Because new medicines have such a high burden of R&D costs to recuperate, their commercialisation in a protected environment is essential to the innovator business model. That protection has historically been offered through the patent system, but for how much longer, given the increasing time taken to innovate?
Patents generally last 20 years, from filing to expiry. In the USA until recently, an anomalous arrangement prevailed whereby patents lasted for 17 years from grant to expiry, reflecting a nominal 3 year period for patent prosecution. But under either arrangement, the research and development, and subsequent regulatory review process, takes a substantial bite out of the available period of monopoly. As a consequence, legislation was put in place decades ago such that patent offices in most developed countries could offer a supplementary certificate of patent protection (SPC) for around 5 years, specifically for marketed pharmaceuticals. In the US, under the Patent Term Restoration Act of 1984 (the so-called ‘Hatch-Waxman’ act), the additional period is calculated by reference to the development and regulatory review time taken by the FDA.
SPCs only apply to marketed drugs, so most patents do not receive them. The Hatch-Waxman extension period is specific for each drug product. Moreover, drugs are often protected by more than just one patent. Famously, ranitidine’s period of exclusivity was extended by a polymorph patent, covering a specific crystalline form of the world’s first blockbuster. The situation is further complicated in today’s litigious world, by the incentive for generic companies to challenge patent validity. Again in the US, this can offer six months of market exclusivity to the first generic company to file an abbreviated new drug application (ANDA) with a claim for invalidity of the innovator patent.
A key question, given this complexity, is what exactly is the period of exclusivity for marketed pharmaceuticals? Given the sensitivity of the pharmaceutical industry to this issue, a subsidiary question is how has this period changed over the past 30 years, a time in which the blue-chip industry stocks have transformed from ones which offered investors primarily growth opportunities to ones which now offer primarily revenue returns.
The availability of SPCs, the multiplicity of patents covering a single product, the openness to invalidity challenge and the widely varying time taken for research, development and regulatory review, present a set of contributory factors that make the period of exclusivity difficult to predict. However, one short-cut way of measuring this, at least as far as the US is concerned, is to look at the FDA Orange book. This online resource reveals the dates of approval of both innovator and generic versions of all approved products since Jan 1982, and can be used to answer precisely the question posed. While other commentators, in particular DiMasi and Paquette (2004) [PharmacoEconomics, Vol. 22, Suppl. 2, pp. 1-14], have posited that the period of exclusivity has declined in recent decades, the hard evidence to support this claim is incomplete.
So, we carried out this analysis on 231 genericised drugs originally approved by the FDA since 04 Jan 1982, and the results are shown in Figure 1 below. We excluded a few products for which the patent(s) have expired but the market is still a monopoly, or which became generic before 1982; we also excluded discontinued drugs. For the rest, we compared the first (innovator) drug approval date with the first generic competitor approval date; between these two dates the innovator has a monopoly. The length of this period is clearly declining, and is fast approaching the magic 5-year barrier. At this point, the exclusivity available on the data effectively controls the product monopoly, rather than the exclusivity available through the patent system.”]As it stands, generic companies are only able to refer in an ANDA to innovator clinical data after the expiry of the data exclusivity period. Doing so enables them to eliminate the cost and time required to conduct their own studies. In the US, the data exclusivity period is 5 years following FDA approval for a new chemical entity, with a lesser period of 3 years for incremental innovations such as new salt forms or stereoisomers, or new uses. The concept of data exclusivity is fundamentally different from patent exclusivity insofar as the latter must be initiated early in the R&D cycle, long before commercialisation can begin (if it ever begins).
If one takes the regression line from the graph in Figure 1, and extends it to 2022, the projected market monopoly is actually less than 5 years. This means that the patent system will be redundant in the protection of pharmaceuticals for drugs approved in 2022. But, think onâ€¦a research programme commencing in 2010 can reasonably be expected not to bear fruit until at least 12 years from initiation, which brings us to the startling conclusion that the patent system as it applies to new pharmaceuticals in the world’s most important market is currently redundant.
This is an important result, if it is right. But the analysis has a problem, because the later points on the graph are a self-selecting set of genericised drugs: those drugs still with patent protection are not represented. We looked carefully at the Orange Book to find the earliest approval of a product that is still patent-protected. This is lodoxamide, from Alcon, which was approved on 23 Sep 1993, and its method of use US Patent 5,457,126 is still in force, as of 05 Mar 2010. If we cut off all later approvals, and just analyse the monopoly periods for products before this date, we end up with the graph shown in Figure 2. The correlation is now significantly poorer (the number of data points is now only 138), but the trend is still downwards, and, if extrapolated, still takes us below the 5 year point. So the earlier conclusion is unchanged, except we cannot be sure exactly what has been happening in recent years; it could be the previous downward trend has levelled out, perhaps as companies anticipate the lack of sufficient patent cover and selectively abandon commercially unattractive programmes. However, it is unlikely to have escaped falling below 10 years, the data exclusivity period for NCEs in Europe.”]Last, I raise a subsidiary point. Data exclusivity has been discussed recently in the US Congress in regard to biologics. Proposals for data exclusivity for biologics range from 7 to 14 years. The argument made is that the patent system only narrowly protects biological inventions, and products of this kind are open to patent-busting strategies. The analysis we have conducted and report here is necessarily focussed on small molecule drugs, since generic versions of large biological products are not yet widely approved. It suggests that the patent system is ineffective in protecting the commercialisation of small molecules, and a similar debate on data exclusivity needs to start for this type of new medicine, and urgently so.
In the meantime, on the assumption that legislation will take some time, even while Rome burns, we call for the industry to do more to improve R&D cycle times by the adoption of more low-risk, rapid strategies to product innovation.