About this study
The populations of the South West Pacific are highly diverse and exhibit a range of red blood cell (RBC) polymorphisms. Within Papua New Guinea (PNG), a variety of red cell variants are found that have geographical patterns paralleling malaria endemicity (Muller I et al, 2003). In particular, Southeast Asian Ovalocytosis (band 3 deletion SLC4A1Δ27 [SAO]) is found in up to 35% of the population in some coastal areas and has been associated with complete protection against cerebral but not other forms of severe Plasmodium falciparum malaria in previous studies in PNG (Genton B et al, 1995; Allen SJ et al, 1999). Alpha+-thalassaemia is found in more than 90% of people and has been associated with protection from severe malaria (Mockenhaupt FP et al, 1997) and severe non-malarial disease (Allen SJ et al, 1997). Finally, polymorphisms in complement receptor 1 protect against severe malaria through reduced red cell rosette formation (Cockburn IA et al, 2004).
Taken together, the high prevalence of these known genetic variants suggests that malaria has exerted strong selective pressure in Melanesian populations. However, similar to African settings, known genetic variations may account for only a small proportion of the total variability in genetic susceptibility to severe malaria in the population as a whole (Mackinnon MJ et al, 2005). The epidemiology and clinical features of severe malaria in Melanesian children differ substantially to African children and the presence of Plasmodium vivax or specific genetic factors are thought to be responsible for a lower mortality observed in a number of studies from this area (Maitland K et al, 1997; Manning L et al, 2011).
As a prelude to a genome wide analysis (GWA), we outline three separate severe malaria case-control studies undertaken at the same field site in Madang province, Papua New Guinea examining the genetic association for 69 candidate single nucleotide polymorphisms (SNPs) (selection based on previous reports of association with severe malaria or on their likely biological role in malaria infection/disease) that were performed during the years 1993-6 (Study A; Allen SJ et al, 1996), 2003-4 (Study B; Karunajeewa HA et al, 2006) and 2006-9 (Study C; Manning L et al, 2011). In the present report we outline the characteristics of the study populations and the methodology for the clinical assessment of cases and controls recruited into each study.
Clinical data and DNA samples were contributed to the MalariaGEN Consortial Project 1 (CP1) along with those of 11 other case-control studies from a total of 11 malaria-endemic countries. As part of the sample handling process, baseline genotyping data was generated for a number of malaria–associated SNPs and the appropriate data has been returned to each site for site-specific analysis.
The present study was conducted in Madang Province on the northern PNG coast where most of the population are subsistence farmers. Malaria transmission is perennial but with seasonal variations. The annual entomological inoculation rate (EIR) for Madang Province has recently been estimated at 37 for P. falciparum and 24 for P. vivax. Malaria is transmitted by a number of mosquito vectors including Anopheles punctulatus complex, A. farauti and A. koliensis (Michon P et al, 2007). During the recruitment period for Study C, healthy, asymptomatic Madang children aged 1-10 years had spleen rates of 13% and the prevalence of asymptomatic parasitaemia by microscopy was 8.2% for P. falciparum (median [interquartile range] parasite density 1360 [453-2881] /µl) and 14.1% (348 [226-727] /µl) for P. vivax (Manning L et al, 2011). These malariometric indices are lower than reported from the same area during the period that Study A was performed (Burkot TR et al, 1987). Approximately 90% of local children have alpha-thalassemia trait (Muller I et al, 2003). The current national human immunodeficiency virus (HIV) seroprevalence is 0.9% (UNAIDS, 2010).
A matched case-control study was conducted. Cases consist of children (aged 5 months-12 years) with signs of severe or uncomplicated malaria. The study participants with severe malaria in this study were recruited at Modilon Hospital, the main referral centre for Madang Province during three separate studies performed during the years 1993-6 (Study A; Allen SJ et al, 1996), 2003-4 (Study B; Karunajeewa HA et al, 2006) and 2006-9 (Study C; Michon P et al, 2007).
The criteria for inclusion of severe malaria cases varied slightly between the three studies. For those recruited into Study A the WHO 1990 (World Health Organization, 1990) definition for severe childhood malaria was used whereas for studies B and C the contemporary WHO 2000 (World Health Organization, 2000) definition was applied. For each study, clinicians or trained research nurses carried out clinical assessments on admission. This included details of immunizations, past medical history and recent treatment with antimalarial drugs and antibiotics, as documented in each child’s hand-held medical record book. The assessment of coma (Blantyre Coma Score [BCS]; Molyneux ME et al, 1989), lactate, glucose and haemoglobin (Hb) concentrations was measured in a standardised manner. This allowed consistent and comparable clinical phenotypes to be derived from each study that could be uploaded to the MalariaGEN CP1 website. A BCS≤2 was considered deep coma and a BCS≤4 as impaired consciousness at 0.5, 1 or 6 hours after correction of hypoglycemia, a seizure or parenteral anticonvulsant therapy, respectively. Respiratory distress was considered present if the child had i) deep breathing, ii) inter-costal in-drawing, iii) sub-costal recession, iv) persistent alar flaring, v) tracheal tug, and/or vi) respiratory rate >60/minute. Other indicators of severe malaria were defined in accordance with the respective WHO definitions. Chest radiography and lumbar puncture were performed in a minority of children whilst blood culture was only available during Study C.
In the latter study, recruitments were restricted to children with severe malaria, a pre-defined parasitaemia threshold (>1000 P. falciparum/µl and >500 P. vivax/µl) and limited to children from Madang, Morobe and Sepik provinces.
For Study A, recruitments were restricted to children with P. falciparum only and were initially treated with intramuscular quinine, followed by oral quinine and a single dose of sulphadoxine/pyrimethamine (SP). In Study B, children with severe malaria were given rectal artesunate or intramuscular quinine as part of a safety and efficacy trial, whilst in Study C children were given intramuscular artemether, followed by SP on day three and oral artesunate. During all 3 studies antibiotics (chloramphenicol 25 mg/kg by intramuscular injection 6-hourly) and intravenous dextrose/saline were given concurrently with antimalarials in accordance with local protocols. Blood transfusion was done at the discretion of the attending physician. The PNG standard treatment guidelines recommend transfusing all children with Hb <40g/l (4g/dl) or at higher concentrations in the presence of cardiovascular compromise (Paediatrics Society of PNG, 2005).
Healthy community controls were matched to severe cases for all 3 studies. During Study A, controls were recruited from the community soon after the recruitment of a severe malaria case and individually matched as closely as possible to a case for ethnicity, age, gender and residence. Controls for Study B were recruited to the study 4 years after the original study had been performed and were matched by age and sex. During Study C, controls were recruited at immunization clinics in villages of severe malaria patients and were matched by age, sex and ethnicity. In the latter two studies (B and C) children matched by age were within 12 months of the index case, in reasonable health as defined by the absence of i) a history of malaria within the previous fortnight, ii) current fever (axillary temperature >37.5°C) plus a positive rapid diagnostic test for malaria, iii) respiratory distress (respiratory rate >40/minute plus in-drawing of chest wall or dyspnea), iv) impaired consciousness (Blantyre Coma Score ≤4), or v) a hemoglobin concentration <50g/L (<5g/dl). In this part of PNG, malarial parasitaemia without acute illness is common and therefore children with asymptomatic parasitaemia were included.
During Study C, when possible an uncomplicated malaria control was also recruited for each severe malaria case. Children with uncomplicated malaria were matched by age, sex and ethnicity and defined by a history of or current fever, either a positive rapid diagnostic test for malaria or plasmodium parasites by light microscopy and had none of the clinical signs indicating severe illness.
The clinical data was double-entered by PNG Institute of Medical Research’s (PNGIMR) data management unit before being uploaded onto secure web-based software developed by MalariaGEN. Here, the integrity of the data was checked, standardised and amalgamated.
For Study A, genomic DNA was extracted on site from whole blood by proteinase K digestion followed by phenol chloroform extraction and shipped frozen to the Institute of Molecular Medicine, Oxford for molecular analysis. Genomic DNA was extracted from whole blood at molecular laboratory of the PNGIMR in Goroka using QIAamp 96 DNA Blood Mini Kit, QIAGEN, Valencia, CA. Aliquots of the DNA samples were shipped to the MalariaGEN Resource Centre in Oxford for further processing and quality control for quantity, quality (by genotyping) and confirming appropriate clinical data was available. Baseline genotype data for 69 malaria-associated SNPs was generated for all contributing samples; briefly, samples underwent a primer-extension pre-amplification (PEP) step (Xu K et al, 1993; Zhang L et al, 1992) prior to genotyping on the Sequenom® MassArray® platform. Following curation, the genotype data were returned to the PI’s for local analyses.
|Number||Gender: n (%)||Age in years: n (%)||Ethnicity: n (%)|
Malaria cases: 805
Male: 442 (55)
Female: 361 (45)
Not recorded: 2 (<1)
<5: 526 (65)
5-15: 132 (16)
Not recorded: 147 (18)
Madang: 440 (55)
Other: 152 (19)
Not recorded: 213 (26)
|Healthy controls: 553||
Male: 300 (54)
Female: 253 (46)
<5: 435 (79)
5-15: 109 (20)
Not recorded: 9 (1)
Madang: 348 (63)
Other: 116 (21)
Not recorded: 89 (16)
The study was approved by the PNG Institute of Medical Research Institutional Review Board (proposal number: IMR IRB 0603) and the Medical Research Advisory Committee of the PNG Health Department (proposal number: MRAC No: 06.21) and conducted according to the principles of the Declaration of Helsinki.
Written informed consent was obtained from parent(s)/guardian(s) of both cases and controls before recruitment. All written materials were available in English and Melanesian Pidgin languages. Trained nursing officers who were fluent in Melanesian Pidgin recruited participants into the study using the language most comfortable for the patient and their family.
- Angela Allen, Weatherall Institute of Molecular Medicine, Oxford University, UK
- Harin Karunajeewa, University of Western Australia, Australia
- Moses Laman, Papua New Guinea Institute of Medical Research, Papua New Guinea
- Peter Siba, Papua New Guinea Institute of Medical Research, Papua New Guinea
- Steve Allen, Swansea University, UK
- Timothy M E Davis, University of Western Australia, Australia
The authors gratefully acknowledge the assistance of staff on the Paediatric Ward at Modilon Hospital, the Papua New Guinea Institute of Medical Research staff at Modilon Hospital and the Yagaum campus, and the patients and their families for their participation.
In addition to the MalariaGEN Consortium, financial support for this study was obtained from The National Health and Medical Research Council (NHMRC) of Australia (grant #513782). In addition, LM was supported by Royal Australasian College of Physicians (Basser) and NHMRC scholarships, ML a Fogarty Foundation scholarship, and TMED an NHMRC Practitioner Fellowship.