Sex Hormone Measurement: How Evolving Technology Has Improved Test Accuracy, Precision and Reproducibility

Swerdloff, Ronald S., MD
Professor of Medicine
David Geffen School of Medicine at UCLA
Chief Division of Endocrinology
Senior Investigator
Los Angeles Biomedical Research Institute
Harbor-UCLA Medical Center
Los Angeles, CA
Also by this Author 

For endocrinologists, laboratory testing is critical for patient diagnosis and monitoring, and assays for sex hormone measurement are at the core of endocrine practice. While the limitations of traditional direct immunoassays have been widely documented, test methodologies based on mass spectrometry (MS) now offer clinicians access to improved accuracy, precision, specificity and sensitivity.1-4

Dr. Ronald Swerdloff, Professor of Medicine, David Geffen School of Medicine at UCLA and Chief Division of Endocrinology, Harbor-UCLA Medical Center and Dr. Nigel Clarke, Senior Science Director for Mass Spectrometry and Automation, Quest Diagnostics Nichols Institute, review the evolution of sex hormone testing and how recent technological developments are enabling clinicians to diagnose and manage patients with a higher level of confidence than with traditional testing methodologies.

The Evolution of Testing Methodologies

“During my career there has been a rapid evolution in assay methodology,” notes Dr. Swerdloff, “from the use of bio-assays as the only available tool to the development of antibody-based assays, such as the radio immunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA), and their development into automated assay systems, to address the need for rapid measurement and lower cost. Unfortunately, in some instances this improvement in efficiency and cost of direct (platform) assays had a significant downside - a reduction in accuracy, precision and reproducibility.”

The limitations of direct immunoassays can have significant implications for clinicians in certain critical situations.  Reduced accuracy is a particular problem when the blood levels of the hormone in question are in a low range. “It has been shown in multiple studies that some of the earlier methodology broke down in its ability to measure hormones at low concentrations, such as in the measurement of estrogens in post-menopausal, hypogonadal or agonadal women,” says Dr. Swerdloff.1-3  “It was also shown to be a problem in the measurement of sex steroids in pre-pubertal children, severely hypogonadal men, castrated men or men who were being treated with drugs to lower their androgen levels because of prostate cancer.”

The Challenge of Cross-Reactivity

“The challenge in the circumstances described above,” says Dr. Clarke, “is that there are a large number of very homologous compounds circulating. You’re not doing any separation, yet you’re expecting one antibody to try and select one steroid without cross-reacting. In fact, using direct immunoassays to measure testosterone levels in these groups has been likened to random number generation.”

“The situation is even more challenging with estradiol,” continues Dr. Clarke, “because it’s present in such low concentrations in males, pediatrics and postmenopausal women, yet you’re trying to measure it in a sea of much higher concentration steroids, so there’s great potential for cross-reactivity. This is the case, for example, with breast cancer patients, such as post-menopausal women taking aromatase inhibitors to reduce the conversion of testosterone produced by the adrenals to estradiol. It’s critical to get an accurate measurement but it’s been shown that immunoassays are unreliable for measuring estradiol at these very low levels.5, 6 The worst case is if you get a false negative and believe everything is fine, when it’s not. But a false positive is also serious as it leads to stress for the patient and additional scans, both of which could have been avoided. A direct immunoassay is simply too unreliable to be used in such cases.”


Precision of assays is equally important, as the clinician needs to know that an assay’s result is reproducible. If an assay has a high degree of variance the results will vary from time to time even if measured at the same laboratory. It has been noted in a recent article in the Journal of Clinical Endocrinology & Metabolism that ‘the method-specific bias of immunoassays, a function of antibody epitope, allows for cross-reactivity of all steroid assays with steroid precursors, metabolites or conjugants, as well as the matrix effects of unextracted direct assays.' 4 “What this means,” says Dr. Swerdloff, “is that the specificity of an immunoassay is dependent upon the antibody that was chosen for the assay. These antibodies vary from laboratory to laboratory and so the level of specificity will also vary depending on the antibody used.”

Mass Spectrometry

Recognizing the shortcomings of direct assays, The Endocrine Society, in a position statement, has recommended that direct immunoassays should be avoided when testing samples with low testosterone concentration.3 It recommends the initial use of extraction and chromatography, followed by either MS or extraction RIA. These methods “are likely to furnish more reliable results and are currently preferred.”3 Dr. Clarke points out that extraction RIA requires a significant amount of manual intervention, is labor intensive and requires radioactivity. “The advantage of MS is that it is very precise, very accurate, and is the same technique used as the reference method at all the standard centers like the National Institute of Standards and Technology (NIST),” he notes. “Not only has the technology evolved over the last twenty years but its costs have come down so that it is being used for clinical diagnostics and not just for research.”

“Liquid chromatography tandem mass spectrometry (LC/MS/MS) has evolved to become the gold standard for the measurement of many hormones,” adds Dr. Swerdloff. “In many ways this methodology meets the criteria of an ideal assay - to be accurate, precise, reproducible, sensitive and highly specific. In some instances the sensitivity of the assay may be less important, and in other instances a high specificity may not be required, but these principles hold in most instances. Furthermore, while a good immunoassay in specific situations can be as good as a MS assay, the evolution of MS clearly is the approach of the present and will increasingly be the approach of the future. While immunoassays have played a valuable role in the past, and are still useful in certain situations, MS is an improved methodology which offers the clinician a better tool across the board.”

A Better Tool for Clinicians

“Evolving technology offers the clinician greater confidence that the results they receive reflect the clinical state,” notes Dr. Swerdloff. “Increased sensitivity in MS allows the measurement of hormones in smaller volumes in biologic fluids, and allows the accurate, precise measurement of hormones at very low levels, which provides a new perspective when dealing with many diseases. The specificity of the assays is really the heart of tandem mass spectrometry, because interfering substances no longer create confusion in the results that are provided. This is particularly so with low concentrations of hormone levels in the blood.”

Test specificity becomes extremely important in separating out and clearly defining such conditions as congenital adrenal hyperplasia, tumor produced hormones, and in conditions where the concentration in the blood of the hormone that you’re trying to measure is particularly low and therefore low concentrations of cross-reacting substances in immunoassays can cause confusion for the clinician.

Another advantage of MS for clinicians, indicates Dr. Swerdloff, is that it enables the measurement of multiple ligands simultaneously. “This will prove to be a huge benefit,” he says, “because you can now measure a spectrum of hormones, which provides new insights into many diseases. An example of this is in patients with congenital adrenal hyperplasia, where multiple hormones in a pathway may be modified, and measuring a complex of hormones gives immediate insight into where the enzyme defect may be.”

Understanding Tandem Mass Spectrometry

A key difference between the LC/MS/MS technology and the traditional immunoassay method, explains Dr. Clarke, is that LC/MS/MS measures the actual physical compound – the molecule – as opposed to a surrogate. “With an antibody test you may be measuring the release of a radioactive tracer or, in a chemiluminescenceassay, looking at a flash of light,” he says. “The problem is you don’t actually know what bound to the antibody to cause that flash of light. You hope it’s testosterone but since these are small non-antigenic molecules that are all very similar, and since antibodies are not very good at distinguishing small molecules, they struggle to tell the difference between say dihydrotestosterone (DHT) and testosterone. So the questions are: Was it really testosterone? Was every flash of light testosterone or was some of it cross-reactivity? Immunoassays measure the surrogate, while MS measures the absolute and that’s where its specificity comes from.”

There are two main steps in the LC/MS process, continues Dr. Clarke. “First, liquid chromatography reduces the complexity of the sample to eliminate all the proteins and peptides and introduce the hormones into the mass spectrometer under very controlled conditions. We can actually separate the hormone from all the other compounds in the sample, something which doesn’t occur in the immunoassay.”

This is followed by MS, which Dr. Clarke describes as “an exquisite set of scales”. “This allows you to measure molecules and actually get back structural information from them,” he continues. “You don’t just weigh the hormone and assume it’s testosterone or estradiol because it has the right mass. You also cause that molecule to break into pieces and obtain information to confirm its structure. By knowing both its mass and structure we can guarantee that what we’re measuring is the specific hormone in question.”

Toward Test Harmonization

One major problem for the clinician is trying to interpret assay reports from different laboratories. If the results are not accurate and precise the results on a given sample may be considerably different from different laboratories. “Interpreting this, both in diagnostic and therapeutic circumstances becomes very difficult,” notes Dr. Swerdloff, “but the issue is now being addressed by experts in the field. There is an effort by leaders in endocrinology to harmonize assays so the results for a given hormone will read the same from all qualified laboratory sites. The Centers for Disease Control and Prevention (CDC) has undertaken the initial efforts to harmonize endocrine assays and has selected the measurement of testosterone and estradiol as two of its target hormone assays.”

“Laboratories have been encouraged to participate in quality control methods through the CDC to determine how close their results are to a CDC standard assay and recommendations may be made about how to calibrate their assays to give reproducible results that are specific, accurate and sensitive. This effort by leaders in endocrinology, the CDC, and many reference laboratories will allow the clinician to compare results from different laboratories and better diagnose and treat patients with endocrine disorders. MS was chosen as the methodology at the CDC because it gives the highest degree of validity to the assays.”

As test harmonization increases and the technology continues to evolve, it is expected that MS will play an increasingly important role in endocrinology.


  1. Rosner W, Auchus R, Azziz R, Sluss P, Raff H. Utility, Limitations, and Pitfalls in Measuring Testosterone: An Endocrine Society Position Statement J Clin Endocrinol Metab 2007. 92(2):405–413.
  2. Rosner W and Vesper H, on behalf of The Endocrine Society and the endorsing organizations. Toward Excellence in Testosterone Testing: A Consensus Statement. J Clin Endocrinol Metab, October 2010, 95(10):4542–4548
  3. Rosner W, Hankinson SE, Sluss PM, Vesper HW, Wierman ME. Challenges to the measurement of estradiol: an Endocrine Society position statement. J Clin Endocrinol Metab. 2013;98:1376–1387.
  4. Handelsman D and Wartofsky L Requirement for Mass Spectrometry Sex Steroid Assays. J Clin Endocrinol Metab. 2013;98 (10)
  5. Jaque J, Macdonald H, Brueggmann D, Patel SK, Azen C, Clarke N and Stanczyk F. Deficiencies in immunoassay methods used to monitor serum Estradiol levels during aromatase inhibitor treatment in postmenopausal breast cancer patients. SpringerPlus 2013, 2:5  doi:10.1186/2193-1801-2-5
  6. Stanczyk F and Clarke N. Measurement of Estradiol—Challenges Ahead. J Clin Endocrinol Metab. doi: 10.1210/jc.2013-2905