3(d) a patent eligibility provision: DHC in Roche v. Cipla

3dIn Part I I’ve discussed the decision of the Division Bench of the Delhi High Court (DB) in the Roche v. Cipla patent infringement suit on the tests for obviousness and infringement of a patent. As explained, Cipla manufactures the drug ‘Erlocip’ which is a particular polymorphic form of the Erlotinib Hydrochloride compound (Polymorph B). Roche’s patent IN ‘744, on the other hand, is for the compound Erlotinib Hydrochloride (EH) which at the time of grant comprised of a mixture of two polymorphic forms (A and B). In this post I discuss the decision of the DB on the impact of the rejection of the separate patent application for B polymorphic form of EH on the ground of Section 3(d).

A crucial question in deciding the patent infringement claim in this case has been: Does a patent for a substance (here, EH) cover variations of that substance (here, Polymorph B of EH)? As explained in Part I, the test for infringement requires determining the meaning and scope of claims in the suit patent and then comparing the properly construed claim with the allegedly infringing product. Since Section 3(d) deals with incremental variations, construing the purpose and scope of the Section 3(d) became necessary in this case.


After patent US ‘498 on EH (corresponding to IN ‘774) was sought, a method to separate and purify the B polymorph of EH was discovered and that compound was separately patented in the US as U.S.’221. Roche began selling a drug by the name ‘Tarceva’ in which the API is the B polymorphic form of EH and is hence covered by US ‘221. Roche’s patent application for the B polymorphic form in India, viz., Patent Application DEL ‘507, however, was rejected by the Controller on the ground that the increased thermodynamic stability of Polymorph B fell below the Section 3(d) standard of enhanced therapeutic efficacy. Cipla argued that Roche, by seeking an independent patent on the B polymorphic form, have clearly sought to distinguish it from the invention disclosed in IN ‘774/US ‘498. Further, the fact that the application failed on Section 3(d) meant that the provision could “not be (then) utilized as a tool to enhance the ambit of a patent to cover even those forms which have either been abandoned by the patentee itself or rejected in India.” Thus, Cipla argued that it was not possible to construe the claims in IN ‘774 to include the B polymorphic form in light of the rejection under Section 3(d).

Section 3(d): A patent eligibility provision

The DB analysed Sections 2(1)(j), 2(1)(ja), 2(1)(l), 2(1)(ta) and Section 3(d) and held:

“61. Section 3 of the Act lays down a threshold for patent eligibility and is not an exception to Section 2(1)(j) as urged by learned Senior counsel for Cipla. Section 2(1)(j) provides a theoretical definition of an invention while Section 3 illustratively outlines what are not inventions. In other words, for subject matter that falls outside the scope of Section 3, a qualitative analysis needs to be employed to ascertain whether it satisfies the conditions of Section 2(1)(j), while for subject matter that falls within the scope of Section 3, an analysis under Section 2(1)(j) need not be employed as it will be rejected at the threshold.”

Thus, the steps in analysing a new chemical entity in respect of which a patent application is made will be:

“A new chemical entity (NCE) that is structurally dissimilar but functionally similar to an existing chemical entity is thus merely a substance under Section 3(d). If the substance has an added layer of enhanced efficacy then it would be treated as a “new product” and would be eligible for assessment under Section 2(1)(j) to ascertain whether its formation involved an inventive step. If the new product involved one or more inventive steps, then it would qualify as a pharmaceutical substance.”

Thus the court negatived the proposition that Section 3(d) is a patentability standard. It must be remembered that the Supreme Court in Novartis v. Union of India did not decide whether Section 3(d) was a patent eligibility or patentability standard since this question did not impact its determination on validity in the facts of that case (It did, however, state that for pharmaceuticals, Section 3(d) would not operate as a patentability standard but as a threshold for ‘invention’, i.e., a patent eligibility criteria). This begs the question: What exactly is the impact of identifying Section 3(d) as a patent eligibility standard versus a patentability standard? It can be argued that this strengthens Section 3(d) itself, since it now becomes a threshold question prior to the tripartite patentability test. It also means that complex questions of anticipation and obviousness would not be mixed up with the Section 3(d) screening and that the latter can operate as a neat and distinct test. For example, since methods of surgical treatment are not patent eligible [Section 3(i)], there is no requirement for the Patent Office to examine whether the method is novel or obvious or capable of use, and the application can be rejected right away. Here, however, the enquiry is slightly more complicated by the requirement to show therapeutic efficacy.

In terms of the conceptual distinction between the two, while the novel/non-obvious/industrial application standard for patentability addresses the worthiness of a product to monopoly rights, Section 3(d) seeks to prevent evergreening and the exploitation of these monopoly rights. Therefore, while the purpose of these two standards can be described slightly differently, the nature of the enquiry involved in both Section 3(d) as a patent eligibility and patentability standard seems to be the same.

Legal Fiction: Infringement in the case of derivatives of substances under patent 

On the impact of the Section 3(d) challenge to the B polymorph patent application on the present infringement determination, the DB held that “merely because an inventor applies for a latter patent that is already objectively included in a prior patent, but which inventor subjectively feels needs a separate patent application, doesn‘t mean it is to be taken at face value and therefore neither Section 3(d) or abandonment of subsequent patent application can be used to read into terms of prior application, which has to be construed on its own terms.” However, the Court goes on to examine Section 3(d) and finds it contains a deeming fiction in relation to ‘salts, esters, ethers, polymorphs, metabolites, pure form, particle size, isomers, mixtures of isomers, complexes, combinations and other derivatives of known substance’, namely, that they are to be considered as the same substance as the known substance.

The DB then holds that the deeming fiction in Section 3(d) implies that when a patent application for a substance is rejected by Section 3(d) because it is a derivative of known substance, that substance would automatically be deemed to be covered and disclosed by the prior art on the basis of which the application was rejected, i.e., the known substance if under patent.

“73. We understand Section 3(d) as a positive provision that in fact recognizes incremental innovation while cautioning that the incremental steps may sometimes be so little that the resultant product is no different from the original. The inherent assumption in this is that an infringement of the resultant product would therefore be an infringement of the original i.e. the known substance and by no stretch of imagination can Section 3(d) be interpreted as constituting a defence to infringement.”

Thus, what Section 3(d) does impact, apart from the prosecution of a patent application, is the construction of claims of the patent of the known substance whose ‘salts, esters, ethers, polymorphs, metabolites, pure form, particle size, isomers, mixtures of isomers, complexes, combinations and other derivatives of known substance’ is sought to be patented independently (but fails because of Section 3(d)). Now, ‘disclosed’ and ‘covered’ by a patent refer to two separate concepts. ‘Covered’ refers to whether a certain product can be said to be within the four corners of the claims of the suit patent and it relevant in an infringement analysis. The ruling of the court would mean that these salts, esters, etc. would be deemed to be covered by the patent and thus any manufacture of these would infringe the original patented product. ‘Disclosed’ would refer to whether the patent adequately teaches a certain aspect of an invention – whether chemical composition, structure, use, etc and is relevant to a validity analysis. Thus, there might be cases, like this one, where a patent covers a salt, ester, etc. for the purpose of infringement, but does not actually teach the distilling out or manufacture of that particular salt, ester, etc.

Would this make it easier for patent holders to bring infringement suits? Will this mean less stringent disclosure (of variations and uses of variations) in patents on chemical compounds? What are other possible impacts? A shout out to our readers to share their comments!

Tags: , , , , ,

About The Author

1 thought on “3(d) a patent eligibility provision: DHC in Roche v. Cipla”

  1. Hi,
    I’m not sure I understand this properly. If product A is already covered under the first Patent why would anyone file another application for a separate patent, stating specifically that it is a new product? And if it is rejected, like it was here, then doesn’t changing the argument to say that it is covered under the previous patent contradicting your own previous claim? Also, for a patent to cover a product, shouldn’t it be explicitly disclosed in the application, considering that you get a patent only for what you disclose?

Leave a Comment

Scroll to Top