In the mid-1980s, amodiaquine was recommended as a malaria prophylaxis for travellers but several reports pointed high level of toxicity, mainly agranulocytosis and hepatotoxicity [1, 2], leading to the removal of amodiaquine monotherapy from the Essential Drug List of WHO in 1990 . Some years later, an updated appraisal of available data suggested that amodiaquine toxicity related with severe liver damages and agranulocytosis was primarily seen in non-Africans and only after several weeks of regular chemoprophylaxis, reinstating this drug as an option for treatment of malaria [4, 5]. Amodiaquine was reintroduced as an important, slowly eliminated partner drug in artemisinin-based combination therapy (ACT), the current global mainstay for the treatment of uncomplicated Plasmodium falciparum malaria. Nowadays, artesunate-amodiaquine (ASAQ), a first-generation ACT, is used as first- or second-line treatment in many countries in Africa . Amodiaquine is also increasingly used in combination with sulfadoxine-pyrimethamine (SP-AQ) in seasonal malaria chemoprevention, i.e., monthly distribution of intermittent preventative treatment in young children during peak malaria transmission, in several countries of the Sahel sub-region [7, 8]. In numerous clinical trials ASAQ efficacy has been high, with an estimated mean of 95.1% cure rate in a large meta-analysis of studies in Africa . Furthermore, treatment (as opposed to prophylaxis) of malaria with amodiaquine has been associated with mild adverse events including gastrointestinal effects, abdominal pain, neutropenia, nausea, dizziness, and pruritus, but typically not with serious adverse events [4, 10–12].
Amodiaquine is short lived (half-life 2–8 hours) and is primarily metabolized by cytochrome P450 2C8 (CYP2C8) to its main, biologically active metabolite desethyl-amodiaquine (DEAQ)  which has a long terminal elimination half-life (9–18 days) . The main antimalarial action of amodiaquine is thus carried out by DEAQ, including an initial immediate treatment effect (parasite clearance), as well as a temporary post-treatment protective effect during the elimination phase of the metabolite. The CYP2C8 gene carries several polymorphisms including the most frequent minor alleles CYP2C8*2 and CYP2C8*3, coding for enzymes with altered activity in comparison with the CYP2C8*1 wildtype . The CYP2C8*2 variant has been associated in vitro with a six-fold lower amodiaquine metabolism activity than the CYP2C8*1 wildtype enzyme . The effect was even greater in the CYP2C8*3 variant, suggesting that any impact of reduced CYP2C8 metabolism would be more pronounced in CYP2C8*3 carriers. CYP2C8*2 is most prevalent in those of African descent, whereas CYP2C8*3 is highly frequent among Caucasians [14, 16–18].
It has been postulated that the impaired conversion of amodiaquine to DEAQ among low activity CYP2C8*2 and CYP2C8*3 carriers is not likely to impact treatment efficacy as both amodiaquine and DEAQ have antimalarial activity, the latter considered the major active component . However, the prolonged pharmacokinetic profile in poor-metabolizers may lead to a non-negligible increased risk of amodiaquine-related adverse events among populations with these specific genotypes [14, 19, 20]. Albeit of interest, only a few studies have investigated the potential association between slow amodiaquine metabolizers and reduced treatment efficacy and/or increased risk of adverse events [3, 20–22]. In vivo data on the impact of the low activity CYP2C8*2 allele are sparse, and almost non-existent among CYP2C8*3 carriers due to the very low CYP2C8*3 allele frequency in the generality of African populations, where ASAQ is primary used [6, 14].
Zanzibar, where ASAQ has been the first line treatment for uncomplicated malaria since 2003, has similar CYP2C8*2 (13.9%) frequencies but higher CYP2C8*3 (2.1%) allele frequency than in other places in sub-Saharan Africa [3, 17]. This latter particular characteristic sets the opportunity to a more complete investigation the effect of CYP2C8 polymorphism on amodiaquine-based antimalarial treatment. We therefore retrospectively assessed the impact of these CYP2C8 polymorphisms on treatment outcome and tolerability in two ASAQ malaria efficacy trials conducted in Zanzibar in 2002–2005, when malaria in these islands was still characterized by high incidence (Bhattarai et al., 2007). More specifically, we assessed if CYP2C8*2 and CYP2C8*3 carriers were at increased risk of new and/or recrudescent infections during the 42-day follow-up period, and if CYP2C8*2 and CYP2C8*3 carriers were at increased risk of experiencing adverse events after ASAQ treatment.