The objective of the studies was to assess the safety and immunogenicity of ChAd63 ME-TRAP and MVA ME-TRAP administered in a heterologous prime boost regimen to healthy malaria-exposed adults.

The first trial (Trial A) was conducted at the Medical Research Council (MRC) Unit field site located within Sukuta Health Centre in Kombo North district of The Gambia, West Africa. Sukuta village has an estimated population of 17,000 (2003 census). The climate is typical of sub-Saharan Africa with a long dry season lasting from December–June followed by a relatively short rainy season from July–November when the majority of P. falciparum malaria transmission occurs [36].

The second trial (Trial B) was conducted in Vipingo, Kilifi County, Kenya, East Africa. Participants were recruited from the Rea Vipingo Sisal Plantation Estates in Kilifi which has over 1000 employees and a land area of 3,950 hectares. In Kilifi, there are two seasons of high transmission of P. falciparum malaria coinciding with the long monsoon rains (April to June) and the short rains (October to December) [37].

Recent studies have reported a decline in malaria transmission in both sites [38], [39] but a surge was recorded during the period of vaccinations in the Sukuta site (M. Afolabi personal communication).

Healthy males aged 18–50 years were invited to participate in the studies. There was no selection of participants on the basis of pre-existing neutralizing antibodies (NAb) to the ChAd63 vector prior to enrolment. Volunteers were considered eligible if they were consenting adult males aged 18–50 years in good health who were likely to remain resident in the study area for the study duration. Exclusion criteria included any evidence of any acute or chronic illness or hematological, renal or hepatic pathology. Specific exclusion criteria included; prior receipt of an investigational malaria vaccine, recent or planned use of any investigational drug, vaccine, immunoglobulin or any blood product, confirmed or suspected immunodeficiency, history of surgical splenectomy, concurrent participation in another clinical trial or within 3 months of this study (see Protocol S1 (Trial A: The Gambia), Protocol S2 (Trial B: Kenya) for the full list of inclusion and exclusion criteria). Blood positivity for P. falciparum at screening was not an exclusion criterion.

We conducted two Phase Ib open-label, dose-escalation malaria vaccine trials (Figure 1). Both clinical trials evaluated low (1×10¹⁰ vp) and high dose (5×10¹⁰ vp) ChAd63 ME-TRAP. Trial B also compared intramuscular and intradermal routes of administration of 2×10⁸ pfu MVA ME-TRAP. The same lot of each vaccine was used in both trials. The trials were conducted independently however the same Data and Safety Monitoring Board (DSMB) was used for each trial. The clinical trial protocols and supporting CONSORT checklist are available as Supplementary Information; see Protocol S1 (Trial A: The Gambia), Protocol S2 (Trial B: Kenya), and Checklist S1.

In Trial A, eligible participants were allocated to receive either ChAd63 ME-TRAP 1×10¹⁰ viral particles (vp) (group 1; n = 6) or ChAd63 ME-TRAP 5×10¹⁰ vp (group 2; n = 10) administered intramuscularly in the deltoid. All participants were subsequently vaccinated in the opposite arm 56 days later with 2×10⁸ plaque forming units (pfu) MVA ME-TRAP administered intramuscularly. The first participant in group 1 to receive ChAd63 ME-TRAP 1×10¹⁰ vp was vaccinated in isolation. 48 hours later, two further participants were enrolled in group 1. Prior to dose escalation of ChAd63 ME-TRAP from 1×10¹⁰ vp to 5×10¹⁰ vp, safety data from group 1 up to 14 days post ChAd63 ME-TRAP was reviewed by the DSMB. There was a protocol-required interval of at least 14 days between immunization of groups 1 and 2. Details of clinical follow-up and safety monitoring are given in Protocol S1 and S2.

In Trial B, eligible participants were allocated to receive either ChAd63 ME-TRAP 1×10¹⁰ viral particles (vp) (group 1; n = 10) or ChAd63 ME-TRAP 5×10¹⁰ vp (group 2; n = 20) administered intramuscularly in the deltoid. All participants were subsequently vaccinated in the opposite arm 56 days later with 2×10⁸ plaque forming units (pfu) MVA ME-TRAP. Participants in each group were randomised 1∶1 to receive MVA ME-TRAP administered intramuscularly (IM) or intradermally (ID). The first 3 participants in group 1 were vaccinated with ChAd63 ME-TRAP 1×10¹⁰ vp 7 days ahead of the remaining 7 participants in this group. There was an 8 day interval between enrolment of group 1 and group 2. Details of clinical follow-up and safety monitoring are given in Protocol S1 and S2.

For both trials, a time window ranging between 1 and 28 days depending on the visit was allowed for vaccination and follow-up visits. Throughout the paper, study day refers to the nominal time point for a group and not the actual day of sampling.

30 participants were systematically allocated to receive either 1×10¹⁰ vp ChAd63 ME - TRAP or 5×10¹⁰ vp dose ChAd63 ME in a ratio of 1∶2. 8 weeks later participants were randomised 1∶1 to receive 2×10⁸ pfu MVA ME-TRAP administered intramuscularly (IM) or intradermally (ID). The randomization sequence was generated by an independent statistician using STATA programme. Group allocations were kept in sealed opaque envelopes stored in a locked cabinet by the study coordinator who gave them to the research nurses on day of vaccination. Participants and clinical study staff were un-blinded to group allocation, however, field workers were blinded to group allocation.

These were observational and descriptive studies to assess the safety and immunogenicity of ChAd63 ME-TRAP and MVA ME-TRAP in malaria exposed adults. The sample sizes were chosen to allow estimation of the magnitude of the primary outcome measures, especially of serious adverse events (AEs) rather than assessment of statistically significant differences between groups.

The clinical trial protocols and associated documents were approved by Gambia Government/MRC Joint Ethics Committee for Trial A and The Kenya Medical Research Institute National Ethics Review Committee for Trial B. Documents for both clinical trials were reviewed and approved by the Oxford Tropical Research Ethics Committee (OXTREC). Regulatory approval was given by the Medicines Board of The Gambia for Trial A and The Pharmacy and Poisons Board of Kenya for Trial B. All participants gave documented informed consent prior to any study procedure being undertaken. The study was conducted according to the principles of the Declaration of Helsinki (2008) and the International Conference on Harmonization (ICH) Good Clinical Practice (GCP) guidelines. An independent DSMB and local safety monitors provided safety oversight and GCP compliance was independently monitored by an external organization at both trial sites (Appledown Clinical Research Ltd, Great Missenden, UK).

Generation of the recombinant vectors has been previously described [40], [41]. Vaccines were manufactured under Good Manufacturing Practice (GMP) conditions by the Clinical Biomanufacturing Facility, University of Oxford (ChAd63 ME-TRAP) and IDT Biologika GmbH, Dessau, Germany (MVA ME-TRAP). Briefly, ChAd63 ME-TRAP was grown in suspension HEK293 cells and purified by caesium chloride density-gradient centrifugation. MVA ME-TRAP was generated in chicken embryo fibroblasts (CEFs) and purified by sucrose density-gradient centrifugation. Each vaccine lot underwent comprehensive quality control analysis to ensure that the purity, identity and integrity of the virus met pre-defined specifications. Vaccine lots were stored at the clinical site in a −70°C freezer and vaccines were temperature monitored when moved.

The antigen ME-TRAP contains a fusion protein of multiple epitopes (ME) and the P. falciparum pre-erythrocytic thrombospondin-related adhesion protein (TRAP). The ‘ME’ is a string of 20 epitopes, mainly CD8 T cell epitopes from P. falciparum pre-erythrocytic antigens, fused to the thrombospondin-related adhesion protein. The individual cytotoxic T lymphocyte (CTL) epitopes which constitute the ‘multiple epitope’ part of ME-TRAP represent six potentially protective target antigens and are included to try to broaden the immune response rate in the vaccinated population. The ME string is fused to the entire sequence of the T9/96 strain of P. falciparum TRAP and the ME-TRAP hybrid is a 2398 base-pair insert which encodes for a single polypeptide of 789 amino acids [21].

In each trial participants were observed for 1 hour post each immunization. Following each immunization participants in both trials were reviewed at home by a trained field worker and findings recorded on standardised case report forms. Local and systemic vaccine reactogenicity was evaluated and graded for severity, outcome and association to vaccination as per the criteria outlined in Tables 1, 2, 3, 4, 5. Local solicited reactions were; pain, discoloration, swelling, warmth, pruritus, scaling or blistering at the injection site. Systemic solicited symptoms were; fever (axillary temperature>37.5°C), feverishness, malaise, arthralgia, headache, myalgia and nausea or vomiting. Unsolicited symptoms that occurred within 30 days of each immunization were assessed, recorded and their relationship to the immunization determined. Serious adverse events (SAEs) were assessed throughout the study period. First response® Rapid Diagnostic kits (Trial A) or blood film microscopy (Trial B) [42] were performed for diagnostic purposes whenever participants presented with symptoms suggestive of malaria at each trial site. In Trial A blood was sampled at all clinic visits post vaccination (days 14, 56, 63, 90 and 300 post ChAd63 ME-TRAP) and full blood count, creatinine and alanine aminotransferase (ALT) measured. In Trial B blood was sampled at all clinic visits post vaccination (days 14, 56, 63, 91 and 308 post ChAd63 ME-TRAP) and full blood count, creatinine and ALT measured.

Blood samples were collected into lithium or sodium heparin-treated vacutainer blood collection tubes (Becton Dickinson, UK). PBMC were isolated and used within 6 hours in fresh assays as previously described [43]. Excess cells were frozen in foetal calf serum (FCS) containing 10% dimethyl sulfoxide (DMSO) and stored in liquid nitrogen. For serum preparation, untreated blood samples were stored at 4°C and then the clotted blood was centrifuged for 5 min (1000 xg). Serum was stored at −80°C.

Peptides were purchased from NEO Peptide (Cambridge, MA, USA). The peptides, 20 amino acids (aa) in length and overlapping by 10 aa covered the entire ME-TRAP insert present in the viral vectored vaccines. Peptides were also synthesised for the sequence of TRAP from the 3D7 strain. Peptides were reconstituted in 100% DMSO at 50–200 mg/mL and combined into various pools for ELISPOT and flow cytometry assays. Peptides are listed in Table S1.

The kinetics and magnitude of the T cell response to ME-TRAP were assessed over time by ex-vivo IFN-γ ELISPOT assays performed on blood samples taken at each clinic review (days 14, 56, 63, 90 and 300 post ChAd63 ME-TRAP in Trial A and days 14, 56, 63, 91 and 308 post ChAd63 ME-TRAP in Trial B). Ex-vivo IFN-γ ELISPOT assays were performed with an 18–20 hour stimulation of PBMC with peptides pools containing up to 10 peptides per pool, including peptides representing the T9/96 and 3D7 strains. Fresh PBMC were used in all ELISPOT assays using a previously described protocol, except that 50 µL/well ME-TRAP peptide pools (final concentration of each peptide 10 µg/mL) were added to duplicate wells, 50 µL/well R10 and DMSO control were added to negative un-stimulated wells, and 50 µL/well Staphylococcal enterotoxin B (SEB) (final concentration 0.02 µg/mL) plus phytohemagglutinin (PHA) (final concentration 10 µg/mL) was added to positive control wells. Spots were counted using an ELISPOT counter (Autoimmun Diagnostika (AID), Germany). Results are expressed as the mean of the duplicate IFN-γ spot-forming units (SFU) per million PBMC. Background responses in un-stimulated control wells were almost always less than 20 spots, and were subtracted from those measured in peptide-stimulated wells. Responses are shown as the summed response to all the ME-TRAP (T9/96) peptide pools (unless otherwise stated).

Data were analyzed using GraphPad Prism version 5.03 for Windows (GraphPad Software Inc., California, USA) and Stata 10.0 (Statacorp LP, Texas, USA). Geometric mean or median responses for each group are described. Significance testing of differences between two groups used the two-tailed Mann-Whitney U test or Wilcoxon signed rank test as appropriate. Correlations were analyzed using Spearman's rank correlation co-efficient (r_s) for non-parametric data. Collated immunology data was analysed by multivariate linear regression using log-transformed ELISPOT results. A value of P<0.05 was considered significant.

Recruitment for Trial A took place in the Gambia between 19^th May 2010 and 9^th June 2010. Sixteen healthy male adult participants were enrolled, immunized and followed up (Figure 1). The mean age of volunteers was 33.3 years (range 23–48 years). All participants were from the Mandinka ethnic group. Vaccinations began in June 2010 and all follow-up visits were completed by May 2011. With the exception of a participant in group 2 who was lost to follow-up after review on Day 90, all volunteers attended all visits as scheduled and completed the study.

Recruitment for Trial B took place in Kenya between 10^th June 2010 and 7^th July 2010. Thirty healthy male adult participants were enrolled, immunized and followed up (Figure 1). The mean age of volunteers was 32.5 years (range 22–50). 47% of participants were from the Mijikenda ethnic group, 27% were Luos and the remaining 26% from other ethnic groups. Vaccinations began in June 2010 and all follow-up visits were completed by May 2011. All volunteers attended all visits as scheduled and completed the study.

No unexpected AEs or SAEs occurred and no volunteers were withdrawn due to AEs. AEs associated with ChAd63 ME-TRAP AEs were all mild in intensity (Figure 2 & Table S2). No clear difference in reactogenicity was noted between participants receiving 1×10¹⁰ vp and 5×10¹⁰ vp ChAd63 ME-TRAP. All AEs resolved without sequelae within 72 hours of immunization. MVA ME-TRAP was more reactogenic than ChAd63 ME-TRAP though still well tolerated with the majority of AEs mild in intensity (Figure 3 & Table S2). Whilst systemic reactogenicity of MVA ME-TRAP was unaffected by route of administration, intradermal injection was associated with increased local reactogenicity (namely injection site swelling, warmth, discoloration, blistering and pain) compared to intramuscular administration. All AEs resolved without sequelae. Local AEs post-MVA ME-TRAP resolved between 1 to 15 days post immunization apart from one case of swelling that lasted for 30 days occurring in an individual who received MVA ME-TRAP intradermally (maximum 35 mm of swelling gradually resolving over time, not associated with a sterile abscess). Systemic AEs post-MVA ME-TRAP resolved within 48 hours of immunization. Minor laboratory abnormalities were seen post immunization. However, all were mild, clinically insignificant and resolved fully (Table S3). In Trial A, two volunteers were diagnosed and treated for P. falciparum malaria by First response® Rapid Diagnostic kits; one volunteer in group 1, 6 months post MVA ME-TRAP and another in group 2, 34 days post MVA ME-TRAP. These volunteers demonstrated no unusual features of the illness and were included in the final analyses. Similarly, 2 volunteers in Trial B were diagnosed and treated for P. falciparum malaria by blood film, one on day 17 and one on day 139 and were included in the final analyses. These volunteers demonstrated no unusual features of the illness.

Heterologous prime boost with ChAd63-MVA ME-TRAP induced high frequencies of antigen-specific IFNγ-secreting T cells in both trials as measured by ex-vivo IFNγ ELISPOT. Peak IFNγ ELISPOT responses were detected 7 days post MVA ME-TRAP when a positive response (defined as responses above the lower limit of detection and at least double the response measured at Day 0) was detected in 90% and 100% of recipients in Trial A and Trial B respectively (Table 6 & Figure 4A). Responses were well maintained post immunization and detectable in 88% of all vaccinees 9 months after the final immunization (median 116 SFC/10⁶ PBMC, 95% CI 133, 268).

In Trial A there was no statistically significant difference in peak IFNγ ELISPOT response between individuals receiving 1×1010 vp ChAd63 ME-TRAP (median 266 SFC/106 PBMC, 95% CI -208, 1310) and individuals receiving 5×1010 vp ChAd63 ME-TRAP (1558 SFC/106 PBMC, 95% CI 550, 2179) (p = 0.11; Mann Whitney U, 2 tailed test). In contrast in Trial B, the median peak IFNγ ELISPOT response elicited in individuals receiving 5×1010 vp ChAd63 (1536 SFC/106 PBMC (95% CI 1230, 2355) was significantly greater than the peak response in individuals receiving 1×1010 vp ChAd63 ME-TRAP (590 SFC/106 PBMC, 95% CI 399, 1314, p = 0.011 2 tailed Mann Whitney test).

In Trial B there was no significant difference in peak immune response between volunteers receiving MVA administered intramuscularly or intradermally for either dose of ChAd63 ME-TRAP (ChAd63 ME-TRAP 1×1010 vp p = 0.22; ChAd63 ME-TRAP 5×1010 vp p = 0.62, 2-tailed Mann Whitney test) (Figure 4B and Table 6).

ELISPOT data was combined from both trials and analysed using a multivariate linear regression model. Data was stratified by dose of ChAd63 ME-TRAP and trial site. Only dose of ChAd63 ME-TRAP had a significant effect with a 2.3 (95% CI 1.4–3.8) fold increase in mean ELISPOT response in the individuals receiving 5×1010 vp. Of note, route of administration and trial site did not have significant effects on outcome (0.8× fold increase [95% CI 0.4–1.4], p = 0.4 for IM versus ID, 1.6×fold increase [95% CI 0.9–3.0], p = 0.1 for Kenya versus The Gambia).