Anh Nguyen Hoang, Long Nguyen Phuoc, Kim Sun Jo, Min Jung Eun, Yoon Sang Jun, Kim Hyung Min, Yang Eugine, Hwang Eun Sook, Park Jeong Hill, Hong Soon-Sun, Kwon Sung Won
College of Pharmacy, Seoul National University, Seoul 08826, Korea.
College of Pharmacy, Ewha Womans University, Seoul 03760, Korea.
Metabolites. 2019 Sep 21;9(10):199. doi: 10.3390/metabo9100199.
Steroidomics, an analytical technique for steroid biomarker mining, has received much attention in recent years. This systematic review and functional analysis, following the PRISMA statement, aims to provide a comprehensive review and an appraisal of the developments and fundamental issues in steroid high-throughput analysis, with a focus on cancer research. We also discuss potential pitfalls and proposed recommendations for steroidomics-based clinical research. Forty-five studies met our inclusion criteria, with a focus on 12 types of cancer. Most studies focused on cancer risk prediction, followed by diagnosis, prognosis, and therapy monitoring. Prostate cancer was the most frequently studied cancer. Estradiol, dehydroepiandrosterone, and cortisol were mostly reported and altered in at least four types of cancer. Estrogen and estrogen metabolites were highly reported to associate with women-related cancers. Pathway enrichment analysis revealed that steroidogenesis; androgen and estrogen metabolism; and androstenedione metabolism were significantly altered in cancers. Our findings indicated that estradiol, dehydroepiandrosterone, cortisol, and estrogen metabolites, among others, could be considered oncosteroids. Despite noble achievements, significant shortcomings among the investigated studies were small sample sizes, cross-sectional designs, potential confounding factors, and problematic statistical approaches. More efforts are required to establish standardized procedures regarding study design, analytical procedures, and statistical inference.
类固醇组学是一种用于挖掘类固醇生物标志物的分析技术，近年来备受关注。本系统综述和功能分析遵循PRISMA声明，旨在对类固醇高通量分析的进展和基本问题进行全面综述和评估，重点关注癌症研究。我们还讨论了基于类固醇组学的临床研究的潜在陷阱并提出了建议。45项研究符合我们的纳入标准，重点关注12种癌症类型。大多数研究集中在癌症风险预测，其次是诊断、预后和治疗监测。前列腺癌是研究最频繁的癌症。雌二醇、脱氢表雄酮和皮质醇在至少四种癌症类型中大多被报道且发生了改变。雌激素和雌激素代谢物与女性相关癌症的关联报道很多。通路富集分析表明，类固醇生成、雄激素和雌激素代谢以及雄烯二酮代谢在癌症中显著改变。我们的研究结果表明，雌二醇、脱氢表雄酮、皮质醇和雌激素代谢物等可被视为肿瘤类固醇。尽管取得了显著成就，但所调查研究中的重大缺点是样本量小、横断面设计、潜在混杂因素和有问题的统计方法。需要更多努力来建立关于研究设计、分析程序和统计推断的标准化程序。
