In a 1981 study entitled ‘Patent-Term Extension and the Pharmaceutical Industry’, the US Congress observed that, ‘although all patented inventions receive protection for the same amount of time, the effective patent terms for the inventions vary’, as they depend on the amount of time elapsed between patent filing and effective commercialization. In relation to pharmaceutical products and medical devices, the study argued that the effective patent term is reduced, due to premarketing and premanufacturing R&D, subject to compulsory regulations. In 2003,B. Lehman, President of the International Intellectual Property Institute, made similar remarks, and disputed the efficiency of current mechanisms of extension of patent terms, which ‘do not equal the time lost in ability to market’ and impose, upon the patent owner, restrictions which allow generic manufacturer to use the product for testing and developing the generic alternative while the patent is still in effect.
A group of highly distinguished scholars (Budish, Roin and Williams) recently showed that the fixed patent term distortion, among US cancer patients diagnosed in 2003, resulted in around 890,000 lost life-years, with an estimated social value of US $ 89 billion per year. They highlighted some findings which may inspire future policy changes: (1) the number of clinical trials is inversely proportional to the patient survival time, which shows that drugs with long clinical trials, hence with a short effective patent term, are less likely to be developed than drugs with a short lag between development and commercialization, (2) patents affect the level and composition of R&D investments, (3) increases in R&D positively translate into improvements in patient health, and (4) the use of valid surrogate endpoints effectively increased R&D investments, reducing commercialization lags and improving patients’ health. In particular, the use of surrogate endpoints was seen as a plausible solution to reduce the commercialization lag, and to reinforce patent protection, without modifying its duration.
However, the insufficient length of the patent term for new medicines (discussed in a debate hosted by The Wall Street Journal here) is not the only potential obstacle to their development. D. Baker, co-director of the Center for Economic and Policy Research, suggested (here) that other issues affect the protection offered by the patent system to pharmaceutical R&D, including (a) ‘excessive marketing expenses, as firms seek to pursue the monopoly profits associated with patent protection - data from the industry suggests that marketing costs are currently comparable to the amount of money spent on research’, (b) investments in duplicative drugs, which involve two thirds of the overall research spending, and (c) insufficient disclosure of information.
Professor B. Roin, in his article on ‘Unpatentable Drugs and the Standards of Patentability’, focused, instead, on the requisites of novelty and non-obviousness, arguing that their application in the context of pharmaceutical cases may prevent the patentability, and thus discourage the development, of socially valuable drugs.
With regard to novelty, the author noted that ‘in the pharmaceutical industry, […] new drugs can cease to be ”novel” inventions long before they have undergone the clinical trials needed to establish their medicinal value and thus can become unpatentable before the public ever gains access to them’. In other words, the link between disclosure and innovation, in this specific field, could be particularly weak, as the mere theoretical possibility of a therapeutical use of a chemical compound does not usually allow the pharmaceutical company to establish its safety and efficacy, which will be determined only at the end of clinical trials. The article refers to the analgesic drug Ultracet as a prime example of the distortions introduced by the novelty requirement: the idea of combining the two drugs of which Ultracet is composed was briefly mentioned in a 1972 publication, but physicians only began prescribing the combination after a company established its safety and efficacy in 2001, receiving FDA approval to market what turned out to be an unpatentable drug. According to Roin, unpatentability for lack of novelty may derive from accidental disclosure or preliminary disclosure of a compound erroneously deemed ineffective, or whose therapeutical value is not demonstrated: these situations explain why ‘82% of universities with large medical-research programs were unable to patent at least one of their life-science inventions during the previous year and 71% of [them] were unable to find a commercial partner to develop one or more of their life-science inventions’. The doctrine of inherent anticipation (particularly after Schering Co. v Geneva Pharmaceuticals Inc. et Al., commented here - see also, in the UK, Merrell Dow v H N Norton & Co), may lead to similar distortions.
The non-obviousness requirement prevents the patentability of drugs whose therapeutic properties are reasonably foreseen, but incurs in the same issue, as the predictability of therapeutic efficacy remains highly speculative until clinical trials are under way. This requirement, according to Roin, ‘denies patent protection to inventions because they seem likely to work while ignoring the question of whether a patent is needed to motivate that invention’s development’ [although this conclusion, in Merpel’s view, conflates the aim of the non-obviousness requirement, with that of the patent itself].
The conclusion drawn by the article is that the inadequacy of the novelty and non-obviousness requirements to distinguish between drugs which deserve patent protection, and me-too drugs which yield little social value, prevents the development of valuable drugs, as pharmaceutical companies regularly, and repeatedly, assess the patentability of a candidate compound before proceeding with its development. ‘The social costs of losing such drugs’, in the author’s view, ‘likely far outweigh any benefits to the public from faster access to inexpensive generics of the unpatentable drugs that actually reach the market’. Possible solutions include a patent reform, which could create an exception in the novelty and non-obviousness standards for drugs in the preclinical stage, direct government funding, or FDA-administered exclusivity periods. Roin focused on the latter solution, and argued that lengthening the period of exclusivity to somewhere between ten and fourteen years, would ‘eliminate the distortions arising from the novelty and non-obviousness requirements’.
Roin’s arguments were partly disputed by K. Outterson (here), who criticized the Ultracet example and observed that companies may be willing to take the risk of proceeding with the development of a drug with weak patent protection or in the public domain, resorting to method-of-use, formulation, or combination patents, or simply relying on their brands’ strength. Outterson recommended a re-evaluation of the possible use of direct government funding to deal with the issue, and suggested charging generic manufacturers with a fee payable to the company who did all the R&D, including the clinical trials and regulatory approval phases. In the UK and US, there is some evidence of direct public funding aimed at supporting research on abandoned compounds, as the UK Medical Research Council and the National Health Institute are both financing similar initiatives (more here).
In a recently published paper, M. Morgan argued that market exclusivity could be a more flexible substitute of patents in the biologics industry, which shares many of the characteristics of the pharmaceutical sector. Savva summarizes Morgan’s findings as follows: (1) ‘FDA-administered exclusivity has the benefit of reducing the barriers caused by upstream “patent thickets” on basic research’, (2) rent dissipation ‘can be prevented by allocating “clinical exclusivity” once a product is ready for clinical trials, and (3) ‘until that stage, the cultural norms of science (which favor early publication of breakthrough research at the “pre-competitive” stage) combined with trade secret protection (until a drug candidate is ready for clinical development), can provide adequate incentives for development of unpatentable drugs’.
Our Katfriend also explains how the current regimes of “regulatory exclusivity” work and why they need to be improved:
It may be argued that current “regulatory exclusivity” regimes - e.g. data exclusivity for new drugs (5 or 12 years in the US for pharmaceuticals or biologics, respectively, and 10 years in EU) and new indications (3 years in US and 1 year in EU) and "market" exclusivity for Orphan Drugs (7 years in the US and 10 years in the EU) - provide adequate incentives for "dormant therapies" in absence of patent protection. However, data exclusivity only protects clinical trial data which does not prevent another competitor from conducting their own trials (at no risk of failure), which may be likely if there is a lucrative market.
Also, Orphan Drug exclusivity only applies to rare diseases (affecting less than 200,000 people in the US and fewer than 5 people per 10,000 in the EU) [and is] shorter than average patent protection (12 years). Further, commentators have noted that drug companies have been known to "game" the Orphan Drug legislation by seeking approval for a drug in a small patient sub-population while generating "blockbuster" revenues through "off-label" use for broader indications (see Kesselheim and Armstrong). Gleevec/Glivec is an example of such a drug.In the US, the MODDERN Cures Act, which died in subcommittee during the 112th Congress, but is due to be resubmitted in the 113th, aims to address some of the concerns raised by dormant therapies. The draft’s preamble clarifies the issue:
(7) In addition, there is reason to believe that potential treatments with tremendous value to patients are never developed or are discontinued during research and development due to insufficiencies in the intellectual property system.
(8) It is in the public interest to address the hurdles that may be precluding new treatments from reaching patients and to remove the disincentives for the development of therapies for these unmet needs.The Act defines ‘dormant therapies’ as those that meet unmet medical needs and have prospectively insufficient patent protection. The first requisite essentially implies that the drug to be investigated should address ‘a need for drugs or biological products for the treatment of one or more life-threatening or other serious diseases or conditions for which no therapy exists’, or guarantee improved effects on serious outcomes already treatable through other therapies. The second requisite is fulfilled if patent protection, for a period of 14 years from the date of first approval, is not reasonably anticipated to ‘provide an adequate scope of protection to prevent the approval of products that would rely upon or reference the dormant therapy, in an application filed under section 505(b)(2) or 505(j) of the Federal Food, Drug,and Cosmetic Act, or section 351(k) of the Public Health Service Act’, or to provide patent protection under the patents, or applications, filed by the sponsor. The request for designation lists the related patents owned by the sponsor, who must agree to waive their enforcement after the end of the exclusivity period, and certifies the prospectively insufficient patent protection. The designation is conceded if all the above requisites are met, provided that the sponsor effectively intends to file an application for approval of the drug with the FDA.
'unmonopolisable...' |
The MODDERN Cures Act appears to be a significant step in the right direction. Savva explains that, although there is the theoretical risk that a competitor may obtain exclusivity over a drug initially developed and investigated by another company, a situation which could discourage the disclosure of information in the public domain, ‘the fact that most companies are incentivized to file patent specifications early with broad claims (subject to potential utility and sufficiency issues) [makes it] unlikely that the Bill would result in any significant change in the disclosure of information or filing practices, although it may discourage “defensive” publications’. This legislation would thus provide the necessary reward for companies who invest in the development of socially valuable drugs, shifting the focus away from the rigid requirements of novelty and non-obviousness, and the distortions induced by insufficient patent protection.
This Kat is of the view that the issues highlighted above, and the reduced efficiency of the patent system in promoting important research investments in the pharmaceutical sector, is certainly worth investigating, as the social benefits that may derive from the commercialization of ‘dormant and unmonopolisable therapies’ should not be wasted. There are, however, some further concerns and doubts that he briefly summarizes as follows:
- the lack of data on specific R&D expenditure, and the profits made through sales of the developed drug at premium prices, prevents a more fair evaluation of the possible solutions to the issues examined, as proposals focus on guaranteeing an exclusivity period similar to that afforded by patents in other industries, on the unproved assumption that this may effectively address the industry’s concern [which is possible, and perhaps even plausible, but not easy to demonstrate];
- based on the data available, there appears to be extreme variability in R&D expenditure (with the cost of each drug varying between US $ 500 million and more than US $ 2,000 million, according to this study) and profitability, with some drugs reportedly generating profits sufficient to recoup not only their research costs, but also those of less successful drugs (as demonstrated by the so-called ‘patent cliff’, and by a study which found that, for new chemical entities introduced between 1988 and 1992, the top 10% of drugs accounted for 56% of the overall sales of all the NCEs); if this variability was confronted, once again, with a rigid solution, it may result in issues of over-protection and, more importantly, under-protection, potentially compensated through extremely high market prices, which may critically inflate the 'deadweight loss' effect;
- the idea of using administrative periods of exclusivity is certainly more reasonable than tampering with the novelty and non-obviousness requirements, which do not serve the scope of rewarding inventors for their innovations, but merely restrict the target of patent law;
- public funding in the pharmaceutical sector is certainly needed to promote the development of drugs for both rare and endemic diseases, and could be necessary to stimulate further research on compounds which already translated into successful drugs; more generally, there should be some recognition for the composite structure of pharmaceutical R&D, which frequently involves many subjects in its various stages (for example, Universities which disclose the basic compound, and companies which investigate its property and therapeutic efficacy, and conduct clinical trials), calling for public co-funding as a primary drive for basic research;
- 'unmonopolisable therapies', an issue which usually remains in the shadow of 'dormant therapies', warrants a cautious approach, as it may be difficult to identify the innovations which really deserve an exclusivity-based protection; further, the development of diets, dietary supplements, lifestyle interventions, surgical methods, “natural” remedies, and many complementary and alternative medicines is frequently the result of the work of individuals or restricted groups, which may be more sensitive to better research funding (i.e. increased "push" incentives) and recognition than to patent-like protection (Savva notes that the latter provides limited incentives for 'unmonopolisable therapies', as monopoly prices cannot be enforced, and suggests that examples of alternative "pull" incentive could comprise prize-based models such as the Health Impact Fund, Advance Market Commitments, priority review vouchers, wildcard patent extensions, prizes for clinical trials, etc.);
- it is necessary to ensure that legislative intervention, such as the MODDERN Cures Act, contain provisions that (1) address the risk of races to approval and rent dissipation, (2) ensure that ‘dormant therapies’ do not turn into ‘hibernated therapies’ [Kat-invented description, which points to the possibility that companies could slow their research efforts after obtaining the designation, knowing that the exclusivity period starts from the regulatory approval - Savva argues that the use of development certifications should be sufficient to minimize the issue], or into the equivalent of ‘submarine patents’, and (3) prevent reliance upon designation as ‘dormant therapies’ from generating a shift towards the development of drugs with high profitability (for which ‘the clock is ticking’, as they do not fulfill the requisites for extended protection), over those that meet urgent and unmet medical needs (for which the clock begins ticking only after approval);
- finally, these issues suggest that, although not welcome by the industry, an administrative authority, equipped with the power to investigate these issues through access to real numbers, and to review the correct use of the system, would not only guarantee that the rights of inventors and patients are rightly balanced, but would ultimately represent a safe way for the industry to make its voice heard, when, despite the high costs of the marketed drugs, the risk of insufficient reward is real.
Finally, our Katfriend is also interested in comments on the use of 'patent evergreening' techniques (e.g. patenting derivatives, metabolites, selection inventions, new formulations or combinations with the same active ingredient, new uses, methods of administration or methods of manufacture) to rescue 'dormant therapies':
For example, if a compound does not have regulatory approval, a method of use patent will still provide adequate market protection against generic competition which cannot obtain approval for any other indication - this is the likely strategy of various "drug repurposing" companies, which try and find new indications for failed drugs which may have passed "safety" testing (in Phase I of clinical trials) but failed in Phase II and/or III (efficacy). Various commentators have noted that such "secondary" or "second generation" patents (which do not cover the drug compound itself) are weaker and more susceptible to obviousness challenges (according to two studies here and here). However, drug repurposing companies may still be willing to take a drug to market on a "secondary patent" due to lower costs of development for repurposed drugs which have already passed pre-clinical and phase I testing. Arguably, there would be no incentives to develop potentially safe and effective drugs which had not undergone such testing (or drugs which cannot even obtain such "secondary patents" due to obviousness).The IPKat is sure that his readers were just waiting for this occasion to showcase their knowledge and helpfulness.
Phew! Where to start?
ReplyDeleteData exclusivity is very powerful, provided the term is reasonable. 10 years in the EU is good, but in the US is unhelpful. No-one conducts there own trials so it provides a de facto monopoly until data expiry. Investment will be made on a 'dormant' compound on the back of such exclusvity. Patent protection will ideally cover the period up until the late-stage trials are underway.
Method of treatment/first medical use patents are as good as new chemical entity patents, provided they are valid, but as pointed out, they are susceptible to attack due to the huge amount of suggestive publications out there. Stand up well when the known compound is found to have a new mechanism of action leading to the new use.
No-one will take a compound through initial clinical studies unless patent protection exists, so such dormant compounds will remain dormant. There are always novel analogues (patentable) that can be prepared however, so provided the mechanism of action is deemed to have potential, work in the area will proceed.
Rarely will 'patent evergreening' techniques provide adequate protection, just as they do not for marketed products. Such patents only block lazy generics from doing even the minimum of development work.
Thanks for your comment!
ReplyDeleteIt is difficult to find empirical data on the extent a company may be able to rely on data exclusivity alone after regulatory approval as opposed to patents to secure investors.
According to a recent survey of biotech executives in the following paper by Roth over a third of biotechs do not rely on patents to secure investors (at p292). However, this might be due to an increased reliance on trade secret protection before an in-licensing deal. My hypothesis is that a major pharmaceutical company would likely require at least 10 or more years of market exclusivity after regulatory approval (secured by patent or market exclusivity) in order to approve an in-licensing deal.
Another point is that if patent protection was redundant, surely there would be a large number of biologics that have been launched without patent protection, particularly in the US and EU which had unlimited data exclusivity before biosimilar reforms?
In addition, according to the following paper by Grabowski et al it is inappropriate to equate data exclusivity with monopoly protection (at p3). As an example, the paper mentions that five of six competing manufacturers producing somatropin-based human growth hormones received product approvals in the FDA by running their own clinical trials. Notably this was also before the US biosimilar pathway reforms which effectively reduced the amount of data exclusivity from an unlimited period to 12 years. However, perhaps if competitors now know that they can just wait until data exclusivity expired, this would make it uneconomical (and unethical) to spend money on socially wasteful duplicate clinical trials, which provides a de facto monopoly for the duration of data exclusivity.
I was (once again) surprised by the quality of your post, Stefano. This is publishable material and I suggest you develop it in an article. I would certainly be interested in reading it.
ReplyDeleteOne of the critical points here is whether data exclusivity or term extension may achieve the result of resuscitating dormant drugs. I am partly skeptikal on this, due to the industry structure. In my experience, companies are primarily eager to develop what you called "blockbuster" medicines, for which revenues usually cover expenses in a short period of time. A lead of two to three years, which is compatible with the average time that it takes generics to market their products, would be sufficient then. All the drugs which cannot generate significant revenues in such a short period are likely to be never researched anyway.
On the other hand, why not doing something different? Like shortening the patent term if no development is made on a substance. Or, instead of shortening, making it subject to compulsory licensing, in favor of someone who is willing to develop it. And completing the scheme with a serious incentive for industry cooperation, which is so widely underestimated?
I stumbled across this today (http://christianengstrom.wordpress.com/2011/03/09/an-alternative-to-pharmaceutical-patents/):
ReplyDelete"The relevant question is which model provides the most efficient and cost effective way of funding pharmaceutical research. Because nobody claims that pharmaceutical research is cheap. The average cost for developing a new drug is just over a billion US dollars.
But considering that ”the government” already provides most of the income for the pharmaceutical companies, a reasonable first step would be to find out how much of that income actually goes to research.
Fortunately, this is very easy to do, as all the big pharmaceutical companies have their annual reports available online. As an example, we can look at the numbers for Novartis (page 143), Pfizer or AstraZeneca.
They all spend around 15% of their revenues on research. The other 85% go to other things, according to their own figures. The numbers are typical for the industry.
So the question is: does the patent system really give us, the taxpayers, the maximum amount of pharmaceutical research for the money we are spending on drugs? Or is there room for improvement, when even the pharma companies themselves admit spending 85% of the money we give them on other things?"
If true, doesn't that change the whole story?
To anon at 11:22, the "government" provides "NOTHING".
ReplyDeleteLast time I checked, nobody GIVES pharma companies anything. The sell drugs. People buy drugs. Drugs make people not-so-dead for a little longer.
I was thinking about small molecules rather than biologics regarding the power of data exclusivity.
ReplyDeleteA big reason many small companies do not rely on patents to secure investment is because they don't have any of great value. Common problems include poorly drafted patent applications, poor quality prosecution, limited territorial filing coverage, and limited patent-term remaining due to very slow development. The poor quality is mainly due to the use of private practice firms. Most do not have the skills and experience to handle pharmaceutical inventions diligently. They are after all, jack of all trades, master of....
Their business-relevant advice is also poor. As always there are the exceptions that prove the rule.
I agree that this stuff would make a good article,, but I think it is actually the subject for a Masters degree first. The research involved in obtaining the true opinions from the people making these decisions will be a difficult task.
In an ideal world, a new compound will have excellent Phase II results, 20 years patent life remaining with the ability to extend, and both data and orphan exclusivity to boot. Realism brings the exclusivity period (patent or otherwise) expected back down to the 10-15 year mark - 10 year patent term withe the ability to obtain PTE/SPC.
As the area gets more exciting from a market-size point of view, reduced exclusivity periods become acceptable.
Also to anon at 11:22:
ReplyDeleteThe link you posted to the Swedish Pirate party MP's page is interesting and certainly has some lively discussion - however, I think the proposal to abolish patents and replace them with government funded research runs into some problems. Firstly, there is a free-riding issue - you would need some kind of global fund into which all countries contribute a percentage of their GDP - this may be politically difficult to implement, although is worth doing (e.g. see Health Impact Fund). There are also the inefficiencies of a non-market mechanism for distributing funds. Payors/governments currently do not spend more on a drug which is no more effective than cheaper drugs already on the market. The market is also optimised to allocate risk to those willing to bear it and has approximately US$30 trillion available to invest, provided that a ROI is possible. Biotech/pharma companies with drugs which don't work are designed to fail early and often. By comparison, researchers working in government departments have limited reason to kill a project, especially if their job depended on it. In addition, government departments have to "pick winners in advance" and the taxpayer foots the bill for clinical failures (which can cost billions). However, I do believe that increased direct or indirect government funding of "unmonopolisable" and "unprofitable" therapies makes sense, particularly where this has the potential to decrease the overal global disease burden and the current market/patent model does not have adequate incentives in place.