A malaria vaccine would provide a much-needed way of alleviating the toll of this disease on the world. But one does not yet exist, and there is no scientific consensus on the best lines of research to pursue. This policy brief outlines the progress and challenges in vaccine development.
Malaria is caused by infection with a microbe called a protozoan, which is transferred to people though the bite of Anopheles mosquitoes. Distinct from viruses and bacteria, protozoans are single-celled eukaryotic micro-organisms. Four species of protozoan parasites cause malaria in humans — Plasmodium falciparum, P. vivax, P. ovale and P. malariae. No vaccine has ever been made against a protozoan that causes the disease in people.
P. vivax, P. malariae and P. ovale infections are as widespread as P. falciparum infection but they are not usually fatal. P. falciparum, however, is highly dangerous. People who are often infected by it — such as those living in rural Africa develop a partial immunity to the disease. But infection can be fatal in other groups, such as children or people without naturally acquired immunity e.g. tourists, aid workers and military personnel.
Methods for controlling and treating malaria have been fiercely debated since the earliest descriptions of malaria more than two thousand years ago. At present, disease control relies on three main efforts:
- Drug treatment of patients
Treatment to kill the Plasmodium infection is the mainstay of control in areas of high transmission in Africa and Asia. But drug resistance, cost, and inadequate infrastructure are severely hampering efforts. More targeted treatment efforts to those most vulnerable to malaria are taking place in the form of intermittent preventative treatment to pregnant women and infants, involving the administration of regularly spaced doses of antimalarial drugs.
- Mosquito control
Measures to prevent the spread of malaria include using insecticide-impregnated bednets, insecticide treatment of mosquito habitats, killing mosquito larvae with chemicals at wetland breeding sites, and suppressing breeding sites by water drainage. Indoor residual spraying of homes with insecticides, although phased-out in many areas following concerns about the use of DDT, is now returning to favour, most notably in southern Africa and India.
- Post-eradication monitoring and treatment
When malaria is brought into areas where it was eradicated in the 20th century (such as Europe and North America), all those who might be carrying infection are rapidly treated to avoid re-establishment of transmission.
Ideally, a malaria vaccine that could greatly improve global public health would induce a rapid and protective immune response that completely eliminates the infection.
Less ideal, but more feasible and still highly desirable, would be a partially protective vaccine that does not completely prevent infection but would at least boost the immune system sufficiently to lessen the severity of disease caused by P. falciparum.
Why a vaccine?
Vaccines are the most cost-effective component of public health services. They are usually given orally or by injection of an inactivated (killed) or attenuated (live, but non-virulent) whole pathogen. When given together with a boosting substance called an adjuvant, a successful vaccination safely induces an immune response that leads the body to recognise and kill the infectious agent.
A malaria vaccine seems to offer the greatest hope of achieving significantly improved malaria control, particularly in Africa, where the ecological habitat is such that effective mosquito control has proved difficult or impossible to maintain.
The success of other vaccination campaigns demonstrate that the control of major infectious diseases is achievable on a global scale. The World Health Organization's (WHO) Expanded Program on Immunisation (EPI) was launched in 1974. This effort has successfully provided subsidised vaccination worldwide against six major childhood diseases: measles, diphtheria, pertussis (whooping cough), tetanus, polio and tuberculosis. Furthermore, smallpox has been eradicated as a result of vaccination, and polio seems close to being eliminated.
Additionally, the WHO immunisation programme has created the global skill base and infrastructure through which other protective vaccines could be administered. If funding were available, existing vaccines against diseases such as Hepatitis B, Haemophilus influenzae type b (Hib) and yellow fever could be introduced. Indeed, in view of the high rates of death and disease in children caused by malaria, great efforts would be made to incorporate a malaria vaccine into the WHO immunisation schedules, if an effective and affordable option were available.
So why is there not yet a malaria vaccine? The lack of a malaria vaccine today can be attributed to several complementary factors including lack of interest by governments and the private sector and significant scientific obstacles. This policy brief outlines the direction that any future malaria-vaccine research policy could take.
Inadequate attention and competing ideas
Until quite recently, malaria vaccine development has not been a global priority. Research in this area has had only a few adherents, who scraped together funding to test small batches of vaccine prototypes in early-stage clinical trials. The research field ran against the grain of conventional 'pure' academic immunological research, and at the same time was too risky to attract potential partners in either the small biotech sector or the established pharmaceutical industry.
Malaria vaccine research involves the unfamiliar combination of vaccinology, immunology and malariology. There are few established theoretical principles in this field to serve as guidelines. Instead, workers must contend with a mass of frequently inconsistent data from field and laboratory studies that confuse attempts to forge ahead in any one direction.
Some malaria researchers have noted the similarities with the effort to develop an HIV vaccine, which has also been affected by competing research priorities in the past 25 years. 
'Whole organism' vaccines
Malaria researchers in the 1960s and 1970s followed the traditional approach of producing attenuated or killed whole organisms for inoculation. This method has yielded live attenuated vaccines that protect against smallpox, yellow fever, measles, mumps, rubella, polio (the Sabin oral vaccine) and varicella zoster (chicken pox).
The same approach has produced killed or inactivated viral vaccines against polio (the Salk injected vaccine) and Hepatitis A. Killed or inactivated whole-bacteria based vaccines are used safely and effectively to protect against pertussis (whooping cough), pneumonia and meningitis caused by Haemophilus influenzae type b (Hib), typhoid, plague and cholera.
All Plasmodium species have distinct forms in both the human and mosquito stages of their life cycle. The first malaria vaccine was made from fragmented and dissolved malaria parasites, chosen in the form that is ready to invade red blood cells (merozoites). Tested in the 1960s, this vaccine gave some protection in experimental monkey models. But inconsistent results, difficulties in producing immunising material, and dependence on toxic adjuvants led to a halt of such trials.
Instead, workers took an alternative approach using the emergent recombinant DNA/genetic engineering technologies to purify individual protective merozoite surface components that could be recognised by the immune system (antigens).
Meanwhile, other investigators attempted to create vaccines using attenuated sporozoites — the form of the malaria parasite that is transferred to humans by mosquitoes and that invades the liver.  Live, infected mosquitoes were used to deliver the vaccine after having first been subjected to X-ray radiation to render the parasites unable to multiply. A very high proportion of volunteers bitten by the irradiated mosquitoes could subsequently fight off infection with normal untreated malaria sporozoites.
Many investigators consider this approach technically too difficult to pursue, and impossible to scale up — although there are advocates of novel attenuated sporozoite production programmes. 
As with whole merozoite vaccines, attention on sporoziotes largely moved to discovering and purifying the major sporozoite protective antigen or antigens. 
The focus of malaria vaccine research, therefore, switched to vaccines based on one or more immunogenic components of the parasite; the so-called subunit vaccine approach. Early precedents for successful subunit vaccines are the bacterially secreted toxins used in the diphtheria and tetanus vaccines. The most influential recent precedent has been the Escherichia coli or yeast-produced Hepatitis B surface-protein vaccine, which is currently the only genetically engineered recombinant vaccine to be successfully produced and deployed.
Subunit vaccines: the challenge of target selection
A key problem for subunit vaccine development is how to choose the parasite component (i.e. the protective antigen) to induce immunity. There are three stages of the malaria parasite's life cycle that seem especially vulnerable to the immune system of infected human hosts and are thus prime vaccine targets. They are the:
- pre-erythrocytic stage — not only the circulating sporozoites transferred by mosquitoes but also those that continue to develop after entry into the liver.
- asexual blood stage — the merozoites that emerge from the liver to invade, and subsequently grow in, red blood cells.
- sexual blood stage (gametocytes) — taken up by the blood-feeding mosquito to continue the protozoan life cycle.
This complex biology gives rise to two target selection problems. First, there are several thousand proteins (plus carbohydrates and lipids) made by malaria parasites during human infection. These compounds can serve as antigen targets for two types of immune responses: the secretion of antibodies to attack parasites floating in the blood stream, and infected red blood cells; and the action of white blood cells known as T cells, which can attack infected liver cells, and also aid in boosting antibody production and specificity.
Second, many antigenic proteins vary between individual parasites within an infected person. Making matters even more complex, individual malaria parasites can also switch the selection of proteins that appear on the surface of infected red blood cells to evade the host's antibodies.
Malaria, thus, poses a formidable challenge to the human immune system. Indeed a successful vaccine might seem an impossible idea, were it not for the knowledge that people clearly can become immune. This natural immunity takes years to develop as people encounter a very diverse population of parasites. In Africa, tropical conditions sustain huge populations of mosquitoes and humans can be bitten by infected mosquitoes once or more every night. Every bite introduces 10 – 100 genetically diverse sporozoites. So far, however, it has not been possible to test the protective efficacy of immunisation with more than a handful of the 6000 or so proteins made by P. falciparum.
Sporozoite-protein based vaccines
At New York University in the late 1980s, Ruth Nussenzweig and her colleagues discovered the major surface antigen of sporozoites, the circumsporozoite protein (CSP). Several clinical trials in the late 1980s with chemically synthesised versions of CSP were unsuccessful.
Since then, a development programme run by the US Army and Smith Kline, now GlaxoSmithKline, has led to the development of the most successful CSP-based malaria vaccine so far — the RTS,S formulation.
The RTS,S formulation consists of 'virus-like particles', produced through recombinant DNA technology. The formulation contains a mixture of Hepatitis B surface antigens and a fragment of CSP and is given with a very potent adjuvant. The adjuvant, ASO2, is an oil-in-water emulsion of a lipid (fat particle) from the cell wall of salmonella bacteria (monophosphoryl lipid A) and a saponin-type detergent, and is essential for the vaccine's protective effect. RTS,S is thought to act by stimulating both the production of antibodies that block the parasite's invasion of the liver and T cells that kill the replicating parasites inside liver cells.
This vaccine has been shown to confer a substantial (40–70 per cent, depending on the trial) but short-lived protection in volunteers deliberately exposed to the bites of infected mosquitoes (usually five infected mosquitoes per volunteer).
However, the vaccine conferred only short periods of protection (less than six months) to naturally exposed adult Gambian volunteers. These volunteers were already semi-immune to malaria, in that they had been naturally exposed to the disease since childhood. In the most recent reported field trial in Africa, the vaccine gave around 30 per cent protection overall against the first clinical attack of malaria in Mozambican children and reduced the incidence of severe disease by 58 per cent .
Multi-epitope peptide vaccines
Also during the late 1980s and early 1990s, attention turned to the idea of creating vaccines containing multiple antigen targets — so-called multi-epitope peptide vaccines. Colombian researchers led by Manuel Pattaroyo in Bogota created such a vaccine formulation known as SPf 66. The vaccine contained chemically synthesised protein fragments (peptides) representing portions of three merozoite surface antigens, linked together with a protein sequence matching CSP (a so-called multi-epitope synthetic vaccine). It showed great promise in Aotus monkey experiments and in early human clinical trials in South America.
However, further field trials in Africa and South East Asia failed to repeat the early success and this candidate has now been largely dropped from the global vaccine development effort. But the concept of multi-epitope vaccines remains valid.
Vaccines to trigger and boost T cells
This class of vaccine aims to elicit cytotoxic T cells that can kill malaria parasites in infected liver cells. Unlike red blood cells, liver cells can alert the immune system to the parasite invasion and thereby render themselves targets for killing.
These 'prime-boost' vaccines involve a two-step procedure in which volunteers are first given a DNA-based vaccine to instruct body cells to produce malaria proteins as if they were infected with the parasite. This first step provides the initial stimulus to T cells that can detect these liver-stage antigen targets.
The second, booster inoculation contains attenuated viruses that carry synthetic portions of various pre-erythrocytic, liver-stage antigens, including CSP and also another sporozoite protein known as TRAP.
This booster step magnifies the numbers of T cells capable of recognising and killing malaria-infected cells, so that in theory, there can be a rapid and powerful response to a parasite infection. However, although results of several Phase I trials have shown safety and some immunogenicity, Phase IIb trials on adult male Gambian volunteers have not shown significant protective efficacy of such vaccines.  Further clinical trials of alternative prime-boost vaccines designed to produce stronger responses are underway.
For tourists or the military, blood-stage vaccines are less appealing than vaccines that would block all infection. Blood-stage vaccines aim to reduce the most damaging aspect of the malaria parasite life cycle — its uncontrolled asexual replication in human red blood cells. Research in this area has focused on proteins that enable the parasite to latch on to, and invade, red blood cells.
The lead candidate proteins of most of these types of vaccines contain fragments of the merozoite surface protein-1 (MSP-1) and the merozoite antigen-1 (AMA-1), produced through genetic engineering. They exist as advanced prototypes in or near clinical trials, alone and in combination.
Other early-stage human trials involve vaccines containing other malaria parasite blood-stage antigens (including glutamate-rich protein (GLURP), Exported Protein 1 (EXP-1), 175 kilo-Dalton erythrocyte binding antigen (EBA-175), serine repeat antigen (SERA), ring-infected erythrocyte surface antigen (RESA), and Merozoite Surface Proteins 2 and 3 (MSP 2 and 3).
There is, as yet, no striking proof that these formulations confer high-level protection against clinical malaria but it remains an active area of research.
Preventing infected cells sticking to human tissue
Other blood-stage malaria vaccines aim to stop malaria-infected red blood-cells sticking to human tissues. Such adhesion can lead to serious and fatal disease. For example, during pregnancy when infected red cells accumulate in the placenta and impede blood supply to the growing foetus.
The adhesion process involves malaria parasite proteins called P. falciparum erythrocyte membrane protein-1 (PfEMP-1) appearing on the surface of infected red cells, causing the cells to bind to protein 'receptors' on the surface of body tissues. This vaccine strategy is based on evidence that natural human immunity to malaria involves acquiring a range of antibodies against PfEMP-1 proteins.
Women who have acquired malaria immunity during childhood, seem to lose this immunity during pregnancy. This loss of immunity, leading to heavy malaria parasite infection of the placenta is a serious cause of low birth weight and infant mortality and morbidity.
The explanation for the loss of resistance, it seems, is that malaria parasites, when they infect pregnant women, begin to express a form of PfEMP-1 molecule that can only bind in the placenta. With no previous encounter with these antigens, women become malaria-susceptible during their first pregnancy.
Other severe malaria syndromes, such as cerebral malaria in children, may also be triggered by similar potentially blockable adhesion processes. Prototypes of PFEMP-1 based vaccines are being produced and their safe testing and delivery is under discussion.  One possibility is to genetically engineer their co-delivery with other vaccines designed to protect unborn children against infectious disease, such as the rubella component of the MMR (measles, mumps and rubella) vaccine.
Transmission blocking vaccines
Although sexual fusion of malaria parasite gametes takes place in the female mosquito after a blood meal, this process could potentially be blocked by serum antibodies produced in the blood of the human before they are bitten.
Vaccines aimed at eliciting such antibodies might, therefore, prevent mosquitoes from becoming infected after feeding on people who have been vaccinated with transmission-blocking vaccines.
These vaccines benefit the entire community rather than a single individual, by reducing transmission from one person to another. Phase I trials of transmission blocking vaccines for both P. falciparum and P. vivax are underway. 
The future: comparative testing of vaccines
Current malaria vaccine development projects concentrate mainly on the small number of parasite surface antigens involved in invasion, adhesion onto host tissue, or development in the liver, and are based on the assumption that such processes are vulnerable to disruption.
There has been much vague talk about the need for more 'outside the box' thinking in vaccine candidate selection, but as yet no significant challenge to the logic of concentrating on triggering antibodies that block receptors, or T cells that kill infected liver cells.
Additionally, an important technical constraint has been the lack of simple, standardised assays that allow vaccine candidates to be compared. There is no clear consensus on which vaccine antigens trigger the strongest immune responses, and what type of response (antibody or T cell), measured in vitro assays, can serve as an accurate correlate of in vivo human immune responses to malaria.
Such assays have been difficult to develop as there is no small animal model — such as mice — of P. falciparum infection. Research in monkeys is possible, but is, on a practical level, increasingly difficult and expensive to do.
Hundreds of research papers report measurements of human antibody or T cell responses to particular antigens after natural infections. Dozens more studies show positive or negative correlations between immune responses to particular antigens and measures of human immune protection such as the absence of, or low concentrations of, parasites in the blood. However, these results are rarely comparable between studies and thus not clear enough to enable the vaccine-research community to focus resources on a short list of promising candidates.
Future progress might be aided by researchers working together to create and develop controlled, comparative immunisation experiments. Such work might allow researchers to compare vaccine-elicited immune responses (such as concentration of serum antibodies, the response to whole parasites, and inhibition of parasite growth) on a small scale.
Such comparisons would allow workers to better prioritise candidate vaccines before beginning the expensive production of vaccines that are fit for human clinical trials. Several groups in the USA and Europe have started to work together to achieve such improved pre-clinical comparative testing.
Despite only small-scale investment, malaria vaccine research has made significant progress in the past decade. Advocacy has increased and major charitable support from the Bill and Melinda Gates Foundation has prompted other foundations, national governments and organisations such as the European Union and the WHO, to consider increasing their financial commitment to finding a vaccine.
There is a widely held consensus on how to proceed: give support to vaccine development projects that are supported by a substantial body of peer-reviewed data; broker novel collaborations, in particular between academia and industry; and let the competitive research environment produce the most effective vaccine.
The pace of vaccine development is quickening, thanks to increased efforts on several fronts: basic research on antigen target selection, industrial research and development to optimise experimental vaccines, and clinical testing of the best candidates so far. As a result, ‘first generation’ malaria vaccines are on the horizon. Such vaccines are likely to give only partial rather than complete protection, however, and will supplement rather than replace vector control and drug treatment.
Although vaccine research looks ready to start contributing to the decline of malaria, it is likely to be gradual, occurring over decades rather than years – as the history of infectious disease indicates. By maintaining a broad-based research effort, the vaccine research community may one day be able to convince political players, philanthropic donors and even more importantly, key target populations such as African mothers, that a vaccine will provide worthwhile protection for their children.
Snow R. S., et al. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature, 434, 214-217 (2005).
United Nations Development Programme. Human Development Report 2003: Millenium Development Goals: a Compact Amongst Nations to End Human Poverty. Oxford University Press, Oxford, 2003.
 Esparza J. The Global HIV Vaccine Enterprise. Internat Microbiol, 8, 93 – 101 (2005).
 Nussenzweig R.S. et al. Protective immunity produced by the injection of X-irradiated sporozoites of Plasmodium berghei. Nature, 216, 160 – 62 (1967)
 Luke T. C., Hoffman S.L. Rationale and plans for developing a non-repPcating, metaboPcally active, radiation-attenuated Plasmodium falciparum sporozoite vaccine. J Experiment Biol 206, 3803 – 08 (2003).
 Alonso P. L. et al. The efficacy of the RTS,S/ASO2A vaccine against Plasmodium falciparum infection and disease in young African children: randomised control trial. Lancet, 364, 1411 – 20 (2004).
 Moorthy V.S. et al. A randomised, double bPnd, controlled vaccine efficacy trial of DNA/MVA ME-TRAP against malaria infection in Gambian adults. PLoS Medicine 1, (2004).
 Salanti A. et al. Selective up-regulation of a single distinctly structured var gene in chondroitin sulphate A- adhering Plasmodium falciparum involved in pregnancy-associated malaria. Mol Microbiol, 49, 179–91 (2003).
 Malkin E. M. et al. Phase 1 vaccine trial of Pvs25H: a transmission blocking vaccine for Plasmodium vivax malaria. Vaccine 23, 3131–38 (2005).