One of the first things any patent practitioner (or inventor filing a patent application) learns is the difference between independent and dependent claims. The former read something like,
“1. A device comprising an X, a Y and a Z connecting the X to the Y”, or
“2. A method for doing/making A comprising doing X, Y and then Z”.
The hallmark of a dependent claim is that it necessarily incorporates all the limitations of the claim from which it depends, and further specifies some additional limitation(s). In the U.S., this is codified in 35 U.S.C. §112, fourth paragraph, which reads,
Subject to the following paragraph, a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
Thus dependent claims usually read something like,
“3. The device of claim 1 wherein the X is shaped like a doughnut”, or
“4. A method according to claim 2 wherein X is conducted at a temperature above 60ºC.”
Sometimes, though, a claim will be worded, e.g.,
“6. A method for treating cancer, comprising administering to a patient in need thereof a compound as recited in claim 5.”
Is that an independent or dependent claim? I don’t know any practitioners who would say it’s a dependent claim: although claim 6 has incorporated the limitations of claim 5, it has done so simply as a form of shorthand. Claim 6 hasn’t added any further limitations regarding the compound recited claim 5. And most importantly, whereas claim 5 is directed to a compound, claim 6 is directed to a different statutory class of patent-eligible subject matter, a method of treatment. So claim 6 must be independent.
Curiously, under US practice, there is at least one situation in which claim 6 would be regarded as dependent. When counting the number of independent claims for purposes of calculating the filing fee due (each independent claim beyond the first 3 costing the applicant big bucks), the USPTO regards a claim like claim 6 above as dependent, thus saving applicants filing fees. At least, that’s been my experience and the experience of other practitioners with whom I’ve spoken. However, for purposes of unity of invention, it’s possible that an examiner would regard claims 5 and 6 as being directed to different inventions that necessitate separate searches, and thus force an applicant to choose between the two, in so doing implicitly finding that claim 6 is an independent claim. In Israel, claim 6 would be regarded as an independent claim (although in Israel, there are no excess fees for excess independent claims per se, only for the 51st claim and up, period).
Why discuss such a basic notion as the difference between independent and dependent claims? Because it arose two weeks ago in a decision from the Board of Patent Appeals and Interferences (BPAI) at the USPTO. The decision was handed down in an old interference (declared in October 1996!) between Abbott GMBH & Co. KG and Yeda Research and Development Co. Ltd., the tech transfer arm of the Weizmann Institute. At issue was the ownership of patent rights over proteins that bind another protein known as tumor necrosis factor alpha (TNFα). The crux of the dispute between the two parties was whether or not Abbott’s US 5,344,915 was entitled to the benefit of Paris Convention priority to a German application filed in 1989. The BPAI ruled it was, hence Abbott had rights in the one count over Yeda’s USSN 07/930,443.
The interesting part of the decision, however, concerns Yeda’s contention that claim 2 of the ‘915 patent was in improper form for a dependent claim, as set forth in 35 U.S.C. §112, ¶4. These claims read as follows:
1. A purified and isolated TNFα-binding protein which has a molecular weight of about 42,000 daltons and has at the N terminus the amino acid sequence
Xaa Thr Pro Tyr Ala Pro Glu Pro Gly Set Thr Cys Arg Leu Arg Glu
where Xaa is hydrogen, a phenylalanine residue (Phe) or the amino acid sequences Ala Phe, Val Ala Phe, Gln Val Ala Phe, Ala Gln Val Ala Phe, Pro Ala Gln Val Ala Phe or Leu Pro Ala Gln Val Ala Phe.
2. A protein as claimed in claim 1 in deglycosylated form.
Yeda argued that since the deglycosylated protein recited in claim 2 had molecular weight that, according to Example 6 of the ‘915 patent, was about 3000 Daltons less than that of the claimed TNFα-binding protein, claim 2 did not include all of the limitations of claim 1, and did not contain a further limitation on the protein of claim 1. Hence, said Yeda, claim 2 was in improper dependent form. In support of its argument, Yeda cited to MPEP ¶ 608.01(n), which sets forth an infringement test for determining compliance with 35 U.S.C. §112, paragraph 4:
The test as to whether a claim is a proper dependent claim is that it shall include every limitation of the claim from which it depends (35 USC 112, fourth paragraph) or in other words that it shall not conceivably be infringed by anything which would not also infringe the basic claim.
Argued Yeda, since the lower molecular weight deglycosylated protein of claim 2 could not conceivably infringe Claim 1, the recitation therein was not a “further limitation” on the subject matter of that claim.
The BPAI majority disposed of this argument by saying that Yeda bore the burden of proof, and had failed to show that a person of ordinary skill in the art would not regard the deglycosylated protein of claim 2 as being “about 42,000 Daltons” as recited in claim 1, and therefore
Yeda has failed to show that the deglycosylated protein of Claim 2 is completely outside the scope of the “about 42,000 dalton” requirement of Claim 1 (Pfizer, 457 F.3d at 1292) or that the deglycosylated form would not be understood to infringe Claim 1. Yeda has not established, prima facie, that Abbott’s Claim 2 does not provide a “further limitation” of the Claim 1 subject matter.
Yeda’s argument was a creative and intriguing one, and the majority’s disposing of it on grounds that 39,000 Daltons is close enough to 42,000 Daltons isn’t really satisfying. Suppose that claim 2 had instead read, “2. The protein of claim 1 having a molecular weight of 39,000 Daltons”. In that case, Yeda’s argument might have been stronger – if a claim specifies a molecular weight if 42 kDa, and the patent itself doesn’t explicitly that 39 kDa is to be considered as 42 kDa, it’s not clear that a 3000 Dalton difference is minor.
One Administrative Patent Judge, however, while concurring in this part of the judgment, did so on different grounds. The first paragraph of his two paragraph concurrence is unremarkable:
While I join the majority in its denial of Yeda’s motion for unpatentability based on 35 U.S.C. 112(4), I do so for a different reason. Paragraph 4 of §112 requires an express reference to the claim from which it depends and an incorporation by reference of the limitations of that claim. It does not follow, however, that the converse is true: reference to another claim does not mean that the claim is intended to be dependent. For at least a generation the Office has permitted such cross references in claims that cannot possibly be dependent. For example, “A widget made by the method of claim 3.” The widget cannot be within the scope of the method. The cross reference simply incorporates the limitations of the cited claim without creating a dependency. Under this practice, the principal effect is that the non-dependent, cross-referencing claim is itself independent and thus the applicant might owe an additional fee if there are more than three independent claims. 35 U.S.C. 41(a)(2)(A). In an age of word processors, one could question whether the convenience of cross referencing is outweighed by the confusion it causes. The question, however, is beyond our purview.
That’s just as discussed above: different statutory classes = no dependency, even if another claim is referenced. Having (in baseball terms) set up the non-dependence of claim 2, the concurrence closes:
I conclude that Abbott's claim 2 is an independent claim. The claim states in its entirety “A protein as claimed in claim 1 in deglycosylated form.” Significantly, the claim does not use the typical dependent form of “The protein of claim 1, wherein the protein is deglycosylated.” The use of “A protein” rather than “The protein” is telling since it deliberately avoids making the protein of Abbott claim 1 a literal antecedent. Since the patent has only three claims, we cannot infer anything from the fees paid. Abbott claim 2 is an independent claim and thus cannot be unpatentable for failing to comply with the statute governing claim dependency.
It’s hard to argue with this APJ’s logic, which also explains why the majority’s view doesn’t sit quite right: claim 2 is not an improperly dependent claim because 42,000 can mean 39,000. It’s not an improperly dependent claim because it’s not a dependent claim at all!
Interestingly, most practitioners I know don’t really care how one begins a dependent claim, “A thing according to claim 1…” or “The thing according to claim 1…”; the particular wording usually comes down to personal preference. It may be that the person who drafted claim 2 intended it to be an independent claim, but it may also be that in this case they got lucky.