Format

Send to

Choose Destination
See comment in PubMed Commons below
Surgery. 2016 Aug;160(2):454-62. doi: 10.1016/j.surg.2016.03.031. Epub 2016 May 27.

Breast cancer biology varies by method of detection and may contribute to overdiagnosis.

Author information

  • 1Departments of Surgery and Radiology, and Yale Comprehensive Cancer Center, Yale University School of Medicine, New Haven, CT.
  • 2Departments of Surgery and Radiology, and Yale Comprehensive Cancer Center, Yale University School of Medicine, New Haven, CT. Electronic address: Donald.lannin@yale.edu.

Abstract

BACKGROUND:

Recently, it has been suggested that screening mammography may result in some degree of overdiagnosis (ie, detection of breast cancers that would never become clinically important within the lifespan of the patient). The extent and biology of these overdiagnosed cancers, however, is not well understood, and the effect of newer screening modalities on overdiagnosis is unknown.

METHODS:

We performed a retrospective review of a prospectively collected database of breast cancers diagnosed at the Yale Breast Center from 2004-2014. The mode of initial presentation was categorized into 5 groups: screening mammogram, screening magnetic resonance imaging, screening ultrasonography, self-detected masses, and physician-detected masses.

RESULTS:

Compared with cancers presenting with masses, cancers detected by image-based screening were more likely to present with ductal carcinoma-in-situ or T1 cancers (P < .001). In addition to a simple stage shift, however, cancers detected by image-based screening were also more likely to be luminal and low-grade cancers; symptomatic cancers were more likely high-grade and triple-negative (P < .001, respectively). On a multivariate analysis, adjusting for age, race, and tumor size, cancers detected by mammogram, US, and magnetic resonance imaging had greater odds of being luminal (odds ratio 1.8, 95% confidence interval, 1.5-2.3; odds ratio 2.2, 95% confidence interval, 1.1-4.7; and odds ratio 4.7, 95% confidence interval, 2.1-10.6, respectively), and low-grade (odds ratio 2.2, 95% confidence interval, 1.6-2.9; odds ratio 4.9, 95% confidence interval, 2.7-8.9; and odds ratio 4.6, 95% confidence interval, 2.6-8.1, respectively) compared with cancers presenting with self-detected masses.

CONCLUSION:

Screening detects cancers with more indolent biology, potentially contributing to the observed rate of overdiagnosis. With magnetic resonance imaging and US being used more commonly for screening, the rate of overdiagnosis may increase further.

PMID:
27241117
DOI:
10.1016/j.surg.2016.03.031
[PubMed - in process]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Support Center