Enhancing precision in breast conserving surgery: introduction to emerging imaging technologies

Introduction

Breast cancer is the most common cancer affecting women worldwide with over 2.3 million women diagnosed annually, and its incidence continues to rise year after year (Giaquinto et al., 2024; Arnold et al., 2022). Breast Conserving Surgery, alongside radiotherapy is the treatment of choice for treating early breast cancers. However, imprecision remains a challenge faced by surgeons when deciding on how much tissue to remove during Breast Conserving Surgery. In response, new and improved imaging technologies are being developed to help guide surgeons and reduce rates of secondary re-excision procedures. Our aim by publishing this special edition of articles is to motivate research and development in this important area and highlight the vast amount of work already being carried by clinical colleagues, engineers and industrial companies across the world. The focus of this collection is exclusively on emerging imaging technologies that can be used by clinicians for intraoperative assessment of oncology margin status in Breast Conserving Surgery.

Positive margins in breast conserving surgery

For Breast Conserving Surgery to be an effective treatment, the margins of the specimen of breast tissue that is removed need to be deemed cancer free, also known as a “clear margin”. Definitions of clear margins can vary slightly depending on organisation, as highlighted in Table 1. In clinical practice, margin status is confirmed post-operatively during histopathological assessment of the specimen. In a recent large systematic review and meta-analysis comprising of over one hundred thousand patients, close margins of <1 mm were shown to be associated with increased local and distal recurrence (Bundred et al., 2022).

TABLE 1. Recommendations on surgical margins for Breast Conserving Surgery from major guidelines.

During Breast Conserving Surgery, it is challenging for surgeons to determine the adequacy of resection of the tumour, due to the imprecision of estimates based on visual and tactile cues. The imprecision of current Breast Conserving Surgery operative platforms manifests in high rates of incomplete excision (known as “positive margins”), whereby the disease comes too close to the resection edge. Approximately 14%–40% of women undergoing Breast Conserving Surgery are reported to have positive margins following their index procedure, requiring further re-excision surgery to ensure clear margins (Kim et al., 2024; Chagpar et al., 2015; National Audit of Primary Breast, 2025). The wide range in patients reported as having margin positivity may be explained by confounding factors including definition of positive margins (see Table 1), operative technique including the use of routine excision of additional tissue (known as cavity shave) and tumour subtype (Kim et al., 2024). Re-operative intervention has been shown to negatively affect patient reported outcomes such as psychological wellbeing and reported appearance of the breast (Baliski and Bakos, 2022). Reoperation is cost inefficient for providers and has financial toxicity, with total treatment costs increasing by up to 24% per patient (Kim et al., 2024; Grant et al., 2019).

Future of intra-operative decision making

In clinical practice, optimization of intra-operative margin assessment should be viewed as a continuous, multidisciplinary, and dynamic decision-making process. Its core concept is to integrate individualized preoperative planning with real-time intraoperative imaging feedback, thereby forming a closed loop among imaging, pathology, and surgical decisions. Surgeons should first formulate a margin management strategy preoperatively, considering tumour biology, localization technique, and planned radiotherapy. During surgery, two-dimensional or three-dimensional specimen radiography should serve as an initial screening tool for rapid margin assessment. At this point, and depending on available resources, advanced high-resolution imaging modalities could be selectively applied to verify margin integrity at the microscopic level. Based on predefined threshold criteria and inter-observer agreement rules, the surgical team can promptly decide whether to perform additional excision or cavity sampling, minimizing positive margins and re-operation rates. Postoperatively, imaging findings should be correlated with final histopathology during imaging-pathology review meetings to accumulate experience, refine interpretation algorithms, and promote standardization of intraoperative evaluation protocols. Through this systematic closed-loop model, intraoperative imaging techniques would be transformed from isolated tools into integral components for enhancing the precision and individualization of breast-conserving surgery.

Current practice

Widely used techniques for intraoperative margin assessment include two-dimensional specimen mammography and immediate histological assessment using frozen section. Frozen section has been shown to have a high sensitivity for assessing breast cancer margins intraoperatively, but main barriers to implementation include lack of pathology availability and increased operating time which makes the technology logistically prohibitive for many institutions (Garcia et al., 2021; St John et al., 2017). Specimen mammography is commonly used intraoperatively and relies on two-dimensional interpretation by the surgeon. It is user-friendly and easily integrates within surgical workflows, however the produced images are of lower resolution (Dowling et al., 2024). A recent meta-analysis suggests this technique may result in lower sensitivity for margins assessment compared to other techniques (pooled sensitivity 0.39) (Dowling et al., 2024). Additionally, specimen mammography is unable to identify non calcified lesions and at times, calcifications that are benign may be overinterpreted as malignant (St John et al., 2017).

Emerging technologies

3D Tomosynthesis Specimen radiography is an imaging technique whereby multiple radiographs are taken of the excised specimen in 1 mm digital slices (Partain et al., 2020). This allows for three-dimensional reconstitution of the specimen and allows for better visuospatial assessment of the specimen and its contents. In a single centre retrospective analysis, it was observed to be superior to two-dimensional radiography given it enabled the surgeons to visualise the orientation of margins map areas that needed additional excision at the index procedure, thus potentially lowering the re-excision rate (Partain et al., 2020).

Micro Computed Tomography (Micro-CT) provides real time ex-vivo three-dimensional volumetric images of intraoperative specimens. Reliable radiographic images are produced, capturing the whole specimen in detail (DiCorpo et al., 2020). Images can then be manipulated and ‘sliced’ into any configuration, which could help guide surgeons on where to excise additional tissue. Preliminary tissue studies have demonstrated that Micro-CT images were able to detect margin positivity in up to 93% of cases deemed positive by histopathologists, in addition to detecting cases where the pathologist was not provided with sections where the margin positivity was present (DiCorpo et al., 2020). However pooled sensitivity and specificity in a systematic review of Micro-CT for margins assessment grouping eight studies were lower - at 0.63 (95% CI: 0.45–0.79) sensitivity and 0.77 (95% CI: 0.68–0.85) specificity (Meshkati Yazd et al., 2025). The main barriers to implementation remain related to lacking evidence from clinical trials and ongoing need for development of image management systems to interpret images (DiCorpo et al., 2020).

Phase Contrast Micro-CT uses imaging technology akin to Micro-CT to produce rapid three-dimensional images. The addition of a ‘mask’ that selects refracted rays, makes small lesions within tissue stand out in a way not possible with conventional radiology (Hagen et al., 2020). In preliminary studies, the size, shape and detail of breast cancers captured on the Phase Contrast Micro-CT images are comparable the gold standard histological examination findings, allowing for precise and rapid intraoperative margins assessment, estimated to be 1.6 times the sensitivity of conventional 2D imaging (Hagen et al., 2020). Further research evaluating sensitivity and specificity within different breast cancer subtypes as well as clinical trial evidence and further prototyping is required to develop this technology for clinical integration.

Optical Coherence Tomography (OCT) uses low coherence interferometry to provide high resolution (typically 1–15 µm axially and 10–20 µm laterally) visualisation of either ex-vivo or in vivo tissue, which can be reconstituted into three-dimensional images (Duan et al., 2023). It can also be used in real time to deliver microscopic images reaching 2–3 mm in depth, making it a suitable technology to assess breast cancer margins intraoperatively, with ongoing studies demonstrating its value in Breast Conserving Surgery (Duan et al., 2023; Nguyen et al., 2009). However, thicker tissue samples that require deeper tissue penetration may not be fully assessed by OCT and furthermore additional training for image interpretation by pathologists remains a clinical barrier (Fan et al., 2024).

Cerenkov Luminescence Imaging (CLI) is an optical imaging modality that detects Cerenkov radiation from decaying radiotracers (Grootendorst et al., 2017). In Breast Conserving Surgery, tracers such as 18- fluorodeoxyglucose (FDG) are injected intravenously pre-operatively. CLI has been shown to have a high sensitivity for detecting areas of high metabolic activity such as cancerous cells, allowing for precise tumour localisation (Grootendorst et al., 2017). This is particularly useful for smaller or impalpable tumours, as well as for residual disease. Intraoperatively, CLI provides real-time imaging of the excised tissue specimen, and this technology has been applied to clinical settings safely without exposing theatre staff to high radiation levels (Grootendorst et al., 2017). Positive margins (indicated by residual radiotracer signal) suggest incomplete resection, prompting further tissue removal. Certain tumours that have low glucose metabolism may however result in false negative findings and preliminary feasibility studies for this technology have also commented on the challenge in timing the 18-FDG injection to provide the optimal operating window (Lambert et al., 2025).

Confocal Fluorescence Microscopy is an advanced imaging technique which combines laser excitation, spatial pinhole filtering and fluorescence to achieve optical sectioning of operative specimens. The generated products are high-resolution cellular images of the surface of the specimen, likened to frozen section or histopathology, without having to physically fix or slice the tissues (Sandor et al., 2022). Normal and malignant breast tissue can be differentiated intraoperatively by surgeons using this technique with high accuracy allowing for precise mapping (Conversano et al., 2023). Current studies are now investigating whether or not commercial confocal imaging (e.g., Histolog™, Samantree, Lausanne, Switzerland) reduces the risk of positive margins (Lux et al., 2025). However, clinical implementation remains far from practice given high costs, ongoing need for regulatory approval and need for training requirement for clinical staff.

Conclusion

With the growing number of women undergoing Breast Conserving Surgery worldwide, it is imperative that more accurate intraoperative imaging technologies are comprehensively evaluated, then systematically integrated into surgical practice. Many barriers for implementation remain, highlighting the need for further research and development of these technologies. In the following collection of articles, we will delve into the development and translational progress of each of these novel imaging modalities in turn to determine the diagnostic accuracy and implementation plan for improving precision in Breast Conserving Surgery.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Andrea Cafarelli, Sant'Anna School of Advanced Studies, Italy

Reviewed by: Hang Su, Université Paris-Saclay, France

Lorenzo Vannozzi, Sant'Anna School of Advanced Studies, Italy