Small Primary Breast Cancers and the Need for Axillary Node Dissection


Synopsis: Patients with small primary breast cancers have a more favorable prognosis, but axillary node metastases are still relatively common. To determine if a subset of patients with primary tumors of less than 1 cm could be identified for whom the likelihood of axillary node involvement was so low that axillary node dissection would not be needed, an analysis of clinical prognostic factors was performed on a prospective database of 919 patients with T1 lesions presenting at a single institution. Although several factors offered predictive value with regard to nodal metastases, none alone, or in combination would preclude the need for axillary node surgery.

Source: Rivadeneira DE, et al. J Am Coll Surg 2000;191:1-8.

A relatively small percentage of women who present with small primary breast cancers will have axillary lymph node metastases (ALNM) and there has been some effort to define those at risk, and spare node dissection surgery for those who are not. The purpose of this report was to attempt to define a subset of those with primary tumors equal to, or less than 1 cm who might be at such low risk for axillary node involvement that surgery would not be required.

Between 1990 and 1996 a total of 919 women with small primary (< 1 cm) invasive breast cancers were operated upon at the Weill Medical College of Cornell University (New York Presbyterian Hospital), or the Strang-Cornell Breast Center in New York. Of these, 199 (21.7%) had T1a tumors (< 0.5 cm) and 720 (78.3%) had T1b tumors (0.6-1.0 cm). All patients had axillary node dissections (levels I and II), and metastases were discovered in 165 patients (18.0%). Of these, 32 (19.4%) had T1a tumors and 133 (80.6%) had T1b tumors. Thus, 32 of 199 T1a patients (16.0%) and 133 of 720 T1b patients (18.5%) had surgically defined metastatic axillary disease.

Characteristics included in this analysis were: age, race, tumor size, palpable vs. non-palpable lesion, histological type and grade, lymphatic or vascular channel invasion (present or absent), and estrogen and progesterone receptor status (positive or negative). The total number of lymph nodes per patient was not evaluated, nor was the influence of the number of nodes examined upon the frequency of metastases assessed.

By univariate analysis, four variables were found to be significant predictors of axillary node involvement. These were: increasing tumor size, poor histologic grade, presence of lymphatic or vascular invasion, and younger age. Under multivariate analysis, an increased risk for axillary lymph node metastases was demonstrated with increasing tumor size (0.1 cm increments), poor histologic grade, and younger age.

Rivadeneira and collegues were unable to use these selected factors to identify a subset of individuals that would have a negligible risk of axillary node metastases. For example, in the 108 patients 50 years and older (favorable) with T1a lesions (favorable, when compared to T1b lesions) that were well differentiated (favorable histologic grade) and without vascular or lymphatic invasion, 14 (13%) were still identified as having axillary node disease. Thus, Rivadeneira et al conclude that for patients with T1a or T1b primary breast cancers, axillary node investigation remains warranted, either by traditional axillary node dissection, or by intraoperative lymphatic mapping and sentinel lymph node biopsy techniques.

COMMENT By William B. Ershler, MD

The take home message from this report is that the commonly appreciated clinical variables do not allow sufficient precision in predicting axillary node status and, to the extent that additional therapy is predicated on the presence or absence of axillary nodes, axillary node dissection remains warranted. Certainly, axillary node status remains the single most important prognostic indicator in these patients.1 As demonstrated in the NSABP-B04 trial, the surgery itself does not offer survival benefit, but the results of axillary node dissection allow the identification of those individuals who are likely to benefit from adjuvant radiation and chemotherapy.

The results of this analysis are in accordance with a tumor registry study of an even larger cohort of patients (n = 2,185 patients) with T1 invasive breast carcinoma.2 In that series the overall frequency of axillary node metastases was 16% and age and nuclear grade (available in 49% of cases) also appeared to have predictive value. Examining patients with all three poor prognostic factors, 34% had axillary node involvement, whereas for those with none of the poor prognostic factors, 7% were found to have axillary nodes positive for tumor.

The rationale for defining axillary node status in patients with T1 lesions can be derived from data provided by investigators at Memorial Sloan Kettering.3 The 20-year recurrence rate for T1 invasive carcinomas that were axillary node negative was 12%, compared with T1 invasive carcinomas that were axillary node positive, in which the 20-year recurrence rate was 39%. Thus, determining nodal status, even for those with these small primary tumors, has important prognostic and therapeutic indications.

Perhaps new approaches, such as intraoperative mapping and sentinel node biopsy will prove useful in this population, and reduce the need for more extensive axillary node dissection.4 However, short of that, it appears that axillary node dissection is still an indicated approach for patients with primary breast cancer, even those with T1 lesions.


1. NIH Consensus Development Panel. Monogr Natl Cancer Inst 1992;11:1-5.

2. Mustafa IA, et al. Arch Surg 1997;132:384-391.

3. Rosen PP, et al. Ann Surg 1981;193:15-25.

4. Hsueh E, et al. J Am Coll Surg 1999;189:207-213.