The Paxlovid Rebound Study: A Prospective Cohort Study to Evaluate Viral and Symptom Rebound Differences Between Paxlovid and Untreated COVID-19 Participants

IntroductionThe uptake of Paxlovid in individuals infected with COVID-19 has been significantly limited by concerns around the Paxlovid rebound phenomenon despite the scarcity of evidence around its epidemiology. The purpose of this study was to prospectively compare the epidemiology of Paxlovid rebound in treated and untreated participants with acute COVID-19 infection MethodsWe designed a decentralized, digital, prospective observational study in which participants who tested positive for COVID-19 using eMed Test-to-Treat telehealth kits and were clinically eligible for Paxlovid were recruited to be evaluated for viral and symptom clearance, as well as rebound. Participants were assigned to a Paxlovid or control group based on their decision to take Paxlovid. Following initial diagnosis based on a telehealth proctored test both groups were provided 12 telehealth proctored rapid antigen home tests and asked to test on a regular frequent schedule for 16 days and answer symptom surveys. Viral rebound based on test results and COVID-19 symptom rebound based on patient reported symptoms were evaluated. ResultsViral rebound incidence was 14.2% in the Paxlovid group (n=127) and 9.3% in the control group (n=43). COVID-19 symptom rebound incidence was higher in the Paxlovid group (18.9%) compared to the control group (7.0%). There were no notable differences in viral rebound by age, gender, pre-existing conditions, or major symptom groups during the acute phase or at the 1-month interval. ConclusionThis preliminary report of our prospective study suggests that rebound after clearance of test positivity or symptom resolution is higher than previously reported. However, we observed a similar rate of rebound in both in the Paxlovid and control groups. Large studies with diverse participants and extended follow-up are needed to better understand the rebound phenomena.

High Rates of Rapid Antigen Test Positivity After 5 Days of Isolation for COVID-19

BackgroundThe emergence of the highly transmissible COVID-19 variant, omicron, has resulted in high numbers of breakthrough infections, including among healthcare workers (HCW). Recent CDC recommendations now allow healthcare workers to return to work after day 5 if symptoms have improved, without a requirement for a negative rapid antigen test (RAT). MethodsFully vaccinated and non-immunocompromised HCW at a large, urban, academic medical center who tested positive for COVID-19 starting in late December, 2021 (when omicron was the predominant circulating strain) were allowed to return to work early if all symptoms had resolved excepting mild, intermittent cough and/or lingering loss of taste/smell, provided a rapid antigen test was negative upon return. Those with negative tests were allowed to return to work with the stipulations that they wear an N95 at all times and take breaks and eat meals apart from others. Those with positive tests on first attempt could return 24-48 hours later to test again for as many days as needed to achieve a negative result or until they completed 10 days of restriction from work. ResultsBetween January 2, 2022 and January 12, 2022 there were 309 total RAT done on 260 separate HCW on day 5-10 of illness. Overall, 43% (134 of 309) of all RAT were positive between days 5-10. The greatest percent positive RAT was noted among HCW returning for their first test on day 6 (58%). The rate of positivity was greatest (58%) among HCW returning for their first test on day 6. HCW returning on day 8 and 9 were less likely to have a positive test (26%, 19/74). In RAT positive HCW returning for their first test on days 5 or 6 (and for which line intensity was recorded) 49% (25/51) were recorded as having the darkest intensity on their RAT. HCW who test positive on their first test most often remained positive on their second test, with 56% of second tests, aggregated across all days 6-10, remaining positive. Over all first tests performed on days 5-10, boosted HCW were nearly twice as likely to test RAT positive: 53% (75 out of 141) of boosted HCW tested positive. DiscussionMore than 40% of vaccinated HCW who felt well enough to work still had positive RAT tests when presenting for a first test between days 5 and 10. Boosted individuals were nearly 3x as likely to result positive on day 5, their first day eligible for return, and [~]2x as likely to result positive on first RAT overall. New guidance provides clearance to exit isolation after 5 days from symptom onset, without the need for a negative rapid antigen test to exit, as long as symptoms are beginning to resolve. Per CDC, the guidance was driven by prior studies, mostly collected before Omicron and before most people were vaccinated or infected, that reported on symptom onset beginning one or more days after peak virus loads. In such an instance, where isolation based on symptom onset often did not begin until peak virus load was already attained, then release from isolation at 5 days would be appropriate. However, reports showing much earlier onset of symptoms, coupled with our own results here demonstrate that the relationship between symptom onset and peak virus load has changed, and 5 days from symptom onset may no longer be an appropriate window to end isolation without a negative rapid antigen test to support safe exit. ConclusionThese results indicate that a substantial proportion of individuals with COVID-19 are likely still contagious after day 5 of illness regardless of symptom status. Early liberation from isolation should be undertaken only with the understanding that inclusion of individuals on day 6-10 of illness in community or work settings may increase the risk of COVID-19 spread to others which, in turn, may undermine the intended benefits to staffing by resulting in more sick workers.

Vaccine nationalism and the dynamics and control of SARS-CoV-2

Vaccines provide powerful tools to mitigate the enormous public health and economic costs that the ongoing SARS-CoV-2 pandemic continues to exert globally, yet vaccine distribution remains unequal between countries. To examine the potential epidemiological and evolutionary impacts of vaccine nationalism, we extend previous models to include simple scenarios of stockpiling. In general, we find that stockpiling vaccines by countries with high availability leads to large increases in infections in countries with low vaccine availability, the magnitude of which depends on the strength and duration of natural and vaccinal immunity. Additionally, a number of subtleties arise when the populations and transmission rates in each country differ depending on evolutionary assumptions and vaccine availability. Furthermore, the movement of infected individuals between countries combined with the possibility of increases in viral transmissibility may greatly magnify local and combined infection numbers, suggesting that countries with high vaccine availability must invest in surveillance strategies to prevent case importation. Dose-sharing is likely a high-return strategy because equitable allocation brings non-linear benefits and also alleviates costs of surveillance (e.g. border testing, genomic surveillance) in settings where doses are sufficient to maintain cases at low numbers. Across a range of immunological scenarios, we find that vaccine sharing is also a powerful tool to decrease the potential for antigenic evolution, especially if infections after the waning of natural immunity contribute most to evolutionary potential. Overall, our results stress the importance of equitable global vaccine distribution.

Recalibrating SARS-CoV-2 antigen rapid lateral flow test relative sensitivity from validation studies to absolute sensitivity for detecting individuals with live virus

Testing for SARS-CoV-2 internationally has focused on COVID-19 diagnosis among symptomatic individuals using reverse transcriptase polymerase chain reaction (PCR) tests. Recently, however, SARS-CoV-2 antigen rapid lateral flow tests (LFT) have been rolled out in several countries for testing asymptomatic individuals in public health programmes. Validation studies for LFT have been largely cross-sectional, reporting sensitivity, specificity and predictive values of LFT relative to PCR. However, because PCR detects genetic material left behind for a long period when the individual is no longer infectious, these statistics can under-represent sensitivity of LFT for detecting infectious individuals, especially when sampling asymptomatic populations. LFTs (intended to detect individuals with live virus) validated against PCR (intended to diagnose infection) are not reporting against a gold standard of equivalent measurements. Instead, these validation studies have reported relative performance statistics that need recalibrating to the purpose for which LFT is being used. We present an approach to this recalibration. We derive a formula for recalibrating relative performance statistics from LFT vs PCR validation studies to give likely absolute sensitivity of LFT for detecting individuals with live virus. We show the differences between widely reported apparent sensitivities of LFT and its absolute sensitivity as a test of presence of live virus. After accounting for within-individual viral kinetics and epidemic dynamics within asymptomatic populations we show that a highly performant test of live virus should show a LFT-to-PCR relative sensitivity of less than 50% in conventional validation studies, which after re-calibration would be an absolute sensitivity of more than 80%. Further studies are needed to ascertain the absolute sensitivity of LFT as a test of infectiousness in COVID-19 responses. These studies should include sampling for viral cultures and longitudinal series of LFT and PCR, ideally in cohorts sampled from both contacts of cases and the general population.

Economic Benefits of COVID-19 Screening Tests with a Vaccine Rollout

This note estimates the costs and benefits of a nationwide COVID-19 screening testing program in the presence of vaccine distribution. Even for an optimistic vaccine rollout scenario, a well-designed federally-funded screening testing program, coupled with self-isolation of those who test positive, pays for itself in terms of increased GDP and is projected to save 20,000 or more lives. The sooner the testing program is put in place, the greater are its net economic benefits. This note updates the December 9, 2020 version to include updated deaths data, later dates for rolling out the screening testing program, and the spread of more contagious variants such as the B.1.1. 7 variant.

Real World Performance of SARS-CoV-2 Antigen Rapid Diagnostic Tests in Various Clinical Settings

BackgroundSeveral uses of Antigen rapid diagnostic tests (Ag-RDT) have been suggested. Analytical studies reported high specificity yet with lower sensitivity for detecting SARS-CoV-2 compared to qRT-PCR. Here, we present the use of these tests as a decision support tool in several settings. MethodsSamples were collected for both Ag-RDT and qRT-PCR in three different clinical settings; 1. Symptomatic patients presenting at the Emergency Departments 2. Asymptomatic patients screened upon hospitalization and 3. Health-care workers (HCW) following SARS-CoV-2 exposure. Positive percent agreement (PPA), negative percent agreement (NPA), positive predictive value (PPV) and negative predictive value (NPV) were calculated. To estimate the association between Ct value, Ag-RDT and the number of days since SARS-CoV-2 exposure or symptomatic COVID-19, a mixed model was applied. ResultsA total of 5172 samples were obtained from 4595 individuals, with Ag-RDT and qRT-PCR results. Of these, 485 samples were positive by qRT-PCR. The PPA of Ag-RDT was greater for lower Ct values, reaching 93% in cases where Ct value was lower than 25 and 85% where Ct value was lower than 30. PPA was similar between symptomatic and asymptomatic individuals. The NPV and PPV were 96.8% and 99.1%, respectively. We observed a significant correlation between Ct value and time from infection onset (p<0.001). Lower Ct values were significantly associated with a positive Ag-RDT (p=0.01). ConclusionsAg-RDT can be used as a decision support tool in various clinical settings and play a major role in early detection of SARS-CoV-2 infected individuals, highly specific and with high sensitivity to the infectious stage of disease, whether symptomatic or asymptomatic.

Mathematical modeling to inform vaccination strategies and testing approaches for COVID-19 in nursing homes

BackgroundNursing home residents and staff were included in the first phase of COVID-19 vaccination in the United States. Because the primary trial endpoint was vaccine efficacy (VE) against symptomatic disease, there are limited data on the extent to which vaccines protect against SARS-CoV-2 infection and the ability to infect others (infectiousness). Assumptions about VE against infection and infectiousness have implications for possible changes to infection prevention guidance for vaccinated populations, including testing strategies. MethodsWe use a stochastic agent-based SEIR model of a nursing home to simulate SARS-CoV-2 transmission. We model three scenarios, varying VE against infection, infectiousness, and symptoms, to understand the expected impact of vaccination in nursing homes, increasing staff vaccination coverage, and different screening testing strategies under each scenario. ResultsIncreasing vaccination coverage in staff decreases total symptomatic cases in each scenario. When there is low VE against infection and infectiousness, increasing staff coverage reduces symptomatic cases among residents. If vaccination only protects against symptoms, but asymptomatic cases remain infectious, increased staff coverage increases symptomatic cases among residents through exposure to asymptomatic but infected staff. High frequency testing is needed to reduce total symptomatic cases if the vaccine has low efficacy against infection and infectiousness, or only protects against symptoms. ConclusionsEncouraging staff vaccination is not only important for protecting staff, but might also reduce symptomatic cases in residents if a vaccine confers at least some protection against infection or infectiousness. SummaryThe extent of efficacy of SARS-CoV-2 vaccines against infection, infectiousness, or disease, impacts strategies for vaccination and testing in nursing homes. If vaccines confer some protection against infection or infectiousness, encouraging vaccination in staff may reduce symptomatic cases in residents.

Epidemiological and evolutionary considerations of SARS-CoV-2 vaccine dosing regimes

As the threat of Covid-19 continues and in the face of vaccine dose shortages and logistical challenges, various deployment strategies are being proposed to increase population immunity levels. How timing of delivery of the second dose affects infection burden but also prospects for the evolution of viral immune escape are critical questions. Both hinge on the strength and duration (i.e. robustness) of the immune response elicited by a single dose, compared to natural and two-dose immunity. Building on an existing immuno-epidemiological model, we find that in the short-term, focusing on one dose generally decreases infections, but longer-term outcomes depend on this relative immune robustness. We then explore three scenarios of selection, evaluating how different second dose delays might drive immune escape via a build-up of partially immune individuals. Under certain scenarios, we find that a one-dose policy may increase the potential for antigenic evolution. We highlight the critical need to test viral loads and quantify immune responses after one vaccine dose, and to ramp up vaccination efforts throughout the world.

Development of at-home sample collection logistics for large scale SARS-CoV-2 seroprevalence studies

In the midst of a pandemic, serologic studies are a valuable tool to understand the course of the outbreak and guide public health and general pandemic management. However, given significant safety constraints including social distancing and stay-at-home orders, sample collection becomes more difficult given traditional phlebotomy protocols. For such studies, a representative sample of the underlying population is paramount to elicit meaningful insights that capture the spread of the infection, particularly when different sub-populations face varying disease burden. We aimed to address these challenges by conducting a fully remote study to investigate the seroprevalence of SARS-CoV-2 in the state of Massachusetts. Leveraging electronic study engagement and at-home self-collection of finger-prick samples, we enrolled 2,066 participants representative of the ethnic and racial composition of Massachusetts. SARS-CoV-2 total IgG seropositivity was 3.15%, and follow-up measurements at days 7, 15, 45, and 90 indicate a generally durable antibody response. A higher risk of infection was observed for healthcare workers and their cohabitants and those with comorbidities, as well as lower-income, less educated, Hispanic, and those in the age groups of 18-29 and 50-59-years-old. High engagement and positive feedback from the participants and quality of self-collected specimens point to the usefulness of this design for future population-level serological studies that more effectively and safely reach a broad representative cohort, thus yielding more comprehensive insights into the burden of infection and disease in populations. Key pointsO_ST_ABSQuestionC_ST_ABSWe aimed to implement a fully remote seroprevalence study for SARS-CoV-2, leveraging electronic methods and at-home self-collection of specimens to engage a representative study population. FindingsThe population enrolled reflected the ethnic and racial composition of Massachusetts, revealing a SARS-CoV-2 seroprevalence of 3.15% and higher risk of previous infection associated with healthcare workers/their cohabitants, those with comorbidities, lower-income, less educated, Hispanic, and those in age groups 18-29 and 50-59 years old. MeaningHigh engagement and positive feedback from participants as well as quality of self-collected specimens point to the usefulness of this design for future population-level serological studies.

Economic Benefits of COVID-19 Screening Tests

We assess the economic value of screening testing programs as a policy response to the ongoing COVID-19 pandemic. We find that the fiscal, macroeconomic, and health benefits of rapid SARS-CoV-2 screening testing programs far exceed their costs, with the ratio of economic benefits to costs typically in the range of 2-15 (depending on program details), not counting the monetized value of lives saved. Unless the screening test is highly specific, however, the signal value of the screening test alone is low, leading to concerns about adherence. Confirmatory testing increases the net economic benefits of screening tests by reducing the number of healthy workers in quarantine and by increasing adherence to quarantine measures. The analysis is undertaken using a behavioral SIR model for the United States with 5 age groups, 66 economic sectors, screening and diagnostic testing, and partial adherence to instructions to quarantine or to isolate.

Estimating epidemiologic dynamics from single cross-sectional viral load distributions

Virologic testing for SARS-CoV-2 has been central to the COVID-19 pandemic response, but interpreting changes in incidence and fraction of positive tests towards understanding the epidemic trajectory is confounded by changes in testing practices. Here, we show that the distribution of viral loads, in the form of Cycle thresholds (Ct), from positive surveillance samples at a single point in time can provide accurate estimation of an epidemics trajectory, subverting the need for repeated case count measurements which are frequently obscured by changes in testing capacity. We identify a relationship between the population-level cross-sectional distribution of Ct values and the growth rate of the epidemic, demonstrating how the skewness and median of detectable Ct values change purely as a mathematical epidemiologic rule without any change in individual level viral load kinetics or testing. Although at the individual level measurement variation can complicate interpretation of Ct values for clinical use, we show that population-level properties reflect underlying epidemic dynamics. In support of these theoretical findings, we observe a strong relationship between the time-varying effective reproductive number, R(t), and the distribution of Cts among positive surveillance specimens, including median and skewness, measured in Massachusetts over time. We use the observed relationships to derive a novel method that allows accurate inference of epidemic growth rate using the distribution of Ct values observed at a single cross-section in time, which, unlike estimates based on case counts, is less susceptible to biases from delays in test results and from changing testing practices. Our findings suggest that instead of discarding individual Ct values from positive specimens, incorporation of viral loads into public health data streams offers a new approach for real-time resource allocation and assessment of outbreak mitigation strategies, even where repeat incidence data is not available. Ct values or similar viral load data should be regularly reported to public health officials by testing centers and incorporated into monitoring programs.

Comparison of viral levels in individuals with or without symptoms at time of COVID-19 testing among 32,480 residents and staff of nursing homes and assisted living facilities in Massachusetts.

BackgroundTransmission of COVID-19 from people without symptoms poses considerable challenges to public health containment measures. The distribution of viral loads in individuals with and without symptoms remains uncertain. Comprehensive cross-sectional screening of all individuals in a given setting provides an unbiased way to assess viral loads independent of symptoms, which informs transmission risks. COVID-19 cases initially peaked in Massachusetts in mid-April 2020 before declining through June, and congregate living facilities were particularly affected during this early surge. We performed a retrospective analysis of data from a large public health-directed outbreak response initiative that involved comprehensive screening within nursing homes and assisted living facilities in Massachusetts to compare nasopharyngeal (NP) viral loads (as measured by RT-PCR cycle threshold (Ct) levels) in residents and staff to inform our ability to detect SARS-CoV-2 in individuals with or without symptoms in the population. MethodsBetween April 9 and June 9, 2020, we tested NP swabs from 32,480 unique individuals comprising staff and residents of the majority of nursing homes and assisted living facilities in Massachusetts. Under the direction of the MA Department of Public Health (MDPH), symptomatology at the time of sampling and demographic information was provided by each facility for each individual to facilitate reporting to health officials. NP swabs were collected, RNA extracted, and SARS-CoV-2 testing performed using quantitative reverse-transcriptase polymerase chain reaction (qRT-PCR). ResultsThe nursing home and assisted living facilities resident cohort (N =16,966) was 65% female with a mean age of 82 years (SD 13 yrs). The staff cohort (N = 15,514) was 76% female with a median age of 45 (SD 15 yrs). A total 2654 residents (15.5%) and 624 staff (4.1%) tested positive for SARS-CoV-2. 12.7% of residents and 3.7% of staff without symptoms tested positive for SARS-CoV-2, compared to 53.1% of residents and 18.2% of staff with symptoms. Of the individuals who tested positive, 70.8% of residents and 92.4% of staff lacked symptoms at the time of testing. In aggregate, the distributions of Cts for viral probes used in the qRT-PCR assay were very similar, with a statistically but not meaningfully different mean ({Delta}Ct 0.71 cycles, p = 0.006) and a similar range (12-38 cycles), between populations with and without symptoms over the entire time period, across all sub-categories examined (age, race, ethnicity, sex, resident/staff). Importantly, the Ct mean values and range were indistinguishable between the populations by symptom class during the peak of the outbreak in Massachusetts, with a Ct gap appearing only later in the survey period, reaching >3 cycles (p [≤] 0.001) for facilities sampled during the last two weeks of the study. ConclusionsIn a large cohort of individuals screened for SARS-CoV-2 by qRT-PCR, we found strikingly similar distributions of viral load in patients with or without symptoms at the time of testing during the local peak of the epidemic; as the epidemic waned, individuals without symptoms at the time of testing had lower viral loads. The size of the study population, including both staff and residents spanning a wide range of ages, provides a comprehensive cross-sectional point prevalence measurement of viral burden in a study spanning 2 months. Because the distributions of viral loads in infected individuals irrespective of symptomatology are very similar, existing testing modalities that have been validated for detection of SARS-CoV-2 RNA in symptomatic patients should perform similarly in individuals without symptoms at the time of testing.

Quantifying the impact of US state non-pharmaceutical interventions on COVID-19 transmission

COVID-19 is an ongoing public health emergency. Without a vaccine or effective antivirals, non-pharmaceutical interventions form the foundation of current response efforts. Quantifying the efficacy of these interventions is crucial. Using mortality data and a classification guide of state level responses, we relate the intensity of interventions to statistical estimates of transmission, finding that more stringent control measures are associated with larger reductions in disease proliferation. Additionally, we observe that transmission increases with population density, but not population size. These results may help inform future response efforts.

Surveillance testing of SARS-CoV-2

The COVID-19 pandemic has created a public health crisis. Because SARS-CoV-2 can spread from individuals with pre-symptomatic, symptomatic, and asymptomatic infections, the re-opening of societies and the control of virus spread will be facilitated by robust surveillance, for which virus testing will often be central. After infection, individuals undergo a period of incubation during which viral titers are usually too low to detect, followed by an exponential viral growth, leading to a peak viral load and infectiousness, and ending with declining viral levels and clearance. Given the pattern of viral load kinetics, we model surveillance effectiveness considering test sensitivities, frequency, and sample-to-answer reporting time. These results demonstrate that effective surveillance depends largely on frequency of testing and the speed of reporting, and is only marginally improved by high test sensitivity. We therefore conclude that surveillance should prioritize accessibility, frequency, and sample-to-answer time; analytical limits of detection should be secondary.

Efficient prevalence estimation and infected sample identification with group testing for SARS-CoV-2

Extensive virological testing is central to SARS-CoV-2 containment, but many settings face severe limitations on testing. Group testing offers a way to increase throughput by testing pools of combined samples; however, most proposed designs have not yet addressed key concerns over sensitivity loss and implementation feasibility. Here, we combine a mathematical model of epidemic spread and empirically derived viral kinetics for SARS-CoV-2 infections to identify pooling designs that are robust to changes in prevalence, and to ratify losses in sensitivity against the time course of individual infections. Using this framework, we show that prevalence can be accurately estimated across four orders of magnitude using only a few dozen pooled tests without the need for individual identification. We then exhaustively evaluate the ability of different pooling designs to maximize the number of detected infections under various resource constraints, finding that simple pooling designs can identify up to 20 times as many positives compared to individual testing with a given budget. We illustrate how pooling affects sensitivity and overall detection capacity during an epidemic and on each day post infection, finding that sensitivity loss is mainly attributed to individuals sampled at the end of infection when detection for public health containment has minimal benefit. Crucially, we confirm that our theoretical results can be accurately translated into practice using pooled human nasopharyngeal specimens. Our results show that accounting for variation in sampled viral loads provides a nuanced picture of how pooling affects sensitivity to detect epidemiologically relevant infections. Using simple, practical group testing designs can vastly increase surveillance capabilities in resource-limited settings.

Estimating the number of undetected COVID-19 cases exported internationally from all of China

Early in the COVID-19 pandemic, when cases were predominantly reported in the city of Wuhan, China, local outbreaks in Europe, North America, and Asia were largely predicted from imported cases on flights from Wuhan, potentially missing imports from other key source cities. Here, we account for importations from Wuhan and from other cities in China, combining COVID-19 prevalence estimates in 18 Chinese cities with estimates of flight passenger volume to predict for each day between early December 2019 to late February 2020 the number of cases exported from China. We predict that the main source of global case importation in early January was Wuhan, but due to the Wuhan lockdown and the rapid spread of the virus, the main source of case importation from mid February became Chinese cities outside of Wuhan. For destinations in Africa in particular, non-Wuhan cities were an important source of case imports (1 case from those cities for each case from Wuhan, range of model scenarios: 0.1-9.8). Our model predicts that 18.4 (8.5 - 100) COVID-19 cases were imported to 26 destination countries in Africa, with most of them (90%) predicted to have arrived between 7th January ({+/-}10 days) and 5th February ({+/-}3 days), and all of them predicted prior to the first case detections. We finally observed marked heterogeneities in expected imported cases across those locations. Our estimates shed light on shifting sources and local risks of case importation which can help focus surveillance efforts and guide public health policy during the final stages of the pandemic. We further provide a time window for the seeding of local epidemics in African locations, a key parameter for estimating expected outbreak size and burden on local health care systems and societies, that has yet to be defined in these locations.