In a recent study published in PLOS One, researchers developed a causal model to analyze the viral load distribution of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) based on the age of patients.
Study: causal, Bayesian and non-parametric modeling of the viral load distribution of SARS-CoV-2 versus patient age. Image Credit: Nhemz/Shutterstock
Background
The true extent to which adolescents and children become infected with SARS-CoV-2 is not well understood. Their role in community transmission of SARS-CoV-2 depends on symptoms, viral load, behavior, susceptibility, and existing mitigation strategies. Viral load is the concentration of virus in the upper respiratory tract and is usually expressed as viral RNA copies per milliliter of sample.
Viral load is inferred from the cycle threshold (Ct) value of a sample in a reverse transcription polymerase chain reaction (RT-PCR) test. Several studies have investigated whether children and adults have differential viral loads during coronavirus disease 2019 (COVID-19). Viral load is a critical variable that could help predict the severity and mortality of COVID-19.
About the study
The present study examined viral load as an indicator of SARS-CoV-2 infectivity and re-analyzed age-stratified data previously reported by another research group through a nonparametric, Bayesian, and causal model. Since the COVID-19 outbreak, efforts have been made to identify whether people in specific age groups are more susceptible to infection than others.
To explain SARS-CoV-2 viral load and age data, a model must integrate fundamental insights into the causal relationship between viral load and age. A nonparametric causal model was developed and applied to the data. RT-PCR viral load data from the Charite Institute for Virology and Work, Germany. These data were obtained with two PCR instruments: Roche Cobas 6800/8800 and Roche LightCycler 480 II.
The cobas data set, denoted by dC, comprised approximately 2,200 data points, while the LC480 (dL) data set comprised approximately 1,350 data points. The analyzed data set consisted of indexed pairs of age (x) and log viral load (y) for each of the ‘N’ infected patients. Two lower data filtering thresholds were defined for viral loads: ymin (3.8) and y’min (5.4) and any data points with a viral load less than ymin and y’min were discarded. The two data sets were first analyzed excluding data below ymin and then below y’min.
recommendations
The authors generally observed a downward trend in the viral load probability distribution for all data sets and age groups. The dC dataset showed significant differences in viral load for different age groups. The distribution had a different maximum for viral loads equal to or greater than 8 (in log units) for patients older than 60 years. The researchers noted that this was not a result of overfitting the sample noise, but was triggered by the data. This meant that the differences in the actual data were not just an effect of gunshot noise.
For the causal structure x → y (age influences viral load), there was evidence in the dC dataset for age-dependent viral load distribution. The log-evidence relationship for the dC data set explicitly favored the dependent model, but decreased when y’min was considered the lower threshold. The log-of-evidence ratio was low for the reverse causal relationship y → x (viral load influences age), indicating that there is no strong y → x structure in the data.
The log-of-evidence ratios for the dL dataset for either threshold favored an independent model. The authors performed data randomization multiple times to generate several random data sets. Randomizations were repeated to validate and calibrate the evidence ratio calculation. The log-evidence ratios between causal (dependent) and independent models for 10 random data sets were much lower than for the original data sets.
Next, the authors investigated whether age difference(s) in viral load distribution would be relevant to the dynamics of infection. To do this, viral load was linked to infectivity, the probability of transmitting infection. The team used a “projected virus isolation success” based on the probit distribution as an indicator of infectivity. There were no major differences in projected infectivity between the different age groups. This meant that at most a 50% difference (more likely a smaller one) in infectivity could be anticipated due to viral load differences between age groups.
Conclusions
In summary, the authors found that differences in SARS-CoV-2 viral load distribution between age groups in the AD dataset were statistically significant. They observed a statistically significant increase in viral load with age, a trend that fits with the generally accepted notion of a weaker immune response with advancing age.
As such, its impact on the infectivity of the different age groups was moderate. Overall, the findings underlined that viral load was only moderately dependent on age, consistent with evidence from the literature. The authors suggested that the models described could be easily adapted for general purposes and can be used for future SARS-CoV-2 variants or pandemics.
Source: www.news-medical.net