The contribution of immunohistochemical staining in tumour diagnosis
- Additional Document Info
- View All
With the increasing use of immunohistochemical stains in the diagnostic laboratory, it becomes relevant to review the contributions of this new technology in the area of tumour diagnosis. During 1986, our laboratory received 21,479 biopsy specimens of which 2013 were tumours. Nine hundred and fifty-eight tumours (47.5%) were subjected to immunohistochemical analysis. The biopsy and immunohistochemistry reports of 200 consecutive tumours, including 41 consultation cases, were retrospectively reviewed and assigned to one of the five categories. Immunohistochemical stains confirmed the preferred H & E diagnosis in 106 cases (53%); made the definitive diagnosis from a list of differential diagnosis in 29 cases (14.5%); provided contributory information in 36 cases (18%); were non-contributory in 27 cases (13.5%); and rendered an unsuspected diagnosis in two cases (1%). Immunohistochemical stains were particularly useful in distinguishing between malignant lymphoma and anaplastic carcinoma and in the identification of amelanotic melanoma. The application of a panel of antibodies chosen in accordance with the differential diagnoses considered was very useful in the typing of anaplastic round cell and spindle cell tumours. In 27 cases (13.5%), immunohistochemical stains were non-contributory. About half of these were referred cases and the failure to demonstrate the relevant antigens in normal tissues which served as in-built controls suggested that part of the problem may be due to differences in methods of fixation which led to sub-optimal preservation of tissue antigens. We conclude that immunohistochemical stains provide important and sometimes essential information for definitive typing of anaplastic tumours. Often the information derived was of therapeutic and prognostic relevance. We argue that this is a cost-effective test although we would caution that in all circumstances the interpretation of immunostaining must be made in the context of the histological as well as the clinical and laboratory data.
has subject area