The value of non-enhanced magnetic resonance imaging based on T2-weighted imaging and apparent diffusion coefficient in the study of three-tiered HER2 status in patients with breast cancer

Introduction

Breast cancer is the most common malignant tumor and the leading cause of cancer death in women worldwide (1,2). Human epidermal growth factor receptor 2 (HER2) is a prototype oncogene in breast cancer. Its role and impact in breast cancer are multifaceted, including promoting tumor cell proliferation, survival, and differentiation (3). Overexpression or gene amplification of HER2 is closely related to the invasion and poor prognosis of breast cancer (4). In recent years, the traditional classification of HER2-negative cases has been refined into two distinct categories: HER2-low expressing status, defined as an immunohistochemistry (IHC) score of 1+ or 2+ in the absence of gene amplification, and HER2-zero expressing status, characterized by an IHC score of 0 (4,5). It is estimated that approximately half of all breast cancers fall into the HER2-low category, which has consistently revealed heterogeneity in terms of clinical outcomes and biological features in the recent literature (5,6). A new generation of HER2-targeting agents, especially antibody-drug conjugates, such as trastuzumab deruxtecan (T-DXd), have shown remarkable efficacy in HER2-low breast cancer (6,7). Therefore, identifying the status of HER2-overexpressing, HER2-low expressing, and HER2-zero expressing in breast cancer patients is crucial for determining precise, personalized treatment plans (4,6).

The preoperative assessment of HER2 status is usually based on pathologic analysis of biopsy specimens. However, percutaneous breast tumor biopsy is an invasive procedure, breast cancer displays considerable heterogeneity at multiple levels, and a limited amount of biopsied tissue may not fully represent the biological and genetic characteristics of the entire tumor (8,9). Therefore, a simple, practical, and noninvasive method is needed to assess the expression status of HER2 in breast cancer patients before treatment. The main advantages of magnetic resonance imaging (MRI) in breast cancer diagnosis include the absence of radiation, high soft tissue resolution, and multi-parameter and multi-sequence imaging capabilities, which can provide morphological and functional information of the tumor (10,11). In the current research, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and radiomics are hot spots. Contrast-enhanced (CE) MRI is the most sensitive tool for detecting breast cancer but has higher costs from contrast agents, allergy risks, and longer scanning time (12). Although radiomics exhibits commendable sensitivity and specificity in breast cancer (13,14), its drawbacks such as complex procedures, low reproducibility, and limited clinical applicability cannot be ignored (15,16). In contrast, non-enhanced MRI based on T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) is more practical in clinical use, cost-effective, and safer. Previous studies have shown that T2WI can identify benign/malignant lesions (11), predict breast cancer patients’ neoadjuvant chemotherapy response (17), and detect breast edema, the degree of which correlates with tumor grade and serves as a breast cancer aggressiveness biomarker (18).

DWI leverages the Brownian motion characteristics of water protons to provide valuable biological insights into tissue cell density. Apparent diffusion coefficient (ADC) is a quantitative parameter calculated from DWI signal, which directly reflects the speed of water molecular diffusion. ADC plays an important role in the qualitative and differential diagnosis of breast lesions and the prediction of vascular tumor thrombosis, as reported in previous studies (19,20). There is a paucity of literature on the application of T2WI and ADC to differentiate HER2 expression statuses in breast cancer, so this study aimed to investigate the feasibility and clinical significance of this approach. We present this article in accordance with the TRIPOD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2547/rc).

Methods

Study population

From February 2015 to December 2023, a total of 482 breast cancer patients in the Third Affiliated Hospital of Sun Yat-sen University (hereinafter referred to as the Third Hospital) were collected. The inclusion criteria were as follows: (I) female individuals with primary breast cancer confirmed by pathological examination; (II) patients who underwent routine breast MRI examination; (III) HER2 status of postoperative specimens was detected by IHC and/or in situ hybridization (ISH). The time interval between the most recent magnetic resonance (MR) examination and the surgery for all the above patients was within one month. Of the 482 patients with primary breast carcinoma who underwent pre-operative MRI at our institution, we excluded those with pathologically confirmed ductal carcinoma in situ DCIS (n=32), incomplete clinical and pathological data (n=36), prior breast treatment (n=40), and poor or incomplete image quality due to movement or hemorrhage artifacts (n=13). Finally, 361 eligible patients were included in the study and were categorized into three groups according to the three-tiered classification of HER2 expression level: HER2-zero expressing (IHC 0; n=63), HER2-low expressing (IHC 1 or 2, FISH-negative; n=181), and HER2-overexpressing (IHC 2+, fluorescence FISH-positive, or IHC 3; n=117) (Figure 1). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Medical Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University (No. 1I2023-285-02). In view of the retrospective design and minimal risk to participants, the Ethics Committee granted a waiver of the requirement to obtain informed consent.

Figure 1 Patient flowchart. ADC, apparent diffusion coefficient; DCIS, ductal carcinoma in situ; HER2, human epidermal growth factor receptor 2; MRI, magnetic resonance imaging; T2WI, T2-weighted imaging.

MRI acquisition

All MRI examinations were performed using a 3.0 tesla (T) superconducting MRI scanner equipped with a special breast phased-array surface coil, with machine models Discovery 750 (GE Healthcare, Chicago, IL, USA), Prisma (Siemens, Erlangen, Germany) and 790 (United Imaging, Shanghai, China). The acquisition parameters of the T2WI fat-suppressed (T2FS) and DWI sequences are detailed in Table 1.

MRI evaluation

MR images were analyzed independently by two radiologists (4 and 15 years of experience, respectively) who were unaware of the clinicopathological information, and disagreements were resolved via consensus-based discussion. In patients with multicentric/multifocal tumors, only the largest lesions were selected. Focal edema, on T2WI, was defined as water-like high SI. With reference to previous literatures (18,21), we used the breast edema score (BES) to classify the presentation of breast edema on T2WI: BES 1, no edema; BES 2, peritumoral edema; BES 3, chest edema; BES 4, subcutaneous edema (Figure 2A-2D).

Figure 2 Examples of breast edema score on fat-suppression T2WI. (A) No peritumoral edema (BES 1). (B) An area of T2WI hyperintensity was seen around the tumor but did not extend into the pectoralis major muscle (BES 2). (C) A water-like signal area is visible around the tumor, extending towards the pectoralis major muscle in a band-like shape (BES 3). (D) An area of severe watery hyperintensity was seen around the mass extending into the subcutaneous area (BES 4). Arrows: peritumoral edema; stars: tumor. BES, breast edema score; T2WI, T2-weighted imaging.

In addition, two radiologists manually delineated the region of interest of breast lesion (L-ROI) on T2FS images, avoiding the cystic degeneration and necrosis areas, and delineated the region of interest of the pectoralis muscle (M-ROI) on the same level of the affected side, and tried to keep the area of the two ROIs close. The SI of the tumor was expressed as the ratio of lesion to pectoralis muscle (L/M) (Figure 3).

Figure 3 Measurement schematic of L/M ratio. In STIR images, the mass showed high signal intensity. The SI was 314.2, and the SI of pectoralis muscle at the same level was 35.2, and L/M ratio was 8.92. L/M, lesion to pectoralis muscle; SI, signal intensity; STIR, short-time inversion recovery.

For the U790 MR and Siemens scanner, ADC plots were automatically generated after DWI sequence scanning, with b values of 50 and 800 s/mm², respectively. For the GE scanner, the DWI images need to be imported into the GE Advanced Workstation to obtain the ADC map. The b values are 0 and 800 or 1,000 s/mm² respectively. The method for measuring ADC values was as follows: Initially, an ROI was delineated on the slice showing the largest solid portion of the lesion, carefully avoiding areas of cystic degeneration or necrosis. This process was then repeated on the immediately adjacent upper and lower slices. The average ADC value from these three ROIs was used for the final analysis.

Pathology

HER2 status was evaluated and reclassified by an experienced radiologist (with 20 years of expertise in breast imaging) based on the original pathology reports, with strict adherence to the 2023 American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) guidelines. Breast cancers were categorized into HER2-overexpressing status (IHC score 3+ or IHC 2+ with ISH amplification), HER2-low expressing status (IHC score 1+ or IHC 2+ without ISH amplification), and HER2-zero expressing status (IHC score 0).

Statistical analysis

The software SPSS 20.0 (IBM Corp., Armonk, NY, USA) was used for statistical analyses in this study. Normally distributed continuous variables were presented as mean ± standard deviation (SD), whereas non-normally distributed variables were presented as medians (interquartile ranges). Categorical variables were expressed in terms of frequency counts and frequencies.

As the continuous variables in this study were all non-normally distributed, comparisons between the two groups were made using the Mann-Whitney U test. For comparisons of continuous variables across multiple groups (e.g., the three HER2 status groups), the Kruskal-Wallis H test was employed. Categorical measures were examined using either the Chi-squared test or Fisher’s exact test. Given the multiple comparisons performed in this study, a Bonferroni correction was applied to control for Type I error inflation. All tests were conducted with a two-tailed approach, and significance was set at P<0.05.

Prediction model establishment and evaluation

The clinicopathological and MRI features of breast cancer patients with different HER2 expression status were evaluated respectively, and univariate logistic regression analysis was performed to determine the potential related factors of HER2 expression level. Variables with a P value <0.1 in univariate analysis were then included in the multivariate logistic regression model using the stepwise method (likelihood ratio) to determine the predictors of HER2 expression levels. The final multivariate regression model was visualized using a nomogram.

Kappa coefficient (k) was used to evaluate the degree of inter-observer agreement. k≥ 0.75 was generally considered to be excellent agreement; 0.40≤ k <0.75 represented good consistency; k<0.40 indicates poor agreement.

Results

Clinicopathologic and radiological characteristics of patients

A total of 361 female breast cancer patients were enrolled in this study. There were 63 (17.5%), 181 (50.1%), and 117 (32.4%) patients in the HER2-zero expressing, HER2-low expressing, and HER2-overexpressing groups, respectively.

Significant differences in Ki-67 level, intra-tumoral T2FS hypersignal, and L/M ratio were observed between the HER2-zero and HER2-low expressing groups (P<0.05). Moreover, significant differences in histological tumor grade, Ki-67 level, intra-tumoral FS T2WI hypersignal, short diameter, ADC value, tumor edema score, and L/M ratio were found between the HER2-low and HER2-overexpressing groups (P<0.05) (Tables 2,3).

Presence of edema and L/M ratio on T2WI

The results of reader 1 and reader 2 showed good consistency in the observation of edema score [k=0.96 (0.94–0.99)].

In the HER2-zero expressing group, the distribution across BES categories was as follows: 34 cases (54.0%) in BES 1, 11 (17.5%) in BES 2, 3 (4.8%) in BES 3, and 15 (23.7%) in BES 4. For the HER2-low expressing group, the distribution was 120 cases (66.3%) in BES 1, 27 (14.9%) in BES 2, 10 (5.5%) in BES 3, and 24 (13.3%) in BES 4. In the HER2-overexpressing group, the distribution was 58 cases (49.6%) in BES 1, 28 (23.9%) in BES 2, 7 (6.0%) in BES 3, and 24 (20.5%) in BES 4.

After pairwise comparison, we found that there was a significant difference in the BES between the HER2-low expressing group and the HER2-overexpressing group (P=0.032). The degree of edema in the HER2-low expressing group was lower than that in the HER2-overexpressing group. The proportion of BES 1 was the highest in both the HER2-low expressing group and HER2-overexpressing group, accounting for 66.3% and 49.6%, respectively, followed by BES 2, accounting for 14.9% and 23.9%, respectively. For BES 4, comprising those with the most severe degree of edema, the percentage in the HER2-overexpressing group (20.5%) was higher than that in the HER2-low expressing group (13.3%). However, there was no statistically significant difference in BES between the other two groups.

We utilized the average L/M of reader1 and reader 2 for group comparisons. The disparity in L/M ratio between the HER2-zero expressing group and the HER2-low expressing group, as well as between the HER2-low expressing group and the HER2-overexpressing group, exhibited statistical significance (P<0.001, P=0.005).

Multivariate analysis

The multivariable logistic analysis revealed that intra-tumoral T2FS hypersignal, L/M ratio, ADC, and Ki-67 served as independent predictors for HER2 expression status (Tables 4,5). The predictive model for distinguishing HER2-zero and HER2-low expression achieved an area under the curve (AUC) of 0.737, whereas the model for differentiating HER2-low and HER2-overexpressing groups yielded an AUC of 0.761 (Figure 4A,4B). Additionally, the model was visualized as a nomogram for ease of use (Figure 4C,4D).

Figure 4 The diagnostic performance of the multivariate regression model for HER2 status. (A) The AUC of the model for predicting HER2-zero expressing and HER2-low expressing groups was 0.737. (B) The AUC for predicting HER2-low expressing and HER2-overexpressing groups was 0.761. (C,D) Visualize the clinical-radiomics model as nomograms. Nomogram (C) illustrates the model for distinguishing between HER2-zero expressing and HER2-low expressing groups. Nomogram (D) presents the model for distinguishing between HER2-low expressing and HER2- overexpressing groups. For T2WI hyperintensity, 1 denotes the presence of high signal, whereas 0 denotes its absence. For Ki-67, 1 indicates positivity and 0 indicates negativity. ADC, apparent diffusion coefficient; AUC, area under the curve; HER2, human epidermal growth factor receptor 2; L/M, ratio of lesion to pectoralis muscle; ROC, receiver operating characteristic; T2WI, T2-weighted imaging.

The graphical predictions of representative cases for both models are shown in Figures 5,6.

Figure 5 Representative case and prediction model of HER2-low expressing breast cancer are presented. (A) Invasive breast cancer (no significant T2WI hyperintensity, L/M =1.40) in a 35-year-old woman. (B) As shown in the nomogram, this case received a total score of 125 points (29+96=125) by summing the scores assigned to the two predictors—T2 hyperintensity and the L/M ratio. This score indicates that the imaging model, composed of these two factors, predicts HER2-low expressing breast cancer with an accuracy exceeding 90%. For T2WI hyperintensity, 1 denotes the presence of high signal, while 0 denotes its absence. HER2, human epidermal growth factor receptor 2; L/M, ratio of lesion to pectoralis muscle; T2WI, T2-weighted imaging.

Figure 6 Representative case and prediction model of HER2-overexpressing breast cancer are presented. (A,B) Invasive breast cancer (ADC =1.71×10⁻³ mm²/s, Ki-67 =90%, L/M =5.53) in a 42-year-old woman. (C) As shown in the nomogram, this case received a total score of 125 points (72+33+20=125) by summing the scores assigned to the three predictors—ADC, L/M ratio, and Ki-67. This score indicates that the imaging-pathology model, composed of these three factors, predicts HER2-overexpressing breast cancer with an accuracy of 80%. For Ki-67, 1 indicates positivity and 0 indicates negativity. ADC, apparent diffusion coefficient; HER2, human epidermal growth factor receptor 2; L/M, ratio of lesion to pectoralis muscle.

Discussion

The findings of this study suggest that the combination of edema degree and SI based on T2WI with ADC can assist in identifying HER2-overexpressing, HER2-low expressing, and HER2-zero expressing states in patients with breast cancer. Based on the current research on the application of T2WI and DWI/ADC in breast cancer, the AUC value of this study is similar to the results of some other articles (22,23), and is higher than most of the previous literature (21,24).

Currently, studies have indicated that multi-parametric DCE-MRI imaging and radiomics models based on it have the potential to predict HER2 status, including the category of HER2-low expression (25-27). However, if MRI plain scan, as a more economical and convenient examination method, can provide effective biological information, the examination time will be significantly reduced, patient discomfort will be alleviated, and the occurrence of contrast agent allergies will be minimized. Previously, it has been shown that diffusion spectral imaging (DSI) is helpful for the prediction of HER2 status in breast cancer (28). Other diffusion models, including DWI, diffusion tensor imaging (DTI), and diffusion kurtosis imaging (DKI), have also been utilized to predict molecular subtypes of breast cancer (22,29). Moreover, non-contrast T2WI features have also been explored for predicting axillary lymph node metastasis in breast cancer (30). Therefore, we attempted to predict HER2-zero, HER2-low, and HER2-overexpressing breast cancer using a model based on non-contrast MRI combined with clinical features.

Our results indicate that there is a significant difference in edema scores between the HER2-low expressing group and the HER2-overexpressing group, with the HER2-overexpressing group having higher edema scores compared to the HER2-low expressing group. In the HER2-low expressing group, the proportion of BES 3–4 (high degree of edema) was 18.8%, whereas in the HER2-overexpressing group, the proportion of BES 3–4 was 26.5%. Breast edema is primarily the result of impaired outflow due to the obstruction or destruction of the lymphatic vessels that drain the breast (31). Peritumoral edema is significantly associated with cancer cells invading lymphatic vessels, blood vessels, and other histopathological changes (32). Lesions with a higher histological grade and HER2 positivity are more likely to exhibit focal edema, and the degree of edema is more severe (18,33). The reason may be that the HER2 gene can induce the expression of vascular endothelial growth factor, which in turn promotes the formation of new blood vessels, increases microcirculation and peritumoral extracellular fluid perfusion within the tumor, leading to edema around the tumor (24). Our study results show greater edema in the HER2-overexpressing group than in the HER2-low group, suggesting higher malignancy and invasiveness of the HER2-overexpressing group, consistent with prior research.

The results of this study showed that the L/M ratio of the HER2-low expressing group was lower than that of the HER2-overexpressing group (P=0.005). Previous research has found that on short-time inversion recovery (STIR) images, the L/M ratio of invasive breast cancer with less fibrous stroma is higher than that of the control group, suggesting that the SI of breast cancer gradually decreases as the fibrous component increases (34). Consequently, HER2-low expressing breast cancer may exhibit higher fiber content compared to HER2-overexpressing subtypes. It should be noted that the available literature lacks HER2 classification details, and there is a difference in principle between the STIR sequence—which suppresses fat by exploiting T1 relaxation time differences—and the T2FS imaging used in our research.

Meanwhile, hypointensity on T2WI is not entirely fibrous component, it can also represent old hemorrhage or coagulative necrosis. The relatively elevated T2 SI in the HER2 overexpressing group may also be due to increased necrosis and edema (35). Although careful measures were taken to exclude necrotic regions during region of interest (ROI) selection, the potential impact of these confounding factors cannot be entirely eliminated.

The study found that HER2-zero breast cancer has a lower hormone receptor (HR) positive rate, a higher histological grade, and is more prone to necrotizing cystic changes (24), which may be one of the reasons for the higher T2 signal in the HER2-zero expressing group than in the HER2-low expressing group. Furthermore, the HER2-zero expressing group had a sample size about one-third of the HER2-low group, and this substantial difference in sample sizes might also partly explain the observed results.

DWI is a non-contrast MRI technique used to measure the diffusion of water molecules in living organisms. Malignant lesions often exhibit impaired water diffusion and lower ADCs on DWI due to their higher cellular architecture compared to benign and healthy tissues (36). In this study, the ADC value of the HER2-low group was lower than that of the HER2-over group (P<0.001), suggesting that the HER2-low group may have more densely packed cellular structures with restricted diffusion. However, it has also been suggested that the HER2-overexpression group may have increased ADC value due to significant tumor edema and higher vascular permeability (37). A meta-analysis including 2,990 tumors showed no significant difference in ADC values between different immunohistochemical subtypes (38). We also found that no statistically significant difference in ADC values between HER2-zero and HER2-low groups. The current literature is inconclusive. We noted the potential of ADC’s quantitative nature to reflect the cellular state of tissue, in that it can complement the morphological information provided by T2FS. However, recent evidence suggests that ADC is not a pure diffusion coefficient, given its substantial modulation by tissue T2 relaxation time (the so-called T2 shine-through effect) (39,40). This merits further exploration in future studies.

Our study still has some limitations. Firstly, the evaluation of tumor edema degree is subjective, but the consistency between our observers is perfect. Secondly, the L/M ratio is based on the SI of T2FS images. There is a certain selection bias in the delineation of ROI, and the images come from different MR machines, which may cause measurement errors. An additional limitation is that the MR images, acquired from three different scanners, exhibit inconsistencies in some of the acquisition b-values. Although this mirrors real-world clinical practice in most hospitals and may influence ADC measurements, the resultant variability warrants further investigation. Moreover, the effect of inter-scanner variability on the reproducibility of quantitative metrics should not be overlooked. The limited sample size in some subgroups is an important limitation of this study, which may lead to unstable effect estimates, excessively wide confidence intervals, or non-estimable values. The confidence intervals for some predictors overlapped, which may be attributable to collinearity or to the fact that they reflect similar underlying pathophysiological processes—an issue that certainly warrants further investigation. The proposed imaging model is not intended to supersede pathological assessment. We also emphasize that clinical implementation would require the following: (I) external validation of cutoffs and performance in multi-center cohorts; (II) prospective studies evaluating the impact of MRI-based stratification on clinical decision-making and patient outcomes compared to standard care.

Future research will focus on expanding the sample size to ensure equitable representation across HER2 groups and reducing technical heterogeneity (scanners and sequences) to yield more definitive conclusions. In our study, we have interpreted the ADC changes as reflecting diffusion alterations. However, we acknowledge that T2 effects may have contributed to the observed SI and quantitative values. Future studies incorporating T2 correction models or advanced diffusion techniques might be necessary to disentangle these effects. Besides, due to the retrospective nature of this single-center study, we did not conduct a centralized pathological review of the HER2 status. Our research relied on existing medical records and pathology reports generated in our institution’s routine clinical practice; however, the HER2 status of all participants was reclassified based on the original pathology reports in strict accordance with the 2023 ASCO/CAP guidelines for HER2 reclassification. Although the 2023 ASCO/CAP guidelines were strictly applied to reclassify HER2 status and ensure consistency, potential inter-observer variability among pathologists may still exist, an issue we recognize and will focus on in future efforts to refine our research design.

Conclusions

As non-invasive imaging modalities, T2WI and ADC demonstrate potential for application in the auxiliary evaluation of HER2 expression status in breast cancer. This study utilized a non-contrast MRI protocol combining T2WI and ADC, offering three distinct advantages: elimination of gadolinium-based contrast agents, making it suitable for patients with renal insufficiency while avoiding contrast-related adverse reactions; significantly reduced scan time and costs compared with DCE-MRI; and minimal post-processing requirements, facilitating easy adoption in primary healthcare settings. This method features operational simplicity and notable cost-effectiveness. Nevertheless, given the exploratory nature of current evidence, more rigorous investigations are warranted to establish its definitive value and appropriate indications for assessing HER2 expression status.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2547/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2547/dss

Funding: This work was supported by Five-Five Project of the Third Affiliated Hospital of Sun Yet-sen University (No. 2023WW605).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1-2547/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Medical Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University (No. 1I2023-285-02). In view of the retrospective design and minimal risk to participants, the Ethics Committee granted a waiver of the requirement to obtain informed consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Hoxha I, Sadiku F, Hoxha L, Nasim M, Christine Buteau MA, Grezda K, Chamberlin MD. Breast Cancer and Lifestyle Factors: Umbrella Review. Hematol Oncol Clin North Am 2024;38:137-170. [Crossref] [PubMed]
2. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin 2024;74:12-49. [Crossref] [PubMed]
3. Denkert C, Seither F, Schneeweiss A, Link T, Blohmer JU, Just M, et al. Clinical and molecular characteristics of HER2-low-positive breast cancer: pooled analysis of individual patient data from four prospective, neoadjuvant clinical trials. Lancet Oncol 2021;22:1151-61. [Crossref] [PubMed]
4. Kang S, Kim SB. HER2-Low Breast Cancer: Now and in the Future. Cancer Res Treat 2024;56:700-20. [Crossref] [PubMed]
5. Ivanova M, Porta FM, D'Ercole M, Pescia C, Sajjadi E, Cursano G, De Camilli E, Pala O, Mazzarol G, Venetis K, Guerini-Rocco E, Curigliano G, Viale G, Fusco N. Standardized pathology report for HER2 testing in compliance with 2023 ASCO/CAP updates and 2023 ESMO consensus statements on HER2-low breast cancer. Virchows Arch 2024;484:3-14. [Crossref] [PubMed]
6. Fan P, Xu K. Antibody-drug conjugates in breast cancer: Marching from HER2-overexpression into HER2-low. Biochim Biophys Acta Rev Cancer 2023;1878:188849. [Crossref] [PubMed]
7. Prat A, Bardia A, Curigliano G, Hammond MEH, Loibl S, Tolaney SM, Viale G. An Overview of Clinical Development of Agents for Metastatic or Advanced Breast Cancer Without ERBB2 Amplification (HER2-Low). JAMA Oncol 2022; Epub ahead of print. [Crossref]
8. Kazerouni AS, Hormuth DA 2nd, Davis T, Bloom MJ, Mounho S, Rahman G, Virostko J, Yankeelov TE, Sorace AG. Quantifying Tumor Heterogeneity via MRI Habitats to Characterize Microenvironmental Alterations in HER2+ Breast Cancer. Cancers (Basel) 2022;14:1837. [Crossref] [PubMed]
9. Yin L, Zhang Y, Wei X, Shaibu Z, Xiang L, Wu T, Zhang Q, Qin R, Shan X. Preliminary study on DCE-MRI radiomics analysis for differentiation of HER2-low and HER2-zero breast cancer. Front Oncol 2024;14:1385352. [Crossref] [PubMed]
10. Wekking D, Porcu M, De Silva P, Saba L, Scartozzi M, Solinas C. Breast MRI: Clinical Indications, Recommendations, and Future Applications in Breast Cancer Diagnosis. Curr Oncol Rep 2023;25:257-67. [Crossref] [PubMed]
11. Li X, Fan Z, Jiang H, Niu J, Bian W, Wang C, Wang Y, Zhang R, Zhang H. Synthetic MRI in breast cancer: differentiating benign from malignant lesions and predicting immunohistochemical expression status. Sci Rep 2023;13:17978. [Crossref] [PubMed]
12. Pötsch N, Sodano C, Baltzer PAT. Performance of Diffusion-weighted Imaging-based Noncontrast MRI Protocols for Diagnosis of Breast Cancer: A Systematic Review and Meta-Analysis. Radiology 2024;311:e232508. [Crossref] [PubMed]
13. Mao C, Hu L, Jiang W, Qiu Y, Yang Z, Liu Y, Wang M, Wang D, Su Y, Lin J, Yan X, Cai Z, Zhang X, Shen J. Discrimination between human epidermal growth factor receptor 2 (HER2)-low-expressing and HER2-overexpressing breast cancers: a comparative study of four MRI diffusion models. Eur Radiol 2024;34:2546-59. [Crossref] [PubMed]
14. Liu W, Yang Y, Wang X, Li C, Liu C, Li X, Wen J, Lin X, Qin J. A Comprehensive Model Outperformed the Single Radiomics Model in Noninvasively Predicting the HER2 Status in Patients with Breast Cancer. Acad Radiol 2025;32:24-36. [Crossref] [PubMed]
15. Conti A, Duggento A, Indovina I, Guerrisi M, Toschi N. Radiomics in breast cancer classification and prediction. Semin Cancer Biol 2021;72:238-50. [Crossref] [PubMed]
16. Qi YJ, Su GH, You C, Zhang X, Xiao Y, Jiang YZ, Shao ZM. Radiomics in breast cancer: Current advances and future directions. Cell Rep Med 2024;5:101719. [Crossref] [PubMed]
17. Harada TL, Uematsu T, Nakashima K, Sugino T, Nishimura S, Takahashi K, Hayashi T, Tadokoro Y, Watanabe J, Nakamoto S, Ito T. Is the presence of edema and necrosis on T2WI pretreatment breast MRI the key to predict pCR of triple negative breast cancer? Eur Radiol 2020;30:3363-70. [Crossref] [PubMed]
18. Chen Y, Wang L, Luo R, Liu H, Zhang Y, Wang D. Focal breast edema and breast edema score on T2-weighted images provides valuable biological information for invasive breast cancer. Insights Imaging 2023;14:73. [Crossref] [PubMed]
19. Chen Y, Wang J, Zhang X, Yang W, Chen H, Bao B, Qiu Y, Tian L. Correlation between apparent diffusion coefficient and pathological characteristics of patients with invasive breast cancer. Ann Transl Med 2021;9:143. [Crossref] [PubMed]
20. Chen P, Zhao S, Guo W, Shao G. Dynamic contrast-enhanced magnetic resonance imaging features and apparent diffusion coefficient value of HER2-positive/HR-negative breast carcinoma. Quant Imaging Med Surg 2023;13:4816-25. [Crossref] [PubMed]
21. Xu Z, Ding Y, Zhao K, Han C, Shi Z, Cui Y, Liu C, Lin H, Pan X, Li P, Chen M, Wang H, Deng X, Liang C, Xie Y, Liu Z. MRI characteristics of breast edema for assessing axillary lymph node burden in early-stage breast cancer: a retrospective bicentric study. Eur Radiol 2022;32:8213-25. [Crossref] [PubMed]
22. Du S, Gao S, Zhang L, Yang X, Qi X, Li S. Improved discrimination of molecular subtypes in invasive breast cancer: Comparison of multiple quantitative parameters from breast MRI. Magn Reson Imaging 2021;77:148-58. [Crossref] [PubMed]
23. Xie S, Tang W, Zhang C, Wang J, Wang M, Zhou Y. Classification of breast edema on T2-weighted imaging for predicting sentinel lymph node metastasis and biological behavior in breast cancer. Clin Radiol 2024;79:e1003-9. [Crossref] [PubMed]
24. Zhao S, Wang S, Li Y, Wu Y, Zhang M, Ning N, Liang H, Dong D, Yang J, Gao X, Guan H, Zhang L. Quantitative Parameters of Intravoxel Incoherent Movement Imaging and Dynamic Contrast Enhancement MRI for the Prediction of HER2-Zero, -Low, and -Positive Breast Cancers. Acad Radiol 2025;32:1851-60. [Crossref] [PubMed]
25. Zheng S, Yang Z, Du G, Zhang Y, Jiang C, Xu T, Li B, Wang D, Qiu Y, Lin D, Zhang X, Shen J. Discrimination between HER2-overexpressing, -low-expressing, and -zero-expressing statuses in breast cancer using multiparametric MRI-based radiomics. Eur Radiol 2024;34:6132-44. [Crossref] [PubMed]
26. Peng Y, Zhang X, Qiu Y, Li B, Yang Z, Huang J, Lin J, Zheng C, Hu L, Shen J. Development and Validation of MRI Radiomics Models to Differentiate HER2-Zero, -Low, and -Positive Breast Cancer. AJR Am J Roentgenol 2024;222:e2330603. [Crossref] [PubMed]
27. Liu W, Liu C, Yang Y, Chen Y, Muhetaier A, Lin Z, Weng Z, Wang X, Zhang P, Qin J. Combining conventional magnetic resonance imaging (MRI) parameters with clinicopathologic data for differentiation of the three-tiered human epidermal growth factor receptor 2 (HER2) status in breast cancer. Clin Radiol 2025;86:106955. [Crossref] [PubMed]
28. Mao C, Jiang W, Huang J, Wang M, Yan X, Yang Z, Wang D, Zhang X, Shen J. Quantitative Parameters of Diffusion Spectrum Imaging: HER2 Status Prediction in Patients With Breast Cancer. Front Oncol 2022;12:817070. [Crossref] [PubMed]
29. Meng N, Wang X, Sun J, Han D, Bai Y, Wei W, Wang Z, Jia F, Wang K, Wang M. A comparative study of the value of amide proton transfer-weighted imaging and diffusion kurtosis imaging in the diagnosis and evaluation of breast cancer. Eur Radiol 2021;31:1707-17. [Crossref] [PubMed]
30. Liu Y, Luo H, Wang C, Chen X, Wang M, Zhou P, Ren J. Diagnostic performance of T2-weighted imaging and intravoxel incoherent motion diffusion-weighted MRI for predicting metastatic axillary lymph nodes in T1 and T2 stage breast cancer. Acta Radiol 2022;63:447-57. [Crossref] [PubMed]
31. Liang T, Hu B, Du H, Zhang Y. Predictive value of T2-weighted magnetic resonance imaging for the prognosis of patients with mass-type breast cancer with peritumoral edema. Oncol Lett 2020;20:314. [Crossref] [PubMed]
32. Gupta SS, Mayrovitz HN. The Breast Edema Enigma: Features, Diagnosis, Treatment, and Recommendations. Cureus 2022;14:e23797. [Crossref] [PubMed]
33. Jirarayapong J, Portnow LH, Chikarmane SA, Lan Z, Gombos EC. High Peritumoral and Intratumoral T2 Signal Intensity in HER2-Positive Breast Cancers on Preneoadjuvant Breast MRI: Assessment of Associations With Histopathologic Characteristics. AJR Am J Roentgenol 2024;222:e2330280. [Crossref] [PubMed]
34. Yamaguchi K, Hara Y, Kitano I, Hamamoto T, Kiyomatsu K, Yamasaki F, Yamaguchi R, Nakazono T, Egashira R, Imaizumi T, Irie H. Relationship between MRI findings and invasive breast cancer with podoplanin-positive cancer-associated fibroblasts. Breast Cancer 2021;28:572-80. [Crossref] [PubMed]
35. Costantini M, Belli P, Distefano D, Bufi E, Matteo MD, Rinaldi P, Giuliani M, Petrone G, Magno S, Bonomo L. Magnetic resonance imaging features in triple-negative breast cancer: comparison with luminal and HER2-overexpressing tumors. Clin Breast Cancer 2012;12:331-9. [Crossref] [PubMed]
36. Gullo RL, Partridge SC, Shin HJ, Thakur SB, Pinker K. Update on DWI for Breast Cancer Diagnosis and Treatment Monitoring. AJR Am J Roentgenol 2024;222:e2329933. [Crossref] [PubMed]
37. Horvat JV, Bernard-Davila B, Helbich TH, Zhang M, Morris EA, Thakur SB, Ochoa-Albiztegui RE, Leithner D, Marino MA, Baltzer PA, Clauser P, Kapetas P, Bago-Horvath Z, Pinker K. Diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) mapping as a quantitative imaging biomarker for prediction of immunohistochemical receptor status, proliferation rate, and molecular subtypes of breast cancer. J Magn Reson Imaging 2019;50:836-46. [Crossref] [PubMed]
38. Meyer HJ, Wienke A, Surov A. Diffusion-Weighted Imaging of Different Breast Cancer Molecular Subtypes: A Systematic Review and Meta-Analysis. Breast Care (Basel) 2022;17:47-54. [Crossref] [PubMed]
39. Xu FY, Xiao BH, Wáng YXJ. The rationale for proposing a magnetic resonance slow diffusion metric and its proof-of-concept testing showing spleen parenchyma and hepatocellular carcinoma have faster diffusion than liver parenchyma. Quant Imaging Med Surg 2025;15:3792-806. [Crossref] [PubMed]
40. Wáng YXJ. An explanation for the triphasic dependency of apparent diffusion coefficient (ADC) on T2 relaxation time: the multiple T2 compartments model. Quant Imaging Med Surg 2025;15:3779-91. [Crossref] [PubMed]