Potential role of ultra-sensitive estradiol assays in estimating the risk of breast cancer and fractures

Richard J. Santen, Jennifer S. Lee, Shuping Wang, Lawrence M. Demers, Nelly Mauras, Hongkun Wang, Ravinder Singh

Research output: Contribution to journalArticlepeer-review

45 Scopus citations


Radioimmunoassays (RIA) for the measurement of estradiol are sufficiently sensitive to assess the reproductive status of pre-menopausal women but lack sufficient sensitivity for low concentrations found in post-menopausal women. Bioassays have been used in the past to measure low estrogen levels but are impractical for handling high volumes of tests, particularly routine and non-research specimens. In this study, we compared results for estradiol using several different methods including bioassay, RIA, and two tandem mass spectrometry methods. At the lower tertile of estradiol measurements by RIA, the overall excellent correlation with results obtained by tandem mass spectrometry (i.e. r = 0.83) was lost (i.e. r = 0.29). In addition, results were much lower with bioassay and mass spectrometry than with RIA suggesting that RIA measures undesired noise or estrogen metabolites. The mass spectrometry methods correlate best with isotopic kinetic methods when assessing aromatase inhibition. On this basis, we conclude that mass spectrometry assays are the best option for measurement of low estradiol concentrations. With such assays, greater discrimination should be achievable when using estradiol levels as a predictor of the risks for breast cancer and for fractures.

Original languageEnglish (US)
Pages (from-to)1318-1321
Number of pages4
Issue number13
StatePublished - Dec 12 2008

All Science Journal Classification (ASJC) codes

  • Biochemistry
  • Molecular Biology
  • Endocrinology
  • Pharmacology
  • Clinical Biochemistry
  • Organic Chemistry


Dive into the research topics of 'Potential role of ultra-sensitive estradiol assays in estimating the risk of breast cancer and fractures'. Together they form a unique fingerprint.

Cite this