Introduction

Canine transmissible venereal tumor (CTVT), also known as infectious lymphosarcoma venereum lymphosarcoma, myxosarcoma, or Sticker’s sarcoma (^{Moulton, 1990}), venereum granuloma, condyloma, or infectious sarcoma (^{Mandewell and Theilen, 1987}), is a transmissible cell neoplasm that affects canines worldwide and is more common in tropical and subtropical areas, particularly urban areas with large populations of stray dogs (^{Mialot, 1984}). It is the oldest known cancer in dogs, originating from a wolf or a dog in East Asia more than 10,000 years ago; nevertheless, the exact date of CTVT appearance is still debated and is being estimated to have occurred between 6,000 and 11,000 years ago (^{Murgia et al., 2006}; ^{Murchison et al., 2014}; ^{Decker et al., 2015}; ^{Ostrander et al., 2016}). Over time and due to a downregulated expression of the major histocompatibility complex, it was transmitted to other dogs (Murgia et al., 2006) and no naturally occurring CTVT cases have been reported in wild canids (^{Strakova and Murchison 2014}). CTVT is transmitted naturally through sexual intercourse and usually affects the genital system in sexually mature animals. However, when the neoplasia is licked or sniffed, it can be transplanted to the skin or nasal mucosa (extragenitally) and, in some severe cases, may metastasize to internal organs ^{(Moulton, 1990}).

Strakova and ^{Murchison (2014)} determined that CTVT is present in at least 90 countries and widely distributed throughout the world. In Colombia, despite of the large number of cases routinely observed, CTVT appears to be underdiagnosed due to the limited literature available on the subject. ^{Sánchez et al. (2014)} reported that the disease had 2.7% prevalence in Boyacá, with genital presentation in all cases. On the other hand, 8.6% cancers diagnosed in dogs at Universidad de los the Llanos corresponded to CTVT, with most dogs affected from urban areas (^{Bravo et al., 2010}). Universidad de la Amazonia reported one case and, although the epidemiologic situation is unclear, its presence is possibly related to stray urban animals (Ramirez et al., 2015). A case of extragenital CTVT presentation was reported by ^{Martinez et al. (2002)} in Monteria in a female dog with a skin lesion and compromised ipsilateral inguinal lymph node. Similarly, ^{Salamanca et al. (2008)} reported a case of pulmonary metastases of a CTVT in Villavicencio.

Review studies are important for improving our understanding of certain diseases. Catanzaro et al. (2016), for example, determined the attitudes, opinions, and knowledge of Italian veterinarians on abdominal visceral pain in canine practice, whereas; ^{Wang et al. (2014)} used a cohort study to determine the health of urban populations, identifying risk factors for cardiovascular disease in Chinese workers. Likewise, ^{Hang et al. (2015)} used online forms and cytological images to determine the degree of knowledge of practitioners on cytological diagnosis and define the accuracy of the diagnoses. Descriptive and analytical clinical epidemiology studies can help identifying disease distribution and presentation, laying the groundwork for research into ways to control or eradicate diseases.

In Colombia, there is limited research on the biological and epidemiological behavior of CTVT. This information is required to establish the endemic nature of the disease and therapeutic possibilities for control. In Brazil, where some degree of resistance to treatment with vincristine sulfate has been reported, resistant CTVT has become a major problem for clinical management, even leading to animal euthanasia. Thus, the aim of the present study was to provide epidemiological and clinical information on CTVT, as well as its distribution and relative prevalence in different regions of Colombia. We also want, to inform veterinarians about the potential risk of resistant CTVT strains in our country and generate hypotheses for future research aimed at understanding the biological behavior of this entity.

Materials and Methods

A cross-sectional study was used to investigate the epidemiological and clinical characteristics of CTVT in Colombia. Using Google Forms®, 18 questions were sent to veterinarians to gather information on variables of clinical interest, such as the degree of knowledge on the disease, diagnostic test, therapies used, treatment resistance, and concomitant diseases. Questions also involved epidemiological information such as gender of affected animals, condition of free-roaming dogs, and whether the dogs were urban or rural. For geographic location, we used official information of the Colombian National Administrative Department of Statistics (DANE), which divides the country into five bioregions. Respondents were selected via Internet search and by contacting the main veterinarian associations in the country. With non-probability sampling, the questionnaire was sent by email to 950 veterinarians, but only 176 forms were completed in the following 4 weeks. Only 152 met the criteria for inclusion in the study, namely, at least 16 complete responses. Data were collected in Excel ® format for easiness of handling.

To determine CTVT prevalence in each region and the human/canine ratio, we used the data reported by respondents and the official DANE data on the projected population density for 2015 (DANE, 2005). Because of lack of data on the dog population in Colombia and taking into account studies conducted in Guatemala, Italy, Ireland, Venezuela, Argentina, and Brazil, we calculated an average rate of 4.6 people per canine (^{Medina et al., 2002}; ^{Alves et al., 2005}; ^{Slater et al., 2008}; ^{Downes et al., 2009}; ^{Zumpano et al., 2011}; ^{Pulczer et al., 2013}). Data on the calculation of canine population is shown in Table 1.

The results are shown in terms of ratios and proportions to reveal the trend in each questionnaire response. In addition, the average number of CTVT cases was statistically compared using univariate ANOVA, with the number of cases as the response variable and several others as factor variables (region, type of organization, clinical support, regression, and treatment). In case of significant differences, the Tukey test was applied to establish the difference. Subsequently, the different regions were compared by analysis of automatic classification (cluster analysis). The XLSTAT 2014 software was used for one- and two-dimensional analyses. SPAD 8.0 software was useful for cluster analysis by region.

Results

Of the 152 eligible surveys, 128 (84.2%) corresponded to private clinics, 15 (9.9%) from Veterinary Medicine schools, 6 (3.9%) from veterinary diagnostic laboratories, and only 3 (2.0%) from foundations dedicated to animal welfare.

Most answers (131) came from the Andean region, whereas 10 forms were obtained from the Caribbean region, 5 from the Amazon, 5 from the Pacific, and 2 from Orinoco. The estimated number of CTVT cases encountered by veterinarians was 1,135. Most occurred in the Andean region with 862 cases (75.9%), followed by the Pacific with 110 cases (9.7%), the Amazon with 79 cases (7.0%), the Caribbean with 64 cases (5.6%), and Orinoco with 20 cases (1.8%) (Figure 1).

Epidemiology of Colombian CTVT presentation

In relation to sex, almost half of the respondents (44.8%) said there was no difference between the number of males and females. However, 35.5% of the veterinarians believed that males were more affected, while 19.7% believed that females were more affected. We found significant differences about the perception of the gender of affected animals according to the veterinarians (p=0.002). Regarding the reproductive status of the animals, 28 of the veterinarians surveyed (18.4%) answered that most of their treated animals were neutered and the remaining 124 (81.6%) reported that these animals were intact.

Figure 1 Map of estimated CTVT prevalence in different regions of Colombia. The greatest number of cases coincided with the largest number of veterinarians surveyed (in the Andean region), as well as with the largest human population in the country. 

Regarding stray dogs in each area, 69.7% of respondents reported the presence of these animals. According to the respondents, most CTVT cases were found in urban areas (75.7%). Similarly, 11.8% of respondents said the cases came from rural areas, 9.2% from semi-rural areas, and 3.3% did not respond.

Clinical behavior of CTVT in Colombia

CTVT cases were reportedly accompanied by parasites; ectoparasite infestation was the most frequent, including fleas, ticks, and lice. Presence of Demodex (canis and folliculorum) infestation was also noted. The respondents also reported the presence of blood parasites such as Trypanosoma sp., Babesia sp., and Ehrlichia canis. Regarding gastrointestinal parasites, the veterinarians reported the presence of Giardia, Ancylostoma caninum, Dipylidium caninum, Toxocara canis, and Coccidia (mainly Isospora sp). Internal or external parasites or both were reported in 32.0, 18.0, and 12.0% of animals, respectively. On the other hand, only two vets said they had seen anemia and thrombocytopenia; other two saw cases accompanied by emaciation and malnutrition associated with abandonment conditions. One respondent said CTVT animals presented some degree of immunosuppression.

Diagnosis tests used for CTVT in Colombia

To diagnose CTVT, a high percentage of respondents (47.4%) said that they used cytology, followed by clinical observation and histopathology at 26.3 and 13.1%, respectively. A small proportion of veterinary surgeons (12.5%) combined clinical diagnosis with cytology and histopathology, whereas 0.7% did not answer this question.

Biological behavior of CTVT in Colombia

An important aspect of the biological behavior of CTVT is the ability of the immune system to generate spontaneous tumor regression. However, most respondents said that they had not seen this type of behavior in any of the animals that they had attended. Additionally, no reported cases of distant metastasis were seen.

With regards to treatment, 91 of the veterinarians (59.9%) said they used vincristine as the only means of treatment, 32 (21.1%) used a combination of vincristine and surgery, and 27 (17.8%) combined more than three procedures among chemotherapies, homeopathy, surgery, and a combination of chemotherapy-immunotherapy. One respondent (0.7%) used surgery alone as treatment while another (0.7%) used chemotherapy with doxorubicin.

Veterinarians who used vincristine (alone or combined with surgery) needed a minimum of two doses and a maximum of eight. In the statistical analysis, a significant difference was found (p<0.001) between the proportion of vets that used vincristine as the only treatment and those who used it in combination with surgery.

Discussion

Although CTVT cases occurred throughout the country, the estimated prevalence in this study did not exceed 7.05% per annum, similar to results obtained of relative diagnostic percent (RDP) in other Colombian studies: at Universidad de Antioquia in Medellín, CTVT comprised 4.54% of all neoplasms in dogs (^{Ferreira and Pedraza 2003}); in Villavicencio, 8.6% of neoplasms in dogs corresponded to this transmissible tumor (^{Bravo et al., 2010}); and in Boyacá, 2.7% of street canines had this tumor (^{Sánchez et al., 2014}). The staff at Universidad de Caldas in Manizales determined 40 animals were affected by CTVT in 2014, corresponding to 1% of all dogs treated at the veterinary hospital and 9.2% of the recorded cases was associated with neoplasms (unpublished data).

According to the Professional Council of Veterinary Medicine in Colombia (COMVEZCOL), 70% of Colombian veterinarians are working in the Andean region, which explains why most surveys originated from this region. Similarly, the regions with the fewest veterinaries (Amazon, Pacific, and Orinoco) showed the lowest number of completed surveys. The highest estimated prevalence of CTVT was found in the Andean region, whereas the Amazon and Pacific regions proportionately had the highest number of cases because the few veterinarians who responded reported they had seen many CTVT cases. This large number of cases may be explained by the fact that they are remote regions, with few veterinarians, without access to medicines (vincristine), and a high population of wandering dogs that keep the disease circulating in the environment.

Significant differences were observed regarding sex of affected animals, with imbalance stating that the disease was more prevalent in males. However, it could be that the disease is easier to diagnose in males than in females, or at least the recovery or lack thereof is easier to verify in males, resulting in fewer females received chemotherapy. Analysis of the presence of stray dogs revealed an apparently high proportion of these animals in Colombia. Given that a high proportion of these dogs are affected by CTVT (^{Strakova and Murchison, 2014}), it is necessary to decide whether CTVT should be included in stray dog control, taking into account that most of the affected animals live in urban areas. Regarding reproductive status, it is known that intercourse is the typical infection route. Nevertheless, there were castrated animals affected by CTVT. It is possible that their lower immunity, which characterizes stray animals due to malnutrition, concomitant diseases, and stress, allows greater susceptibility to the tumor, or they might have been affected before neutering and the tumor was “dormant” for more than one year. In addition, a number of respondents observed ecto and endoparasites in animals affected by CTVT and, as reported by ^{Strakova and Murchison (2014)}, this condition could be related to abandonment or income, because the poorest regions reported the largest number of cases.

On the other hand, clinical observation is still widely used to diagnose CTVT. However, cytology has gained importance as a method for disease diagnosis, as reported by most respondents. The understanding of the techniques available for diagnosis is constantly improving. In addition, the easy access and low cost of reagents make cytology a useful tool for CTVT diagnosis. No differences have been reported in the phenotype of neoplastic cells (CTVT lymphocytoid and plasmacytoid), as suggested by ^{Florez et al., (2012)}.

Regarding biological behavior of CTVT, Colombian veterinarians have not seen spontaneous regression of the tumor. However, considering that the main reason for consultation is genital bleeding, veterinarians should set up an actual treatment regimen at the very least. The owner would most almost certainly expect this. Placing the animal under observation without treatment for several weeks to determine whether spontaneous regression of the lesion occurs is not the appropriate course of action for owners who merely want relief for their pet. In the case of wandering canines, improvement in living conditions (food, deworming, tranquility) could be enough to allow their immune system to trigger spontaneous regression of the lesion. However, only clinical trials can determine what happens in this regard. Some veterinarians suggest that spontaneous regression of lesions could explain the absence of sick animals in the same geographical area where the animals with CTVT were treated. However, there are no reports of such regression.

A high percentage of respondents depend on vincristine (alone or in combination) for treating CTVT. In Brazil, there are reports that cancer cells have acquired resistance because of chemotherapy mismanagement, either due to lower than necessary frequency of administration or administration of smaller than recommended amounts of chemotherapy drugs. For complete tumor elimination, a weekly dose of vincristine of 0.5 to 0.75 mg/m² body surface area is required for as long as 4 to 8 weeks. However, one or two doses limit tumor growth and eliminate the main clinical sign (genital bleeding), which could mistakenly be interpreted as a “cure” and ultimately leave the animal without a complete treatment course. In addition, the use of lower than therapeutic doses can generate chemotherapeutic agent resistance (^{Sousa et al., 2000}, ^{Montoya et al., 2014}). Similarly, there have been reports of the coincidence of resistance with plasmacytoid cytomorphology, leading to hypotheses about cell transformation triggering resistance to vincristine (^{Eze et al., 2013}). In our study, very few veterinarians reported encountering some type of resistance using the therapeutic dose, suggesting that the phenomenon of resistance is generally not a serious problem in Colombia. We are now trying to establish whether there are differences in the expression of molecules associated with resistance in samples from Brazilian and Colombian canines. This information could help identifying a solution to the resistance phenomenon affecting our neighboring country. In Colombian provinces such as Valle and Risaralda, homeopathic products and homotoxicology therapy derivatives were used as an alternative to difficult- to-obtain chemotherapy, whereas remote sites tended to opt for surgery or euthanasia. Surgery is not the treatment of choice due to the fragility of the tissue, the gaping wound, difficult healing and the tumour is easily transplanted to surgical wounds (^{Das, 2000}). Moreover, no studies have shown the mechanism of action of homeopathic agents or homotoxicology and their usefulness in treating CTVT is unproven.

Our results indicate a moderate prevalence and incidence of CTVT in Colombia and provide key clinical veterinary information on the medical approach and diagnosis of the disease. Our data also generate alternatives for establishing effective measures aimed at controlling and reducing disease prevalence in the country.