A randomized feasibility trial comparing four antimalarial drug regimens to induce Plasmodium falciparum gametocytemia in the controlled human malaria infection model

Isaie J Reuling; Lisanne A van de Schans; Luc E Coffeng; Kjerstin Lanke; Lisette Meerstein-Kessel; Wouter Graumans; Geert-Jan van Gemert; Karina Teelen; Rianne Siebelink-Stoter; Marga van de Vegte-Bolmer; Quirijn de Mast; André J van der Ven; Karen Ivinson; Cornelus C Hermsen; Sake de Vlas; John Bradley; Katharine A Collins; Christian F Ockenhouse; James McCarthy; Robert W Sauerwein; Teun Bousema

doi:10.7554/eLife.31549

. 2018 Feb 27;7:e31549. doi: 10.7554/eLife.31549

A randomized feasibility trial comparing four antimalarial drug regimens to induce Plasmodium falciparum gametocytemia in the controlled human malaria infection model

Isaie J Reuling ¹, Lisanne A van de Schans ¹, Luc E Coffeng ², Kjerstin Lanke ¹, Lisette Meerstein-Kessel ¹, Wouter Graumans ¹, Geert-Jan van Gemert ¹, Karina Teelen ¹, Rianne Siebelink-Stoter ¹, Marga van de Vegte-Bolmer ¹, Quirijn de Mast ³, André J van der Ven ³, Karen Ivinson ⁴, Cornelus C Hermsen ¹, Sake de Vlas ², John Bradley ⁵, Katharine A Collins ⁶, Christian F Ockenhouse ⁴, James McCarthy ⁶, Robert W Sauerwein ^1,^†, Teun Bousema ^1,^†,^✉

Editor: Ben Cooper⁷

PMCID: PMC5828662 PMID: 29482720

Abstract

Background:

Malaria elimination strategies require a thorough understanding of parasite transmission from human to mosquito. A clinical model to induce gametocytes to understand their dynamics and evaluate transmission-blocking interventions (TBI) is currently unavailable. Here, we explore the use of the well-established Controlled Human Malaria Infection model (CHMI) to induce gametocyte carriage with different antimalarial drug regimens.

Methods:

In a single centre, open-label randomised trial, healthy malaria-naive participants (aged 18–35 years) were infected with Plasmodium falciparum by bites of infected Anopheles mosquitoes. Participants were randomly allocated to four different treatment arms (n = 4 per arm) comprising low-dose (LD) piperaquine (PIP) or sulfadoxine-pyrimethamine (SP), followed by a curative regimen upon recrudescence. Male and female gametocyte densities were determined by molecular assays.

Results:

Mature gametocytes were observed in all participants (16/16, 100%). Gametocytes appeared 8.5–12 days after the first detection of asexual parasites. Peak gametocyte densities and gametocyte burden was highest in the LD-PIP/SP arm, and associated with the preceding asexual parasite biomass (p=0.026). Male gametocytes had a mean estimated circulation time of 2.7 days (95% CI 1.5–3.9) compared to 5.1 days (95% CI 4.1–6.1) for female gametocytes. Exploratory mosquito feeding assays showed successful sporadic mosquito infections. There were no serious adverse events or significant differences in the occurrence and severity of adverse events between study arms (p=0.49 and p=0.28).

Conclusions:

The early appearance of gametocytes indicates gametocyte commitment during the first wave of asexual parasites emerging from the liver. Treatment by LD-PIP followed by a curative SP regimen, results in the highest gametocyte densities and the largest number of gametocyte-positive days. This model can be used to evaluate the effect of drugs and vaccines on gametocyte dynamics, and lays the foundation for fulfilling the critical unmet need to evaluate transmission-blocking interventions against falciparum malaria for downstream selection and clinical development.

Funding:

Funded by PATH Malaria Vaccine Initiative (MVI).

Clinical trial number:

NCT02836002.

Research organism: <i>P. falciparum</i>

eLife digest

The parasite that causes malaria, named Plasmodium falciparum, has a life cycle that involves both humans and mosquitoes. Starting in the saliva of female Anopheles mosquitoes, it enters a person’s bloodstream when the insects feed. It then moves to the person’s liver, where it infects liver cells and matures into a stage known as schizonts. The schizonts then divide to form thousands of so-called merozoites, which burst out of the liver cells and into the bloodstream. The merozoites infect red blood cells, producing more schizonts and yet more merozoites, which continue the infection.

To complete its life cycle, the parasite must return to a mosquito. Some of the parasites in the person’s blood transform into male and female cells called gametocytes that are taken up by a mosquito when it feeds on that person. Inside the mosquito, male and female parasites reproduce to create the next generation of parasites. The new parasites then move to the mosquito’s salivary glands, ready to begin another infection. Stopping the parasite being transmitted from humans to mosquitoes will stop the spread of malaria in the population. Yet it has proven difficult to study this part of the life cycle from natural infections.

Here, Reuling et al. report a new method for generating gametocytes in human volunteers that will enable closer study of the biology of malaria transmission. The method is developed using the Controlled Human Malaria Infection (CHMI) model. Healthy volunteers without a history of malaria are bitten by mosquitoes infected with malaria parasites. Shortly afterwards, the volunteers are given a drug treatment to control and reduce their symptoms. The gametocytes form during this phase of the infection. At the end of the experiment, all the volunteers receive a final treatment that completely cures the infection.

Reuling et al. recruited 16 volunteers and assigned them to four groups at random. Each group received a different drug regime. Roughly a week after the mosquito bites, all participants showed malaria parasites in their blood, and between 8.5 and 12 days later, mature gametocytes started to appear. This early appearance suggests that the parasites start to transform into gametocytes when they first emerge from the liver. The experiment also revealed that female gametocytes stay in the blood for a longer period than their male counterparts.

These results are proof of principle for a new way to investigate malaria infection. The new model provides a controlled method for studying P. falciparum gametocytes in people. In the future, it could help to test the impact of drugs and vaccines on gametocytes. Understanding more about these parasites’ biology could lead to treatments that block malaria transmission.

Introduction

Malaria, a disease caused by Plasmodium parasites, continues to be a public health burden. Despite a reduction in the malaria case incidence of ~40%, and mortality by 62% over the last decade, malaria caused ~429,000 deaths in 2015 (World Health Organization, 2016). Apart from the direct health implications, malaria is a substantial contributor to ongoing poverty in affected countries. Recently, the spread of artemisinin-resistant parasites has emerged as a global health concern. Both the recent gains in malaria control and concerns about artemisinin resistance have stimulated programs to eliminate malaria (World Health Organization, 2016). Novel interventions may support malaria elimination efforts in endemic settings (Griffin et al., 2010) that are further dependent on political and financial commitments to maximize coverage with currently available interventions and improve surveillance systems to optimize disease notification and treatment (Moonen et al., 2010).

A major challenge to eliminating malaria is its highly efficient transmission by Anopheles mosquitoes. Transmission to mosquitoes starts when a small proportion of asexual parasites commit to form male and female gametocytes. It is currently unclear what stimulates gametocyte commitment and when gametocyte commitment first occurs (Nilsson et al., 2015). Upon commitment, maturation of gametocytes takes place predominantly in the bone marrow, and requires 7 days (range 4–12) of development. (Eichner et al., 2001) Subsequently, mature gametocytes (parasites that are not associated with clinical disease) appear in the peripheral blood, where they may circulate for an average of 6 days (Eichner et al., 2001; Bousema et al., 2010). During this period, blood-feeding Anophelines may ingest gametocytes where, after a sporogonic development phase, sporozoites reach the mosquito salivary gland rendering the mosquito infectious to humans upon its next bite. Early work based on the microscopic evaluation of experimental P. falciparum infection (malariatherapy) studies reported that gametocytes may make their appearance in small numbers around 10 days following the first day of fever (Shute and Maryon, 1951; Ciuca et al., 1937).

The renewed focus on malaria elimination requires a thorough understanding of malaria transmission dynamics - when mature male and female gametocytes are first produced upon infection and how long they circulate in peripheral blood (Sinden, 2017). These parameters are difficult to measure in naturally acquired infections where frequent super-infections, immunity and other factors dictate parasite and gametocyte dynamics (Bousema and Drakeley, 2011). Interventions that specifically aim to reduce gametocyte development, circulation time or infectivity are highly desirable in the context of malaria elimination and require effective models for the early clinical evaluation.

The controlled human malaria infection (CHMI) model allows the induction of parasitemia under highly standardized conditions and plays an important role in the assessment of safety and efficacy of novel antimalarial drugs and vaccines (Sauerwein et al., 2011). Preliminary evidence for the induction of female gametocytes in CHMI studies with blood stage inoculum was recently demonstrated using piperaquine monotherapy (Pasay et al., 2016; Farid et al., 2017).

In this study, we aimed to develop a CHMI transmission model to induce gametocyte carriage after mosquito bite infection. The primary objective of the current trial was to safely induce gametocytemia in study participants by the use of different (sub)curative drug regimens based on sulfadoxine-pyrimethamine (Bousema and Drakeley, 2011; Butcher, 1997) and piperaquine (Adjalley et al., 2011).

Results

From a total of 49 screened candidate participants, 16 volunteers were included in a first cohort and randomly assigned to four study arms prior to challenge (Figure 1). After observed transient liver enzyme elevations in the first cohort, the study was temporarily put on hold and the already initiated infections in the second cohort of 13 participants were abrogated by curative treatment on day 3 post challenge. The hold was lifted after reviewing safety data. Participants from the first cohort completed all study visits, and form the basis of the current manuscript; their baseline characteristics are shown in Table 1. After exposure to bites of a standard protocol of five P. falciparum infected mosquitoes, all participants developed parasitemia on days 6.5–12 post-challenge; peak parasite densities ranged from 1050 to 63113 Pf/mL (Figure 2; Figure 2—figure supplement 1; Table 2; Supplementary file 1). Due to asexual recrudescence in seven of the eight participants after a subcurative treatment (T1) with LD-PIP, a curative treatment (T2) had to be administered before day 21 post challenge. The median period between T1 and T2 was 9.1 (range of 7.7–11.7), 10.0 (range of 9.2–10.2), 4.7 (range of 2–10.7), and 2.5 (range of 1.5–5.0) days for study arms LD-SP/SP, LD-SP/PIP, LD-PIP/PIP, and LD-PIP/SP, respectively. In participants receiving a subcurative LD-SP as T1, no recrudescent infection occurred and T2 was initiated on day 21 per protocol. One participant from treatment arm LD-PIP/PIP developed asexual recrudescence after T2, and received end treatment with atovaquone/proguanil on day 36. The remaining participants did not develop recrudescent infections after T2, and were treated with atovaquone/proguanil on day 42 as per protocol.

Figure 1. — ECG = electrocardiography, BMI = body mass index, AST = aspartate aminotransferase, ALP = alkaline phosphatase

Table 1. Baseline characteristics of the participants included in analysis.

Table 1—source data 1. Source data for Table 1.

elife-31549-table1-data1.xlsx^{(10.6KB, xlsx)}

		LD-SP/SP	LD-SP/PIP	LD-PIP/PIP	LD-PIP/SP
No. subjects		n = 4	n = 4	n = 4	n = 4
Treatment 1 (T1)		Sulfadoxine-pyrimethamine 500 mg/25 mg	Sulfadoxine-pyrimethamine 500 mg/25 mg	Piperaquine 480 mg	Piperaquine 480 mg
Treatment 2 (T2)		Sulfadoxine-pyrimethamine 1000 mg/50 mg	Piperaquine 960 mg	Piperaquine 960 mg	Sulfadoxine-pyrimethamine 1000 mg/50 mg
Sex
Male	n (%)	2 (50%)	0 (0%)	1 (25%)	1 (25%)
Female	n (%)	2 (50%)	4 (100%)	3 (75%)	3 (75%)
Age	Mean (range)	24.5 (21–29)	24 (21–28)	21.5 (20–24)	22.5 (20–27)
BMI (kg/m²)	Mean (range)	21 (18–23)	22 (19–25)	24.5 (21–27)	26.5 (24–29)

Open in a new tab

Figure 2. — Black line represents 18S qPCR asexual parasitemia. Black dotted-line represents 18S qPCR after treatment 1. Red line represents Pfs25 qRT-PCR gametocytemia.

Figure 2—figure supplement 1. — Black line represents 18S qPCR asexual parasitemia. Black dotted-line represents 18S qPCR after treatment 1. Red line represents Pfs25 qRT-PCR gametocytemia.

Table 2. Treatment and parasitological data per study group.

		LD-SP/SP	LD-SP/PIP	LD-PIP/PIP	LD-PIP/SP
Time to T1 (days)	Median (range)	13 (9.3–12.8)	10.8 (0.8–11.8)	10.3 (10.3–12.3)	12.8 (12.3–14.3)
Time between T1-T2 (days)	Median (range)	9.1 (7.7–11.7)	10 (9.2–10.2)	4.7 (2–10.7)	2.5 (1.5–5.0)
Area under the curve (AUC)*	Median (range)
Asexual		6490 (1120–16337)	13280 (2773–43777)	14347 (5408–24898)	12747 (4572–82973)
Sexual		280 (27–3640)	271 (64–848)	784 (316–1274)	6624 (1515–10244)
Peak parasite density (Pf/mL)	Median (range)	6467 (1050–20261)	16376 (2590–50210)	11603 (2408–21565)	8491 (3976–63113)
Peak gametocyte density (gct/mL)	Median (range)	38 (11–368)	30 (13–101)	83 (46–99)	627 (199–1285)
Day of gametocyte detection after infection (days)	Mean (SD)	18.3 (1.0)	18.5 (1.0)	17.3 (1.5)	19.4 (1.3)
Time to gametocyte detection relative to first asexual parasites^† (days)	Mean (SD)	10.5 (1.3)	11.5 (1.0)	10.1 (1.3)	10.1 (1.2)
Proportion of days gametocyte positive (%)^‡	Mean (SD)	27.4 (6.7)	35.9 (7.6)	51.4 (7.9)	48.3 (8.1)
Duration gametocytemia^§ (days)	Median (range)	7.5 (1–24)	6 (2–14)	17 (12–25)	24.5 (17–25)

Open in a new tab

^*The area under the curve (AUC) represents the total parasite exposure over time (asexual- or sexual parasite load).

^†Time to gametocyte detection is calculated as the day of the detection of gametocytes (≥5 gct/mL) minus the day of first peak asexual parsitaemia.

^‡The proportion of gametocyte positive days is calculated as all days with ≥5 gct/mL by Pfs25-qRT-PCR divided by all days where Pfs25 qRT-PCR was performed.

^§Maximum number of consecutive days of Pfs25 qRT-PCR measured gametocytemia ≥5 gct/mL.

All participants also developed gametocytemia as determined by Pfs25 qRT-PCR (Figure 2; Figure 3A; Figure 2—figure supplement 1). Gametocytes were first detected 8.5–12 days after the initial peak of asexual parasites with no statistically significant difference in time to gametocyte appearance between study arms (p=0.26) (Table 2). The median peak density of gametocytes was 83 gametocytes/mL (range 11–1285) when all study participants were considered. Peak gametocyte densities were higher in the study arm randomised to LD-PIP/SP, with a median of 627 gametocytes/mL (range of 199–1285), compared to 38 gametocytes/mL (range of 11–368), 30 gametocytes/mL (range of 13–101), 83 gametocytes/mL (range of 46–99), for arms LD-SP/SP, LD-SP/PIP, and LD-PIP/PIP, respectively (Figure 2; Figure 2—figure supplement 1; Table 2).

Thirteen (81%, 13/16) participants showed gametocytes on at least 5 consecutive days. The mean number of consecutive gametocyte-positive days was 24.5 (range of 17–25) for the LD-PIP/SP arm and was higher than for other arms (Table 2; Figure 2). Using multi-level logistic regression (random effect for within-group variation), we estimated that the average proportion of days that individuals tested positive for gametocytes was 27.4% (LD-SP/SP), 35.9% (LD-SP/PIP), 51.4% (LD-PIP/PIP), and 48.3% (LD-PIP/SP) (Table 2). The LD-PIP/PIP and LD-PIP/SP arms (i.e. those receiving ‘low dose PIP’) each had significantly higher average proportions of gametocyte-positive days than both arms LD-SP/SP and LD-SP/PIP (posterior probability 90.8% and 86.1%, respectively; 81.1% joint probability of arms LD-PIP/PIP and LD-PIP/SP both being higher than both LD-SP/SP and LD-SP/PIP). Furthermore, the area under the curve (AUC) for gametocyte density showed a statistically significant difference between arms (p=0.04). The LD-PIP/SP arm had a significantly higher gametocyte load (area under the curve) than each of the other three treatment arms (94.4% posterior probability of being the highest; Figure 3B). After correction for the asexual AUC, the probabilities of the gametocyte AUC in the LD-PIP/SP arm being higher than the other three decreased to 97.2%, 96.3%, and 96.2% (from 99,1%, 98.9%, and 95.4%), and the probability of LD-PIP/SP being higher than all other study arms decreased to 94.0%.

Both female and male gametocytes were detected in 14/16 (88%) participants (Figure 4; Figure 4—figure supplement 1). Gametocyte sex-ratio’s and circulation times have to be interpreted with caution since they rely on two separate qRT-PCR assays with differences in assay sensitivity (Figure 5; Supplementary file 2, 3). On average 2.5 times as many female gametocytes were observed compared to male gametocytes per measured time-point (Figure 4; mean ratio 2.5 (SD = 2.5)). Combining all treatment arms, the best estimate of gametocyte half-life was 5.1 days (95% CI 4.1–6.1) for female gametocytes and 2.7 days (95% CI 1.5–3.9) for male gametocytes (Figure 4—figure supplement 2).

Figure 5. — Standard curves (Mean, SD) obtained using 10-fold dilutions of cultured gametocytes. The highest concentration was enumerated by two independent expert microscopists. The mean and standard deviation of 54, 28, 72 replicates of the standard curve during the study was determined for the Pfs 25-, PfMGET, and 18S target genes, respectively. For PfMGET, six points starting from 10⁶ pure male gametocytes/mL were measured. 10¹ was positive in 6/28 replicates (black dot).

Figure 5—figure supplement 1. — Standard curves (Mean, SD) obtained using 10-fold dilutions of cultured gametocytes. The highest concentration was enumerated by two independent expert microscopists. The mean and standard deviation of 54, 28, 72 replicates of the standard curve during the study was determined for the Pfs 25-, PfMGET, and 18S target genes, respectively. For PfMGET, six points starting from 10⁶ pure male gametocytes/mL were measured. 10¹ was positive in 6/28 replicates (black dot).

Gametocytes are produced from their asexual progenitors, and hence asexual parasite kinetics and gametocyte kinetics are related. The AUC of asexual parasitemia was statistically significantly associated with the AUC of gametocytemia (r² = 0.31, p=0.026), as shown in Figure 3C. The mean time-window between the first asexual parasites and the first appearance of gametocytes was 10.6 (SD = 0.65) days, see Table 2. Membrane feeding experiments were performed as an exploratory objective, and confirmed infectivity of gametocytes in three mosquitoes from three study arms on days 25 (LD-PIP/SP and LD-SP/SP arms) and 31 (LD-SP/PIP arm) post-infection. Mean gametocyte densities at those time-points were 106 gametocytes/mL (SD = 175), and 28 gametocytes/mL (SD = 47), respectively. Expressed as a proportion of all examined mosquitoes, 0.0002% (3/14400) of mosquitoes became infected in these exploratory assessments. Possible and probable related adverse events after challenge infection are shown in Figure 6 and Table 3. The most frequently reported adverse events were fatigue, malaise, headache, fever, nausea, and chills. Grade three adverse events were reported in 14/16 (88%) participants, and were predominated by headache (n = 8), chills (n = 6), and nausea (n = 5). All possible and probable related adverse events resolved by the end of study. No serious adverse events occurred. The median number of adverse events was 20.5 per individual; the median number of adverse events with a grade three severity score was 1.5 per individual. There was no evidence for a difference between study arms in the occurrence of adverse events (p=0.49) or grade three adverse events (p=0.28).

Figure 6. — (A) Adverse events per study arm (B) Total no. of adverse events and time course.

Figure 6—figure supplement 1. — (A) Adverse events per study arm (B) Total no. of adverse events and time course.

Table 3. List of adverse events possibly or probably related to the trial.

Adverse events	Total	LD-SP/SP			LD-SP/PIP			LD-PIP/PIP			LD-PIP/SP
	Number of subjects	Number of subjects	Number of episodes	Mean duration in days (SD)	Number of subjects	Number of episodes	Mean duration in days (SD)	Number of subjects	Number of episodes	Mean duration in days (SD)	Number of subjects	Number of episodes	Mean duration in days (SD)
Fatigue, Malaise	16	4	10	3.6 (4.5)	4	15	2.0 (3.0)	4	10	2.9 (1.1)	4	6	6.8 (8.1)
Headache	15	3	12	1.0 (1.5)	4	25	1.2 (1.2)	4	17	1.3 (1.2)	4	21	1.6 (1.4))
Fever	15	4	9	0.4 (0.4)	4	10	0.3 (0.4)	3	11	0.4 (0.3)	4	13	0.7 (0.4)
Nausea	14	4	12	0.6 (0.8)	4	15	1.1 (1.6)	3	8	1.2 (1.5)	3	10	0.7 (1.0)
Chills	14	3	4	1.7 (1.0))	3	5	1.7 (2.0)	4	10	1.2 (1.3)	4	6	0.9 (1.1)
Myalgia	11	3	5	3.2 (3.3)	3	9	2.1 (1.9)	3	5	1.2 (1.0)	2	3	2.2 (2.6)
Abdominal pain	10	2	5	0.3 (0.2)	3	3	0.6 (0.9)	2	8	1.1 (1.3)	3	3	1.6 (2.4)
Pruritis	6	2	3	0.6 (0.8)	2	2	3.3 (0.5)	1	2	0.3 (0.4)	1	1	3.6
Athralgia	5	1	1	2.2	2	4	1.5 (1.8)	0	-	-	2	2	5.1 (3.6)
Diarrhoea	5	1	1	0.8	1	1	0.1	2	2	1.7 (2.1)	1	1	4
Diziness	3	1	1	0.1	0	-	-	2	5	0.5 (0.7)	0	-	-
Reflux	2	0	-	-	2	2	2.9 (1.8)	0	-	-	0	-	-
Pyrosis	1	0	-	-	0	-	-	0	-	-	1	1	8.6
Aspecific chest pain	1	1	2	0.0 (0.0)	0	-	-	0	-	-	0	-	-
Syncope	1	0	-	-	1	1	0.0	0	-	-	0	-	-
Mouth ulcera	1	1	1	10.0	0	-	-	0	-	-	0	-	-
Grade 3 adverse events
Total	14	3			4			3			4
Headache	8	0	-	-	2	2	0.3 (0.2)	2	2	0.6 (0.1)	4	4	1.1 (1.3)
Chills	6	1	1	0.9	2	2	1.7 (2.0)	2	2	0.3 (0.3)	1	1	2.2
Nausea	5	1	1	0.1	2	3	0.3 (0.6)	1	1	0.7	1	1	0
Fever	4	0	-	-	0	-	-	2	5	0.5 (0.4)	2	5	0.7 (0.5)
Fatigue, malaise	4	0	-	-	3	4	0.8 (0.4)	1	1	2	0	-	-
Abdominal pain	1	1	1	0.5	0	-	-	0	-	-	0	-	-

Open in a new tab

Laboratory abnormalities during the study are shown in Table 4. Most prevalent abnormalities were elevated transaminases (ALT/AST) (n = 16), decreased lymphocytes (n = 15), decreased neutrophils (n = 13), and decreased platelets (n = 12). The only grade three laboratory abnormalities were elevated ALT (n = 8), and elevated AST (n = 7). 16/16 (100%) volunteers showed mild to severe ALT/AST elevations. 5/16 (31%) mild (grade 1); 3/16 (19%) moderate (grade 2), and 8/16 (50%) severe (grade 3) (up to 25 x ULN) ALT/AST elevations. These derangements were transient, and returned to baseline values within the normal range before the end of the study. A detailed overview of these liver function test derangements can be found in the supporting information (Figure 6—figure supplement 1). These unexpected safety findings were reported to the Safety Monitoring Committee (SMC) and CCMO, and thoroughly reviewed.

Table 4. Laboratory abnormalities per study arm.

	LD-SP/SP			LD-SP/PIP			LD-PIP/PIP			LD-PIP/SP
	N (% of total) of grade 1	N (% of total) of grade 2	N (% of total) of grade 3	N (% of total) of grade 1	N (% of total) of grade 2	N (% of total) of grade 3	N (% of total) of grade 1	N (% of total) of grade 2	N (% of total) of grade 3	N (% of total) of grade 1	N (% of total) of grade 2	N (% of total) of grade 3
Any lab. abnormality	15 (14)	7 (7)	2 (2)	13 (12)	10 (9	3 (3)	16 (15)	9 (8)	2 (2)	13 (12)	8 (8)	8 (8)
Decreased hemoglobin	0	0	0	1 (14)	2 (29)	0	1 (14)	1 (14)	0	1 (14)	1 (14)	0
Decreased WBC	1 (8)	3 (23)	0	1 (8)	2 (15)	0	1 (8)	2 (15)	0	1 (8)	2 (15)	0
Decreased neutrophils	3 (23)	1 (8)	0	2 (15)	0	0	3 (23)	1 (8)	0	3 (23)	0	0
Decreased lymphocytes	3 (20)	1 (7)	0	1 (7)	3 (20)	0	3 (20)	1 (7)	0	1 (7)	2 (13)	0
Decreased platelets	3 (25)	0	0	2 (17)	0	0	4 (33)	0	0	1 (8)	2 (17)	0
Elevated ALT	2 (13)	1 (6)	1 (6)	2 (13)	0	2 (13)	1 (6)	2 (13)	1 (6)	0	0	4 (25)
Elevated AST	1 (7)	1 (7)	1 (7)	2 (13)	1 (7)	1 (7)	1 (7)	2 (13)	1 (7)	0	0	4 (27)
Elevated yGT	1 (11)	0	0	1 (11)	1 (11)	0	2 (22)	0	0	3 (33)	1 (11)	0
Elevated ALP	0	0	0	0	1 (33)	0	0	0	0	2 (67)	0	0
Elevated total bilirubin	1 (50)	0	0	0	0	0	0	0	0	1*(50)	0	0
Elevated creatinine	0	0	0	1 (100)	0	0	0	0	0	0	0	0
Elevated BUN	0	0	0	0	0	0	0	0	0	0	0	0

Open in a new tab

Number of subjects with the highest grade reported for a laboratory abnormality. Grading based on WHO toxicity grading scale. No grade four abnormalities were reported. Lymphocytes (10⁹/l) were graded based on grade 1: 0.9–0.6; grade 2: 0.3–0.5; grade 3:<0.3.

Liver function tests were graded based on grade 1: 1.1.–2.5X ULN, grade 2: 2.6–5.0x ULN, grade 3:>5.0X ULN. WBC, white blood count; ALT, alanine aminotransferase; AST, aspartate aminotransferase; yGT, glutamyl transpeptidase; ALP, alkaline phosphatase;.

See Figure 6—figure supplement 1 for a detailed overview of liver function test abnormalities.

BUN, blood urea nitrogen. T1, treatment 1; T2, treatment 2.*Subject showed elevated total bilirubin at baseline.

Discussion

Here, we present a CHMI model to induce mature gametocytes after mosquito bite infection in malaria-naive study participants. The timing of the first appearance of gametocytes suggests that a fraction of the first wave of asexual parasites commit to the production of male and female gametocytes. With the use of antimalarial drugs that attenuates asexual stage infections but leave (developing) gametocytes unaffected, we determined biologically plausible half-lives of male and female gametocytes, and show preliminary evidence of the potential of this model to complete the lifecycle of malaria in mosquito feeding assays.

Malaria elimination efforts require a thorough understanding of the transmissibility of infections. Gametocyte commitment occurs for a fraction of asexual parasites under regulation of the transcription factor AP2-G with the entire progeny of a sexually committed schizont forming either male or female gametocytes (Kafsack et al., 2014). Our findings, based on novel sex-specific gametocyte qRT-PCR, confirm earlier work from malariatherapy studies where gametocytes were first detected by microscopy at 9–11 days after asexual parasites (Ciuca et al., 1937; Shute and Maryon, 1951). These data indicate very early gametocyte commitment and are in line with our earlier observations that Pfs16 mRNA, the earliest gametocyte transcript, is detectable at the moment of peak parasitemia in CHMI models (Schneider et al., 2004). This timing is highly relevant for understanding gametocyte transmission biology. The circulation of mature gametocytes has not been reported in previous CHMI trials using curative regimens of chloroquine, artemether-lumefantrine, or atovaqoune-proguanil, and our data illustrate the differential impact of antimalarial drugs on developing gametocytes. Once treatment is initiated, gametocyte production ceases abruptly (in the case of artemisinins), remains unaffected, or may even be stimulated under drug pressure as suggested for sulfadoxine-pyrimethamine and piperaquine (Bousema and Drakeley, 2011; Butcher, 1997; Adjalley et al., 2011). In our study, we aimed for a protracted low density of asexual parasitemia demonstrating that early abrogation of asexual infections by both sulfadoxine-pyrimethamine and piperaquine permits successful mature gametocyte development. SP has long been associated with a rapid appearance of gametocytes that is too early to be explained by de novo gametocyte production upon drug pressure and has thus been hypothesized to reflect an efflux of sequestered gametocytes upon treatment (Butcher, 1997). Evidence for the permissiveness of piperaquine to (developing) gametocytes is more recent (Pasay et al., 2016; Farid et al., 2017; Adjalley et al., 2011). In the current study, group sizes are limited and comparisons between treatment arms have to be interpreted with caution. CHMI studies are logistically challenging and the number of volunteers that can be monitored to ensure participant safety is an important consideration when defining the study size. Our sample size calculation was based on the optimistic assumption that the vast majority of volunteers would develop mature gametocytes; an assumption that was supported by the current data. With our limited study size, our findings indicate that none of the study drugs prevented the appearance of gametocytes after treatment, thereby suggesting limited or no effect of PIP and SP on developing or mature gametocytes (Bolscher et al., 2015). We hypothesized that slow acting drugs may promote the development of gametocytes (Méndez et al., 2002), potentially via microvesicles that are derived from infected erythrocytes (Nilsson et al., 2015) and differences between drug regimens in the rate at which asexual parasites are cleared upon T1 and T2 would result in different gametocyte dynamics. Although our findings indicate highest gametocyte concentrations in the LD-PIP/SP arm, more observations and thus additional studies are needed to allow the construction of a model that allows a quantification of gametocyte commitment at different time-points during the study (e.g. prior to T1, during the phase of parasite recrudescence and following T2). One hypothesis would be similar gametocyte commitment in all arms after T1 but a more rapid release of gametocytes that accumulated in the bone marrow between T1 and T2.

We present the novel evidence that both male and female gametocytes appear early, upon infection. Our findings suggest an earlier appearance of female gametocytes (18.8 days (SD 1.8) compared to male gametocytes 20.3 days (SD 1. 2)) and a longer circulation time of female gametocytes that is in line with previous estimates from naturally infected individuals (Bousema et al., 2010; Ciuca et al., 1937). Whilst both male and female gametocytes are consistently detected at densities of 0.1 gametocyte/µL (Stone et al., 2017), the highly abundant Pfs25 mRNA makes the female gametocyte qRT-PCR more sensitive than the male PfMGET qRT-PCR. Differences in gametocyte dynamics between male and female gametocytes should therefore be interpreted with caution. Gametocyte densities remained below the threshold of detection by microscopy throughout the study period and were strongly associated with the preceding densities of the asexual progenitors. Participants in the LD-PIP/SP study arm showed the highest gametocytes densities, and a mean female/male sex ratio of 4.1 (SD = 5.1), in line with gametocyte sex-ratios in natural infections (~3 to 5 females to one male) (Ciuca et al., 1937; Delves et al., 2013). We confirmed the infectivity of gametocytes in three mosquitoes from three study arms. The very low rate of infected mosquito corroborates observations from naturally acquired infections where mosquito infection becomes highly unlikely below 1000–10,000 gametocytes/mL (Gonçalves et al., 2016). The sporadic mosquito infections thus demonstrate that mature gametocytes in sex-ratios supportive of mosquito infections can be achieved in CHMI transmission models. Studies on the evaluation of TBIs will need a further optimized protocol aimed to achieve higher gametocyte densities by increasing duration and load of the asexual parasite burden. For the evaluation of gametocytocidal interventions in the CHMI transmission model, gametocyte densities should be sufficiently high to quantify an intervention-associated reduction in gametocyte appearance or gametocyte half-life. For the evaluation of interventions that reduce the transmissibility of gametocytes, higher mosquito infections should be achieved at proportions that allow the detection of meaningful reductions in mosquito infection rates in experimental arms. Low infectivity in membrane feeding assays may be overcome by achievement of higher gametocyte densities in the model, and the use of gametocyte concentration methods (Reuling et al., 2017), or by direct skin feeding assays (Bousema et al., 2012).

In line with recent findings, we observed recrudescent infections in 7/8 participants treated with LD-PIP (Pasay et al., 2016). Recrudescent infections were not observed in arms that first received LD-SP, suggesting that this dose, although 1/3 of the standard curative dose of sulfadoxine-pyrimethamine, is curative at asexual parasite densities observed in our participants. It has been hypothesized that the prolonged parasitemia under drug pressure increases gametocyte commitment (WWARN Gametocyte Study Group, 2016). The duration of parasite multiplication between T1 and T2 was relatively short in this study (2–5 days) for subjects with recrudescent infections, and the contribution of drug pressure may thus have been limited. The current findings suggest that further lowering the SP dose may be considered to prolong asexual parasite exposure.

The liver enzyme elevations found in our study led to a structured risk analysis, and review by independent experts. Transient, asymptomatic liver function test (LFT) derangements have been reported in volunteers in previous CHMI studies, and are likely to be related to the asexual stage parasitemia, and subsequent treatment.

Detailed studies on gametocyte biology and dynamics, and the early development of novel drugs and vaccines that target malaria transmission (TBIs) are currently restricted to in vitro assays, such as drug sensitivity assays, and standard membrane feeding assays (SMFA) (Bousema and Drakeley, 2011; Wells et al., 2009). Recently, a humanized mice model has been developed to investigate P. falciparum sexual commitment that could, therefore, bridge in vitro assays to in vivo animal studies that take into account drug metabolism and gametocyte sequestration (Duffier et al., 2016). Also, an experimental Plasmodium vivax transmission model in human has been reported (Griffin et al., 2016). However, mechanisms underlying P. falciparum gametocytogenesis and dynamics have never been addressed in a controlled clean system in humans.

Here, we present a novel CHMI transmission model for P. falciparum that can be used to study gametocyte biology and dynamics providing novel insights and tools in malaria transmission and elimination efforts. The dynamics of gametocyte commitment, maturation, sex ratio, and sequestration found in our model reflect parasite dynamics found in naturally acquired infections, although parasite densities are much lower than in many endemic settings. This model can be used to evaluate the effect of drugs and vaccines on gametocyte dynamics and sex ratios. With its current performance, the CHMI transmission model may allow testing of vaccination strategies that reduce the production of gametocytes from their asexual progenitors or accelerate their clearance from the blood stream (Stone et al., 2016), and the testing of gametocytocidal drugs (White, 2013). To allow testing of sterilizing effect of drugs on circulating gametocytes (White et al., 2014) or the effect of antibodies that interfere with gametocyte fertilisation inside the mosquito gut (Stone et al., 2016), the model needs to be optimized to achieve considerably higher mosquito infection rates. The current work lays the foundation for fulfilling the critical unmet need to evaluate transmission-blocking interventions against falciparum malaria for downstream selection and clinical development.

Materials and methods

Study design

This single centre, open-label randomised trial was conducted at the Radboud university medical center (Radboudumc), Nijmegen, the Netherlands. Healthy malaria-naive male and female participants aged 18–35 years were recruited from June until November 2016. Screening included physical examination, electrocardiography (ECG), hematology and biochemistry parameters and serology for human immunodeficiency virus (HIV), hepatitis B and C, and asexual stages of P. falciparum. Informed consent was provided by all participants at screening visit. The central committee for research involving human subjects (CCMO), and the Western Institutional Review Board (WIRB) approved the protocol for this study (NL56659.091.16). The trial was conducted according to the principles outlined in the Declaration of Helsinki and Good Clinical Practice standards, and registered at ClinicalTrials.gov, identifier NCT02836002 (Supplementary file 4; Reporting Standard 1).

Randomisation

A total of 16 participants were included in the analysis of this study. After inclusion, study participants were randomly allocated to one of the four different treatment arms (n = 4 per group) with low-dose (LD) of either piperaquine (PIP) or sulfadoxine-pyrimethamine (SP), followed by curative regimen of piperaquine or sulfadoxine-pyrimethamine upon recrudescence; (i) LD-SP/SP, (ii) LD-SP/PIP, (iii) LD-PIP/SP, or (iv) LD-PIP/SP. Randomisation was done by a computer-generated random number table (Microsoft Excel 2007, Redmond, WA).

Procedures

All study participants were subjected to a standard CHMI with five female Anopheles stephensi mosquitoes infected with the P. falciparum strain 3D7 (Sauerwein et al., 2011; Cheng et al., 1997). P. falciparum 3D7 asexual and sexual blood stages were cultured in a semi-automated culture system and used to infect mosquitoes by standard membrane feeding as described previously (Ponnudurai et al., 1986; Ponnudurai et al., 1989). The 3D7 lineage that was used in the current study is based on a 3D7 bank described in detail in Cheng et al. (1997). To examine molecular markers of drug resistance, we used available Illumina whole genome sequencing data (https://www.ebi.ac.uk/ena/data/view/PRJEB12838); aligning reads to the P. falciparum reference genome v3 (plasmoDB) with bowtie2 (sourceforge) and obtaining consensus sequences for dhps and dhfr genes with samtools. No mutations were identified in the dhfr gene; the only detected mutation was dhps A437G which, by itself, is not associated with sulfadoxine-pyrimethamine resistance (Staedke et al., 2004). Plasmepsin II/III duplication events are associated with piperaquine resistance (Witkowski et al., 2017) but were not observed although the sequence similarities with neighboring genes Plasmepsin I and IV suggest that unambiguous quantification may require more specific gene targeting. Importantly, piperaquine sensitivity of our 3D7 lineage was previously confirmed by in vivo experiments (Pasay et al., 2016). We conclude that the lineage used was sensitive to both sulfadoxine-pyrimethamine and piperaquine.

Participants were monitored twice daily on an outpatient basis from day 6 after exposure to infected mosquitoes until malaria parasites were detected at a density of ≥5000 parasites per milliliter (Pf/mL) by qPCR or a positive thick blood smear, upon which they were treated with a subcurative dose of 500 mg/25 mg sulfadoxine-pyrimethamine (Roche, Boulogne-billancourt, FR) or 480 mg of piperaquine phosphate (PCI Pharma Services, Tredegar, UK). After the first treatment (T1), participants continued to visit the study center twice daily for another 4 days to monitor the initial clearance of parasitemia by qPCR, after which they were monitored once a day for recrudescence. On day 21 or upon parasite density reaching ≥1500 Pf/mL, participants received a second treatment (T2), consisting of 1000 mg/50 mg sulfadoxine-pyrimethamine or 960 mg of piperaquine phosphate. After the second treatment, participants were monitored daily for 3 days, then three times a week until final treatment with atovaquone/proguanil (Malarone) on day 42. Adverse events were recorded, and blood sampling was performed to monitor parasitemia and blood safety parameters. Symptoms of malaria were treated with acetaminophen up to 4000 mg daily, and nausea with metoclopramide up to 30 mg daily, if necessary.

Parasite density was determined by quantitative PCR (qPCR) targeting the multicopy 18S rRNA gene (Hermsen et al., 2001); samples collected in the morning were processed immediately, evening samples 12 hr later. Thick blood smears were taken during evening visits, double-read and considered positive if two or more parasites were detected in 0.5 $µ$ µL (Laurens et al., 2012). The presence of gametocytes was monitored in samples from day 7.5 after challenge until end of study by quantitative reverse-transcriptase PCR (qRT-PCR) targeting female-specific Pfs25 mRNA and male specific PfMGET (Pf3D7_1469900) and using sex-specific trendlines (Stone et al., 2017; Pett et al., 2016). All samples with an estimated gametocyte density ≥5 gametocytes per mL (gametocytes/mL) were considered gametocyte positive. The duration of gametocyte carriage as an indicator of stable gametocyemia was defined as the maximum number of consecutive days with detectable gametocytemia above the threshold for detection. Direct Membrane Feedings Assays (DMFA) were performed as exploratory measures on days 21, 25 and 31 post-infection with ~300 mosquitoes per feed per participant (total of ~14,400 mosquitoes) (Bousema et al., 2013; Lensen et al., 1998; Ouédraogo AL et al., 2013). Mosquito infection status was determined on day 12 by circumsporozoite (CSP) ELISA(Stone et al., 2015) followed by qPCR confirmation of mosquitoes where the OD exceeded the mean +3 standard deviations of control mosquitoes (Graumans et al., 2017).

Adverse events were recorded and graded by the research physician as mild (easily tolerated, grade 1), moderate (interfering with daily activity, grade 2) or severe (preventing daily activity, grade 3), and in the case of fever as mild (38.0–38.4°C), moderate (38.5–38.9,°C) or severe (≥39°C). Safety blood tests were performed daily, including full blood counts, LDH and highly sensitive troponin-T. Biochemistry tests including liver function test were assessed at screening, inclusion, 2 days after every treatment and at the end of study, and on additional days if considered relevant for clinical decision-making.

Pfs25 and PfMGET RNA quantification

For the quantification of the P. falciparum Pfs25 transcript levels total NA was RQ1 DNaseI treated according to the manufacturer’s protocol. 2 µL of DNaseI-treated material was run in a total volume of 25 µL of TaqMan RNA-to-Ct qRT-PCR reaction mixture (Applied Biosystems, Foster City, California). For the quantification of the P. falciparum male gametocyte enriched transcript (PfMGET), cDNA was synthesized from Total NA with the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Samples were added in a 1: one ratio to the mastermix. 2 µL of cDNA was run in a total volume of 20 µL making use of the GoTaq qPCR Master Mix (Promega, Madison, Wisconsin). Male P. falciparum gametocytes were quantified using a standard curve of serially diluted StageV male gametocytes from the transgenic PfDynGFP/P47mCherry line (Lasonder et al., 2016). Detailed information on the validation and performance characteristics of the assays can be found in the supporting materials (Figure 5; Supplementary file 2, 3; Figure 5—figure supplement 1, 2).

Study outcome

The primary study outcomes were the frequency and magnitude of adverse events, and the prevalence of gametocytes by Pfs25 qRT-PCR. The prevalence of gametocytes is the presence of female gametocytes as measured by qRT-PCR targeting female-specific Pfs25 mRNA at any of the twice daily measurements from day 6. Secondary outcomes were the peak density and time-point of peak density of male and female gametocytes, the AUC of gametocyte density, and assessment of the dynamics of gametocyte commitment, maturation and sex-ratio. The AUC of gametocyte density represents the total gametocyte exposure over time (gametocyte load). Assessment of gametocyte infectivity to Anopheles stephensi mosquitoes by DMFA was an exploratory study endpoint.

Statistical analysis

The sample size was calculated based on preliminary data that > 95% of the participants would develop gametocytemia. Conservatively, we considered the approach unsuitable for gametocyte induction if <50% of individuals developed mature gametocytes. We, therefore, powered the trial to estimate a 90% confidence interval around the proportion of gametocytaemic individuals that excludes 50%. If eight individuals (allowing for one dropout per arm), and 6/7 or 7/7 of these individuals become gametocytaemic, we would be able to estimate this proportion with a lower limit of the 90% Wilson confidence interval ≥54.8% (the lower limit of the 95% confidence interval being 48.7%). Differences between study arms were assessed by comparing mean values using a one-way ANOVA or non-parametric equivalents.

To further identify which study arm(s) potentially deviated from others, we jointly estimated the differences between all four arms in a Bayesian framework (standard linear regression model, no mixed effects), using Hamiltonian Monte Carlo as implemented in the R package rstanarm, and using an uninformative (uniform) prior for the explained variation (R^2) (see R codes used in Source code 1) (Team SD, 2016). For discrete variables (e.g. the number of positive assays), the chi-squared test or Fisher’s exact test was used (two-tailed). The total number of adverse events and total number of grade three adverse events were calculated per individual and compared by non-parametric Kruskal Wallis test.

A previously developed model was used to estimate gametocyte half-life for female and male gametocytes separately (Bousema et al., 2010). For this analysis, gametocyte observations were included from 12 days after the last detection of asexual parasites until the end of study. This was based on the gametocyte sequestration time of 10–12 days in this study, and the assumption that the number of newly released gametocytes would thus be minimal in this observation period. All model fittings were carried out using the PROC NLMIXED procedure in SAS (Version 9, SAS Institute Inc) and included no covariates other than time (see Source code 2 for SAS code). The AUC was computed by GraphPad Prism 5 (USA) with the (X2-X1)*(Y1 +Y2)/2 formula (X = days post challenge; Y = gametocytes per mL (≥5 gametocytes/mL)) as used repeatedly for each adjacent pair of points defining the curve; the total AUC was used.

Acknowledgements

We would like to thank the staff from the Clinical Research Centre Nijmegen. We also thank Daphne Smit and Annemieke Jansens for their assistance and project support during the trial. We thank the following individuals for their assistance during the trial: Laura Pelser, Jolanda Klaassen, Astrid Pouwelsen, and Jacqueline Kuhnen for the mosquito infection and dissection work, Foekje Stelma, and all the thick smear microscopists for reading many smears. We acknowledge Gheorghe Pop for his help in evaluation of electrocardiograms. This trial was supported mainly with funds from Path Malaria Vaccine Initiative (MVI). TB is supported by a fellowship from the European Research Council (ERC-2014-StG 639776). Finally and foremost, we would like to thank the study volunteers who participated in this trial. The funder, PATH’s Malaria Vaccine Initiative, was involved in the study design, analysis and interpretation of the data, the preparation of the report, but not data collection. IJR, LS, TB, and RWS had full access to all study data with final responsibility for the decision to submit the report for publication.

Funding Statement

The funder, PATH's Malaria Vaccine Initiative, was involved in the study design, analysis and interpretation of the data, the preparation of the report, but not data collection. IJR, LS, TB, and RWS had full access to all study data with final responsibility for the decision to submit the report for publication.

Contributor Information

Teun Bousema, Email: teun.bousema@radboudumc.nl.

Ben Cooper, Mahidol Oxford Tropical Medicine Research Unit, Thailand.

Funding Information

This paper was supported by the following grants:

PATH Malaria Vaccine Initiative to Isaie J Reuling, Lisanne A van de Schans, Kjerstin Lanke, Wouter Graumans, Geert-Jan van Gemert, Karina Teelen, Rianne Siebelink-Stoter, Marga van de Vegte-Bolmer, Quirijn de Mast, André J van der Ven, Karen Ivinson, Cornelus C Hermsen, Katharine A Collins, Christian F Ockenhouse, James McCarthy, Robert W Sauerwein, Teun Bousema.
H2020 European Research Council Fellowship to Teun Bousema.
European Research Council ERC-2014-StG 639776 to Teun Bousema.

Additional information

Competing interests

No competing interests declared.

Author contributions

Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing.

Formal analysis, Investigation, Visualization, Writing—original draft, Writing—review and editing.

Formal analysis, Writing—review and editing.

Methodology, Writing—review and editing.

Data curation, Methodology, Writing—review and editing.

Methodology, Writing—review and editing.

Investigation, Writing—review and editing.

Resources, Validation, Project administration, Writing—review and editing.

Methodology, Writing—review and editing.

Formal analysis, Writing—review and editing.

Conceptualization, Writing—review and editing.

Conceptualization, Resources, Writing—review and editing.

Conceptualization, Supervision, Investigation, Writing—review and editing.

Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing—original draft.

Ethics

Clinical trial registration NCT02836002.

Human subjects: Informed consent was provided by all participants at screening visit. The central committee for research involving human subjects (CCMO), and the Western Institutional Review Board (WIRB) approved the protocol for this study (NL56659.091.16). The trial was conducted according to the principles outlined in the Declaration of Helsinki and Good Clinical Practice standards, and registered at ClinicalTrials.gov, identifier NCT02836002.

Additional files

Source code 1. R codes used for Bayesian statistical analysis.

elife-31549-code1.zip^{(9.9KB, zip)}

Source code 2. SAS code used for estimation of gametocyte half-life for gametocytes.

elife-31549-code2.docx^{(16KB, docx)}

Supplementary file 1. Individual data of the participants included in analysis.

elife-31549-supp1.docx^{(16.5KB, docx)}

Supplementary file 2. Selected P. falciparum gene targets and primers of qRT PCR assays.

elife-31549-supp2.docx^{(11.7KB, docx)}

Supplementary file 3. Quality parameters of qRT PCR and qPCR.

The table shows for each target: limit of detection (LOD; defined as lowest pathogen concentration with reproducible detection); limit of quantification (LOQ; defined as lowest pathogen concentration where the CV was <5%), slope, efficiency (E), and the coefficient of correlation of combined trendlines (R²).

elife-31549-supp3.docx^{(12KB, docx)}

Supplementary file 4. Clinical trial protocol.

elife-31549-supp4.pdf^{(480KB, pdf)}

Reporting standard 1. CONSORT extension for Pilot and Feasibility Trials Checklist.

elife-31549-repstand1.docx^{(41KB, docx)}

Transparent reporting form

elife-31549-transrepform.docx^{(246.7KB, docx)}

Data availability

The following dataset was generated:

Reuling IJ, Bousema T, Sauerwein RW. 2017. Data from: Induction of Plasmodium falciparum gametocytemia in the Controlled Human Malaria Infection model: a randomised trial comparing four antimalarial drug regimens. Available at Dryad Digital Repository under a CC0 Public Domain Dedication.

References

Adjalley SH, Johnston GL, Li T, Eastman RT, Ekland EH, Eappen AG, Richman A, Sim BK, Lee MC, Hoffman SL, Fidock DA. Quantitative assessment of Plasmodium falciparum sexual development reveals potent transmission-blocking activity by methylene blue. PNAS. 2011;108:E1214–E1223. doi: 10.1073/pnas.1112037108. [DOI] [PMC free article] [PubMed] [Google Scholar]
Bolscher JM, Koolen KM, van Gemert GJ, van de Vegte-Bolmer MG, Bousema T, Leroy D, Sauerwein RW, Dechering KJ. A combination of new screening assays for prioritization of transmission-blocking antimalarials reveals distinct dynamics of marketed and experimental drugs. Journal of Antimicrobial Chemotherapy. 2015;70:1357–1366. doi: 10.1093/jac/dkv003. [DOI] [PubMed] [Google Scholar]
Bousema T, Churcher TS, Morlais I, Dinglasan RR. Can field-based mosquito feeding assays be used for evaluating transmission-blocking interventions? Trends in Parasitology. 2013;29:53–59. doi: 10.1016/j.pt.2012.11.004. [DOI] [PubMed] [Google Scholar]
Bousema T, Dinglasan RR, Morlais I, Gouagna LC, van Warmerdam T, Awono-Ambene PH, Bonnet S, Diallo M, Coulibaly M, Tchuinkam T, Mulder B, Targett G, Drakeley C, Sutherland C, Robert V, Doumbo O, Touré Y, Graves PM, Roeffen W, Sauerwein R, Birkett A, Locke E, Morin M, Wu Y, Churcher TS. Mosquito feeding assays to determine the infectiousness of naturally infected Plasmodium falciparum gametocyte carriers. PLoS One. 2012;7:e42821. doi: 10.1371/journal.pone.0042821. [DOI] [PMC free article] [PubMed] [Google Scholar]
Bousema T, Drakeley C. Epidemiology and infectivity of Plasmodium falciparum and Plasmodium vivax gametocytes in relation to malaria control and elimination. Clinical Microbiology Reviews. 2011;24:377–410. doi: 10.1128/CMR.00051-10. [DOI] [PMC free article] [PubMed] [Google Scholar]
Bousema T, Okell L, Shekalaghe S, Griffin JT, Omar S, Sawa P, Sutherland C, Sauerwein R, Ghani AC, Drakeley C. Revisiting the circulation time of Plasmodium falciparum gametocytes: molecular detection methods to estimate the duration of gametocyte carriage and the effect of gametocytocidal drugs. Malaria Journal. 2010;9:136. doi: 10.1186/1475-2875-9-136. [DOI] [PMC free article] [PubMed] [Google Scholar]
Butcher GA. Antimalarial drugs and the mosquito transmission of Plasmodium. International Journal for Parasitology. 1997;27:975–987. doi: 10.1016/S0020-7519(97)00079-9. [DOI] [PubMed] [Google Scholar]
Cheng Q, Lawrence G, Reed C, Stowers A, Ranford-Cartwright L, Creasey A, Carter R, Saul A. Measurement of Plasmodium falciparum growth rates in vivo: a test of malaria vaccines. The American Journal of Tropical Medicine and Hygiene. 1997;57:495–500. doi: 10.4269/ajtmh.1997.57.495. [DOI] [PubMed] [Google Scholar]
Ciuca MBL, Chearescu M, Lavrinenko N. Contribution À L’étude De L’infection Expérimentale Au Plasmodium Falciparum. Hamburg: Festchrift Institute Bernard Nocht; 1937. pp. 81–101. [Google Scholar]
Delves MJ, Ruecker A, Straschil U, Lelièvre J, Marques S, López-Barragán MJ, Herreros E, Sinden RE. Male and female Plasmodium falciparum mature gametocytes show different responses to antimalarial drugs. Antimicrobial Agents and Chemotherapy. 2013;57:3268–3274. doi: 10.1128/AAC.00325-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
Duffier Y, Lorthiois A, Cisteró P, Dupuy F, Jouvion G, Fiette L, Mazier D, Mayor A, Lavazec C, Moreno Sabater A. A humanized mouse model for sequestration of Plasmodium falciparum sexual stages and in vivo evaluation of gametocytidal drugs. Scientific Reports. 2016;6:35025. doi: 10.1038/srep35025. [DOI] [PMC free article] [PubMed] [Google Scholar]
Eichner M, Diebner HH, Molineaux L, Collins WE, Jeffery GM, Dietz K. Genesis, sequestration and survival of Plasmodium falciparum gametocytes: parameter estimates from fitting a model to malariatherapy data. Transactions of the Royal Society of Tropical Medicine and Hygiene. 2001;95:497–501. doi: 10.1016/S0035-9203(01)90016-1. [DOI] [PubMed] [Google Scholar]
Farid R, Dixon MW, Tilley L, McCarthy JS. Initiation of gametocytogenesis at very low parasite density in Plasmodium falciparum infection. The Journal of Infectious Diseases. 2017;215:1167–1174. doi: 10.1093/infdis/jix035. [DOI] [PMC free article] [PubMed] [Google Scholar]
Gonçalves BP, Drakeley C, Bousema T. Infectivity of microscopic and submicroscopic malaria parasite infections in areas of low malaria endemicity. Journal of Infectious Diseases. 2016;213:1516–1517. doi: 10.1093/infdis/jiw044. [DOI] [PubMed] [Google Scholar]
Graumans W, Tadesse FG, Andolina C, van Gemert GJ, Teelen K, Lanke K, Gadisa E, Yewhalaw D, van de Vegte-Bolmer M, Siebelink-Stoter R, Reuling I, Sauerwein R, Bousema T. Semi-high-throughput detection of Plasmodium falciparum and Plasmodium vivax oocysts in mosquitoes using bead-beating followed by circumsporozoite ELISA and quantitative PCR. Malaria Journal. 2017;16:356. doi: 10.1186/s12936-017-2011-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Griffin JT, Hollingsworth TD, Okell LC, Churcher TS, White M, Hinsley W, Bousema T, Drakeley CJ, Ferguson NM, Basáñez MG, Ghani AC. Reducing Plasmodium falciparum malaria transmission in Africa: a model-based evaluation of intervention strategies. PLoS Medicine. 2010;7:e1000324. doi: 10.1371/journal.pmed.1000324. [DOI] [PMC free article] [PubMed] [Google Scholar]
Griffin P, Pasay C, Elliott S, Sekuloski S, Sikulu M, Hugo L, Khoury D, Cromer D, Davenport M, Sattabongkot J, Ivinson K, Ockenhouse C, McCarthy J. Safety and reproducibility of a clinical trial system using induced blood stage plasmodium vivax infection and its potential as a model to evaluate malaria transmission. PLOS Neglected Tropical Diseases. 2016;10:e0005139. doi: 10.1371/journal.pntd.0005139. [DOI] [PMC free article] [PubMed] [Google Scholar]
Hermsen CC, Telgt DS, Linders EH, van de Locht LA, Eling WM, Mensink EJ, Sauerwein RW. Detection of Plasmodium falciparum malaria parasites in vivo by real-time quantitative PCR. Molecular and Biochemical Parasitology. 2001;118:247–251. doi: 10.1016/S0166-6851(01)00379-6. [DOI] [PubMed] [Google Scholar]
Kafsack BF, Rovira-Graells N, Clark TG, Bancells C, Crowley VM, Campino SG, Williams AE, Drought LG, Kwiatkowski DP, Baker DA, Cortés A, Llinás M. A transcriptional switch underlies commitment to sexual development in malaria parasites. Nature. 2014;507:248–252. doi: 10.1038/nature12920. [DOI] [PMC free article] [PubMed] [Google Scholar]
Lasonder E, Rijpma SR, van Schaijk BC, Hoeijmakers WA, Kensche PR, Gresnigt MS, Italiaander A, Vos MW, Woestenenk R, Bousema T, Mair GR, Khan SM, Janse CJ, Bártfai R, Sauerwein RW. Integrated transcriptomic and proteomic analyses of P. falciparum gametocytes: molecular insight into sex-specific processes and translational repression. Nucleic Acids Research. 2016;44:6087–6101. doi: 10.1093/nar/gkw536. [DOI] [PMC free article] [PubMed] [Google Scholar]
Laurens MB, Duncan CJ, Epstein JE, Hill AV, Komisar JL, Lyke KE, Ockenhouse CF, Richie TL, Roestenberg M, Sauerwein RW, Spring MD, Talley AK, Moorthy VS, Consensus Group on Design of Clinical Trials of Controlled Human Malaria Infection A consultation on the optimization of controlled human malaria infection by mosquito bite for evaluation of candidate malaria vaccines. Vaccine. 2012;30:5302–5304. doi: 10.1016/j.vaccine.2012.04.088. [DOI] [PubMed] [Google Scholar]
Lensen A, Mulder L, Tchuinkam T, Willemsen L, Eling W, Sauerwein R. Mechanisms that reduce transmission of Plasmodium falciparum malaria in semiimmune and nonimmune persons. The Journal of Infectious Diseases. 1998;177:1358–1363. doi: 10.1086/515263. [DOI] [PubMed] [Google Scholar]
Moonen B, Cohen JM, Snow RW, Slutsker L, Drakeley C, Smith DL, Abeyasinghe RR, Rodriguez MH, Maharaj R, Tanner M, Targett G. Operational strategies to achieve and maintain malaria elimination. The Lancet. 2010;376:1592–1603. doi: 10.1016/S0140-6736(10)61269-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
Méndez F, Muñoz A, Carrasquilla G, Jurado D, Arévalo-Herrera M, Cortese JF, Plowe CV. Determinants of treatment response to sulfadoxine-pyrimethamine and subsequent transmission potential in falciparum malaria. American Journal of Epidemiology. 2002;156:230–238. doi: 10.1093/aje/kwf030. [DOI] [PubMed] [Google Scholar]
Nilsson SK, Childs LM, Buckee C, Marti M. Targeting human transmission biology for malaria elimination. PLoS Pathogens. 2015;11:e1004871. doi: 10.1371/journal.ppat.1004871. [DOI] [PMC free article] [PubMed] [Google Scholar]
Ouédraogo AL, Guelbéogo WM, Cohuet A, Morlais I, King JG, Goncalves BP, Bastiaens GJH, Vaanhold M, Sattabongkot J, Wu Y, Coulibaly M, Ibrahima B, Jones S, Morin M, Drakeley C, Dinglasan RR, Bousema T. A protocol for membrane feeding assays to determine the infectiousness of P. falciparum naturally infected individuals to Anopheles gambiae. MalariaWorld Journal. 2013;4:16. doi: 10.5281/zenodo.10926272. [DOI] [PMC free article] [PubMed] [Google Scholar]
Pasay CJ, Rockett R, Sekuloski S, Griffin P, Marquart L, Peatey C, Wang CY, O'Rourke P, Elliott S, Baker M, Möhrle JJ, McCarthy JS. Piperaquine monotherapy of drug-susceptible plasmodium falciparum infection results in rapid clearance of parasitemia but is followed by the appearance of gametocytemia. Journal of Infectious Diseases. 2016;214:105–113. doi: 10.1093/infdis/jiw128. [DOI] [PMC free article] [PubMed] [Google Scholar]
Pett H, Gonçalves BP, Dicko A, Nébié I, Tiono AB, Lanke K, Bradley J, Chen I, Diawara H, Mahamar A, Soumare HM, Traore SF, Baber I, Sirima SB, Sauerwein R, Brown J, Gosling R, Felger I, Drakeley C, Bousema T. Comparison of molecular quantification of Plasmodium falciparum gametocytes by Pfs25 qRT-PCR and QT-NASBA in relation to mosquito infectivity. Malaria Journal. 2016;15:539. doi: 10.1186/s12936-016-1584-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
Ponnudurai T, Lensen AH, Meis JF, Meuwissen JH. Synchronization of Plasmodium falciparum gametocytes using an automated suspension culture system. Parasitology. 1986;93:263–274. doi: 10.1017/S003118200005143X. [DOI] [PubMed] [Google Scholar]
Ponnudurai T, Lensen AH, Van Gemert GJ, Bensink MP, Bolmer M, Meuwissen JH. Infectivity of cultured Plasmodium falciparum gametocytes to mosquitoes. Parasitology. 1989;98:165–173. doi: 10.1017/S0031182000062065. [DOI] [PubMed] [Google Scholar]
Reuling IJ, Stone WJR, van de Vegte-Bolmer M, van Gemert GJ, Siebelink-Stoter R, Graumans W, Lanke K, Bousema T, Sauerwein RW. Concentration of Plasmodium falciparum gametocytes in whole blood samples by magnetic cell sorting enhances parasite infection rates in mosquito feeding assays. Malaria Journal. 2017;16:315. doi: 10.1186/s12936-017-1959-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Sauerwein RW, Roestenberg M, Moorthy VS. Experimental human challenge infections can accelerate clinical malaria vaccine development. Nature Reviews Immunology. 2011;11:57–64. doi: 10.1038/nri2902. [DOI] [PubMed] [Google Scholar]
Schneider P, Schoone G, Schallig H, Verhage D, Telgt D, Eling W, Sauerwein R. Quantification of Plasmodium falciparum gametocytes in differential stages of development by quantitative nucleic acid sequence-based amplification. Molecular and Biochemical Parasitology. 2004;137:35–41. doi: 10.1016/j.molbiopara.2004.03.018. [DOI] [PubMed] [Google Scholar]
Shute PG, Maryon M. A study of gametocytes in a West African strain of Plasmodium falciparum. Transactions of the Royal Society of Tropical Medicine and Hygiene. 1951;44:421–438. doi: 10.1016/S0035-9203(51)80020-8. [DOI] [PubMed] [Google Scholar]
Sinden RE. Developing transmission-blocking strategies for malaria control. PLoS Pathogens. 2017;13:e1006336. doi: 10.1371/journal.ppat.1006336. [DOI] [PMC free article] [PubMed] [Google Scholar]
Staedke SG, Sendagire H, Lamola S, Kamya MR, Dorsey G, Rosenthal PJ. Relationship between age, molecular markers, and response to sulphadoxine-pyrimethamine treatment in Kampala, Uganda. Tropical Medicine and International Health. 2004;9:624–629. doi: 10.1111/j.1365-3156.2004.01239.x. [DOI] [PubMed] [Google Scholar]
Stone W, Grabias B, Lanke K, Zheng H, Locke E, Diallo D, Birkett A, Morin M, Bousema T, Kumar S. A comparison of Plasmodium falciparum circumsporozoite protein-based slot blot and ELISA immuno-assays for oocyst detection in mosquito homogenates. Malaria Journal. 2015;14:451. doi: 10.1186/s12936-015-0954-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
Stone W, Sawa P, Lanke K, Rijpma S, Oriango R, Nyaurah M, Osodo P, Osoti V, Mahamar A, Diawara H, Woestenenk R, Graumans W, van de Vegte-Bolmer M, Bradley J, Chen I, Brown J, Siciliano G, Alano P, Gosling R, Dicko A, Drakeley C, Bousema T. A molecular assay to quantify male and female plasmodium falciparum gametocytes: results from 2 randomized controlled trials using primaquine for gametocyte clearance. The Journal of Infectious Diseases. 2017;216:457–467. doi: 10.1093/infdis/jix237. [DOI] [PMC free article] [PubMed] [Google Scholar]
Stone WJ, Dantzler KW, Nilsson SK, Drakeley CJ, Marti M, Bousema T, Rijpma SR. Naturally acquired immunity to sexual stage P. falciparum parasites. Parasitology. 2016;143:187–198. doi: 10.1017/S0031182015001341. [DOI] [PubMed] [Google Scholar]
Stone WJ, Eldering M, van Gemert GJ, Lanke KH, Grignard L, van de Vegte-Bolmer MG, Siebelink-Stoter R, Graumans W, Roeffen WF, Drakeley CJ, Sauerwein RW, Bousema T. The relevance and applicability of oocyst prevalence as a read-out for mosquito feeding assays. Scientific Reports. 2013;3:3418. doi: 10.1038/srep03418. [DOI] [PMC free article] [PubMed] [Google Scholar]
Team SD rstanarm: Bayesian applied regression modeling via Stan 2016
Wells TN, Alonso PL, Gutteridge WE. New medicines to improve control and contribute to the eradication of malaria. Nature Reviews Drug Discovery. 2009;8:879–891. doi: 10.1038/nrd2972. [DOI] [PubMed] [Google Scholar]
White NJ, Ashley EA, Recht J, Delves MJ, Ruecker A, Smithuis FM, Eziefula AC, Bousema T, Drakeley C, Chotivanich K, Imwong M, Pukrittayakamee S, Prachumsri J, Chu C, Andolina C, Bancone G, Hien TT, Mayxay M, Taylor WR, von Seidlein L, Price RN, Barnes KI, Djimdé A, ter Kuile F, Gosling R, Chen I, Dhorda MJ, Stepniewska K, Guérin P, Woodrow CJ, Dondorp AM, Day NP, Nosten FH. Assessment of therapeutic responses to gametocytocidal drugs in Plasmodium falciparum malaria. Malaria Journal. 2014;13:483. doi: 10.1186/1475-2875-13-483. [DOI] [PMC free article] [PubMed] [Google Scholar]
White NJ. Primaquine to prevent transmission of falciparum malaria. The Lancet Infectious Diseases. 2013;13:175–181. doi: 10.1016/S1473-3099(12)70198-6. [DOI] [PubMed] [Google Scholar]
Witkowski B, Duru V, Khim N, Ross LS, Saintpierre B, Beghain J, Chy S, Kim S, Ke S, Kloeung N, Eam R, Khean C, Ken M, Loch K, Bouillon A, Domergue A, Ma L, Bouchier C, Leang R, Huy R, Nuel G, Barale JC, Legrand E, Ringwald P, Fidock DA, Mercereau-Puijalon O, Ariey F, Ménard D. A surrogate marker of piperaquine-resistant Plasmodium falciparum malaria: a phenotype-genotype association study. The Lancet Infectious Diseases. 2017;17:174–183. doi: 10.1016/S1473-3099(16)30415-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
World Health Organization . Eliminating Malaria. Vol 24. World Health Organization; 2016. [Google Scholar]
World Health Organization . World Malaria Report. World Health Organization; 2016. [Google Scholar]
WWARN Gametocyte Study Group Gametocyte carriage in uncomplicated Plasmodium falciparum malaria following treatment with artemisinin combination therapy: a systematic review and meta-analysis of individual patient data. BMC Medicine. 2016;14:79. doi: 10.1186/s12916-016-0621-7. [DOI] [PMC free article] [PubMed] [Google Scholar]

eLife. doi: 10.7554/eLife.31549.sa1

Decision letter

Editor: Ben Cooper¹

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Thank you for submitting your article "Induction of Plasmodium falciparum gametocytemia in the Controlled Human Malaria Infection model: a randomised trial comparing four antimalarial drug regimens" for consideration by eLife. Your article has been favorably evaluated by Prabhat Jha (Senior Editor) and three reviewers, one of whom, Ben Cooper (Reviewer #3), is a member of our Board of Reviewing Editors. The following individual involved in review of your submission has agreed to reveal their identity: Nicholas J White (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

This paper describes a novel controlled human malaria infection model for Plasmodium falciparum that consistently and safely induces gametocytaemia using four different drug-regimens based on sulfadoxine-pyrimethamime and piperaquine. Estimated mean gametocyte circulation times were about 3 and 5 days for male and female gametocytes respectively and the timing of gametocyte appearance indicated probable commitment to gametocyte production in the first wave of asexual parasites. This work has importance for understanding gametocyte dynamics and evaluating transmission blocking interventions.

Essential revisions:

1) Clarification of sample size calculations is required (see reviewer 1 comments).

2) The section on parasite DNA and RNA quantification needs strengthening.a) The performance characteristics of the 18S DNA and mRNA assays needs presenting in detail or references provided to detailed validation – accuracy at different densities, reproducibility, linearity, criteria for limits of detection etc.

b) The volume of blood taken and assayed needs stating.

c) Validation of gametocyte quantification needs support – particularly with reference to stability of transcript numbers per cell and any assumptions made in the derivation of a gametocyte density (see reviewer 2 comments).

3) Details of statistical analysis performed and reporting of results of the analysis are inadequate. See comments from all three reviewers for more details. The response should include addressing reviewer 2's concerns about modelling the decline in gametocyte densities. It would also be helpful for authors to provide the code used for data analysis as recommended in eLife's transparent reporting form.

4) The CONSORT guidelines for trial reporting should be followed (see reviewer 1 comments).

5) More details of the membrane feeding experiments need to be reported (see reviewer 1 and 3 comments).

6) Figures 2, 4 and (ideally) 5 should be improved to better convey information (see reviewer 1 and 3 comments).

Reviewer #1:

Deliberate infections treated with SP and piperaquine have been shown to be followed by gametocytaemia. This paper reports a small study with 16 participants with the objective of determining which of four drug combinations work best for inducing gametocytaemia in CHMI.

Introduction, last paragraph. The aim is to "induce stable gametocyte carriage". What is considered to be stable is buried as a footnote to Supplementary file 1. The optimal characteristics (density, duration) are also implied rather than stated.

The primary study outcome (subsection “Study outcome”) is prevalence of gametocytes by Pfs25 qRT-PCR. Was this the presence of female gametocytes at any of the twice daily measurements from day 6?

The sample size calculation (subsection “Statistical analysis”, first paragraph) appears to set the bar very low. Even so, I could not quite reproduce the calculation: an exact binomial 90% confidence interval for 6/7 would have lower bound 47.6%. It might be that you used an approximation, but since the numbers are small an exact CI seems more appropriate. Why did you choose a 90% confidence interval rather than the more conventional 95%? (with 95% CI, the lower bound for 6/7 would be 42.1%). The numbers are very small, and nowhere is it really explained why more data would be hard to collect.

There was an odd dissonance between the simple comparisons for hypothesis tests (ANOVA, Fisher's Exact test) and a more sophisticated analysis to "identify which study arm potentially deviated from the others". I'm not sure what this involves since few details were given, but it includes a prior for R^2 which seems unintuitive, and, given the limited sample size, could be stretching the data further than it can bear. It was also written without clarity in the Results e.g. "96.5% probability of being the highest" – the LD-PIP/SP mean was anyway the highest in the trial. Do you mean the chance of the population mean of LD-PIP/SP being the highest?

It would be useful to briefly state the assumptions in the previously developed model to estimate the gametocyte half-life.

It is not obvious that the AUC is the most useful measure – a short spike in gametocytes could be equal to longer low-level gametocytaemia. Clarification of what the optimal characteristics for drug development are would help. The AUC as a measure may also suffer from the non-independence of each time-point from the previous time-points – from a high peak (which may have occurred by chance or not due to the drugs), then further high values might tend to follow. A suggestion only: it might be possible to take into account the non-independence between time-points and use all the data, by using a statistical time-series model with a lag from the asexual densities or a simple mechanistic model fitted to the data. You might need more than four participants per arm for that but you might also be able to borrow strength for some parameters.

The Results section focuses on differences between the arms but it is unfortunate that the AUC is the only reported measure of gametocytaemia which is (borderline) significant between the arms for gametocytaemia (or the other p-values are just not mentioned). Looking at the graphs, I think the effect is plausible, but there is very limited evidence due to the small sample size.

That three mosquitoes were infected cannot be interpreted without the numbers of mosquitoes fed (Results, fifth paragraph).

The male and female gametocytes were measured by different assays which may have different levels of detection. The results for the sex-ratio should be interpreted with caution.

The CONSORT guidelines for reporting should be followed for eLife. Since this is a phase 1/2 trial with a small sample of healthy participants, the most appropriate CONSORT guidelines would be: CONSORT 2010 statement: extension to randomised pilot and feasibility trials (http://www.bmj.com/content/355/bmj.i5239). The reporting guidelines cover design, randomization, intervention, sample size, bias, generalisability and access to the pilot protocol.

Given that the emphasis in the results is on the differences between arms, Figure 4, where all groups are combined, is not easy to interpret.

There is little mention of the next steps for the induced gametocytes. If they are to be used in testing interventions that reduce gametocyte development, how would they be used?

It is not discussed why the different combinations of drug might have different effects.

Reviewer #2:

This is a very interesting and informative study of Plasmodium falciparum malaria gametocyte dynamics which exploits the opportunities provided by the resurgence of interest in human challenge studies and the development of methodologies for sensitive quantitation of nucleic acid concentrations. In general this is a very good piece of work but more details on the validation of qPCR gametocyte quantitation are essential if it is to be published. If there is one disappointment it is the complete absence of reference to the early observations of gametocytaemia in human challenge studies – notably the work of Shute and Cuica whose conclusions were in broad agreement with the current paper. The earliest reference is from 1986.

A list of questions or comments in the order they appear in the manuscript:

1) "It is widely accepted that malaria elimination is unlikely to be attainable in the majority of endemic settings with currently available resources and tools". Perhaps this could be attenuated? Many think the obstacles are primarily political, organisational and financial.

2) "maturation of gametocytes takes place predominantly in the bone marrow".

3) The 3D7 "lineage" is well known, but it is also well known to be different in different laboratories! Perhaps a few additional sentences on this particular lineage would be valuable. In particular whether there is any evidence that serial passage through volunteers and these laboratory reared mosquitoes has altered its biological properties – notably infectivity over the years? Please also confirm wild type PfDHFR and DHPS and single copy plasmepsin 2/3.

4) "until malaria parasites were detected at a density of {greater than or equal to}5000 parasites per milliliter (Pf/mL) by qPCR or a positive thick blood smear" – – were any volunteers symptomatic at this density?

5) Piperaquine base or piperaquine phosphate?

6) Safety seems rightly to have been a major concern. Did anyone look at the ECG if metoclopramide was given to piperaquine recipients? Was there any additional QTc prolongation?

7) The section on parasite DNA and RNA quantitation needs considerable strengthening as it is the central component of the paper. This is important.a) The performance characteristics of the 18S DNA and mRNA assays needs presenting in detail or references provided to detailed validation – accuracy at different densities, reproducibility, linearity, criteria for limits of detection etc.

b) The volume of blood taken and assayed needs stating.

c) Validation of gametocyte quantitation needs support – particularly with reference to stability of transcript numbers per cell and any assumptions made in the derivation of a gametocyte density.

8) Bousema et al., 2010 is referred to – which if I understand correctly assumes a single first order decline in gametocyte densities. Is this justified? It would be valuable to describe, if possible, the residuals around the model fits and comment on any heteroscedasticity. This is particularly important considering that a major finding of this report is that male gametocytes appear to be cleared more rapidly than female gametocytes. If in fact the gametocyte clearance profile is more complex (e.g. multiexponential) then the lower density male gametocytes may appear to be more rapidly eliminated whereas, in fact, the slower terminal phase of elimination is below the limit of accurate detection. A similar problem has bedevilled assessment of the pharmacokinetic properties of slowly eliminated antimalarials.

9) It is not very clear what modelling approach was used to fit the model – was this a mixed effect model? Were any covariates incorporated – if so which? What programme was used?

10) Is there any explanation for the liver function test abnormalities?

11) "SP has long been associated with a rapid appearance of gametocytes that is too early to be explained by de novo gametocyte production upon drug pressure and has thus been hypothesized to reflect an efflux of sequestered gametocytes upon treatment". True – but it is not very clear from the text whether this study's results supports this hypothesis. Could the authors be explicit here? The implication is that such early released gametocytes should be more immature- and thus the period during which Plasmodium falciparum gametocytes are not infectious (after release into the circulation) should be longer in these circumstances. Is there any evidence for this?

12) "The highly abundant Pfs25 mRNA makes the female gametocyte qRT-PCR more sensitive than the male PfMGET qRT-PCR." Could this be elaborated upon- or at least referenced?

Reviewer #3:

This paper describes a novel controlled human malaria infection model that consistently achieves gametocyte carriage. This work is important with implications for understanding of P. falciparum transmission dynamics and broad significance for the future study of transmission blocking interventions.

The paper is, in most respects, clearly written. There are, however, some aspects of reporting where it is not entirely clear exactly what was done and why (particularly in relation to statistical methods).

Results, fifth paragraph. Details of these membrane feeding experiments are lacking from the Materials and methods. In particular, how many mosquitos on each feeding day? Even if full details are presented in the references describing the protocols for these experiments, it would be useful to summarise some of the key numbers in the manuscript to give the reader an idea of what these numbers mean. See also Discussion, third paragraph – "the very low rate". Since only the numerators are reported, there is no information on what this rate actually is.

Subsection “Statistical analysis”, second paragraph. More details need to be given here. What non-parametric tests were used and what exactly was the statistical model implemented in the Bayesian framework (code for this does not appear to have been provided). Also, the R package rstanarm is mentioned, but appears not to be acknowledged in the references. It probably should be along with the appropriate references for Stan and R.

Table 3 is confusing. Why are some adverse events listed twice in the leftmost column? Duration time units should be given.

eLife. 2018 Feb 27;7:e31549. doi: 10.7554/eLife.31549.sa2

Author response

Essential revisions:

1) Clarification of sample size calculations is required (see reviewer 1 comments).

We have clarified the sample size calculation in response to the comments of reviewer 1. We used Wilson confidence intervals and have explained this in more detail in the comment of reviewer 1 and the revised manuscript.

2) The section on parasite DNA and RNA quantification needs strengthening.a) The performance characteristics of the 18S DNA and mRNA assays needs presenting in detail or references provided to detailed validation – accuracy at different densities, reproducibility, linearity, criteria for limits of detection etc.

We have provided all the requested details of our assay performance in the supporting information of the revised manuscript. We agree that this is a considerable improvement of the original data presentation and important for the reader. Please find our detailed response to this question below in the section related to comments of reviewer 2.

b) The volume of blood taken and assayed needs stating.

Similarly, we have provided details on blood volume taken, used for extraction and pathogen quantification. The detailed response is given below in the section of reviewer 2.

c) Validation of gametocyte quantification needs support – particularly with reference to stability of transcript numbers per cell and any assumptions made in the derivation of a gametocyte density (see reviewer 2 comments).

We have provided these details using new data related to the current manuscript and a re-analysis of data of a recently published study where we originally presented the methodology and determined the stability of gametocyte transcripts in maturing male and female gametocytes (Stone et al., 2017). We were also able to provide details on the number of Pfs25 transcripts per female gametocyte and the number of PfMGET transcripts per male gametocyte.

3) Details of statistical analysis performed and reporting of results of the analysis are inadequate. See comments from all three reviewers for more details. The response should include addressing reviewer 2's concerns about modelling the decline in gametocyte densities. It would also be helpful for authors to provide the code used for data analysis as recommended in eLife's transparent reporting form.

We have provided the details on the statistical analysis methods used and the reporting of the analysis (see reviewer comments below). In addition, the annotated codes used for data analysis are now added as supplementary information in the manuscript.

4) The CONSORT guidelines for trial reporting should be followed (see reviewer 1 comments).

We have added the CONSORT extension for Pilot and Feasibility Trials Checklist to the dossier, and adjusted the manuscript as per CONSORT guidelines.

5) More details of the membrane feeding experiments need to be reported (see reviewer 1 and 3 comments).

Membrane feeding experiments were an exploratory outcome in the current study and in the initial submission we provided limited details in order to focus on the primary and secondary endpoints. In the revision, we provide all details on the Direct Membrane Feedings Assays (DMFA) that was performed on days 21, 25 and 31 post-infection with ~300 mosquitoes per feed per participant (total of ~14.400 mosquitoes). This information has been added to the Materials and methods and Results sections.

6) Figures 2, 4 and (ideally) 5 should be improved to better convey information (see reviewer 1 and 3 comments).

We have adjusted the figures in the manuscript to accommodate the requests of reviewers 1 and 3. Please also find our responses below.

Reviewer #1:

[…] Introduction, last paragraph. The aim is to "induce stable gametocyte carriage". What is considered to be stable is buried as a footnote to Supplementary file 1. The optimal characteristics (density, duration) are also implied rather than stated.

We agree that this was not clear in the original manuscript. We did not define ‘stable gametocytaemia’ a priori. We have thus modified our wordings in the Materials and methods section on the calculation of gametocyte positive days, peak gametocyte density and area under the curve of gametocyte density versus time. In the Discussion section, we comment on the optimal characteristics of the model, including stable gametocyte carriage, for different use scenarios and the extent to which the current data achieve these optimal characteristics.

The primary study outcome (subsection “Study outcome”) is prevalence of gametocytes by Pfs25 qRT-PCR. Was this the presence of female gametocytes at any of the twice daily measurements from day 6?

The reviewer is correct in this assumption. The prevalence of gametocytes was defined as the presence of female gametocytes as measured by qRT-PCR targeting female-specific Pfs25 mRNA at any of the twice daily measurements from day 6. This information has been added to the Materials and methods subsection “Study outcome”.

The sample size calculation (subsection “Statistical analysis”, first paragraph) appears to set the bar very low. Even so, I could not quite reproduce the calculation: an exact binomial 90% confidence interval for 6/7 would have lower bound 47.6%. It might be that you used an approximation, but since the numbers are small an exact CI seems more appropriate. Why did you choose a 90% confidence interval rather than the more conventional 95%? (with 95% CI, the lower bound for 6/7 would be 42.1%). The numbers are very small, and nowhere is it really explained why more data would be hard to collect.

The sample size calculation was used to plan the study and ensure we had sufficient power to determine whether the majority of volunteers would develop detectable levels of mature gametocytes upon treatment. Controlled Human Malaria Infection (CHMI) studies are logistically challenging and this CHMI study was particularly challenging in terms of the number of study participants that were examined in exploratory mosquito feeding assays. As a consequence, the number of volunteers was informed by both feasibility and study power and a 90% confidence interval was chosen to find a balance between feasibility and the precision in study endpoints.

We used a Wilson confidence interval in the planning phase of this study. The Wilson confidence interval is particularly informative in case of a small number of observations. The Wilson confidence interval is considered an improvement over the normal approximation interval in that the actual coverage probability is closer to the nominal value. We used the command ‘cii 7 6, wilson level(90)’ in STATA, giving the confidence limit of 54.8 if we assumed 6/7 volunteers per arm would develop gametocytes. A 95% CI would have given a limit of 48.7. We have clarified this in the revised Materials and methods section.

Whilst our sample size calculations were optimistic, our study data indicate that this optimism was justified with 100% of all volunteers developing mature gametocytes (16/16, 90% Wilson CI 85.5-100% or 95% CI 80.6-100%.

We have further clarified the limitations of our modest sample size in the Discussion section, especially notable when comparing study arms.

“In the current study, group sizes are limited and comparisons between treatment arms have to be interpreted with caution. […] Our sample size calculation was based on the optimistic assumption that the vast majority of volunteers would develop mature gametocytes; an assumption that was supported by the current data. With our limited study size, our findings indicate…”

There was an odd dissonance between the simple comparisons for hypothesis tests (ANOVA, Fisher's Exact test) and a more sophisticated analysis to "identify which study arm potentially deviated from the others". I'm not sure what this involves since few details were given, but it includes a prior for R^2 which seems unintuitive, and, given the limited sample size, could be stretching the data further than it can bear. It was also written without clarity in the Results e.g. "96.5% probability of being the highest" – the LD-PIP/SP mean was anyway the highest in the trial. Do you mean the chance of the population mean of LD-PIP/SP being the highest?

We have clarified our statistical methods in the revised manuscript. There was no statistically significant difference in time to appearance of gametocytes between the study arms. (p = 0.26). A simple ANOVA for gametocyte density (AUC) did however suggest a statistical significant difference between arms (p = 0.04). Following this conventional frequentist analysis, we used a Bayesian model to determine whether there was evidence for a difference between study arms. With this Bayesian model, this difference was identified as being caused by the LD-PIP/SP arm, which had significantly higher mean gametocyte density (in terms of AUC) than the other three arms. Because the model is Bayesian, we are actually able to calculate the probability that this conclusion is true (i.e. the posterior probability) given the model, which we calculated to be 94.4% (94.0% after correction for asexual density AUC). Regarding the “unintuitive” prior for R^2, we did not add any prior information to the Bayesian model, which in the case of standard linear regression in the package rstanarm takes its prior information in the form of a prior for R^2. Essentially, this means that we allowed model parameters to take on any value and that these values were solely informed by the data.

We now also provide further details in the Results section.

“The LD-PIP/SP arm had significantly higher gametocyte concentrations (area under the curve, AUC) than each of the other three arms with a posterior probability of 99.1% (compared to the LD-SP/SP arm), 98.9% (LD-SP/PIP), and 95.4% (LD-PIP/PIP), respectively. […] After correction for the asexual AUC, the probabilities of the gametocyte AUC in the LD-PIP/SP arm being higher than the other three decreased to 97.2%, 96.3%, and 96.2%, and the probability of LD-PIP/SP being higher than all other groups decreased to 94.0%.”

In addition, we expanded our Bayesian analysis to also look at the number of gametocyte positive days per individual by means of a multi-level logistic regression model (including a random effect for individual deviations from the four group means). The model was fitted to data on the number of days that each individual was tested using for Pfs25mRNA (denominator) and the number of days that individuals tested positive for gametocytes by Pfs25 qRT-PCR (≥ 5 gametocytes/mL) at least once (numerator). Consistent with the results from the AUC analysis, the average proportion of days that individuals in each group tested positive for Pfs25 mRNA was estimated at 27.4% (LD-SP/SP), 35.9% (LD-SP/PIP), 51.4% (LD-PIP/PIP), and 48.3% (LD-PIP/SP). The LD-PIP/PIP and LD-PIP/SP arms (i.e. those receiving “low dose PIP”) each had significantly higher average proportions of gametocyte-positive days than both arms LD-SP/SP and LD-SP/PIP (posterior probability 90.8% and 86.1%, respectively; 81.1% joint probability of arms LD-PIP/PIP and LD-PIP/SP both being higher than both LD-SP/SP and LD-SP/PIP). This information has been added to the Results.

It would be useful to briefly state the assumptions in the previously developed model to estimate the gametocyte half-life.

We have provided the requested details in the revised supplemental information and expanded the Materials and methods section. The model was a mixed effects model that had no covariates (other than time) fit in SAS 9.4. The annotated model code, including all assumptions are now provided in detail.

It is not obvious that the AUC is the most useful measure – a short spike in gametocytes could be equal to longer low-level gametocytaemia. Clarification of what the optimal characteristics for drug development are would help. The AUC as a measure may also suffer from the non-independence of each time-point from the previous time-points – from a high peak (which may have occurred by chance or not due to the drugs), then further high values might tend to follow. A suggestion only: it might be possible to take into account the non-independence between time-points and use all the data, by using a statistical time-series model with a lag from the asexual densities or a simple mechanistic model fitted to the data. You might need more than four participants per arm for that but you might also be able to borrow strength for some parameters.

We appreciate these comments that make a case for a more comprehensive model of gametocyte commitment, maturation and longevity as a function of asexual parasitaemia. This is a longer term objective of our group once we have optimized our model that will require a larger number of participants to accurately parameterize gametocyte commitment and release rates. For the current purposes (i.e. to establish the CHMI transmission model, to select an optimal treatment regimen and to determine the time of first appearance of gametocytes), we consider the presented analyses appropriate.

For the comparison of drug regimens, we consider the AUC a useful summary measure as gametocyte density patterns looked very similar between individuals (i.e. similar levels of variation over time and not different types of patterns of variation, e.g. “short spikes” vs. “continuous stable levels”).We have, however, also incorporated analyses on the number of gametocyte positive days (showing the same patterns between groups as the AUC) and provide a more detailed presentation of individual parasite and gametocyte curves.

Importantly, and this was not clearly described in the original submission, a future study will examine infectivity to mosquitoes in more detail. For that, the current data are crucial since they inform us on the time till gametocyte appearance and the plateau phase in gametocyte density. A short spike in gametocytes (without a plateau phase of gametocyte densities that are sufficiently high to measure gametocyte clearance or infectivity) would be operationally very challenging since it would be nearly impossible to define the optimal moment for mosquito feeding assays and/or initiation of gametocytocidal drugs. This is clarified in the revised Discussion section.

The reviewer’s concern of correlation over time within individuals would indeed be an issue when one would consider the observed densities at individual time points as independent identically distributed variables. For this specific reason, we reduced the time series data to one data point per individual (i.e. the AUC), which essentially relaxes the assumption of inter individual differences observations. So in our analysis, despite the fact that the AUC may be higher in some individuals by shear chance (which the model captures in the error term of the linear predictor), the posterior probability that the AUC is highest in the LD-PIP/SP arm on average is still a very convincing 94.4% (if groups where identical, one would expect a posterior probability of about 1 / 4 = 25%).

The Results section focuses on differences between the arms but it is unfortunate that the AUC is the only reported measure of gametocytaemia which is (borderline) significant between the arms for gametocytaemia (or the other p-values are just not mentioned). Looking at the graphs, I think the effect is plausible, but there is very limited evidence due to the small sample size.

As indicated above, we have clarified the limitations of our study size in the Discussion section. We have further extended our analysis to the proportion of days that individuals test positive for gametocytes, showing that individuals in study arms LD-PIP/PIP and LD-PIP/SP on average experience more gametocyte-positive days, with little difference between LD-PIP/PIP and LD-PIP/SP. The previous AUC analysis then further shows that individuals in LD-PIP/SP have higher gametocyte concentrations than in all other groups.

We would also like to clarify that the finding of 94.4%% probability is not borderline as it represents a posterior probability, i.e. the probability that there really is a difference between groups. In a Bayesian context, if in reality there is no real difference between four groups (and this is reflected by the data), the posterior probability of one group being the highest would be estimated around 25%. Therefore, our estimate of 94.4% probability of group 4 being the highest is very convincing.

That three mosquitoes were infected cannot be interpreted without the numbers of mosquitoes fed (Results, fifth paragraph).

We have now provided full details of this exploratory endpoint. Expressed as a proportion of all examined mosquitoes, 0.0002% (3/14400) of mosquitoes became infected in these exploratory assessments.

The male and female gametocytes were measured by different assays which may have different levels of detection. The results for the sex-ratio should be interpreted with caution.

We fully agree. We have now provided full details on our qRT-PCR performance, sensitivity and differences in sensitivity between arms. We have also rephrased our results on the sex ratio to highlight uncertainties with these estimates that rely on a ratio of two separate qRT-PCRs.

The CONSORT guidelines for reporting should be followed for eLife. Since this is a phase 1/2 trial with a small sample of healthy participants, the most appropriate CONSORT guidelines would be: CONSORT 2010 statement: extension to randomised pilot and feasibility trials (http://www.bmj.com/content/355/bmj.i5239). The reporting guidelines cover design, randomization, intervention, sample size, bias, generalisability and access to the pilot protocol.

We thank the reviewer for its comment and have added the CONSORT extension for Pilot and Feasibility Trials Checklist to the dossier, and adjusted the manuscript as per CONSORT guidelines. In addition, the clinical trial protocol has been added to the submission.

Given that the emphasis in the results is on the differences between arms, Figure 4, where all groups are combined, is not easy to interpret.

We appreciate this comment. Figure 4 represents the total female and male gametocyte density of all participants to give an impression of the female-biased sex-ratio per time-point and its time course. Figure 4—figure supplement 1, shows the individual PCR data per study arm. In addition, we have now provided an illustration of male and female gametocyte clearance dynamics for individual study participants (Figure 4—figure supplement 2).

There is little mention of the next steps for the induced gametocytes. If they are to be used in testing interventions that reduce gametocyte development, how would they be used?

We have now better explained this in the Discussion section. We describe the ideal gametocyte characteristics for different use scenarios of the model (e.g. gametocytocidal drugs or transmission-blocking vaccines) and end the Discussion section with a paragraph on future scenarios.

“For the evaluation of gametocytocidal interventions in the CHMI transmission model, gametocyte densities should be sufficiently high to quantify an intervention-associated reduction in gametocyte appearance or gametocyte half-life. […] Low infectivity in membrane feeding assays may be overcome by achievement of higher gametocyte densities in the model, and the use of gametocyte concentration methods (Reuling et al., 2017), or by direct skin feeding assays (Bousema et al., 2012).”

And:

“Here, we present a novel CHMI transmission model for P. falciparum that can be used to study gametocyte biology and dynamics providing novel insights and tools in malaria transmission and elimination efforts. […] The current work lays the foundation for fulfilling the critical unmet need to evaluate transmission-blocking interventions against falciparum malaria for downstream selection and clinical development.’

It is not discussed why the different combinations of drug might have different effects.

We have clarified this in the revised Discussion section.

“With our limited study size, our findings indicate that none of the study drugs prevented the appearance of gametocytes after treatment, thereby suggesting limited or no effect of PIP and SP on developing or mature gametocytes (Bolscher et al., 2015). […] One hypothesis that may explain differences between study arms would be similar gametocyte commitment in all arms after T1 but a more rapid release of gametocytes that accumulated in the bone marrow between T1 and T2.”

Reviewer #2:

This is a very interesting and informative study of Plasmodium falciparum malaria gametocyte dynamics which exploits the opportunities provided by the resurgence of interest in human challenge studies and the development of methodologies for sensitive quantitation of nucleic acid concentrations. In general this is a very good piece of work but more details on the validation of qPCR gametocyte quantitation are essential if it is to be published. If there is one disappointment it is the complete absence of reference to the early observations of gametocytaemia in human challenge studies – notably the work of Shute and Cuica whose conclusions were in broad agreement with the current paper. The earliest reference is from 1986.

We appreciate this comment. One of the joys of malaria research is that there is a wealth of excellent early data and we agree that this deserves referencing and attention in the manuscript. We have now added these references in the manuscript and described some of this excellent early work in detail in the Introduction., e.g.:

“Early work based on the microscopic evaluation therapeutic controlled P. falciparum infections reported that gametocytes may make their appearance in small numbers around 10 days following the first day of fever (Shute and Maryon, 1951; Ciuca, Chearescu and Lavrinenko, 1937).”

A list of questions or comments in the order they appear in the manuscript:

1) "It is widely accepted that malaria elimination is unlikely to be attainable in the majority of endemic settings with currently available resources and tools". Perhaps this could be attenuated? Many think the obstacles are primarily political, organisational and financial.

We agree with the reviewer and have adjusted the sentence:

“Novel interventions may support malaria elimination efforts in endemic settings (Griffin et al., 2010) that are further dependent on political and financial commitments to maximize coverage with currently available interventions and improve surveillance systems to optimize disease notification and treatment (Moonen et al., 2010).”

2) "maturation of gametocytes takes place predominantly in the bone marrow".

We agree with the reviewer and have adjusted this sentence:

“maturation of gametocytes takes place predominantly in the bone marrow”.

3) The 3D7 "lineage" is well known, but it is also well known to be different in different laboratories! Perhaps a few additional sentences on this particular lineage would be valuable. In particular whether there is any evidence that serial passage through volunteers and these laboratory reared mosquitoes has altered its biological properties – notably infectivity over the years? Please also confirm wild type PfDHFR and DHPS and single copy plasmepsin 2/3.

The 3D7 lineage that was used in the current study is based on a 3D7 bank described in detail in Cheng et al., AJTMH 1997. The material has thus not been in constant passage. Whilst differences in infectivity exist between parasite lines (McCall, Sci Transl Med 2017 as recent example), the material we used was generating gametocyte densities not dissimilar from our routine NF54 line and we achieved high mosquito infection rates and oocyst and sporozoite densities for inoculation in the current study.

We have clarified the origin of the 3D7 material, including the original reference, in the revised manuscript. In addition, we used available Illumina Whole genome sequencing data for this lineage (https://www.ebi.ac.uk/ena/data/view/PRJEB12838). Reads were aligned to the P. falciparum reference genome v3 (plasmoDB) with bowtie2 (sourceforge) and consensus sequences for dhps and dhfr genes were obtained with samtools. Read depth (coverage) was assessed using bedtools. With this approach, we determined dhfr/dhps mutations in our 3D7 parasites; only one mutated locus in the dhps gene was found (PfDHPS A437G), which is not associated with treatment failure as single mutation (Staedke et al. 2004). WGS analysis thus rules out sulfadoxine-pyrimethamine drug resistance of the 3D7 lineage we used. Plasmepsin II/III duplication events are associated with piperaquine resistance (Witkowski et al., 2017) but were not found in the 3D7 WGS data. However, as the neighboring genes Plasmepsin I and IV have sequence similarities, unambiguous quantification is challenging and a final assessment would involve more specific targeting, e.g. by qPCR.

gene	locus	3D7 codon	amino acid	mutated
DHPS	A437G	GGT	G	yes
DHPS	K540E	AAA	K	no
DHFR	N51I	AAT	N	no
DHFR	C59R	TGT	C	no
DHFR	S108N	AGC	S	no

Open in a new tab

Importantly, piperaquine sensitivity of the 3D7 lineage was previously confirmed by in vivo experiments (Pasay et al., 2016).

We included the following information in the revised manuscript in the section describing the parasite lineage:

“The 3D7 lineage that was used in the current study is based on a 3D7 bank described in detail in Cheng et al., (Cheng et al., 1997). […] We conclude that the lineage used was sensitive to both sulfadoxine-pyrimethamine and piperaquine.”

4) "until malaria parasites were detected at a density of {greater than or equal to}5000 parasites per milliliter (Pf/mL) by qPCR or a positive thick blood smear" – were any volunteers symptomatic at this density?

The majority of the volunteers became symptomatic (1-2 day) before the initiation of treatment. As expected from previous CHMI studies the majority of symptoms were experienced after the first treatment. The time course of adverse events is shown in Figure 6B, and Supplementary file 1. shows the day of the first treatment per volunteer.

5) Piperaquine base or piperaquine phosphate?

We have now added the full nonproprietary name of piperaquine phosphate to the Materials and methods – subsection “Procedures”.

6) Safety seems rightly to have been a major concern. Did anyone look at the ECG if metoclopramide was given to piperaquine recipients? Was there any additional QTc prolongation?

Piperaquine recipients received an ECG, 4-12 hours after treatment of each dose of piperaquine (based on the expected maximal piperaquine concentrations after oral dosing). The maximum dose of piperaquine used in our study is approximately only 1/3 of the curative dose used in the dihydroartemisinin piperaquine combination therapy (label use), therefore, the risk of QT-prolongations in this study, next to administration on an empty stomach (to reduce the peak concentrations) is assessed to be very low. The combination of the very rarely described QT-prolongations with metoclopramide and the sporadic, and mostly single-dose (10mg) metoclopramide use in our study participants made an additional risk of QT-prolongations highly unlikely. Therefore, additional ECG assessments after metoclopramide were not deemed necessary at discretion of the investigators. No QT-prolongations were found after or between piperaquine treatments in this study.

7) The section on parasite DNA and RNA quantitation needs considerable strengthening as it is the central component of the paper. This is important.a) The performance characteristics of the 18S DNA and mRNA assays needs presenting in detail or references provided to detailed validation – accuracy at different densities, reproducibility, linearity, criteria for limits of detection etc.

We appreciate this comment and have provided all details of our molecular assay performance in the revised manuscript and supporting information. We agree that this is essential, especially since the lower sensitivity of the male-specific PfMGET qRT-PCR is relevant for the interpretation of the study results. Our assay characteristics that are now also included in the revision (Figure 5, Supplementary file 2, Supplementary file 3, Figure 5—figure supplement 1, Figure 5—figure supplement 2). The limit of detection was defined as the lowest concentration where all trendline samples were still positive, the limit of quantification was defined as the lowest concentration where the coefficient of variation was <15% and where duplicate study samples showed good agreement (Figure 5—figure supplement 2).

b) The volume of blood taken and assayed needs stating.

We have now added more information on the molecular assays used in the study, including the volume of blood used per assay.

“For the quantification of the P. falciparum Pfs25 transcript levels total NA was RQ1 DNaseI treated according to the manufacturer’s protocol. […] Detailed information on the validation and performance characteristics of the assays can be found in the supporting materials (Figure 5; Supplementary file 2, 3; Figure 5—figure supplement 1, 2).”

c) Validation of gametocyte quantitation needs support – particularly with reference to stability of transcript numbers per cell and any assumptions made in the derivation of a gametocyte density.

Full details on the stability of transcripts are now provided in the supplemental information to the revised manuscript. In our original presentation of the methodology (Stone et al., 2017) we determined the stability of transcripts in maturing male and female gametocytes. Culturing of NF54 was done without prior gametocyte sex-sorting and female (top panel) and male (bottom panel) transcripts were quantified in the same source culture material using Illumina RNAseq with paired reads, 37bp in length. Transcript counts after normalization for library size are presented for d4 till d16 after synchronization; stage IV gametocytes were observed on day 5 and stage V gametocytes on day 7. These results indicate that for both male and female markers transcript abundance is low in stage IIb gametocytes and peaks in stage V gametocytes after which transcript numbers appear stable (Pfs25) or may decline at late time-points (PfMGET).

8) Bousema et al., 2010 is referred to – which if I understand correctly assumes a single first order decline in gametocyte densities. Is this justified? It would be valuable to describe, if possible, the residuals around the model fits and comment on any heteroscedasticity. This is particularly important considering that a major finding of this report is that male gametocytes appear to be cleared more rapidly than female gametocytes. If in fact the gametocyte clearance profile is more complex (e.g. multiexponential) then the lower density male gametocytes may appear to be more rapidly eliminated whereas, in fact, the slower terminal phase of elimination is below the limit of accurate detection. A similar problem has bedevilled assessment of the pharmacokinetic properties of slowly eliminated antimalarials.

The reviewer is correct that the model assumes a single first order decline in male gametocyte densities. The decline is on the linear scale; but the error structure is a Tobit distribution on the log scale (the SAS code is now provided in the supplemental information). The fact that the error structure is on the log scale means that the heteroscedasticity is less likely to be a problem than if the errors were on the linear scale. The residuals for the non-zero female densities is displayed in Author response image 1.

Author response image 1. — Log female gametocyte densities at different days of follow-up (left) and residuals of non-zero female log gametocyte densities as a function of time of follow-up (right).

The left panel shows a linear decline on the log scale consistent with the model. The right panel shows no evidence of heteroscedasticity. The residuals of non-zero measurements are larger to the right of the graph, but this reflects the fact that zero measurements are more likely in that region and only residuals for non-zero measurements are shown.

For female gametocytes 74% (98/133) of observations were non-zero so it is straightforward to assess heteroscedasticity. For males the proportion of non-zero measurements was much lower (28%, 38/135) so it is much harder to assess the fit of the model and in particular heteroscedasticity. The equivalent graph is shown in Author response image 2.

Author response image 2. — Log male gametocyte densities at different days of follow-up (left) and residuals of non-zero male log gametocyte densities as a function of time of follow-up (right).

The left panel of figure shows a linear decline on the log scale consistent with the model, but the smaller number of non-zero points makes it harder to assess. There is no evidence of heteroscedasticity in the right panel but there is a clear trend of larger residuals to the right of the graph. But this is explained by the fact there are more zero measurements in this region. To assess whether the model would show a better fit to the data if male gametocytes had been higher, we re-analysed data from a trial in high density gametocyte carriers in Mali (NCT02831023; Dicko et al. Lancet Infect Dis in press) that used the same PfMGET qRT-PCR. We restricted analysed to the treatment arms in this study that received non-gametocytocidal drugs dihydroartemisinin-piperaquine or amodiaquine plus sulfadoxine-pyrimethamine.

Author response image 3. — Log male gametocyte densities at different days of follow-up (left) and residuals of non-zero male log gametocyte densities as a function of time of follow-up (right).

In this trial, with a much larger number of non-zero values for PfMGET male gametocytes, model fit was appropriate and not different from the above model fit for Pfs25 female gametocytes.

In summary, the large number of zero measurements make it hard to assess the fit of the model to the male gametocyte densities, but we think its use if justified in the case of males by 1) The fact the model fits the female gametocyte densities well; 2) The fact the model fits the male gametocyte densities in a trial with high starting gametocyte densities well; 3) The model has been widely applied in the past to many other datasets with good fit. We accept that this model, like all models, relies on assumptions and we have made this clearer in the Discussion. We have also plotted male and female gametocyte decay curves for all volunteers with >. We observe that whilst individual gametocyte clearance lines appear to be parallel for some volunteers (Figure 4—figure supplement 2), others show evidence for a faster disappearance of male gametocytes.

9) It is not very clear what modelling approach was used to fit the model – was this a mixed effect model? Were any covariates incorporated – if so which? What programme was used?

We have now added this information in the Materials and methods section. The model was a mixed effect model that included time as only covariate. The code is now given in the supporting information so the methodology in accessible for everyone in SAS and the model assumptions are described in detail.

10) Is there any explanation for the liver function test abnormalities?

There is no clear explanation for the transient elevated transaminases in CHMI studies. There is no clear relationship between parasitemia densities and liver abnormalities, and it also seems unlikely to be directly related to anti-malarials due to variety of drugs used in CHMI. Similarly, the use of paracetamol does not support a clear relationship. Rather a combination of the above mentioned factors, and individual susceptibility may have triggered the observed abnormalities. The transient asymptomatic liver function test abnormalities are most likely directly related to the asexual parasitemia, and subsequent treatment. The exact mechanism of the transient organ injury during the malaria and its treatment is yet to be understood and ongoing work. We are currently reviewing all our CHMI data, including the many trials where treatment was initiated at lower parasite densities, and intend to write this up as a separate manuscript to describe the prevalence, severity and transient nature of liver function abnormalities upon controlled malaria infections.

11) "SP has long been associated with a rapid appearance of gametocytes that is too early to be explained by de novo gametocyte production upon drug pressure and has thus been hypothesized to reflect an efflux of sequestered gametocytes upon treatment". True – but it is not very clear from the text whether this study's results supports this hypothesis. Could the authors be explicit here? The implication is that such early released gametocytes should be more immature- and thus the period during which Plasmodium falciparum gametocytes are not infectious (after release into the circulation) should be longer in these circumstances. Is there any evidence for this?

We agree with the reviewer that early released gametocytes are possibly more immature, and therefore, possibly, less infectious. However, our preliminary mosquito infectivity results do not provide any evidence for this since its very low mosquito infection rates. We have now added additional information in the Discussion on the hypothesis of an efflux of sequestered gametocytes upon treatment:

“We hypothesized that slow acting drugs may promote the development of gametocytes (Mendez et al., 2002), potentially via microvesicles that are derived from infected erythrocytes (Nilsson et al., 2015) and differences between drug regimens in the rate at which asexual parasites are cleared upon T1 and T2 would result in different gametocyte dynamics. […] One hypothesis would be similar gametocyte commitment in all arms after T1 but a more rapid release of gametocytes that accumulated in the bone marrow between T1 and T2.’

12) "The highly abundant Pfs25 mRNA makes the female gametocyte qRT-PCR more sensitive than the male PfMGET qRT-PCR." Could this be elaborated upon- or at least referenced?

We clarified this in response to the earlier comment on assay performance.

Reviewer #3:

[…] The paper is, in most respects, clearly written. There are, however, some aspects of reporting where it is not entirely clear exactly what was done and why (particularly in relation to statistical methods).

Results, fifth paragraph. Details of these membrane feeding experiments are lacking from the Materials and methods. In particular, how many mosquitos on each feeding day? Even if full details are presented in the references describing the protocols for these experiments, it would be useful to summarise some of the key numbers in the manuscript to give the reader an idea of what these numbers mean. See also Discussion, third paragraph – "the very low rate". Since only the numerators are reported, there is no information on what this rate actually is.

We have now presented all data, including an estimate of the (very low) proportion of infected mosquitoes.

Subsection “Statistical analysis”, second paragraph. More details need to be given here. What non-parametric tests were used and what exactly was the statistical model implemented in the Bayesian framework (code for this does not appear to have been provided). Also, the R package rstanarm is mentioned, but appears not to be acknowledged in the references. It probably should be along with the appropriate references for Stan and R.

We have added details to the Materials and methods – statistical analysis, including references and codes used. We used a standard linear regression model (no mixed effects) for the Bayesian model for gametocyte density AUC as there was only one data point (the AUC) considered per individual. The linear regression model reported in the Results section only included trial arm as a predictor. Findings did not change when individuals’ asexual density AUC was included as predictor: the probability of the LD-PIP/SP group having a higher gametocyte density AUC was still 94.0%, compared to 94.4% when asexual density AUC was not included as a predictor.

Table 3 is confusing. Why are some adverse events listed twice in the leftmost column? Duration time units should be given.

We have adjusted Table 3 accordingly, with shading, and underlines for clarification. Furthermore, the duration time units (in days) are added to the table.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Citations

Reuling IJ, Bousema T, Sauerwein RW. 2017. Data from: Induction of Plasmodium falciparum gametocytemia in the Controlled Human Malaria Infection model: a randomised trial comparing four antimalarial drug regimens. Available at Dryad Digital Repository under a CC0 Public Domain Dedication. [DOI]

Supplementary Materials

Table 1—source data 1. Source data for Table 1.

elife-31549-table1-data1.xlsx^{(10.6KB, xlsx)}

Source code 1. R codes used for Bayesian statistical analysis.

elife-31549-code1.zip^{(9.9KB, zip)}

Source code 2. SAS code used for estimation of gametocyte half-life for gametocytes.

elife-31549-code2.docx^{(16KB, docx)}

Supplementary file 1. Individual data of the participants included in analysis.

elife-31549-supp1.docx^{(16.5KB, docx)}

Supplementary file 2. Selected P. falciparum gene targets and primers of qRT PCR assays.

elife-31549-supp2.docx^{(11.7KB, docx)}

Supplementary file 3. Quality parameters of qRT PCR and qPCR.

elife-31549-supp3.docx^{(12KB, docx)}

Supplementary file 4. Clinical trial protocol.

elife-31549-supp4.pdf^{(480KB, pdf)}

Reporting standard 1. CONSORT extension for Pilot and Feasibility Trials Checklist.

elife-31549-repstand1.docx^{(41KB, docx)}

Transparent reporting form

elife-31549-transrepform.docx^{(246.7KB, docx)}

Data Availability Statement

The following dataset was generated:

[bib1] Adjalley SH, Johnston GL, Li T, Eastman RT, Ekland EH, Eappen AG, Richman A, Sim BK, Lee MC, Hoffman SL, Fidock DA. Quantitative assessment of Plasmodium falciparum sexual development reveals potent transmission-blocking activity by methylene blue. PNAS. 2011;108:E1214–E1223. doi: 10.1073/pnas.1112037108. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib2] Bolscher JM, Koolen KM, van Gemert GJ, van de Vegte-Bolmer MG, Bousema T, Leroy D, Sauerwein RW, Dechering KJ. A combination of new screening assays for prioritization of transmission-blocking antimalarials reveals distinct dynamics of marketed and experimental drugs. Journal of Antimicrobial Chemotherapy. 2015;70:1357–1366. doi: 10.1093/jac/dkv003. [DOI] [PubMed] [Google Scholar]

[bib3] Bousema T, Churcher TS, Morlais I, Dinglasan RR. Can field-based mosquito feeding assays be used for evaluating transmission-blocking interventions? Trends in Parasitology. 2013;29:53–59. doi: 10.1016/j.pt.2012.11.004. [DOI] [PubMed] [Google Scholar]

[bib4] Bousema T, Dinglasan RR, Morlais I, Gouagna LC, van Warmerdam T, Awono-Ambene PH, Bonnet S, Diallo M, Coulibaly M, Tchuinkam T, Mulder B, Targett G, Drakeley C, Sutherland C, Robert V, Doumbo O, Touré Y, Graves PM, Roeffen W, Sauerwein R, Birkett A, Locke E, Morin M, Wu Y, Churcher TS. Mosquito feeding assays to determine the infectiousness of naturally infected Plasmodium falciparum gametocyte carriers. PLoS One. 2012;7:e42821. doi: 10.1371/journal.pone.0042821. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib5] Bousema T, Drakeley C. Epidemiology and infectivity of Plasmodium falciparum and Plasmodium vivax gametocytes in relation to malaria control and elimination. Clinical Microbiology Reviews. 2011;24:377–410. doi: 10.1128/CMR.00051-10. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib6] Bousema T, Okell L, Shekalaghe S, Griffin JT, Omar S, Sawa P, Sutherland C, Sauerwein R, Ghani AC, Drakeley C. Revisiting the circulation time of Plasmodium falciparum gametocytes: molecular detection methods to estimate the duration of gametocyte carriage and the effect of gametocytocidal drugs. Malaria Journal. 2010;9:136. doi: 10.1186/1475-2875-9-136. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib7] Butcher GA. Antimalarial drugs and the mosquito transmission of Plasmodium. International Journal for Parasitology. 1997;27:975–987. doi: 10.1016/S0020-7519(97)00079-9. [DOI] [PubMed] [Google Scholar]

[bib8] Cheng Q, Lawrence G, Reed C, Stowers A, Ranford-Cartwright L, Creasey A, Carter R, Saul A. Measurement of Plasmodium falciparum growth rates in vivo: a test of malaria vaccines. The American Journal of Tropical Medicine and Hygiene. 1997;57:495–500. doi: 10.4269/ajtmh.1997.57.495. [DOI] [PubMed] [Google Scholar]

[bib9] Ciuca MBL, Chearescu M, Lavrinenko N. Contribution À L’étude De L’infection Expérimentale Au Plasmodium Falciparum. Hamburg: Festchrift Institute Bernard Nocht; 1937. pp. 81–101. [Google Scholar]

[bib10] Delves MJ, Ruecker A, Straschil U, Lelièvre J, Marques S, López-Barragán MJ, Herreros E, Sinden RE. Male and female Plasmodium falciparum mature gametocytes show different responses to antimalarial drugs. Antimicrobial Agents and Chemotherapy. 2013;57:3268–3274. doi: 10.1128/AAC.00325-13. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib11] Duffier Y, Lorthiois A, Cisteró P, Dupuy F, Jouvion G, Fiette L, Mazier D, Mayor A, Lavazec C, Moreno Sabater A. A humanized mouse model for sequestration of Plasmodium falciparum sexual stages and in vivo evaluation of gametocytidal drugs. Scientific Reports. 2016;6:35025. doi: 10.1038/srep35025. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib12] Eichner M, Diebner HH, Molineaux L, Collins WE, Jeffery GM, Dietz K. Genesis, sequestration and survival of Plasmodium falciparum gametocytes: parameter estimates from fitting a model to malariatherapy data. Transactions of the Royal Society of Tropical Medicine and Hygiene. 2001;95:497–501. doi: 10.1016/S0035-9203(01)90016-1. [DOI] [PubMed] [Google Scholar]

[bib13] Farid R, Dixon MW, Tilley L, McCarthy JS. Initiation of gametocytogenesis at very low parasite density in Plasmodium falciparum infection. The Journal of Infectious Diseases. 2017;215:1167–1174. doi: 10.1093/infdis/jix035. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib14] Gonçalves BP, Drakeley C, Bousema T. Infectivity of microscopic and submicroscopic malaria parasite infections in areas of low malaria endemicity. Journal of Infectious Diseases. 2016;213:1516–1517. doi: 10.1093/infdis/jiw044. [DOI] [PubMed] [Google Scholar]

[bib15] Graumans W, Tadesse FG, Andolina C, van Gemert GJ, Teelen K, Lanke K, Gadisa E, Yewhalaw D, van de Vegte-Bolmer M, Siebelink-Stoter R, Reuling I, Sauerwein R, Bousema T. Semi-high-throughput detection of Plasmodium falciparum and Plasmodium vivax oocysts in mosquitoes using bead-beating followed by circumsporozoite ELISA and quantitative PCR. Malaria Journal. 2017;16:356. doi: 10.1186/s12936-017-2011-9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib16] Griffin JT, Hollingsworth TD, Okell LC, Churcher TS, White M, Hinsley W, Bousema T, Drakeley CJ, Ferguson NM, Basáñez MG, Ghani AC. Reducing Plasmodium falciparum malaria transmission in Africa: a model-based evaluation of intervention strategies. PLoS Medicine. 2010;7:e1000324. doi: 10.1371/journal.pmed.1000324. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib17] Griffin P, Pasay C, Elliott S, Sekuloski S, Sikulu M, Hugo L, Khoury D, Cromer D, Davenport M, Sattabongkot J, Ivinson K, Ockenhouse C, McCarthy J. Safety and reproducibility of a clinical trial system using induced blood stage plasmodium vivax infection and its potential as a model to evaluate malaria transmission. PLOS Neglected Tropical Diseases. 2016;10:e0005139. doi: 10.1371/journal.pntd.0005139. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib18] Hermsen CC, Telgt DS, Linders EH, van de Locht LA, Eling WM, Mensink EJ, Sauerwein RW. Detection of Plasmodium falciparum malaria parasites in vivo by real-time quantitative PCR. Molecular and Biochemical Parasitology. 2001;118:247–251. doi: 10.1016/S0166-6851(01)00379-6. [DOI] [PubMed] [Google Scholar]

[bib19] Kafsack BF, Rovira-Graells N, Clark TG, Bancells C, Crowley VM, Campino SG, Williams AE, Drought LG, Kwiatkowski DP, Baker DA, Cortés A, Llinás M. A transcriptional switch underlies commitment to sexual development in malaria parasites. Nature. 2014;507:248–252. doi: 10.1038/nature12920. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib20] Lasonder E, Rijpma SR, van Schaijk BC, Hoeijmakers WA, Kensche PR, Gresnigt MS, Italiaander A, Vos MW, Woestenenk R, Bousema T, Mair GR, Khan SM, Janse CJ, Bártfai R, Sauerwein RW. Integrated transcriptomic and proteomic analyses of P. falciparum gametocytes: molecular insight into sex-specific processes and translational repression. Nucleic Acids Research. 2016;44:6087–6101. doi: 10.1093/nar/gkw536. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib21] Laurens MB, Duncan CJ, Epstein JE, Hill AV, Komisar JL, Lyke KE, Ockenhouse CF, Richie TL, Roestenberg M, Sauerwein RW, Spring MD, Talley AK, Moorthy VS, Consensus Group on Design of Clinical Trials of Controlled Human Malaria Infection A consultation on the optimization of controlled human malaria infection by mosquito bite for evaluation of candidate malaria vaccines. Vaccine. 2012;30:5302–5304. doi: 10.1016/j.vaccine.2012.04.088. [DOI] [PubMed] [Google Scholar]

[bib22] Lensen A, Mulder L, Tchuinkam T, Willemsen L, Eling W, Sauerwein R. Mechanisms that reduce transmission of Plasmodium falciparum malaria in semiimmune and nonimmune persons. The Journal of Infectious Diseases. 1998;177:1358–1363. doi: 10.1086/515263. [DOI] [PubMed] [Google Scholar]

[bib23] Moonen B, Cohen JM, Snow RW, Slutsker L, Drakeley C, Smith DL, Abeyasinghe RR, Rodriguez MH, Maharaj R, Tanner M, Targett G. Operational strategies to achieve and maintain malaria elimination. The Lancet. 2010;376:1592–1603. doi: 10.1016/S0140-6736(10)61269-X. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib24] Méndez F, Muñoz A, Carrasquilla G, Jurado D, Arévalo-Herrera M, Cortese JF, Plowe CV. Determinants of treatment response to sulfadoxine-pyrimethamine and subsequent transmission potential in falciparum malaria. American Journal of Epidemiology. 2002;156:230–238. doi: 10.1093/aje/kwf030. [DOI] [PubMed] [Google Scholar]

[bib25] Nilsson SK, Childs LM, Buckee C, Marti M. Targeting human transmission biology for malaria elimination. PLoS Pathogens. 2015;11:e1004871. doi: 10.1371/journal.ppat.1004871. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib26] Ouédraogo AL, Guelbéogo WM, Cohuet A, Morlais I, King JG, Goncalves BP, Bastiaens GJH, Vaanhold M, Sattabongkot J, Wu Y, Coulibaly M, Ibrahima B, Jones S, Morin M, Drakeley C, Dinglasan RR, Bousema T. A protocol for membrane feeding assays to determine the infectiousness of P. falciparum naturally infected individuals to Anopheles gambiae. MalariaWorld Journal. 2013;4:16. doi: 10.5281/zenodo.10926272. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib27] Pasay CJ, Rockett R, Sekuloski S, Griffin P, Marquart L, Peatey C, Wang CY, O'Rourke P, Elliott S, Baker M, Möhrle JJ, McCarthy JS. Piperaquine monotherapy of drug-susceptible plasmodium falciparum infection results in rapid clearance of parasitemia but is followed by the appearance of gametocytemia. Journal of Infectious Diseases. 2016;214:105–113. doi: 10.1093/infdis/jiw128. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib28] Pett H, Gonçalves BP, Dicko A, Nébié I, Tiono AB, Lanke K, Bradley J, Chen I, Diawara H, Mahamar A, Soumare HM, Traore SF, Baber I, Sirima SB, Sauerwein R, Brown J, Gosling R, Felger I, Drakeley C, Bousema T. Comparison of molecular quantification of Plasmodium falciparum gametocytes by Pfs25 qRT-PCR and QT-NASBA in relation to mosquito infectivity. Malaria Journal. 2016;15:539. doi: 10.1186/s12936-016-1584-z. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib29] Ponnudurai T, Lensen AH, Meis JF, Meuwissen JH. Synchronization of Plasmodium falciparum gametocytes using an automated suspension culture system. Parasitology. 1986;93:263–274. doi: 10.1017/S003118200005143X. [DOI] [PubMed] [Google Scholar]

[bib30] Ponnudurai T, Lensen AH, Van Gemert GJ, Bensink MP, Bolmer M, Meuwissen JH. Infectivity of cultured Plasmodium falciparum gametocytes to mosquitoes. Parasitology. 1989;98:165–173. doi: 10.1017/S0031182000062065. [DOI] [PubMed] [Google Scholar]

[bib31] Reuling IJ, Stone WJR, van de Vegte-Bolmer M, van Gemert GJ, Siebelink-Stoter R, Graumans W, Lanke K, Bousema T, Sauerwein RW. Concentration of Plasmodium falciparum gametocytes in whole blood samples by magnetic cell sorting enhances parasite infection rates in mosquito feeding assays. Malaria Journal. 2017;16:315. doi: 10.1186/s12936-017-1959-9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib32] Sauerwein RW, Roestenberg M, Moorthy VS. Experimental human challenge infections can accelerate clinical malaria vaccine development. Nature Reviews Immunology. 2011;11:57–64. doi: 10.1038/nri2902. [DOI] [PubMed] [Google Scholar]

[bib33] Schneider P, Schoone G, Schallig H, Verhage D, Telgt D, Eling W, Sauerwein R. Quantification of Plasmodium falciparum gametocytes in differential stages of development by quantitative nucleic acid sequence-based amplification. Molecular and Biochemical Parasitology. 2004;137:35–41. doi: 10.1016/j.molbiopara.2004.03.018. [DOI] [PubMed] [Google Scholar]

[bib34] Shute PG, Maryon M. A study of gametocytes in a West African strain of Plasmodium falciparum. Transactions of the Royal Society of Tropical Medicine and Hygiene. 1951;44:421–438. doi: 10.1016/S0035-9203(51)80020-8. [DOI] [PubMed] [Google Scholar]

[bib35] Sinden RE. Developing transmission-blocking strategies for malaria control. PLoS Pathogens. 2017;13:e1006336. doi: 10.1371/journal.ppat.1006336. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib36] Staedke SG, Sendagire H, Lamola S, Kamya MR, Dorsey G, Rosenthal PJ. Relationship between age, molecular markers, and response to sulphadoxine-pyrimethamine treatment in Kampala, Uganda. Tropical Medicine and International Health. 2004;9:624–629. doi: 10.1111/j.1365-3156.2004.01239.x. [DOI] [PubMed] [Google Scholar]

[bib37] Stone W, Grabias B, Lanke K, Zheng H, Locke E, Diallo D, Birkett A, Morin M, Bousema T, Kumar S. A comparison of Plasmodium falciparum circumsporozoite protein-based slot blot and ELISA immuno-assays for oocyst detection in mosquito homogenates. Malaria Journal. 2015;14:451. doi: 10.1186/s12936-015-0954-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib38] Stone W, Sawa P, Lanke K, Rijpma S, Oriango R, Nyaurah M, Osodo P, Osoti V, Mahamar A, Diawara H, Woestenenk R, Graumans W, van de Vegte-Bolmer M, Bradley J, Chen I, Brown J, Siciliano G, Alano P, Gosling R, Dicko A, Drakeley C, Bousema T. A molecular assay to quantify male and female plasmodium falciparum gametocytes: results from 2 randomized controlled trials using primaquine for gametocyte clearance. The Journal of Infectious Diseases. 2017;216:457–467. doi: 10.1093/infdis/jix237. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib39] Stone WJ, Dantzler KW, Nilsson SK, Drakeley CJ, Marti M, Bousema T, Rijpma SR. Naturally acquired immunity to sexual stage P. falciparum parasites. Parasitology. 2016;143:187–198. doi: 10.1017/S0031182015001341. [DOI] [PubMed] [Google Scholar]

[bib40] Stone WJ, Eldering M, van Gemert GJ, Lanke KH, Grignard L, van de Vegte-Bolmer MG, Siebelink-Stoter R, Graumans W, Roeffen WF, Drakeley CJ, Sauerwein RW, Bousema T. The relevance and applicability of oocyst prevalence as a read-out for mosquito feeding assays. Scientific Reports. 2013;3:3418. doi: 10.1038/srep03418. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib41] Team SD rstanarm: Bayesian applied regression modeling via Stan 2016

[bib42] Wells TN, Alonso PL, Gutteridge WE. New medicines to improve control and contribute to the eradication of malaria. Nature Reviews Drug Discovery. 2009;8:879–891. doi: 10.1038/nrd2972. [DOI] [PubMed] [Google Scholar]

[bib43] White NJ, Ashley EA, Recht J, Delves MJ, Ruecker A, Smithuis FM, Eziefula AC, Bousema T, Drakeley C, Chotivanich K, Imwong M, Pukrittayakamee S, Prachumsri J, Chu C, Andolina C, Bancone G, Hien TT, Mayxay M, Taylor WR, von Seidlein L, Price RN, Barnes KI, Djimdé A, ter Kuile F, Gosling R, Chen I, Dhorda MJ, Stepniewska K, Guérin P, Woodrow CJ, Dondorp AM, Day NP, Nosten FH. Assessment of therapeutic responses to gametocytocidal drugs in Plasmodium falciparum malaria. Malaria Journal. 2014;13:483. doi: 10.1186/1475-2875-13-483. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib44] White NJ. Primaquine to prevent transmission of falciparum malaria. The Lancet Infectious Diseases. 2013;13:175–181. doi: 10.1016/S1473-3099(12)70198-6. [DOI] [PubMed] [Google Scholar]

[bib45] Witkowski B, Duru V, Khim N, Ross LS, Saintpierre B, Beghain J, Chy S, Kim S, Ke S, Kloeung N, Eam R, Khean C, Ken M, Loch K, Bouillon A, Domergue A, Ma L, Bouchier C, Leang R, Huy R, Nuel G, Barale JC, Legrand E, Ringwald P, Fidock DA, Mercereau-Puijalon O, Ariey F, Ménard D. A surrogate marker of piperaquine-resistant Plasmodium falciparum malaria: a phenotype-genotype association study. The Lancet Infectious Diseases. 2017;17:174–183. doi: 10.1016/S1473-3099(16)30415-7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib46] World Health Organization . Eliminating Malaria. Vol 24. World Health Organization; 2016. [Google Scholar]

[bib47] World Health Organization . World Malaria Report. World Health Organization; 2016. [Google Scholar]

[bib48] WWARN Gametocyte Study Group Gametocyte carriage in uncomplicated Plasmodium falciparum malaria following treatment with artemisinin combination therapy: a systematic review and meta-analysis of individual patient data. BMC Medicine. 2016;14:79. doi: 10.1186/s12916-016-0621-7. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

A randomized feasibility trial comparing four antimalarial drug regimens to induce Plasmodium falciparum gametocytemia in the controlled human malaria infection model

Isaie J Reuling

Lisanne A van de Schans

Luc E Coffeng

Kjerstin Lanke

Lisette Meerstein-Kessel

Wouter Graumans

Geert-Jan van Gemert

Karina Teelen

Rianne Siebelink-Stoter

Marga van de Vegte-Bolmer

Quirijn de Mast

André J van der Ven

Karen Ivinson

Cornelus C Hermsen

Sake de Vlas

John Bradley

Katharine A Collins

Christian F Ockenhouse

James McCarthy

Robert W Sauerwein

Teun Bousema

Roles

Abstract

Background:

Methods:

Results:

Conclusions:

Funding:

Clinical trial number:

eLife digest

Introduction

Results

Figure 1. Trial profile.

Table 1. Baseline characteristics of the participants included in analysis.

Figure 2. Asexual parasitemia and gametocytemia.

Figure 2—figure supplement 1. Asexual parasitemia and gametocytemia per study participant.

Table 2. Treatment and parasitological data per study group.

Figure 3. Gametocyte kinetics between study arms.

Figure 4. Total female and male gametocyte density of all participants.

Figure 4—figure supplement 1. Female and male gametocytes per study arm.

Figure 4—figure supplement 2. Female and male gametocyte clearance dynamics per participant included in analysis.

Figure 5. Standard curves of qRT-PCR and qPCR.

Figure 5—figure supplement 1. Standard curves of Pfs25 qRT-PCR – low-density trendlines.

Figure 5—figure supplement 2. Correlation of duplo Pfs25 qRT-PCR measurements in all study samples.

Figure 6. Adverse events.

Figure 6—figure supplement 1. Liver function test derangements.

Table 3. List of adverse events possibly or probably related to the trial.

Table 4. Laboratory abnormalities per study arm.

Discussion

Materials and methods

Study design

Randomisation

Procedures

Pfs25 and PfMGET RNA quantification

Study outcome

Statistical analysis

Acknowledgements

Funding Statement

Contributor Information

Funding Information

Additional information

Competing interests

Author contributions

Ethics

Additional files

Data availability

References

Decision letter

Roles

Author response

Author response image 1. Model fit for CHMI-trans data female gametocytes.

Author response image 2. Model fit for CHMI-trans data male gametocytes.

Author response image 3. Model fit for male gametocytes in trial in Malian gametocyte carriers treated with non-gametocytocidal drugs dihydroartemisinin-piperaquine or amodiaquine plus sulfadoxine-pyrimethamine (NCT02831023).

Associated Data

Data Citations

Supplementary Materials

Data Availability Statement

ACTIONS