Applying In re Bilski to Diagnostic Method Claims

    By Kevin E. Noonan --

An astute reader has asked, regarding our post concerning EXACT Sciences' announcement of the issuance of U.S. Patent No. 7,485,420, whether the claims of the '420 patent would be invalid under the Federal Circuit's In re Bilski standard for patentable subject matter.  The independent claims of the '420 patent are:

1.  A method for detecting the likelihood that a human patient has colon neoplasia, comprising:
    a)  obtaining a sample from a human patient; and
    b)  assaying said sample for the presence or absence of methylation within a nucleotide sequence selected from the group consisting of SEQ ID NOs: 40-45,
    wherein the presence of methylation within any one of said nucleotide sequence is indicative of the likelihood that the human patient has colon neoplasia.

13.  A method for detecting the likelihood that a human subject has colon neoplasia, comprising detecting vimentin protein or nucleic acid expression level in a sample from the human subject, wherein reduced expression level of vimentin protein or nucleic acid relative to a control sample from a healthy subject is indicative of the likelihood that the human subject has colon neoplasia.

17.  A method for monitoring over time colon neoplasia in a human subject, comprising:
    a)  detecting the methylation status of a vimentin nucleotide sequence selected from the group consisting of SEQ ID NOs: 40-45 in a sample from the human subject for a first time; and
    b)  detecting the methylation status of any one of said vimentin nucleotide sequence of step (a) in a sample from the same human subject at a later time;
    wherein absence of methylation in the vimentin nucleotide sequence taken at a later time and presence of methylation in the vimentin nucleotide sequence taken at the first time is indicative of regression of colon neoplasia; wherein presence of methylation in the vimentin nucleotide sequence taken at a later time and absence of methylation in the vimentin nucleotide sequence taken at the first time is indicative of progression of colon neoplasia.

Since the Supreme Court has not yet spoken on the issue, the standard for patentability of process (method) claims is the one announced by the en banc Federal Circuit in Bilski:  for a process to be eligible for patenting (i.e., to be deemed patentable subject matter under 35 U.S.C. § 101), it must be claimed to include one or the other of these elements:  (1) it is tied to a particular machine or apparatus, or (2) it transforms a particular article into a different state or thing.

The invention in Bilski was a method for hedging risk in commodities trading, subject matter seemingly far-removed from diagnostic claims such as EXACT's colorectal cancer detection method. 

However, the Federal Circuit has had the occasion, since its Bilski decision, to consider the patentability of claims more similar to EXACT's claims than those invalidated in Bilski.  In Classen Immunotherapeutics, Inc. v. Biogen Idec, the CAFC affirmed a district court decision invalidating the claims in Classen's U.S. Patent No. 5,723,283:

1.  A method of determining whether an immunization schedule affects the incidence or severity of a chronic immune-mediated disorder in a treatment group of mammals, relative to a control group of mammals, which comprises immunizing mammals in the treatment group of mammals with one or more doses of one or more immunogens, according to said immunization schedule, and comparing the incidence, prevalence, frequency or severity of said chronic immune-mediated disorder or the level of a marker of such a disorder, in the treatment group, with that in the control group.

The Federal Circuit in Classen made short work of any analysis of how Bilski's patentability rubrics apply to Classen's claims:  the Court's decision read in its entirety:

In light of our decision in In re Bilski, 545 F.3d 943 (Fed. Cir. 2008) (en banc), we affirm the district court's grant of summary judgment that these claims are invalid under 35 U.S.C. § 101.  Dr. Classen's claims are neither "tied to a particular machine or apparatus" nor do they "transform[] a particular article into a different state or thing."  Bilski, 545 F.3d at 954.  Therefore we affirm.

There are distinctions that can be drawn, of course, between Classen's claims and EXACT's claims, one of which is that the Classen claims are directed to an immunization schedule, and expressly recite that the results of the immunization procedure be "compared" between an experimental and a control group.  This "comparing" step implicates the deficiency found by three Supreme Court justices (Breyer, Souter, and Stevens) in their dissent from dismissal of the appeal in the Laboratory Corp. v. Metabolite Labs., Inc. (LabCorp) case.  There, the claim at issue read as follows:

13.  A method for detecting a deficiency of cobalamin or folate in warm-blooded animals comprising the steps of:  assaying a body fluid for an elevated level of total homocysteine; and correlating an elevated level of total homocysteine in said body fluid with a deficiency of cobalamin or folate.

In their dissent, the Justices opined:

The researchers who obtained the present patent found that an elevated level of homocysteine in a warm-blooded animal is correlated with folate and cobalamin deficiencies.  As construed by the Federal Circuit, claim 13 provides those researchers with control over doctors' efforts to use that correlation to diagnose vitamin deficiencies in a patient.  Does the law permit such protection or does claim 13, in the circumstances, amount to an invalid effort to patent a "phenomenon of nature"?  I concede that the category of non-patentable "phenomena of nature," like the categories of "mental processes," and "abstract intellectual concepts," is not easy to define.  See Flook, supra, at 589 ("The line between a patentable 'process' and an unpatentable 'principle' is not always clear"); cf. Nichols, 45 F. 2d, at 122 ("[W]e are as aware as anyone that the line [between copyrighted material and non-copyrightable ideas], wherever it is drawn, will seem arbitrary").  After all, many a patentable invention rests upon its inventor's knowledge of natural phenomena; many "process" patents seek to make abstract intellectual concepts workably concrete; and all conscious human action involves a mental process.  See generally 1 Chisum §1.03, at 78–295.  Nor can one easily use such abstract categories directly to distinguish instances of likely beneficial, from likely harmful, forms of protection.  Cf. FTC, To Promote Innovation: The Proper Balance of Competition and Patent Law and Policy, ch. 3, p. 1 (Oct. 2003) (hereinafter FTC) (collecting evidence that "issues of fixed cost recovery, alternative appropriability mechanisms, and relationships between initial and follow-on innovation" vary by industry); Burk & Lemley, Policy Levers in Patent Law, 89 Va. L. Rev. 1575, 1577–1589 (2003) ("Recent evidence has demonstrated that this complex relationship [between patents and innovation] is . . . industry-specific at each stage of the patent process").

But this case is not at the boundary.  It does not require us to consider the precise scope of the "natural phenomenon" doctrine or any other difficult issue.  In my view, claim 13 is invalid no matter how narrowly one reasonably interprets that doctrine.

                                                *    *    *

Even were I to assume (purely for argument's sake) that claim 13 meets certain general definitions of process patentability, however, it still fails the one at issue here:  the requirement that it not amount to a simple natural correlation, i.e., a "natural phenomenon."  See Flook, supra, at 588, n. 9 (even assuming patent for improved catalytic converter system meets broad statutory definition of patentable "process," it is invalid under natural phenomenon doctrine); Diehr, 450 U. S., at 184–185 (explaining that, even if patent meets all other requirements, it must meet the natural phenomena requirement as well).

At most, respondents have simply described the natural law at issue in the abstract patent language of a "process."  But they cannot avoid the fact that the process is no more than an instruction to read some numbers in light of medical knowledge.  Cf. id., at 192 (warning against "allow[ing] a competent draftsman to evade the recognized limitations on the type of subject matter eligible for patent protection").  One might, of course, reduce the "process" to a series of steps, e.g., Step 1: gather data; Step 2: read a number; Step 3: compare the number with the norm; Step 4: act accordingly.  But one can reduce any process to a series of steps.  The question is what those steps embody.  And here, aside from the unpatented test, they embody only the correlation between homocysteine and vitamin deficiency that the researchers uncovered.  In my view, that correlation is an unpatentable "natural phenomenon," and I can find nothing in claim 13 that adds anything more of significance.

The Supreme Court remanded the LabCorp case without reaching a decision, and Chief Justice Roberts took no role in the decision to remand on the grounds that certiorari had been improvidently granted, so there is no way to know whether Justice Breyer's opinion in dissent is truly limited to the Justices who joined him in dissent.  Bilski has petitioned for certiorari, so the Court may address the propriety of the Federal Circuit's application of the Bilski formula to the Classen claims.   

EXACT's claims avoid a direct recitation of the "buzzwords" like "compared" and "correlated," but that may not be enough:  after all, saying that the level of methylation in vimentin gene sequences is "reduced" poses the question "compared to what?" and therein may be the problem.  Of course, "comparison" is required in many claims that, prior to Bilski, would have been thought to be unambiguously patentable, for example, claim 1 of U.S. Patent No. 7,464,002:

A diagnosis method for boiler degradation, comprising:  calculating a first thermal transmission rate from a difference in thermal quantity of gas or steam between an entrance and exit of a heat exchanger with respect to at least one heat exchanger constituting the boiler; calculating a second thermal transmission rate based upon a thermal conductivity rate of pipes and thermal conductivity transfer rate of steam and gas of the heat exchanger; comparing the first and second thermal transmission rates; and carrying out diagnosis of the progress of degradation, wherein the diagnosis is displayed on a display device, wherein the boiler is a heat recovery boiler for recovering heat from a gas turbine.

The trivial answer is that this claim is directed at "diagnosing" a machine, and so fall under the "machine" prong of the test.  But a better reading of Bilski is that the method must be performed by a machine, not on a machine, and so that line of reasoning does not appear to be a valid distinction.  The level of uncertainty produced by the Federal Circuit in Bilski, encouraged or at least catalyzed by Justice Breyer's dissent in LabCorp, is the polar opposite to the avowed purpose of Congress in establishing this specialized appellate court, i.e., to bring greater certainty to patent law than had been achieved by the several regional circuit Courts of Appeal.  It is up to the Supreme Court (if it deigns to) to decide whether to overturn the Federal Circuit's mechanical, bright-line test for patentability in Bilski (which is frankly in direct conflict with the Supreme Court's own view of what is patentable enunciated in Diamond v. Chakrabarty), and thus to follow its decade-long pattern of imposing on the Federal Circuit its conception of the metes and bounds of U.S. Patent Law.

For information regarding this or other related topics, please see:
• "Patent Profile: EXACT Sciences Announces Issuance of Colorectal Cancer Screening Patent," February 4, 2009
• "Classen Immunotherapies, Inc. v. Biogen Idec (Fed. Cir. 2008)," December 22, 2008
• "Personalized medicine: A Patent Office Perspective," December 11, 2008
• "The Relevance of In re Bilski to the Patentability of the Metabolite Claim," November 17, 2008

Comments

[This Isn't Terribly Difficult]

Claim 13 is dead in the water. I just spit in my sink and I noticed that the level of nucleic acid in my spit sample is reduced. That's all it takes because the rest of mental stepping.

Even Kevin Noonan will not stand behind the validity of this claim. I dare you, Kevin. Go ahead. Make my day.

The patentability of the other two claims under 101 depends (and I know people hate to hear this) on whether the data-collecting steps are admittedly old. If the data collecting steps are admittedly old, then the jig is up and what's being claimed is thinking about data collected from an old step. That's not a patentable invention.

But given how recently this patent issued, surely there are good claims that recite significant post-solution steps such as reporting the diagnosis to the patient or testing a group of patients and treating only those patients whose levels were reduced.

Right.

[Kevin E. Noonan]

Dear Terribly:

The point of the piece was to set out where we are in this debate; neither of us can say whether anything is dead or alive, except under the Classen decision (which by definition would sink all process claims unless they satisfy the mechanical, bright line Bilski test).

I disagree (as does Judge Rich, by the way) with the idea that whether something is patentable subject matter or not should be based on whether the data collecting steps are old or not. Read In re Bergy: it's putting the cart before the horse to make the decision on this basis. First decide whether it's patentable subject matter, then decide whether it's novel. You may not like that test, but it was (and I think still is) the law. Moreover, it avoids all the drama about patentable subject matter, since if it isn't novel it isn't patentable - so what's the point?

I also agree that the disconnect between PTO processes CAFC decisions means there are going to be many process patents of dubious validity - but I don't think that's a good thing.

The real question is a policy one: do we want to extend patent protection to medical diagnostic claims or not? The answer may be no, but it's a good idea to consider the consequences. Medical research will go on, much of it government (i.e., taxpayer) funded, but without any recourse to patent protection. The research will be published, since it will be mostly academic research and their concerns are not primarily patent protection. Which means that corporations, foreign and domestic, will be able to freely exploit the fruits of this research - the U.S. taxpayer will thus be funding the R&D efforts of companies. And without patent protection, those companies will be able to cut down on collaborations and funding medical research - a win-win for them.

All of this will be a consequence of a passion your post exhibits - an zeal to characterize patent protection as something pursued by inventors who haven't invented and attorneys without scruples. Such caricatures are fun (if you aren't an inventor or a patent attorney - although a reasonably high proportion of people with this passion are patent attorneys), but besides the point. As Judge Rich said, there is always an inventor - it's just that what she invented may not be patentable. And there are plenty of ways (like "old" data collection methods) to deny patentability to those inventions that should not be patented, without categorically excluding them from being considered as patentable subject matter.

And if you think all this isn't terribly difficult, I don't think you're paying attention.

Thanks for the comment.

[This Isn't Terribly Difficult]

Kevin "The real question is a policy one: do we want to extend patent protection to medical diagnostic claims or not?"

No, that's not the "real" question. The real question was "whether the claims of the '420 patent would be invalid under the Federal Circuit's In re Bilski standard for patentable subject matter". That's a direct quote. I gave my sincere answer to the question.

Your alleged "real" question is very very different. It's also a strawman because I never argued, nor did the court in Bilski, nor did the Supreme Court in LabCorp., nor did the district court or the Federal Circuit in Classen, that claims to diagnostic methods should be unpatentable. Moreover, even if such methods were deemed unpatentable, a novel nonobvious kit for performing the diagnosis would certainly still be patentable (and would encourage their manufacture and development more than a broad patent on the method claim would).

"Medical research will go on, much of it government (i.e., taxpayer) funded, but without any recourse to patent protection."

Again, that's simply false. There are recourses to patent protection still available.

"Which means that corporations, foreign and domestic, will be able to freely exploit the fruits of this research - the U.S. taxpayer will thus be funding the R&D efforts of companies."

If this is perceived to be a real problem, then it is best addressed legislatively. Applauding the issuance of claims to abstract concepts and/or methods of thinking about a correlation observed in nature is not the correct response.

"if it isn't novel it isn't patentable - so what's the point"

That's a good question. If the alleged novelty of an invention resides solely in a mental step, then the invention is not patentable, whether under 101 or 102 or 103. Where's the drama?

[This Isn't Terribly Difficult]

Kevin "neither of us can say whether anything is dead or alive"

We can say whatever we want. I'm not giving legal advice. I'm just expressing my reasonable view on the validity of a government grant.

Let me ask it this way, Kevin. I'm a doctor. I have some patients. I took some samples from my patients, some healthy, some not, before this patent issued. We ran some tests. I didn't check the nucleic acid levels in the samples but we did check some levels of other molecules that, in my experience, correlate perfectly with nucleic acid levels. If I look at those measurements now and see that the levels of some samples are higher in patients who I know are healthy, am I infringing the claim?

Why or why not?

And just to make it a bit more realistic, let's say I own a company whose business it is to diagnose patients and I have a lot of money and you're the patentee and you need money. Am I infringing the claim?

[Kevin E. Noonan]

Dear Terribly:

To respond to the last post first:

If you were a doctor, you would be immune from infringement liability by 35 USC 287(c). So you don't have to worry about it.

To address your hypothetical, if you did not assay the nucleic acid or vimentin, I would argue you were not within the scope of the claim, so again no worries for you.

Whether you are a deep pocket corporation may affect whether the patentee offers to license the patent to you, but where do you think you got all that money? Maybe from reading the published patent and designing around it, and then charging your customers/patients for the test you sell them? The results, after all, are just the result of your "mental step" - but you make money, don't you? Do you think that makes you better than the inventor who actually discovered how to make the diagnosis in the first place? And while you may benefit, and your corporation with a lot of money may benefit, I'm not sure the rest of us do.

Thanks for the comment

[Kevin E. Noonan]

Dear Terribly:

Well, the drama was (and is) certainly in the sarcasm and self-righteousness of your tone. That may be a source of amusement on other blogs, but here, it just gets in the way.

To address the "strawman" charge, first you are correct that kit claims are an alternative, and that may be one way to protect these types of inventions. But they are not, and have not, been the only way. And the Bilski decision has the effect of invalidating claims otherwise relied upon by inventors for many years.

As for dealing with the problem "legislatively," that's been done - under Diamond v. Chakrabarty, the Supreme Court said that Congress had determined that "anything under the sun made by man" was patentable. The Court focused on the word "any" so the idea that Congress must deal with it legislatively is the strawman - it's been done.

If you think the Bilski decision doesn't have the potential to invalidate every diagnostic assay claim, write one that's worth anything - since tying it to a machine will permit practicing without infringing. And while you're at it, tell me why the claim of the '002 patent should be patentable?

Thanks for the comment.

[This Isn't Terribly Difficult]

"To address your hypothetical, if you did not assay the nucleic acid or vimentin, I would argue you were not within the scope of the claim,"

Well, go ahead. Let's hear the argument.

[This Isn't Terribly Difficult]

"If you think the Bilski decision doesn't have the potential to invalidate every diagnostic assay claim, write one that's worth anything - since tying it to a machine will permit practicing without infringing. And while you're at it, tell me why the claim of the '002 patent should be patentable?"

I think communicating of the diagnostic result to a patient who is subsequently treated or to a physician who is treating a patient can be a significant transformation.

Note that this is a quite different transformation from merely displaying a "number" calculated from some other "number". A **properly drafted** method effects the transformation of a sample into a treatment, whereas an improperly drafted claim reads on the mere transformation of information or data into a treatment, or reads on the transformation of a sample into a mental step. Those transformations are not sufficient to meet 101.

Your argument re Diamond v. Diehr is weak. Anything under the sun? Maybe. Any thought under the sun? No way.

[Kevin E. Noonan]

Dear Terribly:

I have some affinity for the "patenting thought" part of the argument. But your solution, which seems to be the PTO's solution (tying the diagnosis to treatment) raises Muni Auction problems. That is, there are no joint tortfeasors in patent law unless one of them is working under the direction and control of the other, and so you could be left with one actor (the diagnostician) and another actor (the treating doctor) and no infringer of a diagnostic treatment claim.

Thinking isn't patentable, of course, but I don't think this is thinking - it's acting, insofar as a diagnosis is more than a thought (it wouldn't be much good if it wasn't translated into action).

Interestingly, Justice Breyer did not base his opinion that the Metabolite claim was not patentable on a mental step; he thought it was merely recognition of a natural phenomenon. But both arguments get to the same place: a correlation between a diagnostic marker and a disease should not be patentable. I agree with that: the correlation per se should not be, but the act of making a diagnosis may be.

I think your argument, about communicating the diagnostic result, may be one way to address the issue, provided that the Supreme Court agrees that this would constitute a "significant transformation."

And, seriously, I can't see how the '002 claim is patentable under Bilski.

Thanks for the comment.

[Kevin E. Noonan]

Dear Terribly:

I missed the "let's hear the argument" comment. If you assay molecule X, which "from your experience" is positively correlated with vimentin gene methylation, you are not directly assaying what is recited in the claim - assaying vimentin gene methylation. So I think you don't infringe the claim, even if the correlation was known. Because the claim says you have to assay vimentin gene methylation.

Thanks for the comment.

[EG]

"At most, respondents have simply described the natural law at issue in the abstract patent language of a "process." But they cannot avoid the fact that the process is no more than an instruction to read some numbers in light of medical knowledge."

Kevin,

That quote from the dissent in LabCorp speaks volumes about the illogic of Justice Breyer when it comes to patent matters, and how it really reflects a complete and utter "anti-patent" bias. It's on par with his comment about "raccoon inventions" in KSR International. The fact is, Claim 13 of LabCorp reflects a diagnostic method (and thus within 35 USC 101), as does Claim 1 of the Exact patent. Frankly, I grow weary of this "rhetorical nonsense" that SCOTUS spawns on what is subject matter under 35 USC 101, and that is now being unfortunately perpetuated by the Bilski decision.

BTW, I also concur in your defense against Dr. TD.

[moocow]

Gentlemen:
I think the time has come to move on from the LabCorp dissent. Justice Breyer's generosity in imparting his views on patent-eligibility was unneccessary, probably confusing, and certainly uncalled-for. That's why the Bilski cert petition ought to scare the dickens out of the biotech community. Just imagine what could happen if Bilski were to be taken up on cert? The only reason for them to take the case would be to say that the machine-or-transformation test is not the appropriate, or not the only, test for "preemption," that the Bilski-test is too rigid, and that we need a "flexible" analysis under which no claim would be safe.

I would submit that a pretty darn "rigid" test for detecting ineligible subject matter is exactly what we need. Bilski, with its "insignificant extra-solution activity" language currently sweeps in too many claims and needs a lot of refinement - but we're better off working in the Federal Circuit rather than the Supreme Court. Let's give Bilski a chance - maybe the CAFC will soon give us some of the urgently-needed guidance they declined to provide in Classen.

[Fan]

Kevin is absolutely right that the bigger test is the policy issue of whether medical diagnostic claims are patentable. If a process is tied to a machine or apparatus, patent practioners need to come up with a generic device that satisfies the 112 and draft diagnostic claims slightly differently. Ofcourse, the patent office would love to limit the claims to the tested device e.g., PCR machine or something like that.

[Bilki and Diehr]

TD said "If the alleged novelty of an invention resides solely in a mental step, then the invention is not patentable, whether under 101 or 102 or 103."

I don't think that is true either under an analysis of Bilski or Diehr. Below is claim 11 from the patent in question in Diehr.

"A method of manufacturing precision molded articles from selected synthetic rubber compounds in an openable rubber molding press having at least one heated precision mold, comprising:
(a) heating said mold to a temperature range approximating a predetermined rubber curing temperature,

(b) installing prepared unmolded synthetic rubber of a known compound in a molding cavity of predetermined geometry as defined by said mold,

(c) closing said press to mold said rubber to occupy said cavity in conformance with the contour of said mold and to cure said rubber by transfer of heat thereto from said mold,

(d) initiating an interval timer upon the closure of said press for monitoring the elapsed time of said closure,

(e) heating said mold during said closure to maintain the temperature thereof within said range approximating said rubber curing temperature,

(f) constantly determining the temperature of said mold at a location closely adjacent said cavity thereof throughout closure of said press,

(g) repetitively calculating at frequent periodic intervals throughout closure of said press the Arrhenius equation for reaction time of said rubber to determine total required cure time v as follows:

ln v = cz + x

wherein c is an activation energy constant determined for said rubber being molded and cured in said press, z is the temperature of said mold at the time of each calculation of said Arrhenius equation, and x is a constant which is a function of said predetermined geometry of said mold,

(h) for each repetition of calculation of said Arrhenius equation herein, comparing the resultant calculated total required cure time with the monitored elapsed time measured by said interval timer,

(i) opening said press when a said comparison of calculated total required cure time and monitored elapsed time indicates equivalence, and

(j) removing from said mold the resultant precision molded and cured rubber article."

The novelty lies in the application of the cited equation to determine when to open the press and remove the cured rubber. However, the other steps like heating the mold, etc. were certainly old. In fact the Court says in Diehr:

"When a claim containing a mathematical formula implements or applies the formula in a structure or process which, when considered as a whole, is performing a function which the patent laws were designed to protect (e.g., transforming or reducing an article to a different state or thing), then the claim satisfies § 101's requirements" and "The machinery pointed out as suitable to perform the process may or may not be new or patentable"

You have to take the claim as a whole. You cannot dissect it and throw out everything but the mental step and then assert that the claim is directed only to mental processes.

I also tend to disagree with the assertion by Kevin that Bilski should be read to mean that the process has to be "performed" by a machine. In the claim above, most if not all of the steps could be performed manually without a machine. For example, opening and closing the press, initiating a timer, etc. Even the novel step could be performed without a machine. However, the process is tied to a machine for curing rubber because it can't be performed without one even though the machine for curing rubber does not perform the steps of the process.

[Kevin E. Noonan]

Dear Moo:

You have a point, but I would have been much more sanguine about the CAFC cleaning up its own house if a panel that included Judge Newman hadn't affirmed Classen in 69 words. I think Judge Newman may be giving up, with senior status looming.

Not that the Supreme Court would do anything other than what you suggest, but at least there would be room for argument. Classen suggests the Federal Circuit may not be listening.

Thanks for the comment.

[BO]

Would claim 1 of US7171311 be unvalid on the Bilski decision?

[This Isn't Terribly Difficult]

Kevin, it's a fun argument. I think you are making the best points in favor of your position that can be made. I understand where you are coming from with your interpretation of Breyer's dissent, but I think your ancillary point is more important: it does not matter whether you argue that the claim effectively claims a
"natural phenomenon", and "abstract concept" or "a mental process."

To me, the most relevant line in Breyer's dissent is this: "the process is no more than an instruction to read some numbers in light of medical knowledge." That's where these claims (and many other so-called diagnostic claims) fail. A commenter upthread referred to the comment as "anti-patent". I disagree. It's a statement of objective fact.

Earlier, Kevin, you made comments about the policy implications of the PTO allowing these poorly drafted claims (I'm shocked that they continue to do so -- if I was a competitor I'd throw these claims in re-exam lickety split).

I presented you with a hypothetical wherein the practical implications of these claims are palpable. Claim 13 says NOTHING about assaying ANYTHING. It certainly is not limited to detecting vimentin directly. For a patentee with a broad claim who needs money, Kevin, your interpretation of your own claim is, frankly, unbelievable. That's not how the world works.

The claim recites simply "detecting" a feature of a sample, i.e., vimentin expression. Is vimentin a newly discovered protein, never observed before by humans? No, it's not. So I could have been assaying vimentin for years. Now someone comes along and publishes a paper stating that "vimentin expression is linked to this disease." As soon as I comprehend that sentence, I become an infringer of claim 13, even though I am doing nothing differently from what I did before. Except thinking.

"I don't think this is thinking - it's acting, insofar as a diagnosis is more than a thought"

It can be more than a thought, but it's not necessarily more than a thought unless you expressly indicate otherwise in your specification. If you remember, that fact is what turned LabCorp into a Supreme Court case: the patentee was on the record stating that the claim covered the "act" of THINKING about the result. Again, that is how patentees behave in real life when they have nothing to lose and money to gain.

[Kevin E. Noonan]

Dear Terribly:

Well, not exactly. The claim reads (so we both don't have to scroll back and forth):

13. A method for detecting the likelihood that a human subject has colon neoplasia, comprising detecting vimentin protein or nucleic acid expression level in a sample from the human subject, wherein reduced expression level of vimentin protein or nucleic acid relative to a control sample from a healthy subject is indicative of the likelihood that the human subject has colon neoplasia.

So it isn't just detecting vimentin (which has been detected before) but detecting for a purpose, i.e. a specific application of detecting. To use the '002 claim as an example,the step of "a difference in thermal quantity of gas or steam between an entrance and exit of a heat exchanger" had been done before but not as it was done in the claim.

I recognize that the way the claim is written tries to avoid a "comparing" step, maybe post Labcorp, but the point remains that the clinical observation captured in the claim (that reduced vimentin expression predicts a likelihood that a patient has cancer) is (probably) new - I haven't looked specifically.

So, the useful result of this concept would be to use it to screen individuals to identify those who should be further screened by more expensive and invasive means (colonoscopy, etc.). There is clearly an economic advantage to such a diagnostic test.

And just thinking about it isn't enough. That "thinking" is done in a context, of hospitals and health insurance and kit-makers and other providers, all of whom have an economic stake in the outcome. And remember, the one really bad outcome - that a physician would avoid "thinking" because of infringement liability - is off the table under 35 USC 287(c).

I know Justice Breyer thinks it just about thinking about numbers. But if there is anything we should have learned in recent years, it's that the Supreme Court has an interesting but perhaps not so informed understanding of patent law.

I agree the discussion is a good one, because I also agree that "thought" isn't patentable. Now we just need to figure out how to protect both of the important principles in a way that promotes innovation and, more importantly, disclosure of that innovation.

Thanks for the comments.

[Kevin E. Noonan]

Dear BO:

So everyone is on the same page (so to speak), claim 1 of US7171311 reads:

1. A method of assigning treatment to a breast cancer patient, wherein said breast cancer patient is a human breast cancer patient, comprising: (a) classifying said breast cancer patient as having a prognosis selected from the group consisting of a first prognosis, a second prognosis, and a third prognosis on the basis of a first expression profile comprising the nucleic acid levels of expression of at least five genes listed in Table 5 in a clinically relevant cell sample from said breast cancer patient by a method comprising (a1) determining the similarity between said first expression profile and a first good prognosis expression profile comprising the nucleic acid levels of expression of said at least five genes to obtain a patient similarity value, wherein said nucleic acid levels of expression in said first good prognosis expression profile represent the nucleic acid levels of expression of said genes in patients having no distant metastases within five years of initial diagnosis; and (a2) classifying said breast cancer patient as having said first prognosis if said patient similarity value exceeds a second similarity threshold value, said second prognosis if said patient similarity value exceeds a first similarity threshold value but does not exceed said second similarity threshold value, and said third prognosis if said patient similarity value does not exceed said first similarity threshold value, wherein said second similarity threshold indicates greater similarity to said first good prognosis expression profile than does said first similarity threshold; and (b) assigning said breast cancer patient a treatment without adjuvant chemotherapy if the breast cancer patient is lymph node negative and is classified as having said first prognosis or said second prognosis, or assigning said breast cancer patient a treatment with adjuvant chemotherapy if said breast cancer patient (b1) is lymph node positive and is classified as having said first prognosis, said second prognosis, or said third prognosis, or (b2) is lymph node negative and is classified as having said third prognosis.

A complex claim, but the key might be the "treatment" step at the end, which could (by some commentators' reasoning) bring the claim outside the scope of the "just thinking about numbers" criticism by Justice Breyer. This step could also be a "transformation" step under Bilski.

The caveat on these claims is the joint tortfeasor problem from Muni Auction, but that's infringment not patentability.

Thanks for the comment.

[Gary Johnston (from Team America)]

Let's say that I discover that reduced vimentin expression is correlated to a statistically significant level with, say, ear wax production. I set up a lab, and offer a clinical diagnostic to physicians that consists of assaying vimentin expression. The stated purpose of my test (in my promotional/marketing materials, that is) is not to predict colon cancer, but rather to predict ear wax production. Do I infringe the above patent? Forget about this joint tortfeasor stuff. What I want to know is, what effect should be given to the preamble and the wherein clause?

[This Isn't Terribly Difficult]

Kevin: "the point remains that the clinical observation captured in the claim (that reduced vimentin expression predicts a likelihood that a patient has cancer) is (probably) new"

Yes. And that's the problem, Kevin: the claim captures AN OBSERVATION. And you can cite Diehr until your blue in the face but "anything under the sun" doesn't include "observations." Observations aren't statutory subject matter. And there's nothing new about that. The only thing that is "new" here is that occasionally (tho not in this case) the PTO or a defendant actually points this out to the applicant or patentee.

"So it isn't just detecting vimentin (which has been detected before) but detecting for a purpose,"

Again, this is improper. A properly drafted method claim is not limited by its "purpose", i.e., the actions in the recited PROCESS are what is invented. Whether I am practicing the claim for the purpose of identifying patients to take off life support, or for the purpose of identifying patients to treat can not be a source of patentability.

"There is clearly an economic advantage to such a diagnostic test."

I have no idea why you are bringing this up.

"That "thinking" is done in a context"

Unfortunately, none of that context is recited in the claim.

"the one really bad outcome - that a physician would avoid "thinking" because of infringement liability - is off the table under 35 USC 287(c)."

I submit that there is no good outcome to patenting a method of thinking a new thought about an old result. Furthermore, what if I send my sample to a lab and have it tested? I'm not a licensed doctor. 35 USC 287(c) doesn't protect me from infringing claim 13 by thinking about the results.

"if there is anything we should have learned in recent years, it's that the Supreme Court has an interesting but perhaps not so informed understanding of patent law"

I think they are far more informed about it now than they were when they decided Diamond v. Diehr and State Street. Is that what you mean? If not, I have no idea what you are referring to.

Read Breyer's dissent again. He is not opposed to patenting new methods of diagnosing patients. He is opposed to claims written so poorly that they capture thinking about the correlation, i.e., the patentee effectively owns the correlation.

As I tried to illustrate in my hypothetical (which is derived from a real life example, surprise, surprise), patentees can not resist abusing these open-ended "detecting X, wherein X determines Y" claims. They can be asserted against just about anybody. All you need is a modestly creative attorney and a list of weak (but solvent) targets.

[This Isn't Terribly Difficult]

Gary Johnston is probably also speaking from experience. I submit, Gary, that even if the preamble is limiting (which it seems that it should be in this case but only the Federal Circuit knows for sure) that you could be non-frivolously sued for infringement under the theory that you and your purchasers know that the *real* purpose of your test is to "determine the likelihood" of colon cancer. And shouldn't the patentee at least be allowed to subject you to the joys of litigation discovery to find out if that is the case?

The preamble in this claim, of course, gives the game away. It's not a method for treating colon cancer. It's a method for "determining the likelihood" that a conclusion is more likely true than false. Note that the claim is not limited to any degree of certainty in this conclusion. It's a "likelihood." I can probably sit here on my computer and find information about people named Noonan that would allow me to make such a determination about Kevin, i.e., if I find a lot of Noonan's on the internet with colon cancer, then there's a greater likelihood that Kevin will get colon cancer compared to someone whose name (or family history) is not associated with colon cancer. Kevin says that such information is "economically valuable." I submit that the market does not agree. The PATENT to a broad claim covering that method is economically valuable, sure. But the information itself: no. No informed person would pay for that information.

And your hypo, Gary, raises another issue that we've discussed a bit upthread: if I "discover" that ear wax production is linked to vimention expression, am I allowed to publish that information? Because now anybody can "determine the likelihood" of colon cancer by evaluating their own ear wax production (indirectly detecting vimentin expression) compared to healthy people. By publishing the information, I am inducing infringement.

This is Breyer's point. Breyer's doesn't understand patent law? It's a ridiculous assertion. Breyer's detractors -- all of whom are patent bulldogs -- are upset because Breyer understands more about patent law than they want him to understand.

[Kevin E. Noonan]

Dear Gary:

I think you are safe predicting earwax but not cancer. That conclusion stems from the assumption that merely detecting vimentin expression is not patentable.

Thanks for the comment.

[Kevin E. Noonan]

Dear Terribly:

I don't want to start arguing past each other. I think the mere observation is not patentable. I don't think that's what the claim says.

Moreover, if you practice the invention, you won't be sued - de minimis non curat lex. What the claim prevents is a big diagnostic lab using this test and charging patients for the privilege, without supporting the basic research.

We will just have to disagree with how informed Supreme Court justices are about patent law.

I would be fascinated to hear about the "real world" example you reference, off-line if you prefer.

I think publishing the link between earwax and cancer wouldn't raise infringement liability, unless there was evidence that the earwax test was being sold only to oncology centers.

Thanks for the comment.