Early Prediction of Breast Cancer Recurrence for Patients Treated with Neoadjuvant Chemotherapy: A Transfer Learning Approach on DCE-MRIs

Cancer treatment planning benefits from an accurate early prediction of the treatment efficacy. The goal of this study is to give an early prediction of three-year Breast Cancer Recurrence (BCR) for patients who underwent neoadjuvant chemotherapy. We addressed the task from a new perspective based on transfer learning applied to pre-treatment and early-treatment DCE-MRI scans. Firstly, low-level features were automatically extracted from MR images using a pre-trained Convolutional Neural Network (CNN) architecture without human intervention. Subsequently, the prediction model was built with an optimal subset of CNN features and evaluated on two sets of patients from I-SPY1 TRIAL and BREAST-MRI-NACT-Pilot public databases: a fine-tuning dataset (70 not recurrent and 26 recurrent cases), which was primarily used to find the optimal subset of CNN features, and an independent test (45 not recurrent and 17 recurrent cases), whose patients had not been involved in the feature selection process. The best results were achieved when the optimal CNN features were augmented by four clinical variables (age, ER, PgR, HER2+), reaching an accuracy of 91.7% and 85.2%, a sensitivity of 80.8% and 84.6%, a specificity of 95.7% and 85.4%, and an AUC value of 0.93 and 0.83 on the fine-tuning dataset and the independent test, respectively. Finally, the CNN features extracted from pre-treatment and early-treatment exams were revealed to be strong predictors of BCR.

Download Full-text

Early prediction of neoadjuvant chemotherapy response by exploiting a transfer learning approach on breast DCE-MRIs

Scientific Reports ◽

10.1038/s41598-021-93592-z ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Maria Colomba Comes ◽

Annarita Fanizzi ◽

Samantha Bove ◽

Vittorio Didonna ◽

Sergio Diotaiuti ◽

...

Keyword(s):

Neoadjuvant Chemotherapy ◽

Transfer Learning ◽

Early Treatment ◽

Fine Tuning ◽

Svm Classifier ◽

Learning Approach ◽

Early Prediction ◽

Low Level ◽

Independent Test ◽

Pre Treatment

AbstractThe dynamic contrast-enhanced MR imaging plays a crucial role in evaluating the effectiveness of neoadjuvant chemotherapy (NAC) even since its early stage through the prediction of the final pathological complete response (pCR). In this study, we proposed a transfer learning approach to predict if a patient achieved pCR (pCR) or did not (non-pCR) by exploiting, separately or in combination, pre-treatment and early-treatment exams from I-SPY1 TRIAL public database. First, low-level features, i.e., related to local structure of the image, were automatically extracted by a pre-trained convolutional neural network (CNN) overcoming manual feature extraction. Next, an optimal set of most stable features was detected and then used to design an SVM classifier. A first subset of patients, called fine-tuning dataset (30 pCR; 78 non-pCR), was used to perform the optimal choice of features. A second subset not involved in the feature selection process was employed as an independent test (7 pCR; 19 non-pCR) to validate the model. By combining the optimal features extracted from both pre-treatment and early-treatment exams with some clinical features, i.e., ER, PgR, HER2 and molecular subtype, an accuracy of 91.4% and 92.3%, and an AUC value of 0.93 and 0.90, were returned on the fine-tuning dataset and the independent test, respectively. Overall, the low-level CNN features have an important role in the early evaluation of the NAC efficacy by predicting pCR. The proposed model represents a first effort towards the development of a clinical support tool for an early prediction of pCR to NAC.

Download Full-text

Early assessment of shear wave elastography parameters foresees the response to neoadjuvant chemotherapy in patients with invasive breast cancer

Breast Cancer Research ◽

10.1186/s13058-021-01429-4 ◽

2021 ◽

Vol 23 (1) ◽

Author(s):

Juanjuan Gu ◽

Eric C. Polley ◽

Max Denis ◽

Jodi M. Carter ◽

Sandhya Pruthi ◽

...

Keyword(s):

Breast Cancer ◽

Neoadjuvant Chemotherapy ◽

Shear Wave ◽

Invasive Breast Cancer ◽

Shear Wave Elastography ◽

Early Prediction ◽

Early Assessment ◽

Significant Difference ◽

Mass Size ◽

Response To Neoadjuvant Chemotherapy

Abstract Background Early prediction of tumor response to neoadjuvant chemotherapy (NACT) is crucial for optimal treatment and improved outcome in breast cancer patients. The purpose of this study is to investigate the role of shear wave elastography (SWE) for early assessment of response to NACT in patients with invasive breast cancer. Methods In a prospective study, 62 patients with biopsy-proven invasive breast cancer were enrolled. Three SWE studies were conducted on each patient: before, at mid-course, and after NACT but before surgery. A new parameter, mass characteristic frequency (fmass), along with SWE measurements and mass size was obtained from each SWE study visit. The clinical biomarkers were acquired from the pre-NACT core-needle biopsy. The efficacy of different models, generated with the leave-one-out cross-validation, in predicting response to NACT was shown by the area under the receiver operating characteristic curve and the corresponding sensitivity and specificity. Results A significant difference was found for SWE parameters measured before, at mid-course, and after NACT between the responders and non-responders. The combination of Emean2 and mass size (s2) gave an AUC of 0.75 (0.95 CI 0.62–0.88). For the ER+ tumors, the combination of Emean_ratio1, s1, and Ki-67 index gave an improved AUC of 0.84 (0.95 CI 0.65–0.96). For responders, fmass was significantly higher during the third visit. Conclusions Our study findings highlight the value of SWE estimation in the mid-course of NACT for the early prediction of treatment response. For ER+ tumors, the addition of Ki-67improves the predictive power of SWE. Moreover, fmass is presented as a new marker in predicting the endpoint of NACT in responders.

Download Full-text