PROGNOSTIC VALUE OF QUANTITATIVE CYTOMETRY IN A SERIES OF 415 T1T2 N0N1/M0 BREAST-CANCER PATIENTS - PRELIMINARY-RESULTS/

Identifying prognostic markers in local regional breast carcinomas rem ains an important challenge today. DNA content obtained by flow cytome try, has been found to be of prognostic value; results with other meth ods remain less clear. This report describes DNA image cytometry patte rns which are assessed with respect to disease-free survival. From Jun e 1982 to December 1992, 415 patients under 75 years of age, without a ny previous or synchronous carcinoma, suffering from an invasive breas t cancer classified as T1 (52.8%), T2 (47.2%), N0 (65.1%) N1 (34.9%), MO according to clinical TNM staging, were enrolled in this study. The median age was 53 (28-75) and 58.8% of the patients were premenopausa l; 85.3% underwent a breast conservative procedure and 14.7% a modifie d radical mastectomy followed by postoperative irradiation. Histologic al axillary lymph node status, Scarff-Bloom grade and/or cytological g rade and, oestrogen receptor content were used in decision-making for adjuvant treatment: hormonotherapy (48%) or chemotherapy (18.8%). Impr ints were taken from the macroscopically visible lesion at the time of surgery, and a Feulgen staining was carried out on air dried smears t o be analysed using the Samba 200 cell image processor (Alcatel TITN, France). Five parameters were systematically assessed: proliferation i ndex; DNA histogram, integrated optical density, DNA malignancy grade, ploidy balance. With a median follow-up of 36 months (0-105), prolife ration index (P = 0.0008), DNA histogram (P = 0.0017), integrated opti cal density (IOD) (P = 0.018) and DNA malignancy grade (P = 0.017) had a significant prognostic value on disease-free survival estimated by the Kaplan-Meier method. When these parameters were included in a Cox proportional regression hazards model, PR (P = 0.01), Scarff-Bloom his tological grading (P = 0.02), axillary clearance (P = 0.04) were signi ficant; however, in the same model, taking into account the axillary l ymph node histological status, IOD was significant for pN- patients (P = 0.03), and proliferation index (P = 0.03) was significant for pN+. Such results need to be updated with a longer median follow-up, but th ey suggest that the mean DNA content, as measured by the integrated op tical density (IOD), should be considered when deciding on medical adj uvant treatment with respect to patients with a negative axillary clea rance. Copyright (C) 1996 Elsevier Science Ltd