Severe Mpox Among People With Advanced Human Immunodeficiency Virus Receiving Prolonged Tecovirimat in New York City.

Severe mpox has been observed in people with advanced human immunodeficiency virus (HIV). We describe clinical outcomes of 13 patients with advanced HIV (CD4 <200 cells/µL), severe mpox, and multiorgan involvement. Despite extended tecovirimat courses and additional agents, including vaccinia immune globulin, cidofovir, and brincidofovir, this group experienced prolonged hospitalizations and high mortality.

The role of socio-economic disparities in the relative success and persistence of SARS-CoV-2 variants in New York City in early 2021.

Socio-economic disparities were associated with disproportionate viral incidence between neighborhoods of New York City (NYC) during the first wave of SARS-CoV-2. We investigated how these disparities affected the co-circulation of SARS-CoV-2 variants during the second wave in NYC. We tested for correlation between the prevalence, in late 2020/early 2021, of Alpha, Iota, Iota with E484K mutation (Iota-E484K), and B.1-like genomes and pre-existing immunity (seropositivity) in NYC neighborhoods. In the context of varying seroprevalence we described socio-economic profiles of neighborhoods and performed migration and lineage persistence analyses using a Bayesian phylogeographical framework. Seropositivity was greater in areas with high poverty and a larger proportion of Black and Hispanic or Latino residents. Seropositivity was positively correlated with the proportion of Iota-E484K and Iota genomes, and negatively correlated with the proportion of Alpha and B.1-like genomes. The proportion of persisting Alpha lineages declined over time in locations with high seroprevalence, whereas the proportion of persisting Iota-E484K lineages remained the same in high seroprevalence areas. During the second wave, the geographic variation of standing immunity, due to disproportionate disease burden during the first wave of SARS-CoV-2 in NYC, allowed for the immune evasive Iota-E484K variant, but not the more transmissible Alpha variant, to circulate in locations with high pre-existing immunity.

Reasons for Discordance between 68Ga-PSMA-PET and Magnetic Resonance Imaging in Men with Metastatic Prostate Cancer.

BACKGROUND: PSMA PET has emerged as a "gold standard" imaging modality for assessing prostate cancer metastases. However, it is not universally available, and this limits its impact. In contrast, whole-body MRI is much more widely available but misses more lesions. This study aims to improve the interpretation of whole-body MRI by comparing false negative scans retrospectively to PSMA PET. METHODS: This study was a retrospective sub-analysis of a prospectively collected database of patients who participated in a clinical trial of PSMA PET/MRI comparing PSMA PET and whole-body MRI from 2018-2021. Subjects whose separately read PSMA PET and MRI diagnostic reports showed discrepancies ("false negative" MRI cases) were selected for sub-analysis. The cases were reviewed by the same attending radiologist who originally read the scans. The radiologist noted specific features on MRI indicating metastatic disease that were initially missed. RESULTS: Of 263 cases, 38 (14%) met the inclusion criteria and were reviewed. Six classes of mpMRI false negatives were identified: anatomically normal (18, 47%), atypical MRI appearance (6, 16%), mischaracterization (1, 3%), undercall (6, 16%), obscured (4, 11%), and no abnormality on MRI (3, 8%). Considering that the atypical and undercalled cases could have been adjusted in retrospect, and that 4 additional cases had positive lesions to the same extent and 11 further cases had disease confined to the pelvis, only 11 (4%) of the original 263 would have had disease outside of a conventional radiation treatment plan. CONCLUSION: Notably, almost 50% of the cases, including most lymph node metastases, were anatomically normal using standard criteria. This suggests that current anatomic criteria for evaluating prostate cancer lymph node metastases are not ideal, and there is a need for improved criteria. In addition, 32% of cases involved some element of human interpretive error, and, therefore, improving reader training may lead to more accurate results.

SARS-CoV-2 lineage assignments using phylogenetic placement/UShER are superior to pangoLEARN machine-learning method.

With the rapid spread and evolution of SARS-CoV-2, the ability to monitor its transmission and distinguish among viral lineages is critical for pandemic response efforts. The most commonly used software for the lineage assignment of newly isolated SARS-CoV-2 genomes is pangolin, which offers two methods of assignment, pangoLEARN and pUShER. PangoLEARN rapidly assigns lineages using a machine-learning algorithm, while pUShER performs a phylogenetic placement to identify the lineage corresponding to a newly sequenced genome. In a preliminary study, we observed that pangoLEARN (decision tree model), while substantially faster than pUShER, offered less consistency across different versions of pangolin v3. Here, we expand upon this analysis to include v3 and v4 of pangolin, which moved the default algorithm for lineage assignment from pangoLEARN in v3 to pUShER in v4, and perform a thorough analysis confirming that pUShER is not only more stable across versions but also more accurate. Our findings suggest that future lineage assignment algorithms for various pathogens should consider the value of phylogenetic placement.

From Full Day Learning to 30 Minutes a Day: A Descriptive Study of Early Learning During the First COVID-19 Pandemic School Shutdown in Ontario.

When the COVID-19 pandemic was declared in March 2020, the lives of families all over the world were disrupted. Many adults found themselves working from home while their children were unable to go to school. To better understand the potential impact of these educational disruptions, it is important to establish what learning looked like during the first school shutdown in the spring of 2020, particularly for the youngest learners who may feel the longest lasting impacts from this pandemic. Therefore, the purpose of the current descriptive study was to gather information on how kindergarten teaching and learning occurred during this time, what the biggest barriers were, and what concerns educators had regarding returning in person to the classroom setting. The sample for the current study was 2569 kindergarten educators (97.6% female; 74.2% teachers, 25.8% early childhood educators) in Ontario, Canada. Participants completed a questionnaire consisting of both quantitative scales and qualitative open-ended questions. Educators reported that parents most often contacted them regarding technological issues or how to effectively support their child. The largest barrier to learning was the ability of both parents and educators to balance work, home life, and online learning/teaching. With regards to returning to school, educators were most concerned about the lack of ability of kindergarten aged children to do tasks independently and to follow safety protocols. Our findings highlight unique challenges associated with teaching kindergarten during the pandemic, contributing to our understanding of the learning that occurred in Ontario during the first COVID-19 shutdown. Supplementary Information: The online version contains supplementary material available at 10.1007/s10643-021-01304-z.

A high-dimensional mediation model for a neuroimaging mediator: Integrating clinical, neuroimaging, and neurocognitive data to mitigate late effects in pediatric cancer.

Pediatric cancer treatment, especially for brain tumors, can have profound and complicated late effects. With the survival rates increasing because of improved detection and treatment, a more comprehensive understanding of the impact of current treatments on neurocognitive function and brain structure is critically needed. A frontline medulloblastoma clinical trial (SJMB03) has collected data, including treatment, clinical, neuroimaging, and cognitive variables. Advanced methods for modeling and integrating these data are critically needed to understand the mediation pathway from the treatment through brain structure to neurocognitive outcomes. We propose an integrative Bayesian mediation analysis approach to model jointly a treatment exposure, a high-dimensional structural neuroimaging mediator, and a neurocognitive outcome and to uncover the mediation pathway. The high-dimensional imaging-related coefficients are modeled via a binary Ising-Gaussian Markov random field prior (BI-GMRF), addressing the sparsity, spatial dependency, and smoothness and increasing the power to detect brain regions with mediation effects. Numerical simulations demonstrate the estimation accuracy, power, and robustness. For the SJMB03 study, the BI-GMRF method has identified white matter microstructure that is damaged by cancer-directed treatment and impacts late neurocognitive outcomes. The results provide guidance on improving treatment planning to minimize long-term cognitive sequela for pediatric brain tumor patients.

The molecular epidemiology of multiple zoonotic origins of SARS-CoV-2.

Understanding the circumstances that lead to pandemics is important for their prevention. We analyzed the genomic diversity of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) early in the coronavirus disease 2019 (COVID-19) pandemic. We show that SARS-CoV-2 genomic diversity before February 2020 likely comprised only two distinct viral lineages, denoted "A" and "B." Phylodynamic rooting methods, coupled with epidemic simulations, reveal that these lineages were the result of at least two separate cross-species transmission events into humans. The first zoonotic transmission likely involved lineage B viruses around 18 November 2019 (23 October to 8 December), and the separate introduction of lineage A likely occurred within weeks of this event. These findings indicate that it is unlikely that SARS-CoV-2 circulated widely in humans before November 2019 and define the narrow window between when SARS-CoV-2 first jumped into humans and when the first cases of COVID-19 were reported. As with other coronaviruses, SARS-CoV-2 emergence likely resulted from multiple zoonotic events.

Introduction and Establishment of SARS-CoV-2 Gamma Variant in New York City in Early 2021.

BACKGROUND: Monitoring the emergence and spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants is an important public health objective. We investigated how the Gamma variant was established in New York City (NYC) in early 2021 in the presence of travel restrictions that aimed to prevent viral spread from Brazil, the country where the variant was first identified. METHODS: We performed phylogeographic analysis on 15 967 Gamma sequences sampled between 10 March and 1 May 2021, to identify geographic sources of Gamma lineages introduced into NYC. We identified locally circulating Gamma transmission clusters and inferred the timing of their establishment in NYC. RESULTS: We identified 16 phylogenetically distinct Gamma clusters established in NYC (cluster sizes ranged 2-108 genomes); most of them were introduced from Florida and Illinois and only 1 directly from Brazil. By the time the first Gamma case was reported by genomic surveillance in NYC on 10 March, the majority (57%) of circulating Gamma lineages had already been established in the city for at least 2 weeks. CONCLUSIONS: Although travel from Brazil to the United States was restricted from May 2020 through the end of the study period, this restriction did not prevent Gamma from becoming established in NYC as most introductions occurred from domestic locations.

Detection of SARS-CoV-2 intra-host recombination during superinfection with Alpha and Epsilon variants in New York City.

Recombination is an evolutionary process by which many pathogens generate diversity and acquire novel functions. Although a common occurrence during coronavirus replication, detection of recombination is only feasible when genetically distinct viruses contemporaneously infect the same host. Here, we identify an instance of SARS-CoV-2 superinfection, whereby an individual was infected with two distinct viral variants: Alpha (B.1.1.7) and Epsilon (B.1.429). This superinfection was first noted when an Alpha genome sequence failed to exhibit the classic S gene target failure behavior used to track this variant. Full genome sequencing from four independent extracts reveals that Alpha variant alleles comprise around 75% of the genomes, whereas the Epsilon variant alleles comprise around 20% of the sample. Further investigation reveals the presence of numerous recombinant haplotypes spanning the genome, specifically in the spike, nucleocapsid, and ORF 8 coding regions. These findings support the potential for recombination to reshape SARS-CoV-2 genetic diversity.

Utilizing Multilingual Methods and Rapid Analysis for Global Qualitative Research During a Pandemic.

Historically, qualitative research has complemented quantitative biologic and epidemiologic studies to provide a more complete understanding of pandemics. The COVID-19 pandemic has generated unique and novel challenges for qualitative researchers, who have embraced creative solutions including virtual focus groups and rapid analyses to continue their work. We present our experience conducting a multilingual global qualitative study of healthcare resilience among teams of pediatric oncology professionals during the COVID-19 pandemic. We provide an in-depth description of our methodology and an analysis of factors we believe contributed to our study's success including our use of technology, engagement of a large multilingual team, global partnerships, and framework-based rapid analysis. We hope these techniques may be useful to qualitative researchers conducting studies during the current pandemic, as well as for all pediatric oncology studies including multiple languages or geographically disparate subjects.

Alzheimer's Disease Classification Through Imaging Genetic Data With IGnet.

The application of deep learning techniques to the detection and automated classification of Alzheimer's disease (AD) has recently gained considerable attention. The rapid progress in neuroimaging and sequencing techniques has enabled the generation of large-scale imaging genetic data for AD research. In this study, we developed a deep learning approach, IGnet, for automated AD classification using both magnetic resonance imaging (MRI) data and genetic sequencing data. The proposed approach integrates computer vision (CV) and natural language processing (NLP) techniques, with a deep three-dimensional convolutional network (3D CNN) being used to handle the three-dimensional MRI input and a Transformer encoder being used to manage the genetic sequence input. The proposed approach has been applied to the Alzheimer's Disease Neuroimaging Initiative (ADNI) data set. Using baseline MRI scans and selected single-nucleotide polymorphisms on chromosome 19, it achieved a classification accuracy of 83.78% and an area under the receiver operating characteristic curve (AUC-ROC) of 0.924 with the test set. The results demonstrate the great potential of using multi-disciplinary AI approaches to integrate imaging genetic data for the automated classification of AD.

Epidemiological characteristics of the B.1.526 SARS-CoV-2 variant.

To characterize the epidemiological properties of the B.1.526 SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) variant of interest, here we used nine epidemiological and population datasets and model-inference methods to reconstruct SARS-CoV-2 transmission dynamics in New York City, where B.1.526 emerged. We estimated that B.1.526 had a moderate increase (15 to 25%) in transmissibility, could escape immunity in 0 to 10% of previously infected individuals, and substantially increased the infection fatality risk (IFR) among adults 65 or older by >60% during November 2020 to April 2021, compared to estimates for preexisting variants. Overall, findings suggest that new variants like B.1.526 likely spread in the population weeks before detection and that partial immune escape (e.g., resistance to therapeutic antibodies) could offset prior medical advances and increase IFR. Early preparedness for and close monitoring of SARS-CoV-2 variants, their epidemiological characteristics, and disease severity are thus crucial to COVID-19 (coronavirus disease 2019) response.

Association between neighbourhood composition, kindergarten educator-reported distance learning barriers, and return to school concerns during the first wave of the COVID-19 pandemic in Ontario, Canada.

Introduction: Research to date has established that the COVID-19 pandemic has not impacted everyone equitably. Whether this unequitable impact was seen educationally with regards to educator reported barriers to distance learning, concerns and mental health is less clear. Objective: The objective of this study was to explore the association between the neighbourhood composition of the school and kindergarten educator-reported barriers and concerns regarding children's learning during the first wave of COVID-19 related school closures in Ontario, Canada. Methods: In the spring of 2020, we collected data from Ontario kindergarten educators (n = 2569; 74.2% kindergarten teachers, 25.8% early childhood educators; 97.6% female) using an online survey asking them about their experiences and challenges with online learning during the first round of school closures. We linked the educator responses to 2016 Canadian Census variables based on schools' postal codes. Bivariate correlations and Poisson regression analyses were used to determine if there was an association between neighbourhood composition and educator mental health, and the number of barriers and concerns reported by kindergarten educators. Results: There were no significant findings with educator mental health and school neighbourhood characteristics. Educators who taught at schools in neighbourhoods with lower median income reported a greater number of barriers to online learning (e.g., parents/guardians not submitting assignments/providing updates on their child's learning) and concerns regarding the return to school in the fall of 2020 (e.g., students' readjustment to routines). There were no significant associations with educator reported barriers or concerns and any of the other Census neighbourhood variables (proportion of lone parent families, average household size, proportion of population that do no speak official language, proportion of population that are recent immigrants, or proportion of population ages 0-4). Conclusions: Overall, our study suggests that the neighbourhood composition of the children's school location did not exacerbate the potential negative learning experiences of kindergarten students and educators during the COVID-19 pandemic, although we did find that educators teaching in schools in lower-SES neighbourhoods reported more barriers to online learning during this time. Taken together, our study suggests that remediation efforts should be focused on individual kindergarten children and their families as opposed to school location.

Natural Language Processing and Machine Learning Methods to Characterize Unstructured Patient-Reported Outcomes: Validation Study.

BACKGROUND: Assessing patient-reported outcomes (PROs) through interviews or conversations during clinical encounters provides insightful information about survivorship. OBJECTIVE: This study aims to test the validity of natural language processing (NLP) and machine learning (ML) algorithms in identifying different attributes of pain interference and fatigue symptoms experienced by child and adolescent survivors of cancer versus the judgment by PRO content experts as the gold standard to validate NLP/ML algorithms. METHODS: This cross-sectional study focused on child and adolescent survivors of cancer, aged 8 to 17 years, and caregivers, from whom 391 meaning units in the pain interference domain and 423 in the fatigue domain were generated for analyses. Data were collected from the After Completion of Therapy Clinic at St. Jude Children's Research Hospital. Experienced pain interference and fatigue symptoms were reported through in-depth interviews. After verbatim transcription, analyzable sentences (ie, meaning units) were semantically labeled by 2 content experts for each attribute (physical, cognitive, social, or unclassified). Two NLP/ML methods were used to extract and validate the semantic features: bidirectional encoder representations from transformers (BERT) and Word2vec plus one of the ML methods, the support vector machine or extreme gradient boosting. Receiver operating characteristic and precision-recall curves were used to evaluate the accuracy and validity of the NLP/ML methods. RESULTS: Compared with Word2vec/support vector machine and Word2vec/extreme gradient boosting, BERT demonstrated higher accuracy in both symptom domains, with 0.931 (95% CI 0.905-0.957) and 0.916 (95% CI 0.887-0.941) for problems with cognitive and social attributes on pain interference, respectively, and 0.929 (95% CI 0.903-0.953) and 0.917 (95% CI 0.891-0.943) for problems with cognitive and social attributes on fatigue, respectively. In addition, BERT yielded superior areas under the receiver operating characteristic curve for cognitive attributes on pain interference and fatigue domains (0.923, 95% CI 0.879-0.997; 0.948, 95% CI 0.922-0.979) and superior areas under the precision-recall curve for cognitive attributes on pain interference and fatigue domains (0.818, 95% CI 0.735-0.917; 0.855, 95% CI 0.791-0.930). CONCLUSIONS: The BERT method performed better than the other methods. As an alternative to using standard PRO surveys, collecting unstructured PROs via interviews or conversations during clinical encounters and applying NLP/ML methods can facilitate PRO assessment in child and adolescent cancer survivors.

Detection and characterization of the SARS-CoV-2 lineage B.1.526 in New York.

Wide-scale SARS-CoV-2 genome sequencing is critical to tracking viral evolution during the ongoing pandemic. We develop the software tool, Variant Database (VDB), for quickly examining the changing landscape of spike mutations. Using VDB, we detect an emerging lineage of SARS-CoV-2 in the New York region that shares mutations with previously reported variants. The most common sets of spike mutations in this lineage (now designated as B.1.526) are L5F, T95I, D253G, E484K or S477N, D614G, and A701V. This lineage was first sequenced in late November 2020. Phylodynamic inference confirmed the rapid growth of the B.1.526 lineage. In concert with other variants, like B.1.1.7, the rise of B.1.526 appears to have extended the duration of the second wave of COVID-19 cases in NYC in early 2021. Pseudovirus neutralization experiments demonstrated that B.1.526 spike mutations adversely affect the neutralization titer of convalescent and vaccinee plasma, supporting the public health relevance of this lineage.

Magnetic Resonance Imaging Radiomics-Based Machine Learning Prediction of Clinically Significant Prostate Cancer in Equivocal PI-RADS 3 Lesions.

BACKGROUND: While Prostate Imaging Reporting and Data System (PI-RADS) 4 and 5 lesions typically warrant prostate biopsy and PI-RADS 1 and 2 lesions may be safely observed, PI-RADS 3 lesions are equivocal. PURPOSE: To construct and cross-validate a machine learning model based on radiomics features from T2 -weighted imaging (T2 WI) of PI-RADS 3 lesions to identify clinically significant prostate cancer (csPCa), that is, pathological Grade Group ≥ 2. STUDY TYPE: Single-center retrospective study. POPULATION: A total of 240 patients were included (training cohort, n = 188, age range 43-82 years; test cohort, n = 52, age range 41-79 years). Eligibility criteria were 1) magnetic resonance imaging (MRI)-targeted biopsy between 2015 and 2020; 2) PI-RADS 3 index lesion identified on multiparametric MRI; (3) biopsy performed within 1 year of MRI. The percentages of csPCa lesions were 10.6% and 15.4% in the training and test cohorts, respectively. FIELD STRENGTH/SEQUENCE: A 3 T; T2 WI turbo-spin echo, diffusion-weighted spin-echo echo planar imaging, dynamic contrast-enhanced MRI with time-resolved T1-weighted imaging. ASSESSMENT: Multislice volumes-of-interest (VOIs) were drawn in the PI-RADS 3 index lesions on T2 WI. A total of 107 radiomics features (first-order histogram and second-order texture) were extracted from the segmented lesions. STATISTICAL TESTS: A random forest classifier using the radiomics features as input was trained and validated for prediction of csPCa. The performance of the machine learning classifier, prostate specific antigen (PSA) density, and prostate volume for csPCa prediction was evaluated using receiver operating characteristic (ROC) analysis. RESULTS: The trained random forest classifier constructed from the T2 WI radiomics features good and statistically significant area-under-the-curves (AUCs) of 0.76 (P = 0.022) for prediction of csPCa in the test set. Prostate volume and PSA density showed moderate and nonsignificant performance (AUC 0.62, P = 0.275 and 0.61, P = 0.348, respectively) for csPCa prediction in the test set. CONCLUSION: The machine learning classifier based on T2 WI radiomic features demonstrated good performance for prediction of csPCa in PI-RADS 3 lesions. EVIDENCE LEVEL: 4 TECHNICAL EFFICACY: 2.

Rapid Emergence and Epidemiologic Characteristics of the SARS-CoV-2 B.1.526 Variant - New York City, New York, January 1-April 5, 2021.

Recent studies have documented the emergence and rapid growth of B.1.526, a novel variant of interest (VOI) of SARS-CoV-2, the virus that causes COVID-19, in the New York City (NYC) area after its identification in NYC in November 2020 (1-3). Two predominant subclades within the B.1.526 lineage have been identified, one containing the E484K mutation in the receptor-binding domain (1,2), which attenuates in vitro neutralization by multiple SARS-CoV-2 antibodies and is present in variants of concern (VOCs) first identified in South Africa (B.1.351) (4) and Brazil (P.1).* The NYC Department of Health and Mental Hygiene (DOHMH) analyzed laboratory and epidemiologic data to characterize cases of B.1.526 infection, including illness severity, transmission to close contacts, rates of possible reinfection, and laboratory-diagnosed breakthrough infections among vaccinated persons. Preliminary data suggest that the B.1.526 variant does not lead to more severe disease and is not associated with increased risk for infection after vaccination (breakthrough infection) or reinfection. Because relatively few specimens were sequenced over the study period, the statistical power might have been insufficient to detect modest differences in rates of uncommon outcomes such as breakthrough infection or reinfection. Collection of timely viral genomic data for a larger proportion of citywide cases and rapid integration with population-based surveillance data would enable improved understanding of the impact of emerging SARS-CoV-2 variants and specific mutations to help guide public health intervention efforts.

Detection and characterization of the SARS-CoV-2 lineage B.1.526 in New York.

Wide-scale SARS-CoV-2 genome sequencing is critical to tracking viral evolution during the ongoing pandemic. Variants first detected in the United Kingdom, South Africa, and Brazil have spread to multiple countries. We developed the software tool, Variant Database (VDB), for quickly examining the changing landscape of spike mutations. Using VDB, we detected an emerging lineage of SARS-CoV-2 in the New York region that shares mutations with previously reported variants. The most common sets of spike mutations in this lineage (now designated as B.1.526) are L5F, T95I, D253G, E484K or S477N, D614G, and A701V. This lineage was first sequenced in late November 2020 when it represented <1% of sequenced coronavirus genomes that were collected in New York City (NYC). By February 2021, genomes from this lineage accounted for ~32% of 3288 sequenced genomes from NYC specimens. Phylodynamic inference confirmed the rapid growth of the B.1.526 lineage in NYC, notably the sub-clade defined by the spike mutation E484K, which has outpaced the growth of other variants in NYC. Pseudovirus neutralization experiments demonstrated that B.1.526 spike mutations adversely affect the neutralization titer of convalescent and vaccinee plasma, indicating the public health importance of this lineage.

Revealing Opinions for COVID-19 Questions Using a Context Retriever, Opinion Aggregator, and Question-Answering Model: Model Development Study.

BACKGROUND: COVID-19 has challenged global public health because it is highly contagious and can be lethal. Numerous ongoing and recently published studies about the disease have emerged. However, the research regarding COVID-19 is largely ongoing and inconclusive. OBJECTIVE: A potential way to accelerate COVID-19 research is to use existing information gleaned from research into other viruses that belong to the coronavirus family. Our objective is to develop a natural language processing method for answering factoid questions related to COVID-19 using published articles as knowledge sources. METHODS: Given a question, first, a BM25-based context retriever model is implemented to select the most relevant passages from previously published articles. Second, for each selected context passage, an answer is obtained using a pretrained bidirectional encoder representations from transformers (BERT) question-answering model. Third, an opinion aggregator, which is a combination of a biterm topic model and k-means clustering, is applied to the task of aggregating all answers into several opinions. RESULTS: We applied the proposed pipeline to extract answers, opinions, and the most frequent words related to six questions from the COVID-19 Open Research Dataset Challenge. By showing the longitudinal distributions of the opinions, we uncovered the trends of opinions and popular words in the articles published in the five time periods assessed: before 1990, 1990-1999, 2000-2009, 2010-2018, and since 2019. The changes in opinions and popular words agree with several distinct characteristics and challenges of COVID-19, including a higher risk for senior people and people with pre-existing medical conditions; high contagion and rapid transmission; and a more urgent need for screening and testing. The opinions and popular words also provide additional insights for the COVID-19-related questions. CONCLUSIONS: Compared with other methods of literature retrieval and answer generation, opinion aggregation using our method leads to more interpretable, robust, and comprehensive question-specific literature reviews. The results demonstrate the usefulness of the proposed method in answering COVID-19-related questions with main opinions and capturing the trends of research about COVID-19 and other relevant strains of coronavirus in recent years.