Ipsilateral invasive cancer risk after diagnosis with ductal carcinoma in situ (DCIS): Comparison of patients with and without index surgery.

Authors

null

Marc D Ryser

Duke University School of Medicine, Durham, NC

Marc D Ryser , Laura Hendrix , Samantha M. Thomas , Thomas Lynch , Anne McCarthy , Zahed Mohammed , Amanda B. Francescatti , Elizabeth S. Frank , Ann H. Partridge , Alastair Mark Thompson , Terry Hyslop , Eun-Sil Shelley Hwang

Organizations

Duke University School of Medicine, Durham, NC, Duke Cancer Institute, Durham, NC, Duke University Medical Center, Durham, NC, Duke University, Durham, NC, American College of Surgeons, Chicago, IL, Dana-Farber Cancer Institute, Lexington, MA, Dana-Farber Cancer Institute, Boston, MA, Baylor College of Medicine, Houston, TX, Department of Biostatistics and Bioinformatics, Duke Cancer Institute, Durham, NC

Research Funding

Other
PCORI/PCS-1505-30497

Background: Most women diagnosed with ductal carcinoma in situ (DCIS) undergo surgical resection, potentially leading to overtreatment of patients who would not develop clinically significant breast cancer in the absence of locoregional treatment. We compared the risk of ipsilateral invasive breast cancer (iIBC) between DCIS patients who received breast conserving surgery (BCS) for their index diagnosis of DCIS (BCS group) and patients who did not receive any locoregional treatment within 6 months of diagnosis (surveillance [SV] group). Methods: A treatment-stratified random sample of patients diagnosed with screen-detected and biopsy-confirmed DCIS in 2008-14 was selected from 1,330 Commission on Cancer-accredited facilities (20/site). Excluding patients who received a mastectomy ≤6 months, the final analytic cohort contained 14,245 (88.2%) BCS and 1,914 (11.8%) SV patients. Subsequent breast events were abstracted up to 10 years after diagnosis. Primary outcome was the 8-year absolute difference in iIBC risk between BCS and SV; a subgroup analysis was performed for grade I/II patients. A propensity score (PS) model for treatment was fitted with sampling design (SD) weighting and random effects for patients within facilities. Absolute risk differences were estimated using PS-SD-weighted Kaplan Meier estimators. Results: Overall, median age at diagnosis was 61 years (IQR: 52-69) and median follow-up was 5.8 years (95% CI 5.7-6.1). The majority of patients were Caucasian (81.9%), with estrogen receptor-positive (80.6%), and nuclear grade I/II (54.5%) DCIS. The fraction of patients with a Charlson comorbidity score of ≥2 was higher in SV (14.2%) compared to BCS (6.4%, p < 0.001). The 8-year risk of iIBC was 3.0% (95% CI: 2.4%-3.6%) for BCS and 7.7% (95% CI: 4.9%-10.5%) for SV, with an absolute risk difference of 4.7% (95% CI: 4.5%-4.9%; log-rank p < 0.001). Among patients with grade I/II tumors, the 8-year risk of iIBC was 3.1% (95% CI: 2.3%-4.0%) for BCS and 6.1% (95% CI: 2.5%-9.8%) for SV; difference: 3.0% (95% CI: 2.7%-3.2%; p = 0.005). Conclusions: Despite an increased risk of iIBC in SV patients compared to BCS patients, the 8-year risk did not exceed 10% in either group. The risk of recurrence in BCS patients was comparable to previously reported estimates. These data demonstrate a considerable degree of overtreatment among patients with non-high grade DCIS. Prospective clinical trials will help determine the tradeoffs between universally directed as opposed to selectively applied surgery for low risk DCIS.

Disclaimer

This material on this page is ©2024 American Society of Clinical Oncology, all rights reserved. Licensing available upon request. For more information, please contact licensing@asco.org

Abstract Details

Meeting

2020 ASCO Virtual Scientific Program

Session Type

Poster Discussion Session

Session Title

Breast Cancer—Local/Regional/Adjuvant

Track

Breast Cancer

Sub Track

Local-Regional Therapy

Citation

J Clin Oncol 38: 2020 (suppl; abstr 519)

DOI

10.1200/JCO.2020.38.15_suppl.519

Abstract #

519

Poster Bd #

11

Abstract Disclosures