On Wednesday, we reported on an article, entitled "The dangers of diagnostic monopolies," that appeared in the March 26th edition of Nature (see "Genetic Diagnostic Testing: The 'Anti-Commons' Revisited?"). Today, we received an e-mail from one of the authors of the article, Robert Cook-Deegan (at right), providing some additional comments regarding the article:
This is a very thoughtful take on our article. I disagree with much of it, but not for the reasons you might predict. I think if you read through the case studies that underlie this summary commentary in today's Nature, you'll find the reason for specifying the "problems" as we did.
My main point here: the problems we specify are identified as such by the companies, health professionals who order the tests, and those who get tested -- everyone says they want to avoid them, and the companies say their informal policies are to, for example, permit unfettered basic and clinical research, and to ensure that everyone gets a second opinion and has their test paid for by their insurance or health plan. If you're prepared to defend patent rights when they cause these problems -- and sometimes they do -- then that's where we can have productive disagreement.
Two other key points where patents in this area differ -- and here I really do urge you to read the case studies and see if you agree. The word "monopoly" is not a scare word when it describes the reality. The companies in these case studies -- Athena, Myriad, and Clinical Data/PGxHealth -- do have effective monopolies for testing for some of the conditions (and not others, the point of comparisons); that is their business plan. It is completely legal, and as the case studies make plain, it's not a slam dunk for either the pro-patent camp or the anti-patent camp (and believe me, we'll get shot at from the other side too). The tests have been on the market for over ten years in many cases, and in no case has someone "worked around" the key patents when those center on DNA sequences for the consensus sequence or the most common mutations associated with disease. Myriad has been in effect the sole provider in the US. (One little aside -- the reason BRCA testing options in Europe are different are because Myriad's patent position has been much weaker there -- so I didn't quite follow your train of argument on that point.) So it is simply misleading to say such patents do not lead to monopoly effects. Sometimes they do; and sometimes they don't (other cases also covered in the case studies -- see comparison to colon). When they do, the problems we note can arise.
The peculiar dynamics of genetic testing explain why patents on one or a few common mutations can dominate testing for an entire condition caused by many different genes or different mutations in the same gene. An example: if you have long-QT syndrome, you don't know which mutation or even which gene might be causing it. You send your sample for testing to find that out. Until recently, PGxHealth had a monopoly for such testing by having exclusive rights to Utah patents for some of the most common mutations associated with long-QT. It was the only place you could send samples to get tested for those and other mutations. BioReference Labs Inc. has now secured countervailing exclusive rights on other mutations, so that monopoly probably won't continue. It seems to be an instance of mutually blocking patents -- because neither lab can know in advance whether they are infringing the others' patents until they find out which mutations that particular patient has. Until BRLI acquired contervailing exclusive rights, no one else was offering testing because PGxHealth had the exclusive rights to common mutations. Through exclusive rights on the common mutations, PGxHealth has gotten all the business for those mutations as well as patients with long-QT caused by unpatented variants and mutations in undiscovered genes. This is not necessarily bad (arguments on both sides), and it is unlikely to be catastrophic -- the companies can cross-license or find some other accommodation. But the situation until recently is fairly and accurately described as a monopoly, and a monopoly not only on the patented inventions but extending to other causes of the same clinical condition. So I don't agree there is an inherent contradiction in our logic; monopolies can indeed form (but they can also prove unstable in the face of countervailing exclusive rights).
The point about the necessity of the patent incentive is particular to this use of patents, not patents in general. We do not make these arguments in therapeutics, vaccines, or instruments. The evidence that a patent incentive is not needed comes from reference labs and academic labs that already offer a test when the exclusively licensed test hits the market. In that case, it is tautological that the patent incentive was not needed. The role of patents in that case is to shut down competitors. That is, of course, a common use of patents and a reason they exist and how they often work; but it is nonetheless evidence that the patent was not needed to create the product or service. This is not Pfizer keeping Lilly and Merck from making Viagra or Lipitor. These are commercial labs eliminating testing by reference labs and academic health centers. We are careful not to judge whether that is a good or bad thing -- there are plenty of arguments on both sides -- but the social benefit of patents in those cases is not that they create tests that would otherwise never come to be.
I do think you're on thin ice and won't be very persuasive if you rest your case on "just live with it" arguments about access to health goods and services. Here I stick to my guns. If we can specify problems that could be solved by weakening monopoly rights a little at the margin when they conflict with patient access, then we should do so. At root, this rests on a premise that patents are instrumental rights balancing net social benefit, not an inherent right, and so constraining them is fair game when doing so promotes social benefit.
But thanks for being first out of the gate in rekindling a debate on this. Really. This is great. And thanks for keeping it civil.
My main point here: the problems we specify are identified as such by the companies, health professionals who order the tests, and those who get tested -- everyone says they want to avoid them, and the companies say their informal policies are to, for example, permit unfettered basic and clinical research, and to ensure that everyone gets a second opinion and has their test paid for by their insurance or health plan. If you're prepared to defend patent rights when they cause these problems -- and sometimes they do -- then that's where we can have productive disagreement.
Two other key points where patents in this area differ -- and here I really do urge you to read the case studies and see if you agree. The word "monopoly" is not a scare word when it describes the reality. The companies in these case studies -- Athena, Myriad, and Clinical Data/PGxHealth -- do have effective monopolies for testing for some of the conditions (and not others, the point of comparisons); that is their business plan. It is completely legal, and as the case studies make plain, it's not a slam dunk for either the pro-patent camp or the anti-patent camp (and believe me, we'll get shot at from the other side too). The tests have been on the market for over ten years in many cases, and in no case has someone "worked around" the key patents when those center on DNA sequences for the consensus sequence or the most common mutations associated with disease. Myriad has been in effect the sole provider in the US. (One little aside -- the reason BRCA testing options in Europe are different are because Myriad's patent position has been much weaker there -- so I didn't quite follow your train of argument on that point.) So it is simply misleading to say such patents do not lead to monopoly effects. Sometimes they do; and sometimes they don't (other cases also covered in the case studies -- see comparison to colon). When they do, the problems we note can arise.
The peculiar dynamics of genetic testing explain why patents on one or a few common mutations can dominate testing for an entire condition caused by many different genes or different mutations in the same gene. An example: if you have long-QT syndrome, you don't know which mutation or even which gene might be causing it. You send your sample for testing to find that out. Until recently, PGxHealth had a monopoly for such testing by having exclusive rights to Utah patents for some of the most common mutations associated with long-QT. It was the only place you could send samples to get tested for those and other mutations. BioReference Labs Inc. has now secured countervailing exclusive rights on other mutations, so that monopoly probably won't continue. It seems to be an instance of mutually blocking patents -- because neither lab can know in advance whether they are infringing the others' patents until they find out which mutations that particular patient has. Until BRLI acquired contervailing exclusive rights, no one else was offering testing because PGxHealth had the exclusive rights to common mutations. Through exclusive rights on the common mutations, PGxHealth has gotten all the business for those mutations as well as patients with long-QT caused by unpatented variants and mutations in undiscovered genes. This is not necessarily bad (arguments on both sides), and it is unlikely to be catastrophic -- the companies can cross-license or find some other accommodation. But the situation until recently is fairly and accurately described as a monopoly, and a monopoly not only on the patented inventions but extending to other causes of the same clinical condition. So I don't agree there is an inherent contradiction in our logic; monopolies can indeed form (but they can also prove unstable in the face of countervailing exclusive rights).
The point about the necessity of the patent incentive is particular to this use of patents, not patents in general. We do not make these arguments in therapeutics, vaccines, or instruments. The evidence that a patent incentive is not needed comes from reference labs and academic labs that already offer a test when the exclusively licensed test hits the market. In that case, it is tautological that the patent incentive was not needed. The role of patents in that case is to shut down competitors. That is, of course, a common use of patents and a reason they exist and how they often work; but it is nonetheless evidence that the patent was not needed to create the product or service. This is not Pfizer keeping Lilly and Merck from making Viagra or Lipitor. These are commercial labs eliminating testing by reference labs and academic health centers. We are careful not to judge whether that is a good or bad thing -- there are plenty of arguments on both sides -- but the social benefit of patents in those cases is not that they create tests that would otherwise never come to be.
I do think you're on thin ice and won't be very persuasive if you rest your case on "just live with it" arguments about access to health goods and services. Here I stick to my guns. If we can specify problems that could be solved by weakening monopoly rights a little at the margin when they conflict with patient access, then we should do so. At root, this rests on a premise that patents are instrumental rights balancing net social benefit, not an inherent right, and so constraining them is fair game when doing so promotes social benefit.
But thanks for being first out of the gate in rekindling a debate on this. Really. This is great. And thanks for keeping it civil.
Kevin Noonan (at right), of course, provides a response:
I don't doubt that there are circumstances and situations where the problems you outline in your commentary could occur. My critique stems from my understanding that the evidence strongly suggests that the problems do not (at present) exist.
As you mention, I have not reviewed the underlying case studies, and we will blog on the HHS White Paper in a future post. But turning to what I do understand, there are good reasons to believe that the "problems" you outline are not problems at present. For example, if companies permit basic and clinical research, or ensure that "everyone" gets a second opinion, then the informality of the policies does not trump the consequences -- exactly what you propose be done under threat of patent unenforceability. If there is an example of a company that has sued a university researcher engaged in basic research, or has refused a patient from getting a second opinion, by all means identify it and let's talk about how to solve the problem. Revising patent law isn't the only option -- good corporate citizenship seems to be working effectively to prevent the types of predation your commentary warns against.
My reaction to using the term "monopoly" is several-fold. First, even if not intended, it has a negative connotation. If that is your opinion, fine, but I will react to it based on its use in anti-patent polemic. It is also inaccurate because patents by definition have only a limited right -- to exclude others from making, using, selling, offering to sell or importing the patented article or method in the U.S. during the term of the patent. It does not prevent someone from designing around the patent, and the patent by definition discloses the identity of the gene involved in disease pathology. All this information is fair game for a competitor to look for other, unpatented ways to achieve the same diagnosis. You are right that the exclusivity is important for diagnostic companies' business plans -- as I mentioned in my piece, the cost of validating these tests so that they can be reliably used with actual human patients is typically much higher than the cost of the original genetic observation, and without patent exclusivity no one would take the risk on that type of investment. The fact that no one has "worked around" the patents you cite is, as you much recognize, a very small sample for interrogating a very big question: it is possible that the earliest genetic diagnostic tests are just robust enough to be hard to design around. But these tests are ten years old -- how long until they are "off patent" and then fair game for competitors to use for free in perpetuity?
The situation you described between PGxHealth and Bioreference Labs is precisely that which the patent system envisions -- the development of blocking patents between competitors that require cross-licensing. But as you admit, any "monopoly" is both unstable and short-lived -- the quid pro quo of having the companies disclose their inventions so they can be practiced freely after the patents expire. The alternatives -- and they are potentially myriad -- are instances where the underlying basis for the diagnostic assay is undisclosed as a trade secret. In that case (which weakening the patent system would encourage) there would be no disclosure and thus no end to the (non-patent based) monopoly.
I don't intend to rely on "just live with it" arguments. I merely contend that you identify a host of theoretical or potential problems that, even if they existed would, at their worst, be inherently time-limited and associated with positive benefits for the public. These benefits include disclosure of the underlying disease-associated polymorphisms, development of the genetic information into robust diagnostic tests, and incentives for others to design-around the patented tests. I disagree that your proposals would weaken patents at the margins -- the proposals in your commentary would impose your concepts of morality and public benefit -- patient rights -- for mine, or others. There are certainly instances where that may be appropriate; the Federal government has "march-in" rights for any test developed with Federal grant money, as well as an inherent right to license any patent to any third party for its own use (with compensation through the Court of Claims). These rights have not been lightly exercised, for good reason, but they exist for instances when they are necessary. But I think there is little need to impinge on patent rights that underlie investment in genetic diagnostics to solve problems -- such as patient access and insurance coverage -- better addressed by other routes.
Thanks for your response.
As you mention, I have not reviewed the underlying case studies, and we will blog on the HHS White Paper in a future post. But turning to what I do understand, there are good reasons to believe that the "problems" you outline are not problems at present. For example, if companies permit basic and clinical research, or ensure that "everyone" gets a second opinion, then the informality of the policies does not trump the consequences -- exactly what you propose be done under threat of patent unenforceability. If there is an example of a company that has sued a university researcher engaged in basic research, or has refused a patient from getting a second opinion, by all means identify it and let's talk about how to solve the problem. Revising patent law isn't the only option -- good corporate citizenship seems to be working effectively to prevent the types of predation your commentary warns against.
My reaction to using the term "monopoly" is several-fold. First, even if not intended, it has a negative connotation. If that is your opinion, fine, but I will react to it based on its use in anti-patent polemic. It is also inaccurate because patents by definition have only a limited right -- to exclude others from making, using, selling, offering to sell or importing the patented article or method in the U.S. during the term of the patent. It does not prevent someone from designing around the patent, and the patent by definition discloses the identity of the gene involved in disease pathology. All this information is fair game for a competitor to look for other, unpatented ways to achieve the same diagnosis. You are right that the exclusivity is important for diagnostic companies' business plans -- as I mentioned in my piece, the cost of validating these tests so that they can be reliably used with actual human patients is typically much higher than the cost of the original genetic observation, and without patent exclusivity no one would take the risk on that type of investment. The fact that no one has "worked around" the patents you cite is, as you much recognize, a very small sample for interrogating a very big question: it is possible that the earliest genetic diagnostic tests are just robust enough to be hard to design around. But these tests are ten years old -- how long until they are "off patent" and then fair game for competitors to use for free in perpetuity?
The situation you described between PGxHealth and Bioreference Labs is precisely that which the patent system envisions -- the development of blocking patents between competitors that require cross-licensing. But as you admit, any "monopoly" is both unstable and short-lived -- the quid pro quo of having the companies disclose their inventions so they can be practiced freely after the patents expire. The alternatives -- and they are potentially myriad -- are instances where the underlying basis for the diagnostic assay is undisclosed as a trade secret. In that case (which weakening the patent system would encourage) there would be no disclosure and thus no end to the (non-patent based) monopoly.
I don't intend to rely on "just live with it" arguments. I merely contend that you identify a host of theoretical or potential problems that, even if they existed would, at their worst, be inherently time-limited and associated with positive benefits for the public. These benefits include disclosure of the underlying disease-associated polymorphisms, development of the genetic information into robust diagnostic tests, and incentives for others to design-around the patented tests. I disagree that your proposals would weaken patents at the margins -- the proposals in your commentary would impose your concepts of morality and public benefit -- patient rights -- for mine, or others. There are certainly instances where that may be appropriate; the Federal government has "march-in" rights for any test developed with Federal grant money, as well as an inherent right to license any patent to any third party for its own use (with compensation through the Court of Claims). These rights have not been lightly exercised, for good reason, but they exist for instances when they are necessary. But I think there is little need to impinge on patent rights that underlie investment in genetic diagnostics to solve problems -- such as patient access and insurance coverage -- better addressed by other routes.
Thanks for your response.
Noonan: 1
Cook-Deegan: 0
Posted by: Pacific Reporter | March 29, 2009 at 12:30 AM
Kevin "I disagree that your proposals would weaken patents at the margins -- the proposals in your commentary would impose your concepts of morality and public benefit -- patient rights -- for mine, or others."
This is rather confused. First of all, we do live in a representative democracy. Nobody, not even Cook-Deegan, is proposing a dictatorship whereby Cook-Deegan's views are substituted for our own. If anything, the morality of those who presently rely on the patent system for monetary gain is very much imposed on everyone, like it or not, and that is not likely to change in our lifetime. You protest too much, Kevin.
Also, it is not clear why Cook-Deegan's proposals would have much effect other than weakening the patent system "at the margins" to achieve the results that Cook-Deegan deems are preferred. Do you believe, Kevin, that efforts to identify genetic tests for disease will cease if Cook-Deegan's proposals are implemented? If so, why do you believe that?
Posted by: I Lurv Freedim | March 30, 2009 at 07:59 PM
Dear Lurv:
I try not to protest too much, but I also don't think the patent system is imposed on us (any more than anything else the government does is imposed - you take a confusing position yourself for someone reminding me that we live in a representative democracy).
I don't think Mr. Cook-Deegan's proposals act at the margins - they would preclude patent enforceability unless a patentee complied with his ideas of how they should behave. I think those behaviors are socially desirable, but I don't think they should be enforced by putting the patent right at risk.
While I don't think that research into genetic tests will cease if Mr. Cook-Deegan's proposals were implemented, I do think they would not be developed into robust clinical tests. This is because it takes a great deal to establish the reliability of such tests before we permit them to form the basis for patient or physician decisions about treatment. This is absolutely how it should be - we don't want un-verified tests to be used for such important decisions.
But the fact is that there must be some exclusivity involved to justify the investment risk. One way to do this is patents and another is trade secret. The former has the advantage that the method is disclosed (fully, in the U.S. due to the best mode requirement), so that the rest of us get the unfettered benefit of the technology once the patent expires. I think Mr. Cook-Deegan's proposal could drive the technology into trade secret protection, and then the public would not benefit.
So ultimately I am trying to take the long view - short-term exclusivity to promote funding of test development, followed by unrestricted use after patent expiry. I think Mr. Cook-Deegan's approach is short-sighted, because it imposes additional costs that could (and I expect would) inhibit investment in these technologies.
All the research in the world cannot substitute for an actual clinical test. I think that's what everyone wants; we just disagree on how to get there.
Thanks for the comment.
Posted by: Kevin E. Noonan | March 30, 2009 at 11:40 PM
Kevin:
Would you be in favor of a marketing exclusivity scheme for new clinical diagnostics? Right now the laws are such that drugs which are new chemical entities are entitled to a period of marketing exclusivity. This typically doesn't mean very much in the context of drugs because the patents on the drugs extend well beyond any such exclusivity. But this would seem to solve the incentive problem for development of new clinical diagnostics, no? Obviously the compromise is that such diagnostics would need to be subject to a more rigorous approval process than they are now, but is this so terrible? Wouldn't it help to weed out all the junk tests that are out there? I understand that this is complete heresy to many diagnostics companies, but it certainly appears that the days of obtaining and enforcing broad process patents for biomarker correlations are probably over.
Posted by: Gary Johnston from Team America | April 01, 2009 at 08:57 PM
Dear Gary:
An interesting suggestion. It is another way to get to the same place - through regulation - and may be an attractive alternative should the Bilski/Breyer/Classen wing of patent jurisprudence prevail.
Thanks for the comment.
Posted by: Kevin E. Noonan | April 01, 2009 at 11:33 PM