ABSTRACT
Background: Camostat, a serine protease inhibitor, prevents activation of the SARS-CoV-2 spike protein and blocks SARS-CoV-2 infection in vitro. We studied the safety and antiviral and clinical efficacy of orally administered camostat in non-hospitalized adults with mild-moderate COVID-19. Methods: ACTIV-2/A5401 is a platform trial to evaluate therapies for non-hospitalized adults with mild-moderate COVID-19. In a Phase II portion of the study, participants were enrolled within 10 days of COVID-19 related symptom onset and randomized to camostat 200 mg orally every 6 hours for 7 days or the pooled placebo group. Objectives were to evaluate the safety and efficacy of camostat to reduce the duration of COVID-19 symptoms and increase the proportion of participants with SARS-CoV-2 RNA below the lower limit of quantification (LLoQ) from nasopharyngeal (NP) swabs on days 3, 7, and 14. Participants completed a study diary from day 0 to day 28 scoring COVID-19 symptoms as absent, mild, moderate, or severe. Results: Of the 224 participants enrolled from 54 US sites, 215 participants (108 camostat, 107 placebo) initiated study intervention and formed the modified intent-to-treat population. Fifty-four percent were female, >99% cis-gender, 85% White, 9% Black, and 51% Latinx. Median age was 37 years;47% reported ≤5 days of symptoms at study entry and 26% met the protocol definition of higher risk of progression to severe COVID-19. Most frequent symptoms on day 0 were cough (86%), fatigue (85%), nasal obstruction/congestion (71%) and body/muscle aches (71%). There was no significant difference between camostat and placebo arms in grade 3 or higher adverse events (7.4% vs. 6.5%, respectively). Median (Q1, Q3) time to symptom improvement was 9 days for both camostat (5, 20) and placebo (6, 19). There were no significant differences in the proportion of participants with NP SARS-CoV-2 RNA<="" div=""> Conclusion: Camostat was well-tolerated. Despite compelling in vitro data, camostat did not show evidence of antiviral or clinical efficacy in ACTIV-2/A5401. This highlights the critical importance of randomized controlled trials in the evaluation of therapies for COVID-19.
ABSTRACT
The automated transcription of spoken language, and meetings, in particular, is becoming more widespread as automatic speech recognition systems are becoming more accurate. This trend has significantly accelerated since the outbreak of the COVID-19 pandemic, which led to a major increase in the number of online meetings. However, the transcription of spoken language has not received much attention from the NLP community compared to documents and other forms of written language. In this paper, we study a variation of the summarization problem over the transcription of spoken language: given a transcribed meeting, and an action item (i.e., a commitment or request to perform a task), our goal is to generate a coherent and self-contained rephrasing of the action item. To this end, we compiled a novel dataset of annotated meeting transcripts, including human rephrasing of action items. We use state-of-the-art supervised text generation techniques and establish a strong baseline based on BART and UniLM (two pretrained transformer models). Due to the nature of natural speech, language is often broken and incomplete and the task is shown to be harder than an analogous task over email data. Particularly, we show that the baseline models can be greatly improved once models are provided with additional information. We compare two approaches: one incorporating features extracted by coreference-resolution. Additional annotations are used to train an auxiliary model to detect the relevant context in the text. Based on the systematic human evaluation, our best models exhibit near-human-level rephrasing capability on a constrained subset of the problem. © 2021 Association for Computational Linguistics