Over the last decades, many studies based on gene or protein expression in vitro and in vivo have evaluated the role of biomarkers in determinations of thyroid tumor malignancy [6–10], but candidate biomarkers such as galectin-3 or MET have not shown the sensitivity and specificity needed for a preoperative thyroid nodules screening tool . We previously evaluated metabolic changes in urinary steroid levels in premenopausal and postmenopausal women and men with PTCs by quantitative steroid profiling, as PTC pathogenesis and development might be affected by androgens and estrogens . However, these results did not reveal information about the changes in estrogen metabolism that could distinguish PTCs from benign thyroid tumors. Therefore, this study focused more on the metabolic differences between benign tumors and early and advanced-stage malignant PTCs in postmenopausal women.
The 2-MeO-E2 levels significantly decreased in both early and advanced PTCs, suggesting that this could serve as a potential biomarker to differentiate benign from malignant thyroid neoplasms. 2-MeO-E2, which acts as a potential antitumor agent in several types of cancers [29, 30], blocks cell growth and induces apoptosis in thyroid carcinoma cells by activating the p38 mitogen-activated protein kinase . Based on its cellular effects, 2-MeO-E2 might be expected to attenuate the progression of tumorigenesis and protect against the development of malignant PTCs from benign adenomas.
Estrogen metabolism was shown to potentiate the growth of several cancers, including breast, prostate, endometrial, and thyroid cancers [12–14, 20–23, 25, 31]. For E2, the C-2/C-16α-hydroxylation ratio and level of 16α-hydroxylation could be used as predictive biomarkers for breast cancer [20–22], suggesting that the major estrogen metabolism pathways have different biological actions such as the antiproliferative effect of 2-hydroxylation and proliferative effect of 16α-hydroxylation in hormone-related cancers [12, 18–23]. In this study, 16α-hydroxylation of both E1 and E2 was markedly increased in both early and advanced-stage malignant PTCs. In particular, the influence of estrogen 16α-hydroxylation tended to be in clear contrast to that of 2-hydroxylation in postmenopausal low- and high-grade PTCs (Table 2).
Increased 16α-hydroxylation activity preceded clinical evidence of cancer and therefore might represent a significant risk factor for estrogen-dependent tumor development [20–23]. In this study, the 2-OH-E1-to-16α-OH-E1 ratio was not statistically significantly different between the malignant PTC and benign groups, despite the lower activity in the former group. However, a decreased 2-OH-E2-to-E3 ratio was observed. For E2, the ratio of C-2 and C-16α hydroxylation in the malignant PTC groups might reflect the observed effects of increased 16α-hydroxylation on malignant PTC in postmenopausal women. This observation is consistent with previous studies that examined the extent of the interplay between 2-hydroxylation and 16α-hydroxylation and reflected that the estrogen metabolite ratio could be a distinguishing feature between benign and malignant thyroid tumors .
Another interesting estrogen profile pattern was detected between the benign and malignant tumors. The expression of reductive 17β-HSD in patients with both early and late-stage PTCs was significantly higher than that in the benign patients; this affected the conversions of E1 to E2 and 16α-OH-E1 to E3 (Table 2). Although the underlying mechanism is not clear, the E3-to-16α-OH-E1 ratio progressively increased from the benign group to the early stage malignant PTC group and finally to the advanced-stage malignant PTC group, and had a more significant P value than the E2 to E1 ratio. In recent, the implications of 17β-HSD activity in tumorigenesis have been investigated [32–34]. In these studies, high 17β-HSD activity was related to increased cell proliferation, and it was important to the understanding of the mechanisms that underlie breast cancer development [32, 33]. Reductive 17β-HSD is also overexpressed in endometrial cancer, compared with control tissues, and was found to increase tumor cell proliferation . To our knowledge, this is the first study to demonstrate an association between 17β-HSD expression and pathological features in patients with PTCs.
This study had 2 limitations. First, a small number of postmenopausal women with thyroid disease (only 27 individuals) limited the statistical power with which to distinguish between benign and malignant tumors. Despite these drawbacks, however, very few fundamental studies on this topic have been performed in postmenopausal women. Although we could not conclusively determine the associations between estrogen metabolism and pathological features in patients with PTCs, we have demonstrated statistically significant estrogen metabolic patterns and differentiated metabolic pathways. Second, high variability in the urinary estrogen levels was detected in all postmenopausal patients, including those of advanced age. Most circulating estrogens in elderly women are derived from adipose tissues and can be affected by additional factors such as nutrition, smoking, alcohol intake, and physical activity . There is considerable intra- and inter-patient estrogen variability among postmenopausal women in response to estrogen replacement therapy . In contrast to the metabolic ratios, the urinary levels of the individual estrogens, except 2-MeO-E2, were not significantly different between benign and malignant thyroid tumor patients because of the high variability in the urinary estrogen levels. Further studies with a large population of individuals with thyroid tumors are needed to better define the changes in individual estrogens in PTC patients.