Fluorescence guided surgery imaging systems for breast cancer identification: a systematic review

Abstract

Significance: Breast-conserving surgery (BCS) is limited by high rates of positive margins and re-operative interventions. Fluorescence-guided surgery seeks to detect the entire lesion in real time, thus guiding the surgeons to remove all the tumor at the index procedure.

Aim: Our aim was to identify the optimal combination of a camera system and fluorophore for fluorescence-guided BCS.

Approach: A systematic review of medical databases using the terms "fluorescence," "breast cancer," "surgery," and "fluorescence imaging" was performed. Cameras were compared using the ratio between the fluorescent signal from the tumor compared to background fluorescence, as well as diagnostic accuracy measures, such as sensitivity, specificity, and positive predictive value.

Results: Twenty-one studies identified 14 camera systems using nine different fluorophores. Twelve cameras worked in the infrared spectrum. Ten studies reported on the difference in strength of the fluorescence signal between cancer and normal tissue, with results ranging from 1.72 to 4.7. In addition, nine studies reported on whether any tumor remained in the resection cavity (5.4% to 32.5%). To date, only three studies used the fluorescent signal for guidance during real BCS. Diagnostic accuracy ranged from 63% to 98% sensitivity, 32% to 97% specificity, and 75% to 100% positive predictive value.

Conclusion: In this systematic review, all the studies reported a clinically significant difference in signal between the tumor and normal tissue using various camera/fluorophore combinations. However, given the heterogeneity in protocols, including camera setup, fluorophore studied, data acquisition, and reporting structure, it was impossible to determine the optimal camera and fluorophore combination for use in BCS. It would be beneficial to develop a standardized reporting structure using similar metrics to provide necessary data for a comparison between camera systems.

Keywords: breast conserving surgery; fluorescence guided surgery; fluorescent cameras; near-infrared; optical filters.

Fig. 1

FGS in breast cancer. (a) The patient is administered a fluorescent agent either via an oral solution or an injection (either into the tumor or into the systemic circulation). This fluorophore then targets the tumor actively (i.e., by targeting receptors or enzymes) or passively (i.e., by leaking into the tumor). (b) A light source emits a specific range of wavelengths of light to excite that agent. Images of the operative area are acquired using a camera sensitive to fluorescence. These images are taken of the tumor in situ, with the surgeon’s view of the operating field undisturbed. The image displayed on the top right screen is the fluorescence camera processed image wherein the likely site of the tumor (green) is superimposed onto the color image. A visual depiction of the areas of fluorescence is available to the operating surgeon for improved intraoperative decision making.

Fig. 2

Properties of light in tissue. Illustration of light–tissue interactions. Upon illumination of the tissue, part of the incident light (i) is reflected from the tissue surface without changing its initial properties (spectral shape or polarization state). This reflection is called “specular reflection” (ii), whereby both incident and reflected light are coplanar and at the same angle to the surface normal (perpendicular to the surface direction). Part of incident light can also be scattered (iii) in the tissue and re-emerge from the surface. This light is called “diffuse reflection” (iv) and its direction/spectral shape and polarization state are altered compared to the incident light. Finally, part of the incident light can be absorbed by a fluorophore (v), whereby part of the initial energy will be emitted as fluorescence (vi), or absorbed by a chromophore (viii), whereby no subsequent fluorescence emission occurs. Fluorescence light can be absorbed or scattered as well prior to its emergence from surface. Image reproduced with the permission of publisher.

Fig. 3

Mechanism of action for targeting tumors using fluorophores. (a) Passive targeting through the EPR effect whereby the fluorophore leaks into tissue due to the porous vasculature and has impaired lymphatic outflow. Panels (b) and (c) illustrate active targeting. (b) The targeting of specific receptors overexpressed in tumor. (c) The targeting of specific enzymes present in the tumor microenvironment by requiring the probe activation by that enzyme.