Apparent treatment failure of praziquantel and pyrantel pamoate against anoplocephalid tapeworms

Abstract

Anoplocephalid tapeworms are commonly occurring in grazing horses around the world. Two currently available anthelmintics have documented high efficacy against Anoplocephala perfoliata; praziquantel in various dosages ranging from 1.0 to 2.5 mg/kg and pyrantel pamoate administered at 13.2 mg base/kg. Anthelmintic resistance has not been reported in A. perfoliata, but anecdotal reports made during 2022 have suggested a possible loss of efficacy for both actives. This paper reports fecal egg count data from a Thoroughbred operation in Central Kentucky in 2023. Fifty-six yearlings were first dewormed with a combination of ivermectin (200 μg/kg) and praziquantel (1.5 mg/kg) and subsequently treated with pyrantel pamoate (13.2 mg base/kg). Fecal egg counts were determined at the day of treatment and again 14 days post-treatment. Two groups of mares (n = 39 and 45) were also treated with ivermectin/praziquantel and examined pre- and post-treatment. Low efficacy of ivermectin and pyrantel pamoate was demonstrated against strongylid parasites in the yearlings with mean Fecal Egg Count Reductions (FECRs) at 75.6% or below and upper 95% credible interval (CI) limits below 90% in all cases. Overall anti-cestodal FECR levels in the yearlings were 23.5% (95% CI: 11.2–48.0) for praziquantel and 50.9% (20.5–72.0) for pyrantel pamoate. Praziquantel eliminated anoplocephalid eggs from three of 17 yearlings, but another 5 yearlings went from negative to positive status following treatment. Pyrantel pamoate failed to eliminate anoplocephalid eggs from any of 14 treated tapeworm-positive yearlings. Nine of 84 mares tested positive for anoplocephalid eggs, and seven of these were still positive post praziquantel treatment. These findings sharply contrast data from historic field efficacy studies conducted for both actives and raise concern about anthelmintic resistance having possibly developed. This emphasizes the need for developing and refining antemortem methodologies for evaluating anti-cestodal treatment efficacy and for searching for possible alternative treatment options.

Graphical abstract

Highlights

• •Praziquantel and pyrantel pamoate reduced anoplocephalid egg counts by 23.5 and 50.9%, respectively.
• •Praziquantel eliminated anoplocephalid eggs from 3 of 17 horses.
• •Pyrantel pamoate did not eliminate anoplocephalid eggs from any of the 14 tapeworm-positive horses.
• •Strongylid egg count data suggested resistance to both ivermectin and pyrantel pamoate.

1. Introduction

Equine tapeworms of the anoplocephalid family are widely common in grazing horses around the world. Three species, Anoplocephala perfoliata, A. magna and Anoplocephaloides mamillana are described with A. perfoliata being, by far, the most common, while the two other species are rarely reported (Borgsteede and van Beek 1998; Meana et al., 2005; Rehbein et al., 2003; Rehbein et al., 2013). Prevalence of A. perfoliata has been reported in the range of 7–80% depending on location, season, and diagnostic method used (Rehbein et al., 2003; Meana et al., 2005; Tomczuk et al., 2015; Engell-Sørensen et al., 2018; Hreinsdóttir et al., 2019). The role of A. perfoliata in equine gastrointestinal disease has been well documented through case-control studies demonstrating an association between the parasite and conditions related to the íleum and cecum (Proudman and Edwards, 1993; Proudman et al., 1998).

Anti-cestodal efficacy has been demonstrated for two anthelmintic classes currently on the market for usage in equines; pyrantel pamoate/embonate and praziquantel. Early work demonstrated that pyrantel pamoate paste administered at the dosage labelled for strongyle treatment (6.6 mg base/kg) had a mean efficacy of 88% against A. perfoliata (Lyons et al., 1989), and subsequent work demonstrated that doubling the dosage to 13.2 mg base/kg yielded efficacy levels above 95% (Reinemeyer et al., 2006). The efficacy of praziquantel has been evaluated at dosages ranging from 0.25 to 2.5 mg/kg (Lyons et al., 1995, 1998; Slocombe, 2006; Slocombe et al., 2007) and has been reported to be >99% at dosages above 1.0 mg/kg (Slocombe 2006).

While the above-referenced studies were all terminal studies with efficacy estimates based on worm counts, it is also relevant to evaluate anthelmintic performance in field settings without conducting necropsies. Several field studies have been conducted with both anthelmintic classes and have demonstrated good efficacy in reducing A. perfoliata egg counts in fecal samples. Pyrantel pamoate administered at the 13.2 mg base/kg dosage was demonstrated to reduce egg counts by 95% or above starting at 8 days post administration (Craig et al., 2003; Marchiondo et al., 2006). Similarly, various anthelmintic formulations of praziquantel have been evaluated in field studies at dosages ranging from 1.0 to 2.5 mg/kg and found to eliminate >99% of A. perfoliata eggs at 14 days post administration (Craig et al., 2003; Grubbs et al., 2003; Lyons et al., 2017) and in some cases as early as 6 (Slocombe et al., 2007), 8 (Rehbein et al., 2003), and 10 days post treatment (Roelfstra et al., 2006).

Anthelmintic resistance is deemed biologically plausible in A. perfoliata, and this notion is supported by reports of apparent praziquantel resistance in Dipylidium caninum in dogs (Chelladurai et al., 2018) and Schistosoma spp. infecting humans (Vale et al., 2017). To date there have been no reports of apparent drug resistance in A. perfoliata, although one report of four colic cases associated with this parasite in yearlings that had been routinely treated with pyrantel pamoate did raise the suspicion of possible treatment failure (Peregrine et al., 2008). However, no other reports have been published raising similar suspicions. Nonetheless, over the course of 2022, this author received several reports from veterinarians in Central Kentucky of apparent failure to eliminate A. perfoliata eggs from fecal samples in horses treated with praziquantel or pyrantel pamoate, prompting further investigation.

The aim with this report is to provide data on field performance of praziquantel and pyrantel pamoate against A. perfoliata in Central Kentucky.

2. Materials and methods

2.1. Horses

The data presented herein were generated through regular routine deworming and testing procedures in place for a Thoroughbred operation located in Central Kentucky during February–April 2023. The study population was a group of US-bred Thoroughbred yearlings located in four different barns as well as two groups of mares located at two different facilities within the operation (Table 1). The yearlings were all born on a different facility in the same area belonging to the operation and then moved to the yearling facilities in the first week of November 2022. During 2022, they were first treated with a combination of piperazine (110 mg/kg) and fenbendazole (10 mg/kg) at approximately 2 and 4 months of age, followed by a combination of ivermectin (200 μg/kg) and praziquantel (1.5 mg/kg) at approximately 5 months of age. Subsequently, they were all treated with a combination of moxidectin (400 μg/kg) and praziquantel (2.5 mg/kg) on November 23, 2022. Mares are generally treated in the spring with a combination of ivermectin (200 μg/kg) and praziquantel (1.5 mg/kg) and in the autumn with a combination of moxidectin (400 μg/kg) and praziquantel (2.5 mg/kg). High strongylid shedders (exceeding 500 eggs per gram) are treated during summer with ivermectin (200 μg/kg) as well. In addition, foaling mares are treated with ivermectin (200 μg/kg) within five days of foaling.

Table 1.

Study population. All were Thoroughbreds. The yearlings were all born at the operation in 2022.

Yearlings
	N	Sex	Birth date range	Weaning age range (days)
Barn A	14	Fillies	April 8 - May 14	145–187
Barn B	16	Fillies	February 10 - March 31	149–231
Barn C	16	Colts	February 4 - April 6	126–197
Barn D	10	Colts	April 8 - May 20	148–189
Mares
	N		Age range
Facility 1	39		6–12 years
Facility 2	45		3–4 years

2.2. Anthelmintics

All horses were weighed on electronic scales on a monthly basis and anthelmintic products were administered orally according to weight. The yearlings were treated with a combination of ivermectin (200 μg/kg) and praziquantel (1.5 mg/kg) on February 7, 2023 (Equimax, Bimeda Animal Health, Outbrook Terrace, IL, USA). On February 27, 2023, they were treated with pyrantel pamoate (Strongid P, Zoetis, Parsipanny, NJ, USA) at 13.2 mg base/kg. The two groups of mares were treated with the same ivermectin (200 μg/kg) and praziquantel (1.5 mg/kg) combination on March 3rd or March 30th, 2023, respectively.

2.3. Egg counting technique

From the yearlings, fresh fecal samples were collected either rectally or from samples deposited on stall floors on February 6, 21, and March 13, 2023. Pre- and post-treatment samples were collected from the two groups of mares on March 3rd and 28th or March 9th and April 20th, respectively. Samples were refrigerated in airtight containers and analyzed within five working days.

Fecal egg counts (FEC) were determined onsite by a farm laboratory technician using the OvaTector system (BG Medical Products Inc, Venice, Florida, USA) with a multiplication factor of 10 and sodium nitrate (specific gravity 1.25) as flotation medium. The manufacturer's protocol was followed, but briefly, 2 g of feces were added to the provided cylinder and were suspended and mixed thoroughly. A 22 × 22 mm coverslip was placed on top of the cylinder and it was left for passive floatation for 15 minutes. The cover slip was then transferred to a microscope slide and examined under the microscope. Strongylid and tapeworm eggs were counted and recorded.

2.4. Data analysis

With the yearlings, anthelmintic efficacy was determined using an online web interface providing a Bayesian hierarchical model analysis of fecal egg count data with determination of model-estimated mean FECRs and 95% Credible Intervals (Torgerson et al., 2014; Wang et al., 2018). Data were analyzed using the two samples paired (with individual efficacy) option.

For classification of anti-strongylid efficacy, the newly published World Association for the Advancement of Veterinary Parasitology guidelines were followed for the clinical testing protocols for ivermectin and pyrantel (Kaplan et al., 2023). Based on these guidelines, the upper and lower efficacy thresholds for ivermectin were 99.9 and 92.0%, respectively, while the corresponding thresholds for pyrantel pamoate were 98.0 and 80.0%, respectively. The classification outcomes also followed WAAVP guidelines and were as follows:

No evidence of reduced efficacy: The lower credible limit is above the lower efficacy threshold.

Evidence of reduced efficacy: The upper credible limit is below the upper efficacy threshold.

Inconclusive: Neither criterion met.

There are no guidelines on FECR testing for efficacy against anoplocephalid parasites, so tapeworm data were not classified according to resistance status.

3. Results

3.1. Strongylid data

All four yearling barns fulfilled the criteria for the WAAVP clinical protocol for ivermectin efficacy testing against strongylids. With a minimum of five treated horses present in each barn, the WAAVP guidelines require a minimum of 200 eggs counted pre-treatment for the group (Kaplan et al., 2023). As can be seen in Table 2, the minimum number of eggs counted was 431, and three of the barns (A, C, and D) fulfilled the criteria for the more stringent research protocol requiring a minimum of 7 horses and 280 eggs counted. For each of the 4 b, the upper credible limit fell below the upper efficacy threshold of 99.9%, which means that the data are indicative of reduced ivermectin efficacy (Kaplan et al., 2023).

Table 2.

Efficacy of ivermectin and pyrantel pamoate against strongylid parasites in the four yearling barns in the study.

	Ivermectina				Pyrantel pamoateb
	N	Eggsc	FECRd	CIe	N	Eggsc	FECRd	CIe
Barn A	8	2,221	53.6	[26.6–71.5]	4	703	38.6	[13.3–66.3]
Barn B	5	431	30.3	[12.4–59.4]	4	261	75.6	[43.4–89.4]
Barn C	9	1,583	70.5	[49.6–82.3]	4	238	38.5	[14.7–67.8]
Barn D	10	2,731	48.9	[24.2–67.1]	2	413	62.3	[21.0–85.8]

^a200 μg/kg.

^b13.2 mg base/kg.

^cTotal number of eggs counted under the microscope for the group pre-treatment.

^dFecal Egg Count Reduction (%).

^e95% Credible Intervals.

For the pyrantel pamoate treatments, neither of the 4 b fulfilled the minimum group requirement of 6 horses for the clinical protocol, although 3 b exceeded the minimum eggs counted requirement of 240 (Table 2). Thus, results should be interpreted with caution. However, upper credible limits fell below the upper efficacy threshold of 98%, which does suggest evidence of reduced pyrantel efficacy.

3.2. Anoplocephalid data

Tapeworm fecal egg count ranges and the proportion of positive samples at each time point in the yearlings are presented in Table 3, while FECR calculations for anoplocephalid egg counts are presented in Table 4, along with the total number of eggs counted pre-treatment for each group. Given the limited group sizes and number of eggs counted, overall FECRs were also calculated across the 4 b for each drug.

Table 3.

Anoplocephalid egg counts pre and 14 days post anthelmintic treatment in yearlings treated with two different anthelmintic products.

	Praziquantela				Pyrantel pamoateb
	Pre-treatment		Post-treatment		Pre-treatment		Post-treatment
	Positivec	Ranged	Positivec	Ranged	Positivec	Ranged	Positivec	Ranged
Barn A	2/8	10–40	6/8	20–70	4/4	20–70	4/4	10–50
Barn B	3/4	10–40	4/4	30–60	4/4	30–60	4/4	10–40
Barn C	9/9	10–80	7/9	10–110	4/4	30–110	4/4	10–30
Barn D	3/3	10–40	2/3	30–40	2/2	30–40	2/2	10–20

^a1.5 mg/kg.

^b13.2 mg base/kg.

^cThe number of tapeworm positive horses relative to the number of horses tested.

^dAnoplocephalid fecal egg counts in eggs per gram of feces.

Table 4.

Observed efficacy of ivermectin and pyrantel pamoate against anoplocephalid parasites in the four yearling barns in the study.

	Praziquantela				Pyrantel pamoateb
	N	Eggsc	FECRd	CIe	N	Eggsc	FECRd	CIe
Barn A	2	5	34.0	[12.6–74.4]	4	19	46.8	[15.4–78.6]
Barn B	4	9	29.3	[11.8–61.9]	4	15	47.3	[15.0–79.9]
Barn C	9	29	34.1	[13.3–65.4]	4	22	56.6	[18.3–84.3]
Barn D	3	7	41.4	[14.4–81.2]	2	7	57.1	[17.1–90.2]
Overall	17	50	23.5	[11.2–48.0]	14	63	50.9	[20.5–72.0]

^a1.5 mg/kg.

^b13.2 mg base/kg.

^cTotal number of eggs counted under the microscope for the group pre-treatment.

^dFecal Egg Count Reduction (%).

^e95% Credible Intervals.

It should be noted that Barn A had five horses that were tapeworm negative pre praziquantel treatment but became positive post treatment. Due to the negative pre-treatment samples, these five horses were not included in the FECR calculation.

3.3. Mares

The FEC data from the mares are presented in Table 5. Given that pre-treatment counts were not determined with the standard two-week interval pre- and post-treatment, FECR calculations were not carried out. At Facility 1, 5 of 39 mares tested positive for anoplocephalid eggs, while 4 of 45 tested positive at Facility 2. One additional mare at Facility 1 tested positive in the post-treatment sample. Of the 9 tapeworm-positive mares, 7 remained positive post-treatment.

Table 5.

Pre- and post-treatment anoplocephalid and strongylid fecal egg counts in two populations of brood mares. All mares were treated with a combination of ivermectin (200 μg/kg) and praziquantel (1.5 mg/kg).

Mares	Anoplocephalid		Strongylid
	Pre-treatment	Post-treatment	Pre-treatment	Post-treatment
Facility 1a
A	10	0	2,640	190
B	10	10	950	0
C	30	20	610	20
D	10	10	1,030	0
E	0	20	640	0
F	10	0	1,660	0
Facility 2b
A	30	10	70	0
B	20	20	0	0
C	10	0	0	0
D	40	20	30	20

^aPre-treatment samples collected on March 3rd, 2023. Treatment on March 3rd, 2023. Post-treatment samples collected on March 28th, 2023.

^bPre-treatment samples collected on March 9th, 2023. Treatment on March 30th, 2023. Post-treatment samples collected on April 20th, 2023.

4. Discussion

The data presented herein are the first to suggest reduced anthelmintic efficacy against equine anoplocephalid tapeworms. Given the numerous reports this author has received over the past year, there are reasons to be concerned that this could be the first signs of another lurking anthelmintic resistance crisis. Since A. perfoliata is well-recognized as a significant equine pathogen (Nielsen, 2016), a loss of viable treatment options could have dire consequences.

It should be emphasized that FECR testing has not been standardized for equine tapeworms, and that the hierarchical Bayesian approach for determining FECRs and credible intervals is developed for strongylid egg counts and may well not be suitable for tapeworm data. Furthermore, it is well established that standard fecal egg counting techniques perform very poorly for detecting equine tapeworm infection (Meana et al., 2005; Tomczuk et al., 2014; Hreinsdóttir et al., 2019), and egg counts determined are typically very low. Given that the statistical power of the FECRT is dependent on the number of eggs counted pre-treatment, it is a real challenge to find sufficient egg count positive horses contributing enough eggs counted to make the test reliable and meaningful. As illustrated in Table 4, only 50 and 63 eggs were counted in the yearlings across the 4 b prior to the two anthelmintic treatments, which is in sharp contrast to the hundreds of strongylid eggs counted in the same horses (Table 2). This problem was even more pronounced with the mares (Table 5), where only 9 of 84 tested positive for tapeworms and only 17 eggs were counted in total pre-treatment. The WAAVP minimum requirements for the number of ascarid and strongylid eggs counted are typically over 100 except scenarios with very large group sizes (Kaplan et al., 2023), and statistical requirements for tapeworm eggs are likely to be broadly similar, if not higher. The egg counting technique employed in this study is used by a large equine practice in the area, but validation data are lacking, so it is not known how reliable it might be for tapeworm egg counting. This demonstrates very clearly that further development and refinement of fecal egg counting techniques for equine tapeworm eggs is warranted before we can develop a framework for appropriate treatment efficacy testing. Having said this, however, it should be acknowledged that several field efficacy trials for both praziquantel and pyrantel pamoate have demonstrated that tapeworm fecal egg counts can provide some indication of anti-cestode efficacy despite the above-described limitations. These studies made use of a variety of different techniques such as Wisconsin (Grubbs et al., 2003; Marchiondo et al., 2006; Slocombe et al., 2007), McMaster (Roelfstra et al., 2006), Proudman (Slocombe et al., 2007), as well as other egg counting techniques (Rehbein et al., 2003; Craig et al., 2003; Marchiondo et al., 2006), and each provided a useful indication of field efficacy despite counts being low. Thus, more field studies should be encouraged in different locations to monitor anti-cestodal treatment success.

Given the above-referenced historic field efficacy studies, which demonstrated near complete elimination of anoplocephalid eggs from equine fecal samples following administration of either praziquantel (Rehbein et al., 2003; Grubbs et al., 2003; Roelfstra et al., 2006; Slocombe et al., 2007; Lyons et al., 2017) or pyrantel pamoate (Craig et al., 2003; Marchiondo et al., 2006), the data presented herein strongly suggest a loss of efficacy for both actives. This was particularly pronounced for praziquantel, where some of the horses were negative pre-treatment and then became positive post-treatment. It is also noticeable that while the majority of tapeworm positive horses were still positive post praziquantel treatment, pyrantel pamoate failed to clear fecal samples from tapeworm eggs in every single case. Again, this contrasts the findings reported in the above-referenced field efficacy trials (Rehbein et al., 2003; Grubbs et al., 2003; Craig et al., 2003; Marchiondo et al., 2006; Roelfstra et al., 2006; Slocombe et al., 2007; Lyons et al., 2017). Although the data from the mares are sparse, they do indicate that the apparent loss of praziquantel efficacy is not limited to the yearling age group. It should be acknowledged that previous studies have demonstrated that tapeworm quantity in fecal samples increases during the first 24–48 h following anti-cestodal treatment (Slocombe 2004; Sanada et al., 2009; Elsener and Villeneuve, 2011), which is speculated to be due to the decomposition of dead worms releasing more eggs within the intestinal tract. However, these elevated egg counts did not extend into the second or third weeks post-treatment, when all samples were either negative or low positives (Slocombe 2004; Sanada et al., 2009; Elsener and Villeneuve, 2011). Given that post-treatment samples were collected and analyzed at two weeks post-treatment or later in this study, it is highly unlikely that this phenomenon contributed to the positive tapeworm egg counts observed following treatment. It should also be noted that serum and saliva ELISAs for detection of anti-A. perfoliata antibodies are available (Proudman and Trees, 1996; Lightbody et al., 2016), but that these are unlikely to be useful for treatment efficacy testing due to the relatively long half-life of the antibodies and the continued exposure to the parasites. A coproantigen ELISA was once developed and showed promise (Kania and Reinemeyer, 2005; Skotarek et al., 2010), but it remains unknown if it could be useful for treatment efficacy testing.

Not only is finding evidence of possible anti-cestodal resistance remarkable, it is even more noteworthy that apparent treatment failure was observed simultaneously to two different anthelmintic classes with different modes of action. There are several possible explanations for this observation and more data will undoubtedly help unravel these. But it is perfectly plausible that anthelmintic resistance could have been developing in equine tapeworms to both anthelmintic classes over the course of the past several years without being noticed due to an overall lack of fecal egg count monitoring. This could have been further complicated by the previously referenced diagnostic limitations of standard and widely used fecal egg count techniques for tapeworm monitoring (Meana et al., 2005; Tomczuk et al., 2014; Hreinsdóttir et al., 2019). Thus, it is possible that resistance could have finally reached a level where even techniques with moderate to poor diagnostic sensitivities return positive results. It appears unlikely that anthelmintic resistance would develop simultaneously and at the same pace to two different anthelmintic classes, although it cannot be ruled out that cross-resistance could occur between them. It seems more plausible that resistance could have developed at different paces over the course of several years, if not decades, before it finally was noticed at a stage where resistance had fully developed to both actives.

Beyond praziquantel and pyrantel pamoate, there are currently no alternative anti-cestodal anthelmintic classes marketed for equines in Europe or North America. Thus, this and other equine operations experiencing similar signs of lack of efficacy do not have viable treatment alternatives, and it is unknown if the pharmaceutical industry may be working to develop new compounds with anti-cestodal activity for equines. One product, a paste formulation of bithionol, appears to have been licensed and marketed for equines in Japan with documented efficacy against A. perfoliata (Toguchi et al., 2004; Sanada et al., 2009). If this product is still in production, it could be worth exploring in other parts of the world. One study carried out in the US evaluated the field efficacy of nitazoxanide in horses and suggested good anti-cestodal efficacy of this compound as well (Craig et al., 2003), so this may represent another possible treatment alternative in the future, although proper efficacy and safety testing would be required. Similarly, early work indicated some activity of niclosamide (Slocombe, 1979) and closantel (Guerrero et al., 1983) against A. perfoliata, and these could be worth exploring as well.

Although it was not a primary aim, this study also provided information on anti-strongylid efficacy of ivermectin and pyrantel pamoate in the yearlings. The ivermectin efficacy estimates indicated clear evidence of resistance with mean FECRs in the 30–70% range and upper credible limits below 85% for all 4 b (Table 2). This is in agreement with previously reported ivermectin efficacy levels in yearlings in this area (Nielsen et al., 2020, 2022) and confirms global trends of strongylid resistance to this drug class (Felippelli et al., 2015; Flores et al., 2020; Abbas et al., 2021; Bull et al., 2023). Although group sizes fell below WAAVP requirements for pyrantel pamoate FECRT testing, the data clearly suggested anthelmintic resistance to this drug class as well with mean FECRs in the 38–75% range and all upper credible limits below 90% (Table 2). It should be noted that these low efficacy estimates were observed despite the dosage of pyrantel pamoate being twice the labelled dosage for strongylid treatment, which demonstrates that there are no benefits to doubling the dose, once resistance is fully developed.

The findings reported herein raise several immediate questions and research needs. The anti-cestodal activity of praziquantel and pyrantel pamoate should be closely monitored in equine establishments around the world. This, in turn, raises a need for developing a standardized approach for FECRT testing of equine tapeworms, which includes identifying a suitable fecal egg counting technique and developing an evidence-based statistical approach for analyzing and interpreting the data. Only when this has been established can we reliably compare results between studies and determine if the efficacy has reduced from original levels. Certainly, if similar findings of apparent reduced anti-cestodal efficacy are subsequently made in other parts of the world, there will be an urgent need to develop and test new anti-cestodal products for use in equines.

In summary, this study represents one example of several observations of an apparent loss of anti-cestodal efficacy of praziquantel and pyrantel pamoate made over the course of 2022 and the early part of 2023. Given the lack of FECRT standards for equine tapeworms and limitations of available fecal egg counting techniques, the data presented herein do not allow the conclusion of anthelmintic resistance at this stage. Nonetheless, the apparent lack of efficacy of both praziquantel and pyrantel pamoate is deeply concerning and warrants further research.

Declaration of competing interest

The author declares no conflicts of interest.