Issue: December 2020
Disclosures: Auchus reports he serves on advisory boards for or received consultant fees or grants from Adrenas Therapeutics, Corcept Therapeutics, Janssen, Millendo Therapeutics, Neurocrine Biosciences, Novartis, Quest Diagnostics, Selenity Therapeutics, Spruce Biosciences and Strongbridge Biopharma, and has served as a research investigator for Corcept Therapeutics, Millendo Therapeutics, Neurocrine Biosciences, Novartis, Spruce Biosciences and Strongbridge Biopharma. Vesper and Wierman report no relevant financial disclosures.
December 18, 2020
6 min read
Save

Very low estrogen levels during menopause complicate assay use, measurement

Issue: December 2020
Disclosures: Auchus reports he serves on advisory boards for or received consultant fees or grants from Adrenas Therapeutics, Corcept Therapeutics, Janssen, Millendo Therapeutics, Neurocrine Biosciences, Novartis, Quest Diagnostics, Selenity Therapeutics, Spruce Biosciences and Strongbridge Biopharma, and has served as a research investigator for Corcept Therapeutics, Millendo Therapeutics, Neurocrine Biosciences, Novartis, Spruce Biosciences and Strongbridge Biopharma. Vesper and Wierman report no relevant financial disclosures.
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Clinicians using mass spectrometry to measure hormone levels should exercise caution when assessing data from postmenopausal women, when very low estradiol levels are difficult to measure accurately.

Accurate measurement of very low levels of circulating estradiol in postmenopausal women is important to investigate sex steroid action in target tissues. However, immunoassays are inaccurate, and conventional mass spectrometry-based measurement is often too insensitive at typical postmenopausal levels of estradiol, which can run below 5 pg/mL.

“For estradiol, assays were optimized for reproductive-age women to be able to measure estrogen across the menstrual cycle,” Margaret E. Wierman MD, professor in medicine, integrative physiology, director of the Pituitary Adrenal and Neuroendocrine Tumor Program at the University of Colorado Anschutz Medical Campus, and chief of endocrinology at Rocky Mountain Regional VAMC, told Endocrine Today. “The immunoassays are not as sensitive as our old radioimmunoassays, but they work fairly well. The problem is that they are not standardized to measure the estradiol levels in pubertal children or postmenopausal women.”

In clinical practice, the goal is often to look for “too much” of a hormone in question — too much testosterone for a women who may have hyperandrogenism, or too much estradiol for women undergoing induction of ovulation, Wierman said. That can complicate how an assay is calibrated, she said.

Margaret E. Wierman

“If the assay is designed to be very sensitive on the ‘low’ side, samples at higher levels must be diluted,” Wierman said. “Still, because of the rapid turnaround, most places continue to use immunoassay for measuring estradiol.”

Mass spectrometry, though, also brings caveats for clinicians when measuring steroid levels in the low, postmenopausal range, according to Richard J. Auchus, MD, PhD, professor of pharmacology and internal medicine in the division of metabolism, endocrinology and diabetes at the University of Michigan.

“When looking at low values like this, we must be cautious with the interpretation of the data,” Auchus told Endocrine Today. “Most postmenopausal women are down around the 15 pg/mL area, so it is going to be difficult to draw conclusions for what estradiol is doing down at those levels without using better methods.”

Pros, cons of mass spectrometry

Mass spectrometry — an analytical technique that measures the mass-to-charge ratio of ions — has several advantages when measuring serum samples, Auchus said. Mass spectrometry tends to eliminate immunoassay cross-reactivity that can invalidate the low levels measured during menopause, and multiple steroids can be simultaneously measured with a single sample, improving sensitivity.

The technology also brings disadvantages, Auchus said: It is expensive and technically demanding, and the instruments, on the best days, are fragile.

“Mass spectrometry is not a turnkey method,” Auchus said. “It is real people in the lab, running experiments on a day-to-day basis, getting funny results, and trying to repeat them again. Maybe we need to change the column; maybe we need to clean the instrument or replace a part or run some solvent through it over the weekend to clean it up. That is why you cannot necessarily get results one day. It is not the sensitivity that limits you. It is the noise that is the problem.”

Mass spectrometry machines tend to be delicate, which is a problem particularly when measuring sex steroid hormones, Wierman said.

Richard J. Auchus

“Sex steroids are sticky,” Wierman said. “They circulate in low abundance, apart from dehydroepiandrosterone, and the free hormones are very hard to measure by any assay. It comes down to the question of what you are looking for. Estradiol in an immunoassay is useful, but the estradiol is more useful when measured by mass spectrometry, especially for anything related to pubertal development or to menopause.”

Mass spectrometry used today is based on a quadrupole design — four metal rods with direct and alternating currents tuned to create an oscillating magnetic field, Auchus said. Ions that come in with the right mass-to-charge ratio will oscillate through the quadrupole and go on to the next chamber; all other ions will be expelled through the sides of the quadrupole.

“This is how we choose the ions we want to measure,” Auchus said.

In measuring samples from the Study of Women’s Health Across the Nation (SWAN) study, a longitudinal study of women across the menopause transition that utilized immunoassay, researchers took several steps to measure estradiol levels more accurately among women using mass spectrometry. Steps included increasing serum volume (extracting 0.2 mL), which was resuspended in a smaller volume and then injecting a large aliquot, allowing a sixfold increase in sensitivity over a pilot method, Auchus said. Researchers also used a longer column and slower flow rate, optimized the source for estradiol and used scheduled isolated time segments to improve signal-to-noise.

“Other groups have developed high-sensitivity estradiol methods using derivatizing reagents, but we run it as straight estradiol,” Auchus said.

In assessing approximately 400 SWAN samples, researchers observed correlation between the immunoassay results and the mass spectrometry results at high levels of estradiol; however, as estradiol levels declined, researchers noted a deviation in the curve, with the immunoassay breaking down at about 20 pg/mL.

“The correlation was excellent until you got down to the very low levels, and then we found that the mass spectrometry assay was reading down to roughly 3 pg/mL, but there was a gradual deterioration of the correlation,” Auchus said. “When you got down to about 15 pg/mL, it started to deteriorate, even though the assay was said to read down to 6 pg/mL.”

Average menopausal testosterone levels are similarly not accurate when measured using immunoassay, he said.

“In reproductive-aged women, estradiol immunoassay is fine, but in the menopausal women, it is not reliable,” Auchus said.

Need for standardization

Laboratory measurements are the sole source of information for diagnosis and treatment of many endocrine diseases, and hormone tests must be accurate and precise, according to Hubert W. Vesper, PhD, director of clinical standardization programs at the CDC’s National Center for Environmental Health. Measurements of testosterone and estradiol are two of eight laboratory measurements the CDC has worked to optimize for high accuracy and precision. Vesper said reference intervals for estradiol and other hormones are needed, especially for postmenopausal women.

“One of the main challenges with measuring low-level sex steroids is analytical specificity and sensitivity, for example, making sure that an assay measures only estradiol or testosterone and no other compound,” Vesper told Endocrine Today. “The CDC’s Clinical Standardization Programs for estradiol and testosterone are assessing the measurement accuracy and reliability of estradiol and testosterone measurements, among other biomarkers. Those laboratories and assay manufacturers successfully participating in these CDC programs are listed on the CDC website.”

If an assay manufacturer is listed on the CDC website, this means the assay is appropriately calibrated and one can see how well an assay determines these hormones in individual blood samples, Vesper said.

“The higher the individual sample pass rate, the less inaccurate measurements are on different individual blood samples,” Vesper said. “In other words, the individual sample pass rate provides some information about the specificity of an assay.”

The Endocrine Society has been actively involved with the CDC in the effort to standardize assays for more accurate hormone measurement, Wierman said.

“One mass spectrometry assay is not the same as another mass spectrometry assay,” Wierman said. “Mass spectrometry is not perfect. There still are some very sensitive immunoassays, for example, for estradiol, that actually are very good. There are situations where they are useful.”

Vesper noted that some laboratories and assay manufacturers perform extra steps to ensure and improve the quality of hormone assays by voluntarily participating in the CDC’s clinical standardization programs or by participating in accuracy-based surveys offered by the College of American Pathologists.

“It is important to support laboratories with these efforts, as standardized, high-quality measurements will improve patient care,” Vesper said.

Use a conservative number

Auchus said the discrepancies between assays raise the question, “What is the limit of quantification in your assay?’” The answer is not simple, he said.

“There is not a straight answer for that because it depends on the day and it depends on the samples,” Auchus said. “You would think that human serum, even though it is highly complex, would have the same major components for most people. There could be some operator error; there could be variability in sample preparations. Some just have more noise in the mass spectrometry range where we are trying to look than others.”

Rather than trying to “dig the values out of the noise,” Auchus said his laboratory does not report a level as measurable if there is too much noise relative to the signal. If someone wants to give a limit of detection, they use a conservative number that can be relied on most days, he said.

“This is something that is difficult for people to get their head around,” Auchus said. “They may think the machine operates the same way every day and with every sample. That is not the case. People need to be aware of this.”