👤 Carlos H Barrios

Coronary angiography is the primary procedure for diagnosis and management decisions in coronary artery disease (CAD), but ad-hoc visual assessment of angiograms has high variability. Here we report a fully automated approach to interpret angiographic coronary artery stenosis from standard coronary angiograms. Using 13,843 angiographic studies from 11,972 adult patients at University of California, San Francisco (UCSF), between April 1, 2008 and December 31, 2019, we train neural networks to accomplish four sequential necessary tasks for automatic coronary artery stenosis localization and estimation. Algorithms are internally validated against criterion-standard labels for each task in hold-out test datasets. Algorithms are then externally validated in real-world angiograms from the University of Ottawa Heart Institute (UOHI) and also retrained using quantitative coronary angiography (QCA) data from the Montreal Heart Institute (MHI) core lab. The CathAI system achieves state-of-the-art performance across all tasks on unselected, real-world angiograms. Positive predictive value, sensitivity and F1 score are all ≥90% to identify projection angle and ≥93% for left/right coronary artery angiogram detection. To predict obstructive CAD stenosis (≥70%), CathAI exhibits an AUC of 0.862 (95% CI: 0.843-0.880). In UOHI external validation, CathAI achieves AUC 0.869 (95% CI: 0.830-0.907) to predict obstructive CAD. In the MHI QCA dataset, CathAI achieves an AUC of 0.775 (95%. CI: 0.594-0.955) after retraining. In conclusion, multiple purpose-built neural networks can function in sequence to accomplish automated analysis of real-world angiograms, which could increase standardization and reproducibility in angiographic coronary stenosis assessment. Show less

In an interim analysis of this phase 3 trial, the addition of pembrolizumab to chemotherapy resulted in longer progression-free survival than chemotherapy alone among patients with advanced triple-negative breast cancer whose tumors expressed programmed death ligand 1 (PD-L1) with a combined positive score (CPS; the number of PD-L1-staining tumor cells, lymphocytes, and macrophages, divided by the total number of viable tumor cells, multiplied by 100) of 10 or more. The results of the final analysis of overall survival have not been reported. We randomly assigned patients with previously untreated locally recurrent inoperable or metastatic triple-negative breast cancer in a 2:1 ratio to receive pembrolizumab (200 mg) every 3 weeks plus the investigator's choice of chemotherapy (nanoparticle albumin-bound paclitaxel, paclitaxel, or gemcitabine-carboplatin) or placebo plus chemotherapy. The primary end points were progression-free survival (reported previously) and overall survival among patients whose tumors expressed PD-L1 with a CPS of 10 or more (the CPS-10 subgroup), among patients whose tumors expressed PD-L1 with a CPS of 1 or more (the CPS-1 subgroup), and in the intention-to-treat population. Safety was also assessed. A total of 847 patients underwent randomization: 566 were assigned to the pembrolizumab-chemotherapy group, and 281 to the placebo-chemotherapy group. The median follow-up was 44.1 months. In the CPS-10 subgroup, the median overall survival was 23.0 months in the pembrolizumab-chemotherapy group and 16.1 months in the placebo-chemotherapy group (hazard ratio for death, 0.73; 95% confidence interval [CI], 0.55 to 0.95; two-sided P = 0.0185 [criterion for significance met]); in the CPS-1 subgroup, the median overall survival was 17.6 and 16.0 months in the two groups, respectively (hazard ratio, 0.86; 95% CI, 0.72 to 1.04; two-sided P = 0.1125 [not significant]); and in the intention-to-treat population, the median overall survival was 17.2 and 15.5 months, respectively (hazard ratio, 0.89; 95% CI, 0.76 to 1.05 [significance not tested]). Adverse events of grade 3, 4, or 5 that were related to the trial regimen occurred in 68.1% of the patients in the pembrolizumab-chemotherapy group and in 66.9% in the placebo-chemotherapy group, including death in 0.4% of the patients in the pembrolizumab-chemotherapy group and in no patients in the placebo-chemotherapy group. Among patients with advanced triple-negative breast cancer whose tumors expressed PD-L1 with a CPS of 10 or more, the addition of pembrolizumab to chemotherapy resulted in significantly longer overall survival than chemotherapy alone. (Funded by Merck Sharp and Dohme; KEYNOTE-355 ClinicalTrials.gov number, NCT02819518.). Show less