Taryn Bessen1, Dorothy M K Keefe2, Jonathan Karnon3. 1. Department of Medical Imaging,Royal Adelaide HospitalTaryn.Bessen@health.sa.gov.au. 2. SA Cancer Service;Department of Cancer Medicine,University of Adelaide;Sansom Institute,University of South Australia;Royal Adelaide Hospital. 3. School of Population Health,University of Adelaide.
Abstract
OBJECTIVES: International guidelines recommend annual mammography after early breast cancer, but there is no randomized controlled trial evidence to support this schedule over any other. Given that not all women have the same risk of recurrence, it is possible that, by defining different risk profiles, we could tailor mammographic schedules that are more effective and efficient. METHODS: A discrete event simulation model was developed to describe the progression of early breast cancer after completion of primary treatment. Retrospective data for 1,100 postmenopausal women diagnosed with early breast cancer in South Australia from 2000 to 2008 were used to calibrate the model. Women were divided into four prognostic subgroups based on the Nottingham Prognostic Index of their primary tumor. For each subgroup, we compared the cost-effectiveness of three different mammographic schedules for two different age groups. RESULTS: Annual mammographic follow-up was not cost-effective for most postmenopausal women. Two yearly mammography was cost-effective for all women with excellent prognosis tumors; and for women with good prognosis tumors if high compliance rates can be achieved. Annual mammography for 5 years and 2 yearly surveillance thereafter (a mixed schedule) may be cost-effective for 50- to 69-year-old women with moderate prognosis tumors, and for women aged 70-79 years with poor prognosis tumors. For younger women with poor prognosis tumors, annual mammography is potentially cost-effective. CONCLUSIONS: Our results suggest that mammographic follow-up could be tailored according to risk of recurrence. If validated with larger datasets, this could potentially set the stage for personalized mammographic follow-up after breast cancer.
OBJECTIVES: International guidelines recommend annual mammography after early breast cancer, but there is no randomized controlled trial evidence to support this schedule over any other. Given that not all women have the same risk of recurrence, it is possible that, by defining different risk profiles, we could tailor mammographic schedules that are more effective and efficient. METHODS: A discrete event simulation model was developed to describe the progression of early breast cancer after completion of primary treatment. Retrospective data for 1,100 postmenopausal women diagnosed with early breast cancer in South Australia from 2000 to 2008 were used to calibrate the model. Women were divided into four prognostic subgroups based on the Nottingham Prognostic Index of their primary tumor. For each subgroup, we compared the cost-effectiveness of three different mammographic schedules for two different age groups. RESULTS: Annual mammographic follow-up was not cost-effective for most postmenopausal women. Two yearly mammography was cost-effective for all women with excellent prognosis tumors; and for women with good prognosis tumors if high compliance rates can be achieved. Annual mammography for 5 years and 2 yearly surveillance thereafter (a mixed schedule) may be cost-effective for 50- to 69-year-old women with moderate prognosis tumors, and for women aged 70-79 years with poor prognosis tumors. For younger women with poor prognosis tumors, annual mammography is potentially cost-effective. CONCLUSIONS: Our results suggest that mammographic follow-up could be tailored according to risk of recurrence. If validated with larger datasets, this could potentially set the stage for personalized mammographic follow-up after breast cancer.
Entities:
Keywords:
Cost-effectiveness; Early breast cancer; Follow-up; Mammography
Authors: Koen Degeling; Hendrik Koffijberg; Mira D Franken; Miriam Koopman; Maarten J IJzerman Journal: Med Decis Making Date: 2019-01 Impact factor: 2.583