Tuesday, October 18, 2005


Once again, the relationship between patent rights and public health has made mainstream media –and this time, in the context of the notorious AVIAN FLU, which is spreading rapidly.

A recent news item states that Roche, which holds the patent over Tamiflu is unwilling to license it, despite the fact that it cannot produce enough quantities to cater to an outbreak, should one occur. It notes:

“Tamiflu, a pricey antiviral pill invented in a Bay Area lab and made in part from a spice used in Chinese cookery, has emerged as the world's first line of defense against bird flu should the deadly strain begin its feared spread among human beings

As nations begin to stockpile the drug in anticipation of a flu pandemic, calls are mounting for countries to sidestep patents on the drug -- as Brazil first did for AIDS medications -- and make their own generic versions

But Swiss pharmaceuticals giant Roche, which acquired rights to the drug from Gilead Sciences Inc. of Foster City in 1996, said Wednesday it had no intention of letting others make it

Roche ... fully intends to remain the sole manufacturer of Tamiflu,'' said company spokesman Terry Hurley

The immediate problem is not the cost of Tamiflu, which runs about $60 for a 10-pill course of treatment, but a staggering gap between the sudden demand for it and the capacity of its sole manufacturer to produce it.”

This is a rare case, where the price of a patented drug is not as much of an issue, as the fear of under-production and the ability to satisfy demand in a timely fashion. It therefore presents an easier case for the issuing of a compulsory licensing, a point that I will come to in a moment.

Enter, CIPLA, an Indian pharmaceutical company, which once again attempts to play Robinhood by threatening to go ahead and producing generic versions (Oseltamivir), despite Roche’s patents. A NY Times report states:

“A major Indian drug company announced yesterday that it would start making a generic version of Tamiflu, the anti-influenza drug that is in critically short supply in the face of a possible epidemic of avian flu."Right or wrong, we're going to commercialize and make oseltamivir," said Dr. Yusuf K. Hamied, chairman of Cipla of Bombay, using the drug's generic name and acknowledging that he might face a fight in the Indian courts with Roche, the Swiss pharmaceutical giant that holds the patent.”

Can CIPLA do so? On the one hand, there’s the technical issue of whether it can reverse-engineer TAMIFLU and manufacture quantities quick enough to cater to rising demands. Leaving this aside for a moment, lets focus on the issue of issue of legality¾an issue of far more relevance to this blog. Can CIPLA do this without violating patent laws? This is where we come to issues such as compulsory licensing.

First, lets take the position in India. I’m not sure that there’s any serious threat of legal action in India. The new regime patent regime that grants patents for pharmaceutical products came into force only on 1 January 2005 (with retrospective effect) and it will be a while before Roche’s application goes through the rigmarole of examination, opposition etc in order to be granted a patent. Till the patent is granted, (likely to take a year, at the very least), CIPLA is free to produce this drug. Of course, once the patent is granted, CIPLA can be injuncted. But will it be liable for damages for the ‘interim’’ infringement? Not so, under the new 2005 regime¾since the section that provides for retrospective damages does not apply to pharmaceutical ‘mailbox’ applications.

Lets assume for a moment that Indian law does prohibit from CIPLA from manufacturing this in India. What are its options?

1. The Indian Patent Act has one of the most comprehensive compulsory licensing regimes and the recent 2005 amendments have only expanded the scope of this regime. Assuming that Roche cannot adequately cater to the demands of the Indian market (if an when an outburst occurs), then CIPLA could invoke the following grounds:
a. Section 84 (1) (a) of the Patents Act, arguing that “the reasonable requirements of the public with respect to the invention have not been satisfied”.
b. If 60 dollars is not an affordable price for the common Indian, CIPLA could invoke section 84 (1) (b) arguing that “that the patented invention is not available to the public at a reasonably affordable price.”
c. I’m guessing that Tamiflu is not being manufactured in India (either directly by Roche or by any of its partners in India). If that is indeed the case, then CIPLA could also invoke the ground under 84 (1) (c) that “the patented invention is not worked in the territory of India.” However, it is not clear whether such a provision is TRIPS compliant.
d. The 2005 amendments to the Indian Patent Act provide an automatic compulsory license to generic companies, in so far as pharmaceutical mailbox applications are concerned. Thus for example, if Roche filed a mailbox application covering Tamiflu and if CIPLA produced a generic version of this prior to 2005, then it could ask for a compulsory license on a ‘reasonable royalty” basis. However, since CIPLA did not produce any generic version of TAMIFLU prior to 2005, it cannot invoke this ground.

2. A more creative option is for CIPLA to relocate its Tamiflu operations to Bangladesh (or tie up with companies there) and manufacture the generic version from there. It wouldn’t even have to pay any royalties in this case.

Being a Least Developed Country, Bangladesh has time till 2015 (under TRIPS) to implement product patents for pharmaceuticals. Contrary to popular notions of Bangladesh being so poor that it cannot support a domestic pharmaceutical industry, one would be surprised to learn that Bangladesh does have a robust industry.

CIPLA could even export the product from Bangladesh to India. You might ask: isn’t parallel importation a patent violation? (assuming that there is a patent in India). Not in this case, and certainly, not after the recent changes to India’s patent regime.

The earlier section 107A (b) provided that it was not an infringement to import a patented product, provided such import was from an exporter, who was ‘duly authorised by the patentee to sell or distribute the product’. The 2005 Act now makes such import easier by dispensing with the authorisation required from the patentee¾it mandates that the exporter of such patented product only be ‘duly authorised under the law to produce and sell or distribute the product’.

Under this amended provision, CIPLA could set up base in Bangladesh to manufacture and export medicines to India. Absent a patent in Bangladesh, and/or any other law barring manufacture/exports, such company would presumably be ‘duly authorised’ under the laws of Bangladesh to ‘sell or distribute the product’.

Therefore, currently, CIPLA faces negligible risk while manufacturing, selling or importing into India. But what of its exports to the US or EU markets? This is a little tricky, since a patent right in these jurisdictions means that Roche can prevent CIPLA from importing TAMIFLU versions. Here again, the only way out is through the grant of a compulsory license. This is precisely what a US senator and a CPTech letter propose.

Interestingly enough, the US does not have any compulsory licensing clauses within its patent regime that can be invoked by private parties directly¾rather this has to be worked into the ‘takings clause’ in 28 USC §1498 and perhaps the ‘eminent domain’ principle enshrined in the fifth amendment to the US constitution. Under these principles, the government could use the patent either directly or indirectly (through third parties) without the permission of the patentee, subject only to payment of some reasonable compensation.

The US threatened to do something similar with Bayer’s patent during the Anthrax crisis and I don’t see why it wouldn’t step in now, particularly since Roche is yet to don its humanitarian robes and commit to some serious licensing.

Unlike the US, most European member states, and the UK in particular, have compulsory licensing provisions within their patent regime. Section 48A of the UK Patent Act provides that a compulsory licence can be granted if “a demand in the UK for the patented product is not being met on reasonable terms”

As I mentioned earlier, apart from the legal issues, there is also the technical issue of whether CIPLA can do this in good time. While a WHO official and Roche sound sceptical about the ability to reverse engineer Tamiflu in a short time and come up with a generic substitute, CIPLA’s President, Hamid birms with confidence. I extract from the news item:

WHO flu chief Stohr is not optimistic that generic producers would be able to make Tamiflu. He told reporters in San Francisco that the drug takes a full year to make and involves a potentially explosive process that would drive out all but the most sophisticated manufacturers. It would take a generic supplier at least "two years" to put a plant into action.

Roche claims that “Making the drug involves 10 complex steps…. and the company believes that it will take another company two to three years, starting from scratch to produce it.”Dr. Hamied, Chairman of CIPLA however dismissed that claim, saying that he initially thought it would be too hard but that his scientists had finished reverse-engineering the drug in his laboratories two weeks ago. He said he could have small commercial quantities available as early as January.Asked if he thought Dr. Hamied was making an idle boast, Mr. Hurley declined to comment.”

I would wager that Dr Hamid is right here—and being an excellent chemist himself and not just an MBA laden proprietor, I’m sure he knows what he is talking about.

I’d advise Roche to hire a new PR division and rethink their strategy, as this is a very sensitive issue. With the flak that the pharmaceutical industry seems to receiving on an almost daily basis for placing profits over human lives, they should immediately commit to a widespread licensing scheme with low royalty rates. A good licensing strategy would at least ensure that they make some money (through licensing fees). If the government decides to take over their patent or worse still “revoke” it (as happened with Agracetus’ patent covering transgenic cotton in India), they would get nothing.

This is not to say to say that Roche is not doing anything to help. Indeed, today’s news item states that it is donating three million packs of Tamiflu to the WHO and a small quantity to Romania. However, what is perturbing is it’s intention to remove the sole producer.

"Roche and its partners fully intend to remain the only manufacturer of Tamiflu and are best qualified to scale up production," spokesman Daniel Piller said on Friday.

If it does indeed have the wherewithal to supply, this is not an issue. However, at this stage, one cannot state this for sure, particularly since Roche refuses to release production figures.

A guaranteed and timely supply of this drug is critical, particularly, since Tamiflu is thought to work best within 36 hours of symptoms. What is interesting is that Roche has stated that if need be, it would outsource production¾so perhaps Indian generics would gain either way. I would wager that with CIPLA’s heroic statements about breaking its patent, CIPLA would be way down in the list, when it comes to Roche chosing outsourcing partners!!

All the above is only to show that even in major healthcare crises of this sort, we still have to grapple with property rights. A recent article in nature states that “We have the means to make a vaccine against pandemic flu. But quarrels over money, science and politics mean it could come too late”. You can bet that patents make up for a large chunk of the “money” equation.

Also, I’m wondering that if we have so many issues with Roche’s existing patent, what are we going to do when the virus mutates or develops resistance?

"There are now resistant H5N1 strains appearing, and we can't totally rely on one drug (Tamiflu)," said William Chui, honorary associate professor with the department of pharmacology at the Queen Mary Hospital in Hong Kong.

We need to quickly think of innovative ways to decrease the incidence of patent blockages and of incentivising rapid research in this area. Tamiflu is only the first line of defence and is not a vaccine—and yet, its patents threaten to pose problems.

The actions of an Indian research institute are exemplary in this regard, where, IGIB, headed by Dr SK Brahmachari used an in-house “gene decipher” software to determine the functions of 3 genes of the SARS virus¾rather than applying for patents, they made these results available via their website, so as to speed up research on a potential SARS vaccine. This generosity was spurred in some part by the fact that the genome of the SARS virus itself was available in a public database. Do we see an open source model being built here? Whatever the case, we need more such commitments in times like this and not a strict insistence on property rights.
Post a Comment