Deep learning with biopsy whole slide images for pretreatment prediction of pathological complete response to neoadjuvant chemotherapy in breast cancer：A multicenter study

IntroductionPredicting pathological complete response (pCR) for patients receiving neoadjuvant chemotherapy (NAC) is crucial in establishing individualized treatment. Whole-slide images (WSIs) of tumor tissues reflect the histopathologic information of the tumor, which is important for therapeutic response effectiveness. In this study, we aimed to investigate whether predictive information for pCR could be detected from WSIs.Materials and methodsWe retrospectively collected data from four cohorts of 874 patients diagnosed with biopsy-proven breast cancer. A deep learning pathological model (DLPM) was constructed to predict pCR using biopsy WSIs in the primary cohort, and it was then validated in three external cohorts. The DLPM could generate a deep learning pathological score (DLPs) for each patient; stromal tumor-infiltrating lymphocytes (TILs) were selected for comparison with DLPs.ResultsThe WSI feature-based DLPM showed good predictive performance with the highest area under the curve (AUC) of 0.72 among the cohorts. Alternatively, the combination of the DLPM and clinical characteristics offered a better prediction performance (AUC >0.70) in all cohorts. We also evaluated the performance of DLPM in three different breast subtypes with the best prediction for the triple-negative breast cancer (TNBC) subtype (AUC: 0.73). Moreover, DLPM combined with clinical characteristics and stromal TILs achieved the highest AUC in the primary cohort (AUC: 0.82) and validation cohort 1 (AUC: 0.80).ConclusionOur study suggested that WSIs integrated with deep learning could potentially predict pCR to NAC in breast cancer. The predictive performance will be improved by combining clinical characteristics. DLPs from DLPM can provide more information compared to stromal TILs for pCR prediction.     

Cardiac-type total anomalous pulmonary venous return is not benign

ObjectiveThe objective of this study was to investigate the association between morphological variation and postsurgical pulmonary vein (PV) stenosis (PPVS) in patients with cardiac total anomalous pulmonary venous connection (TAPVC).MethodsThis single-center, retrospective study included 168 pediatric patients who underwent surgical repair of cardiac TAPVC from 2013 to 2019 (connection to the coronary sinus [CS], n = 136; connection directly to the right atrium [RA], n = 32). Three-dimensional computed tomography modeling and geometric analysis were performed to investigate the morphological features; their relevance to the PPVS was examined.ResultsThe connection type had no association with PPVS (CS type: 18% vs right atrial type: 19%; P = .89) but there was a higher incidence of PPVS in patients with a single PV orifice than > 1 orifice (P < .001). Confluence-to-total PV area ratio (hazard ratio, 4.78, 95% CI, 1.86-12.32; P = .001) and length of drainage route (hazard ratio, 1.22; 95% CI, 1.14-1.31; P < .001) had a 4- and 1-fold increase in the risk for PPVS in the CS type after adjustment for age and preoperative pulmonary venous obstruction. In the right atrial type, those with anomalous PV return to the RA roof were more likely to develop PPVS than to the posterior wall of the RA (P < .001).ConclusionsThe number of inter-junction PV orifice correlated with PPVS development in cardiac TAPVC. The confluence-to-total PV ratio, length of drainage route, and anomalous PV return to the RA roof are important predictors for PPVS. Morphological subcategorization in this clinical setting can potentially assist in surgical decision-making.