Local Recurrence in Women With Stage I Breast Cancer: Declining Rates Over Time in a Large, Population-Based Cohort

• Published in: International journal of radiation oncology, biology, physics Vol. 88; no. 1; pp. 80 - 86
• Main Authors: Canavan, Joycelin, Truong, Pauline T., Smith, Sally L., Lu, Linghong, Lesperance, Mary, Olivotto, Ivo A.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.01.2014
• Subjects: Aged; Antineoplastic Agents - therapeutic use; Breast Neoplasms - chemistry; Breast Neoplasms - epidemiology; Breast Neoplasms - pathology; Breast Neoplasms - therapy; CHEMOTHERAPY; Cohort Studies; Combined Modality Therapy - trends; DIAGNOSIS; ELDERLY PEOPLE; ESTROGENS; Female; Follow-Up Studies; Hematology, Oncology and Palliative Medicine; Humans; Incidence; MAMMARY GLANDS; Mastectomy, Segmental; Middle Aged; Multivariate Analysis; Neoplasm Recurrence, Local - epidemiology; Neoplasm Staging; NEOPLASMS; Radiology; RADIOLOGY AND NUCLEAR MEDICINE; RADIOTHERAPY; Radiotherapy Dosage; Receptors, Estrogen - analysis; SURGERY; Time Factors; WOMEN
• ISSN: 0360-3016; 1879-355X; 1879-355X
• DOI: 10.1016/j.ijrobp.2013.10.001

To evaluate whether local recurrence (LR) risk has changed over time among women with stage I breast cancer treated with breast-conserving therapy. Subjects were 5974 women aged ≥50 years diagnosis with pT1N0 breast cancer from 1989 to 2006, treated with breast-conserving surgery and radiation therapy. Clinicopathologic characteristics, treatment, and LR outcomes were compared among 4 cohorts stratified by year of diagnosis: 1989 to 1993 (n=1077), 1994 to 1998 (n=1633), 1999 to 2002 (n=1622), and 2003 to 2006 (n=1642). Multivariable analysis was performed, with year of diagnosis as a continuous variable. Median follow-up time was 8.6 years. Among patients diagnosed in 1989 to 1993, 1994 to 1998, 1999 to 2002, and 2003 to 2006, the proportions of grade 1 tumors increased (16% vs 29% vs 40% vs 39%, respectively, P<.001). Surgical margin clearance rates increased from 82% to 93% to 95% and 88%, respectively (P<.001). Over time, the proportions of unknown estrogen receptor (ER) status decreased (29% vs 10% vs 1.2% vs 0.5%, respectively, P<.001), whereas ER-positive tumors increased (56% vs 77% vs 86% vs 86%, respectively, P<.001). Hormone therapy use increased (23% vs 23% vs 62% vs 73%, respectively, P<.001), and chemotherapy use increased (2% vs 5% vs 10% vs 13%, respectively, P<.001). The 5-year cumulative incidence rates of LR over the 4 time periods were 2.8% vs 1.7% vs 0.9% vs 0.8%, respectively (Gray's test, P<.001). On competing risk multivariable analysis, year of diagnosis was significantly associated with decreased LR (hazard ratio, 0.92 per year, P=.0003). Relative to grade 1 histology, grades 2, 3, and unknown were associated with increased LR. Hormone therapy use was associated with reduced LR. Significant changes in the multimodality management of stage I breast cancer have occurred over the past 2 decades. More favorable-risk tumors were diagnosed, and margin clearance and systemic therapy use increased. These changes contributed to the observed declining LR rates among patients treated with breast-conserving therapy.