Study design

A randomized, open-label clinical trial was designed to standardize a P. vivax sporozoite challenge model in malaria-naïve volunteers, specifically, the relationship between the number of infectious mosquito bites and the likelihood of developing patent parasitemia detectable by Giemsa-stained thick blood smear (TBS). The study was divided into two steps. Step A was designed to produce mature P. vivax sporozoites suitable for inoculation into humans; and Step B was designed to assess the safety and reproducibility of the sporozoite challenge. For Step A, patients with infective P. vivax gametocytes detected by TBS were recruited from the outpatient clinics at the Immunology Institute (IDIV) in Cali and Buenaventura, Colombia. Patients donated 35 mL of whole blood, which was screened for co-infections that could potentially represent a threat to the health of volunteers. The blood was then artificially fed to Anopheles mosquitoes.¹⁴ For Step B, malaria-naive subjects were exposed to the bites of 3 ± 1, 6 ± 1, or 9 ± 1 infected mosquitoes to determine the prepatent period of the infection and its reproducibility.

Blood donation and blood quality assurance

The EDTA vacutainer tubes (Becton Dickinson, Franklin Lakes, NJ) were used to collect 35 mL of whole blood from each infected patient, which were divided into two aliquots: a 15 mL sample for mosquito feeding and a 20 mL sample that was transported to the blood bank of the Valle del Lili Clinic for routine screening for a panel of common infectious agents (viral, bacterial, parasitic), including confirmation of Plasmodium species ( P. vivax, P. falciparum, P. malariae) by polymerase chain reaction (PCR) (Table 2).¹⁵ The rationale behind the blood screening and direct mosquito feeding on human volunteers was based on two premises: 1) standard blood bank screening is universally accepted as a safety procedure that ensures that no pathogens are transmitted from human-to-human during a blood transfusion. Therefore, blood samples determined to be negative for pathogens other than P. vivax by standard blood bank screening should be safe enough to be transferred from human-to-human through a mosquito bite. 2) Anopheles mosquitoes have not been reported to trans mit any pathogen other than Plasmodium. Before the study, we carried out an extensive bibliographic search and consulted with U.S. and Colombian experts on virology, parasitology, and ento mology about potential pathogens that might be trans mitted by Anopheles and could represent a threat for the volunteers exposed to challenge. There was a consensus that as yet no other pathogen had been reported to be transmitted by Anopheles mosquitoes, including those comprised in the blood bank screening. However, mosquitoes were discarded using a biosafety procedure if after feeding blood was confirmed to have any co-infection.

Mosquito infection

Anopheles albimanus mosquitoes Buenaventura strain, have been reared under laboratory conditions at IDIV in Cali for about 10 years. This mosquito colony has been successfully used to regularly produce sporozoites^14,16 and for studies on malaria transmission-blocking immunity.¹⁷ Here, mosquitoes were fed within 2–3 hours after the P. vivax-infected blood (gametocytemia ≥ 0.1%) was collected in Buenaventura. The 15 mL blood was centrifuged at 500 × g for 5 minutes at room temperature, plasma was removed, and cells were washed once with RPMI 1640 medium (Gibco Cell Culture Systems, Invitrogen, Grand Island, NY ). Parasitized erythrocytes were reconstituted to 50% hematocrit with pooled human AB non-immune, complement-inactivated serum obtained from the Red Cross blood bank and were used to infect lots of 4,000 female mosquitoes (3–4 days old) using a water-jacketed membrane apparatus at 37°C, described previously.¹⁴ Infected mosquitoes were maintained under strict biosafety conditions at 27 ± 1°C and relative humidity of 82% and a sugar solution supplemented with 0.05% para-aminobenzoic-acid was provided.¹⁸ Samples of fed mosquitoes were dissected and microscopically examined 7–8 days after the blood meal to determine the presence of oocysts in the midgut, and on Days 14 and 15 to assess the presence of sporozoites in salivary glands.¹⁹ Mosquito infections were graded as 1+ (1–10 spz), 2+ (11–100 spz), 3+ (101–1000 spz), and 4+ (> 1001 spz).⁷

Sporozoite challenge

The proposed infective biting dose was to be 3, 6, and 9 mosquitoes, respectively. However, a range of plus or minus one bite per dose was allowed. Therefore, screen-meshed boxes (7 × 7 × 7 cm) were filled with 4, 7, and 10 mosquitoes, to have a better chance of achieving the targeted mosquito dose in a single biting round. Participants who did not complete the minimal targeted dose were subjected to a second round of biting. Eighteen malaria-naïve volunteers were randomly assigned to one of three groups (N = 6) and were exposed by group, from the fewest (3 bites) to the greatest (9 bites) number of bites. Sporozoite challenge was carried out under strict adherence to experimental protocol in a secure room in the entomology unit at the IDIV. Volunteers were asked not to use any topical chemicals (e.g., soap, deodorant, perfume) that could affect mosquito feeding. Mosquitoes were allowed to bite the flexor side of the forearm for a 10 to 15-minute period, previously determined to be sufficient for full An. albimanus engorgement. After biting, all mosquitoes were dissected to confirm the presence of blood meal and sporozoites. Mosquitoes were considered positive if any number of sporozoites (> 1) was detected microscopically. Study participants were observed for 1 hour at the entomology unit, followed up by phone 8 hours later, and examined again 24 hours thereafter to assess the response to challenge.

Malaria diagnosis and patient follow-up

From Day 7 post-challenge onward, volunteers had daily follow-up visits during which symptoms and signs of malaria were assessed and blood was collected for TBS and Plasmodium PCR (the latter performed retrospectively).

The TBS were performed using 50 μL of whole blood collected by finger prick that were spread over a rectangular area of approximately 1.5 sq cm and were stained with freshly prepared 10% Giemsa stain.²⁰ A total of 300 microscopic fields were examined before a slide was considered negative. When TBS were found positive, parasitemia was quantified by estimating the number of parasites counted in presence of 300 leukocytes and calculating absolute parasitemia according the leukocyte counts per μL. Smears were read by experienced microscropists and a random sample of TBS was subjected to quality control by a microscopist from the malaria control program.

Participants were treated with anti-malarial drugs as soon as parasites were detected by TBS. Treatment consisted of chloroquine (1,500 mg chloroquine base provided orally in divided doses: 600 mg initially followed by 450 mg given 24 and 48 hours later) and primaquine (two 15 mg doses given once per day for 14 days). Volunteers were provided with intravenous fluids, analgesics, and antiemetics as needed. Signs and symptoms consistent with malaria were assessed by clinical examination and were graded 1–5 according to their severity following the National Cancer Institiute (NCI) criteria for adverse events (CTCAE ) as follows: Grade 1 = Low, Grade 2 = Moderate, Grade 3 = Severe, Grade 4 = Serious, and Grade 5 = Death. Pruritus was graded from 1 to 3 (1 = mild or localized; 2 = intense or widespread; and 3 = intense or widespread and interfering with activities of daily living). Because there is no common terminology AE agreement, for edema and erythema following mosquito bites, an adaptation from the CTCAE v 3 for “injection site/extravasation changes” was used as follows: Grade 1 = local pain/edema and erythema; Grade 2 = pain or swelling with inflammation or phlebitis; Grade 3 = ulceration or necrosis. Pain was scored as Grade 1 = mild not interfering with function; Grade 2 = moderate pain (pain or analgesics interfering with function, but not interfering with activities of daily living); Grade 3 = severe pain (pain or analgesics severely interfering with activities of daily living).

Clinical laboratory tests

Clinical laboratory screening in -cluding an electrocardiogram confirmed the health status of volunteers 1 month before challenge and after the study was completed (Table 1). Any abnormality found in the electrocardiogram was considered as an exclusion criterion. Tests for hemoglobin, white blood cell count, platelet count, and total bilirubin were performed again on Days 11 and 53 post-challenge.

Mosquito infection

The 15 patients provided written informed consent and served as donors of infected blood. Whole blood samples were examined using standard blood bank screening for co-infections; of those a total of four presented co-infections such as P. falciparum, Hepatitis B, and Hepatitis C and were discarded (Table 3). Blood used for mosquito feeding had parasite densities ranging from 80–8,720 (mean 3,693) P. vivax asexual erythrocytic stage parasites/μL and gametocytemia ranging from 40–3,360 (mean 934) parasites/μL.

Four mosquito lots were discarded because of co-infection in blood used for feeding. One parasite donor had a mixed P. vivax/P. falciparum infection and three blood samples carried the hepatitis B or hepatitis C virus; patients were treated and referred for medical follow-up. Six of the remaining 11 lots of mosquitoes showed infections with > 50% of the mosquitoes having sporozoites in salivary glands, whereas the other five were considered ineligible for challenge because of low percentage (< 50%) of sporozoite-infected mosquitoes. On the day before challenge, two mosquito lots examined microscopically showed sporozoite rates of 97% and 86%, respectively. The 97% lot (3 + sporozoite score in salivary glands) was selected for use.

Sporozoite challenge

Sixteen volunteers were exposed to infection during the first day and the other two (from Group C), were challenged on Day 2. The number of mosquitoes required to complete the biting dose and the number of cycles required for each dose is indicated in Table 4. Some individuals developed minor, transient discomfort that was not associated with the number of bites. Local pruritus was observed most frequently in Group B (four of six volunteers) but disappeared within 2 days. One volunteer from Group C developed mild arm pain and localized pruritus that lasted for 5 days (Table 5).

Prepatent period and clinical follow-up

Seventeen volunteers developed malaria as confirmed by TBS and PCR, and one volunteer subjected to eight mosquito bites was determined negative for malaria in a total of 20 thick smears and 15 PCR performed during the 30-day follow-up period. We believe that this volunteer who was a paramedic, surreptitiously used anti-malarials; however, as serum levels of anti-malarial drugs were not measured, we are not able to make an unequivocal confirmation. The latter (volunteer no. 213) was given anti-malarial treatment on Day 30 and was followed up as per protocol (Table 4). Patent parasitemia was confirmed between 9 and 13 days (geometric mean 11 ± 2 days) after the mosquito bites without any correlation between the challenge dose and the prepatent periods, which were 11.1 days, confidence interval (CI) 95% (9.3–11), for Group A; 10.8 days, CI 95% (9.8–11.9), for Group B; and 10.6 days, CI 95% (8.7–12.4), for Group C. No statistically significant difference was observed in the prepatent period of the different groups (Kruskal-Wallis P = 0.70) (Table 1). In five cases, parasites were detected earlier by PCR than by TBS, whereas in 11 cases both methods were equally sensitive. In one case, parasites were detected earlier by TBS than by PCR. Parasitemia ranged from 75 to 420 parasites/μL (Table 4).

All 17 infected-volunteers started to develop signs and symptoms consistent with malaria on Day 9 post-challenge, but some of the clinical manifestations appeared only 4 to 5 days later. The most frequent were chills, fever, headache, arthralgia, myalgia, and malaise followed by weakness and nausea (Table 6). Fever (axillary temperature > 38°C) was documented in 10 of the volunteers and six more reported having had fever without confirmation. All individuals cleared parasitemia between 24 and 48 hours after initiating anti-malarial treatment, and all successfully recovered within 2 to 3 days without any severe or serious adverse events (Table 6). No parasites including gametocytes were observed after treatment. The most frequent symptoms associated with treatment were epigastralgia with duration of 1 to 21 days; nausea, 1–5 days; and vomiting, 1–3 days.

Table 6

Symptoms and signs related with P. vivax infection

	A			B			C
Adverse events	Frequency %	Initial day*	Duration (hour)†	Frequency %	Initial Day*	Duration (hour)†	Frequency %	Initial day*	Duration (hour)†
Associated to infection†
Symptoms
Myalgias	6/6	9	74.4	6/6	10	43.2	5/5	9	57.6
Chills	4/8	9	64.8	6/6	10	50.4	5/5	9	48.0
Headache	6/6	9	108.0	6/6	9	60.0	5/5	9	67.2
Malaise	5/6	9	76.8	6/6	10	62.4	5/5	9	52.8
Weakness	5/6	10	62.4	5/6	10	81.6	4/5	10	52.8
Signs
Temperature > 38°C	2/6	9	48.0	4/6	10	28.8	3/5	9	55.2
Jaundice	0/6	-	-	3/6	12	48.0	0/6	-	-
Tachycardia	1/6	9	24.0	0/6	-	-	1/5	9	24.0
Splenomegaly	0/6	-	-	0/6	-	-	1/5	10	24.0
Laboratories
Lymphopenia	5/6	10	NA§	1/6	14	NA	2/5	10	NA
Thrombocytopenia	1/6	13	NA	NA	-	-	NA	-	-
GPT increase	1/6	11	NA	2/6	14	NA	2/5	12	NA
Hyperbilirrubinemia	-	-	-	-	-	-	1/5	10	NA
BUN¶	2/6	53	NA	-	-	-	-	-	-
Associated with treatment‡
Abdominal pain	2/6	17	96.0	0/6	-	-	2/5	1	24.0
Dizziness	3/6	15	134.4	2/6	15	72.0	3/5	16	72.0
Nausea	2/6	16	67.2	4/6	15	96.0	1/6	15	24.0
Epigastric pain	2/6	22	192.0	4/6	16	102.0	3/5	15	247.2
Vomiting	2/6	18	24.0	2/6	16	48.0	0/6	-	-
Blurred vision	4/6	19	60.0	1/6	15	48.0	0/5	-	-

Open in a separate window

^*Days after parasite challenge, expressed as geometric mean.

^†Geometric mean.

^‡Days after treatment initiation expressed as geometric mean.

^§NA = not applied.

GPT = glutamic pyruvic transaminase.

^¶BUN = blood urea nitrogen.

In addition, five patients developed blurred vision that lasted for 2 to 3 days after treatment initiation (Table 6). None of the volunteers required hospitalization, but because of nausea and vomiting, seven required fluid therapy that was administered at the MVDC outpatient clinic. The volunteer who did not become infected was treated with the anti-malarial protocol after a month of follow-up.

The investigators express their sincere gratitude to the volunteers who participated in this study. We also thank the Fundación Clínica Valle del Lili (FCVL) for the services of their blood bank and emergency room. We wish to give special acknowledgment to the team at MVDC that supported the work for this study, particularly A. C. Londoño, F. Zamora, L. Acuna, A. Jordan, A. Rincon, and F. Yasnot.

Financial support: This work was supported by World Health Organization Initiative for Vaccine Research (grant no. LA35735G), National Institute of Allergy and Infectious Diseases (NIAID grant no. AI49486-05/TMRC), Colombian National Research Council, COLCIENCIAS, and the Malaria Vaccine and Drug Development Center Foundation. The contribution of U.S. Navy staff was supported by Work Unit Number 6000.RAD1.F.A309.