Does the placement of surgical clips within the excision cavity influence local control for patients treated with breast-conserving surgery and irradiation?

Purpose: A number of authors have demonstrated the importance of using surgical clips to define the tumor bed in the treatment planning of early-stage breast cancer. The clips have been useful in delineating the borders of the tangential fields, especially for very medial and very lateral lesions as well as the boost volume. If surgical clips better define the tumor bed, then a reduction in true or marginal recurrences should be appreciated. We sought to compare the incidence of breast recurrence in women with and without surgical clips, controlling for other recognized prognostic factors. Methods and Materials: Between 1980 and 1992, 1364 women with clinical Stage I or II invasive breast cancer underwent excisional biopsy, axillary dissection, and definitive irradiation. Median follow-up was 60 months. Median age was 55 years. Seventy-one percent of patients were path N0, 22% had one to three nodes, and 7% had > four nodes. Sixty-one percent were ER positive and 49% PR positive. Margin status was negative in 62%, positive in 10%, close in 9%, and unknown in 19%. Fifty-seven percent of women underwent a reexcision. Adjuvant chemotherapy + tamoxifen was administered in 29%, and tamoxifen alone in 17%. Surgical clips were placed in the excision cavity in 556 patients, while the other 808 did not have clips placed. All patients had a boost to the tumor bed. Patients had their boost planned with CT scanning or stereo shift radiographs. No significant differences between the two groups were noted for median age, T stage, nodal status, race, ER/ PR receptor status, region irradiated, or tumor location. Patients without clips had negative margins less often, a higher rate of unknown or positive margins and more often received no adjuvant therapy compared to patients with surgical clips. Results: Twenty-five and 27 patients with and without surgical clips, respectively, developed a true or marginal recurrence in the treated breast. The actuarial probability of a breast recurrence was 2% at 5 years and 5% at 10 years for patients without clips compared to 5 and 11%, respectively, for patients with clips (p = 0.01). Comparing the breast recurrence rates for patients with and without clips there was no significant difference for the following factors: chemotherapy, tamoxifen, negative, positive or close margins, reexcision, N1, and central or inner primary. Increased rates of breast recurrence were noted for patients with clips for the following variables: no adjuvant treatment (p < 0.001), unknown margins (p < 0.001), a single excision (p = 0.003), path N0 (p = 0.001), and outer location (p = 0.02). A forward stepwise multivariate analysis for all 1364 patients was performed using the aforementioned variables as well as the presence or absence of surgical clips and the primary surgeon. The surgeon (p = 0.03) and no adjuvant treatment (p = 0.01) significantly influenced breast recurrence. For patients with surgical clips the 10 year isolated breast recurrence rate was 21% for a single surgeon vs. 6% in the remainder of the group (p = 0.01). For patients with clips, this surgeon had unknown margins in 48% of cases compared to 10% overall (p = 0.001). Excluding this surgeon from analysis the isolated breast recurrence for patients with clips was 6 vs. 5% for patients without clips (p = 0.18). Conclusions: Overall, there was a significant difference in the 10-year breast recurrence rate favoring women without clips despite more adverse prognostic factors. There was no difference in the breast recurrence rate for patients with or without surgical clips if careful attention to margin status was addressed. Failure to ink the surgical specimen resulting in unknown margins cannot be compensated for with the placement of surgical clips or treatment planning using CT to delineate the surgical bed.