Abstract PS4-08-08: Comparison of localization techniques for non-palpable breast cancer: real world data

Abstract Background: Accurate tumor localization is critical in performing successful breast-conserving surgery (BCS). Multiple localization methods are available to identify and target non-palpable lesions. This multi-national study aims to compare various localization techniques, by evaluating the impact on margin status, re-operation and recurrence. Methods: We conducted a multi-center retrospective study of 1481 patients with Tis-1N0 breast cancer undergoing BCS from 2015-2024 across 28 international centers. Patients were grouped by localization technique: radioactive seed, RSL (n=711, 48%), wire-guided, WGL (n=561, 37.9%), ultrasound-guided, UGL (n=104, 7%), radio-guided occult lesion, ROLL (n=68, 4.6%), radar reflectors (n=22, 1.5%), magnetic marker (n=7, 0.5%), and radiofrequency identification RFID (n=8, 0.5%). Patients were then categorized by margin status as positive (n=54, 3.6%), <2 mm (n=102, 6.9%), or ≥2 mm (n=1325, 89.5%). Cross-tabulation and Pearson Chi-Square test assessed the association between techniques and margin status. Multivariate logistic regression analyzed re-operation and recurrence. Results: In 1266 cases (85.5%), surgeon preference was the primary factor guiding choice of localization technique, followed by availability of localization methods in 173 cases (11.7%). Tumor size was classified according to standard staging, pTis, pT1a, pT1b, and pT1c, with distributions of 305 (20.6%), 154 (10.4%), 383 (25.9%), and 639 (43.1%) cases respectively. The dominant histologic subtype was invasive ductal carcinoma (n=850, 57.4%). Molecular subtypes were predominantly luminal A (n=705, 47.6%) and luminal B (n=492, 33.2%). Re-operation was required in 100 (6.8%) patients. Local, regional, and systemic recurrence occurred in 13 (0.9%), 2 (0.1%) and 6 (0.4%) patients respectively. There was a significant association between localization technique and margin status (p < 0.001). WGL yielded the highest rate of positive margins (n=48, 8.6%), leading to additional surgeries. In contrast RSL had the lowest positive margin rate (n=2, 0.3%) and the highest proportion of margins ≥2 mm (n=705, 99.2%). Radar reflectors, magnetic markers, RFID, and ROLL had no positive margins, although ROLL showed a relatively high proportion of margins <2 mm (n=14, 20.6%), suggesting limitations in precision. The observation of a lower positive margin rate with UGL (n = 4, 3.8%) compared to WGL, with a substantially elevated re-operation risk (OR = 4.326, p < 0.001) indicates the impact of a low overall event rate and confounding variables. Among patients undergoing WGL, 100 (17.8%) had in situ disease, with 21 requiring re-operation. In the UGL group, only 4 (3.8%) patients had in situ disease and none underwent re-operation. This proportion of in situ disease may explain the higher re-operation rate observed with WGL, despite lower rate of positive margins, given current margin recommendation for in situ disease of ≥ 2mm. With median follow-up of 28.2 months (range 12.1-108.2 months) local recurrence was observed in 4 patients (5.9%) with ROLL, 6 patients (0.8%) with RSL and in 3 patients (0.5%) with WGL, while no local recurrence was reported with the other localization techniques, although logistic regression models were limited by low event rates and perfect separation. Conclusion: Our findings highlight significant differences in margin status and re-operation rates among localization techniques utilized in BCS for non-palpable breast cancer across diverse clinical practices. RSL demonstrated superior accuracy with the lowest positive margin and re-operation rates. Wide variability among localization techniques underscores the need to consider institutional factors including availability, cost and geographic standards in the selection of localization method to optimize surgical and oncologic outcomes. Citation Format: B. Yigit, M. Tokocin, K. Lupinacci, B. Demirors, N. Zafer Utkan, M. Masta, E. Ozkurt, U. Kesici, A. Akan, E. Sen, A. Kamali Polat, H. Karanlik, B. Comcali, A. Soyder, L. Dogan, A. Salamat, M. Onur Kulturoglu, K. Senol, J. Alazhri, A. Dag, N. Cabioglu, G. Maralcan, H. Valiyeva, M. Levhi Akin, B. Balci Topuz, B. Citgez, D. Can Trabulus, G. Ozan Kucuk, S. Salimoglu, B. Ozcinar, O. Aytac, B. Goktepe, C. Mathelin, V. Ozmen, A. Soran, U. Evrensel. Comparison of localization techniques for non-palpable breast cancer: real world data [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS4-08-08.