In a previous post, I’d promised to bring a more detailed analysis of the problematic patent decision in the Sovaldi (Sofosbuvir) post, wherein Gilead won a hard fought patent battle against some 8 odd opponents. Apart from the rather thin analysis on section 3(d) which I highlighted, the Deputy Controller also appears to have conflated the novelty and inventive step test frameworks. Essentially, he’s imported the “novelty” test framework into the “non obvious” or inventive step analysis. Or pioneered a rather “novel” frame for assessing whether the claimed invention is “non obvious’ or “inventive”.
Novelty vs Inventive Step (Non-Obviousness)
To save my time and yours, let me simply reproduce what I’d written in response to a comment received on the earlier post:
“You’re right. I was a bit harsh on him (Rajesh Dixit, the Deputy Controller) initially in my email message to subscribers. I tempered down a bit while posting to the blog. But am still appalled at the way in which the section 3(d) argument was dealt with. If he thought that section 3(d) was wholly inapplicable (since the earlier known substance [from Merck’s patent] was not really a “derivative” or that it was an “imaginary” compound (as he called it, borrowing liberally from the Gilead submissions filed before him), then he should have stated so explicitly and left it at that. And not ventured out and still held that there was an “added layer of efficacy”. Efficacy when compared with what? And even his labelling of the prior known compound (in Merck’s prior art patent) as “imaginary” and not “enabled” is hugely problematic. How did he arrive at this factual finding–for you can’t simply state that something is not enabled just because a party to your proceeding (in this case Gilead) says so. He has to reason out and make a factual finding on this count (based on expert testimony etc) that this prior known substance (listed as one of the embodiments in Mercks’ patents) was not enabled. And cannot therefore be considered for section 3(d). Unfortunately, no reasoning at all on this count!
His inventive step analysis is even more problematic. For he essentially does a “novelty” framework analysis here, holding that since there is no mention of fluorination in any of the prior art documents cited, this alleged invention is “inventive” or non obvious. But if the precise “fluorine” (2 down position) was in fact mentioned explicitly in a prior art document, then this would go to “novelty”-i.e. the claimed compound would be fully and completely anticipated! He says that a person skilled in the art would not be able to obtain this claimed compound without experimentation. But mere experimentation does not confer inventive step. Rather all it does (if the experimentation qualifies as “undue”) is to take it out of the anticipation challenge and vest it with “novelty”.
Section 3(d) and Burden of Proof
More worryingly, the Controller also states that the opponent has not been able to demonstrate that the allegedly “imaginary” compound in the Merck patent works as stated and therefore qualifies as a known substance? Why should the opponent have to prove this? The burden of proof for section 3(d) is on the patent applicant; not the opponent! Sadly all this was lost on the Controller, leaving huge gaping holes in the decision. Making this a sitting duck for an appeal. Also on the technical expertise front, I believe he has a chemistry background. And reading the decision also suggested to me that he had some degree of proficiency in the underlying science. Wish I could say that about the law or his powers of reasoning. But hopefully with more such cases thrown his way and with more legal/judicial training, we will see better from him in the future. I don’t mean him any ill will. And perhaps he is a mere pawn in a larger issue of institutional malaise where we simply don’t have enough resources at the office and leave our officers without significant legal or judicial training. And burden them with a huge case load! But then again, this is a hugely important case for public health and to have it reasoned so poorly makes ones blood boil! For this is a decision that will ultimately impact the lives and well being of several thousand patients and the future of the innovation ecosystem as well (given that this is one drug that has finally woken up US law makers to go after drug pricing!). All in all, a wasted opportunity for developing and clarifying patent law jurisprudence.”
Amendment Without Notice to Opponents
Apart from the above, the Deputy Controller also permits Gilead to amend their patent specification at the last minute without notifying the opponents, a ground that has been taken up in a writ challenge by IMAK and others who’ve challenged the problematic patent decision.
Merck vs Gilead: Whose Drug is it Anyway?
Meanwhile across the shores, a very interesting development is spewing. As some may be aware, the key prior art relied on by opponents to challenge Gilead’s patent application is a published patent by Merck. Merck sued Gilead (claiming that Sovaldi infringes upon this patent) and won in the US (at least at the initial stage, with the jury awarding Merck 200 million dollars, after the patent was found to be valid. Gilead conceded on the issue of infringement). It now turns out that Gilead has submitted evidence that this patent by Merck may have been built upon confidential disclosures by Pharmasset researchers to a Merck scientist. A judge found prima facie merit in this contention by Gilead and permitted them to bolster this claim by submitting additional evidence. Remember that the Sovaldi patent initially belonged to Pharmasset which was then taken over by Gilead. More on this in the next post. Stay tuned.
13 thoughts on “The Sovaldi Saga: A “Novel” Non-Obvious Standard?”
I think this is not good analysis. Controller has given his reasonally good analysis for novelty, inventive step and Section 3 (d). I have seen many orders from IPO, and to me it looks detailed and good one. Earlier controller, Mr. Karar allowed this case on novelty and inventive step but refused on Section 3 (d). This controller has taken all factors into account. I suggest you to read his whole Section 3 (d) analysis, you will appreciate that controller had mentioned what is a known substance under section 3 (d) as identied by Opponent, and how gilead has met this requirement in view of Glieevec and Roche decision.
I think you are taking it too far without reasonable analysis. This is not fair and look like putting pressure on govt officials on post grant proceedings. Very unfair to Dr. Dixit, a well recognised, experienced and respected Controller.
Please don’t mix public issues with patent law. I feel there are many Indian generic companies who can still use multiple provisions under Indian patent act such as post grant, revocation and compulsory licence etc. Further, we have law such as drug price order act etc which can deal with this pricing issue. Further government has role to provide insurance and other facilities to people of India. They can’t make innovators look like villan all the times.
Requesting you to be fair in your analysis and avoid mixing issues as you have done here.
Indian Patent Attorneys
Well, what can I say? Apart from going on and on about how good the Controllers’ decision is (“reasonally good analysis as you put it), you provide no logical basis for your rather conclusory statements. How is the order “reasonally good”? I had pointed out at least 5 defects with the order (comprising reasoning around section 3(d) and the problematic conflation of novelty and non obviousness, quite apart from the last minute amendments that were sneaked in without notice to opponents). Do provide similar substantiation for your points. Else you will be wasting your time, mine and those of our readers. Thanks!
I have a question, not exactly related to the “criticism” above, but just a food for thought, but that does hinges on how 3d continues to get interpreted in light new/novel drugs (“novel” here used in a broader/layman term sense, and not as used in patent law).
Typically the way much “drug discovery” works is that known, older molecules are researched on. Different “parts” of a molecule are “tweaked” – a methyl group here is replaced by isopropyl, a different glycosyl group out there, phosphate added/removed from”position 3″ etc. Or put another way, new molecules/drugs are often NOT discovered from scratch (the latter would be more in the nature of “breakthrough”/”landmarks” etc.). However for various, and rather obvious reasons (such, competition), patents covering such new molecules need to be filed asap. But at the time of filing, the understanding of the molecule, viz. its entire gamut of chemical and biological properties, would still be in a relative early stage. For example, from the initial studies, a molecule may be shown to have anti-proliferative effects on in-vitro cell lines, or binding to and “switching on” a “G-receptor” that triggers a specific “signal-transduction pathway”. All these are very promising leads, and involves a great deal of research, but still quite far away from even a providing a hint on what the actual indications are likely to be, it’s safety, efficacy (including vis a vis existing approved molecules etc.). Much of this latter information will only start getting available when tox, animal and human safety and efficacy studies get initiated (and as you would be aware, most potential drugs fail this).
My reason for this detour on how new drugs are discovered/researched is that the initial patent filings, covering a molecule, are quite unlikely to have information on how a new molecule is a “real improvement” over another one, from the section 3d stand point. In other words, molecules that pass the novelty and inventive step criteria (now I mention these from the patent law perspective !!!), may still fail 3d, for sheer lack of information at the time of filing,, unless all drug discoverers are ready to wait many more years, post the “discovery stage”, to file patents !!!!! And this, as you imagine, is unlikely to happen.
Any thoughts on this ?
The history of science and technology reveals that often times inventions occur in parallel. With different inventors in different parts of the world coming up with the same idea/invention. Let the best man or woman win by reaching the patent office first. But only after they have adduced enough evidence of the utility/efficacy of the claimed substance (after section 3(d). So long as the goal post is the same distance for all, this should work okay. Perhaps we might delay the patent publication by a couple of months or a year, but far better than wrongly granting to speculative utility.
It doesn’t really work that way “on the ground”. 1st, given the nature of publish/patent or perish world, you need to do the formar as soon as you get sufficient results. Otherwise you are a goner!!!! And in basic biomedical research, a clear “proof of concepet” may sufficient. This , for ex, could be a showing in an in vitro system, or a mouse transgenic model. You need not neccessarily have to show effficy in humans to get a solid publication, say in Nature Medicine. And btw, most institutions/labs are not equiped to initiate clinical studies. My point, all this may be more then enough to get a great publication/s, pass novelty/ inventive step requirement, yet fail 3d !!!! Please be aware, from a basic research lab to a clinical setting is not just “couple of months” or “a year”. We are talking of “a/about a decade” !!! My apologies if I sound condescending, but this sir, is the realty of biomedical research.
please read my response again. i specifically said: that once the goal post is teh same for all, your argument does not hold. If no patent application will stand good without some basic establishment of efficacy (and this need not be clinical trial proof–as i’d argued in one of the first pieces on section 3(d)), then none are disadvantaged. they will all wait to establish some efficacy before rushing to the patent office. LIke I said, you dont need clinical trial efficacy to establish efficacy under section 3(d). I would argue that even in vitro lab testing and some decent results should suffice. But your comparator should be the right one (prior known substance) and your data should be reasonably convincing. “Things on the ground” respond well to sensible laws, as I’ve always found. So lets not take it as a “given” that can never be changed or one that should always be deferred to.
Here is another follow up thought. A corollary to the above, in future a “molecule” may be granted patent, long before it is approved, before its market price decided by companies, and by extension, before it becomes “blockbuster”. Or to put it another way, long before we realize that the drug is “out of reach” of most people !!!
Will we then use hindsight 20-20 to knock of a patent ?!!!! That is deciding it should never have been granted (because it is “out of reach”), and hence should now be revoked !!!
I really don’t know the answer to this. So again, any thoughts ?
but pricing is never a consideration for the grant or non grant of a patent…the supreme court also pooh poohed this argument in the novartis glivec case. this patent (over sovaldi) should be judged simply on technical merits. was it inventive over and above the merck molecule or would the skilled person have tried an obvious array of permutations and combinations to arrive at it?
Yes this is so. But you know, human nature often risks conflating the two 🙂
Btw, sorry about the typos in my earlier “reply”. The result of trying to use stubby fingers over a tab key board !!!!
I am responding to your response just a mintue ago (since I don’t see any other options).
I am afraid that I have to maintain that the reality of biomedical research is different, and cannot be moulded by one country’s laws, no matter how sensible
Well, lets just see how this pans out in future. Interesting times !!!!
Unfortunately the goal post is at a farther distance in India than other countries for the same invention in view of Sec 3(d). Would one then wait to generate the efficacy data and not file patent application early once proof of concept is established? That will be a bigger risk….it will be safer not to file in India.
on another note with regards to comparison of efficacy: let’s say there is a “X” molecule with modifications designated as “A”, “B”, “C” and “D” for which a patent has been filed and granted. The specification provides efficacy data of “X” with modification “A” (let’s say X-A) but not for all the others which are covered as embodiments (X-B; X-C; X-D). If one files a patent application with additional changes to the molecule covered by one of the embodiment (let’s call this X-C-Z) against which prior art molecule one need to show enhanced efficacy, X-A or X-C? There is no efficacy data in the prior art for X-C and the molecule(X-C-Z) is closer to X-C.
Sec 3(d) will throw up many such challenges. My view is that there should be some ways to incentivise such inventions which truly impacts patient healthcare either in terms of efficacy or other benefits.
Thanks for your substantial arguments.
Just one question in respect of your Sec. 3(d) argument that the opponent does not have to prove that the compound at issue qualifies as a known substance: It is my understanding that the burden of proof in an opposition proceeding (Sec. 25(1)) is on the opponent. The evidential burden may shift according to the state of the evidence. It seems to me that Controller held that the burden did not shift although that is not mentioned expressis verbis. Therefore, it is my understanding that according to Sec. 25(1)(f) the opponent has to prove that subject of any claim of the specification is not an invention within the meaning of the Act and consequently it has to prove that the invention is within the prohibition of Sec. 3(d). Do you disagree?
Thought-provoking piece ! I learned a lot from the insight ! Does anyone know where I might be able to locate a sample DD 689 form to edit ?