key: cord-0737396-2kyy9jdb
authors: Althaus, C. L.; Baggio, S.; Reichmuth, M. L.; Hodcroft, E. B.; Riou, J.; Neher, R. A.; Jacquerioz, F.; Spechbach, H.; Salamun, J.; Vetter, P.; Williamson, C.; Hsiao, N.-y.; Preiser, W.; Davies, M.-A.; Lessells, R. J.; de Olivera, T.; Kaiser, L.; Eckerle, I.
title: A tale of two variants: Spread of SARS-CoV-2 variants Alpha in Geneva, Switzerland, and Beta in South Africa
date: 2021-06-15
journal: nan
DOI: 10.1101/2021.06.10.21258468
sha: 64d245cb0b6b10cd4a6d646c8b6ab04ad1d0a14f
doc_id: 737396
cord_uid: 2kyy9jdb

Several SARS-CoV-2 variants of concern (VOC) are spreading rapidly in different regions of the world. The underlying mechanisms behind their transmission advantage remain unclear. We measured viral load in 950 individuals and found that infections with variant Alpha exhibit a higher viral load and longer viral shedding compared to non-VOC. We then used a transmission model to analyze the spread of variant Alpha in Geneva, Switzerland, and variant Beta in South Africa. We estimated that Alpha is either associated with a 37% (95% compatibility interval, CI: 25-63%) increase in transmissibility or a 51% (95% CI: 32-80%) increase of the infectious duration, or a combination of the two mechanisms. Assuming 50% immune evasion for Beta, we estimated a 23% (95% CI: 10-37%) increase in transmissibility or a 38% (95% CI: 15-78%) increase of the infectious duration for this variant. Beta is expected to outgrow Alpha in regions where the level of naturally acquired immunity from previously circulating variants exceeds 20% to 40%. Close monitoring of Alpha and Beta in regions with different levels of immunity will help to anticipate the global spread of these and future variants.

Novel SARS-CoV-2 variants emerged independently in different geographic regions of the world. For Alpha, several studies have estimated an increased transmissibility between 40% to 64 100% in the United Kingdom, the United States, Denmark, and Switzerland (Leung et al.,

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 15, 2021. ; https://doi.org/10.1101/2021.06.10.21258468 doi: medRxiv preprint

To track the spread of Alpha in Geneva, Switzerland, we relied on the identification of Alpha 127 described above. To cover the period of November and December 2020, we used sequence data 128 from randomly chosen samples from Geneva that were submitted to GISAID by the Swiss SARS- Competitive spread between variant and non-variant ('wild-type') strains of SARS-CoV-2 can be described within the susceptible-infected-recovered (SIR) framework by the following two ordinary differential equations:

where W and V are individuals infected with wild-type and variant, respectively, and S the 3. Immune evasion: The variant can partially evade the acquired immunity from previous 144 infections by the wild-type (1 − S). Immune evasion can vary from complete cross-145 protection (ε = 0) to full evasion (ε = 1).

One can show that the proportion of the variant among all infections increases according to logistic growth (Marée et al., 2000) :

where µ = W (0)/V (0) and ρ corresponds to the difference in the net growth rates between the variant and the wild-type:

Eq. 4 can be solved algebraically for τ, κ or ε. If the transmission advantage acts via a single 147 mechanism only, we obtain the following simplified solutions. First, the increased transmissibility . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 15, 2021. ; https://doi.org/10.1101/2021.06.10.21258468 doi: medRxiv preprint evasion (τ = ε = 0). Finally, assuming there is no change in transmissibility nor the infectious 153 duration (τ = κ = 0), the level of immune evasion is given by ε = ρ/(β Ω) = ρD(1 − Ω)/(ΩR w ), 154 where Ω = 1 − S corresponds to the proportion of the population with previously acquired 155 immunity against earlier variants, i.e., the cumulative incidence or seroprevalence, at the time 156 the variant starts to grow. 157 We estimated ρ by fitting a logistic growth model (binomial regression) to the proportion 158 p(t) of Alpha in Switzerland, Geneva, and Beta in South Africa. To propagate the uncertainty, 159 we constructed 95% compatibility intervals (CIs) for τ, κ and ε from 10,000 parameter samples 

We analyzed viral load in a total of 950 specimens from Geneva, Switzerland (604 non-VOC, 176 346 Alpha). We found a higher mean viral load for Alpha compared to non-VOC (7.4 vs. 6.9

177 SARS-CoV-2 log10 RNA copies/ml, p < 0.001) (Figure 2A ). Analyzing viral load by day post 178 onset of symptoms showed a delayed decrease in viral load for Alpha compared to non-VOC 179

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 15, 2021. ; https://doi.org/10.1101/2021.06.10.21258468 doi: medRxiv preprint ( Figure 2B) . Notably, viral load for non-VOC fell below the threshold for presence of infectious 180 (culturable) virus (10 6 SARS-CoV-2 RNA copies/ml) at day 6 to 11. In contrast, viral load 181 remained above that threshold for B.1.1.7 in samples taken from day 6 to 11 post onset of 182 symptoms. Together, these data suggest that Alpha exhibits a transmission advantage that is 183 mediated by either an increased transmissibility or a longer infectious duration, or a combination . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 15, 2021. Figure 5 . Growth advantage of Alpha and Beta over 'wild-type' variants of SARS-CoV-2. For Alpha, we assumed no immune evasion (ε = 0), i.e., the growth advantage is constant and independent of the seroprevalence. For Beta, we assumed immune evasion of 25% (A), 50% (B), and 75% (C). The growth advantage is relative to a 'wild-type' variant with R w = 1. Lines and shaded areas correspond to the median and 95% compatibility intervals.

the level of population immunity.

Our study has a number of limitations. First, we inferred specimens from Geneva, Switzer- Immune evasion is arguably of bigger concern than increases in transmissibility or the in-265 fectious duration, especially if there is similar evasion of vaccine-elicited immunity. Similar to 266 Gamma, Beta appears to exhibit partial immune evasion. The finding that Beta is not expected Europe that experience a high cumulative incidence of SARS-CoV-2 in 2020.

Our study helps to understand the consequences of the altered transmission characteristics . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 15, 2021. ; https://doi.org/10.1101/2021.06.10.21258468 doi: medRxiv preprint

Nasopharynx of Symptomatic Neonates, Children, and Adolescents. Emerg 381 Infect Dis

Estimating relative fitness 383 in viral competition experiments

CMMID COVID-19 working group