Multivariate analyses of flow cytometric S-phase and ploidy as node-negative breast cancer prognostic factors : an international and multi-center study

Recently a set of ten adjustments that optimizes the prognostic strength of both DNA ploidy (P) and S-phase (S) was published (Cytometry, 46(3), 2001). Also presented was an optimal method of combining P and S (P+S) that stratifies node-negative patients into highly significant risk groups. The adjustments compensate for many unappreciated complexities in categorizing P into low and high risk groups and eliminate unwanted correlation between P and S. The purpose of this study is to examine P+S in the context of other well-known prognostic factors such as primary size (pT), estrogen and progesterone receptor (ER,PR) and menopausal status (MS). Methods: DNA histograms derived from frozen primary tumors and clinical databases were provided by Baylor College, n=935; Sweden, n=210 (Lund, Linkoping, Stockholm) and France, n=220 (Angers, Marseille, Saint Cloud, Tours). Time to metastasis was the tested clinical outcome. Results: Cox proportional hazards analysis of theBaylor data revealed P+S, p<0.000002, and pT, p<0.003, as independent significant prognostic factors. The Sweden study also showed P+S the mostsignificant prognostic factor, p<0.002, as well as MS, p<0.004 and ER, p<0.007. The French study results were MS, p<0.0005, P+S, p<0.002 and pT, p<0.007.A P+S, MS and pT prognostic model stratified patients in all studies into highly significant categories, Baylor, p<0.000005, Sweden, p<0.00001, and French, p<0.000005, with low and high risk 10-year relapse-free survival fractions of 0.92-0.69, 0.95-0.58 and 0.96-0.60 respectively. Conclusion: A combined P+S, MS and pT prognostic model is a powerful and reliable method of stratifying node-negative breast cancer patients into highly significant prognostic groups.