Genes and Information: The Problem of Disease-specific Databases

By Kevin E. Noonan --

One of the many untruthful positions taken by the ACLU in the AMP v. Myriad Genetics case was that DNA is merely information, like a computer program, and as such Myriad's patent claims were invalid as reciting patent-ineligible subject matter.  The ACLU also alleged (in counts of their declaratory judgment complaint legally avoided by the District Court, the Federal Circuit, and the Supreme Court) that claims reciting methods for determining the sequence of a woman's BRCA genes were a violation of the First and Fourteenth Amendments as interfering with a woman's right to her genetic information.  And the strategy was effective:  for example, the District Court distinguished DNA as subject matter from all other biologically derived materials (such as antibiotics or antibodies) because DNA is the "physical embodiment of genetic information," and Justice Thomas stated in the Court's opinion that "[i]t is undisputed that Myriad did not create or alter any of the genetic information encoded in the BRCA1 and BRCA2 genes" and "[w]e merely hold that genes and the information they encode are not patent eligible under §101 simply because they have been isolated from the surrounding genetic material."

DNA is not information, of course; it is "but a chemical compound, albeit a complex one."  While the informational content of the molecule determines its chemical structure, the sequence information is not encompassed by the patent claims, and the information can be used without infringement of the isolated DNA claims (either those invalidated or those upheld by the Court's opinion).  However, the diagnostic method claims determine the genetic information of a patient's BRCA genes, and one of the enduring conundrums of the Myriad patent situation is the fate of the genetic information discerned by Myriad in the exclusive practice of their patented methods (putting aside the issue of whether these claims are patent-eligible, either specifically or as a genus).  Patient privacy prevents Myriad from releasing that information to anyone other than the patient's physician, genetic counselor, or the patient herself, and laws such as Health Insurance Portability and Accountability Act (HIPAA) and Genetic Information Non-disclosure Act (GINA) exist to prevent unauthorized disclosure.  But collectively, the genetic information from the plurality of women tested by Myriad (and anyone else doing such testing) produces a higher order, population genetic picture of the mutations responsible for increased risk of cancer that has its own value and in which the public has an interest.

Recently, Myriad's decision (since 2004) to maintain this collective genetic information in a proprietary database has become another battleground in the debate on how best to promote progress in the genetic diagnostic method arts.  Groups are calling for a grass roots effort by individual women to submit their genetic information to a common (albeit privately held) database in an effort to reconstitute Myriad's database (see "Consortium Launches Public Database of BRCA Data" and "Myriad Genetic Database Under Siege").  This information, while a product of the patented methods is not limited to them and is not subject to the disclosure requirements imposed as part of the quid pro quo of patent protection.  This leads to the anomaly that Myriad's patents will expire, permitting competition by others using Myriad's primers, probes, and methods, while Myriad will continue to enjoy a competitive advantage due to its proprietary database.  The consequences are readily envisioned -- a doctor will be faced with two possibilities:  one that a patient has a published BRCA gene mutation so that a test by either Myriad or a lower-cost competitor will provide the same diagnostic information, or that the patient has a mutation whose significance is known only by consideration of the information in Myriad's database.  Faced with this choice, the doctor will either have her patient (or the insurance company or governmental payor) incur a first cost (from the competitor) and a second cost (from Myriad) or one cost (from Myriad).  Depending on the frequency of informative mutations in Myriad's database but not otherwise known, the cost-benefit analysis may (and likely will) tip in favor of using Myriad's test, either preferentially or exclusively.  And this will lead to a de facto extension of the competitive advantage Myriad has enjoyed via its patents for the past twenty years.

This is an important policy issue but limiting patents for genetic diagnostic testing will, paradoxically, exacerbate rather than solve this problem.  The absence of patent protection will provide strong incentives for all genetic diagnostic information to be undisclosed, kept as a trade secret and otherwise not shared with patients, their doctors, or genetic researchers.  With patent protection, the policy-based obligation to disclose the diagnostically relevant mutations arises from the patent grant itself and the quid pro quo of disclosure in exchange for such exclusivity.  Abolishing the exclusivity would abolish the obligation, and would provide incentives for companies to disclose as little information as possible.  Instead of identifying the mutations, for example, future (unpatented) genetic diagnostic methods could come merely with the information that a patient has a genetic mutation associated with an increased risk of a particular disease, with statistics on survival and treatment options and their success rates but no information on which mutation(s) were involved.  While this outcome would be even more inconsistent with the way genetic diagnostics is and has been developing than Myriad's behavior bemoaned by many, what patients and their doctors want most are reliable diagnostic results, and a company offering such advanced diagnostics would be in a position to adopt a "take it or leave it" attitude regarding disclosure.  This possibility is even more likely for the majority of diseases that, unlike breast cancer and BRCA are multivariate and less readily reverse engineered.  Solving this conundrum is important, in ways proponents of the ACLU's lawsuit have not provided and which the recent "victory" over Myriad's "gene patents" does not achieve.

The key to a solution may be to strengthen, not weaken, patent protection for genetic diagnostic methods.  This is because the exclusivity of patent protection provides a single actor motivated to take advantage of the limited exclusivity period and provides the return on investment making development of the testing possible.  Because the government provides this exclusivity and the profits stemming from it, the public nature of the patent grant may justify the government imposing conditions for the practice of these patented methods.  One of the reasons the DOJ took the somewhat ridiculous position it did in the Myriad case, and that the Obama Administration has been antagonistic to patenting generally, is that the Federal government is the largest payor for drugs and (now or in future) genetic diagnostic testing, which should (ultimately) be beneficial to ameliorating increased medical care costs due to the potential for early treatment or other interventions that are less costly than a patient developing breast cancer or other disease with a genetic component.  And perhaps the leverage that stems from government payment for genetic diagnostic testing is the one that could be used to prevent the anomaly of a company having a valid patent that produces genetic information but that can extend at least a portion of its exclusivity past the expiration date of the patent by accumulating patent-ineligible information in a proprietary database.  The government could, presumably, require that the genetic information for every patient produced by the practice of a patented genetic diagnostic method be deposited (in a HIPAA- and GINA-compliant way) with the NIH or other (public) agency, and that database would be freely available after the patent has expired, been found invalid or unenforceable or licensed.  A precedent exists for patented cells, antibodies and other biological material:  the ATCC has a patent depository that requires patentees to agree to maintain their deposit of patented material for a term (30 years) even longer than the patent term, and will give requestors samples (subject to notification of the patentee) once the patent has granted.  (As a non-government agency the ATCC is in no position to monitor whether a requestor is entitled to the sample, but notice to the patentee permits the patent owner to pursue legal remedies against any requestor who obtains the deposit sample improperly.)

Such a scheme is more consistent with achieving a balance between the benefits of disclosure provided by the patent system and the needs of the public for effective disclosure of disease-related genetic information than all the litigations, gene patent-banning bills, Congressionally mandated (and still to be released) "second opinion" studies, and speculative conclusions of blue-ribbon commissions that have led us to where we now are -- a situation where less, not more, genetic information disclosure is likely to be in our future.

Hat tip to Jennifer Gottwald and Victoria Sutton for discussions on this topic.

Comments

[Dan Feigelson]

"While the informational content of the [DNA] molecule determines its chemical structure..."

Kevin, I think it's the other way around.

[Thomas]

Interesting article. But it's not an anomaly to have patent and trade secret rights (or even other copyright data rights) in a technology area for a particular business. The anomaly only relates to the overlap, if any. And patent misuse does not appear to be an decided issue here, does it?

At least because some patents were still upheld or not appealed, right??

[Kevin E. Noonan]

Dear Dan:

I see your point philosophically, but in this argument you can think of it either way (and certainly the way the post states it is relevant for synthetic oligomers used in the practice of the genetic diagnostic methods).

Thanks for the comment.

[Michael]

Very interesting article, as always. Thanks.

Re the ATCC, I had looked into this briefly awhile ago but couldn't find the answer quickly and didn't have time to look further--their policy seems to be that samples will be given out once the patent has granted, but in some instances it seems they will only give out samples that are covered *by the claims.* That is, if something is disclosed in the specification but not covered by the patent claims, the ATCC may not send it out. But I could not find a legal basis for either position (e.g., sending things out that were covered by the claims versus disclosed in the specification)--all the papers I saw referred to the Budapest Treaty, but that did not seem to answer the question. Nor was the legal connection between the ATCC and the requirements of the Budapest Treaty clear. If anyone knows the answer (or about an article discussing it), I would love to hear it.

[Townhouse23]

The issue could be solved relatively simply, by accepting that each person has a right to the information stored in their own genome, and to do with it as they see fit. Thus whilst there would remain a proprietary right in a compiled database of SNPs the company would be obliged to provide the a patient with a copy of any sequence data obtained from their genome. This could then be submitted to any public database they chose.

The result of this would be that the sequencing company would be entitled to the IP that they had generated (the database), but the effect of this right would be fairly offset, by the right of the patient to have their own sequence.

In effect a fair natural balance of rights, that would prevent unreasonable exploitation of a dominant IP position.

[RosalindF]

"The key to a solution may be to strengthen, not weaken, patent protection for genetic diagnostic methods."

Keep digging, Kevin! Maybe you can get Congress to overturn Prometheus and allow patents on methods of thinking new thoughts about old results.

"here we now are -- a situation where less, not more, genetic information disclosure is likely to be in our future."

You have no evidence to support this prediction and, quite frankly, it's surely wrong. Less genetic information in our future? Seriously? Maybe you know something about an apocalyptic nuclear war that is coming in the next couple years because, barring that, the genetic information explosion is going to continue with or without patents protecting that information or the use of that information to "determine the likelihood" of future events.

If anything, without patents that cover (1) the genetic information itself or (2) any attempt to use conventional techniques to acquire that information, there is going to be more information about human genetics than anyone is going to know what to do with. This is true not just for genetic information, Kevin, but for all kinds of information, and applies equally to all the patents that are drafted to prevent such information from being used by others.

[RosalindF]

"what patients and their doctors want most are reliable diagnostic results, and a company offering such advanced diagnostics would be in a position to adopt a "take it or leave it" attitude regarding disclosure"

Indeed. And Myriad is free to do exactly that. But Myriad is not free to escape criticism from the public, of course. And such criticism is inevitable when life and death are at stake.

Maybe Myriad should threaten to destroy its "proprietary database" if it doesn't get its way with the offensive (=revolting) positions it takes towards people who wish to practice old conventional sequencing methods on regions of the genome which Myriad seems to believe that it owns (how else can they make the arguments they make about the primers that people can use to perform old sequencing methods?). You know, Republican party tactics. That'll win Myriad lots of public approval, right? Because it'll show everybody that they won't be "disrespected."

[Kevin E. Noonan]

Dear RosalindF:

Well, I agree that Myriad is not a warm and fuzzy company, but they aren't the AntiChrist, either. And I don't think they have threatened to destroy their proprietary database, just profit from it.

You seem to miss the point that the piece is meant to discuss changes in the way things have been done up until now, to prevent (or at least disincentive) the construction of a proprietary genetic database based on patents to genetic diagnostic methods. Recall that the Myriad Court was very careful not to write a broad brush opinion negating patent protection for all DNA or genetic diagnostic methods. And Mayo, for all its sloppy language was concerned with a situation where everything in the claim except the correlation was in the prior art. So there is little support for the position that either case, or a combination, will preclude patenting genetic diagnostic methods under the right circumstances (acknowledging that we don't yet know what the right circumstances will be).

But if we permit patenting genetic diagnostic methods we will need to address the proprietary database problem, and the post was directed towards one possible avenue for doing so.

As for the explosion of genetic information, while that is undoubtedly true what is not so clear is that there will be an explosion of useful genetic information. And while the past would suggest such progress may continue, the posited situation is different from the past, if we assume genetic diagnostic methods will not be eligible for patenting.

But the wonderful thing about discussing the future is that it provides a natural experiment. We shall see and then know who among us, if any, had the correct intuition.

[Kevin E. Noonan]

Dear Town:

You describe the current situation, which you must admit is not only inefficient but substitutes one proprietary database for another - after all, while patients may be induced to submit their genetic information to the "shadow" database, it is only when that information is complied into the collective that it becomes valuable, and no individual has any right to all the information, just their by itself insignificant piece of it.

I think something more is needed. One way to address the issue is to ban genetic diagnotic patents - a bad idea, in my view. The piece proposes another alternative, which has advantages (the principle one being it doesn't involve a ban) and disadvantages (since it involves government coercion in the patent context). But in toto the good outweighs the bad, I think, which is why I proposed it.

Thanks for the comment.

[Skeptical]

Dr. Noonan,

I admire your patience in dealing with the recurring infection known by his original name of Malcolm Mooney, and here appearing as RosalindF.

Get well soon.

[RosalindF]

Kevin: " the wonderful thing about discussing the future is that it provides a natural experiment. We shall see and then know who among us, if any, had the correct intuition."

That requires that you actually put your cards on the table and make a prediction that is specific and testable.

Here's my prediction: the noose will tighten on claims like Myriad's that restrict access to genetic information. Myriad's claims, and claims like Myriad's, restrict access to genetic information because they prevent skilled artisans from using old methods to study aspects of a person's genome, all of which are useful (in whole or part or combination) at least for the purpose of providing a partial or complete "genetic ID" of that person.

As a result of this, at least as much information about variety in the human genome and "what it means" will become available (i.e., readily accessible) to the public more quickly than it otherwise would. That's because there will be no restrictions on what sections of ones genome one can study (individually or in combination) using conventional (i.e., unpatentable or off patent) methods and/or compositions.

In spite of any dicta in the Supreme Court's incomprehensible Myriad decision (which you seem to take comfort in for some reason), the straightforward logic in Prometheus is going to control. In short, you can't protect ineligible subject matter (whether it's a "correlation" or "information about a person's genome" or "DNA molecules with sequences in the human genome") merely by folding in some obvious or old subject matter. That's because people have an uncontestable right to use old, obvious or conventional methods to study any piece of human DNA or combinations thereof for any purpose whatsoever. If I wish to do that and your patent puts up a roadblock in my path, then your patent is ineligible. That's Prometheus (with help from Myriad).

By the way, I noticed that cert was denied by the Supreme Court where the Federal Circuit had found a biodiagnostic method ineligible in Intema Ltd. v. PerkinElmer, Inc. That was predictable, and correct, in view of Prometheus. Just another data point for you.

[Gary Johnston]

Kevin:

As you know, Myriad cannot and will not prevail in its lawsuits. They are sham lawsuits, and any reasonably knowledgeable patent attorney with an understanding of biotechnology knows this (even if he/she will not openly admit this due to, for example, fear of loss of income). Unfortunately, Myriad does have an advantage over competitors due to its database, but the advantage will erode as more samples are tested by competitors. The days of obtaining patents that seek to make infringers out of those who practice old methods and then consider or think about the results of those methods, are over.

So, we don't need to enact legislation one way or the other in response to Prometheus or Myriad. The only thing we really need to do is move on with our lives.