Patent Proliferation: A 30-year increase in the number of patents per drug
September 12, 2016
Over the past 30 years, there has been a marked increase in the average number of patents covering a single drug. Will it change the way pharmaceutical patents are litigated?
It was once common wisdom about the pharmaceutical industry that a single product was normally covered by a single patent.[1] This one-product one-patent model distinguished pharmaceuticals from other industries such as software or cellphones, where a single product could be covered by dozens or hundreds of patents.[2]
To some extent, the one-drug one-patent model is valid, and this situation has contributed to the strength of pharmaceutical patents, giving them the power to exclude competing products.[3] It has also shaped pharmaceutical litigation—making commercial-success arguments a mainstay of drug patent suits. Courts have long held that, under certain conditions, a product’s commercial success can be an indicator of its patent’s validity. One of those conditions is that the product is protected by a reasonably small number of patents. When that condition holds—which is often true for drug products—the inference from commercial success to patent validity can be quite strong. So it is not surprising that drug patent suits often involve arguments about the drug’s commercial success.
But does the one-product one-patent rule still hold sway in the pharmaceutical industry? Our analysis suggests that it does so less and less. Over the past 30 years, the average number of patents per drug has been increasing. This shift may usher in changes to the way pharmaceutical patents are litigated.
Increase in the Number of Patents per Drug
Before turning to the data analysis, consider the example of Humira.
Humira, a drug sold by AbbVie, is used to treat arthritis and other autoimmune conditions. It is a biologic, which means it is a large-molecule drug produced by a living system like a bacteria or a plant cell. In 2015, Humira was the top selling drug in the world,[4] but the U.S. patent for its active ingredient (adalimumab) is set to expire in December 2016. Some might expect plummeting sales for Humira in
2017, as biosimilars (the biologic equivalent of generic drugs) flood the market, no longer blocked by Humira’s active ingredient patent. But for AbbVie’s CEO the future is looking up. Why? Because, according to AbbVie, Humira is protected by 70 additional patents, and he believes these patents will prevent biosimilar entry until 2022.[5]
Seventy patents. This is far from the paradigm of one-drug one-patent. AbbVie has sent its product into the world not as a single invention but as a complex, multifaceted product. Its 70 patents include 14 formulation patents, 24 manufacturing method patents, and 22 method of treatment patents, which claim particular indications and dosing regimens to treat them.[6] These patents have been represented as a “maze” that will lure in the attorneys of biosimilar companies and trap them for years.[7]
Humira is not a typical case, but it is emblematic of the changes in the pharmaceutical industry. Our analysis of data from the FDA Orange Book shows that the average number of patents per drug has increased dramatically from 1985 to the present.
Our analysis builds on a prior peer-reviewed article, Hemphill and Sampat (2011), which computed changes over time in the number of patents per drug.[8] That study analyzed data from the FDA Orange Book, which is a continually updated list of all FDA-approved drugs and the patents asserted for each. Using that data, Hemphill and Sampat computed the average and median number of patents per drug for each 3-year approval cohort between 1985 and 2002. They found that, between the first cohort (1985–1987) and the last cohort (2000–2002), the average number of patents per drug increased from
1.9 to 3.9, an increase of 105%, and the median increased from 1.5 to 2.5, an increase of 67%.[9]
Wondering whether this trend had continued, we updated Hemphill and Sampat’s results by collecting current FDA Orange Book data and replicating their methodology. We found that, indeed, the average number of patents per drug had continued to climb. The results are presented in the graph below, which shows the average and median number of patents per drug for drugs approved between 1985 and 2014. The graph shows a steep increase in both the average and median number of patents per drug over time. The average number of patents per drug went from 1.9 for drugs approved between 1985 and 1987 to
6.1 for drugs approved between 2012 and 2014—an increase of 321%. During the same time period, the median number of patents per drug went from 1.5 to 4.0—an increase of 267%.[10]
Patents per Drug for All Drugs Available in Orange Book, 3-Yr Approval Cohorts from 1985 to 2014
Why does this matter? Because, as more drugs are protected by multiple patents, there could be changes in the way drug patent cases are litigated. For one thing, there may be fewer opportunities to offer a drug’s commercial success as evidence of patent validity.
There is a paradoxical effect of having multiple patents for the same product. Drug companies build a large patent portfolio around a drug because they believe it will be a barrier to the launch of generic equivalents. This is AbbVie’s strategy for Humira. The company’s reasoning is that, the larger the number of patents asserted for a drug, the more time and resources will be required for a generic company to challenge the drug’s exclusivity, and the greater the chances that at least one patent will be found valid by the courts.
While this reasoning is valid, this approach also has a paradoxical effect: a large patent portfolio can present certain barriers to proving patent validity in court. In particular, it can make it more difficult for the patent-holder to prove patent validity using commercial success as evidence.
If a commercially successful drug is covered by only one patent, proof of a nexus between the commercial success and the patent can be straightforward. But if that drug is covered by four patents, there may be no simple way to establish a nexus between the commercial success and any one of the patents. If each of the patents is litigated separately, then the plaintiffs face the challenge of explaining why the commercial success should be attributed to this patent rather than one of the other three. Thus, in a world where most drugs are protected by multiple patents, commercial-success arguments may be used less frequently—or alternatively, patent-holders’ attorneys and experts will have to develop analytical methods for linking the commercial success to one or more of the patents.
These considerations are well known to participants in the software and cellphone markets, and they may become increasingly familiar to those in the pharmaceutical market. Drug companies should be aware of them, and attorneys and experts should develop strategies for meeting the new challenges posed by the changing market.
[1] Dan L. Burk and Mark A. Lemley, 2003, “Policy levers in patent law,” Virginia Law Review 89(7): 1575– 1696, at 1590–91; Lisa L. Ouellette, 2010, “How many patents does it take to make a drug: Follow-on pharmaceutical patents and university licensing,” Michigan Telecommunications & Technology Law Review 17(1): 299–336.
[2] Dan L. Burk and Mark A. Lemley, 2003, “Policy levers in patent law,” Virginia Law Review 89(7): 1575– 1696, at 1590–91.
[3] On the strength of patents in the pharmaceutical industry, see Lisa L. Ouellette, 2010, “How many patents does it take to make a drug: Follow-on pharmaceutical patents and university licensing,” Michigan Telecommunications & Technology Law Review 17(1): 299–336, at p. 303.
[4] Hadia Taqi, “AbbVie and Gilead Sciences Lead 2015 Drug Sales with Humira & Harvoni,” (February 11, 2016), http://www.bidnessetc.com/63342-abbvie-and-gilead-sciences-lead-2015-drug-sales-with- humira-harvoni/.
[5] Dan Stanton, “AbbVie: Humira’s patent maze will keep US biosimilars away until at least 2022,” (November 3, 2015), http://www.biopharma-reporter.com/Markets-Regulations/AbbVie-Humira-s- patent-maze-to-keep-US-biosimilars-at-bay-until-2022.
[6] Dan Stanton, “AbbVie: Humira’s patent maze will keep US biosimilars away until at least 2022,” (November 3, 2015), http://www.biopharma-reporter.com/Markets-Regulations/AbbVie-Humira-s-patent-maze-to-keep-US-biosimilars-at-bay-until-2022.
[7] Dan Stanton, “AbbVie: Humira’s patent maze will keep US biosimilars away until at least 2022,” (November 3, 2015), http://www.biopharma-reporter.com/Markets-Regulations/AbbVie-Humira-s- patent-maze-to-keep-US-biosimilars-at-bay-until-2022.
[8] Hemphill, C. Scott and Bhaven N. Sampat. “When do Generics Challenge Drug Patents.” Journal of Empirical Legal Studies 8, no. 4 (December 2011): 613-649, at pp. 619-620.
[9] Hemphill, C. Scott and Bhaven N. Sampat. “When do Generics Challenge Drug Patents.” Journal of Empirical Legal Studies 8, no. 4 (December 2011): 613-649, at pp. 619-620. A comparable study published a year earlier had similar results (Lisa L. Ouellette, “How Many Patents Does it Take to Make a Drug – Follow-on Pharmaceutical Patents and University Licensing,” Michigan Telecommunications and Technology Law Review 17, no. 1 (2010): 299-336.
[10] To compute the values for 2003 to 2014, we used a similar methodology to Hemphill and Sampat (2011). The only difference is that our analysis uses data only from the 2016 Orange Book and does not incorporate historical editions of the Orange Book published between 2003 and 2014. As a result of using only 2016 Orange Book data, it is likely that there are some patents that may not be included in our analysis because they expired and were removed from the Orange Book. Therefore, the calculations for 2003 to 2014 may be an underestimate of the actual average and median number of patents per drug.