FIRST AMVTN MEETING, ARUSHA, TANZANIA, 22-24 FEBRUARY, 1995
THE RELATIONSHIP BETWEEN FUNCTIONAL IMMUNE STATUS AND IMMUNE REACTIVITIES
Lars Hviid
Centre for Medical Parasitology, Rigshospitalet, Copenhagen, Denmark
It is possible to induce complete protection against malaria through vaccination with attenuated sporozoites, as has been shown in a number of studies in experimental animals and human volunteers. Nevertheless, a vaccine that is safe, which can induce a high degree of protection, and which is suitable for mass vaccination is not yet available.
An important reason for the relatively slow progress towards such a vaccine may lie in our lack of understanding of the protective mechanisms operating in naturally acquired immunity, and our consequent inability to target our vaccines towards a particular response.
It is evident that protective immunity can be developed through natural exposure to the malaria parasites. In areas of intense transmission, the bulk of morbidity and mortality as a result of malaria infection occurs in infants and children; the severity and frequency of malaria attacks gradually declining with age until only relatively mild and sporadic episodes occur from adolescence onwards. Even then, the acquired immunity is less than solid, and occasional episodes of parasitaemia and disease persist throughout life.
Ideally, immunity in these vaccinated against malaria should induce solid, long-lasting immunity those vaccinated. However, the lesson from the natural acquisition of malarial immunity is that solid immunity may be difficult to induce, although it can be argued that some or even all of the obstacles hindering the development of naturally acquired immunity can be bypassed or at least minimised through artificial immunisation. However, as hinted at before, it is a major problem in this context, that we are still largely ignorant regarding what kind of immune reactivities should be enhanced and which should be avoided, and classical malariometric indices remain the only reliable ways of assessing the efficiency of a given vaccine. However, although these methods provide direct and informative ways of assessing the level of protection afforded at the population level, they say little about the degree of protection afforded at the individual level.
Furthermore, even at best they are unlikely to give more than hints about the immune reactivities and effector mechanisms involved in protection. This is critical, since detailed knowledge about such effector mechanisms and ways of measuring them in immunological assays would be extremely helpful as guides towards optimisation of vaccine efficacy.
Nevertheless, the embarrassing fact is that very little is known about the mechanisms involved in naturally acquired, let alone artificially induced, protective immunity in humans, and no reliable assays are available to measure them.
This depressing situation remains despite much effort put into investigating potential mechanisms by which the immune system can be thought to eliminate or control infection. Indeed, a whole array of such potential mechanisms have been identified in experimental animal systems, and/or in (in vitro) systems employing human material and 'human" malaria parasites.
While there is thus a comprehensive catalogue of effector mechanisms that have all been shown to work in one system or the other, the relative importance of each of these mechanisms in the naturally acquired or artificially induced immunity to malaria is still largely unknown.
Numerous studies have been performed with a view to establish a causal link between various immune reactivities of potential importance and functional immunity. However, many, if not most, of these studies have been unable to demonstrate any convincing relationship between the immune parameters and clinical protection.
Thus, to summarise the evidence, there appears remarkably little correlation between assays of potentially important immune reactivities and actual functional immune status. At best, some co-variation is present at the population level, but to claim an actual causal link between anything measurable in a current immunological assay and clinical protection would be somewhat pretentious in most cases.
As long as clinical or parasitological markers of individual protection remain unavailable, only long-term monitoring can establish clinical protection at the individual level with any reasonable certainty.
Several difficulties are inherent to such an endeavour. Firstly, genuine correlations have to be separated from spurious ones, a task that requires studies designed specifically to confirm (or reject) apparent correlations observed, preferably under the same circumstances under which the original observation was made. Few such studies exist. Secondly, it is very difficult to establish any causal relationship between a given immune parameter and functional immune status in humans, even given that such a relationship exists. As the direct (and invasive) approaches that are available and have been used in experimental anim21 model systems are out of the question for obvious ethical reasons, one has to resort to indirect ways, which almost invariably turn out to be cumbersome, and to yield much less precise results.
Another type of problem is that despite all our efforts, we may not have been looking for the critical kind of reactivity, or even worse: there may be no key reactivity as such. It is possible that critical key parameters of protective immunity have escaped our attention so far, but just as likely, or maybe even more likely, is it that there are no key parameters as such. Perhaps the immune system of each individual finds its own way of establishing a certain degree of immune protection, without inducing too much concomitant immunopathology. What choice of response, or compromise between responses is settled upon is likely to be based on previous immune status to malaria and other infectious agents, to be depending on the available immune capabilities of the individual which are likely to be under genetic control, not to mention parasite properties and a host of other possible variables. It is furthermore likely to involve in unknown combinations of the mechanisms outlined previously. If that is the case, it is not surprising that no single immune parameter can boast a convincing relation to functional immune status, and the search for such a relationship may remain unrewarded because the number of confounding factors is simply unmanageable.
Finally, it is a distinct possibility that although we have actually been looking for the critical key response, we have been looking at the wrong place for it. While this possibility is perhaps less likely when considering antibody-mediated responses, at least those not involving effector cells. There is much reason to worry that the only cells generally available for analysis in humans, i.e., those present in the peripheral circulation, may not be those most important in the protective immune response. Only some of the cell types of the immune system are present in the peripheral circulation in appreciable numbers, and even amongst those that are, only a minor fraction is present at any given time. Recirculation patterns of such cells seem highly depending on cell maturity and specificity, in addition to the degree of pressure imposed on the immune system, in particular by infectious agents. Other cell types are completely sessile, for example endothelial cells; a cell type which is coming increasingly into focus as a key player in the immune response against malaria as well as other diseases. The composition of cells in the periphery is thus likely to be quite different from that in other, and perhaps more relevant, compartments of the immune system. This possibility is likely to be of particular relevance in malaria-endemic areas, where individuals are challenged regularly through natural infection, with consequences for the distribution of cells of the immune system in various compartments, not least in the cell-wise relatively small peripheral circulation. If this is the case, reactivity of peripheral cells in any given assay may not be particularly useful, since they may not give a representative picture of the in vivo response to infection.
In summary, as much as reliable assays of functional immune status are desirable both as monitoring tools in vaccine trials, and as guiding tools in vaccine optimisation, such assays are not yet available, and possibly never will be. Nevertheless, as the presence of such assays would obviously be of immense importance, the search for important immune reactivities and ways of measuring them must continue, and should perhaps even be strengthened. One approach is investigation of available experimental models in an attempt not only to identify the immune mechanisms and reactivities important for protection as well as for pathology, but also to find ways of monitoring them by methods applicable to humans.
Another approach is to search for more appropriate animal models of human malaria, since presently available models suffer from substantial inherent problems of one sort or the other.
A final and extremely important approach is to continue investigations on naturally acquired immunity in endemic populations. While traditional simple cross-sectional studies may not yield substantial new insights, sustained longitudinal studies may be a powerful tool in improving our understanding on what mechanisms are important in both protective and pathogenic immune responses to malaria. Such studies will undoubtedly require collaboration both South-South and South-North, and the inclusion of basic malaria-immunological research should be seriously considered as an important part of any malaria vaccination network.