Failure of pyriproxyfen at recommended application frequency and doses to control Aedes mosquitoes in Thailand

Thipruethai Phanichat; Vincent Corbel; Benedicte Fustec; Chamsai Pientong; Kesorn Thaewnongiew; Neal Alexander; Hans J. Overgaard

doi:10.1371/journal.pntd.0013042

Abstract

Background/objectives

Mosquito-borne diseases like malaria, dengue fever, and Zika virus remain global health concerns. Pyriproxyfen is effective in controlling mosquitoes by disrupting their development. This study seeks to assess pyriproxyfen’s ability to prevent Aedes aegypti emergence from water sources. It is part of a trial evaluating pyriproxyfen’s impact on reducing mosquito infestation and dengue transmission, verifying its persistence and effectiveness in real-world and laboratory conditions.

Methods

The study was conducted in Khon Kaen province (northeastern region) and Prachuap Khiri Khan province (western region) of Thailand. We assessed pyriproxyfen residual effectiveness, inhibition of mosquito larval emergence and active ingredients among batches in a pyriproxyfen-based mosquito control trial in Khon Kaen. In Prachuap Khiri Khan we evaluated pyriproxyfen effectiveness across various water sources. The active ingredients in two pyriproxyfen batches were analyzed in a Sumitomo laboratory and in an independent laboratory.

Results

Thirty days after field water containers were treated with pyriproxyfen the inhibition of mosquito larval emergence declined to ~60% and 60 days post-treatment the inhibition of emergence was just ~10%. Two batches of pyriproxyfen tested in the laboratory had > 85% inhibition of emergence and the active ingredient concentrations varied from 0.45-0.52%, close to the manufacturer’s specifications of 0.5%. In laboratory experiments, the inhibition of mosquito emergence of pyriproxyfen in different water sources started declining after 42 days. Rain- and groundwater had higher inhibition rates (20–30%) than tap water (~10%) after 98 days. Emergence inhibition rates correlated negatively with water pH (F(1,118) = 5.626, p < 0.001) and positively with total dissolved solids, conductivity, and salinity of the water (F(1,118) = 48.302, p < 0.001), (F(1,118) = 37.022, p < 0.001), and (F(1,118) = 36.699, p < 0.001), respectively.

Conclusions

Pyriproxyfen failed to control Aedes mosquitoes at the recommended application frequency and doses in the field. The potential reasons for lack of effectiveness may be caused by environmental factors, such as pH, water source, and other water characteristics or social factors, such as homeowners’ behaviors and water storage practices. The study underscores the importance of understanding environmental and social factors to tailor application strategies and ensuring sustained efficacy through regular monitoring, particularly in diverse contexts.

Author summary

We examined the effectiveness of pyriproxyfen, a chemical used to control immature mosquitoes by preventing their development, particularly focusing on the Aedes aegypti mosquito, which spreads diseases like dengue. The research was conducted in Thailand and aimed to evaluate how well pyriproxyfen prevents adult mosquitoes to emerge in different water sources and under real-world conditions. Field tests showed that pyriproxyfen’s ability to stop mosquito larvae from emerging decreased over time. After 30 days, the inhibition rate was about 60%, and by 60 days, it dropped to just 10%. In contrast, laboratory tests showed higher effectiveness, with over 85% inhibition. The effectiveness of pyriproxyfen also declined in various water sources, with rain and groundwater maintaining better inhibition rates than tap water. Environmental factors like water pH and dissolved solids also affected the effectiveness of pyriproxyfen. Additionally, human behaviors, such as water storage practices, may have influenced the results. Overall, the findings suggest that pyriproxyfen may not be effective enough in field conditions without adjustments to dosage and application strategies. Regular monitoring and consideration of both environmental and social factors are recommended to improve its performance in mosquito control efforts.

Citation: Phanichat T, Corbel V, Fustec B, Pientong C, Thaewnongiew K, Alexander N, et al. (2025) Failure of pyriproxyfen at recommended application frequency and doses to control Aedes mosquitoes in Thailand. PLoS Negl Trop Dis 19(12): e0013042. https://doi.org/10.1371/journal.pntd.0013042

Editor: Amy C. Morrison, University of California Davis School of Veterinary Medicine, UNITED STATES OF AMERICA

Received: September 13, 2024; Accepted: April 9, 2025; Published: December 22, 2025

Copyright: © 2025 Phanichat et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are in the manuscript and its supporting information files.

Funding: The Research Council of Norway (FRIPRO project no. 250443) and the Norwegian University of Life Sciences funded this studyto HJO. TP received a New Researcher Grant, Mahidol University, Thailand supporting part of Assessing the impact of water quality on the effectiveness of pyriproxyfen. The funders did not play any role in study design, data collection and analysis, decision to publish, and preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Several Aedes mosquito species, particularly those in the subgenus Stegomyia are responsible for transmitting viruses to humans. Globally, Aedes aegypti (L.) is a primary vector of dengue, chikungunya, Zika and yellow fever viruses. Aedes albopictus (Skuse) is regarded a secondary vector and is also important for transmitting these arboviruses. Both species are common in Thailand. They generally utilize household and peridomestic water storage containers for oviposition and development of immature stages. One key strategy for vector control is larval source management either by source reduction (habitat elimination) or container treatment with a chemical or biological agent. In Thailand, larval vector control generally consists of applying an inexpensive organophosphate insecticide (temephos) in water storage containers [1,2].

Effective larval control measures are critical for reducing vector populations and disease transmission. Traditional control strategies, such as the use of larvicides, face challenges due to the increasing prevalence of insecticide resistance among Aedes larvae. Studies in Thailand have reported widespread resistance to commonly used chemical insecticides, including temephos and pyrethroids, emphasizing the need for alternative vector control tools [3,4].

Aedes aegypti has developed operationally significant resistance to temephos and other compounds in Thailand [5] and other areas of Southeast Asia [6–8]. Alternative insecticides and compounds are needed and require validation as replacement chemicals. In the quest for effective mosquito control, pyriproxyfen has emerged as a promising tool. Pyriproxyfen is an insect growth regulator that disrupts the development of mosquitoes, preventing them from reaching maturity and reproducing. This compound exhibits a unique mode of action by mimicking juvenile hormone, interfering with the normal metamorphosis process in mosquitoes without affecting other non-target organisms [9]. The long-lasting effects of pyriproxyfen make it a valuable component in integrated mosquito management programs. Furthermore, its low toxicity to humans and other non-target species adds to its appeal as a safe and sustainable solution for mosquito control [10] and is recommended by WHO for vector control [11,12]. Pyriproxyfen has also been approved by WHO for use in drinking water storage containers [13]. Pyriproxyfen has been shown to be effective in controlling mosquito populations. A systematic review showed that pyriproxyfen granules effectively inhibited emergence by 90–100% for up to 90 days [14]. Pyriproxyfen is applied in various formulations, including sprays, pellets, and insecticide-treated nets. These formulations enable flexible and targeted application methods, allowing for customized mosquito control strategies based on local conditions. The versatility of pyriproxyfen makes it an integral part of integrated vector management approaches, complementing other control measures such as insecticide-treated bed nets and larvicides. Aedes mosquitoes in Thailand have shown various levels of susceptibility to pyriproxyfen. While the insect growth regulator has been effective in some cases [15,16], there have been reports of reduced efficacy [17], highlighting the importance of ongoing surveillance and the need for integrated vector management strategies. However, there is limited research evaluating its residual efficacy and effectiveness under real-world conditions in tropical regions like Thailand. Furthermore, most studies focus on laboratory-based evaluations or short-term trials, leaving gaps in understanding its long-term performance and effectiveness across different socio-environmental settings. This study aims to address these gaps by assessing the characterization and residual effectiveness of pyriproxyfen under field conditions in Thailand. By evaluating its performance in diverse environments, this research provides valuable insights into the factors influencing pyriproxyfen efficacy, which can guide its optimized application in larval source management.

In our cluster randomized controlled trial [18] the pyriproxyfen failed to perform as expected; [19] requiring a need to better understand the intrinsic and extrinsic causes of this failure. We intended to validate the effectiveness and persistence of pyriproxyfen in the field and laboratory. Our specific objectives were to: 1) determine the residual efficacy of pyriproxyfen in inhibiting the emergence of Ae. aegypti in water from key domestic artificial containers in the field; 2) determine the field efficacy of two batches of pyriproxyfen in inhibiting Ae. aegypti emergence; 3) determine the concentration of the active ingredient in two batches of pyriproxyfen from different manufacturing date and expiration dates in the laboratory; and 4) determine the efficacy of pyriproxyfen over three months in water from different types of domestic water supplies. These objectives were addressed in four separate experiments corresponding to the stated objectives. The first three experiments: residual effectiveness in the field; batch effectiveness; and active ingredients were carried out in the context of a cluster randomized controlled trial aiming to assess a pyriproxyfen intervention in reducing mosquito infestation and dengue transmission [18]. The fourth experiment on water quality was set in Prachuap Khiri Khan province investigating whether differences in water quality and sources could impact on the effectiveness and persistence of pyriproxyfen. Such information is important to guide vector control operations, hence ensuring adequate use of pyriproxyfen in different environments.

Methods

Ethics statement

The project was approved by the Khon Kaen University Ethics Committee (KKUEC) (Record No. 4.4.01: 29/2017, Reference No. HE601221, 1 September 2017); the London School of Hygiene and Tropical Medicine Ethical Committee, UK (LSHTM Ethics Ref: 14275, 16 August 2017); the Regional Committee for Medical and Health Research Ethics, Section B, South East Norway (REK Ethics ref: 2017/1826b, 03 March 2018); the Ethics Committee of the Faculty of Tropical Medicine (MUTM Ethics Ref: 2020-045-1, 21 July 2020); and the Ethics Committee of the Faculty of Tropical Medicine—Animal Care and Use Committee (FTM-ACUC Ethics Ref: 023/2020, 22 July 2020).

Study site

This study was conducted in two areas

1). Northeast Thailand: field work was carried out from September 2019 to January 2020 in Khon Kaen district (S1 Fig) in households included in the treatment arm of the cluster randomized controlled trial (RCT), but after the trial was completed. Details of the trial, on which the current study is based, have been described elsewhere [18]. Briefly, the trial was carried out in Khon Kaen and Roi Et urban districts in northeastern Thailand during 2018–2019, where 18 out of 36 clusters, 10 houses per cluster, were treated with pyriproxyfen, with nine treatment clusters in each district. Pyriproxyfen was applied to all permanent indoor and outdoor household containers that contained water. The recommended WHO dose of 0.01 mg/L active ingredient (a.i.) pyriproxyfen was used with a single treatment applied every 3 months [20] according to manufacturer’s guidelines starting in June 2018, followed by treatments in September 2018, December 2018, March 2019, and June 2019. For these treatments Sumilarv 0.5G sachets (Sumitomo Chemical Group Co. Japan) were used with an active ingredient content of 0.5% w/w, manufactured in July 2016 and expired in July 2019 (batch no. 6710F4).
2). In Western Thailand, the effectiveness of pyriproxyfen in various water sources was assessed outside of the RCT. Prachuap Khiri Khan is situated along the western coastline of the Gulf of Thailand (S1 Fig), with a total area of 6,368 km² encompassing agricultural fields (45%), forested areas (42%), and residential areas (7%). Due to rural development initiatives, basic infrastructure such as main roads, electricity, and tap water are now prevalent throughout the province, fostering tourism and industrialization, particularly in production of preserved fruits. Economic growth has reshaped social dynamics, transforming rural areas into semi-urban and urban areas. These social changes, coupled with a decentralized dengue control policy, may have contributed to increased dengue incidence. From 2005 to 2009, the Ministry of Public Health, Thailand, designated Prachuap Khiri Khan as a high-risk area for dengue [21]. The province features an earth dam that blocks the Pran Buri River, forming a large reservoir. Processed water from the reservoir is distributed as tap water by the Provincial Waterworks Authority (PWA) to urban and some semi-urban areas. However, water management in rural and semi-urban areas falls under the purview of subdistrict administrative organizations. Nong Ta Team subdistrict, Pran Buri District, Prachuap Khiri Khan was chosen for water collection due to its semi-urban nature and diverse water supply sources.

Residual effectiveness in the field

Key container types. A key container type is defined here as a category of container that is generally mosquito positive, producing high numbers of immatures (larvae and pupae) per container. Key container types were determined using entomological data collected during 2017–2019 from a total of 180 households in Khon Kaen. The key container determination was based on proportion of container types that were mosquito positive and mean number of immatures per container. Two types of key containers were identified: a) medium-sized round-shaped, glazed, fired, earthen ‘dragon’ jars and b) large-sized square-shaped cement containers (Fig 1). Ten houses were randomly selected from five of the nine clusters included in the intervention arm of the RCT. One container of each type was selected in each household, resulting in 20 key containers for treatment and follow-up (Table 1). Container selection was done by convenience sampling in houses that had both dragon jars and tanks and whose household owners allowed water to be collected during the study period.

Download:

Table 1. General container characteristics of 20 containers used in the study.

https://doi.org/10.1371/journal.pntd.0013042.t001

Download:

Fig 1. Left panel: Example of medium-sized round dragon jar (ca. < 50 cm height).

Right panel: Example of large-sized square containers (ca. > 50 cm in height and ca. > 100 cm long). Photo credits by authors.

https://doi.org/10.1371/journal.pntd.0013042.g001

Container treatment with pyriproxyfen

These 20 containers were treated on 29 September 2019 with 0.01 mg/L active ingredient pyriproxyfen following WHO recommendations [7]. The pyriproxyfen used was Sumilarv 0.5G (Sumitomo Chemical Group Co. Japan) with an active ingredient content of 0.52% w/w and a manufacture date of March 2018 and expiry date of March 2020 (batch no.’s 8305F4 and 8306F4). These treatments were done after the cluster randomized controlled trial (RCT) had finished in June 2019.

Water collections

Water samples were collected at eight time points, once before treatment and seven times after treatment (Table 2), from each of the 20 selected containers using a water dipper and a measuring cylinder. Water was collected between 8.00 and 11.00 am to ensure consistency in sampling conditions. At each time point, 1500 ml of water was taken from each container. For convenience this was divided between 15 samples, each taken after stirring the container. Water samples were stored in clean (sterile) sealable plastic bags (Whirl-Pak. Nasco, Madison, WI, USA) and were sent to the Department of Entomology, Faculty of Tropical Medicine, Mahidol University, Bangkok where larval bioassays were conducted within 4–5 days of sample collection. At each time point, the temperature, pH, conductivity, total dissolved solids (TDS), and salinity of the water were measured in each container.

Download:

Table 2. Times of water collections in the field after treatment with pyriproxyfen and start of larval bioassays in laboratory. T-1 = before treatment. T0 = within 24 hours after treatment.

https://doi.org/10.1371/journal.pntd.0013042.t002

Larval bioassays

Larval bioassays were performed on 3^rd instar laboratory-reared Ae. aegypti larvae (Bora-Bora strain F57, susceptible strain) exposed to the field-collected water, according to WHO guidelines [22]. For each bioassay, four plastic test cups (450 ml) filled with 200 mL (5–10 cm depth) of water from the field and four cups of commercial bottled water serving as untreated (negative) controls. Twenty larvae were placed in each cup using a clean pipette. Larval and pupal mortalities were observed daily until emergence of adults. Any physical deformities that occurred in either the successive development (growth, molting) of instars or in emerging adults were observed. Water from each container was tested. The experiment terminated when all control larvae or pupae had either died or successfully emerged as adults.

At the end of the observation period (approximately 10–15 days), pyriproxyfen impact was expressed as percentage of inhibition of emergence (IE) based on the number of larvae that failed to develop and emerge successfully into viable adults by counting both adults and empty pupal cases (see data analysis section below for more details). Moribund and dead larvae and pupae, as well as adult mosquitoes not completely separated from the pupal case were recorded as ‘not having survived’. The time duration of the test required larvae be provided with a small amount of food (ground fish food in suspension) at a concentration of 10 mg/l at two-day intervals until final counts were made. The powdered food was suspended in water to make a slurry and one or two drops added per cup. Each cup was covered with nylon netting to prevent escape of emerging adults. The test containers were held at 25–28°C with a 12 h photoperiod light:dark cycle.

Batch effectiveness

The applied pyriproxyfen came in two different packaging formats: granules in teabag sachets and granules in bulk packaged 1-kg bags. The teabag sachets were used in the RCT and bulk granules were used in residual effectiveness in the field (during T-1 to T0). We evaluated the IE of both packaging types using the same bioassay procedures as described above and using the reverse osmosis water in this experiment. The tea bag sachets were manufactured in July 2016 with an expiry date of July 2019 (batch 6710F4). Two batches of granules both manufactured in March 2018 had an expiry date of March 2020 (Table 3). Bioassays were repeated three times (100 larvae per time) for each batch in a similar manner as described above. Bioassays were performed at the Department of Entomology, Faculty of Tropical Medicine, Mahidol University.

Download:

Table 3. Compounds and formulations analyzed.

https://doi.org/10.1371/journal.pntd.0013042.t003

Active ingredients

In this experiment the concentration of active ingredient in two formulations (three batches) of pyriproxyfen with different manufacture and expiry dates were assessed (Table 3). Analyses were performed by two mutually independent laboratories. Tests were carried out in July 2020 by Sumitomo Chemical Environ-Agro Asia Pacific SDN BHD, Singapore and in November 2020 by Central Laboratory (Thailand) Co. Ltd., Chachoengsao branch, Thailand. Two replicates of each batch were processed. Both companies used the CIPAC method 715/GR/M with minor in-house modifications to analyze the pyriproxyfen compounds [23].

Water source and quality

The manufacturer recommends using 0.01 mg/L active ingredient (a.i.) pyriproxyfen (applied as a 0.5% granule formulation). However, water qualities like turbidity and pollution are criteria for dose adjustment [24].

Water was collected from ‘dragon’ jars (Fig 1, left panel) containing water from four different sources: 1) Government tap water 2) Local tap water, 3) Rainwater, and 4) Groundwater. The jars were selected randomly from households in the study site. Government tap water provision and quality is overseen by the Provincial Waterworks Authority (PWA). Water is pumped from a dam nearby, followed by adding alum and quicklime. Sedimentation tanks allow particles to settle, while filters remove impurities. Chlorine is then added to transform the water into a clean, safe resource for communities [25]. Local tap water is provided by the subdistrict administrative organizations and quality treatment is typically simplified. Rainwater is generally harvested locally from roofs and other water contraptions and stored in dragon jars (Fig 1). Groundwater in the study area is extracted using pumping systems. Wells are equipped with pumps, which vary in type depending on the well’s depth and the required water volume. For daily use, people typically pump groundwater and store it in jars.

Water was collected using 1L NASCO Whirlpak plastic bags and water quality recorded including temperature, conductivity, pH, TDS and salinity. Water was brought back to the Department of Medical Entomology, Faculty of Tropical Medicine, Mahidol University and treated with pyriproxyfen using doses recommended by WHO for Aedes immature mosquito control as described above. Bioassays were repeated three times using 100 larvae per time in a similar manner as described in Residual effectiveness in the field. New sets of larvae were changed every two weeks and repeated until 13 weeks.

Data analysis

Inhibition of adult emergence (IE).

The percent inhibition emergence (IE%) was calculated for residual effectiveness in the field, batch effectiveness and water quality using the following formula [27]:

where T = percentage survival or emergence in treated batches and C = percentage survival or emergence in the control. Adult emergence in the controls was never less than 80%, so no tests were discarded. Similarly, controls generally produced 100% emergence (out of all 160 tests, only 9 gave 99% emergence and 1 test gave 97% emergence) so no correction such as Abbot’s [26] was required.

Difference in emergence inhibition between container types and time points.

In residual effectiveness in the field we used summary measures for the statistical analysis [27], rather than repeated measures analysis of variance (ANOVA) [28]. Most data values were close to zero at later times, inconsistent with the sphericity (constant variance) assumption of the ANOVA [28], meaning that its use would not have been reliable. So, in terms of summary measures, we summarized the IE as the area under its curve (AUC) over time, from day 1 onwards. Lower values of AUC indicate lower IE over time. This AUC was then compared by the non-parametric Mann-Whitney test between container types (jar or tank), and between daily and weekly frequency of use.

Variables such as temperature, which change over time, were compared with deviations from the expected value of IE over time for that container type, to account for the pronounced general decrease in IE over time. This decrease was modelled as a logistic curve [29]. The expected value was obtained from one fitted logistic curve for each container type, with the upper asymptote (value approached as time tends to positive infinity) being set to zero, and the following three parameters being fitted: lower asymptote (value approached as time tends to negative infinity), the steepness of the slope, and the time corresponding to IE halfway between the upper and lower asymptotes [29] (S2 Fig). The time to reach 50% IE was estimated from this curve, and its confidence interval estimated by bootstrapping [30] over containers with 1000 replicates. Then, for temperature and the measures of water quality, and for each container, the Spearman correlation [31] was calculated with the deviations from the expected value from the logistic curve. Finally, the global association was estimated by combining the observed container-level correlations and standard errors [32] by meta-analysis methods [33], i.e., the container-level results were combined in the same way as study-level results in a meta-analysis.