Intra- and Peritumoral Radiomics for Predicting Equivocal HER2 (IHC2+) Status of Breast Cancer on Contrast-Enhanced Mammography

Intra- and Peritumoral Radiomics for Predicting Equivocal HER2 (IHC2+) Status of Breast Cancer on Contrast-Enhanced Mammography | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Intra- and Peritumoral Radiomics for Predicting Equivocal HER2 (IHC2+) Status of Breast Cancer on Contrast-Enhanced Mammography Cong Xu, Juan Qiu, Shijie Zhang, Yuqian Chen, Ziyin Li, Xiaodong Wang, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7828102/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background Identification of Human epidermal growth factor receptor 2 (HER2) status is significant for the treatment and prognosis of breast cancer patients. The study aimed to evaluate the equivocal HER2 (IHC 2+) status of breast cancer using intra- and peritumoral radiomics features of contrast-enhanced mammography (CEM). Methods A total of 131 breast cancer patients with equivocal HER2 (IHC 2+) status of breast cancer were enrolled in the study and divided into training (n = 84), internal test (n = 22) and prospective test (n = 25) cohorts. Radiomics features were extracted from intratumoral and peritumoral regions on CEM and were selected using low variance and least absolute shrinkage and selection operator regression (LASSO). Five radiomics signatures were established based on different intratumoral and peritumoral regions. The nomogram was constructed using the selected signatures and clinical factors by logistic regression analysis. Its predictive performance was compared with the radiomics model and the clinical model. The area under the receiver operator characteristic curve (AUC), sensitivity, specificity, accuracy, the calibration curve, and decision curve analysis (DCA) were used to evaluate predictive performance of the models. Results The intratumoral signature, 5mm-peritumoral signature, and tumor diameter were used to establish nomogram. Compared to the radiomics model and the clinical model, the nomogram achieved optimal predictive performance, with an AUC of 0.893 in the internal test cohort and an AUC of 0.840 in the prospective test cohort. The calibration curves and DCA showed favorable predictive performance of the nomogram. Conclusions The nomogram incorporated the intratumoral and peritumoral radiomics signatures of CEM and clinical risk variables has the potential to predict equivocal HER2 (IHC 2+) status of breast cancer preoperatively. breast cancer equivocal HER2 status radiomics intratumoral peritumoral Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Human epidermal growth factor receptor 2 (HER2)-positive breast cancer is mainly an aggressive subtype of breast cancer that accounts for about 15–20% of all breast cancers[ 1 ]. The HER2 gene is an independent prognostic factor for breast cancer recurrence and survival, a series of studies confirmed that the overexpression and amplification of the HER2 gene are closely associated with the effectiveness of breast cancer treatment[ 2 – 4 ]. Randomized clinical trials have demonstrated that adding anti-HER2-positive therapy to the treatment of HER2-positive breast cancer patients is an important factor for improving patient outcomes, which were reflected in an increase in the pathological complete response (pCR) rate, overall survival rate, and disease-free survival rate, as well as a reduction in the risk of disease recurrence or death[ 5 – 7 ]. Therefore, the timely identification of HER2 status is of paramount importance in order to inform the subsequent course of treatment. Immunohistochemical (IHC) detection of protein overexpression and fluorescence in situ hybridization (FISH) analysis of HER2 gene amplification are two principal methods for determining HER2 status[ 8 ]. HER2 positivity is defined either by protein overexpression as defined by IHC3 + or equivocal protein expression (IHC2+) with evidence of HER2 gene amplification[ 9 ]. IHC 2 + was equivocal and required FISH to ascertain whether it was amplified[ 10 ]. In the clinic, the IHC test is relatively straightforward. However, additional analysis, such as FISH, undoubtedly results in time-consuming procedures and associated additional costs. It also requires specialized equipment and technical expertise. Consequently, a convenient and non-invasive modality to predict equivocal HER2 (IHC 2+) status is required. Imaging is an essential tool in medical science and is routinely used in clinical practice for tumor detection and treatment guidance[ 11 ]. Contrast-enhanced mammography (CEM) represents a cutting-edge technique that utilizes dual energy exposure, which has the advantage of low cost, time-saving, and acceptable tolerance in breast cancer patients. It could not only display the morphological characteristics, but also provide the blood supply information of the lesions. Moreover, CEM has higher sensitivity and specificity in the diagnosis of breast cancer than mammography and comparable to dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI)[ 12 – 14 ]. However, it is difficult to assess equivocal HER2 (IHC 2+) status using only the image features visible to the naked eye. Radiomics is a promising tool that can convert medical images into mineable data by extracting high-throughput quantitative features[ 15 ]. As a non-invasive and effective method, radiomics has been widely used to assess whole tumor heterogeneity[ 16 ], and some researches have attempted to use radiomics to predict HER2 status in breast cancer[ 17 – 19 ]. However, these studies were conducted based on MRI or mammography, not CEM. Besides, the tumor microenvironment plays a role in the development and progression of breast cancer. Some literatures have explored the value of peritumoral radiomics in breast cancer diagnosis and prognosis[ 20 – 22 ]. Empirical evidence implicates that the microenvironment may contain information that is relevant to the treatment of HER2-positive breast cancer[ 23 , 24 ]. However, the value of peritumour radiomics based on CEM in predicting equivocal HER2 (IHC2+) status is unclear. Therefore, the study aims to explore the value of intratumoral and peritumoral radiomics based on CEM for predicting equivocal HER2 (IHC2+) status in breast cancer patients. Methods Study population Ethical approval was obtained for retrospective study, and the requirement for patient informed consent requirement was waived. A prospective study was approved by the institutional ethics committee of the hospital. Written informed consents were obtained from patients whose CEM images were prospectively collected. (Clinical trial number: not applicable) We retrospectively screened the CEM images between January 2021 and January 2022 in the institution from the picture archiving and communication system. The inclusion criteria for the retrospective study were as follows: (1) breast cancer patients pathologically confirmed by biopsy or surgical specimens; (2) underwent CEM less than 2 weeks before surgery or other treatment options; (3) patients with a single lesion; and (4) HER2 score of 2 + verified by IHC and have the results of FISH. The exclusion criteria were as follows: (1) patients who underwent breast radiotherapy, chemotherapy, or hormone treatment before CEM examination; (2) incomplete clinical, pathological, IHC or FISH information of patients; and (3) poor image quality. A total of 106 female patients were ultimately included in this study and divided into training cohort and internal test cohort at a ratio of 8:2. In addition, 25 patients were prospectively evaluated at our hospital from March 2022 to June 2022. For the prospective study, the inclusion criteria were as follows: (1) patients were confirmed by biopsy as breast cancer; (2) enrolled patients were planned to receive standard treatment or surgery in our hospital; (3) CEM examinations were available and performed within two weeks before surgery or other therapeutic regimes; and (4) HER2 score of 2 + verified by IHC and have the results of FISH. The exclusion criteria were as follows: (1) patients who had any previous history of cancer and treatment; (2) patients with incomplete clinical, pathological, IHC or FISH information; and (3) inadequate image quality or non-mass lesions which affected the observation and delineation. Information regarding the clinical characteristics of the patients, including age and other parameters, were obtained from the electronic medical record. CEM examination All patients underwent CEM examination using a GE Senographe Essential mammography unit (GE Healthcare, Milwaukee, WI, USA). The contrast agent Omnipaque 300 (GE Healthcare, Inc., Princeton, NJ) was injected into the upper arm vein with the dose of 1.5 ml/kg and the injection flow rate of 3.0 ml/s. Two minutes after the contrast injection, images were obtained in the following order: craniocaudal (CC) and mediolateral oblique (MLO) views of the suspicious breast, and then CC and MLO views of the less suspicious breast. For each mammographic projection, a pair of high-energy (HE) and low-energy (LE) exposures were consecutively performed to obtain HE and LE images, and a recombined (RC) image was generated automatically from LE and HE images using a dual-energy weighted logarithmic subtraction technique[ 25 ]. Pathological Assessment According to the recommendations by the American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) guidelines, IHC analyses were performed to determine the expression levels of HER2 in each breast cancer patient. A HER2 staining intensity score of 3 + was considered positive, while a score of 0 or 1 + was considered negative. A HER2 staining intensity score of 2+, with confirmation of gene amplification by FISH, was also deemed positive HER2 [ 10 , 26 , 27 ]. Image segmentation All radiologists who participated in the image segmentation were blinded to the clinical and histopathological information. A dedicated radiologist (Segmenter 1) with seven years’ experience of breast imaging annotated tumor boundaries as the intratumoral region (ITR) in the LE and RC images with standard CC view via the ITK-SNAP (version 3.8.0; www.itksnap.org ) software. All the contours were reviewed by a radiologist with 15 years of experience. Meanwhile, 40 patients randomly selected from the training cohort were used to assess the consistency of the inter- and intra-observer segmentation. Another 2 two radiologists (Segmenter 2 and 3) with 8 and 10 years of experience, respectively, performed the segmentation work utilizing the same methods. Additionally, the radiologist (Segmenter 1) repeated the segmentation with an interval of two weeks. The Dice similarity coefficient (DSC) was used to evaluate the agreement both inter- and intra-observer segmentation. Average DSCs of 0.873 and 0.925 were achieved for inter- and intra-observer segmentation performances, respectively. Meanwhile, the largest diameters of the lesions were measured independently by the two radiologists in CC view images. The mean values were calculated as the final tumor diameter. After the ITR mask was annotated, a morphologic operation of dilation was performed to capture the peritumoral regions (PTR) outside the tumor of 5 mm and 10 mm using Python (version 3.6.6). If the contours of peritumoral regions exceeded the breast parenchyma after extension, the beyond portion was manually removed[ 21 ]. Additionally, we define the intra- and peritumoral region in the same ROI named as IPTR. Finally, for each lesion in each image, five ROIs, namely ITR, PTR5 (5-mm peritumoral region), PTR10 (10-mm peritumoral region), IPTR5 (intratumoral region + 5-mm peritumoral region) and IPTR10 (intratumoral region + 10-mm peritumoral region), were generated to extract radiomics features. Example segmentations are presented in Fig. 1 . Radiomics feature extraction Image preprocessing was conducted by Python (version 3.6.6) before radiomics feature extraction, including image resampling and normalization, and gray-level discretization. A total of 1316 quantitative radiomics features were extracted from each ROI in each image using the radiomics function RadiomicsFeatureExtractor toolkit provided by Pyradiomics[ 28 ], including 252 first-order statistics, 14 shape features, 336 gray-level co-occurrence matrix (GLCM), 224 gray-level run length matrix (GLRLM), 224 gray-level size zone matrix (GLSZM), 196 gray-level dependence matrix (GLDM), 70 neighboring gray-tone difference matrix (NGTDM). Since LE-CC and RC-CC images were analyzed, a total of 2632 radiomics features were extracted from each ROI (ITR, PTR5, PTR10, IPTR5, IPTR10). The intraclass correlation coefficient (ICC) was used to evaluate the reproducibility of manual radiomics feature extraction. To calculate the intra- and inter-observer agreement of feature extraction, two radiologists (Radiologist 1 and Radiologist 2) firstly extracted the radiomics features using 40 randomly chosen patients to calculate the ICCs, respectively. Two weeks later, the procedure was repeated by Radiologist 1 and the remaining images were also analyzed by Radiologist 1. In this study, the radiomics features with ICCs of 0.80 or greater were selected and considered a mark of satisfactory intra- and inter-observer agreement. Feature selection and radiomics signature building All feature selection process was performed in the training cohort. Low variance was first used to filter features, the variance of each feature is calculated, and if it is below a threshold (0.80) we filter it out. Then, the least absolute shrinkage and selection operator (LASSO) logistic regression method was applied to select and identify the most stable and predictive features and to construct the radiomics signatures. To avoid over-fitting, the best parameter of the LASSO regularization parameter (α) was determined via 10-fold cross-validation. The radiomics signature score reflecting the respective equivocal HER2 (IHC2+) status for each patient was calculated through a linear combination of selected features weighted by their respective coefficients. On the basis of this procedure, five signatures were constructed from the five ROIs. The sensitivity and specificity values for evaluating the performance of the radiomics signatures in all cohorts were plotted to generate a receiver operator characteristic (ROC) curve, and the area under the curve (AUC) was calculated. Radiomics nomogram construction and comparison Firstly, the potential predictors among clinical variables and five radiomics signatures were identified using univariate logistic regression. Then these predictors were integrated into the multivariate logistic regression analysis with backward-stepwise selection based on minimal Akaike information criteria, which was used to select independent predictors of positive equivocal HER2 (IHC2+) status. A radiomics nomogram was established based on the independent predictors. For comparison, a clinical model was also built using only the selected clinical variables, and a radiomics model was built using only the selected radiomics signatures. To prove the generalization of the nomogram, a prospective cohort was used to test the models. The AUC, sensitivity, specificity, and accuracy were used to evaluate the predictive performance of the models. The study flow chart is shown in Fig. 2 . Statistical analysis All statistical analyses were conducted with Python (version 3.6.6) and R software (version 4.0.3). Continuous variables were compared by two-sample t-test, while qualitative variables were analyzed by Chi-square test. LASSO regression and ROC curve analysis were conducted using the “LassoCV” function and “roc_curve” packages. The ROC curves were plotted to evaluate the performance of the models and the AUCs were calculated. The Youden index was chosen as the optimal cut-off value [ 29 ]. Decision curve analysis (DCA) was employed to assess the net benefit of the models in a clinical context. The Hosmer-Lemeshow goodness of fit test was used to evaluate the calibration of the models and the calibration curve was plotted. DCA and calibration curve were performed using the “rmda” and “rms” packages. The DeLong test [ 30 ] was applied to compare the AUCs of different models. P < 0.05 was considered a statistically significant difference. Results Clinicopathological characteristics A total of 131 patients were enrolled in this study. The training cohort included 84 patients (28 patients with positive IHC 2 + status, 56 patients with negative IHC 2 + status), the average age of patients was 54.2 ± 8.2 years old (range, 33–74 years old). The internal test cohort included 22 patients (8 patients with positive IHC 2 + status, 14 patients with negative IHC 2 + status), the average age of patients was 55.1 ± 8.9 years old (range, 38–75 years old). The prospective test cohort included 25 patients (10 patients with positive IHC 2 + status, 15 patients with negative IHC 2 + status), the average age of patients was 57.2 ± 11.1 years old (range, 32–80 years old). The results of the clinicopathological features between the patients with negative and positive equivocal HER2 (IHC 2+) status are listed in Table 1 . Table 1 Characteristics in the training, internal test, and prospective test cohorts Training cohort (n = 84) Internal test cohort (n = 22) Prospective test cohort (n = 25) HER2 2+ Positive (n = 28) HER2 2+ Negative (n = 56) p HER2 2+ Positive (n = 8) HER2 2+ Negative (n = 14) p HER2 2+ Positive (n = 10) HER2 2+ Negative (n = 15) p Age, years (mean ± SD) 52.3 ± 8.8 54.9 ± 7.8 0.267 53.8 ± 4.8 55.8 ± 10.4 0.060 56.2 ± 10.8 57.8 ± 11.2 0.799 Diameter, cm (mean ± SD) 2.6 ± 1.1 3.1 ± 0.9 0.090 2.2 ± 0.7 3.5 ± 0.7 < 0.010 3.7 ± 1.3 2.3 ± 0.9 0.001 ER 0.341 0.439 0.727 Negative 2 8 0 1 1 1 Positive 26 48 8 13 9 14 PR 1.000 0.674 0.105 Negative 6 12 1 1 4 2 Positive 22 44 7 13 6 13 Ki-67 0.701 0.339 < 0.01 Negative 5 12 0 1 0 7 Positive 23 44 8 13 10 8 Note: HER2, human epidermal growth factor receptor 2; ER, estrogen receptor; PR, progesterone receptor; SD: standard deviation. Performance of the radiomics signatures The extracted features were identified as being highly reproducible based on the intra- and inter-observer ICCs ranging from 0.915 to 0.945 and 0.901 to 0.935, respectively. After selecting radiomics features, a total of 10, 8, 9, 7, and 6 features were selected as the most valuable features from ITR, PTR5, PTR10, IPTR5, and IPTR10, respectively, and then the five radiomics signatures were constructed. The detailed features and their respective coefficients are shown in Supplementary Materials (Table S1 ). All five radiomics signatures showed encouraging results, with AUC values varying from 0.643 (95% confidence interval [CI], 0.364–0.921) to 0.866 (95% CI, 0.714-1.000) (Fig. 3 A and 3 B). Among them, the radiomics signature based on ITR yielded the highest AUC value of 0.866 (95% CI, 0.714-1.000) in the internal test cohort. The detailed results are presented in Supplementary Materials (Table S2). Performance of the nomogram, radiomics model, and clinical model To ascertain the value of radiomics signatures and clinical variables, univariate and multivariate logistic regression analyses were conducted. Two radiomics signatures (constructed from ITR and PTR5) and the diameter were identified as significant predictors for identifying equivocal HER2 (IHC 2+) status. The detailed results are represented in Table 2 . Then, the diameter, signature ITR and signature PTR5 were used to constructed the nomogram (Fig. 4 ). In addition, the clinical model based on diameter and the radiomics model based on signature ITR and signature PTR5 were also constructed, respectively. Table 2 Risk factors for predicting equivocal HER2 (IHC2+) status Variables Univariate logistic regression Multivariate logistic regression OR (95% CI) p OR (95% CI) p Signature ITR 8.240(4.397–15.440) < 0.001* 8.600(3.868–19.123) < 0.001* Signature PTR5 31.629(12.919–77.434) < 0.001* 3.490(1.062–11.467) 0.042* Signature PTR10 3.259(2.104–5.046) < 0.001* 0.878(0.495–1.558) 0.658 Signature IPTR5 37.303(11.509-120.909) < 0.001* 0.857(0.204–3.599) 0.833 Signature IPTR10 29.807(7.724-115.025) < 0.001* 1.140(0.307–4.233) 0.845 Age 0.987(0.976–0.999) 0.044* 0.993(0.983–1.003) 0.630 Diameter 0.891(0.807–0.983) 0.024* 1.125(1.008–1.256) 0.039* Note: OR, odds ratio; CI, confidence interval; ITR, intratumoral region; PTR5, 5-mm peritumoral region; PTR10, 10-mm peritumoral region; IPTR5, intratumoral region + 5-mm peritumoral region); IPTR10, intratumoral region + 10-mm peritumoral region * p < 0.05 was considered statistically significant. As shown in Fig. 5 A-C, the nomogram achieved relatively good discrimination. In the internal test cohort, the AUC of the nomogram reached 0.893 (95CI%: 0.756-1.000), higher than the radiomics model with 0.821 (95CI%: 0.641-1.000) (Delong test, P = 0.292) and the clinical model with 0.866 (95CI%: 0.681-1.000) (Delong test, P = 0.314). In the prospective test cohort, the AUC of the nomogram was 0.840 (95CI%: 0.652-1.000), higher than the radiomics model with 0.819(95CI%: 0.649–0.999) (Delong test, P = 0.676) and the clinical model with 0.774 (95CI%: 0.547-1.000) (Delong test, P = 0.878). In the prospective test cohort, the nomogram achieved the highest sensitivity of 0.923. Detailed results are presented in Table 3 . The calibration curves indicated that the radiomics nomogram had good calibration in the three cohorts (Fig. 6 A). Figure 6 B and 6 C showed the nomogram achieved a higher net benefit than the radiomics model and the clinical model in the internal test cohort. Table 3 Performances of Radiomics model, Clinical model, and Nomogram AUC(95%CI) SEN(95%CI) SPE(95%CI) ACC(95%CI) Training cohort Radiomics model 0.876(0.802–0.950) 0.741(0.601–0.846) 0.862(0.674–0.955) 0.783(0.679–0.866) Clinical model 0.675(0.534–0.816) 0.889(0.767–0.954) 0.586(0.391–0.759) 0.781(0.679–0.866) Nomogram 0.891(0.823–0.959) 0.870(0.745–0.942) 0.759(0.561–0.890) 0.831(0.733–0.905) Internal test cohort Radiomics model 0.821(0.641-1.000) 0.785(0.488–0.942) 0.750(0.355–0.955) 0.772(0.546–0.921) Clinical model 0.866(0.681-1.000) 1.000(0.732-1.000) 0.750(0.355–0.955) 0.909(0.708–0.988) Nomogram 0.893(0.756-1.000) 0.857(0.561–0.974) 0.875(0.466–0.993) 0.864(0.651–0.971) Prospective test cohort Radiomics model 0.819(0.649–0.99) 0.866(0.584–0.976) 0.700(0.353–0.919) 0.800(0.593–0.931) Clinical model 0.774(0.547-1.00) 0.846(0.536–0.973) 0.583(0.285–0.835) 0.720(0.506–0.879) Nomogram 0.840(0.652-1.00) 0.923(0.620–0.996) 0.667(0.354–0.887) 0.800(0.593–0.931) Note: AUC, area under curve; SEN, sensitivity; SPE, specificity; ACC, accuracy; CI, confidence interval Discussion To achieve effective therapeutic efficacy for HER2-targeted treatment, precise equivocal HER2 (IHC 2+) status identification prior to treatment is indispensable. Although both IHC and FISH are commonly used approaches to determine equivocal HER2 (IHC 2+) status, their clinical applications are limited mainly by biopsy availability, procedure complexity and low reproducibility. In this study, we developed a nomogram using the selected radiomics signatures based on CEM and clinical risk factors to predict equivocal HER2 (IHC 2+) status, which achieved the highest AUC compared to the radiomic model and the clinical model. The results indicated that the nomogram yielded good discrimination and calibration. Several studies have investigated that some abnormal imaging features, such as microcalcifications, breast density, or a spiculated mass on mammography are significantly associated with the HER2 status[ 31 , 32 ]. However, the performance of these features in predicting the HER2 status is limited and may be influenced by the subjective judgment of radiologists. Radiomics is an emerging field that can translate medical images into quantitative data for target task prediction. Multiple imaging methodologies, such as mammography and MRI, were revealed to be able to predict HER2 status for breast cancer based on radiomic features[ 17 – 19 ]. However, they did not focus on equivocal HER2 (IHC 2+) status and CEM. At present, there is no study on the prediction of equivocal HER2 (IHC 2+) status based on CEM. To obtain more valuable radiomics features, this study extracted the radiomics features of LE-CC and RC-CC images to establish radiomics signatures, and integrated clinical risk factors to construct a nomogram, which achieved promising predictive performance. This demonstrates the value of CEM in predicting equivocal HER2 (IHC 2+) status of breast cancer. Previous findings have shown that the peritumoral area can provide useful information to assist in the diagnosis of breast cancer, which is associated with lymphatic invasion and vascular infiltration[ 33 , 34 ]. Some researches demonstrated that radiomics features extracted from the peritumoral region could provide information different from those in the intratumoral region[ 24 , 35 , 36 ]. In this study, five radiomics signatures were established based on different intratumoral and peritumoral regions. The signature ITR achieved the best predictive performance, followed by the PTR5 signature in the internal test cohort, indicating that peritumoral radiomics based on CEM could provide valuable information for predicting equivocal HER2 (IHC 2+) status. The results showed that PTR10 had a relatively lower predictive performance than PTR5, which may be related to the fact that a larger peritumoral area contains more fat or mammary glands and less tumor information. Li et al[ 37 ] explored radiomics features of intratumoral and peritumoral regions on breast DCE-MRI to predict HER2 2 + status, and achieved a promising performance in the validation cohort (AUC = 0.840). However, it did not consider the role of clinical risk factors in the prediction task. In this study, the nomogram was developed based on radiomics features of intratumoral and peritumoral of CEM and clinical risk factor. Its predictive performance is superior to the single radiomics model or clinical model, indicating that both radiomics and clinical features have important value in predicting equivocal HER2 (IHC2+) status. Several recent studies have reported similar results, demonstrating the value of the nomogram established using radiomics signatures and clinical factors to evaluate pathological outcomes[ 38 , 39 ].In addition, considering the clinical applicability of the nomogram, an independent prospective test cohort was enrolled to prove the generalization ability of the nomogram. Good predictive power was achieved, demonstrating that the nomogram has promising prospects for clinical application. This work represents a preliminary success in the pre-treatment prediction of equivocal HER2 (IHC2+) status using intra- and peritumoral radiomics, which could potentially assist in guiding personalized treatment. However, our study had several limitations. First, this study is a single-center study with relatively inadequate sample size, although it used the internal and prospective test cohorts to test the nomogram and achieved promising results. A larger sample size from different centers is needed to further prove the generalization of the nomogram. Second, a manual segmentation method was employed in this study. Although favorable intra- and interobserver ICCs were obtained and an automated method was used for segmentation of peritumoral regions, which may have higher stability and be less time-consuming. Third, predictive models based on radiomics features were developed only from images of the CC view. It is valuable to explore whether the extracting features in different views impact the performance of the prediction models, such as MLO views and the combination of CC and MLO views. Conclusions In conclusion, the nomogram combined with CEM-based intratumoral and peritumoral radiomics signatures and clinical variables could predict equivocal HER2(IHC2+) status noninvasively and conveniently. This may effectively guide the personalized treatment of patients with breast cancer in clinical practice. Abbreviations CEM Contrast-enhanced mammography HER2 Human epidermal growth factor receptor IHC Immunohistochemical FISH Fluorescence in situ hybridization DSC Dice similarity coefficient ICC intraclass correlation coefficient CC Craniocaudal MLO Mediolateral oblique LASSO Least absolute shrinkage and selection operator ROC Receiver operator characteristic AUC Area under the curve DCA Decision curve analysis Declarations Ethics approval and consent to participate The study was conducted in accordance with the ethical standards of the Declaration of Helsinki. The study was approved by the ethics committee of Yantai Yuhuangding Hospital, the retrospective study waived the informed consent of patients, and the prospective study obtained the informed consent of patients. Consent for publication Not applicable. Competing interest The authors declare that they have no competing interests. Funding No funding. Author Contribution C.X., J.Q. and S.Z. contributed to the design of the concept and write the original draft. J.G., R.C. and J.L. reviewed and edited the draft. Y.C., H.Z., Q.W. and T.C. conducted data analysis. Z.L., X.W. and N.Q. collected the data information. H.X. and H.M. conducted data assessment. All authors read and approved the final manuscript. C.X., J.Q. and S.Z. contributed equally to this work. J.G., R.C. and J.L. contributed equally to this work. Acknowledgement We thank all participants for their hard work in this study. Data Availability The data presented in this study are available from the corresponding author by request. References Litton JK, Burstein HJ, Turner NC. Molecular Testing in Breast Cancer. American Society of Clinical Oncology educational book American Society of Clinical Oncology Annual Meeting. 2019, 39:e1-e7. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, Goldhirsch A, Untch M, Smith I, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med. 2005;353(16):1659–72. Slamon D, Eiermann W, Robert N, Pienkowski T, Martin M, Press M, et al. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365(14):1273–83. SLAMON DJ, LEYLAND-JONES B, SHAK S, FUCHS H, PATON V PHARMD, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med. 2001;344:783–92. Waks AG, Winer EP. Breast Cancer Treatment: A Review. JAMA. 2019;321(3):288–300. Rugo HS, Chien AJ. HER2-positive breast cancer: is more treatment better? Lancet Oncol. 2016;17(3):268–70. Cameron D, Piccart-Gebhart MJ, Gelber RD, Procter M, Goldhirsch A, de Azambuja E, et al. 11 years' follow-up of trastuzumab after adjuvant chemotherapy in HER2-positive early breast cancer: final analysis of the HERceptin Adjuvant (HERA) trial. Lancet. 2017;389(10075):1195–205. Moelans CB, de Weger RA, Van der Wall E, van Diest PJ. Current technologies for HER2 testing in breast cancer. Crit Rev Oncol/Hematol. 2011;80(3):380–92. Rakha EA, Pinder SE, Bartlett JM, Ibrahim M, Starczynski J, Carder PJ, et al. Updated UK Recommendations for HER2 assessment in breast cancer. J Clin Pathol. 2015;68(2):93–9. Wolff AC, Hammond MEH, Allison KH, Harvey BE, Mangu PB, Bartlett JMS, et al. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update. J Clin Oncol. 2018;36(20):2105–22. Lawson MB, Herschorn SD, Sprague BL, Buist DSM, Lee SJ, Newell MS, et al. Imaging Surveillance Options for Individuals With a Personal History of Breast Cancer: AJR Expert Panel Narrative Review. AJR Am J Roentgenol. 2022;219(6):854–68. Ghaderi KF, Phillips J, Perry H, Lotfi P, Mehta TS. Contrast-enhanced Mammography: Current Applications and Future Directions. Radiographics. 2019;39(7):1907–20. Lee-Felker SA, Tekchandani L, Thomas M, Gupta E, Andrews-Tang D, Roth A, et al. Newly Diagnosed Breast Cancer: Comparison of Contrast-enhanced Spectral Mammography and Breast MR Imaging in the Evaluation of Extent of Disease. Radiology. 2017;285(2):389–400. Sumkin JH, Berg WA, Carter GJ, Bandos AI, Chough DM, Ganott MA, et al. Diagnostic Performance of MRI, Molecular Breast Imaging, and Contrast-enhanced Mammography in Women with Newly Diagnosed Breast Cancer. Radiology. 2019;293(3):531–40. Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P et al. Radiomics: extracting more information from medical images using advanced feature analysis. European journal of cancer (Oxford, England: 1990). 2012, 48(4):441–446. Cao B, Mi K, Dai W, Liu T, Xie T, Li Q et al. Prognostic and incremental value of computed tomography-based radiomics from tumor and nodal regions in esophageal squamous cell carcinoma. Chinese journal of cancer research = Chung-kuo yen cheng yen chiu. 2022, 34(2):71–82. Zhou J, Tan H, Li W, Liu Z, Wu Y, Bai Y, et al. Radiomics Signatures Based on Multiparametric MRI for the Preoperative Prediction of the HER2 Status of Patients with Breast Cancer. Acad Radiol. 2021;28(10):1352–60. Bitencourt AGV, Gibbs P, Rossi Saccarelli C, Daimiel I, Lo Gullo R, Fox MJ, et al. MRI-based machine learning radiomics can predict HER2 expression level and pathologic response after neoadjuvant therapy in HER2 overexpressing breast cancer. EBioMedicine. 2020;61:103042. Zhou J, Tan H, Bai Y, Li J, Lu Q, Chen R, et al. Evaluating the HER-2 status of breast cancer using mammography radiomics features. Eur J Radiol. 2019;121:108718. Niu S, Jiang W, Zhao N, Jiang T, Dong Y, Luo Y, et al. Intra- and peritumoral radiomics on assessment of breast cancer molecular subtypes based on mammography and MRI. J Cancer Res Clin Oncol. 2022;148(1):97–106. Mao N, Shi Y, Lian C, Wang Z, Zhang K, Xie H, et al. Intratumoral and peritumoral radiomics for preoperative prediction of neoadjuvant chemotherapy effect in breast cancer based on contrast-enhanced spectral mammography. Eur Radiol. 2022;32(5):3207–19. Wang S, Sun Y, Li R, Mao N, Li Q, Jiang T, et al. Diagnostic performance of perilesional radiomics analysis of contrast-enhanced mammography for the differentiation of benign and malignant breast lesions. Eur Radiol. 2022;32(1):639–49. Salgado R, Denkert C, Campbell C, Savas P, Nuciforo P, Aura C, et al. Tumor-Infiltrating Lymphocytes and Associations With Pathological Complete Response and Event-Free Survival in HER2-Positive Early-Stage Breast Cancer Treated With Lapatinib and Trastuzumab: A Secondary Analysis of the NeoALTTO Trial. JAMA Oncol. 2015;1(4):448–54. Braman N, Prasanna P, Whitney J, Singh S, Beig N, Etesami M, et al. Association of Peritumoral Radiomics With Tumor Biology and Pathologic Response to Preoperative Targeted Therapy for HER2 (ERBB2)-Positive Breast Cancer. JAMA Netw Open. 2019;2(4):e192561. Bhimani C, Matta D, Roth RG, Liao L, Tinney E, Brill K, et al. Contrast-enhanced Spectral Mammography: Technique, Indications, and Clinical Applications. Acad Radiol. 2017;24(1):84–8. Allison KH, Hammond MEH, Dowsett M, McKernin SE, Carey LA, Fitzgibbons PL, et al. Estrogen and Progesterone Receptor Testing in Breast Cancer: ASCO/CAP Guideline Update. J Clin Oncol. 2020;38(12):1346–66. Nielsen TO, Leung SCY, Rimm DL, Dodson A, Acs B, Badve S, et al. Assessment of Ki67 in Breast Cancer: Updated Recommendations From the International Ki67 in Breast Cancer Working Group. J Natl Cancer Inst. 2021;113(7):808–19. van Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational Radiomics System to Decode the Radiographic Phenotype. Cancer Res. 2017;77(21):e104–7. Marcus D, Ruopp NJP, Brian W, Whitcomb, Enrique F. Schisterman. Youden index and optimal cut-point estimated from observations affected by a lower limit of detection. Biom J. 2008;50(50):419–30. R DE, M DD, L C-PD. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(3):837–45. Li E, Guida JL, Tian Y, Sung H, Koka H, Li M, et al. Associations between mammographic density and tumor characteristics in Chinese women with breast cancer. Breast Cancer Res Treat. 2019;177(2):527–36. Shin HJ, Kim HH, Huh MO, Kim MJ, Yi A, Kim H, et al. Correlation between mammographic and sonographic findings and prognostic factors in patients with node-negative invasive breast cancer. Br J Radiol. 2011;84(997):19–30. Schoppmann SF, Bayer G, Aumayr K, Taucher S, Geleff S, Rudas M, et al. Prognostic value of lymphangiogenesis and lymphovascular invasion in invasive breast cancer. Ann Surg. 2004;240(2):306–12. Ejlertsen B, Jensen MB, Rank F, Rasmussen BB, Christiansen P, Kroman N, et al. Population-based study of peritumoral lymphovascular invasion and outcome among patients with operable breast cancer. J Natl Cancer Inst. 2009;101(10):729–35. Zhou J, Zhang Y, Chang KT, Lee KE, Wang O, Li J, et al. Diagnosis of Benign and Malignant Breast Lesions on DCE-MRI by Using Radiomics and Deep Learning With Consideration of Peritumor Tissue. J Magn Reson Imaging. 2020;51(3):798–809. Wu J, Li B, Sun X, Cao G, Rubin DL, Napel S, et al. Heterogeneous Enhancement Patterns of Tumor-adjacent Parenchyma at MR Imaging Are Associated with Dysregulated Signaling Pathways and Poor Survival in Breast Cancer. Radiology. 2017;285(2):401–13. Li C, Song L, Yin J. Intratumoral and Peritumoral Radiomics Based on Functional Parametric Maps from Breast DCE-MRI for Prediction of HER-2 and Ki-67 Status. J Magn Reson Imaging. 2021;54(3):703–14. Yu Y, Tan Y, Xie C, Hu Q, Ouyang J, Chen Y, et al. Development and Validation of a Preoperative Magnetic Resonance Imaging Radiomics-Based Signature to Predict Axillary Lymph Node Metastasis and Disease-Free Survival in Patients With Early-Stage Breast Cancer. JAMA Netw Open. 2020;3(12):e2028086. Tang TY, Li X, Zhang Q, Guo CX, Zhang XZ, Lao MY, et al. Development of a Novel Multiparametric MRI Radiomic Nomogram for Preoperative Evaluation of Early Recurrence in Resectable Pancreatic Cancer. J Magn Reson Imaging. 2020;52(1):231–45. Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7828102","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":551232155,"identity":"86c76914-eaff-49be-a06c-8da29398b9ec","order_by":0,"name":"Cong Xu","email":"","orcid":"","institution":"Physical Examination Center, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Cong","middleName":"","lastName":"Xu","suffix":""},{"id":551232157,"identity":"394c623e-b661-4029-9053-03b944b9f876","order_by":1,"name":"Juan Qiu","email":"","orcid":"","institution":"Department of Breast, Guilin Hospital of traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Juan","middleName":"","lastName":"Qiu","suffix":""},{"id":551232166,"identity":"da24f3a8-e560-4bac-a3f8-dc325389840e","order_by":2,"name":"Shijie Zhang","email":"","orcid":"","institution":"Department of Radiology, Huangshi Central Hospital, Affiliated Hospital of Hubei Polytechnic University","correspondingAuthor":false,"prefix":"","firstName":"Shijie","middleName":"","lastName":"Zhang","suffix":""},{"id":551232168,"identity":"5f1fc40c-2ffe-4127-ba74-c8f6bbc83cc6","order_by":3,"name":"Yuqian Chen","email":"","orcid":"","institution":"Big Data and Artificial Intelligence Laboratory, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Yuqian","middleName":"","lastName":"Chen","suffix":""},{"id":551232171,"identity":"dfcace22-7a42-488e-9ff7-d4a648e82183","order_by":4,"name":"Ziyin Li","email":"","orcid":"","institution":"Department of Radiology, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Ziyin","middleName":"","lastName":"Li","suffix":""},{"id":551232173,"identity":"bd880601-d940-4336-a8c7-f7a45493abd2","order_by":5,"name":"Xiaodong Wang","email":"","orcid":"","institution":"Department of Radiology, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Xiaodong","middleName":"","lastName":"Wang","suffix":""},{"id":551232174,"identity":"c628a4dd-9c0f-4c14-9420-fc0b877031f8","order_by":6,"name":"Haicheng Zhang","email":"","orcid":"","institution":"Big Data and Artificial Intelligence Laboratory, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Haicheng","middleName":"","lastName":"Zhang","suffix":""},{"id":551232176,"identity":"dc76bba0-2f20-49fa-b926-302c77cbb03b","order_by":7,"name":"Qi Wang","email":"","orcid":"","institution":"Big Data and Artificial Intelligence Laboratory, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Qi","middleName":"","lastName":"Wang","suffix":""},{"id":551232177,"identity":"ecc6f01c-8983-44f2-a83c-e935a5968523","order_by":8,"name":"Tongpeng Chu","email":"","orcid":"","institution":"Big Data and Artificial Intelligence Laboratory, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Tongpeng","middleName":"","lastName":"Chu","suffix":""},{"id":551232178,"identity":"76091049-6a1e-4278-a6b1-f2fd20bdb2dd","order_by":9,"name":"Haizhu Xie","email":"","orcid":"","institution":"Department of Radiology, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Haizhu","middleName":"","lastName":"Xie","suffix":""},{"id":551232179,"identity":"caedeca9-b614-4611-af30-02829bfa6c07","order_by":10,"name":"Heng Ma","email":"","orcid":"","institution":"Department of Radiology, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Heng","middleName":"","lastName":"Ma","suffix":""},{"id":551232182,"identity":"5765b9e7-078e-4de9-bab1-0d461e684993","order_by":11,"name":"Nina Qu","email":"","orcid":"","institution":"Department of Ultrasound, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":false,"prefix":"","firstName":"Nina","middleName":"","lastName":"Qu","suffix":""},{"id":551232185,"identity":"e2de74fc-54e8-458a-9135-5271fcf07a63","order_by":12,"name":"Jing Liu","email":"","orcid":"","institution":"Department of Interventional Operating Room, Yantaishan Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Liu","suffix":""},{"id":551232186,"identity":"39f06729-9013-413b-8755-aeb49951a87a","order_by":13,"name":"Runpeng Chen","email":"","orcid":"","institution":"Binzhou Medical University","correspondingAuthor":false,"prefix":"","firstName":"Runpeng","middleName":"","lastName":"Chen","suffix":""},{"id":551232187,"identity":"87798c58-8dac-43ad-a51e-c0443119ebee","order_by":14,"name":"Jing Gao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8klEQVRIiWNgGAWjYJACgwQwxdjAwFAhISdPtBYesJYzFsaGDcRaxQO2qK0ikeEAAZXy7ocPFDzcUWtvL324TeLjPIkExgbmh49u4NFieCYtwSDxzPHEHr7ENsmZ2yTy2BnYjI1z8GlpyDEwSGw7lsDDw9hszLtNopixgYdNGq+W/jdgLfZgLX/nSCQ2HCCgRV4CbEsNYw8PY+NjxgYitBhIPAP6pe1AYs8ZxsaHPcckjA2bCfhFvj/5mOHPtjp79h72Bwd+1NTJybM3P3yM15YDDGwGDAyHkYSY8SgH29LAwPyAgaGOgLJRMApGwSgY0QAAqkRK3KwXzQ8AAAAASUVORK5CYII=","orcid":"","institution":"Department of Radiology, Yantai Yuhuangding Hospital, Qingdao University","correspondingAuthor":true,"prefix":"","firstName":"Jing","middleName":"","lastName":"Gao","suffix":""}],"badges":[],"createdAt":"2025-10-10 14:38:13","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7828102/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7828102/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":97140138,"identity":"67d65640-0945-4134-9cdf-4756a84ea217","added_by":"auto","created_at":"2025-12-01 10:03:54","extension":"png","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1451076,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/70f6069d45b6880610a1492c.png"},{"id":97139501,"identity":"0e092a6f-92a3-4156-ba43-d84153db07ea","added_by":"auto","created_at":"2025-12-01 10:00:31","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":117496,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/711f80ee821a26ac30ad28dc.docx"},{"id":97096007,"identity":"6f92a64e-197a-45d5-848f-69efdf916e5a","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":202535,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/048112bb7db6391cf0a0d62a.png"},{"id":97096015,"identity":"9518f101-a9fe-41d4-8a5a-cbd0b58233d4","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1229006,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/aaaf6abca2dc0b8b99b78b59.png"},{"id":97140638,"identity":"5e71b48a-6a77-416c-8a3c-93f1a0f3f055","added_by":"auto","created_at":"2025-12-01 10:05:26","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29952,"visible":true,"origin":"","legend":"","description":"","filename":"Table.docx","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/b5ffddf7806ac856e1212d58.docx"},{"id":97096030,"identity":"1f7ecd20-0487-4cb9-9498-d1b642af9af1","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1229006,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/6c5902c4858fe655c5fa04fd.png"},{"id":97096010,"identity":"dafb6d17-ad99-4adb-b529-039f2240ce51","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10676,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/ce4833510e4ba9a037cbfc1b.png"},{"id":97096018,"identity":"e2e6d190-f8fc-4f18-a5bb-289a288e8171","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32338,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/97ff8d5feea39eaf3b61b062.png"},{"id":97096009,"identity":"e47917f8-f4fb-44a0-8c25-b65b85f906d2","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30108,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/35615ff9983252dac31cae11.png"},{"id":97096028,"identity":"06251728-592c-4474-8fe0-998b7c5f05b5","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28742,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5C.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/00c6f16ef607aa924d17bde0.png"},{"id":97141947,"identity":"f7b34da7-6442-4ce6-b3aa-24dd835f723b","added_by":"auto","created_at":"2025-12-01 10:07:11","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10102,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/0af2043fd001bc2e01547e11.png"},{"id":97096022,"identity":"c7e469fc-88eb-4e09-9bb0-25c107f31845","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30341,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/80df09403fcb7dd63cdcd032.png"},{"id":97096025,"identity":"32928176-4186-4036-acd7-32df931f6ccd","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17701,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6C.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/3f057d32f0a171d26b5dfdd5.png"},{"id":97096041,"identity":"eb79718d-9228-4c8c-b28a-783d27cb4676","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"json","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14473,"visible":true,"origin":"","legend":"","description":"","filename":"69419d74a0c64a8faf9133fa9cc8bb3d.json","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/73bdf5f6ba91147e7842777e.json"},{"id":97140181,"identity":"0b8112d6-9b74-42fe-a9c1-8bef446d4501","added_by":"auto","created_at":"2025-12-01 10:04:04","extension":"docx","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":27079,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/344d22d9fa4dc22918c822fc.docx"},{"id":97096048,"identity":"95990ad0-14fd-4a3b-b3bf-52391c4d682c","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"xml","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":133143,"visible":true,"origin":"","legend":"","description":"","filename":"69419d74a0c64a8faf9133fa9cc8bb3d1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/95a5e8af3314cb2cfa3eff07.xml"},{"id":97141031,"identity":"4708db8e-63bd-438d-a568-aa2521b1f48f","added_by":"auto","created_at":"2025-12-01 10:06:09","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1451076,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/d4e7169353529e43dcf94e1f.png"},{"id":97096020,"identity":"1d704898-fc79-426e-aec1-a639546bc391","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":202535,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/0a516ebc2840f47e148b19f9.png"},{"id":97141001,"identity":"4861e602-6d93-4774-8db0-43bc48a568d0","added_by":"auto","created_at":"2025-12-01 10:06:06","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1229006,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/3acb2c3eafbde85bdeef9ce7.png"},{"id":97096019,"identity":"c7ee9c39-95a1-460d-ad4c-aa3a951b434e","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1229006,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/e3948545817e987228b96965.png"},{"id":97140101,"identity":"b5260061-8050-4083-98e1-9bf5a455e3da","added_by":"auto","created_at":"2025-12-01 10:03:45","extension":"png","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10676,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/fa0fa55a6cc9fed1251defb1.png"},{"id":97096023,"identity":"fd018590-e816-4f13-863e-14a2250f5095","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":21,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":32338,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/3ea526d9b9b2f7b8e91fe999.png"},{"id":97141295,"identity":"7f164e7e-a590-4321-a7b8-06bf4cc519c2","added_by":"auto","created_at":"2025-12-01 10:06:32","extension":"png","order_by":22,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30108,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/90b80b08a61afe1cbb62a706.png"},{"id":97139497,"identity":"a8517273-09c1-434c-a936-52f77e569c9e","added_by":"auto","created_at":"2025-12-01 10:00:31","extension":"png","order_by":23,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":28742,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5C.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/a02635ff0b1d15ab82d9a068.png"},{"id":97096047,"identity":"fb6cec57-275c-40fe-8306-3158de777595","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":24,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10102,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/004cb81e90ead38f9910c69e.png"},{"id":97096024,"identity":"6baeb207-8cf7-47a9-a953-1ea8e809aa67","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":25,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":30341,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/6f9e1274803596fba8169666.png"},{"id":97096035,"identity":"610ae813-cedd-46d3-9537-52cdb38bb428","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":26,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17701,"visible":true,"origin":"","legend":"","description":"","filename":"Figure6C.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/c80f3f9624515af0c50d5a37.png"},{"id":97140696,"identity":"008a89b7-48be-46a3-9ca6-c4439d6afffe","added_by":"auto","created_at":"2025-12-01 10:05:36","extension":"png","order_by":27,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":289032,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/dd636fa3dcc686d4ad8dfdf4.png"},{"id":97140342,"identity":"0888ed98-3a2e-42b6-b5de-ef16a51092ea","added_by":"auto","created_at":"2025-12-01 10:04:41","extension":"png","order_by":28,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":75687,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/9c501773ba9ed876b745eed9.png"},{"id":97096021,"identity":"237277e2-507a-4311-9289-18463afa7b58","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":29,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7612,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/32632057f6ef96d25430308f.png"},{"id":97096029,"identity":"f5edbd01-b4c6-4687-ab2e-43740af36380","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":30,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7201,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/f54877609c010b165586f103.png"},{"id":97096026,"identity":"a56ca745-acce-43a5-97bd-1a9c41817479","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":31,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9378,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/1690d821f364f74d2c7ba745.png"},{"id":97140950,"identity":"64d2cfaf-4888-46da-a811-fb3399eac82b","added_by":"auto","created_at":"2025-12-01 10:05:59","extension":"png","order_by":32,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9280,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/d462d13685ab6589bdc098e0.png"},{"id":97096027,"identity":"0d31c9e9-99a3-46f2-b09f-9d697e89d851","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":33,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8157,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/d307851aa6fb0daac6ee3643.png"},{"id":97141026,"identity":"6f954ca1-5b87-4d3f-816a-290616f81509","added_by":"auto","created_at":"2025-12-01 10:06:08","extension":"png","order_by":34,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7978,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5C.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/1d207419b9fafe645dc15b8a.png"},{"id":97096044,"identity":"cab1c473-4794-454f-9297-48516f72f0dc","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":35,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10053,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure6A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/4142e06657bbba5bce39ef86.png"},{"id":97096037,"identity":"2daa17e7-428f-4fcc-a1eb-b6179b91920b","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":36,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":19288,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure6B.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/8eb42f2c84c9d5ad73566e00.png"},{"id":97096032,"identity":"10f3bf36-228f-4784-ab47-4b4bc247aba0","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"png","order_by":37,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17143,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure6C.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/4c6307b7b6238b85d5d74570.png"},{"id":97096040,"identity":"a0ac3e35-12c4-480f-9a41-b7f30eeb9d62","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"xml","order_by":38,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":130720,"visible":true,"origin":"","legend":"","description":"","filename":"69419d74a0c64a8faf9133fa9cc8bb3d1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/740585fa164d5447aa5c5a35.xml"},{"id":97096042,"identity":"264c21c0-3ddb-4b82-81aa-1867ba6f971c","added_by":"auto","created_at":"2025-11-30 23:27:31","extension":"html","order_by":39,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":144923,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/3fad2e2b58684c13af880cb6.html"},{"id":97096001,"identity":"1467e4e5-beed-4585-b0bc-10598538b50d","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1451076,"visible":true,"origin":"","legend":"\u003cp\u003eThe sketch maps of different ROIs on RC-CC view of CEM image. (1a, 1A) ITR, intratumoral region; (1b, 1B) PTR5, 5-mm peritumoral region; (1c, 1C) PTR10, 10-mm peritumoral region; (1d, 1D) IPTR5: intratumoral region plus 5-mm peritumoral region; (1e, 1E) IPTR10, intratumoral region plus 10-mm peritumoral region.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/a9b1e9d0f0cd54ed3da8f300.png"},{"id":97096008,"identity":"2aebe48b-e477-4223-aab5-6ce58f6e466f","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":202535,"visible":true,"origin":"","legend":"\u003cp\u003eStudy flow chart.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/16cbc85a19c50b67dbe3262e.png"},{"id":97139461,"identity":"6fb42aa7-4420-4148-9f3a-21487a936a0c","added_by":"auto","created_at":"2025-12-01 10:00:26","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":36755,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves of the five radiomics signatures for the prediction of equivocal HER2 (IHC 2+) status in the training cohort (A) and internal test cohort (B).\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/35a15119ac7c176cdbc92719.png"},{"id":97096016,"identity":"eec9f465-58b0-4c38-b23d-bcb9684b4713","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":10676,"visible":true,"origin":"","legend":"\u003cp\u003eThe nomogram for the prediction of equivocal HER2 (IHC 2+) status.\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/fc13ecfcd7dfeeb6a8eb1410.png"},{"id":97096014,"identity":"68aa0de5-24b6-485a-96b9-810f4844e0d4","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":83990,"visible":true,"origin":"","legend":"\u003cp\u003eROC curves of the models for the prediction of equivocal HER2 (IHC 2+) in the training (A), internal test (B) and prospective test (C) cohorts.\u003c/p\u003e","description":"","filename":"Figure5A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/bbc225147e6c7b0eae9465b5.png"},{"id":97140979,"identity":"51b80f2c-72d3-48fd-a387-f620dc45b5ce","added_by":"auto","created_at":"2025-12-01 10:06:03","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":67333,"visible":true,"origin":"","legend":"\u003cp\u003e(A) Calibration curve of the nomogram in the three cohorts. The decision curve analysis of three models in the in the internal test (B) and prospective test (C) cohorts.\u003c/p\u003e","description":"","filename":"Figure6A.png","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/364e4f8d13256c8ce5f327b4.png"},{"id":98621817,"identity":"b01251fc-8419-4e99-8bdb-47067930d60e","added_by":"auto","created_at":"2025-12-19 16:23:23","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2686103,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/e58a5644-5871-4a5c-82ba-0c39cefd7f42.pdf"},{"id":97096006,"identity":"779ece6a-615a-4353-8dd7-c925585b703d","added_by":"auto","created_at":"2025-11-30 23:27:30","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":27079,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-7828102/v1/13603305b9a5701d2be9893d.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Intra- and Peritumoral Radiomics for Predicting Equivocal HER2 (IHC2+) Status of Breast Cancer on Contrast-Enhanced Mammography","fulltext":[{"header":"Background","content":"\u003cp\u003eHuman epidermal growth factor receptor 2 (HER2)-positive breast cancer is mainly an aggressive subtype of breast cancer that accounts for about 15\u0026ndash;20% of all breast cancers[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The HER2 gene is an independent prognostic factor for breast cancer recurrence and survival, a series of studies confirmed that the overexpression and amplification of the HER2 gene are closely associated with the effectiveness of breast cancer treatment[\u003cspan additionalcitationids=\"CR3\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Randomized clinical trials have demonstrated that adding anti-HER2-positive therapy to the treatment of HER2-positive breast cancer patients is an important factor for improving patient outcomes, which were reflected in an increase in the pathological complete response (pCR) rate, overall survival rate, and disease-free survival rate, as well as a reduction in the risk of disease recurrence or death[\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Therefore, the timely identification of HER2 status is of paramount importance in order to inform the subsequent course of treatment. Immunohistochemical (IHC) detection of protein overexpression and fluorescence in situ hybridization (FISH) analysis of HER2 gene amplification are two principal methods for determining HER2 status[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. HER2 positivity is defined either by protein overexpression as defined by IHC3\u0026thinsp;+\u0026thinsp;or equivocal protein expression (IHC2+) with evidence of HER2 gene amplification[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. IHC 2\u0026thinsp;+\u0026thinsp;was equivocal and required FISH to ascertain whether it was amplified[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In the clinic, the IHC test is relatively straightforward. However, additional analysis, such as FISH, undoubtedly results in time-consuming procedures and associated additional costs. It also requires specialized equipment and technical expertise. Consequently, a convenient and non-invasive modality to predict equivocal HER2 (IHC 2+) status is required.\u003c/p\u003e\u003cp\u003eImaging is an essential tool in medical science and is routinely used in clinical practice for tumor detection and treatment guidance[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Contrast-enhanced mammography (CEM) represents a cutting-edge technique that utilizes dual energy exposure, which has the advantage of low cost, time-saving, and acceptable tolerance in breast cancer patients. It could not only display the morphological characteristics, but also provide the blood supply information of the lesions. Moreover, CEM has higher sensitivity and specificity in the diagnosis of breast cancer than mammography and comparable to dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI)[\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. However, it is difficult to assess equivocal HER2 (IHC 2+) status using only the image features visible to the naked eye. Radiomics is a promising tool that can convert medical images into mineable data by extracting high-throughput quantitative features[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. As a non-invasive and effective method, radiomics has been widely used to assess whole tumor heterogeneity[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], and some researches have attempted to use radiomics to predict HER2 status in breast cancer[\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. However, these studies were conducted based on MRI or mammography, not CEM. Besides, the tumor microenvironment plays a role in the development and progression of breast cancer. Some literatures have explored the value of peritumoral radiomics in breast cancer diagnosis and prognosis[\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Empirical evidence implicates that the microenvironment may contain information that is relevant to the treatment of HER2-positive breast cancer[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. However, the value of peritumour radiomics based on CEM in predicting equivocal HER2 (IHC2+) status is unclear.\u003c/p\u003e\u003cp\u003eTherefore, the study aims to explore the value of intratumoral and peritumoral radiomics based on CEM for predicting equivocal HER2 (IHC2+) status in breast cancer patients.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eStudy population\u003c/h2\u003e\u003cp\u003eEthical approval\u003c/strong\u003e was obtained for retrospective study, and the requirement for patient informed consent requirement was waived. A prospective study was approved by the institutional ethics committee of the hospital. Written informed consents were obtained from patients whose CEM images were prospectively collected. (Clinical trial number: not applicable)\u003c/p\u003e\u003c/p\u003e\u003cp\u003eWe retrospectively screened the CEM images between January 2021 and January 2022 in the institution from the picture archiving and communication system. The inclusion criteria for the retrospective study were as follows: (1) breast cancer patients pathologically confirmed by biopsy or surgical specimens; (2) underwent CEM less than 2 weeks before surgery or other treatment options; (3) patients with a single lesion; and (4) HER2 score of 2\u0026thinsp;+\u0026thinsp;verified by IHC and have the results of FISH. The exclusion criteria were as follows: (1) patients who underwent breast radiotherapy, chemotherapy, or hormone treatment before CEM examination; (2) incomplete clinical, pathological, IHC or FISH information of patients; and (3) poor image quality. A total of 106 female patients were ultimately included in this study and divided into training cohort and internal test cohort at a ratio of 8:2. In addition, 25 patients were prospectively evaluated at our hospital from March 2022 to June 2022. For the prospective study, the inclusion criteria were as follows: (1) patients were confirmed by biopsy as breast cancer; (2) enrolled patients were planned to receive standard treatment or surgery in our hospital; (3) CEM examinations were available and performed within two weeks before surgery or other therapeutic regimes; and (4) HER2 score of 2\u0026thinsp;+\u0026thinsp;verified by IHC and have the results of FISH. The exclusion criteria were as follows: (1) patients who had any previous history of cancer and treatment; (2) patients with incomplete clinical, pathological, IHC or FISH information; and (3) inadequate image quality or non-mass lesions which affected the observation and delineation. Information regarding the clinical characteristics of the patients, including age and other parameters, were obtained from the electronic medical record.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eCEM examination\u003c/h3\u003e\n\u003cp\u003eAll patients underwent CEM examination using a GE Senographe Essential mammography unit (GE Healthcare, Milwaukee, WI, USA). The contrast agent Omnipaque 300 (GE Healthcare, Inc., Princeton, NJ) was injected into the upper arm vein with the dose of 1.5 ml/kg and the injection flow rate of 3.0 ml/s. Two minutes after the contrast injection, images were obtained in the following order: craniocaudal (CC) and mediolateral oblique (MLO) views of the suspicious breast, and then CC and MLO views of the less suspicious breast. For each mammographic projection, a pair of high-energy (HE) and low-energy (LE) exposures were consecutively performed to obtain HE and LE images, and a recombined (RC) image was generated automatically from LE and HE images using a dual-energy weighted logarithmic subtraction technique[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e].\u003c/p\u003e\n\u003ch3\u003ePathological Assessment\u003c/h3\u003e\n\u003cp\u003e According to the recommendations by the American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) guidelines, IHC analyses were performed to determine the expression levels of HER2 in each breast cancer patient. A HER2 staining intensity score of 3\u0026thinsp;+\u0026thinsp;was considered positive, while a score of 0 or 1\u0026thinsp;+\u0026thinsp;was considered negative. A HER2 staining intensity score of 2+, with confirmation of gene amplification by FISH, was also deemed positive HER2 [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e\n\u003ch3\u003eImage segmentation\u003c/h3\u003e\n\u003cp\u003eAll radiologists who participated in the image segmentation were blinded to the clinical and histopathological information. A dedicated radiologist (Segmenter 1) with seven years\u0026rsquo; experience of breast imaging annotated tumor boundaries as the intratumoral region (ITR) in the LE and RC images with standard CC view via the ITK-SNAP (version 3.8.0; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003ca href=\"http://www.itksnap.org\" target=\"_blank\"\u003ewww.itksnap.org\u003c/a\u003e\u003c/span\u003e\u003cspan address=\"http://www.itksnap.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) software. All the contours were reviewed by a radiologist with 15 years of experience. Meanwhile, 40 patients randomly selected from the training cohort were used to assess the consistency of the inter- and intra-observer segmentation. Another 2 two radiologists (Segmenter 2 and 3) with 8 and 10 years of experience, respectively, performed the segmentation work utilizing the same methods. Additionally, the radiologist (Segmenter 1) repeated the segmentation with an interval of two weeks. The Dice similarity coefficient (DSC) was used to evaluate the agreement both inter- and intra-observer segmentation. Average DSCs of 0.873 and 0.925 were achieved for inter- and intra-observer segmentation performances, respectively. Meanwhile, the largest diameters of the lesions were measured independently by the two radiologists in CC view images. The mean values were calculated as the final tumor diameter.\u003c/p\u003e\u003cp\u003eAfter the ITR mask was annotated, a morphologic operation of dilation was performed to capture the peritumoral regions (PTR) outside the tumor of 5 mm and 10 mm using Python (version 3.6.6). If the contours of peritumoral regions exceeded the breast parenchyma after extension, the beyond portion was manually removed[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Additionally, we define the intra- and peritumoral region in the same ROI named as IPTR.\u003c/p\u003e\u003cp\u003eFinally, for each lesion in each image, five ROIs, namely ITR, PTR5 (5-mm peritumoral region), PTR10 (10-mm peritumoral region), IPTR5 (intratumoral region\u0026thinsp;+\u0026thinsp;5-mm peritumoral region) and IPTR10 (intratumoral region\u0026thinsp;+\u0026thinsp;10-mm peritumoral region), were generated to extract radiomics features. Example segmentations are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eRadiomics feature extraction\u003c/h3\u003e\n\u003cp\u003eImage preprocessing was conducted by Python (version 3.6.6) before radiomics feature extraction, including image resampling and normalization, and gray-level discretization. A total of 1316 quantitative radiomics features were extracted from each ROI in each image using the radiomics function RadiomicsFeatureExtractor toolkit provided by Pyradiomics[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], including 252 first-order statistics, 14 shape features, 336 gray-level co-occurrence matrix (GLCM), 224 gray-level run length matrix (GLRLM), 224 gray-level size zone matrix (GLSZM), 196 gray-level dependence matrix (GLDM), 70 neighboring gray-tone difference matrix (NGTDM). Since LE-CC and RC-CC images were analyzed, a total of 2632 radiomics features were extracted from each ROI (ITR, PTR5, PTR10, IPTR5, IPTR10).\u003c/p\u003e\u003cp\u003eThe intraclass correlation coefficient (ICC) was used to evaluate the reproducibility of manual radiomics feature extraction. To calculate the intra- and inter-observer agreement of feature extraction, two radiologists (Radiologist 1 and Radiologist 2) firstly extracted the radiomics features using 40 randomly chosen patients to calculate the ICCs, respectively. Two weeks later, the procedure was repeated by Radiologist 1 and the remaining images were also analyzed by Radiologist 1. In this study, the radiomics features with ICCs of 0.80 or greater were selected and considered a mark of satisfactory intra- and inter-observer agreement.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eFeature selection and radiomics signature building\u003c/h2\u003e\u003cp\u003eAll feature selection process was performed in the training cohort. Low variance was first used to filter features, the variance of each feature is calculated, and if it is below a threshold (0.80) we filter it out. Then, the least absolute shrinkage and selection operator (LASSO) logistic regression method was applied to select and identify the most stable and predictive features and to construct the radiomics signatures. To avoid over-fitting, the best parameter of the LASSO regularization parameter (α) was determined via 10-fold cross-validation. The radiomics signature score reflecting the respective equivocal HER2 (IHC2+) status for each patient was calculated through a linear combination of selected features weighted by their respective coefficients. On the basis of this procedure, five signatures were constructed from the five ROIs. The sensitivity and specificity values for evaluating the performance of the radiomics signatures in all cohorts were plotted to generate a receiver operator characteristic (ROC) curve, and the area under the curve (AUC) was calculated.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eRadiomics nomogram construction and comparison\u003c/h3\u003e\n\u003cp\u003eFirstly, the potential predictors among clinical variables and five radiomics signatures were identified using univariate logistic regression. Then these predictors were integrated into the multivariate logistic regression analysis with backward-stepwise selection based on minimal Akaike information criteria, which was used to select independent predictors of positive equivocal HER2 (IHC2+) status. A radiomics nomogram was established based on the independent predictors. For comparison, a clinical model was also built using only the selected clinical variables, and a radiomics model was built using only the selected radiomics signatures. To prove the generalization of the nomogram, a prospective cohort was used to test the models. The AUC, sensitivity, specificity, and accuracy were used to evaluate the predictive performance of the models. The study flow chart is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eAll statistical analyses were conducted with Python (version 3.6.6) and R software (version 4.0.3). Continuous variables were compared by two-sample t-test, while qualitative variables were analyzed by Chi-square test. LASSO regression and ROC curve analysis were conducted using the \u0026ldquo;LassoCV\u0026rdquo; function and \u0026ldquo;roc_curve\u0026rdquo; packages. The ROC curves were plotted to evaluate the performance of the models and the AUCs were calculated. The Youden index was chosen as the optimal cut-off value [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Decision curve analysis (DCA) was employed to assess the net benefit of the models in a clinical context. The Hosmer-Lemeshow goodness of fit test was used to evaluate the calibration of the models and the calibration curve was plotted. DCA and calibration curve were performed using the \u0026ldquo;rmda\u0026rdquo; and \u0026ldquo;rms\u0026rdquo; packages. The DeLong test [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e] was applied to compare the AUCs of different models. \u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered a statistically significant difference.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eClinicopathological characteristics\u003c/h2\u003e\u003cp\u003eA total of 131 patients were enrolled in this study. The training cohort included 84 patients (28 patients with positive IHC 2\u0026thinsp;+\u0026thinsp;status, 56 patients with negative IHC 2\u0026thinsp;+\u0026thinsp;status), the average age of patients was 54.2\u0026thinsp;\u0026plusmn;\u0026thinsp;8.2 years old (range, 33\u0026ndash;74 years old). The internal test cohort included 22 patients (8 patients with positive IHC 2\u0026thinsp;+\u0026thinsp;status, 14 patients with negative IHC 2\u0026thinsp;+\u0026thinsp;status), the average age of patients was 55.1\u0026thinsp;\u0026plusmn;\u0026thinsp;8.9 years old (range, 38\u0026ndash;75 years old). The prospective test cohort included 25 patients (10 patients with positive IHC 2\u0026thinsp;+\u0026thinsp;status, 15 patients with negative IHC 2\u0026thinsp;+\u0026thinsp;status), the average age of patients was 57.2\u0026thinsp;\u0026plusmn;\u0026thinsp;11.1 years old (range, 32\u0026ndash;80 years old). The results of the clinicopathological features between the patients with negative and positive equivocal HER2 (IHC 2+) status are listed in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCharacteristics in the training, internal test, and prospective test cohorts\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"21\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c12\" colnum=\"12\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c13\" colnum=\"13\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c14\" colnum=\"14\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c15\" colnum=\"15\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c16\" colnum=\"16\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c17\" colnum=\"17\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c18\" colnum=\"18\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c19\" colnum=\"19\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c20\" colnum=\"20\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c21\" colnum=\"21\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"5\" nameend=\"c6\" namest=\"c2\"\u003e\u003cp\u003eTraining cohort (n\u0026thinsp;=\u0026thinsp;84)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"5\" nameend=\"c13\" namest=\"c9\"\u003e\u003cp\u003eInternal test cohort (n\u0026thinsp;=\u0026thinsp;22)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"6\" nameend=\"c20\" namest=\"c15\"\u003e\u003cp\u003eProspective test cohort (n\u0026thinsp;=\u0026thinsp;25)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"1\" nameend=\"c21\" namest=\"c21\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHER2 2+\u003c/p\u003e\u003cp\u003ePositive (n\u0026thinsp;=\u0026thinsp;28)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003eHER2 2+\u003c/p\u003e\u003cp\u003eNegative (n\u0026thinsp;=\u0026thinsp;56)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c9\"\u003e\u003cp\u003eHER2 2+\u003c/p\u003e\u003cp\u003ePositive (n\u0026thinsp;=\u0026thinsp;8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c11\" namest=\"c10\"\u003e\u003cp\u003eHER2 2+\u003c/p\u003e\u003cp\u003eNegative (n\u0026thinsp;=\u0026thinsp;14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c13\" namest=\"c12\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c14\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c16\" namest=\"c15\"\u003e\u003cp\u003eHER2 2+\u003c/p\u003e\u003cp\u003ePositive (n\u0026thinsp;=\u0026thinsp;10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c18\" namest=\"c17\"\u003e\u003cp\u003eHER2 2+\u003c/p\u003e\u003cp\u003eNegative (n\u0026thinsp;=\u0026thinsp;15)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c20\" namest=\"c19\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c21\" namest=\"c21\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge, years\u003c/p\u003e\u003cp\u003e(mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003e52.3\u0026thinsp;\u0026plusmn;\u0026thinsp;8.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e54.9\u0026thinsp;\u0026plusmn;\u0026thinsp;7.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u003cp\u003e0.267\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e53.8\u0026thinsp;\u0026plusmn;\u0026thinsp;4.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u003cp\u003e55.8\u0026thinsp;\u0026plusmn;\u0026thinsp;10.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u003cp\u003e0.060\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003e56.2\u0026thinsp;\u0026plusmn;\u0026thinsp;10.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u003cp\u003e57.8\u0026thinsp;\u0026plusmn;\u0026thinsp;11.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u003cp\u003e0.799\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDiameter, cm (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003e2.6\u0026thinsp;\u0026plusmn;\u0026thinsp;1.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e3.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u003cp\u003e0.090\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e2.2\u0026thinsp;\u0026plusmn;\u0026thinsp;0.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u003cp\u003e3.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.010\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003e3.7\u0026thinsp;\u0026plusmn;\u0026thinsp;1.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u003cp\u003e2.3\u0026thinsp;\u0026plusmn;\u0026thinsp;0.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u003cp\u003e0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eER\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u003cp\u003e0.341\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u003cp\u003e0.439\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u003cp\u003e0.727\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNegative\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePositive\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003e26\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u003cp\u003e14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u003cp\u003e1.000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u003cp\u003e0.674\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u003cp\u003e0.105\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNegative\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u003cp\u003e2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePositive\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003e22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eKi-67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u003cp\u003e0.701\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u003cp\u003e0.339\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.01\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNegative\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePositive\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003e23\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003e44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c10\" namest=\"c8\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c12\" namest=\"c11\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c15\" namest=\"c13\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c17\" namest=\"c16\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c19\" namest=\"c18\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c21\" namest=\"c20\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"21\"\u003eNote: HER2, human epidermal growth factor receptor 2; ER, estrogen receptor; PR, progesterone receptor; SD: standard deviation.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003ePerformance of the radiomics signatures\u003c/h2\u003e\u003cp\u003eThe extracted features were identified as being highly reproducible based on the intra- and inter-observer ICCs ranging from 0.915 to 0.945 and 0.901 to 0.935, respectively. After selecting radiomics features, a total of 10, 8, 9, 7, and 6 features were selected as the most valuable features from ITR, PTR5, PTR10, IPTR5, and IPTR10, respectively, and then the five radiomics signatures were constructed. The detailed features and their respective coefficients are shown in Supplementary Materials (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). All five radiomics signatures showed encouraging results, with AUC values varying from 0.643 (95% confidence interval [CI], 0.364\u0026ndash;0.921) to 0.866 (95% CI, 0.714-1.000) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Among them, the radiomics signature based on ITR yielded the highest AUC value of 0.866 (95% CI, 0.714-1.000) in the internal test cohort. The detailed results are presented in Supplementary Materials (Table S2).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003ePerformance of the nomogram, radiomics model, and clinical model\u003c/h2\u003e\u003cp\u003eTo ascertain the value of radiomics signatures and clinical variables, univariate and multivariate logistic regression analyses were conducted. Two radiomics signatures (constructed from ITR and PTR5) and the diameter were identified as significant predictors for identifying equivocal HER2 (IHC 2+) status. The detailed results are represented in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Then, the diameter, signature ITR and signature PTR5 were used to constructed the nomogram (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). In addition, the clinical model based on diameter and the radiomics model based on signature ITR and signature PTR5 were also constructed, respectively.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eRisk factors for predicting equivocal HER2 (IHC2+) status\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e\u003cp\u003eUnivariate logistic regression\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e\u003cp\u003eMultivariate logistic regression\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eOR (95% CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eOR (95% CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSignature ITR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8.240(4.397\u0026ndash;15.440)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8.600(3.868\u0026ndash;19.123)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSignature PTR5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e31.629(12.919\u0026ndash;77.434)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.490(1.062\u0026ndash;11.467)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.042*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSignature PTR10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3.259(2.104\u0026ndash;5.046)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.878(0.495\u0026ndash;1.558)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.658\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSignature IPTR5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e37.303(11.509-120.909)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.857(0.204\u0026ndash;3.599)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.833\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSignature IPTR10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e29.807(7.724-115.025)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.140(0.307\u0026ndash;4.233)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.845\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.987(0.976\u0026ndash;0.999)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.044*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.993(0.983\u0026ndash;1.003)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.630\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDiameter\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.891(0.807\u0026ndash;0.983)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.024*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.125(1.008\u0026ndash;1.256)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.039*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003eNote: OR, odds ratio; CI, confidence interval; ITR, intratumoral region; PTR5, 5-mm peritumoral region; PTR10, 10-mm peritumoral region; IPTR5, intratumoral region\u0026thinsp;+\u0026thinsp;5-mm peritumoral region); IPTR10, intratumoral region\u0026thinsp;+\u0026thinsp;10-mm peritumoral region\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003e* \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAs shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA-C, the nomogram achieved relatively good discrimination. In the internal test cohort, the AUC of the nomogram reached 0.893 (95CI%: 0.756-1.000), higher than the radiomics model with 0.821 (95CI%: 0.641-1.000) (Delong test, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.292) and the clinical model with 0.866 (95CI%: 0.681-1.000) (Delong test, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.314). In the prospective test cohort, the AUC of the nomogram was 0.840 (95CI%: 0.652-1.000), higher than the radiomics model with 0.819(95CI%: 0.649\u0026ndash;0.999) (Delong test, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.676) and the clinical model with 0.774 (95CI%: 0.547-1.000) (Delong test, \u003cem\u003eP\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.878). In the prospective test cohort, the nomogram achieved the highest sensitivity of 0.923. Detailed results are presented in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. The calibration curves indicated that the radiomics nomogram had good calibration in the three cohorts (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Figure\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC showed the nomogram achieved a higher net benefit than the radiomics model and the clinical model in the internal test cohort.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003ePerformances of Radiomics model, Clinical model, and Nomogram\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAUC(95%CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSEN(95%CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSPE(95%CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eACC(95%CI)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\u003cp\u003eTraining cohort\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRadiomics model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.876(0.802\u0026ndash;0.950)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.741(0.601\u0026ndash;0.846)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.862(0.674\u0026ndash;0.955)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.783(0.679\u0026ndash;0.866)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClinical model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.675(0.534\u0026ndash;0.816)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.889(0.767\u0026ndash;0.954)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.586(0.391\u0026ndash;0.759)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.781(0.679\u0026ndash;0.866)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNomogram\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.891(0.823\u0026ndash;0.959)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.870(0.745\u0026ndash;0.942)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.759(0.561\u0026ndash;0.890)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.831(0.733\u0026ndash;0.905)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eInternal test cohort\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRadiomics model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.821(0.641-1.000)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.785(0.488\u0026ndash;0.942)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.750(0.355\u0026ndash;0.955)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.772(0.546\u0026ndash;0.921)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClinical model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.866(0.681-1.000)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.000(0.732-1.000)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.750(0.355\u0026ndash;0.955)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.909(0.708\u0026ndash;0.988)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNomogram\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.893(0.756-1.000)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.857(0.561\u0026ndash;0.974)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.875(0.466\u0026ndash;0.993)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.864(0.651\u0026ndash;0.971)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e\u003cp\u003e\u003cb\u003eProspective test cohort\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eRadiomics model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.819(0.649\u0026ndash;0.99)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.866(0.584\u0026ndash;0.976)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.700(0.353\u0026ndash;0.919)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.800(0.593\u0026ndash;0.931)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eClinical model\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.774(0.547-1.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.846(0.536\u0026ndash;0.973)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.583(0.285\u0026ndash;0.835)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.720(0.506\u0026ndash;0.879)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNomogram\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.840(0.652-1.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.923(0.620\u0026ndash;0.996)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.667(0.354\u0026ndash;0.887)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e0.800(0.593\u0026ndash;0.931)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"5\"\u003eNote: AUC, area under curve; SEN, sensitivity; SPE, specificity; ACC, accuracy; CI, confidence interval\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eTo achieve effective therapeutic efficacy for HER2-targeted treatment, precise equivocal HER2 (IHC 2+) status identification prior to treatment is indispensable. Although both IHC and FISH are commonly used approaches to determine equivocal HER2 (IHC 2+) status, their clinical applications are limited mainly by biopsy availability, procedure complexity and low reproducibility. In this study, we developed a nomogram using the selected radiomics signatures based on CEM and clinical risk factors to predict equivocal HER2 (IHC 2+) status, which achieved the highest AUC compared to the radiomic model and the clinical model. The results indicated that the nomogram yielded good discrimination and calibration.\u003c/p\u003e\u003cp\u003eSeveral studies have investigated that some abnormal imaging features, such as microcalcifications, breast density, or a spiculated mass on mammography are significantly associated with the HER2 status[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. However, the performance of these features in predicting the HER2 status is limited and may be influenced by the subjective judgment of radiologists. Radiomics is an emerging field that can translate medical images into quantitative data for target task prediction. Multiple imaging methodologies, such as mammography and MRI, were revealed to be able to predict HER2 status for breast cancer based on radiomic features[\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. However, they did not focus on equivocal HER2 (IHC 2+) status and CEM. At present, there is no study on the prediction of equivocal HER2 (IHC 2+) status based on CEM. To obtain more valuable radiomics features, this study extracted the radiomics features of LE-CC and RC-CC images to establish radiomics signatures, and integrated clinical risk factors to construct a nomogram, which achieved promising predictive performance. This demonstrates the value of CEM in predicting equivocal HER2 (IHC 2+) status of breast cancer.\u003c/p\u003e\u003cp\u003ePrevious findings have shown that the peritumoral area can provide useful information to assist in the diagnosis of breast cancer, which is associated with lymphatic invasion and vascular infiltration[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Some researches demonstrated that radiomics features extracted from the peritumoral region could provide information different from those in the intratumoral region[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. In this study, five radiomics signatures were established based on different intratumoral and peritumoral regions. The signature ITR achieved the best predictive performance, followed by the PTR5 signature in the internal test cohort, indicating that peritumoral radiomics based on CEM could provide valuable information for predicting equivocal HER2 (IHC 2+) status. The results showed that PTR10 had a relatively lower predictive performance than PTR5, which may be related to the fact that a larger peritumoral area contains more fat or mammary glands and less tumor information.\u003c/p\u003e\u003cp\u003eLi et al[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e] explored radiomics features of intratumoral and peritumoral regions on breast DCE-MRI to predict HER2 2\u0026thinsp;+\u0026thinsp;status, and achieved a promising performance in the validation cohort (AUC\u0026thinsp;=\u0026thinsp;0.840). However, it did not consider the role of clinical risk factors in the prediction task. In this study, the nomogram was developed based on radiomics features of intratumoral and peritumoral of CEM and clinical risk factor. Its predictive performance is superior to the single radiomics model or clinical model, indicating that both radiomics and clinical features have important value in predicting equivocal HER2 (IHC2+) status. Several recent studies have reported similar results, demonstrating the value of the nomogram established using radiomics signatures and clinical factors to evaluate pathological outcomes[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e].In addition, considering the clinical applicability of the nomogram, an independent prospective test cohort was enrolled to prove the generalization ability of the nomogram. Good predictive power was achieved, demonstrating that the nomogram has promising prospects for clinical application. This work represents a preliminary success in the pre-treatment prediction of equivocal HER2 (IHC2+) status using intra- and peritumoral radiomics, which could potentially assist in guiding personalized treatment.\u003c/p\u003e\u003cp\u003eHowever, our study had several limitations. First, this study is a single-center study with relatively inadequate sample size, although it used the internal and prospective test cohorts to test the nomogram and achieved promising results. A larger sample size from different centers is needed to further prove the generalization of the nomogram. Second, a manual segmentation method was employed in this study. Although favorable intra- and interobserver ICCs were obtained and an automated method was used for segmentation of peritumoral regions, which may have higher stability and be less time-consuming. Third, predictive models based on radiomics features were developed only from images of the CC view. It is valuable to explore whether the extracting features in different views impact the performance of the prediction models, such as MLO views and the combination of CC and MLO views.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eIn conclusion, the nomogram combined with CEM-based intratumoral and peritumoral radiomics signatures and clinical variables could predict equivocal HER2(IHC2+) status noninvasively and conveniently. This may effectively guide the personalized treatment of patients with breast cancer in clinical practice.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCEM\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eContrast-enhanced mammography\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eHER2\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eHuman epidermal growth factor receptor\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIHC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eImmunohistochemical\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eFISH\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eFluorescence in situ hybridization\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eDSC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eDice similarity coefficient\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eICC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eintraclass correlation coefficient\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eCraniocaudal\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eMLO\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eMediolateral oblique\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eLASSO\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eLeast absolute shrinkage and selection operator\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eROC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eReceiver operator characteristic\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eAUC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eArea under the curve\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eDCA\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eDecision curve analysis\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics approval and consent to participate\u003c/h2\u003e\u003cp\u003e The study was conducted in accordance with the ethical standards of the Declaration of Helsinki. The study was approved by the ethics committee of Yantai Yuhuangding Hospital, the retrospective study waived the informed consent of patients, and the prospective study obtained the informed consent of patients.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003cp\u003eNot applicable.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eCompeting interest\u003c/h2\u003e\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eNo funding.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eC.X., J.Q. and S.Z. contributed to the design of the concept and write the original draft. J.G., R.C. and J.L. reviewed and edited the draft. Y.C., H.Z., Q.W. and T.C. conducted data analysis. Z.L., X.W. and N.Q. collected the data information. H.X. and H.M. conducted data assessment. All authors read and approved the final manuscript. C.X., J.Q. and S.Z. contributed equally to this work. J.G., R.C. and J.L. contributed equally to this work.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe thank all participants for their hard work in this study.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe data presented in this study are available from the corresponding author by request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eLitton JK, Burstein HJ, Turner NC. Molecular Testing in Breast Cancer. American Society of Clinical Oncology educational book American Society of Clinical Oncology Annual Meeting. 2019, 39:e1-e7.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePiccart-Gebhart MJ, Procter M, Leyland-Jones B, Goldhirsch A, Untch M, Smith I, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med. 2005;353(16):1659\u0026ndash;72.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSlamon D, Eiermann W, Robert N, Pienkowski T, Martin M, Press M, et al. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365(14):1273\u0026ndash;83.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSLAMON DJ, LEYLAND-JONES B, SHAK S, FUCHS H, PATON V PHARMD, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med. 2001;344:783\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWaks AG, Winer EP. Breast Cancer Treatment: A Review. JAMA. 2019;321(3):288\u0026ndash;300.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRugo HS, Chien AJ. HER2-positive breast cancer: is more treatment better? Lancet Oncol. 2016;17(3):268\u0026ndash;70.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCameron D, Piccart-Gebhart MJ, Gelber RD, Procter M, Goldhirsch A, de Azambuja E, et al. 11 years' follow-up of trastuzumab after adjuvant chemotherapy in HER2-positive early breast cancer: final analysis of the HERceptin Adjuvant (HERA) trial. Lancet. 2017;389(10075):1195\u0026ndash;205.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMoelans CB, de Weger RA, Van der Wall E, van Diest PJ. Current technologies for HER2 testing in breast cancer. Crit Rev Oncol/Hematol. 2011;80(3):380\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRakha EA, Pinder SE, Bartlett JM, Ibrahim M, Starczynski J, Carder PJ, et al. Updated UK Recommendations for HER2 assessment in breast cancer. J Clin Pathol. 2015;68(2):93\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWolff AC, Hammond MEH, Allison KH, Harvey BE, Mangu PB, Bartlett JMS, et al. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update. J Clin Oncol. 2018;36(20):2105\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLawson MB, Herschorn SD, Sprague BL, Buist DSM, Lee SJ, Newell MS, et al. Imaging Surveillance Options for Individuals With a Personal History of Breast Cancer: AJR Expert Panel Narrative Review. AJR Am J Roentgenol. 2022;219(6):854\u0026ndash;68.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGhaderi KF, Phillips J, Perry H, Lotfi P, Mehta TS. Contrast-enhanced Mammography: Current Applications and Future Directions. Radiographics. 2019;39(7):1907\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLee-Felker SA, Tekchandani L, Thomas M, Gupta E, Andrews-Tang D, Roth A, et al. Newly Diagnosed Breast Cancer: Comparison of Contrast-enhanced Spectral Mammography and Breast MR Imaging in the Evaluation of Extent of Disease. Radiology. 2017;285(2):389\u0026ndash;400.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSumkin JH, Berg WA, Carter GJ, Bandos AI, Chough DM, Ganott MA, et al. Diagnostic Performance of MRI, Molecular Breast Imaging, and Contrast-enhanced Mammography in Women with Newly Diagnosed Breast Cancer. Radiology. 2019;293(3):531\u0026ndash;40.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P et al. Radiomics: extracting more information from medical images using advanced feature analysis. European journal of cancer (Oxford, England: 1990). 2012, 48(4):441\u0026ndash;446.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCao B, Mi K, Dai W, Liu T, Xie T, Li Q et al. Prognostic and incremental value of computed tomography-based radiomics from tumor and nodal regions in esophageal squamous cell carcinoma. Chinese journal of cancer research\u0026thinsp;=\u0026thinsp;Chung-kuo yen cheng yen chiu. 2022, 34(2):71\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou J, Tan H, Li W, Liu Z, Wu Y, Bai Y, et al. Radiomics Signatures Based on Multiparametric MRI for the Preoperative Prediction of the HER2 Status of Patients with Breast Cancer. Acad Radiol. 2021;28(10):1352\u0026ndash;60.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBitencourt AGV, Gibbs P, Rossi Saccarelli C, Daimiel I, Lo Gullo R, Fox MJ, et al. MRI-based machine learning radiomics can predict HER2 expression level and pathologic response after neoadjuvant therapy in HER2 overexpressing breast cancer. EBioMedicine. 2020;61:103042.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou J, Tan H, Bai Y, Li J, Lu Q, Chen R, et al. Evaluating the HER-2 status of breast cancer using mammography radiomics features. Eur J Radiol. 2019;121:108718.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNiu S, Jiang W, Zhao N, Jiang T, Dong Y, Luo Y, et al. Intra- and peritumoral radiomics on assessment of breast cancer molecular subtypes based on mammography and MRI. J Cancer Res Clin Oncol. 2022;148(1):97\u0026ndash;106.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMao N, Shi Y, Lian C, Wang Z, Zhang K, Xie H, et al. Intratumoral and peritumoral radiomics for preoperative prediction of neoadjuvant chemotherapy effect in breast cancer based on contrast-enhanced spectral mammography. Eur Radiol. 2022;32(5):3207\u0026ndash;19.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang S, Sun Y, Li R, Mao N, Li Q, Jiang T, et al. Diagnostic performance of perilesional radiomics analysis of contrast-enhanced mammography for the differentiation of benign and malignant breast lesions. Eur Radiol. 2022;32(1):639\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSalgado R, Denkert C, Campbell C, Savas P, Nuciforo P, Aura C, et al. Tumor-Infiltrating Lymphocytes and Associations With Pathological Complete Response and Event-Free Survival in HER2-Positive Early-Stage Breast Cancer Treated With Lapatinib and Trastuzumab: A Secondary Analysis of the NeoALTTO Trial. JAMA Oncol. 2015;1(4):448\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBraman N, Prasanna P, Whitney J, Singh S, Beig N, Etesami M, et al. Association of Peritumoral Radiomics With Tumor Biology and Pathologic Response to Preoperative Targeted Therapy for HER2 (ERBB2)-Positive Breast Cancer. JAMA Netw Open. 2019;2(4):e192561.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBhimani C, Matta D, Roth RG, Liao L, Tinney E, Brill K, et al. Contrast-enhanced Spectral Mammography: Technique, Indications, and Clinical Applications. Acad Radiol. 2017;24(1):84\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAllison KH, Hammond MEH, Dowsett M, McKernin SE, Carey LA, Fitzgibbons PL, et al. Estrogen and Progesterone Receptor Testing in Breast Cancer: ASCO/CAP Guideline Update. J Clin Oncol. 2020;38(12):1346\u0026ndash;66.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNielsen TO, Leung SCY, Rimm DL, Dodson A, Acs B, Badve S, et al. Assessment of Ki67 in Breast Cancer: Updated Recommendations From the International Ki67 in Breast Cancer Working Group. J Natl Cancer Inst. 2021;113(7):808\u0026ndash;19.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational Radiomics System to Decode the Radiographic Phenotype. Cancer Res. 2017;77(21):e104\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMarcus D, Ruopp NJP, Brian W, Whitcomb, Enrique F. Schisterman. Youden index and optimal cut-point estimated from observations affected by a lower limit of detection. Biom J. 2008;50(50):419\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eR DE, M DD, L C-PD. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(3):837\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi E, Guida JL, Tian Y, Sung H, Koka H, Li M, et al. Associations between mammographic density and tumor characteristics in Chinese women with breast cancer. Breast Cancer Res Treat. 2019;177(2):527\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShin HJ, Kim HH, Huh MO, Kim MJ, Yi A, Kim H, et al. Correlation between mammographic and sonographic findings and prognostic factors in patients with node-negative invasive breast cancer. Br J Radiol. 2011;84(997):19\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchoppmann SF, Bayer G, Aumayr K, Taucher S, Geleff S, Rudas M, et al. Prognostic value of lymphangiogenesis and lymphovascular invasion in invasive breast cancer. Ann Surg. 2004;240(2):306\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEjlertsen B, Jensen MB, Rank F, Rasmussen BB, Christiansen P, Kroman N, et al. Population-based study of peritumoral lymphovascular invasion and outcome among patients with operable breast cancer. J Natl Cancer Inst. 2009;101(10):729\u0026ndash;35.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou J, Zhang Y, Chang KT, Lee KE, Wang O, Li J, et al. Diagnosis of Benign and Malignant Breast Lesions on DCE-MRI by Using Radiomics and Deep Learning With Consideration of Peritumor Tissue. J Magn Reson Imaging. 2020;51(3):798\u0026ndash;809.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu J, Li B, Sun X, Cao G, Rubin DL, Napel S, et al. Heterogeneous Enhancement Patterns of Tumor-adjacent Parenchyma at MR Imaging Are Associated with Dysregulated Signaling Pathways and Poor Survival in Breast Cancer. Radiology. 2017;285(2):401\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi C, Song L, Yin J. Intratumoral and Peritumoral Radiomics Based on Functional Parametric Maps from Breast DCE-MRI for Prediction of HER-2 and Ki-67 Status. J Magn Reson Imaging. 2021;54(3):703\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYu Y, Tan Y, Xie C, Hu Q, Ouyang J, Chen Y, et al. Development and Validation of a Preoperative Magnetic Resonance Imaging Radiomics-Based Signature to Predict Axillary Lymph Node Metastasis and Disease-Free Survival in Patients With Early-Stage Breast Cancer. JAMA Netw Open. 2020;3(12):e2028086.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTang TY, Li X, Zhang Q, Guo CX, Zhang XZ, Lao MY, et al. Development of a Novel Multiparametric MRI Radiomic Nomogram for Preoperative Evaluation of Early Recurrence in Resectable Pancreatic Cancer. J Magn Reson Imaging. 2020;52(1):231\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"breast cancer, equivocal HER2 status, radiomics, intratumoral, peritumoral","lastPublishedDoi":"10.21203/rs.3.rs-7828102/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7828102/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eIdentification of Human epidermal growth factor receptor 2 (HER2) status is significant for the treatment and prognosis of breast cancer patients. The study aimed to evaluate the equivocal HER2 (IHC 2+) status of breast cancer using intra- and peritumoral radiomics features of contrast-enhanced mammography (CEM).\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eA total of 131 breast cancer patients with equivocal HER2 (IHC 2+) status of breast cancer were enrolled in the study and divided into training (n\u0026thinsp;=\u0026thinsp;84), internal test (n\u0026thinsp;=\u0026thinsp;22) and prospective test (n\u0026thinsp;=\u0026thinsp;25) cohorts. Radiomics features were extracted from intratumoral and peritumoral regions on CEM and were selected using low variance and least absolute shrinkage and selection operator regression (LASSO). Five radiomics signatures were established based on different intratumoral and peritumoral regions. The nomogram was constructed using the selected signatures and clinical factors by logistic regression analysis. Its predictive performance was compared with the radiomics model and the clinical model. The area under the receiver operator characteristic curve (AUC), sensitivity, specificity, accuracy, the calibration curve, and decision curve analysis (DCA) were used to evaluate predictive performance of the models.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eThe intratumoral signature, 5mm-peritumoral signature, and tumor diameter were used to establish nomogram. Compared to the radiomics model and the clinical model, the nomogram achieved optimal predictive performance, with an AUC of 0.893 in the internal test cohort and an AUC of 0.840 in the prospective test cohort. The calibration curves and DCA showed favorable predictive performance of the nomogram.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eThe nomogram incorporated the intratumoral and peritumoral radiomics signatures of CEM and clinical risk variables has the potential to predict equivocal HER2 (IHC 2+) status of breast cancer preoperatively.\u003c/p\u003e","manuscriptTitle":"Intra- and Peritumoral Radiomics for Predicting Equivocal HER2 (IHC2+) Status of Breast Cancer on Contrast-Enhanced Mammography","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-30 23:27:25","doi":"10.21203/rs.3.rs-7828102/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4b219672-d7c1-4984-abc6-d5c4312aff2d","owner":[],"postedDate":"November 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-03T08:09:22+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-30 23:27:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7828102","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7828102","identity":"rs-7828102","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}