|Home | About | Journals | Submit | Contact Us | Français|
Sarcoids are the most common tumor of the equine skin but only 1 study describing the epidemiology of sarcoids in Canadian horses has been published. The records of 5 veterinary diagnostic laboratories in western Canada were searched to identify submissions of sarcoids from horses. The submission records and diagnostic reports of 802 separate submissions of equine sarcoids were reviewed for age, breed, and gender of the horse and the number, location, and clinical type of sarcoid. From these records, the 307 submissions to laboratories in Saskatchewan were compared to a reference group to test for breed and gender predisposition. Based on clinical history and lesion descriptions, 5 clinical types of sarcoids were identified. Horses of various ages and 23 equine breeds were affected; donkeys were over-represented. Polymerase chain reaction (PCR) for the bovine papillomavirus (BPV) DNA was performed on formalin-fixed paraffin-embedded tissues from a stratified subset of 96 of the different clinical types; BPV2 was present in 60 of 74 (81%) for which a PCR product was obtained. Unlike other areas in the world, in western Canada, equine sarcoids are most commonly associated with BPV type 2.
Épidémiologie des sarcoïdes équins chez des chevaux de l’Ouest canadien. Les sarcoïdes représentent le type de tumeur de la peau équine le plus fréquent mais seulement 1 étude décrivant l’épidémiologie des sarcoïdes chez les chevaux canadiens a été publiée. Une recherche a été réalisée dans les dossiers de 5 laboratoires diagnostiques vétérinaires de l’Ouest canadien afin d’identifier les soumissions de sarcoïdes provenant de chevaux. Les dossiers et les rapports diagnostiques de 802 soumissions séparées de sarcoïdes équins ont été examinés pour déterminer l’âge, la race et le sexe du cheval ainsi que le nombre, l’emplacement et le type clinique de sarcoïdes. À partir de ces dossiers, les 307 soumissions à des laboratoires en Saskatchewan ont été comparées à un groupe témoin pour réaliser des tests sur la race et la prédisposition des sexes. En se fondant sur les antécédents cliniques et les descriptions, 5 types cliniques ont été identifiés. Des chevaux d’âges divers et de 23 races équines ont été affectés; les ânes étaient surreprésentés. La réaction d’amplification en chaîne par la polymérase (ACP) pour l’ADN du papillomavirus bovin (BPV) a été effectuée sur des tissus fixés dans du formol et enrobés dans de la paraffine provenant d’un sous-groupe stratifié de 96 des différents types cliniques; le BPV2 était présent dans 60 de 74 (81 %) des échantillons pour lesquels un produit d’ACP a été obtenu. Contrairement aux autres régions du monde, dans l’Ouest canadien, les sarcoïdes équins sont le plus communément associés avec le BPV de type 2.
(Traduit par Isabelle Vallières)
Equine sarcoids are the most common skin tumor of horses and are found in horses worldwide. A wide variety of other equids including zebras, donkeys, and mules are affected (1–5). The term “sarcoid” was first used to describe these tumors in 1936, in South Africa, to distinguish them from other fibroblastic skin tumors such as fibromas, fibrosarcomas, and papillomas (1). These tumors do not metastasize, but they can be invasive, are regarded as cosmetic defects, can become ulcerated and infected and, when occurring near the eyes or on the eyelids, can impair vision. The treatment of horses for equine sarcoids can represent a considerable expense for horse owners (6).
Descriptive epidemiologic features of equine sarcoids (age, breed, clinical type, location of tumor, presence of multiple tumors) have not been well-defined in Canadian horses and, to the best of our knowledge, only a single study has been published (7). Regional estimates of prevalence based on records from diagnostic centers in Europe and the United States vary considerably (8–10). The age at diagnosis in published studies varies, with 1 study suggesting that sarcoids rarely develop in animals older than 7 y (11), while a second study, from the Pacific Northwest of the United States, determined that the mean age at diagnosis was 9 y (10).
Equine sarcoids are associated with the presence of 2 types of bovine papillomaviruses (BPV), BPV-type 1 (BPV1) and BPV-type 2 (BPV2) (12). Much of the work on the molecular pathogenesis of sarcoids has been conducted in Europe and performed exclusively on animals from which BPV1 was recovered from lesions. BPV2 is reportedly much more common in sarcoids from horses in the western United States (13). It is unknown if the previous results of research on BPV1-associated sarcoids apply to those associated with BPV2 (13).
The objective of the following studies was to address the knowledge gap in the descriptive epidemiology of equine sarcoids in horses in western Canada. Submission records and diagnostic reports from multiple veterinary diagnostic laboratories in western Canada were used to describe the epidemiology of equine sarcoids. From these records, a subset of cases that had been submitted to laboratories in Saskatchewan were compared to a reference group to test for breed and gender predisposition. A second subset of the most recently submitted cases, stratified by province of origin, and representing different clinical types of sarcoids, was selected and the associated formalin-fixed paraffin-embedded tissues utilized. Polymerase chain reaction (PCR) was performed on this subset to detect the presence of BPV, and the type of BPV was identified by reverse fragment length polymorphism (RFLP) on PCR positive samples. Confirmation that the PCR product was either BPV1 or BPV2 was determined by nucleotide sequencing on 4 samples.
Surgical biopsies of equine sarcoids diagnosed during the 12-year period between January 1, 1996 and December 31, 2007, inclusive, were identified by computer-based record searches at 5 veterinary diagnostic laboratories in western Canada: IDEXX Reference Laboratories Ltd., Langley, British Columbia (formerly Central Laboratory for Veterinarians), and their associated laboratories in Edmonton and Calgary, Alberta; the Department of Veterinary Pathology at the University of Saskatchewan; and Prairie Diagnostic Services (PDS) in Regina and Saskatoon, Saskatchewan. The submission records and diagnostic reports were obtained for each case and from selected cases, the formalin-fixed paraffin embedded tissues were retrieved for molecular-based testing.
Submission records and diagnostic reports for these biopsies were used to determine breed, gender, age at the time of submission, number of sarcoids on the animal, location of the sarcoids on the animal and in which province the animal was located. The submitted history, gross appearence and information in the biopsy description were used to determine the clinical type of sarcoid. Sarcoids were classified using the clinical classification scheme proposed by Knottenbelt (14) that consisted of: fibroblastic, nodular, occult, verrucose, mixed, and malevolent.
Data on all equine skin biopsies not diagnosed as sarcoids within the same time period were collected from the records of the Department of Veterinary Pathology at the University of Saskatchewan, and PDS in Regina and Saskatoon, Saskatchewan. These served as the reference population for identifying potential risk factors associated with the development of sarcoids. Only those sarcoids identified at these same diagnostic laboratories were used for comparisons.
Gender (gelding, mare, stallion), age (in years), and breed were the risk factors considered for associations with the development of sarcoids. Comparisons of median age of horses with different clinical classifications of sarcoids in this study were performed using Kruskal-Wallis equality of populations rank test and Dunn’s multiple comparison test. Gender and breed information on horses diagnosed with 1 or more sarcoids was compared with this same information on horses with other skin conditions diagnosed by Saskatchewan veterinary diagnostic laboratories and the reference population was evaluated for these risk factors using a chi-squared (χ 2) goodness of fit test. Relative risk of the occurrence of sarcoid in different breeds and genders was based on comparisons with the occurrence in the breed or gender with the lowest risk of disease. Significance was set at P < 0.05. Analysis was performed with the aid of a statistical software package (Stata/IC 10.1 for Windows; StataCorp LP, College Station, Texas, USA).
For each clinical type of sarcoid, the 20 most recently submitted biopsies from horses, stratified by province of origin, were selected from the submissions to the various veterinary diagnostic centers. The formalin-fixed paraffin-embedded tissue blocks from these samples were used for PCR as previously described (12). The single primer set that amplifies a 244 base pair (bp) sequence spanning portions of the E2 gene, the intergenic spacer region and a small portion of the E5 gene of BPV1, or a similarly located 248 bp sequence of BPV2, was used. Previous studies using this set of primers have identified the amplified sequence as the E5- open reading frame (13,15,16), but comparisons to BPV sequence data in GenBank (National Center for Biotechnology Information, National Institutes of Health, USA) revealed that the above description of the sequence is more accurate.
Following amplification, RFLP (12) was used to type the BPV in the PCR product. The BPV1 amplicon was cleaved by BstXI into 130 bp and 114 bp fragments. The BPV2 amplicon was cleaved by Hinf I into 129 bp and 119 bp fragments. RFLP was performed in parallel with the 2 restriction endonucleases. To confirm that the PCR was amplifying BPV, 4 samples of PCR product, 2 of each identified as BPV1 and BPV2 were sequenced.
Equine sarcoids (n = 802) were submitted to the veterinary diagnostic laboratories during the 12-year period. Samples were from horses resident in Alberta (n = 231), British Columbia (n = 378), and Saskatchewan (n = 193). IDEXX laboratories processed 495 samples and 307 samples were seen by diagnostic laboratories in Saskatchewan. All laboratories received sarcoids from horses in each of the 3 provinces. Where the number of sarcoids on the animal could be determined, 204 of 722 (28.3%) animals had multiple sarcoids at the time of submission.
Five clinical types of sarcoids were present in the submitted biopsy samples; the malevolent type of sarcoid was not present in any samples. The most common sarcoid classification was the fibroblastic type (Table 1).
Data on the age of horses with each of the 5 clinical types of sarcoids present in this sample are summarized in Table 1. Median ages ranged from 5 y for fibroblastic sarcoids to 7 y for occult, nodular, and mixed sarcoids. The median age of horses with fibroblastic sarcoids was significantly less than for horses with mixed, nodular, and occult types of sarcoids.
There was a wide range of ages in horses diagnosed with sarcoids. The mean age of affected horses was 6.8 y and ages ranged from 0.5 to 31 y. Reference population horses that did not have sarcoids had a mean age of 8.9 y and an age range of 1 d to 30 y. Horses with sarcoids in the study group were significantly younger than horses whose skin was biopsied for other reasons (P < 0.01).
The location of 746 sarcoids on the bodies of 686 horses was determined (Table 2). The most common location from which sarcoids were biopsied was the head with 307, followed by the limbs (149), neck and shoulder (116), abdomen (81), axilla and chest (59), and the paragenital region (34). Some horses (38) had sarcoids in more than 1 of these locations.
The reference population comprised 443 horses with non-sarcoid skin conditions diagnosed by Saskatchewan veterinary diagnostic laboratories over the same time period. The 307 sarcoids diagnosed in these laboratories during this 12-year period were used as a study group. Sarcoids represented 41.9% of all equine skin submissions seen in these laboratories for this period of time.
Twenty-three breeds of horses were affected. The most commonly identified breeds were Quarter horses and Thoroughbreds (Table 3). Breed relative risks for the presence of sarcoids in horses diagnosed in Saskatchewan veterinary laboratories as compared to reference population horses were determined (Table 3). The breed with the lowest risk of development of sarcoids was Warmbloods. Compared to the risk of development of sarcoids in Warmbloods, donkeys were the only group that had a significantly elevated risk (3.1× more likely to develop, P = 0.01).
Geldings, mares, and stallions were all affected with the most commonly affected gender being geldings (Table 3). No significant difference in the risk for the development of sarcoids was identified with any of these groups.
Polymerase chain reaction for amplification of BPV was performed on 96 biopsies that were selected from veterinary diagnostic laboratories across western Canada and represented each of the 5 clinical types seen (Table 4). Bovine papillomavirus DNA was amplified from 74 of 96 biopsies (77%). Reverse fragment length polymorphism on the PCR product from all BPV positive cases resulted in 14 of 74 BPV1 positive (19%), 59 of 74 BPV2 positive (80%), and 1 positive for both BPV1 and BPV2 (1%).
Nucleotide sequences from 2 BPV1-positive and 2 BPV2-positive samples were compared to sequences from other papillomaviruses using the Basic Local Alignment Search Tool (BLAST) from GenBank (17). Multiple sequence alignments were performed (EMMA multiple alignment program, Emboss Version 5.0.0; http://haruspex.usask.ca/emboss/). The 2 BPV1 positive samples (GenBank accession numbers FJ895875, FJ895876) showed > 98% identity with the previously described British II BPV1 sequence recovered from equine sarcoid in the United Kingdom (accession number AY232263.1) (18). Bovine papillomavirus type 2 PCR products from sarcoids (accession numbers FJ895874 and FJ895877) showed 100% and over 98% identity with 3 nucleotide variations, respectively, with the Swiss V sample found in equine sarcoids from Switzerland (accession number AY232264.1) (18).
Sarcoids on horses have been described since historical times (19) and have been recognized as a clinical entity since 1936 (1). Although there has been abundant research on sarcoids in Europe and the United States there has not been, until now, an attempt to describe the epidemiology of sarcoids in horses in western Canada. In this study, 802 horses with sarcoids were identified based on biopsy submission records to multiple veterinary diagnostic laboratories in western Canada.
As sarcoids are the most common skin tumor of horses (10,20) it is not surprising that horses with sarcoids were present in large numbers in each of the 3 provinces. Submissions from the largest number of horses were from British Columbia (378), followed by Alberta (231), and Saskatchewan (193). According to a 2003 report on horse numbers in Canada, Alberta had the largest number of horses with 299 753, British Columbia had 141 410, and Saskatchewan had 126 252 (21). While results from this study may suggest that sarcoids are more common on a per animal basis in British Columbia, the data were derived from submission of biopsies and the results may be biased based on the willingness of owners to spend money on veterinary diagnostic services. The same report states that owners in British Columbia spend an average of $379 annually on veterinary care per horse compared with $255 in Alberta; this may explain the apparent difference in frequency of diagnosis per capita (21).
As noted in other studies on sarcoids, affected horses frequently have multiple sarcoids at the time of biopsy (22,23). In this study, horses with multiple sarcoids at the time of biopsy represented 28.7% of the submissions and sarcoids composed 41.9% of all biopsies involving equine skin at Saskatchewan veterinary diagnostic laboratories, making them the most frequently diagnosed skin condition.
Of the 6 clinical classifications of sarcoids suggested by Knottenbelt (14), 5 were found (Table 1). No malevolent sarcoids were found in this study and this clinical type has only been reported by 1 group (24). All other clinical types were present in relatively equal proportions with the exception of the fibroblastic type which was present at approximately twice the frequency of other types. While this may represent the actual clinical distribution of cases, it may also represent the greater frequency with which these larger and more aggressive tumors are likely to receive diagnostic attention (20).
Most sarcoids occurred on the head, followed by the limbs, and the neck and shoulder (Table 2). This distribution differs from those in other studies which found the lower abdomen to be the most common region for sarcoids to develop (22,25,26), but is similar to that of another study (27). Interestingly, all studies in which the abdomen was the most frequent location of sarcoid development were European, while the study with findings similar to the current study was from the Pacific Northwest of the United States. The variability in sites may be due to differences in the use of horses in different areas. It has been postulated that horses in North America are more likely to injure their legs on rough ground than are those in Europe (25), but this remains speculative and has not been scientifically examined.
Twenty-three equine breeds were diagnosed with sarcoids in this study. Several studies have examined relative risk for the development of sarcoids among breeds (8,9,28). Two of these studies found that among horses diagnosed with sarcoids in New York State, Quarter horses were more than twice as likely to develop sarcoids compared with Thoroughbreds, which in turn were twice as likely to develop sarcoids as Standardbreds (8,9). In one of these studies, Arabians were identified as the breed with the highest risk of sarcoid development, more than 3 times higher than Thoroughbreds (9). This breed disposition for development of sarcoids was not present in the current study. Only donkeys were at a statistically significantly higher risk to be diagnosed with a sarcoid (Table 3). The increased risk of donkeys developing sarcoids has not been previously identified. Warmbloods, Standardbreds, and Arabians were among the breeds at the lowest risk of developing sarcoids in the current study; however, the risk reduction was not statistically significant in any of these breeds. This difference in results may represent a variation based on the geographic location and genetic heritage of the horses involved, but a more likely explanation may be the reference populations used to make these comparisons. The reference population of the earlier study was comprised of horses with corneal ulcers, indigestion, carpal bone chip fractures, and tendonitis (8), while the current study’s reference population was comprised of horses with skin conditions other than sarcoids.
More geldings than mares or stallions were diagnosed with sarcoids in this study; however, no differences in relative risk were noted (Table 3). In previous studies where a gender predilection has been shown, geldings were found to be at an increased risk compared to stallions and mares (5,9). In these studies, it was suggested that castration surgery is a risk factor for the development of sarcoids, but the lack of a gender predilection and the relative rarity of paragenital location of sarcoids in this study would not support this conclusion.
Sarcoids have been reported to be a disease of young horses (20,25), and the mean age of horses with sarcoids in the study group was significantly younger than those of the reference population, but affected animals ranged from 6 mo to 31 y old. Of particular note, is that unlike previous reports which suggested that sarcoids rarely develop in horses older than 7 y (22,25,29,30), the median age of all horses in the current study with several clinical types of sarcoids (nodular, occult, and mixed) was 7 y, suggesting that half of the horses with this diagnosis were this age or older at the time of diagnosis. Of interest, is that the fibroblastic type classification, the most aggressive clinical type, was also the type with the youngest median age, statistically significantly younger than other types (Table 1). It has been suggested that sarcoids develop initially as one of the less aggressive types such as occult or verrucose sarcoids and then transform into the more aggressive fibroblastic form (24,25,31). If this were the case, it seems counterintuitive that the most aggressive form of sarcoids is seen more often in younger horses than are the less aggressive forms from which it may derive.
Bovine papillomavirus has been reported as the cause of equine sarcoids (12,32,33) and in this study BPV DNA was amplified from 77% of 96 samples. The amplification rates of BPV DNA from sarcoids in various studies has ranged from 73% to 100%, with studies with lower rates using formalin-fixed paraffin-embedded samples as in the current study (12,13,34–36). Confirmation that BPV DNA was being recovered was obtained by sequencing 4 randomly chosen PCR products. The significance of the slight nucleotide change in one sequence is uncertain, but nucleotide variation in other regions within these genes has been previously found in equine sarcoids (18).
Bovine papillomavirus was recovered from all clinical types of sarcoids examined (Table 4). The lowest recovery rates were in occult sarcoids with 12 of 19 positive (63%). Whether this represents lower levels of BPV in these types of sarcoids or some of these biopsies were not truly sarcoids is unclear. Occult sarcoids have the most subtle changes histologically of any of the sarcoid types and also have the smallest lesions by volume, often confined to a small proliferation of fibroblasts beneath the skin surface. It is possible that some of these lesions were misdiagnosed as sarcoids or that the smaller volume of neoplastic tissue in occult sarcoid as compared to other types of sarcoids reduced the success of DNA amplification resulting in a lower BPV positivity rate in this type of sarcoid.
Most sarcoids with amplified BPV DNA in this study contained solely BPV2 DNA based on RFLP (80% of the total). Bovine papillomavirus type 1 was found as the only amplified BPV type in 19% while a single sample had both BPV1 and BPV2 amplified. This is dramatically different from most studies of sarcoids. In Europe, BPV1 accounts for a vast majority of DNA detected with only a small number of sarcoids containing BPV2 (12,32,37). Studies in the United States have found BPV1 and BPV2 in roughly equal proportions in the eastern United States (12), while BPV2 predominates in the western United States representing 63% of the amplified DNA (13). No previous studies have found as high a rate of BPV2 infection as this current study. Most of the work conducted on equine sarcoids was performed on horses infected with BPV1 rather than BPV2, the type much more commonly found in this study in western Canada. No clinical or pathogenetic differences between infection with BPV1 and BPV2 have been reported, and the applicability of studies performed on horses with BPV1 to horses with BPV2 is not known (13).
In summary, the study herein is the largest study of equine sarcoids focused on western Canadian horses. Sarcoids appear to be a common lesion of the skin of horses in western Canada. There are several noteworthy features identified in this study. Specifically, donkeys appear to be at an increased risk of sarcoid development, older horses are more commonly affected by sarcoids than some previous reports suggest and, perhaps most importantly, BPV2, not BPV1 is the predominant viral type detected in sarcoids in these horses. More research is needed to determine the importance of BPV2 as a cause of sarcoids and to determine if there are different clinical implications in sarcoids caused by BPV2 versus those caused by BPV1 or differences in the pathogenesis of these lesions. CVJ
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (gro.vmca-amvc@nothguorbh) for additional copies or permission to use this material elsewhere.
This research was supported by funds from the Western College of Veterinary Medicine Equine Health Research Fund and personal support for Dr. Wobeser was provided by a Western College of Veterinary Medicine Interprovincial Graduate Student Fellowship.