scholarly journals Multi-Center Evaluation of Artificial Intelligent Imaging and Clinical Models for Predicting Neoadjuvant Chemotherapy Response in Breast Cancer

2021 ◽  
Author(s):  
Tan Hong Qi ◽  
Ong Hiok Hian ◽  
Arjunan Muthu Kumaran ◽  
Tira J. Tan ◽  
Ryan Shea Tan Ying Cong ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Neoadjuvant Chemotherapy ◽  
Pathological Complete Response ◽  
Complete Response ◽  
Chemotherapy Response ◽  
Clinical Parameters ◽  
Model Based ◽  
Response To Chemotherapy ◽  
Clinical Models

Abstract Background: Neoadjuvant chemotherapy (NAC) plays an important role in the management of locally advanced breast cancer. It allows for downstaging of tumours, potentially allowing for breast conservation. NAC also allows for in-vivo testing of the tumours’ response to chemotherapy and provides important prognostic information. There are currently no clearly defined clinical models that incorporate imaging with clinical data to predict response to NAC. Thus, the aim of this work is to develop a predictive AI model based on routine CT imaging and clinical parameters to predict response to NAC. Methods: The CT scans of 324 patients with NAC from multiple centers in Singapore were used in this study. Four different radiomics models were built for predicting pathological complete response (pCR): first two were based on textural features extracted from peri-tumoral and tumoral regions, the third model based on novel space-resolved radiomics which extract feature maps using voxel-based radiomics and the fourth model based on Deep Learning (DL). Clinical parameters were included to build a final prognostic model. Results: The best performing models were based on space-resolved and deep learning approaches. Space-resolved radiomics improves the clinical AUCs of pCR prediction from 0.743 (0.650 to 0.831) to 0.775 (0.685 to 0.860) and our DL model improved it from 0.743 (0.650 to 0.831) to 0.772 (0.685 to 0.853). The tumoral radiomics model performs the worst with no improvement of the AUC from the clinical model. The peri-tumoral combined model gives moderate performance with an AUC of 0.765 (0.671 to 0.855). Conclusions: Radiomics features extracted from diagnostic CT augments the predictive ability of pathological complete response when combined with clinical features. The novel space-resolved radiomics and deep learning radiomics approaches outperformed conventional radiomics techniques.

scholarly journals Multi-Center Evaluation of Artificial Intelligent Imaging And Clinical Models For Predicting Neoadjuvant Chemotherapy Response In Breast Cancer

2021 ◽  
Author(s):  
Tan Hong Qi ◽  
Ong Hiok Hian ◽  
Arjunan Muthu Kumaran ◽  
Tira J. Tan ◽  
Ryan Shea Tan Ying Cong ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Neoadjuvant Chemotherapy ◽  
Pathological Complete Response ◽  
Complete Response ◽  
Chemotherapy Response ◽  
Clinical Parameters ◽  
Model Based ◽  
Response To Chemotherapy ◽  
Clinical Models

Abstract Background:Neoadjuvant chemotherapy (NAC) plays an important role in the management of locally advanced breast cancer. It allows for downstaging of tumours, potentially allowing for breast conservation. NAC also allows for in-vivo testing of the tumours’ response to chemotherapy and provides important prognostic information. There are currently no clearly defined clinical models that incorporate imaging with clinical data to predict response to NAC. Thus, the aim of this work is to develop a predictive AI model based on routine CT imaging and clinical parameters to predict response to NAC. Methods:The CT scans of 324 patients with NAC from multiple centers in Singapore were used in this study. Four different radiomics models were built for predicting pathological complete response (pCR): first two were based on textural features extracted from peri-tumoral and tumoral regions, the third model based on novel space-resolved radiomics which extract feature maps using voxel-based radiomics and the fourth model based on Deep Learning (DL). Clinical parameters were included to build a final prognostic model. Results:The best performing models were based on space-resolved and deep learning approaches. Space-resolved radiomics improves the clinical AUCs of pCR prediction from 0.743 (0.650 to 0.831) to 0.775 (0.685 to 0.860) and our DL model improved it from 0.743 (0.650 to 0.831) to 0.772 (0.685 to 0.853). The tumoral radiomics model performs the worst with no improvement of the AUC from the clinical model. The peri-tumoral combined model gives moderate performance with an AUC of 0.765 (0.671 to 0.855). Conclusions:Radiomics features extracted from diagnostic CT augments the predictive ability of pathological complete response when combined with clinical features. The novel space-resolved radiomics and deep learning radiomics approaches outperformed conventional radiomics techniques.

Are BRCA1 mutations a predictive factor for anthracycline-based neoadjuvant chemotherapy response in triple-negative breast cancers?

2007 ◽  
Vol 25 (18_suppl) ◽  
pp. 580-580 ◽  
Author(s):  
T. Petit ◽  
M. Wilt ◽  
J. Rodier ◽  
D. Muller ◽  
J. Ghnassia ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Neoadjuvant Chemotherapy ◽  
Triple Negative ◽  
Pathological Complete Response ◽  
Complete Response ◽  
Her2 Overexpression ◽  
Breast Cancers ◽  
Platinum Compounds ◽  
Response To Chemotherapy ◽  
Chemotherapy Efficacy

580 Background: BRCA1 being involved in DNA repair and apoptosis, its mutations may influence response to chemotherapy. In vitro studies demonstrated that loss of BRCA1 function increased sensitivity to platinum compounds and induced resistance to anthracyclines. BRCA1-related breast cancers tend to be ductal carcinomas with high tumor grade, absence of hormonal receptors and no HER2 overexpression, so called triple-negative. We retrospectively analyzed anthracycline-based neoadjuvant chemotherapy efficacy in triple- negative tumors according to BRCA1 status. Methods: 393 breast cancer pts were treated with FEC100 neoadjuvant chemotherapy (FU 500 mg/m2, epirubicine 100 mg/m2, cyclophosphamide 500 mg/m2) between 1/2000 and 12/2006. Out of them, 14% had a triple-negative phenotype (55 pts). Patients with young age at diagnosis or family history of breast cancer were offered genetic testing for BRCA1 and BRCA2 mutations. Twelve of these patients had a BRCA1 deleterious mutation with a triple-negative tumor. Characteristics of these 12 pts at diagnosis were: median age = 38, tumor stage = 7 T2N0, 2 T2N1, 2 T3N0, 1 T3N1. Results: Pathological complete response was defined as absence of invasive tumor in breast and axillary nodes. After 6 cycles of FEC100, 42% of patients with triple-negative tumors (23/55) had a pathological complete response, compared to 17% (2/12) with a BRCA1 mutation. Only one of the 12 BRCA1 patients had an axillary node involvement. Conclusions: In our series, BRCA1 deleterious mutations decreased anthracycline-based chemotherapy efficacy in triple- negative breast cancers. Platinum compounds should be evaluated in these BRCA1-related tumors. No significant financial relationships to disclose.

scholarly journals Deep learning-based predictive biomarker of pathological complete response to neoadjuvant chemotherapy from histological images in breast cancer

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Fengling Li ◽  
Yongquan Yang ◽  
Yani Wei ◽  
Ping He ◽  
Jie Chen ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Deep Learning ◽  
Neoadjuvant Chemotherapy ◽  
Cancer Patients ◽  
Pathological Complete Response ◽  
Predictive Biomarkers ◽  
Predictive Performance ◽  
Complete Response ◽  
Breast Cancer Patients ◽  
Histological Images

Abstract Background Pathological complete response (pCR) is considered a surrogate endpoint for favorable survival in breast cancer patients treated with neoadjuvant chemotherapy (NAC). Predictive biomarkers of treatment response are crucial for guiding treatment decisions. With the hypothesis that histological information on tumor biopsy images could predict NAC response in breast cancer, we proposed a novel deep learning (DL)-based biomarker that predicts pCR from images of hematoxylin and eosin (H&E)-stained tissue and evaluated its predictive performance. Methods In total, 540 breast cancer patients receiving standard NAC were enrolled. Based on H&E-stained images, DL methods were employed to automatically identify tumor epithelium and predict pCR by scoring the identified tumor epithelium to produce a histopathological biomarker, the pCR-score. The predictive performance of the pCR-score was assessed and compared with that of conventional biomarkers including stromal tumor-infiltrating lymphocytes (sTILs) and subtype. Results The pCR-score derived from H&E staining achieved an area under the curve (AUC) of 0.847 in predicting pCR directly, and achieved accuracy, F1 score, and AUC of 0.853, 0.503, and 0.822 processed by the logistic regression method, respectively, higher than either sTILs or subtype; a prediction model of pCR constructed by integrating sTILs, subtype and pCR-score yielded a mean AUC of 0.890, outperforming the baseline sTIL-subtype model by 0.051 (0.839, P  =  0.001). Conclusion The DL-based pCR-score from histological images is predictive of pCR better than sTILs and subtype, and holds the great potentials for a more accurate stratification of patients for NAC.

Predicative value of [18F]-FDG PET scan for pathological complete response to neoadjuvant chemotherapy in breast cancer

2007 ◽  
Vol 25 (18_suppl) ◽  
pp. 505-505 ◽  
Author(s):  
L. Favier ◽  
A. Berriolo-Riedinger ◽  
B. Coudert ◽  
C. Touzery ◽  
J. Riedinger ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Logistic Regression ◽  
Neoadjuvant Chemotherapy ◽  
Cancer Patients ◽  
Predictive Value ◽  
Pathological Complete Response ◽  
Complete Response ◽  
Chemotherapy Response ◽  
Breast Cancer Patients ◽  
Fdg Uptake

505 Background: To evaluate, in breast cancer patients treated by neoadjuvant chemotherapy, the early predictive value of the FDG uptake decrease for the assessment of the pathological complete response (pCR). Methods: Forty seven women with non metastatic with conventional imaging, non inflammatory, large or locally advanced breast cancer were included. Pathological tumour regression determined on surgical resection specimens served as the gold standard for the assessment of the neoadjuvant chemotherapy response. According to the Sataloff classification, patients were classified in two groups: patients with a pathological complete response (pCR) and patients with a pathological non complete response (non pCR). FDG uptake of breast lesions was evaluated before and after the first course of neoadjuvant chemotherapy, using Standard Uptake Value maximum (SUV) corrected by body surface area and glycaemia. Relations between baseline [18F]-FDG uptake and clinical, histopathological and biological parameters were assessed by Mann-Whitney test. Predictive value of the FDG decrease for the assessment of the pCR was studied with logistic regression analysis. Results: An elevated baseline SUV was found independently associated with a high mitotic activity (p<0.002), tumour grading (p<0.004), high score of nuclear pleomorphism (p= 0.03) and positive hormonal receptor status (p<0.005). After completion of chemotherapy, 11 (23%) of the 47 breast tumours examined at surgery showed a pCR while 36 (77%) showed a non pCR. The relative decrease (ΔSUV) after the first course of neoadjuvant chemotherapy was significantly greater in the pCR group than in the non pCR group (p< 10-4). A SUV decrease of 85.4% ± 21.9% in pCR patients versus 22.6% ± 36.6% in non pCR patients was found. ΔSUV<-60% predicted pCR with an accuracy of 87%. With multivariate logistic regression analyses, ΔSUV<-60% was the only predictive factor of the pCR Conclusions: In breast cancer patients treated by neoadjuvant chemotherapy, the FDG uptake decrease, after only one course of treatment, is an early and powerful predictor of the pCR. No significant financial relationships to disclose.

Impact of serum vitamin D on the response and the prognosis in breast cancer patients treated with neoadjuvant chemotherapy.

2019 ◽  
Vol 37 (15_suppl) ◽  
pp. e12088-e12088
Author(s):  
Eriko Tokunaga ◽  
Takanobu Masuda ◽  
Hideki Ijichi ◽  
Chinami Koga ◽  
Wakako Tajiri ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Vitamin D ◽  
Neoadjuvant Chemotherapy ◽  
Pathological Complete Response ◽  
Serum Vitamin ◽  
Complete Response ◽  
Chemotherapy Response ◽  
Free Survival ◽  
Premenopausal Patients ◽  
The Impact

e12088 Background: Neoadjuvant chemotherapy (NAC) is one of the standard therapeutic strategies for operable early breast cancer (BC). A relationship between chemotherapy response and the prognosis has been suggested, especially in HER2+ or triple negative breast cancer (TNBC). Recently, the impact of vitamin D (VD) on pathological complete response (pCR) after NAC has been reported, however, there are no data regarding the relationship between VD level and the efficacy of NAC in Japanese BC patients. Methods: This study included 236 patients with BC at clinical stage I-III, who were treated with anthracycline and taxane-based NAC. HER2-directed therapies were used for HER2+ tumors. Serum 25, hydroxyl VD levels were evaluated using by ELISAusing blood samples obtained before NAC. VD deficiency was defined as <20 ng/ml. Results: Pathological complete response (pCR) was obtained in 52 patients (22.0%). The pCR rate was significantly higher in the patients with HER2+ or TNBC compared with hormone receptor (HR)+/HER2-tumors (p<0.0001). Median VD level was 10.7 ng/ml, and only 7 patients (3.0%) had a VD level ≥20 ng/ml. These results show that most Japanese women are deficient for VD. VD levels increase with age, and the VD levels are significantly higher in postmenopausal patients than in the premenopausal patients (p=0.0029). Patients were dichotomized into VD-low and -high groups at the median value (10.7 ng/ml). Serum VD was not associated with pCR in the all cohort, however, pCR rate was higher in VD-high group in premenopausal patients, especially in the patients withHER2+ or TNBC (p=0.0328). High VD was significantly associated with better prognosis in terms of relapse-free survival (p=0.0327) and distant metastasis-free survival (p=0.0051). This association was observed both in pre- and post-menopausal patients, especially in the patients with HER2+ or TNBC. Conclusions: High VD levels were associated with high pCR rate in premenopausal patients with HER2+ or TNBC. High VD levels were also associated with better prognosis operable BC patients treated with NAC, although VD levels of most Japanese BC patients were lower than the cut-off value.

scholarly journals Association between the nucleosome footprint of plasma DNA and neoadjuvant chemotherapy response for breast cancer

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Xu Yang ◽  
Geng-Xi Cai ◽  
Bo-Wei Han ◽  
Zhi-Wei Guo ◽  
Ying-Song Wu ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Gene Expression ◽  
Neoadjuvant Chemotherapy ◽  
Cancer Patients ◽  
Cell Receptor ◽  
Chemotherapy Response ◽  
Breast Cancer Patients ◽  
Plasma Dna ◽  
Transcription Start Sites ◽  
Response To Chemotherapy

AbstractGene expression signatures have been used to predict the outcome of chemotherapy for breast cancer. The nucleosome footprint of cell-free DNA (cfDNA) carries gene expression information of the original tissues and thus may be used to predict the response to chemotherapy. Here we carried out the nucleosome positioning on cfDNA from 85 breast cancer patients and 85 healthy individuals and two cancer cell lines T-47D and MDA-MB-231 using low-coverage whole-genome sequencing (LCWGS) method. The patients showed distinct nucleosome footprints at Transcription Start Sites (TSSs) compared with normal donors. In order to identify the footprints of cfDNA corresponding with the responses to neoadjuvant chemotherapy in patients, we mapped on nucleosome positions on cfDNA of patients with different responses: responders (pretreatment, n = 28; post-1 cycle, post-3/4 cycles, and post-8 cycles of treatment, n = 12) and nonresponders (pretreatment, n = 10; post-1 cycle, post-3/4 cycles, and post-8 cycles of treatment, n = 10). The coverage depth near TSSs in plasma cfDNA differed significantly between responders and nonresponders at pretreatment, and also after neoadjuvant chemotherapy treatment cycles. We identified 232 TSSs with differential footprints at pretreatment and 321 after treatment and found enrichment in Gene Ontology terms such as cell growth inhibition, tumor suppressor, necrotic cell death, acute inflammatory response, T cell receptor signaling pathway, and positive regulation of vascular endothelial growth factor production. These results suggest that cfDNA nucleosome footprints may be used to predict the efficacy of neoadjuvant chemotherapy for breast cancer patients and thus may provide help in decision making for individual patients.

Sign in / Sign up

Export Citation Format

Share Document