Abstract PO1-23-04: The IBRAnet localisation study; a UK National multi-centre cohort study comparing the safety and effectiveness of guidewire, RFID, and magnetic seed- guided localisation for impalpable breast lesions

Abstract

Abstract Background Wire-guided localisations (WGL) of breast cancers have been the gold-standard for pre-operative localisation of non-palpable breast cancers for many years but have significant limitations. Newer localisation techniques have been introduced including Radio Frequency ID tags (RFID) and magnetic seeds which aim to overcome these limitations. The UK iBRA-net group have designed a multi-centre platform study aiming to compare the safety and effectiveness of wire, RFID, magnetic seeds, and radar-based localisation. A second aim was the qualitative assessment of shared learning as users gained experience. Methods A UK national multi-centre prospective observational platform study was designed to compare a control arm of WGL for women undergoing breast conserving surgery with two interventional arms of RFID tags and magnetic seeds. Patient data were collected between August 2018 and July 2022 in UK centres. The primary outcome was accurate excision of the index lesion at the time of surgery. Secondary outcomes were defined as; resection margins, breast reoperation rate (planned and unplanned), complications, cancellation rate on day of surgery, and duration of the surgery. To account for lesion size in the assessment of excision weight, we used size as a denominator over two dimensions, reporting this as weight/size2, in g/mm2. Data was collected prospectively on a secure REDcap database following approval from local audit departments. To calculate the sample size a power calculation was employed, with n=950 patients per group the upper limit of the observed one-sided 95% confidence interval for the difference between identification rates (intervention vs WGL) is expected to be less than 0.9% with 80% power, assuming the two methods both have an expected identification rate of 99.4%. Simple summary statistics were calculated for each outcome and data were tested for distribution and differences between groups using unpaired t-tests, Mann-Whitney U tests, and Chi squared tests as appropriate. Analyses were conducted using Stata® IC version 14 (StataCorp, College Station, Texas, USA). Results Data were accrued from 3484 patients in 55 units. This included 1293 patients having WGL, 1188 RFID tags, and 1003 magnetic seeds. RFID had a lower rate of identification of the index lesion compared to magnetic seed and wire (97.9% vs. 99.8% vs. 99.1% respectively). When comparing the reasons behind failure of the primary outcome, the majority were not thought to be related to the modality of localisation. Data from 1560 patients with unifocal, unilateral breast lesions were included in subgroup analysis, excluding those having neo-adjuvant chemotherapy and those having therapeutic mammoplasty. Positive margins were lower in RFID excisions compared to wire (11.2% vs. 15% respectively, p< 0.05), whereas magnetic seed were equivalent (11.2 % vs 13.3%, P=0.22). Routine cavity shaves were performed in 64.5% of WGL vs 48.3% of RFID vs 64.3% of magnetic guided excisions. There was no difference in specimen weight/size2 (0.138 WGL vs. 0.131 RFID vs. 0.15 magnetic seed). Re-excision (for positive or close margins) was equivalent across the three groups (13.2% of wire vs 15% of RFID vs 12.3% of magnetic seed guided excisions, p=0.3). There have been four shared learning educational events focussing on three key themes relating to preoperative, intraoperative, and postoperative learning outcomes in addition to 130 database shared learning entries. This may aid other surgeons as they adopt these techniques. Conclusions All three localisation devices demonstrate high levels of localisation accuracy. There was a higher rate of device dislodgement in the RFID group, which was also identified in the shared learning events. However, re-excision rates were equivalent across the three groups. This study has demonstrated a robust platform for the comparative evaluation of new localisation technologies and of sharing learning and experience. Citation Format: Rachel Foster, James Harvey, Rajiv Dave, Anthony Maxwell, Senthurun Mylvaganam, Matthew Gardiner, Sue Down, Nicola Barnes, Mihir Chandarana, Shelley Potter, Yazan Masannat, Chris Holcombe, Tahir Masudi, Amtul Carmichael, Robert Milligan, Jenna Morgan. The IBRAnet localisation study; a UK National multi-centre cohort study comparing the safety and effectiveness of guidewire, RFID, and magnetic seed- guided localisation for impalpable breast lesions [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-23-04.