Phyllodes tumours of the breast--a retrospective study from 1982-2000 of 50 cases in Portsmouth.

INTRODUCTION: This study is a retrospective analysis of 50 phyllodes tumours to determine the optimal surgical procedure for these types of tumours. We have also reviewed rates of recurrence, metastases and mortality based on choice of procedure and histological type. PATIENTS AND METHODS: Cases were ascertained from pathology databases and clinical details extracted from the hospital records. Fifty patients with phyllodes tumours were identified. These comprised 29 benign, 12 borderline and 9 malignant phyllodes tumours. RESULTS: All benign phyllodes tumours were treated with breast-conserving surgery, these included 16 tumours over 40 mm. Borderline and malignant lesions were treated by breast-conserving surgery or mastectomy. The median follow-up period was 35 months (range, 4-96 months). The recurrence rate for all tumours was 14%. Malignant and borderline phyllodes tumours had a recurrence rate of 28%. Tumours excised with a wide margin did not seem to recur. Breast-conserving surgery appeared to be as effective as mastectomy. The choice of procedure was less important than the width of the excision margin. Recurrence occurred in 1/29 benign tumours. Excision margin width did not influence rate of recurrence. One patient died of metastases after mastectomy. CONCLUSIONS: Breast-conserving surgery is the treatment of choice for all benign lesions. For borderline and malignant lesions, excision with a wide margin reduces the rate of recurrence. If a diagnostic local excision biopsy or enucleation is performed, it should be followed by a definitive wider excision.

Detection of micrometastases in lymph nodes from patients with breast cancer.

BACKGROUND: Sentinel node biopsy affords the opportunity of focused examination of lymph nodes, including the use of the reverse transcriptase-polymerase chain reaction (RT-PCR). The mammaglobin gene is expressed by breast cancers but has not been detected in histologically normal lymph nodes. This study compared mammaglobin RT-PCR with routine histology in the sentinel and non-sentinel nodes of patients with breast cancer. METHODS: Patients with breast cancer underwent tumour excision, sentinel node biopsy and axillary dissection. All nodes were bisected and half of each node was sent for routine histological examination. The other half underwent RNA extraction and mammaglobin RT-PCR. RESULTS: Sentinel node biopsy was successful in 50 (96 per cent) of 52 patients. Mammaglobin expression was detected in nine (8 per cent) of 119 histologically negative sentinel nodes (Clopper-Pearson 95 per cent confidence interval (c.i.) 4 to 14 per cent) and in 13 (5 per cent) of 247 histologically negative non-sentinel nodes (95 per cent c.i. 3 to 9 per cent). Mammaglobin expression was detected in four (13 per cent) of 31 patients with histologically negative sentinel nodes (95 per cent c.i. 4 to 30 per cent) and in six (14 per cent) of 44 patients with histologically negative non-sentinel nodes (95 per cent c.i. 5 to 27 per cent). The false-negative rate for sentinel node biopsy was zero using histology results and 10 per cent using RT-PCR. CONCLUSION: RT-PCR screening of axillary nodes for mammaglobin expression increased the detection of breast cancer metastases compared with routine histology.