Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in China in late 2019 and then spread rapidly, causing the first pandemic of the 21st century. Since then, epidemic spread has been sustained by the continued emergence of new variants that combine increased transmissibility1 and antigenic switch2. We are currently seeing the replacement of the Omicron BA.1 variant with a new subline, BA.2, which also emerged in South Africa in late 2021. Omicron BA.2 has fewer mutations than BA.1 in the glycoprotein of tip, some of which are shared with BA.1 and some of which are original3 (Fig. 1). In 2022, two new sub-lineages, BA.4 and BA.5, emerged from the BA.2 lineage in South Africa4. Both BA.4/BA.5 share the same mutations in the spike glycoprotein and mainly differ from BA.2 with respect to the 69-70del, L452R which was present in the Delta variant, F486V and the reversion of the original amino acid at Q4934(Fig. 1). These new mutation patterns have the potential to modify the activity of therapeutic monoclonal antibodies currently used in the clinic.
In the present study, we tested the neutralizing activity of therapeutic antibodies against the clinical strains of the BA.1, BA.2 and BA.5 sublines of the B.1.1.529 Omicron variant, using the European ancestral strain BavPat1 (lineage B.1, D614G) and a Delta variant (B.1.617.2) as reference. We tested currently used therapeutic antibodies that have been shown to retain neutralizing activity against BA.15. All target the spike receptor binding domain (RBD)6.7 (Cilgavimab/AZD1061 and Tixagevimab/AZD8895, part of the Evusheld/AZD7442 cocktail) and more specifically the central region6 for Sotrovimab/Vir-7831.
We used a standardized methodology for the evaluation of antiviral compounds based on the reduction in RNA yield8,9,10which has been applied to SARS-CoV-211,12,13,14,15. The assay was performed in VeroE6 TMPRSS2 cells and the amount of viral RNA in the supernatant medium was quantified by qRT-PCR 48 h post-infection to determine the 50% effective concentration (EC50) (Fig. 2).
Our results support previous studies indicating that sotrovimab retains some neutralizing activity against the BA.1 sublineage in vitro.2,11,16,17. In the case of variant BA.2 (Table 1, Fig. 2), with an EC50 passing from 46.0 (B.1) to 441.0 (BA.2) ng/mL, a decrease in neutralizing activity by a factor of ~*9.6 (Table 1) compared to the strain ancestral B.1, and ~1.4 compared to BA.1. This result is consistent with data from Vir Biotechnology using pseudotype virus harboring all Omicron BA.2 spike mutations and with live virus3,18,19. For BA.5, there is another decrease in Sotrovimab activity with an EC50 increasing to 858.2 ng/ml resulting in an 18.7 decrease in neutralizing activity compared to B.1, ~2.7 compared to BA.1 and ~1.9 compared to BA.2. This loss of activity was recently reported using neutralization with the pseudo-virus BA.4/BA.5 spike20.
The neutralizing activity of Tixagevimab is very weak against BA.1, BA.2 and BA.5 (EC50 > 5000 ng/mL, see Table 1). On the other hand, Cilgavimab regains a neutralizing power against BA.2 and BA.5 with an EC50 increasing only from 19.2 (B.1) to 49.8 ng/mL (BA.2) and 23.5 ng/mL (BA.5), representing a very limited loss of neutralizing activity (ratio B .1/BA.2: ~ 2.6 and B.1/BA.5 ratio: ~ 1.2 Table 1). In comparison, an 84.2-fold reduction in B.1/BA.1 neutralizing activity was observed with this monoclonal antibody. In short, this indicates that Cilgavimab showed 32 times greater activity against BA.2 compared to BA.1 in our tests. This could be due to the absence in the BA.2 and BA.5 RBD of the G446S mutation (Fig. 1), which is located in a region identified as critical for the neutralizing activity of Cilgavimab6. When Cilgavimab has been tested in combination with Tixagevimab, as offered in the therapeutic cocktail Evusheld22the EC50 changed from 20.2 (B.1) to 37.4 ng/mL (BA.2), that’s to say a 1.9-fold decrease in neutralization activity when comparing BA.2 with B.1, but a 15-fold increase when comparing BA.2 with BA.1 (Table 1). Regarding BA.5, there is a slight loss of cocktail activity compared to BA.2 with a 1.4-fold decrease, but there is still a 10-fold increase when comparing BA.5 to BA.1. For the BA.2 sub-variant these results are perfectly in line with recent studies with live viruses and different reading techniques3.19. For BA.5, our findings were also confirmed by recently produced results using the BA.5/BA.5 spike protein pseudovirus20.
The analysis of our results must be made in the context of the actual treatments administered to patients at risk of developing severe forms of Covid-19. Sotrovimab is registered in the European Union for the early treatment of infections with a single intravenous injection of 500 mg and the Evusheld AZD7442 cocktail for the prophylaxis of infection with a single dose of 300 mg (150 mg Tixagevimab + 150 mg Cilgavimab , IM administration) but the possibility of curative use in double doses (300 mg of Tixagevimab + 300 mg of Cilgavimab, IV injection) was left open. As previously described5CE-based50 values, we calculated the neutralizing capacity of each treatment expressed in MNU50(Table 1). This allows for a realistic comparison between treatments of neutralization ability against each variant.
For Evusheld/AZD7442, restoration of cilgavimab activity against BA.2 results in a significant improvement in activity per treatment compared to BA.1 (53.5 MNU50 versus 3.4 MNU50 ). For BA.5 the activity is also preserved despite the slight decrease with 35.3 MNU50 . When the activity of a 300 mg dose of Evusheld/AZD7442 is compared to that of a 500 mg dose of Sotrovimab, the advantage goes to Sotrovimab for the BA.1 variant (10.6 MNU50 versus 7.6 MNU50 (BA.2) and 3.9 MNU50 (BA.5), but at Evusheld/AZD7442 for the BA.2 (53.5 MNU50 versus 7.6 MNU50 for Sotrovimab) and the BA.5 variant (35.3 MNU50 versus 3.9 MNU50for sotrovimab). This latter finding was due to a combination of increased activity of Evusheld/ZD7442 against BA.2 and BA.5, but also slightly lower activity of Sotrovimab against BA.2 and BA.5 compared to BA. 1 (7.6 (BA.2) and 3.9 (BA.5) versus 10.6 MNU50 ).
We conclude that Sotrovimab 500mg retains partial neutralizing activity against BA.2 and BA.5 despite two successive steps of decreased activity which should be closely monitored to ensure that the 500mg dose is sufficient to provide benefit. therapy against Omicron BA.2 and BA. .5. The activity of a 300 mg dose of Evusheld/AZD7442 against BA.1 is limited in vitro and in vivo23leading to a recent FDA recommendation to use a 600mg dose instead24. The restored activity of cilgavimab against BA.2 and BA.5 allows Evusheld /AZD7442 to regain significant activity against this variant. If a dose of 600 mg becomes the norm, the expected activity against BA.2 would be on the order of 107 MNU50 and 71 UNM50for BA.5, which are close to the activity initially observed with a 300 mg treatment against the European variant B.1 (~99 MNU50). However, since the neutralizing activity of Tixagevimab is not restored against BA.2 and BA.5, it remains to be assessed by in vivo experiments whether the combination Cilgavimab and Tixagevimab is still relevant compared to Cilgavimab alone, and in what measure AZD7442 acts against BA.2 and BA.5 as monotherapy or in combination of antibodies.