A DNA Patenting Thought Experiment

    By Kevin Noonan

Much has been written in the comments to my response to Michael Crichton's recent OpEd piece in The New York Times.  In order to open up the debate to those who haven't read the comments, here is a thought experiment:

You isolate a petrochemical with excellent lubrication qualities.  It is used in very expensive machinery so you can charge a lot of money for it.

You isolate Vitamin B12 from beef muscle, and use it as a cure for anemia.  You charge a lot of money for it, because anemia kills people.

You isolate a new drug, penicillin, from a mixture of naturally-produced chemicals made by a mold.  You use it to cure, among other things, syphillis, which otherwise kills people.  You charge a lot of money for it.

You isolate erythropoietin from urine and use it for kidney dialysis patients, who otherwise die of anemia.  You charge a lot of money for it.

You isolate a gene responsible for Gaucher's disease, a lipid storage disease, and use the gene to make the missing protein.  It is efficiently gobbled up by macrophages in the affected childrens' blood, which cures the disease that otherwise kills them at a young age.  You charge a lot of money for it.

Now, if there is a logical position why the last of these is different from the rest, we would like to hear it.  And if the problem devolves into objections to inventors getting to charge a lot of money for cures for fatal diseases, or any other reason, then we would like to hear about that, too.

Comments

[Duff Samoa]

The 4 initial examples you give are different from the last example. In the 3 biological examples you are talking about a product which directly has use to cure a particular ailment, whereas in the fourth example you are talking about a gene which is used to make a product which directly cures an ailment.

I think a much better example for comparison would be isolating a particular growth factor directly from say, urine, which is then used to induce growth in an industrial bacterial strain which produces a life saving protein at a greater rate than without the growth factor.

[Kevin E. Noonan]

Dear Duff:

Your distinction is valid, but what I meant was, is there anything patentably different - that the gene would be unpatentable.

You can think of the gene as an intermediate, or catalyst, that produces the protein.

Thanks for the comment

[Duff Samoa]

I agree with you Kevin in that isolated genes should be allowed to be patented for the intended purpose given in the example you wrote.

I think though that you have worded your question above in such a way that it doesn't cover the more contentious applications of gene patents which seem to upset people. Now I hope I am correct in my understanding of patent law but why should a company be allowed to patent a/the disease causing allele/s of a gene and then charge someone money in order for them to work out their own sequence of the same gene to see if they are carrying one of the diseased alleles?

I don't think it's fair if I want to pay a company to sequence a region of my own genome that they should have to pay companies who own the rights to the genes in the region I wish to sequence. It seems unfair that I do not essentially have a right to know my own DNA sequence because companies have patented my genes. It's equivalent to getting a urine test for drugs and making the company doing the test pay money to those companies who own the patents to the drugs in my system. Do you think that should be allowed? if so, why?

[Kevin E. Noonan]

Dear Duff:

I think it might depend on the claim. If the claim is:

A method of diagnosing disease X, by identifying mutation Y in a human's chromosomal DNA,

then you might be infringing under your scenario. However, what is really the point is that you aren't doing it, but a company is selling a kit or service where a doctor takes a blood sample, and the company does the detecting. This is what the patent is there to protect: commercialization of the test.

This is equivalent to any other medical test where a physician can't identify the problem by merely looking at you. If I invent an MRI machine, couldn't you make the same argument - why don't you have the right to know what's happening in your liver, so why do you have to pay $1000 for the test (that helps pay the cost of the patent machine) instead of $100 (which pays the cost of the technician, and electricity, etc.).

Genetic information isn't any different than any other personal information that you need a test to detect. And remember, the clock is ticking - the term of these patents will expire in our lifetimes, and the genetic information will be available forever.

As for the urine example, that would probably fall under the principle of exhaustion - the drug companies were paid their royalties when you bought the drugs. But if the company doing the test had it patented, then they could charge insurance companies and other more for the urine test that otherwise.

Thanks for the comment.

[Duff Samoa]

Thanks for your reply Kevin.

I have a little trouble with using your MRI exam example as an analogy to a gene test. I extended it to cover the problem a little better, I hope it fits correctly.

So, Thinking about this a little further, I imagine an example with closer relevance to "a diagnostic test for disease X by mutation Y" would be if it was found that a certain known metal led to brighter images in an MRI, hence better diagnosis of a liver disease (e.g. cancer). For me, I have no problem paying a company for the use of the MRI machine and it's associated costs, the cost of the technician and also the cost of the doctor to interpret my results. What I'm stuck on is whether or not it's fair to also pay a royalty to the company that owns the patent for use of the known metal in the MRI exam process. I guess thinking about it further, I must say I think it would be only fair for the company to be paid a royalty as they would have spent money on the research in order to come to their result.

With this in mind, I guess it is harder for me to justify my initial position that a company shouldn't be allowed to patent the mutant alleles which are known to lead to disease X. I guess, just like the company researching the use of the known metal in an MRI, the company researching the genetic link of mutation Y with disease X should be able to get some restitution for their discovery.

I am curious though, if it turns out that another disease, disease Z, is caused by the same mutation, where does the company that holds the inital patent stand? Especially if the reason that someone can have diseae X but not disease Z or vice versa is due to the combination of the remaining genes in their genome.

As well, based on the fact that a patent may be awarded for "diagnosis of disease X from mutation Y", is there any problem with a company providing you with just the sequence of the gene in question but no diagnosis? Would you then be able to take your now known sequence and compare it to the diseased alleles in the database yourself? Or in order to do that (comparing the sequence yourself) would you need to pay a fee to the company who owns the diagnostic test for mutation Y, even if you are doing it for a non-commercial purpose?

[Kevin E, Noonan]

Dear Duff:

If it turns out that another disease, disease Z, is caused by the same mutation, and this is not disclosed in the original patent, then the only way for the company that holds the inital patent to have a basis for an infringement suit would be if the mutation was claimed without reference to disease X, something that happens rarely if at all (although it would be good to check). With regard to the combination aspect, many diseases will be multivariate and that will make it murkier, but the claim "diagnosing disease X with mutation Y" will be infringed if you perform the test for that diagnosis but not for diagnosis of disease Z.

Your second scenario points to a consequence of the informational content of DNA and the database age. If someone were to provide you with the sequence of a known gene, particularly in the region of a known disease-causing mutation, but provide you with nothing else, could they avoid liability? The more this is done with a wink, or with a booklet describing known mutations, or with a website to such information, the more the company could be liable for inducing infringement (by you). But again, it would be the company as deep pocket who would be sued (although Donald Trump should probably stay away from "at-home" gene diagnoses like the one you describe). The one-off type of situation you posit wouldn't show up on the radar, but I can envision a pregnancy-type test that would be used by a consumer but would cause this type of inducing infringement liability.

Thanks for the comment.

[smount]

I don't know that much about patent law (I'm trained as a geneticist, not as a lawyer), but I agree with Crichton that patents should not be granted for things that are in no way inventions. My impression is that in the examples given above patents were awarded for the use of a natural substance, not for all possible uses of that substance. Most gene patents were awarded prior to knowledge of the disease associations and prior to development of the tests being used. Why should a company that owns a patent based solely on the sequence of a gene be granted a licensing fee for a genetic test that was developed after (and independent of) their discovery? The patent-holders are generally NOT the people who develop the tests (except insofar as others have been persuaded not to) and the gene, its mutant alleles, and natural variants would be (and generally have been) discovered in identical form by anyone who looked.

I like the analogy to patenting an element. Putting aside the issue of expiration, does it make sense that Lavoisier, who discovered silicon as an elemental component of flint, and had no idea of electronic computers, should collect fees based on its application to semiconductors?

[Kevin E, Noonan]

Smount:

The problem is that Dr. Crichton knows about as much patent law as you do, and yet you agree with his pronouncements on what should be patentable.

The gene/element dichotomy is simply incorrect. Elements are viewed as being unpatentable for historical reasons and are different from genes in more ways than they are similar. My post/Thought Experiment sets out the framework: genes are no different than other natural products that are certainly patentable, and it is the isolation that makes them so (just as a vitamin or a protein is patentable - there is not real distinction).

In fact, if inventor 1 identified a gene and inventor 2 indentifies a disease-associated mutatnt, each can get a patent on their inventions, and neither can practice the other's without a license. That is the patent system for everything else under the sun made by man, where "made" is equivalent to "isolated." A belief that genes shouldn't be patented would be more persuasive if you could distinguish genes from things like proteins, but there is no rational distinction - they are chemicals, and like any other chemical they can be patented.

Your idea that the association between a gene and a disease is also false - although it is known there is a genetic component, the history of the past 20 years has been identifying which gene is associated with which disease, and how.

Lavoisier couldn't patent silicon, but he could have patented a silicon computer chip.

Thanks for the comment.

[Jeff Steck]

Given the analogy to patenting an element, it is worth pointing out claim 1 of U.S. Patent No. 3,156,523: "1. Element 95."