21 April 2012
Interview with Shamnad Basheer, Intellectual Property Law Professor at NUJS.
SHAMNAD BASHEER: “Drug companies have been reluctant to disclose their costs of drug discovery and development to the public.” PROFESSOR Shamnad Basheer joined the National University of Juridical Sciences (NUJS), Kolkata, in November 2008 as the first Ministry of Human Resource Development Chaired Professor in Intellectual Property Law. Before this, he was Frank H. Marks Visiting Associate Professor of Intellectual Property Law at the George Washington University law school and a research associate at the Oxford Intellectual Property Research Centre (OIPRC). He is the founder of several initiatives, including SpicyIP (one of India’s leading intellectual property, or IP, blogs), IDIA (an initiative to foster access to legal education for the underprivileged) and P-PIL (an initiative to promote public interest lawyering through synergies between legal academia and the legal profession).
Basheer graduated from the National Law School of India University, Bangalore. He did his postgraduate studies at the University of Oxford, where he completed his BCL and M.Phil with distinction as a Wellcome Trust scholar. He was nominated recently as an expert on the IP global advisory council (GAC) of the World Economic Forum (WEF).
In this interview to Frontline, Basheer explains several issues of relevance to the drug prices in India.
In your written submission to the Supreme Court in the Novartis case, you have supported the Madras High Court’s and the Intellectual Property Appellate Board’s (IPAB) interpretation of Section 3(d) of the Indian Patents Act. Do you hold the view that its phraseology is correct? Critics have said that it is vague.
The Madras High Court was right in interpreting “efficacy” to mean therapeutic efficacy, but its reasoning on this front was rather peripheral and could have been stronger. It did not weigh in more significantly on this theme, since the issue of interpreting efficacy was never directly before the court. Rather, it was adjudicating a constitutional issue: was Section 3(d) so vague and ambiguous as to violate Article 14? The court, in this context, expressed some views on Section 3(d), which could arguably constitute obiter and not the real ratio.
In my intervention in the Supreme Court, I have tried to present stronger grounds for relying on therapeutic efficacy as the appropriate standard. I rely primarily on the structure of Section 3(d) as also its parliamentary history to support my interpretation. I note as below:
“The structure of Section 3(d) as also its legislative history supports a narrow reading of the term ‘efficacy’. Illustratively, the Explanation to Section 3(d) clearly states that all pharmaceutical derivatives would be considered the same ‘substance’, unless ‘ they differ significantly in properties with regard to efficacy’.”
The above clause refers to only those “properties” that have some bearing on “efficacy” and not all properties. If “all properties” were to qualify, it would effectively render the term “efficacy” redundant. A statute cannot be interpreted in a manner as to render any of the terms in it redundant. Had Parliament intended “any property” to qualify under Section 3(d), the Explanation would simply have stated “unless they differ significantly in properties”. And the main part of Section 3(d) would have been rephrased as “the mere discovery of a new form of a known substance which does not result in the enhancement of the known properties of that substance”.
Therefore, not all advantageous properties of a new form (such as improved processability or flow characteristics, storage potential, etc.) ought to qualify under Section 3(d) but only those properties that have some bearing on efficacy.
Although this precise line of argument pointing to the phrase “properties with regard to efficacy” does not appear to have been explicitly made by either the Madras High Court or the IPAB to support their conclusion, it is one that compellingly supports a restrictive interpretation of the term “efficacy”.
This interpretation is further buttressed by the objectives of the Act, which suggest that Section 3(d) was introduced to prevent ever-greening.
Although the term evergreening does not have a scientific definition as yet, it is widely understood to mean an inappropriate extension in patent monopoly which does not convert to a significant benefit for the patient.
Put another way, it is a patenting strategy “consisting of acquiring patents on minor, often trivial, modifications of existing pharmaceutical products or processes in order to indirectly extend the period of patent protection over previously patented compounds”.
Do generic knock-offs stifle innovation by drug makers?
This issue essentially depends on the costs associated with bringing a new drug to the market. What is absolutely shocking is that despite the centrality of this question to debates around pharmaceutical innovation, we are still uncertain as to what the true cost of bringing a new drug to the market is. Drug majors have been extremely secretive about this in the past and the only study to have addressed this issue is a highly contested one. I deal with this issue partly in my recently completed PhD thesis and note:
A pharmacologist checks the reaction on a nude mouse after applying the SNB cytotoxic drugs for anti-tumour cancer, inside the biosafety cabinet at Natco Research Centre in Hyderabad on March 13. India’s move to strip German drug maker Bayer of its exclusive rights to a cancer drug has set a precedent that could extend to other medicines, including modern HIV/AIDS drugs, in a major blow to global pharmaceutical firms, experts say.
“The most cited study in this regard (hereinafter “DiMasi study”) estimated that it would take approximately U.S. $802 million to produce a marketable drug. These estimates have increased, with the most recent figure amounting to approximately U.S. $1.3 billion. However, given that drug companies have been reluctant to disclose their costs of drug discovery and development to the public, these costs remain a highly contested issue.
“In 2001, Public Citizen, a civil society group in the U.S., contested the above figures on several grounds, the most pertinent of which are highlighted below:
i) R&D costs should be regarded as an expense and not an investment. Therefore, the costs of capital are irrelevant.
ii) The estimated cost did not take into account the decrease in Food and Drug Administration (FDA) review time, lower clinical trial periods, and the contribution of new technologies such as genomics and combinatorial chemistry, which are believed to have assisted in the lowering of costs involved in creating drug leads.
iii) If one were to assume the veracity of the pharmaceutical industry’s self-reported total R&D figures between 1994 and 2000 and divide this amount by the total number of approved drugs (after controlling for time lag), one would arrive at a figure of U.S. $108 million per new drug before tax benefits and U.S. $71 million after, which is significantly lower than what the DiMasi study cites.
“Donald Light, Professor of comparative health care at the University of Medicine and Dentistry of New Jersey, United States, additionally critiques the study on the ground that the numbers are based on a small sample of large pharmaceutical firms which were non-randomly selected, and include only new molecular entities (hereinafter “NME”), the most costly sub-group of pharmaceuticals which constitute only one-third of new drug approvals. He argues that if the costs of other “incremental” drugs (constituting three-fourths of all drugs) were taken into account, the average cost of a drug would fall to U.S. $400. Light also offers several other correctives to argue in favour of a far lower cost estimate for clinical trials. First, the number of subjects involved in clinical trials as per FDA data was about one-third or less than those used by DiMasi, suggesting that the rest of the trials were done by firms to primarily bolster their marketing materials. Second, he cites data from the National Institutes of Health (NIH) indicating the costs per trial to be only a quarter of the figures used by DiMasi.”
Importantly, the costs of R&D will differ from drug to drug, depending on the complexity of the science involved, the intensity of clinical trials, etc. The current patent system, which provides a uniform patent protection of 20 years to all new drugs, is, therefore, a deeply flawed one since it assumes that all drugs deserve equal protection. Further, many drugs also benefit from public funding and in any investment estimate this should be subtracted from overall costs.
In any case, without these figures, it is difficult to ascertain the worth of patent incentives and to what extent generic entry impacts these incentives. However, we do know three things:
that drug discovery and development require considerable investment (how much, we’re not sure);
that the major drug companies invest primarily in drugs that have lucrative markets in the West. The amount of investment in developing countries or Third World diseases is appallingly low;
and that drug majors recoup almost all of their investments and profits from their main Western markets.
Would compulsory licences (CL) in favour of generics affect incentives to innovate?
Thus far, there is no empirical work demonstrating that licences stultify the rate of innovation. The few studies available indicate that licences do not have any significant demonstrable effect on the rate and pace of innovation. Illustratively, Colleen Chien empirically tests the rates of patenting and other measures of inventive activity before and after six compulsory licences over drug patents issued in the 1980s and 1990s. She observed no uniform decline in innovation by companies affected by compulsory licences and found very little evidence of a negative impact.
More importantly, one needs to ask whether all countries in the world need to contribute equally to Bayer’s R&D efforts. Or whether countries such as India with significant numbers of poor patients can devise policies to induce lower-priced drugs in the market without worrying excessively about Bayer’s incentives to innovate, given that such incentives are more than adequately provided by Western markets such as the U.S. and the European Union [E.U.]. After all, some of the biggest innovators today benefited from lax IP regimes in the past. Illustratively, Switzerland, which houses some of the world’s leading drug originators, refused to introduce product patents until 1977, and it was only after considerable pressure and bullying from Germany that it finally yielded.
Lastly, it must be borne in mind that a compulsory licence is not an evisceration of the patent, as is sometimes made out to be in media reports. Rather, it embodies what Calabresi and Melamed, in their seminal piece, describe as a “liability” rule, where the patentee continues to hold the patent and is entitled to a reasonable royalty from every new player entering the market. This advantage (in terms of compensation through royalties) cannot be understated, particularly in a market like India, where the consumer market is highly differentiated in terms of purchasing power. Drug originators typically cater to very-high-income consumers, while generics are able to tap into middle- and low-income consumer segments as well. Consequently, the possibility of a new generic entrant entering a market segment hitherto untapped by the originator rather than simply displacing the patentee’s existing customer base is high. To this extent, a compulsory licence may permit an innovator to profit from newer, untapped markets.
Is global supply of inexpensive medicines to treat acquired immune deficiency syndrome (AIDS), cancer and other diseases inconsistent with IP rights?
Cipla was the first company to sell AIDS medication at a fraction of the cost sold by drug innovators. Given that the rate of R&D and investments into HIV drugs has not slowed down significantly after that, one might hazard a guess that low-priced generics (albeit legal ones) do not necessarily impact innovation incentives.
India’s strong generic pharma industry is attributed to the process patent regime introduced in the Patents Act, 1970. However, our obligation to implement the Agreement on TRIPS has forced us to reintroduce the product patent regime. Has this limited our ability to produce technologies through reverse engineering? Have our policymakers been able to exploit the flexibilities that exist in the framework provided by the agreement on TRIPS?
My own view is that India strategically exploited TRIPS’ flexibilities to the hilt. It introduced higher standards for pharmaceutical patentability, a very potent opposition mechanism where any member of the public could effectively oppose a patent grant and some of the widest compulsory licensing norms that the world has ever known.
What, according to you, are the implications of Article 39.3 of the TRIPS Agreement requiring fixed period market exclusivity for the pioneer firms? Should India have to agree to comply with this Article and to the protection of test and other data submitted for obtaining marketing approval? Is there a way out?
Article 39.3 is loosely worded, leaving enough scope for India to continue rejecting data exclusivity as a norm. A good middle-path solution would, however, be a compensatory liability model, where data can be used by generics after the payment of appropriate remuneration (similar to the compulsory licensing model).
You have said that under Section 3(d), the applicant has to demonstrate a reasonable correlation between the efficacy claimed and the data provided in support of this. Such reasonable evidence of the correlation can be established by relying on, inter alia, statistically relevant data documenting the activity of the new form and/or known substance, documentary evidence (for example, articles in scientific journals), data generated using in vitro assays, or from testing in an animal model, other preclinical test data or any combination thereof. However, in 2007, the U.S. Supreme Court, in the KSR vs Teleflex case, cautioned against overemphasis on the importance of published articles. It argued that “diversity of inventive pursuits and of modern technology counsels against limiting the analysis in this way”. It pointed out that in many fields it may be that there is little discussion of obvious techniques or combinations, and it often may be the case that market demand, rather than scientific literature, will drive design trends.
KSR related to the standards for determining the obviousness of an invention – for which the court suggested that even common sense would effectively do, rather than demonstrating the existence of some specific published article pointing to the allegedly inventive combination. It is a very different context and a very different theme from the present issue under consideration.
Essentially, my argument is that you cannot insist on proof by way of clinical trials under Section 3(d), since patents are filed at the “discovery” stage and clinical trials are conducted much later. More importantly, forcing patentees to compare known substances with their new discoveries through clinical trials in order to gain a patent may be unethical. Consider the Novartis case. Apparently, the imatinib free base (the previously known substance) cannot be used as a drug since it is not stable, etc. Therefore, it has to be converted to the salt form. Forcing human beings (clinical trial volunteers) to try an ineffective drug (the known imatinib free base) only to demonstrate that the new substance for which the patent has been claimed (beta crystalline form of imatinib mesylate) is more effective is unethical, to say the least.
You have said that efficacy ought to be interpreted to mean a definite “therapeutic advantage”. Can a patent examiner, who is not well trained in pharmaceuticals, pronounce on this convincingly? You have, for instance, refused to believe Novartis’ claim that the beta crystalline form demonstrates a 30 per cent increase in bioavailability. You have said that this by itself does not demonstrate any therapeutic advantage in relation to the patient and that it has to be established independently. Can you elaborate how it can be done? But on the question of known substance, you have sought the appointment of an expert by the Supreme Court to make a determination. Are these not contradictory?
Bioavailability only means that the rate of dispersion of the drug through the body is faster. As to whether or not this speedier dispersion also impacts the speed of the cure is not known. Therefore, Novartis has to independently establish this if therapeutic efficacy is used as the standard.
The Supreme Court is free to appoint an expert to determine the issue of “efficacy” as well, but only after it has outlined the standard of efficacy (whether Section 3(d) means therapeutic efficacy or not). My focus has been on helping it evolve a standard in this regard. On the issue of known substance and novelty, etc, the legal position is established. Only the factual determination needs to be done – which is why I’d asked for the appointment of an expert. Experts under the Patents Act can only be appointed for factual determinations and not for legal determinations.
You have described the IPAB’s decision, which relies on excessive price of the drug as the ground for denying patent to Novartis, as a ludicrous legal proposition. But excessive price of the drug is a relevant factor in the grant of compulsory licence. Are these inconsistent? Some would say excessive pricing is a form of commercial exploitation of the invention, which according to you needs to be prevented. The alternative to compulsory licensing is riddled with many improbables.
A patent grant ought to be based primarily on “technical” criteria, that is, whether the invention has technological merit or not. Pricing, etc., come at a much later stage. At the stage of patenting, one may not even have a product to begin with. To me, denying patents on the basis of pricing concerns is simply bad policy. Further, such a provision may also contravene TRIPS.
Much like a knife, a patent can be put to either good or bad use. It makes sense to regulate the alleged abuse of a patent only after the grant (ex-post use). Such abuse of a patent in terms of pricing, etc., can be regulated through price controls and compulsory licensing. We’ve just had the first case now, so let us wait and watch to see if the CL policy works in helping us regulate excessive pricing by engendering more competition in the market. The first spate of pharma patents were granted only in 2006 or 2007. Under the Act, CL could be applied for only after three years from the date of grant – effectively only from around 2010 or so – therefore, there’s been a real delay of around only two years or so in the grant of the first CL.
How many applications have been filed by generic manufacturers for the grant of a CL? Can you cite some prominent examples other than Bayer vs Natco?
Cipla has filed some CLs (for Isentress). Natco filed one for sunitinib last year (for export to Nepal). But their application was very faulty and was, therefore, dismissed.
Is the law on compulsory licensing unsatisfactory? Does it require any amendment to make it easier to grant?
It does give some scope for patentees to strategically delay the proceedings and frustrate the application. Such loopholes must be plugged.
Pre-grant opposition in patent proceedings has been criticised by some. Should India remove the relevant provision?
Not at all. Opposition is one of the most effective ways to ensure that the high standards under the Indian Patents Act are necessarily adhered to. Given that the Indian patent office is terribly understaffed and under-resourced, the likelihood of wrong decisions is high. An external source that can forward relevant prior art to the office (through an opposition mechanism) is useful for the process and for ensuring that the grant process is a stringent one.
The provisions relating to royalty to the patentee have also come in for criticism for being too vague. Your comments.
In the specific Bayer case, given that Bayer itself did not elucidate its cost and break-up, etc., to the Controller, he had nothing to go by. The easiest option was to pick a figure advocated by the UNDP, which he did. However, had Bayer submitted costs and break-up, it would have been far more difficult to simply apply a round figure that had no bearing on its cost of R&D (computed in proportion to the Indian market) and the paying capacity of Natco.
Perhaps, future cases would have to deal with this in a more elaborate manner. And, perhaps, the patent office might need to appoint economic experts as well to help with royalty computations.