Epidemic model for estimating relative transmissibility and immune escape of SARS-CoV-2 Omicron variant in South Africa

As of December 28th, 2021, the Omicron variant of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been detected in more than 110 countries and territories worldwide. SARS-CoV-2 is the causal agent of the coronavirus disease 2019 (COVID-19) pandemic, which has ravaged human lives and the global economy.

Study: Preliminary modeling estimates of the relative transmissibility and immune escape of the Omicron SARS-CoV-2 variant of concern in South Africa. Image Credit: anushkaniroshan/Shutterstock

The Omicron variant was first discovered in South Africa and was officially named a variant of concern (VOC) on November 26th, 2021. Scientists are working extensively to understand the properties of this new variant to inform public health measures appropriately.

A new study, published on the medRxiv* preprint server, developed a stochastic, multi-strain, compartmental epidemic model to estimate the relative transmissibility and immune escape capacity of the Omicron variant.

Background

The widespread community transmission of Omicron was first observed in South Africa – particularly in the province of Gauteng. The country had experienced a severe wave of the Delta VOC earlier in the year; however, cases and deaths were declining since August 2021. As of late November, only 24% of the population of South Africa was fully vaccinated – suggesting that the mitigation of the Delta wave was mainly driven by high levels of natural immunity across the population.

In November, the rapid rise of cases indicated that the new variant had a significant growth advantage over the Delta VOC. Based on early statistical analysis, scientists stated that the fast spread of Omicron could be explained by increased transmissibility, immune escape, or a combination of both.

A new study

In the current study, scientists developed a multi-strain, stochastic, compartmental epidemic model for South Africa to identify some of the features of the new variant compatible with epidemiological observations. The inputs to the model are several important factors, such as demographics, age-stratified contact patterns, non-pharmaceutical interventions (NPIs), vaccine rollout, etc. Subsequently, a multi-stage calibration was performed by applying an Approximated Bayesian Computation (ABC) method.

Scientists explored a parameter space defined by combining the relative transmissibility of the Omicron variant compared to the Delta VOC. They also studied the immune escape of Omicron with respect to both naturally acquired immunity and vaccines. The next step was to obtain a joint posterior distribution of these parameters, compatible with the number of confirmed cases until December 13th, 2021.

Main findings

The results obtained in this study are in line with early statistical analysis stating that Omicron can reinfect individuals at rates higher than previous VOCs, and vaccines might be less effective against infection. Researchers obtained a joint posterior distribution for the relative transmissibility with respect to Delta VOC and immune escape of the Omicron variant.

One challenge with the existing data is that it does not allow to identify both parameters uniquely, because of which scientists defined a region where a large spreading advantage might be compensated by a limited immune escape and vice versa.

Another interesting observation was that the assumed generation time of Omicron had a significant influence on the results. Short generation times (i.e., 3.5 days) with respect to the Delta variant (i.e., 5.5 days) shifted the joint posterior distribution to a region with smaller values of transmissibility advantage for Omicron. The findings obtained in this study reinforce the rapid spread of the Omicron VOC, which was the dominant circulating strain since the second week of November.   

Limitations

The first limitation is with respect to the compartmental structure, which is relatively simple and does not account explicitly for asymptomatic transmission and different degrees of disease severity. Secondly, the data were also limited on the exact number of different vaccines administered. Thirdly, further mutations and divergence of Omicron could affect some of these values. Lastly, the model did not account for geographical heterogeneity.

Conclusion

The results presented in this study confirm that more data is necessary to estimate the key characteristics of the Omicron variant. However, the preliminary analysis suggests that the Omicron variant could also cause new pandemic waves in regions with high attack rates from previous strains and/or vaccination rates. Data on the severity of the Omicron variant, compared to the Delta VOC, will be essential to gauge the impact on the healthcare systems of countries affected by a rise in cases driven by the Omicron variant.

*Important notice

medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.