One key fact that is missing or is at least unclear from the opinion is when Stanford got the fed funding for the research. All that the opinion says is this: "Stanford received government funding for its HIV research through the National Institutes of Health (“NIH”). If the date of the fed funding is after Holodniy signed the Cetus VCA and did his PCR work at Cetus, Linn's opinion might hold up, especially if the invention was complete during the PCR work at Cetus. But if the the fed funding occurred before, then the assignment from Holodniy to Stanford under the CPA takes on a whole different meaning because of Bayh-Dole. In other words, you can't read this assignment in the CPA in the normal manner of a "promise to assign," but as confirming that title in the "subject inventions" (whenever they were created) would go to Stanford. Even without the formal CPA assignment, if the the fed funding occurred before the Cetus VCA/work, Bayh-Dole should allow Stanford to trump on the title as Holodniy had no title he could transfer in any future "subject inventions" if those "subject inventions" based wholly (or partly) on the fed funding.

[Looks correct to me]

EG, something must cause title to the invention to be transfered from the inventor. Like an assignment. What instrument exactly effects transfer of title under your view of Bayh-Dole? I assume it is the research grant? And what if a particular inventor was not a party to the research grant? Are you saying that the act of expending grant money effects a transfer of invention rights by operation of law? Just because of the "spirit" of Bayh-Dole? I would think such a legal transer would require more than spirit. And besides, as Kevin hinted, the spirit of Bayh-Dole is to get ideas commercialized. Good luck with that if companies get all of their inventions sucked back to the university if they even breathe on grant money during collaborations.

PS: I don't understand why you are alarmed about the difference between a present and future transfer. That distinction has been in the caselaw for quite some time now. And it is a legally relevant difference that contract drafters may want to rely on--it should not be undermined.

[Kevin E. Noonan]

Dear EG:

The earliest priority date for the patents-in-suit is May 1992. The claims all recite methods for using PCR-based detection of HIV RNA in blood samples as a way to assess antiretroviral drug efficacy. So I think it reasonable to assume that the claims were not reduced to practice until after the work done at Cetus.

The problem is that this argument cuts both ways. If the invention was conceived at the time the work was being done at Cetus (thus giving Inventor Holodniy rights in an invention that could be assigned to Cetus), then the invention should have been sufficiently definite that the University's rights under Bayh-Dole should have attached. And the converse also applies.

Also, remember that the court ruled that the Cetus VCA assigned the rights per se (as opposed to the Stanford CPA which was an agreement to assign future rights). Thus, Holodniy could assign to Cetus rights in inventions limited to those he possessed at the time, and arguably not future inventions.

Thanks for the comment.

[Michael Finney]

Dear All,

I realise that this thread is now "old" but given the Supreme Court's interest in hearing this, I thought I would re-open the discussion if I may.

I have limited knowledge of US law, however, as a lawyer from a common law country, I thought I would ask what seems to me an obvious question.

The key Q seems to be whether Holodniy owned rights which he purportedly assigned to Cetus in the VCA. If so, the courts opinion seems to follow, if not, then Cetus acquired nothing and Stanford hold an undivided title to the 3 patents.

Leaving aside arguments on Bahr-Dole and the effectiveness of Stanford's CPA (agreement to assign future rights at some time), I wanted to know the position at common law. That is, even without reliance upon Bahr-Dole or the CPA, why didn't Stanford argue the common law position - that it holds title to the invention as it was an invention reduced to practice by an employee in the course of their employment duties?

I guess my Q is what is the common law position in the USA? Although reliance upon the common law is not the optimal way to regulate ownership, amongst other things, would this not be a way for Stanford to assert ownership? In the UK and Australia, such an argument may well be successful.

[Kevin E. Noonan]

Dear Michael:

It isn't a common law question. Congress has the power to grant patents to inventors - indeed, even in the corporate world, the inventor is the applicant, not the corporation. The problem the Supreme Court will address is whether Bayh-Dole changed this for university inventors. Does the inventor have the right until she assigns, or does the university have the right ab initio?

Should be interesting.

Thanks for the comment.

Thanks for the comment.

[Michael Finney]

Kevin:

Yes, I agree with your analysis and what will be argued in the Supreme Court. And yes, very interesting.

My question however is why Stanford didn't run another argument originally (in addition to the construction of Bayh-Dole) in which they asserted ownership at common law citing Supreme Court authorities Standard Parts Co. v. Peck, 264 U.S. 706 (1933), and Speck v. North Carolina dairy Foundation, 307 S.E. 2d 785 (N.C. 1983) rev'd 319 S.E.2d 139 (1984)?

I would have thought running an argument that Holodniy was "employed to invent" may have been persuasive and worth a go.

If that common law argument was successful, Stanford would have held the ownership rights in priority to Cetus. Cetus would never have acquired any ownership and would presumably have lost the patent infringement action.

If I am right, there must have been a good reason not to run the common law aregument. I may be wrong however as to the common law position in the US.

Michael