ONE OR MULTIPLE SAMPLINGS FOR FLOW CYTOMETRIC DNA ANALYSES IN BREAST-CANCER PROGNOSTIC IMPLICATIONS

Flow cytometric assessments of DNA ploidy status and the S-phase fract ion (SPF) have been shown to yield prognostic information in breast ca ncer. The aim of the present investigation was to elucidate the reprod ucibility of results with regard to tumor DNA heterogeneity, and to as certain whether the prognostic value of DNA measurements might be enha nced by analyzing two pieces of a tumor instead of one. Agreement with regard to ploidy status (diploid versus non-diploid) was obtained in 90% of cases (71/79) when two adjacent sections of the tumor were inve stigated, and in 77% of cases (10/13) when four biopsies from differen t quadrants of the tumor specimen were investigated. The corresponding figures for agreement in SPF (divided into three categories, < 7.0%, 7.0-11.9%, and greater-than-or-equal-to 12%) were 75% (59/79; 2-sample series) and 55% (7/13; 4-biopsy series). The main reason for variance in ploidy results was the difficulties in distinguishing near diploid cell populations. Discrepancies in SPF categories could be explained by minor fluctuations in SPF values near the cut-off levels, or by var iance in ploidy status, the fraction of non-diploid nuclei, and backgr ound noise due to cell debris. There was a stepwise increase in recurr ence rate (RR) among patients with increasing SPF category (RR: 20%, 4 1%, and 53%). Patients whose SPF categories varied, from low or interm ediate in one part of the tumor to high in another, seemed to have a p oor prognosis (RR = 57%). We conclude that the variation in ploidy sta tus and SPF between different parts of the same tumor specimen is larg ely to be explained by uncertainty in the interpretation of the DNA hi stograms and by difficulties inherent in the estimation of SPF, rather than by the existence of DNA tumor heterogeneity. Evaluation of a dif ferent part of the same tumor specimen may be of value, especially in the case of SPF, as this may yield additional prognostic information. These findings also emphasize the need for good quality control of FCM DNA analysis.