E cdip/15/inf/2 original: english date: january 8, 2015 Committee on Development and Intellectual Property (cdip) Fifteenth Session Geneva, April 20 to 24, 2015

II. Literature Review

The economic evidence on the impact of TRIPS and more generally the patentability of pharmaceutical products is at best mixed. This is probably best illustrated by the example of India. The India Patents Act of 1970 denied patentability to pharmaceutical products.¹⁰ Chaudhuri et al. (2006) suggest that the absence of pharmaceutical product patents had indeed a strong effect on maintaining low drug prices in India. However, when India entered the WTO in January 1995, it also adopted the TRIPS agreement which forced India to recognize the patentability of pharmaceutical product patents, although India was allowed –as were all other developing country members- to postpone their introduction until January 2005.¹¹ In 2002 India brought its legal system further in line with TRIPS requirements by among other things introducing a 20-year patent validity term counting from the patent application date. Arora et al. (2009; 2011) analyze the response by Indian pharmaceutical companies to this regime change. Their findings indicate a strong increase in R&D intensity among domestic pharmaceutical companies following the introduction of pharmaceutical product patents. However, this increase is explained not as much by Indian companies beginning to invent their own new drugs, but instead is the result of investment in process innovations and improvements on existing drugs. These improvements enabled Indian producers to successfully compete in the world-wide generics and bulk drugs market. Arora et al. (2009) argue that this development can only be in part attributed to the strengthening of patents in pharmaceuticals in India. Kyle and McGahan (2012) also arrive at a negative assessment of the TRIPS effect on developing countries. While they confirm a positive association between patent protection and R&D investment in developed economies, they do not find any evidence that R&D spending on diseases that are disproportionately more prevalent in developing countries is driven by the implementation of TRIPS in lower income economies.

Based on the evidence for India in Chaudhuri et al. (2006), Goldberg (2010) argues that the most important concern with regard to pharmaceutical patents and TRIPS is access to patented drugs in developing countries. She argues that multinationals have incentives to delay or forego entry in developing countries because of “cross-country reference pricing.” That is, to avoid regulators using lower prices for drugs marketed in a developing country to set prices in developed countries, multinationals wait until prices are set in the lucrative developed economies before they consider entry in developing countries. Pharmaceutical patents assume a critical role if foreign originator companies use the patent system to strategically delay their entry in a developing country while keeping any potential generic competition at bay.
Kyle and Qian (2013) look at the decision by foreign originator companies to introduce a drug in a developing country and the quantity of drugs sold depending on whether the originator has obtained patent protection. Their results contradict Goldberg’s (2010) concerns; using drug-level data from IMS Health for 1990-2011, they find that products are launched faster in markets that allow for pharmaceutical product patents. In fact, they find higher sales for patented drugs, and a lower price premium of patented drugs post-TRIPS in lower- and middle-income economies. That said, their analysis cannot rule out alternative influences, such as price controls, the threat of compulsory licensing etc. Cockburn et al. (2014) address some of these concerns by including broad measures of the regulatory environment (including price controls) in the countries included in their sample. Their findings still suggest that companies are significantly more likely to launch a new drug in a country that has strong patent protection.

III. Primary and secondary patents

Patents play a crucial role in the pharmaceutical industry. Patents are usually filed already during the research phase in the development of a new drug. These early patents are filed to protect potential active ingredients that form the basis of the new drug. Since the early stages of drug development are characterized by an enormous amount of uncertainty (the European Commission suggests that 1 in 5,000-10,000 test active ingredients results in a successful drug (EU Commission, 2009)), early patent filings reflect this. Patents on active ingredients are referred to as primary patents. In later phases of the drug development, patents are filed on other aspects of active ingredients such as different dosage forms, formulations, production methods etc. These patents are referred to as secondary patents. Secondary patents also emerge from changes to formulations and dosages or applications in new therapeutic classes, discovered during clinical trials. Hutchins (2003b) reports that the usual filing strategy is to file many and broad primary patent applications and then to surround them with secondary patent applications.

A critical issue regarding the secondary patents is whether they protect genuine follow-on innovation or whether they represent primarily a form of strategic patenting. That said, strategic patenting and the use of patents to protect follow-on innovation are not necessarily mutually exclusive. There is little discussion about the innovation contained by new active ingredients, but for example the new uses of existing active ingredients in new therapeutic areas, new formulations, new modes of delivery, new combinations of known active ingredients etc. may be regarded as incremental innovation. In this case, secondary patents represent a way of incentivizing and protecting potentially valuable follow-on innovation. This may be particularly valuable for generics producers that want to develop proprietary drugs by modifying existing active ingredients as a lower risk strategy. Take for example a new formulation that allows administering an active ingredient in form of a temperature-stable pill instead of a temperature-sensitive soft-gel version (Amin and Kesselheim, 2012). It is clear that the pill has no added therapeutic benefit over the soft-gel version; at the same time the pill represents an improvement over the soft-gel in terms of ease of drug storage and administration. On the other hand, secondary patents may also be used to extend the time of market exclusivity and to maintain or even expand the market that the product covers during market exclusivity.¹² These objectives can be supported by specific patenting strategies, in particular the creation of patent fences and clusters. Burdon and Sloper (2003) put it bluntly “a key element of any life cycle management strategy is to extend patent protection beyond the basic patent term for as long as possible by filing secondary patents which are effective to keep generics off the market.”
The scarce available evidence on secondary patents suggests that secondary patents are pervasive and that they seem to be used overwhelmingly as a strategic tool. For example, the European Commission found in its 2009 pharmaceutical sector inquiry a primary to secondary patent ratio of 1:7 (EU Commission, 2009: 164). This ratio is higher for pending than granted patents (1:13 vs. 1:5), which suggests that a large number of secondary patent filings are not granted, presumably because they do not meet the statutory patentability requirements or because they are not pursued by the applicant, having served their purpose of increasing uncertainty. The inquiry shows that 57% of secondary patent filings protect formulations, 7% devices, 7% combinations of known active ingredients, 5% polymorphic forms, 4% salts, and the remaining 20% are accounted by a range of claims, such as hydrates or solvates (EU Commission, 2009: Table 20). The study also reveals that if the validity of an originator’s patent is challenged either through post-grant opposition or an invalidation action in court, the majority of secondary patents is invalidated as a result (or their claims restricted) (EU Commission, 2009: 191). Kapczynski et al. (2012) conduct a similar study for the U.S. They look specifically at patenting associated with 342 new active ingredients approved by the U.S. FDA between 1991 and 2005. They find that around 50% of drugs are protected by secondary patents. There is an increase in the share of drugs with secondary patents over time whereas the share of drugs protected by primary patents remains constant. This filing pattern was also found by Sternitzke (2013) who studies the patenting behavior of companies that market Phosphodiesterase Type 5 inhibitors (for the treatment of erectile dysfunction). He also finds that the originator companies included in his study, Pfizer, Bayer and Ely Lilly, file a large amount of secondary patents during later stages of the life-cycle of a drug. This is suggestive of the fact that secondary patents are filed later in the life cycle of a drug to extend the patent life. In fact, the data by Kapczynski et al. (2012) reveals that compound patents are filed before FDA approval whereas secondary patents are filed mostly after approval. The authors estimate that secondary patents generate between 4-5 years of additional patent life on top of compound patents associated with a drug. The mean masks considerable variation. For example, Amin and Kesselheim (2012) found for their case study of two HIV drugs that secondary patents extend patent protection up to 12 years beyond the lifetime of the original primary patents. Another example is Sanofi Aventis’s ARAVA arthritis drug in Australia. Sanofi Aventis effectively extended exclusivity by 10 years through secondary patents.¹³ Other examples of blockbuster drugs are GlaxoSmithKline’s antidepressant Paxil or Pfizer’s cholesterol-lowering Lipitor. In both cases, secondary patents extend patent protection by several years (Hutchins, 2003a, 2003b) relative to the original compound patents. It is, therefore, not surprising that the available evidence indicates a positive correlation between the number of secondary patents for a given drug and higher sales.
An important element in the filing strategy of secondary patents is the creation of legal uncertainty. For example, in their study of HIV drugs Amin and Kesselheim (2012) found overlapping patent claims for a number formulation patents. They also show that some of the formulation patents protect variations of known excipients (for example on new flavors such as peppermint or vanilla), or combinations of known excipients. According to their assessment, these patents are likely invalid. Burdon and Sloper (2003) report the case of AstraZeneca’s Prilosec. While courts in the U.S. upheld secondary patents that AstraZeneca had filed to extend the time of patent protection on Prilosec, the Patents Court in the U.K. invalidated the same formulation patents. This case illustrates that the question of validity of granted secondary patents is particularly unclear. Hemphill and Sampat (2012) even conclude from their analysis of patent challenges by generics companies in the U.S. that challenges target secondary patents and are thus mostly used to restrain attempts by originator companies to extend patent terms beyond the original active ingredient patents through secondary patents.¹⁴
While the literature reviewed tarnishes the use of secondary patents, recent research suggests that the larger the share of generic drug sales within a given therapeutic class, the lower the number of new compounds that enter pre-clinical and early-stage clinical trials (Branstetter et al., 2012). The evidence also suggests that the effect varies depending on “cross-molecular substitutability” within therapeutic classes. This would suggest that restricting secondary patenting might dampen investment in R&D on new compounds. That said, while such effects are conceivable in the world’s most important markets for pharmaceuticals, it is unlikely that generic entry in small emerging economies has any incentive effects on drug companies.
Although there is some recent evidence on secondary patenting by originator companies in Europe and the U.S., so far there is no empirical evidence on the effect of secondary patents on the behavior of generics producers and competition, neither in developed nor developing economies.