Application of a coproantigen ELISA as an indicator of efficacy against multiple life stages of Fasciola hepatica infections in sheep

Accepted Manuscript Title: Application of a coproantigen ELISA as an indicator of efficacy against multiple life stages Fasciola hepatica infections in sheep Authors: S.D. George, K. Vanhoff, K. Baker, L. Lake, P.F. Rolfe, W. Seewald, D.L. Emery PII: DOI: Reference:

S0304-4017(17)30368-0 http://dx.doi.org/10.1016/j.vetpar.2017.08.028 VETPAR 8460

To appear in:

Veterinary Parasitology

Received date: Revised date: Accepted date:

28-5-2017 18-8-2017 24-8-2017

Please cite this article as: George, S.D., Vanhoff, K., Baker, K., Lake, L., Rolfe, P.F., Seewald, W., Emery, D.L., Application of a coproantigen ELISA as an indicator of efficacy against multiple life stages Fasciola hepatica infections in sheep.Veterinary Parasitology http://dx.doi.org/10.1016/j.vetpar.2017.08.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Application of a coproantigen ELISA as an indicator of efficacy against multiple life stages Fasciola hepatica infections in sheep S.D. George a,b*, K. Vanhoff b, K. Baker b, L. Lake b, P.F. Rolfe b, W. Seewald c, D.L. Emery a a

Faculty of Veterinary Science, University of Sydney, 425 Werombi Road, Camden, NSW 2570, Australia

b

ELANCO Animal Health, Yarrandoo R&D Centre, 245 Western Road, Kemps Creek, NSW 2178, Australia

c

ELANCO Animal Health, CH-4058 Basel, Switzerland

* Corresponding author. Tel.: +612 9826 3319 E-mail address: [email protected] (S.D. George).

Graphical Abstract

Application of a coproantigen ELISA as an indicator of efficacy against multiple life stages of Fasciola hepatica infections in sheep. Highlights    

Coproantigen ELISA (cELISA) correlated strongly with adult fluke burden (R2=0.78). cELISA and FWEC combined can provide a non-lethal endpoint (R2=0.84). Treatment failure against immature stages could not be detected until 6 weeks post-treatment. Detection of emerging resistance detection requires cELISA ≥ 6 weeks post-treatment.

ABSTRACT At present diagnosis of true resistance and determination of drug efficacy in Fasciola hepatica infection rely solely on terminal experiments. The coproantigen ELISA (cELISA) has been reported previously as a sensitive and specific tool appropriate to detect treatment failure, and potentially drug resistance. Two studies were conducted to determine whether the cELISA was appropriate for 1

on-farm efficacy and resistance testing in Australian Merino sheep. In Study 1 sheep were infected orally with 50 F. hepatica metacercariae on three occasions, twelve, six and two weeks prior to a single flukicide treatment with triclabendazole, closantel or albendazole. Sheep were sampled weekly for a further seven weeks prior to necropsy. Following effective treatment, no faecal antigen was detected from 1 week. When immature stages (≤ 6 weeks) survived treatment, coproantigen reappeared from 6 weeks post-treatment. Therefore, cELISA conducted 1–4 weeks after treatment will demonstrate obvious treatment failure against adult F. hepatica, but is not sufficiently sensitive to detect survival of immature fluke until these reach maturity. In study 2, fluke burdens of sheep necropsied 13 weeks post single infection were compared to fecal worm egg counts (FWEC) and cELISA at necropsy. Regression analysis demonstrated that cELISA correlated strongly with fluke burden, whilst FWEC correlated weakly with cELISA. The correlation between FWEC and fluke burden was also weak, although stronger than that of FWEC with cELISA. The cELISA is an appropriate tool for monitoring effectiveness of treatments against Fasciola hepatica if an adult infection is present, however when immature stages of the parasite are present it is not as reliable. Where immature parasites are present it is recommended that initial cELISA be followed with a secondary cELISA at least 6 weeks after treatment to ensure resistance to immature stages is detected. Further testing is justified for monitoring the effectiveness of control programs by detecting adult populations that have survived a treatment regime. Keywords: Coproantigen ELISA Efficacy Fasciola hepatica Triclabendazole Drug-resistant 1.

Introduction

2

Globally 600–700 million ruminants are at risk of developing fasciolosis, caused by the common liver fluke; Fasciola hepatica (Martinez-Perez et al., 2012; Mas-Coma et al., 2009). Production losses occur when acute, sub-acute and chronic disease causes mortalities, condemnation of livers, reduced weight gains, reduced milk production, degradation of fleece quality and quantity, and inferior reproductive performance (Rojo-Vazquez et al., 2012). Given that the combined costs of control and production losses are estimated at $2–3 billion (Diab et al., 2010; Elliott et al., 2015; McManus and Dalton, 2006) it is imperative the sensitivity, specificity and accuracy of diagnostic methods used to guide management decisions are sensitive, specific and reliable. Currently, the diagnosis of clinical fasciolosis is hindered by the non-specific symptoms of the disease and variable faecal worm egg counts (FWEC), while acute disease in sheep occurring prior to the prepatent period of 10-12 weeks is rarely diagnosed without a post-mortem assessment (Brunsdon, 1967; Rojo-Vazquez et al., 2012). Diagnostic methods have evolved over the last seven decades from traditional faecal sedimentation and flotation techniques (Happich and Boray, 1969; MAFF, 1971; McCaughey and Hatch, 1964; Thienpont et al., 1979) to serological and faecal enzyme linked immunosorbent assays (ELISA) (Dumenigo et al., 2000; Mezo et al., 2004; Mezo et al., 2007; Pfister, 1990; Zimmerman et al., 1982) and molecular diagnostics (Carnevale et al., 2015; Martinez-Perez et al., 2012). Faecal egg counts, while inexpensive and only requiring basic equipment, are less time-efficient per sample for large sample sizes (Gordon et al., 2012b). The use of ELISA techniques allow for a greater number of samples to be run concurrently and are not limited to faecal samples as antibody responses in both milk and blood may be measured. In situations where a high number of samples are to be analyzed at one time (such as in central diagnostic labs), the increase in price per sample is relatively low compared to the labor required. However, serological responses in sheep remain elevated at least 2–4 months after treatment of initial infection, so may bear no relation to current infestations (Gaasenbeek et al., 2001; SanchezAndrade et al., 2001). 3

The development and commercialization of the coproantigen ELISA (cELISA) has provided a more efficient option for the diagnosis of F. hepatica infection prior to patency at 10-12 weeks; detecting the presence of the parasite enzymes in sheep and cattle within 6-8 weeks after a single dose infection (Brockwell et al., 2013; Flanagan et al., 2011a; Martinez-Perez et al., 2012; Mezo et al., 2004; Valero et al., 2009). Several authors have recommended the cELISA as a standard diagnostic tool for sheep and cattle and as a method to assess the efficacy of flukicides posttreatment (Brockwell et al., 2014; Flanagan et al., 2011b; Gordon et al., 2012b). As the cELISA can detect infection prior to patency, utilization of this tool as a measure for resistance would be particularly relevant for drugs, which affect the early stages (<8 weeks postinfection (wpi)) of F. hepatica; i.e. triclabendazole (TCBZ) in sheep and cattle, and closantel (CLOS) in sheep, as the survivors may not have reached patency. Of particular interest is the application for diagnosis of TCBZ resistance which has been increasingly found since the first globally reported case in Australia (Overend and Bowen, 1995). Countries now reporting TCBZ resistance include the United Kingdom, Ireland, the Netherlands, Spain and Peru (Alvarez-Sanchez et al., 2006; Brennan et al., 2007; Brockwell et al., 2014; Fairweather, 2009; Gordon et al., 2012a; Kelley et al., 2016; Martinez-Valladares et al., 2010; Ortiz et al., 2013; Overend and Bowen, 1995; Rojo-Vazquez et al., 2012). The cELISA, using the MM3 monoclonal antibody (Mezo 2004), is both highly specific and sensitive (up to 100%) with no cross-reactivity against Paramphistomum cervi, Taenia hydatigena, Dicrocelium dendriticum, Moniezia spp., Echinococcus spp. and various gastro-intestinal nematodes in sheep and cattle (Brockwell et al., 2013; Charlier et al., 2008; Kajugu et al., 2012; Mezo et al., 2004; Valero et al., 2009). Experimental infection trials have shown that faecal antigen is reduced to non-significant levels by 7-17 days post-treatment in sheep and cattle (Brockwell et al., 2013; Flanagan et al., 2011a; Flanagan et al., 2011b). These results were supported in field investigations (Martinez-Valladares et al. (2010), although some false (positive and negative) results have been noted in sheep (Gordon et al., 2012b; Novobilsky et al., 2012). 4

The literature clearly describes the quantitative phenology of faecal antigen for naturally infected animals and those receiving single experimental infections. However, there is a distinct lack of data for repeated experimental infections during the pre-patent period that reflect more closely, the ingestion of metacercariae and development of fluke, and associated antigen production in areas with endemic fasciolosis. The objective of the current study was to examine the effect of repeated metacercariae infection at 4-6 week intervals in order to understand establishment rates, total fluke numbers and development, together with the kinetics of faecal coproantigen production. This was designed specifically to link the data reported from single age and natural infection, to provide a basis to interpret situations where multiple life stages of F. hepatica are present and resistance is suspected. 2. 2.1

Materials and methods Experimental design A controlled randomized study was conducted to assess the use of a commercially available

cELISA (Bio-X Diagnostics, Jemelle, Belgium) against multi-age fluke infections (Study 1). Twenty-six sheep were infected with 50 metacercariae at 0, 6 and 10 weeks so that by 12 weeks, the flukes had developed to 2, 6 and 12 weeks of age. At 12 weeks, and on the basis of FWECs at 11 weeks, sheep were randomly allocated to 4 treatment groups, each of 6 animals. Sheep in the respective groups were given the flukicides TCBZ (Group 2), CLOS (Group 3) or albendazole (ALB; Group 4), while Group 1 remained as untreated controls (Table 1). On the day before treatment, two sentinel animals were necropsied to confirm successful infection and determine the establishment rates from the ratios of developmental stages present before treatment. Coproantigen was assayed in faecal samples collected pre-infection and weekly from week 3 after initial infection while FWEC was conducted on samples collected at 11, 15 and 19 weeks postinfection (wpi). At week 19, the remaining 24 sheep were euthanized and their livers (with gallbladders intact) were recovered for F. hepatica counts. Efficacy was calculated by comparing F. hepatica burdens of treated animals with those of untreated sheep (Group 1). 5

In Study 2, coproantigen levels, FWECs and fluke burdens were assessed at necropsy from an additional 50 sheep that had been orally dosed on a single occasion with between 45-150 metacercariae and then necropsied 13 wpi. Nine of these 50 sheep, infected at a rate of 150 metacercariae, were faecal sampled at 0, 3, 4, 5, 6, 7, 9, 11 and 13 wpi to enable comparison of coproantigen development in single age and multi-age infections. These data were compared to that of sheep enrolled in Study 1 to compare the development of antigen expression between single and multi-age infections. Pooled statistical analyses were undertaken across necropsy data from all 76 animals (Study 1 and 2) to determine correlations between fluke counts, FWECs and cELISAs as well as effects of infection level, age and type. 2.2

Animals, management and allocation to treatment groups The sheep enrolled in Study 1 (n=26) were 33 month old, mixed-sex, commercially sourced

Merinos. The sheep were confirmed as F. hepatica-naïve prior to experimental infection with metacercariae by serum ELISA (IDEXX fasciolosis verification kit). Animals were allocated to treatment groups on the basis of week 11 FWECs and week 12 body weight using ‘Sheep-sorter randomization program version 4.0.4’, an in-house validated randomization program. They were maintained in an indoor housing facility for the duration of the study and fed a lucerne hay/oaten hay/straw/oats chaff mix. From week 9 to week 12 two animals were penned together due to weight loss and were supplemented with a lucerne-based concentrate pellet. Water was available ad libitum via the town supply system. The sheep were inspected at least daily during the study. Sheep in Study 2, (n=50) were 6-12 month old Merinos, which were either commercially sourced or bred on site. These lambs were confirmed as F. hepatica-naïve (by serum ELISA), before being enrolled in randomized, controlled, experimental infection studies. They were necropsied 13 wpi. 2.3

Fasciola hepatica isolates

6

The triclabendazole susceptible ‘Sunny Corner’ isolate utilized in Study 1 was isolated by the NSW Department of Primary Industries (Australia) in 1989 (Fairweather, 2011a). Prior to acquisition by Yarrandoo R & D Centre in 2011 this isolate had been maintained at the Elizabeth Macarthur Agriculture Institute, Menangle, NSW, Australia. Metacercariae used in this study were released from snail cultures, 1-3 months prior to the sheep infections and, at the time of infection, were suspended in 5 mL of 0.4% carboxymethylcellulose. Approximately 50 viable metacercariae were orally dosed by syringe to each sheep (Study 1). The 50 sheep in Study 2 were infected with 100 (n=20) or 150 (n=9) ‘Sunny Corner’ metacercariae, or 45 (n=21) ‘Oberon’ metacercariae. The triclabendazole resistant ‘Oberon’ isolate was isolated by the NSW Department of Primary Industries in 1999 (Fairweather, 2011a) and acquired by Yarrandoo R & D Centre in 2012 from Veterinary Health Research, Armidale, NSW, Australia. 2.4

Anthelmintics and dosing

The anthelmintics and dose rates utilized in Study 1 are described in Table 1. The sheep in Groups 2–4 were treated at 12 wpi, the dose based on individual body weights determined 2 days previously. All doses were measured by volume, rounded to the nearest 0.2 mL and administered orally from appropriately sized (3, 5 or 20 mL) disposable plastic syringes; a new syringe was used for each animal. 2.5

Faecal trematode egg counts Faecal samples collected at weeks 11, 15 and 19 of Study 1, and prior to necropsy from the

50 sheep in Study 2 were analyzed by a modified sedimentation test (Happich and Boray, 1969) and recorded as eggs per gm-1 faeces (epg). In brief, 3 g of faeces were hydrated with 10 mL of water and homogenized prior to rinsing through a 75 µm sieve into a 250 mL conical flask. The contents were sedimented for 3 minutes prior to siphoning to 25 mL. The flask was then re-filled and sedimented for a further 3 minutes at which point the diluent was reduced to 10 mL. Methylene blue was added to the samples prior to assessment on a grid-scored counting tray and each egg observed represented 1 epg. 7

2.6

Recovery and counting of F. hepatica at necropsy Livers and gall bladders were recovered intact from sheep at necropsy. The gall bladder was

removed and opened out to check for presence of adult F. hepatica. Bile ducts were massaged to expel the adult parasites prior to slicing the entire organ (gold standard approach) into1–2 cm transverse segments, which were individually squeezed to expel parasites. Incubation in warm saline media at 37°C for 2 hours occurred prior to an additional pressing to recover any remaining F. hepatica in the parenchyma, while the saline was examined carefully, under dissecting microscope, for juvenile parasites. The recovered F. hepatica were counted and measured. Measurements of ≤ 3 mm, 4-10 mm and ≥ 12 mm were classified as 2, 6 and 12 weeks of age F. hepatica respectively. The same method was utilized to recover F. hepatica from Study 2 but measurements were not undertaken. 2.7

Coproantigen ELISA Faecal samples were collected directly from the rectum of each sheep enrolled in the study

prior to infection and weekly from week 3 to week 19. A sub-sample of 0.5 g faeces was stored at 20ºC until analysis. Individual samples were analyzed using a commercially available cELISA (BIOK 201; BIO-X Diagnostics, Belgium) as per the manufacturer’s direction with overnight antigen extraction as described by Brockwell et al. (2013). Optical densities (ODs) were corrected by averaging the negative values of individual plates and subtracting this value from individual samples. Analysis of data was completed on the corrected OD value. Approximately 30% of samples analyzed were replicated at a later date. Results of replicated samples were averaged prior to statistical analysis. 2.8

Calculation of efficacy and statistics (Study 1) Efficacy was calculated using both arithmetic and geometric mean (utilizing a constant of 1)

parasite reductions in treated groups when compared to the untreated control group. The geometric mean constant reflected the sensitivity of counting procedures of F. hepatica and FWEC. 8

Percentage efficacy was determined from the F. hepatica counts using the formula: Efficacy [%] = 100  (C–T)/C, where C and T are the F. hepatica count means (or FWEC means) for the control and treated groups, respectively (Abbott, 1925). Adequacy of infection was checked statistically on raw F. hepatica counts of control sheep at week 19 necropsy (Vercruysse et al., 2001). Fasciola hepatica counts and FWEC were transformed prior to ANOVA or Mann-Whitney U test for comparisons between groups. Analysis of ELISA results used untransformed OD data. One-way analysis of variance (ANOVA) was utilized for week 12 and 19 ELISAs, week 11 FWECs and F. hepatica burdens as this was deemed more valid than a non-parametric analysis for these parameters. All remaining parameters were statistically compared by Mann-Whitney U tests. The level of significance was p=0.05. Pearson correlation coefficients were calculated for various pairs of variables, both per group and for all groups pooled. For selected pairs of variables, scatterplots were produced. For each group, as well as for all groups pooled, a ‘correlation ellipse’ was drawn at one standard deviation (containing approximately 40% of the bivariate normal density). SAS Version 9.2.2 (SAS 2008) was used for all Study 1 calculations. 2.9

Pooled statistical evaluation (Study1 and 2) Counts of F. hepatica and FWEC were transformed before Mann-Whitney U tests were

applied to examine differences in effects of type of infection (single vs. mixed age) and level of infection. For examining level of infection, results were grouped by the number of flukes recovered in the following ranges; 0 F. hepatica, 1-20 F. hepatica, 21-50 F. hepatica, 51-100 F. hepatica and >100 F. hepatica. Linear regression was applied to the pooled data, separated into infection type and also excluding ranges of infection. The linear regression model for pooled data was expanded further to include additional terms of isolate, infection rate, infection type, source of sheep, animal age and age of metacercariae, alongside the existing FWEC and ELISA variables. The least significant fitted terms were dropped from the model until only significant effects remained. GenStat Version 16.1 was used for these analyses. 9

Additionally KAPPA analysis (http://vassarstats.net.au/kappa.html ) and calculations of sensitivity and specificity (http://vassarstats.net.au/clin1.html ) utilizing cut-offs as determined by the manufacturer and as published by Brockwell et al. (2014; 2013) were performed. 2.10

Animal ethics The Novartis Animal Health Australasia Pty Limited Animal Ethics Committee (AEC)

approved the conduct of this study (approval YAR-13-065). Samples obtained at necropsy from single age infection studies were approved by the same Ethics Committee (approvals YAR-12-058, YAR-13-043, YAR-13-024 and NAH-14-020). The Sydney University AEC also ratified approvals. 3. 3.1

Results Sentinel sheep

The numbers of F. hepatica recovered from the two sentinel sheep were 11 and 86 (7.3% to 57.3% of metacercariae given, respectively). From the measurements of 92 whole F. hepatica collected from these sheep, 84.8%, 9.8% and 5.4%, were classified as 12 weeks, 6 weeks and 2 weeks of age, respectively. The challenge model was deemed successful, as all stages were present at week 11 and patency was demonstrated for all infections. 3.2

FWEC

In Study 1 control animal FWECs ranged from 10–49 epg at 15 weeks and 32–99 epg at 19 weeks post-infection (Table 2). In Study 2, the FWECs ranged from 0–1533, with a mean of 176.3 and a median of 10.5. 3.3

Fasciola hepatica counts and efficacy

Study 1. The numbers of F. hepatica at necropsy and calculated efficacies of treatment groups are presented in Table 2. At necropsy, 19 weeks following initial F. hepatica infection, F. hepatica counts from untreated animals in Group 1 ranged from 38–86 with a geometric mean of 60 (F. hepatica recovery equivalent to 25.3% to 57.3% of metacercariae given). The efficacy of the treatments was 10

within expected label claims against mixed age F. hepatica of the Sunny Corner isolate. Fasciola hepatica counts from treated animals ranged between 0–4, 0–55 and 43–89, with geometric means of 1, 8 and 58 F. hepatica, for Groups 2, 3 and 4, respectively (Table 2). Linear measurements of fluke are summarized in Figure 1. For untreated animals, 96.7% of the 338 F. hepatica measured were >12 mm, indicative of fluke greater than 12 weeks of age. Similarly, 96% of the 328 F. hepatica recovered from Group 4 were >12 mm. Groups 2 and 3 had 8 and 89 F. hepatica measured, of which 87.5% and 71.9%, respectively, were adults >12 mm. Fasciola hepatica of <8 mm were recovered in Group 3 only (4.5%), which also had the highest percentage (23.6%), of flukes 8-12 mm in length. Study 2. The F. hepatica burdens in the 50 animals ranged from zero to 136, with a mean of 22.6 and a median of 4.5. 3.4

cELISA results The levels of coproantigen detected over time in the two studies are presented in Figure 2.

Utilizing a cut-off of 1.3% proposed by Brockwell et al. (2014), positive cELISA results initially occurred for individual sheep at 4 wpi in a single infection and 7 wpi for the multi-age infection (Figure 2). However, complete positive results for entire treatment groups were not achieved until 6 wpi (single age infection) and 9 wpi (multi-age infection). When the manufacturer’s cut-offs were applied, the time to initial positive readings were the same, but a positive result from all infected sheep did not occur until 7wpi for a single infection and 11 wpi for a multi-age infection. In Study 1, after treatment at 12 weeks, the cELISA was negative for all sheep in the TCBZ (Group 2) and CLOS (Group 3) treatment groups 1-2 weeks following treatment (Figure 2). In Group 2, (TCBZ) one animal with one 12mm F. hepatica returned a positive ELISA (2.21%) at 4 weeks post-treatment. This same animal and a second sheep with 4 F. hepatica 15–22mm in length were positive respectively, at 6 (23.7%, 1.66%) and 7 (15.4%, 1.66%) weeks post-treatment. The third sheep in Group 2 (TCBZ) had 4 surviving 13–15mm F. hepatica but remained a false-negative for the entire post-treatment period. A single sheep with seven 7–8 mm F. hepatica in Group 3 11

(CLOS) returned a positive cELISA (2.20%) at 3 weeks post-treatment. Three of five CLOS animals with recovered F. hepatica 7–18mm in length were positive at 6 weeks (1.60–6.3%) and 7 weeks (5.20–44.6%) post-treatment. The two remaining sheep with 8–19 mm F. hepatica in Group 3 returned false-negative cELISAs for the entire post-treatment duration. A single sheep in this group returned a false-positive result (1.88%) at 7 weeks post-treatment. Although five of six ABZ treated sheep in Group 4 demonstrated a clear reduction in coproantigen a week after treatment, only two of these could be deemed coproantigen negative (i.e. <1.14%). All animals in Group 4 became and remained coproantigen positive from 2 weeks after treatment (from week 14; Figure 2). The mean OD of treatment groups was not significantly different at treatment (week 12). Post-treatment, no significant differences in OD were demonstrated between Groups 1 (untreated) and 4 (ABZ) or Groups 2 (TCBZ) and 3 (CLOS). Significant differences in OD were demonstrated between CLOS and TCBZ against the untreated controls, with treated animals having the lower values (p=0.002 from weeks 13-18, p <0.001 week 19). In sheep treated with closantel and TCBZ there was a significantly lower OD compared to the ABZ treatment (p= 0.0022 weeks 13 through 18; except week 18 for TCBZ p= 0.0043, p <0.001 week 19). There were significant differences in OD between CLOS and TCBZ treated groups compared to the untreated group and ABZ treated group, with lower values in the former treatments for week 13 and 19 FWEC and F. hepatica burden (Table 2). In addition TCBZ F. hepatica burdens were significantly lower than those of CLOS treated sheep (Table 2). In Study 2 at the point of necropsy, 47 sheep were positive in the cELISA. Three sheep returned a false-positive cELISA, two of these were also false-positive when the manufacturer’s cutoff was applied. Three sheep, constituting 5.3% of the cohort, with mixed age infections showed a false-negative result on the cELISA when the 1.3% cut-off was applied. The number of falsenegatives increased to 10 (17.5% of cohort) under the manufacturer’s cut-off and 50% of these were mixed age infections of 4–9 F. hepatica. The remaining five sheep were single age infections with 1-2 F. hepatica recovered. 12

3.5

Correlations between cELISA, FWEC and F. hepatica burdens (Study 1) Correlations between cELISA results, FWECs and F. hepatica burdens are presented in

Table 3 and scatterplots in Figures 3a, 3b and 3c. Fasciola hepatica burdens correlated strongly with OD of the cELISA results from all sheep at week 13 of the study and strengthened through to week 19 (R=0.885). When individual treatment groups were assessed, the strength and presence of correlations was inconsistent over the study period. The untreated group had positive correlations at five of the 17 time points, with the strongest occurring at week 10 (R=0.862). The strongest correlation for the TCBZ group was R=0.608 at week 15, while the F. hepatica burdens in the CLOS group correlated from week 17 to its peak at week 19 (R=0.982). No correlation was found between cELISAs and F. hepatica burdens in the ABZ group. The strongest correlation between FWECs and cELISA were found at week 11 and 15 in the CLOS treated group (R=0.976, R=0.993, respectively). The correlation for all sheep increased over the three time points and was highest at week 19 (R=0.873). At week 19, the untreated group and TCBZ treated group was highly correlated with FWECs and cELISAs (R=0.826, R=0.931 respectively, Table 3). Across all groups correlations were found between post-treatment FWECs and F. hepatica burdens; highest at week 15 (R=0.778). A weak correlation was demonstrated in the control animals for week 11 FWECs and F. hepatica burdens (R=0.625; Table 3). 3.6

Pooled statistical analysis (Study 1 and 2) Although F. hepatica burdens were found to be significantly different (p=0.043) between single

aged (Study 2) and multi-age infections (Study 1), with the multi-age infections having higher burdens, no significant differences were found between the cELISA or FWEC results. Significant differences were demonstrated between all ranges examined for F. hepatica burdens and all except comparisons between F. hepatica burdens of 1-20 and 21-50 for cELISAs (p=0.172) and FWECs (p=0.106, Table 4). The difference between cELISA and FWEC results for F. hepatica burdens

13

between 21-50 and 51-100, although significant (p=0.046), was noticeably weaker than other comparisons between ranges. Regression models (Table 5) with the response variate being F. hepatica burden demonstrated strong interaction when ELISA was the fitted term (R2=0.777). Separating infection type marginally improved this relationship, although little difference between the two was found with single age infection being the stronger of the two (R2=0.783, R2=0.782). When FWEC was fitted for F. hepatica burden the results were less consistent with single age infections accounting for the greatest variance (R2=0.854), followed by the total dataset (R2=0.571) and mixed age infections (R2=0.505). If both diagnostic tools were used, the variance explained by the model increased for the total cohort (R2=0.835) and single age infection (R2=0.927) but not mixed age infections (R2=0.774). Exclusion of zero values did not improve any of the models for predicting F. hepatica burden (Table 5). By further restricting the dataset, dropping lower ranges, the model weakened for cELISA and the combined model. The model utilizing FWEC actually improved by removing lower burdens until only those animals with >100 F. hepatica were included at which point the variance accounted for decreased dramatically. Restrictions of the higher burdens out of the data resulted in declines in model strength, although this was not consistent with the fit being weaker when burdens of 50 or less were included as compared with burdens of 20 or less. Utilising the expanded regression model variables of infection rate, source of sheep and age of metacercariae were found to be insignificant. The model incorporating terms of FWEC, ELISA, F. hepatica isolate, infection type and animal age further improved the variance able to be explained (R2 = 0.893, p < 0.001). The weighted KAPPA for the manufacturer cELISA cut-off was 0.6308, whilst the cut-off of 1.3% returned a value of 0.789 for F. hepatica burden. Faecal egg count returned a weighted KAPPA for F. hepatica burden of 0.619. Sensitivity was greater utilizing a cut-off of 1.3% (93.1%) compared to the manufacturer’s recommendation (82.5%), however specificity improved under the 14

manufacturer cut-off (89.5%) rather than 1.3% (84.2%). Sensitivity and specificity were both calculated as 84.2% for FWEC against F. hepatica burden. 4.

Discussion

These studies addressed whether the detection of F. hepatica antigen in faeces of infected sheep by the cELISA adequately represented the actual F. hepatica infestation in sheep as the infections progressed and whether treatment failure could be identified under field conditions as previously proposed for cattle and sheep (Brockwell et al., 2014; Flanagan et al., 2011a). The time to antigen reappearance after effective removal of adult F. hepatica populations, and the comparisons of the kinetics and quantity of coproantigen in single and multi-age infections was determined. The kinetics of reappearance of coproantigen and fluke counts at necropsy was consistent with the characteristics of the “Sunny Corner” isolate, in that the isolate was susceptible to TCBZ and CLOS and exhibited some resistance to albendazole (Brockwell et al., 2013; Fairweather, 2011a). Results of Study 1 indicated that the establishment of successive infections progressively declines, with immature burdens (≤6 weeks of age) comprising substantially less than two-thirds of the population recovered at necropsy in sentinel animals. As gold standard methods for fluke recovery and complete microscopic examination of incubates were undertaken, the diagnostic sensitivity of these results is very high. In contrast to nematodes, where establishment limits have been well defined (Dobson et al., 1990) very little data regarding establishment rates during repeated (trickle) infections for F. hepatica are available. In sheep, there is no effective immunity to F.hepatica (Spithill et al., 1997) so, as in cattle, a combination of escalating inflammation and hepatic fibrosis probably acts as a mechanical barrier to reduce establishment success and growth (Boray, 1969; Rojo-Vazquez et al., 2012; Young et al., 2010). Perez et al. (2005) demonstrated overall higher (25.3 vs. 13.5%), but non-significant, establishment rates of F. hepatica in sheep trickle infected with 25 metacercariae per day for 7 days when compared to two single infections, each of 200 metacercariae, given 18 weeks apart. Differences in establishment for subsequent trickle or challenge doses were not assessed in this study as necropsy occurred once F. hepatica 15

were adult, however, as has been noted for abomasal nematodes, it is possible that the presence of established adult parasites or their inflammatory sequellae, reduces the establishment rates for newly ingested metacercaria. Efficacy results of Study 1 confirmed that the Sunny Corner isolate was susceptible (98.5%) to triclabendazole in experimental mixed age infections consistent with reported results for 6 week old (98.4%; sheep) and adult (100%; sheep and cattle) F. hepatica (Brockwell et al., 2013; Fairweather, 2011a). The closantel result of 86.8% is also within the expected efficacy spectrum for both sheep and cattle of 50-90% at 6-8 weeks, and 91-99% against >8 weeks (Boray, 1986; Dobbins and Wellington, 1982; Fairweather and Boray, 1999b; Guerrero, 1984). Albendazole is known to be ineffective against immature fluke and even at 12 weeks efficacy spectrum in sheep (4.75) and cattle (10 mg/kg) remains at 80-99% (Campbell and Hall, 1979; Fairweather and Boray, 1999a). For study 1, the efficacy of albendazole was extremely low at 3.3% given that 84.8% of flukes recovered and measured from sentinel sheep at treatment were classed as 12 weeks of age. Whilst individual animal populations could vary, based on these results the number of adult parasites present at treatment would be expected to form the larger proportion of the treated population. It is expected that a higher efficacy should have been present (40-60%) so the result may be indicative of albendazole resistance. Pearson’s correlation of F. hepatica count and cELISA OD for all sheep in study 1 was moderate (>0.6) from week 13 (7 wpi for secondary infection) and steadily increased. Once all three infections were producing detectable antigen (week 17), the correlation exceeded 0.8. The strong correlation at necropsy of R=0.885, was similar to that (R=0.889) reported at 18 wpi by Mezo et al. (2004) in 15 experimentally infected lambs. These trends indicate that the cELISA adequately and quantitatively detects predominantly adult infection, but if largely immature burdens (as may be the case in the field) are present, diagnostic sensitivity is reduced. Like many other quantitative ELISAs the diagnostic sensitivity (DSn) of the cELISA approaches 100% at 8-10

16

weeks after single infections, as more of the incumbent parasites mature (Mezo et al., 2004); this occurred at 11 weeks in Study 1. Interestingly, when separated into treatment groups, the correlation between F. hepatica burden and ELISA in Study 1 became inconsistent. For the untreated control, the strongest correlation (R = 0.862) was seen in week 10. This perhaps is reflective of the strong establishment rate of the initial infection and the likely source of antigen detected. Further notable correlations were not seen in this group until week 14 and these fluctuated from weeks 16 and 19. Potentially, the lack of correlation at weeks 11–13 was related to physiological disruption in the liver as the final challenge migrated into parenchyma combined with the initial antigen rise from the second experimental F. hepatica challenge. By week 14, the antigen expression would have established and the liver stabilised from both challenges, enabling correlation. Conversely, for the ALB treatment group, significant correlations were not observed at any time point. For the TCBZ treatment, only at weeks 3 and 15 did the correlation exceed 0.6, while the correlation in the CLOS group was highest from weeks 17-19. The delay in any correlation was expected, reflecting the maturation of the reduced numbers of surviving F. hepatica. The poor correlation in the TCBZ treated group raises concern that this test will not adequately detect early resistance. From the current data (Study 1), it would appear that the cELISA is incapable of differentiating between effective (TCBZ) and partially effective treatment (CLOS) at 1-2 weeks post-treatment as previously suggested in cattle and sheep (Brockwell et al., 2013; Flanagan et al., 2011a; Gordon et al., 2012b). In clear-cut cases of treatment failure (i.e. ALB treatment) the cELISA clearly demonstrates treatment failure. It is notable that F. hepatica recovered from both TCBZ and CLOS treatment groups were ≤ 22 mm in length, suggesting that the vast majority of survivors were either 2 or 6 weeks of age at treatment (Behm and Sangster, 1998; Boray, 1969). The inability of cELISA to detect surviving immature F. hepatica until 6 weeks post-treatment, and not differentiate effectiveness of the treatment for a further week, is a major limitation for any field use. As it has 17

been previously suggested, resistance is likely to develop initially in juvenile stages of F. hepatica, for example < 6 weeks old (Boray, 1990; Fairweather and Boray, 1999a). Thus it is imperative that immature survivors are identified and managed on farm to avoid subsequent resistant populations due to the high egg output of F. hepatica (Brunsdon, 1967; Fairweather, 2011b; Rojo-Vazquez et al., 2012). Although the mean result for TCBZ and CLOS in study 1 reflected the presence of survivors by 6 weeks post-treatment, individual results were not reliable. A false-negative rate of 33% (TCBZ) and 40% (CLOS) was demonstrated using a cut-off of 1.3% out to 7 weeks post-treatment, suggesting an even longer re-test interval may be necessary to detect surviving immatures. Early reports by Mezo et al. (2004) suggested the cELISA was capable of reliably detecting burdens of single F. hepatica as early as 4 weeks post-infection in sheep, but the current data (Study 1) revealed false-negatives for burdens of 4-7 F. hepatica at least nine weeks post-infection. However, single experimental infections in study 2 did not return false-positives at 3 weeks posttreatment with a 1.3% cut-off, yet when the manufacturer’s recommendations were followed 5 animals and an extra 2 from the multiple infection group were also false-positives. False-negative results have been reported previously in both field and experimental infections (Charlier et al., 2008; Gordon et al., 2012b). This supports a reduced cut-off as suggested for cattle by many authors (Brockwell et al., 2014; Brockwell et al., 2013; Charlier et al., 2008) are more appropriate, reflected by the improved DSn in the current analysis. Recent modifications to the MM3-COPRO ELISA, on which the commercialized Bio-K201 is based, suggest the utilization of streptavidin-polymerised horseradish peroxidase (as opposed to avidin-horseradish peroxidase) along with shaking at 750rpm half hour incubations significantly lowers the limit of detection of the parent ELISA and negates false negatives in low burden cattle samples (Martínez-Sernández et al., 2016). The impact of these modifications on the commercial kit would be worth investigating and may have improved the false reads in the current data, potentially improving the correlations with burdens of treated animals. 18

The strongest correlation of worm egg count with F. hepatica burden for Study 1 was at week 19 (7 weeks post-treatment), in contrast to guidance for faecal sampling in F. hepatica fecal egg count reduction tests (FECRT) of 3 weeks post-treatment (Daniel et al., 2012). As with coproantigen, treatment group FWECs did not correlate well with F. hepatica burden posttreatment, suggesting that for both tests, variance within small group sizes is too great to provide reliable quantitation of residual F. hepatica burden or efficacy. This has obvious implications for sample size in field investigations of putative flukicide resistance. A strong correlation (R2=0.777) occurred between the F. hepatica burden (0–136) and cELISA of 76 sheep at necropsy. This was close to the R2=0.87 reported by Brockwell et al. (2013) from six cattle infected with 15–117 F. hepatica. However, the correlation of R2=0.571 between FWEC and cELISA in the current study was much lower than the R2=0.84 observed by Brockwell et al. (2013). The amount of bias and type of distribution in the respective datasets may have influenced the strength of the correlation. With this hypothesis in mind, the exclusion of zero burdens might be expected to have improved the correlation, yet the correlations weakened when the dataset was restricted; suggesting inherent variance was more influential on the outcome. Whilst the necropsy correlations of F. hepatica and cELISA for necropsy data as described above are promising, the effect is strengthened by addition of FWEC results to cELISA (R2=0.835). The strength of this correlation indicates potential of a combined non-terminal test for efficacy, and possibly reduced sample sizes in the field. However, given the erratic and reduced correlations demonstrated for individual treatment groups in Study 1, the number of animals required for reliable results in laboratory or efficacy studies is likely to be substantially greater. This has ethical considerations, so that further examination of much larger datasets would be required to truly identify whether FWEC and cELISA can adequately replace terminal studies. The level of antigen detected in sheep enrolled in Study 1 were slower to develop over time and had lower OD expression levels, when compared to 9 sheep monitored in study 2 with a single infection. Positive ODs were detected on the cELISA 4–6wpi for the single age infections and 7–9 19

wpi for multi-age infection. The detection of antigen in sheep for single age infection is consistent with previous reports (Dumenigo et al., 2000; Flanagan et al., 2011b; Mezo et al., 2004; Valero et al., 2009), whilst the multi-age infection was later; similar to what has been demonstrated in natural field infections (Gordon et al., 2012b). The overall quantity of antigen detected was greater in single infections than in control multi-age infection, likely reflective of the mean F. hepatica burdens; 104 and 60 F. hepatica respectively. The kinetics of antigen expression appears to be affected by the presence of multi-age F. hepatica with plateaus and dips present whilst single infection demonstrates a smooth linear increase. 5.

Conclusion The Bio-X ELISA is an appropriate tool for monitoring the effectiveness of treatments against

Fasciola hepatica if an adult infection is present. However the low correlations with fluke burdens (especially immature stages) limit its ability to define efficacy in traditional terms, so it is unsuited for experimental studies with small group sizes. In instances where immature parasites are present it is recommended that an initial cELISA be followed with a secondary cELISA at least 6 weeks after treatment to ensure early resistance is detected. In instances where resistance is suspected and where necropsy cannot be undertaken, the use of FWEC combined with cELISA may improve the quantification of F. hepatica burden. Further optimization of such a protocol would be needed. Conflict of interest statement Several authors are paid employees of ELANCO Animal Health, Australia or ELANCO Animal Health, Switzerland. The latter funded this study. Acknowledgements The authors acknowledge the support of their colleagues in the Animal Management and Facilities, Biological Management and Efficacy sections at Yarrandoo R&D Centre. Navneet Dhand is thanked for his advice on the statistical analysis of Study 2. Meat and Livestock Australia (MLA) is thanked for the award of a technical assistance grant. Barry Hosking and Tim Elliot are thanked for their constructive comments on the draft manuscript. 20

References Abbott, W.S., 1925. A method of computing the effectiveness of an insecticide. J. Econ. Entomol. 18, 265-267. Alvarez-Sanchez, M.A., Mainar-Jaime, R.C., Perez-Garcia, J., Rojo-Vazquez, F.A., 2006. Resistance of Fasciola hepatica to triclabendazole and albendazole in sheep in Spain. Vet. Rec. 159, 424-425. Behm, C.A., Sangster, N.C., 1998. Pathology, Pathophysiology and Clinical Aspects, In: Dalton, J.P. (Ed.) Fasciolosis. CABI publishing, Wallingford, UK, pp. 185-224. Boray, J.C., 1969. Experimental fascioliasis in Australia. Adv. Parasitol. 7, 95-210. Boray, J.C., 1986. Trematode infections of domestic animals, In: Campbell, W.C., Rew, R.S. (Eds.) Chemotherapy of parasitic diseases. Plenum Press, New York, pp. 401-425. Boray, J.C., 1990. Drug Resistance in Fasciola hepatica, In: Boray, J.C., Martin, P.J., Roush R.T. (Ed.) Resistance to Antiparasitic Drugs. MSD AGVET, New Jersey, pp. 51-90. Brennan, G.P., Fairweather, I., Trudgett, A., Hoey, E., McCoy, McConville, M., Meaney, M., Robinson, M., McFerran, N., Ryan, L., Lanusse, C., Mottier, L., Alvarez, L., Solana, H., Virkel, G., Brophy, P.M., 2007. Understanding triclabendazole resistance. Exp. Mol. Pathol. 82, 104109. Brockwell, Y.M., Elliott, T.P., Anderson, G.R., Stanton, R., Spithill, T.W., Sangster, N.C., 2014. Confirmation of Fasciola hepatica resistant to triclabendazole in naturally infected Australian beef and dairy cattle. Int. J. Parasitol. Drugs Drug Resist. 4, 48-54. Brockwell, Y.M., Spithill, T.W., Anderson, G.R., Grillo, V., Sangster, N.C., 2013. Comparative kinetics of serological and coproantigen ELISA and faecal egg count in cattle experimentally infected with Fasciola hepatica and following treatment with triclabendazole. Vet. Parasitol. 196, 417426. Brunsdon, R.V., 1967. Liver fluke Fasciola hepatica in sheep and cattle in New Zealand and its control. N. Z. Vet. J. 15, 12-23. Campbell, N.J., Hall, C.A., 1979. The anthelmintic efficacy of albendazole against Fasciola hepatica and benzimidazole resistant strains of Haemonchus contortus and Trichostrongylus colubriformis in sheep. Res Vet Sci 26, 90-93. 21

Carnevale, S., Pantano, M.L., Kamenetzky, L., Malandrini, J.B., Soria, C.C., Velasquez, J.N., 2015. Molecular diagnosis of natural fasciolosis by DNA detection in sheep faeces. Acta. Parasitol. 60, 211-217. Charlier, J., De Meulemeester, L., Claerebout, E., Williams, D., Vercruysse, J., 2008. Qualitative and quantitative evaluation of coprological and serological techniques for the diagnosis of fasciolosis in cattle. Vet. Parasitol. 153, 44-51. Daniel, R., van Dijk, J., Jenkins, T., Akca, A., Mearns, R., Williams, D.J., 2012. Composite faecal egg count reduction test to detect resistance to triclabendazole in Fasciola hepatica. Vet. Rec. 171, 153, 151-155. Diab, T.M., Mansour, H.H., Mahmoud, S.S., 2010. Fasciola gigantica: Parasitological and scanning electron microscopy study of the in vitro effects of ivermectin and/or artemether. Exp. Parasitol. 124, 279-284. Dobbins, S., Wellington, A., 1982. Comparison of the activity of some fasciolicides against immature liver fluke in calves. Veterinary Record 111, 177-178. Dobson, R.J., Waller, P.J., Donald, A.D., 1990. Population dynamics of Trichostrongylus colubriformis in sheep: the effect of infection rate on the establishment of infective larvae and parasite fecundity. Int. J. Parasitol. 20, 347-352. Dumenigo, B.E., Espino, A.M., Finlay, C.M., Mezo, M., 2000. Kinetics of antibody-based antigen detection in serum and faeces of sheep experimentally infected with Fasciola hepatica. Vet. Parasitol. 89, 153-161. Elliott, T.P., Kelley, J.M., Rawlin, G., Spithill, T.W., 2015. High prevalence of fasciolosis and evaluation of drug efficacy against Fasciola hepatica in dairy cattle in the Maffra and Bairnsdale districts of Gippsland, Victoria, Australia. Vet. Parasitol. 209, 117-124. Fairweather, I., 2009. Triclabendazole progress report, 2005-2009: an advancement of learning? J. Helminthol. 83, 139-150. Fairweather, I., 2011a. Liver fluke isolates: a question of provenance. Vet. Parasitol. 176, 1-8. Fairweather, I., 2011b. Reducing the future threat from (liver) fluke: realistic prospect or quixotic fantasy? Vet. Parasitol. 180, 133-143.

22

Fairweather, I., Boray, J.C., 1999a. Fasciolicides: efficacy, actions, resistance and its management. Vet. J. 158, 81-112. Fairweather, I., Boray, J.C., 1999b. Mechanisms of fasciolicide action and drug resistance in Fasciola hepatica, In: Fasciolosis. CABI publishing, Wallingford, UK, pp. 225-276. Flanagan, A., Edgar, H.W., Gordon, A., Hanna, R.E., Brennan, G.P., Fairweather, I., 2011a. Comparison of two assays, a faecal egg count reduction test (FECRT) and a coproantigen reduction test (CRT), for the diagnosis of resistance to triclabendazole in Fasciola hepatica in sheep. Vet. Parasitol. 176, 170-176. Flanagan, A.M., Edgar, H.W., Forster, F., Gordon, A., Hanna, R.E., McCoy, M., Brennan, G.P., Fairweather, I., 2011b. Standardisation of a coproantigen reduction test (CRT) protocol for the diagnosis of resistance to triclabendazole in Fasciola hepatica. Vet. Parasitol. 176, 3442. Gaasenbeek, C.P.H., Moll, L., Cornelissen, J.B.W.J., Vellema, P., Borgsteede, F.H.M., 2001. An experimental study on triclabendazole resistance of Fasciola hepatica in sheep. Veterinary parasitology 95, 37-43. Gordon, D., Zadoks, R., Skuce, P., Sargison, N., 2012a. Confirmation of triclabendazole resistance in liver fluke in the UK. Vet. Rec. 171, 159-160. Gordon, D.K., Zadoks, R.N., Stevenson, H., Sargison, N.D., Skuce, P.J., 2012b. On farm evaluation of the coproantigen ELISA and coproantigen reduction test in Scottish sheep naturally infected with Fasciola hepatica. Vet. Parasitol. 187, 436-444. Guerrero, J., 1984. Closantel: A review of its antiparasitic activity. Preventive Veterinary Medicine 2, 317-327. Happich, F.A., Boray, J.C., 1969. Quantitative diagnosis of chronic fasciolosis. 1. Comparative studies on quantitative faecal examinations for chronic Fasciola hepatica infection in sheep. Aust. Vet. J. 45, 326-328. Kajugu, P.E., Hanna, R.E., Edgar, H.W., Forster, F.I., Malone, F.E., Brennan, G.P., Fairweather, I., 2012. Specificity of a coproantigen ELISA test for fasciolosis: lack of cross-reactivity with Paramphistomum cervi and Taenia hydatigena. Vet. Rec. 171, 502.

23

Kelley, J.M., Elliott, T.P., Beddoe, T., Anderson, G., Skuce, P., Spithill, T.W., 2016. Current Threat of Triclabendazole Resistance in Fasciola hepatica. Trends in parasitology 32, 458-469. MAFF, 1971. Technical Bulletin No. 18: Manual of Veterinary Parasitological Laboratory Techniques. MAFF, London, 131 p. Martinez-Perez, J.M., Robles-Perez, D., Rojo-Vazquez, F.A., Martinez-Valladares, M., 2012. Comparison of three different techniques to diagnose Fasciola hepatica infection in experimentally and naturally infected sheep. Vet. Parasitol. 190, 80-86. Martínez-Sernández, V., Orbegozo-Medina, R.A., González-Warleta, M., Mezo, M., Ubeira, F.M., 2016. Rapid Enhanced MM3-COPRO ELISA for Detection of Fasciola Coproantigens. PLoS neglected tropical diseases 10, e0004872. Martinez-Valladares, M., del Rosario Famularo, M., Fernandez-Pato, N., Castanon-Ordonez, L., Cordero-Perez, C., Rojo-Vazquez, F.A., 2010. Efficacy of nitroxynil against Fasciola hepatica resistant to triclabendazole in a naturally infected sheep flock. Parasitol. Res. 107, 12051211. Mas-Coma, S., Valero, M.A., Bargues, M.D., 2009. Chapter 2. Fasciola, lymnaeids and human fascioliasis, with a global overview on disease transmission, epidemiology, evolutionary genetics, molecular epidemiology and control. Adv. Parasitol. 69, 41-146. McCaughey, W.J., Hatch, C., 1964. Routine faecal examination for the detection of fluke Fasciola hepatica eggs. An evaluation of some techniques. . Ir. Vet. J. 18, 181-187. McManus, D.P., Dalton, J.P., 2006. Vaccines against the zoonotic trematodes Schistosoma japonicum, Fasciola hepatica and Fasciola gigantica. Parasitol. 133 Suppl, S43-61. Mezo, M., Gonzalez-Warleta, M., Carro, C., Ubeira, F.M., 2004. An ultrasensitive capture ELISA for detection of Fasciola hepatica coproantigens in sheep and cattle using a new monoclonal antibody (MM3). J. Parasitol. 90, 845-852. Mezo, M., Gonzalez-Warleta, M., Ubeira, F.M., 2007. The use of MM3 monoclonal antibodies for the early immunodiagnosis of ovine fascioliasis. J. Parasitol. 93, 65-72. Novobilsky, A., Averpil, H.B., Hoglund, J., 2012. The field evaluation of albendazole and triclabendazole efficacy against Fasciola hepatica by coproantigen ELISA in naturally infected sheep. Vet. Parasitol. 190, 272-276. 24

Ortiz, P., Scarcella, S., Cerna, C., Rosales, C., Cabrera, M., Guzman, M., Lamenza, P., Solana, H., 2013. Resistance of Fasciola hepatica against Triclabendazole in cattle in Cajamarca (Peru): a clinical trial and an in vivo efficacy test in sheep. Vet. Parasitol. 195, 118-121. Overend, D.J., Bowen, F.L., 1995. Resistance of Fasciola hepatica to triclabendazole. Aust. Vet. J. 72, 275-276. Perez, J., Ortega, J., Bravo, A., Diez-Banos, P., Morrondo, P., Moreno, T., Martinez-Moreno, A., 2005. Phenotype of hepatic infiltrates and hepatic lymph nodes of lambs primarily and challenge infected with Fasciola hepatica, with and without triclabendazole treatment. Veterinary research 36, 1-12. Pfister, K., 1990. Serodiagnosis of fasciolosis in ruminants. Rev. Sci. Tech. 9, 511-518. Rojo-Vazquez, F.A., Meana, A., Valcarcel, F., Martinez-Valladares, M., 2012. Update on trematode infections in sheep. Vet. Parasitol. 189, 15-38. Sanchez-Andrade, R., Paz-Silva, A., Suarez, J.L., Panadero, R., Pedreira, J., Diez-Banos, P., Morrondo, P., 2001. Effect of fasciolicides on the antigenaemia in sheep naturally infected with Fasciola hepatica. Parasitology research 87, 609-614. Spithill, T.W., Piedrafita, D., Smooker, P.M., 1997. Immunological approaches for the control of fasciolosis. Int. J. Parasitol. 27, 1221-1235. Thienpont, D., Rochette, F., Vanparijs, O., 1979. Diagnosing helminthiasis through coprological examination. Jansenn Research Foundation, Beerse, Belgium. Valero, M.A., Ubeira, F.M., Khoubbane, M., Artigas, P., Muino, L., Mezo, M., Perez-Crespo, I., Periago, M.V., Mas-Coma, S., 2009. MM3-ELISA evaluation of coproantigen release and serum antibody production in sheep experimentally infected with Fasciola hepatica and F. gigantica. Vet. Parasitol. 159, 77-81. Vercruysse, J., Holdsworth, P., Letonja, T., Barth, D., Conder, G., Hamamoto, K., Okano, K., 2001. International harmonisation of anthelmintic efficacy guidelines. Vet. Parasitol. 96, 171-193. Young, N.D., Hall, R.S., Jex, A.R., Cantacessi, C., Gasser, R.B., 2010. Elucidating the transcriptome of Fasciola hepatica - a key to fundamental and biotechnological discoveries for a neglected parasite. Biotechnology advances 28, 222-231.

25

Zimmerman, G.L., Jen, L.W., Cerro, J.E., Farnsworth, K.L., Wescott, R.B., 1982. Diagnosis of Fasciola hepatica infections in sheep by an enzyme-linked immunosorbent assay. Am. J. Vet. Res. 43, 2097-2100.

26

Figure legends Figure 1 Numbers of Fasciola hepatica of various lengths recovered by treatment (Study 1). Recovered Fasciola hepatica at 19 weeks post infection from Group 1 untreated controls (dots), Group 2 treated with 10 mg/kg triclabendazole (light grey), Group 3 treated with 7.5 mg/kg closantel (stripes) and Group 4 treated with 4.75 mg/kg albendazole (dark grey). Figure 2 Mean faecal antigen (OD) over time (± s.e.) - Study 1. Optical density (450nm) at each week post infection of untreated control single infection (purple bar), Group 1 untreated control multiple infection (black bar), Group 2 treated with 10 mg/kg triclabendazole (blue bar), Group 3 treated with 7.5 mg/kg closantel (green bar) and Group 4 treated with 4.75 mg/kg albendazole (red bar). Arrow indicates timing of treatments. Figure 3a Scatterplot of ELISA OD and fluke burden at necropsy (Study 1; week 19). Correlation of cELISA OD (450nm) with fluke burden at 19 wpi of Group 1 untreated controls (unfilled diamond; R2=0.404), Group 2 treated with 10 mg/kg triclabendazole (blue diamond; R2=-0.032), Group 3 treated with 7.5 mg/kg closantel (green triangle; R2=0.982), Group 4 treated with 4.75 mg/kg albendazole (red triangle; R2=0.395) and pooled data (dotted line; R2=0.885). Figure 3b

27

Scatterplot of FWEC and fluke burden at necropsy (Study 1; week 19). Correlation of FWEC with fluke burden at 19 wpi of Group 1 untreated controls (unfilled diamond; R2=-0.017), Group 2 treated with 10 mg/kg triclabendazole (blue diamond; R2=0.182), Group 3 treated with 7.5 mg/kg closantel (green triangle; R2=0.120), Group 4 treated with 4.75 mg/kg albendazole (red triangle; R2=-0.211) and pooled data (dotted line; R2=0.688). Figure 3c Scatterplot of ELISA OD and FWEC at necropsy (Study 1; week 19). Correlation of cELISA (450nm) with FWEC at 19 wpi of Group 1 untreated controls (unfilled diamond; R2=0.826), Group 2 treated with 10 mg/kg triclabendazole (blue diamond; R2=0.931), Group 3 treated with 7.5 mg/kg closantel (green triangle; R2=-0.012), Group 4 treated with 4.75 mg/kg albendazole (red triangle; R2=0.457) and pooled data (dotted line; R2=0.873)

28

Figr-1

29

Figr-2

30

Figr-3

31

32

33

Table 1 Treatment groups and anthelmintic information Group Treatment allocation

n=

Dose (mg/kg)

Trade name

1

Untreated control

8*

2

Triclabendazole

6

10 mg/kg Fasinex®

Novartis Animal Health Australasia Pty Limited

3

Closantel

6

7.5 mg/kg

Western Stock Distributors

4

Albendazole

6

WSD Closantel

Registrant

4.75 Alben Virbac Australia Pty Limited mg/kg *Two sentinel animals (necropsied at week 11) were not included in the calculation of efficacy

34

Table 2 Fluke counts, fecal worm egg counts and efficacy of three anthelmintic treatments against mixed aged infections of Fasciola hepatica (sheep infected 12, 6 and 2 weeks before treatment) - Study 1 p-values a (group comparison) Arithmetic Geometric Arithmetic Geometric Group Group Group 2 3 4 37.7 33.8 0.0022 0.0022 0.0649 Mean counts Parameter Group Treatment FWEC, 3 weeks post treatment (week 15)

FWEC, 7 weeks post treatment (week 19)

F. hepatica counts, 7 weeks

1 2 3 4 1 2

Untreated control Triclabendazole 0.2 Closantel

0.2

Albendazole 20.0 Untreated 62.0 control Triclabendazole 0.5

Efficacy (%)

2

0.1

1.0000 0.0022 99.6

99.6

99.6 46.9 -

99.6 46.3 -

4 1 2

Closantel

3.3

Albendazole 52.7 Untreated 62.3 control Triclabendazole 1.5

2

0.1 18.2 57.1

0.0022 3

0.0022 0.0022 0.4848 2

0.3

3

1.0000 0.0022 99.2

3

3

2

99.4

0.7 47.2 59.7

0.0022 94.6 15.1 -

98.8 17.4 -

3

<.0001 0.0006 0.9449 2

0.9

3

0.0040 <.0001 97.6

98.5 35

2

2

post treatment (week 19)

3

Closantel

16

7.9

0.0007 74.3

86.8

3

4 Albendazole 59.7 57.7 4.3 3.3 a FWEC, faecal worm egg counts expressed as eggs/g faeces; next to each significant comparison, the group with the lower fluke count is indicated (superscript).

36

Table 3 Correlations of ELISA OD, FWEC and F. hepatica burden (Study 1)

Parameter

Wee ELISA k week FWEC 11 11 (vs. ELISA 15 15 OD) 19 19 FWEC (vs. F. hepatica count week 19) F. hepatica count (vs. ELISA OD)

Pearson’s correlation coefficient All Group 1 Group 2 groups untreated TCBZ 0.146 0.667 0.625 0.258 -0.235 0.752 0.873 0.826 0.931

Group 3 CLOS 0.976 0.993 -0.012

Group 4 ABZ 0.224 0.284 0.457

0.778

0.624 0.481

-0.340 -0.372

-0.116 -0.216

0.036 0.313

11 15 19

-

0.372

0.688

-0.017

0.182

0.120

-0.211

19

0 3 4 5 6 7 8 9 10 11 12 13 14

-0.218 0.313 -0.111 0.093 0.209 0.075 -0.034 0.023 0.203 0.073 0.323 0.667 0.709

-0.412 0.465 0.297 0.069 0.412 0.374 0.136 0.146 0.862 0.318 0.309 -0.015 0.616

-0.377 0.618 0.074 -0.582 -0.592 0.085 -0.021 -0.387 -0.079 -0.243 -0.253 -0.087 -0.035

0.152 -0.169 -0.252 0.422 -0.555 -0.103 -0.110 -0.359 -0.273 -0.286 -0.333 0.529 0.215

-0.143 0.110 -0.094 -0.115 0.538 0.268 -0.516 -0.179 -0.623 -0.276 -0.528 0.468 -0.199

37

15 16 17 18 19

0.748 0.797 0.845 0.837 0.885

0.837 0.235 0.777 0.780 0.404

0.608 0.065 0.546 -0.021 -0.032

-0.168 0.569 0.691 0.906 0.982

38

-0.673 0.237 0.358 0.142 0.395

Table 4 Summary statistics and comparisons (p-value) of F. hepatica burden, ELISA and FWEC result by level of infection (Study 2)

Parameter Fluke burden

ELISA

FWEC

Fluke burden 1-20 21-50 51-100 >100 1-20

n= Min.

Mean

Max.

SD

32 6 13 6 32

6.7 34.8 73.0 115.5 0.415

17 43 95 136 1.239

4.6 10.5 13.5 14.7 0.397

Comparison (p-value) by level of infection 0 1-20 21-50 51100 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 0.002 <.001 <.001

21-50 51-100 >100 1-20 21-50 51-100 >100

6 13 6 32 6 13 6

0.771 1.471 2.039 27.2 41.3 180.8 1034

1.756 2.406 2.238 173 99 778 1533

0.647 0.472 0.157 40.7 30.6 250.4 455.9

<.001 <.001 <.001 <.001 <.001 <.001 <.001

1 21 54 101 0.002 0.131 0.632 1.811 0 18 0 411

0.172 <.001 <.001

0.046 0.002

0.009

0.106 <.001 <.001

0.046 0.002

<.001

39

Table 5 Percentage variance of fluke burden predicted by linear regression of pooled necropsy data (Study 1 and 2) Fitted term variance (R2) Response variate Fluke burden

Range of burden Single age infection Mixed age infection All >0 >20 >50 >100 <100 <50 <20

n=

ELISA

FWEC

0.783

pvalue <.001

50

0.854

pvalue <.001

ELISA+FWEC pvalue 0.927 <.001

26

0.782

<.001

0.505

<.001

0.774

<.001

76 57 25 19 6 70 57 51

0.777 0.732 0.594 0.442 0.287 0.708 0.412 0.546

<.001 <.001 <.001 <.001 0.157 <.001 <.001 <.001

0.571 0.553 0.591 0.642 0.381 0.282 0.172 0.283

<.001 <.001 <.001 <.001 0.114 <.001 <.001 <.001

0.835 0.800 0.760 0.716 0.384 0.707 0.405 0.538

<.001 <.001 <.001 <.001 0.225 <.001 <.001 <.001

40