“Harmonized Reference Ranges for Circulating Testosterone Levels in Men of Four Cohort Studies in the USA and Europe”
Thomas G. Travison, Hubert W. Vesper, Eric Orwoll, Frederick Wu, Jean Marc Kaufman, Ying Wang, Bruno Lapauw, Tom Fiers, Alvin M. Matsumoto, Shalender Bhasin
ONLINE: January 2017 – The Journal of Clinical Endocrinology and Metabolism
Introduction and Goal
Reference ranges are useful for making determinations about a patient’s health. In the case of testosterone, measuring hormone levels and evaluating them as “low” or “normal” can help clinicians diagnose hypogonadism and, in turn, provide the proper treatment. However, it is unclear whether reference ranges that apply to one population can be applied to another. Inter-assay, inter-laboratory, biological, and environmental factors should be considered.
This study aimed to generate consensus reference ranges for total testosterone using data from four epidemiologic studies of men in the United States and Europe. The studies included the following:
• The Framingham Heart Study (FHS)
• The European Male Aging Study (EMAS)
• The Osteoporotic Fractures in Men Study (MrOS)
• The Sibling Study of Osteoporosis (SIBLOS)
Combined, the four studies included 9,054 qualifying participants.
Originally, testosterone levels for each study were measured using different assays, which used different calibrators.
For this study, a fasting morning serum sample was taken from 100 randomly-chosen men from each cohort. Each sample was frozen and sent to the Centers for Disease Control and Prevention (CDC) Clinical Reference Laboratory in Atlanta, Georgia, USA. Testosterone concentrations were measured using liquid chromatography tandem mass spectrometry (LC-MS/MS), described by the authors as “a higher order (a reference method against which other methods are compared).”
Normalizing equations were then developed so that measurements could be “translated” from the original cohort values to the CDC standard, thus generating harmonized values. Next, the harmonized measurements were used to “derive standardized, age-specific reference ranges in each of the four cohorts and overall,” the authors wrote.
Three independent analyses of harmonized data were conducted:
|Generation of Reference Ranges in Healthy, Nonobese, Young Men||Data for 1,185 men aged 19-39 years with body mass indexes (BMIs) of <30 kg/m2 and free of major comorbidities were included.|
|Age-specific Reference Ranges in Nonobese Men||Data for individuals with BMIs of <30 kg/m2 were computed for the following age ranges: 19-39, 40-49, 50-59, 60-69, 70-79, and 80-99. This category included 6,933 men.|
|Age-specific Reference Ages in All Men||Using combined data from all cohorts, estimates were computed for all age ranges for all men regardless of obesity status.|
Results and Discussion
The harmonized range for healthy nonobese young men (aged 19 to 39 years) was 264 to 916 ng/dL. Values for percentiles were as follows:
|2.5th percentile||264 ng/dL|
|5th percentile||303 ng/dL|
|50th percentile||531 ng/dL|
|95th percentile||852 ng/dL|
|97.5th percentile||916 ng/dL|
The cross-calibration of assays using the CDC standard “provides substantial reduction in inter-cohort variation,” the authors wrote.
However, much of the data used to compute the harmonized ranges came from white men in the United States and Europe. Studies that include men from different races, ethnicities, and geographic locations should be developed to see if the results of this study can be applied to broader populations, the authors noted.
Future research might also consider the following questions:
• Should the reference range be based on a sample of healthy young men (T-score approach) or should it be age-adjusted (Z-score approach)?
• Should the reference sample include only healthy nonobese men or all men in a specific age range? Because obesity and comorbidities can affect testosterone levels, including men with these conditions could affect results.
These harmonized reference ranges should be validated using outcomes data from longitudinal studies and randomized trials. “Eventually the specificity, sensitivity and predictive value of these harmonized reference ranges should be evaluated in clinical populations of men seeking medical care,” the authors wrote.
“We conclude that standardized hormone measurements calibrated to a higher order benchmark, such as that offered by the Centers for Disease Control Clinical Reference Laboratory, provide a rational and feasible approach to generating harmonized reference ranges for testosterone and possibly other analytes,” they added.