Never say never – Why the High Income Countries that opted-out from the Art. 31bis WTO TRIPS system must urgently reconsider their decision in the face of the Covid-19 pandemic

To limit the ability of patent owners to charge excessive prices for patented medicines, countries can grant compulsory licences which permit others to produce or import (competing) generic versions of the patented medicine. Even the suggestion that a compulsory licence might be granted will impact patent holders’ pricing decisions. The Financial Times (of London) recently said that: ‘[Facing the Covid-19 pandemic emergency] The world has an overwhelming interest in ensuring [drugs and vaccines effective against the virus] will be universally and cheaply available. Fortunately, trade rules allow compulsory licensing. If necessary, it must be used.’ It is therefore worthwhile to take a moment to understand how compulsory licensing has become complicated by the ‘outsourcing’ of Active Pharmaceutical Ingredient (API) manufacture to a small number of countries, notably India and China, and why it means that the High Income Countries that opted-out from the ‘Art. 31bis system’, which is now a part of the World Trade Organization (WTO) Trade-Related Aspects of Intellectual Property Rights (TRIPS) Agreement, must urgently reconsider their decision in the face of the Covid-19 pandemic.

What does ‘outsourcing’ API manufacture mean? There are at least two key steps in the production of a finished medicine product. Firstly, the manufacture of the necessary Active Pharmaceutical Ingredient (API) from raw materials. Secondly, the appropriate formulation of the API to make the particular medicine product. These two steps can either be completed by a single company or by two different companies. Although there are many companies in the world capable of formulating APIs into finished medicine products, there are far fewer who specialise in manufacturing APIs and many of these are now located in India and China. Following the entry into force of the TRIPS Agreement, patent holders are able to obtain the most far-reaching types of patents, which cover the Active Pharmaceutical Ingredient (API) (new chemical entity) itself, in both India and China. Major pharmaceutical companies wishing to increase their organisational flexibility and drive down costs therefore now routinely outsource the production of their APIs (under patent licence) to lower cost Contract Manufacturing Organisations (CMOs) in these countries. Although such outsourcing was already occurring before the entry into force of the TRIPS Agreement, its pace increased and carried on increasing afterwards. To take just one illustrative example,  following ‘the general trend within the pharma sector’, AstraZeneca, announced their intention in 2007 to begin withdrawing from API manufacture and outsource it instead to CMOs in China. It is now often suggested that up to 80% of the APIs used in the US come from India and China (see, for example, US Senate Finance Committee Chairman Chuck Grassley’s letter to the Department of Health and Human Services (HHS) and the Food and Drug Administration (FDA)) but, given the lack of transparency often associated with outsourced API manufacture, it is difficult to be certain.

To see how this affects the use of compulsory licensing, let’s take the example of a (WTO Member) country that wants to make use of compulsory licensing to supply its population with a generic version of a patented medicine (Art. 31 TRIPS) but that lacks domestic API manufacturing capability. This country could use compulsory licensing to import a finished medicine product, probably from a country where the API is manufactured. Or it could use compulsory licensing to import the necessary API from a country where the API is manufactured so that one of its own domestic generic manufacturers could formulate and finish the medicine product. The problem is that, whilst a country can grant a compulsory licence which permits the import of the medicine / API, it cannot command the medicine / API producing country to permit its export. If there is a patent in the medicine / API producing country, which is now likely for any commercially important new medicine, then a compulsory licence to permit the manufacture and export of that medicine / API will have to be granted in that country too. Here the TRIPS Agreement raises a significant problem. Art. 31 (f) TRIPS requires that any compulsory licences granted by WTO Members must be ‘predominantly for the supply of the domestic market of the Member authorizing such use’. This is not an inherent limitation of compulsory licensing but is an imposed constraint. It presents a severe limitation on the ability of medicine / API producing countries, including India and China, to grant a compulsory licence permitting the export of the medicine / API to other countries. Even though India and China have very large domestic markets, the ‘non-predominant’ portion of the medicine / API production will be insufficient to supply many other countries, much less the rest of the world. Thus, even though the (WTO Member) country in our example has granted a compulsory licence to supply its population with a generic version of a patented medicine, the Art. 31(f) export restriction may prevent it from having any practical effect.

This problem was just one of a number of concerns about the impact of the TRIPS Agreement on public health which continued to simmer in the late 1990’s. Addressing at least some of them, a WTO Ministerial Conference agreed the ‘Doha Declaration on the TRIPS Agreement and Public Health’ in 2001. Paragraph 6 of the Declaration addressed the Art. 31(f) problem by promising that an ‘expeditious solution’ would be found to permit ‘WTO Members with insufficient or no manufacturing capacities in the pharmaceutical sector’ to make effective use of compulsory licensing under the TRIPS Agreement. A partial waiver of Art. 31(f) was subsequently agreed under the ‘Decision of the General Council of 30 August 2003’ which was formalised in 2017 with the Art. 31bis amendment to the TRIPS Agreement. The system established under the waiver / Art. 31bis amendment permits WTO Members to grant a ‘special’ compulsory licence for export without the Art. 31(f) limitation to ‘predominant’ domestic supply. However, the core provision is hedged with potentially cumbersome procedures and it can only be regarded as a limited solution to the Art. 31(f) at best. There has only been a single example of the use of the system so far (under the waiver) and the slow pace at which it occurred has been strongly criticised. Nevertheless, Art. 31bis is now the only attempt to address the Art. 31(f) problem yet agreed and there is perhaps room, with political will, to make it work more effectively. The remarkable thing is that several HICs (‘Australia, Canada, the European Communities with…its member States, Iceland, Japan, New Zealand, Norway, Switzerland, and the United States’) unilaterally committed that they would not make use of the system as importers (General Council Decision Art, 1(b), footnote 3; Art. 31bis, Annex, Art. 1(b), footnote 3). This not only impacts access to affordable generic medicines in these opt-out HICs but, since the economies of scale that could have been harnessed by exporting to these comparatively wealthy HICs are curtailed, the price of the generic medicines that could be produced for other WTO Members may well be higher than it could have been. Even more remarkably, this opt-out extends even to ‘situations of national emergency or other circumstances of extreme urgency’. So even in these situations, the opt-out HICs are stuck with reliance on only the ‘non-predominant’ portion of ‘ordinary’ compulsory licensed medicine / API production in India and China. This is entirely unsatisfactory.

The opt-out HICs should have provided themselves with a way to opt-back into the Art 31bis system if they needed to, even if just in ‘situations of national emergency or other circumstances of extreme urgency’. The system is framed in terms of ‘WTO Members with insufficient or no manufacturing capacities in the pharmaceutical sector’. Whether or not such manufacturing capacities are sufficient must be regarded as a dynamic question, the answer to which may change over time (Art. 31bis, Appendix to the Annex: ‘Member has some manufacturing capacity…found that…currently insufficient for the purposes of meeting its needs…’). If the pharmaceutical manufacturing capacity of the opt-out HICs has become heavily reliant on the importation of API from countries such as India and China then it arguably no longer represents the ‘sufficient (end-to-end) manufacturing capacity’ which supports the effective independent use of compulsory licensing. For these purposes, and even if only in some key sub-sectors, it may have become ‘insufficient’.

This consideration is given urgent force by the present Covid-19 pandemic emergency. It is not clear which medicines (or diagnostics, vaccines or medical devices) will prove to be most potent in combatting Covid-19. Whichever they are, they will hopefully be swiftly available at the necessary scale and affordable price either via the intellectual property right holders themselves or in ‘non-conflictual’ generic versions (whether through the licensing of rights, perhaps via a Covid-19 intellectual property pool, the waiving of rights or where there are no rights existing at all). However, if not, governments must have practically effective compulsory licensing powers at their immediate disposal. Take the example of remdesivir, one of the medicines in clinical trials for use in the treatment of Covid-19. Gilead’s instinctive response to the outbreak has been to try to reinforce their exclusive rights over remdesivir by persuading the FDA to grant them an orphan drug designation in the United States (now hastily relinquished), rather than reassure about universal access. Who knows what their commercial strategy will be if remdesivir proves effective? Gilead are currently trying to increase their manufacturing capacity (in conjunction with their licensed CMOs) but have recently struggled to maintain patient access to remdesivir even through their compassionate use program. Whilst other API manufacturers will be technically capable of manufacturing the remdesivir API, they will be prevented from supplying it in the absence of either a voluntary licence (from Gilead) or, if necessary, a compulsory licence (for example, in India or China), in which latter case the Art. 31(f) problem immediately presents itself.

Governments should therefore be urgently reviewing their ability to use their compulsory licensing powers, in practice not just on paper. National compulsory licensing systems (including ‘public non-commercial use’ powers for the government itself) should be scrutinised and, if necessary, action taken to remove potential roadblocks, as has already now happened in Germany, Canada, Chile and Ecuador. Israel has already used its compulsory licensing powers. At the international compulsory licensing level, the Art. 31(f) problem remains the elephant in the room. By ruling out the use of the Art. 31bis system for importation in all circumstances, even in ‘situations of national emergency or other circumstances of extreme urgency’, the opt-out HICs have left themselves dangerously exposed to the Art. 31(f) problem and, in the face of the Covid-19 pandemic emergency, such a dangerous exposure could not have come at a worse time.

The broader (and, it has to be said, long-standing) concerns about the vulnerability of international medicine manufacturing supply chains will have to wait for another time. Commercial API production cannot be turned on and off at the flick of a switch. The Covid-19 outbreak in China has already caused significant interruptions to the international supply of APIs, not least to India. Facing knock-on shortages, India responded by restricting the export of more than twenty essential APIs (and formulations made from them). The US President, Donald Trump, had to call the Indian President, Narendra Modi, to ask in person if the particular restriction on hydroxychloroquine (another of the medicines in clinical trials for use in the treatment of Covid-19) could be lifted to permit supply to the US. Hydroxychloroquine is a comparatively old medicine, however, and there appear to be no patent barriers to its manufacture and export. Remdesivir would be different.

At the instant moment, the opt-out HICs will just have to deal with the threat posed by the Covid-19 pandemic emergency as best they can. It is in their vital interests to make sure that they have a workable solution to the Art. 31(f) problem. Following the call in a widely supported open letter, the opt-out HICs should therefore urgently change course and work with other WTO Members to find a way to opt back into the Art. 31bis system, even if limited to ‘situations of national emergency or other circumstances of extreme urgency’. If that fails, they must work with other WTO Members to find a different solution, whether looking to previously suggested alternatives (for example, based on Art. 30 TRIPS) or supervening emergency powers at national and international levels (for example, based on Art. 73 (b) TRIPS). International cooperation, rather than division, will be key. The scale and seriousness of the threat posed by the Covid-19 pandemic emergency means that all governments, including those of the presently opted-out HICs, must have practically effective compulsory licensing powers ready and able to be used should they need them.