Guidance in breast-conserving surgery

Tumour localization

Localization techniques involve radiological insertion of a tumour guidance system, followed by surgical resection. These identify/mark the tumour core, which offers surgeons an approximate geolocation, yet they are unable to provide information about the tumour extent or tumour–normal tissue interface, and are therefore associated with a high average reoperation rate1. By definition, all localization techniques depend on accurate identification of the target lesion by the radiologist. Challenges include localization and mapping of ductal carcinoma in situ that is not detected by ultrasound imaging6. Similarly, lobular carcinoma is difficult to detect on mammography and ultrasound examination, and thus MRI guidance is often required6.

In the standard wire-guided localization (WGL) technique, a wire is placed in the core of the tumour. Bracketing wires can be used to describe the approximate size of the lesion or multiple lesions within proximity2^,7^,8. Patients, however, find wires uncomfortable, they are prone to displacement and need to be inserted immediately before surgery. This strain on radiology services disrupts patient flow between imaging departments and theatres, delaying operations2^,7^,8.

A variety of alternative localization techniques have been developed to address these flaws. Radio-occult lesion localization (ROLL) comprises injection of a radioactive colloid into the tumour, whereas a radioactive seed is implanted in radioactive seed localization (RSL)2^,3^,7^,8. A Cochrane Review7 confirmed that WGL is comparable to RSL and ROLL in terms of positive margins and reoperation rates, but the latter are associated with improved patient comfort, hospital flow, or surgical precision. More recently, a ferromagnetic seed (Magseed^®; Endomag, Cambridge, UK), a reflector to electromagnetic radar signals (SAVI SCOUT^®; Meritmedical, South Jordan, UT, USA), a radiofrequency tag (Localizer™; Hologic, Marlborough, MA, USA), and an electromagnetic signature for triangulation (Elucent Smart Clip™; Elucent Medical, Eden Prairie, MN, USA; NCT04604561) have entered the market2^,3^,8–10. Seldom used techniques include cryo-assisted localization (freezing) and carbon track localization (intralesional carbon injection), which enable tactile and visual feedback, respectively8. Seeds are somewhat difficult for localization of multifocal lesions in close proximity (less than 2 cm), as independent signals would be indistinguishable2. Thus, localization techniques help pinpoint the tumour core; but they fail to provide a visual representation of the healthy cancer tissue boundary (Table S1).

Tumour identification

There is an urgent need for techniques that can identify, on a macroscopic scale, tumour location, size, and invasiveness, thus demarcating where disease stops and healthy breast starts. Potential imaging solutions include IOUS, iMRI or FGS3–5. IOUS is highly operator-dependent and iMRI (preoperative MRI and intraoperative optical imaging) requires the patient to be in the same position as during the preoperative scan. Both delay the operation, while scanning takes place, so FGS might be a more suitable option2 (Table S1).

During FGS, a near-infrared dye is administered that targets the tumour, and a light source and dual colour and infrared camera system are used to detect the signal, with the images being displayed on a screen. The images are provided in real time and are not dependent on the skill of the operator nor patient position. As infrared light is used, the healthy tissue–tumour border can be visualized without affecting the surgeon’s view of the operative site. Anecdotal evidence suggests a penetration depth of 4 mm in FGS, which enables both intraoperative guidance and provision of an adequate margin11^,12. Furthermore, a lack of fluorescence in the resection cavity could be used to verify the adequacy of resection.

There have been multiple studies using US Food and Drug Administration (FDA)-approved non-selective fluorophores, such as indocyanine green (ICG). After injection, ICG leaks into tumours as a result of the enhanced permeability and retention (EPR) effect that capitalizes on the tumour’s porous vasculature and compromised lymphatic outflow. This passive mechanism of action, however, lacks accuracy. The most recent study11 combining image analysis with texture metrics surpassed results of previous studies, achieving a sensitivity of 66–82% and specificity of 90–93%.

The advent of targeting fluorophores has made the delineation between tumour and normal tissue clearer (Fig. 1). Eleven targeting fluorophores (8 targeting receptors and 3 targeting enzymes) are currently undergoing clinical trials, and more are being developed13. Of these, eight are currently undergoing trials in BCS, with favourable results. For example, bevacuzimab-800 (vascular endothelial growth factor), LUM015 (cathepsin), and EC17 (folate) were able to identify 100% of the lesions to be resected, with sensitivities of 88–98, 100, and 100%, respectively14–17. These targeting fluorophores need to gain FDA approval; however, many are already in the latter stages of clinical trials18^,19.

Open multimedia modal