Narrower negative lumpectomy margins were associated with slightly higher 10-year ipsilateral breast recurrence rates among postmenopausal patients with hormone receptor–positive ductal carcinoma in situ, although margin width was not statistically significantly associated with recurrence after adjustment for age, endocrine therapy assignment, and tumor size.
Researchers conducted an ancillary analysis of prospectively collected margin data from the phase 3, double-blind NRG Oncology/National Surgical Adjuvant Breast and Bowel Project B-35 trial, according to findings published in JAMA Surgery. The parent trial randomized patients to tamoxifen or anastrozole; it did not randomly assign margin width or reexcision management.
The trial enrolled 3,104 postmenopausal patients with hormone receptor–positive ductal carcinoma in situ (DCIS) and tumor-free margins from January 2003 to June 2006. Patients underwent lumpectomy and were assigned to receive 5 years of tamoxifen or anastrozole. Whole-breast irradiation was specified by the protocol, although partial-breast irradiation was permitted among patients concurrently enrolled in another trial.
The ancillary analysis evaluated ipsilateral breast tumor recurrence (IBTR) using protocol-defined margin groups of less than 1 mm vs at least 1 mm and guideline-based groups of less than 2 mm vs at least 2 mm. Margins of less than 1 mm were close but did not have tumor on ink. Patients whose pathology forms reported positive margins were excluded.
Following exclusions for unavailable breast cancer follow-up, positive or unevaluable margins, and missing pathology forms, 2,707 patients were included in the 1-mm analysis. Of those, 2,546 had documented closest-margin measurements for the 2-mm analysis.
Researchers calculated 10-year cumulative incidence while accounting for competing risks. Adjusted models included age, randomized endocrine therapy assignment, and pathological tumor size.
IBTR occurred in 24 of 502 patients with margins of less than 1 mm and 66 of 2,205 patients with margins of at least 1 mm. The corresponding 10-year cumulative incidences were 5.6% and 4.0%, a statistically significant unadjusted difference.
In the 2-mm analysis, IBTR occurred in 39 of 879 patients with margins of less than 2 mm and 49 of 1,667 patients with margins of at least 2 mm. Ten-year cumulative incidence was 5.3% and 3.8%, respectively, also representing a statistically significant unadjusted difference.
After adjustment for age, endocrine therapy assignment, and tumor size, margin width was not statistically significantly associated with IBTR at either threshold. Margin width was not randomized, and the study did not use a noninferiority or equivalence framework.
Tumor size remained associated with recurrence. Patients with tumors of at least 1 cm had approximately twice the adjusted hazard of IBTR compared with those with smaller tumors.
Just over one-third of ipsilateral recurrences were invasive. Ten-year cumulative incidence of all breast cancer events was 9.5% among patients with margins of less than 1 mm and 8.5% among those with margins of at least 1 mm. Rates were 10% and 8.2%, respectively, using the 2-mm threshold; neither difference was statistically significant.
Results were similar when researchers restricted the analyses to the 98.7% of patients who received whole-breast irradiation and among the 82.2% who received a radiation boost. Overall, 65% completed 5 years of endocrine therapy, and treatment adherence did not differ by margin group.
Current consensus guidance from the Society of Surgical Oncology, American Society for Radiation Oncology, and American Society of Clinical Oncology recommends a 2-mm negative margin for patients with DCIS undergoing breast-conserving surgery and whole-breast irradiation.
The researchers wrote that omission of reexcision for margins of less than 1 mm or less than 2 mm could be reconsidered in appropriately selected patients. They further stated that their findings supported potential modification of margin-width criteria to reduce reexcision among postmenopausal patients with hormone receptor–positive DCIS planning breast irradiation and endocrine therapy.
In an accompanying invited commentary, Rakhshanda Layeequr Rahman, MD, of Creighton University School of Medicine, argued that the lack of molecular risk stratification limits application of the findings to contemporary practice. Genomic profiles may provide prognostic and predictive information beyond conventional clinical characteristics, she wrote. Rahman also noted that genomic-test validation was conducted in margin-negative tumors.
The study researchers similarly acknowledged that genomic tools may help identify recurrence risk but do not address adequate margin width.
Rahman cited ongoing trials evaluating active surveillance and omission of surgery among selected patients with low-risk DCIS. She noted that those studies generally define low risk using conventional clinical factors because obtaining a genomic profile ordinarily requires surgical tissue.
The researchers identified several limitations. Margin width was not randomized, and the protocol included no intervention based on margin width. Margin status was assessed by local institutional pathologists without real-time central review, although recurrences were reviewed centrally. Among the excluded patients, 84 had positive margins, 180 had margins that could not be evaluated, and 106 were missing pathology review forms.
Tumor size was estimated rather than directly measured in 34% of patients and was unknown in 3%. The researchers also stated that the analysis did not address premenopausal patients or those with hormone receptor–negative DCIS. The analyzed population had tumor-free margins, and 98.7% received whole-breast irradiation. The researchers framed their conclusions around postmenopausal patients with hormone receptor–positive DCIS planning breast irradiation and endocrine therapy.
“Personalized decision-making is warranted regarding the need for repeat margin reexcision based on a narrow lumpectomy margin,” wrote lead study author Irene L. Wapnir, MD, of Stanford Cancer Institute, and colleagues.
Disclosures: The National Cancer Institute supported the research and was involved in the design and conduct of the parent trial and in review and approval of the ancillary manuscript. It was not involved in data collection, management, analysis, or interpretation; manuscript preparation; or the decision to submit the manuscript for publication. Several researchers reported grants, consulting or advisory fees, honoraria, royalties, data-monitoring fees, or travel support from government, nonprofit, and industry entities. Rahman reported no conflicts of interest.
Source: JAMA Surgery
