TWN Info Service on Health Issues No. 14

9 November 2005

Roche's Secret, Sub-Licenses for Tamiflu will not help poor

As governments scramble to stockpile Tamiflu in view of the avian flu threat, the patent holder Roche has been busy trying to control pricing and profits. Roche is into secret licensing agreements to outsource manufacturers (sub-licensees). The following article explains that sub licensing will increase Roche’s market share for Tamiflu in the rich countries but will not ensure access for the poor in the developing countries.

With best wishes
Evelyne Hong


Roche's Secret, Sub-Licenses for Tamiflu Will Not Bring Poor People in from the Cold

Brook K. Baker, Health GAP

October 31, 2005

After stumbling badly in its initial efforts to tightly control production of Tamiflu, an antiviral important in containing possible human bird-flu pandemic, Roche is now attempting to brighten its tarnished image with secret negotiations for sub-licenses.  These licenses, while they might speed stockpiles for rich countries and pad profits for Roche, will do little to ensure access for poor people in developing countries.

At first Roche insisted on using its patent-based rights to limit production to its manufacturing facilities and three favored, but secret, out-source manufacturers (sub-licensees).  When public outcry revealed that it would take Roche ten years, even with expanded capacity, to satisfy minimal requests for international stockpiles, Roche made widely publicized offers to "enter into discussion with any party who is able to fully or partially produce substantial quantities of Tamiflu, for emergency pandemic use, which [sic] a specialized timeline and in accordance with appropriate quality specifications, safety and regulatory guidelines." 

So far, Roche has revealed discussions with Teva, Barr, Mylan, and Ranbaxy, all major generic producers, and it reports that it is "in discussions" with many anxious countries, including China, Taiwan, and Thailand. At the same time that Roche opened its doors to discussions, it made threats against use of compulsory licenses, arguing that the 10-step manufacturing process is time-consuming, difficult, and dangerous - indeed "potentially explosive."

Few, if any details, have emerged about the proposed terms of Roche's secret licensing agreements.  However, public statements from Roche indicate that it is offering time-limited sub-licenses for public-sector, emergency stockpiling only.  By entering into sub-licensees, whereby it strictly controls and has rights to the pills produced, Roche will retain its exclusive access to all lucrative markets.  Although government purchasers may well get a volume discount, Roche will continue to control pricing and profits.

At present, the price of a 10-pill course of Tamiflu treatment in the U.S. is $60 ($6 pill).  Roche expects to make as much as $925 million from Tamiflu sales in 2005, up dramatically from $266 million in 2004.  Future profits look even brighter based on Roche's eight-fold expansion of production capacity. An interesting aspect of Roche's proposal to add more sub-licenses is that it may reap even larger profit margins.  These new sub-licenses are likely to reduce costs in two ways:  (1) the price of active pharmaceutical ingredients may well fall as additional manufacturers of key ingredients (like shikimic acid) enter the market and reach economies of scale and (2) lower-cost producers, particularly generic producers in India that have 1/7 the overhead, will pass their costs savings on to Roche.

If Roche succeeds in controlling large segments of the market by sub-licensing, not only will its profits improve, but it will also deter generic entry even if compulsory licenses are issued. By capturing all of the government-stockpile and private-sector markets in rich countries, Roche will succeed in leaving the poorer and smaller developing country markets to generic producers.  These markets will be less attractive for several reasons:  (1) they have less purchasing power; (2) their medicine transportation and distribution systems are weaker; (3) the costs of obtaining marketing approval are more onerous in proportion to market size; and (4) in many of the smaller countries, Roche has not even registered Tamiflu, making registration of a generic equivalent even more costly and time-consuming.

A Roche-controlled alternative to sub-licenses, sought by some countries, would be voluntary licenses, whereby the licensee would have a right to sell the pills it produced, but Roche could nonetheless tightly control sales with marketing and pricing restrictions.  For example, Roche could preclude sales to the private sector and/or it could exclude certain markets altogether, e.g. the U.S., Europe, and Japan.  These voluntary licenses might increase the quantity of generic oseltamivir being produced, but they might not create global access or increased affordability.

A much better alternative, from a public health perspective, is a government issued compulsory license or government use order (assuming that licensees can reverse-engineer quickly or that Roche can be compelled to disclose technical know-how).  If such a license were granted to respond to a public health emergency, to allow government, non-commercial use, or to remedy an anti-competitive practice, there would be no obligation to enter into preliminary negotiations with the patent-right-holder.  Instead, Roche would be entitled to a reasonable royalty only. The licenses, in turn, would be permitted to engage in large-scale production for all internal markets (except government use order would be for government, non-commercial use only) and for export (non-predominant quantities for ordinary licenses and unlimited quantities for competitive-based licenses).

This option works well for countries that have local pharmaceutical capacity to make Tamiflu, assuming they can produce at efficient economies of scale.

Countries with insufficient manufacturing capacity, however, would need to rely on imports either from countries where Tamiflu was not patented or from countries granting export licenses.  Because Tamiflu is a post-1995 discovery, it is entitled to patent protection in most countries where the patent holder had filed a patent application.  In some countries, like India, 1995-2005 patent applications were placed in a "mailbox" awaiting patent consideration under new WTO-mandated patent rules effective January 1, 2005.  Thus, countries with Tamiflu patents and Tamiflu patents pending might need to rely on the 30 August 2003 Paragraph 6 Implementation Decision in order to grant import and export licenses.

The 30 August Decision system, however, is terribly flawed because it requires new national laws and a complex array of notices and licenses on a product-by-product, country-by-country, and quantity-by-quantity basis. WTO Member States are currently negotiating a permanent solution to the import/export problem and the bird-flu scenario provides a perfect example of how poorly the proposed solution works.  Instead of relying on complex, duplicative procedures that add expense and delay, countries should simply be able to access generic medicines via a single compulsory license that would authorize either domestic production or importation. If such a license were granted, the exporter should simply be able to supply the importer's need and pay a modest royalty to the patent holder.

Governments should not get tricked into thinking that Roche's sub-licenses and voluntary licenses will solve global need.  The production of antivirals like Tamiflu has to increase dramatically and the price has to fall precipitously so that the international community can prepare itself for the threat of a global bird-flu pandemic.  Unless governments and multilateral organizations act to increase quantity and reduce cost, poor people, once again, will be left out in the cold.