Three-dimensional quantitative ultrasound for detecting lymph node metastases

Abstract

Purpose: Detection of metastases in lymph nodes (LNs) is critical for cancer management. Conventional histological methods may miss metastatic foci. To date, no practical means of evaluating the entire LN volume exists. The aim of this study was to develop fast, reliable, operator-independent, high-frequency, quantitative ultrasound (QUS) methods for evaluating LNs over their entire volume to effectively detect LN metastases.

Methods: We scanned freshly excised LNs at 26 MHz and digitally acquired echo-signal data over the entire three-dimensional (3D) volume. A total of 146 LNs of colorectal, 26 LNs of gastric, and 118 LNs of breast cancer patients were enrolled. We step-sectioned LNs at 50-μm intervals and later compared them with 13 QUS estimates associated with tissue microstructure. Linear-discriminant analysis classified LNs as metastatic or nonmetastatic, and we computed areas (Az) under receiver-operator characteristic curves to assess classification performance. The QUS estimates and cancer probability values derived from discriminant analysis were depicted in 3D images for comparison with 3D histology.

Results: Of 146 LNs of colorectal cancer patients, 23 were metastatic; Az = 0.952 ± 0.021 (95% confidence interval [CI]: 0.911-0.993); sensitivity = 91.3% (specificity = 87.0%); and sensitivity = 100% (specificity = 67.5%). Of 26 LNs of gastric cancer patients, five were metastatic; Az = 0.962 ± 0.039 (95% CI: 0.807-1.000); sensitivity = 100% (specificity = 95.3%). A total of 17 of 118 LNs of breast cancer patients were metastatic; Az = 0.833 ± 0.047 (95% CI: 0.741-0.926); sensitivity = 88.2% (specificity = 62.5%); sensitivity = 100% (specificity = 50.5%). 3D cancer probability images showed good correlation with 3D histology.

Conclusions: These results suggest that operator- and system-independent QUS methods allow reliable entire-volume LN evaluation for detecting metastases. 3D cancer probability images can help pathologists identify metastatic foci that could be missed using conventional methods.

Figure 1

Receiver-operator characteristic curves for LNs of the three cancer types. Colorectal cancer, blue line; gastric cancer, black line; breast cancer, green line. The receiver-operator characteristic curves indicate superior classification performance for colorectal and gastric cancers compared to breast cancer readily apparent.

Figure 2

Illustrative quantitative ultrasound (QUS) images: three-dimensional (3D) cross-sectional parametric images displaying effective scatterer-size estimates. Figure 2-a shows 3D cross-sectional parametric images of a benign locoregional lymph node (LN) from a colon cancer patient. Figure 2-b shows 3D cross-sectional parametric images of a locoregional LN with diffusely metastatic adenocarcinoma from a different colon cancer patient. Parametric cross-sectional B-mode images are shown with overlaid color-coded effective scatterer-size estimates. Bottom right panel of each figure displays the co-registered histology of the bottom left panel.

Figure 3

Illustrative quantitative-ultrasound (QUS) images: three-dimensional (3D) cross-sectional parametric images displaying effective scatterer-size estimates. Figure 3-a shows 3D cross-sectional parametric images of a benign locoregional lymph node (LN) from a gastric cancer patient. Figure 3-b shows 3D cross-sectional parametric images of a locoregional LN with diffusely metastatic adenocarcinoma from a different gastric cancer patient. Parametric cross-sectional B-mode images are shown with overlaid color-coded effective scatterer-size estimates. Bottom right panel of each figure displays the co-registered histology of the bottom left panel.

Figure 4

1: 3D Interactive GUI with cancer-probability images of a locoregional lymph node (LN) with partially metastatic adenocarcinoma from a patient with colorectal cancer. The LN is 9.54 mm in its largest dimension, and the metastasis is 5.09 mm in its largest dimension. The graphical user interface (GUI) displays three orthogonal gray-scale B-mode cross-sections from a three-dimensional (3D) rendering in 4-1a through 4-1c. The cross sections depict color-encoded cancer-probability values using red to indicate a probability greater than 75%, orange to indicate a probability between 25 and 75%, and green to indicate a probability less than 25%. Figure 4-1d shows a co-registered hematoxylin and eosin (H&E) stained histology photomicrograph that corresponds to the same section as in 4-1c. These images show that excellent concurrence is achieved between the red cancer-probability region and the definitive histology result shown in 4-1d showing the demarcated metastatic tumor. 2: 3D Interactive GUI with cancer-probability images of a benign locoregional lymph node (LN) with from a patient with colorectal cancer. The LN is 4.41 mm in its largest dimension. The graphical user interface (GUI) displays three orthogonal gray-scale B-mode cross-sections from a three-dimensional (3D) rendering in 4-2a through 4-2c. Figure 4-2d shows a co-registered hematoxylin and eosin (H&E) stained histology photomicrograph that corresponds to the same section shown in 4-2c. The cross sections depict color-encoded cancer-probability values using red to indicate a probability greater than 75%, orange to indicate a probability between 25 and 75%, and green to indicate a probability less than 25%. These images show that excellent concurrence is achieved between the green cancer-probability region and the definitive histology result of the benign LN shown in 4-2d.

Figure 5

1: Cancer-probability images of an axillary sentinel-lymph node (LN) of invasive ductal breast cancer patient. The LN is 5.86 mm in its largest dimension, and it contains two micrometastatic foci. The bigger focus is 1.82 mm in its largest dimension. The graphical user interface (GUI) displays three orthogonal gray-scale B-mode cross-sections from a three-dimensional (3D) rendering in 5-1a through 5-1c. The cross sections depict color-encoded cancer-probability values using red to indicate a probability greater than 75%, orange to indicate a probability between 25 and 75%, and green to indicate a probability less than 25%. Figure 5-1d shows a co-registered hematoxylin and eosin (H&E) stained histology photomicrograph that corresponds to the same section shown in 5-1c. Like Figure 4-1, Figure 5-1 shows excellent concurrence between the red high-probability region and the corresponding metastatic region in the histology result. 2: Cancer-probability images of a benign axillary sentinel-lymph node (LN) of invasive ductal breast cancer patient. The LN is 5.51 mm in its largest dimension. The cross sections depict color-encoded cancer-probability values using red to indicate a probability greater than 75%, orange to indicate a probability between 25 and 75%, and green to indicate a probability less than 25%. Like Figure 4-2, Figure 5-2 shows excellent concurrence is achieved between the green cancer-probability region and the definitive histology result of the benign LN shown in 5-2d.