the conjugation process, however, offer the prospect of a more affordable conjugate and should be pursued.¹⁰

There are a number of specific scientific issues that must be considered as conjugate vaccines continue along the development pipeline. For instance, researchers need to determine how many pneumococcal polysaccharide antigens can be incorporated into a single vaccine using the same carrier proteins. They would also like to know what sorts of immunization schedules and routes of immunization will be most practical. Another important question is whether more than one type of protein carrier can be combined into one vaccine for several diseases (for example, Hib glycoconjugate plus S. pneumoniae).

In the long term, however, it is the complexity of making pneumococcal conjugates that seems to pose the greatest challenge in the developing-world context. A number of developing nations are now able to make DTP reliably, and some are working collaboratively with developed-world manufacturers to combine this trivalent “platform” vaccine with newer products, such as hepatitis B. However, the DTP vaccine is in flux: a new acellular pertussis component will soon be substituted for the current whole-cell version, and the implications of this change for the D and T components are unclear. As the number of available vaccines increases and the costs of newer formulations rise, donor agencies are less able to provide vaccine. This creates pressure for developing nations to build domestic manufacturing capacity. The movement toward vaccine self-sufficiency (whether through direct purchase of vaccine or indigenious manufacturer) is proceeding despite significant quality control and regulatory concerns. (See page 29.)

Developing countries may be reluctant to incorporate pneumococcal conjugates into their immunization programs, given the limited serotype coverage¹¹ of candidate vaccines. The median coverage of two proposed seven-valent pneumococcal vaccines (the Merck and Lederle-Praxis formulations) is below 70 percent in the group of developing-world countries for which data are available.¹² Given the mortality and morbidity caused by pneumococcal infection, however, even an incomplete vaccine may have a substantial impact on adverse outcomes. Coverage could be increased through the addition of more serotypes, but this would probably add substantially to the cost of the final product. Each additional serotype adds a diminishing amount of coverage, since rarer serotypes cause a small proportion of overall disease. Vaccine cost might