PET imaging allowed accurate differentiation between benign and malignant breast tumors providing a high specificity. Sensitivity for detection of small breast cancer ( < 1 cm) was limited due to partial volume effects. Quantitative image analysis combined with partial volume correction may be necessary to exploit fully the diagnostic accuracy. PET imaging may be helpful as a complimentary method in a subgroup of patients with indeterminate results of conventional breast imaging.

Our results provide evidence that the presence of single lung carcinoma cells in lymph nodes is an independent indicator of the disseminatory capacity of an individual primary tumor. Immunohistochemical assessment of micrometastases in lymph nodes is recommended for current tumor staging in NSCLC, as it might lead to better stratification of patients for adjuvant therapy.

In clinical practice, PET imaging cannot substitute for histopathologic analysis in detecting axillary lymph node metastases in breast cancer patients. PET imaging, however, improves the preoperative staging of the disease in breast cancer patients and, therefore, might alter therapeutic regimen options.

Summary MIBl proliferation rate (MIBl-PR) and total S-phase fraction (SPF) were retrospectively determined in formalin-fixed, paraffinembedded sections of 90 primary node-negative breast carcinomas. None of the patients had received adjuvant systemic therapy. Median follow-up in patients still alive at the time of analysis was 37.5 months (1 6-72 months). lmmuno6taining of Ki-67 antigen was performed using the monoclonal antibody MIBl and the APAAP technique. An adjacent 50-gm paraffin section was used for flow cytometric S-phase determination. Results were compared to established clinicopathological prognostic factors. MIBl -PR was significantly correlated to grading (P = 0.018); SPF was significantly correlated with tumour size (P = 0.041) and inversely with steroid hormone receptor status (P = 0.03). A significant correlation between MIBl -PR and SPF was found in aneuploid (P = 0.025) but not in diploid tumours (P = 0.164). In univariate analysis, both MIBl -PR (optimized cut-off of 25%) and SPF (optimized cut-off of 8%) were significant prognostic factors for disease-free survival (DFS) (MIBl -PR, P = 0.0224; SPF, P = 0.0028). In multivariate analysis, however, only SPF remained significant; it was the strongest prognostic factor for DFS (P = 0.0073), stronger than MIBl -PR or established clinicopathological prognostic factors. We thus conclude that MIBl-PR and SPF provide additional prognostic information in node-negative breast cancer. However, in our study, flow cytometrically determined SPF had the greater prognostic impact.

No abstract