Background
Nearly half the world’s population is at risk of contracting malaria [
1]. Since the year 2000, the prevalence of its most common and dangerous causative parasite,
Plasmodium falciparum, has more than halved, leading to the prevention of an estimated 500 million clinical cases of malaria between 2000 and 2016 [
2]. This progress has been largely attributed to the scaling-up of vector control tools (VCTs), predominantly long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS), both of which are now used in malaria-endemic regions across the globe [
3]. However, there has been growing concern that an increase in resistance among mosquito vectors to the pyrethroid-based insecticides used in LLINs and IRS is hampering further progress [
2]. In addition, it has been suggested that in response to the scaling up of LLINs and IRS which target mosquitoes attempting to feed on humans indoors, the mosquito vectors may be modifying their feeding behaviour to outdoor feeding around dawn and dusk, thus reducing the efficacy of LLINs and IRS [
4]. This has led to calls for the development of non-pyrethroid-based insecticides [
3] as well as the development and adoption of more effective VCTs specifically targeting outdoor biting [
5].
Plant sugars are an essential dietary component for female and male mosquitoes, with female mosquitoes combining this with protein obtained from blood meals to metabolize egg development [
6]. Targeting this aspect of the mosquito life cycle using attractive targeted sugar baits (ATSB) [
7,
8] (also referred to in the literature as attractive toxic sugar baits) has, therefore, been proposed as a potential strategy that may complement LLINs and IRS in suppressing mosquito vector populations [
9‐
11]. ATSBs provide a manufactured sugar-based alternative to plant sugars for female and male mosquitoes with the addition of a toxin that rapidly kills on ingestion or contact. Previous studies have found the effect of ATSBs to be two-fold. First, ATSBs suppress the overall mosquito population by reducing the numbers of female and male mosquitoes available for reproduction. Second, ATSBs diminish the number of mosquitoes living long enough to pass on the malaria parasite, since many are killed before completing the extrinsic incubation period [
12‐
14]—the length of time between a mosquito biting an infectious human and becoming infectious themselves, typically of the order of ~ 10 days [
15]. The suppression of the mosquito population (entomological endpoint) due to the use of ATSBs is expected to result in the reduction of malaria prevalence and clinical incidence (epidemiological endpoints). Although several VCTs have confirmed the entomological endpoint (reduction in catch numbers of several mosquito vector species) attributable to ATSBs [
10‐
14], there is currently no empirical data demonstrating a link between the entomological and epidemiological endpoints for this tool.
A previous study [
8] by Marshall et al
. used a mathematical modelling approach, parameterized using data from a previous ATSB entomological field study in Mali [
7], to understand the entomological impact of ATSBs in a West African setting. Mosquito catch number reductions of ~ 80% were projected over a timescale of a few weeks during a time period when the mosquito catch numbers at a control location remained approximately constant. One of the key parameters determining the efficacy of ATSBs on entomological endpoints was the bait feeding rate parameter, which describes the probability of a particular mosquito ingesting the bait on a given day. This was estimated to be ~ 0.4–0.5 per day—a rate substantially higher than the baseline mosquito death rate of ~ 0.1 per day in this setting. Given that mosquito lifetime following ATSB consumption was estimated to be just a few hours [
8], this implies a very large reduction in average mosquito lifespan in the presence of ATSBs.
Data from the first cluster-randomized entomological study [
16] of ATSBs in Africa was combined with mathematical modelling to explore the potential utility of this new tool to reduce
P. falciparum malaria prevalence and clinical incidence in humans. The cluster-randomized entomological study was undertaken in southern Mali between April 2016 and December 2017, with the efficacy measured through one-day tests using non-toxic stained bait (to estimate the bait feeding rate), monthly mosquito catches in intervention and control villages and monthly estimates of the entomological inoculation rate (EIR)—the number of infective bites received per person per unit time (to estimate onward infectivity to humans). To estimate the subsequent impact on human endpoints, a mathematical model [
17‐
19] of the transmission of
P. falciparum malaria to incorporate the presence of ATSBs was adapted, with the field study data being used to estimate key parameters for the model. The model was applied across a range of malaria transmission settings capturing different transmission intensity and seasonality to evaluate the potential utility of ATSBs as an additional VCT.
Methods
Entomological study of ATSB
A Phase II entomological study (previously reported in this journal [
16]) was undertaken in 14 villages in central Mali. The climate in this region is highly seasonal, with high rainfall in the rainy season (peaking in September) and very dry conditions in the dry season (December–March). Full details of the study and outcomes are reported elsewhere and are summarized here for completeness.
Fourteen villages were selected to participate in the study. In the first year of the study (April 2016 to May 2017), baseline entomological data were collected in all 14 villages. They were randomly sorted into two groups of seven, with one group designated as the intervention (ATSBs + standard of care) group and one as the control (standard of care) group. ATSBs were then deployed in the intervention villages in June 2017 with two bait stations containing the insecticide dinotefuran being placed on the outer walls of each building, and entomological data collected through to December 2017. To estimate the feeding rate on the ATSBs, 1-day tests using stained bait were carried out at monthly intervals in the control villages. As a means of estimating the bait-feeding rate, simple tests were carried out in which attractive sugar baits without toxic additives (ASBs) were temporarily introduced to villages where ATSBs were not used. These baits contained a harmless dye which allowed captured mosquitoes in the relevant villages to be separated into those which had fed on the ASBs and those which had not.
Mosquitoes were collected monthly in each village using Centre for Disease Control (CDC) UV light traps, Malaise traps and pyrethroid spray catch (PSC) inside houses. Here the data from CDC traps is used as a measure of mosquito density. In addition, human landing catch (HLC) measurements were carried out indoors and outdoors by four volunteers (two indoors in separate homes and two outdoors at least 5 m apart from the indoor volunteers). A random sample of the captured mosquitoes were examined to determine the proportion containing viable sporozoites and, therefore, onwardly infectious; this sporozoite rate is multiplied by the number caught per volunteer per unit time in HLC experiments to estimate the entomological inoculation rate (EIR).
Estimating the impact of ATSBs on mosquito density and EIR
To estimate the impact of the ATSBs on the two entomological endpoints—mosquito count and EIR—a non-linear model was formulated to capture the seasonal variation in outcomes in addition to the effect of the intervention whilst accounting for the village cluster-level variability. The non-linear model with mosquito count outcomes from the CDC light traps had convergence difficulties. Mi denotes the mosquito population count in village i where i = 1, 2…14 (7 treatment and 7 control villages) at time t [indexing the month of the year in which count of mosquito outcome was obtained with July being the first month (t = 1) and November the fifth month (t = 5)].
Equation
1a shows the trigonometric function, which captures the seasonal variation in the mosquito population density for an individual village where
aM, bM, cM, dM are parameters to be estimated. The term
RM denotes the treatment effect coefficient for ATSB (the fractional decrease in population density) while
δT is the predictor that identifies treatment assignment at village level (coded as 1 and 0 for treatment and control villages respectively). The variation in count between villages was captured by a Poisson distribution with its mean based on Eq.
1a (Eq.
1b).
$$M\left( {\text{t}} \right) = \left( {1 - {R_M}{\delta_T}} \right)({{\text{a}}_M}\sin \left( {{{\text{b}}_M}{\text{t}} - {{\text{c}}_M}} \right) + {{\text{d}}_M})$$
(1a)
$${{\text{M}}_{\text{i}}}\left( {\text{t}} \right)\ \sim{\text{Poi}}({\uplambda } = {{\text{e}}^{\left( {{\text{r}}{_{{\text{M}},{\text{i}}}} + \left( {1 - {{\upbeta }_M}{{\updelta }_{\text{T}}}} \right)} \right)}} \cdot \left( {{{\text{a}}_M}\sin \left( {{{\text{b}}_M}{\text{t}} - {{\text{c}}_M}} \right) + {{\text{d}}_M})} \right)$$
(1b)
The terms
rM,i denotes the random intercept that captures the correlation of mosquito population count at village level assumed to be be normally distributed as
rM,i ~
N(0,σM,r2) where
σM,r2 is a level 2 variance component to be estimated. The reduction
RM in mosquito count can be approximated by the formula
\({R}_{M}\sim \left(\mathrm{exp}\left(1\right)-\mathrm{exp}(1-{\beta }_{M})\right)/\mathrm{exp}(1)\) where
βM is the effect size parameter estimated from fitting the statistical model from Eq.
1b.
The EIR is modelled by a normal distribution with its parameters captured by the model shown in Eq.
2. The mean of the normal distribution was defined by the term
\(\left(1-{R}_{EIR}{\delta }_{T}\right)({a}_{EIR}\mathrm{sin}\left({b}_{EIR}t-{c}_{EIR}\right)+{d}_{EIR})+ {r}_{EIR,i}\) and the residual error variance by
εi.
$$EI{R_i}\left( t \right) = \left( {1 - {R_{EIR}}{\delta_T}} \right)({a_{EIR}}\sin \left( {{b_{EIR}}t - {c_{EIR}}} \right) + {d_{EIR}}) + \;{r_{EIR,i}} + {\varepsilon_i}$$
(2)
Due to the nature of the data, the time t was measured in months in which the EIR value was obtained. The months were coded as t = 1 to 7 representing the months from June to December. As with Eq. (1), the fixed effect part of Eq. 2 containing the term aEIR sin(bEIRt – cEIR) + dEIR is a trigonometric function that captures the seasonal variation in the EIR where aEIR, bEIR, cEIR, dEIR are parameters to be estimated. The term REIR denotes the treatment effect coefficient for ATSB (the fractional decrease in EIR) while δT is the predictor that identifies treatment assignment at village level (coded as 1 and 0 for treatment and control villages, respectively). The terms rEIR,i and εi denote the random intercept that captures the correlation of EIR at village level and the residual variability, respectively. These are assumed to be normally distributed as rEIR,i ~ N(0,σEIR,r2) and εij ~ N(0,σ EIR,e2) where σ EIR,r2 and σ EIR,e2 are variance components to be estimated.
In both cases, the model was fitted to the mosquito catch and EIR data using PROC NLMIXED with the adaptive Gauss-Hermite quadrature method [
20] in Statistical Analysis System (SAS) software version 9.4 [
21] to obtain the point estimate for the parameter
R together with its corresponding 95% confidence interval based on a two-sided p-value for the null hypothesis
\({H}_{0} :R=0\) versus the alternative
\({H}_{A} :R\ne 0\). The parameter values extrapolated from population and EIR data are shown in Additional file
1.
Estimating the excess mortality
Equation 3a expresses the rate of change of the mosquito catch in ATSB villages
MEXP following the introduction of ATSBs. This expression, based on the approach taken by Marshall et al
. [
8], is a simplified version of the more detailed mosquito population model, used here as a means of relating the function fitted to the observed data (Eq.
1) to mosquito mortality parameters.
$$\frac{{d{M_{EXP}}\left( t \right)}}{dt} = {\mu_{BASE}}{M_{EQ}}\left( t \right) - \left( {{\mu_{BASE}} + {\mu_{ATSB}}} \right){M_{EXP}}\left( t \right)$$
(3a)
$$\frac{{d{M_{CON}}\left( t \right)}}{dt} = {\mu_{BASE}}{M_{EQ}}\left( t \right) - {\mu_{BASE}}{M_{CON}}\left( t \right)$$
(3b)
Here
MEQ is the seasonally varying equilibrium mosquito catch,
µBASE is the baseline adult mosquito appearance and death rate in the absence of ATSBs, and
µATSB is the excess mortality due to ATSBs.
µBASE is given by the natural mosquito death rate
µNAT added to any additional mortality due to vector control interventions present in both control and ATSB villages. In control villages, the mosquito catch
MCON is given by the same equation with
µATSB set to 0 (Eq.
3b).
From Eq.
1a, the average mosquito catch rate in the control and ATSB arms can be written as shown in Eqs.
4a, b.
$${M_{CON}}\left( t \right) = {a_M}\;{\text{sin}}\left( {{b_M}t - {c_M}} \right) + {d_M}$$
(4a)
$${M_{EXP}}\left( t \right) = \left( {1 - {R_M}} \right){M_{CON}}\left( t \right) = \left( {1 - R} \right)\left[ {{a_M}\;{\text{sin}}\left( {{b_M}t - {c_M}} \right) + {d_M}} \right]$$
(4b)
The relationship between
MEXP and
MCON (Eq. 4b) can be substituted into Eq.
3a (Eq.
5a). Equation 5a and Eq. 3b can then re-arranged to give two different expressions for
MEQ (Eqs.
5b–c). These can then be equated in order to express the relationship between
MCON,
RM,
µBASE and
µATSB (Eq.
5d).
$$\left( {1 - {R_M}} \right)\frac{{d{M_{CON}}}}{dt} = {\mu_{BASE}}{M_{EQ}} - \left( {1 - {R_M}} \right)\left( {{\mu_{BASE}} + {\mu_{ATSB}}} \right){M_{CON}}$$
(5a)
$${M_{EQ}} = \frac{{1 - {R_M}}}{{{\mu_{BASE}}}}\left( {\frac{{d{M_{CON}}}}{dt} + \left( {{\mu_{BASE}} + {\mu_{ATSB}}} \right){M_{CON}}} \right)$$
(5b)
$${M_{EQ}} = \frac{1}{{{\mu_{BASE}}}}\frac{{d{M_{CON}}}}{dt} + {M_{CON}}$$
(5c)
$${\mu_{ATSB}} = \frac{{R_M}}{{1 - {R_M}}}\left( {\frac{1}{{{M_{CON}}}}\frac{{d{M_{CON}}}}{dt} + {\mu_{BASE}}} \right)$$
(5d)
Equation 5d can be rewritten as follows with Eq. 4a used to substitute for
MCON, to give an estimate of
µATSB in terms of the estimated parameters
aM,
bM,
cM,
dM,
RM and the base death rate:
$${\mu_{ATSB}} = \frac{{R_M}}{{1 - {R_M}}}\left( {\frac{{{a_M}{b_M}\;{\text{cos}}\left( {{b_M}t - {c_M}} \right)}}{{{a_M}\;{\text{sin}}\left( {{b_M}t - {c_M}} \right) + {d_M}}} + {\mu_{BASE}}} \right)$$
(6)
Estimating the impact of ATSBs on malaria prevalence and incidence
An existing detailed model [
17‐
19,
22] of malaria was used for simulations of the effects of ATSBs on malaria infection levels in human populations. In the model, individuals begin life susceptible to
P. falciparum infection and are exposed to infectious bites at a rate that depends on local mosquito density and infectivity. Newborn infants passively acquire maternal immunity, which decays in the first 6 months of life. After exposure, individuals are susceptible to clinical disease and may progress through a range of infection categories (clinical infection, asymptomatic infection, subpatent infection, treated and prophylaxis). As they age, the risk of developing disease declines through natural acquisition of immunity, at a rate that depends on the rate of continued exposure. At older ages, parasitaemia levels fall so that a high proportion of asymptomatic infections become sub-microscopic. Full mosquito-population dynamics were included in the model to capture the effects of vector control in preventing transmission, killing adult female mosquitoes, and the resulting reduction in egg-laying. The model has previously been fitted to existing data on the relationship between rainfall (the seasonal parameter found to give best fit to data [
22]), mosquito abundance, entomological inoculation rate (the rate at which people receive infectious bites), parasite prevalence and clinical disease incidence in order to establish parameter values. Full mathematical details of the model and a complete parameter list are included in Additional file
1.
The effect of ATSBs was included in the model by modifying the death rate of mosquitoes from
µBASE to
µBASE +
µATSB as shown in the previous section. Note that this differs from the modelling of other common vector control interventions such as LLINs and IRS, where direct reduction in biting rate must also be incorporated and additional mortality is affected by biting rate [
23].
The initial conditions for a study were created by generating characteristics (proportions of humans in different infection categories, immunity levels) at steady state under particular levels of adult mosquito density, then after an extended period of time with particular seasonal variation in adult mosquito density. ATSBs were then introduced to modify the mosquito death rate, resulting in reduced mosquito populations due to direct death and reduced larval birth rate. As noted above, the population of infectious mosquitoes decreases more significantly than the overall population, due to increased death rates causing fewer infected mosquitoes to survive for the duration of the parasite incubation period. This in turn caused reductions in EIR which in turn reduced the number of new infections. Benchmark data values including malaria prevalence and clinical incidence were recorded at regular intervals and the results compared with the same data values under control conditions (where the mosquito death rate is simply equal to the natural value µNAT) to measure the effectiveness of ATSBs.
Discussion
The results from the first cluster-randomized entomological field study of ATSBs [
16] demonstrate the potential of this new tool to significantly suppress
Anopheles catch numbers, confirming results from earlier studies [
8,
11,
13]. Using this data, a statistically significant reduction in the mosquito count in the villages with ATSBs and LLINs is estimated compared to those with LLINs alone of 57% (95% CI 33–72%) over a 1-year follow-up period. Notably, this effect is most apparent in the reduction in the seasonal peak mosquito catch rate concomitant with the period of highest malaria transmission. Furthermore, a greater estimated reduction in onward transmission as captured by the EIR of 91% (95% CI 75–100%) is obtained, reflecting the impact that this intervention is likely to have in reducing the lifespan of mosquitoes and hence the potential for mosquitoes to survive the extrinsic incubation period. These modelling results suggest that these large reductions in vector populations should translate to significant public health impacts, with > 30% reductions in both parasite prevalence and clinical incidence predicted across a wide range of transmission settings and across a range of potential values for the excess mortality.
While theoretically, reductions in entomological endpoints should lead to reductions in epidemiological endpoints, this relationship is not always clear empirically. However, it is worth noting that the observed reductions in mosquito catch and EIR are similar to or greater than those observed to date in cluster-randomized studies of other VCTs. For example, a large cluster-randomized trial of ITNs in Western Kenya [
24] showed a 58.5% reduction in
Anopheles gambiae and 90% reduction in
Anopheles funestus captured using pyrethrum spray collection
. This same trial resulted in a reduction in malaria incidence and prevalence in young children of 60% and 19% respectively [
25,
26]. However, extrapolation from entomological endpoints alone cannot be made for ITNs since their efficacy will represent both direct and indirect protection. Whilst there are no cluster RCTs of IRS alone [
27], in recent years the impact of LLINs and IRS on mosquito populations has been assessed in two large cluster-randomized trials comparing the benefits of combining IRS and LLINs. In the first in the Gambia (in which IRS did not show any additional epidemiological benefit in addition to LLINs), mosquito counts were 33% lower in the LLIN + IRS group compared to the LLIN-only group, but this difference was not statistically significant. In this study, the EIR was low and not statistically different between intervention arms. In the second trial in Tanzania mosquito counts were reduced by 29% and EIR by 87% in the LLIN + IRS arm compared to the LLIN-only arm, although these differences were of marginal statistical significance. However, this translated to a 50% reduction in parasite prevalence in the LLIN + IRS arm compared to the LLIN-only arm. The projected epidemiological impact of ATSBs based on the observed entomological endpoints therefore appears plausible but requires confirmation in epidemiological randomized trials.
One of the key parameters determining the likely efficacy of ATSBs is the excess mortality, which, given the lethality of the toxin, is primarily determined by the rate at which mosquitoes feed on the bait. From the observed reduction in mosquito catch numbers, this excess mortality is estimated to be in the range ~ 0.07–0.11/day, effectively at least doubling the natural death rate of
Anopheles mosquitoes. These estimates are notably lower than the estimates of the bait feeding rate obtained by Marshall et al
. [
8] (0.40/day) and the values estimated here in the one-day staining experiments using dyed, non-toxic bait (0.28–0.38/day in the period of ATSB use) although they are more consistent with the bait feeding rate in the control arm of the study by Marshall et al
. (0.15/day for female mosquitoes).
There are a number of possible explanations for this discrepancy. Firstly, the base mosquito mortality may be higher than the natural mortality rate due to the presence of other vector control interventions, as noted in Sect. 3.2. This can lead to an under-estimation of the ATSB mortality rate, as shown in Fig.
2b, where the extrapolated ATSB mortality rate increases if the base mortality rate is increased. In addition, short-term experiments may give rise to higher values than those observed over longer time periods due to variations in the bait-feeding rate between mosquitoes—if some mosquitoes are more disposed to feed on the bait than others, these will be killed early on and over time the bait-feeding rate may decline.
Another possibility is that migration of mosquitoes from areas unaffected by the ATSBs acts to mitigate the population reduction and produce a lower apparent increase in the death rate. The villages in which the data reported here was gathered are located in the flood plain of the Niger river with rice paddies and other breeding sites nearby. In contrast, the previous small-scale village studies in central Mali [
8] were carried out at the end of the rainy season without significant nearby breeding sites and this difference may account for the higher apparent excess mortality. Thus, whilst the one-day staining experiments can continue to provide useful information, particularly on the variation in the bait feeding rate between different ecological locations [
11,
12,
14,
28], it will be important to continue to collect entomological outcomes in future RCTs of this new intervention.
The bait feeding rate estimated here from the one-day staining experiments was found to be negatively correlated with the rainfall level, suggesting that the ATSB efficacy may vary seasonally. This seasonal effect may be due to greater availability of natural sugar sources such as flowers and fruit inside villages during the wet season, providing alternative attractant sugar sources for both male and female mosquitoes. Alternatively, the observed feeding rates of mosquitoes inside the villages may be lowered by large numbers of already sugar fed mosquitoes invading the villages from nearby breeding sites. If this is the case, it suggests that the overall efficacy of ATSBs can be expected to vary significantly between ecological and/or geographical settings, with locations where natural sugar sources are more abundant showing reduced efficacy. In this context, the impact of invasive plants flowering during the dry season also needs to be considered [
28]. Previous studies of ATSB in arid environments with varying levels of natural sugar availability suggest that natural sugar sources have only a delaying effect on ATSB efficacy [
13]. However, further data are needed from the range of malaria-endemic environments to confirm this.
There are a number of limitations to this study. Firstly, the projected impact of ATSBs made here are based on the results of a single field study in a single ecological zone. These results should, therefore, be interpreted as indicative rather than predictive. Secondly, as noted above, the excess mortality can be expected to vary between different ecological environments and at different times of the year, depending on the availability of alternative sugar sources. Thus, whilst a major impact on both malaria and clinical incidence is predicted across a range of potential excess mortality values, if the feeding rate on the bait is very low then a sharp reduction in efficacy would be expected. Thirdly, trials have not yet been carried out to assess the effect of ATSBs on epidemiological outcomes in the field. Evidence from past VCT trials has demonstrated that it can be difficult to extrapolate epidemiological outcomes based on entomological outcomes alone given that many other factors could vary between settings. Finally, the large impact of ATSBs on mosquito catch numbers observed in the study in Mali, if replicated elsewhere, could be expected to exert a strong selective pressure. Adaptation of mosquito behaviour in response to the presence of ATSBs could reduce efficacy in the longer-term, making it difficult to predict the sustainability of this new tool, although it may be possible to mitigate this through the use of multiple toxins.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.