To evaluate the prognostic significance of the axillary apex biopsy and its impact on clinical practice, a retrospective analysis was performed in 875 patients with clinically operable breast cancer who underwent this procedure from 1977 to 1985 (165 TNM stage I; 512 TNM stage II; 198 TNM stage IIIA). Apex biopsy is performed as a staging procedure. Apex biopsy positive patients are treated by radiotherapy alone, while apex biopsy negative patients are treated with breast conserving therapy or mastectomy, both including complete axillary dissection. The apex biopsy was tumour positive in 4% of TNM stage I patients; 17% of stage II patients and 40% of stage IIIA patients. Among patients with clinically node-negative disease, the apex biopsy was positive in 12%; in patients with palpable suspected lymph nodes this figure was 45%. Actuarial 8 y survival rates for patients with stage I, II and III disease and a negative apex biopsy were 83, 70 and 50%, respectively. The corresponding figures for patients with a positive apex biopsy were 60, 28 and 14%. In a multivariate analysis, a positive apex biopsy, clinical N classification and T classification were independent prognostic factors for survival (P < 0.0001). We conclude that a positive apex biopsy is rare in clinical stage I breast cancer, and that in patients with TNM stage II and III disease the procedure is an important tool to assess prognosis pre-operatively.