Clear cell renal cell carcinoma detection by multimodal photoacoustic tomography

Abstract

There is a need for accurate and rapid detection of renal cancer in clinic. Here, we integrated photoacoustic tomography (PAT) with ultrasound imaging in a single system, which achieved tissue imaging depth about 3 mm and imaging speed about 3.5 cm²/min. We used the wavelength at 1197 nm to map lipid distribution in normal renal tissues and clear cell renal cell carcinoma (ccRCC) tissues collected from 19 patients undergone nephrectomy. Our results indicated that the photoacoustic signal from lipids was significantly higher in ccRCC tissues than that in normal tissues. Moreover, based on the quantification of lipid area ratio, we were able to differentiate normal and ccRCC with 100 % sensitivity, 80 % specificity, and area under receiver operating characteristic curve of 0.95. Our findings demonstrate that multimodal PAT can differentiate normal and ccRCC by integrating the morphologic information from ultrasound and lipid amount information from vibrational PAT.

Keywords: Clear cell renal cell carcinoma; Lipid accumulation; Photoacoustic tomography.

Fig. 1

Schematic of the multimodal PAT system and the procedures of imaging. (a) The PAT system generates the laser output at 1197 nm. The collinear probe combines the optical fiber and the ultrasonic transducer to transmit and receive signals, which are processed by a high-frequency ultrasound imaging system. (b) Schematic to show acquisition of multimodal PAT images. US: ultrasound; PAT: photoacoustic tomography.

Fig. 2

Characterization of multimodal PAT system performance. (a-b) Ultrasound and PAT images of five polyethylene tubes in an agarose gel phantom at different depths. The PAT images of polyethylene tubes were taken by excitation at 1197 nm. (c) Signal-to-noise ratios of PAT images of polyethylene tubes at different depths. (d) PAT image of a 50 μm tungsten ∼1 mm beneath the probe in water. (e) Axial resolution measured at 1 mm depth. (f) Lateral resolution measured at 1 mm depth. The color bar shows the PA amplitude and the unit for the color bar is arbitrary unit. Black spot: experimental data; Red curve: Gaussian fitting curve; US: ultrasound; PAT: photoacoustic tomography.

Fig. 3

Representative multimodal PAT images of normal and ccRCC tissues. (a-b) Multimodal PAT images of the normal tissue. (c) H&E image of the corresponding normal tissue. (d-e) Multimodal PAT images of the ccRCC tissue. (f) H&E image of the corresponding ccRCC tissue. Grey color: US image; Green color: PAT image at 1197 nm; Scalar bars, 1 mm. ccRCC: clear cell renal cell carcinoma; The color bar shows the PA amplitude and the unit for the color bar is arbitrary unit. US: ultrasound; PAT: photoacoustic tomography.

Fig. 4

Differentiation between normal and ccRCC tissues by multimodal PAT system. (a) The lipid area ratios in normal and ccRCC tissues, respectively. The scattered dots represent area ratios for each tissue. Error bars represent the standard error of the mean (SEM). *** p < 0.0001; n.s., not significant. (b) The ROC curve illustrating the ability of using the lipid area ratio to distinguish normal tissues and ccRCC tissues (orange curve). The closer the curve comes to the 45-degree diagonal (black line), the less accurate the diagnostic test. ROC: receiver operating characteristic; ccRCC: clear cell renal cell carcinoma; AUC: area under curve.