Metabolic Subtyping of Adrenal Tumors: Prospective Multi-Center Cohort Study in Korea

被引：13

作者：

Ku, Eu Jeong ^{[1
]}

Lee, Chaelin ^{[2
]}

Shim, Jaeyoon ^{[2
]}

Lee, Sihoon ^{[3
]}

Kim, Kyoung-Ah ^{[4
]}

Kim, Sang Wan ^{[5
]}

Rhee, Yumie ^{[6
]}

Kim, Hyo-Jeong ^{[7
]}

Lim, Jung Soo ^{[8
]}

Chung, Choon Hee ^{[8
]}

Chun, Sung Wan ^{[9
]}

Yoo, Soon-Jib ^{[10
]}

Ryu, Ohk-Hyun ^{[11
]}

Cho, Ho Chan ^{[12
]}

Hong, A. Ram ^{[13
]}

Ahn, Chang Ho ^{[14
]}

Kim, Jung Hee ^{[15
]}

Choi, Man Ho ^{[2
]}

机构：

[1] Chungbuk Natl Univ, Chungbuk Natl Univ Hosp, Dept Internal Med, Coll Med, Cheongju, South Korea

[2] Korea Inst Sci & Technol, Mol Recognit Res Ctr, 5 Hwarang Ro 14 Gil, Seoul 02792, South Korea

[3] Gachon Univ, Dept Internal Med, Coll Med, Incheon, South Korea

[4] Dongguk Univ, Coll Med, Dept Internal Med, Ilsan Hosp, Goyang, South Korea

[5] Seoul Natl Univ, Dept Internal Med, Seoul Metropolitan Govt, Boramae Med Ctr,Coll Med, Seoul, South Korea

[6] Yonsei Univ, Dept Internal Med, Coll Med, Seoul, South Korea

[7] Eulji Univ, Nowon Eulji Med Ctr, Dept Internal Med, Seoul, South Korea

[8] Yonsei Univ, Dept Internal Med, Wonju Coll Med, Wonju, South Korea

[9] Soonchunhyang Univ, Coll Med, Dept Internal Med, Cheonan Hosp, Cheonan, South Korea

[10] Catholic Univ Korea, Coll Med, Dept Internal Med, Div Endocrinol & Metab,Bucheon St Marys Hosp, Bucheon, South Korea

[11] Hallym Univ, Coll Med, Dept Internal Med, Chuncheon Sacred Heart Hosp, Chunchon, South Korea

[12] Keimyung Univ, Dept Internal Med, Sch Med, Daegu, South Korea

[13] Chonnam Natl Univ, Dept Internal Med, Med Sch, Gwangju, South Korea

[14] Seoul Natl Univ, Bundang Hosp, Dept Internal Med, Coll Med, Gwangju, South Korea

[15] Seoul Natl Univ, Seoul Natl Univ Hosp, Dept Internal Med, Coll Med, 101 Daehak Ro, Seoul 03080, South Korea

来源：

ENDOCRINOLOGY AND METABOLISM | 2021年 / 36卷 / 05期

基金：

新加坡国家研究基金会;

关键词：

Steroid metabolism; Supervised machine learning; Adrenal neoplasm; Cushing syndrome; Primary hyperaldosteronism; PRIMARY ALDOSTERONISM; CUSHINGS-SYNDROME; DIAGNOSIS; 18-HYDROXYCORTISOL; 18-OXOCORTISOL; SOCIETY; MANAGEMENT; SECRETION; MS/MS;

D O I：

10.3803/EnM.2021.1149

中图分类号：

R5 [内科学];

学科分类号：

1002 ; 100201 ;

摘要：

Background: Conventional diagnostic approaches for adrenal tumors require multi-step processes, including imaging studies and dynamic hormone tests. Therefore, this study aimed to discriminate adrenal tumors from a single blood sample based on the combination of liquid chromatography-mass spectrometry (LC-MS) and machine learning algorithms in serum profiling of adrenal steroids. Methods: The LC-MS-based steroid profiling was applied to serum samples obtained from patients with nonfunctioning adenoma (NFA. n=73). Cushing's syndrome (CS, n=30), and primary aldosteronism (PA, n=40) in a prospective multicenter study of adrenal disease. The decision tree (DT), random forest (RF), and extreme gradient boost (XGBoost) were performed to categorize the subtypes of adrenal tumors. Results: The CS group showed higher scrum levels of 11-deoxycortisol than the NFA group, and increased levels of tctrahydrocorti-sone (THE), 20 alpha-dihydrocortisol, and 60-hydroxycortisol were found in the PA group. However, the CS group showed lower levels of dehydroepiandrosterone (DHEA) and its sulfate derivative (DHEA-S) than both the NFA and PA groups. Patients with PA expressed higher serum 18-hydroxycortisol and DHEA but lower THE than NFA patients. The balanced accuracies of DT, RF, and XGBoost for classifying each type were 78%, 96%, and 97%, respectively. In receiver operating characteristics (ROC) analysis for CS, XGBoost, and RF showed a significantly greater diagnostic power than the DT However, in ROC analysis for PA, only RF exhibited better diagnostic performance than DT. Conclusion: The combination of LC-MS-based steroid profiling with machine learning algorithms could be a promising one-step diagnostic approach for the classification of adrenal tumor subtypes.

引用

页码：1131 / 1141

页数：11

共 37 条

[31] An LC-MS/MS method for steroid profiling during adrenal venous sampling for investigation of primary aldosteronism [J].