# Multiparametric Breast MRI for the Differentiation of Atypical Ductal Hyperplasia and Ductal Carcinoma In Situ in Postmenopausal Women

## TL;DR

This study explores whether breast MRI can distinguish between two similar breast lesions in postmenopausal women but finds no significant differences.

## Contribution

The study is the first to compare MRI parameters for differentiating ADH and DCIS in postmenopausal women using a matched cohort.

## Key findings

- No significant differences were found in MRI parameters like lesion size, ADC, or PEI between ADH and DCIS.
- TIC types were similarly distributed between the two groups, with no washout patterns observed.
- The results suggest MRI may not reliably differentiate ADH and DCIS in this population.

## Abstract

Introduction

Atypical ductal hyperplasia (ADH) is a proliferative epithelial breast lesion associated with an increased risk of progression to ductal carcinoma in situ (DCIS) or invasive carcinoma. Although breast MRI is highly sensitive for lesion detection, its ability to differentiate biologically related non-mass enhancement (NME) lesions such as ADH and DCIS remains incompletely defined.

Aim

The aim of this exploratory study was to investigate whether predefined morphologic, dynamic, and functional breast MRI parameters, such as lesion size, postcontrast signal intensity increase (wash-in), time-intensity curve (TIC) type, apparent diffusion coefficient (ADC), and positive enhancement integral (PEI), demonstrate measurable differences between ADH and DCIS in postmenopausal women.

Materials and methods

This retrospective, single-center study included 40 postmenopausal women with histopathologically confirmed ADH (n = 20) or DCIS (n = 20) who underwent standardized 1.5-Tesla (T) breast MRI. Quantitative MRI parameters evaluated included lesion size, early postcontrast signal intensity increase (wash-in), time-intensity curve (TIC) type, apparent diffusion coefficient (ADC), and positive enhancement integral (PEI). Groups were matched for age and lesion size. Statistical comparisons were performed using appropriate parametric tests, with a two-sided significance threshold of p < 0.05.

Results

No statistically significant differences were observed between ADH and DCIS across the evaluated MRI parameters, including lesion size (p = 0.119), wash-in (p = 0.484), ADC (p = 0.141), or PEI (p = 0.360). TIC Types 1 and 2 were similarly distributed between groups (p = 0.342), with no lesions demonstrating a washout pattern.

Conclusion

In this exploratory cohort of postmenopausal women, ADH and DCIS demonstrated substantial overlap in morphologic, dynamic, and functional breast MRI parameters. While no statistically significant differences were identified, these findings should be interpreted cautiously given the limited sample size and potential for Type II error. Larger, adequately powered studies incorporating advanced quantitative imaging techniques are required to determine whether multiparametric MRI can reliably differentiate between these entities.