Real-time sentinel lymph node biopsy guidance using combined ultrasound, photoacoustic, fluorescence imaging: in vivo proof-of-principle and validation with nodal obstruction

Abstract

Precise sentinel lymph node (SLN) identification is crucial not only for accurate diagnosis of micro-metastases at an early stage of cancer progression but also for reducing the number of SLN biopsies (SLNB) to minimize their severe side effects. Furthermore, it is desirable that an SLNB guidance should be as safe as possible in routine clinical use. Although there are currently various SLNB guidance methods for pre-operative or intra-operative assessment, none are ideal. We propose a real-time SLNB guidance method using contrast-enhanced tri-modal images (i.e., ultrasound, photoacoustic, and fluorescence) acquired by a recently developed hand-held tri-modal probe. The major advantage of tri-modal imaging is demonstrated here through an in vivo study of the technically-difficult case of nodal obstruction that frequently leads to false-negative results in patients. The results in a tumor model in rabbits and normal controls showed that tri-modal imaging is capable of clearly identifying obstructed SLNs and of indicating their metastatic involvement. Based on these findings, we propose an SLNB protocol to help surgeons take full advantage of the complementary information obtained from tri-modal imaging, including for pre-operative localization, intra-operative biopsy guidance and post-operative analysis.

Figure 1. Schematic diagram of the tri-modal imaging system and probe.

WL: white light, SHG: second-harmonic generator, OPO: optical parametric oscillator, DAQ: data acquisition board.

Figure 2

(a) US and (b) PA cross-sectional in vivo images of control (left) and tumor-bearing (right) rabbits acquired in the pre-operative localization session. SLNs are indicated by the white arrows in the US images. The PA images, taken at 90 s after dye injection, covered the region of interest indicated by the dotted rectangle in the US images (see Fig. S1 for the PA images acquired at 0, 30, 60 and 90 s.). The white scale bar in the US image indicates 1 cm. (c) PA intensity as a function of time following contrast injection, averaged over the region of interest indicated by white dotted circles in the PA images. BG denotes resting-state background.

Figure 3

FL images obtained during the intra-operative guidance session: (a) control and (b) tumor-bearing; PT-primary tumor, LC-lymphatic channel.

Figure 4

Post-operative analysis: ex vivo FL validation of the resected SLNs from (a) control and (b) tumor-bearing rabbits. The blue arrow in (b) indicates the direction of FL imaging available during in vivo surgical guidance. The black rectangles indicate the regions-of-interest for FL intensity measurements. H&E stained sections are shown for the corresponding (c) control and (d) tumor-bearing SLN: MT-metastatic tumor, EL-efferent lymph vessel, ALV-afferent lymph vessel, NLN-normal lymph node tissue.

Figure 5. Flowchart of proposed SLNB guidance using contrast-enhanced tri-modal imaging.