Testosterone Testing Increasing Dramatically; Reference Range, Technological Standardization Remain Challenges

Testosterone testing has increased dramatically in the United States over the last decade, but this increase has largely been among men with normal levels, according to a study published in the March issue of the Journal of Clinical Endocrinology & Metabolism (JCEM). This surge may be due to increased marketing efforts by supplement makers and wider recognition of naturally declining testosterone levels in older men, particularly in those who are obese or with chronic disease. However, these tests may be of limited clinical value, and there are calls for technological improvements on the part of laboratories to improve the reliability of biochemical diagnosis of hypogonadism. While testosterone treatment is recognized for men with diagnosed hypogonadism, controversy exists over the necessity and safety of treating men who may have decreasing testosterone levels without meeting diagnostic criteria for hypogonadism. Diagnosis is confounded by a lack of agreement over the range defining testosterone deficiency, particularly at the lower normal range (200 to 350 ng/dL). Testosterone assay reference ranges have frequently been determined in populations of healthy, younger men, which may not be applicable to older men experiencing natural age-related declines. Furthermore, interpretation of results is complicated by assay variation between testing facilities. In the JCEM study the researchers evaluated commercial and Medicare insurance claims from the United States (MarketScan Commercial Claims and Encounters and Medicare Supplementary and Coordination of Benefit databases for 410,019 men) and general practitioner health care records from the United Kingdom (Clinical Practice Research Datalink registry for 6,858 men) from 2000 through 2011. For a subset of patients for whom outpatient laboratory assays were processed by a large national lab, assay results were available (2007 to 2011) and were classified using the following reference ranges: low, <300 mg/dL (10.4 nmol/L); normal, 300 to 849 ng/dL (10.4 to 25.4 nmol/L); and high, ≥850 ng/dL (29.5 nmol/L). The researchers found that testing has increased dramatically over the last decade in both the United States and the United Kingdom (from 39.6 per 10,000 person-years in 2000 to 170 per 10,000 person-years in 2010 in the United States and from three per 10,000 person-years to 46.4 per 10,000 person-years in the United Kingdom over the same period). However, testing in the United States is often performed in men at normal levels, with the proportion of normal results increasing from 2007 to 2011 from 64.5 percent to 73.2 percent. By contrast, in the United Kingdom testing increasingly identified men with low levels, as seen by the increase from 18.9 percent in 2000 to 26.7 percent in 2011. Additionally, the researchers found that many men initiate testosterone treatment without recent testing. In the United Kingdom, 53.8 percent of treatment initiators did not have a total testosterone measurement in the 180 days before initiation (32.7 percent had one test), while in the United States, 40.2 percent of initiators did not have a baseline test and 50 percent had one test. “Heavy direct-to-consumer marketing of newer testosterone formulations in the United States may have led to a much wider interest in testosterone levels and hypogonadism symptoms, resulting in wider testing of men with nonspecific symptoms but normal levels rather than targeted testing of symptomatic individuals,” write the authors, led by J. Bradley Laton, Ph.D., from University of North Carolina at Chapel Hill. This proliferation of testing, particularly in men with a borderline medical indication for treatment, is complicated by technical difficulties that continue to plague testosterone measurement, according to a review published online in Urologyon Feb. 17. The difficulties include the failure to establish clinically relevant normal assay ranges and harmonization of assay performance across different platforms. Serum testosterone levels vary widely and are subject to temporal variation, as well as impact from certain illnesses, medications, and risk factors, including nutrition, alcohol consumption, and smoking. Assays vary in their performance characteristics and limits of detection, as well as normal ranges based on variation in populations utilized to derive these ranges. “There is no large population-based study of testosterone values from healthy, fertile men with normal sexual activity and reproductive function assessed by commonly accepted validation methods,” write the review authors, led by Darius Paduch, M.D., Ph.D., from Weill Cornell Medical College (New York City). “The lack of these types of studies confuses clinical decision making and impairs comparison of assays on the same subject obtained in different laboratories.” A collaborative between the American Urological Association, the Endocrine Society, and the U.S. Centers for Disease Control and Prevention is under way to optimize assay platforms and to create evidence-based normal assay ranges to guide clinical decisionmaking. “Until such standardization is commonplace in clinical laboratories, the decision to treat should be based on the presence of signs and symptoms in addition to serum testosterone measurements. Rigid interpretation of T ranges should not dictate clinical decision making or define coverage of treatment by third party payers,” the authors write. “We encourage urologists and andrologists to discuss methodology and source of reference values with laboratory directors at their institutions to better understand limitations and advantages of local assays and to improve patient care.” Takeaway: While testosterone testing has increased dramatically over the past decade, these measurements, particularly in borderline low cases, may be of questionable clinical utility until improvements in the standardization of testosterone testing are implemented.