Extended AI therapy reduces breast cancer recurrence risk, ups fracture risk

SAN ANTONIO – Extending aromatase inhibitor (AI) therapy an additional 5 years reduces breast cancer recurrence risk, particularly in patients with node involvement, but the benefits vary based on prior treatment and must be weighed against the risk of bone fracture, according to findings from a meta-analysis involving more than 22,000 women.

The rate of any recurrence after 10 years in almost 7,500 women treated with 5 years of tamoxifen and then randomized to 5 additional years of AI treatment was reduced by 35%, compared with the rate in those who did not receive 5 additional years of AI therapy (recurrence rate, 10.7% vs. 7.1%, respectively; relative risk, 0.67), and the difference was “very highly significant,” Richard Gray, MD, reported at the San Antonio Breast Cancer Symposium.

The distant recurrence rate and mortality rate were also significantly improved in those who received 5 years of AI therapy (rr, 0.77 for each), but the difference in mortality was of borderline significance, Dr. Gray, professor of medical statistics at the University of Oxford, London, reported on behalf of the Early Breast Cancer Trialists’ Collaborative Group.

However, in many of the trials included in the analysis, control group patients crossed over to the treatment group, which likely reduced the effect, he noted.

In about 12,000 women who were treated with 2-3 years of tamoxifen followed by 2-3 years of an AI and who were then randomized to an additional 3-5 years of AI therapy, the effects were less pronounced, with about a 20% reduced risk of any recurrence after 10 years vs. the rate in those without extended therapy (recurrence rate, 9.2% vs. 7.1%), he said.

The differences in the rates of distant recurrence and breast cancer mortality were not statistically significant in this group, but again, follow-up was short, he said.

Similarly, in about 3,300 women treated with an AI followed by an additional 5 years of AI therapy, recurrence risk was reduced by about 25% vs. the rate in those who did not receive extended therapy, and no difference was seen in the rates of distant recurrence or breast cancer mortality.

Of note, the benefits in those who received tamoxifen were seen immediately, whereas the benefits in those receiving AIs in the first 5 years emerged after about 2 years of extended therapy, Dr. Gray said, explaining that this was likely due to “carry-over benefits” of the earlier AI therapy.

The downside with extended AI therapy was a 25% increase in fracture risk, as well as bone pain, which can reduce quality of life.

Therefore, decisions about extended therapy should involve careful risk-benefit analyses, he said, adding that the findings of this meta-analysis of 12 trials, which included postmenopausal women – 99% of whom had estrogen receptor–positive disease – provide “the most reliable assessment of the available evidence ... [for] guiding decisions about endocrine therapy.”

In this video interview, he further discussed the details and limitations of the study, the effects of nodal status on outcomes, implications of the findings for clinical practice, the need for further follow-up on all of the studies included in the analysis, and plans for incorporating new data from the AERAS trial, which were also presented at the symposium and which complement and reinforce the current findings.

Dr. Gray reported having no disclosures.

SOURCE: Gray R et al., SABCS 2018: Abstract GS3-03.

SAN ANTONIO – Extending aromatase inhibitor (AI) therapy an additional 5 years reduces breast cancer recurrence risk, particularly in patients with node involvement, but the benefits vary based on prior treatment and must be weighed against the risk of bone fracture, according to findings from a meta-analysis involving more than 22,000 women.

The rate of any recurrence after 10 years in almost 7,500 women treated with 5 years of tamoxifen and then randomized to 5 additional years of AI treatment was reduced by 35%, compared with the rate in those who did not receive 5 additional years of AI therapy (recurrence rate, 10.7% vs. 7.1%, respectively; relative risk, 0.67), and the difference was “very highly significant,” Richard Gray, MD, reported at the San Antonio Breast Cancer Symposium.

The distant recurrence rate and mortality rate were also significantly improved in those who received 5 years of AI therapy (rr, 0.77 for each), but the difference in mortality was of borderline significance, Dr. Gray, professor of medical statistics at the University of Oxford, London, reported on behalf of the Early Breast Cancer Trialists’ Collaborative Group.

However, in many of the trials included in the analysis, control group patients crossed over to the treatment group, which likely reduced the effect, he noted.

In about 12,000 women who were treated with 2-3 years of tamoxifen followed by 2-3 years of an AI and who were then randomized to an additional 3-5 years of AI therapy, the effects were less pronounced, with about a 20% reduced risk of any recurrence after 10 years vs. the rate in those without extended therapy (recurrence rate, 9.2% vs. 7.1%), he said.

The differences in the rates of distant recurrence and breast cancer mortality were not statistically significant in this group, but again, follow-up was short, he said.

Similarly, in about 3,300 women treated with an AI followed by an additional 5 years of AI therapy, recurrence risk was reduced by about 25% vs. the rate in those who did not receive extended therapy, and no difference was seen in the rates of distant recurrence or breast cancer mortality.

Of note, the benefits in those who received tamoxifen were seen immediately, whereas the benefits in those receiving AIs in the first 5 years emerged after about 2 years of extended therapy, Dr. Gray said, explaining that this was likely due to “carry-over benefits” of the earlier AI therapy.

The downside with extended AI therapy was a 25% increase in fracture risk, as well as bone pain, which can reduce quality of life.

Therefore, decisions about extended therapy should involve careful risk-benefit analyses, he said, adding that the findings of this meta-analysis of 12 trials, which included postmenopausal women – 99% of whom had estrogen receptor–positive disease – provide “the most reliable assessment of the available evidence ... [for] guiding decisions about endocrine therapy.”

In this video interview, he further discussed the details and limitations of the study, the effects of nodal status on outcomes, implications of the findings for clinical practice, the need for further follow-up on all of the studies included in the analysis, and plans for incorporating new data from the AERAS trial, which were also presented at the symposium and which complement and reinforce the current findings.

Dr. Gray reported having no disclosures.

SOURCE: Gray R et al., SABCS 2018: Abstract GS3-03.

SAN ANTONIO – Extending aromatase inhibitor (AI) therapy an additional 5 years reduces breast cancer recurrence risk, particularly in patients with node involvement, but the benefits vary based on prior treatment and must be weighed against the risk of bone fracture, according to findings from a meta-analysis involving more than 22,000 women.

The rate of any recurrence after 10 years in almost 7,500 women treated with 5 years of tamoxifen and then randomized to 5 additional years of AI treatment was reduced by 35%, compared with the rate in those who did not receive 5 additional years of AI therapy (recurrence rate, 10.7% vs. 7.1%, respectively; relative risk, 0.67), and the difference was “very highly significant,” Richard Gray, MD, reported at the San Antonio Breast Cancer Symposium.

The distant recurrence rate and mortality rate were also significantly improved in those who received 5 years of AI therapy (rr, 0.77 for each), but the difference in mortality was of borderline significance, Dr. Gray, professor of medical statistics at the University of Oxford, London, reported on behalf of the Early Breast Cancer Trialists’ Collaborative Group.

However, in many of the trials included in the analysis, control group patients crossed over to the treatment group, which likely reduced the effect, he noted.

In about 12,000 women who were treated with 2-3 years of tamoxifen followed by 2-3 years of an AI and who were then randomized to an additional 3-5 years of AI therapy, the effects were less pronounced, with about a 20% reduced risk of any recurrence after 10 years vs. the rate in those without extended therapy (recurrence rate, 9.2% vs. 7.1%), he said.

The differences in the rates of distant recurrence and breast cancer mortality were not statistically significant in this group, but again, follow-up was short, he said.

Similarly, in about 3,300 women treated with an AI followed by an additional 5 years of AI therapy, recurrence risk was reduced by about 25% vs. the rate in those who did not receive extended therapy, and no difference was seen in the rates of distant recurrence or breast cancer mortality.

Of note, the benefits in those who received tamoxifen were seen immediately, whereas the benefits in those receiving AIs in the first 5 years emerged after about 2 years of extended therapy, Dr. Gray said, explaining that this was likely due to “carry-over benefits” of the earlier AI therapy.

The downside with extended AI therapy was a 25% increase in fracture risk, as well as bone pain, which can reduce quality of life.

Therefore, decisions about extended therapy should involve careful risk-benefit analyses, he said, adding that the findings of this meta-analysis of 12 trials, which included postmenopausal women – 99% of whom had estrogen receptor–positive disease – provide “the most reliable assessment of the available evidence ... [for] guiding decisions about endocrine therapy.”

In this video interview, he further discussed the details and limitations of the study, the effects of nodal status on outcomes, implications of the findings for clinical practice, the need for further follow-up on all of the studies included in the analysis, and plans for incorporating new data from the AERAS trial, which were also presented at the symposium and which complement and reinforce the current findings.

Dr. Gray reported having no disclosures.

SOURCE: Gray R et al., SABCS 2018: Abstract GS3-03.