Prediction of SARS-CoV-2-positivity from million-scale complete blood counts using machine learning.

Background: The Complete Blood Count (CBC) is a commonly used low-cost test that measures white blood cells, red blood cells, and platelets in a person's blood. It is a useful tool to support medical decisions, as intrinsic variations of each analyte bring relevant insights regarding potential diseases. In this study, we aimed at developing machine learning models for COVID-19 diagnosis through CBCs, unlocking the predictive power of non-linear relationships between multiple blood analytes. Methods: We collected 809,254 CBCs and 1,088,385 RT-PCR tests for SARS-Cov-2, of which 21% (234,466) were positive, from 900,220 unique individuals. To properly screen COVID-19, we also collected 120,807 CBCs of 16,940 individuals who tested positive for other respiratory viruses. We proposed an ensemble procedure that combines machine learning models for different respiratory infections and analyzed the results in both the first and second waves of COVID-19 cases in Brazil. Results: We obtain a high-performance AUROC of 90 + % for validations in both scenarios. We show that models built solely of SARS-Cov-2 data are biased, performing poorly in the presence of infections due to other RNA respiratory viruses. Conclusions: We demonstrate the potential of a novel machine learning approach for COVID-19 diagnosis based on a CBC and show that aggregating information about other respiratory diseases was essential to guarantee robustness in the results. Given its versatile nature, low cost, and speed, we believe that our tool can be particularly useful in a variety of scenarios-both during the pandemic and after.

Assessment of initial SARS-CoV-2 seroprevalence in the most affected districts in the municipality of São Paulo, Brazil.

BACKGROUND: São Paulo city has been one of the regions most affected by the COVID-19 pandemic in Brazil. Frequent asymptomatic and oligosymptomatic infections and poor access to diagnostic tests make serosurveys crucial to monitor the magnitude of the epidemic and to inform public health policies, such as vaccination plans. OBJECTIVES: To estimate, early in the epidemic, the seroprevalence of antibodies to SARS-CoV-2 in adults living in the six most affected districts in São Paulo city, and to assess potential associated risk factors. METHODS: This was a cross-sectional population-based survey of 1,152 households randomly selected from 72 census tracts. During the period May 4-12, 2020, 463 participants completed a questionnaire on sociodemographic characteristics and history of symptoms in the past two weeks, and provided a blood sample. Prevalence of SARS-CoV-2 antibodies was the outcome of interest and was estimated based on results of two immunoassays, Maglumi SARS-CoV-2 chemiluminescence assay Immunoglobulin (Ig) M (IgM) and IgG, and Roche electrochemiluminescence assay total Ig. Serum samples reactive to either assay were considered positive. RESULTS: Weighted overall seroprevalence was 6% (95%CI 3.9-8.3%). No association was observed between seropositivity and sex, age group or education level. Participants who reported black and brown skin color showed a 2.7 fold higher prevalence than people with white skin (p = 0.007). Among the 30 seropositive individuals, 14 (46.6%) reported no COVID-19 compatible symptoms in the past two weeks. CONCLUSION: This study represents the first assessment of SARS-CoV-2 seroprevalence in the city of São Paulo and 6% is the baseline estimate of a series of population-based seroprevalence surveys. Serological screening using sound serological assays is the key tool to monitoring temporal and geographic changes in the spread of the virus through an important epicenter of the COVID-19 pandemic in Brazil. Ultimately, it may inform prevention and control efforts.

Rationale and design of the "Tocilizumab in patients with moderate to severe COVID-19: an open-label multicentre randomized controlled" trial (TOCIBRAS). / Justificativa e delineamento do estudo "Tocilizumabe em pacientes com COVID-19 moderado a grave: estudo aberto, multicêntrico, randomizado, controlado" (TOCIBRAS).

INTRODUCTION: Pro-inflammatory markers play a significant role in the disease severity of patients with COVID-19. Thus, anti-inflammatory therapies are attractive agents for potentially combating the uncontrolled inflammatory cascade in these patients. We designed a trial testing tocilizumab versus standard of care intending to improve the outcomes by inhibiting interleukin-6, an important inflammatory mediator in COVID-19. METHODS AND ANALYSIS: This open-label multicentre randomized controlled trial will compare clinical outcomes of tocilizumab plus standard of care versus standard of care alone in patients with moderate to severe COVID-19. Two of the following four criteria are required for protocol enrolment: D-dimer > 1,000ng/mL; C reactive protein > 5mg/dL, ferritin > 300mg/dL, and lactate dehydrogenase > upper limit of normal. The primary objective will be to compare the clinical status on day 15, as measured by a 7-point ordinal scale applied in COVID-19 trials worldwide. The primary endpoint will be assessed by an ordinal logistic regression assuming proportional odds ratios adjusted for stratification variables (age and sex). ETHICS AND DISSEMINATION: The TOCIBRAS protocol was approved by local and central (national) ethical committees in Brazil following current national and international guidelines/directives. Each participating center had the study protocol approved by their institutional review boards before initiating protocol enrolment. The data derived from this trial will be published regardless of the results. If proven active, this strategy could alleviate the consequences of the inflammatory response in COVID-19 patients and improve their clinical outcomes.

INTRODUÇÃO: Os marcadores pró-inflamatórios desempenham papel importante na severidade de pacientes com COVID-19. Assim, terapêuticas anti-inflamatórias são agentes interessantes para potencialmente combater a cascata inflamatória descontrolada em tais pacientes. Delineamos um ensaio para testar tocilizumabe em comparação com o tratamento padrão, tendo como objetivo melhorar os desfechos por meio da inibição da interleucina 6, um importante mediador inflamatório na COVID-19. MÉTODOS E ANÁLISES: Este será um estudo aberto multicêntrico, randomizado e controlado, que comparará os desfechos de pacientes tratados com tocilizumabe mais tratamento padrão com o tratamento padrão isoladamente em pacientes com COVID-19 moderada a grave. Como critérios de inclusão, serão exigidos dois dos quatro critérios a seguir: dosagens de dímero D acima de 1.000ng/mL, proteína C-reativa acima de 5mg/dL, ferritina acima de 300mg/dL e desidrogenase lática acima do limite superior do normal. O objetivo primário será comparar a condição clínica no dia 15, conforme avaliação por meio de escala ordinal de 7 pontos aplicada nos estudos de COVID-19 em todo o mundo. O desfecho primário será avaliado por regressão logística ordinal assumindo razões de propensão proporcionais ajustadas pelas variáveis de estratificação (idade e sexo). ÉTICA E DISSEMINAÇÃO: O TOCIBRAS foi aprovado pelos comitês de ética locais e central (nacional) do Brasil em conformidade com as atuais diretrizes e orientações nacionais e internacionais. Cada centro participante obteve aprovação do estudo por parte de seu comitê de ética em pesquisa, antes de iniciar as inscrições no protocolo. Os dados derivados deste ensaio serão publicados independentemente de seus resultados. Se tiver sua efetividade comprovada, esta estratégia terapêutica poderá aliviar as consequências da resposta inflamatória na COVID-19 e melhorar os resultados clínicos.

Establishing a mass spectrometry-based system for rapid detection of SARS-CoV-2 in large clinical sample cohorts.

The outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is pressing public health systems around the world, and large population testing is a key step to control this pandemic disease. Here, we develop a high-throughput targeted proteomics assay to detect SARS-CoV-2 nucleoprotein peptides directly from nasopharyngeal and oropharyngeal swabs. A modified magnetic particle-based proteomics approach implemented on a robotic liquid handler enables fully automated preparation of 96 samples within 4 hours. A TFC-MS system allows multiplexed analysis of 4 samples within 10 min, enabling the processing of more than 500 samples per day. We validate this method qualitatively (Tier 3) and quantitatively (Tier 1) using 985 specimens previously analyzed by real-time RT-PCR, and detect up to 84% of the positive cases with up to 97% specificity. The presented strategy has high sample stability and should be considered as an option for SARS-CoV-2 testing in large populations.