EDP Sciences logo
Submit your paper
Search
Menu
All issues Volume 33 (2026) Parasite, 33 (2026) 33 Full HTML
Open Access
Issue
Parasite
Volume 33, 2026
Article Number 33
Number of page(s) 11
DOI https://doi.org/10.1051/parasite/2026029
Published online 03 June 2026

© A. Sibari et al., published by EDP Sciences, 2026

Licence Creative CommonsThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Introduction

Leishmaniosis is a deadly vector-borne disease spreading on different continents, including Europe. Both humans and animals can be affected by the disease that can be fatal in immunocompromised patients and in untreated susceptible dogs. The pathogen responsible for leishmaniosis is a protozoan called Leishmania that is mainly transmitted through the bites of female phlebotomine sand flies (Diptera: Psychodidae) [2, 26]. To complete their life cycle, phlebotomine females require blood meals. They are hematophagous for the development and maturation of their eggs, feeding multiple times during the vectorial season, which impacts pathogen maturation and transmission [30]. One of the first lines of canine leishmaniosis (CanL) prevention is protection against vector bites. Repellent formulations with anti-feeding efficacy have been developed specifically for dog protection. To assess the effectiveness of these products, experimental dogs are traditionally exposed to numerous and repeated vector bites in controlled conditions. Alternative artificial models have been developed to reproduce the natural conditions of blood feeding for parasites [7] such as ticks [10, 31], fleas [5], mosquitoes [12, 19] and sand flies [13, 16, 21, 35]. The first membrane feeding model for sand flies was described in 1927 with rabbit skin [1]. Over the years, different models have been developed to characterize sand fly feeding behavior and to stimulate factors such as blood source [14] or temperature [9]. Artificial feeders were also historically tested to study infection by and transmission of Leishmania spp. [13, 22, 35, 37]. However, the only direct comparison between ex vivo and in vivo models was performed on hamsters [27] and no alternative to the dog in vivo infestation model has yet been developed to assess the efficacy of an ectoparasiticide against Ph. perniciosus, more in alignment with the 3Rs principle (reduce, refine, replace). The aim of this study was therefore to compare an in vivo and an ex vivo model of sand fly infestation in dogs through the efficacy assessment of a repellent product.

Materials and methods

Ethics statement

The study was conducted after approval by the ethics committee at the National Veterinary School of Toulouse, France (APAFIS#21563-2019071916306501 v6). The study complied with Good Clinical Practices (VICH GL9, June 2000). All technicians were blinded to the groups, except for the personnel involved in treatment and dog follow-up post treatment.

Sand fly Phlebotomus perniciosus colony

The sand fly Phlebotomus perniciosus strain used in this study originated from Lisbon, Portugal, and had been maintained under laboratory conditions at École Nationale Vétérinaire de Toulouse (ENVT), France, for 20 years without being exposed to any insecticide. The sand fly females are fed on rabbit blood, and the larvae are maintained on a specific medium. The cycle from egg to adult lasts for 5–7 weeks, at optimal conditions of ambient temperature (25–30 °C) and relative humidity (70–80%). Sand fly exposures described below were performed using unfed females aged between 8 to 10 days, according to our internal procedures.

Experimental dogs

Fourteen healthy Beagle dogs, males (5) and females (9), originating from the kennel of the Parasitology Department of the ENVT were enrolled in the study. They had not been treated with any ectoparasiticide and/or repellent within 3 months before the start of the animal phase. Prior to allocation, all dogs underwent a physical examination performed by the study veterinarian to ensure their health status. Dogs were housed by pair of 2 in indoor enclosures in compliance with European animal welfare regulations, under controlled environmental conditions with a 12-hour light/12-hour dark cycle. Each animal was identified by a subcutaneously implanted microchip. They received a commercially available dry diet in appropriate daily rations and had ad libitum access to water. All animals were handled and maintained with due consideration for their welfare and were acclimated to the facility for two weeks prior to the initiation of treatment. Throughout the study period, the dogs were monitored daily for general health status and remained in good health during the entire animal phase.

Dog study design

The study was a randomized, controlled, unicenter trial with a parallel design. The study followed a randomized block design based on the number of engorged (dead + alive) Ph. perniciosus sand flies collected from each dog after the pre-treatment infestation. The pre-infestation phase was conducted with 14 dogs, from which the two dogs with the lowest number of engorged females were removed, leaving 12 dogs for the trial. These twelve dogs (12.9 ± 1.3 kg body weight) were then allocated into 6 replicates of 2 animals each. The two dogs with the highest number of engorged Ph. perniciosus (dead + alive) females formed replicate 1, the next 2 formed replicate 2, and so on. Within each replicate, dogs were randomly assigned to one of two groups: a Dinotefuran-Permethrin-Pyriproxyfen [DPP, VECTRA® 3D] treated group (n = 6) or an untreated control group (n = 6) (Figure 1). To facilitate the workload for both in vivo and ex vivo infestations over time, each group was further divided into two sub-groups A and B: the first 3 dogs were allocated to sub-group A, and the remaining 3 to sub-group B.

Thumbnail: Figure 1 Refer to the following caption and surrounding text. Figure 1

Study design diagram. 12 dogs were allocated to either an untreated control group or a treated group receiving Vectra 3D®. Each group was further divided into two sub-groups (A and B) to allow staggered experimental procedures. Treatment was performed on Days 0 and 2 for sub-groups A and B, respectively. Dogs were combed for hair, which was then used for assessments performed on Days 1, 7, 14, 21, and 28 post-treatment for sub-group A and on Days 3, 9, 16, 23 and 30 post-treatment for sub-group B.

In vivo model

Infestations of dogs were performed as previously described [17]. The first infestation was conducted prior to treatment at day-7 and was used for allocation purposes. Five post-treatment infestations were conducted at 1, 7, 14, 21, and 28 days post-treatment. The day before each infestation, 50 (±5) Ph. perniciosus females and approximately 5 males were placed in each sand fly-proof net, with cotton soaked in sugar-water at their disposal. Sand flies were fasted for 2 h before exposure to the dogs. Prior to infestation, animals were weighed and sedated by intramuscular injection of a mixture of dexmedetomidine (at the recommended dose of 25 μg/kg) (Dexdomitor®, Elanco Santé Animale, Lilly, Suresnes, France) and ketamine (at the recommended dose of 5 mg/kg) (Clorketam®, Laboratoire Vetoquinol S.A., Lure, France) and diazepam (at the recommended dose of 0.45 mg/kg) (Valium®) to facilitate muscular relaxation, following the dose recommended by the manufacturers, and then placed in each individual infestation net. After 60 (±5) minutes of exposure, dogs were carefully taken out of the nets and examined for any dead sand flies, then placed back in their individual boxes and supervised until they regained full consciousness. At the end of the exposure period, all live sand flies were aspirated from the nets using a vacuum pump and recorded as live-engorged or live-unengorged. All dead sand flies found on the dog’s body or in the net were collected, counted, and recorded as dead-unengorged or dead-engorged. Engorgement status was determined by observing abdominal distension and the presence of blood with the naked eye. To facilitate counting, live sand flies recovered from individual animals after exposure were placed in separate net cages (one net per animal). Each cage was labelled with the animal number. Once the counting of alive sand flies was completed, all remaining sand flies were discarded.

Ex vivo model

Hair collection

The hair specimens were collected from each individual dog by combing the whole body until approximately 1 g of hair was obtained per dog. The hair collected from each dog was then placed into aluminium foil and stored in an individual plastic bag identified with the dog’s number, the date, and, when relevant, the treatment group number. Hair combing was performed once on Day-7 and then after treatment and before anesthesia, on Days 1, 7, 14, 21, and 28 for dogs from Groups 1A and 2A, and on Days 3, 9, 16, 23, and 30 for dogs from Groups 1B and 2B.

Infestation

The day before each infestation, 100 (±5) Ph. perniciosus females and approximately 5 males, between 8 and 10 days old, were placed in each small sand fly-proof net, with cotton soaked in sugar-water at their disposal. Each net was identified with the corresponding individual number of each artificial feeder. Sand flies were fasted for approximately two hours before exposure to the feeders.

The individual glass feeders were closed at the bottom with a thin Parafilm membrane (Parafilm® M; Pechiney Plastic Packaging, Chicago, IL, USA) and identified with the corresponding individual number of each dog (one glass feeder represented one dog). They were filled with cattle blood derived from residual clinical samples collected at the ENVT Large Animal Hospital, treated with lithium heparin (approximately 5 mL to cover the Parafilm membrane). To stimulate the sand flies to feed on the device, a constant temperature of 38.5 °C (±0.5 °C), mimicking the host’s body temperature, was maintained by water circulating through the glass feeder. Once the blood was added and the water system started, approximately 0.09 g of hair (corresponding to 0.007 g of hair/cm2 following the ex vivo model from Tahir et al. [31]) collected from each dog was placed on top of each individual net, and the corresponding artificial feeder (4 cm in diameter, surface of approximately 12.6 cm2) was placed on top of it, allowing the sand flies to feed directly onto the blood through the hair and the Parafilm membrane + surface feeder (Figure 2). The system was left for 60 min (±5 min). At the end of the exposure period, all sand flies were counted and recorded as live/dead and engorged/unengorged.

Thumbnail: Figure 2 Refer to the following caption and surrounding text. Figure 2

Diagram representation of the ex vivo feeding system. The system is composed of a glass feeder containing cattle blood and covered at its base with a synthetic membrane. Dog hair samples were placed between the feeder and a net enclosing approximately 100 P. perniciosus sand flies, allowing assessment of feeding under ex vivo conditions.

The sand flies were exposed to the artificial devices on Day-7, and then on Days 1, 7, 14, 21, and 28 for dogs from Groups A and on Days 3, 9, 16, 23, and 30 for dogs from Groups B.

Product efficacy assessment

The product efficacy was assessed based on various criteria such as the feeding rate and survival rate of sand flies, calculated for each dog at each time point. The primary efficacy criterion was the anti-feeding (also called repellent) effect. The 1-hour insecticidal efficacy against the total number of sand flies and the 1-hour insecticidal efficacy against fed sand flies only were determined as secondary efficacy criteria.

Feeding rate (%):

100×Total number offedsand fliesTheoretical number of sand flies released.Mathematical equation: $$ 100\times \frac{\mathrm{Total}\ \mathrm{number}\ \mathrm{of}\ \mathrm{fed}\ \mathrm{sand}\ \mathrm{flies}}{\mathrm{Theoretical}\ \mathrm{number}\ \mathrm{of}\ \mathrm{sand}\ \mathrm{flies}\ \mathrm{released}\ }. $$

Anti-feeding efficacy (%):

100×MC−MTMC,Mathematical equation: $$ 100\times \frac{\mathrm{MC}-\mathrm{MT}}{\mathrm{MC}}, $$

where MC and MT are the arithmetic mean of engorged sand flies in the control and treated groups, respectively.

Survival rate (%):

100×Total number of live sand fliesTheoretical number of sand flies released.Mathematical equation: $$ 100\times \frac{\mathrm{Total}\ \mathrm{number}\ \mathrm{of}\ \mathrm{live}\ \mathrm{sand}\ \mathrm{flies}}{\kern0.5em \mathrm{Theoretical}\ \mathrm{number}\ \mathrm{of}\ \mathrm{sand}\ \mathrm{flies}\ \mathrm{released}\ }. $$

Insecticidal efficacy (%):

100×MC−MTMC,Mathematical equation: $$ 100\times \frac{\mathrm{MC}-\mathrm{MT}}{\mathrm{MC}}, $$

where MC and MT are the arithmetic mean of live sand flies in the control and treated groups, respectively.

Data were entered using Excel and statistical analysis was performed using R software R version 4.0.5 (https://www.R-project.org/.) The parameters were compared at each time point between control and treated groups, as well as between the in vivo and ex vivo models for each dog. A Wilcoxon test was applied. A p-value of <0.05 was considered statistically significant. Multiple comparisons were performed using the Holm adjustments to manage multiplicity testing.

Results

The product was well-tolerated by the dogs and no adverse reactions were observed. Our study was aimed to determine whether the effect of the treatment was comparable when using the reference in vivo dog model vs. an alternative ex vivo model.

Feeding rate and survival rate of sand flies in vivo

Feeding rate of sand flies on dogs

The average feeding rate of sand flies on dogs was 82.5 ± 6.9% in the control group. In the treated group, the feeding rate on dogs ranged between 2% (day 7) to 15% (day 28). On day 28, an outlier female dog in the treated group was identified with a feeding rate of 46%. At each time point post-treatment, the difference in the feeding rate status of Ph. perniciosus females was higher (p < 0.01) in the control than in the treated group (Table 1).

Table 1

Comparative analysis of feeding and survival rates of sand flies (± standard deviation) in vivo over time.

Survival rate of sand flies after exposure to dogs

The average survival rate of sand flies on dogs was 92.6 ± 4.0 % in the control group. In the treated group, the survival rate on dogs ranged between 17% (day 1) to 55% (day 28). On day 28, the outlier female dog was identified in the treated group with a survival rate of 70%. At each time point post-treatment, the survival rate of Ph. perniciosus females was higher (p < 0.01) in the control than in the treated group (Table 1).

Feeding rate and survival rate ex vivo

Feeding rate of sand flies on membranes

The average feeding rate of sand flies on membranes was above 40.1 ± 15.2 % in the control group (Table 2). In the treated group, the feeding rate ranged between 1% (day 1) to 4% (day 21). At each time point post-treatment, the difference in the feeding rate status of Ph. perniciosus females was higher (p < 0.01) in the control than in the treated group (Table 2).

Table 2

Comparative analysis of feeding and survival rates of sand flies (± standard deviation) ex vivo over time.

Survival rate of sand flies after exposure to membranes

The average survival rate of sand flies on dogs was above 96.7 ± 3.2 % in the control group (Table 2). In the treated group, the survival rate ranged between 51% (day 28) to 77% (day 21). At each time point post-treatment, the survival rate of Ph. perniciosus females was higher (p < 0.01) in the control than in the treated group (Table 2).

Efficacy assessment

In the treated group, the application of DPP demonstrated an anti-feeding effect in dogs lasting for one month, with efficacy exceeding 83% (Table 3). Similar results were observed with the artificial feeder, achieving anti-feeding efficacy greater than 91% over time. For both models, the onset of efficacy was observed within 24 h of product administration (D1) and maximum efficacy was reached at Day 7 with an anti-feeding effect above 97%. The insecticidal effect was also evaluated in both models (Table 3). In the in vivo model, efficacy ranged from 80.7% on Day 1 to 40.3% on Day 28, whereas in the ex vivo model it ranged from 45.7% (Day 1) to 20.5% (Day 28) (Table 3). A comparison of anti-feeding and insecticidal efficacies between the two models is presented in Figure 3 and Supplementary Figure 1. In addition, the insecticidal efficacy of the product against fed sand flies was evaluated, with results presented in Figures 4 and 5.

Table 3

Evaluation of anti-feeding and insecticidal efficacy (%) of Vectra 3D® against sand flies from D1 to D28.

Thumbnail: Figure 3 Refer to the following caption and surrounding text. Figure 3

Comparison of anti-feeding and insecticidal efficacy (arithmetic mean) of Vectra 3D® between in vivo & ex vivo models. Anti-feeding efficacy (left panel) and insecticidal efficacy (right panel) are presented as arithmetic means at each time point (Days 1, 7, 14, 21, and 28 post-treatment) for both ex vivo and in vivo models. Efficacy was calculated relative to the untreated control group.

Thumbnail: Figure 4 Refer to the following caption and surrounding text. Figure 4

Insecticidal efficacy against fed sand flies with the in vivo model. The total number of alive and fed sand flies in the in vivo model is shown for the untreated control and treated groups at each time point (Day -7, and Days 1, 7, 14, 21, and 28 post-treatment). Anti-feeding efficacy (%) is indicated above the bars and represented by black triangles.

Thumbnail: Figure 5 Refer to the following caption and surrounding text. Figure 5

Insecticidal efficacy against fed sand flies with the ex vivo model. The total number of alive and fed sand flies in the ex vivo model is shown for the untreated control and treated groups at each time point (Day -7, and Days 1, 7, 14, 21, and 28 post-treatment). Anti-feeding efficacy (%) is indicated above the bars and represented by black triangles.

Discussion

The objective of this study was to develop an alternative model to dog infestations by sand flies used for the efficacy assessment of parasiticide repellent products. To the best of the authors’ knowledge, this work documents for the first time a direct comparison between in vivo and ex vivo models of sand fly infestation in dogs for the assessment of ectoparasiticidal product efficacy.

Methodological considerations

The methodology is a key success factor in the development of this alternative ex vivo model. In this study, certain methodological aspects affected model performance and some could be improved. We will discuss the impact of methodological choices on both feeding and survival rates, such as the number of sand flies, the type of membrane, the exposure time, and the feeding assessment of the sand flies.

Feeding rate

In the present study, we observed that the mean feeding rate in the untreated control group using the in vivo model (82.5% ±6.9) was 49% higher than with the ex vivo model (40.1% ±15.2), suggesting that the current methodology could be further optimized. Importantly, the feeding rate is generally considered satisfactory for inclusion of dogs prior to treatments when it is above 50% [17]. This improvement should be achievable as a previous comparison between in vivo (hamster, >60%) and ex vivo (membrane, >70%) did not detect significative differences in feeding rates of sand flies between the two models [27].

Number of sand flies

The number of sand flies per challenge in this study (50 in vivo, 100 ex vivo) may have influenced the feeding rate. Of note, a study has demonstrated a positive correlation between the number of phlebotomine sand flies and the feeding rate [29]. Evidence suggests that cooperative feeding improves feeding efficiency, as single-feeding flies take 70.5% longer to obtain a blood meal compared to those feeding in groups [32]. This cooperative behavior significantly influences sand fly behavior during blood meals and affects Leishmania transmission and development by enhancing the effectiveness of saliva components [32]. However, this parameter might be dependent on other factors. Similarly, in a previous study [17], the use of twice the number of female sand flies released with dogs (100 vs. 50) did not lead to a higher absolute value feeding rate (74%) than in our study with the in vivo model (82%). The number of sand flies should also be considered with the surface or volume available in the model. While this would not be limiting with the in vivo model using the full body of dogs, it might represent a limiting factor with the ex vivo model. The optimum ratio of sand flies to surface area or volume remains a critical parameter for understanding sand fly feeding behavior. In previous studies, different feeding surface areas of 3.14 cm2 [37] and 7.1 cm2 [29] were used, but the number of sand flies was not well documented. In the present study, the surface area available for sand flies to feed on in the ex vivo model was approximately 12.6 cm2.

Type of membrane and blood source

A parafilm membrane was chosen in this study as the non-animal membrane and its application for artificial blood feeders is well-documented in the literature [8, 11, 15]; it has been used extensively in other in vitro systems developed by the ENVT collaborators with other insect models [6, 18]. In studies comparing different membrane compositions, animal membranes such as chicken skin membranes were the most successful for sand fly feeding in species such as Lutzomyia longipalpis [37], Ph. papatasi [9, 27], and Ph. perniciosus [15]. Hošková et al. compared different membrane types in a recent study, and the results showed that only 5% of Ph. perniciosus females fed through a synthetic membrane, compared to 55.7% with a chicken skin membrane. However, feeding rates increased to 25.5% with the synthetic membrane when coagulated blood plasma was applied to its outer surface [15].

In a previous study [27], the authors compared chicken-skin membrane with an in vivo model (live hamster) and the results showed no difference between both models. However, the use of animal skins poses ethical question as they are euthanized for the methodology [29]. In the same study, the relationship between membrane thickness and feeding rate was investigated, but the results demonstrated no correlation [29].

Cattle blood was provided as a blood source for the ex vivo model. Previous research has shown that the origin of the blood, whether from animals, humans, or a mixture does not influence the feeding rate of other species such as Ph. papatasi [9, 14]. Furthermore, the ENVT sand fly colony is occasionally fed with cattle blood, in addition to rabbit blood, for colony maintenance. Additionally, dog hair was applied to the membrane system to act as a mechanical and chemical stimuli, imitating the presence of the host animal to trigger their feeding behavior.

Exposure time

Exposure time is also a parameter to consider regarding the difference in feeding rate between the two models. In this study, the sand flies were left to feed for approximately 1 h on the ex vivo model, but this length of time was questioned in a previous study comparing two models with live hamster and Ph. papatasi species [27]. In the study, the researchers extended the initial exposure time of sand flies with the ex vivo model by up to 3 h of additional time to optimize engorgement and observed no significant difference between the two methods. During the present study, the feeding behavior of female sand flies was also monitored. It was observed that females typically initiated feeding within the first 30 min of exposure to both the live animal hosts and the membrane feeding system (authors’ personal observations). Consequently, extending the duration of exposure to the feeding system would likely not have resulted in a significant increase in the overall feeding rate within the control group.

Feeding assessment

In this study, feeding was determined by visual assessment of the abdominal distension of sand flies and its change of color. This method is easy, rapid, and cost-effective to perform. However, the visual assessment is subjective and may not detect a small bloodmeal. The mean blood volume taken by Ph. perniciosus females from live rabbits has been estimated to be 0.51 μL in a previous study [36], but this is generally detected without microscopic examination, with the naked eye. It remains unknown whether partial blood meals occur and whether they can be reliably detected through visual assessment alone. Molecular techniques, such as real-time PCR and ELISA, are available and allow for accurate identification of phlebotomine sand fly blood meals [20, 28]. The detection limit of host DNA using the real-time PCR method for a 2 μL sand fly blood meal varied depending on the host species, with a threshold of 1 pg for canine blood [28]. Future studies should consider combining visual inspection with molecular approaches to more precisely evaluate feeding rates.

Survival rate

The viability of sand flies is a critical element of their feeding activity, as a higher survival rate directly correlates with an increased feeding rate, allowing for more frequent and successful blood meals. In the present study, the survival rates in the control group with the in vivo model remained above 90%, which is consistent with other product trials [4, 24, 25], while it was >94% with the ex vivo model.

Product efficacy

The main goal of repellent parasiticide products is to protect dogs from the bites of sand flies. This is demonstrated through the anti-feeding efficacy. To a lesser extent, these products usually also kill sand flies through their insecticidal efficacy. For both parameters, it is important to document the onset of action, which is the time interval between treatment and first documentation of efficacy, and the duration of this efficacy over time, which will indicate the treatment intervals.

Anti-feeding efficacy

Regardless of the model, the overall anti-feeding effect of DPP for one month measured in this study against Ph. perniciosus sand flies was above 83% and was consistent with previous trials using in vivo models showing efficacy >83% [17] and >92% [33, 34]. The onset of anti-feeding efficacy was documented 24 h after administration at 94.9% in dogs and 93.5% ex vivo. In previous studies, within 24 h of administration, we observed with DPP similar anti-feeding efficacy above 80.6% [3] and 96.9% [17].

The duration of the repellent effect of DPP over time was 4 weeks with a maximum reached at Day 7 with both in vivo an ex vivo models: 97.7% and 97.8%, respectively. The study confirmed that DPP protects dogs from sand fly bites for at least 4 weeks.

Insecticidal efficacy

Because of its strong repellent effect, the product prevents most sand flies from landing and from remaining on dogs. Therefore, their exposure to the treated animal is not long enough to trigger a repeatable and strong insecticidal effect. In fact, it was demonstrated in a previous study that contact is required between treated dogs and vectors to induce insecticidal effect [23]. The insecticidal efficacy remained above >40.3% with the in vivo model, while this value was above >20.5% with the ex vivo model. The difference could be explained by a possible higher attraction from the whole dog body compared to the membrane with hair.

In this study, the onset of insecticidal action of DPP was assessed 24 h after administration at 80.7% in dogs and 45.7% ex vivo. In previous studies, insecticidal efficacy greater than 97.6% was observed within 24 h of treatment with DPP in dogs [17], indicating that direct contact between treated animals and the vector is necessary for the effect to occur [23]. The insecticidal effect of DPP differed significantly between the two models, with values obtained from the in vivo model being at least 1.6 times higher than those observed in the ex vivo model.

This result showed that improvements in the methodology are achievable to obtain equivalence between the two models. The insecticidal efficacy against fed sand flies was also assessed. A fed sand fly can be infected and then transmit pathogens to another host. In the present study, the results showed that the insecticidal efficacy against fed sand flies remained above 85.5% and 93% with the in vivo and ex vivo models, respectively, over 4 weeks. These results demonstrate that the DPP product could play a significant role in global prevention against the transmission of canine leishmaniosis by efficiently controlling vector bites.

Conclusion

This study highlights a promising ex vivo alternative to the traditional in vivo dog infestation model for evaluating the efficacy of ectoparasiticide repellents against phlebotomine sand flies. The development of this artificial model aligns with the 3Rs principles, aiming to reduce and refine the use of animals in research. While further optimization of the methodology is needed, the model has already yielded consistent efficacy results when compared to the in vivo reference method and to previously reported product performances [17, 33]. Additionally, regardless of the model used, the anti-feeding efficacy of DPP against sand flies reached 93.5% within 24 h post-administration and remained above 83.1% for up to one month. This ex vivo model should be considered by regulatory authorities as a viable tool for future efficacy assessments.

Acknowledgments

The authors wish to thank Martine Roques and Sonia Gounaud, zootechnicians at ENVT, for their professionalism in handling experimental dogs and their fantastic work with the maintenance of the Ph. perniciosus colony over the past 25 years.

Conflicts of interest

EB and EL declare collaborations with Ceva Santé Animale in the past 5 years (clinical studies conducted in collaboration, expert input). AS, TB, and MV are currently employees of Ceva Santé Animale.

The study was funded by Ceva Santé Animale.

Supplementary material

Thumbnail: Supplementary Figure 1: Refer to the following caption and surrounding text. Supplementary Figure 1:

Anti-feeding efficacy (left panel) and insecticidal efficacy (right panel) are presented as geometrical means at each time point (Days 1, 7, 14, 21, and 28 post-treatment) for both the ex vivo and in vivo models. Efficacy was calculated relative to the untreated control group.

References

  1. Adler S, Theodor O. 1927. The behaviour of cultures of Leishmania tropica, L. infantum, and L. braziliense in the sandfly, Phlebotomus papatasii. Nature, 119, 48–49. [Google Scholar]
  2. Akhoundi M, Kuhls K, Cannet A, Votýpka J, Marty P, Delaunay P, Sereno D. 2016. A historical overview of the classification, evolution, and dispersion of Leishmania parasites and sandflies. PLOS Neglected Tropical Diseases, 10, e0004349. [Google Scholar]
  3. Bongiorno G, Bosco A, Bianchi R, Rinaldi L, Foglia Manzillo V, Gizzarelli M, Maurelli MP, Giaquinto D, El Houda Ben Fayala N, Varloud M, Crippa A, Oliva G, Gradoni L, Cringoli G. 2022. Laboratory evidence that dinotefuran, pyriproxyfen and permethrin combination abrogates Leishmania infantum transmissibility by sick dogs. Medical and Veterinary Entomology, 36, 81–87. [Google Scholar]
  4. Bouhsira E, Deuster K, Lienard E, Le Sueur C, Franc M. 2018. Evaluation of the anti-feeding and insecticidal effects of a topically administered combination of imidacloprid and permethrin (Advantix®) against Phlebotomus (Larroussius) perniciosus (Newstead, 1911) in dogs following monthly administration. Parasites & Vectors, 11, 120. [Google Scholar]
  5. Bouhsira E, Ferrandez Y, Liu M, Franc M, Boulouis H-J, Biville F. 2013. Ctenocephalides felis an in vitro potential vector for five Bartonella species. Comparative Immunology, Microbiology and Infectious Diseases, 36, 105–111. [Google Scholar]
  6. Bouhsira E, Franc M, Boulouis H-J, Jacquiet P, Raymond-Letron I, Liénard E. 2013. Assessment of persistence of Bartonella henselae in Ctenocephalides felis. Applied and Environmental Microbiology, 79, 7439–7444. [Google Scholar]
  7. Costa-da-Silva A, Carvalho D, Kojin B, Capurro M. 2014. Implementation of the artificial feeders in hematophagous arthropod research cooperates to the vertebrate animal use Replacement, Reduction and Refinement (3Rs) principle. Journal of Clinical Research & Bioethics, 05. [Google Scholar]
  8. Costa-da-Silva AL, Navarrete FR, Salvador FS, Karina-Costa M, Ioshino RS, Azevedo DS, Rocha DR, Romano CM, Capurro ML. 2013. Glytube: A conical tube and Parafilm M-based method as a simplified device to artificially blood-feed the dengue vector mosquito, Aedes aegypti. PLoS ONE, 8, e53816. [Google Scholar]
  9. Fatemi M, Saeidi Z, Noruzian P, Akhavan AA. 2018. Designing and Introducing a new artificial feeding apparatus for sand fly rearing. Journal of Arthropod-Borne Diseases, 12, 426–431. [Google Scholar]
  10. Fourie JJ, Stanneck D, Luus HG, Beugnet F, Wijnveld M, Jongejan F. 2013. Transmission of Ehrlichia canis by Rhipicephalus sanguineus ticks feeding on dogs and on artificial membranes. Veterinary Parasitology, 197, 595–603. [Google Scholar]
  11. Friend WG, Smith JJ. 1987. The study of insect blood-feeding behaviour. 1: Feeding equipment, physical and endogenous factors, dose effect analysis, and diet destination. Memórias do Instituto Oswaldo Cruz, 82 Suppl 3, 11–17. [Google Scholar]
  12. Gerberg E, Kutz F. 1971. A large-scale artificial feeding technique for infecting mosquitoes and its application to screening antimalarial chemicals. Journal of Medical Entomology, 8, 610–612. [Google Scholar]
  13. Ghosh KN. 1994. A modified artificial membrane feeding method for the study of the transmission dynamics of leishmaniasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 88, 488–489. [Google Scholar]
  14. Harre JG, Dorsey KM, Armstrong KL, Burge J, Kinnamon K. 2001. Comparative fecundity and survival rates of Phlebotomus papatasi sandfies membrane fed on blood from eight mammal species. Medical and Veterinary Entomology, 15, 189–196. [Google Scholar]
  15. Hošková A, Vojtková B, Stejskalová M, Polanská N, Jančářová M, Da Costa LM, Sant´Anna MRV, Volf P, Sádlová J. 2025. Evaluation of various membranes for blood-feeding in nine sand fly species and artificial feeding challenges in Sergentomyia minuta. Parasites & Vectors, 18, 119. [Google Scholar]
  16. Lawyer P, Killick-Kendrick M, Rowland T, Rowton E, Volf P. 2017. Laboratory colonization and mass rearing of phlebotomine sand flies (Diptera, Psychodidae). Parasite, 24, 42. [Google Scholar]
  17. Liénard E, Bouhsira E, Jacquiet P, Warin S, Kaltsatos V, Franc M. 2013. Efficacy of dinotefuran, permethrin and pyriproxyfen combination spot-on on dogs against Phlebotomus perniciosus and Ctenocephalides canis. Parasitology Research, 112, 3799–3805. [Google Scholar]
  18. Liénard E, Salem A, Jacquiet P, Grisez C, Prévot F, Blanchard B, Bouhsira E, Franc M. 2013. Development of a protocol testing the ability of Stomoxys calcitrans (Linnaeus, 1758) (Diptera: Muscidae) to transmit Besnoitia besnoiti (Henry, 1913) (Apicomplexa: Sarcocystidae). Parasitology Research, 112, 479–486. [Google Scholar]
  19. Luo Y-P. 2014. A novel multiple membrane blood-feeding system for investigating and maintaining Aedes aegypti and Aedes albopictus mosquitoes. Journal of Vector Ecology, 39, 271–277. [Google Scholar]
  20. Maleki-Ravasan N, Oshaghi M, Javadian E, Rassi Y, Sadraei J, Mohtarami F. 2009. Blood meal identification in field-captured sand flies: Comparison of PCR-RFLP and ELISA assays. Iranian Journal of Arthropod-Borne Diseases, 3, 8–18. [Google Scholar]
  21. Mann RS, Kaufman PE. 2010. Colonization of Lutzomyia shannoni (Diptera: Psychodidae) utilizing an artificial blood feeding technique. Journal of Vector Ecology, 35, 286–294. [Google Scholar]
  22. Maroli M. 1985. The artificial feeding of laboratory reared palearctic sandflies (Diptera : Psychodidae) for studies on the transmission of disease agents. Annales de Parasitologie Humaine et Comparée, 60, 631–634. [Google Scholar]
  23. McCall J, Hodgkins E, Ramiro V, Varloud M. 2016. Contact is required between dogs treated with Vectra® 3D and Aedes aegypti mosquitoes for insecticidal efficacy. ESCCAP Symposium Vector-Borne Diseases, Granada, Spain, October 2016. [Google Scholar]
  24. Molina R, Espinosa-Góngora C, Gálvez R, Montoya A, Descalzo MA, Jiménez MI, Dado D, Miró G. 2012. Efficacy of 65% permethrin applied to dogs as a spot-on against Phlebotomus perniciosus. Veterinary Parasitology, 187, 529–533. [Google Scholar]
  25. Molina R, Miró G, Gálvez R, Nieto J, Descalzo MA. 2006. Evaluation of a spray of permethrin and pyriproxyfen for the protection of dogs against Phlebotomus perniciosus. Veterinary Record, 159, 206–209. [Google Scholar]
  26. Ready PD. 2013. Biology of phlebotomine sand flies as vectors of disease agents. Annual Review of Entomology, 58, 227–250. [Google Scholar]
  27. Rowton ED, Dorsey KM, Armstrong KL. 2008. Comparison of in vitro (chicken-skin membrane) versus in vivo (live hamster) blood-feeding methods for maintenance of colonized Phlebotomus papatasi (Diptera: Psychodidae). Journal of Medical Entomology, 45, 9–13. [Google Scholar]
  28. Sales KGDS, Costa PL, De Morais RCS, Otranto D, Brandão-Filho SP, Cavalcanti MDP, Dantas-Torres F. 2015. Identification of phlebotomine sand fly blood meals by real-time PCR. Parasites & Vectors, 8, 230. [Google Scholar]
  29. Sánchez Uzcátegui YDV, Dos Santos EJM, Matos ER, Silveira FT, Vasconcelos Dos Santos T, Póvoa MM. 2022. Artificial blood-feeding of phlebotomines (Diptera: Psychodidae: Phlebotominae): is it time to repurpose biological membranes in light of ethical concerns? Parasites & Vectors, 15, 399. [Google Scholar]
  30. Serafim TD, Coutinho-Abreu IV, Oliveira F, Meneses C, Kamhawi S, Valenzuela JG. 2018. Sequential blood meals promote Leishmania replication and reverse metacyclogenesis augmenting vector infectivity. Nature Microbiology, 3, 548–555. [CrossRef] [PubMed] [Google Scholar]
  31. Tahir D, Geolier V, Dupuis S, Lekouch N, Ferquel E, Choumet V, Varloud M. 2024. Comparative evaluation of the efficacy of two ectoparasiticides in preventing the acquisition of Borrelia burgdorferi by Ixodes scapularis and Ixodes ricinus: A canine ex vivo model. Microorganisms, 12, 202. [Google Scholar]
  32. Tripet F, Clegg S, Elnaiem D-E, Ward RD. 2009. Cooperative blood-feeding and the function and implications of feeding aggregations in the sand fly, Lutzomyia longipalpis (Diptera: Psychodidae). PLoS Neglected Tropical Diseases, 3, e503. [Google Scholar]
  33. Varloud M, Moran C, Grace S, Chryssafidis A, Kanaki E, Ramiro M, Ferrari G, Pinho A. 2015. Residual repellency after administration of a topical ectoparasiticide Vectra® 3D (dinotefuran–permethrin–pyriproxyfen) to dogs exposed to Phlebotomus perniciosus sandflies weekly for 6 weeks. Poster presented at SEVC-AVEPA, Barcelona, Spain, October 2015. [Google Scholar]
  34. Varloud M, Warin S, Murphy M, Moran C, McGrath S, Ferrari G. 2014. Immediate and residual anti-feeding efficacy of a dinotefuran-permethrin- pyriproxyfen topical administration against sandfly (Phlebotomus perniciosus) infested dogs under laboratory conditions. Poster presented at XXVIII Congresso Nazionale Societa Italiana di Parassitologia, Rome, Italy, June 2014. [Google Scholar]
  35. Volf P, Volfova V. 2011. Establishment and maintenance of sand fly colonies. Journal of Vector Ecology, 36, S1–S9. [Google Scholar]
  36. Volfová V, Jančářová M, Volf P. 2024. Sand fly blood meal volumes and their relation to female body weight under experimental conditions. Parasites & Vectors, 17, 360. [Google Scholar]
  37. Ward RD, Lainson R, Shaw JJ. 1978. Some methods for membrane feeding of laboratory reared, neotropical sandflies (Diptera: Psychodidae). Annals of Tropical Medicine & Parasitology, 72, 269–276. [Google Scholar]

Cite this article as: Sibari A, Weis-Servat G, Liénard E, Salem A, Blondel T, Varloud M & Bouhsira E. 2026. A novel 3R-compliant ex vivo approach to assess ectoparasiticide repellency against Phlebotomus perniciosus in dogs. Parasite 33, 33. https://doi.org/10.1051/parasite/2026029.

All Tables

Table 1

Comparative analysis of feeding and survival rates of sand flies (± standard deviation) in vivo over time.

In the text
Table 2

Comparative analysis of feeding and survival rates of sand flies (± standard deviation) ex vivo over time.

In the text
Table 3

Evaluation of anti-feeding and insecticidal efficacy (%) of Vectra 3D® against sand flies from D1 to D28.

In the text

All Figures

Thumbnail: Figure 1 Refer to the following caption and surrounding text. Figure 1

Study design diagram. 12 dogs were allocated to either an untreated control group or a treated group receiving Vectra 3D®. Each group was further divided into two sub-groups (A and B) to allow staggered experimental procedures. Treatment was performed on Days 0 and 2 for sub-groups A and B, respectively. Dogs were combed for hair, which was then used for assessments performed on Days 1, 7, 14, 21, and 28 post-treatment for sub-group A and on Days 3, 9, 16, 23 and 30 post-treatment for sub-group B.
In the text
Thumbnail: Figure 2 Refer to the following caption and surrounding text. Figure 2

Diagram representation of the ex vivo feeding system. The system is composed of a glass feeder containing cattle blood and covered at its base with a synthetic membrane. Dog hair samples were placed between the feeder and a net enclosing approximately 100 P. perniciosus sand flies, allowing assessment of feeding under ex vivo conditions.
In the text
Thumbnail: Figure 3 Refer to the following caption and surrounding text. Figure 3

Comparison of anti-feeding and insecticidal efficacy (arithmetic mean) of Vectra 3D® between in vivo & ex vivo models. Anti-feeding efficacy (left panel) and insecticidal efficacy (right panel) are presented as arithmetic means at each time point (Days 1, 7, 14, 21, and 28 post-treatment) for both ex vivo and in vivo models. Efficacy was calculated relative to the untreated control group.
In the text
Thumbnail: Figure 4 Refer to the following caption and surrounding text. Figure 4

Insecticidal efficacy against fed sand flies with the in vivo model. The total number of alive and fed sand flies in the in vivo model is shown for the untreated control and treated groups at each time point (Day -7, and Days 1, 7, 14, 21, and 28 post-treatment). Anti-feeding efficacy (%) is indicated above the bars and represented by black triangles.
In the text
Thumbnail: Figure 5 Refer to the following caption and surrounding text. Figure 5

Insecticidal efficacy against fed sand flies with the ex vivo model. The total number of alive and fed sand flies in the ex vivo model is shown for the untreated control and treated groups at each time point (Day -7, and Days 1, 7, 14, 21, and 28 post-treatment). Anti-feeding efficacy (%) is indicated above the bars and represented by black triangles.
In the text
Thumbnail: Supplementary Figure 1: Refer to the following caption and surrounding text. Supplementary Figure 1:

Anti-feeding efficacy (left panel) and insecticidal efficacy (right panel) are presented as geometrical means at each time point (Days 1, 7, 14, 21, and 28 post-treatment) for both the ex vivo and in vivo models. Efficacy was calculated relative to the untreated control group.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.