br studies have shown that increased blood
studies have shown that increased blood levels of steroids including es-trogens, androgens and adrenal precursors are associated with in-creased risk of EC [4–6]. For instance, obesity is an important EC risk factor as the adipose tissue represents a major source of circulating es-trogens in postmenopausal women who comprise the majority of EC patients. It is estimated that almost 40% of EC cases can be linked to overweight and obesity [7,8]. Factors like the obesity epidemic and the higher life expectancy contribute explaining the observed increasing in-cidence of EC . Adding to the effects of increased 6-Hydroxydopamine levels in the blood, active steroid hormones and estrogens are generated locally from blood precursors and in turn influence the final exposure of EC tissues to hormone stimulation and cancer risk [10–12].
Although EC is a common cancer type with increasing incidence, im-provements in treatment options have been slow. Clinical applicable and robust biomarkers for better prognostication and risk stratification are needed to better guide post-surgical adjuvant treatment and patient care. Several studies have investigated the potential prognostic value of steroid receptors and key factors in the local steroid metabolism in EC [13–18], and estrogen and progesterone receptors in primary tumor are at present the best validated prognostic biomarkers, although not widely used in the routine clinical setting. However, the potential of cir-culating hormone levels as biomarkers in EC is less thoroughly studied, despite the fact that recent investigations indicated that several steroids are altered in EC patients and may represent novel biomarkers predic-tive of clinical characteristics including the risk of relapse and overall survival .
In this study we investigated if the plasma concentration of a broad range of steroids including estrogens, progestogens, androgens, gluco-corticoids and adrenal precursors was associated with outcome in EC. Since adipose tissue is an important source of steroids, circulating ste-roid levels were compared to fat distribution in patients. Finally, tran-scription profiling was performed to explore whether differences in plasma steroid concentrations were reflected in transcriptional alter-ations in the tumor, and to chart the local steroid metabolism in EC tumors.
2. Materials and methods
A population based patient series was prospectively collected from 2001 to 2015 including patients diagnosed with EC in Hordaland County (Norway). Clinical data were collected as previously described  and patients were surgically staged according to the International Federa-tion of Gynecology and Obstetrics (FIGO) 2009 criteria. From this pa-tient series a subgroup (n = 19, blood samples and mRNA) comprising postmenopausal EC patients with stage 1 (grade 1–3) and stage 2 (grade 2) disease were selected based on short survival time. All these patients recurred within 27 months (average 10 months, stan-dard deviation (SD) 6.8) after primary surgery, and died within 35 months of primary surgery (average 18 months, SD 9.9). These pa-tients were matched with comparable patients (n = 19 for blood sam-ples and n = 14 for mRNA; having the same type, stage and grade, and similar parity, age and BMI) who had no signs of recurrence after a mean (SD) follow-up of 66 (11) months. All patients were postmenopausal.
The study has been approved according to Norwegian legislation by the Western Regional Committee for medical and health Research Ethics (REK 2009/2315). All included patients had given written in-formed consent.
2.2. Steroid metabolite analysis by LC-MS/MS
EDTA-blood was obtained from 38 patients with EC before primary surgery. The blood samples were centrifuged at 1600g for 15 min and the plasma was stored at −80 °C. The following steroids were measured using liquid-chromatography tandem mass-spectrometry (LC-MS/MS),
employing three different protocols. Androsterone (AN), dihydrotestos-terone (DHT), dehydroepiandrosterone-sulphate (DHEAS), cortisol and cortisone were measured with commercially available and ISA certified AbsoluteIDQ® Stero17 kit assay (Biocrates Life Sciences AG, Innsbruck, AU) according to the manufacturer's instruction . Samples were pre-pared through a solid phase extraction in a 96 well plate-format for pre-cleaning and pre-concentration of the target steroid hormones. LC-MS/ MS was then performed in multiple reaction monitoring mode (MRM) using a SCIEX API 4000 QTrap, and steroids were quantified by 7-point external calibration curves and 13 isotope-labeled internal standards. These analyses were performed at the Biocrates laboratory.
Estrone (E1), 17β-estradiol (E2), progesterone (P4), 17OH-progesterone (17OH-P4), 21OH-progesterone (21OH-P4), pregneno-lone (P5), 17OH-pregnenolone (17OH-P5), androstenedione (A4), de-hydroepiandrosterone (DHEA), aldosterone (ALDO), 11-deoxycortisol and corticosterone (CORT) were measured with Agilent 1290 UHPLC - Agilent 6495 QQQ as described earlier . In short, the samples were spiked with isotope-labeled steroids as internal standards, and ex-tracted with toluene prior to LC-MS/MS analysis. Underivatized steroids were separated using a Kinetex biphenyl (100 × 2.1 mm, 1.7u), (Phenomenex) with methanol gradient and 0.2 mM NH4F as eluent additive.