HIGH-RESOLUTION IMAGE-CYTOMETRIC DIFFERENTIATION OF HORMONE-SENSITIVEAND HORMONE INSENSITIVE PROSTATIC CARCINOMAS

Citation
U. Schenck et al., HIGH-RESOLUTION IMAGE-CYTOMETRIC DIFFERENTIATION OF HORMONE-SENSITIVEAND HORMONE INSENSITIVE PROSTATIC CARCINOMAS, Analytical cellular pathology, 7(2), 1994, pp. 139-152
Citations number
23
Categorie Soggetti
Cytology & Histology",Pathology
ISSN journal
09218912
Volume
7
Issue
2
Year of publication
1994
Pages
139 - 152
Database
ISI
SICI code
0921-8912(1994)7:2<139:HIDOH>2.0.ZU;2-U
Abstract
Until the present it was not possible to predict hormone sensitivity o f prostatic carcinoma. Based on studies correlating image cytometric r esults of hormone receptor negative and hormone receptor positive brea st carcinomas, the present study aims at separating responders and non -responders to hormone therapy in metastatic prostatic carcinoma. From May-Grunwald-Giemsa stained slides of fine needle aspirates of 23 pat ients with metastasizing prostatic carcinoma about 100 nuclei per slid e were taken by TV camera for image-cytometric processing. One thousan d and twenty-two nuclei came from 10 patients who showed tumour regres sion for at least 36 months and who all survived for more than 5 years . One thousand three hundred and thirty-two nuclei were from prostatic aspirates of patients who showed a continuous tumour progression desp ite receiving hormone therapy. All patients of the latter group died w ithin 5 years. A correct classification of the patient groups of respo nders and non-responders was possible in 19-21 of 23 cases by means of high resolution image analysis including nuclear structural features. It was found that even simple planimetric features, like the nuclear perimeter, or densitometric features, such as the total nuclear extinc tion, differed markedly between the two groups. The data show that nuc lei from hormone sensitive prostatic carcinoma are distinct from those of non-sensitive ones in the present series. The interpretation of re sults must take into account that the very strict criteria for hormone sensitivity leads to a highly selected patient group. The application of the method to an unselected patient group can be presumed to yield a higher rate of false classifications.