Feline Chondrosarcoma: A Retrospective Study of 67 Cats (1987–2005)
The histories of 67 cats diagnosed with chondrosarcoma (CSA) from 1987 to 2005 were reviewed. The mean age was 9.6 years, and males were 1.9 times more likely to be affected than females. Chondrosarcomas were diagnosed in the following sites: appendicular and axial skeleton, nasal cavity, facial bones, and extraskeletal sites. Of the 46 (70%) CSA associated with bone, 63% arose in long bones and 37% arose in flat bones. The remaining (30%) CSA arose in the subcutis. In cases available for follow-up (n=24), no definitive evidence of metastases was found. Cats that underwent radical surgical therapies were more likely to achieve long-term control or cure.
Introduction
Chondrosarcoma (CSA) is a malignant tumor in which the neoplastic cells produce chondroid and varying amounts of fibrillar matrix, but never osteoid.1 Feline CSA appears to grow slowly and invade locally, but it rarely metastasizes. Survival time seems to be lengthened by aggressive surgeries such as amputation, and information on therapies such as radiation is limited.2,3 The purposes of this study are to detail the signalment and tumor location in 67 cats diagnosed with CSA and to summarize the treatment methods for the 24 cats that were available for follow-up.
The characteristics and biological behavior of CSA in cats are not well defined. In one survey, only two of 256 neoplasms in cats were CSA.4 Feline CSA has been reported in studies of primary bone tumors, non-lymphoma tumors of the vertebral canal, and nasal tumors; also, several case reports have described CSA in cats, including two reports of CSA arising in multilobular chondromas.2,3,5–10 In one study of primary bone tumors in 24 cats, osteosarcoma (OSA) was the most common, followed by juxtacortical OSA and CSA.5 The CSAs reported in this study arose from the mandible, proximal tibia, and scapula.Asubsequent study of primary bone tumors in cats indicates that CSA is as common as OSA, with tumors arising from the femur, ischium, metacarpal bones, and scapula.2
Chondrosarcomas also comprise a subset of sarcomas that arise in soft tissues. In a study of 23 cats with “soft-tissue sarcomas,” two cases were classified as CSA.11 In another study by Hendrick and Brooks, the histology and immunohistochemistry of sarcomas arising in vaccination sites were examined.12 Only one (2%) of the 46 postvaccinal sarcomas was a CSA; most of these sarcomas were fibrosarcomas and malignant fibrous histiocytomas. A similar incidence (2%) of CSA in vaccine-associated sarcomas was found in another immunohistochemical study.13
In contrast to CSA in cats, CSA in dogs is fairly well characterized. It accounts for approximately 10% of primary bone tumors in dogs and is, therefore, second only to OSA in incidence of primary bone tumors.14,15 In dogs, 61% of CSAs originate in flat bones; in one study, 50% of CSAs arose in the nasal cavity.16,17 Local recurrence, not metastasis, is the primary cause of treatment failure in dogs.16
Materials and Methods
Surgical pathology reports dating between 1987 and 2005 were obtained for all cats diagnosed with CSA by the Laboratory of Pathology and Toxicology at the University of Pennsylvania School of Veterinary Medicine. Tissues were fixed in 10% buffered formalin, embedded in paraplast, cut, and stained with hematoxylin and eosin. Histopathology slides of all 67 cases were reviewed by one veterinary pathologist to confirm CSA. Chondrosarcomas are composed of interwoven bundles, sheets, and nests of cells within a lightly basophilic, chondroid matrix. Cells have a moderate amount of amphophilic cytoplasm, indistinct borders, and large ovoid nuclei. Cellular and nuclear pleomorphism, binucleation, and frequent mitoses are common features. Tumor cells may produce neoplastic chondroid, fibrillar matrix, and bone formed by endochondral ossification of tumor cartilage [see Figure]. Cats with tumors producing varying amounts of both tumor chondroid and tumor osteoid were excluded from the study.
Information about age, sex, breed, and tumor location was obtained from the biopsy reports. Tumors were classified as arising from the nasal cavity, appendicular skeleton (i.e., humerus, radius/ulna, carpus, femur, tibia, tarsus, digit), axial skeleton (i.e., scapula, sternum, rib, vertebra, pelvis), facial bones (i.e., skull, maxilla, mandible), and extraskeletal regions (i.e., subcutaneous tissues, subcutis). Bone tumors were also categorized as those involving bones that develop via endochondral ossification (long bones) or those involving bones that develop from intramembranous ossification (flat bones). Tumors within the subcutis were further categorized as originating from the axilla, interscapular region, thoracic region, flank, hind limb, or lumbar area.
Additional information was obtained for all cases from the form submitted by the referring veterinarian that accompanies every biopsy sample. Information obtained from this form included species, breed, sex, birth date, case history, gross appearance of the tumor, number of lesions, size, duration and rate of growth, treatment, and location.A section of the form was available to indicate the type of sample submitted (e.g., wedge, entire specimen, Tru-Cut, punch, fragment, endoscopic, or other), and yes/no questions were provided to indicate lymph node involvement, tumor encapsulation, and excisional biopsy. Follow-up information was obtained from hospital records and questionnaires that were sent to the referring veterinarians. Additional clinical data (i.e., current status, survival time, radiographs, blood work, clinical signs, and treatment) were available for 24 cats. For the purposes of this study, treatment options included incisional and excisional biopsies, radical surgery (e.g., amputation), or chemotherapy and/or radiation therapy. Notation was made if the tumor was located in a previous vaccination site.
To assess differences in age, breed, and sex, cats with CSA were compared to the general population of cats with surgical accessions during the same time period. Categorical data were compared using either the chi-square or Fisher’s exact test, and Student’s t-test was used for continuous data. Analysis of variance was used to examine tumor location with regard to age. The chi-square test was used to examine tumor location with regard to sex. The Kaplan-Meier product limit method was used to determine the median survival time for cats with CSA.When applicable, data are presented as odds ratios (ORs) with 95% confidence limits (CIs). All analyses were performed using SAS statistical software.a A P value <0.05 was considered statistically significant.
Results
Age Distribution
The mean age of affected cats (n=67) was 9.6 years, and the median age was 9 years (range 2 to 18 years). No significant difference was found in mean age with respect to the general population of cats seen in the surgical pathology service during the same time period (9.6 years versus 8.8 years; P=0.09); also, no significant difference was found in age with respect to tumor location.
Breed
Most (n=57; 85%) of the cats were domestic shorthairs, followed by domestic longhairs (n=5; 7.5%), unspecified rare breeds (n=2; 3%), Siamese (n=2; 3%), and Persian (n=1; 1.5%). When compared to the general population of cats submitted to the surgical pathology service, no breed predilection was noted.
Sex Distribution
Of the 67 affected cats, 42 (63%) were males and 25 (37%) were females. When compared to the surgical pathology population, males were more frequently affected than females (OR 1.9; 95% CIs 1.1, 3.1; P=0.014). No statistically significant difference was seen between males and females with respect to tumor location.
Tumor Morphology/Histology
Tumor location was reported in 66 of the 67 cases. Of the 41 cases in which encapsulation was noted on the submission form, 26 (63%) of the tumors were encapsulated, and 15 (37%) were not. Involvement of regional lymph nodes, based on physical examination, was reported in 58 cases. Of these 58 cases, 56 (97%) had no lymph node involvement, and two (3%) had possible involvement, although this was not confirmed via biopsy in either case. Follow-up data were available for one cat, which is still alive with no recurrence or metastatic disease. The tumors were generally described (by the referring veterinarians) as white-tan, often cavitated, glistening, and firm. Histologically, these tumors were composed of bundles, nests, and sheets of neoplastic chondrocytes; some contained areas of necrosis, endochondral ossification, or mineralization. Aggressive histological features included severe pleomorphism, bi- or trinucleation, and numerous mitotic figures (approximately two to four per high-power field).
Tumor Location
Forty-six (70%) of the tumors were associated with bone. Thirteen (28%) tumors arose from the axial skeleton and included the following sites: scapula (15%), vertebra (6.5%), pelvis (2.2%), and sternum (4.3%). Appendicular CSAs accounted for 52% (n=24), and sites included the humerus (6.5%), radius/ulna (4.3%), femur (13%), tibia (8.7%), carpus (4.3%), tarsus (2.2%), and digits (13%). Six (13%) CSAs arose from facial bones, including the skull bones (4.3%) and mandibular/maxillary bones (8.7%). Three (6.5%) CSAs arose from the nasal cavity [Table 1].
The locations of tumors were further subdivided into those from bones that develop via endochondral ossification (i.e., vertebra, sternum, ribs, humerus, radius, ulna, femur, tibia, carpus, tarsus, and digits) and those from bones that undergo intramembranous ossification (i.e., facial bones, nasal turbinates, scapula, and pelvis). Of the CSAs that involved bones, 63% involved bones arising from endochondral ossification, and 37% involved bones arising from intramembranous ossification [Table 2].
Twenty (30%) cases of CSA involved subcutaneous sites and were categorized based on the following locations: axilla (5%), interscapular region (15%), thoracic region (10%), flank (15%), lumbar area (5%), hind limb (45%), and unspecified subcutaneous locale (5%). Histopathology reports regarding several of the subcutaneous tumors indicated debate as to whether these tumors represented vaccine-site sarcomas. In one report, the presence of grey-blue material within macrophages, which is consistent with vaccine product, was noted. The overall site distributions of CSAs were 44% in long bones, 26% in flat bones, and 30% in the subcutis [Table 2].
Follow-up Information
Information about the current status of the cat was available for 24 cases from 1998 to 2005. The remainder of the medical records had been discarded. Eleven (45.8%) cats are still alive with no recurrence or metastases.Most of these 11 cats had radical surgeries performed, including amputation, hemipelvectomy, and mandibulectomy. Only one of these 11 cats had an excisional biopsy only, in which the margins were clean; this particular tumor arose from the subcutis. The durations of follow-up for these 11 cats ranged from 0 to 2118 days (mean 490 days).
Four (16.7%) cats are currently alive but have progression of the primary tumor. All four of these cats had incisional biopsies only. One of these cats with local recurrence had a limb amputation performed 108 days later; this cat is still alive with no recurrence after the amputation, and it is included in the two survival categories. The durations of follow-up for these four cats ranged from 31 to 302 days (mean 167 days). (The cat that underwent amputation was excluded.)
Nine (37.5%) cats were euthanized for reasons related to the CSA, including decreased quality of life and poor prognosis. Five of these euthanized cats had incisional biopsies; one had a laminectomy, one had chemotherapy and radiation in addition to the excisional biopsy, and two had amputations (both of which had “dirty” surgical margins) [Table 3]. The survival times from date of diagnosis of euthanized cats ranged from 0 to 315 days (mean 85 days). The five euthanized cats that only had incisional biopsies survived 0 to 40 days. The cat that had the laminectomy was euthanized the day after surgery; the two that underwent amputations survived 137 and 237 days; and the cat that had an excisional biopsy and adjuvant therapy survived 315 days.
In summary, in the nine cats that received incisional biopsies only (i.e., effectively no treatment), four had progression of the tumor, and five were euthanized for reasons related to CSA. Fifteen cats received surgical treatment. Of the 12 cats that had radical surgery performed, 10 (83%) are still alive, and two (17%) were euthanized (the two euthanized cats had “dirty” surgical margins). Three treated cats had excisional biopsies performed; one with clean margins is still alive, and the two with “dirty” margins were euthanized. Median survival time could not be calculated, because 50% of the cats are alive, and the overall occurrence rate for this particular tumor is low.
Thoracic radiographs were available for 16 of the 24 follow-up cases. Almost all thoracic radiographs were taken at the time of diagnosis; radiographs were taken in one cat every 2 months, with no signs of metastatic disease. Possible lung metastases were reported in only three (18.7%) cats—two of which are still alive with no recurrence of the tumor. These cats were rechecked by the veterinarian at 563 and 918 days after radiographs were available, and no abnormalities were noted. The third cat with possible lung metastases is still alive with local progression of tumor; a recheck evaluation 31 days later found no clinical signs associated with pulmonary disease. Radiographs of the tumor were taken in 19 cases; the most commonly reported findings included soft-tissue swelling and bone lysis.
Results of blood work, reported in 18 of the 24 cases, were normal in all but three cats. One of these three cats that had been treated with chemotherapy was leukopenic, and two of these cats had mildly elevated amylase.
In three of the 24 cats available for follow-up, the veterinarians reported that the locations corresponded to vaccination sites.
Discussion
Although CSA is one of the more common bone tumors in cats, its overall incidence is relatively low.2,4,5 To date, published reports on its clinical characteristics and prognosis have been few. Much of the literature on CSA in cats consists of single case reports. In the present study, the signalment, tumor locations, and, when available, treatment protocols and outcomes were examined.
The mean age (9.6 years) of all cats with CSA in this study was slightly older than that in previous reports of CSA in cats.2 The finding that most affected cats were domestic shorthairs (85%) is consistent with the normal breed distribution in the feline population. A significant difference was seen regarding sex distribution; males were 1.9 times more likely to develop CSA than females. This finding is in contrast to the sex distribution in dogs with CSA, in which females are affected 1.5 times more frequently than males.16
Although CSAs have been reported in both long bones and flat bones, the published data are inadequate to determine site predilection in cats.2,5,15 In this study, most (46 of 67) CSAs were associated with bone; 63% involved long bones, and 37% involved flat bones. These findings are again in sharp contrast to site predilection in dogs, in which 61% of CSAs arise in flat bones.16 Of the long-bone CSAs in cats, the femur and digits were the most common sites (13% each), followed by the tibia (8.7%). The scapula was the most commonly affected flat bone as well as the most commonly affected of all bones (15%).
Feline CSAs do arise in vaccination sites, although fibrosarcomas are more widely reported.12 Thirty percent of CSAs in cats involved the soft tissues, with locations reflecting vaccination sites, including interscapular, lateral thorax/axilla, flank, lumbar, and hind limb. In this study, the hind-limb subcuticular region was the most commonly affected site. Interestingly, the hind-limb region is often the vaccination site reserved for rabies vaccine (right rear) and feline leukemia virus (left rear) in accordance with the Vaccine-Associated Feline Sarcoma Task Force guidelines set forth in 1996. In this study, vaccine product was seen in one of these tumors, and most of the subcuticular CSAs were suspected to be injection-related sarcomas based on their location and often aggressive histological features.
Because only a limited number of cases were available for follow-up, it is difficult to ascertain whether subcutaneous CSAs are more clinically aggressive than primary bone CSAs. In fact, in the cases available, survival appeared to depend more on the type and success of therapeutic intervention. Of the 24 cats for which follow-up information was available, 45.8% were still alive with no recurrence of the primary tumor or metastases. It is important to note that follow-up evaluation for evidence of metastasis was limited. Most of the cats that were still alive with no recurrence or metastases had undergone radical surgeries, including amputation, hemipelvectomy, and mandibulectomy; only one of these cats had an excisional biopsy performed, and the margins were clean. All four of the cats with local progression of the primary tumor had only incisional biopsies performed. Five of the nine euthanized cats had only incisional biopsies as well. One cat had an incisional biopsy and adjuvant therapy, including radiation and chemotherapy. Two of the nine euthanized cats had amputations, and one had a laminectomy; in all cases, the surgical margins were “dirty.” Moreover, the survival time from diagnosis in euthanized cats that had only an incisional biopsy was 0 to 40 days. This may reflect an unwillingness on the part of the owner to elect more radical surgery and instead opting for euthanasia. Interestingly, the survival times of the two cats with amputations increased to 137 and 237 days; the cat that received adjuvant therapy lived 315 days.
Of the three cats with reported lung masses (two of which were followed for 1.5 to 2.5 years), none showed any clinical signs of pulmonary disease, and no lung biopsies were performed. No cats in this study were euthanized because of lung metastasis or metastatic disease; this is in contrast to the situation in dogs, in which 20.5% of deaths (either from the disease or euthanasia) are caused by lung metastases.16
Conclusion
The overall incidence of CSA in the feline population is low, and the cases available were subject to the difficulties inherent in retrospective studies and the retrieval of records from multiple veterinary clinics. In spite of these limitations, some conclusions as to treatment and prognosis can be drawn from the available data. In summary, CSA in cats tends to be locally invasive but not prone to metastasize. Cats in this study that underwent aggressive surgery had a higher likelihood of achieving full recoveries.
Version 9.1; SAS Institute, Cary, NC 27513
Acknowledgments
The authors thank Dr. Frances Shofer for providing the statistical analyses for this paper.
Citation: Journal of the American Animal Hospital Association 44, 3; 10.5326/0440124
Neoplastic cells are arranged in nests and sheets and surrounded by lightly basophilic, chondroid matrix. Note marked nuclear and cellular pleomorphism and binucleate cells. Neoplastic cells have moderate amounts of finely vacuolated, amphophilic cytoplasm; large, oval, finely stippled nuclei; and prominent nucleoli. Mitoses are often variable. (Hematoxylin and eosin stain, A 10×; B 20×; C 40×.)
