Development and external evaluation of predictions models for mortality of COVID-19 patients using machine learning method.

To predict the mortality of patients with coronavirus disease 2019 (COVID-19). We collected clinical data of COVID-19 patients between January 18 and March 29 2020 in Wuhan, China . Gradient boosting decision tree (GBDT), logistic regression (LR) model, and simplified LR were built to predict the mortality of COVID-19. We also evaluated different models by computing area under curve (AUC), accuracy, positive predictive value (PPV), and negative predictive value (NPV) under fivefold cross-validation. A total of 2924 patients were included in our evaluation, with 257 (8.8%) died and 2667 (91.2%) survived during hospitalization. Upon admission, there were 21 (0.7%) mild cases, 2051 (70.1%) moderate case, 779 (26.6%) severe cases, and 73 (2.5%) critically severe cases. The GBDT model exhibited the highest fivefold AUC, which was 0.941, followed by LR (0.928) and LR-5 (0.913). The diagnostic accuracies of GBDT, LR, and LR-5 were 0.889, 0.868, and 0.887, respectively. In particular, the GBDT model demonstrated the highest sensitivity (0.899) and specificity (0.889). The NPV of all three models exceeded 97%, while their PPV values were relatively low, resulting in 0.381 for LR, 0.402 for LR-5, and 0.432 for GBDT. Regarding severe and critically severe cases, the GBDT model also performed the best with a fivefold AUC of 0.918. In the external validation test of the LR-5 model using 72 cases of COVID-19 from Brunei, leukomonocyte (%) turned to show the highest fivefold AUC (0.917), followed by urea (0.867), age (0.826), and SPO2 (0.704). The findings confirm that the mortality prediction performance of the GBDT is better than the LR models in confirmed cases of COVID-19. The performance comparison seems independent of disease severity. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at(10.1007/s00521-020-05592-1).

A Randomized, Double-Blind, Placebo-Controlled Phase II Study to Evaluate the Efficacy and Safety of Ivarmacitinib (SHR0302) in Adult Patients with Moderate to Severe Alopecia Areata.

BACKGROUND: Alopecia areata (AA) is a CD8+ T cell mediated autoimmune disease characterized by non-scarring hair loss. Ivarmacitinib, a selective oral Janus kinase 1 (JAK1) inhibitor, may interrupt certain cytokine signaling implicated in the pathogenesis of AA. OBJECTIVE: To evaluate the efficacy and safety of ivarmacitinib in adult AA patients who have ≥25% scalp hair loss. METHODS: Eligible patients were randomized 1:1:1:1 to receive ivarmacitinib 2 mg, 4 mg, or 8 mg QD or placebo for 24 weeks. The primary endpoint was percentage change from baseline in Severity of Alopecia Tool (SALT) score at week 24. RESULTS: A total of 94 patients were randomized. At week 24, the least squares mean (LSM) difference in percentage change from baseline in SALT score for ivarmacitinib 2 mg,4 mg, 8 mg, and placebo groups were -30.51% (90% confidence interval [CI]: -45.25, -15.76), -56.11% (90% CI: -70.28, -41.95), -51.01% (90% CI: -65.20, -36.82) and -19.87% (90% CI: -33.99, -5.75), respectively. Two SAEs, follicular lymphoma, and COVID-19 pneumonia were reported. LIMITATIONS: Small sample size limits the generalizability of the results. CONCLUSION: Treatment with ivarmacitinib 4 mg and 8 mg doses in moderate and severe AA patients for 24 weeks was efficacious and generally tolerated.

Self-Masked Aldehyde Inhibitors of Human Cathepsin L Are Potent Anti-CoV-2 Agents.

Cysteine proteases comprise an important class of drug targets, especially for infectious diseases such as Chagas disease (cruzain) and COVID-19 (3CL protease, cathepsin L). Peptide aldehydes have proven to be potent inhibitors for all of these proteases. However, the intrinsic, high electrophilicity of the aldehyde group is associated with safety concerns and metabolic instability, limiting the use of aldehyde inhibitors as drugs. We have developed a novel class of compounds, self-masked aldehyde inhibitors (SMAIs) which are based on the dipeptide aldehyde inhibitor (Cbz-Phe-Phe-CHO, 1), for which the P1 Phe group contains a 1'-hydroxy group, effectively, an o-tyrosinyl aldehyde (Cbz-Phe-o-Tyr-CHO, 2; (Li et al. (2021) J. Med. Chem. 64, 11,267-11,287)). Compound 2 and other SMAIs exist in aqueous mixtures as stable δ-lactols, and apparent catalysis by the cysteine protease cruzain, the major cysteine protease of Trypanosoma cruzi, results in the opening of the lactol ring to afford the aldehydes which then form reversible thiohemiacetals with the enzyme. These SMAIs are also potent, time-dependent inhibitors of human cathepsin L (K i = 11-60 nM), an enzyme which shares 36% amino acid identity with cruzain. As inactivators of cathepsin L have recently been shown to be potent anti-SARS-CoV-2 agents in infected mammalian cells (Mellott et al. (2021) ACS Chem. Biol. 16, 642-650), we evaluated SMAIs in VeroE6 and A549/ACE2 cells infected with SARS-CoV-2. These SMAIs demonstrated potent anti-SARS-CoV-2 activity with values of EC50 = 2-8 µM. We also synthesized pro-drug forms of the SMAIs in which the hydroxyl groups of the lactols were O-acylated. Such pro-drug SMAIs resulted in significantly enhanced anti-SARS-CoV-2 activity (EC50 = 0.3-0.6 µM), demonstrating that the O-acylated-SMAIs afforded a level of stability within infected cells, and are likely converted to SMAIs by the action of cellular esterases. Lastly, we prepared and characterized an SMAI in which the sidechain adjacent to the terminal aldehyde is a 2-pyridonyl-alanine group, a mimic of both phenylalanine and glutamine. This compound (9) inhibited both cathepsin L and 3CL protease at low nanomolar concentrations, and also exerted anti-CoV-2 activity in an infected human cell line.

Self-Masked Aldehyde Inhibitors of Human Cathepsin L Are Potent Anti-CoV-2 Agents

Cysteine proteases comprise an important class of drug targets, especially for infectious diseases such as Chagas disease (cruzain) and COVID-19 (3CL protease, cathepsin L). Peptide aldehydes have proven to be potent inhibitors for all of these proteases. However, the intrinsic, high electrophilicity of the aldehyde group is associated with safety concerns and metabolic instability, limiting the use of aldehyde inhibitors as drugs. We have developed a novel class of compounds, self-masked aldehyde inhibitors (SMAIs) which are based on the dipeptide aldehyde inhibitor (Cbz-Phe-Phe-CHO, 1), for which the P1 Phe group contains a 1′-hydroxy group, effectively, an o-tyrosinyl aldehyde (Cbz-Phe-o-Tyr-CHO, 2;(Li et al. (2021) J. Med. Chem. 64, 11,267–11,287)). Compound 2 and other SMAIs exist in aqueous mixtures as stable δ-lactols, and apparent catalysis by the cysteine protease cruzain, the major cysteine protease of Trypanosoma cruzi, results in the opening of the lactol ring to afford the aldehydes which then form reversible thiohemiacetals with the enzyme. These SMAIs are also potent, time-dependent inhibitors of human cathepsin L (Ki = 11–60 nM), an enzyme which shares 36% amino acid identity with cruzain. As inactivators of cathepsin L have recently been shown to be potent anti-SARS-CoV-2 agents in infected mammalian cells (Mellott et al. (2021) ACS Chem. Biol. 16, 642–650), we evaluated SMAIs in VeroE6 and A549/ACE2 cells infected with SARS-CoV-2. These SMAIs demonstrated potent anti-SARS-CoV-2 activity with values of EC50 = 2–8 μM. We also synthesized pro-drug forms of the SMAIs in which the hydroxyl groups of the lactols were O-acylated. Such pro-drug SMAIs resulted in significantly enhanced anti-SARS-CoV-2 activity (EC50 = 0.3–0.6 μM), demonstrating that the O-acylated-SMAIs afforded a level of stability within infected cells, and are likely converted to SMAIs by the action of cellular esterases. Lastly, we prepared and characterized an SMAI in which the sidechain adjacent to the terminal aldehyde is a 2-pyridonyl-alanine group, a mimic of both phenylalanine and glutamine. This compound (9) inhibited both cathepsin L and 3CL protease at low nanomolar concentrations, and also exerted anti-CoV-2 activity in an infected human cell line.

PaCAR: COVID-19 Pandemic Control Decision Making via Large-Scale Agent-Based Modeling and Deep Reinforcement Learning.

BACKGROUND: Policy makers are facing more complicated challenges to balance saving lives and economic development in the post-vaccination era during a pandemic. Epidemic simulation models and pandemic control methods are designed to tackle this problem. However, most of the existing approaches cannot be applied to real-world cases due to the lack of adaptability to new scenarios and micro representational ability (especially for system dynamics models), the huge computation demand, and the inefficient use of historical information. METHODS: We propose a novel Pandemic Control decision making framework via large-scale Agent-based modeling and deep Reinforcement learning (PaCAR) to search optimal control policies that can simultaneously minimize the spread of infection and the government restrictions. In the framework, we develop a new large-scale agent-based simulator with vaccine settings implemented to be calibrated and serve as a realistic environment for a city or a state. We also design a novel reinforcement learning architecture applicable to the pandemic control problem, with a reward carefully designed by the net monetary benefit framework and a sequence learning network to extract information from the sequential epidemiological observations, such as number of cases, vaccination, and so forth. RESULTS: Our approach outperforms the baselines designed by experts or adopted by real-world governments and is flexible in dealing with different variants, such as Alpha and Delta in COVID-19. PaCAR succeeds in controlling the pandemic with the lowest economic costs and relatively short epidemic duration and few cases. We further conduct extensive experiments to analyze the reasoning behind the resulting policy sequence and try to conclude this as an informative reference for policy makers in the post-vaccination era of COVID-19 and beyond. LIMITATIONS: The modeling of economic costs, which are directly estimated by the level of government restrictions, is rather simple. This article mainly focuses on several specific control methods and single-wave pandemic control. CONCLUSIONS: The proposed framework PaCAR can offer adaptive pandemic control recommendations on different variants and population sizes. Intelligent pandemic control empowered by artificial intelligence may help us make it through the current COVID-19 and other possible pandemics in the future with less cost both of lives and economy. HIGHLIGHTS: We introduce a new efficient, large-scale agent-based epidemic simulator in our framework PaCAR, which can be applied to train reinforcement learning networks in a real-world scenario with a population of more than 10,000,000.We develop a novel learning mechanism in PaCAR, which augments reinforcement learning with sequence learning, to learn the tradeoff policy decision of saving lives and economic development in the post-vaccination era.We demonstrate that the policy learned by PaCAR outperforms different benchmark policies under various reality conditions during COVID-19.We analyze the resulting policy given by PaCAR, and the lessons may shed light on better pandemic preparedness plans in the future.

Zinc-Loaded Black Phosphorus Multifunctional Nanodelivery System Combined With Photothermal Therapy Have the Potential to Treat Prostate Cancer Patients Infected With COVID-19.

Since 2019, coronavirus disease 2019 (COVID-19) has swept the world and become a new virus threatening the health of all mankind. The survey found that prostate cancer accounts for one in three male cancer patients infected with COVID-19. This undoubtedly makes prostate cancer patients face a more difficult situation. Prostate cancer is the second most harmful malignant tumor in men because of its insidious onset, easy metastasis, and easy development into castration-resistant prostate cancer even after treatment. Due to its high immunogenicity and a small number of specific infiltrating T cells with tumor-associated antigens in the tissue, it is difficult to obtain a good therapeutic effect with immune checkpoint blocking therapy alone. Therefore, in the current study, we developed a platform carrying Doxorubicin (DOX)-loaded black phosphate nanometer combined with photothermal therapy (PTT) and found this drug combination stimulated the immungentic cell death (ICD) process in PC-3 cells and DC maturation. More importantly, zinc ions have a good immunomodulatory function against infectious diseases, and can improve the killing ability of the nanosystem against prostate cancer cells. The introduction of Aptamer (Apt) enhances the targeting of the entire nanomedicine. We hope that this excellent combination will lead to effective treatment strategies for prostate cancer patients infected with COVID-19.

Venous thromboembolism risk assessment of surgical patients in Southwest China using real-world data: establishment and evaluation of an improved venous thromboembolism risk model.

BACKGROUND: Venous thromboembolism (VTE) risk assessment in surgical patients is important for the appropriate diagnosis and treatment of patients. The commonly used Caprini model is limited by its inadequate ability to discriminate between risk stratums on the surgical population in southwest China and lengthy risk factors. The purpose of this study was to establish an improved VTE risk assessment model that is accurate and simple. METHODS: This study is based on the clinical data from 81,505 surgical patients hospitalized in the Southwest Hospital of China between January 1, 2019 and June 18, 2021. Among the population, 559 patients developed VTE. An improved VTE risk assessment model, SW-model, was established through Logistic Regression, with comparisons to both Caprini and Random Forest. RESULTS: The SW-model incorporated eight risk factors. The area under the curve (AUC) of SW-model (0.807 [0.758, 0.853], 0.804 [0.765, 0.840]), are significantly superior (p = 0.001 and p = 0.044) to those of the Caprini (0.705 [0.652, 0.757], 0.758 [0.719, 0795]) on two test sets, but inferior (p < 0.001 and p = 0.002) to Random Forest (0.854 [0.814, 0.890], 0.839 [0.806, 0.868]). In decision curve analysis, within threshold range from 0.015 to 0.04, the DCA curves of the SW-model are superior to Caprini and two default strategies. CONCLUSIONS: The SW-model demonstrated a higher discriminative capability to distinguish VTE positive in surgical patients compared with the Caprini model. Compared to Random Forest, Logistic Regression based SW-model provided interpretability which is essential in guarantee the procedure of risk assessment transparent to clinicians.

Potent Anti-SARS-CoV-2 Activity by the Natural Product Gallinamide A and Analogues via Inhibition of Cathepsin L.

Cathepsin L is a key host cysteine protease utilized by coronaviruses for cell entry and is a promising drug target for novel antivirals against SARS-CoV-2. The marine natural product gallinamide A and several synthetic analogues were identified as potent inhibitors of cathepsin L with IC50 values in the picomolar range. Lead molecules possessed selectivity over other cathepsins and alternative host proteases involved in viral entry. Gallinamide A directly interacted with cathepsin L in cells and, together with two lead analogues, potently inhibited SARS-CoV-2 infection in vitro, with EC50 values in the nanomolar range. Reduced antiviral activity was observed in cells overexpressing transmembrane protease, serine 2 (TMPRSS2); however, a synergistic improvement in antiviral activity was achieved when combined with a TMPRSS2 inhibitor. These data highlight the potential of cathepsin L as a COVID-19 drug target as well as the likely need to inhibit multiple routes of viral entry to achieve efficacy.

A Clinical-Stage Cysteine Protease Inhibitor blocks SARS-CoV-2 Infection of Human and Monkey Cells.

Host-cell cysteine proteases play an essential role in the processing of the viral spike protein of SARS coronaviruses. K777, an irreversible, covalent inactivator of cysteine proteases that has recently completed phase 1 clinical trials, reduced SARS-CoV-2 viral infectivity in several host cells: Vero E6 (EC50< 74 nM), HeLa/ACE2 (4 nM), Caco-2 (EC90 = 4.3 µM), and A549/ACE2 (<80 nM). Infectivity of Calu-3 cells depended on the cell line assayed. If Calu-3/2B4 was used, EC50 was 7 nM, but in the ATCC Calu-3 cell line without ACE2 enrichment, EC50 was >10 µM. There was no toxicity to any of the host cell lines at 10-100 µM K777 concentration. Kinetic analysis confirmed that K777 was a potent inhibitor of human cathepsin L, whereas no inhibition of the SARS-CoV-2 cysteine proteases (papain-like and 3CL-like protease) was observed. Treatment of Vero E6 cells with a propargyl derivative of K777 as an activity-based probe identified human cathepsin B and cathepsin L as the intracellular targets of this molecule in both infected and uninfected Vero E6 cells. However, cleavage of the SARS-CoV-2 spike protein was only carried out by cathepsin L. This cleavage was blocked by K777 and occurred in the S1 domain of the SARS-CoV-2 spike protein, a different site from that previously observed for the SARS-CoV-1 spike protein. These data support the hypothesis that the antiviral activity of K777 is mediated through inhibition of the activity of host cathepsin L and subsequent loss of cathepsin L-mediated viral spike protein processing.

Bioinformatic analyses suggest augmented interleukin-17 signaling as the mechanism of COVID-19-associated herpes zoster.

Herpes zoster results from latent varicella zoster virus reactivation in the dorsal root ganglia, causing blistering rash along the dermatomal distribution and post-herpetic neuralgia. Increasing studies indicated that there may be a correlation between herpes zoster and COVID-19. Nevertheless, the detailed pathophysiological mechanism is still unclear. We used bioinformatic analyses to study the potential genetic crosstalk between herpes zoster and COVID-19. COVID-19 and herpes zoster were associated with a similar subset of genes involved in "cytokine-cytokine receptor interaction," "Jak-STAT signaling pathway," and "IL-17 signaling pathway," including TNF, IL10, ESR1, INFG, HLA-A, CRP, STAT3, IL6, IL7, and IL17A. Protein-protein interaction network assay showed that the combined gene set indicated a raised connectivity as compared to herpes zoster or COVID-19 alone, particularly the potentiated interactions with APOE, ARSA, CCR2, CCR5, CXCL13, EGFR, GAL, GP2, HLA-B, HLA-DRB1, IL5, TECTA, and THBS1, and these genes are related to "cytokine-cytokine receptor interaction". Augmented Th17 cell differentiation and the resulting enhanced IL-17 signaling were identified in both COVID-19 and herpes zoster. Our data suggested aberrant interleukin-17 signaling as one possible mechanism through which COVID-19 could raise the risk of herpes zoster.

Self-Masked Aldehyde Inhibitors: A Novel Strategy for Inhibiting Cysteine Proteases.

Cysteine proteases comprise an important class of drug targets, especially for infectious diseases such as Chagas disease (cruzain) and COVID-19 (3CL protease, cathepsin L). Peptide aldehydes have proven to be potent inhibitors for all of these proteases. However, the intrinsic, high electrophilicity of the aldehyde group is associated with safety concerns and metabolic instability, limiting the use of aldehyde inhibitors as drugs. We have developed a novel class of self-masked aldehyde inhibitors (SMAIs) for cruzain, the major cysteine protease of the causative agent of Chagas disease-Trypanosoma cruzi. These SMAIs exerted potent, reversible inhibition of cruzain (Ki* = 18-350 nM) while apparently protecting the free aldehyde in cell-based assays. We synthesized prodrugs of the SMAIs that could potentially improve their pharmacokinetic properties. We also elucidated the kinetic and chemical mechanism of SMAIs and applied this strategy to the design of anti-SARS-CoV-2 inhibitors.

Performance of Diagnostic Model for Differentiating Between COVID-19 and Influenza: A 2-Center Retrospective Study.

BACKGROUND COVID-19 and influenza share many similarities, such as mode of transmission and clinical symptoms. Failure to distinguish the 2 diseases may increase the risk of transmission. A fast and convenient differential diagnosis between COVID-19 and influenza has significant clinical value, especially for low- and middle-income countries with a shortage of nucleic acid detection kits. We aimed to establish a diagnostic model to differentiate COVID-19 and influenza based on clinical data. MATERIAL AND METHODS A total of 493 patients were enrolled in the study, including 282 with COVID-19 and 211 with influenza. All data were collected and reviewed retrospectively. The clinical and laboratory characteristics of all patients were analyzed and compared. We then randomly divided all patients into development sets and validation sets to establish a diagnostic model using multivariate logistic regression analysis. Finally, we validated the diagnostic model using the validation set. RESULTS We preliminarily established a diagnostic model for differentiating COVID-19 from influenza that consisted of 5 variables: age, dry cough, fever, white cell count, and D-dimer. The model showed good performance for differential diagnosis. CONCLUSIONS This initial model including clinical features and laboratory indices effectively differentiated COVID-19 from influenza. Patients with a high score were at a high risk of having COVID-19, while patients with a low score were at a high risk of having influenza. This model could help clinicians quickly identify and isolate cases in the absence of nucleic acid tests, especially during the cocirculation of COVID-19 and influenza. Owing to the study's retrospective nature, further prospective study is needed to validate the accuracy of the model.

A cysteine protease inhibitor blocks SARS-CoV-2 infection of human and monkey cells.

K777 is a di-peptide analog that contains an electrophilic vinyl-sulfone moiety and is a potent, covalent inactivator of cathepsins. Vero E6, HeLa/ACE2, Caco-2, A549/ACE2, and Calu-3, cells were exposed to SARS-CoV-2, and then treated with K777. K777 reduced viral infectivity with EC50 values of inhibition of viral infection of: 74 nM for Vero E6, <80 nM for A549/ACE2, and 4 nM for HeLa/ACE2 cells. In contrast, Calu-3 and Caco-2 cells had EC50 values in the low micromolar range. No toxicity of K777 was observed for any of the host cells at 10-100 µM inhibitor. K777 did not inhibit activity of the papain-like cysteine protease and 3CL cysteine protease, encoded by SARS-CoV-2 at concentrations of ≤ 100 µM. These results suggested that K777 exerts its potent anti-viral activity by inactivation of mammalian cysteine proteases which are essential to viral infectivity. Using a propargyl derivative of K777 as an activity-based probe, K777 selectively targeted cathepsin B and cathepsin L in Vero E6 cells. However only cathepsin L cleaved the SARS-CoV-2 spike protein and K777 blocked this proteolysis. The site of spike protein cleavage by cathepsin L was in the S1 domain of SARS-CoV-2 , differing from the cleavage site observed in the SARS CoV-1 spike protein. These data support the hypothesis that the antiviral activity of K777 is mediated through inhibition of the activity of host cathepsin L and subsequent loss of viral spike protein processing.

Can we predict the severity of coronavirus disease 2019 with a routine blood test?

INTRODUCTION: The ongoing worldwide pandemic of coronavirus disease 2019 (COVID­19) has posed a huge threat to global public health. However, the issue as to whether routine blood tests could be used to monitor and predict the severity and prognosis of COVID­19 has not been comprehensively investigated so far. OBJECTIVES: This study aimed to provide an overview of the association of markers in the routine blood test with the severity of COVID­19. METHODS: PubMed, Embase, Cochrane Library, Wanfang, and China National Knowledge Infrastructure (CNKI) databases were searched to identify studies reporting data on markers in the routine blood test and the severity of COVID­19, published until March 20, 2020. The STATA software was used for meta­analysis. RESULTS: A total of 15 studies with 3090 patients with COVID­19 were included in this analysis. Patients in the nonsevere group, compared with those in the severe group, had lower counts of white blood cells (weighted mean difference [WMD], -0.85 [×109/l]; 95% CI, -1.54 to -0.16; P = 0.02) and neutrophils (WMD, -1.57 [×109/l]; 95% CI, -2.6 to -0.54; P = 0.003), greater counts of lymphocytes (WMD, 0.29 [×109/l]; 95% CI, 0.22-0.36; P <0.001) and platelets (WMD, 19.05 [×109/l]; 95% CI, 3.04-35.06; P = 0.02), and a lower neutrophil­to­lymphocyte (NLR) ratio (WMD, -2.48; 95% CI, -3.81 to -1.15; P <0.001). There was no difference in the monocyte count (WMD, 0.01 [×109/l]; 95% CI, -0.01 to 0.03; P = 0.029) between these 2 groups. Sensitivity analysis and meta­analysis based on standard mean difference did not change the conclusions regarding neutrophils, lymphocytes, and NLR, but yielded inconsistent results for white blood cells and platelets. CONCLUSIONS: Severe patients had more neutrophils, higher NLR level, and fewer lymphocytes than non-severe patients with COVID-19. Measurement of these markers might assist clinicians to monitor and predict the severity and prognosis of COVID-19.