What Is Optimal Endocrine Therapy in Early Breast Cancer?

Linda Brookes, MSc


November 16, 2015

It is becoming more and more difficult to decide which treatment schedule to give premenopausal women with estrogen receptor (ER)-positive breast cancer following the results from the TEXT and SOFT trials," said Vivianne Tjan-Heijnen, MD, PhD, professor in the department of medical oncology at Maastricht University Medical Centre in Maastricht, The Netherlands.

The Tamoxifen and Exemestane Trial (TEXT)[1] and the Suppression of Ovarian Function Trial (SOFT)[2] raised a number of questions about the value of ovarian function suppression and the role of adjuvant aromatase inhibitors (AIs) in these patients, commented Dr Tjan-Heijnen. The findings of both trials have been incorporated into recent breast cancer management guidelines.[3,4,5] In a 2015 European Cancer Congress session on early breast cancer management, Dr Tjan-Heijnen, who was not involved with either trial, reviewed the questions raised by TEXT and SOFT, drawing conclusions that did not always concur with those of the investigators.

Benefit of Ovarian Function Suppression

Dr Tjan-Heijnen recalled that earlier studies showed that very young women with ER-positive early breast cancer have a higher risk for relapse compared with older premenopausal women with ER-positive tumors or women with ER-negative tumors.[6] It was also observed that chemotherapy-induced amenorrhea occurred less frequently in women younger than 35 years of age compared with older premenopausal women.

In the National Surgical Adjuvant Breast and Bowel Project-B30 study, chemotherapy-induced amenorrhea improved overall survival but only in patients with ER-positive early breast cancer, not in patients with ER-negative disease.[7] "That implies that dose intensity or chemotherapy concentration in the patient cannot be an alternative explanation. This was another indication that ovarian function suppression might be important," concluded Dr Tjan-Heijnen. A meta-analysis of data from studies with luteinizing hormone-releasing hormone (LHRH) agonists showed reductions in breast cancer recurrence especially in patients younger than 40 years,[8] although the older studies included in the analysis had limitations, most importantly that many patients were not treated with tamoxifen.

SOFT addressed the question of whether ovarian suppression is of benefit in premenopausal women with hormone receptor-positive early breast cancer when combined with tamoxifen.[2] A total of 3066 premenopausal women, stratified according to prior receipt or nonreceipt of chemotherapy, were randomly assigned to receive 5 years of tamoxifen, tamoxifen plus ovarian function suppression, or exemestane (an AI) plus ovarian function suppression. Women randomly assigned to ovarian suppression in either arm had the choice of monthly injections of triptorelin, surgical removal of the ovaries, or radiation. In the primary analysis, disease-free survival (DFS) at 5 years was improved with tamoxifen plus ovarian suppression compared with tamoxifen alone, but the difference was small, at only 2% for all patients included, Dr Tjan-Heijnen noted.

"In a preplanned subgroup analysis, however, there was an important difference between patients who did and did not receive prior chemotherapy," she commented. "Those who did not receive prior chemotherapy had an excellent outcome, indicating that ovarian suppression is not of benefit in patients who have an excellent prognosis. On the other hand, in patients who had worse tumor characteristics and who received adjuvant chemotherapy, the difference in DFS was larger, at nearly 5%, with an almost significant hazard ratio of 0.78." The SOFT investigators concluded that in women who were at sufficient risk for recurrence, the addition of ovarian suppression improved breast cancer outcome. "This gives one clue for selecting patients that might benefit from additional ovarian suppression—the tumor profile," Dr Tjan-Heijnen explained.

Patient age, as indicated by the SOFT findings, may prove another selection factor. In the patients with no prior chemotherapy, only 10% were younger than 40 years of age compared with nearly half of the patients (49%) who had received prior chemotherapy.

"As we know, patients who are older than 40 will have a 'natural' menopause within the next few years," commented Dr Tjan-Heijnen. "Importantly, patients who receive chemotherapy and who remain premenopausal have been shown to undergo natural menopause as a late side effect of chemotherapy. We could expect that some patients on the control arm of the SOFT trial may have undergone menopause, so that there may have been a shorter time difference between the 5 years of ovarian suppression between the experimental arm and the control arm," Dr Tjan-Heijnen proposed. "If this is true, we should see an age relation with the outcome, and in the elderly premenopausal patients there was only a small benefit with the addition of ovarian suppression," she noted, "whereas in the very young patients (aged <35 years) there was a big advantage from ovarian suppression."

Dr Tjan-Heijnen noted that at the 2014 San Antonio Breast Cancer Symposium, the SOFT investigators reported that the most striking benefit from adding ovarian suppression to hormonal therapy was found among women younger than 35 years, who had an 11% advantage in DFS compared with the 5% average.[9]

"Apart from tumor characteristics, age is also important in deciding how big the advantage is for a specific patient, with very young premenopausal patients having the largest benefit from ovarian suppression," Dr Tjan-Heijnen said. "OS was significantly improved by nearly 3.6%, so the most important endpoint was also positive. We can conclude from the SOFT results that the addition of ovarian suppression improves both DFS and OS breast cancer outcome and that the largest impact from ovarian suppression is seen in the very young and in unfavorable risk groups."

AIs Instead of Tamoxifen?

Because AIs have resulted in about a 3% survival benefit in postmenopausal breast cancer patients,[10] TEXT and SOFT addressed this issue in premenopausal women, Dr Tjan-Heijnen explained. In TEXT, women were randomly assigned to receive exemestane plus ovarian function suppression or tamoxifen plus ovarian function suppression. In a combined analysis of data from SOFT and TEXT,[1] there was a difference of about 4% in 5-year DFS in favor of exemestane plus ovarian suppression (hazard ratio, 0.72). "This is a positive result," Dr Tjan-Heijnen acknowledged. "However, in contrast to the investigators, I am not convinced that this is the way we should go for our own patients in daily practice. Patients actually did not benefit in overall survival (OS) with AIs. You might say, the time for follow-up was too short, but what was important is that the hazard ratio went in the wrong direction, above 1 (1.14). Even though it was not statistically significantly above 1, I am still worried about what this means."

in contrast to the investigators, I am not convinced that this is the way we should go...

Dr Tjan-Heijnen recalled that the Austrian Breast and Colorectal Cancer Study Group (ABCSG) showed, using the same study design,[11] that among premenopausal women who received zoledronic acid and either tamoxifen or anastrozole, those who received anastrozole had a significantly worse OS (hazard ratio, 1.63). "I personally would not start giving AIs in premenopausal women when I do not understand the underlying mechanism," she said.

When the ABCSG investigators analyzed their findings, they observed that in patients who were overweight, anastrozole was less effective for DFS and for OS compared with normal-weight patients.[12] "This might be one explanation. With tamoxifen it is not relevant, because the results were the same in patients who were overweight and those who were of normal weight," Dr Tjan-Heijnen added. The ABCSG investigators hypothesized that the effect might be the result of incomplete estrogen level suppression by AIs in overweight women. "This seems a logical suggestion, but it has to be tested further," commented Dr Tjan-Heijnen.

A similar finding to that of the ABCSG study was the observation from a British study[13] that young women with chemotherapy-induced amenorrhea had higher estradiol blood levels at all time points compared with patients older than 60 years of age. "Perhaps we cannot extrapolate results from older postmenopausal patients automatically to younger premenopausal patients, at least with a diagnosis of breast cancer," Dr Tjan-Heijnen suggested. "The same British study showed that patients who experienced ovarian function recovery had the highest estradiol levels, even before regaining ovarian function. This is the case in about one quarter of all patients above 40 years of age."

AIs may promote recovery of ovarian function in some women with chemotherapy-induced amenorrhea.

"Especially in patients who have chemotherapy-induced amenorrhea, I would say, do not give AIs unless you really regularly check estradiol levels." she advised. Dr Tjan-Heijnen and her colleagues have found consistently elevated levels of follicle-stimulating hormone (FSH) in breast cancer patients with chemotherapy-induced amenorrhea (unpublished data). Patients on tamoxifen initially had the highest FSH levels, but these levels declined after 1 year; in patients on AIs, the FSH levels remained elevated. In some patients, commented Dr Tjan-Heijnen, elevated FSH levels may lead to stimulation of the ovaries and recurrence of ovarian function. "That is an important issue to remember," she noted.

The role of AIs as an adjuvant endocrine treatment strategy for premenopausal women remains uncertain, concluded Dr Tjan-Heijnen. "AIs may promote recovery of ovarian function in some women with chemotherapy-induced amenorrhea, and I believe they should not be used or only used with caution," she said.

The Other Side of Endocrine Therapy: Side Effects

In the Adjuvant Tamoxifen: Longer Against Shorter (ATLAS) study, longer duration of treatment (10 vs 5 years) improved breast cancer outcome.[14] The risk for recurrence was 12% in women who continued treatment vs 15% in those who did not, regardless of whether patients had node-negative or node-positive disease. "In our center," Dr Tjan-Heijnen explained, "we select the patients with the worst tumor profile characteristics to offer longer duration tamoxifen treatment." The absolute benefit is larger, and the difference is only apparent after about 15 years from diagnosis, she said.

"Of course there is another side to the coin with these endocrine treatments—the side effects," Dr Tjan-Heijnen cautioned. In SOFT and TEXT, patients on tamoxifen reported more hot flashes and heavier night sweats whereas sexual functioning and joint and bone pain were worse for patients treated with AIs.[15] "Even after 5 years, many patients were still reporting these side effects, so they did not diminish over time," she noted.

The treatment burden was perceived as high, resulting in a high percentage of patients who stopped treatment prematurely because of the side effects: 1 of 6 (16%) on AI plus ovarian suppression and 1 of 9 (11%) on tamoxifen plus ovarian suppression. "I think that we should try to manage these symptoms by lifestyle interventions or coping strategies, although symptoms and patient selection need to be addressed with every patient individually," Dr Tjan-Heijnen advised. Clinical toxicities are also important, but even in postmenopausal women, the numbers appear to be low for both AIs and tamoxifen,[16] she noted.

A significant number of early breast cancer patients have bone fractures after 5 years of hormonal therapy (8.2% for AIs vs 5.5% for tamoxifen), according to a recent meta-analysis.[17] "It is important to think of what can we do to lower these figures," Dr Tjan-Heijnen said. In the ABCSG Trial 12, patients were randomly assigned to additional or no zoledronic acid from the beginning of the trial.[11] Without zoledronic acid, there was significant bone loss with only a partial recovery 2 years after completing treatment. In patients who received zoledronic acid, bone loss was prevented during therapy, and bone marrow density was improved even at 5 years. "Perhaps this is too little an argument to support the routine use of zoledronic acid, but I think these new data might give extra support for treating a patient with zoledronic acid or another bisphosphonate," Dr Tjan-Heijnen suggested.

According to another meta-analysis, patients who were menopausal when treatment began, including those who were initially premenopausal and who received an LHRH agonist, administration of adjuvant bisphosphonate was associated with a 2% reduction in bone recurrence and a 2% reduction in breast cancer mortality.[18] This benefit was obtained with only a few infusions, Dr Tjan-Heijnen noted.


Tamoxifen or ovarian suppression is an endocrine standard of care in premenopausal women with hormone receptor-positive early breast cancer, according to Dr Tjan-Heijnen. One single treatment modality (tamoxifen or ovarian function suppression) might be a good alternative option in some patients, depending on age, estimated prognosis, and tolerance, she suggested. In high-risk patients, longer tamoxifen (10 years) might be considered. The value of AI in combination with ovarian function suppression in premenopausal women remains unclear. Supportive measures are recommended because endocrine treatment causes significant side effects with risk for early treatment cessation. "This is a very important issue for daily care," Dr Tjan-Heijnen stressed. Addition of bisphosphonates should be considered for premenopausal women undergoing ovarian function suppression to prevent bone loss and to improve breast cancer outcome.

Dr Tjan-Heijnen reports no relevant financial relationships.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.