HISMAR: Intraoperative Margin Assessment Using the Histolog® Scanner to Reduce Reoperation in Breast-Conserving Surgery

BACKGROUND Breast cancer is the most common cancer in women in Switzerland, with approximately 6,500 new cases annually. Breast-conserving surgery (BCS, lumpectomy) is the preferred surgical approach for localised tumours and achieves equivalent survival outcomes to mastectomy when negative margins are obtained. However, re-excision rates of 20-30% have been reported in unselected BCS populations.

The risk of positive margins is particularly elevated in specific histological subtypes: published data indicate a relative risk of positive margins of 2.21 for ductal carcinoma in situ (DCIS) and 1.44 for invasive lobular carcinoma (ILC) compared to NST carcinoma without in situ component. A retrospective analysis of 266 consecutive eligible BCS cases at CHVR (2021-2024) identified a reoperation rate of 11.3% (30/266). A prospective study in the literature reported a reduction from 30% to 10% when intraoperative histological scanning was added to standard of care.

DEVICE The Histolog® Scanner (SamanTree Medical SA, Lausanne; CE IVD Class A, UDI 764017998003GV, software v3.4.0) enables real-time confocal fluorescence imaging of fresh lumpectomy specimen surfaces ex vivo, without tissue processing or staining, providing H\&E-quality images intraoperatively. The device is already in routine clinical use at CHVR.

STUDY OBJECTIVE To evaluate whether systematic use of the Histolog® Scanner in addition to standard of care reduces the reoperation rate by 50% or more - from 11.3% to ≤5.6% - in patients with ILC, pure DCIS, or NST+DCIS undergoing BCS at CHVR.

STUDY DESIGN Prospective, single-arm, single-centre post-market performance follow-up (PMPF) study (IVDR 2017/746). Comparator: validated historical institutional control (CHVR 2021-2024, same surgical team, n=266, reoperation rate 11.3%). A randomised design was not ethically justifiable as the device is already in routine use at CHVR.

PROCEDURE Following lumpectomy, the fresh specimen undergoes standard palpation then systematic Histolog® Scanner imaging (ex vivo, on the specimen surface). At the surgeon's discretion, standard radiological assessment is also performed. Intraoperative recuts are decided based on combined standard of care and Histolog® Scanner findings. The specimen is sent to the institutional pathology laboratory (ICH-CHVR) for definitive histopathology. The primary endpoint (reoperation recommendation) is determined by the multidisciplinary tumour board based on final histopathology, independently of intraoperative findings.

OUTCOMES Primary: reoperation rate (proportion of patients for whom the tumour board recommends re-excision), assessed within 4 weeks of BCS.

Secondary: (1) subgroup reoperation rates (ILC, pure DCIS, NST+DCIS); (2) intraoperative recut rate and localisation; (3) cost analysis; (4) exploratory concordance between intraoperative and final pathological margin status.

ELIGIBILITY Inclusion: women ≥18 years; CNB/VAB confirming ILC, pure DCIS, or NST+DCIS; planned BCS at CHVR; signed eIC.

Exclusion: pregnancy; prior neo-adjuvant treatment; mastectomy indication; acriflavine hypersensitivity; incapacity to consent.

SAMPLE SIZE n=228 patients (205 evaluable minimum; 10% dropout). Calculated by one-sample z-test (Fleiss 1981): α=0.05 two-sided, power=80%, p₀=11.3%, p₁=5.6%. Estimated recruitment: 3-3.5 years.

STATISTICAL ANALYSIS Primary: one-sample z-test for a proportion (two-sided, α=0.05), per-protocol population, with exact 95% Clopper-Pearson CI. Pass criterion: p<0.05 and observed rate ≤5.6%. Sensitivity analysis on ITT population. No interim analysis planned.

ETHICAL / REGULATORY Category A1 (ClinO-MD Art. 6a). CEC: Commission cantonale d'éthique du Valais (CCVS). No Swissmedic approval required. Electronic informed consent (eIC). Declared conflict of interest: PI holds a consultancy arrangement with SamanTree Medical SA (institutional fee); mitigation: independent tumour board endpoint, independent biostatistician.

FUNDING Institutional budget of CHVR. No industry funding.