US Enablement Case: Relevance for Novartis Patent Case in India

Harold Wegner’s timely email updates brought an excellent opinion by Judge Kimberly Moore to my attention. I’ve sat in her court once and she is fabulous (a razor sharp mind combined with a keen sense for the broad policy issues–she’d been an academic prior to her career as a judge). Hal’s email states:

“Today in Pharm. Resources, Inc. v. Roxane Laboratories, Inc. (Fed. Cir. 2007), (Moore, J.), in her first important pharmaceutical opinion, the newest member of the court authored an affirmance of a summary judgment of invalidity under 35 USC § 112, ¶ 1, against a claim to “[a]n oral pharmaceutical composition in the form of a stable flocculated suspension in water comprising: (a) megesterol acetate; (b) at least two compounds selected from the group consisting polyethylene glycol, propylene glycol, glycerol, and sorbitol; and (c) a surfactant.”

Citing Liebel-Flarsheim Co. v. Medrad, Inc., 481 F.3d 1371, 1380 (Fed. Cir. 2007), the court said that the patentee “sought extremely broad claims in a field of art that it acknowledged was highly unpredictable, therefore, [the patentee] has set a high burden that its patent disclosure must be to satisfy the requisite quid pro quo of patent enablement.”

(To receive Hal’s email updates, you can email and request him).

The judgment is a very lucid one–and more importantly, is very concise and to the point –rare for a US patent judgment. Perhaps our Madras HC judges (who wrote a good 40 pages in the Novartis patent case without much lucidity or analysis) could learn a lesson or two!!.

For those interested in this decision, please see here. Also, Patently O has posted the following summary of the case:

“Par’s megestrol acetate suspension is prescribed to people who lose their appetites — often during treatment for cancer or AIDS. Interestingly, the company created this popular drug while attempting to design around a megestrol patent owned by BMS.

After obtaining patent protection, Par sued Roxane for infringement. The district court, however, granted summary judgment of invalidity under 35 USC 112 ¶ 1 — finding that “Par is not entitled to the broad claims it asserts.”

Enablement analysis begins with the presumptions that an issued claim is enabled and that a challenge to enablement requires clear and convincing evidence. Unlike its close analog written description, enablement is reviewed by the CAFC on a de novo basis.

A claim is enabled when a PHOSITA can make and use the claim without undue experimentation. Broader claims, of course, require broader disclosure to ensure that their “full scope” is enabled. The unpredictable nature of an area of technology often serves patentees well as they argue non-obviousness. (It cannot be obvious if the results could not have been predicted). However, the enablement requirement demands more disclosure for unpredictable arts.

Broad Claims: Here, Par’s claim includes the following elements: “(a) megestrol acetate; … and (c) a surfactant.” The CAFC finds this claim very broad because it would “allow the choice of any surfactant in any concentration.” (emphasis in original). In its disclosure, Par described three working examples and only one new surfactant. As a matter of law, the CAFC found that “these three working examples do not provide an enabling disclosure commensurate with the entire scope of the claims.”

So how is this relevant for our pending Novartis patent case? Those that have looked at the facts carefully will notice that there is a an enablement issue here. Novartis’s invention consists of the following steps:

i) Synthesizing the Imatinib free base, a compound that was patented in the US, EU and several other countries. However, this could not be patented in India, owing to the fact that in 1993, India did not provide product patents for pharmaceutical substances.

ii) The conversion of the free base to a salt form — imatinib mesylate by adding methanesulfonic acid.

iii) The obtaining of the beta crystalline form of imatinib mesylate, a polymorphic form that is allegedly the most stable form of the salt. A patent application was filed for this and it is this application that is the subject matter of dispute in India

iv) The making of Glivec, a drug that is based upon the above beta crystalline form of imatinib mesylate.

Novartis claims that Glivec (beta crystalline form of imatinib mesylate) is more effective than the Imatinib free base, since it displays better bio-availability properties i.e. it is absorbed more easily into the blood.

Novartis claims all acceptable pharmaceutical salt forms in its main 1993 patent (step (i) above). This would ordinarily include Imatinib Mesylate as well. The question however is: does the 1993 patent “enable” the making of imatinib mesylate? (step (ii) above). I personally think it does–but am happy to be corrected. If it does not, then imatinib mesylate cannot form part of the claims. It also cannot taken to be a “known” substance. Section 3(d) requires that the new form demonstrate increased efficacy over the “known” substance. If Imatinib Mesylate is not a “known” substance, then the comparison under section 3(d) cannot be with Imatinib Mesylate. And to this extent, the case against Novartis on section 3(d) becomes a weak one.

However, as I said earlier, I personally think the patent enables the making of Imatinib Mesylate and the case on section 3(d) therefore still remains a strong one. Would be interested to hear what readers think.

Shamnad Basheer

Shamnad Basheer

Prof. (Dr.) Shamnad Basheer founded SpicyIP in 2005. He's also the Founder of IDIA, a project to train underprivileged students for admissions to the leading law schools. He served for two years as an expert on the IP global advisory council (GAC) of the World Economic Forum (WEF). In 2015, he received the Infosys Prize in Humanities in 2015 for his work on legal education and on democratising the discourse around intellectual property law and policy. The jury was headed by Nobel laureate, Prof. Amartya Sen. Professional History: After graduating from the NLS, Bangalore Prof. Basheer joined Anand and Anand, one of India’s leading IP firms. He went on to head their telecommunication and technology practice and was rated by the IFLR as a leading technology lawyer. He left for the University of Oxford to pursue post-graduate studies, completing the BCL, MPhil and DPhil as a Wellcome Trust scholar. His first academic appointment was at the George Washington University Law School, where he served as the Frank H Marks Visiting Associate Professor of IP Law. He then relocated to India in 2008 to take up the MHRD Chaired Professorship in IP Law at WB NUJS, a leading Indian law school. Later, he was the Honorary Research Chair of IP Law at Nirma University and also a visiting professor of law at the National Law School (NLS), Bangalore. Prof. Basheer has published widely and his articles have won awards, including those instituted by ATRIP, the Stanford Technology Law Review and CREATe. He was consulted widely by the government, industry, international organisations and civil society on a variety of IP issues. He also served on several government committees.


  1. AvatarSushant

    Dear Shamnad,

    Great analysis of Novartis patent. I actually searched yesterday night to get a hold of Gleevec patent in US but couldn’t find one. Can you tell me how you actually found these details on Novartis patent?

    A quick note on this post. “Enablment” requirement of a patent is not directly linked with
    the non-obvious requirement. Quoting from
    “The information contained in the disclosure of an application must be sufficient to inform those skilled in the relevant art how to both make and use the claimed invention.”

    The idea is that protection in the form of patent is provided for disclosing details for making the compound (and it is not kept trade-secret). Pharma Resources lost the case because it did not disclose sufficient details to manufacture what was claimed. And the judge ruled that “these three working examples do not provide an enabling disclosure commensurate with the entire scope of the claims.”

    The judgment actually links the enablement and the non-obviousness requirements of a patent. As she writes: “The unpredictable nature of an area of technology often serves patentees well as they argue non-obviousness.”

    So here I am a bit confused. You are saying that Novartis patent can be rejected because

    1. details for making the salt was already disclosed in 1993 patent.

    2. it can be rejected because they claimed all salts but did not disclose making it

    3. it is obvious to construct a more stable salt from the free base. This is different from enablement requirement.


  2. AvatarTahir


    I know we’ve discussed this issue in the past, but your post and the judgement made me re-visit the question you raise. I have the following thoughts.

    Before going into the point of whether the 1993 patent is an enabling disclosure for making imatinib mesylate (IM), the case Pharm. Resources v Roxane Labs Inc refers to the case Re Wands 1988 and 8 factors relevant for considering enablement which I think are as useful a guide I have seen for this topic. The 8 factors being:

    (1) the quantity of experimentation necessary, (2) the amount of direction or guidance presented, (3) the presence or absence of working examples, (4) the nature of the invention, (5) the state of the prior art, (6) the relative
    skill of those in the art, (7) the predictability or unpredictability of the art, and (8) the breadth of the claims.

    On the point of the 1993 patent -US 5521184- (‘184) the key parts of the specification that lend themselves to an enabling disclosure and anticipation (and as argued by the parties in the actual case) are:

    1. Col 3, lines 22-43 – list over 35 suggested acids as salt forming groups for the compound (formula 1 – Imatinib). Line 39 specifically mentions methane as a possible aliphatic sulfonic acid.

    2. Col 19, lines 53-58 state that ‘acid addition salts of compounds of formula I can be obtained in customay manner e.g treatment with an acid or suitable anion exchange reagent.’

    3.Claim 23 – ‘the compound according to claim 1 of the formula I or a pharmaceutically acceptable salt thereof. (It is noticeable that claim 1 uses the wording ‘or a salt of such a compound having at least one salt forming group).

    What is noticeable in the ‘184 patent is that none of the examples exemplify how to make any of the salts – but only the free form. Indeed, this point is specifically mentioned by Novartis (as it would) in the later patent for the B-crystalline form (see under the heading ‘Background to Invention’). This seems to eliminate one of the factors for enablement in Re Wands i.e No.3 the presence or absence of examples. However, that might not be critical (though it was in the Roxane case) as there is ample case law e.g Synthon v SKB and Bayer Diastereomers EPO/TBA 1982 to suggest that common general knowledge in the art may be enough to fill the gaps in a disclosure that is not explicit.

    I think the argument of whether ‘184 is an enabling disclosure will depend on how the IPAB views, despite the lack of explicit disclosure, whether a skilled person would know how get a salt for the compound of formula I in question. Col 19, lines 53-58 seem to suggest that a salt can be obtained in a customary manner. Novartis (and the pharma industry as a whole) may argue that getting a pharmaceutically acceptable salt of a compound is an unpredictable exercise in light of the number of possible salts possible (and given that Novartis listed a possible 30-40 salts in ‘184). However, there is ample prior art dating from 1977-1995 that goes against such an argument.

    This question seems a fine line between enabling disclosure of an anticipatory form and whether something is obvious. It will be interesting to see what the IPAB says. If it says getting a salt is not an unpredictable art – it could bode well for all the salt selection patent oppositions e.g TDF.

    I also note from the Roxane decision how the wording in the claims ‘a surfactant’ was considered too broad because it could mean any on of 100 surfactants. Could that be the case with the claims in ‘184 and the language ‘or a salt of a compound’ ‘or a pharmaceutically acceptable salt’? If so, then may be Novartis would not be able to claim infringement of its 93 patent if someone made IM (but that is a moot point at least for India) – though it might help them in this case.

    Tahir Amin (

  3. Avatarmadhav

    Dear All,

    Does anyone have any idea of what salt was applied for FDA approval and when? which was the salt in the clinical trials first used by Novartis anywhere in the world? This could probably throw some light on the issue.


  4. AvatarTahir

    Just to add, it is worth noting that when the EPO was examining the B-crystalline patent, the examination report held that the ‘184 patent disclosed the methanesulfonic acid additional salts for formula (I) – but the application was considered novel because the ‘184 patent did not disclose the b-crystalline form.

    Tahir Amin (

  5. AvatarShamnad Basheer

    Dear Tahir,

    As always, thanks for the incisive and detailed analysis. I don’t see anything in US “enabling disclosure” law to rule out a consideration of “common general knowledge”–i.e. if salt selection methodologies were known in the art, then not specifically mentioning them in the patent application will not prejudice a claim that the application does not enable. If this were not the case, then one could make the argument that unless every aspect of science (including aspects well known to skilled people in teh art) were disclosed in the application, one could never moot an “enabling disclosure” objection… which case, we may as well bury such a requirement and not have it at all…

    Very interesting point you raise on the ’93 patent and whether its claim covering “all pharma acceptable salts” is liable to be struck down since none of hte salts are “disclosed” in an enabling manner (under the Roxanne ruling). I honestly dont think this will make a difference to Novartis worldwide, as the ’93 patent covers Imatinib and the moment someone makes a generic version of IM, they are anyway infringing the “Imatinib” patent.

    As for India, the ’93 patent does not exist….so the question of enabling disclosure will not matter –except to the extent that we wish to analyse what the “known” substance is under section 3(d).

  6. AvatarShamnad Basheer

    Dear Madhav,

    My understanding is that Gleevec contains the “beta crystalline version of Imatinib Mesylate” and that this was the substance that was put through clinical trial testing.

  7. Avatarmadhav

    if imatinib mesylate was used in trials before 1st January 1995, then no question of granting patent in India arises. Some document disclosing that could be useful.

  8. AvatarTahir


    I agree that if something is common general knowledge then there is no need to specifically explain the steps/directions for carrying out for it to be enabling. Question is whether getting a salt of the free base imatinib is common general knowledge that requires no directions for one skilled in the art. I think it is based on practices in the field and literature available – and lines 53-58 of Col 19 of ‘184 seem to suggest that.

    But lets say for argument sake the IPAB were to find that the ‘184 patent does not enable IM. Would it still then be possible to compare the B-crystalline form to IM for determining efficacy by arguing that because it would have been obvious to make the methanesulfonic acid (even if not enabled by ‘184) that IM was still a known substance? There are some advocates out there that have adopted this line i.e that ‘known’ implies ‘obvious. I think such an interpretation (as useful as it would be for patent challenges) is a stretch too far. Be interested to hear thoughts on this.

    Your’re absolutely right that the 93 patent would capture anyone making IM – even if the all salt claims were considered too broad.

    Tahir Amin (

  9. Avatarvanshika

    1. Novartis claimed free base (imatinib) in ‘184 patent but generically claimed the salts. Typically the enablement is proved if by reading the patent specification the person ordinary skilled in an art was able to reduce the invention to practice or not without undue experimentation. If the patent application has not explicitly mentioned the synthetic route and the best mode then it should not be considered as enabled.

    2. Pls. correct me if I go wrong. Just a thought came to my mind: Why are we debating on whether the imatinib mesylate is enabled in the patent to prove that it is the known substance? Shamnad, as you already stated that the clinical trial were done on the said salt only and the drug Gleevec contains imatinib mesylate salt which patients are consuming thus it itself means that the substance is known irrespective of the fact whether it is claimed (generically or specifically) and enabled in the patent.

  10. AvatarShamnad Basheer

    Dear Madhav and Vanishka,

    In most cases, drug trials are conducted only after the patent application is filed. The relevant patent application covering the beta crystalline form was filed only in 1998. To the best of my knowledge, drug trials on gleevec (which embodies the beta crystalline form) was conducuted only after this patent was filed. So “beta crystalline” form and even imatinib mesylate were not “known” to the public prior to 1998.

    Even assuming the “clinical trial” testing preceded the application for patent, in some cases it will not prejudice the “novelty” of the patent application, since it falls under “experimental use” that is not “known” to the pubic.

    On another note: it is because patents are typically filed prior to initiating clinical trials, that it will be too onerious a requirement if we insisit on “clinical trials” as proof of efficacy under section 3 (d).

  11. AvatarShamnad Basheer

    Dear Vanshika,

    So long as a skilled person in the art can arrive at Imatinib Mesylate by reading the ’93 patent and using common general knowledge, it will be taken to be “enabled”, whether it is procured using the “best mode” or not. And if it is so enabled, then it will be taken to be “known” under section 3(d).

  12. AvatarShamnad Basheer

    Dear Tahir,

    You’re absolutely right– I don’t think something that is only “obvious” from existing prior art should be construed as a “known” substance under section 3(d).

    As you rightly say, such a reading would be a stretch too far and would militate against the known meaning of “known” 🙂

  13. Avatarvanshika

    Dear Tahir and Shamnad,

    Well it will be right to imply “known” as “obvious” in some inventions, wherein there is some inherent characteristic of the substance.

    For example, in a very well known case of Loratidine, the patent of desloratidine was revoked on the grounds that it is an inherent characteristic of loratidine to get converted into major metabolite desloratidine after injestion. The inherent characteristic of a substance make the end product obvious and hence novelty destroying.

    While in Novartis case the term “obvious” may not be exchanged with “novel” as the third party would have to play real chemical gymnastic to arrive at the pharmaceutically acceptable salt which is most stable over the other. Though in various salts of the same drug the main pharmacophore is the same but the biological activities of said salts may vary alot. For example in case of amlodipine, amlodipine maleate is sticky, less stable and is less bioavailable as compared to amlodipine besylate.

  14. AvatarHafiz Aziz

    Thanks Shamnad for providing us with such an eloquent analysis continuously. We have been discussing Novartis judgement mainly from Patent law perspective and understandably so as this discussion is going on IPSpicy. But, I am just curious to know about an international appraisal of Madras High Court judgement, especially, keeping in view the historical background of application and treatment of international law in Indian Courts. In Novartis, High Court has very easily waived its jurisdiction without any tear! Is it something compatible with earlier Indian case law and what are the future implications in this regard? Please just refer me somewhere if such kind of debate is going on Novartis case from IL perspective.

  15. AvatarTahir

    Dear Vanshika,

    I think its important not to confuse something that is inherent with the question of whether something is obvious/inventive. The example you give of active metabolites would usually fall under the category of inherent anticipation (otherwise novelty) – in other words something already existing or known to exist in the body even if not enabled/disclosed. I believe that was the case with loratidine. See also the House of Lords in Merrell Dow v Norton and the issure of the metabolite of terfenadine.

    As for the point you make re salt selection – there are cases where getting the best salt may not be ‘known’ at the outset even if disclosed in a group of possible salts. That said, most skilled formulators working in labs will have tested various salts against all sorts of base compounds during their work experience (i.e basic compounds like nucleotides, compounds with poor water solubility, hygroscopicity etc etc) and would be able to narrow the most likely salts to formulate the drug with without having to go through a huge batch. Indeed, there is substantial literature providing endless guidance on salt selection from 1977 for different types of compounds. That said – and as I mentioned above- that might be more a case of obviousness than novelty. Simplying listing a salt may(emphasis) not be enabling (and therefore known for the purpose of 3d) – but it has a high likelihood of being obvious. I think it really depends on the base compound in question and its properties – a lot can be understood and determined from that.

    Tahir Amin (

  16. AvatarShamnad Basheer

    Dear Hafiz,

    Along with Prashant, I am doing an article exploring the constitutional law/international law angle of the Mad HC judgment. I will send you a copy soon –if you give me your email.

  17. Avatarvanshika

    I have a doubt.
    If I file a provisional patent application in IN and within one year from the date of filing such application I file PCT application (US as receiving office) and designate IN and wish to enter national phase in India after the expiry of 31 months. Would my application be considered as abandpned in IN because as per section 9(1) I have not filed complete spec in IN before the expiry of 12 months?

    Also, what will happen in the above case if I file PCT application wherein IB is receiving office.


Leave a Reply

Your email address will not be published.