The 2-/16alpha-Hydroxylated Estrogen Ratio-Breast Cancer Risk Hypothesis: Insufficient Evidence for its Support: Running title: 2-/16alpha-Hydroxyestrone-Breast Cancer Risk Hypothesis.

Abstract During the past 25 years or so a number of studies have been carried out to address the hypothesis that the ratio of 2-hydroxyestrone (2-hydroxy-E1) to 16α-hydroxyestrone (16α-hydroxy-E1) is associated with breast cancer risk. The rationale for this hypothesis is based on data from studies that suggest a tumorigenic and genotoxic effect of 16α-hydroxy-E1 and a protective effect of 2-hydroxy-E1 regarding breast cancer risk. The adverse effect of 16α-hydroxy-E1 has been attributed to its potential to form covalent adducts with macromolecules. Initial studies used radiometric assays and enzyme immunoassays to test the hypothesis. However, concerns about the accuracy of these assays led to the development of a liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay that is capable of measuring 5 unconjugated and 15 conjugated endogenous estrogens, which include 2- and 16-hydroxylated estrogen metabolites, in serum or urine. The conjugated estrogens are quantified following a deconjugation (hydrolysis) step to remove the sulfate and glucuronide groups. Epidemiologic studies have been using the LC-MS/MS assay to determine whether there is an association between breast cancer risk and the ratio of the sum of the concentrations of metabolites in the 2-hydroxylated estrogen pathway and in the 16-hydroxylated estrogen pathway. However, the validity of the pathways as biomarkers was not evaluated. The 16-hydroxylated estrogen pathway includes estriol, 16-epiestriol, 17-epiestriol and 16-ketoestradiol, in addition to 16α-hydroxy-E1. However, with the exception of 16α-hydroxy-E1, there is no evidence that any of the other estrogens in the pathway have tumorigenic or genotoxic properties, and they do not form covalent adducts with macromolecules. Another deficiency in the epidemiological studies pertains to the accuracy of estrogen metabolite measurements obtained after the hydrolysis step in the LC-MS/MS assays. No validation was performed to demonstrate that a constant efficiency of hydrolysis is found for all the different structural forms of sulfated and glucuronidated conjugates. Other deficiencies in the assays include the need for greater sensitivity so that the very low concentrations of unconjugated 2-hydroxy-E1, 2-hydroxy-E2, and 16α-hydroxy-E1 can be measured in serum. There is also a need to develop assays to measure intact forms of conjugated estrogens in both serum and urine, particularly the sulfates and glucuronides of 2-hydroxylated, 2-methoxylated, and 16α-hydroxylated E1 and E2. This will avoid inaccuracies that stem from hydrolysis procedures. Improvements in LC-MS/MS assay methodology to obtain accurate measurements of unconjugated and conjugated 2-hydroxylated, 2-methoxylated, and 16α-hydroxylated estrogen metabolites are needed. This should provide valuable data for testing the 2-/16α-hydroxylated estrogen-breast cancer risk hypothesis.