A co-registration method to validate in vivo optical coherence tomography in the breast surgical cavity

Breast-conserving surgery accompanied by adjuvant radiotherapy is the standard of care for patients with early-stage breast cancer. However, re-excision is reported in 20–30 % of cases, largely because of close or involved tumor margins in the specimen. Several intraoperative tumor margin assessment techniques have been proposed to overcome this issue, however, none have been widely adopted. Furthermore, tumor margin assessment of the excised specimen provides only an indirect indication of residual cancer in the patient following excision of the primary tumor. Handheld optical coherence tomography (OCT) probes and their functional extensions have the potential to detect residual cancer in vivo in the surgical cavity. Until now, validation of in vivo OCT has been achieved through correlation with ex vivo histology performed on the specimen removed during surgery that is adjacent to the tissue scanned in vivo. However, this indirect approach cannot accurately validate in vivo imaging performance. To address this, we present a method for robust co-registration of in vivo OCT scans with histology performed, not on the main specimen, but on cavity shavings corresponding directly to the tissue scanned in vivo. In this approach, we use ex vivo OCT scans as an intermediary, surgical sutures as fiducial markers, and extend the in vivo field-of-view to 15 × 15 mm2 by acquiring partially overlapping scans. We achieved successful co-registration of 78 % of 139 in vivo OCT scans from 16 patients. We present a detailed analysis of three cases, including a case where a functional extension of OCT, quantitative micro-elastography, was performed. [Display omitted] •In vivo imaging of the breast surgical cavity may improve the assessment of residual cancer.•OCT scanning of breast tissue shavings provides a direct means of validating in vivo imaging.•Co-registration between in vivo OCT and histology is improved by intermediary ex vivo OCT imaging.•Tissue features and the proportion of adipose tissue impact co-registration accuracy.