A liquid chromatography-mass spectrometry method for the simultaneous measurement of 15 urinary estrogens and estrogen metabolites: assay reproducibility and interindividual variability

Abstract

Background: Accurate, reproducible, and sensitive measurements of endogenous estrogen exposure and individual patterns of estrogen metabolism are needed for etiologic studies of breast cancer. We have developed a high-performance liquid chromatography-tandem mass spectrometry method to quantitate simultaneously 15 urinary estrogens and estrogen metabolites (EM): estrone; estradiol; 3 catechol estrogens; 5 estrogens in the 16alpha pathway, including estriol; and 5 methoxy estrogens.

Methods: Overnight urines were obtained from 45 participants. For the reproducibility study, two blinded, randomized aliquots from 5 follicular and 5 luteal premenopausal women, 5 naturally postmenopausal women, and 5 men were assayed in each of four batches. Assay coefficients of variation and intraclass correlation coefficients were calculated with ANOVA models. Data from the additional 25 participants were added to compare EM levels by menstrual/sex group and assess interindividual variability.

Results: For each EM, overall coefficients of variation were < or = 10%. Intraclass correlation coefficients for each menstrual/sex group were generally > or = 98%. Although geometric mean EM concentrations differed among the four groups, rankings were similar, with estriol, 2-hydroxyestrone, estrone, estradiol, and 16-ketoestradiol accounting for 60% to 75% of total urinary EM. Within each group, interindividual differences in absolute concentrations were consistently high; the range was 10- to 100-fold for nearly all EM.

Conclusion: Our high-performance liquid chromatography-tandem mass spectrometry method for measuring 15 urinary EM is highly reproducible, and the range of EM concentrations in each menstrual/sex group is quite large relative to assay variability. Whether these patterns persist in blood and target tissues awaits further development and application of this method.

Figure 1

Box plots of concentrations for urinary estrogens and estrogen metabolites (EM) in urine (y axis) for 10 premenopausal follicular phase women, 10 premenopausal luteal phase women, 15 postmenopausal women, and 10 men (x axis). Figure 1a includes estrone (E₁) and estradiol (E₂); Figure 1b, the catechol estrogens: 2-hydroxyestrone (2-OHE₁), 2-hydroxyestradiol (2-OHE₂), and 4-hydroxyestrone (4-OHE₁); Figure 1c, the 16α pathway estrogens: 16α-hydroxyestrone (16α-OHE₁), 17-epiestriol 17-epiE₃), estriol (E3), 16-ketoestradiol 16-ketoE₂) and 16-epiestriol (16epiE₃); Figure 1d, the methoxy estrogens: 2-methoxyestrone (2-MeOE₁), 2-methoxyestradiol (2-MeOE₂), 2-hydroxyestrone-3-methyl ether (3-MeOE₁), 4-methoxyestrone (4-MeOE₁), and 4-methyoxyestradiol (4-MeOE₂). The y-axis shows pg EM/mg creatinine on a logarithmic scale. The horizontal line within the box represents the median value of the distribution; the top and bottom of the box represent, respectively, the 75^th and 25^th percentiles of the distribution. Outliers are represented as stars [(>1.5 but ≤3 times the interquartile range (IQR) and open circles (> 3 times the IQR).

Figure 2

Results of the assay reproducibility study for 16α-hydroxyestrone (16α-OHE₁). The concentration of 16α-OHE₁ (pg metabolite / ml urine) is plotted on a logarithmic scale on the y axis, and the x axis identifies the four batches. Duplicate readings are shown in each batch for 5 premenopausal follicular women (Figure 2a); 5 premenopausal luteal women (Figure 2b), 5 postmenopausal women (Figure 2c), and 5 males (Figure 2d). Each symbol in each of the plots corresponds to a different subject in that menstrual/sex group. If the values for the duplicate readings for an individual in a batch overlapped, only one symbol is shown. Similar plots for the other 14 estrogens and estrogen metabolites (EM) are presented in the online Appendix, Figures 3–16.