Postvaccinal Sarcomas in the Cat: Histology and Immunohistochemistry

Abstract

Recently, we reported an increased incidence of soft tissue sarcomas in a biopsy population of cats at the University of Pennsylvania School of Veterinary Medicine.8 Epidemiologic evidence indicated that this increase coincided with enactment of a Pennsylvania state law requiring rabies vaccination of cats, and the majority of these sarcomas were in areas routinely used by veterinarians for vaccination (dorsal neck/ interscapular, dorsolateral thorax, hindlimb, dorsal lumbar). Many of the sarcomas were surrounded and partially infiltrated by an inflammatory infiltrate of lymphocytes and macrophages. A similar inflammatory infiltrate had previously been reported to occur subsequent to subcutaneous injection of rabies vaccines in cats and dogs.' Furthermore, aluminum was identified in macrophages surrounding the sarcomas by electron probe x-ray microanalysis. Aluminum, in the form of aluminum hydroxide or aluminum phosphate, is an adjuvant in approximately 20% of vaccines for feline use. It was our interpretation that persistence of the inflammatory and immunologic reactions associated with the presence of the aluminum in the vaccination sites predisposes the cat to a derangement of its fibrous connective tissue repair response, occasionally leading to neoplasia. It is now apparent that vaccine-associated sarcomas are being recognized throughout the United S t a t e ~ . ~ . ' ~ Here we describe the histology and immunohistochemical marker staining of postvaccinal sarcomas. We hope this will increase the recognition of these sarcomas by pathologists and shed some light on the pathogenesis of this intriguing syndrome. The information presented here is based on 46 feline sarcomas in the surgical pathology files of the School of Veterinary Medicine of the University of Pennsylvania. These sarcomas were received during 1991 and 1992 and were documented by Veterinarians responding to an epidemiological survey to have occurred at sites at which there had been previous vaccination (unpublished data). Sarcomas were fixed in 10% formalin and routinely processed for histologic examination. Five-micrometer sections were cut from paraffinembedded blocks and stained with hematoxylin and eosin. A previously described avidin-biotin immunoperoxidase complex technique was used on representative sarcoma^.^ Briefly, paraffin-embedded sections, 5 pm in thickness, were cut and carefully melted at 58-60 C. After deparaffinization and rehydration, slides were incubated in 0.3% H 2 0 2 in absolute methanol for 45 minutes. Sequential incubations in 20% normal goat serum (30 minutes), primary antiserum (1 hour at room temperature or overnight at 4 C), secondary biotinylated antibody (45 minutes), and avidin-biotin complex reagent (45 minutes) followed. Sections were then exposed to the chromagen reaction solution (0.035% diaminobenzidine in 10 ml Tris buffer, filtered, and brought to 0.03% H202) for 5 minutes. Sections were counterstained in Mayer's hematoxylin, dehydrated, cleared, and mounted. For certain antisera (see below), sections were pretreated with a 0.1% solution of trypsin (Sigma Co., St. Louis, Missouri) in phosphate-buffered saline with 0.1% CaCl for 30 minutes at 37 C. Sources of reagents and dilutions of antisera and modifications were as follows: avidin-biotin reagents (Vector Laboratory System, Burlingame, California); diaminobenzidine (Sigma Co.); antibody to cytokeratin (1 / 100, trypsinized, Lab Systems, Chicago, Illinois); desmin (1/450,2 hour room temperature; Dako, Santa Barbara, California); vimentin (1/20, Dako); muscle-specific actin (MSA, 1/4,000, Enzo, New York, New York); smooth muscle actin (SMA, 1/50, Dako); Factor VIII antigen (1/750, trypsinized, Dako); SlOO antigen (1/1,000, Dako); Factor 13a (1/1,000, Dako); epithelial membrane antigen (EMA, 1/50, Dako); and antibody to CD34 (1/25, Becton-Dickinson, Rutherford, New Jersey). Both negative and substitution serum controls and positive tissue controls were employed. A photograph of a post-vaccinal sarcoma in the thigh of a cat is shown in Fig. 1. The majority of the tumors were well demarcated and some were partially encapsulated. Those in the subcutis of the dorsal neck or thorax were often just ventral to and/or continuous with the panniculus carnosus muscle (Fig. 2). Histologically, the 46 post-vaccinal sarcomas included fibrosarcomas (23), malignant fibrous histiocytomas (1 3), osteosarcomas (7), rhabdomyosarcomas (2), and chondrosarcoma (1). Fibrosarcomas (FSA) and malignant fibrous histiocytomas (MFH) made up the bulk ofthe tumors, and there was much overlap between fibrosarcomas with