6th ESACP Congress, Heidelberg, April 7-11, 1999

A134
THE RELEVANCE OF DNA PLOIDY, BIOCHEMICAL MARKERS AND CLINICOPATHOLOGICAL FEATURES FOR PROGNOSIS PREDICTION IN BREAST CANCER
Theissig F, Kotzsch M, Dimmer V, Haroske G, Luther TH, Kunze KD

Institute of Pathology, Dresden University of Technology, Germany

The DNA cytometry has been established as a valuable tool for prognostic evaluaton of tumor material. From 799 primary breast cancers we analyzed simultaneously the DNA content of tumor cells by both image and flow cytometry. The overall coincidence between the determined DNA pIoidy with both methods was 71 % indicating a good correlation (r=0.68, p<0.001). In addition, we compared the DNA ploidy and the S-phase-fraction (SPF) to the lymph node status (LNS) at the time of operation and to the uPA antigen levels of tumor tissue extracts, since these are prognostic parameters with well documented prognostic impact. As expected we found with both methods a significant correlation (Chi Square, p<0.0007) to LNS with a higher percentage of aneuploid breast cancers with lymph node metastasis (60%) as compared to diploid tumors (42 %). The SPF determined by flow cytometry was also significantly correlated with the LNS (Mann Withney, p < 0.001). The uPA antigen level of both tetraploid (3.99 ng/mg protein) and aneupIoid (4.72 ng/mg) tumors was interestingly significantly higher as compared to diploid (3.39 ng/mg) tumors (Kruskall Wallis, p<0.001). To analyze the prognostic relevance of DNA measurements, the resuIts were evaluated in a subgroup of 99 patients with primary invasive ductal carcinomas (median follow up 29 ( 15 month). Whereas we found SPF as an independent prognostic parameter in univariate analysis (p = 0.037, cut off SPF=10 %), DNA-ploidy failed tightly to be significant (p=0.06). In the analyzed patient group, however, the usual clinicopathological prognostic markers such as LNS (p=0.004), Bloom-Richardson-Grading (p=0.0002) as well as the biochemical markers such as uPA- (p=0.038) and PAI-1-antigen levels (p=0.007) showed also significant correlation on disease-free survival in univariate analysis. In conclusion, the determination of DNA content by flow cytometric techniques in comparison to other prognostic parameters seems to be useful for prognostic evalution in breast cancer.