P3-12-01: ROLL Trial – Radio-Guided Occult Lesion Localization Versus Wire Guided Localization in Breast Conserving Surgery for Non-Palpable Breast Cancer: A Multicenter Randomized Controlled Trial.

Abstract

Abstract Background: For the management of non-palpable breast cancer, accurate preoperative localization is essential to achieve complete resection with acceptable cosmetic results. ROLL takes advantage of the intratumorally injected radiotracer, that is already used for the sentinel node procedure, to localize the primary tumor during surgery. In a multicenter randomized controlled trial, we determined if ROLL is superior to the standard of care (i.e. wire guided localization, WGL) for preoperative tumor localization. Methods: Women (>18 yrs.) with histologically proven non-palpable breast cancer and eligible for breast conserving treatment (BCT) with sentinel node procedure were randomized to ROLL or WGL. Patients allocated to ROLL received an intra-tumoral dose of 120 Mbq Technetium99 nanocolloid. Guided by a gamma detection probe, the surgeon excised the primary tumor and the sentinel node(s). In the WGL group, patients received a similar intra-tumoral or periaureolair dose of technetium in order to allow sentinel node biopsy. Ultrasound or mammography guided insertion of a hooked wire provided surgical guidance for excision of the primary tumor. Primary outcome measure was the proportion of complete tumor excisions (i.e. with negative margins). Furthermore, the proportion of patients requiring re-excision was assessed. Data were analyzed according to intention to treat analysis. Results: Three hundred and fourteen patients with 316 invasive breast cancers were enrolled. There were no significant differences in proportion of complete tumor removal with free margins and re-excision rate; complete tumor removal with negative margins was seen in 87.7% of patients in the ROLL group versus 87.5% (p=0.97) of patients in the WGL group Re-excision was required in 11.7% of patients in the ROLL group versus 9.2% (p=0.47) in the WGL group. Incorrect preoperative localization (i.e. failure to identify the correct localization of the lesion) occurred in 5 (3%) patients in the ROLL group versus 0 patients in the WGL group (p=0.029). Differences in the volume of the excised specimen, duration of the procedure, success rate of the sentinel node procedure, surgeons’ preferences and patients’ pain perception are currently being analyzed. Conclusion: With this multicenter randomized controlled comparison, the first of its kind in patients with histologically proven breast cancer, we show that ROLL is not superior to WGL in terms of complete tumor excision and re-excision rates. Furthermore incorrect lesion localization occurred more frequently when applying ROLL. Data on long term (6 months) cosmetic results and quality of life (6 weeks, 3 months, 6 months) are awaited. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P3-12-01.