A DNA CYTOMETRIC PROLIFERATION INDEX IMPROVES THE VALUE OF THE DNA-PLOIDY PATTERN AS A PROGNOSTICATING TOOL IN PATIENTS WITH CARCINOMA OF THE PROSTATE

Citation
G. Ahlgren et al., A DNA CYTOMETRIC PROLIFERATION INDEX IMPROVES THE VALUE OF THE DNA-PLOIDY PATTERN AS A PROGNOSTICATING TOOL IN PATIENTS WITH CARCINOMA OF THE PROSTATE, Urology, 50(3), 1997, pp. 379-384
Citations number
18
Categorie Soggetti
Urology & Nephrology
Journal title
ISSN journal
00904295
Volume
50
Issue
3
Year of publication
1997
Pages
379 - 384
Database
ISI
SICI code
0090-4295(1997)50:3<379:ADCPII>2.0.ZU;2-Q
Abstract
Objectives. A still controversial issue is whether the results of a cy tometric assessment of the DNA distribution pattern of the nuclei of t he neoplastic parenchymal cells of a prostatic adenocarcinoma has addi tional prognostic value to that of the stage and grade of the disease. To increase the accuracy of the DNA ploidy assessments, image cytomet ry (ICM) has been used and combined with the determination of an ICM p roliferation index (PI) to increase its value as an additional prognos ticating tool. Methods. We investigated 96 patients, followed up since diagnosis in 1980/1981 until death or, in 11 surviving patients, for an average of 14.5 years. Survival analysis was made by the convention al Kaplan-Meier method. Fine-needle aspiration biopsy was used as the major diagnostic tool. The neoplastic cell nuclei were classified as I CM DNA diploid, tetraploid, or aneuploid by means of the ploidy-establ ishing peak in the ICM DNA histograms, as well as the fraction of turn er cells in the S-phase. Scattered cells to the right of the ploidy-es tablishing peak, the S-phase fraction, and those in the G2M area of th e ICM DNA histograms were counted as percent of the total number of tu mor cells; this percentage was defined as the PI. Arbitrarily, tumors with a PI less than 5% were classified as having a low proliferation r ate, those with a PI greater than 10% were considered highly prolifera ting, and those with a PI between 5% and 10% as carcinomas with an int ermediate proliferation potency. Results. By univariate analyses, clin ical stage, cytodiagnostic grade, cytometric DNA ploidy pattern and PI all had significant prognostic value. By multivariate analyses, the P I was found to add prognostic information to that of the ICM DNA ploid y pattern variable, giving it an increase in its statistical P value f rom 0.002 to 0.0005. As a consequence, the combination of these two va riables was found to give rise to three new patient groups with regard s to their prognosis: DNA group I had tumors with a diploid ICM DNA pa ttern with a low PI; DNA group II had tumors with a diploid or tetrapl oid ICM DNA tumor cell nuclei pattern with an intermediate PI; and DNA group III had a diploid or tetraploid ICM DNA pattern with high PI an d all tumors with an aneuploid pattern. By multivariate analysis, incl uding tumor grade and clinical stage, these new DNA groups (P = 0.0004 ) and M stage disease (P = 0.0006) were the only significant prognosti c variables. Conclusions. A DNA cytometric PI improves the prognostica ting value of DNA ploidy. Patients with prostatic adenocarcinomas, cla ssified as DNA group I, have a low risk of death from their neoplastic disease with deferred or hormonal treatment only. (C) 1997, Elsevier Science Inc. All rights reserved.