Mucinous breast carcinoma (MBC) is a rare histological type of breast cancer, accounting for about 1% to 6% of all invasive breast cancers, prevalent among perimenopausal and postmenopausal women [1,2]. It has a better prognosis and long-term survival rate than breast cancer without specific types but a lower incidence of lymph node involvement [1,2]. In addition, MBC is pathologically characterized by a large amount of extracellular mucin and abnormal cells that make up the tumor “float” in the mucin pool [2]. MBCs can be divided into two subgroups: pure mucinous breast carcinoma (PMBC) and mixed mucinous breast carcinoma (MMBC) [1]. PMBCs comprise at least 90% extracellular mucin, whereas MMBC contains 50–90% extracellular mucin and mixed lobular or ductal breast cancer [1,2]. The high water content of this myxoid stroma accounts for the strong high-signal intensity of PMBC on T2-weighted imaging (T2WI) (T2-SHi) [3,4]. In addition, unlike MMBCs with invasive components, PMBCs exhibit benign features (e.g., well-defined borders, increased posterior echogenicity, etc.) on mammography and ultrasound [5]. They can be misdiagnosed as other benign tumors with T2-SHi on breast magnetic resonance imaging (MRI).

Fibroadenoma (FA) is the most common benign breast tumor, with a peak age of onset of 15 to 35 years [6]. Histopathologically, FAs have several variants, such as cellular, juvenile, complex, and myxoid FAs, of which myxoid FAs are mainly composed of myxoid stroma, whereby the number of interstitial compartment cells is reduced [6]. In addition, approximately 40% of FAs have edematous or myxoid histological changes [6]. These edematous or mucinous components in FAs have a longer T2 relaxation time in breast MRI and manifest as T2-SHi. Accordingly, such FAs have been defined as FAs with T2-SHi. In clinical practice, FAs with T2-SHi are easily confused with PMBCs, and it is difficult to give an accurate diagnosis by a single T2WI sequence alone.

Indeed, the dilemma lies in accurately distinguishing between PMBCs and FAs with T2-SHi prior to surgery, as this decision determines the need for resection. Importantly, FAs may slowly regress, often requiring no treatment but necessitating clinical follow-up [7,8]. In contrast, the most commonly used treatment modality for PMBCs is breast-conserving therapy plus sentinel lymph node biopsy [4].

Multiparametric breast MRI, usually including T1-weighted dynamic contrast-enhanced imaging (DCE), T2WI, and diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) maps, has been demonstrated to be useful in the differential diagnosis of breast tumors and the evaluation of the efficacy of neoadjuvant chemotherapy [9,10]. It is increasingly used in clinical practice as the examination indications expand. Over the last decade, few studies have applied MRI to the differential diagnosis of PMBCs and FAs with T2-SHi [3,[11], [12], [13], [14]]; however, almost all have used qualitative assessment methods. For example, Qu et al. [3] analyzed the qualitative features of 35 PMBCs and 70 FAs with T2-SHi on DCE images and found that the delayed enhancement pattern was a reliable feature to differentiate PMBCs from FAs with T2-SHi, with the delayed local edge enhancement of PMBCs being significantly greater than that of FAs with T2-SHi. In addition, Igarashi et al. [15] reported the value of breast imaging reporting and data system (BI-RADS) MRI descriptors (5th Edition) in distinguishing MBCs from FAs. They showed that irregular margins, delayed heterogeneous enhancement, and intratumoral septation could help differentiate MBCs from FAs. Although these results have clinical significance, they are based on subjective assessment. Therefore, an objective, quantitative, and comprehensive assessment tool is urgently needed to discriminate PMBCs and FAs with T2-SHi.

Histogram, as a first-order statistical method, can quantitatively assess tumor heterogeneity by measuring the probability distribution of pixel values on medical images. In addition, unlike small 2D-region of interest (ROI) or single-slice ROI, a whole-tumor-based histogram can more objectively and comprehensively reflect the heterogeneous characteristics of the entire tumor with good reproducibility [[16], [17], [18]]. A previous study by Guo et al. [19] demonstrated that the parameters based on whole-lesion ADC histogram can effectively identify subtypes of MBCs (e.g., PMBCs vs. MMBCs, etc.). Multiparametric MRI can provide more abundant lesion information; however, no study has investigated the utility of a whole-tumor histogram method based on multiparametric MRI in differentiating PMBCs from FAs with T2-SHi.

Therefore, we hypothesized that a whole-tumor histogram based on multiparametric MRI could better assess tumor heterogeneity. We aimed to investigate the utility of whole-tumor histogram analysis using multiparametric MRI to distinguish PMBCs from FAs with T2-SHi. Furthermore, we investigated the relationship between whole-tumor histogram parameters and the Ki-67 index of PMBCs.