Poster Topical Area: Biomarkers

Location: Auditorium

Poster Board Number: 139

P02-005 - Analysis of 16 estrogen metabolites (EMs) in serum and plasma by LC-MS/MS: Method Validation

Sunday, Jun 10
8:00 AM – 6:00 PM

Objective: To establish an LC-MS/MS method for the measurement of 16 estrogen metabolites in human serum/plasma and to validate the method.

250 µL of sample is combined with 500 µL of acetate buffer (pH 4.6), 50 µL of 50 mg/mL ascorbic acid, and glucuronidase/sulfatase. Incubate at 37 ⁰C overnight. 12 isotopically labeled EMs are added as Internal Standards. Samples are mixed and equilibrated for 1 hour at room temperature. Acetonitrile (500 µL) is added to the samples to precipitate the proteins and free any protein bound EMs. Samples are vortexed and centrifuged (4500 rpm, 10 min). Supernatant is transferred to a 10 mL screw cap tube, mixed with 5 mL of dichloromethane and capped. Tubes are rotated at 40 rpm for 1 hour. Then, samples are frozen at -40 ⁰C for 30 minutes followed by centrifugation (4000 rpm, 1 hour, 4 ⁰C). The dichloromethane extract is transferred to a test tube and dried in a Speedvac™. The dried extract is held at -40 ⁰C for at least 1 hour prior to analysis. Finally, samples are dansylated prior to LC-MS/MS analysis.

EMs were separated using a Synergi 2.5 µm Hydro-RP 100Ȧ 100 x2 mm column (Phenomenex) and a 50 minute gradient of Water/MeOH/Acetone all containing 0.1% formic acid. MS conditions for each analyte were optimized.

A proof-of-performance study was conducted to access the reproducibility of the method for total EMs and unconjugated EMs. Samples included matching serum and heparin-plasma from 5 postmenopausal, 3 premenopausal women, and 2 men. Two batches of serum and two of heparin-plasma were measured in blinded duplicates for all subjects.

Results: 11 of the 16 measured EMs were consistently detectable in the total EM samples. 3-MeO-Estrone and 4-MeO-Estradiol were not detectable in any samples. For the unconjugated EM samples, only estrone and estradiol were consistently measured with CVs

Total EM (within and between batch) CVs are < 8% for combined and individual parent estrogens. Spearman and Pearson correlations between serum and heparin-plasma EMs are excellent, consistently 0.9 or higher. Accuracy of unconjugated estradiol was established by the CDC HOST Program.

Conclusion: The method is sufficiently accurate and precise for measuring EMs in both serum and heparin plasma.

Funding Source: Funded in-part by the Div of Epidemiology and Genetics, National Cancer Institute.

Matthew K. Fleshman

Mass Spec Group Leader
Eurofins Craft Technologies Inc
Wilson, North Carolina