Oversimplification of the role of immunological processes in the pathogenesis of malaria

The pathogenic basis of malaria has been reviewed many times recently^1,2,3,4, as it has come to be recognized that, even in the case of cerebral malaria, tissues and organs other than the brain might be affected. Indeed, it is argued that the acute disease, including coma, caused by severe falciparum malaria has the same fundamental pathogenesis at a cellular level as severe sepsis, whether bacterial or viral in origin³.

That the pathology of malaria depends on an inflammatory response associated with a number of cytokines was first explored by Clark et al.^5,6, with regard to tumour-necrosis factor (TNF) in rodent malaria. Later, parasite glycosylphosphatidylinositol (GPI) that induced the production of TNF was identified as a malarial toxin^7,8, and was then considered to have potential as a vaccine against disease⁹. As discussed in their Review¹⁰, Schofield and co-workers have since pursued this topic (without acknowledgement of any of this groundwork, although with references to earlier work by others in 1924 and 1939), but dismiss justifiable concerns that such a GPI-based vaccine might actually exacerbate disease by inhibiting responses that limit parasite replication. They argue that it is “difficult to envisage why parasites would produce a molecule to cause their own demise”¹⁰ and ignore the fact that it is the host, not the parasite, that produces such molecules.

In contrast to their pro-inflammatory function and therefore pathogenic role, it is generally accepted that soluble mediators, such as TNF and interferon-γ, are also important in protection against infection and have an antiparasitic effect^2,3. Furthermore, TNF is required for the development of an efficient memory B-cell response and the production of antibody¹¹, and might also contribute to the establishment of cell-mediated immunity, through the activation of Toll-like receptors that recognize parasite GPI and induce the production of nitric-oxide synthase¹². Whereas individuals in areas where Plasmodium falciparum malaria is endemic are reported to develop IgG antibody that is specific for parasite GPI — commonly in an age-dependent manner — there is conflicting evidence that the antibody correlates with protection against symptoms⁴, and it did not seem to confer protection against chronic placental infection and poor birth outcomes⁴. It is also possible that parasite GPI is not the only toxin that is associated with the symptoms of malaria⁴.

Although there is general agreement that much pathology in malaria has an immunological basis, it should also be recognized that to use a anti-GPI antitoxic vaccine to immunize people against malarial disease, as if immunizing them against tetanus toxin, is a dangerous proposition. Vaccinated individuals might not develop certain symptoms of the disease, but their parasitaemia would be uncontrolled¹², leaving them highly susceptible to severe anaemia^3,4. It is noteworthy that infections, such as tuberculosis, salmonellosis and listeriosis, become fulminant in patients with rheumatoid arthritis or Crohn's disease when they are given antibody against TNF13, and patients need to be carefully monitored.

We feel that the potentially harmful effect a GPI-based vaccine might have by interfering with innate cellular responses and the development of cell-mediated immunity should not be ignored^3,4, and that these issues have not been addressed as they should have been in this Review¹⁰.