A study was carried out to test the accuracy and consistency of veterinary pathologists, not specialists in hematopathology, in applying the World Health Organization (WHO) system of classification of canine lymphomas. This study represents an initiative of the ACVP Oncology Committee, and the classification has been endorsed by the World Small Animal Veterinary Association (WASVA). Tissue biopsies from cases of canine lymphoma were received from veterinary oncologists, and a study by pathologists given only signalment was carried out on 300 cases. Twenty pathologists reviewed these 300 cases with each required to choose a diagnosis from a list of 43 B and T cell lymphomas. Three of the 20 were hematopathologists who determined the consensus diagnosis for each case. The 17 who formed the test group were experienced but not specialists in hematopathology, and most were diplomates of the American or European Colleges of Veterinary Pathology. The overall accuracy of the 17 pathologists on the 300 cases was 83%. When the analysis was limited to the 6 most common diagnoses, containing 80% of all cases, accuracy rose to 87%. In a test of reproducibility enabled by reintroducing 5% of cases entered under a different identity, the overall agreement between the first and second diagnosis ranged from 40 to 87%. The statistical review included 43,000 data points for each of the 20 pathologists.

Epidermal growth factor receptor (EGFR) overexpression occurs in a significant percentage of cases of glioblastoma multiforme (GBM), and amplification has been found in approximately 40% of these neoplasms. Controversy exists as to the prognostic significance of EGFR gene amplification: some reports have indicated that amplification is associated with a poor prognosis, while other authors have reported no relationship between gene amplification and prognosis. Some reports have found a poor prognosis to be associated with amplification of the EGFR gene in patients of all ages with GBM, while other authors have found EGFR amplification to be an independent predictor of prolonged survival in patients with GBM who are older than 60 years of age. The authors studied a series of 34 specimens (32 patients) with histologically proven GBM by immunohistochemistry for the presence of EGFR overexpression and by fluorescence in situ hybridization (FISH) for gene amplification of the EGFR gene. Results of these studies and data on patient age, sex, functional status, therapy, and survival were correlated to determine which variables were predictive of survival. p53 expression was also determined by immunohistochemistry and correlated with the other variables and survival.

We reviewed fine-needle aspiration (FNA) samples of metastatic tumor in the pancreas from nonhematologic neoplasms over a 5-year period. In 1,050 total procedures, 20 metastases were diagnosed: 9 renal-cell carcinomas (RCCs), 3 melanomas, 2 pulmonary small-cell carcinomas, 2 breast carcinomas, 1 prostate carcinoma, 1 colon adenocarcinoma, 1 pulmonary squamous-cell carcinoma, and 1 gastrointestinal stromal tumor. A wide range of latency from primary diagnosis was noted; the longest was RCC at 12.6 years (range, 5-28). Sites of involvement were: 13 heads, 4 bodies, and 3 tails. Eighteen cases presented as a solitary mass. The average size was 4.7 cm (range, 1.5-9.8), and a case of RCC (9.8 cm) was the largest. In seven cases, the clinical and radiographic impression was of a pancreatic primary. We conclude that metastases to the pancreas are rarely diagnosed by FNA and may clinically mimic a pancreatic primary.