E-interview with Marc Coosemans, senior full professor, medical entomology unit, Institute of Tropical Medicine, Antwerp, Belgium. The team and collaborators of Prof. Coosemans were recently awarded a $ 3 million grant by the Gates Foundation, to study the effects of community-wide use of topical repellents on malaria in Cambodia, when used in conjunction with insecticide-treated bednets.
E-interview with Prof. Marc Coosemans (Belgium)
Q: First, congratulations with receiving this grant. It will be interesting to know what led to your choice of adding this novel intervention (community-wide use of repellents) to bednets as a means to protect against outdoor and early evening biting vectors. Historically, outdoor control efforts would have focused on larval source management, so why the choice for repellents?
A: This is a challenging project prospecting a new paradigm in the control of malaria: to raise evidence that malaria can substantially be reduced by controlling early / outdoor biting vectors in addition to bednets. The scaling up of Insecticide treated nets (ITNs), particularly Long Lasting Insecticidal Nets (LLINs) and the expansion of Indoor Residual Spraying (IRS) has contributed to a significant decrease of malaria worldwide. However, these control methods tackle only indoor and night biting vectors. The proportion of transmission occurring before sleeping hours and outdoors referred to as “residual transmission” is steadily increasing and may compromise the efforts towards malaria elimination. Outdoor and early biting malaria vectors are of particular importance in Southeast Asia and present a major challenge in eliminating malaria and especially artemisinin resistant malaria at the Thai-Cambodian border. Our previous entomological studies have shown that about 40-50% of vector bites occur before sleeping time although high variation may occur from one site to the other. Consequently, people are at high risk of being bitten by infected mosquitoes during their evening activities. This means that the protective effect of insecticide-treated nets is limited and that something has to be added. Larval source management can be an effective additional control method but is facing some constraints as a result of the wide spread and the large range of breeding sites for some vector species. In Southeast Asia, the major vector An. dirus is a typical jungle mosquito maintaining malaria foci even if the vector density is very low. Breeding places are temporary and wide spread within the tropical forest making larval control not feasible. The community protection against early biting and/or outdoor biting vectors seems to be a more appropriate option.
Q: Ten years ago, Mark Rowland and colleagues conducted a study in Pakistan where they added a repellent (DEET) to soap (Mosbar). They recorded an additional 23% greater reduction in the odds of malaria when soap usage was combined with bednets (69%), against bednets alone (46%). On what grounds do you foresee that the combined use of nets and repellents in Cambodia can have a larger effect and lead to elimination of malaria as you mentioned on the Belgian radio?
A: Mark Rowland and collaborators published two interesting studies using the Mosbar soap, containing DEET 20% and permethrin 0.5%: one was a household randomized trial using a placebo in the control arm, the other was a case control study. More recently Hill and collaborators did a similar study in Bolivia using a double blind household randomized placebo design. Both authors demonstrate indeed a substantial decrease of malaria infection when repellents are used in combination with ITNs. However these studies addressed the individual protection of repellents against malaria. In Cambodia we will use a randomised community based design to assess protection at the community level . We hypothesize that a mass use of repellents in the community will further affect the vector population remaining outdoors (mass effect), thus also inducing a reduction of the residual malaria transmission and protecting those that refuse or forget to use the repellent .
Q: Even if people are instructed carefully about the use and benefits of topical repellents, and even if they are made to understand the rationale for using them, coverage always wanes after some time. How will you ensure maximum coverage and usage of repellents during the study?
A: Unlike LLINs, repellents were not used in these communities before the start of this study. Therefore, campaigns on correct and regular use of LLINs and repellents in the intervention arm and LLIN use in the control arm are organized by the local teams at the beginning of the trial and are repeated at regular intervals. Leaflets are widely distributed. To avoid any bias we must ensure that the sensitization effort is similar in both arms. Repellent distributors visited all households every two weeks and evaluate the household consumption. Of course nobody will be forced to use the repellents or the LLINs, however in communities where indeed the consumption of repellents or the reported use of LLINs decrease, a special EIC team will try to mobilize the inhabitants. Acceptability, adherence and adequacy of the repellents and LLIN will be estimated in a social science study triangulating qualitative and quantitative methods. For quantitative methods a randomized sample will be used in each community.
Q: If people are left untreated (with repellent) they will receive the bites diverted from those protected. Are you not concerned that you will be diverting bites from treated to untreated people and still remain with the same number of infectious bites, now disproportionately affecting those that forget or refuse to use repellent?
A: This diversion effect is exactly what happens in randomised household design where untreated persons are close to treated ones. Moreover in household based designs, people receiving a placebo realize quiet quickly that their non-repellent lotion is not protecting them against mosquito bites. Switching treatments (placebo for repellent) is then found to be recurrent. The use of a placebo may also create a false perception of protection. This will not occur in a community based trial where the repellent in the treated arm is available for everyone. In the control arm, no placebo repellent will be used because randomization is carried out on community level and this in contrast with a household design where a placebo would be essential. The purpose in the Cambodian study is to measure the additional effect of repellents on the use of ITNs and not the effect of the repellent on its own.
Q: You are going to use picaridin as a repellent, widely hailed as a very safe repellent. Recently, it was shown by James Logan's group that resistance to DEET can be developed within a few generations. Are you not concerned that use of this repellent on such a large scale (you mention 700 thousand bottles of picaridin) will lead to resistance in the vector population?
A: DEET alters the fine-tuning of the insect olfactory system in a very broad way. The spatial configuration of DEET and picaridin molecules are very similar, and probably their mode of action, although other receptors may be present in Anopheles. However, at comparable dose Picaridin gives a longer protection against Anopheles gambiae then DEET showing that the two repellent are not completely the same in terms of performance. Nevertheless, if insensitivity to DEET can be selected in a colony of Aedes aegypti there is no reason to believe that this will not occur in a wild malaria vector population under constant repellent pressure. However, unlike insecticide resistance, this may take more time in nature as repellents will not kill the exposed insects. The efficacy in personal protection should then clearly be monitored over time.
Many thanks for your answers!