Financing and Institutional Arrangements for New Product Development
Research that contributes to the discovery and development of drugs, vaccines, and diagnostics occurs in public, private, and mixed settings, each with different funding mechanisms.
In most high-income countries (HICs), government funding from tax revenues is generally targeted to basic research—that is, research that advances understanding of underlying disease processes but is unlikely to yield commercially viable products in the near term. The research may be done in government institutions or in academic and other not-for-profit research institutions. Governments also stimulate private sector R&D through tax credits.
Private For-Profit Sector
Applied research that targets specific products is generally undertaken by the private sector using equity financing. Firms that rely on equity financing must provide a return to their investors comparable to returns on other potential investments, hereafter referred to as a competitive return. This requirement applies for multinational pharmaceutical companies, for biotechnology firms, and for firms in developing countries, unless they receive public subsidies. Start-up firms in HICs generally rely on equity capital from venture capitalists and other private investors, whereas established firms issue shares in the broadly based public equity markets but finance most of their R&D from retained earnings on existing products. The need to provide a competitive return to shareholders means that commercial firms can invest only in products that they expect will generate sufficient revenues to cover all costs, including the costs of R&D. In practice, commercial firms have focused on products with a potential market in industrial countries because of their residents' ability to pay prices sufficient to cover costs.
Differential, or "Tiered", Pricing
For global products—that is, products targeting diseases that occur in all countries, such as cardiovascular diseases—revenues generated in HICs and in the more affluent sectors of middle-income countries are sufficient to recoup the investment in R&D to the extent that, ideally, prices in LICs need to cover only the incremental or marginal costs of production for these countries.
Even with pricing at marginal cost, medicines may still be unaffordable for the poorest populations, particularly for drugs with high manufacturing costs, in which case additional subsidies may be necessary. However, the important conclusion is that for drugs for global diseases, the existence of a market in industrial countries attracts private sector investment in R&D; thus, differential (tiered) pricing provides a finance mechanism for developing new drugs that can achieve both dynamic efficiency (appropriate incentives for R&D) and static efficiency (appropriate incentives for use of existing products) (Danzon and Towse 2005).
By contrast, for drugs and vaccines that target diseases that occur predominantly in LICs, no HIC market exists in which to recoup the costs of R&D, and patents and differential pricing will not suffice to attract R&D for products that cannot expect to generate sufficient revenue to cover their development costs. In 2002, annual per capita spending on drugs alone in member states of the Organisation for Economic Co-operation and Development was US$279, while developing countries typically spent less than US$20 per capita for all health services (Sachs 2001; Troullier and others 2002). Per capita health spending on drugs by the poorest individuals, who may be the majority of patients for communicable diseases, is even lower. Thus, for products that target LIC-only diseases, even if millions of patients are in need, expected revenues are insufficient to attract private sector investment for developing new products without additional public subsidies.
For HIV/AIDS, even though the majority of the disease burden is in LICs, the markets in HICs have been sufficient to attract private sector companies to develop several drugs and to undertake considerable investment in an AIDS vaccine, albeit with little success to date. In 2001, the GATB increased the estimated size of the TB market from US$150 million to US$450 million per year, with the potential to grow to US$700 million per year (GATB 2001). This amount is within the range normally considered necessary to attract private investment. However, estimated potential revenues for anti-malarials and treatments for other LIC diseases are still well below this threshold. In addition to the limited ability to pay, some developing countries still lack the health care infrastructure necessary for conducting clinical trials and for delivering medicines and vaccines effectively, which further reduces incentives for R&D investment.
Given the low potential revenues and lack of necessary infrastructure, R&D for tropical diseases and TB for the past 25 years has been far less, relative to need, than for global diseases. The number of NCEs per million disability-adjusted life years lost (a proxy for research relative to need) was 0.55 for infectious and parasitic diseases but roughly 1.25 to 1.44 for cardiovascular system diseases (Troullier and others 2002). Between 1975 and 1999, just 16 of the 1,393 NCEs registered were for tropical diseases or TB (Troullier and others 2002).5 Several of these products were fortuitous by-products of commercial research efforts initially intended for the oncology or veterinary market (Ridley 2003).
Multinational companies appear to be showing some signs of increasing their investment in tropical disease R&D. For example, GlaxoSmithKline, AstraZeneca, and Novartis have recently announced or established research centers devoted to tropical disease. AstraZeneca's facility in Bangalore, India, will focus on TB treatments and receive a commitment of personnel and US$40 million in investment during 2003-8. The nonprofit Novartis Institute for Tropical Diseases in Singapore is a US$122 million joint venture between Singapore and Novartis that will focus on dengue fever and TB. GlaxoSmithKline has established a research institute for TB and malaria in Spain ("Drugs for the Poor" 2003).
Orphan Drug Acts
Orphan drug acts provide additional stimulus for private sector R&D for diseases that afflict only small populations in HICs. The U.S. Orphan Drug Act grants orphan status to drugs to treat diseases that affect 200,000 or fewer patients per year in the United States. Orphan drug status provides additional R&D tax credits and seven years of market exclusivity, during which the FDA cannot approve another drug to treat the same condition unless it uses a novel mechanism of action. Such market exclusivity enhances the orphan drug's market power, enabling the developer to charge high prices that to some extent offset the low sales volumes, thereby covering the costs of R&D. The U.S. act has stimulated a sharp increase in the number of drugs developed to treat orphan conditions since its passage. The European Union recently adopted similar legislation.
The potential for orphan status in the United States and the European Union may provide some additional stimulus for commercial firms to develop drugs and vaccines for LIC diseases, but the effects are likely to be minor for several reasons. First, after one product has acquired market exclusivity, firms have few incentives to develop other products to treat the same disease. Second, the value of orphan drug status in terms of annual revenue per patient is greatest for drugs to treat chronic diseases that require daily or weekly treatment. Potential revenues for treatments for acute diseases, for which each patient needs only a short course of treatment, are likely to be smaller. Thus, though orphan drug acts may create some additional stimulus for R&D for LIC diseases, other institutional and financing mechanisms are essential. Of these, PPPs are the most promising.
"Pull" Financing Mechanisms
Since the late 1990s, organizations such as the Bill & Melinda Gates Foundation and the Rockefeller Foundation have increased their funding commitments to fight diseases in developing countries. This new funding, including funding coordinated through the Global Fund to Fight AIDS, Tuberculosis, and Malaria, is allocated primarily to paying for vaccinations and treatment. By paying for vaccines and drugs, such financing could provide additional revenues to suppliers of these products and, hence, stimulate R&D. However, for the financing of vaccines and therapeutics to serve as an effective pull mechanism for future R&D, such financing must be sustained and must pay originators enough that they can recoup the costs of R&D. Thus, purchasers such as the United Nations Children's Fund or the Global Fund face a tradeoff between paying the lowest possible prices so as to maximize their ability to supply existing medicines to current patients and paying somewhat higher prices so as to create incentives for future R&D.
Creating effective pull financing incentives for R&D is probably best done by means of explicit purchasing commitments for specific products. Some progress has been made in identifying the contractual and legal requirements of such commitments to enter into future contracts. The most promising candidates for initial implementation would be products or vaccines that are already in late stages of development or have been approved for industrial countries but for which additional purchasing commitments are needed to induce the investment necessary to undertake clinical trials and build the manufacturing capacity required to extend these products to LICs. Possible candidates are the pneumococcal vaccine and the rotavirus vaccine. For both these products, accelerated development and introduction plans have been created in the Global Alliance for Vaccines and Immunization to address the many practical issues surrounding the implementation of an advance purchase contract. When advance purchasing commitments have been successfully demonstrated on products in the late stages of development, extending this promising approach to products at earlier stages of development may be possible.
In recent years, a growing number of initiatives involving partnerships between industry and government, nonprofit, and philanthropic organizations have been set up to stimulate tropical disease R&D. One of the oldest is the World Health Organization, World Bank, and United Nations Development Programme TDR, which has worked with industry, academia, and research institutions to spur R&D and has contributed to half the new drugs developed for neglected tropical diseases during the past 25 years (Ridley 2003; Troullier and others 2002) (see table 6.2 for examples of the program's initiatives). The TDR is a relatively small program, with contributions of US$30 million in 2002.
Since the late 1990s, increased government and foundation funding, particularly from the Rockefeller Foundation and the Bill & Melinda Gates Foundation, has stimulated the growth of product development PPPs, giving a "push" stimulus to R&D (Varmus and others 2004). According to the Initiative on Public-Private Partnerships for Health, about 20 PPPs were involved in product development as of 2004. Although a few focus on a specific project, most adopt a portfolio approach with multiple candidates. The latter include five targeting HIV/AIDS vaccines or microbicides; three working with malaria therapeutics or vaccines; three investigating TB therapeutics, vaccines, or diagnostics; and at least six targeting drugs for other neglected diseases (Widdus 2004).
The PPPs are heterogeneous in terms of their objectives, structure, and financing. In general, their goal is to develop products for use in developing countries with a public health rather than a commercial goal. Their sources for promising compounds include modifications of existing compounds; continued development of compounds previously abandoned because of a lack of commercial potential; and totally new initiatives coming out of academia, industry, or government laboratories. If a PPP acquires a product from another firm, the other firm typically retains patent rights in HICs and middle-income countries, and the PPP commits to noncommercial pricing in developing countries.
PPPs draw on financing from foundations and, to a lesser extent, from governments. They work closely with private industry, including large pharmaceutical and biotechnology firms, obtaining a range of in-kind contributions, including promising compounds; useful technologies; patent rights; and expertise and advice on discovery, clinical trials, manufacturing, market estimation, regulatory requirements, and so on. They operate largely as "virtual" firms, usually contracting out actual operations to other firms or to contract research or service organizations. As compounds move into human trials, PPPs must also liaise closely with disease-endemic countries regarding clinical trials, regulatory requirements, and product delivery. Thus, they face significant scientific, managerial, financing, and operational challenges.
Table 6.3 lists the leading product development PPPs and their committed funding as of early 2004. Several have received grants of US$50 million or less, with significantly larger amounts for the International AIDS Vaccine Initiative and the Malaria Vaccine Initiative. Several of the organizations rely heavily on the Bill & Melinda Gates Foundation and the Rockefeller Foundation for both their initial and continued funding (Widdus 2004). Note that the dollar funding amounts shown exclude in-kind contributions from industry and other sources, whose worth is difficult to calculate because the value to the PPPs is presumably greater than the cost to the donor.
Table 6.4 shows the product development PPPs' portfolios of products as of early 2004. The percentage of products still in preclinical development is higher for vaccines than for drugs, which may reflect the scientific challenges of developing vaccines for LIC diseases. In general, comparing funding amounts with the number of products in development across PPPs is inappropriate as an indicator of performance because the different PPPs target different problems and have received varying in-kind contributions. Also, some products are modest extensions of existing therapies, whereas others are more innovative and, hence, more risky approaches.
As table 6.3 shows, aggregate committed funding for the product development PPPs as of early 2004 was US$1.2 billion, excluding in-kind contributions. A comparison of these funding amounts to the costs per NCE suggests that current rates of investment will produce some progress, but not rapid advances. Assuming optimistically that future funding for PPPs will be US$300 million per year, that private industry will invests similar amount, and that other sources will provide US$100 million a year (all of which are probably generous estimates) would imply total investment of US$700 million per year. If this level of investment were sustained over time, it might result in two or three NCEs per year, using the conservative cost estimates of US$200 million to US$300 million per NCE. This development level would be significant progress, although it still leaves a large shortfall, given the number of diseases for which no good treatment or vaccine is available and the threat of resistance developing to existing treatments. It is also far short of the Commission on Macroeconomics and Health's target for 2006 of US$3 billion in R&D spending for diseases in developing countries (Sachs 2001).
Industry in Developing Countries
Pharmaceutical firms in developing countries have traditionally focused on the generic sector, making use of their expertise in engineering and other skills needed for efficient drug manufacturing. More recently, the adoption of product patents has created incentives for LIC firms to invest in R&D. For example, India adopted product patents as of 2005, and several leading generic firms are already developing new products. However, assuming that these firms will focus their efforts on developing drugs for tropical diseases would be a mistake. As for-profit firms, they face similar incentives to those of commercial firms in any country, which means focusing on the global diseases that offer the greatest expected net revenues rather than diseases specific to LICs. Nonetheless, several policies might help target R&D efforts in these countries toward tropical diseases. These policies include collaboration with the product development PPPs, provision of special government funding or tax credits for products that target LIC diseases, and provision of subsidies targeting the development and scientific testing of products derived from local products and other traditional medicines.
Other Proposed Mechanisms for Increasing Affordability
In evaluating other proposals for making drugs or vaccines available in developing countries, distinguishing proposals to stimulate new product development from proposals to increase the affordability of existing drugs is critical. One proposal pertaining to affordability is that multinational companies should voluntarily license production rights to LIC producers. Experience with generic markets across countries indicates that necessary conditions for such out-licensing to reduce prices to consumers are (a) the existence of competition between multiple licensees, (b) the licensees having lower production costs than the originator firms, and (c) a mechanism that prevents middlemen and retailers from capturing any potential savings. In practice, these conditions may not be met. The more probable scenario of licensing to only one local generic manufacturer is unlikely to reduce prices to consumers.
Another proposal is that governments should purchase patent rights, paying the originator firm the estimated value of the drug (net of production costs) and then selling the product to consumers at the marginal cost of production. This proposal has several disadvantages. First, because the government would presumably have to raise taxes to pay for the patents, the tax-induced efficiency loss could offset any efficiency gain in the pharmaceutical market, so the net effect on efficiency is unclear. Second, the presumption that patents result in suboptimal drug consumption because of monopoly pricing ignores the widespread prevalence of insurance in HICs and middle-income countries, so that, in practice, consumers face out-of-pocket prices that are already close to marginal cost. Third, and most important, is the difficulty of estimating the value of a product before its use in the market, because both positive features (additional uses) and negative features (side effects) may be discovered. In addition, distortions in the amounts paid for patent rights would distort incentives for R&D. Moreover, the proposal would reduce originator firms' incentives to invest in post launch improvements.