Meeting reportMeasurement of malaria vaccine efficacy in phase III trials: Report of a WHO consultation
Introduction
On October 4–5, 2006, WHO convened a meeting in Montreux, Switzerland of an expert study group to make recommendations on the measurement of vaccine efficacy in the context of pivotal trials of malaria vaccines. The group included leading statisticians, epidemiologists and clinical trialists with experience of malaria intervention trials and representatives from industry, registration authorities and donor organisations. Discussions at the meeting covered a wide range of issues arising in the conduct of malaria vaccine trials, but focused on case definitions and design and analysis considerations for phase III pivotal trials of malaria vaccines. While recognising that an important component of phase III trials would be safety assessment, this was not a focus of the meeting. We present here a short summary of the discussions and conclusions of the meeting.
The public health burden of malaria is one of the greatest of any infectious agent and the disease is among the most important causes of illness and death for children under the age of 5 years in sub-Saharan Africa. Plasmodium falciparum malaria kills around 1 million children and causes 300–500 million clinical episodes of malaria annually [2], [3]. While the deployment of existing intervention measures, such as vector control, insecticide-treated bed-nets and anti-malarial therapy, can substantially reduce the burden of disease caused by malaria [4], the development and widespread use of an effective malaria vaccine appropriate for young children in Africa would greatly enhance the prospects for malaria control.
Malaria vaccines in development can be categorized into three broad categories representing the three life-cycle stages in the human host: pre-erythrocytic, blood-stage and transmission-blocking vaccines. Morbidity from malaria arises as a consequence of the replication of parasites in red blood cells. Establishment of this blood-stage follows the injection of the sporozoite form of the parasite by female Anopheline mosquitoes and subsequent development through the liver-stage before progressing into the blood. Pre-erythrocytic vaccines target the sporozoite and liver-stages and while such vaccines would prevent blood-stage infection if fully effective, where partially effective they will prevent a proportion of blood-stage infections and may also decrease the risk of disease by reducing the initial parasite density of each blood-stage infection [5], [6]. Blood-stage vaccines target antigens expressed on the surface of infected red blood cells and are designed to inhibit parasite replication in red blood cells. Although not preventing blood-stage infection these vaccines may reduce parasite density sufficiently to prevent or ameliorate development of clinical disease in infected individuals [7], [8]. Transmission-blocking vaccines target the sexual stages of P. falciparum and aim to induce antibodies in humans that may act directly against the sexual stages or, when ingested by mosquitoes, will inhibit the development of infectious parasites in the vector, thus preventing further onward transmission to humans [9].
Several candidate malaria vaccines are progressing through clinical trials [10], [11] and many more are in pre-clinical development. The most advanced malaria vaccine candidate is RTS,S, a pre-erythrocytic vaccine, for which pivotal phase III trial design and site preparation is underway [12], [13], [14], [15], [16], [17]. Other pre-erythrocytic and blood-stage candidates are under evaluation in early phase II field trials. No vaccine against the sexual stage of the parasite is currently undergoing field trial evaluation.
From the information available currently from early field trials it seems possible that none of the malaria vaccine candidates that are at a relatively advanced stage of evaluation will offer high protective efficacy against malaria. However, it is considered that the magnitude of the malaria problem justifies the pursuit of such vaccines, even if the efficacy of these first-generation vaccines may be of the order of 30–50%. Most vaccines in common use against other diseases have efficacy in excess of 70–80% and thus the situation with respect to the present generation of potential malaria vaccines is unusual, though not unique. For example, current potential vaccines against HIV infection are expected by many to have, at best, low efficacy and efficacy of pneumococcal conjugate vaccine against radiological pneumonia is <40% [18] though their efficacy is much higher against disease due to pneumococcal serotypes included in the vaccine.
A critical issue in measuring vaccine efficacy is the selection of an appropriate primary endpoint for pivotal phase III studies. Ideally, such a choice should balance scientific, regulatory and public health considerations, such that the interval between licensure following phase III studies and public health use is minimal. The choice of the primary endpoint for pivotal phase III studies may be influenced by the type of vaccine and its presumed mode of action, as well as the clinical development plan and results of the exploratory phase of clinical testing of each specific candidate.
Early clinical trials might evaluate the extent to which the proportion of children who show signs of malaria parasites in their blood is reduced at different times after vaccination. However, in highly endemic areas such parasitaemia may be essentially asymptomatic and it is unclear how partial protection conferred against this endpoint will translate into a reduction in episodes of symptomatic disease. A proportion of children infected with malaria parasites will develop clinical disease, a smaller proportion of these will develop severe life-threatening disease and a smaller proportion still will die as the end result of the infection. The likelihood of severe disease decreases as a consequence of the development of natural immunity through repeated infections, but other than this it remains largely unknown what determines the severity of an infection with malaria. The most severe effect of malaria is premature death and ideally the definitive evaluation of efficacy of a vaccine would be against death from malaria. However, the confident assignment of causes of death in situations in which many deaths occur outside of hospital, even in the context of a trial, is generally impossible and this endpoint could generally not be measured with sufficient specificity to be used as the primary endpoint in a pivotal phase III trial. As malaria constitutes a high proportion of all child deaths in highly endemic areas, an alternative measure of impact would be against all-cause mortality, but the size of trials that would be required against this endpoint for vaccines that may have relatively low efficacy against malaria would be very large.
Therefore, it would probably be necessary for the primary endpoint for pivotal phase III trials of malaria vaccines to be chosen from clinical endpoints which are more directly measured consequences of malaria. The two primary endpoints for pivotal trials that have been most discussed are clinical malaria and severe, life-threatening malaria. The public health burden of both of these consequences of malaria infection was considered to be sufficiently great for either to be used as the primary endpoint in a pivotal vaccine trial, though it was considered that introduction and use of a vaccine post-registration might be speeded if evidence was already available on its impact against severe disease.
While it was appreciated that as far as possible the endpoints that drive licensure should also provide compelling evidence for implementation, the primary focus of regulatory decision-making is on the safety and efficacy database. For regulatory authorities, demonstration of efficacy must include convincing evidence of clinical benefit. Where the point estimate of efficacy for the vaccine is less compelling than for traditional vaccines, it will be critical to demonstrate convincingly that this level of efficacy has clinically relevant benefit. Whatever the primary endpoint selected, robust case definitions and standardised methods of design and analysis are essential.
However, even the definition of the two conditions under consideration is not straightforward. In highly endemic areas a high proportion of children have malaria parasites in their blood and it is often difficult to ascertain whether children who present with clinical features such as fever and also with parasitaemia have the fever as a consequence of the parasites or whether the parasitaemia is incidental and the true cause of the fever is another childhood infection. Methods of defining clinical malaria and severe malaria are considered below, together with other factors that are likely to be of importance in the design of phase III malaria vaccine trials.
Section snippets
Clinical malaria
The advantage of clinical malaria for use as the primary endpoint in a pivotal trial is that malaria is a common disease in many endemic countries and in many communities children may have several episodes of malaria each year. Thus, the size of a trial to demonstrate even a low level of efficacy is considerably smaller than will be required for much less common outcomes.
Severe malaria
In the context of a phase III trial potential cases of severe malaria are likely to be ascertained from among those who report to a health facility. While there has been considerable recent work on defining severe malaria in several sites in Africa as part of the severe malaria in African children (SMAC) network [20], the emphasis has been on a sensitive case definition to ensure that potential cases are not missed for instituting treatment. However, as for clinical malaria, in the evaluation
Regulatory considerations
In addition to the critical importance of the case definitions, additional major issues of concern from a regulatory perspective include the heterogeneity of malaria in terms of age, geographic location, transmission intensity, and the interaction with co-morbidities. It will also be expected that a rational approach will be demonstrated for measuring vaccine performance and impact in the context of increasing use of other malaria control interventions. Trials will likely be expected to
Multiple episodes of malaria
The consensus of the study group was that the primary endpoint in a trial should be time to first episode of clinical malaria (or severe malaria) rather than the total number of episodes of malaria in trial participants, even though the total burden of malaria in a community might be better measured by the number of episodes. There are several reasons for preferring the former measure. Firstly, there is a considerable statistical complexity involved in analysing multiple episode data in the
Discussion
Given that current malaria vaccines are expected to have relatively low efficacy, the choice of which endpoint to use in a pivotal phase III trial is not straightforward. From a regulatory perspective, efficacy of 30–50% may be sufficient to establish clinical benefit. However, robust case definition, trial design and analysis methods will be central to the demonstration of clinical benefit. The robustness of efficacy data will depend on standardisation and validation of case definitions.
References (26)
- et al.
Malaria: current status of control, diagnosis, treatment, and a proposed agenda for research and development
Lancet Infect Dis
(2002) - et al.
Phase 1 vaccine trial of Pvs25H: a transmission blocking vaccine for Plasmodium vivax malaria
Vaccine
(2005) - et al.
Malaria vaccine developments
Lancet
(2004) - et al.
Efficacy of RTS, S/AS02 malaria vaccine against Plasmodium falciparum infection in semi-immune adult men in The Gambia: a randomised trial
Lancet
(2001) - et al.
Efficacy of the RTS, S/AS02A vaccine against Plasmodium falciparum infection and disease in young African children: randomised controlled trial
Lancet
(2004) - et al.
Duration of protection with RTS, S/AS02A malaria vaccine in prevention of Plasmodium falciparum disease in Mozambican children: single-blind extended follow-up of a randomised controlled trial
Lancet
(2005) - et al.
Efficacy of nine-valent pneumococcal conjugate vaccine against pneumonia and invasive pneumococcal disease in The Gambia: randomised, double-blind, placebo-controlled trial
Lancet
(2005) - et al.
Standardized data collection for multi-center clinical studies of severe malaria in African children: establishing the SMAC network
Trans R Soc Trop Med Hyg
(2006) - et al.
Attributable fraction estimates and case definitions for malaria in endemic areas
Stat Med
(1994) - et al.
Estimating mortality, morbidity and disability due to malaria among Africa's non-pregnant population
Bull World Health Organ
(1999)
The ears of the hippopotamus: manifestations, determinants, and estimates of the malaria burden
Am J Trop Med Hyg
Pre-erythrocytic malaria vaccines: towards greater efficacy
Nat Rev Immunol
Modeling a field trial of the RTS, S/AS02A malaria vaccine
Am J Trop Med Hyg
Cited by (57)
R21/Matrix-M: a second malaria vaccine?
2021, The LancetNew dimensions in the field of antimalarial research against malaria resurgence
2019, European Journal of Medicinal ChemistryRecent Developments in Malaria Vaccinology
2015, Advances in ParasitologyCitation Excerpt :In 2009, the first Phase III malaria vaccine trial was initiated in collaboration with the Clinical Trial Partnership Committee (CTPC, representing several leading research institutes and academic partners from Africa, the EU and the USA) as well as in consultation with African, European and US regulatory authorities. Recommendations of the WHO were also incorporated in the study design (Moorthy et al., 2007, 2009): 6537 infants (6–12 weeks old) and 8923 children (5–17 months old) were randomised in 11 centres of seven African countries and were followed up for 32 months (Leach et al., 2011). Participants were subdivided into three study groups: One group received RTS,S/AS01 in months 0, 1, 2 (in infants EPI vaccines were also co-administered); the second group was vaccinated with the same vaccines, but additionally received a booster dose 18 months after completion of primary vaccination; the third group served as a comparator (Leach et al., 2011).
- 1
See Appendix A.