The study protocol was reviewed and approved by the Institutional Review Boards of the Malaria Vaccine and Drug Development Center (CECIV-MVDC) and Centro Médico Imbanaco (CMI # 0992304-493-26202) (Appendix 1). The study complied with the Declaration of Helsinki principles, International Conference on Harmonization, Good Clinical Practices guidelines, and all pertinent Colombian regulations. All participants provided written informed consent (IC) and were advised that they were free to withdraw from the study at any time. Volunteers were excluded if they had diseases or medical conditions that would alter the vaccine’s assessment or any condition that could increase the risk of adverse outcomes.
Study design and participants
This was a phase IIa/b randomized, double-blind, comparative, controlled trial to evaluate the protective efficacy and safety of P. vivax CS protein formulated in Montanide ISA-51 adjuvant. Thirty-five healthy, Duffy positive (Fy+) men and non-pregnant women (19-44 years of age) were recruited from a larger group (n=121) and allocated into two groups: healthy malaria-naïve (n=17) and malaria semi-immune (n=18) volunteers previously exposed to P. vivax. Participants were randomly (simple) assigned in a 2:1 ratio. A blinded data manager controlled the allocation to receive the vaccine (Experimental; Exp, n=25) or placebo (Control; Ctrl, n=10) (Figure 1). Access to the randomization code was strictly controlled at the pharmacy. The naïve group was further divided into Exp (n=12) and Ctrl (n=5) and the semi-immune group into Exp (n=13) and Ctrl (n=5). Naïve volunteers were from Cali (Colombia), a non-malaria endemic city, with eligibility based on no history of malaria and negative serology against P. vivax blood stages. Semi-immune volunteers were recruited from Buenaventura, a low to moderate malaria-endemic area of Colombia; eligible volunteers should have had a history of malaria and antibodies against P. vivax blood stages with titers higher than 1:20 by IFAT or higher than 1:200 by ELISA using a recombinant P. vivax MSP1 (Pv200L) protein .
The three LSP (N, R, and C) synthesized under good laboratory practices (GLP) conditions at the Biochemistry Institute, University of Lausanne, Switzerland, were packaged, lyophilized, and then tested for sterility and apyrogenicity. As previously described , the N polypeptide corresponded to N-terminal amino acids (aa) 20-96, the C peptide to C-terminal aa 301-372. In contrast, the R peptide VK210 (type I) corresponded to a construct based on the first central non-peptide repeat (aa 96-104) in tandem three times and collinearly linked to a universal T-cell epitope (ptt-30) derived from tetanus toxin [22-24]. A 1:1:1 peptide mixture (50μg/each peptide) was emulsified in Montanide ISA-51 (Seppic Inc., Paris, France) in the same proportion according to manufacturer recommendations on the same day of subject immunizations. Saline solution (Baxter, Deerfield, IL) was emulsified with the same adjuvant and used as a placebo.
The primary outcome measure was to assess the P. vivax CS LSP vaccine's protective efficacy to the P. vivax CHMI in malaria-naïve and semi-immune volunteers, and the secondary outcome, the immune response associated with protection. Eligible participants were enrolled to receive three doses of vaccines at months 0, 2, and 6 containing the mixture of LSP (150μg/dose) or placebo, formulated in the Montanide ISA-51 adjuvant by i.m. injection in the deltoid muscle with a volume of 0.5mL. The first immunization dose contained a mixture of peptides N and C only (50μg/peptide; total dose 100μg/dose), whereas for immunizations second and third, the doses comprised peptides N, R and C (50μg peptide/dose; a total of 150μg protein/dose). Control groups were immunized with the placebo (saline solution) formulated in the same adjuvant’s total volume. Vaccines were prepared by staff researchers not involved with patient care.
During the hour following immunization, volunteers were under direct medical supervision to detect any adverse reaction to the vaccine injection, after which a physical examination was performed. Eight hours after vaccination, volunteers’ physical status was assessed by a telephone call. Also, a personal follow-up was performed one week before the following immunizations. Clinical laboratory tests were performed to evaluate vaccine tolerability and safety at months 0, 1, 2, 3, 6, 7, 9, and 10. Blood samples were collected to determine Fy blood group, G6PD deficiency test, complete blood count (CBC), prothrombin time (PT), partial thromboplastin time (PTT), alanine aminotransferase (ALT), aspartate aminotransferase (AST), bilirubin, alkaline phosphatase, blood urea nitrogen (BUN), creatinine, and a pregnancy test in women.
Volunteers were also under observation for one hour after the CHMI and then by phone monitoring eight hours after and once a day until day four. Volunteers were then evaluated daily for clinical manifestations and microscopic patent parasitemia from days five to 30 after challenge and later on, every second day until day 60. Two experienced, independent microscopists evaluated parasitemia by counting the number of asexual P. vivax parasites per 400 white blood cells (WBC), assuming normal WBC counts (8000 cell/mL). Samples were considered negative after observing 200 microscopic fields, and qPCR was performed subsequently for retrospective analysis. Adverse Events (AE) were recorded, graded and classified according to FDA recommendations .
Whole blood (15mL) was collected by venipuncture (Vacutainer tubes, Becton Dickinson, NJ, USA) from patients diagnosed with VK210 P. vivax in Leticia, Colombia, and used to infect colonized Anopheles albimanus mosquitoes. Fed mosquito batches were dissected and microscopically examined for the presence of oocysts in the midgut (day 7) and sporozoites in salivary glands (day 14) as previously described . Only batches with >60% sporozoite infection rates were considered acceptable for CHMI to evaluate the protective efficacy. CHMI was performed three months after the last immunization by volunteer’s exposure to 2-5 P. vivax infected-mosquito bites, as previously described [15,16]. After human-biting, each mosquito's dissection confirmed blood in midguts and sporozoites in salivary glands .
Antibody response was assessed using blood samples collected on months 0, 1, 2, 3, 6, 7 and 10 and measured by enzyme-linked immunosorbent assay (ELISA) using as antigen the N, R, or C peptides (1μg/mL), as described previously . Controls were selected from a pool of sera from semi-immune blood donors (positive ctrl) and a pool of sera from malaria-naïve donors (negative ctrl). Also, peptide-specific IgG isotypes were determined using sera collected at months 0, 7, and 10 from immunized volunteers by ELISA ; titers <1:200 were considered negative. In addition, anti-LSP antibodies' parasite recognition was determined at 0-, 7-, and 10-months using an immunofluorescence assay (IFAT) as described before .
IFN-g ELIspot production
Peripheral blood mononuclear cells (PBMC) collected at months 0, 1, 2, 3, 6, 7, and 10 were separated from whole blood using Ficoll-Histopaque (Sigma-Aldrich, St. Louis, MO) density gradients and used to determine the IFN-γ-producing cells as previously described . Fresh PBMCs (4×105 /well) were mixed with 10 μg/mL of each LSP, and after 40 h culture, spots were counted with an ELISpot reader (AID Autoimmun Diagnostika GmbH, Germany), and the results expressed as the mean number of IFN-γ spot-forming cells (sfc) per 106 PBMC. Volunteers were considered responders if the number of sfc in their samples have increased from their baseline level (before immunization on day 0); any increase > 5 sfc was considered positive .
Data were collected and managed using REDCap (Nashville, TN, USA) electronic data capture tools, analyzed using SPSS version 16.0 software (SPSS Inc., Chicago, IL, USA), and plotted using Graph Pad Prism version 6.0 (GraphPad Software, San Diego, California, USA). The main outcome evaluated the frequency of P. vivax infection in volunteers vaccinated with PvCS LSP formulated in Montanide ISA-51. The sample size was calculated with a confidence level of 95%, an error of 5% and an estimated prevalence of the Fy+ genotype of 30% with a population census of 5,603 in afro-descendent subjects from Buenaventura and 78% from Cali . All naïve and semi-immune participants were Fy+. Immunization’s protective efficacy was assessed at a 5% significance level and 80% power. Nominal variables were analyzed using descriptive statistics. Mann-Whitney U or the Kruskal-Wallis tests were used as needed. Fisher's exact test was used to compare proportions. Spearman’s rank correlation was used to assess the correlation between numeric variables. Incubation and prepatent periods were determined by microscopy and qPCR and visualized using Kaplan-Meier estimator. Antibody titers and IFN-g production were compared among groups at several points in the study using Wilcoxon signed-rank and the Kruskal-Wallis test. Two-tailed, non-parametric p values ≤ 0.05 were considered significant.