Efficacy and Bio-availability of Artemether-Lumefantrine in Severely Malnourished Children

Study hypothesis:

We hypothesize that AL efficacy might be impaired in severely malnourished children, due to impaired bio-availability of antimalarial drugs in this population.

Objectives:

The general objective of the study is to answer to the question: "Is the current dose of Artemether-Lumefantrine (AL) less efficacious in the severely malnourished compared to the non-severely malnourished children, because of impaired bio-availability?" We aim to assess whether the current treatment dose should be adjusted for this specific population.

The primary objective of the study is to compare the rates of treatment failure (after PCR correction) between severely malnourished and non-severely malnourished children.

The secondary objectives are:

• To compare the bio-availability of lumefantrine between severely malnourished and non-severely malnourished children (AUC, concentration at day 7…)
• To compare other efficacy parameters (such as the rates of early parasitological failure, early clinical failure, late therapeutic failure, reinfection and recrudescence) and safety parameters between severely malnourished and non-severely malnourished children.
• Additionally, If the rate of early parasitological failure appears high in severely malnourished children, we will measure the bio-availability of the artemisinin derivative and will compare it between severely malnourished and non-severely malnourished children.

Settings:

This study will be conducted in two sites in Mali and Niger, two country most affected by malaria and malnutrition, with a high malaria transmission roughly from July to December, corresponding to the malnutrition peak.

Study population:

This study will enrol children with uncomplicated P. falciparum malaria aged between 6 and 59 months, according to the WHO standardized protocol for the surveillance of antimalarial drug efficacy .

The definition of severe acute malnutrition will be weight-for-height <-3 z-scores to the reference WHO 2006 growth standards or middle upper arm circumference (MUAC) <115 mm.

The definition of non-severe acute malnutrition will be a weight-for-height z-score ≥-3 and MUAC ≥115 mm.

Severely stunted children (height-for-age z -score <-3) will be excluded from this study, which is primarily focused on wasted children.

The following modifications of the inclusion criteria will be applied, compared to the standardised WHO protocol:

• Decreased lower level of parasite density (PD) from 2000 to 1000 parasites per micro litre, as previous studies showed frequent low parasitemias in these patients.
• In the malnourished children arms, malnutrition <-3 z-score will be an inclusion criterion rather than an exclusion criterion.
• Children with complications of acute malnutrition needing intensive treatment will be excluded.

Study procedures:

Children will be screened at the renutrition center and pediatric ward of each site.

Two non-malnourished children will be enrolled soon after the enrolment of each malnourished child (ideally, during the same week), to allow carrying out inclusions in parallel for the two study groups, with benefit in terms of representativeness of the study population throughout the malaria season. The ratio of 2 non-severely malnourished children for 1 severely malnourished has been chosen to lower the total number of severely malnourished children to be recruited.

Treatment intake will be supervised and children will be followed-up for 42 days, according to the WHO standardized protocol4. PCR will be used to distinguish recrudescence and re-infection.

In malnourished children, the standardized nutritional rehabilitation program will also be conducted in parallel, and weekly follow-up will include the assessment of nutritional rehabilitation.

Treatment adverse events will be monitored and a Data Safety Monitory Board (DSMB) will be instituted.

To assess the bio-availability of the study drugs, a population pharmacokinetics approach will be used to reduce the number of blood sampling. Capillary blood will be collected on filter paper for measurement of lumefantrine concentrations, 5 times between Day 0 and day 7. Assays will use a liquid chromatography technique with tandem mass spectrometry.

Statistical considerations:

Efficacy analysis:

The primary analysis will compare the efficacy of AL in malnourished children versus non-malnourished children.

It will be designed to detect a minimum crude difference of 8% in Adequate Clinical and Parasitological Response (ACPR) rate in the population of malnourished children (87% versus 95% in non-malnourished children for both treatment combinations), with a statistical power of 80% and a 2-sided significance level of 5%. 160 severely malnourished children and 320 non-severely malnourished children will be enrolled. An additional 10% will be added to account for losses to follow-up and exclusions, yielding to 180 malnourished children and 360 non-malnourished children for each treatment evaluated. To summarize, a total of 540 children will be recruited as a whole.

To answer to the primary objective of the study, we will use multivariate models in which ACPR will be explained by malnutrition status after adjustment for baseline parasitemia, age, and other baseline characteristics that differ between the two populations of children. Cox proportional hazards models and logistic regression models will be used.

Pharmacokinetics (PK) analysis:

A population pharmacokinetics approach will be used, according to the WHO guidelines recommending 5 sampling windows for lumefantrine. Ideally, each sampling window should contain an equal number of samples randomly distributed over the duration of the window. Alternatively, we will use fixed, pre-determined sampling times within the suggested windows, which will fit other samplings required by the protocol. The 5 blood samples will be collected in the 180 severely malnourished children enrolled, and in 180 non-severely malnourished children (randomly chosen), which will allow us to obtain data representative of the studied populations5.

To limit blood sampling in severely malnourished children, we choose to initially restrict the PK assessment to lumefantrine, which exposure is the most correlated to overall treatment response5.

We will have the possibility to add the dosage of artemisinin derivatives in a second phase, in case of "relatively high" observed rate of early parasitological failures or impaired parasite clearance. Indeed, the main pharmacodynamics effect initially (during the first 3 days of treatment) is that of the artemisinin component5. If parasites are still present at day 3, it could be due to under-exposure to the artemisinin component. Otherwise, there is no reason to suspect such under-exposure.

The DSMB will have the responsibility to monitor closely all cases of early parasitological failure, and to recommend adding artemether PK assessment if they reckon that the rate is "unexpectedly high".

PK data will be analysed using nonlinear mixed-effects modelling (with external partnership).

Benefits and risks:

Benefit for the participants will be to receive free of charge treatment for a confirmed diagnosis of malaria. Participants will also benefit from regular visits that will ensure that the treatment cleared the malaria infection and any recurrent malaria episode or emergence of new diseases will be treated quickly. Malnourished patients will benefit from the standardized nutritional rehabilitation program.

The child will be asked to come to the study site for regular check and will be exposed to a series of blood sampling that would not happen if he was not enrolled in the study.

Ethical and regulatory considerations:

The study has been funded by the MSF innovative fund. The protocol is submitted to the approval of MSF Ethics Review Board and the Malian Institutional Ethics Committee.

A DSMB will closely monitor early parasitological failures, and follow serious adverse events.

An external monitoring will be performed during the study.