Patenting a correlation between two variables?

http://arstechnica.com/tech-policy/n...al-patents.ars

Quote:

---

The case focuses on a patent that covers the concept of adjusting the dosage of a drug, thiopurine, based on the concentration of a particular chemical (called a metabolite) in the patient's blood. The patent does not cover the drug itself—that patent expired years ago—nor does it cover any specific machine or procedure for measuring the metabolite level. Rather, it covers the idea that particular levels of the chemical "indicate a need" to raise or lower the drug dosage.

---

What the fucking fuck

It sounds bad, but these things are almost always misreported in my experience. Does anyone have a link to the actual patent? I didn't see one in the article, and I can't search for them.

TBH I'm not sure about this one. There are lots of method patents out there (e.g. a way of making X that's much more efficient/cheaper/faster/whatever), and I'm not sure this is all that different. Many method patents rely on pretty basic underlying science, but they aren't just lying around for the taking - someone needs to have a good idea plus a lot of hard work to develop a new method. Is this really patenting the correlation itself, or the method of determining dosage?

In general, I'd say method patents are legal, but should be only carefully awarded. In this specific case, I think it may be in society's best interest to provide some exceptions for medical uses in particular, though managing such exceptions would undoubtedly be complex. IANAL, though, nor have I looked up this specific case outside of your link.

BTW, Wraith, looking at the actual patent is probably not very helpful. Every patent on the planet has claims that are ridiculously broad.

Quote:

---

Originally Posted by wiggin

BTW, Wraith, looking at the actual patent is probably not very helpful. Every patent on the planet has claims that are ridiculously broad.

---

That's the thing though - they're always reported as ridiculously broad, but most of the time when I actually look at the patent itself, it turns out to be extremely narrow. Most of my experience is in software patents though, so maybe it's different it medical patents.

Yeah, that's probably it. I've looked at a lot of medical device and drug patents, and they often claim the world - every concentration imaginable, similar but untested chemistries, extremely broad applications, etc. It's kinda normal to do this as a way to protect your core idea with a wall of other patents. Then the lawyers chip away at them as much as possible if a competitor wants to use the same space.

Quote:

---

Originally Posted by Wraith

It sounds bad, but these things are almost always misreported in my experience. Does anyone have a link to the actual patent? I didn't see one in the article, and I can't search for them.

---

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnet ahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=6,355,62 3.PN.&OS=PN/6,355,623&RS=PN/6,355,623

And I read the claims. It's really as bad as the article makes it out to be.

The summary of the patent: "If A below B, increase dosage of C - if A above B, decrease dosage of C".

Laughable. If they allow patenting that then every doctor becomes a patent law breaker. "You may have illness A. Let's see if I licensed the proper patents to be able to determine whether something in your blood levels needs adjusting."

I sometimes wonder why stuff like that is even allowed to see a court. Probably due to the equally idiotic notion of software patents which is one the same moronic level.

I don't think that patent link is correct (even with the VB tag edited out) - it leads to something about some mechanical tools.

edit: Here we go. At issue are http://patents.com/us-6335623.html, covering an MRI component or something, and http://patents.com/us-6680302.html, which is what's specifically being talked about. Reading the later now.

edit 2: So my reading of it is that this is actually a method of determining whether and how to adjust dosage. They figure it out by checking for certain chemicals in the blood to determine metabolization rate. They're basically patenting this process when treating for certain diseases and using chemical signs they've discovered, with floors and ceilings set by research they've conducted. It's about how to tell if dosage is too high or too low, not just saying "if dosage is too low, raise it." They're trying to collect a reward for figuring out that the metabolites are important, and what their values should be.

There's also stuff in there about age and gender, but as far as I could tell from my skimming, it was only supporting that their numbers are correct across a large number of cases. It can be hard reading technical talk mixed in with legalese, so Wiggin can correct me if I'm wrong and it actually has any bearing on the patent's breadth.

There's NO method in the actual patent. It really is only "If A below B, increase dosage of C - if A above B, decrease dosage of C".

Nowhere in the whole patent is a new "how" - they're only using already known diagnostic instruments (HPLC in this case).

Again, this is moronic. If they can actually patent this, expect a whole new slew of medical patents which patent every diagnostic process under the sun. As if you guys did not have an inefficient medical system before. This will make it worse.

Just don't tell anyone and it'll be okay!

http://www.nytimes.com/2012/03/21/bu...osages.html?hp

Finally, some common sense in all this patent idiocy. Now, if they could only throw software patents out of the window as well...

Only concern here is the incentive to try different methods will be different.