THE IMPACT OF HISTOPATHOLOGY ON NODAL METASTASES IN MINIMAL BREAST-CANCER

Objective: To establish a model based on risk factor analysis to guide selective axillary lymph node dissection in patients with Tla and Tlb breast cancers.Design: Retrospective review to determine histopatholo gic features and patient demographic profiles that may influence the i ncidence of nodal metastases. Setting: Primary care and referral cente rs in Rhode Island and Massachusetts. Patients: Women with invasive br east cancers with nodal status reported to the statewide tumor registr y, the Hospital Association of Rhode Island, and the tumor registry at Baystate Medical Center, Springfield, Mass, between January 1984 and December 1995. There were 12 030 patients with breast cancer reported; 2185 (18%) of these had tumors 1 cm or less in diameter. Intervention s: None. Main Outcomes Measure: Axillary node metastases. Results: The nodal status of 377 patients with Tla tumors and 1808 patients with T lb tumors was studied. Seventy-five percent had axillary dissections, and 16% were found to have nodal metastases. Thirty-one percent (29/93 ) of patients younger than 40 years had positive nodes compared with 1 5% (241/1546) of older patients (P=.001). The Tla tumors had fewer met astases than the Tlb tumors did (11% vs 17%; P=.02). Nuclear grade was available in 49% of cases. Nuclear grades 2 and 3 were associated wit h nodal involvement twice as often as grade 1 tumors were (P=.002). Pa tients with no poor prognostic factors had a 7% or less chance of noda l involvement, while patients with all 3 poor prognostic indicators ha d a 33.5% chance of nodal involvement. Conclusions: Selective nodal di ssection may be possible through risk factor analysis. Prospective reg istration of complete histopathologic information will allow more comp rehensive analysis and may further enhance the selective treatment of patients with minimally invasive breast cancer.