A New Blood Test to Measure Breast Cancer Drugs and Estrogen Levels in Postmenopausal Women

Table of Contents

• Key Points
• Why This Research Matters
• What the Researchers Set Out to Do
• How the Study Was Conducted
• How the Test Was Validated
• Comparison With Existing Laboratory Methods
• Key Findings in Breast Cancer Patients
• Clinical Implications for Patients
• Limitations of the Study
• Recommendations for Patients
• Frequently Asked Questions
• Source Information

Key Points

• A new blood test measures all three aromatase inhibitors and ultra-low estrogen levels in one sample from postmenopausal breast cancer patients.
• The test is sensitive enough to detect estradiol and estrone below 1 pmol/L, which standard hospital tests cannot measure.
• In a study of 51 patients, the test revealed that 3 of 26 letrozole patients had drug levels outside the expected range.
• Two of four anastrozole patients had estradiol levels above 2 pmol/L, suggesting possible inadequate suppression.
• Exemestane degrades at room temperature, so blood samples should be refrigerated or frozen promptly for accurate measurement.

Why This Research Matters

Breast cancer is the most common cancer affecting women worldwide. In Europe, it is the leading cause of death among women in the 30-to-59 age group. Approximately 75% of breast cancers express the estrogen receptor and/or progesterone receptor, meaning the cancer cells rely on hormones to grow. These tumors are candidates for endocrine therapy—treatment that blocks or lowers hormones.

In postmenopausal women, the ovaries no longer produce significant amounts of estrogen. Instead, estrogens are produced through a process called peripheral aromatization, where circulating androgens (male hormones) are converted into estrogens by an enzyme called aromatase. This conversion happens in fat tissue, muscle, skin, and breast tissue.

Aromatase inhibitors (AIs) are drugs that block this enzyme. The third-generation AIs—letrozole (marketed as Femara), anastrozole (Arimidex), and exemestane (Aromasin)—have become the recommended first-line endocrine therapy for postmenopausal breast cancer patients with hormone receptor-positive disease. These medications work in different ways:

• Anastrozole and letrozole are nonsteroidal "type-II" inhibitors—they bind reversibly to the aromatase enzyme and are the parents compounds that do the work, since their main breakdown products (metabolites) are inactive.
• Exemestane is a steroidal "type-I" aromatase inactivator—it binds irreversibly to the substrate binding site of the aromatase enzyme, permanently disabling it.

While these drugs appear to have comparable treatment efficacy overall, letrozole seems to be somewhat more potent than anastrozole when it comes to suppressing estrogen. There is less data on estrogen levels in patients treated with exemestane, partly because exemestane's breakdown products interfere with traditional immunochemical estrogen tests, causing falsely elevated readings that require extensive pre-cleaning before analysis.

The main breakdown product of exemestane, 17-hydroxyexemestane (17HEXE), may be just as effective at blocking aromatase as exemestane itself. A recent evaluation suggested that letrozole may produce lower "overall estrogenic activity" than exemestane during treatment, and this is consistent with earlier findings from this research group.

The degree of estrogen suppression directly affects treatment outcomes. A clinical case-control study by Ingle and colleagues showed that postmenopausal women on AI therapy who did not achieve sufficiently suppressed levels of estradiol and estrone had an increased risk of an early breast cancer recurrence. However, standard hospital laboratory tests are not sensitive enough to measure estrogen levels this low—often below 1 pmol/L—during AI treatment. This makes it difficult for doctors to know whether the medication is truly working in an individual patient.

There's also an issue of treatment compliance. Some patients, particularly premenopausal women receiving a combination of AI drugs with luteinizing hormone-releasing hormone (LHRH) analogues, can experience "ovarian suppression escape"—meaning the ovaries start producing estrogen again despite treatment. Monitoring both drug levels and estrogen levels could catch these problems early.

What the Researchers Set Out to Do

Currently, no single laboratory test can simultaneously measure the blood levels of all three third-generation aromatase inhibitors (letrozole, anastrozole, and exemestane) along with the extremely low estrogen levels expected during AI treatment. Up to now, testing required separate methods for drug levels and estrogen levels.

The goal of this study was to develop and validate a new laboratory method using liquid chromatography-tandem mass spectrometry (LC-MS/MS)—a highly precise analytical technique that identifies and measures molecules by their mass and electrical charge. The new test was designed to:

1. Measure all three AIs plus the main exemestane metabolite (17HEXE) in one run
2. Simultaneously measure estradiol (E2) and estrone (E1) at subpicomolar levels (below 1 pmol/L)
3. Be robust and practical enough for use in a routine clinical laboratory
4. Provide proof of concept by testing blood from real patients on AI treatment

How the Study Was Conducted

Patient Samples

The researchers used serum samples (the liquid part of blood after clotting factors are removed) from postmenopausal patients with estrogen receptor-positive breast cancer who were receiving standard-dose AI treatment. These samples came from the Prospective Breast Cancer Biobank, a population-based research biobank at Haukeland University Hospital and Stavanger University Hospital in Norway. All participants gave written informed consent, and the study was approved by the Norwegian Regional Ethical Committee (approval numbers 2010/1957, 2011/2161, 172359, and 255133).

Additionally, the team reanalyzed 16 serum samples from postmenopausal breast cancer patients (8 on letrozole, 8 on exemestane) that had been used in earlier published studies. These samples were used for comparing the new method against existing accredited methods.

Sample Preparation

Preparing blood samples for analysis is a multi-step process. The researchers automated the entire process using a Hamilton Microlab STAR Liquid Handling System, a robotic pipetting instrument. Key steps included:

1. Pipetting 10 microliters (µL) of internal standard (a known amount of a labeled reference compound) into a deep-well plate
2. Adding 500 µL of calibrators, serum samples, or quality controls and mixing
3. Incubating for 1 hour at room temperature
4. Adding 1000 µL of extraction solvent (hexane and methyl tert-butyl ether in a 75:25 ratio) and mixing
5. Adding 75 µL of separation solvent (hexane and 2-propanol, 75:25) to create clean separation of the two liquid layers
6. Centrifuging for 2 minutes at 4000g
7. Transferring 700 µL of the organic (top) phase to a new plate with glass vials
8. Evaporating the solvent under nitrogen gas at 40°C
9. Reconstituting the dried sample in 70 µL of water:methanol (75:25)
10. Storing the plate at 5°C overnight before analysis

That overnight storage step is an important trick: the researchers had previously observed that it increases the signal for estradiol and estrone and lowers background noise. This improved sensitivity by approximately 20% to 25% for both estrogens.

The Mass Spectrometry Setup

For separation, the team used a Shimadzu Nexera UPLC system with two chromatography columns connected in series—a phenyl column and a C8 column—heated to 60°C. The mobile phases (the liquid that carries the sample through the columns) were water with 0.08% ammonium hydroxide and methanol with 0.02% ammonium hydroxide, run at a flow rate of 0.250 mL/min with a gradient from 40% to 82% over 5.95 minutes.

The mass spectrometer was a QTRAP 6500+ (SCIEX), operated in positive electrospray ionization (ESI) mode for anastrozole, exemestane, and 17HEXE, and in negative ESI mode for letrozole, estradiol, and estrone. The temperature was set at 500°C.

One technical challenge was that anastrozole and letrozole have very similar retention times (they come off the column at nearly the same time). The researchers solved this by using polarity switching—rapidly switching the mass spectrometer between positive and negative ionization modes, with a settling time of 30 milliseconds. This allowed both drugs to be measured accurately in the same run.

How the Test Was Validated

Before a laboratory test can be used on patients, it must be thoroughly validated to prove that the results are reliable and reproducible. The research team performed a comprehensive set of validation experiments.

Measurement Range and Sensitivity

The measurement range of a test is defined from the lower limit of quantification (LLOQ)—the lowest concentration that can be measured with acceptable precision—to the upper limit of quantification (ULOQ). The limit of detection (LOD) is the lowest concentration that can be distinguished from a blank sample. Here are the key numbers:

• Estrone (E1): LOD 0.17 pmol/L, LLOQ 0.2 pmol/L, ULOQ 2400 pmol/L (expandable to 12,000 pmol/L)
• Estradiol (E2): LOD 0.56 pmol/L, LLOQ 0.8 pmol/L, ULOQ 2594 pmol/L (expandable to 13,000 pmol/L)
• 17HEXE (exemestane metabolite): LOD 2.3 pmol/L, LLOQ 8.0 pmol/L, ULOQ 25,077 pmol/L (expandable to 125 nmol/L)
• Exemestane (EXE): LOD 5.8 pmol/L, LLOQ 13 pmol/L, ULOQ 40,520 pmol/L (expandable to 203 nmol/L)
• Letrozole (LET): LOD 9.8 pmol/L, LLOQ 14 pmol/L, ULOQ 140,000 pmol/L (expandable to 701 nmol/L)
• Anastrozole (ANA): LOD 32 pmol/L, LLOQ 95 pmol/L, ULOQ 300,000 pmol/L (expandable to 1500 nmol/L)

The estrogen measurement ranges are particularly notable because they cover the extremely low estrogen levels expected during AI therapy while also reaching up into the normal postmenopausal range. All calibration curves had an R² value of 0.997 or higher (R² measures how well the calibration curve fits the data, where 1.0 is a perfect fit).

Accuracy and Precision

Accuracy refers to how close a measured value is to the true value. Precision refers to how reproducible the measurements are when the same sample is tested multiple times.

Mean accuracy against known nominal concentrations was 100 ± 5% for all calibration points. Accuracy was confirmed against certified reference materials (CRMs)—highly purified standards with known concentrations—for anastrozole, letrozole, estradiol, estrone, and exemestane. Each level was analyzed in triplicate across 3 runs, for a total of 9 analyses per level. The coefficient of variation (CV) of the slope of the regression line was below 5% for all analytes. Overall accuracies were within 100 ± 8% at all levels.

Precision was evaluated using quality controls at 4 concentration levels, analyzed in quadruplicate over 6 days (one run per day, n = 24 per level). All analytes had precision within ± 13%, with results varying by analyte and concentration:

• E1: CV 5-6% across quality control levels
• E2: CV 2-13% (the 13% was at the lowest level, 1.3 pmol/L, which is expected for such low concentrations)
• 17HEXE: CV 4-8%
• EXE: CV 2-4%
• LET: CV 2-3%
• ANA: CV 4-8%

Selectivity and Interference Testing

Selectivity means the test only measures what it is supposed to measure, without interference from other substances in the blood. The researchers tested pooled serum samples spiked with high concentrations (100-2000 nmol/L) of potentially interfering compounds.

Results showed that at a lipid (fat) index of 1735 mg/dL—extremely fatty blood—the measured value of 17HEXE increased by 14%, but none of the other analytes were affected. High concentrations (2000 nmol/L) of 17-α-estradiol and ethinylestradiol (a synthetic estrogen used in birth control) interfered with low-concentration measurements of E2 and E1, but these are not expected to be present in postmenopausal women on AI therapy. No carryover (residual contamination from a previous sample) was detected for any analyte, and inspection of chromatograms from more than 400 patient samples showed no interfering peaks.

A particular concern was that anastrozole and letrozole co-elute (come off the column at the same time). However, even at concentrations high enough to saturate the detector, no spectral interference was observed between the two drugs. This was confirmed by overlaying the chromatograms of both compounds measured in different ionization modes.

Matrix Effects

Matrix effects occur when other components in the blood affect the measurement. The researchers used an "add-in recovery" method: 10 unique patient serum pools from postmenopausal women were analyzed before and after spiking with known amounts of analytes. Recovery was calculated as the post-spiked value times 100 divided by the expected value (endogenous plus added concentration). Results showed negligible matrix effects:

• ANA: mean add-in recovery 102 ± 2%
• LET: 101 ± 2%
• E2: 95 ± 4%
• E1: 102 ± 2%
• EXE: 101 ± 2%
• 17HEXE: 96 ± 11%

These numbers indicate that the test performs consistently regardless of the individual patient's blood composition.

Stability Testing

Stability testing determines how long samples can be stored before the drug or hormone levels degrade. The researchers tested samples under three conditions, using 5 patient samples in each experiment:

• Refrigerated at 5°C for 7 days: All analytes were stable, with mean deviation of ± 5%
• Room temperature for 7 days: Exemestane showed degradation—dropping to -11% deviation by day 3 and -14% by day 7. All other analytes were stable with mean deviation between +3% and -3%.
• Freeze/thaw cycles (3 cycles): All analytes were stable, with mean deviation of -1% to 6%

The key practical takeaway: blood samples should be refrigerated or frozen promptly, especially if exemestane is being measured.

Comparison With Existing Laboratory Methods

The new method's estrogen measurements were compared against two accredited (ISO 15189:2012) laboratory methods to confirm its accuracy. These accredited methods are the current "gold standard" used in clinical practice:

• Against the accredited in-house LC-MS/MS method for estradiol (measuring range 13-2722 pmol/L): the mean difference was only 2%, with an R² of 0.996 (n = 40)
• Against the ultra-sensitive accredited LC-MS/MS method for both estradiol and estrone (measuring range 0.6-224 pmol/L for E2 and 0.3-234 pmol/L for E1): the mean deviation was less than 6% for both compounds, with R² greater than 0.997

There was no alternative method available to cross-check the AI drug level measurements, since no other clinical laboratory test measures all three AIs simultaneously.

Key Findings in Breast Cancer Patients

To prove the method works in real clinical situations, the researchers analyzed serum samples from 51 postmenopausal breast cancer patients receiving standard-dose AI treatment. The results revealed important patterns—and some concerning findings.

Patients on Letrozole (26 patients)

Letrozole suppressed estradiol and estrone to subpicomolar levels (below 1 pmol/L) in all but one patient. This confirms letrozole's potent estrogen-suppressing effect.

Drug level monitoring showed that 23 of the 26 patients had letrozole concentrations in the range of 180 to 380 nmol/L. This leaves 3 patients whose levels fell outside this range—potentially indicating poor compliance, absorption problems, or other individual differences in drug metabolism.

Patients on Exemestane (8 patients in the reanalysis cohort, plus others)

In the exemestane-treated group, only one patient had detectable estradiol. Estrone levels were also very low, although all but one patient had quantifiable estrone peaks.

Measured exemestane concentrations ranged from 0.6 to 90.2 nmol/L, and the metabolite 17HEXE ranged from 0.3 to 8.4 nmol/L. Notably, exemestane and 17HEXE levels correlated strongly with each other (correlation coefficient R = 0.88), with an average exemestane-to-17HEXE ratio of 6:1. This means the parent drug is present at about 6 times the level of its active metabolite.

Interestingly, there was no correlation between exemestane levels and estrone levels—meaning higher drug levels did not necessarily translate into lower estrogen levels in individual patients. This suggests that factors beyond drug concentration, such as the patient's own aromatase activity, may influence treatment response.

Patients on Anastrozole (4 patients)

All 4 anastrozole-treated patients had quantifiable estrone peaks, and 2 of the 4 had relatively high estradiol levels of 2.3 and 4.5 pmol/L. These are levels that might be considered inadequate suppression for optimal treatment. The measured anastrozole concentrations ranged from 6.5 to 112 nmol/L.

Clinical Implications for Patients

This new test could change how breast cancer treatment is monitored. Here's what it means for patients:

It can confirm the drug is actually working. The goal of AI therapy is to suppress estrogens to subpicomolar levels. The new test can accurately measure whether this goal is being achieved—something that standard hospital estrogen tests often cannot do because they lack sensitivity.

It can monitor treatment compliance. For the first time, doctors can check whether the correct amount of AI drug is present in a patient's blood. In this study, 3 of 26 letrozole patients had drug levels outside the expected 180-380 nmol/L range, which could indicate missed doses, poor absorption, or drug interactions worth investigating.

It may identify patients needing different treatment. The 2 anastrozole patients with estradiol levels above 2 pmol/L despite treatment might benefit from a different AI or a different dosing strategy. The link between inadequate estrogen suppression and increased risk of early breast cancer recurrence (shown in the Ingle study) makes this information potentially life-saving.

It can detect ovarian suppression escape. For premenopausal women on combined LHRH analogue plus AI therapy, a rise in estradiol levels can signal that the ovaries are escaping suppression. This test could detect that escape earlier than current methods.

It solves the exemestane measurement problem. Exemestane's metabolites have historically interfered with estrogen immunoassays, causing falsely elevated readings. The mass spectrometry approach eliminates this interference, giving doctors reliable estrogen data for exemestane-treated patients for the first time.

It offers practical advantages. Because drug and estrogen levels can be measured in one test from one blood sample, it saves patients time and reduces laboratory costs.

Limitations of the Study

While this is a well-designed validation study, it has several limitations that should be acknowledged:

• Small clinical sample size: The proof-of-concept testing included only 51 patients, including just 4 on anastrozole. The findings about anastrozole patients with relatively high estradiol levels need confirmation in larger studies.
• No long-term outcome data: This study validated the test method but did not follow patients over time to see whether drug levels or estrogen levels predicted disease recurrence. That would require a separate, longer prospective study.
• No alternative method for AI drug verification: The drug level measurements could not be compared against an independent accredited method because no such method exists yet. The validation therefore relied on spiking experiments and precision/accuracy assessments.
• Limited premenopausal population: The test was developed and validated for postmenopausal patients. Its utility in premenopausal patients on combined LHRH analogue plus AI therapy, where ovarian escape monitoring is important, has not yet been directly demonstrated.
• Single overnight storage protocol: The method relies on an overnight storage step at 5°C to boost estrogen signal. This may slow turnaround time for urgent clinical results.

Recommendations for Patients

Based on this research, here are some practical points for breast cancer patients and their families:

1. Talk to your oncologist about drug monitoring. If you are taking an aromatase inhibitor, ask whether blood tests to measure drug levels and estrogen suppression might be valuable in your case—especially if you have concerns about side effects, missed doses, or treatment effectiveness.
2. Take your medication consistently. This study showed that some patients have drug levels outside the expected range. If your AI levels were ever measured and found low, consistent daily intake at the same time may help.
3. Don't skip blood tests. Even though estrogen tests during AI treatment may show very low numbers, these measurements are clinically important. The degree of suppression correlates with outcomes.
4. Handle sample collection properly. This study found that exemestane degrades at room temperature (losing about 11% by day 3 and 14% by day 7). If you're having blood drawn for drug level testing, make sure the sample is promptly refrigerated or processed.
5. Keep taking your AI even if you feel fine. The fact that estrogens are suppressed to undetectable levels by your treatment is exactly what doctors want to see. If you have side effects, discuss them with your medical team rather than stopping the medication.
6. Ask about exemestane specifically. If you are on exemestane, know that older estrogen tests may have given inaccurate results due to interference. The new mass spectrometry method used in this study eliminates that problem, so ask whether your hospital uses LC-MS/MS for estrogen measurement.

This research represents an important step toward personalized medicine in breast cancer care. The ability to simultaneously measure both the medication and its intended biological effect (estrogen suppression) in a single blood sample gives doctors a powerful tool to optimize treatment for each individual patient. While more research is needed to establish clinical guidelines for acting on these measurements, the method itself is now available and validated for routine laboratory use.

Frequently Asked Questions

What is this new blood test for?

This new blood test measures all three third-generation aromatase inhibitor drugs—letrozole, anastrozole, and exemestane—and ultra-low estrogen levels in a single blood sample. It was developed for postmenopausal breast cancer patients with hormone receptor-positive disease who are taking aromatase inhibitors, to help doctors check if treatment is working effectively.

Why do doctors need to measure such low estrogen levels?

Aromatase inhibitor treatment aims to suppress estrogen to very low levels, often below 1 pmol/L. Standard hospital tests cannot measure these tiny amounts. Previous research linked inadequate estrogen suppression to a higher risk of early breast cancer recurrence, so knowing whether levels are truly low helps doctors personalize treatment.

How is this test different from standard estrogen tests?

Standard hospital estrogen tests are not sensitive enough to measure the extremely low estrogen levels expected during aromatase inhibitor therapy. The new method uses liquid chromatography-tandem mass spectrometry, allowing it to measure estradiol and estrone below 1 pmol/L, while also measuring drug levels in the same blood sample.

Will this test be available for my routine care?

The researchers validated this test for routine laboratory use, but it is not yet widely available. If you are on an aromatase inhibitor, ask your oncologist whether drug level and estrogen monitoring might be helpful in your case, especially if you have concerns about side effects or effectiveness.

Does this test work for premenopausal women?

No, the test was developed and validated specifically for postmenopausal breast cancer patients. Its usefulness in premenopausal women who are receiving combined LHRH analogue plus aromatase inhibitor therapy has not yet been directly demonstrated, although ovarian escape monitoring could be a future application.

Source Information

Original Article: "Simultaneous Quantification of Aromatase Inhibitors and Estrogens in Postmenopausal Breast Cancer Patients"

Authors: Bjørn-Erik Bertelsen, Kristin Viste, Thomas Helland, Magnus Hagland, Håvard Søiland, Jürgen Geisler, Tone Hoel Lende, Per Eystein Lønning, Jørn V. Sagen, Gunnar Mellgren, and Bjørg Almås

Journal: The Journal of Clinical Endocrinology & Metabolism, 2022, Volume 107, pages 1368-1374

DOI: 10.1210/clinem/dgab923

Publication Date: Advance access publication 27 December 2021

Funding/Institutional Affiliations: Haukeland University Hospital, University of Bergen, Stavanger University Hospital, Akershus University Hospital, and University of Oslo, Norway.

This patient-friendly article is based on peer-reviewed research published in a leading endocrinology journal. It is intended for educational purposes and does not replace individualized medical advice from your healthcare team.