Eleftherios (Terry) P. Mamounas, MD, discusses whether 10 years was superior to the widely practiced 5 years of adjuvant hormone therapy for patients with breast cancer.
Terry Mamounas, MD
Recent findings have suggested that 10 years versus 5 years of adjuvant hormone therapy reduces recurrence in women with primary breast cancer; however, the topic has been widely debated among researchers.
For example, results of the NSABP B-42 clinical trial, which were presented at the 2016 San Antonio Breast Cancer Symposium, found that extending letrozole therapy in women with early-stage, hormone receptor (HR)—positive breast cancer who completed 5 years of prior hormone therapy did not yield a statistically significant improvement in either disease-free or overall survival. However, data also showed that prolonged use of the aromatase inhibitor may be beneficial in some subgroups of women with a higher risk of recurrence.
In his presentation on duration of adjuvant endocrine therapy at the PER® 16th Annual International Congress on the Future of Breast Cancer® (East), Eleftherios (Terry) P. Mamounas, MD, discussed whether 10 years was superior to the widely practiced 5 years of adjuvant hormone therapy.
In an interview with OncLive during the meeting, Mamounas, medical director of the Comprehensive Breast Program at the University of Florida Health Cancer Center, shared his insight on recent studies of adjuvant endocrine therapy and their impact on clinicians’ choice in therapy for pre- and postmenopausal women with primary breast cancer.Mamounas: As you know, for many years, the thinking was that 5 years of tamoxifen was standard, because many clinical trials have shown that there is a benefit. However, after 5 years, the early studies showed that additional therapy up to 10 years was not better. Therefore, 1 to 5 years was considered the standard.
The European investigators thought that perhaps longer duration of tamoxifen would be important, so they ran these big trials—the ATLAS and aTTom trials. In the early publication, it showed no significant benefit in the first 5 years, but then with longer follow-up to 15 years, both of the studies showed a significant improvement in disease-free survival (DFS) with 10 years of tamoxifen. ATLAS also saw a significant improvement in survival. After that data were disclosed, the guidelines changed and 10 years of tamoxifen became the standard for patients who are premenopausal.
At the same time, we had introduced aromatase inhibitors, which is a whole different class of drugs with lower estrogen levels for postmenopausal patients. We also had data to indicate that if you take 5 years of tamoxifen and then 5 years of an aromatase inhibitor, the outcomes are better. For patients who already are or become postmenopausal—after 5 years of tamoxifen—most will be treated with an aromatase inhibitor rather than go on to 10 years of tamoxifen, because that has shown benefit.
The last question that we had to address was, “If you started with an aromatase inhibitor, would 10 years of the aromatase inhibitor be better than 5 years?” This was addressed recently in 2 studies—one of which is the MA.17R study that was presented that the 2016 Annual ASCO Meeting. [Regarding the] significant benefit of when you take an aromatase inhibitor for 10 years versus 5 years, the majority of those patients have had tamoxifen for 5 years because they were participating in a clinical trial.
At the 2016 San Antonio Breast Cancer Symposium, we also presented the results of the NSABP B-42 trial that recruited almost 4000 patients with postmenopausal stage I to III breast cancer. They were randomized to 5 years versus 10 years of an aromatase inhibitor-based therapy, such as letrozole. Most of these patients have had 5 years of an aromatase inhibitor previously or tamoxifen followed by an aromatase inhibitor for 5 years. In that study, we showed a borderline improvement; it did not formally meet statistical significance by definition, though the P value is .048.
Therefore, they were a little bit of discordant results—1 very positive study and 1 that showed a borderline significant benefit. As I explained in my lecture, the endpoints that these 2 studies used—the DFS used in the MA.17R trial included only recurrences of contralateral breast cancer. Our definition of DFS in NSABP B-42 also included deaths from other causes and nonbreast malignancies, which is a more traditional definition of DFS. When the MA.17R trial looked at the same endpoint without the nonbreast malignancies or deaths from other causes, then the difference was not statistically significant; it was about a 20% reduction.
These studies are actually more similar. If you look at the overall DFS endpoint and then get a 15% to 20% reduction, it is borderline or not statistically significant. However, if, biologically, you only look at the recurrence of contralateral breast cancer, then you get a significant reduction. We also showed that if you get a significant reduction in distant recurrence, it is in the range of about 28%.
Therefore, biologically, extended adjuvant therapy works. The big question is, clinically, “Does this make a big difference?” Some of these patients are older and have comorbidities. Some die from other causes or complications, potentially from the treatment of long-term estrogen deprivation. You actually have to be very careful of who you select for adjuvant therapy.
That is part of our effort now, to identify additional factors. It is not only clinical or pathologic factors, but perhaps genomic classifiers that put patients at a higher or lower risk for late recurrence, increasing or decreasing the rationale to give them endocrine therapy, and more importantly, to find factors that predict benefit from extended adjuvant endocrine therapy. There may be subgroups of patients who have a bigger benefit and others with very little benefit.
However, up until that point, we triage patients based on their clinical tumor sign, pathologic nodal status, and how they tolerated the aromatase inhibitor. If a patient is begging to get off of treatment, it is hard to document. In this case, you can actually persist with tamoxifen if you want extended therapy; that strategy has not been tested a lot but has been shown to actually work. We are becoming more selective, and hopefully more research will shed some more light into this.It is a problem because we know that some patients do not respond. We have gotten closer now to finding out who these patients are, either by running tests before including genomic profiling tests. For example, if you do the 21-Gene Recurrence Score assay, and you see that the patients have a high recurrence score, the proportion of these patients do not seem to benefit from endocrine therapy. They may be sensitive to chemotherapy if they have a high recurrent score, but they appear to be resistant to endocrine therapy.
The other way to do it is an in vivo way, meaning we give them neoadjuvant endocrine therapy and see who can decrease proliferation at an early time point—it is usually about 4 to 10 weeks of endocrine therapy. If you see the Ki-67 proliferation going down, then you know they are sensitive to endocrine therapy. If it is not, then they are usually resistant to endocrine therapy, and then you have to change your strategies.
In the metastatic setting, it becomes a little more complicated because people tend to respond to endocrine therapy and then eventually become resistant. You can actually change the hormonal therapy and then they respond to another endocrine therapy. That usually happens 2 to 3 times, and eventually the patients become truly resistant to endocrine therapy. At that point, it is time to switch to chemotherapy-based regimens. Nowadays, with drugs such as CDK 4/6 and mTOR inhibitors, it is believed that you can sensitize patients who are resistant to endocrine therapy to actually become sensitive again by inhibiting these additional pathways.
For HER2-positive patients, who are technically more resistant to endocrine therapy, another approach would be to give them anti-HER2 therapy. You shut down the HER2 pathway, so you may actually sensitize them to endocrine therapy. There are ways of circumventing this endocrine resistance.We think that genomic profiling can help. We have data from at least 4 or 5 assays which have shown predictors of late recurrence. Again, the “holy grail” is whether they can predict a benefit with extended endocrine therapy. That may be a different pathway all together.
One of these assays showed that, in the MA.17R trial, there was an interaction in terms of predictive endocrine therapy benefit. If there was a high homeobox 13 and interleukin-17B receptor ratio, those patients tended to have higher benefit from endocrine therapy. We would like to validate this in another big data set; we are looking into doing that in the NSABP B-42 trial as well. The take-home message is that we have to be selective in terms of who we recommend extended adjuvant endocrine therapy to when it comes to the aromatase inhibitors. When it comes to tamoxifen, when patients are premenopausal, 10 years of tamoxifen is the arguable standard.
However, for the postmenopausal patients, starting with an aromatase inhibitor will need to be selective. This is not because the treatment doesn’t work well, but because of additional comorbidities, as most of these aromatase inhibitor trials haven’t shown any survival benefit. At the end of the day, we don’t stop people from dying of breast cancer by giving this extended adjuvant therapy. Obviously more follow-up is needed, but, certainly, the benefit does not seem to be huge at this point. We have to have a discussion with our patients and make a joint decision on what to do.
Mamounas EP, Bandos H, Lembersky BC, et al. A randomized, double-blinded, placebo-controlled clinical trial to evaluate extended adjuvant endocrine therapy (5 years of letrozole) in postmenopausal women with hormone-receptor positive breast cancer who have completed previous adjuvant endocrine therapy: Initial results of NRG oncology/NSABP B-42. Presented at: 2016 San Antonio Breast Cancer Symposium; San Antonio, TX; December 6-10, 2016. Abstract S1-05.