Thyroid Cancer in Dogs: An Update Based on 638 Cases (1995-2005) The goal of this study was to update the descriptive statistics of thyroid cancer by using data from multiple institutions collected through the Veterinary Medical Database (VMDB). Information was collected and reported from cases of canine thyroid cancer submitted to the VMDB between January 1, 1995 and December 31, 2005. Odds ratio (OR) analysis was performed on breeds that had 3% of the total number of dogs with thyroid cancer; ORs for each age category were also determined. Thyroid cancer represented 1.1% of all neoplasms during the time period of interest. Golden retrievers, beagles, and Siberian huskies all had significantly increased ORs for developing thyroid cancer. No sex predisposition was evident, but dogs between 10 and 15 years of age had a significantly increased chance of developing thyroid disease. Carcinomas and adenocarcinomas represented 90% of thyroid cancers, while adenomas represented 9.3%. Thyroid carcinoma and adenocarcinoma continue to be uncommon in our canine population. Older dogs are still more commonly affected, and this study is in agreement with previous studies that golden retrievers and beagles are overrepresented. A new finding is that Siberian huskies are also overrepresented. Carcinomas represent a much higher proportion of thyroid cancers than previously reported, and adenomas are likely incidental findings on necropsy. Thyroid cancer should be high on the list of differentials for a neck mass in older, large-breed dogs, as they make up 1.1% of the cancer cases reported. The overwhelming majority of thyroid cancers are carcinomas, and they are most common in golden retrievers, beagles, and Siberian huskies. J Am Anim Hosp Assoc 2010;46:249-254. Katja L. Wucherer, DVM Vicki Wilke, DVM, PhD, Diplomate ACVS RS From the Department of Veterinary Clinical Sciences, College of Veterinary Medicine, University of Minnesota, 1352 Boyd Avenue, St. Paul, Minnesota 55108. Introduction The incidence of thyroid cancer in the dog has been reported to be between 1% and 4%, 1-5 making it the most common endocrine tumor in dogs. 6 Carcinomas are far more common than adenomas and represent between 60% and 90% of thyroid tumors. 1,2,4,6-11 Thyroid carcinomas are considered highly malignant, with 16% to 60% of dogs historically reported to have evidence of distant metastasis in regional lymph nodes or pulmonary parenchyma at the time of diagnosis. 1,6,10,12,13 Left untreated, 60% to 80% of cases have evidence of distant metastasis at necropsy. 6,10,12,14 Benign adenomas tend to be incidental findings, with up to 80% found postmortem. 2,6,10 The median age previously reported for dogs at the time of diagnosis was 9 to 10.5 years, with a range of 5 to 18 years. 1-5,8,11,12,15,16 A study by Leav et al revealed that the average age of dogs with thyroid adenoma was 10.7 years, while the average age of dogs with thyroid carcinoma was 9 years. 10 A study performed by Haley et al on a colony of beagles noted the dogs were much older, with a mean age of 16 years at the time of diagnosis. 17 JOURNAL of the American Animal Hospital Association 249
250 JOURNAL of the American Animal Hospital Association July/August 2010, Vol. 46 As opposed to humans in which thyroid nodules are about four times more likely in women than in men, numerous studies have not found a sex predisposition. 1-5,8,10,12,14,15,18 Authors have found a variety of trends in breeds. Early studies revealed that beagles, boxers, and golden retrievers were overrepresented. 1-3,5,6,10 Furthermore, a study performed by Liptak et al found that boxers are at an increased risk for benign tumors, while golden retrievers and beagles are at an increased risk for malignant tumors. 19,20 Subsequent studies have shown that mixed-breed dogs are overrepresented 16 or that no breed predisposition exists. 12 Much of the data used to cite descriptive statistics in dogs with thyroid tumors are several decades old. The goal of this study was to update descriptive statistics and the prevalence of dogs with thyroid cancer, using a much larger number of cases than has been previously reported. Recently, we have noted an increased recognition of the disease in our hospital and therefore hypothesized that a higher incidence of thyroid cancer would be present than previously described. In addition, we have not appreciated an increased incidence in the breeds that have been previously cited as having a predisposition to thyroid cancer; therefore, we also hypothesized that given a larger sample size and accounting for breed popularity, a breed predisposition toward thyroid cancer would not be seen. Materials and Methods Study Design This was a retrospective study designed to update descriptive data on thyroid cancer. Information obtained through the Veterinary Medical Database (VMDB) included the institution where the animal was seen, date of discharge, weight, gender, age, breed, discharge status (alive, died, or euthanized), and diagnostic code. Data were reported as both the absolute number of cases with the variable of interest as well as the percentage of the total number of dogs diagnosed with thyroid cancer. Statistical Analysis A Web-based statistical program was used to calculate odds ratios (ORs). 21 Odds ratios were performed on breeds that represented 3% of the total number of dogs with thyroid cancer. An OR was also obtained for each age category. Case Selection Eligible cases were retrieved from a computer search of the VMDB for dogs presented during January 1, 1995 through December 31, 2005 that had the diagnostic code of follicular thyroid carcinoma, with or without metastasis, as well as thyroid adenomas. Dogs that had metastasis from a different tumor to the thyroid gland were included. Duplicate cases as a result of multiple visits were eliminated. Statistical analysis included the total number of dogs diagnosed with any type of cancer, as well as the total dogs seen for any reason, broken down by breed and age. Results Thirteen VMDB-participating institutions reported 638 cases of dogs with thyroid cancer (out of a total of 275,347 cases submitted) that were discharged between January 1, 1995 through December 31, 2005. This represents 1.1% of all neoplasms seen during the same time period. Of these 638 cases, 54% were female (343 total; 38 intact, 305 spayed), and 46% were male (295 total; 83 intact, 212 castrated). Six (<1%) dogs were <4 years of age at the time of diagnosis; 45 (7%) were between 4 and 7 years of age; 205 (32%) were between 7 and 10 years of age; 364 (57%) were between 10 and 15 years of age; 17 (3%) were >15 years of age; and one dog was of unknown age [Table 1]. Dogs in the 10- to 15-year age category had an OR of 6.9 (95% confidence interval [CI]: 5.90<OR<8.15), which was statistically significant (P values <0.005). Twenty-four beagles were diagnosed during this time period. Of those 24 dogs, 12 (50%) were 7 to 10 years of age at the time of diagnosis; 10 (42%) were between 10 and 15 years of age; and only one was either <1 year of age or >15 years of age. Five (<1%) dogs were <2.27 kg; 24 (4%) dogs were between 2.27 and 6.81 kg; 64 (10%) dogs were between 6.81 and 13.63 kg; 119 (19%) dogs were between 13.63 and 22.72 kg; 151 (24%) dogs were between 22.72 and 34.09 kg; 92 (14%) dogs were between 34.09 and 45.45 kg; 26 (4%) dogs were >45.45 kg; and for 157 (25%) dogs, a weight was not reported [Table 2]. Golden retrievers (n=80) and Labrador retrievers (n=74) represented 13% and 12%, respectively, of the breeds reported with a diagnosis of thyroid cancer. Beagles (n=24) were 4% of dogs diagnosed with thyroid cancer, while Shetland sheepdogs (n=18) and Siberian huskies (n=16) represented 3% each. Boxers (n=13) were only 2% of dogs reported with thyroid cancer. The proportions of each of these breeds seen during this time period for any diagnosis were: golden retrievers 5.9% (n=16,497); Labrador retrievers 10.1% (n=27,837); beagles 1.3% (n=3579); Shetland sheepdogs 1.6% (n=4588); Siberian huskies 1.08% (n=2994); and boxers 2.6% (n=4634). Only golden retrievers (OR 2.2; 95% CI: 1.75<OR<2.84), beagles (OR 3.0; 95% CI: 2.0<OR<4.57), and Siberian huskies (OR 2.5; 95% CI: 1.51<OR<4.08) were significantly (P values <0.005) overrepresented [Table 3]. Dogs reported as mixed breeds (n=159) were 26% of the cases, and all the rest of the breeds (n<13) were reported as <2%. Ninety-three (14.6%) dogs had unclassified thyroid masses, which leaves 545 cases that were classified as either adenoma, adenocarcinoma, carcinoma, or cancer not of thyroid origin. Thyroid adenoma (including Hürthle cell tumors) was diagnosed in 51 (9.3%) dogs, while thyroid adenocarcinoma or carcinoma was diagnosed in 490 (90.0%) dogs. Four (<1%) dogs had cancer elsewhere that had metastasized to the thyroid. Two of these had squamous cell carcinoma, one had hemangiosarcoma, and the other had transitional cell carcinoma (primary location not specified). Carcinoma and adenocarcinoma were diagnosed in 90% of the tumors in golden retrievers, in 95% of the tumors in beagles, and in 85% of the tumors in Siberian
July/August 2010, Vol. 46 Thyroid Cancer in Dogs 251 Table 1 Age Distribution, Based on Categories Established by the VMDB, * of Dogs Diagnosed With Thyroid Cancer No. of Dogs (%) With No. of Dogs (%) Thyroid With Any Age (y) Cancer Diagnosis P Value Odds Ratio 2-4 y 4 (1%) 43,546 (16%) <0.0001 0.04 95% CI: 0.01<OR<0.1 4-7 y 45 (7%) 55,014 (16%) <0.0001 0.32 95% CI: 0.24<OR<0.43 7-10 y 205 (32%) 49,441 (18%) <0.0001 2.15 95% CI: 1.81<OR<2.55 10-15 y 364 (57%) 44,249 (16%) <0.0001 6.94 95% CI: 5.90<OR<8.10 15+ y 17 (3%) 3257 (1%) <0.0004 2.27 95% CI: 1.39<OR<3.75 * VMDB=Veterinary Medical Database CI=confidence interval; OR=odds ratio Table 2 Weight Distribution, Based on Categories Established by the VMDB, * of Dogs Diagnosed With Thyroid Cancer Weight No. of (kg) Dogs % 0-2.27 5 <1 2.27-6.81 24 4 6.81-13.63 64 10 131.63-22.72 119 19 22.72-34.09 151 24 34.09-45.45 92 14 45.45 26 4 Unknown 157 25 * VMDB=Veterinary Medical Database huskies. This was not significantly different from the finding that 90% of dogs with thyroid cancer have adenocarcinoma (P values >0.1). At the time of discharge, 519 (81%) dogs were alive, 15 (2%) had died, and 104 (16%) were euthanized. Table 4 lists the status of dogs at discharge, depending on their diagnosis. All dogs that were diagnosed with thyroid adenoma and had either a discharge status of euthanized (n=29) or died (n=2) also had necropsies performed. Discussion Previous descriptive data on thyroid carcinoma in dogs have typically been on smaller sample sizes, with all dogs being from one or a few institutions or hospitals. In the literature, estimates on the incidence of thyroid cancer in dogs were based on a small handful of studies performed over 2 decades ago. This current study revealed that thyroid cancers accounted for 1.1% of all neoplasms diagnosed in dogs between January 1, 1995 and December 31, 2005. This continues to be a consistent finding when compared to the earlier studies, particularly those performed by Brodey and Kelly (1.6%, 1967), Harari et al (1.2%, 1986), and Birchard and Roesel (1.2%, 1981). 1-3 Based on our findings and contradictory to our hypothesis, we conclude that an increase has not occurred in the incidence of thyroid cancer in dogs.
252 JOURNAL of the American Animal Hospital Association July/August 2010, Vol. 46 Table 3 Breeds With >3% of the Total Number of Dogs With Diagnosis of Thyroid Cancer That Were Presented Between January 1,1995 Through December 31, 2005 Total No. (%) of Total No. (%) of Dogs With Any Dogs With Breed Diagnosis Thyroid Cancer Odds Ratio * P Value Golden retriever 16,497 (6%) 80 (13%) 2.2 <0.001 95% CI: 1.74<OR<2.84 Labrador retriever 27,837 (10%) 74 (12%) 1.24 0.09 95% CI: 0.97<OR<1.58 Beagle 3579 (1%) 24 (4%) 3.01 <0.001 95% CI: 1.98<OR<4.57 Shetland sheepdog 4588 (2%) 18 (3%) 1.7 0.04 95% CI: 1.05<OR<2.77 Boxer 4634 (3%) 13 (2%) 1.28 0.34 95% CI: 0.74<OR<2.23 Siberian husky 2994 (1%) 16 (3%) 2.47 <0.005 95% CI: 1.5<OR<4.07 Total no. of dogs 275,347 (100%) 638 (100%) * CI=confidence interval; OR=odds ratio Table 4 Status at Discharge Depending on Diagnosis as Categorized by the VMDB * Status Adenoma Carcinoma Metastatic Unclassified All Alive 20 (39%) 418 (85%) 1 (25%) 80 (86%) 519 (81%) Died 2 (4%) 11 (2%) 1 (25%) 1 (1%) 15 (2%) Euthanized 29 (57%) 61 (12%) 2 (50%) 12 (13%) 104 (16%) Total 51 490 4 93 638 * VMDB=Veterinary Medical Database Metastatic disease included squamous cell carcinoma, hemangiosarcoma, and transitional cell carcinoma. This study is in agreement with previous studies that note that gender does not appear to be a risk factor for the development of thyroid carcinoma. The age of presentation of dogs with thyroid cancer may be older than previously reported. In this study, the majority of dogs are between 10 and 15 years of age older than the 9 to 10 years of age typically cited. One possibility for this change is that a delay may be between onset of clinical signs and clinical diagno-
July/August 2010, Vol. 46 Thyroid Cancer in Dogs 253 sis. However, only a few reports have shown a delay of 1 to 2 years between owner recognition of clinical signs and time of diagnosis; most studies note that the average time to diagnosis is 1 to 2 months. 6,16,22 Another possibility is that all of the institutions reporting are referral hospitals; therefore, the diagnosis was potentially made earlier by the referring veterinarian. Dogs weighing between 22.72 and 34.09 kg made up almost 25% of the cases identified in this study. This makes intuitive sense, considering golden retrievers and Labrador retrievers collectively made up 25% of the dogs diagnosed with thyroid cancer. However, of those two breeds, only golden retrievers are overrepresented when breed popularity was taken into account. These findings are consistent with previous reports. 3,4,13 Beagles were also overrepresented in this report, but, unlike the earlier study by Haley et al in which the average age of a beagle with a thyroid tumor was 16.2 years, most of the beagles in this study were between 7 and 15 years of age, with only one beagle being >15 years of age. In this study, a larger proportion of the dogs with thyroid cancers were boxers, but this was not statistically significant when compared to all the boxers seen in the same time period. We recognize that the Shetland sheepdog has a P value of 0.04, lower than the traditional P value of 0.05; however, after correcting for multiple testing, this P value is actually rendered not statistically significant. Surprisingly, and never before reported to our knowledge, the Siberian husky was overrepresented. Theories as to why this is the case are sparse, as the Siberian husky does not seem to have a predisposition to thyroid disease. A study performed in 2006 by Lee et al describes a predominantly Alaskan malamute pedigree with familial medullary thyroid carcinoma. 23 The first reported occurrence of familial thyroid carcinoma was in this group of dogs. From the data obtained, we cannot tell if a similar phenomenon is at play in our population of interest. Alaskan malamutes are closely related genetically to Siberian huskies. 24 Possibly dogs reported as Siberian huskies were not true Siberian huskies but instead predominantly consisted of that breed; if so, this could have skewed the results. Medullary thyroid carcinoma arises from the parafollicular cells, which produce calcitonin; such tumors may be underdiagnosed because of the need for special staining. 12 Many studies have not conclusively been able to support that the histological subtype of the cancer affects prognosis in dogs. 6 However, at least one study revealed that medullary carcinoma can comprise up to 36% of thyroid cancers and may carry a more favorable prognosis than other types of thyroid cancer. 12 We reject our second hypothesis that there is no breed predisposition to thyroid cancer in favor of stating that golden retrievers, beagles, and Siberian huskies have a statistically significant increased risk of developing thyroid cancer. Adenomas made up a much smaller percentage of thyroid tumors than previously noted, accounting for only 9.3% of thyroid tumors in this study. This is in contrast to humans, where 15% to 40% of thyroid nodules are adenomas and 8% to 17% are carcinomas. The rest, about 42% to 77%, are nonneoplastic colloid nodules. 14 Perhaps given the benign nature of adenomas, they are simply not being diagnosed as frequently. Another possibility is that adenomas are being reported as adenocarcinomas to the VMDB, thus the true percentage of adenomas versus adenocarcinomas is unknown. The VMDB does not clarify if the animal died or was euthanized from the diagnosis submitted or from some other diagnosis; therefore, it is very plausible that the higher percentage of adenomas in dogs that were euthanized or had died was an incidental finding at necropsy. This is supported by the finding that all adenomas diagnosed on dogs that had a discharge status of euthanized or died also underwent necropsy examination. Carcinomas and adenocarcinomas made up 90% of the classified thyroid tumors, a percentage that is consistent with some of the higher estimates reported in the past. 4,12 Overrepresented breeds did not significantly deviate from this finding. While uncommon, the thyroid is a place of distant metastasis, which is not surprising considering its highly vascular nature. The advantage of using the VMDB for descriptive data is that it provides a large number of cases from a variety of geographic regions over a large time span. The VMDB started in 1961 as a National Cancer Institute-sponsored project to track cancer in animals. Since then, it has expanded to include all cases seen at any of the veterinary teaching hospitals, totaling over seven million cases from 26 different institutions in the United States. Participating institutions submit procedure and diagnostic codes for each case that is presented for evaluation or care only (i.e., excludes pharmaceutical transactions). These codes are readily made available to participating institutions and are matched to the diagnosis and procedures in the animal s medical record provided by the clinician overseeing the case. The diagnosis is based on the overseeing clinician s discretion and may or may not include histopathology. Reported along with the diagnostic and procedure codes are the animal s medical record number, date of birth, gender, specie, breed, chip identification, postal code, country, weight, admission date, discharge date, discharge status, and whether it was an initial or recheck visit. The disadvantage is that the information obtained is limited, and we cannot always provide information on metastasis, survival time, and cause of death. In this study, for example, further classification of thyroid tumors into follicular or medullary (parafollicular) tumors would have been helpful, since (as stated previously) medullary tumors seem to carry a better prognosis. Future studies using hospital records would need to be performed to update this specific information. Only one other paper has used the VMDB to study thyroid neoplasia; this was a study performed by Hayes and Fraumeni in 1975. 4 One hundred forty-four cases of thyroid cancer were collected between March 1964 and June 1974 from the 500,000 total cases submitted during this time period. Almost twice as many cases with any type of diagnosis were submitted during that time period when compared to
254 JOURNAL of the American Animal Hospital Association July/August 2010, Vol. 46 the 275,347 cases that were submitted during the time period of interest for our study. This is most likely because the coding system for diagnoses that was used in both studies was the Standard Nomenclature of Veterinary Disease and Operations (SNVDO). Submissions using SNVDO slowed in the late 1990s, and several schools have since switched over to a newer system. Databanks such as SNVDO are rare in veterinary medicine and provide a reservoir of information that far exceeds cases from even the largest multicenter study; thus, we should strive to accurately and consistently contribute to these databanks. Conclusion Thyroid cancer continues to make up about 1.1% of all neoplasms diagnosed in our canine population. No sex predisposition is seen, and most dogs are between 10 and 15 years of age at the time of diagnosis. Golden retrievers, beagles, and Siberian huskies are overrepresented, even when breed popularity is taken into account. Ninety percent of thyroid cancers are carcinomas, while 9.3% are adenomas. Adenomas are likely incidental findings at necropsy, but further studies are needed to confirm this. Based on other studies that have shown carcinomas to be highly malignant with a high rate of metastasis at the time of diagnosis, we recommend that palpation of the neck be part of every physical examination, especially in geriatric animals. Data submitted to the VMDB provide invaluable tools for research; therefore, institutions and hospitals should continue to strive to update the databank. 11. Scarlett JM. Epidemiology of thyroid diseases of dogs and cats. Vet Clin North Am Small Anim Pract 1994;24:477-486. 12. Carver JR, Kapatkin A, Patnaik AK. A comparison of medullary thyroid carcinoma and thyroid adenocarcinoma in dogs: a retrospective study of 38 cases. Vet Surg 1995;24:315-319. 13. Page RL. Tumors of the endocrine system. In: Withrow SJ, Macwen EG, eds. Small Animal Clinical Oncology. 3rd ed. Philadelphia: Saunders, 2001:423-427. 14. Mazzaferri EL. Management of a solitary thyroid nodule. N Eng J Med 1993;328:553-559. 15. Turrel JM, McEntee MC, Burke BP, et al. Sodium iodide I131 treatment of dogs with non resectable tumors: 39 cases. J Am Vet Med Assoc 2006;229:542-548. 16. Klein MK, Powers BE, Withrow SJ, et al. Treatment of thyroid carcinoma in dogs by surgical resection alone: 20 cases (1981-1989). J Am Vet Med Assoc 1995;206:1007-1009. 17. Haley PJ, Hahn FF, Muggenburg BA, et al. Thyroid neoplasms in a colony of beagle dogs. Vet Pathol 1989;26:438-441. 18. Fineman LS, Hamilton TA, de Gortari A, et al. Cisplatin chemotherapy for treatment of thyroid carcinoma in dogs: 13 cases. J Am Anim Hosp Assoc 1998;34:109-112. 19. Liptak JM. Canine thyroid carcinoma. Clin Tech Small Anim Pract 2007;22:75-81. 20. Loar AS. Canine thyroid tumors. In: Kirk RW, ed. Current Veterinary Therapy IX. Philadelphia: Saunders, 1986:1033-1039. 21. http://www.quantitativeskills.com/sisa/statistics/t-test.htm 22. Feldman EC, Nelson RW. Canine thyroid tumors and hyperthyroidism. In: Feldman EC, Nelson RW, eds. Canine and Feline Endocrinology and Reproduction. 3rd ed. St. Louis: Saunders, 2004:219-249. 23. Lee J, Larsson C, Lui W, et al. A dog pedigree with familial medullary thyroid cancer. Int J Oncol 2006;29:1173-1182. 24. Parker HG, Kim LV, Sutter NB, et al. Genetic structure of the purebred domestic dog. Science 2004;304:1160-1164. Acknowledgments We thank the VMDB for all the help and support provided, as well as for assimilating the data used in this study. References 11. Birchard SJ, Roesel OF. Neoplasia of the thyroid gland in the dog: a retrospective study of 16 cases. J Am Anim Hosp Assoc 1981;17:369-372. 12. Brodey RS, Kelly DF. Thyroid neoplasms in the dog. A clinicopathologic study of fifty seven cases. Cancer 1968;22:406-416. 13. Harari J, Patterson J, Rosenthal R. Clinical and pathologic features of thyroid tumors in 26 dogs. J Am Vet Med Assoc 1986;188: 1160-1164. 14. Hayes HM, Fraumeni JF Jr. Canine thyroid neoplasms: epidemiologic features. J Natl Cancer Inst 1975;55:931-934. 15. Mitchell M, Hurov LI, Troy GC. Canine thyroid carcinomas: clinical occurrence, staging by means of scintiscans and therapy of 15 cases. Vet Surg 1979;8:112-118. 16. Barber LG. Thyroid tumors in dogs and cats. Vet Clin North Am Small Anim Pract 2007;37:756-773. 17. Radlinsky MG. Thyroid surgery in dogs and cats. Vet Clin North Am Small Anim Pract 2007;37:789-798. 18. Worth AJ, Zuber RM, Hocking M. Radioiodide (I131) therapy for the treatment of canine thyroid carcinoma. Aust Vet J 2005;83: 208-214. 19. Ettinger SJ, Feldman EC. Thyroid neoplasia. In: Ettinger SJ, Feldman EC, eds. Textbook of Veterinary Internal Medicine: Diseases of the Dog and Cat. St. Louis: Saunders, 2005:1558-1560. 10. Leav I, Schiller AL, Rijnberk A, et al. Adenomas and carcinomas of the canine and feline thyroid. Am J Pathol 1976;83:61-93.