Anand, Praveen

INTRODUCTION: Screening for Barrett's esophagus (BE) is suggested in those with risk factors, but remains underutilized. BE/esophageal adenocarcinoma (EAC) risk prediction tools integrating multiple risk factors have been described. However, accuracy remains modest (area under the receiver-operating curve [AUROC] ≤0.7), and clinical implementation has been challenging. We aimed to develop machine learning (ML) BE/EAC risk prediction models from an electronic health record (EHR) database. METHODS: The Clinical Data Analytics Platform, a deidentified EHR database of 6 million Mayo Clinic patients, was used to predict BE and EAC risk. BE and EAC cases and controls were identified using International Classification of Diseases codes and augmented curation (natural language processing) techniques applied to clinical, endoscopy, laboratory, and pathology notes. Cases were propensity score matched to 5 independent randomly selected control groups. An ensemble transformer-based ML model architecture was used to develop predictive models. RESULTS: We identified 8,476 BE cases, 1,539 EAC cases, and 252,276 controls. The BE ML transformer model had an overall sensitivity, specificity, and AUROC of 76%, 76%, and 0.84, respectively. The EAC ML transformer model had an overall sensitivity, specificity, and AUROC of 84%, 70%, and 0.84, respectively. Predictors of BE and EAC included conventional risk factors and additional novel factors, such as coronary artery disease, serum triglycerides, and electrolytes. DISCUSSION: ML models developed on an EHR database can predict incident BE and EAC risk with improved accuracy compared with conventional risk factor-based risk scores. Such a model may enable effective implementation of a minimally invasive screening technology.

Abstract The raging COVID-19 pandemic in India and reports of “vaccine breakthrough infections” globally have raised alarm mandating the characterization of the immuno-evasive features of SARS-CoV-2. Here, we systematically analyzed over 1.3 million SARS-CoV-2 genomes from 178 countries and performed whole-genome viral sequencing from 53 COVID-19 patients, including 20 vaccine breakthrough infections. We identified 116 Spike protein mutations that increased in prevalence during at least one surge in SARS-CoV-2 test positivity in any country over a three-month window. Deletions in the Spike protein N-terminal domain (NTD) are highly enriched for these ‘surge-associated mutations’ (Odds Ratio = 18.2, 95% CI: 7.53-48.7; p=1.465x10-18). In the recent COVID-19 surge in India, an NTD deletion (ΔF157/R158) increased over 10-fold in prevalence from February 2021 (1.1%) to April 2021 (15%). During the recent surge in Chile, an NTD deletion (Δ246-253) increased rapidly over 30-fold in prevalence from January 2021 (0.86%) to April 2021 (33%). Strikingly, these simultaneously emerging deletions associated with surges in different parts of the world both occur at an antigenic supersite that is targeted by neutralizing antibodies. Finally, we generated clinically annotated SARS-CoV-2 whole genome sequences and identified deletions within this NTD antigenic supersite in a patient with vaccine breakthrough infection (Δ156-164) and other deletions from unvaccinated severe COVID-19 patients that could represent emerging deletion-prone regions. Overall, the expanding repertoire of Spike protein deletions throughout the pandemic and their association with case surges and vaccine breakthrough infections point to antigenic minimalism as an emerging evolutionary strategy for SARS-CoV-2 to evade immune responses. This study highlights the urgent need to sequence SARS-CoV-2 genomes at a larger scale globally and to mandate a public health policy for transparent reporting of relevant clinical annotations (e.g. vaccination status) in order to aid the development of comprehensive therapeutic strategies.

Understanding temporal dynamics of COVID-19 symptoms could provide fine-grained resolution to guide clinical decision-making. Here, we use deep neural networks over an institution-wide platform for the augmented curation of clinical notes from 77,167 patients subjected to COVID-19 PCR testing. By contrasting Electronic Health Record (EHR)-derived symptoms of COVID-19-positive (COVID pos ; n = 2,317) versus COVID-19-negative (COVID neg ; n = 74,850) patients for the week preceding the PCR testing date, we identify anosmia/dysgeusia (27.1-fold), fever/chills (2.6-fold), respiratory difficulty (2.2-fold), cough (2.2-fold), myalgia/arthralgia (2-fold), and diarrhea (1.4-fold) as significantly amplified in COVID pos over COVID neg patients. The combination of cough and fever/chills has 4.2-fold amplification in COVID pos patients during the week prior to PCR testing, in addition to anosmia/dysgeusia, constitutes the earliest EHR-derived signature of COVID-19. This study introduces an Augmented Intelligence platform for the real-time synthesis of institutional biomedical knowledge. The platform holds tremendous potential for scaling up curation throughput, thus enabling EHR-powered early disease diagnosis.