Monitoring of efficacy and safety of artemisinin-based anti-malarials for treatment of uncomplicated malaria: a review of evidence of implementation of anti-malarial therapeutic efficacy trials in Tanzania

Following the recent reports of emergence of P. falciparum artemisinin-resistant field isolates in Southeast Asia [1,2,19] and the threat of such parasites spreading to other malaria-endemic countries, country-specific evidence based on reliable data are urgently required to monitor the efficacy of the drugs and support timely review and implementation of malaria treatment guidelines. Surveillance of anti-malarial efficacy is crucial to enable early detection of emergence of drug resistance when it happens before it spreads in most of the populations, as happened for CQ and SP [33]. Information generated from such surveillance provides evidence to relevant national and international authorities for policy formulation and review. This review was undertaken to assess the implementation of efficacy testing for monitoring of therapeutic efficacy of ACT for treatment of uncomplicated falciparum malaria in Tanzania before and after policy changes (in 2006).

Apparently, due to limited resources, especially funding, NMCPs in most endemic countries have not been able to implement regular anti-malarial drug efficacy monitoring at sentinel sites and there has been a strong call for regional networks to facilitate the implementation [42]. The former East African Network for monitoring anti-malarial treatment (EANMAT) forged a partnership between the ministries of health and the research community in East African countries and facilitated monitoring of anti-malarial drug resistance in the region [28]. However, EANMAT collapsed as a result of many factors, including dependence on short-term donor funding.

The findings of this review showed that nine clinical trials have been conducted to monitor the efficacy of ACT before and after Tanzania adopted ACT for treatment of uncomplicated malaria. Of these, only one study [39] was conducted within the in vivo efficacy testing framework of Tanzania NMCP/EANMAT with financial support from NMCP and EANMAT. This might be partly due to lack of funding, or due to complacency attributed to perceived high therapeutic efficacy of ACT. However, in Tanzania, NMCP-supported TETs have resumed since 2011, although the findings have not yet been published.

The present review has shown that the efficacy of AL, which is the first-line anti-malarial drug for treatment of uncomplicated falciparum malaria in Tanzania, was high even after unsupervised treatment. The PCR-corrected cure rate on day 28 was >91% and this is in line with findings from other studies in eastern Africa [43,44]. The high cure rate (100%) reported in Muheza in 2004 [39] prior to the official adoption of AL was similar to the cure rate reported in other African countries [45,46]. The efficacy of AS + AQ, which is currently the first-line anti-malarial drug for treatment of uncomplicated falciparum malaria in Zanzibar (an island part of the United Republic of Tanzania) [30], was also high although comparably lower than that of AL (PCR-corrected cure rate was 88.8 - 93.8% for AS + AQ compared to 91 -100% for AL). The lower cure rates of AS + AQ compared to AL could be attributed to the fact that AQ had been extensively used in the country and was also adopted as a second-line anti-malarial drug together with SP (which was the first-line) in 2001 [25]. Since AQ resistance had been reported in Tanzania [27,47], addition of an artemisinin to AQ was unlikely to make a combination with high therapeutic efficacy. Similar failure rates have been reported in other SSA countries that used AQ extensively prior to introduction of AS + AQ [48].

The cure rates of AQ + AS in the present review compares well with those reported elsewhere in East Africa whereby day-28 adequate clinical and parasitological response (ACPR) in children treated with AS + AQ was 90.2% in Kenya [49], 90.3% in Rwanda [50] and 91.7% in Uganda [51]. These rates were comparatively lower than those of AL. However, AS + AQ was selected and is still being used as the first-line anti-malarial in some countries when other factors apart from efficacy were considered. In Burundi, prior to policy changes, day-14 ACPR in children treated with AS + AQ was reported to be 95.3% compared to 99.3% for AL. However, considering other factors such as acceptability by users, adherence and cost, AS + AQ was chosen as the first-line anti-malarial for treatment of uncomplicated falciparum malaria in Burundi [10,12]. Similarly, Zanzibar adopted AS + AQ as the first-line anti-malarial despite lower day-28 ACPR compared to AL (97% versus 91% for AS + AQ) [30].

There is a concern about the limited post-treatment prophylactic effects of both AL and AS + AQ in high transmission areas. In fact in one trial, more than half of the recruited patients had recurrent infections within the 42-day follow-up period after treatment with AL. However, the majority of recurrent infections were due to re-infections which suggests that the partner drug cannot give prolonged protection despite high therapeutic efficacy [36]. Similar high re-infections rates have been reported in other high transmission areas in Africa after AL treatment [43,52]. Studies that compared the efficacy of AL and AS + AQ showed a significantly lower risk of re-infection after treatment with AL compared to AS + AQ [30,40] suggesting that AL confers a longer prophylactic effect than AS + AQ. The difference in prophylactic effect of the two drugs could be attributed to the longer half-life of lumefantrine compared to AQ. Thus, the concentration of the active amodiaquine metabolite might be lower or completely absent when a re-infection occurs compared to lumefantrine concentration. This observation has also been reported elsewhere in Africa where re-infection rates were higher after AS + AQ treatment than after AL [48,53,54]. However, a recent study has reported high level of resistance to lumenfatrine in the Democratic Republic of Congo [55] that threatens the therapeutic usefulness of AL and further monitoring is urgently needed in all malaria-endemic countries where AL is the first-line anti-malarial drug.

In most of the studies, a great majority of the recurrent infections were due to re-infections, when assessed with a step-wise PCR genotyping protocol, which signifies that the drugs are still efficacious and the high rates of re-infections could only be attributed to high malaria transmission. In terms of clinical practice, the high re-infection rates are of great concern among clinicians. Clinicians should be clearly guided on what to expect and how to handle such cases with recurrent infections within a period of three to eight weeks post-treatment. The observed high re-infection rates after ACT treatment underscores the importance of integrating treatment with vector control interventions, including use of long-lasting insecticide-treated nets so as to effectively block malaria transmission and prevent recurrent infections [56].

The study which tested AZ + AS showed that the drug had low efficacy (28 days ACPR = 68%) and could not be considered a potential anti-malarial drug in Tanzania and other malaria-endemic countries [41]. It is plausible that since AZ is a common antibiotic in the treatment of trachoma, the local parasites might have been exposed to the drug leading to development of resistance [57]. This could have possibly compromised the efficacy of AZ + AS combination. An alternative explanation for the observed lower efficacy of AZ + AS compared to adults in Asia is that the effective dose of AZ absorbed in often-malnourished African children might not be sufficient to achieve adequate cure rates. Malnourishment is known to reduce drug absorption [8] and cure rates among patients treated with different drugs. Furthermore, a recent review of AZ across continents for treating uncomplicated malaria revealed that AZ has low efficacy as a monotherapy for treatment of uncomplicated malaria and when used in combinations with other anti-malarials, it may need to be used at high doses which may affect tolerability to the drug [58].

Measurement and reporting of parasite clearance on day 3 after treatment with ACT is particularly important, as this is one of the first signals of emergence of parasite tolerance/resistance to artemisinin [23]. In the present review, two studies reported day-3 parasitaemic cases of 1.1 and 1.4% after treatment with AL [36,38] and the day 3 parasite positivity rates were lower than what has been previously reported [21]. However, the parasite positivity rate reported on day 2 in one of the studies conducted in Muheza district with moderate malaria transmission was higher than the rates reported in previous studies [21]. Thus, more studies will be required to confirm these findings and their role in possible emergence of artemisinin resistance. Although the proportion of patients with detectable parasitaemia on day 3 serves as a simple measure of parasite clearance time at the population level [21], it is often influenced by the baseline parasite density and the timing of parasite sampling, which can vary within and across studies. On the contrary, parasite clearance half-life doesn’t depend on baseline parasite density and is thus considered a more reliable indicator of changes in parasite susceptibility to artemisinin. Measurement of parasite clearance at six-, eight- or 12-hourly intervals for the first 72 hours, as it is currently recommended [59,60], provides a population level profile and useful data of parasite sensitivity to artemisinin. More accurate estimates of parasite clearance through frequent parasite counts are recommended [59,60]. However, the studies reviewed in this article were based on 24-hour sampling, which is not the recommended method for assessing parasite clearance and detection of tolerance/resistance to artemisinins.

Artemisinins are known to be highly potent anti-malarial drugs that are active against immature gametocytes and are useful in the reduction of malaria transmission and elimination/eradication agenda [61]. In clinical trials reviewed in this paper, it was shown that in fact AL and AS + AQ have potentials to reduce gametocyte carriage [30,34,37,40]. However, the unusual increase in gametocytes from four on day-0 to 68 sexual parasites per 500 leucocytes on day-2 post AL treatment, as reported in one of the trials [34], needs to be further evaluated in the light of changes in the parasite sensitivity to ACT. Gametocyte clearance by ACT has also been documented by other studies in East Africa [62] and elsewhere [63,64] where significant reduction of gametocytes by day 14 after treatment with AL or AS + AQ was observed, indicating the potential advantages of ACT over non-artemisinin-based anti-malarials.

It is well established that the efficacy of AL combination is strongly influenced by variations in the pharmacokinetics of lumefantrine among individuals [8]. The maximum therapeutic cure rate is achieved when the plasma drug concentration is adequately available for at least six days [65]. Measurements of day-7 plasma lumefantrine levels are particularly important in unsupervised trials as a measure of adherence to treatment, rather than the use of questionnaires [66]. Day-7 lumefantrine concentrations were significantly lower in unsupervised patients suggesting lower adherence to the drug dosage or fat intake advice. However these differences did not affect the cure rates and high therapeutic efficacy was achieved even in the unsupervised group, indicating that the parasites are highly susceptible to lumefantrine. The observation by other studies in East Africa which showed the median lumefantrine levels were significantly lower in unsupervised patients, but without any effects on the cure rates [67,68], lends support to the findings of this review. It is clear that a high day-28 AL cure rate can be achieved despite low plasma lumefantrine levels, even among unsupervised patients. However care should be taken to avoid exposure of parasites to sub-therapeutic levels of the drugs and creating favourable conditions for emergence of lumefantrine resistance [69]. Monitoring of lumefantrine tolerance/resistance should also be implemented in order to safeguard usefulness of AL.

The present review showed no unexpected adverse events and overall, AL, AS + AQ and AZ + A were well tolerated. Admittedly, the few studies that reported safety profile in the present review (e.g., only one study reported safety data on AS + AQ while the rest reported AL safety) would not enable a firm comparison of safety of different anti-malarials. However, other studies in Africa have shown that certain mild or moderate adverse events, such as vomiting and anaemia, were more frequent in patients treated with AS + AQ than those treated with AL [70,71]. This review has shown that respiratory infection, including cough, was the most frequent adverse event in children treated with AL. This is in line with previous findings which showed that respiratory infections were common in African children with malaria [72,73]. The findings from a recent review on the safety of AL with other ACT in children [73] showed that adverse events were attributed to symptoms or progression of malaria and not directly to the drugs, and this lends support to the findings of this review. Thus, AL and AS + AQ are safe when used for treatment of uncomplicated malaria.