The Southwestern Surgical Congress
Breast cancer tumor size: correlation between magnetic resonance imaging and pathology measurements

Presented at the 60th Annual Meeting of the Southwestern Surgical Congress, Acapulco, Mexico, March 30–April 2, 2008
https://doi.org/10.1016/j.amjsurg.2008.07.028Get rights and content

Abstract

Background

As physicians increasingly use magnetic resonance imaging (MRI) for the evaluation of newly diagnosed breast cancers, a review of the correlation between MRI and pathology tumor size is imperative.

Methods

A retrospective review of 91 breast tumors comparing preoperative MRI tumor size to final pathology tumor size was performed.

Results

MRI and pathology tumor size were positively correlated (R = .650), but with an average overestimation by MRI of .63 cm (P <.0001). When stratified by MRI tumor size (≤2.0 cm and >2.0 cm), a significant difference was found only in tumors greater than 2.0 cm (average overestimation = 1.06 cm; P <.0001). This trend continued for the histological subtypes of ductal carcinoma in situ (DCIS), invasive ductal carcinoma (IDC), and invasive lobular carcinoma (ILC).

Conclusions

MRI tumor size correlates with pathology size; however, a significant overestimation exists, particularly for tumors >2.0 cm. Clinicians should therefore use caution in relying on MRI tumor size in determining candidacy for breast conservation therapy (BCT).

Section snippets

Methods

We performed a retrospective chart review of all women who had undergone a breast MRI from January 2000 through August 2007 within a breast surgery specialty practice. This study was approved for implementation by the Human Subjects Committee of the University of Kansas School of Medicine–Wichita. Patients were included in the study if they had a newly diagnosed, biopsy-proven breast cancer with a positive MRI before definitive surgical treatment. Evaluation with MRI led to the identification

Results

Demographics, including age, family history, menopausal status, and abnormal mammograms, are presented in Table 1. We identified 77 patients with 91 tumors that met the study criteria. There were 79 initial cancer procedures performed on the 77 patients, with 1 patient undergoing bilateral mastectomies for bilateral cancers and a second patient undergoing bilateral lumpectomies for bilateral cancers. Initial surgical treatment consisted of 38 mastectomies and 41 lumpectomies. An additional 11

Comments

The current standards of surgical care for the treatment of breast cancer include BCT and mastectomy. The long-term survival associated with these 2 options has been proven to be equal.1 The dilemma for today's physicians centers on the determination of eligibility for BCT while at the same time minimizing local recurrence. The traditional tool for determining candidacy for BCT is mammography, but there has been some evidence supporting the addition of MRI to determine optimal operative

References (20)

There are more references available in the full text version of this article.

Cited by (93)

  • Correlation Between Preoperative Radiological and Postoperative Pathological Tumor Size in Patients With HER2<sup>+</sup> Breast Cancer After Neoadjuvant Chemotherapy Plus Trastuzumab and Pertuzumab

    2022, Clinical Breast Cancer
    Citation Excerpt :

    RECIST criteria are then applied to establish response to therapy.9 Scientific studies, that investigated the accuracy of radiological measurements for detection of radiological complete response (rCR) during chemotherapy, came to divergent results on which should be the modality of choice.11–1710–17 MRI has been proposed as the standard for early response,6 but several studies showed that also MRI might underestimate or overestimate residual tumor.17

  • Accuracy and inter-reader agreement of breast MRI for cancer staging using 0.08 mmol/kg of gadobutrol

    2021, Clinical Imaging
    Citation Excerpt :

    When the absolute difference was between 5 mm and −5 mm, the MRI measurement was considered correct; when the absolute difference was over 5 mm, the MRI measurement was considered overestimated; when the absolute difference was smaller than 5 mm, the MRI measurement was considered underestimated. The choice of the 5 mm cut-off was determined by its adoption by previous studies and considering its clinical relevance [19,20]. Inter-reader reproducibility was estimated using the Bland–Altman method [21].

View all citing articles on Scopus
View full text