One-year post-acute COVID-19 syndrome and mortality in South Korea: a nationwide matched cohort study using claims data.

Background: Current understanding of post-COVID-19 syndrome in South Korea is primarily based on survey studies or research targeting specific patient groups, such as those hospitalized. Moreover, the majority of relevant studies have been conducted in European and North American populations, which may limit their applicability to the South Korean context. To address this gap, our study explores the one-year outcomes of COVID-19, focusing on the potential post-acute syndrome and all-cause mortality in South Korea. Methods: This retrospective cohort study used nationwide claims data in South Korea, including adults aged >18 with records between January 20, 2020, and February 25, 2021. Patients were classified into COVID-19 and non-COVID-19 groups and matched 1:1 based on propensity scores. Primary outcomes were 12-month post-acute COVID-19 syndrome and all-cause mortality. Results: The study involved 34,802 matched patients. The COVID-19 group had significantly elevated risks of coagulopathies (OR = 2.70 [2.24, 3.28]; p < 0.001), chronic lower respiratory diseases (OR = 1.96 [1.80, 2.14]; p < 0.001), symptoms of the circulatory and respiratory systems (OR = 1.91 [1.80, 2.04]; p < 0.001), mood disorders (OR = 1.67 [1.51, 1.86]; p < 0.001), cardiac diseases (OR = 1.39 [1.21, 1.59]; p < 0.001), and symptoms of cognition, perception, emotional state, and behavior (OR = 1.15 [1.04, 1.27]; p = 0.005). All-cause mortality was higher in the COVID-19 group during the 6 months (OR = 1.34 [1.06, 1.69]; p = 0.015), but gradually decreased, reaching an OR of 0.996 ([0.83, 1.19]; p = 0.964) at 1 year. Conclusion: In South Korea, the 12-month post-acute COVID-19 syndrome includes coagulopathies, respiratory issues, mood disorders, and cardiac diseases. The risk of all-cause mortality post-COVID-19 is heightened for up to 6 months, then significantly decreases and resolves within a year.

Deep learning-based long-term risk evaluation of incident type 2 diabetes using electrocardiogram in a non-diabetic population: a retrospective, multicentre study.

Background: Diabetes is a major public health concern. We aimed to evaluate the long-term risk of incident type 2 diabetes in a non-diabetic population using a deep learning model (DLM) detecting prevalent type 2 diabetes using electrocardiogram (ECG). Methods: In this retrospective study, participants who underwent health checkups at two tertiary hospitals in Seoul, South Korea, between Jan 1, 2001 and Dec 31, 2022 were included. Type 2 diabetes was defined as glucose ≥126 mg/dL or glycated haemoglobin (HbA1c) ≥ 6.5%. For survival analysis on incident type 2 diabetes, we introduced an additional variable, diabetic ECG, which is determined by the DLM trained on ECG and corresponding prevalent diabetes. It was assumed that non-diabetic individuals with diabetic ECG had a higher risk of incident type 2 diabetes than those with non-diabetic ECG. The one-dimensional ResNet-based model was adopted for the DLM, and the Guided Grad-CAM was used to localise important regions of ECG. We divided the non-diabetic group into the diabetic ECG group (false positive) and the non-diabetic ECG (true negative) group according to the DLM decision, and performed a Cox proportional hazard model, considering the occurrence of type 2 diabetes more than six months after the visit. Findings: 190,581 individuals were included in the study with a median follow-up period of 11.84 years. The areas under the receiver operating characteristic curve for prevalent type 2 diabetes detection were 0.816 (0.807-0.825) and 0.762 (0.754-0.770) for the internal and external validations, respectively. The model primarily focused on the QRS duration and, occasionally, P or T waves. The diabetic ECG group exhibited an increased risk of incident type 2 diabetes compared with the non-diabetic ECG group, with hazard ratios of 2.15 (1.82-2.53) and 1.92 (1.74-2.11) for internal and external validation, respectively. Interpretation: In the non-diabetic group, those whose ECG was classified as diabetes by the DLM were at a higher risk of incident type 2 diabetes than those whose ECG was not. Additional clinical research on the relationship between the phenotype of ECG and diabetes to support the results and further investigation with tracked data and various ECG recording systems are suggested for future works. Funding: National Research Foundation of Korea.

Automatic Extraction of Comprehensive Drug Safety Information from Adverse Drug Event Narratives in the Korea Adverse Event Reporting System Using Natural Language Processing Techniques.

INTRODUCTION: Concerns have been raised over the quality of drug safety information, particularly data completeness, collected through spontaneous reporting systems (SRS), although regulatory agencies routinely use SRS data to guide their pharmacovigilance programs. We expected that collecting additional drug safety information from adverse event (ADE) narratives and incorporating it into the SRS database would improve data completeness. OBJECTIVE: The aims of this study were to define the extraction of comprehensive drug safety information from ADE narratives reported through the Korea Adverse Event Reporting System (KAERS) as natural language processing (NLP) tasks and to provide baseline models for the defined tasks. METHODS: This study used ADE narratives and structured drug safety information from individual case safety reports (ICSRs) reported through KAERS between 1 January 2015 and 31 December 2019. We developed the annotation guideline for the extraction of comprehensive drug safety information from ADE narratives based on the International Conference on Harmonisation (ICH) E2B(R3) guideline and manually annotated 3723 ADE narratives. Then, we developed a domain-specific Korean Bidirectional Encoder Representations from Transformers (KAERS-BERT) model using 1.2 million ADE narratives in KAERS and provided baseline models for the task we defined. In addition, we performed an ablation experiment to investigate whether named entity recognition (NER) models were improved when a training dataset contained more diverse ADE narratives. RESULTS: We defined 21 types of word entities, six types of entity labels, and 49 types of relations to formulate the extraction of comprehensive drug safety information as NLP tasks. We obtained a total of 86,750 entities, 81,828 entity labels, and 45,107 relations from manually annotated ADE narratives. The KAERS-BERT model achieved F1-scores of 83.81 and 76.62% on the NER and sentence extraction tasks, respectively, while outperforming other baseline models on all the NLP tasks we defined except the sentence extraction task. Finally, utilizing the NER model for extracting drug safety information from ADE narratives resulted in an average increase of 3.24% in data completeness for KAERS structured data fields. CONCLUSIONS: We formulated the extraction of comprehensive drug safety information from ADE narratives as NLP tasks and developed the annotated corpus and strong baseline models for the tasks. The annotated corpus and models for extracting comprehensive drug safety information can improve the data quality of an SRS database.

An annotated corpus from biomedical articles to construct a drug-food interaction database.

MOTIVATION: While drug-food interaction (DFI) may undermine the efficacy and safety of drugs, DFI detection has been difficult because a well-organized database for DFI did not exist. To construct a DFI database and build a natural language processing system extracting DFI from biomedical articles, we formulated the DFI extraction tasks and manually annotated texts that could have contained DFI information. In this article, we introduced a new annotated corpus for extracting DFI, the DFI corpus. RESULTS: The DFI corpus contains 2270 abstracts of biomedical articles accessible through PubMed and 2498 sentences that contain DFI and/or drug-drug information (DDI), a substantial amount of information about drug/food entities, evidence-levels of abstracts and relations between named entities. BERT models pre-trained on the biomedical domain achieved a F1 score 55.0% in extracting DFI key-sentences. To the best of our knowledge, the DFI corpus is the largest public corpus for drug-food interaction. AVAILABILITY AND IMPLEMENTATION: Our corpus is available at https://github.com/ccadd-snu/corpus-for-DFI-extraction.

A population pharmacokinetic-pharmacodynamic model of YH12852, a highly selective 5-hydroxytryptamine 4 receptor agonist, in healthy subjects and patients with functional constipation.

YH12852, a novel, highly selective 5-hydroxytryptamine 4 (5-HT4 ) receptor agonist, is currently under development to treat patients with functional constipation. In this study, we aimed to develop a pharmacokinetic (PK)-pharmacodynamic (PD) model that adequately described the time courses of the plasma concentrations of YH12852 and its prokinetic effect as assessed by the Gastric Emptying Breath Test (GEBT) and to predict the prokinetic effect of YH12852 at higher doses through PD simulation. We used the plasma concentrations of YH12852 from patients with functional constipation and healthy subjects and the GEBT results from healthy subjects obtained from a phase I/IIa trial. The PK-PD modeling and covariate analysis were performed using NONMEM software. The prokinetic effect of YH12852 was described using a semimechanistic multicompartment PD model and an empirical model by Ghoos et al. A two-compartment model with first-order absorption adequately described the observed concentration-time profiles of YH12852. The semimechanistic multicompartment PD model and the revised Ghoos model with two slope parameters adequately described the observed kPCDt (the percent dose of 13 C excreted in the exhaled air at minute t after completing the test meal, multiplied by 1000) values. YH12852 accelerated gastric emptying even at low doses of 0.05-0.1 mg, and its prokinetic effect was greater in subjects suffering from more severe functional constipation. The PD simulation experiments revealed that the change from baseline in the half time for gastric emptying induced by YH12852 increased in a dose-dependent manner at 0.05-5 mg although the results at doses >0.1 mg were extrapolated. We also showed that the empirical Ghoos model is a special case of the general semimechanistic multicompartment PD model for gastric emptying.

A critical review of the United States regulatory pathways for determining the equivalence of efficacy between CT-P13 and original infliximab (Remicade®).

We evaluated the appropriateness of various equivalence margins for CT-P13, an infliximab biosimilar, in the PLANETRA clinical trial. The 95-95% method was used to independently determine an equivalence margin by pooling the historical clinical trials with original infliximab versus placebo, identified in a systematic literature search. The constancy assumption with the PLANETRA trial was assessed for each identified historical clinical trial to decide which study was scientifically justifiable to be pooled. A sensitivity analysis was performed for each study-pooling scenario. As a result, we identified two historical clinical trials that were deemed appropriate, whereas the PLANETRA trial pooled three additional studies to determine an equivalence margin, which was accepted by the United States Food and Drug Administration. However, those extra clinical trials did not meet the constancy assumption in baseline characteristics, methotrexate dose, and efficacy assessment time. The clinically more appropriate equivalence margin was 5.7 percentage points, which was much narrower than the 12 percentage points applied in the approval of CT-P13. In conclusion, the equivalence claim for CT-P13 to original infliximab in patients with rheumatoid arthritis did not appear to be supported when the constancy assumption was strictly assessed. The equivalence margin for biosimilars could be determined more conservatively.