After breast cancer (BC) screening was introduced for all Swedish women in 1989-90 the number of early invasive BC, clin N0 rose dramatically. For these BC it was known that only 10 % or less was N+ after axillary dissection and the morbidity for axillary surgery was substantial. Omitting axillary dissection in a randomized trial was ruled out because of anticipated few events. A national cohort was decided for invasive unifocal BC, histological grade 1 or 2, T=max 10mm, free margins, clin node neg. No axillary dissection should be undertaken. This was before the Sentinel Node era. Adjuvant treatment could be given according to regional guidelines for early breast cancer. Not uniform concerning postoperative radiotherapy and antihormonal treatment. Follow-up was clinical the first 5 years with mammography and then by hospital records and national Death register. Primary aim: Number of axillary recurrence and BC specific survival.

After national breast cancer screening was introduced in Sweden in 1989-90 the number of early invasive BC, clin N0 rose dramatically. Only 10 % or less of these was N+ after axillary dissection with a substantial morbidity after the axillary intervention. Omitting axillary dissection in a randomized trial was ruled out because of anticipated few events. A national cohort was decided for invasive unifocal BC, Screening or clinical diagnosis were eligible, only histological grade 1 or 2, T=max 10mm, free margins , clinical node negative. No axillary dissection should be undertaken, only radical removal of the breast cancer. (Note; This was before the Sentinel Node era). Adjuvant treatment could be given according to regional guidelines for early breast cancer as no regular national guidelines existed then concerning postoperative radiotherapy and antihormonal treatment. Chemotherapy was not recommended to this type of breast cancer. Follow-up was done by the treating surgeon the first 5 years including mammography. Then hospital records including mammography and reports from Pathology and Oncology and Death register were followed at intervals 5, 10 and 15 years. Primary aim: Number of axillary recurrence and distant metastases/ BC specific survival. Correlation to adjuvant therapy. Inclusion started in 1997 and ended in 2002. 1584 patients were included. If the axillary recurrence rate was more than 1% per year the first five years the inclusion should be stopped.

Inclusion Criteria:

• Invasive breast cancer, unifocal
• T=max 10mm
• Clinical node negative and clinical M0
• Histological grade 1 or 2
• Both screening / clinical detected accepted
• Radical operation = free margins in breast specimen
• Partial mastectomy and mastectomy accepted
• Fit to understand inclusion criteria

Exclusion Criteria:

• Previous breast cancer diagnosis, including carcinoma in situ
• Previous ipsilateral axillary dissection
• Non radical operation = no free margins
• Multifocal cancer
• Previous treatment for other cancer.
• Unable to understand information (informed consent)
• Bilateral breast cancer at diagnosis

In five out of six regions in Sweden (West was involved in another international trial) patients were selected at the local breast cancer conference as eligible in this cohort (study approved by the Ethical board and the Swedish Breast cancer group) and then informed by the surgeon. If the patient accepted the patient was included (registered at the regional Cancer Registry) and the primary breast surgery was done without any axillary dissection. Follow-up included regular visits to the breast cancer unit and yearly mammography for 5-10 years and then biannually to 74 years of age. Thereafter only hospital records (including surgery, oncology and pathology) were reviewed at intervals as were the national Death registry. Median age at study start was 60 years.