This study investigated the efficacy and residual activity of two WHO-prequalified biolarvicides, VectoBac® GR and VectoMax® FG 21, against the two major malaria vectors in Tanzania, An. arabiensis and An. funestus in habitats that were either shaded or unshaded. The larval development period of each species was also determined. The data are considered necessary to inform potential larviciding dosing and retreatment intervals for future studies.
Overall, this study found that the two biolarvicides had very high efficacies (> 98%) against both early and late instars of An. funestus and An. arabiensis within 72 hours of habitat treatment. However, they had short residual effects, lasting just one week in sun-exposed habitats. Increasing the treatment dose in sun-exposed habitats did not increase the residual efficacy of biolarvicides. However, increased residual efficacy, with approximately 60% reduction of L3-L4 of An. arabiensis and An. funestus during the second week was observed in shaded habitats treated with VectoMax®. These findings are consistent with previous research in Sub-Saharan Africa that reported low residual efficacy of biolarvicides in sun-exposed habitats 27,30–32. This is a limitation of biolarvicides and may have implications for their use in Tanzania and elsewhere since not all habitats are shaded or vegetated 28,29.
In this study, we also aimed to assess the larval development period of An. funestus and An. arabiensis in both natural and semi-natural habitats. The estimated minimum and mean larval development period were similar for An. arabiensis in natural [minimum 6 days, mean 8.32 (CI: 5.0-11.6) days] and semi-natural habitats [minimum 6 days, mean 8.2 (CI: 5.8–10.6) days]. There were slight differences for An. funestus, which had a minimum of 8 days and a mean of 10 days (CI: 6.6–13.5) in natural habitats and a minimum of 9 days and a mean of 13.2 days (CI: 10.4–16.0) in semi-natural habitats. The assessment of larval development in natural habitats was initiated with the first instars found in the habitats. Since we did not know their exact time after hatching, several first-instar larvae might likely have already spent more than one day in the natural habitats before we started our assessments, which could have led to an underestimation of the larval development period. Factors such as temperature, food, and predators may affect larval growth 33–35 and could also have led to the differences in larval development time between the habitats for An. funestus.
In both natural and semi-natural habitats, we observed a large difference in the larval development periods of the two vectors, with An. funestus larvae spending an average of thirteen days before pupating, compared to eight days for An. arabiensis. This observation, alongside other logistical and environmental constraints, is an important factor when planning the retreatment schedules for larvicide applications in the aquatic habitats of these two vector species. Since early instar larvae were constantly observed in the natural habitats, despite the development of larvae, this indicated that new oviposition events were frequent in the study area. Thus, effectively targeting such habitats requires that the larviciding frequency does not exceed the larval development period of the target vectors. For An. funestus, any retreatment every two weeks would ensure that all larvae in the habitats contact the larvicide at some point in the period before pupation. On the other hand, for An. arabiensis mosquitoes, which mature faster and have far shorter larval development periods, weekly applications would ensure that the larvae are exposed to the biolarvicide at some point during their development before pupation. As such, in the south-eastern Tanzania villages, where this study was conducted, and where An. funestus typically dominates transmission 36–38, biweekly biolarvicide treatments would be required. That consideration notwithstanding, it should also be noted that any higher frequency of treatment would still probably be more impactful against An. funestus and may be desirable if the two species occur in the same area, as is common in east and southern Africa (Msugupakulya et al, submitted). Mathematical modeling, incorporating these observations, could provide a quick way to explore the impact of such a strategy before deployment.
This study also suggests that several other factors must be considered when deciding on the frequency of larviciding. For example, the sun-exposed habitats, such as those usually frequented by some Anopheles mosquitoes such as An. gambiae and An. arabiensis may present challenges due to the reduced persistence of biolarvicides and may require more frequent retreatment and alternative larval source management strategies. In contrast, shaded habitats, such as the vegetated streams and wells under tree canopies, both preferred by An. funestus, may be more amenable to biweekly biolarviciding since in shaded habitats efficacy remained at about 60% in the second week with VectoMax®.
The efficacy of biolarvicides and the larval development period are also important factors to consider when choosing a biolarvicide for a larviciding program. Indeed, a desirable product is one that effectively and persistently targets different larval instars as this would minimize the operational costs of larviciding 39,40. Our study used biolarvicides that are relatively short-acting, VectoBac® and VectoMax®. When considering products such as these, it is important to consider the larval development period of local vectors in choosing the optimal retreatment frequency of habitats. A few longer-acting biolarvicides are also available, which require less frequent retreatment and for which larval development time will play a lesser role, such as LL3 and FourStar® 40. However, when selecting longer-acting biolarvicides, it is important to consider the nature of habitats in the target area. This is because these biolarvicides have reduced efficacy in flowing habitats and may not be suitable for areas with vectors that prefer flowing waters such as streams 40.
Some authors have suggested weekly treatments 25,32, which match the life cycle parameters for An. arabiensis in this study, and would potentially be more impactful against An. funestus, which has slower larval growth. They also urge that weekly application is especially important in areas with dynamic habitats and during the rainy seasons to ensure effective dose even for new temporary habitats that form, and crucial for the larviciding team to get familiar with the target area 25. While this is true, it may not be practical nor effective to implement larviciding during the rainy season as habitats are typically numerous and unstable, and larvicides are more likely to be diluted and washed away by rainwater 25. The implementation of larviciding may be more feasible and effective in the dry season 13,14, because during this season new habitats hardly form, and dilution or washing away of larvicides is not expected. Besides, the decline in populations of immature mosquitoes in natural habitats observed in this study, particularly for the An. arabiensis experiment implies that the densities of immature mosquitoes decline over time during the dry season. Therefore, larval control efforts during this period are likely to be more effective than in wet seasons and would add further stress to the already declining population.
Although successful, this study was not without limitations: First, there are several factors, including the physicochemical parameters of water, which may have affected the efficacy of biolarvicides 23. The design of this study did not permit the assessment of such other factors affecting the efficacy of biolarvicides. Second, this study was conducted in semi-natural habitats made of plastic containers, which may not reflect the conditions of aquatic habitats in the field as most Afro-tropical Anopheles are not container breeders. Nevertheless, to minimize these potential effects, soil and vegetation were added to the habitats, habitats were sunk into the surrounding soil, and habitats were allowed to condition for several days so that the habitats could better mimic the field environment. Third, because dead larvae play a role in the regeneration of B. sphericus spores in aquatic habitats 41 it is likely that our procedure which involved the removal of dead larvae from the habitats when that was possible may have contributed to the reduced residual efficacy of VectoMax®.
Lastly, the scope of this study is limited to frequencies and retreatment intervals of larviciding, however, an effective larviciding program would also require a proper plan to identify, map, and monitor suitable aquatic habitats of the vectors 13,42. There is also a need to choose suitable methods to effectively deliver larvicides to the target habitats, depending on the available resources and expertise. This could include community participation or aerial technologies 43,44. In addition to being effective, larviciding programs must also be sustainable and acceptable. This requires strong collaboration with different stakeholders and continuous engagement and incorporating their needs and recommendations 13.