RESUMO
BACKGROUND: Large retrospective archival studies of S-phase fraction (SPF) measured by DNA flow cytometry in patients with breast carcinoma have shown its long-term prognostic relevance. However, contradictory results have also been reported, some of them being related to the various methods of SPF calculation using different commercially available software. METHODS: DNA flow cytometric list mode data, initially computed with Cellfit software, were blindly reanalyzed using Modfit and Multicycle software. The data, acquired prospectively between 1990 and 2003 from cytologic fine-needle aspiration biopsy samples of 397 patients with breast carcinoma, were compared with patient outcome with a median follow-up of 99 months (8.2 yrs). RESULTS: Measurement of SPF was successful in 321 (81.7%), 362 (92.1%), and 335 cases (85.2%) by means of the Cellfit, Modfit, and Multicycle software programs, respectively. In 306 cases (77.9%), SPF values were obtained using all 3 methods. Comparisons between SPF measurements showed a good agreement between Modfit and Multicycle computations. In the series of 306 patients, SPF median values of 2.5%, 4.3%, and 5.45% for Cellfit, Modfit, and Multicycle, respectively, were also found to be statistically different. Regardless of the software used, high SPF defined as above the median value was an independent factor of prognosis in a multivariate analysis including all traditional clinicopathologic parameters. It is noteworthy that this was also observed in the subgroups of patients either treated by primary surgery for an early tumor (n = 133) or by neoadjuvant chemotherapy for a locally advanced breast carcinoma (n = 173). CONCLUSIONS: The data in the current study supported the prognostic relevance of SPF measurement in predicting the long-term overall survival of patients with early-stage or locally advanced invasive breast carcinoma.