Noonan's Corner Office: Amgen v. HMR

By Kevin Noonan.

The peripatetic case of Amgen v. HMR has once again been reviewed by the Federal Circuit.  No. 05-1157, decided August 3, 2006.  Once again the Court has failed to lay the several issues in the case to rest.  While affirming infringement and validity of two of the patents in suit, the court has remanded the matter once again to the district court in view of its de novo claim construction, and perhaps predictably disagreed with the lower court and reversed a determination of infringement under the Doctrine of Equivalents.

The case has two interesting aspects.  First, citing the rubrics of Phillips the court came to its own determination of the meaning of the term “therapeutically effective amount” with regard to a claim for recombinant erythropoietin (EPO).  EPO is a naturally-occurring hormone that stimulates the body to produce red blood cells, and its absence (or insufficiency) causes anemia.  The standard clinical measure of anemia is the hematocrit, or the percent of whole blood comprised of red blood cells. In construing the term “therapeutically-effective amount,” the district court had required that EPO falling within the scope of the claim containing the term increase hematocrit and have any other biological properties of naturally-occurring EPO. 

The Federal Circuit disagreed, and in doing so provided a nice illustration that Phillips has not changed the Court’s capacity for arriving at its own idiosyncratic construction.  The Court determined that the lower court had improperly focused on hematocrit, which is one of EPO’s biological properties recited in the specification, to the exclusion of several others, including increasing stimulation of reticulocyte response, development of ferrokinetic effects, erythrocyte mass changes, and stimulation of hemoglobin.  The Court noted in particular that the specification recited that the therapeutic properties of recombinant EPO “included” all of these, and that the specification further stated that recombinant EPO was therapeutically useful even if it lacked some but not all of these properties. 

Of course, the Court seems to have not considered the fact that all of the recited properties are related to the clinical measurement of hematocrit, since they are all part of the biological developmental pathway leading to an increase in the number of red blood cells in blood.  Thus, by cherry-picking the language of the specification, the Court was able to arrive at a facially-reasonable claim construction that seems to run contra to clinical reality.  This portion of the matter has been remanded, for further consideration in view of the Court’s claim construction, on the issue of invalidity over prior art references to EPO preparations that did not increase hematocrit but may have had one of the other recited biological properties.

The second issue is the doctrine of equivalents.  The “predicted” recombinant EPO protein contains 166 amino acids, but as produced naturally (and using HMR’s recombinant process) EPO has only 165 amino acids.  The issue before the Court was whether Amgen was entitled to assert claims to HMR’s recombinant EPO as an equivalent.  The district court twice found in the affirmative, and the Federal Circuit remanded in its first review so the lower court could consider the issue in view of Festo. Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722, 740–41 (2002).  The district court having arrived at the same conclusion, infringement under the doctrine of equivalents, the Federal Circuit reversed, finding estoppel arising during prosecution.  Specifically, the Court held that  preliminary amendments resulting in claim language reciting the 166 amino acid species were not made for reasons tangentially related to the asserted equivalent, as found below, but raised prosecution history estoppel that precluded asserting the claim against HMR’s 165-amino acid species.  The Court’s holding was based in part on Amgen’s cancellation of claim limitations related to “fragments” of the 166-amino acid species, and in part because it found erroneous the district court’s determination that the amendments were made to limit the species to human EPO (the Court noted that if Amgen had intended to so limit the claims the mere addition of the word “human” would have been sufficient).  Finally, the Court rejected the notion that amendments made to avoid a double-patenting rejection were “tangential to patentability.”

For those keeping score, the Court in its two decisions has affirmed the district court’s judgment that two of Amgen’s patents are not invalid and are infringed, also affirmed that another Amgen patent is invalid, reversed a finding of infringement under the doctrine of equivalents (but affirmed that this patent is also not invalid), and for one patent remanded the issue of invalidity to the district court while affirming that if valid this patent is also infringed.  As noted in Judge Michel’s dissent (over the claim construction issue), the district court is now in position to decide whether to grant an injunction, keeping HMR’s competing EPO product off the market until at least the expiration of two of Amgen’s patents.  For the parties, that may be (or have to be) enough.

NOTE: Kevin Noonan, PhD is a partner at MBHB and has extensive experience in biotechnology and the chemical arts.  He represents pharmaceutical and biotechnology companies both large and small, and he is particularly experienced in representing university clients in both patent prosecution and licensing.

Comments

[Michael Risch]

I'm not convinced that the Federal Circuit, but reading both the opinion and the dissent I believe the district court got it wrong.

"Therapeutically effective" was clearly intended to differentiate from prior art amounts that were not enough to do any "good." The specification makes clear that the issue is dosage, not what was to be healed.

Prior art = not enough EPO to be helpful
Patent = enough EPO to be helpful

While the treatment might have been increased hematocrit, that's not what was invented - if EPO is later found to have other effects, there's no reason that shouldn't be covered. It's like saying that a "therapeutically effective" amount of aspirin is an amount that makes a headache go away. But what if it is later learned that aspirin has other effects, like aiding with heart disease? While the dosage question of "therapeutically effective" might be in question, the patent still covers aspirin even if it might be used for other things.

[Lonnie E. Holder]

Okay, I surrender. What does Eli Lilly and Zyprexa have to do with Amgen v. HMR?

[Paul Cole]

"Therapeutically effective" is a limitation by result which is inherently uncertain. It is a limitation of a particularly indefinite and arm-waving character because it does not specify either what disorder is to be considered or how therapeutic effectiveness or its absence is to be determined. Is the therapy against anaemia, depression, irrational cheerfulness, malaria, the common cold, hay fever or what? The description may tell us, but the claim does not define it for us.

Looking at it another way, what technical teaching is added to the claim by the presence of the words "therapeutically effective." The answer is none - the intended ambit of the claim is to cover everything that works for some purpose or other while leaving everything that does not work outside the scope of the claim, and leaving it for the skilled reder to determine which is which. "Nice work if you can get it" as the song goes, but perhaps not surprising that such vague language does not impress a judge, or in the CAFC three judges. It might, perhaps, defend against an allegation of non-enablement, but it is not a good vehicle for a novelty or non-obviousness distinction.

MORAL: Get claims of this general scope if you can: they may terrify the competition. But back them up with additional claims with more specific definitions of disease process, method of determining therapeutic effectiveness, or even just dosage ranges.

[Perplexed patent practitioner]

OK, let me save everyone some time. I found this on the internet. In the Greek, the word "peripatetic" means "walking with." This was the method of teaching that Aristotle used - ie, "walking with" his students in the gardens as he taught them and they questioned him. Hence, the term "peripatetic" is used to describe the followers of Aristotle but in the present context may mean only travelling about.

[Andrew Brenc]

Lonnie - Very well written. I agree.

[Andrew Brenc]

OK, I did not mean "Lonnie", I meant Paul Cole.

Well written Paul Cole, I agree.

The comments dividing lines are poorly placed and it makes it unclear, at first, who has written each piece.

[Kevin E. Noonan]

With regard to "therapeutically effective," here the words were not mere surplusage - they were intended to mean that the EPO worked (mostly because there was enough of it in the recombinant preparation as opposed to the small amounts in the prior art). It was the patentee's description of what that meant that got him/Amgen into trouble with the CAFC - they included all the things that EPO can do.

However, my point is that all those things are related to hematocrit - the hematocrit is what a physician measures, but all the other things are required to get to the hematocrit.

We can quibble about whether the disclosure was clear enough, but if the Court begins to find words like "therapeutically effective" to be somehow indefinite or uncertain, then pharmaceutical claims will be valid only after the FDA determines the therapeutically-effective dose. Not the right result.

[Paul Cole]

Pleased to see that we have a debate!

If you look up the decision, you will see that the patent in issue is US 5955422. It contains just two claims, and here they are:

1. A pharmaceutical composition comprising a therapeutically effective amount of human erythropoietin and a pharmaceutically acceptable diluent, adjuvant or carrier, wherein said erythropoietin is purified from mammalian cells grown in culture.

2. A pharmaceutically-acceptable preparation containing a therapeutically effective amount of erythropoietin wherein human serum albumin is mixed with said erythropoietin.

What was wrong with the way the patent was drafted?

The answer is childishly simple. The business that the inventors were in was the production of erythropoietin for the treatment of anaemia. In particular, the problem had become particularly acute for kidney patients being treated by dialysis.

So should there not be claims that (a) define the therapeutic amount as being for the therapeutic treatment of anaemia and (b) defining the therapeutic amount of the compound as being for the treatment of different forms of anaemia (what may be an effective amount for one form of anaemia may be different from the amount for another form)? Only a few minutes to write, not so difficult to argue with the examiner, but giving good protection for the inventor IN THE FIELD OF ENDEAVOUR IN WHICH HE IS RESEARCHING.

Again, at the time when the patent was written, erythropoietin was a known compound with known effects. So it might have been possible to claim some dosage ranges, which are much easier to construe than limitation-by-result expressions like "therapeutically effective" and do not have to be tied to the treatment of particular diseases.

In the UK, where I practice, I am well known as the author of a paper on the Windsurfer litigation, and more recently for a textbook just published called FUNNDAMENTALS OF PATENT DRAFTING (hopefully Patently Obvious will review it sometime soon). The idea of the inventors of Windsurfer was to put a sail on a surfboard. But their attorneys were so hypnotized by the generality of claiming some kind of vehicle (including land vehicles and ice vehicles) that they neglected to include a claim to a surfboard. Even when they re-issued the patent in the US, they omitted to include a claim to a surfboard.

Some drafting faults recur across all technologies. Omitting to include claims that specify your field of endeavour is one of them, and often one of the most damaging.

[Lonnie E. Holder]

Paul is absolutely correct, as the practitioners on this site will know. Claim drafting 101 says that you want an array of claims, from broad to narrow, so that you can gauge what the USPTO believes to be patentable AND, perhaps more importantly, so you can have a fall back position if your broadest claims are deemed unpatentable, either during prosecution or afterward.

[Paula Cole]

Paul Cole, you make a good point. However, in the US, where I live, I Googled your name and found that you are well known for racehorse training, not for writing a paper. Please refrain from inapplicable superfluous statements.

[Paul Cole]

My personal identity has been stolen!

Try Googling --- "Paul Cole" Windsurfer --- and you should be able to download the paper from Mondaq.