Radiology: The value of breast DWI magnetic resonance imaging in predicting preoperative staging of ductal carcinoma in situ

Ductal carcinoma in situ (DCIS) is a non-invasive breast cancer characterized by malignant proliferation of epithelial cells confined to the duct without a basement membrane of invasion
.
DCIS is usually diagnosed at the time of
a percutaneous needle biopsy before surgery.
However, due to the sampling limitations inherent in biopsy techniques, up to 59% of lesions diagnosed with DCIS at biopsy escalate to invasive cancer
after surgery.
The presence of invasive breast cancer requires adequate axillary lymph node staging to determine the type of
further adjuvant therapy.

At present, breast MRI has been widely used in the preoperative evaluation
of women with breast cancer.
Dynamic augmentation (DCE) MRI is the standard protocol for breast MRI scanning, which is useful in detecting DCIS and invasive breast cancer compared to mammography and ultrasound has the highest sensitivity
.
Although results may be inconsistent between studies, large tumor diameter, mass type and heterogeneity, or edge enhancement have been reported as important predictors of invasive breast cancer on DCE MRI scans
。 DWI is a functional MRI technique that can provide additional information
for DCE MRI by evaluating the movement of water molecules within a tissue to reflect the microstructural properties of the tissue.

Recently, a study published in the journal Radiology established and verified a prediction model for the staging and risk stratification of patients with DCIS based on preoperative breast MRI, which provides support
for accurate clinical risk stratification and prognosis assessment of patients.

This study retrospectively identified patients with serial DCIS (development set) who underwent a 3.
0T breast MRI scan preoperatively and underwent surgery between June 2019 and March 2020, with breast MRI scans, from a medical center These include Dynamic Contrast Enhancement (DCE) MRI and diffusion-weighted imaging (DWI).

The apparent diffusion coefficient of lesions, DCE MRI lesion size and morphological characteristics, milk-mammogram results, age, symptoms, biopsy method, and DCIS grade
at biopsy were collected.
The pathology of invasive breast cancer and the presence
of axillary metastases are determined surgically.
A predictive model was established using multivariate logistic regression and validated
in a prospective internal validation group recruited between July 2020 and April 2021.

57 (41%) of 140 women in the development group (mean age 53 years±11 [SD]) and 43 (41%) of 105 women in the validation group (mean age 53 years ±10) experienced grade improvement
after surgery.
Predictive models combining DWI and clinicopathological factors showed AUC of 0.
87 (95% CI: 0.
80, 0.
92) for the development set and 0.
76 (95% CI: 0.
67, 0.
84)
for the validation set.
The predicted probability of invasive breast cancer showed good inter-observer agreement (intraclass correlation coefficient, 0.
79); The positive predictive value was 85% (28 out of 33) and the negative predictive value was 92% (22 out of 24).

Figure 66-year-old woman diagnosed by needle biopsy with intermediate ductal carcinoma in situ (DCIS).

(A) The right longitudinal oblique mammography shows segmental, fine polymorphic calcifications (arrows)
in extremely dense breast tissue.
(B) A 6.
3 cm segmental clump-like strengthening area (arrow) is observed in the right upper lateral breast (arrow) on MRI scans of the enhanced axial (left) and sagittal (right)
positions before surgery.
(C) Axial diffusion-weighted image with a b value of 1200 sec/mm² (left) and sagittal maximal intensity projection image (right) do not show hypersignal
in the area corresponding to the enhanced lesion.
A total mastectomy and sentinel biopsy of axillary lymph nodes were performed, pathological analysis showed pure DCIS and no axillary lymph node metastases
.
The probability of predicting aggressive cancer is 0.
15

The proposed preoperative breast MRI model combines diffusion-weighted imaging and clinicopathological factors to more accurately predict the stage
of women with biopsy-confirmed ductal carcinoma in situ.
The developed multivariate model is a reproducible imaging parameter for predicting invasive cancer that can be used to screen patients undergoing sentinel lymph node biopsy at the time of initial surgery, thus avoiding unnecessary axillary surgery and preventing delays
in secondary surgery.

Original source:

Shin Ae Lee,Youkyoung Lee,Han Suk Ryu,et al.
Diffusion-weighted Breast MRI in Prediction of Upstaging in Women with Biopsy-proven Ductal Carcinoma in Situ.
DOI:10.
1148/radiol.
213174