Predictors of reexcision findings and recurrence after breast conservation.

PURPOSE

To identify predictors of reexcision findings and local recurrence in the setting of breast-conserving therapy with radiation.

METHODS

The records of 535 patients who underwent breast-conserving surgery followed by radiation for Stage I or II cancer between 1972 and 1996 were reviewed. The mean follow-up period for surviving patients without evidence of recurrence is 6 years. Various clinical and pathologic prognostic factors were examined for significance with regard to reexcision findings and recurrence rates. Pathologic margin status was classified as negative, close (<or=2 mm), positive, or indeterminate.

RESULTS

The pathologic margin status was the most important predictor of local recurrence. The freedom from local relapse (FFLR) at 6 years was 97% for patients with negative pathologic margins and 86% for all others (p < 0.0001). There was no significant difference in recurrence rates among patients with close, positive, or indeterminate margins. However, the use and sequencing of systemic therapy affected recurrence rates among these patients. For patients with close, positive, or indeterminate margins, the crude risk of local recurrence was 4% among patients who received tamoxifen or received chemotherapy integrated with or after radiation. The risk of local recurrence was 16-29% among the patients with close, positive, or indeterminate margins who did not receive systemic therapy or who received radiation after completion of chemotherapy. Local recurrence rates were low in patients with negative margins (2-8%) regardless of the use of systemic therapy or its timing. The presence or absence of residual disease at reexcision did not predict recurrence as long as the final margins were negative. Among patients who underwent reexcision before radiation, extensive intraductal component (EIC) (p = 0.0001) and young patient age (p = 0.03) were predictive of residual disease in the specimen. Patients with initially close margins and no EIC had a low risk of residual disease at the time of reexcision, as did patients older than age 65 without EIC.

CONCLUSION

Pathologic margin status is the most important predictor of local recurrence after breast conservation with radiation. Patient age and EIC were significant predictors of residual disease at reexcision. The use and timing of systemic therapy appear to influence the risk of local recurrence in patients who do not have negative lumpectomy margins.