Taking Exemestane Less Often: Could a 3-Times-Weekly Dose Be as Effective as Daily for Breast Cancer?

Table of Contents

• Key Points
• Understanding the Problem: Why This Research Matters
• How the Study Was Conducted
• Who Took Part in the Trial
• Main Result: How Well Did Each Schedule Lower Estrogen?
• Tissue Biomarkers, Other Hormones, and Cholesterol
• Side Effects and Tolerability
• What These Findings Mean for Patients
• Study Limitations
• Recommendations for Patients
• Frequently Asked Questions
• Source Information

Key Points

• A 2023 study of 180 postmenopausal women found 25 mg exemestane three times weekly suppressed estrogen as well as daily dosing, but only among women who took at least 80% of scheduled pills.
• Once-weekly exemestane was less effective, reducing serum estradiol by 60% versus 89% with daily dosing in the intention-to-treat analysis.
• No significant differences in side effects were seen among daily, three-times-weekly, and once-weekly exemestane schedules during the 4–6 week presurgical treatment period.
• Do not change your exemestane dose on your own; the three-times-weekly schedule is not yet approved and should be discussed with your oncologist.
• Adherence is critical: the finding of noninferiority for three-times-weekly dosing applied only to compliant participants who took their medication as directed.

Understanding the Problem: Why This Research Matters

Breast cancer incidence is increasing, and it remains the leading cause of cancer-related burden worldwide, even though mortality rates are declining. Modern prevention strategies are risk-based—they include personalized screening, lifestyle changes, and medications that block or reduce estrogen's effects on breast tissue. Two major classes of preventive drugs are selective estrogen receptor modulators (SERMs), such as tamoxifen, and aromatase inhibitors (AIs), which include exemestane and anastrozole.

Aromatase inhibitors work by blocking the aromatase enzyme, which is responsible for producing estrogen in the body. Exemestane is a steroidal aromatase inhibitor, meaning it binds irreversibly to the aromatase enzyme, causing permanent inactivation even after the drug has been cleared from the bloodstream. In a phase 1 study of postmenopausal volunteers, a single 5 mg dose was already effective, and 25 mg was identified as the minimal dose that provides maximal estrogen suppression. This effect was reached by day 3 and persisted for up to 7 days—a key finding that raised the possibility of less frequent dosing.

In the prevention setting, exemestane has shown impressive results: a 65% relative risk reduction in the annual incidence of invasive breast cancer compared with placebo. However, a major obstacle to its use is adverse events (side effects). Fear of side effects is the primary reason for low uptake of preventive therapy. In the adjuvant setting (treatment given after surgery to prevent recurrence), many women stop taking AIs because of side effects, and this nonadherence increases the risk of breast cancer recurrence.

The idea behind this study was straightforward: if lower or less frequent dosing could maintain efficacy while reducing side effects, more women might be willing to take—and stick with—these medications. This concept had already shown promise in another trial: low-dose tamoxifen (5 mg/day for 3 years) halved disease recurrence in women with previously diagnosed intraepithelial neoplasia (precancerous changes). The current trial, a presurgical phase 2b randomized clinical trial, aimed to apply similar dose-de-escalation logic to exemestane.

How the Study Was Conducted

This was an international, presurgical, 3-arm, double-blind, noninferiority phase 2b trial. "Presurgical" means that women took the study drug for 4 to 6 weeks between their diagnostic biopsy and their breast cancer surgery—a window of time commonly used to test how well a drug works against tumor tissue. "Double-blind" means that neither the participants nor the researchers knew which dosing schedule each woman was receiving. "Noninferiority" means the study was designed to test whether the experimental schedules were not worse than the standard daily dose by more than a predefined margin.

The study was conducted at multiple centers in Italy, the United States, and Norway. It was approved by the National Cancer Institute Central Institutional Review Board, local Italian institutional review boards, and the Regional Committees for Medical and Health Research Ethics in Western Norway. All participants provided written informed consent, and the study followed the CONSORT (Consolidated Standards of Reporting Trials) reporting guideline.

Who Could Participate

Main inclusion criteria were:

• Postmenopausal women with confirmed estrogen receptor–positive (ER-positive) breast cancer, meaning the cancer's growth is fueled by estrogen
• Cancer stage cT0 to cT2 or cN0 to cN1 (early-stage disease that has not spread extensively)
• Blood tests within normal laboratory limits, or with alterations that were not clinically relevant

Women were excluded if their body mass index (BMI) was below 18.5, or if they had previous breast cancer treatment, uncontrolled illness, recent diagnoses of other cancers, or severe osteoporosis.

Randomization and Blinding

Women were randomly assigned (1:1:1) to one of three arms:

1. Exemestane 25 mg once daily (the standard dose)
2. Exemestane 25 mg three times per week
3. Exemestane 25 mg once weekly

Treatment lasted 4 to 6 weeks before surgery. Randomization was stratified by center and by BMI (less than 25 vs 25 or higher). To maintain blinding, weekly blister packs were manufactured containing 7 pills: 7 active pills for the once-daily arm, 3 active pills plus 4 placebos for the 3-times-weekly arm, and 1 active pill plus 6 placebos for the once-weekly arm. Pills were numbered 1 to 7 in each blister pack.

The final visit was scheduled on the day of surgery or the day before. It included assessment for toxic effects, review of concomitant medications, blood collection, and review of the pill diary. Participants continued taking the study drug until the night before surgery. Surgery was ideally performed on day 29, or alternatively on day 36 or 43, to maintain the same treatment schedule in each arm.

How Biomarkers Were Measured

Morning fasting blood samples were collected at baseline and at the final visit. After clotting, the tubes were centrifuged at 2000 times gravity for 10 minutes at room temperature. Serum was stored at −80°C until analysis.

The primary endpoint was the percentage change in serum estradiol (the main form of estrogen in the blood) from baseline to post-treatment. Estradiol and estrone were measured using solid-phase extraction (SPE) followed by liquid chromatography–tandem mass spectrometry (LC-MS/MS) detection—a highly accurate method. The lower limit of quantification was 1.0 pg/mL for estradiol and 5 pg/mL for estrone.

Because estradiol levels are extremely low in postmenopausal women treated with aromatase inhibitors, researchers also analyzed serum estradiol and estrone with an ultrasensitive LC-MS/MS method using automated liquid-liquid extraction (LLE) without derivatization. This method could detect estradiol down to 0.8 pmol/L (0.22 pg/mL) and estrone down to 0.2 pmol/L (0.05 pg/mL).

Other measurements included:

• Sex hormone–binding globulin (SHBG), a protein that carries sex hormones in the blood, measured by chemiluminescent immunoassay
• Lipid profiles (total cholesterol, HDL cholesterol, LDL cholesterol, and triglycerides), determined locally at baseline and final visit
• Androstenedione and testosterone, measured by LC-MS/MS
• Ki-67, a protein that indicates how quickly cells are dividing (cell proliferation), assessed by immunohistochemistry using the Mib-1 monoclonal antibody
• Estrogen receptor, progesterone receptor (PgR), and HER2 expression, also determined by immunohistochemistry

Pretreatment and post-treatment tissue measurements were centralized at the pathology division of the European Institute of Oncology to minimize variability among centers. Toxic effects were evaluated using the National Cancer Institute terminology criteria (Common Terminology Criteria for Adverse Events, version 4.0.3). Menopausal symptoms were assessed using a self-administered questionnaire called the Menopause-Specific Quality of Life Questionnaire, comparing pre-treatment and post-treatment answers.

Statistical Design

The study was designed to test whether the reduction in serum estradiol with the two lower-dose schedules was noninferior to the standard dose, using a noninferiority margin of −6%. In plain terms, the experimental schedule would be accepted as "not worse" if the estradiol reduction was within 6 percentage points of the daily dose's reduction.

The sample size calculation assumed that the daily dose would reduce estradiol by at least 80%. A total of 162 participants was needed to have 80% power to detect noninferiority of −6% in the mean percentage change, assuming the true difference was 0% and the standard deviation was 11%. Accounting for an anticipated 10% dropout rate, 180 participants were randomized. The significance level for the main endpoint was set at P = .025 to account for multiple comparisons (two experimental arms vs the standard arm). For secondary endpoints, a two-tailed P value of less than .05 was considered statistically significant.

The primary analysis was intention-to-treat (ITT), meaning it included all randomized participants who provided blood samples. A planned per-protocol (compliant) analysis was also conducted, defined as participants who received at least 80% of the active scheduled pills and underwent blood testing according to the protocol schedule.

Who Took Part in the Trial

From February 1, 2017, to August 31, 2019, a total of 230 women were screened for eligibility. Of these, 25 were not eligible, 20 withdrew consent, and 5 were excluded for other reasons. This left 180 women who were randomized into one of the three arms: 59 to once-daily, 58 to 3-times-weekly, and 63 to once-weekly exemestane.

The median ages were similar across arms:

• Once-daily arm: 66 years (interquartile range [IQR], 60–71 years)
• 3-times-weekly arm: 63 years (IQR, 60–69 years)
• Once-weekly arm: 65 years (IQR, 61–70 years)

Four participants did not start the treatment (2 for personal reasons, 1 due to an adverse event, and 1 due to a serious adverse event unrelated to the study treatment), and 4 dropped out during the study. The final number of evaluable participants for the primary endpoint was 55 in the once-daily arm, 56 in the 3-times-weekly arm, and 60 in the once-weekly arm—a total of 171 women in the intention-to-treat population.

Drug adherence was high overall: 153 participants (89%) took at least 80% of their pills, with no significant differences among the three arms. A total of 47, 52, and 54 participants in the once-daily, 3-times-weekly, and once-weekly arms, respectively, underwent blood testing according to the protocol schedule. Eighty-eight women had surgery exactly after 4 weeks, 44 after 5 weeks, and 16 after 6 weeks.

Main Result: How Well Did Each Schedule Lower Estrogen?

The primary measure of drug effectiveness was the percentage reduction in serum estradiol (measured by the SPE method) from baseline to the end of treatment. The results differed between the overall intention-to-treat analysis and the analysis restricted to compliant participants.

Intention-to-Treat Analysis

In the ITT population (n = 171), the least square mean percentage change in serum estradiol was:

• −89% for exemestane once daily (n = 55)
• −85% for exemestane 3 times weekly (n = 56)
• −60% for exemestane once weekly (n = 60)

The difference between the once-daily and 3-times-weekly arms was −3.6% (P for noninferiority = .37), which did not meet the noninferiority criterion. The once-weekly arm clearly fell short, with a difference of −28.8% compared with daily dosing.

Compliant Participants (Per-Protocol Analysis)

Among the 153 compliant participants who took at least 80% of their scheduled active pills, the results were more favorable for the 3-times-weekly schedule:

• −91% for once daily
• −92% for 3 times weekly
• −69% for once weekly

The difference between the once-daily and 3-times-weekly arms was 2.0% (97.5% lower confidence limit, −5.6%; P for noninferiority = .02). Because this lower confidence limit did not cross the −6% threshold, the 3-times-weekly schedule was declared noninferior to the once-daily schedule among compliant participants—meaning it worked just as well.

Confirmatory Ultra-Sensitive Analysis

A secondary analysis using the more sensitive liquid-liquid extraction (LLE) method confirmed the overall pattern. In the ITT analysis, estradiol reductions were −96%, −91%, and −72% for once-daily, 3-times-weekly, and once-weekly dosing, respectively. In compliant participants, the corresponding reductions were −96%, −92%, and −74%. The difference between once-daily and 3-times-weekly was −4.9% (P = .28) in the ITT population and −3.8% (P = .11) in compliant participants—again indicating comparable effectiveness of the two schedules, with the once-weekly schedule clearly underperforming.

Suppression Below Detection Limits

The percentage of participants whose estradiol levels fell below the detection limit of the assay was another way of measuring the completeness of estrogen suppression:

• SPE method: 69.0% (once daily), 65.4% (3 times weekly), and 17.2% (once weekly)
• LLE method: 77.7% (once daily), 47.2% (3 times weekly), and 3.4% (once weekly)

The difference between once-daily and 3-times-weekly was not statistically significant (P = .78), while the once-weekly schedule was clearly less effective.

Tissue Biomarkers, Other Hormones, and Cholesterol

Ki-67: A Measure of Cancer Cell Growth

Ki-67 is a protein that indicates how rapidly cells are multiplying. Higher Ki-67 levels generally mean more aggressive tumor growth. In this study, Ki-67 was measured in biopsy tissue before treatment and in the surgical specimen after treatment. The median absolute percentage change in Ki-67, adjusted for baseline values, was:

• −7.5% in the once-daily arm
• −5.0% in the 3-times-weekly arm
• −4.0% in the once-weekly arm

The reductions were not statistically significantly different between arms (once daily vs 3 times weekly, P = .31; once daily vs once weekly, P = .06). By intention-to-treat, the Ki-67 reduction was similar when only compliant participants were analyzed. As an exploratory analysis, Ki-67 was also measured in normal tissue adjacent to the tumor—because expression in normal tissue is very low, no modulation was observed.

Progesterone Receptor (PgR) Expression

PgR is another hormone receptor whose expression often decreases when estrogen signaling is blocked. The median absolute change in PgR was:

• −17.0% in the once-daily arm
• −9.0% in the 3-times-weekly arm
• −7.0% in the once-weekly arm

Again, these differences did not reach statistical significance (once daily vs 3 times weekly, P = .44; once daily vs once weekly, P = .06).

Other Hormones and Lipids

For estrone, total estrone, and estrone sulfate (other forms of estrogen), there were no statistical differences between the once-daily and 3-times-weekly arms, whereas the once-weekly (lower-dose) schedule was less effective. No major changes were observed for androstenedione or testosterone with any dose.

However, a dose-response pattern was noted for sex hormone–binding globulin (SHBG), a protein that binds to and transports sex hormones. The absolute changes were:

• −12.8 nmol/L for once daily
• −7.0 nmol/L for 3 times weekly
• −2.3 nmol/L for once weekly

A greater reduction in SHBG means less of this protective biomarker is available in the blood. Regarding the lipid profile, there was an expected reduction in HDL cholesterol (the "good" cholesterol) with once-daily dosing, but only a marginal effect with the two lower-dose regimens. The changes were −10, −4, and +1 mg/dL for once daily, 3 times weekly, and once weekly, respectively. This suggests that the less frequent schedules may have a more favorable impact on cholesterol levels.

Side Effects and Tolerability

Overall, the treatment was well tolerated, with a total of 358 adverse events reported. Of these, 255 (71%) were grade 1, meaning they were mild. No statistically significant differences in side effects were detected among the three arms.

Women's perception of menopausal symptoms—such as hot flashes, night sweats, and other menopause-related quality-of-life concerns—showed no clinically relevant differences among the three arms, as measured by the Menopause-Specific Quality of Life Questionnaire.

These findings are important because they suggest that while the 3-times-weekly schedule offers comparable estrogen suppression, it does not appear to increase side effects. Conversely, the side-effect profile was not clearly better with the reduced-frequency schedules in this short presurgical window, although the study was not specifically designed to detect differences in long-term toxicity.

What These Findings Mean for Patients

The study provides evidence that exemestane, 25 mg, given three times per week, was noninferior to the standard once-daily dose in reducing circulating estradiol among compliant (adherent) patients. The compliant participants represented 89% of the whole study population—a high rate of adherence that was likely encouraged by the structured clinical trial setting.

Several points from the discussion are worth highlighting for patients:

• The difference in results between the intention-to-treat and per-protocol analyses was not due to differences in baseline characteristics between compliant and noncompliant participants. Rather, it likely reflects the lower variability in estradiol levels among women who adhered to their medication schedule.
• Estrone, total estrone, and estrone sulfate showed no differences between once-daily and 3-times-weekly dosing, reinforcing that the 3-times-weekly schedule profoundly suppresses estrogen in multiple forms.
• A dose-response relationship was noted for SHBG: the daily dose significantly reduced this protective biomarker, whereas the 3-times-weekly dose had a smaller effect, which could potentially be beneficial.
• The once-weekly dose was clearly less effective and should not be considered an equivalent alternative.

The authors note that the benefit-risk ratio of existing drugs could be improved by dose-de-escalation studies that search for the minimal effective dose. The findings from this trial suggest that a 3-times-weekly schedule of exemestane warrants further study in prevention trials and in women who do not tolerate the daily dose in the adjuvant (post-surgery) setting.

If confirmed in larger studies, a less frequent dosing schedule could improve quality of life and adherence for many women. Since nonadherence to aromatase inhibitors is common in the adjuvant setting and increases the risk of breast cancer recurrence, any strategy that makes these drugs easier to take has the potential to improve real-world outcomes. Additionally, prior studies by the same research group showed that low-dose tamoxifen (5 mg daily) halved disease recurrence rates in women with intraepithelial neoplasia—further supporting the concept that "less can be more" when it comes to hormonal therapy dosing.

Study Limitations

Like all research, this trial has important limitations that should be considered when interpreting the results:

• The intention-to-treat analysis did not demonstrate noninferiority. The noninferiority result was achieved only in the per-protocol analysis of compliant participants. This means the 3-times-weekly schedule was "proven" equivalent only among women who took their medication as directed.
• Short treatment duration. The presurgical window was only 4 to 6 weeks, which is long enough to measure changes in estrogen levels and tumor biomarkers (like Ki-67) but not long enough to assess long-term clinical outcomes such as cancer recurrence, survival, or long-term side effects like bone loss and joint pain.
• Surrogate endpoint. The study used serum estradiol suppression as a surrogate marker of efficacy, not actual breast cancer outcomes. While estrogen suppression is a well-established indicator of aromatase inhibitor potency, it is not the same as demonstrating reduced cancer incidence or recurrence.
• Small sample size for subgroup comparisons. Although 180 women were randomized, the study was not powered to detect differences in side-effect profiles or long-term tolerability among the three arms.
• High adherence in a trial setting. The 89% adherence rate observed in this clinical trial may not reflect real-world medication-taking behavior, where adherence to aromatase inhibitors is often much lower.
• Laboratory differences. Biomarkers were measured in different laboratories between April 2020 and December 2021, though centralization of tissue analyses and standardized methods were used to minimize variability.

Recommendations for Patients

Based on this study, here are some practical takeaways for patients and their doctors:

1. Do not change your dose on your own. If you are currently taking exemestane daily, do not reduce the frequency without speaking with your oncologist. The 3-times-weekly schedule is not yet an approved or standard alternative outside of clinical trials.
2. Ask your doctor about the latest evidence. If you are struggling with side effects from daily exemestane, this study provides a scientific basis for discussing whether a reduced-frequency schedule might be appropriate for your situation, particularly in the context of a clinical trial or shared decision-making.
3. Adherence matters enormously. The study's key finding—that 3-times-weekly dosing works as well as daily—applies only to women who took at least 80% of their scheduled pills. The most important thing is to take your medication exactly as prescribed, whatever the schedule.
4. Once-weekly dosing is not recommended. The once-weekly schedule produced significantly less estrogen suppression (60% vs 89% reduction in the ITT analysis) and should not be considered an equivalent option based on current evidence.
5. Watch for side effects and report them. Side effects are a major reason women discontinue aromatase inhibitors, which increases the risk of recurrence. Talking openly with your care team about side effects can lead to strategies that help you stay on treatment.
6. Stay informed about ongoing research. The study authors call for further research on the 3-times-weekly schedule in prevention studies and in the adjuvant setting. Future trials may provide more definitive guidance on alternative dosing.

Ultimately, the goal of this research is to make effective breast cancer prevention and treatment more tolerable, so that more women can benefit from these life-saving medications. For now, the standard daily dose of exemestane remains the proven and approved option—but the evidence for a 3-times-weekly alternative is growing.

Frequently Asked Questions

Can I take exemestane only three times a week instead of every day?

In a study of 180 postmenopausal women with early-stage, estrogen-sensitive breast cancer, those taking 25 mg exemestane three times weekly had estrogen suppression similar to the daily dose—but only if they took at least 80% of their pills. This reduced-frequency schedule is not yet approved. Talk to your oncologist before changing your dose.

Who was eligible for the exemestane dosing study?

Participants were postmenopausal women with confirmed estrogen receptor–positive breast cancer, stage cT0 to cT2 or cN0 to cN1, and blood tests within normal limits. Women were excluded if they had previous breast cancer treatment, uncontrolled illness, recent other cancers, or severe osteoporosis. The study lasted 4–6 weeks before breast cancer surgery.

What side effects were seen with the different exemestane schedules?

Overall, treatment was well tolerated. There were 358 adverse events, of which 71% were mild (grade 1). No statistically significant differences in side effects were detected among the once-daily, three-times-weekly, and once-weekly arms. Menopausal symptoms were similar across all three schedules in this short presurgical study.

What did the study measure to determine if lower-dose exemestane worked?

The main measure was the percentage change in serum estradiol from baseline to after treatment, using highly accurate mass spectrometry. They also measured estrone, sex hormone–binding globulin, cholesterol, and Ki-67, a marker of how quickly cancer cells are dividing. Estrogen suppression was used as a surrogate marker of drug effectiveness.

What does this study mean for women taking exemestane after surgery?

For women who struggle with side effects from daily exemestane, this study provides a scientific basis to discuss a reduced-frequency schedule with their doctor, particularly in a clinical trial or shared decision-making. However, the standard daily dose remains the proven and approved option. Larger studies are needed before three-times-weekly dosing is recommended.

Source Information

Original article title: Efficacy of Alternative Dose Regimens of Exemestane in Postmenopausal Women With Stage 0 to II Estrogen Receptor–Positive Breast Cancer

Authors: Davide Serrano, MD; Sara Gandini, PhD; Parjhitham Thomas, MD; Katherine D. Crew, MD, MS; Nagi B. Kumar, PhD, RD; Lana A. Vornik, MHA, MS; J. Jack Lee, PhD; Paolo Veronesi, MD; Giuseppe Viale, MD; Aliana Guerrieri-Gonzaga, MSc; Matteo Lazzeroni, MD; Harriet Johansson, PhD; Mauro D'Amico, MD; Flavio Guasone, MD; Stefano Spinaci, MD; Bjørn-Erik Bertelsen, MSc; Gunnar Mellgren, MD, PhD; Isabelle Bedrosian, MD; Diane Weber, RN; Tawana Castile, BS, CCRP; Eileen Dimond, RN, MS; Brandy M. Heckman-Stoddard, PhD, MPH; Eva Szabo, MD; Powel H. Brown, MD, PhD; Andrea DeCensi, MD; Bernardo Bonanni, MD

Journal: JAMA Oncology, May 2023, Volume 9, Number 5, pages 664–672. doi:10.1001/jamaoncol.2023.0089. Published online March 23, 2023.

Trial registration: ClinicalTrials.gov Identifier NCT02598557; EudraCT: 2015-005063-16

Funding: The study was conducted with support from the National Cancer Institute (as part of the NCI Community Oncology Research Program) and other institutional funders, with biomarker analyses performed at centralized laboratories in Italy and Norway.

This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and should not replace individualized medical advice from a qualified health care professional. Patients should always consult their oncology team before making any changes to their cancer treatment or prevention plan.