![]() Patients with PCR or low RCB scores are more likely to be candidates for breast-conserving surgery sparing a full mastectomy and are also likely to have longer progression-free survival (PFS) and overall survival. Pathological complete response (PCR) (defined as the absence of any residual disease) and residual cancer burden (RCB) (defined on a scale of 0 to 3 with increasing residual disease burden) are often used to assess NAC response via pathological analysis of biopsied or dissected tissue at the end of the NAC treatment course. Neoadjuvant chemotherapy (NAC) is often used to reduce tumor size prior to breast cancer surgery and to minimize distant metastasis with remarkable success. Human epidermal growth factor receptor 2 PReLUs, Receiver operating characteristic area under curve 18FDG, įunding: The author(s) received no specific funding for this work.Ĭompeting interests: The authors have declared that no competing interests exist. I-SPY-1 data used in this paper are available via the. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.ĭata Availability: Codes are available via. Received: MaAccepted: DecemPublished: January 6, 2023Ĭopyright: © 2023 Dammu et al. ![]() PLoS ONE 18(1):Įditor: Sriparna Saha, Indian Institute of Technology Patna, INDIA This approach may prove useful for treatment selection, planning, execution, and mid-treatment adjustment.Ĭitation: Dammu H, Ren T, Duong TQ (2023) Deep learning prediction of pathological complete response, residual cancer burden, and progression-free survival in breast cancer patients. Deep learning using longitudinal multiparametric MRI, demographics, and molecular subtypes accurately predicts PCR, RCB, and PFS in breast cancer patients. Using the best model and data combination, PCR prediction yielded an accuracy of 0.81☐.03 and AUC of 0.83☐.03 RCB prediction yielded an accuracy of 0.80☐.02 and Cohen’s κ of 0.73☐.03 PFS prediction yielded a mean absolute error of 24.6☐.7 months (survival ranged from 6.6 to 127.5 months). The combined pre- and post-neoadjuvant chemotherapy data outperformed either alone. Inclusion of both MRI and non-MRI data outperformed either alone. The Integrated approach outperformed the “Stack” or “Concatenation” CNN. Three (“Integrated”, “Stack” and “Concatenation”) CNN were evaluated using receiver-operating characteristics and mean absolute errors. ![]() The inputs were dynamic-contrast-enhanced (DCE) MRI, and T2- weighted MRI as three-dimensional whole-images without the tumor segmentation, as well as molecular subtypes and demographics. In the I-SPY-1 TRIAL, 155 patients with stage 2 or 3 breast cancer with breast tumors underwent neoadjuvant chemotherapy met the inclusion/exclusion criteria. The goal of this study was to employ novel deep-learning convolutional-neural-network (CNN) to predict pathological complete response (PCR), residual cancer burden (RCB), and progression-free survival (PFS) in breast cancer patients treated with neoadjuvant chemotherapy using longitudinal multiparametric MRI, demographics, and molecular subtypes as inputs. ![]()
0 Comments
Leave a Reply. |