PURPOSE: To assess the prognostic value of DNA ploidy and nuclear morphometry in metastatic prostate cancer after androgenic deprivation treatment. METHODS: Fifty four patients with prostate cancer and bone metastases who had undergone androgenic suppression treatment were retrospectively studied. The deoxyribonucleic acid (DNA) content was analysed by flow cytometry. Nuclear morphometry characterized 14 nuclear descriptors. The study also included age, Gleason score, T classification, haematocrite, serum albumin, serum alkaline phosphatase, serum prostatic acid phosphatase and the amount of metastatic foci detected during radioisotope bone scan. Univariate survival analyses were performed and Cox's proportional hazards model was used to identify significant prognostic factors. To assess how the experimental factors improve the capacity of the classical factors for predicting the patients who reach median survival, logistic regression multivariate analysis was performed for the classical prognostic factors only and after added experimental variables (DNA content and Nuclear Area). RESULTS: The univariate survival analyses assigned a prognostic value to T category, level of albumin, alkaline phosphatase, Gleason score, bone scan, DNA ploidy and mean nuclear area. In the case of the Cox regression model only Gleason score, bone scan, mean nuclear area and DNA ploidy provided independent prognostic information. In logistic regression for classic prognostic factors only Gleason score is significant (sensibility 89.3%, specificity 64%). However, when the experimental factors are added, in addition to Gleason score, radioisotope bone scan and DNA ploidy are of prognostic value (sensibility 90% and specificity 72%). CONCLUSIONS: The study of DNA content and nuclear morphometry in the primitive tumor provides independent prognostic information in survival analysis for patients with metastatic prostate cancer. However, there is limited improvement with respect to the classical factors in predicting survival. This questions its utility in the daily clinical usage.
PURPOSE: To assess the prognostic value of DNA ploidy and nuclear morphometry in metastatic prostate cancer after androgenic deprivation treatment. METHODS: Fifty four patients with prostate cancer and bone metastases who had undergone androgenic suppression treatment were retrospectively studied. The deoxyribonucleic acid (DNA) content was analysed by flow cytometry. Nuclear morphometry characterized 14 nuclear descriptors. The study also included age, Gleason score, T classification, haematocrite, serum albumin, serum alkaline phosphatase, serum prostatic acid phosphatase and the amount of metastatic foci detected during radioisotope bone scan. Univariate survival analyses were performed and Cox's proportional hazards model was used to identify significant prognostic factors. To assess how the experimental factors improve the capacity of the classical factors for predicting the patients who reach median survival, logistic regression multivariate analysis was performed for the classical prognostic factors only and after added experimental variables (DNA content and Nuclear Area). RESULTS: The univariate survival analyses assigned a prognostic value to T category, level of albumin, alkaline phosphatase, Gleason score, bone scan, DNA ploidy and mean nuclear area. In the case of the Cox regression model only Gleason score, bone scan, mean nuclear area and DNA ploidy provided independent prognostic information. In logistic regression for classic prognostic factors only Gleason score is significant (sensibility 89.3%, specificity 64%). However, when the experimental factors are added, in addition to Gleason score, radioisotope bone scan and DNA ploidy are of prognostic value (sensibility 90% and specificity 72%). CONCLUSIONS: The study of DNA content and nuclear morphometry in the primitive tumor provides independent prognostic information in survival analysis for patients with metastatic prostate cancer. However, there is limited improvement with respect to the classical factors in predicting survival. This questions its utility in the daily clinical usage.
Authors: Girish Venkataraman; Vijayalakshmi Ananthanarayanan; Gladell P Paner; Rui He; Saeedeh Masoom; James Sinacore; Robert C Flanigan; Eva M Wojcik Journal: Virchows Arch Date: 2006-08-03 Impact factor: 4.064