Journal of multidisciplinary healthcare · 2026
Interpretable Dynamic MRI-ITH Model for Predicting Neoadjuvant Chemotherapy Response in Breast Cancer: A Multicenter Study.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Wang Mengshen, Liu Xiaohua, Ding Wei, Xu Kai, Feng Jingyan, Lyu Di
Аннотация
OBJECTIVE: This study aimed to develop an interpretable model integrating dynamic MRI intratumoral heterogeneity (ITH) scores for early assessment of pathological complete response (pCR) in breast cancer patients receiving neoadjuvant chemotherapy (NAC).
METHODS: A total of 400 breast cancer patients from three centers were prospectively enrolled. Among them, 300 patients from the Affiliated Hospital of Xuzhou Medical University were randomly assigned in a 7:3 ratio to a training set (n = 210) and an internal validation set (n = 90), while 50 patients from Xuzhou Cancer Hospital and 50 patients from Xuzhou First People's Hospital constituted the external validation set. Clinicopathological characteristics and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data were collected. The baseline MRI-ITH score (ITH0) and dynamic changes at 2 weeks (MRI-Delta-ITH1) and 4 weeks (MRI-Delta-ITH2) after treatment initiation were calculated. Seven predictive models were constructed using logistic regression, and SHAP analysis was used to interpret feature contributions.
RESULTS: The model integrating clinical features and MRI-ΔITH2 yielded the best performance, with area under the receiver operating characteristic curve (AUC) values of 0.940, 0.873, and 0.917 in the training, internal validation, and external validation sets, respectively. SHAP analysis revealed that MRI-ΔITH2 (31.7%), PR status (24.3%), and HER-2 status (18.8%) were the core predictive factors.
CONCLUSION: A predictive model integrating dynamic MRI-ITH scores with clinicopathological features demonstrated favorable performance for early assessment of pCR after NAC in breast cancer patients. Further multicenter validation is warranted before clinical translation.
METHODS: A total of 400 breast cancer patients from three centers were prospectively enrolled. Among them, 300 patients from the Affiliated Hospital of Xuzhou Medical University were randomly assigned in a 7:3 ratio to a training set (n = 210) and an internal validation set (n = 90), while 50 patients from Xuzhou Cancer Hospital and 50 patients from Xuzhou First People's Hospital constituted the external validation set. Clinicopathological characteristics and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data were collected. The baseline MRI-ITH score (ITH0) and dynamic changes at 2 weeks (MRI-Delta-ITH1) and 4 weeks (MRI-Delta-ITH2) after treatment initiation were calculated. Seven predictive models were constructed using logistic regression, and SHAP analysis was used to interpret feature contributions.
RESULTS: The model integrating clinical features and MRI-ΔITH2 yielded the best performance, with area under the receiver operating characteristic curve (AUC) values of 0.940, 0.873, and 0.917 in the training, internal validation, and external validation sets, respectively. SHAP analysis revealed that MRI-ΔITH2 (31.7%), PR status (24.3%), and HER-2 status (18.8%) were the core predictive factors.
CONCLUSION: A predictive model integrating dynamic MRI-ITH scores with clinicopathological features demonstrated favorable performance for early assessment of pCR after NAC in breast cancer patients. Further multicenter validation is warranted before clinical translation.