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 de ned 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): rst 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 nal prognostic model. Results:The best performing models were based on space-resolved and deep learning approaches. Spaceresolved 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.

PurposeThis study was undertaken to examine the impact of screening and race on breast cancer outcomes in Singapore.MethodsAn institutional database was reviewed, and invasive ductal carcinoma (IDC) and ductal carcinoma in situ (DCIS) data were analyzed separately. Overall survival (OS), disease-free survival (DFS), and cancer-specific survival (CSS) were assessed.ResultsThe study included 6,180 IDC and 1,031 DCIS patients. The median follow-up time was 4.1 years. Among IDC patients, Malay women were the youngest when first diagnosed, and were more likely to present with advanced stage disease. Malay women also had the highest proportion of T3 and T4 tumors at 14.2%, compared with Chinese women at 8.7% and Indian women at 9.6% (p<0.001). Malay women had a higher incidence of node-positive disease at 58.3% compared with Chinese women at 46.4% and Indian women at 54.9% (p<0.001). Malay subjects also had higher-grade tumors; 61.8% had grade 3 tumors compared with 45.8% of Chinese women and 52% of Indian women (p<0.001). Furthermore, tumors in Malay subjects were less endocrine-sensitive and more human epidermal growth factor receptor 2 enriched. Malay women had the lowest 5- and 10-year OS, DFS, and CSS rates (p<0.001). After separating clinically and screen-detected tumors, multivariate analysis showed that race was still significant for outcomes. For screen-detected tumors, the OS hazard ratio (HR) for Malay women compared to Chinese women was 5.78 (95% confidence interval [CI], 2.64–12.64), the DFS HR was 2.18 (95% CI, 1.19–3.99), and the CSS HR was 5.93 (95% CI, 2.15–16.39). For DCIS, there were no statistically significant differences in the tumor size, grade, histology subtypes, or hormone sensitivity.ConclusionMalay race is a poor prognostic factor in both clinically and screen-detected IDC. Special attention should be given to the detection and follow-up of breast cancer in this group.

No 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.