Background
Plasmodium falciparum is the most prevalent malaria parasite in Africa. However,
Plasmodium vivax has a wider geographical distribution. In 2018, there were approximately 75 million cases of malaria due to
P. vivax, accounting for 50% of cases in South East Asia and 75% of cases in the Americas [
1].
In the absence of a method for continuous in vitro culture of P. vivax, parasites are usually sourced ex vivo from infected humans. This limits many aspects of the study of P. vivax research, including the development of interventions to control and eliminate P. vivax, such as diagnostics, drugs and vaccines. For example, a reliable source of P. vivax sporozoites is required to test and develop new drugs targeting the dormant liver-stage parasites—the hypnozoites. Currently this entails an expensive, logistically complex and unreliable process of sourcing P. vivax-infected mosquitoes from endemic areas. In addition to the practical issues, parasites sourced in this way are not genetically homogenous. Thus, experiments are subject to possible effects of strain variability.
Experimental infection of human subjects with malaria termed volunteer infection studies (VIS) or controlled human malaria infection (CHMI) studies are increasingly being used for drug and vaccine development [
2‐
7]. Infections can be induced by mosquito bite, injection of sporozoites or blood-stage parasites. The latter is used in the induced blood stage malaria (IBSM) model, where healthy subjects are injected with
Plasmodium infected red blood cells. The IBSM model is being increasingly used for clinical studies of
P. vivax [
7‐
9].
It has recently been shown that
P. vivax can be successfully transmitted from healthy subjects to
Anopheles stephensi mosquitoes [
9,
10]. In this study sporozoites were harvested from the infected mosquitoes that were able to infect human hepatocytes in vitro [
9]. Although this system offers the potential to study the biology of
P. vivax malaria transmission and liver stage parasites, it is not a sustainable large-scale source of sporozoites for downstream work.
Apheresis is the removal of a specific component of an individual’s blood. Currently, centrifugal apheresis is the preferred method whereby blood components are separated based on buoyancy. Computer-controlled automated apheresis systems undertake continuous removal, separation of the target component, and then return the remaining blood to the individual [
11]. Automated erythrocytapheresis, also known as red cell exchange (RCE) has been used in the past for treatment of severe
P. falciparum malaria with the rationale of reducing the parasitized red blood cell concentration by replacing
Plasmodium-infected red blood cells with normal donor red blood cells [
12]. However, the rapid parasite clearance resulting from artesunate therapy has negated the need for RCE as a treatment for severe malaria [
13,
14].
Reported below are the results of a P. vivax clinical trial where apheresis was used as a means to harvest and concentrate all blood stages of P. vivax, including gametocytes, from human subjects experimentally infected with blood-stage P. vivax parasites.
Methods
Study design
The study presented here is a Phase 1 exploratory study that was conducted in four sequential single subject cohorts (ANZCTR Trial ID: ACTRN12617001502325) and performed at Q-Pharm Pty Ltd, Brisbane, Australia and the Apheresis Unit at the Royal Brisbane and Women’s Hospital (RBWH), Australia between October 2017 and May 2019. The primary objective of the study was to determine the safety of the P. vivax infection in healthy subjects following inoculation with blood-stage parasites, and the safety of apheresis for collection of P. vivax parasites from experimentally infected subjects. Secondary objectives were to assess the feasibility of apheresis as a method of harvesting, concentrating and subsequently cryopreserving P. vivax parasites from healthy subjects. Exploratory objectives were to evaluate the potential for apheresis to be used as a method for producing a P. vivax human malaria parasite bank, to evaluate the transmission of P. vivax gametocytes to mosquitoes and to collect and store plasma and peripheral blood mononuclear cells harvested using apheresis.
Specific modifications to the study protocol, such as the apheresis procedure, were required between subjects in an attempt to optimize the procedure and meet the objectives. All changes made between subjects were based on the findings from previous subjects.
Study subjects
Healthy adult males and females, aged between 18 and 55 years who met all inclusion criteria and none of the exclusion criteria were eligible for participation. Subjects were required to be malaria-naïve, Duffy blood group positive and have blood type O. Female subjects had to be Rh(D) positive. All subjects had to be available for a safety follow up period of three months. A full list of the inclusion/exclusion criteria for this study is included in the study protocol located in Additional file
1.
Study conduct
Pre-clinical component
A pre-clinical experiment was conducted prior to the clinical trial in order to confirm the feasibility of harvesting
Plasmodium parasites using apheresis. The
P. falciparum NF54 clone was used in these experiments [
15] due to limited availability of
P. vivax parasites.
Plasmodium. falciparum infected red blood cells (17.6 ml; 16 ml blood with 0.1% asexual parasitaemia and 1.67 ml blood with 0.01% gametocytaemia) were added to 450 ml of fresh venous whole blood and subjected to ex vivo apheresis. Samples were collected from the 1%, 2%, 3%, 5% and 7% haematocrit (HCT) layers as determined by visualizing the colour saturation of the apheresis product. An automated haematology analyser (Sysmex XN-3000; Sysmex UK) was used retrospectively to confirm the HCT of samples collected during apheresis. Presence of parasites was assessed in each layer by 18S qPCR [
16] and microscopy.
Clinical component
Following intravenous injection of
P. vivax (day 0), subjects were monitored by daily telephone calls until day 4, when subjects visited the clinical unit daily until the day of apheresis. Subjects were monitored for adverse events (AEs), signs and symptoms of malaria infection, and blood was collected for 18S qPCR measurement of parasitaemia. The severity of AEs were determined by the common terminology of clinical trial adverse events (CTCAE) v. 4.03 [
17].
The threshold for commencement of apheresis and treatment with artemether–lumefantrine was within 24 h of a parasitaemia > 20,000 parasites/mL, or the Malaria Clinical Score reaching > 6 [
10], or at the Investigator’s discretion. The morning that this threshold was reached (anticipated based on previous studies to be Day 10 or 11 [
9], subjects were admitted to the clinical unit (Q-Pharm) for initial safety assessments before being escorted to the Apheresis Unit at RBWH by Q-Pharm staff. The Apheresis Unit is located in the Haematology Department at RBWH where patients are subject to donor or therapeutic apheresis. At the Apheresis Unit the subjects underwent the apheresis procedure as per the Standard Operating Procedure (Additional files
2,
3,
4 and
5) whilst being supervised by the apheresis specialist nurse and under the supervision of the responsible clinical haematologist (GK). The same apheresis nurse performed the apheresis procedure for all four subjects. The apheresis procedure lasted 1–4 h. Subjects were then escorted back to the clinical unit and began treatment with artemether–lumefantrine (Riamet®, Novartis Pharmaceuticals Australia Pty Ltd). Treatment consisted of six doses of 4 tablets at 12 hourly intervals (each tablet contains 20 mg artemether and 120 mg of lumefantrine). Subjects remained confined within the clinical unit for 48–72 h for safety monitoring. Following release from confinement, subjects attended protocol specified visits until three months post-treatment to monitor for signs of recrudescent parasitaemia and to assess late safety signals. Relapse is not a concern in the
P. vivax IBSM studies as liver infection is bypassed and hypnozoites are not produced. A schematic of the study design is shown in Additional file
6: Fig. S1.
This study used an iterative adaptive design approach where subject safety and outcome data were analysed after each subject and modifications made to improve the chances of meeting the exploratory objectives in the subsequent subject. A summary of the changes instituted is shown in Table
1.
Table 1
Summary of main study design differences between subjects
Apheresis procedure | CMNC | CMNC | CMNC | Red cell depletion followed by CMNC on red cell depletion product |
HCT layers sampled* | 1%, 2%, 3%, 5%, 7% | 1%, 2%, 3%, 5%, 7% | 0.5%, 1%, 2%, 3%, 5%, 7%, 11%+ 2–3%, 5–7%, 1–7% | From the primary apheresis (HCT): Intermediate (64%) From the secondary apheresis (HCT): Final (3%) Spare (5%) Waste (42%) |
Apheresis timepoint | 10 | 10 | 11 PM | 11 AM |
Mosquito feeding assay samples | Pre-apheresis (with-Percoll enrichment), 1%, 2%, 3% HCT layers | Pre-apheresis (with-Percoll enrichment) | Pre-apheresis (with and without-Percoll enrichment) | Pre-apheresis (without-Percoll enrichment), Intermediate, Final, Waste |
Whole blood:citrate ratio during apheresis | 15:1 | 8:1 | 8:1 | 13:1 |
Citrate added to apheresis collection bags | No | Yes | Yes | Yes |
Biological duplicates* | No | No | Yes | Yes |
CMNC; continuous mononuclear cell collection, HCT; haematocrit. Protocols for subjects 1 to 4 and all experiments can be found in Additional files
7,
8,
9 and
10. *Biological duplicates involved repeat 18S qPCR testing from two separate blood samples from each HCT layer collected using apheresis.
When a HCT range is included the sample was taken from multiple HCT layers e.g. 5–7% = 5%, 6% and 7% HCT. +Originally aimed to sample 8% HCT layer but actual sample consisted of 11% HCT. #During cohort 4 a red cell depletion was carried out, producing an intermediate bag sample, followed by a second apheresis procedure on the red cell depletion product. The second apheresis procedure involved sampling of ~ 100 ml of the lowest HCT layers of the sample (final bag) followed by ~ 100mls of the subsequent lowest HCT layers (spare bag) and then the remainder ~ 100mls (waste bag).
Malaria challenge agent
The
P. vivax human malaria parasite (HMP) bank HMP013 was derived from blood group O rhesus positive blood donated from a returned traveller from India who presented with clinical manifestations of malaria [
9]. The inoculum was prepared as previously described [
18].
Measurement of parasitaemia by qPCR
Parasitaemia was quantified using 18S qPCR targeting the highly conserved
Plasmodium 18S ribosomal RNA gene [
16,
19]. Quantitative reverse transcriptase PCR (qRT-PCR) assays were used to quantify gametocyte levels with assays targeting the
P. falciparum pfs25 (female) and
pfMGET (male) gametocyte mRNA transcripts [
20] and
P. vivax pvs25 (female) gametocyte mRNA transcripts [
21].
Flow cytometry
Flow cytometry was performed to characterize cell populations present in samples collected during the apheresis process. A combination of stains and antibodies were used to identify cells containing DNA/RNA (SYBR Green I), white blood cells (WBCs) (CD45 antibody) and/or reticulocytes (CD71 antibody). Samples from subject 1 were stained with SYBR Green I (Molecular Probes); samples from subjects 2 and 3 were stained with SYBR Green I and CD45-PacificBlue; and samples from subject 4 were stained with SYBR Green I, CD45-Pacific Blue and/or CD71-APC. Samples were kept on ice or at 4–8 °C until analysed by flow cytometry.
SYBR Green I staining
A volume of 2.5 μl or 1 × 106 cells from each sample was stained with 30–50 μl of SYBR Green I at 10× for 30 min in the dark. After incubation, 200 μl of FACS buffer (2% fetal bovine serum in phosphate buffered saline) was added.
Antibody staining
Approximately 1 × 106 cells were stained with 5–10 μg/ml of CD45-Pacific Blue or 2.5 μl of CD71 stock solution for 30 min at 4–8 °C in the dark. Cells were washed twice with PBS by centrifugation at 1455×g for 4 min, at 4 °C. After the last wash 200 μl of FACS buffer was added to the cells.
Double staining with SYBR Green I and CD45-Pacific Blue: A volume of 30 μl of SYBR Green I at 10× was added to pelleted cells that were previously stained with CD45-Pacific Blue (as mentioned above) for 30 min at 4–8 °C, in the dark. After incubation a volume of 200 μl of FACS buffer was added.
Triple staining with SYBR Green I, CD45-Pacific Blue and CD71-APC: Approximately 1 × 106 cells were stained with 10 μg/ml of CD45-Pacific Blue, 2.5 μl of CD71 stock solution and 30 μl of SYBR Green I at 10x. Samples were incubated for 30 min in the fridge (4–8 °C) in the dark. Cells were washed twice with PBS by centrifugation at 1455×g for 4 min, at 4ºC. After incubation a volume of 200 μl of FACS buffer was added.
Flow cytometry analysis
Samples from subjects 1, 2 and 3 were acquired on a FACS CANTO II (BD Biosciences), using the 488 nm and 405 nm lasers. SYBR Green I positive cells were detected using a 530/30 nm band-pass filter and CD45-Pacific Blue positive cells were detected using a 450/50 nm band-pass filter. Samples from subject 4 were acquired on a LSR FORTESSA (BD Biosciences), using the 488 nm, 640 nm and 405 nm lasers. SYBR Green I positive cells were detected using a 530/30 nm filter, CD45-Pacific Blue positive cells were detected using a 450/50 nm filter and CD71-APC positive cells were detected using a 670/14 nm filter. Flow cytometry data was analysed using FlowJo® software (version 10.8, Tree Star Inc, Oregon, USA).
Microscopy
Thick and thin smears were stained with Giemsa and examined under a 100× oil immersion objective by level 1 or 2 WHO certified malaria microscopists. Apheresis samples were expected to have a significantly different composition in terms of proportions of RBCs and WBCs when compared to whole blood (e.g. RBCs make up 1% and approximately 46% of 1% HCT and whole blood samples respectively). As such, standard parasitaemia measures by microscopy were not feasible. It was decided that the expert microscopists would estimate parasitaemia based on sample composition.
Mosquito feeding assays
Transmissibility of pre-apheresis samples and post-apheresis samples to
An. stephensi was evaluated using membrane feeding assays (MFA) [
9,
22]. For enriched MFA, gametocytes present in 80 mL of whole blood (pre-apheresis) were enriched in 70% Percoll gradient. For direct MFA (DMFA), 650µL of pellet from whole blood (pre-apheresis) or from each apheresis sample was reconstituted to 50% haematocrit with malaria naïve AB + serum. Infection in midguts was assessed by qPCR [
23] 8 days after the feeding assays. For logistic reasons, enriched MFA was not carried out in subject 4. Following consideration of gametocyte levels measured by qRT-PCR targeting
pvs25, DMFA was not carried out in subject 2.
Apheresis procedures
Apheresis was carried out using a Spectra Optia v11.3 apheresis system (Terumo BCT, Inc Tokyo Japan) as detailed in the Additional file
11.
The continuous mononuclear cell collection procedure was used to sample from the targeted HCT layers from the blood of subjects 1 to 3. The targeted HCT layers that were sampled from these subjects ranged from the platelet rich layer through to the red cell rich layer.
A double stage procedure was used to collect the targeted HCT layers from subject 4. In the first stage, a red cell depletion procedure was used to collect approximately 500 ml of packed red blood cells from the subject. Targeted HCT layers were then collected from the red cell concentrate using the polymorphonuclear (PMN) collection procedure. The starting product (red cell concentrate) for the second stage of this procedure had a significantly higher HCT than the whole blood of subjects 1 to 3 and sampling focussed on the higher HCT layers. This was the rationale for using a PMN collection in subject 4 rather than the CMNC collection used in subjects 1 to 3.
Statistical analysis
Continuous data was summarized using descriptive statistics (mean and standard deviation, or median and interquartile range). Categorical data was presented using N and %. Descriptive statistics were produced using Microsoft Excel® (version 1903). GraphPad® Prism was used for the construction of all figures.
Discussion
Using apheresis it was possible to achieve modest concentration of both asexual and gametocyte stages of P. vivax. However, the modest level of parasite enrichment (4.9-fold and 1.45-fold for asexual parasites and gametocytes respectively) was deemed to be insufficient for downstream research. Furthermore the relatively low parasite levels, particularly of gametocytes, meant that the figures may be subject to chance variations in parasite levels.
Results of this study suggest that apheresis in healthy subjects infected with blood-stage
P. vivax parasites is safe. No serious adverse events were encountered, with all adverse events having resolved by the end of study. The majority of adverse events were malaria related, and in line with previous
P. vivax IBSM studies [
7,
8]. Adverse events related to apheresis consisted largely of asymptomatic transient reductions in haematology parameters.
After correction for red blood cell number, parasite quantitation by qPCR suggested that both asexual parasites and gametocytes were preferentially concentrated in the lower HCT layers (Additional file
6: Figs. S2 and S4). Furthermore, the pre-clinical experiment demonstrated a selective concentration of
P. falciparum gametocytes compared to asexual parasites in lower HCT layers (Additional file
6: Fig. S6A and Table S4). However as the pre-clinical experiment did not involve collection of blood from an infected subject they do not fully replicate those collected ex vivo, for example being subject to host interactions such as sequestration, thus limiting the utility of this work in predicting what would be the situation in natural infection. Findings from the pre-clinical and clinical experiments were consistent with the previously published observations of the buoyancy of
Plasmodium parasites [
24].
Several scenarios were considered to explain why concentration of parasites using apheresis was lower than expected. Firstly, as gametocyte concentrations in apheresis samples were around the level of detection of the
pvs25 by qRT-PCR, minor variation in concentration may have been difficult to quantify [
25]. Secondly, infected RBCs containing magnetic haemozoin [
26] may have attached to ferromagnetic components of the apheresis apparatus. Thirdly, it is possible that lysis of asexual parasites and gametocytes occurred during the apheresis procedure. This latter hypothesis is supported by a recent report suggesting that parasite maturation results in increasing fragility of
P. vivax infected red blood cells [
27]. Although it may have been possible to assess for low level haemolysis during the procedure, for example by measuring haptoglobin levels, controlling for a range of other variables would have been difficult.
An enhanced level of transmission to mosquitoes compared to whole blood samples collected pre-apheresis was only observed on one occasion (final sample bag [3% HCT] in subject 4), corresponding to the higher gametocyte concentration in this sample compared to the pre-apheresis whole blood (Fig.
4b). A possible explanation of the low success in the transmission studies was the difficulty in maintaining tight temperature control to prevent exflagellation of male gametocytes [
28], thereby negatively impacting gametocyte infectivity [
29]. Blood was most vulnerable to a temperature drop whilst in the apheresis equipment itself. It was not possible to heat apheresis equipment, and it was deemed impractical to heat the room where apheresis took place to > 35 °C. Temperature monitoring was not possible during the experiments. However, the demonstration of transmission success in the final bag (3% HCT) indicates that at least some gametocytes were maintained within a temperature range that did not trigger exflagellation. Regardless of the underlying cause, recent reports of success in improving concentration of gametocytes and enhanced transmission by either Percoll [
9,
10] or magnetic bead [
30] enrichment suggests that such methods are superior for concentration of gametocytes for mosquito transmission experiments.
Each of the vials from the HMP bank used in this study to infect subjects contain 2.08 × 10
6 parasites. Based on the greatest level of asexual enrichment per ml of sample observed (7% HCT, subject 1), calculations suggest that apheresis alone can create parasites vials with a maximum of 7.92 × 10
4 parasites (Additional file
6). Therefore, to create a HMP bank using an apheresis approach equivalent to the one used to infect volunteers in this study a > 25-fold increase in pre-apheresis parasitaemia would be required, equating to > 650,000 parasites/ml. Attaining such high parasitaemia would likely lead to significant discomfort in study volunteers, and therefore raise significant ethical concerns. Therefore, unless significant improvements in enrichment can be attained, apheresis should not be used to create HMP banks, and the current practice of collecting blood by venesection is preferable.
Strengths of the study include the wide sampling across HCT layers (1%, 2%, 3%, 5% and 7%) and the use of multiple methods to enumerate parasites (flow cytometry, microscopy and qPCR). Although only a small number of subjects were studied using this approach, the lack of promising data meant that continuation of the trial was deemed inappropriate by the safety review team and the study was terminated.
Acknowledgements
Indera Govender provided technical advice regarding the apheresis procedure, Adam Potter assisted with protocol PICF and manuscript preparation, Helen Jennings provided technical advice on safety and quality control of blood products produced during the study, Katherine Trenholme provided assistance and technical advice during laboratory experiments, Renee Atkinson provided clinical nursing supervision to volunteers and oversaw data collection, Sue Mathison provided clinical nursing supervision to volunteers and oversaw data collection, Matthew Adams provided assistance and technical advice during laboratory experiments, LTCOL Ken Lilley provided microscopy input, LTCOL Ivor Harris provided microscopy input, CAPT Jo Kizu provided microscopy input, Stephan Chalon acted as an independent medical monitor, Dennis Shanks acted as an independent medical monitor.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.