AT ACOG 2017
SAN DIEGO – Continuation of tamoxifen for an additional 5 years is a cost-effective strategy that does not increase all-cause mortality for premenopausal women with estrogen receptor–positive breast cancer, based on an analysis using sophisticated computational modeling techniques.
“For premenopausal women with an early estrogen receptor–positive breast cancer who have completed 5 years of tamoxifen as initial treatment, another 5 years of tamoxifen is preferable to ovarian ablation with an aromatase inhibitor as extended endocrine treatment,” Janice Kwon, MD, said at the annual meeting of the American College of Obstetricians and Gynecologists.
The researchers sought to answer a key clinical question: “What is the optimal endocrine strategy for premenopausal women who have completed 5 years of tamoxifen? Another 5 years of tamoxifen? An aromatase inhibitor preceded by ovarian ablation? Or no further treatment?”
Dr. Kwon and her coinvestigators used a Markov Monte Carlo simulation to project adverse events that would occur with each of the three treatments in a hypothetical cohort of 18,000 premenopausal women with estrogen receptor–positive breast cancer. They also conducted sensitivity analyses to ascertain the point at which a given treatment would become cost effective. The investigators used a time horizon of 40 years in the Monte Carlo simulation, which uses repeated random sampling of a large data set to model the probability of a variety of outcomes. The primary outcome measure used to compare the three treatment strategies was the incremental cost-effectiveness ratio (ICER).
For the no further treatment strategy, the average costs were $1,074, for an average life expectancy gain of 16.69 years. Compared with this strategy, 5 more years of tamoxifen would cost $3,550 for an average life expectancy gain of 17.31 years, yielding an ICER of $4,042. The strategy of performing a bilateral salpingo-oophorectomy (BSO), followed by 5 years of aromatase inhibitor therapy, was more costly at $14,312 and yielded a shorter life expectancy gain at an average of 17.06 years, eliminating it as a feasible strategy in the ICER analysis.
Using the Monte Carlo simulation to assess treatment-related mortality, Dr. Kwon and her colleagues found that no further treatment would result in the most deaths from breast cancer, at 7,358. For this, and each of the other two strategies, the investigators also modeled deaths from other causes and from early BSO, using the Nurses’ Health Study hazard ratios. No further treatment would result in 5,878 deaths from other causes and none from early BSO, for a total of 13,236.
Another 5 years of tamoxifen, the model showed, would result in 6,227 deaths from breast cancer, 6,330 from other causes, and none from BSO, for a total of 12,557.
The BSO–aromatase inhibitor strategy was modeled to have the fewest deaths from breast cancer (5,504) and from other causes (5,834) but would result in an additional 1,897 deaths from the early BSO. The BSO–aromatase inhibitor strategy thus resulted in a virtually identical number of deaths over a 40-year period as no treatment at all, at 13,235.
An aromatase inhibitor is frequently considered as a treatment strategy for women with estrogen receptor–positive breast cancer. However, using an aromatase inhibitor is predicated on the patient being menopausal, so ovarian ablation is recommended for patients who have, or who may have, intact ovarian function.
Nearly 3 decades’ worth of data from the Nurses’ Health Study showed an overall hazard ratio of 1.41 for premenopausal oophorectomy without hormone therapy, said Dr. Kwon of the gynecologic oncology division at the University of British Columbia, Vancouver. Increased rates of osteoporosis, stroke, and coronary heart disease contributed to the increased risk, with 80% of the excess deaths occurring within 15 years of oophorectomy. The analysis yielded a number needed to harm for the procedure of eight.
The study’s results have also been substantiated by a recent meta-analysis, said Dr. Kwon, that also saw “fewer disease-free events but more deaths with aromatase inhibitor versus tamoxifen” (Breast Cancer Res Treat. 2017;161:185-90). However, she said, the long-term outcomes of breast cancer over many decades are unknown, and the analysis did not include costs for treatment of recurrent breast cancer.
No external funding sources were reported, and Dr. Kwon reported having no relevant financial disclosures.
On Twitter @karioakes