Distinguishing metastases from benign adrenal masses: what can CT texture analysis do?

Abstract

Background

Computed tomography texture analysis (CTTA) has gained an increasing role in oncology and has successfully demonstrated to reflect biological associations with glucose metabolism, hypoxia, angiogenesis, and even genetic variation.

Purpose

To determine whether quantitative CTTA can be used to differentiate metastases from benign adrenal masses on single energy CT images.

Material and Methods

A total of 225 patients with 265 histologically confirmed adrenal masses (101 metastases, 98 pheochromocytomas, and 66 lipid-poor adenomas) were included in this retrospective study. CTTA was performed and six texture parameters (including mean, SD of pixel distribution histogram, mean of positive pixels, entropy, kurtosis, skewness) across six spatial scaling factor (SSF) were recorded on both unenhanced and contrast-enhanced CT images. Receiver operating characteristic (ROC) analysis was performed and the area under the ROC curve (AUC) was calculated using the significant texture parameters for the objective. Diagnostic performance was evaluated using the cut-off values of texture parameters by ROC analysis. The optimal discriminative texture parameters were used to produce support vector machine (SVM) classifiers. Diagnostic accuracy and 10-fold cross-validation was performed.

Results

Compared to benign adrenal masses, metastases had significantly lower mean gray-level intensity, SD, entropy, mean of positive pixels and kurtosis on unenhanced images (P < 0.0083). On contrast-enhanced CT images, except for skewness and kurtosis, the other four texture texture-quantifiers were lower in the metastatic compared to the other group (P < 0.0083). A model incorporating mean, SD, entropy and mean value of positive pixels produced an AUC of 0.85 ± 0.03 with an SVM accuracy of 77% to identify metastatic from benign adrenal masses.

Conclusion

Differentiation of metastatic and benign adrenal masses can be achieved accurately with CTTA on contrast-enhanced CT images.

Keywords

Adrenal metastasis CT texture analysis

Get full access to this article

View all access options for this article.

References

Boland

GWL

Blake

Hahn

et al. Incidental adrenal lesions: principles, techniques, and algorithms for imaging characterization. Radiology 2008; 249:756–775.

Lam

. Metastatic tumours of the adrenal glands: a 30-year experience in a teaching hospital. Clin Endocrinol 2002; 56:95–101.

Sahdev

Willatt

Francis

et al. The indeterminate adrenal lesion. Cancer Imaging 2010; 10:102–113.

Fergany

. Adrenal masses: A urological perspective. Arab J Urol 2016; 14:248–255.

Hönigschnabl

Gallo

Niederle

et al. How accurate is MR imaging in characterisation of adrenal masses: Update of a long-term study. Eur J Radiol 2002; 41:113–122.

Johnson

Horton

Fishman

. Adrenal imaging with multidetector CT: evidence-based protocol optimization and interpretative practice. RadioGraphics 2009; 29:1319–1331.

Schieda

Al Dandan

Kielar

et al. Pitfalls of adrenal imaging with chemical shift MRI. Clin Radiol 2014; 69:1186–1197.

Choi

Kim

Park

et al. Evaluation of adrenal metastases from renal cell carcinoma and hepatocellular carcinoma: Use of delayed contrast-enhanced CT. Radiology 2013; 190:2018–2019.

Taner

Schieda

Siegelman

. Pitfalls in adrenal imaging. Semin Roentgenol 2015; 50:260–272.

10.

Garrett

Nepute

Hayek

et al. Adrenal incidentalomas: clinical controversies and modified recommendations. Am J Roentgenol 2016; 206:1170–1178.

11.

Hodgdon

Mcinnes

MDF

Schieda

et al. Can quantitative CT texture analysis be used to differentiate fat-poor renal angiomyolipoma from renal cell carcinoma on unenhanced CT images? Radiology 2015; 276:787–796.

12.

Lubner

Stabo

Abel

et al. CT textural analysis of large primary renal cell carcinomas: pretreatment tumor heterogeneity correlates with histologic findings and clinical outcomes. Am J Roentgenol 2016; 207:96–105.

13.

Hayano

Tian

Kambadakone

et al. Texture analysis of non-contrast-enhanced computed tomography for assessing angiogenesis and survival of soft tissue sarcoma. J Comput Assist Tomogr 2015; 39:607–612.

14.

Dang

Lysack

Wet al

. MRI texture analysis predicts p53 status in head and neck squamous cell carcinoma. Am J Neuroradiol 2005; 36:166–170.

15.

Ganeshan

Miles

. Quantifying tumour heterogeneity with CT. Cancer Imaging 2013; 13:140–149.

16.

Guan

Chen

et al. Adrenal incidentaloma: machine learning-based quantitative texture analysis of unenhanced CT can effectively differentiate sPHEO from lipid-poor adrenal adenoma. J Cancer. 2018; 19:3577–3582.

17.

Miles

Ganeshan

Hayball

. CT texture analysis using the filtration-histogram method: What do the measurements mean? Cancer Imaging 2013; 13:400–406.

18.

Chang

Lin

. LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol 2011; 2:1–27.

19.

Kuo

Jamshidi

. Behind the numbers: decoding molecular phenotypes with radiogenomics—guiding principles and technical considerations. Radiology 2014; 270:320–325.

20.

Suo

Cheng

Cao

. Assessment of heterogeneity difference between edge and core by using texture analysis: differentiation of malignant from inflammatory pulmonary nodules and masses. Acad Radiol 2016; 23:1115–1122.

21.

Campbell DH, Barrett M, Ramanathan RK, et al. Quantitative Textural Analysis (QTA) in CT imaging: identifying markers for genetic instability and overall survival in cohort of previously treated metastatic pancreatic cancer (mPC). AACR Annual Meeting 2014. 2; April 5–9. San Diego, CA.

22.

Miles

Ganeshan

Hayball

. CT texture analysis using the filtration-histogram method: What do the measurements mean? Cancer Imaging 2013; 13:400–406.

23.

Andersen

Harders

Ganeshan

et al. CT texture analysis can help differentiate between malignant and benign lymph nodes in the mediastinum in patients suspected for lung cancer. Acta Radiol 2015; 57:669–676.

24.

Bayanati

E Thornhill

Souza

et al. Quantitative CT texture and shape analysis: Can it differentiate benign and malignant mediastinal lymph nodes in patients with primary lung cancer? Eur Radiol 2015; 25:480–487.

25.

Warda

MHA

Shehata

Zaiton

. Chemical-shift MRI versus washout CT for characterizing adrenal incidentalomas. Clin Imaging 2016; 40:780–787.

26.

Kozarski

Ganeshan

et al. Assessment of tumor heterogeneity by CT texture analysis: Can the largest cross-sectional area be used as an alternative to whole tumor analysis? Eur J Radiol 2013; 82:342–348.

27.

Ganeshan

Miles

Young

et al. Three-dimensional textural analysis of brain images reveals distributed grey-matter abnormalities in schizophrenia. Eur Radiol 2010; 20:941–948.

28.

Zhao

Oxnard

Moskowitz

et al. A pilot study of volume measurement as a method of tumor response evaluation to aid biomarker development. Clin Cancer Res 2010; 16:4647–4653.