The study design and methodology for the parent clinical trial have been published previously [9]. In summary, PWH over 18 years of age on antiretroviral therapy were recruited into a randomized open – label phase III clinical trial. Participants were randomized in a 1:1:1 ratio to (1) continue standard of care of daily TS prophylaxis, (2) discontinue TS prophylaxis and begin weekly CQ prophylaxis, or (3) discontinue TS prophylaxis. Participants were followed up every 4–12 weeks, and whenever they were ill.
Study participants were recruited from two sites in Malawi: Ndirande research clinic in Blantyre, an urban centre, and Tisungane ART clinic at Zomba Central Hospital, located in a more rural setting. Of note, the two sites have different malaria burdens: malaria parasite prevalence in children under five years old is 4% in Ndirande versus 28% in Zomba [10]. Potential participants underwent informed consent before any study-related procedures. At screening, all participants had complete medical history, full physical examination, and blood samples collected for complete blood count, alanine aminotransferase (ALT), creatinine, CD4 count and HIV viral load. Consenting adults were recruited if they had: (1) been on ART for at least 6 months, (2) undetectable HIV viral load of < 400 copies/mm3, and (3) CD4 count of at least 250/mm3. Additional inclusion/exclusion criteria are published with the protocol [9] .
Malaria diagnosis and species differentiation
Participants who presented with malaria symptoms were evaluated at the study clinic and had blood collected for microscopy. Uncomplicated malaria was defined as objective fever (temperature ≥ 37.5ºC) or history of fever within the past 48 hours and/or other symptoms such as nausea, vomiting and diarrhoea, headache, back pain, chills, or myalgia, plus detection of any malaria parasites in blood by microscopy. Participants with danger signs such as reduction or loss of consciousness and difficulty breathing were referred to the hospital and not enrolled in the therapeutic efficacy study. Two trained microscopists identified and quantified malaria parasitemia from thick blood smears and determined Plasmodium species using thin blood smears. Only infections with P. falciparum are included here.
Malaria treatment and follow up
Participants diagnosed with uncomplicated P. falciparum malaria were treated with 80 mg artemether and 480 mg lumefantrine twice daily for three days. First doses were directly observed. All participants were followed up on days 1, 2, 3, 7, 14, 21 and 28 according to WHO standard for monitoring therapeutic efficacy, and blood smears and dried blood spots were collected at each visit.
In cases of recurrent malaria infection occurring on or after day 14, dried blood spots collected on filter paper from enrolment day and the day of recurrence of infection underwent extraction to distinguish new from recrudescent infection by genotyping merozoite surface protein-1 (MSP-1), MSP-2 and the glutamate-rich protein (GLURP) according to the publicly available protocol [11].
Participants who consented to participate in a sub-study of lumefantrine drug concentration measurements were selected to submit blood specimens for day 7 drug levels. These participants were given 250 ml of milk to drink with each dose to ensure optimal and consistent lumefantrine absorption [12]. Doses 1, 3 and 5 of antimalarial treatment were administered under direct observation. On day 7 of follow up for monitoring therapeutic efficacy, 5 ml of blood was collected and centrifuged at 3000 rpm for 10 minutes to collect 2 ml of plasma. Immediately after centrifugation, plasma was stored in a refrigerator at -20ºC. Once a week, plasma samples were transferred to a -80ºC refrigerator for storage at the central laboratory.
Plasma was shipped to the Parikh laboratory at the Yale School of Public Health (New Haven, CT) on dry ice, and immediately transferred to NorthEast BioLab (Hamden, CT) for drug level analysis of lumefantrine and desbutyl-lumefantrine using liquid chromatography-tandem mass spectrometry on an API 5000 triple-quadruple mass spectrometer (Applied Biosystems/MDS SCIEX, Foster City, CA). The calibration range was 10.9–3,785 ng/mL for lumefantrine, and 1.9–1,130 ng/mL for desbutyl-lumefantrine with the lower limit of quantification (LLOQ) at 10.0 and 1.9 ng/mL for lumefantrine and desbutyl-lumefantrine, respectively. The coefficient of variation was 1.4% and 5.6% for lumefantrine and desbutyl-lumefantrine, respectively.
Definitions
Treatment failure was diagnosed according to WHO criteria which classify outcomes in mutually exclusive groups [13]. Early treatment failure was defined as any of the following: danger signs or severe malaria on day 1, 2 or 3 in the presence of parasitaemia; parasitaemia on day 2 higher than on day 0, irrespective of axillary temperature; parasitaemia on day 3 with axillary temperature ≥ 37.5 ºC; parasitaemia on day 3 ≥ 25% of count on day 0.
Late clinical failure was defined as any of the following: danger signs, severe malaria, or axillary temperature ≥ 37.5°C in the presence of parasitaemia on any day between day 4 and day 28 in patients who did not previously meet early treatment failure criteria.
Late parasitological failure was defined as any of the following: presence of parasitaemia on any day between day 7 and day 28 with axillary temperature < 37.5 ºC in patients who did not previously meet early treatment failure or late clinical failure criteria.
Adequate clinical and parasitological response (ACPR) was defined as: absence of parasitaemia on days 28, irrespective of axillary temperature, in patients who did not previously meet early treatment failure, late clinical failure or late parasitological failure criteria.
After analysis of genotyping results, an infection was recrudescent if any genotype identified in the recurrent infection was also present at the initial infection. In PCR-adjusted results, only recrudescent infections were considered treatment failures.
Statistical analysis
Kaplan Meier survival analysis was used to determine the cumulative success rate defined as not reaching a failure point during the time under observation. Individuals diagnosed with malaria who did not have a 28-day follow-up visit and did not meet treatment failure definitions were censored at their last visit before 26 days.
We estimated the hazard ratio of failure using frailty proportional hazards model comparing prophylaxis regimen to no treatment to determine if malaria prophylaxis altered the risk of AL treatment failure. We combined the CQ and TS treatment arms for this analysis due to few malaria episodes among individuals on prophylaxis. A random effect for individual was included to account for multiple observations among the same individuals. We also examined the effects of other possible predictors of treatment failure, including age, sex, ART regimen, presentation with fever, parasite density at presentation, and study site.
To examine the association between lumefantrine and desbutyl lumefantrine concentrations and treatment failure, we first used a Wilcoxon Rank Sum test to compare the distribution of lumefantrine concentration between patients with treatment failure and those with ACPR. Because the majority of desbutyl lumefantrine levels were below the level of detection, we classified these levels as detectable or undetectable. For this analysis, participants who were followed up to at least 14 days without treatment failure were classified as treatment successes. As a day 7 lumefantrine concentration of 200 ng/ml has been associated with higher likelihood of treatment success, we next examined whether this cut-off point was associated with treatment failure in our cohort using Pearson’s Chi – squared test.
Ethical considerations
The study was approved by the Kamuzu University of Health Sciences Research Ethics Committee and the University of Maryland Baltimore Institutional Review Board. All participants provided written informed consent.