Gold H, Dick MA, Griggs J; Academy for Health Services Research and Health Policy. Meeting.
Abstr Acad Health Serv Res Health Policy Meet. 2002; 19: 7.
University of Rochester School of Medicine & Dentistry, 227 North College Street, Carlisle, PA 17013; Tel: (717) 245-0341; E-mail: heather_gold@urmc.rochester.edu
RESEARCH OBJECTIVE: To evaluate the relative outcomes of treatment decisions for ductal carcinoma in situ (DCIS) on recurrence rates to DCIS and invasive breast cancer. No prior population-based or clinical studies compare three alternative treatments of breast-conserving surgery (BCS), BCS with radiotherapy, and mastectomy, yet the incidence of DCIS has increased dramatically since the 1980s, with concurrent shifts in practice patterns toward the use of less-extensive treatment. DCIS is not associated with mortality unless it is followed by invasive breast cancer. Therefore, the reduction in the probability of recurrence via effective treatment is of importance for women diagnosed with this disease. STUDY DESIGN: Retrospective, observational, population-based econometric survival analysis of women diagnosed with DCIS. The analysis is comprised of three parts: 1) summary statistics of outcomes by treatment type and empirical (Kaplan-Meier) hazard and survivor function estimates by treatment type; 2) investigation of the comparability of women undergoing each treatment type; and 3) multivariate survival models to eliminate the confounding effects of population differences on outcomes. POPULATION STUDIED: Women ages 65 and over with a first diagnosis of unilateral DCIS according to the linked SEER-Medicare database from 1991-1996. The final sample includes 2,223 women who were treated surgically based on Medicare fee-for-service claims data. Women could have a recurrence as early as 6 months post-treatment. PRINCIPAL FINDINGS: Of the 2,223 women in the sample, 950 (43%) had BCS alone, 459 (21%) had BCS with radiotherapy (BCS-XRT), and 814 (37%) had mastectomy. Of women receiving BCS-XRT, 78% were below age 75, compared to 59% undergoing BCS alone and 65% receiving mastectomy, making the BCS-XRT group the youngest (p=0.000 for difference in ages). From 1991 through 1996, there is an increased use of BCS alone from 41% in the first half of the study to 44% in the second. The risk of subsequent DCIS or invasive breast cancer (ipsilateral or contralateral) is lowest in women undergoing mastectomy, followed by BCS-XRT and then BCS alone. After controlling for demographic and socioeconomic covariates, multivariate models show that BCS alone has a 1-year recurrence-free rate of 95.4%, compared to BCS-XRT at 98.5% and mastectomy at 98.6%. The 5-year recurrence-free rate is 87.6% for BCS alone, 93.8% for BCS-XRT, and 96.2% for mastectomy. CONCLUSIONS: Many factors are involved in the initial treatment choice for DCIS, but because BCS with radiotherapy and mastectomy have lower recurrence rates than BCS alone, they should be considered the best treatments to decrease the risk of local recurrence. Future studies should account for patient preferences, cost, and utilization differences and tumor characteristics not available in the SEER-Medicare data, including lesion size, pathology, and margin. IMPLICATIONS FOR POLICY, DELIVERY OR PRACTICE: Our results provide a novel analysis of the population-based linked SEER-Medicare data to evaluate outcomes associated with treatment of DCIS. The use of BCS alone ought to be reconsidered, given our findings that BCS alone yields a much higher risk of recurrence to DCIS and invasive breast cancer. Clinical trials of BCS alone in highly-selected patients are underway. PRIMARY FUNDING SOURCE: Foundations
Publication Types:
Keywords:
- Breast Neoplasms
- Carcinoma, Intraductal, Noninfiltrating
- Combined Modality Therapy
- Female
- Humans
- Incidence
- Lymph Node Excision
- Mastectomy
- Radiotherapy, Adjuvant
- SEER Program
- radiotherapy
- surgery
- therapy
- hsrmtgs
Other ID:
UI: 102274283
From Meeting Abstracts