Premium
Nuclear shape and prognosis following orchiectomy in stage D 2 prostate cancer
Author(s) -
Miller John,
Horsfall David J.,
Marshall Villis R.,
Skinner John M.,
Rao Dora M.,
Leong Anthony S. Y.
Publication year - 1994
Publication title -
the prostate
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.295
H-Index - 123
eISSN - 1097-0045
pISSN - 0270-4137
DOI - 10.1002/pros.2990240606
Subject(s) - prostate cancer , medicine , orchiectomy , stage (stratigraphy) , prostate , discriminant function analysis , survival analysis , cancer , mann–whitney u test , urology , oncology , pathology , biology , paleontology , machine learning , computer science
In this study, we have examined whether tumor grade and morphometric nuclear features can predict the outcome of treatment by orchiectomy in patients with stage D 2 prostate cancer. Two outcome groups based on duration of survival postorchiectomy were examined, a bad outcome group of 63 patients who died from prostate cancer within 12 months and a good outcome group of 34 patients who survived beyond 5 years. Tumors were histologically classified as well (17%), moderate (17%), or poorly differentiated (66%). Tumor grade and patient outcome were significantly associated (Mann‐Whitney test; P < 0.005), with 76% of poorly differentiated tumors in the bad outcome group, and 65% of well‐differentiated tumors in the good outcome group. Using discriminant function analysis, tumor grade correctly predicted outcome in 70% of cases. A statistically significant difference was also detected in nuclear shape values between the two outcome groups ( P < 0.05) and histological grades ( P < 0.05). Using discriminant function analysis, 51% of cases were correctly classified into outcome groups using nuclear shape factors, a figure which rose to 65% when all nuclear morphometric features were used. This demonstrates that nuclear morphometric features are of no clinical value in predicting the outcome of treatment in stage D 2 disease. Furthermore, these evaluations cannot select patients who might be spared orchiectomy on the basis of a predicted poor response. However, nuclear shape and variance measurements of benign glandular epithelial cells within cancerous prostates were significantly different from those of malignant cells ( P < 0.005). We conclude that, while video image analysis of prostatic nuclear shape can reliably discriminate between benign and malignant cells, nuclear morphometric features are of minimal prognostic value in men with stage D 2 prostate cancer treated by androgen ablation. © 1994 Wiley‐Liss, Inc.