menting (as opposed to replacing) natural immunity in residents of malaria-endemic areas. For example, among semi-immune recipients, a blood-stage vaccine that lowered parasite density might prove beneficial even if it had little effect on the incidence of infection. Similarly, a partially effective preerythrocytic vaccine might protect against death from malaria just as ITNs do. Whether or not an imperfect malaria vaccine justifies use in a particular setting will depend upon its efficacy relative to other available control tools, its acceptability, and its cost.

Pregnant women in malaria-endemic areas, especially primigravidae, have a special need for enhanced malaria protection because of malaria’s damaging effects in pregnancy. In some areas, it is becoming difficult to provide this protection through any other means (e.g., chemoprophylaxis, or intermittent treatment). A vaccine that prevented sequestration of P. falciparum parasites in the placenta might lessen the chance of a pregnant woman giving birth to a low-birth weight baby, but would not protect her from anemia.

For various reasons, including malaria’s antigenic variation and protein polymorphisms, it is unlikely that any single malaria antigen will be found that meets all of the criteria for a perfect vaccine. Thus, the most effective malaria vaccines are likely to contain a “cocktail” of antigens from the same stage of the parasite’s life cycle, or different antigens from different stages. The most extensively tested combination vaccine thus far is SPf 66, which includes several erythrocytic-stage antigens originally found protective in Colombia (Pattarroyo and Armador, 1999). However, trials using SPf66 in holoendemic areas of Africa showed marginal, if any, protection (Alonso et al., 1994; D’Alessandro et al., 1995), nor did the vaccine confer protection upon Tanzanian infants when administered as part of their initial EPI vaccine package (Acosta et al., 1999). The vaccine also failed to protect Karen children in a malarious region of northwestern Thailand (Nosten et al., 1996).

The first multistage, multiantigen P. falciparum vaccine candidate was NYVAC-Pf7, an attenuated vaccinia virus genetically engineered to include seven P. falciparum genes encoding the pre-erythrocytic antigens CSP and LSA1; the asexual blood stage antigens MSP1, AMA1 and SERA; and the transmission-blocking antigen PFs25 (Tine et al., 1996). Initial studies did not demonstrate protection against P. falciparum in human volunteers, although detectable immune responses were elicited (Ockenhouse et al., 1998). MuStDO (Multi-Stage Malaria DNA Vaccine Operation) is an ongoing collaboration involving several scientific institutions worldwide, including the Naval Medical Research Center (NMRC), the U.S. Agency for