Understanding Estrogen Suppression in Premenopausal Breast Cancer: What the SOFT-EST Study Reveals About Triptorelin Plus Exemestane or Tamoxifen

Table of Contents

• Key Points
• Background: Why Ovarian Function Suppression Matters
• Study Design and Patient Population
• How Estrogen Levels Were Measured
• Key Findings at 3, 6, and 12 Months
• Who Had Suboptimal Estrogen Suppression?
• Exemestane vs. Tamoxifen: Hormone Level Differences
• What This Means for Patients
• Study Limitations
• Recommendations for Patients
• Frequently Asked Questions
• Source Information

Key Points

• In a 12-month study of 116 premenopausal women, triptorelin plus exemestane or tamoxifen produced profound estrogen suppression in most patients.
• At each time point, 17–25% of patients receiving exemestane plus triptorelin had estradiol above 2.72 pg/mL; 34% had an elevation at least once.
• Women under 35, those with higher BMI, and those without prior chemotherapy were more likely to have incomplete ovarian suppression.
• Median estradiol in the exemestane plus triptorelin group was 0.625 pg/mL, lower than typical postmenopausal levels on aromatase inhibitors.
• The study measured hormone levels only, not cancer outcomes, so the clinical relevance of transient estradiol elevations remains uncertain.

Background: Why Ovarian Function Suppression Matters

For over a century, doctors have used ovarian function suppression (OFS) as a treatment strategy for premenopausal women with endocrine-responsive (hormone receptor–positive) breast cancer. The ovaries produce estrogen, which can fuel the growth of hormone-sensitive breast tumors. By suppressing ovarian function, doctors aim to remove this fuel source.

Ovarian suppression can be achieved in three main ways: surgical removal of the ovaries (oophorectomy), radiation to the ovaries, or medications called gonadotropin-releasing hormone agonists (GnRHa), such as triptorelin. In the 1990s, phase III trials reported similar effectiveness between ovarian ablation and GnRHa treatment. Studies testing GnRHa alone, or combined with tamoxifen or aromatase inhibitors (AIs), have shown clinical activity with effective estrogen suppression for most patients.

However, researchers have noticed that some patients treated with GnRHa may have incomplete ovarian suppression—meaning their estrogen levels do not drop as low as expected. This is particularly concerning when GnRHa is combined with an aromatase inhibitor like exemestane, because AIs work less effectively—and may even be stimulatory—when residual ovarian function remains.

The Suppression of Ovarian Function Trial (SOFT) and the Tamoxifen and Exemestane Trial (TEXT) were large international adjuvant trials that demonstrated a significant benefit in disease-free survival for exemestane plus OFS compared with tamoxifen plus OFS. Approximately 95% of patients in these trials used GnRHa as their OFS method. Yet, detailed data about how well GnRHa suppresses estrogen in the adjuvant setting remained limited, especially for the combination of GnRHa plus AIs.

Most earlier reports describing the endocrine effects of GnRHa—whether used alone or with tamoxifen or an AI—had short follow-up periods (3 to 6 months on average), small sample sizes, or inadequate measurement of estradiol (E2) levels. This is why researchers launched the SOFT Estrogen Substudy (SOFT-EST), a prospective substudy of the larger SOFT trial, to answer a critical question: Are there patients who experience suboptimal estrogen suppression while receiving triptorelin plus exemestane?

Study Design and Patient Population

The parent SOFT trial enrolled 3,066 premenopausal women with early breast cancer. These women either remained premenopausal after receiving (neo)adjuvant chemotherapy or were candidates for adjuvant tamoxifen alone. They were randomly assigned to one of three 5-year treatment options:

• Exemestane plus ovarian function suppression (OFS)
• Tamoxifen plus OFS
• Tamoxifen alone

OFS was achieved through the patient's choice of one of three methods: triptorelin acetate (Decapeptyl Depot, 3.75 mg given intramuscularly every 28 ± 3 days), bilateral oophorectomy (surgical ovary removal), or ovarian irradiation (radiation therapy).

For the SOFT-EST substudy specifically, all patients enrolled in SOFT at 24 selected sites who were randomly assigned to tamoxifen plus OFS or exemestane plus OFS, and who chose triptorelin as their OFS method, were offered participation. The target was 120 patients: 30 receiving tamoxifen plus triptorelin and 90 receiving exemestane plus triptorelin.

From March 2009 to January 2011, 123 patients were enrolled (32 in the tamoxifen plus triptorelin group; 91 in the exemestane plus triptorelin group). Of these, 116 patients actually started triptorelin and had at least one blood sample analyzed, forming the analytic cohort. This included:

• 30 patients receiving tamoxifen plus triptorelin
• 86 patients receiving exemestane plus triptorelin

The median age at random assignment was 44 years (interquartile range, 41–48 years). Sixty-four patients (55.2%) had received prior chemotherapy. The median baseline E2 level was 50.6 pg/mL, and the median baseline estrone sulfate (E1S) level was 894 pg/mL.

An important observation: despite meeting the protocol's definition of premenopausal status, 35% of patients had baseline E2 levels consistent with postmenopause (≤20 pg/mL) when measured by the highly sensitive assay. This was supported by higher centrally measured follicle-stimulating hormone (FSH) and luteinizing hormone (LH) levels in those patients. Among women in the exemestane plus triptorelin group, 56% of those who had prior chemotherapy and 8% of those who had not had E2 levels at study entry consistent with postmenopausal status.

How Estrogen Levels Were Measured

Blood samples were collected at baseline (before treatment started) and then at 3, 6, 12, 18, 24, 36, and 48 months. Samples were drawn while patients were fasting and before the triptorelin injection. The study analyzed samples from the first 12 months of treatment for this report.

Serum aliquots were stored at −20°C and shipped to a central laboratory for analysis. Estrogen measurements—specifically estradiol (E2), estrone (E1), and estrone sulfate (E1S)—were performed using gas chromatography tandem mass spectrometry (GC/MS/MS), which is considered a benchmark assay due to its high sensitivity and specificity.

The lower limits of quantification (LLQ) for the assay were:

• E2: 0.625 pg/mL
• E1: 1.56 pg/mL
• E1S: 3.13 pg/mL

Researchers confirmed that there was no cross-reactivity with exemestane in an ad hoc experiment conducted before testing the samples. All samples from the first 12 months were run consecutively without knowledge of treatment assignment, ensuring objectivity.

FSH and LH levels were measured by electrochemiluminescence using a Cobas 6000 automated analyzer, with a measurement range of 0.100 to 200 mIU/mL.

The primary objective of this 12-month analysis was to describe E2, E1, and E1S levels at 3, 6, and 12 months, and to determine the proportion of patients receiving exemestane plus triptorelin who had E2 levels greater than 2.72 pg/mL (10 pmol/L)—a strict threshold considered inconsistent with postmenopausal levels on an AI. Secondary objectives included comparing estrogen, FSH, and LH levels between treatment groups and exploring patient characteristics associated with suboptimal suppression.

Key Findings at 3, 6, and 12 Months

After accounting for missing samples and early discontinuations during the first year, 79 patients treated with exemestane plus triptorelin had at least one post-baseline sample available for analysis, and 27 patients treated with tamoxifen plus triptorelin had the same.

The results showed that estrogen suppression was dramatic and consistent in the exemestane plus triptorelin group. For all three estrogen fractions (E2, E1, and E1S), there was a median reduction from baseline of at least 95% at all time points after treatment initiation.

Estradiol (E2) Levels

Median E2 levels in the exemestane plus triptorelin group were 0.625 pg/mL (the lower limit of quantification) at all post-baseline time points. This represents a median reduction of 96% from baseline at 3, 6, and 12 months.

Estrone (E1) Levels

Median E1 levels were 1.56 pg/mL (also at the lower limit of quantification) at all post-baseline time points, reflecting a median reduction of 95% from baseline.

Estrone Sulfate (E1S) Levels

Estrone sulfate—the most abundant estrogen fraction in plasma—was reduced to 11.7 pg/mL at 3 months, 14.9 pg/mL at 6 months, and 10.6 pg/mL at 12 months. These represent median reductions of approximately 96–98% from baseline.

The Critical Finding: Some Patients Had E2 Above the Threshold

Despite the impressive overall suppression, a notable proportion of patients had E2 levels that rose above the study's predefined threshold of 2.72 pg/mL:

• At 3 months: 25% of patients had E2 levels greater than 2.72 pg/mL
• At 6 months: 24% of patients had E2 levels greater than 2.72 pg/mL
• At 12 months: 17% of patients had E2 levels greater than 2.72 pg/mL

Overall, 27 of 79 patients (34.2%; 95% CI, 23.9% to 45.7%) receiving exemestane plus triptorelin had at least one post-baseline E2 value greater than 2.72 pg/mL during the first year of treatment.

When researchers explored less stringent thresholds, they found that 18% of patients had E2 levels greater than 10 pg/mL and 13% of patients had E2 levels greater than 20 pg/mL at least once during the 12-month period. However, these higher elevations were mostly not persistent: only six women (8%) had E2 values greater than 10 pg/mL and only one woman (1%) had E2 values greater than 20 pg/mL at more than one post-baseline time point.

Looking at the pattern over time among the 27 women who had at least one elevated E2 reading: 14 patients had E2 above the threshold at one post-baseline time point, 9 patients at two time points, and 4 patients—three of whom were younger than 35 years and three of whom had not received prior chemotherapy—had E2 above the threshold at all three post-baseline time points. These four patients represent 8% of the 48 women who had all three post-baseline samples analyzed.

Interestingly, two patients in the exemestane plus triptorelin group experienced vaginal bleeding more than 3 months after starting triptorelin, but significantly elevated E2 (41 pg/mL) was confirmed centrally in only one of these patients.

Who Had Suboptimal Estrogen Suppression?

The researchers compared baseline characteristics between the 27 patients who had on-treatment E2 levels above 2.72 pg/mL at any point and the patients whose E2 levels stayed below the threshold. They identified several factors associated with a higher likelihood of suboptimal suppression:

• No prior chemotherapy (P = .06)—patients who had not received chemotherapy before were more likely to have elevated E2
• Higher body mass index (BMI) (P = .05)—patients with higher BMI had a greater chance of suboptimal suppression
• Lower baseline FSH levels (P = .002)—patients with lower FSH before treatment were more likely to have elevated E2
• Lower baseline LH levels (P = .004)—the same pattern was seen for LH

Being chemotherapy-naïve was notably associated with suboptimal suppression: 46% of chemotherapy-naïve patients had an E2 level above the threshold at least once, compared with a lower percentage among patients who had received prior chemotherapy. Similarly, being younger than 35 years was a risk factor—four of the eight women in this age group (50%) had elevated E2 levels at some point.

Exemestane vs. Tamoxifen: Hormone Level Differences

The study also compared hormone levels between the two treatment groups. The reductions in E2, E1, and E1S were significantly greater in the exemestane plus triptorelin group than in the tamoxifen plus triptorelin group (P < .001 for each post-baseline time point). This makes biological sense: exemestane works by blocking the aromatase enzyme that produces estrogen, while tamoxifen works by blocking estrogen receptors rather than reducing estrogen production itself.

Gonadotropin levels also differed between the groups. Median FSH values were higher in the exemestane plus triptorelin group (P < .001 at each post-baseline time point), while LH values were persistently lower (P ≤ .01 at each post-baseline time point) compared with the tamoxifen plus triptorelin group. Both groups showed marked reductions in FSH and LH after treatment began.

These complex gonadotropin dynamics likely result from several interacting factors: the direct suppressive effect of the GnRHa (triptorelin), the removal of estrogen's normal feedback on gonadotropin production (since FSH is more sensitive than LH to this feedback), and tamoxifen's direct effect on the pituitary gland.

What This Means for Patients

The study provides both reassuring and cautionary messages for patients. On the reassuring side, 66% of premenopausal patients treated with exemestane plus triptorelin achieved a profound, persistent reduction in E2 levels during the first 12 months of treatment. The median E2 levels achieved—0.625 pg/mL—are actually lower than those typically reported in postmenopausal women on aromatase inhibitors.

For context, mean E2 values measured by GC/MS/MS in postmenopausal women are typically 4.0 to 7.3 pg/mL, and in postmenopausal patients on letrozole (another AI), E2 levels were below 0.65 pg/mL in all samples. This means that the women in the exemestane plus triptorelin group who achieved full suppression reached levels comparable to—or even lower than—those seen in postmenopausal women on AI therapy.

However, the cautionary message is equally important: at every time point, at least 17% of patients had E2 levels above 2.72 pg/mL, and 34% had an elevation at least once during the year. While the clinical implications of these ultra-low E2 thresholds are still uncertain, the findings suggest that some patients—particularly those who are younger than 35, have not received prior chemotherapy, or have a higher BMI—may not achieve complete ovarian suppression with triptorelin alone.

It is worth noting that in the postmenopausal setting, small differences in the degree of aromatase inhibition between different third-generation AIs have not translated into clinically meaningful differences in efficacy in head-to-head comparisons. This suggests that minor variations in estrogen suppression may not necessarily affect treatment outcomes, though this remains an open question.

The study's findings are consistent with results from the Hormonal Bone Effects (HOBOE) trial, which randomly assigned patients to triptorelin plus either letrozole or tamoxifen and measured hormone levels at baseline and after 6 months. Like SOFT-EST, the HOBOE trial found lower median E2 levels in the AI group than in the tamoxifen group, with LH levels significantly lower and FSH levels significantly higher in the AI group.

Study Limitations

Patients should understand the limitations of this research when interpreting the findings:

• No outcome data: This substudy measured hormone levels only—it was not designed to determine whether suboptimal estrogen suppression leads to worse breast cancer outcomes. The relationship between E2 levels above the threshold and actual recurrence risk remains unknown.
• Uncertain clinical relevance of the threshold: The 2.72 pg/mL threshold was chosen because it represents E2 levels inconsistent with postmenopausal status on AIs. However, the clinical implication of this ultra-low threshold is still uncertain, and the study's own data show that E2 elevations were mostly non-persistent.
• Limited sample size: The substudy enrolled 116 patients, with only 79 in the exemestane plus triptorelin group having post-baseline samples. Subgroup analyses—such as the finding about women younger than 35—involved very small numbers.
• Missing samples and discontinuations: Some patients stopped triptorelin or had samples not taken during the first year, which could introduce bias.
• Higher-than-expected baseline postmenopausal status: The fact that 35% of patients already had postmenopausal-range E2 levels at baseline—despite meeting premenopausal criteria by local testing—highlights the challenges of assessing menopausal status, especially after chemotherapy.
• Single GnRHa studied: The results apply specifically to triptorelin at the 3.75 mg monthly dose, and may not be generalizable to other GnRHa preparations or doses.

Recommendations for Patients

Based on this study, patients and their doctors may want to consider the following:

1. Discuss monitoring of estrogen levels: For patients receiving exemestane plus triptorelin—particularly those who are under 35, have a higher BMI, or did not receive prior chemotherapy—it may be worth discussing with your oncologist whether checking E2 levels during treatment is appropriate.
2. Report symptoms promptly: Vaginal bleeding more than 3 months after starting triptorelin could be a sign of incomplete ovarian suppression. In this study, one of two patients who experienced this symptom had an elevated E2 level of 41 pg/mL. Any unexpected bleeding should be reported to your care team.
3. Understand that transient elevations may not be alarming: Most E2 elevations in this study were not persistent. Only 8% of patients had E2 above 10 pg/mL and only 1% had E2 above 20 pg/mL at more than one time point. A single elevated reading does not necessarily mean treatment is failing.
4. Maintain a healthy weight: Since higher BMI was associated with suboptimal estrogen suppression, maintaining a healthy weight through diet and exercise may support treatment effectiveness.
5. Continue treatment as prescribed: Despite these findings, the parent SOFT trial demonstrated significant improvements in disease-free survival with exemestane plus OFS compared with tamoxifen plus OFS. The vast majority of patients achieved profound estrogen suppression, and the clinical benefits of these treatments are well-established.
6. Ask about follow-up: The ongoing SOFT-EST study will continue evaluating estrogen levels through 48 months of treatment, which may provide additional insight into whether early elevations resolve or persist over time.

In summary, this study confirms that triptorelin plus exemestane provides profound estrogen suppression for most premenopausal breast cancer patients—but not all. Approximately one-third of patients experienced at least one E2 measurement above the strict threshold during the first year. Future research linking these hormonal findings to actual patient outcomes will help determine whether closer monitoring or alternative suppression strategies are needed for the subgroup of patients with suboptimal suppression.

Frequently Asked Questions

What did the SOFT-EST study measure in premenopausal women with hormone receptor–positive breast cancer?

The study followed 116 premenopausal women for 12 months while they received triptorelin injections plus either exemestane or tamoxifen. Researchers used highly sensitive mass spectrometry to measure estrogen levels at 3, 6, and 12 months. They wanted to see how completely triptorelin suppressed ovarian estrogen production during adjuvant breast cancer treatment.

What percentage of women on exemestane plus triptorelin had estradiol levels above the study threshold?

At 3 months, 25% of patients had estradiol above 2.72 pg/mL; at 6 months, 24%; and at 12 months, 17%. Overall, 34.2% of 79 patients had at least one elevated reading during the first year. Most elevations were not persistent. Only 8% had levels above 10 pg/mL at more than one time point.

Which patients were more likely to have incomplete estrogen suppression with triptorelin plus exemestane?

Patients who were younger than 35, had not received prior chemotherapy, or had a higher body mass index were more likely to have estradiol levels above 2.72 pg/mL. For example, 50% of women under 35 had elevated levels at some point. Lower baseline FSH and LH levels were also associated with suboptimal suppression.

How did estrogen levels compare between the exemestane and tamoxifen groups?

Women receiving exemestane plus triptorelin had significantly greater reductions in estradiol, estrone, and estrone sulfate than women receiving tamoxifen plus triptorelin at every post-baseline time point. Median estradiol in the exemestane group was 0.625 pg/mL, the lower limit of quantification. Tamoxifen blocks estrogen receptors rather than lowering estrogen production.

Should I be worried if I have a single elevated estradiol reading during treatment?

Probably not, based on this study. Most elevations were not persistent: only 8% of patients had estradiol above 10 pg/mL and only 1% had estradiol above 20 pg/mL at more than one time point. The study did not measure whether these temporary increases affect cancer outcomes. Always discuss results with your oncologist.

Should I get a second opinion about my premenopausal breast cancer treatment if I'm taking triptorelin plus exemestane and concerned about incomplete estrogen suppression?

About one in three premenopausal women taking triptorelin plus exemestane had at least one estradiol reading above the strict threshold during the first year, and at every three-month check at least 17% were above it. Women under 35, those with higher BMI, and those who had not had chemotherapy were more likely to have incomplete ovarian suppression. A second opinion can help you decide whether to ask your oncologist about checking estradiol levels during treatment or whether alternative suppression approaches should be considered. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original Article Title: Estrogen Levels in Premenopausal Women With Hormone Receptor–Positive Breast Cancer Receiving Adjuvant Triptorelin Plus Exemestane or Tamoxifen in the Suppression of Ovarian Function Trial (SOFT)- The SOFT-EST Substudy

Authors: Meritxell Bellet, Kathryn P. Gray, Prudence A. Francis, István Láng, Eva Ciruelos, Ana Lluch, Miguel Angel Climent, Gustavo Catalán, Antoni Avella, Uriel Bohn, Antonio González-Martin, Roser Ferrer, Roberto Catalán, Analía Azaro, Agnita Rajasekaran, Josefa Morales, Josep Vázquez, Gini F. Fleming, Karen N. Price, and Meredith M. Regan

Journal: Journal of Clinical Oncology, Volume 34, Number 14, May 10, 2016, pages 1584–1593. Published online ahead of print on January 4, 2016. DOI: 10.1200/JCO.2015.61.2259

Clinical Trial Registration: NCT00975676

Funding: The SOFT-EST substudy was supported by Pfizer International Oncology, Spain and Portugal Cluster, and the International Breast Cancer Study Group (IBCSG).

This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and should not replace medical advice from your oncology care team. Always discuss any questions about your specific breast cancer treatment with your doctor.