Multiple Distinct Malignancies in Dogs: 53 Cases
Despite the clinical recognition of multiple distinct types of neoplasia in individual dogs, a detailed description of such cases has not recently been published. Canine oncology cases that were diagnosed with multiple, confirmed, distinct malignancies were prospectively collected for analysis. Approximately 3% of 1722 dogs that were presented to the oncology service at the Colorado State University Veterinary Medical Center were diagnosed with multiple distinct primary tumors. No significant breed or sex predisposition was apparent. Dogs with mast cell tumor, malignant melanoma, and thyroid carcinoma were significantly overrepresented and thus more likely to be diagnosed with multiple tumor types. These findings emphasize the importance of thorough, whole-body evaluation for dogs presented with mast cell tumor, malignant melanoma, and thyroid carcinoma. Furthermore, because approximately 33% of dogs that were presented with thyroid tumors were found to have additional distinct tumors, complete staging is justified in all dogs presented with thyroid tumors.
Introduction
Despite the fact that some dogs are presented with tumors believed to carry a low metastatic potential, frequent recommendations are for these dogs to undergo advanced staging to rule out preexisting or concurrent neoplasia. In addition, staging for tumors believed to carry a high probability of metastasis sometimes leads to the discovery of incidental or confounding second malignancies. Several studies have reported a relatively high occurrence of second malignancies in dogs with endocrine, hepato-cellular, and intracranial neoplasms.1–3 Yet, only a limited number of studies (the largest of which were published over 30 years ago) have directly examined the occurrence of multiple tumor types (MTT) in dogs.4,5 Therefore, it is not known how common second malignancies are clinically diagnosed using currently available staging modalities.
Several breeds are reported to be overrepresented with regard to the development of certain distinct tumor types, and a heritable basis for some specific canine cancers has been strongly suggested.6–11 Multiple endocrine neoplasia is one of the most well-characterized syndromes of MTT in humans; it results from one of several genetic mutations that ultimately leads to multiple tumors of endocrine and nonendocrine origin.12 While a specific genetic aberration has not been identified in dogs, a similar syndrome has been described in a limited number of dogs.13–17 A recently identified germline mutation in the mesenchymal-epithelial transition factor (MET) proto-oncogene was found in approximately 70% of rottweilers, a breed believed to be predisposed to the development of several cancers.18 Breed also influences tumor karyotypes in canine appendicular osteosarcoma and gene expression patterns in canine hemangiosarcoma.19,20 Based on these recent findings, we questioned whether any particular breeds might be predisposed to the development of multiple distinct malignancies.
Therefore, we set out to prospectively examine and report the occurrence of MTT in individual canine cancer cases in order to 1) determine the occurrence of MTT in dogs; 2) determine whether any particular breeds may be overrepresented within this population of dogs; and 3) determine if any specific tumor types are commonly associated with previous, concurrent, or subsequent primary cancers.
Materials and Methods
Case Selection
Dogs that were presented to the oncology service at the Colorado State University Veterinary Medical Center (CSU-VMC) between April 1, 2006 and December 31, 2007 with a newly diagnosed malignant tumor were eligible for inclusion. Dogs with histories of previous malignant tumor(s) or that received a diagnosis of synchronous malignant tumors (within 1 month) were prospectively collected for inclusion in this study. In addition, dogs diagnosed with a malignant tumor during this time period and that went on to develop a subsequent distinct malignant tumor(s) were also included. Case criteria collected included tumor histological type, breed, sex, age and weight at diagnosis of first malignancy, and history of prior anticancer therapy. Only cases with multiple, histologically or cytologically confirmed distinct malignancies were included in analyses.
Statistical Analysis
Variables assessed included tumor histological type, breed, sex, and age and weight at diagnosis of first malignancy. Median age and weight at diagnosis were determined, and Fisher’s exact test was utilized to determine if histological type, breed, or sex were significantly overrepresented. A P value of <0.05 was considered significant for all analyses.
Results
Case Characteristics
Fifty-three dogs met the criteria for inclusion. The sex, breed, age, and weight at diagnosis of affected dogs are summarized in Table 1. Within this population were 119 cytologically or histologically distinct tumors included in analysis. One dog was diagnosed with four distinct tumor types, nine dogs had three distinct tumor types, and the remaining 43 dogs each had two distinct malignant tumors.
Prior Therapy
Fourteen dogs received chemotherapy prior to being diagnosed with at least one neoplasm (median time from beginning of therapy was 8.6 months). Five dogs received radiation therapy prior to the development of additional neoplasms. Only one dog had a tumor arising near the irradiated field; the tumor was a hemangiosarcoma of the right atrium after undergoing three once-weekly 3-Gy fractions of radiation therapy to one hemithorax. Two dogs underwent investigational treatment with some form of immunotherapy prior to the development of subsequent neoplasms.
Observed Synchronous and Subsequent Tumors
Twenty-one dogs with no previous history of cancer were presented with synchronous primary tumors; they represented approximately 1.2% of cases that were presented to the oncology service at the CSU-VMC during the study period [Table 2]. Three additional dogs were presented with synchronous primary tumors and histories of previously diagnosed third distinct malignancies (one malignant melanoma, one peripheral nerve sheath tumor, and one soft-tissue sarcoma) [Table 2]. The remaining 29 dogs were diagnosed with multiple subsequent tumors [Table 3].
Prevalence of Distinct Tumor Types
During the study period, 1722 dogs were presented for a new appointment to the oncology service at the CSU-VMC, and approximately 3% of these dogs were diagnosed with MTT. Of dogs with tumor types that could be statistically evaluated (soft-tissue sarcomas were not able to be evaluated because of inconsistent coding within the database), those having thyroid carcinoma, malignant melanoma, or mast cell tumor were significantly overrepresented within the population of MTT dogs [Table 4]. No significant breed or sex predilection was found when comparisons were made to the general population of dogs presented as new appointments to the oncology service.
Discussion
We set out to prospectively evaluate client-owned dogs that developed multiple distinct malignant tumors in order to 1) determine the percentage of dogs affected by MTT; 2) determine whether any particular breeds were overrepresented; and 3) determine if any specific primary tumors are commonly associated with previous, concurrent, or subsequent primary cancers. Analysis of this population of dogs revealed that approximately 3% of new cases presented to the oncology service at the CSU-VMC were diagnosed with MTT; however, no significant breed or sex predisposition was apparent. Interestingly, several distinct tumor types were overrepresented within the population of MTT dogs; they included thyroid carcinoma, mast cell tumors, and malignant melanoma.
Because several breeds are predisposed to individual cancers, we questioned whether any breeds may be predisposed to the development of multiple distinct primary tumors. A recently described MET proto-oncogene mutation in the vast majority of rottweilers certainly lends credence to the possibility that distinct hereditary predispositions may exist among specific canine breeds, and such mutations could potentially give rise to multiple distinct malignancies within affected dogs. In fact, no breed predisposition was identified, with approximately 30% of affected dogs being of mixed-breed backgrounds. While spayed females were numerically highest within this population, this finding was also not statistically significant (P=0.21).
Previous treatment with cyclophosphamide in both dogs and humans has been associated with the development of transitional cell carcinoma (TCC) of the urinary bladder, and treatment with alkylating agents or topoisomerase inhibitors can be associated with the development of human leukemias.21–25 Therefore, it was critical to evaluate and report whether dogs had received any chemotherapy prior to the development of subsequent primary tumors. Fourteen dogs received some form of chemotherapy prior to the development of subsequent tumors. The median time from first chemotherapy treatment to the development of a subsequent tumor was 8.6 months in this population of dogs. Only four dogs within this study were diagnosed with TCC of the urinary bladder. Two of these dogs had not received prior chemotherapy; one dog had received vinblastine (48.5 months prior to development of TCC); and the remaining dog had previously been treated with a multidrug lymphoma chemotherapy protocol that included cyclophosphamide (approximately 8 months prior to diagnosis of TCC).
The development of secondary sarcomas resulting from definitive megavoltage radiation therapy has been reported in dogs; however, the resulting tumors were diagnosed long after (5.2 and 8.7 years) undergoing radiation therapy.26 While five dogs within our study underwent some form of external-beam megavoltage radiation therapy prior to the development of subsequent tumors, only one dog developed a tumor near or within the radiation field. This particular dog underwent an investigational clinical trial evaluating the safety and efficacy of immunotherapy combined with external-beam radiation therapy for metastatic pulmonary osteosarcoma. Therefore, this dog received three, 3-Gy fractions of external-beam radiation to one hemithorax. Necropsy performed on this dog revealed primary hemangiosarcoma of the right auricle; however, the tumor was unlikely to have arisen secondary to radiation, because the total dose received (9 Gy) was quite low, and the diagnosis was made within 2.5 months after receiving the first 3-Gy dose.
While 3% of dogs represents a relatively small subpopulation overall, several tumor types were found to be significantly overrepresented within affected dogs. Tumor types included mast cell tumors, malignant melanoma, and thyroid carcinoma [Table 4]. Several breeds have a predisposition for the development of mast cell tumors, and previous studies report that up to 31% of dogs diagnosed with a single mast cell tumor will go on to develop additional mast cell tumors.6,27 In the present study, 25% of dogs diagnosed with mast cell tumors had or went on to develop additional distinct tumors that were not mast cell malignancies. The high risk of developing additional tumors in dogs with mast cell tumors is noteworthy; however, it must be cautioned that the dogs within this study likely represent a selective population of large or aggressive mast cell tumors for which referral is pursued. Alternatively, because mast cell tumors represent the most common skin malignancy of dogs, the high incidence of mast cell tumors within this study is possibly a result of incidentally diagnosed tumors when the dogs were presented to an oncologist for evaluation or treatment of different tumor types.6
In humans, approximately 7% of patients diagnosed with thyroid tumors and 10% diagnosed with melanoma will go on to develop additional distinct malignancies.28 Malignant melanoma and thyroid carcinoma were statistically overrepresented in dogs diagnosed with MTT. Activating mutations in the rearranged during transfection (RET) gene have been associated with the familial form of medullary thyroid carcinoma in humans, and although a dog pedigree with an apparent familial medullary thyroid cancer has been reported, no such mutation was identified within the affected dogs.11 Boxers, beagles, and golden retrievers are reported to be at increased risk of developing thyroid tumors; however, the overrepresentation of thyroid tumors within this report cannot be explained based upon breed predisposition, since only two golden retrievers and one beagle with MTT were diagnosed with thyroid tumors.29
Several dogs that were suspected to have MTT were excluded from analysis because of lack of cytological or histological confirmation. Combined with the fact that not all dogs underwent complete necropsy is the likelihood that the true prevalence of multiple distinct tumor types in dogs is slightly higher than reported. Also worth noting is that certain tumor sites may have a propensity to be underrepresented in a prospective clinical study because of the inherent difficulty associated with obtaining adequate diagnostic tissue from those sites (i.e., heart base, brain, or lung). In fact, one previous postmortem study reported concurrent unrelated tumors in 23% of dogs diagnosed with primary intracranial neoplasia.2 However, we felt it critical to include only dogs with confirmed multiple distinct malignancies in analysis. An additional deficiency within this report was our inability to statistically evaluate the total number of soft-tissue sarcomas diagnosed at the CSU-VMC because of the inconsistent coding of medical records. While this deficiency in no way affects the collection of prospective cases, nor does it impact the statistical analysis of other tumor types, it does preclude the ability to determine whether soft-tissue sarcomas may be overrepresented in this population of canine cancer cases.
Conclusion
The diagnosis of multiple distinct malignancies in dogs was made in approximately 3% of the dogs that were presented to the oncology service at the CSU-VMC. Despite numerical differences, no statistically significant sex or breed predilection was identified. Dogs with mast cell tumor, malignant melanoma, and thyroid carcinoma were overrepresented and thus more likely to be diagnosed with MTT. For the practicing veterinarian, such information should alert clinicians to the possibility of multiple distinct malignancies in dogs. Also, more extensive staging and workup for dogs diagnosed with a thyroid tumor, mast cell tumor, or malignant melanoma may be justified.
