Slashdot Mirror
What Bilski Means For Biotech Patents
eldavojohn writes "Patents aren't just a software thing, and while Bilski's dismissal didn't shake the ground for software, it's certainly making waves in the biotech community. You may recall Prometheus v. Mayo, in which doctors fought a biotech startup's methodology patents. Well, medical method patents are now being reconsidered by order of the Supreme Court. Stocks of biotech startups jumped as this news broke, but questions remain on how the lower Federal Circuit court will rule when it reconsiders these cases of medical testing. It's clear the Supreme Court has 'ruled that judges should be more flexible in determining if methods, rather than objects, are eligible for patents, citing emerging technologies such as medical testing.' So Bilski may result in dire news for medical methods and testing patents."
The SCOTUS really screwed the pooch here. Considering how the oral arguments went, I was very surprised at the opinion they ended up releasing.
Living With a Nerd
I thought that a patent was still permitted if (1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing. It's hard to believe that a drug or gene couldn't meet those qualifications if a piece of software can.
Well, first, the Supreme Court held that the machine-or-transformation test, which you recited, is only a tool or clue to the patentability of a method, not the sole test. Second, the diagnostic patents being discussed here generally take this form:
"A method for treating disease X, comprising performing test A, observing the result, and administering treatment foo, bar, or baz depending on the result."
The issue here is that in many of these patents test A and treatments foo, bar, and baz are all known in the art. The invention is the discovery of the association or correlation between test results and the optimal treatment (e.g., if you have a certain gene then you get this drug, but if you lack it you should get this other drug). The Federal Circuit has hinted that it is the observing or correlating step that must pass the machine-or-transformation test; (possibly) the testing and (more likely) the treatment steps are merely 'insignificant postsolution activity' that can't rescue the patent from unpatentability.
In my opinion, however, such patents should be granted so long as they are new, useful, nonobvious, and adequately specified in the patent. Subject matter is too crude a tool to filter out undesirable patents. As a society we want investment into new diagnostic methods and personalized medicine. For example, the availability of fast, inexpensive genetic testing has opened up new doors to making sure that people are given the best drug at the best dose, but determining which genes match which drugs and doses will require significant studies. We can encourage investment in those kinds of studies by offering patent protection to the resulting diagnostic and therapeutic methods.
Full Disclosure: I worked with the team that wrote the amicus brief for Dr. Ananda Chakrabarty (of Diamond v. Chakrabarty fame) in the Bilski case.
IAAL. When SCOTUS takes a patent case, they're like a bull in a china shop. They're not engineers or patent lawyers, so they undo a lot of stuff that has been carefully worked out by the Patent Office and special patent courts. That said, the area of non-device patents has grown enormously over the last few years, and there was a great need to set some ground-rules to cut back on overreaching claims while giving people with new ideas a chance to make a profit from their innovations. In my view, Bilski is a monster (and necessary) bitch slap for methodology patents, which had gotten out of bounds. (IMO, the next one will come on copyright overreaching.) The Supremes deliberately left the opinion vague to let the experts work out the details.
Why do you think the methods would be unavailable to 'other doctors or hospitals?' The patent holders would want to maximize revenue by making the methods available to as many licensees as possible.
The number of hospitals willing to pay the licensing fee will probably be lower than the number of hospitals that would practice these methods if they weren't patented.
And remember, of course, that the US government has 'march-in rights' to any NIH-funded invention
Which only applies if the NIH funds it. My personal opinion is that nothing that has significant government funding should be eligible for a patent.
This is my signature. There are many like it, but this one is mine.
Sounds like there is prior art for this patent. No?
There are (basically) two ways to invalidate a patent based on prior art: anticipation and obviousness. Anticipation requires that a single piece of prior art (e.g., another patent, a paper in a journal, a web page, etc) describe each and every element of the claimed invention. The prior art must also enable someone to use the invention, which is why a dictionary isn't prior art for every invention imaginable; although a dictionary technically contains all the right words, they're not in a useful order. Clearly the method you describe would not anticipate another therapeutic method unless it was either identical to or a subset of the method you've described.
The other prior art-based invalidation is obviousness, which allows the combination of multiple pieces of prior art that would each by themselves be incomplete. Since the KSR decision, obviousness can also be based on a certain degree of common sense or common ingenuity rather than explicit teachings documented in evidence. So, the method you describe might be a part of a larger body of prior art that could establish the obviousness of a similar therapeutic method. For example, it could be combined with prior art describing a drug similar to levothyroxine to make obvious a method that was the same except for substituting that other drug for levothyroxine. After all, a doctor or researcher having ordinary skill in the art would probably find it obvious that a drug with a similar pharmacokinetic profile to levothyroxine could have its dosage optimized in the same way.
What it would not do, however, is make obvious all therapeutic methods based on measuring a substance, observing the level, and adjusting treatment because the nonobvious part is discovering exactly what the critical levels are and what the appropriate dose response is. Nothing about the method you described would suggest what the optimal level of, say, azathioprine is for someone with rheumatoid arthritis or how much to adjust the dose by given a certain level of azathioprine or its metabolites in the blood.
Why do you think the methods would be unavailable to 'other doctors or hospitals?' The patent holders would want to maximize revenue by making the methods available to as many licensees as possible.
Licensing to as many as possoble may very vell be less profitable then licensing to a few who can pay really good. Thats why 7 miljon people in Africa don't get the medicine against AIDS that they need - they can't afford it.
The pharmaceuticals are making more money selling the AIDS medicine to those that can pay a high price then letting lots of people that really need it get it for a price they all can afford.
So no, its often more profitable to sell a medicine to a select few and let the rest die.
The technical term is the supply and demand curve in micro economics. (google it)
Just saying it like it are.