Maxime Beaulieu 1, Nathanaël Hozé 1, Vincent Vieillefond 2, Thimothée Goetschy 3, Gina Cosentino 3, François Blanquart 4, Florence Débarre 5, Jérémie Guedj 1
1 IAME, Inserm (Paris, France), 2 Biogroup Paris Ouest (Levallois-Perret, France), 3 SCM Biogroup (Levallois-Perret, France), 4 CIRB, Collège de France (Paris, France), 5 IEES, CNRS (Paris, France)
Introduction
During the COVID-19 pandemic, the emergence of SARS-CoV-2 variants of concern and the rollout of mass vaccination campaigns have profoundly modified viral transmission (1–4), and disease severity(5–8). Despite numerous studies, the respective effects of vaccination and Omicron variants on viral load dynamics remain debated, partly due to heterogeneous study designs, specific population included, limited sample sizes, and sparse longitudinal follow-up (9–12). Leveraging routinely collected PCR data from community laboratories to reconstruct within-host viral dynamics could help to overcome these limitations. Nonetheless, substantial methodological challenges remain, including heterogeneous testing practices, missing metadata, and highly skewed sampling times relative to infection, which collectively raise questions about the reliability of such data for this purpose.
Objectives
This study aimed to assess whether sparse and heterogeneous PCR data collected in the general community can be used to reliably infer within-host viral kinetics. Specifically, we sought (i) to evaluate, through simulation, the feasibility and limitations of reconstructing viral dynamics from community testing data, and (ii) to quantify the independent effects of variant of infection, vaccination status, and age on SARS-CoV-2 viral load trajectories during Delta and Omicron circulation in France.
Methods
A simulation study was conducted under three realistic sampling scenarios to compare two modelling strategies and inference frameworks. The approaches consisted in analysing all individuals, and in modelling only individuals with at least one positive PCR test. Bayesian inference using Stan and frequentist inference using the SAEM algorithm in Monolix were evaluated in terms of bias, uncertainty, and computational cost.
Then, we analysed PCR test results collected in Biogroup community laboratories between July 2021 and March 2022 in France. The dataset included over 6.6 million PCR tests, from which 322,218 symptomatic infections with at least one positive test and complete metadata were retained. Viral kinetics were modelled using a piecewise linear mixed-effects model describing viral proliferation, peak viral load, and clearance. The time of infection was reconstructed using reported time since symptom onset. Variant status was classified as pre-Omicron or Omicron, vaccination status as vaccinated (1-2 doses) or unvaccinated, and age as <65 or ≥65 years.
Results
Simulation results showed that key viral kinetic parameters, particularly peak viral load and clearance duration, could be estimated with low bias even under sparse and skewed sampling designs typical of community testing. However, uncertainty in early infection parameters was underestimated when data prior to symptom onset were scarce, and although including all individuals in a Bayesian inference framework reduced both bias and estimation uncertainty, this approach was computationally prohibitive. Given the size of the real dataset, parameter estimation for community data relied on frequentist inference.
Applied to real-world data, the model revealed that age was the main factor that modulated the viral kinetics of SARS-CoV-2. Individuals aged ≥65 years exhibited longer clearance periods, with infections lasting 2–6 days longer than in <65 years individuals. Vaccination consistently shortened viral clearance by 2–4 days across variants and age groups, without altering peak viral load. In contrast, Omicron infections were associated with substantially lower peak viral loads, by approximately 2–3 Ct, and faster clearance compared to pre-Omicron infections, with a maximum reduction of 2 days. The effect of vaccination on clearance was most pronounced in older individuals, particularly during Omicron circulation.
Conclusions
This study demonstrates that large-scale PCR data collected in community laboratories can be used to reconstruct within-host viral dynamics and identify the main determinants of viral load trajectories. Our findings indicate that host-related factors, particularly age and immune status, primarily influence infection duration, while viral variants mainly affect peak viral load. Vaccination accelerates viral clearance without substantially altering peak viral replication. Methodologically, this work highlights both the potential and the limitations of using routinely collected diagnostic data, emphasizing the need for appropriate modelling frameworks and pragmatic modelling strategies. These results support the integration of community laboratory data into epidemic surveillance and preparedness efforts, especially in the context of emerging variants and widespread multiplex testing.
References:
1. Thompson et al. Prevention and Attenuation of Covid-19 with the BNT162b2 and mRNA-1273 Vaccines. New England Journal of Medicine. 2021.
2. Eyre et al. Effect of Covid-19 Vaccination on Transmission of Alpha and Delta Variants. New England Journal of Medicine. 2022.
3. Campbell et al. Increased transmissibility and global spread of SARS-CoV-2 variants of concern as at June 2021. Eurosurveillance. 2021.
4. Lyngse et al. Household transmission of the SARS-CoV-2 Omicron variant in Denmark. Nat Commun. 2022.
5. Nasreen et al. Effectiveness of Coronavirus Disease 2019 Vaccines Against Hospitalization and Death in Canada: A Multiprovincial, Test-Negative Design Study. Clin Infect Dis. 2023.
6. Chemaitelly et al. mRNA-1273 COVID-19 vaccine effectiveness against the B.1.1.7 and B.1.351 variants and severe COVID-19 disease in Qatar. Nat Med. 2021.
7. Twohig et al. Hospital admission and emergency care attendance risk for SARS-CoV-2 delta (B.1.617.2) compared with alpha (B.1.1.7) variants of concern: a cohort study. The Lancet Infectious Diseases. 2022.
8. Menni et al. Symptom prevalence, duration, and risk of hospital admission in individuals infected with SARS-CoV-2 during periods of omicron and delta variant dominance: a prospective observational study from the ZOE COVID Study. The Lancet. 2022.
9. Puhach et al. Infectious viral load in unvaccinated and vaccinated individuals infected with ancestral, Delta or Omicron SARS-CoV-2. Nat Med. 2022.
10. Singanayagam et al. Community transmission and viral load kinetics of the SARS-CoV-2 delta (B.1.617.2) variant in vaccinated and unvaccinated individuals in the UK: a prospective, longitudinal, cohort study. The Lancet Infectious Diseases. 2022.
11. Woodbridge et al. Viral load dynamics of SARS-CoV-2 Delta and Omicron variants following multiple vaccine doses and previous infection. Nature Communications. 2022.
12. Hay et al. Quantifying the impact of immune history and variant on SARS-CoV-2 viral kinetics and infection rebound: A retrospective cohort study. eLife. 2022.
Reference: PAGE 34 (2026) Abstr 11874 [www.page-meeting.org/?abstract=11874]
Poster: Drug/Disease Modelling - Infection