Risk Status Determines Optimal Treatments in Premenopausal HR+ Breast Cancer

Dawn L. Hershman, MD, MS

Risk status should be used to help decide whether premenopausal women with early-stage hormone receptor (HR)—positive breast cancer need more intensive therapy with an aromatase inhibitor (AI), explained Dawn L. Hershman, MD, MS.

“Many clinicians don't really have a good sense of which premenopausal patients should receive ovarian function suppression (OFS) with an AI versus those who should not,” said Hershman. “My take-home message would be to use these risk calculators because the very small portion of women who fall into the highest-risk range may get a 10% absolute benefit. We also have to consider the low-risk patients who may not need [an AI] because they can really suffer from those adverse events (AEs).”

Data from the phase III SOFT and TEXT trials showed that the addition of adjuvant exemestane or tamoxifen to OFS is superior to tamoxifen alone in premenopausal women with HR-positive breast cancer. When investigators evaluated the rate of distant recurrence, they found that patients with a high recurrence risk derived more benefit from an AI than from tamoxifen plus OFS or tamoxifen alone. Among patients with the highest risk, the absolute improvement in 8-year distant recurrence ranged from 10% to 15%.

In an interview during the 2020 OncLive^® State of the Science Summit™ on Breast Cancer, Hershman, professor of medicine and epidemiology and leader of the Breast Cancer Program of the Herbert Irving Comprehensive Cancer Center at Columbia University, discussed current treatment considerations in premenopausal women with early-stage, HR-positive breast cancer.

OncLive: Could you discuss the evolution of treatment for premenopausal women with early-stage HR-positive, HER2-negative disease?

Hershman: Targeting the estrogen receptor (ER) has been one of the most effective precision medicine platforms for women with hormone-sensitive breast cancers for decades. We have various ways of doing that, such as blocking estrogen from getting to cancer cells and lowering estrogen levels. About 65% to 70% of women have ER-positive breast cancers, [so these approaches] are key components of treatment in women with early-stage disease.

Initially, it wasn't really clear if premenopausal women got the same benefit [from endocrine therapy] as postmenopausal women. Through years of research, however, we discovered that premenopausal women received just as much benefit from tamoxifen as postmenopausal women did, even though they continued to menstruate. What was really interesting from that research is we learned that very young women got the bulk of the benefit when their chemotherapy put them into menopause. A lot of work has resulted in various strategies to maximize that effect. A series of trials have looked at suppressing ovaries in premenopausal patients with high-risk, ER-positive breast cancer, and set out to answer the question of whether or not we should give tamoxifen or an AI along with it.

How do the results of the SOFT and TEXT trials contribute to treatment decisions in this space?

The SOFT and TEXT trials are very similar in that they evaluated premenopausal women and the effect of OFS. We discovered that there is a benefit to suppressing a woman's ovaries and giving her an AI along with it compared with tamoxifen alone. However, the combination is very toxic. We always consider the toxicity of chemotherapy, but the toxicities from endocrine therapy can be far worse and they go on for much longer. We give these treatments for 5, sometimes even 10 years, and it's super challenging to keep patients on their therapy.

Research has shown that only about 50% of patients are taking their therapy 80% of the time after 5 years. The reasons why people stop their medications are complicated. Among the most common reasons are the AEs associated with the treatment. When we suppress the ovarian function of young women and prescribe an AI, close to 100% experience AEs, including depression, hot flashes, difficulty sleeping, severe bone pain, and mood problems. Additionally, [AIs] can increase the risk for osteoporosis and sometimes even cardiovascular disease. The therapy is not without significant consequences.

To better treat these women, we need better risk prediction models to figure out who is at highest risk; that way, we give those patients more intensive treatment and spare women who don’t have such high-risk cancers [from unnecessarily toxic treatment]. A group at Dana-Farber Cancer Institute, led by Meredith M. Regan, ScD, created a risk profile based on age, tumor characteristics, grade, and lymph node involvement to determine [the patient’s risk]. If the patient is young and falls into the high-risk category, we should try to give OFS plus an AI. [The same is true if the patient has] HER2-positive disease. If the patient has a smaller, low-grade cancer with no lymph node involvement, maybe we can spare them the toxicity [of intensive therapy]. It's hard to get through tamoxifen in and of itself.

Which patients are the best candidates for extended endocrine therapy?

Of course, there is the question of who needs 5 years of treatment versus 10 years of treatment. A series of studies showed that there's some benefit to giving tamoxifen for 10 years versus 5 years. However, the bulk of the benefit is in preventing new breast cancers. We have to personalize our treatment based on the patient's risk. Young women have a higher likelihood of having bilateral mastectomies. Perhaps they're not going to get the biggest benefit from continued therapy, unless they have a very high-risk breast cancer to begin with. Ultimately, we need to develop better genomic strategies so that we can more precisely figure out who needs longer therapy versus who can get away with 5 years of treatment.

Could you elaborate on the use of genomics in the space?

The biggest impact came from the TAILORx trial, which helped us determine who doesn't need chemotherapy. However, [the study] can be very confusing to patients. Sometimes patients with a lower Oncotype DX score get more benefit from hormonal therapy. Every patient in the TAILORx trial received hormonal therapy, [which is standard of care]. Sometimes patients believe that if they’re low risk, they don’t need chemotherapy or hormone therapy; however, that's not what the study showed. We have to be really careful in our message because just because somebody is low or intermediate risk, it doesn’t mean that they don't need any hormonal therapy at all. [If we’re not careful], we'll start to see a reversal of the progress we’ve made.

Have other assays shown the same predictive value as Oncotype DX?

Other assays have been developed, including the Breast Cancer Index and the PAM50 (Prosigna) assay, that appear to predict late recurrence. Ongoing studies are examining whether or not those approaches can help us better determine who needs 10 years of treatment versus who can get away with 5 years.

Data have demonstrated a benefit with 10 years of tamoxifen. What about 10 years of an AI?

That's an excellent question. We know that if a patient can take tamoxifen for 5 years and then switch to an AI for another 5 years, there's a benefit, compared with taking tamoxifen for 5 years. Some data suggest that taking letrozole for 10 years is better than taking letrozole for 5 years. Because the risks in these patient populations are low, we look at invasive disease-free survival (iDFS); that can mean a new breast cancer as opposed to a distant recurrence. We know there's a benefit in iDFS, but we don't know if there's a benefit in terms of overall survival.

How is clinical judgment still being used?

You order a test to help with difficult decision making. Some patients have a low-grade cancer and a very small tumor size; under no circumstance should they receive chemotherapy. Other patients come in with very poor clinical risk; they have large, high-grade cancers, lymphovascular space invasion, cancers in more than 1 spot in the breast. These patients don't really represent the patients who went on the clinical trial. Even if you get those scores [that recommend a certain therapy], that may not be the appropriate choice for that patient. We need to use our judgment, and we need to apply the results to the patients who are most like those [who were enrolled] in the trial.

Are any preventative strategies in place that can help reduce the toxicities of extended endocrine therapy?

Usually patients who complete 5 years of therapy tolerate it. Many other questions arise for patients depending on where they are in their life course, whether it's a very young woman who is thinking of having children or has issues with pregnancy versus someone who has been on these medications and has had no AEs. Usually, we try to get patients through 5 years [of treatment] because that's hard enough. Once patients have completed 5 years of therapy, we do an assessment and ask, “What are the AEs of the treatment? Can you live with those? What are the various benefits [of therapy] depending on your risk? Can you live with another 5 years of treatment?”

Many women believe that when they get through the chemotherapy, the hard part is over. It’s important to prepare them [for the reality] that endocrine therapy can cause a lot of AEs, but there are also ways of overcoming those toxicities. [You should relay to patients the] different ways of managing each of those AEs, whether it's taking antidepressants for depression or cognitive behavioral therapy for sleep, acupuncture for joint pain, or other medications to reduce hot flashes. There are options that people can [handle]. It’s important to reach out to try to solve those issues so patients can stay on their medication.

Are risk calculators standard in clinical practice?

You can look at the clinical trials, and they go through each of the categories that one can look at. An app is being developed that people should be able to download soon. That will calculate a patient’s risk based on the SOFT and TEXT trials.

Pagani O, Francis PA, Fleming GF, et al. Absolute improvements in freedom from distant recurrence to tailor adjuvant endocrine therapies for premenopausal women: results from TEXT and SOFT [published online October 16, 2019]. J Clin Oncol. doi: 10.1200/JCO.18.01967.