World CAB II
This Time It's Generic
In January 2005, an unprecedented meeting was held in Mumbai, India, between four manufacturers of affordable generic antiretroviral medicines and 30 advocates for HIV treatment access drawn from every region of the world. This was the second meeting of a worldwide HIV community advisory board (CAB) with drug companies; the first was held in San Francisco in 2004 with several multinational pharmaceutical firms. The meetings were organized by iBase of London and GMHC.
The companies meeting in Mumbai included Cipla, the pioneer manufacturer of low-cost antiretroviral (ARV) drugs; Hetero, a large supplier of the bulk drug substance to many other generic drug makers; Strides, a small supplier of ARVs with historical ties to the African market; and Ranbaxy, an emerging powerhouse with ambitions of joining the ranks of the multinational, research-based pharmaceutical industry. The advocates traveled from Africa, South America, the Caribbean, Eastern Europe, Southeast Asia, as well as from India and Nepal, to press for lower prices and assurances of quality products.
Within the span of three years competition from the Indian generic drug industry has caused the price of ARVs to fall from over $8,000 per year, offered by the multinational originator companies, to under $200 per year, thus making large-scale plans for treating millions of people throughout the developing world feasible. But if lowered prices have brought plans to the table, the realities of actually providing therapy to so many have proven formidable and many problems remain. The meeting in Mumbai was intended to explore some of the problems that the generics industry can address, including the need for pediatric formulations of generic ARVs; the need for equally affordable second-line therapy for use when one's initial ARV regimen is not tolerable or has been compromised by resistance; and the need to address sometimes radical disparities in pricing that occur from country to country.
A key topic of discussion was the recent withdrawal of several important ARVs from a quality assurance list maintained by the World Health Organization (WHO) that many governments rely upon when purchasing large quantities of drugs. At issue was the performance of several clinical studies designed to show that the generic drugs are absorbed into the bloodstream as well as the brand name versions. Inspectors from WHO uncovered certain irregularities with the conduct and reporting of several of these bioequivalence studies, which called into question results from all such studies performed by the Indian drug makers. This was a delicate issue for the generic companies, and while no consistent explanation for the problems uncovered by WHO were obtained, the affected companies all pledged a rapid return to the critical list.
The future of low-cost generic ARVs also came under discussion. New patent rules going into effect in India may change the way the industry does business. Low-cost generic drugs have been possible because Indian patent law does not protect the final drug product, only the process of making the product. This means the generic makers have been free to copy expensive Western-developed drugs by coming up with a new manufacturing process. Under the new rules, patent protection will be granted to final products first known after 1995, which means that newer ARV drugs, like Viread and Reyataz may never become available at the kind of prices the generic makers are able to deliver. Products known before 1995 will not be affected. All of the implications of the new patent laws are not clear, but Dr. Hamied of Cipla, the only of the four companies serving the private ARV market within the country, thinks the outcome will hurt India by introducing monopoly patents, "It's going to be a disastrous situation for India ten years down the line."
Representatives from India pressed the manufacturers on one of the most vexing contradictions they face, "Why are drugs made in India cheaper in Africa than at home?" Each company added its piece to the puzzle, variously blaming import duties on raw materials, taxes and an indifferent Indian government. But all agreed that the key barrier to achieving lower prices, whether in India or Nigeria, is the lack of a consistent, growing commercial demand for their production. Despite widely announced plans for scaling-up treatment to reach three million people by the end of 2005, last year, one maker said, only 40,000 new patients were added to the treatment rolls. Yet with Cipla now claiming an average price of $160 per year, it may be that drug cost is no longer the limiting factor. Until the barriers of inadequate healthcare infrastructure, training and finance are removed, many fear that attainment of the kind of large-scale purchases than can bring ARV costs below $100 per year will remain unlikely.
The following are condensed excerpts from the discussions held in Mumbai.
The Price of Drugs in Delhi
Loon: I'm proud India is making these drugs. I am a consumer of these products. But I am paying $280 per year while people in other countries are paying $180 per year. I'm pissed off about this.
Strides: The only way we can get costs down is to get volume and to manufacture the API (Active Pharmaceutical Ingredient), which we are pursuing. The government should remove sales tax and statutory levies that increase the consumer price. Wholesale and retail margins are about 20 percent, but 36 percent is going to agencies of the government. If you supply to the government of India, sales tax is exempted. But if you supply to an NGO, then you are charged 16 percent. The government has to be lobbied. It took us years to remove the taxes for TB products. We need to do that for ARVs.
Loon: Can the price for the stavudine/lamivudine/nevirapine combo be brought down?
Strides: I don't think you will find that the price of Strides, Ranbaxy, and Cipla will be very different from one another. Certain costs are fixed: the API, the manufacturing costs. We have all agreed to keep profits down. Just because it is a generic doesn't mean the manufacturing process is compromised. It is a Rolls-Royce product. You can only hope to reduce price if the volume becomes larger. Only 40,000 patients for ARVs were added last year. The economies of scale have not yet been demonstrated.
Loon: Could you have non-profit prices in India for people who pay out of pocket?
Strides: It is very difficult to make small sales. It would be good if all the communities would get together and have a common basket for procurement. It then becomes much easier for us to work out a better price. If you have a buyer's club, then we can do that.
Thomas: Will you sell to buyer's clubs? What will be the process?
Strides: It's all about numbers. If someone asks for 50 bottles, I may or may not be able to offer it to them. You have different packaging specifications in different countries. In Nigeria it must have a code on it. What I produce in Nigeria, in English and French, I can't sell to Costa Rica. When you register, you include the label in the language and you can't deviate from that. Each country's label is different. You also have to have a minimum batch size. If you all get together, you can get the price advantage and better service, too. If 500 individual buyers get together the freight is the same as for ten.
Marie Mendene: Have you had any discussions about regional procurement?
Strides: Many people talk about this proposal. It would make our life simpler.
The Clinton Price
Marie de Cenival: What is the quantity and the price that was offered to the Indian government under the Clinton Foundation agreement?
Ranbaxy: The Clinton prices are theoretical prices based on volumes that have not materialized. They said they would reach 100,000 patients after 2004, then 200,000 after 2005. We said our price is based on 40,000 patients. This is where we could enter into contracts with our suppliers. But the volumes haven't been realized yet.
Marie: Which countries have really gotten the Clinton price?
Ranbaxy: Very few. In Rwanda the price went from our NGO price to the Clinton price but there was no increase in volume. Only the distributor lost out because his commission shrank.
Cipla: If you want to treat a million patients, we can manufacture 2 million tablets a day. But where will you get the API? The bottleneck in the large scale supply of ARVs in is the manufacture of active ingredients. You need 30 tons of stavudine; 146 tons of nevirapine per year. If you change to efavirenz, you need 220 tons annually. Nelfinavir, you need 900 tons annually for a million patients. Who is there to finance these quantities? One hundred companies like Cipla can not cope with this situation. The problem is the manufacture of the actives.
Gopa: It was a year ago that we heard of the Clinton price at $140 per year. What assumptions went into that pricing? Are those assumptions still in effect?
Cipla: We were told that seven countries were in partnership with Clinton for supplies, and that the total demand would be 2 million patients in the next two years. On that basis we arrived at a price of $140 per patient per year for Triomune. But that was a conditional price. They announced that price but not the conditions.
The conditions were: the price of the APIs were fixed; no payment of royalties or licensing fees; no litigation over patent infringements; the specification of the product was fixed; any variations were higher. Supplies were point to point; no intermediates. These were subject to large confirmed irrevocable orders. Payment terms were to be agreed in advance: advance payment or against supply. Then the fluctuation in currency would be considered. If the dollar went down, the price would increase. The dollar is already down 25 percent. It was without shipping, for only three products in only seven countries.
So they are nice people at the Clinton Foundation, but I don't know how effective as an NGO they are, because no business has developed from them.
Marie de Cenival: So the market is not what we thought. I'm surprised that you believed what Clinton told you.
Cipla: He is a very good talker.
Planning for the Future
Marie. The Clinton deal is not working. Can we talk about what will work?
Cipla: First, who will invest in the API? I'd suggest the Global Fund and World Bank should invest in API. The South African government has given a loan of $50 million to a company to make APIs. I think it is a public mission to fund this.
Marie: What is the best price you can give?
Cipla: Why do you only talk about pricing? There are other issues. What is the cost of medication? The medicine is only a small part of the costs of treating the people.
Loon: Will we ever get below $140 in India?
Cipla: India is in a unique position -- we have a large industry, over 20,000 pharmaceutical companies -- but there is also a public drug sector. I suggest that the public sector takes over ARVs. We have reduced the price dramatically in Triomune over the last four years. If the duties go away, then the price will be lower.
Estella: What are the opportunities for reducing the price of APIs?
Cipla: Twelve to fourteen APIs are being used extensively for ARVs. It is not humanly possible for any one company to produce all of them. Understand the scale of operation: stavudine requires 30 tons but nelfinavir requires 900 tons. If I make Triomune for 1 million patients I need 110 tons of lamivudine. Last year GSK consumed 26 tons of lamivudine, and their sales were $135 million.
We import and make APIs and intermediates. We sell very few APIs. From India, the countries buying APIs are Brazil and maybe Thailand. The problem will come when WHO begins to inspect APIs; that will be a major problem. Before they said they only approve the end product. The responsibility for the API was ours. Now they are going to inspect the API factories. If we have to use FDA approved factories, it will be difficult.
Marie de Cenival: I take your suggestion is to get the World Bank to fund API production.
Cipla: What you should do is have a consortium that sets up a manufacturing base. Choose a least developed country with no patents until 2016. Bangladesh, Ceylon, Mauritius. If governments do not actively participate, then little can be done. You need a long-term partnership because it is a life-time illness. Why doesn't the Indian government use its factories? Let me ask you an important question. What contingency plans do you have to assure that you have medications 20 years down the line? One of our main partners in producing ARVs is China. If China decides to stop supplying chemicals for ARVs then we are all out of business. The China patent rule is that they cannot sell the patented drug in China but they can export it to countries where it is legal to sell, like India. That could change. Today the China currency is linked to the dollar. Tomorrow, if the currency is delinked and starts floating, the prices of all ARV drugs will raise very high. What happens if the companies stop selling at cost? If they leave the field? If China stops selling APIs? What contingency plans do you have?
Asia: We want to talk about the crisis over WHO prequalification. The entire Ranbaxy ARV portfolio has been withdrawn from the WHO pre-qualified list. This has caused disruptions to treatment programs. Nigeria is deregistering all Ranbaxy products. We saw that a new CRO (Contract Research Organization) was contracted for PEPFAR purposes. This seems to establish two tiers of quality. Can you comment on the apparent double standard?
Ranbaxy: Double standard? We didn't realize our delisting would be interpreted that way. PEPFAR was knocking at the generics' door -- due to your pressure. We have good relations with the U.S. FDA. We understand that process very well. They have approved 100 of our products. We were in the middle of assessing data for the FDA and we wanted to harmonize everything to the U.S. market and consolidate manufacturing in one place. If a product is approved by the FDA then it is accepted anywhere. It's not a double standard, but exactly the same product made at the same locations.
Gregg: You need to get these drugs back on the WHO list and not wait for FDA. You are creating havoc until they are re-listed.
Ranbaxy: The two are not linked. The U.S. asks for less stability data than the WHO does, so filing for WHO prequalification could actually happen later. WHO requires certain additional data to be generated which is being generated. There are bioequivalence studies running today and you cannot be 100% sure when the data will come out.
Marie de Cenival: What really happened?
Ranbaxy: Those three products were supported by bioequivalence data from a particular CRO in India, which had been used by several different companies, pharmaceutical and others, doing various kinds of analyses, including bioequivalence studies. It was a very reputable, publicly listed company. The CRO had some issues with GCP (Good Clinical Practice) and GLP (Good Laboratory Practice) and consequently those studies were not passed by WHO. And WHO de-listed the products.
Marie de Cenival: What issues? If we know what happened, then maybe we can calm down the concern that you are selling bad products. Can you be more specific?
Ranbaxy: What I can say is there were issues with noncompliance and handling of data. This issue has caused everyone a lot of trouble. It has been a breach of faith and a breach of contract and once everything settles down we might need to take legal action against the CRO. Anything I say now would be a breach of confidentiality.
Strides: We are happy with our CRO -- it is approved by Brazil, inspected by the FDA and WHO. Monitoring of the bioequivalence protocol is the responsibility of the drug company, not the CRO. No CRO objects to having a company-appointed monitor in place. There are absolutely no issues with our bioequivalence studies.
Let me explain the process of developing a product. We do preformulation studies; we study the compatibility with excipients, do dissolution tests, etc. We compare with the innovator product. There is no pharmacopoeia, so there are no reference standards other than by comparison with the innovator. Twelve months of stability studies are required by WHO. From the day you identify a product to having approval is 18 months to 2 years. We are underway with efavirenz and will be ready by the end of the year.
We have sent stavudine/lamivudine to the FDA. The problem with FDA has been their inability to accept bioequivalence studies that were carried out using European innovator products. They don't have the right of reference. This means every bio study has to be repeated with U.S. approved innovator product. We just completed a bioequivalence study with one.
Marie Mendene: There was news in Cameroon that the drugs were of poor quality. We were embarrassed. What are you doing to address these concerns in our communities?
Hetero: We have traveled to several countries and are educating people. We met with the Ministry of Health in Cameroon. We want to eliminate any ambiguity. We could not survive in this business if they weren't good quality.
Marie Mendene: We are glad to hear that you have resubmitted your products to the WHO. But we are concerned about the other products that you have not yet submitted.
Cipla: Are you aware that WHO hires out inspectors for prequalification? We have been disillusioned by the type of inspectors coming to inspect. There are two types of qualification: the facilities and the products. When we give work to an outside agency, then we need to have them independently certified. We asked WHO, why not certify Contract Research Organizations (CROs) too?
Also, WHO does not qualify the active ingredient going into the end product. So any prequalification is meaningless unless they also qualify the APIs. Now they are going to start qualifying the APIs and the cost will double. We will have to produce every API according to FDA good manufacturing practice (GMP).
Cipla: After India's independence in 1947, the multinational companies were very strong in India and we were following the British Patent Act of 1911. We started fighting to change the patent law and we succeeded in 1972. We decided you can only patent a process, not a product, and then only for 7 years for health and food products. That gave us the legal freedom to make whatever we wanted. Now Indian companies now control 80% of Indian market. Unfortunately, as of January 1, 2005, the situation has changed concerning products invented after 1995. Whatever was known prior to 1995, we can still produce.
Gopa: Have you assessed the impact of the new patent regime on ARVs?
Ranbaxy: Most individual ARV drugs are pre-1995. Only patent applications filed after January 1995 are affected by the new law. The drugs in our portfolio are okay except for tenofovir and abacavir. But certain combinations such as lopinavir/ritonavir might be restricted. Atazanavir will be a problem. If there is a need for efavirenz in a triple combo, there could be a problem, but most products in the portfolio will not be affected. We have to take it as it comes.
German: When world patents come to India that protects products for 20 years, what will this do to the generics industry?
Strides: Well, you have to find ways to manufacture that are non-infringing. The cost will be higher, but you will have higher returns because there will be less competition. But the preferred way will probably be to get into partnership with voluntary licenses. We will live with it. This is going to happen and there's nothing we can do about it. This is happening the world over.
Asia: You say that partnership with the multinationals is preferred but sometimes the terms of licensing will not be acceptable if they don't serve the interests of patients.
Strides: Some multinationals are extremely good and others not so. But getting into litigation is not the way to go. I think non-confrontation way is the way to go. I'm not rejecting compulsory licensing, but I have no comment now. I think we should give voluntary licensing another shot.
Gopa: The other companies said they are going to live with the new patent regime. What will you do?
Cipla: I think amendment of the patents is a big mistake on the part of the government. We require a permanent compulsory licensing system for the developing world. We are willing to pay a 4% royalty, but I cannot allow a monopoly in a country the size of India. It is going to be a disastrous situation for India 10 years down the line. I would prefer an automatic patent: a 5% royalty on net sales is equivalent to a 25% ownership position.
There should be no monopoly. Evergreening of patents is the most dangerous thing. AZT was invented in 1963. It was claimed in 1985 as an AIDS drug with a patent until 2005. GSK has said AZT should only be marketed in combination, with a patent that goes to 2017. That's 54 years of monopoly for AZT. We should oppose this.
Drugs for Children
Elena: In Romania there are many AIDS cases in children; I am the mother of a child with HIV; there are no generics, although the government is obligated to pay for ARVs. I want to know what is planned for pediatric formulations in Romania and the rest of the world?
Ranbaxy: Romania follows the European Union, which requires different kinds of data for registration that we haven't done yet, but we will.
On pediatrics, I want to have your suggestions on what kinds of products we should develop. The costs of pediatric drugs are higher. For example, carrying liquids in glass across the globe raises the cost. We have dedicated resources in R&D that we can put to work on developing pediatric formulations.
Polly: What pediatric formulations do you have so far?
Ranbaxy: We have conventional zidovudine, and are working on a stavudine. These are available in India and produced in Nigeria for local use. Volumes are very low and economy of scale will be slow.
Olive: Are you working on a pediatric triple combination?
Ranbaxy: The ratio of the drugs in the combination changes with the child's weight. We can't get people to agree on the needed doses, so we decided on a product which will exclude some age groups. It will apply to a weight of about 20-30kg, but not above and not below. Some pediatricians agree with this and some don't. If we develop something that is not necessary we will lose a lot of time and resources. Another challenge is to create a formulation that is not a liquid. But then access to clean water becomes an issue.
Simon: You know that in practice, children are being treated by crushing adult formulations. This is due to cost and also lack of availability.
Ranbaxy: That's off-label use. To get it approved, you have to clinically prove it, and that is where the challenge is.
Polly: What are your plans for developing pediatric formulations?
Strides: We made a proposal to WHO a year and a half ago for a triple drug pediatric formulation. WHO never got back to us. We need people to tell us if we are doing the right thing, making the right doses; but not much has happened. We can develop granules in sachet, which UNICEF was excited about. Clean water is still an issue. Temperature and humidity are issues. The granules can't clump together.
What's in the Pipeline?
Svilen: What new medications do you have in the pipeline?
Hetero: The latest is lopinavir/ritonavir and tenofovir.
Simon: Atazanavir is listed in your book. What about the legal problems?
Hetero: The APIs for atazanavir and tenofovir are absolutely ready today, but we have to see what will happen with the patent situation. We can do anything we want in R&D, but for commercialization, patents are an issue.
Gregg: What else is in the pipeline? We need tenofovir, and a better formulation of ddI.
Strides: On tenofovir, it depends on the patent mailbox. We don't want to market these then get into litigation. We want work with the innovator companies. The best thing is if they give voluntary licenses, then we will make them. They need to be more considerate with giving voluntary licenses. There is an advantage for them, too. The cost of making tenofovir in India would be much cheaper. It works both ways.
Svilen: What about older products like abacavir and nelfinavir?
Strides: Nelfinavir is in the pipeline. WHO has said not to pay much attention to abacavir. Also there is less API available. The cost is prohibitive because the raw materials must be from an approved source. The Chinese are good at making materials at a low cost but their standards are not accepted. They don't have DMFs (Drug Master Files) that are acceptable to WHO or the FDA. So we depend on a few Indian API makers.
With our soft gel capabilities, we have ritonavir and lopinavir/ritonavir and saquinavir. Two are prequalified and five more are under evaluation. We expect prequalification in the next three months. Bioequivalence studies are completed.
Loon: What are your new products?
Ranbaxy: We are close to introducing lopinavir/ritonavir and nelfinavir. But these won't be triple fixed dose combinations (FDCs) because there are problems with some combinations. Second-line is based on PIs, which have high doses, and frequent doses.
Gregg: The price of generic efavirenz needs to match the price for nevirapine. Why is there such a great difference?
Ranbaxy: Prices of originator efavirenz were lower than the generic because our price could not match the innovator's cost. If we start getting volume we can improve the price. We sold at a loss for a while to match the innovator until we could lower our costs. But not much was purchased at 95 cents. If you look at the volume of efavirenz supplied to Africa, it is very small at this time.
Simon: What about abacavir, zidovudine, 3TC? The market is for use in TB patients.
Ranbaxy: Abacavir is not widely used because doctors in India don't like to use it. A small number use it but it is growing slowly. We found that in Latin America governments are asking for abacavir.
Gregg: Will the patents for tenofovir and FTC stop you from making them?
Cipla: FTC was known before 1995; tenofovir is 1997. I think we can make it and see how it works out. I'm a firm believer in the automatic patent of right. You don't have to ask, you just pay a fair royalty. Until they joined NAFTA, Canadians were able to copy any drug they wanted and pay a 2 percent royalty.
I think it is very important to work out what drugs you want for the future. What are your criteria? T-20 costs $20,000 per year. I can make T-20 for $5,000 a year. Are you interested in T-20? You have to look at efficacy and affordability side by side. We go by what is the easiest to produce, but that may not be what you need.
Copies of the complete World CAB II Report can be obtained at