The histologic classification of 602 cases of feline lymphoproliferative disease using the National Cancer Institute...

Case information and histologic slides for 688 admissions of feline tissues from 12 veterinary institutions were assembled and reviewed to determine tissues obtained by biopsy or necropsy, age and sex of cat, tumor topography, feline leukemia viral antigen status, histologic frequency of mitoses, diagnosis, presence of necrosis, and presence and degree of sclerosis. Histologic sections were examined to place the lesions in one of the diagnostic categories of the National Cancer Institute working formulation (NCI WF) for lymphomas or lymphoid leukemia. Correlations between the various factors determined were tested using contingency tables and chi-square analysis to provide a statistical comparison between the levels of observations determined by case examination with the numbers expected from chance alone. Significant correlations (P < or = 0.05) were found between diagnosis and tumor topography, the frequency of mitoses, necrosis, sclerosis, and age, between mitoses and necrosis, topography, age, and feline leukemia viral infection status, between topography and necrosis and age, and between leukemia viral status and age. Significant correlations between diagnosis and tumor topography included a greater than expected number of cases of acute and chronic lymphoid leukemia and multicentric distribution of tumor. Small cell lymphomas were more frequent than expected in enteric and cutaneous areas and less frequent than expected in mediastinal, renal, and multicentric areas. In contrast, the high-grade small noncleaved type of lymphomas was found significantly more frequently than expected in the mediastinum and less frequently than expected in enteric tissues. In comparing diagnosis and frequency of mitoses, the lymphomas classified as low grade by the NCI WF were significantly more frequent than expected in the lower categories (0-2/100x) of mitoses, and those classified as high-grade lymphomas were more frequent than expected in the higher categories (4-8/1OOx) of mitoses. In comparing diagnosis and sclerosis, diffuse sclerosis was more frequent than expected for the intermediate grade lymphomas of mixed cell type and for the high-grade lymphomas of the immunoblastic polymorphous type. In comparing diagnosis and locally extensive necrosis, this feature was more frequently observed than expected for cases of intermediate grade lymphoma of the small-cleaved cell category and for the high-grade lymphoma of the immunoblastic cell type. In comparing mitoses and necrosis, the lower grade lymphomas were, in general, characterized by a lower frequency of mitoses and a lower incidence of necrosis then would be expected from chance alone. In contrast, the higher grade lymphomas were characterized by more frequent mitoses and a higher incidence of necrosis. In tests comparing mitoses and tumor topography, lymphomas of the alimentary tract were more frequently observed than expected in the category with the lowest level of mitoses (0-1/100x), whereas lymphomas of the mediastinum and kidney were more frequently observed than expected in the categories with a higher level (4-20/ 100x) of mitoses.