Corporate Self-Interest And Strategic Choices: Gilead Licenses To Medicines Patent Pool

Print This Post Print This Post

The views expressed in this column are solely those of the authors and are not associated with Intellectual Property Watch. IP-Watch expressly disclaims and refuses any responsibility or liability for the content, style or form of any posts made to this forum, which remain solely the responsibility of their authors.

By Prof. Brook K. Baker, Health GAP

Summary: Despite some important advances to the scope of its previous purely voluntary licenses on key antiretroviral medicines, Gilead’s licenses with the UNITAID-sponsored Medicines Patent Pool (MPP) contain unfortunate restrictions that undermine the potential impact of the MPP on access to more affordable generic ARVs of assured quality throughout the developing world. Major defects in Gilead’s licenses include their geographical exclusions of many low- and middle-income countries in North Africa, Asia, and Latin America and the bizarre provisions restricting licenses to Indian generics only and requiring that all licenses source Active Pharmaceutical Ingredients only from Gilead or from licensed Indian suppliers. A careful review of available information on Gilead’s patent landscape reveals that it is relying on weak tenofovir patent claims in India as the basis for collecting royalties on sales in 109 of 111 countries where it has no patent claim whatsoever. Generics companies may wisely choose to delink Gilead’s tenofovir API license from the licenses for its pipeline medicines so that they can supply not only those 109 countries (plus India), but also many other middle-income countries where Gilead holds no blocking patent, such as Brazil. In addition, countries should closely analyze Gilead’s patent claims domestically and in India to ascertain whether they need to issue compulsory licenses on Gilead products. If so, sec. 92A of the Indian Patents Act is easily used to permit export of needed quantities to countries with insufficient local manufacturing capacity. Indeed, excluded countries should act quickly and decisively to signal both to Gilead and other ARV right-holders that refusal to cooperate fully with the Patent Pool will result in self-protective measures to access affordable medicines.

Even though the majority of this analysis is going to focus on a critical and strategic analysis of the Gilead/Medicines Patent Pool license, it is appropriate to start with some praise for Gilead – it is the first pharmaceutical company to join the MPP; it has licensed important pipeline medicines as well as existing ARVs; it will permit combinations with non-Gilead products; it is transferring technical know-how expediting efficient production on a commercial scale; it provides royalty-free licenses and expansive access for pediatric formulations; it permits referencing of regulatory data to fast-track registration of generic equivalents; and the field of use of the tenofovir license includes both HIV and hepatitis B prevention and treatment.

The major defects in Gilead license include its limited geographic scope (and differing geographic scope for each ARV); its unreasonable limitation on the sourcing of Active Pharmaceutical Ingredients (Gilead or MPP licensed Indian API suppliers only); its restriction of product licenses to only Indian generics companies; and its reliance on extraordinarily weak patent claims in India on tenofovir (TDF) as the basis for collecting royalties on sales in 109 countries where it has no viable patent claim.

Fortunately, the Medicines Patent Pool has negotiated favorable license terms that will permit potential licensees (generics manufacturers) and excluded middle-income countries to access affordable TDF-based products. By publishing the Gilead/MPP licenses and the prospective Gilead/MPP/Licensee licenses, analysts can review the licenses in detail. Two provisions stand out as options for improved access: (1) the right of licensees to unbundle the license for a particular ARV from the others, e.g., TDF from licenses for cobicistat (COBI), elvitegravir (EVG), and the Quad product (TDF/FTC/COBI/EVG), and (2) the right of licensees to supply countries where required compulsory licenses have been granted.

This analysis will clarify how generics companies and excluded countries can selectively deploy the Gilead/MPP license, existing TRIPS-compliant flexibilities, and provisions of India’s Patent Act to maximize global access to generic versions of Gilead products. Because of strategic differences between TDF and the other three licensed medicines, these two classes of products will be discussed separately.

Background – How and Why Patents Matter

Patents are government-granted rights to exclude others from manufacturing, selling, distributing, or importing the patented product or using the patented invention for a set period of time (currently 20 years). The inventor does not receive a patent in a particular country unless the inventor files an application in that country (or in some cases to a regional body) and unless the patent application is examined to confirm that the claimed invention is novel, involves an inventive step, is industrially applicable, and is adequately disclosed.

In other words, there is no such thing as a worldwide patent – patents must be filed and granted country by country. Because of uncertainty about the market for the invention and because of the cost of filing and maintaining a patent, many pharmaceutical inventors do not file patent applications in every country. In particular, they tend to exclude countries that are small and poor and that do not have domestic pharmaceutical capacity. Similarly, countries vary in how strictly they interpret patentability criteria (novelty, inventiveness, and industrial applicability) and also vary with respect to their exclusions from patentability (for example the right to patent new uses or new indications of known medicines). Finally, at least some least developed countries have taken advantage of their 2016 TRIPS waiver with respect to granting pharmaceutical product patents and data protection, though many more need to do so. Accordingly, a patent application on a particular medicine might be granted in one country and denied in another.

Complicating matters further, there are often many patents filed and granted on a particular medicine – patents on the active pharmaceutical ingredient, on the dosage, on the formulation and method of delivery, on the use, on combinations of two or more medicines into one pill, on the process for manufacturing the product, etc. Pharmaceutical companies frequently game the system, by seeking patents on minor variations in chemical entities and minor changes in the formulation, new uses, and revised methods of production in order to “evergreen” the exclusivity period for a medicine by staggering successive 20-year patents on the same active pharmaceutical ingredient.

The last background issue to clarify is that determining the patent status of a medicine is important in both the country where it is manufactured and distributed and, in the cases of export/import, in the country where the medicine is imported and consumed. Accordingly, if a country without a blocking patent wants to import a generic version of a medicine, it might still not be able to do so if there is a patent in the country of manufacture. Similarly, even though there is no patent in the country of manufacture, there may be patent blockage in the country of importation.

Compulsory Licenses Can Lawfully Circumvent Patent Blockages

Patent blockages in either the country of manufacture and/or the country of importation/use can be overcome with compulsory licenses. Even though many countries have not yet amended their national legislation to take full advantage of TRIPS flexibilities so as to be able to expeditiously issue compulsory licenses, most developing countries do have compulsory licensing and government use mechanisms that allow them to license local manufacturing and to import from generics manufacturers in other countries.

If an importing country has no blocking patent, but the medicine is patented in the country of manufacture/export, then a compulsory license will need to be issued in compliance with TRIPS and national law in the exporting country. Regular compulsory licenses will allow export of non-predominant quantities (TRIPS Art. 31(f)); competition-based licenses will allow export of unlimited quantities (TRIPS Art. 31(k)); and Paragraph 6/August 30 Decision licenses will permit export of needed quantities if the county lacks sufficient domestic pharmaceutical capacity. If there is a blocking patent not in the country of export, but rather the country of import and use, then a TRIPS-compliant and national-law-compliant compulsory license will have to be issued in importing country to generics manufacturer(s) in the exporting country (an Art. 31 CL or a Paragraph 6/August 30 Decision CL). If there are blocking patents in both the country of import and the country of export, then two compulsory licenses will have to be issued – one in each country.

India’s patent law is of particular significance because the Indian generics industry supplies over 90 percent of generic HIV medicines used in developing countries. India has some special advantages with respect to its system for exporting to satisfy the need for more affordable medicines of assured quality in countries that lack sufficient manufacturing capacity to produce locally. Section 92A reads as follows:
(1) Compulsory licence shall be available for manufacture and export of patented pharmaceutical products to any country having insufficient or no manufacturing capacity in the pharmaceutical sector for the concerned product to address public health problems, provided compulsory licence has been granted by such country or such country has, by notification or otherwise, allowed importation of the patented pharmaceutical products from India.
(2) The Controller shall, on receipt of an application in the prescribed manner, grant a compulsory licence solely for manufacture and export of the concerned pharmaceutical product to such country under such terms and conditions as may be specified and published by him
(3) The provisions of sub-sections (1) and (2) shall be without prejudice to the extent to which pharmaceutical products produced under a compulsory licence can be exported under any other provision of this Act. Explanation.- For the purposes of this section, “pharmaceutical products” means any patented product, or product manufactured through a patented process, of the pharmaceutical sector needed to address public health problems and shall be inclusive of ingredients necessary for their manufacture and diagnostic kits required for their use.

Although one interpretation of this Section 92A might be that it was enacted to operationalize the Paragraph 6/August 30 Decision system within India and thus that it requires all the notifications and special procedures of that complex and labyrinth system, an alternative and superior interpretation based on the literal language of Section 92A is that a country with insufficient manufacturing capacity can trigger a mandatory export compulsory license in India merely by: (1) providing evidence of the grant of an importing country compulsory license or of having otherwise allowed importation of a patented pharmaceutical product from India, via notification or otherwise, and (2) ensuring that its prospective compulsory licensee(s) have filed the prescribed application for an export compulsory license with the Controller of Patents.

To date, it does not appear that the Patents Office has issued any rules or regulations specifying specific required terms or conditions. Accordingly, it might be possible for any country, including those excluded from Gilead’s geographic license, merely to notify the Indian Patent Controller by letter or otherwise that it has chosen to allow importation in the absence of sufficient local manufacturing capacity (or has issued a domestic compulsory license allowing importation from India). Even if such an interpretation would not adhere the to stricter requirements of the Paragraph 6/August 30 Decision System, it would nonetheless arguably be TRIPS-compliant as an Art. 30 limited exception to the predominantly-for-domestic-use rule in Art. 31(f). Gilead would appear to have no contractual basis to challenge this interpretation because the Gilead/MPP/Licensee license merely requires Gilead and Licensee agreement (not to be unreasonably withheld) regarding the existence, scope, and content of any compulsory license, not concerning the legal regime under which it was granted.

Even if the Section 92A is interpreted to require Paragraph 6/August 30 Decision notifications and anti-diversion measures, the India export-CL procedure is very easy to use and results in mandatory issuance of the requested export license.

Gilead’s Patents Landscape Weak on Tenofovir, Stronger on Pipeline

Gilead has not filed patent applications on tenofovir in the vast majority of low-income countries, nor even in many middle-income countries. In fact, of the 111 countries included in the geographical scope of the tenofovir MPP license, Gilead has patent applications pending or granted in only 2 of the licensed countries, India and Indonesia. Moreover, some of Gilead’s patent applications for tenofovir have been rejected In India under its strict definitions of and exclusions from patentability, which prevent patenting of variations of known chemical entities unless they show a significant therapeutic advantage. Given tenofovir’s discovery in 1985, most analysts think that Gilead’s unexamined divisional patent applications concerning the disoproxil prodrug and the fumarate salt will fail under Sec. 3(d) of the India Patent Act and that Indian manufacturers can easily make TDF-based products without violating the one Indian process patent that Gilead has been granted to date.

As revealed in the Patent Status Chart below, not only has Gilead never patented tenofovir in 109 of the 111 countries in the TDF-licensed territories, it has also failed to secure tenofovir patents in a number of excluded middle-income countries with a high HIV-burden including Argentina, Brazil, Chile, Colombia, Malaysia, Peru, Philippines, Ukraine, and Uruguay. Unfortunately, tenofovir is patented in some of the 45 excluded low- and middle-income countries, including important markets in Mexico and China. The patent status of tenofovir in the other 34 excluded countries remains unclear.

The patent landscape for cobicistat and elvitegravir/Quad is a little more complex. Although few patents have actually been granted either in licensed or unlicensed territories, there are a large number of pending patent applications, many of which are likely to be granted, even in India, since both medicines involve a new chemical entity. Even so, there do not appear to be patents pending on cobicistat in 61 of the 102 licensed countries nor on elvitegravir/Quad in 58 of the 99 licensed countries.


TenofovirLicensed Territories (111)Non-licensed Territories (45)
No patent109 countriesArgentina, Brazil, Chile, Colombia, Malaysia, Peru, Philippines, Ukraine, Uruguay
Patents filed or grantedIndia and IndonesiaChina and Mexico
Patent status uncertain34 countries
CobicistatLicensed Territories (102)Non-licensed Territories (54)
No patent filed61 countriesChile, Colombia, Malaysia, Peru, Philippines, Ukraine, Uruguay
Patents filed or grantedARIPO (15 of 17 members), Eurasian Patent Organization (5 of 9 countries), African Union Territories (OAPI 16 countries), India, Viet Nam, South Africa, Bolivia, and PakistanArgentina, Botswana, Brazil, China, Egypt, Mexico, Morocco, Namibia, Azerbaijan, Belarus, Kazakhstan, Russia
Patent status uncertain35 countries
Elvitegravir and QuadLicensed Territories (99)Non-licensed Territories (57)
No patent58 countriesUruguay
Patents filed or grantedARIPO (15 of 17 members), Eurasian Patent Organization (5 of 9 countries), African Union Territories (OAPI 16 countries), India, Viet Nam, Nigeria, South Africa, BoliviaAzerbaijan, Argentina, Azerbaijan, Belarus, Brazil, Botswana (process), Chile, China, Colombia, Kazakhstan, Malaysia, Mexico, Namibia (process), Philippines, Peru, Russia, Thailand, Ukraine
Patent status uncertain38 countries

* This information is compiled from the Medicines Patent Pool Patent Status Database and from Gilead’s patent disclosures in the Gilead/MPP license. Note: Although there is reliable information about where patents have been granted or filed within the licensed territories based on Gilead’s disclosures, Gilead did not disclose information concerning its claimed patents in non-licensed territories. Information on non-licensed territories is from the Medicines Patent Pool patent database, available here:

Why Did Gilead Restrict APIs?

The Gilead/MPP license and Gilead/MPP/Licensee license restrict API and pharmaceutical product licenses to Indian generics manufacturers and further requires Indian licensees to access active pharmaceutical ingredients from: (1) authorized Gilead API suppliers, (2) the company’s own integrated API manufacturing division, or (3) from another Indian Gilead/MPP licensee. By restricting API suppliers to Gilead’s supplier or to Indian licensees, Gilead is excluding potential API suppliers from China (a major API producer) or from Asian, Latin American, or African companies that might emerge as viable producers of assured quality APIs. Similarly, by adopting an all-in-India approach, Gilead is also preventing licensed Indian API producers from exporting APIs to other countries where manufacturers could lawfully formulate generics of assured quality.

Gilead argues that it is limiting suppliers to Indian companies that are GMP certified and that have received regulatory approval from the European Medicines Agency or the US Food and Drug Administration or WHO pre-qualification because it has established relations with 13 such companies via its prior voluntary licensing agreement and because it is confident of the consistent quality and cost-effectiveness of their manufacturing processes. A cynic might reach a different conclusion, namely that Gilead is seeking to benefit from further economies-of-scale from its API suppliers (who could thereby sell APIs even cheaper to Gilead, thereby increasing its expected profits because Gilead is unlikely to lower rich country prices) and because it is seeking to exclude the development of a robust generics industry in China and to prevent API production in a least developed country (like Uganda or Bangladesh) that might benefit from the 2016 LDC TRIPS waiver on patenting of pharmaceutical products (and perhaps an extension thereof).

In any event, by this decision, Gilead has frustrated efforts of the African region to become more self-sufficient with respect to pharmaceuticals, though Gilead might point to its one license with Aspen Pharmacare, which is licensed to sell tenofovir throughout the sub-Saharan region, as evidence of its commitment to African producers.

Strategic Choices

Generics companies and developing countries excluded from the geographic scope of Gilead’s API licenses face important strategic choices in responding to the Gilead/MPP License. This section will first discuss the options facing potential licensees and then the options available to excluded countries.

Tenofovir API License – Consider Rejection

The most important decision generics companies face is whether to reject the API license for tenofovir. Fortunately, they are permitted to “unbundle” the TDF API license and still retain licenses for other Gilead ARVs because of shrewd bargaining by the Patent Pool. Companies could wisely make this rejection decision based on the weakness of Gilead’s TDF patent claims in India and on the strong likelihood that those claims will be rejected by the Indian Patent Office and courts. The grounds for doubting the patentability of TDF in India include: (1) the original tenofovir compound and its antiviral properties were discovered by Czech researchers in 1985, long before the effective date of the TRIPS Agreement, which did not require India to consider patents on pharmaceutical products invented before 1995; and (2) the claims on disoproxil prodrug and the fumarate salt lack novelty and inventive step and both involve variations of existing chemical entities and offer no additional and significant therapeutic efficacy, as defined under Indian law. The one process patent that has been granted does not appear to be a barrier to alternative efficient processes for manufacturing tenofovir-based medicines.

If Indian companies opt out of the TDF API license and if Gilead’s putative patent claims in India are meritless, Indian generics companies would still be able to sell TDF-based medicines in 110 of the existing licensed territories (only Indonesia would be excluded if tenofovir patents are eventually granted there). In addition, TDF license-free companies would face no patent bars to sales in Argentina, Brazil, Chile, Colombia, Malaysia, Peru, Philippines, Ukraine, and Uruguay, and perhaps in other 34 low- and middle-income countries where the patent status of tenofovir is uncertain.

Similarly, rejecting the TDF API license would not prevent companies from co-formulating tenofovir with other Gilead licensed APIs (cobicistat and elvitegravir) at least in the COBI and EVA/Quad licensed territories. Moreover, these Indian companies could be selected as export/import compulsory licensees in countries where tenofovir is patent protected.

Accordingly, except perhaps with respect to the loss of CL-free exports to Indonesia, some residual uncertainty about tenofovir’s ultimate patent status in India, and the loss of automatic data referencing rights in countries where data exclusivity might (or might not) interfere with product registration, there is little or no downside to rejecting the TDF API license.

COBI, EVG/Quad Licenses – Consider Acceptance

On the other hand, given the probable patenting of cobicistat and elvitegravir in India, qualified Indian generics companies would probably accept COBI and EVG API licenses. (Although the Quad product will probably not be separately patented in India because it would be merely a compound or admixture of known substances [Sec. 3(d) and Sec. 3(e)], its manufacture would be blocked unless companies had licensee rights and covenants not to sue with respect to COBI, EVG, and FTC – assuming TDF is not patent protected.) Accepting the licenses would restrict sales to the 103 or 99 licensed countries, but the Gilead/MPP licensees would still be eligible to be export/import manufacturers and suppliers if required compulsory licenses in India and the country of import/use (if needed) were issued.

Excluded Importing Countries – No Domestic Patent Blockage

Countries with no domestic patent blockage on tenofovir, but that are excluded from the Gilead/MPP TDF territory, could chose a generics supplier that is lawfully producing tenofovir because: (1) it is an Indian company that rejected the TDF API license but is manufacturing for domestic use in India and for export because it is satisfied that Gilead’s tenofovir patent applications in India are without merit; (2) it is a generics producer that is lawfully producing in a non-India country (perhaps even an LDC thereby benefiting from the 2016 TRIPS waiver); or (3) the requesting country has insufficient domestic manufacturing capacity and has notified the Indian Controller of Patents as required and secured a mandatory compulsory license for export/import of needed quantities pursuant to Sec. 92A of the Indian Patents Act. With respect to COBI, EVG, and the Quad product, countries might find fewer or no Indian companies that have rejected those API licenses in which cases they would have to source pursuant to options (2) and (3) above. In any event because quality is crucial, importing countries should seek generic ARVs from generics companies that are GMP certified and whose products have received WHO pre-qualification or regulatory approval by a stringent regulatory authority.

Excluded Importing Countries – With Domestic Patent Blockage

If tenofovir, cobicistat, elvitegravir, and/or the Quad product are patented in a country excluded from the relevant Gilead territories, those countries will need to issue a domestic compulsory license allowing both domestic production and importation. If importation is required, the country will need to confirm the Gilead patent landscape in the manufacturing/exporting country to determine whether a compulsory license must also be issued there. If the exporting country is India, the importing country can seek an ordinary CL for importation of non-predominant quantities, a competition-based license for importation of unlimited quantities, or a Sec. 92A license for export of needed quantities to countries with insufficient domestic manufacturing capacity. (Note: whether Sec. 92A requires the full panoply of notifications and anti-diversion measures required by the Paragraph 6/August 30 Decision system is currently uncertain.)


The Gilead/MPP license is far from perfect, but Indian generics companies and excluded developing countries still have options to make strategic choices that will maximize access to more affordable ARVs. Generics companies should at least consider rejecting the tenofovir API license even if they accept the others. Excluded or unlicensed countries could still source medicines from generics suppliers where patent blockages do not exist or where required compulsory licenses are obtained, as needed, in the export country, India, or both. Where compulsory licenses are needed, they are easily obtained in India and hopefully pursuant to domestic legislation. However, excluded countries should act decisively and perhaps even in a coordinated fashion in seeking CLs both to increase access to affordable medicines and to send a message to right-holders that excluding countries from the Patent Pool geographic scope will be met with compensatory strategies.

Brook K. Baker is a law professor at Northeastern University School of Law (US) and an affiliate of its Program on Human Rights and the Global Economy. He is also an honorary research fellow at the University of KwaZulu Natal, Faculty of Law, South Africa. He is a policy analyst for Health GAP (Global Access Project) and writes frequently on IP, trade, and access to medicines issues.

Creative Commons License"Corporate Self-Interest And Strategic Choices: Gilead Licenses To Medicines Patent Pool" by Intellectual Property Watch is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.


Leave a Reply