Parameters that predict nipple involvement in breast cancer

Breast conservation therapy has been shown to produce survival rates equivalent to those seen with modified radical mastectomy. Synchronous occult neoplastic involvement of the nipple may lead to incomplete excision of the tumor in patients undergoing breast conservation therapy, possibly leading to recurrence.

The charts of 803 breast cancer patients treated between 1990 and 1995 at two teaching hospitals were retrospectively reviewed. The patients were divided into three groups: nipple-positive for malignancy (n = 54), nipple-negative for malignancy (n = 404), and nipple-not-removed (n = 345). Ten different clinical and tumor parameters including age, race, primary tumor location, histologic grade, primary tumor size, nodal involvement, TNM stage, estrogen receptor status, DNA ploidy, and S-phase were examined for the ability to predict cancerous nipple involvement.

Overall, the rate of nipple positivity was 12%. In univariate analysis pathologic stage, tumor size, lymph node status, histologic grade, and tumor location were significant predictors of positive nipple involvement. Patients with tumors that were stage III or higher were nearly ten times (odds ratio [OR] = 9.8, 95% confidence interval [CI] = 5.5 to 17.7) more likely to have nipple involvement than patients with early-stage tumors. Patients with a tumor size of 4 cm or greater were nearly eight times (OR = 7.8, 95% CI = 4.2 to 14.5) more likely to have nipple involvement than patients with tumor size less than 4 cm. Patients with positive lymph nodes were five times (OR = 5.0, 95% CI = 2.7 to 9.1) more likely to have nipple involvement than patients with negative lymph nodes. Patients with tumors in a central location or that overlapped quadrants were nearly four times (OR = 3.8, 95% CI = 2.2 to 6.8) more likely to have nipple involvement than patients with tumors in other locations. Patients with grade 3 or undifferentiated tumors were three times (OR = 3.0, 95% CI = 1.4 to 6.4) more likely to have nipple involvement than patients with lower grade tumors. In multivariable analysis, stage > or = 3 (OR = 9.2, 95% CI = 4.2 to 20.3) central/ overlap location (OR = 4.1, 95% CI = 2.0 to 8.7) and grade 3 or undifferentiated (OR = 3.1, 95% CI = 1.3 to 7.5) were the only variables that remained significant predictors of nipple involvement.

The decision to perform breast conservation surgical procedures with nipple preservation can be difficult, particularly in patients with larger, more centrally located tumors. The multivariable model developed in this study may be useful in predicting the risk of cancerous nipple involvement and selecting appropriate breast conservation patients for nipple preservation.