LETTERS 406 VOLUME 10 | NUMBER 4 | APRIL 2004 NATURE MEDICINE Many human T-cell responses specific for epitopes in Plasmodium falciparum have been described, but none has yet been shown to be predictive of protection against natural malaria infection 1 . Here we report a peptide-specific T-cell assay that is strongly associated with protection of humans in The Gambia, West Africa, from both malaria infection and disease. The assay detects interferon-γ-secreting CD4 + T cells specific for a conserved sequence from the circumsporozoite protein, which binds to many human leukocyte antigen (HLA)-DR types 2 . The correlation was observed using a cultured, rather than an ex vivo, ELISPOT assay that measures central memory-`type T cells rather than activated effector T cells 3,4 . These findings provide direct evidence for a protective role for CD4 + T cells in humans, and a precise target for the design of improved vaccines against P. falciparum. Control of malaria would be greatly enhanced by an effective vaccine. Selection of vaccines has been hampered by the lack of an immuno- logical correlate of protection, making it difficult to distinguish an effective vaccine from one that is only immunogenic. Animal models of pre-erythrocytic malarial immunity, especially immunity induced by irradiated sporozoites, indicate that protection against disease is mediated primarily by CD8 + T cells 1 . However, protection induced by malaria subunit vaccines in animals may also be mediated by either CD4 + IFN-γ-secreting T cells 5 or by antibodies 6 . Although similar T-cell-mediated immune (CMI) responses have been seen in humans vaccinated with irradiated sporozoites 7 , no convincing correlation with protective immunity in the field has yet been identified. There is an urgent need for a clearly defined correlate of protection, as vaccines directed against the circumsporozoite protein have usually been selected on the basis of assays that do not effectively distinguish protected from unprotected people (such as antibody induction 8,9 , proliferation 2,8–11 or cytotoxic CD8 + T-cell response 7,12 ). The ELISPOT assay is increasingly being used as an assay to determine vaccine immunogenicity 13–16 , but there is no evidence as yet that it is effective at determining the protection status of humans exposed to malaria 17 . To find an assay that is associated with protection in the field, we studied volunteers from a double-blind, randomized, controlled phase 2b study of the recombinant malaria vaccine RTS,S/AS02 in The Gambia, West Africa 13 . RTS,S/AS02 is the first recombinant pre- erythrocytic-stage vaccine to show significant efficacy in such a ran- domized, controlled study 13 . It consists of the C terminus of the circumsporozoite protein (CS) fused to the hepatitis B surface anti- gen, coexpressed in yeast with the nonrecombinant hepatitis B surface antigen, together with the adjuvant AS02. The vaccine is safe and well tolerated in humans, and induces strong humoral and also CMI responses in malaria-naive and semi-immune individu- als 8,9,13–15,18–20 . Our volunteers were cleared of parasitemia using Fansidar, then exposed to malaria through natural infection over the course of a single malaria transmission season. Infection was assessed by weekly finger-prick blood samples and whenever a volunteer reported malaria-related symptoms. Time to first parasitemia was the primary endpoint for analysis, and Cox regression analysis was used to estimate the effect of immune responses on the risk of infection, including as censored observation data from uninfected volunteers who left the study before the end of the transmission season. CMI responses were measured prevaccination (d0), 2 weeks after the second vaccination (V2+14), 2 weeks after third vaccination (V3+14), and at the end of the malaria transmission season (V3+150). The CMI assays included proliferation, ex vivo ELISPOT, cultured ELISPOT and cytotoxicity assays, directed toward the whole RTS,S and hepatitis B surface antigens and a series of 15-mer peptides spanning the Th2R, Th3R and CS.T3 immunodominant regions of CS (Table 1). This selection of peptides was shown in previous studies to capture the vast majority of T-cell responses to CS induced by this 1 Nuffield Department of Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU, UK. 2 MRC Laboratories, PO Box 273, Fajara, The Gambia, West Africa. 3 GlaxoSmithKline Biologicals, Rue de l’Institut 89, B-1330 Rixensart, Belgium. 4 London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK. 5 Walter Reed Army Institute of Research, 503 Robert Grant Avenue, Silver Spring, Maryland 20910, USA. Correspondence should be addressed to A.V.S.H. (Adrian.Hill@ndm.ox.ac.uk). Published online 14 March 2004; doi:10.1038/nm1009 A CD4 + T-cell immune response to a conserved epitope in the circumsporozoite protein correlates with protection from natural Plasmodium falciparum infection and disease William H H Reece 1 , Margaret Pinder 2 , Philip K Gothard 1 , Paul Milligan 2 , Kalifa Bojang 2 , Tom Doherty 2 , Magdalena Plebanski 1 , Peter Akinwunmi 2 , Simone Everaere 1 , Katherine R Watkins 1 , Gerald Voss 3 , Nadia Tornieporth 3 , Ali Alloueche 4 , Brian M Greenwood 4 , Kent E Kester 5 , Keith P W J McAdam 2 , Joe Cohen 3 , Adrian V S Hill 1 © 2004 Nature Publishing Group http://www.nature.com/naturemedicine