Photoacoustic and fluorescence image-guided surgery using a multifunctional targeted nanoprobe

Abstract

Purpose: A complete surgical excision with negative tumor margins is the single most important factor in the prediction of long-term survival for most cancer patients with solid tumors. We hypothesized that image-guided surgery using nanoparticle-enhanced photoacoustic and fluorescence imaging could significantly reduce the rate of local recurrence.

Methods: A murine model of invasive mammary carcinoma was utilized. Three experimental groups were included: (1) control; (2) tumor-bearing mice injected with non-targeted nanoprobe; and (3) tumor-bearing mice injected with targeted nanoprobe. The surgeon removed the primary tumor following the guidance of photoacoustic imaging (PAI), then inspected the surgical wound and removed the suspicious tissue using intraoperative near-infrared (NIR) fluorescence imaging. The mice were followed with bioluminescence imaging weekly to quantify local recurrence.

Results: Nanoprobe-enhanced photoacoustic contrast enabled PAI to map the volumetric tumor margins up to a depth of 31 mm. The targeted nanoparticles provided significantly greater enhancement than non-targeted nanoparticles. Seven mice in the group injected with the targeted nanoprobes underwent additional resections based upon NIR fluorescence imaging. Pathological analysis confirmed residual cancer cells in the re-resected specimens in 5/7 mice. Image-guided resection resulted in a significant reduction in local recurrence; 8.7 and 33.3 % of the mice in the targeted and control groups suffered recurrence, respectively.

Conclusions: These results suggest that photoacoustic and NIR intraoperative imaging can effectively assist a surgeon to locate primary tumors and to identify residual disease in real-time. This technology has promise to overcome current clinical challenges that result in the need for second surgical procedures.

FIG. 1

Schematic of the a PAI system and b NIR fluorescence imaging system. PAI photoacoustic imaging, NIR near-infrared

FIG. 2

Validation of targeting ability and contrast enhancement for PAI of NIR-830–ATF–IONPs. a NIR fluorescence imaging of mice that received an injection of NIR-830–ATF–IONPs or NIR-830–BSA–IONPs. All tumors from the targeted animal group had strong optical signals and some showed strong signals from the internal organs (kidney and liver), possibly due to the non-specific uptake of the IONP by the RES in the liver and elimination of degraded NIR-830-dye-conjugates in the kidney. PB-stained tumor tissue sections from the mice injected with NIR-830–ATF–IONPs showed a cluster of blue-stained IONPs, while no blue-stained IONPs were found in the mice injected with NIR-830–BSA–IONPs. b A typical three-dimensional photoacoustic image with two tiny markers in the corners. Markers indicated by the red arrows were used to assistant surgeons in locating the tumors in the imaging area. c MAP and cross-sections of photoacoustic image shown in b. The MAP images were used to depict the surgical profile in the imaging area, and cross-sections provided depth information and axial dimension of tumors. NIR near-infrared, IONPs iron oxide nanoparticles, RES reticuloendothelial system, MAP maximum amplitude projection

FIG. 3

Establishment of a protocol for identifying suspicious tissues with residual diseases. a Following the PAI-guided surgery, two pieces of tissues from the area in (white dashed circles) and away from (yellow dashed circle) the surgical bed were resected and imaged with the primary tumor. b Both the primary tumor and the tumor-free specimen resected from the wound (white circle) emitted stronger optical signals compared with the specimen resected from the area away from the surgical wound (yellow circle). The optical intensity difference between two re-resected specimens was caused by the inhomogeneous illumination pattern. In the following experiments, 1.4 was set to be the cutoff value of contrast for suspicious tissues. c Histological section of the tumor-free specimen resected from the wound. Scale bar 1 mm. PAI photoacoustic imaging, H&E hematoxylin and eosin

FIG. 4

Fluorescent detection of residual disease in the surgical bed after incomplete resection and BMI to track local recurrence in the surgical bed. a The mouse was imaged after surgically removing the primary tumor guided by PAI. The residual disease could not be visualized by the surgeon; however, it was clearly detected by the NIR fluorescence imaging system. After additional surgery, no residual disease was detected. The contrast of the suspicious tissue was approximately 3:1 compared with the background normal tissue. The high-power magnification and low-power magnification histological sections confirmed the tumor cells in this specimen. Scale bar 2 mm. b The invisible small residual disease was fluorescently detected by the intraoperative planar fluorescence imaging system. Although the size of the suspicious tissue was much smaller than the primary tumor in this figure, the contrast was almost the same (3:1). H&E staining confirmed the presence of residual cancer cells in the suspicious tissues. Higher magnification microscopic image of the tumor tissue section showed the residual nodule interface. Scale bar 2 mm. c BMI on the 14th, 21st, and 28th day post-operation. Overall, 8.7 and 33.3 % of the mice in the targeted and control animal groups, respectively, had local recurrence, and approximately 30 % of the mice in both groups had lymph node metastasis. In the control group, most recurrent tumors grew larger than 1 cm within 3 weeks post-operation. BMI bioluminescence imaging, PAI photoacoustic imaging, NIR near-infrared, H&E hematoxylin and eosin