Roche vs Cipla: Did the Judges Construe the Claim At All?

I’m still coming to grips with the division bench ruling in the Roche vs Cipla matter. I wanted to put out some initial thoughts in the hope that I can get a bit more clarity from some of our more knowledgeable readers.

The facts, known to most of you are briefly as below:

Roche sued Cipla before the Delhi High Court claiming that “Erlocip”, a generic product by Cipla corresponding to Roche’s originator drug, “Tarceva” violates Roche’s Indian patent (IN196774)–hereafter ‘774 patent. Both drugs are based on a compound that goes by the name of “Erlotinib Hydrocloride”.

Trial Court Ruling

The single judge, Justice Ravindra Bhat declined a temporary injunction on the ground of “public interest” i.e. the lower priced generic version of Cipla would enable “more access” to this life saving drug in India. Although Cipla challenged the validity of Roche’s patent in this suit and even claimed non infringement, the trial judge found that Roche had established a “prima facie” case of infringement. However, the judge curiously noted that Cipla had raised a “credible challenge” to the patent. For our earlier posts on this case, see here.

At the appellate court, Cipla focussed less on the “public interest” part and more on the “non infringement” aspect, as discussed below:

Non Infringement

The judge held that since Roche had not established infringement of its ‘774 patent, there was no “prima facie” case in favour of Roche. And therefore the application for interim injunction was dismissed.

Since this was really the “ratio” of the case, I plan to focus just on this “non infringement” aspect in this post.

The other parts of the judges’ ruling (that the patent in issue faced a credible challenge of invalidity and was therefore vulnerable + that public interest dictated that no injunction ought to be granted) is really “obiter” and will be taken up in a future post.

To appreciate Cipla’s “non infringement” argument, it may help to take a look at the patents that have been implicated in this case.

1. The Suit Patent (774)

Application Number: 537/DEL/1996

Granted patent: IN196774 (‘774)

Grant Date: 23 Feb 2007 or 6 July 2007 (we’re not sure, since the patent office appears to have found the patent to be in order for grant and then sat on it extensively before issuing a certificate–during which time an opposition was filed by Natco and decided in favour of Roche.

This case again demonstrates the value of PH Kurian’s order that no patent certificates shall be kept pending and shall be issued immediatley after the application is placed in order for grant.

Corresponding US patent: US 5747498 (‘498). This was granted in the US in 1998.

2. The Future Patent Application (‘507)

A later patent application claiming a specific polymorph B of Erlotinib hydrochloride was filed in India (IN/PCT/2002/00507/DEL) (hereafter the ‘507 application). This corresponds with US patent 6900221 (the ‘221 patent). This was filed in India in 2002 and rejected in December 2008 after an opposition filed by Cipla on the grounds of section 3(d) primarily.

Cipla’s argument runs thus: The ‘774 patent only disclosed Erltonib Hydrochloride, existing as Polymorphs A+B. Whereas the ‘507 application claims Polymorph B alone. The Tarceva tablet corresponds to polymorph B alone, since this polymorph is much more stable and capable of being made into tablet form (solid form oral dosage).

The judge found in favour of Cipla, holding that Tarceva really corresponded to Polymorph B, which was the subject matter of ‘507, an application that was ultimately rejected. Therefore there was no infringement of the ‘774 patent. The judge then holds that Roche ought to have disclosed the ‘507 application to the patent office whilst prosecuting the ‘774 patent. Had this been disclosed, the judge opines that ‘774 patent may not have been likely to be granted at all! And then goes on to impose costs on Roche in this regard for non disclosure/suppression of material facts.

Concerns with Appellate Ruling

My concerns with the above line of judicial reasoning are as below:

i) The first claim of the ‘774 patent reads as:

“A novel [6,7-bis(2-methoxyethoxy)quinazolin-4-yl]-(3-ethynylphenyl) amine hydrochloride compound of the formula A.”

To me, this appears to be a claim covering the compound: “Erlotinib Hydrochloride”. This compound may exist in several polymorphic forms, but any such forms are likely to be subsumed within this patent.

Therefore assuming Cipla’s drug relies on only one particular polymorphic form of the Erlotinib Hydrochloride compound (Form B), it will still infringe the main patent. For Cipla cannot get to such polymorphic form wihtout in some way implicating the patent covering Erlotinib Hydrocholride.

Therefore, it ought not to matter that the drug “Tarceva” (and Cipola’s generic version) was primarily based on the Polymorph B version (as elucidated in the later ‘507 application). Making the drug will necessarily infringe the parent ‘774 patent.

ii) Not only did the court appear to have missed the above chain of logic, it went out of its way to hold that Roche suppressed material information by not bringing the existence of the ‘507 application to the notice of the court or the patent office (during the course of prosecution of the ‘774 patent). The court insinuated that by withholding this information, Roche was in some way culpable of fraud.

Firstly, if the ‘774 patent is broad enough to subsume all polymorphs, is there a need for Roche to disclose any other accompanying patent application to the court? Of coure, there could be a section 8 issue here, but that is a matter for the patent office and the corresponding oppositions filed. Section 8 requires that all information relating to corresponding patent applications/grants in foreign countries be brought to the notice of the patent office. For an interesting discussion on the contours of this provision, see this post by Sai Deepak.

Secondly, I don’t necessarily see an inconsistency in Roche’s statements at the patent office with respect to the ‘507 application: i.e. that although ‘774 patent subsumes the polymorph, it still does not disclose it adequately enough, meriting another patent application (‘507) that is of the “selection” variety. As to whether this second application is meritorious enough to warrant a separate patent is another issue–and the patent office held (in its December 2008 oppn decision) that it wasn’t meritorious enough.

Does the strength or otherwise of this later patent application that was ultimately rejected by the patent office affect the strength of Roche’s infringement case that was plainly limited to the ‘774 patent?

At best, whether or not Roche committed fraud by withholding pertinent information appears a contentious issue to me? And not a conclusive one or a strong enough one to merit the awarding of “costs” of Rs 5 lakhs. Imposing such costs on them when they were attempting to legitimately fight on the basis of what they thought to be validly granted patent rights appears a bit extreme to me.

In what can only be termed as the worst kind of a double whammy, Roche lost out on both patents (‘774 and ‘507).

Conclusion 1: Whither the Claim?

The most surprising part of the division bench decision is that they do not refer to the claims of the ‘774 patent even once. The key question ought to have been:

What exactly does the ‘774 patent claim? Does it cover Erlotinib Hydrochloride? As I’ve argued earlier, from my tentative understanding of claim construcition and chemistry, the claim covers Erltoinib Hydrochloride. As to whether or not this is a valid claim is a different issue. But assuming it is valid, does not Cipla’s use of a specific form or version of this compound infringe?

Conclusion 2: Deconstructing the Reasoning of the Judges

If my tentative reading of the claim is correct, how did the judges miss it? And why did they go out of their way to impose costs on Roche? I offer three potential reasons and welcome your thoughts on this.

(i) The judges were clearly biased against Roche: In my opinion, this is very very unlikely (not least because I run the risk of contempt here)

(ii) Our judges are still coming to grips with patent law and cannot be expected to master patent nuances on day one. Consequently, there will be mistakes at this stage. However, this would then mean that our courts must, as far as possible, defer to the patent office on issues of claim validity and the like? At least to the extent that such issues of validity, turn not on issues of law but issues of fact, such as determining the scope of the prior art and determining whether something is obvious or not to a person skilled in the art. And particularly when such patent has already been subjected to an opposition. I will come back to this point in a later post.

(iii) Roche’s counsel and strategy were no match for that deployed by Cipla.

I personally think the decision came out the way it did owing to reasons number (ii) and (iii) (with reason number (iii) being the predominant reason). I had already suggested this in a number of earlier posts. I also happen to have sat in court during some of the hearings and must say that the combination of Pratibha Singh and Arun Jaitley was simply terrific.

However, notwithstandind the relative deficiences in choice of counsel and in terms of legal strategy, my own view is that Roche has a perfectly appealable case (on merits) to the Supreme Court. If not for anything else, it must get that little black mark of “suppressing material facts” expunged.

As to whether it will be able to obtain a temporary restraining order from the Supreme Court is unclear. Given that two lower courts denied the injunction and Cipla has been selling the drug since December 2007, the apex court is unlikely to grant an injunction. However, there is a good chance that it may just get its claim construction and chemistry right. What do our readers think?

Tags: , , , ,

About The Author

11 thoughts on “Roche vs Cipla: Did the Judges Construe the Claim At All?”

  1. Our Insightful Anonymous commentator (IAC from henceforth) who has already sent us fabulous comments on the petition to Mr Kurian has now given us some fabulous insights into the Roche vs Cipla reasoning.

    As always, I was struck by his mastery of patent nuances and his deep knowledge of Indian cases. Here is what he opines:

    “From the way you have described it, it seems to me that the 507 application can be treated as having been rejected for double patenting by treating the 774 patent as adequately subsuming the B Form selection.

    If so, the 507 rejection actually clarifies the Form B selection-inclusive construction placed on 774 by the Patent Office during the 507 rejection. The 507 ruling if cited would have enured to the patentee’s benefit; hardly making for any materiality in concealment.

    This is not the first time we are seeing courts in India floundering on the basics of claim construction and on the basics of infringement analysis. Your reasoning is sound.

    In the Bajaj-TVS case also, you will find that a lengthy judgement of the Madras High Court barely reproduces some claims and then just not know what to do with it. In that case, the real point – whether the claim phraseology in claim 1, concerning a restriction on the intended use of the invention, was a limitation or not, and whether the claim so read presented a point of novelty or not, was completely lost on the court. In other judgements we have seen the Delhi High Court apply some kind of a pith and substance approach to the invention as a whole (all claims construed together in a consolidated manner instead of claim-by-claim; with an ultimate finding on invention-infringement instead of claimwise infringement findings).

    I have yet to see a single Indian court ruling that has undertaken any claim construction and analysis by actually deconstructing a claim to clearly identify each of its elements (/steps/components/parts) and also clearly identify each of the interconnections (/relationships) that each element has with which other elements, proceeding element-by-element; and then using a check-box analysis for infringement to see element-by-element and interconnection-by-interconnection whether each is present in the accused article(/apparatus/method/composition) or not.

    I have come to believe now that for Indian courts, indeed for ‘less-patent-experienced’ courts anywhere, the challenge is just too great. I think we should modify our system to relegate patent invalidations exclusively to the IPAB with only appeals from it to the High Court. Courts should focus purely on infringement issues and not deal with it in combination with a revocation (invalidation) action. China’s system has some of these features.

    But it would be a mistake to think that our Patent Office is much better when it comes to allowing claim language during grants. In the Bajaj -TVS case, just see, for example, with what flourish the rule of requiring antecedents is breached in the allowed claims. Not just that, the Patent Office allowed the main independent claim in a language that confronted the court with the subtle nuance of limitation-identification in the intended-use phrase. As we all know, strictness in applying the rules of claim drafting before allowing claims directly facilitates claim construction in infringement anaysis subsequently. And how the Patent Office by allowing faulty claim language, makes the court’s job of claim construction even more difficult. But the Patent Office should exercise stricter scrutiny of the ‘grammar andd syntax’ of claim language before allowing them viz. grant, not only to avoid disservice to the court later but also to itself during its examination of patentability.

    Here is a sugggestion that is self-evidently radical to generate debate. Should the Patent Office make its job of examining patentability easier by requiring an applicant to draft and present alongwith the RFE, a single traditional German style Jepson format improvement-focussing claim?”

  2. My reply to IAC:

    Dear IAC,

    As always thanks very much for your very thoughtful comments on this case. As for the ‘507 application, the rejection was on the grounds of section 3(d) and insufficient disclosure. So not really rejected on double patenting grounds. Interestingly, the initial attack by the opponents was on anticipation and inventive step as well–however, after amendment of claims (in response to FER) the opponent limited challenge to insufficiency and section 3.d.

  3. Dear IAC,

    I’m also thinking aloud here and would like your inputs on this issue:

    As I’ve argued in the post, the ‘774 patent claims “Erlotinib Hydrochloride” (EH). Assume that the invention when worked would only yield EH in a certain specific form i.e. EH as Polymorph B or EH as polymorph A+B (which actually seems to be Cipla’s argument in this case). If EH is obtainable from the patent only as Polymorphs A+B, does the patent cover EH in all its forms?

    Or is it limited to an EH version that is A+B. In other words, would an amorphous version of EH or a separate crystalline form (let us assume EH in polymorphic form C fall outside the scope of the claim)?

    What do you think?

  4. Dear Sir,
    I agree with IAC to a large extent and would like to add to what he has said. It is true that rarely do Indian courts resort to claim construction or any analysis worth being termed so. Unfortunately, claim construction and that too in Indian Courts is still a few years away because our Courts are yet to get a hang of the statute, leave alone intricate doctrines which require nuanced understanding of the application of such doctrines. This requires a fairly balanced blend of technological and legal knowledge, i.e. the Court needs to understand how a drafter goes about his work. When a Court thinks aloud in the open the very purpose of a spark plug (not even twin spark plugs) as it did in the Bajaj-TVS case, though I appreciate its honesty, it should indicate the state of affairs.

    Also, the finest of judgments in any branch of the law are usually the product of the subtle arguments presented by the Counsels which are then mulled over by the Judge and finally find place in the judgment. At the end of the day, it is the counsels who put forward their take on the law and help the Courts arrive at a decision. This means that if counsels had raised any points about claim construction that are worth mentioning, they would have certainly found their way in the judgment. The absence of any such analysis in judgments probably indicates lack of such arguments in the submissions of counsels. In the name of claim construction what one gets to see is mere verbatim replication of what is given in the patent specification followed by a “bald averment” on the position of the law. Until such time, it is realised that this branch of law needs a very rigorous and thoroughly application-based approach with every word and sentence in the patent specification being put to withering scrutiny, all we shall get to see in patent cases here would be a repetition of the established principles of prima face analysis, mere recitation of the provisions of the statute and citing precedents without specifically explaining their relevance to the case on hand. Having said that, I think we need to give ourselves and particularly the Courts some time because to judge them now would be premature and harsh. It should also be appreciated that there is an increase in the Courts comprehension of the law which is reflected in the judgments. Also, it is not that Indian case law is completely bereft of good judgments on patent law; the 1969 decision of the Bombay High Court in Farbwerke V. Unichem is one good example where the Court has gone deep into the technology and the law. I don’t say this because it runs into several pages, but because the Court did really put quite a bit of effort in making sense of the issues involved.

    Therefore, let us hope that both counsels and courts emphasise more on applying claim construction or deconstruction techniques often.

    J.Sai Deepak.

  5. Not discussing patent claims or patent related arguments is not the hallmark only of the Courts.. take a look at most of the Patent Office rejections… they are replete with 3D based rejections and not focusing on obviousness positions or lack of inventive step arguments.
    Let me clarify – for most crystal form/ hydrates, the rejections have been on 3d… while I would have preferred that the PO would reject them initially on lack of inventive step and additionally on lack of data u/s 3d.

    The Patent office and counsels are more than happy to argue 3d… the counsels don’t want to put really coherent arguments on why crystal forms/ solvates are not inventive.. and so is the patent office staff who don’t want to give a reasoned rejection order in most cases… only bald rejections on 3d in most cases.

    The rejection of hydrates/ crystalline forms even when done on lack of inventive step has been very tangential/ illusory, to say the least.

    When our PO staff does not want to give coherent obviousness based rulings, where is the question of HC judges coming out with coherent claim construction based rulings?

    Frequently Anon.

  6. Dear FA,

    You’re right that patent office decisions also leave much to be desired. But where properly reasoned, I think they must be given some level of deference, without opening up everything de novo at the stage of assessing whether a prima facie case has been established or not.

  7. Charanjit Sehgal

    I understand that patent 774 pertains to erlotinib molecule without any relationship to a specific polymorph. In such a scenario any solid crystalline material of any polymorphic form will be within the scope of the claim of 774 patent. On the contrary 507 patent pertains to a specific polymorph B. as regards novelty there is no relationship despite the fact the 507 might be predated to 774 in terms of date of filing. However, non citation of 507 in 774 and also not sharing the information about 507 during prosecution of 774 may call for invalidation under undue conduct. As per the understanding it may be perceived that the patent can be considered for invalidation and if patent is considered valid by the court then use of any polymorphic form infringes the 774 patent

  8. Shamnad,

    Unfortunately the judge failed to understand and even comprehend simple and pretty straight forward facts pertaining to pharmaceutical patents and that also after hearing the matter for four years. Such facts even a fresher in Pharma comprehend well in 4-6 weeks (smarter guys may comprehend in couple of days).

    How ridiculous it sounds – patent claiming compound erlotinib hydrochloride per se stands valid but not infringed by generic manufacturer merely because the generic product contain FORM B variant of erlotinib hydrochloride, not erlotinib hydrochloride. And this bizarre decision was hailed as landmark ruling by few (obviously with vested interests).

    This decision is not technically flawed and reflection of poor and miserable state of our judiciary in even straight forward patent matters. Nowadays it is not unusual to see lows in every sphere of India be it governance, policy making or judiciary.

    I too second to Sai about the importance of having good counsel arguments but what if the judge considerably lack technical blend of mind to even acknowledge basic technical parameters.

  9. If the judge would have the basic understanding of pharma patents then this case doesn’t even require much arguments on infringement issue. The only issue that needed discussion and attention was the validity of patent. If patent found valid then generic product was infringing and if invalid then there would have been no infringement.

  10. Great Article. Thanks for the info. Does anyone know where I can find a blank patent cover form to fill out?

Leave a Comment

Scroll to Top