Risk–benefit ratio requires ‘highly individualized’ plan for extended aromatase inhibitor therapy

Approximately 70% of breast cancers are hormone receptor positive.

Risk for recurrence among this population can persist as long as 25 years after diagnosis.

Five years of adjuvant endocrine therapy, or hormone therapy, can reduce recurrence risk by about 50% and mortality risk by 30%. However, the optimal duration of therapy, especially beyond 5 years, has been the subject of large-scale trials over the last 2 decades.

Results of the ATLAS trial — published in The Lancet — showed extending tamoxifen therapy to 10 years reduced risk for breast cancer recurrence, as well as disease-specific and overall mortality.

However, the ideal length of treatment with aromatase inhibitors such as anastrozole and letrozole — often used instead of or after tamoxifen to stop estrogen production in postmenopausal women — remains controversial.

The MA.17R trial — results of which appeared last year in The New England Journal of Medicine — showed 10 years of therapy with letrozole reduced local recurrences and contralateral breast cancers compared with 5 years. However, the difference in distant recurrences was modest, suggesting the impact on survival may be negligible.

Subsequent studies — several of which were presented in December at San Antonio Breast Cancer Symposium (SABCS) — showed less benefit from extended aromatase inhibitor therapy. Also, adverse events and lower compliance occur more frequently among those assigned extended therapy.

“We are left with equivocal data as to what to do,” Debu Tripathy, MD, chair of breast medical oncology at The University of Texas MD Anderson Cancer Center and a HemOnc Today Editorial Board member, said in an interview.

“Providers will use different criteria to choose whether extended therapy with aromatase inhibitors is best for their patients,” Tripathy said. “Although lowering of overall events is apparent, the impact on metastases and mortality is questionable. Local recurrences and contralateral breast cancers are important events, but they must be balanced against side effects — especially those that may be silent with a long natural history, such as lower bone density and fracture risk.”

HemOnc Today spoke with breast cancer specialists about the conflicting data related to extended aromatase inhibitor therapy, the risks for adverse events, the potential for low patient compliance and the need to identify which patients are most likely to benefit.

Conflicting data

Trials designed to evaluate extended aromatase inhibitor therapy have yielded variable results on benefits and harms.

Postmenopausal women with early breast cancer often undergo adjuvant treatment with 5 years of an aromatase inhibitor, either as upfront monotherapy or after tamoxifen.

In the randomized phase 3 MA.17R trial (n = 1,918), Goss and colleagues evaluated the effect of extending letrozole treatment for another 5 years.

Results showed extended therapy improved 5-year DFS (see Table) and reduced the annual incidence rate of contralateral breast cancer (0.21% vs. 0.49%; HR = 0.42; P = .007). However, researchers reported no significant differences in 5-year OS (93% vs. 94%; HR = 0.97), distant recurrence (4.4% vs. 5.5%) or quality of life.

“The most striking change in this study was the decrease in new breast cancers,” Douglas Yee, MD, director of Masonic Cancer Center, professor of medicine and pharmacology at University of Minnesota, and a HemOnc Today Editorial Board member, said in an interview. “It is an important finding, but from a medical oncology standpoint, what was a little less impressive was the lack of changes in distant recurrences. Breast cancer mortality is caused by distant metastases.”

The NRG Oncology/NSABP B-42 clinical trial — presented by Terry Mamounas, MD, MPH, and colleagues at SABCS — showed no survival benefit with extended aromatase inhibitor therapy.

The analysis included 3,923 postmenopausal women who completed 5 years of endocrine therapy. Women who received 5 additional years of letrozole had a 3% improvement in DFS (see Table), which did not reach the pre-defined level of significance. OS also was not significantly different at 7 years (HR = 0.72; 95% CI, 0.99-1.44).

However, at 7 years, extended letrozole conferred a 28% reduction in cumulative incidence of distant recurrence (HR = 0.72; 95% CI, 0.53-0.97) and a 29% reduction in cumulative incidence of breast cancer–free interval events (HR = 0.71; 95% CI, 0.56-0.89).

Due to the contradictory data, clinicians must evaluate the pros and cons of extended therapy for each patient.

“What is the risk for recurrence?” Mamounas told HemOnc Today when the study was presented. “What are significant comorbidities? What is the age of the patient? How did they tolerate the primary aromatase inhibitor for the first 5 years, and what is the bone mineral density? All of these factors figure into the potential decision of [whether to extend] therapy beyond 5 years.”

Two other phase 3 studies presented at SABCS evaluated extended aromatase inhibitor therapy.

In the DATA study, Tjan-Heihnen and colleagues measured adapted DFS — or DFS beyond 3 years after randomization to aromatase inhibitor therapy — among 1,912 postmenopausal women who received either 3 or 6 years of anastrozole therapy after completion of 2 to 3 years of adjuvant tamoxifen.

The DFS rate was again improved with extended anastrozole therapy group at 5 years (see Table), but it did not reach the study’s efficacy threshold (HR = 0.6). Researchers reported no difference in 5-year adapted OS between groups.

The IDEAL trial, conducted by Blok and colleagues, included 1,824 women with hormone receptor–positive breast cancer who underwent 5 years of endocrine therapy. Researchers randomly assigned study participants to an additional 2.5 years or 5 years of letrozole. This study also failed to show improvement in 5-year DFS with longer treatment (see Table).

Although none of these studies reached their primary efficacy endpoints, the NSABP B-42 and DATA trials showed trends toward improvements in DFS. However, the absolute benefit from extended endocrine therapy was small in all studies, Lajos Pusztai, MD, PhD, professor of medicine at Yale University and chief of the breast medical oncology section at Yale Cancer Center, told HemOnc Today.

“For example, in the NSABP B-42 trial, there were 26 fewer distant metastatic events,” he said. “This may be statistically insignificant — indeed the P value of .045 did not meet the primary endpoint — however, these 26 patients likely avoided death from breast cancer, since metastatic breast cancer remains an incurable disease. Results from these trials underscore the importance of patient selection for extended endocrine therapy, so that we do not need to treat 1,800 women to save 26 lives, as in the B-42 trial.”

Risk assessments

Individuals who undergo breast cancer treatment — including adjuvant endocrine therapy — can remain cancer free for many years, but they still remain at risk for recurrence or metastases.

“The risk for recurrence never goes to zero,” Tara Sanft, MD, assistant professor of medical oncology and medical director of adult survivorship for Connecticut Challenge Survivorship Clinic at Yale Cancer Center, told HemOnc Today. “We see women who are 5, 10 or even 20 years out from having breast cancer who may have recurrences.

“Traditionally, we identify high risk by the initial size of the tumor and number of lymph nodes involved, but the genetics of cancer can change,” Sanft added. “We are learning there are probably women with small tumors who are still at risk for recurrence. Better tests are needed to understand which patients are at highest risk.”

This potential for recurrence contributes to the dilemma among clinicians and patients about extended aromatase inhibitor therapy.

“The benefits are quite clear in the literature,” Sanft said. “Aromatase inhibitors do decrease the risk for recurrence in women with estrogen receptor–positive breast cancer. People widely accept these results as negative because they didn’t reach a statistically significant endpoint, but to just interpret it as that and move on is to miss out on the opportunity to help those at highest risk.”

Other clinicians contend there are not enough positive data to support the use of extended aromatase inhibitor therapy for all patients.

Photo credit: Karsten Moran.

“The benefit of extended aromatase inhibitor therapy is that it decreases the risk for recurrence in patients with hormone receptor–positive breast cancer, but it is a relatively small improvement suggesting that only a subset of patients will benefit from this approach,” Francisco J. Esteva, MD, PhD, director of the breast medical oncology program at NYU Langone Perlmutter Cancer Center, told HemOnc Today. “The MA.17R trial showed 10 years of aromatase inhibitor therapy was better than 5 years, but there were several studies presented at [SABCS] that drew some questions into the mix.”

“The NSABP B-42 clinical trial did not show much of a benefit, so it is still unclear who exactly will benefit from 10 or even 15 years of extended aromatase inhibitor therapy,” he added. “There are some patients who will benefit — because we do see recurrences at 10 and 15 years after diagnosis — but I do not think it is something that should be recommended for all patients.”

V.K. Gadi, MD, PhD, medical oncologist at Seattle Cancer Care Alliance and associate professor of medicine at University of Washington, agreed results from the DATA, IDEAL and NSABP B-42 trials do not suggest a sufficiently significant benefit.

“For those women who started tamoxifen for 5 years, data from the last several years firmly allow us to say there is benefit to continuing for another 5 years with either tamoxifen or, if postmenopausal, transitioning to an aromatase inhibitor,” Gadi said. “There is now limited-to-no statistically significant support for continuing aromatase inhibitor–based therapy beyond 5 years for the purpose of preventing recurrences and deaths for those patients who received some amount of aromatase inhibitor therapy in their first 5 years of endocrine therapy.”

Data show reduced recurrence risk primarily can be attributed to reductions in local tumors and contralateral breast cancer, not metastatic disease, some clinicians contend.

“Most recurrences prevented are not metastatic, but rather local recurrence of the breast — which are still curable with surgery — or totally new breast cancers in the other breast,” Tripathy said. “The important recurrences we want to prevent are distant recurrences or metastatic disease, because patients die of these. [These] were only lowered by 1% in the trials.”

No trial showed a decrease in mortality rate, which is “the most important endpoint,” Tripathy said.

Still, benefit can vary by patient. For example, in the DATA study, the adjusted DFS benefit was greater among patients with node-positive disease (HR = 0.71; 95% CI, 0.53-0.96), those with both ER– and PR–positive disease (HR = 0.68; 95% CI, 0.51-0.9, and among patients who were both ER and PR positive with node-positive disease who received chemotherapy (5-year DFS, 76% vs 86%; HR= 0.58; P = .01).

“The underlying risk for late recurrence and continued endocrine therapy sensitivity together determine if one observes a benefit from extended adjuvant therapy, or not,” Pusztai said. “Both clinical variables, like nodal status and tumor size, as well molecular assays such as the Breast Cancer Index [Biotheranostics], Prosigna [NanoString] or EndoPredict [Myriad Genetics] may help identify patients who may have greater-than-average benefit from extended endocrine therapy.”

Risks and compliance

Aromatase inhibitor therapy also is associated with adverse events, such as osteoporosis, fractures, bone and joint pain, vaginal dryness and hot flashes.

“Lower estrogen levels accelerate bone mineral loss, so the risk for osteoporosis can increase substantially,” Tripathy said. “Fracture rates are higher in patients who take aromatase inhibitors if you add them all up over 5 years. If patients are put on extended therapy, the fracture rate could double.”

There are ways to reduce risk for bone fracture and osteoporosis through surgery. However, clinicians must consider the risk–benefit ratio for each patient, Yee said.

“We understand through this therapy that we do some harm by increasing bone fracture risk, and this is clearly the downside,” he said. “If one woman has good bone density going into year 5 of therapy, they can think about continuing therapy because the medication did not have a detrimental effect.”

Extended aromatase inhibitor therapy is appropriate if the benefits outweigh the risks, Pusztai said.

“Clearly, if one has a 30% risk for metastatic recurrence vs. a 5% extra risk for bone fractures from extended endocrine therapy, the expected benefits of therapy would outweigh the risk,” Pusztai said. “However, what makes this decision difficult is that we never have precise risk estimates — this is where more objective molecular tests could help decision-making.”

Cardiovascular events also are a concern.

Anne H. Blaes, MD, MS, associate professor of medicine at University of Minnesota and a HemOnc Today Editorial Board member, presented a study at SABCS that showed postmenopausal women with locally advanced, curative-intent breast cancer prescribed an aromatase inhibitor had significantly worse endothelial function as measured by EndoPAT (median ratio, 0.8 vs. 2.7; P < .0001) — a predictor of cardiovascular disease — than healthy controls.

Better biomarkers are needed to determine if there is actual risk for cardiovascular events, Blaes said.

“Most of [the trials] that have been done so far actually look at end events, such as myocardial infarction,” Blaes told HemOnc Today. “Our study suggests that there are precursors.”

More than 90% of the 36 study participants had no history of heart disease, Blaes noted.

“We all feel very comfortable that aromatase inhibitors have a place in early-stage breast cancer, particularly in the first 5 years of treatment,” she said. “There is a lot of controversy about whether we should stay on prolonged endocrine therapy and, if so, which medication should we use and who is the most appropriate patient group? ... In that extended period, it is important that we weigh the risks and benefits. If there is no OS advantage and there is a potential for cardiac events, we really need to take that into consideration.”

Although this study and others have raised concerns, more data are needed, Gadi said.

“Large trials of extended therapy will likely need to encounter more follow-up for any cardiovascular events to emerge,” Gadi said. “Until this matures for analysis, doctors in the trenches should offer advice to patients in survivorship that emphasizes lifestyle, diet and exercise changes.”

The study by Blaes and colleagues used a sophisticated test that cardiologists frequently use to estimate long-term cardiovascular risk, Yee said.

“[Blaes and colleagues] found women on aromatase inhibitors had worse scores on this test, but that doesn’t necessarily mean there is increased risk,” he said. “We need to follow people for longer to sort it out, but this is always a question you should think about, too: Are we assessing cardiovascular risks the way we should?”

Study data do not confirm elevated risk for cardiovascular events, Tripathy said.

“The first studies that came out evaluated tamoxifen, which is actually protective against heart disease due to the lowering of cholesterol levels,” he said. “Cardiac events while on tamoxifen are slightly lower, so comparing aromatase inhibitors to tamoxifen, it was not so much that aromatase inhibitors were causing the risk, but that the tamoxifen had been lowering it. In prevention studies of placebo vs. aromatase inhibitors, there wasn’t an increase in cardiac risk.”

Still, concern about side effects may discourage patient compliance, thereby reducing the success of extended therapy.

“Compliance over 5 years of therapy is only about 50% to 60%. If you extend therapy to 10 years, another 10% to 20% of patients will discontinue their medication,” Tripathy said. “People may forget or they feel better one day, so they do not end up taking their medication. When you are first diagnosed, you are scared and follow everything by the book, but that can change over time, especially over 10 years.”

A study by Sanft and colleagues, published in 2015 in Breast Cancer Research and Treatment, showed 69% of patients with a history of ER–positive, stage I to stage III breast cancer who received endocrine therapy for at least 3.5 years were more likely to be compliant on extended aromatase inhibitor therapy after they received their Breast Cancer Index score. The test predicts the risk for node-negative, HR–positive breast cancer recurrence 5 to 10 years after diagnosis.

“These medications are not easy to take,” Sanft said. “After results of this test, patients had less anxiety about their decisions and whether they would take medication. They experienced a greater confidence in their decision and were more likely to be compliant.”

‘The right patients’

The real reason clinical trials have yet to reflect a greater benefit with extended aromatase inhibitor therapy may be because recurrences often are endocrine resistant, Gadi said.

“As we evaluate adjuvant treatment with agents such as mTOR and CDK4/6 inhibitors in large clinical trials, it will be interesting to note if late recurrences also will be impacted, [as] these agents may overcome endocrine resistance,” Gadi said. “Late recurrences may arise from dormant populations [of cancer cells], which may genetically be distinct from the actual treated breast tumors.”

Patients begin extended aromatase therapy due to the failure of initial treatment to eliminate dormant tumor cells, Yee said.

“I would like us to evolve in this area [to a point] where we can actually kill those dormant cells,” he said. “Breast cancer does not have that many antigens toward which response can be mounted, but we can try to increase immunity that will target dormant cells so we do not have to treat women for 5 or 10 years down the road.”

Until therapies more effectively kill dormant breast cancer cells, clinicians continue to rely on assays to identify high-risk patients who can benefit from extended aromatase inhibitor therapy.

“Perhaps we should be reserving [endocrine therapy] for our higher-risk patients, but the studies do not point us to exactly who it should be,” Tripathy said. “It will take a little more time to analyze studies and pool them together to get better results. I suspect, in the coming years, we will hear about more follow-up and will get more clarity.”

The Breast Cancer Index is the only assay available that characterizes a patient’s risk for late recurrence.

“This is where a lot of decision-making comes in,” Sanft said. “We need higher uptake and greater understanding of how this impacts recurrence rates.”

ASCO guidelines endorse the use of extended aromatase inhibitor therapy, but they do not specify how to select molecular-risk testing of specific patients for extended therapy, Pusztai said.

“The testing guidelines will likely change as more data accumulate supporting the use, value and cost-effectiveness of molecular assays to identify high-risk patients for whom the risk–benefit ratio is favorable,” he said.

The availability of more tests could facilitate the “highly individualized” decision to undergo extended aromatase inhibitor therapy, Gadi said.

“Tools to better identify which recurrences we can actually prevent would be welcome,” Gadi said. “I look forward to seeing analyses including assays such as MammaPrint [Agendia] or Breast Cancer Index to better counsel patients for considering extended therapy.”

Designing prospective studies to follow thousands of patients for 10 to 15 years is no longer practical, Esteva said.

“We might be able to use molecular tests as predictive markers to identify patients who are more or less likely to relapse in 10 years or beyond,” Esteva said. “Retrospective studies showed that certain tests may be able to predict late relapses. Additional studies are needed to determine which molecular tests are clinically useful for selecting the right patients for extended endocrine therapy.” – by Melinda Stevens

Click here to read the , “Does the efficacy of bisphosphonates support the use of extended aromatase inhibitor therapy in women at risk for bone fractures?”

References:

Colleoni M, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2015.62.3504.

Davies C, et al. Lancet. 2013;381:805-816.

Goss PE, et al. N Engl J Med. 2016;doi:10.1056/NEJMoa1604700.

Rugo HS, et al. J Clin Oncol. 2016;doi:10.1200/JCO.2016.67.1487.

Sanft T, et al. Breast Cancer Res Treat. 2015;doi:10.1007/s10549-015-3631-9.

The following were presented at San Antonio Breast Cancer Symposium; Dec. 6-10, 2016; San Antonio, Texas:

Blaes AH, et al. Abstract S5-07.

Blok EJ, et al. Abstract S1-04.

Mamounas P, et al. Abstract S1-05.

Sanft T, et al. Abstract P2-09-15.

Tjan-Heijnen VC, et al. Abstract S1-03.

For more information:

Francisco J. Esteva, MD, PhD, can be reached at 160 East 34th St. 4th Floor, New York, NY 10016.

V.K. Gadi, MD, PhD, can be reached at Seattle Cancer Care Alliance, 825 Eastlake Ave. E, G3-630, Seattle, WA 98109-1023.

Lajos Pusztai, MD, PhD, can be reached at lajos.pusztai@yale.edu.

Tara Sanft, MD, can be reached at tara.sanft@yale.edu.

Debu Tripathy, MD, can be reached at dtripathy@mdanderson.org.

Douglas Yee, MD, can be reached at yeexx006@umn.edu.

Disclosure: Pusztai reports consultant roles with Celgene and Clovis Oncology, honoraria from Biotheranostics and Pfizer, and research funding from Foundation Medicine, Genentech and Merck. Sanft reports institutional funding from Biotheranostics. Tripathy reports a consultant role with and research support from Novartis. Esteva, Gadi and Yee report no relevant financial disclosures.