Hypothesis: Possible influence of antivector immunity and SARS‐CoV‐2 variants on efficacy of ChAdOx1 nCoV‐19 vaccine

The present work provides arguments for the involvement of anti‐vector immunity and of SARS‐CoV‐2 variants on the efficacy of ChAdOx1 nCoV‐19 vaccine. First, it is suggested that anti‐vector immunity takes place as homologous vaccination with ChAdOx1 nCoV‐19 vaccine is applied and interferes with vaccine efficacy when the interval between prime and booster doses is less than 3 months. Second, longitudinal studies suggest that ChAdOx1 nCoV‐19 vaccine provides suboptimal efficacy against SARS‐CoV‐2 Alpha variant, which appears to have an increased transmissibility among vaccinated people. At the moment, ChAdOx1 nCoV‐19 vaccine is able to reduce the severity of symptoms and transmissibility. However, if the vaccinated individuals do not maintain physical preventive measures, they could turn into potential spreaders, thus suggesting that mass vaccination will not quickly solve the pandemic. Possible consequences of SARS‐CoV‐2 evolution and of repeated anti‐SARS‐CoV‐2 vaccinations are discussed and adoption of an influenza‐like vaccination strategy is suggested.


| CHADOXNCOV-19 VACCINE CLINICAL TRIALS
A vaccine is a special drug that people do not take every day but only once or a few times. It primes the immune system to fight off an infection. However, like any drug, vaccines can vary in probability of both effectiveness and side effects, and they can induce drug/vaccine resistance; benefits and risks that can differ depending on age, comorbidities and other genetic and/or environmental factors. The widespread mortality and morbidity associated with the COVID-19 pandemic has induced the development of several vaccines (Kyriakidis et al., 2021), some of which have recently received emergency use authorisation. Among them, ChAdOx1 nCoV-19 vaccine was authorised with a regimen of two standard doses (SDs) given with an interval of 4-12 weeks on the basis of the interim analysis data (Voysey et al., 2021b). Following regulatory approval, the optimal dose interval was assessed in a recent report through post hoc exploratory analyses (Voysey et al., 2021a). The ChAdOx1 nCoV-19 vaccine consists of a replication-deficient chimpanzee adenoviral vector containing the full-length SARS-CoV-2 spike glycoprotein gene, which was tested across different studies (Folegatti et al., 2020;Ramasamy et al., 2021;Voysey et al., 2021b). Based on previous experience with ChAdOx1 MERS (van Doremalen, Haddock, et al., 2020), the vaccination studies (COV001-UK, COV002-UK, COV003-Brazil and COV005-South Africa) were initially designed to assess a single dose (5 Â 10 10 viral particles) of ChAdOx1 nCoV-19 (Folegatti et al., 2020;Ramasamy et al., 2021;Voysey et al., 2021b), although other vaccination protocols consist of first (i.e., prime) and second (i.e., boost) doses. Differently from human virus-vectored vaccines, for which preexisting anti-vector immunity could reduce the vaccine immunogenicity, a chimpanzee adenovirus-vectored vaccine can bypass this possibility. However, after the first (priming) dose, tere is the possibility to develop an anti-vector immunity, which could inhibit the potency of the booster dose. Preliminary data showed that vaccination of rhesus macaques with a single dose of ChAdOx1 nCoV-19 was able to protect against SARS-CoV-2 infection, indicating the efficacy of the single-dose strategy .
However, once the studies were underway, the analysis of immune responses and other factors led to amendments to the trials including groups receiving different vaccination protocols in the analysis. Initially, low dose (LD) of viral particles (2.2 Â 10 10 viral particles) due to an inaccurate quantification of viral particles by spectrophotometric methods was administered, and further doses were adjusted to the SD (5 Â 10 10 viral particles), using a more accurate qPCR assay (Voysey et al., 2021b). Induction of both spike-specific neutralising antibody titres and T cell responses has been shown to provide protection against viral infections in animal models (van Doremalen, Haddock, et al., 2020;, and the immunogenicity data from Phase 1 (COV001-UK, begun on 23 April 2020) showed a substantial increase in SARS-CoV-2 spike neutralising antibodies (but not in interferon-γ ELISpot T cell response to SARS-CoV-2 spike peptides) with a second dose of vaccine given after 28 days (Folegatti et al., 2020). Based on this observation, the trial protocols were modified to a regime of two doses administered 28 days apart (Folegatti et al., 2020). While it was ongoing, the above protocol changes were applied to a second study (COV002-UK), which included participants who received a low dose (LD) of the vaccine (2.2 Â 10 10 viral particles) as their first dose and were boosted with a SD (3.5-6.5 Â 10 10 virus particles), called LD/SD group, and subsequently participants who were vaccinated with two SD vaccines (SD/SD group). The LD/SD cohort was enrolled between 31 May and 10 June 2020, whereas the SD/SD cohort (aged 18-55 years) was enrolled later from 9 June to 20 July 2020 (Voysey et al., 2021b).
Boosting began on 3 August 2020, resulting in a longer gap between prime and booster vaccines in LD/SD cohort (median 84 days, interquartile range, IQR, 77-91) than for those in SD/SD cohorts (median 69 days, IQR 50-86) (Voysey et al., 2021b). Indeed, most participants in the LD/SD group received a second dose around 12 weeks after the first, whereas the interval between doses for the SD/SD group (target 28 days) was both lower and more heterogeneous because of an insufficient production of the vaccine (Voysey et al., 2021b). Differently, a trial in Brazil (COV003), which began on 23 June 2020, included a SD/SD group with the majority of participants receiving a second dose within 6 weeks of the first (median 36 days) (Voysey et al., 2021b). Finally, some participants who received a low first dose (originally planned as single-dose cohort) chose not to receive the second dose and constituted a cohort of low single-dose recipients (Voysey et al., 2021a;Voysey et al., 2021b). These situations provide the opportunity to analyse the vaccine efficacy of a single dose and the effect of different dose intervals. Unfortunately, there was no overlap in enrolment of participants in these cohorts, and participants of LD/SD cohort and single LD cohort were vaccinated (prime dose) before those of SD/SD cohort (Voysey et al., 2021b).
Interestingly, vaccine efficacy against symptomatic or asymptomatic disease in participants (COV002-UK) who received a LD as their first dose of vaccine (LD/SD) was significantly higher than that of participants who received SD/SD vaccines (Voysey et al., 2021a(Voysey et al., , 2021b. Indeed, vaccine efficacies in LD/SD group was 90.0% (95% CI 67.4-97.0) and 58.9% (95% CI 1.0-82.9) against symptomatic and asymptomatic (evaluated by mean of weekly self-swab) disease, respectively, whereas they were, respectively, 60.3% (95% CI 28.0-78.2) and 3.8% (95% CI À72.4 to 46.3) in SD/SD group (data cut-off on 4 November 2020) (Voysey et al., 2021b), indicating that the two trial protocols produced significantly different protection from SARS-CoV-2 symptomatic and asymptomatic disease and transmission. Moreover, the SD/SD cohort in Brazil displayed a relatively low protection, 64.2% (95% CI 30.7-81.5), which was similar to vaccine efficacy of SD/SD UK cohort (60.3%). These surprising data might suggest that a low first dose would induce a longer and/or a higher SARS-CoV-2 immune protection. However, other factors such as dose interval are likely to be involved in determining the significant differences between LD/SD and SD/SD cohorts. In this regard, both the UK (COV002) and Brazil (COV003) SD/SD cohorts, which displayed relatively low vaccine efficacies against primary symptomatic COVID-19, had shorter dose intervals than LD/SD cohort (Voysey et al., 2021b), suggesting that the longer dose intervals of LD/SD group might give higher protection. Notably, a subsequent analysis (data cut-off on 7 December 2020) (Voysey et al., 2021a) revealed that when SD/SD group was restricted to those who received their vaccines more than 84 days (12 weeks) between the two doses (a dose interval similar to LD/SD group), vaccine efficacy of SD/SD against primary symptomatic (but not against asymptomatic) SARS-CoV-2 infection in the first 90 days after vaccination, with no significant waning of protection during this period, thus supporting the approach to delay second doses (Voysey et al., 2021a). As indicated in the report, participants were removed from the analysis of single-dose efficacy at the time of their booster dose (Voysey et al., 2021a). However, most participants in the single-dose analysis received a second dose within 90 days after the first dose. That means that the data analysed for participants from 22 to 90 days since the first dose were collected before the data cut-off date indicated in the report (7 December 2020), possibly between August and October. Instead, the group of participants reaching 91 and 120 days since the first dose is likely to represent subjects who never received a second dose, for which vaccine efficacy was assessed at the data cut-off date (7 December 2020). During this last 30-day period, the vaccine efficacy of the single dose appeared to wane, reaching only 31.6% protection (95% CI À141.8 to 80.7). This is possibly due to a progressive decrease of anti-SARS-CoV-2 spike IgG responses (64% by Day 180, GMR 0.36 [0.27-0.47]) from the peak at Day 28 and/or other factors (e.g., SARS-CoV-2 variants emerging during the month of November, see later). Altogether, the data suggested that a 3-month dose interval provided better protection after a second dose without compromising protection in the period before the booster dose is administered. This conclusion was supported by immunogenicity data that showed that in both LD/SD and SD/SD cohorts, participants who received a second SD of vaccine more than 84 days after the first had anti-SARS-CoV-2 spike IgG titres more than twofold higher than those who received the second dose within 42 days of their initial vaccination. Assuming there is a relationship between the humoral immune response and vaccine efficacy, this evidence suggested that long (≥84 days) dose intervals were more efficacious than shorter dose intervals and could induce a long protection from SARS-CoV-2 (Voysey et al., 2021a). These data were recently discussed in a report (Voysey et al., 2021a). However, the possible mechanism(s) underlying this observation was not discussed.
In this context, the likelihood of developing an anti-vector immunity on homologous boosting has been raised and this eventuality could explain the reduced efficacy of the booster dose when it was administered earlier than 84 days after the first dose. Indeed, it is likely that immunity against the antigenic proteins of simian adenovirus vector tends to wane during the time (as well as that against spike proteins), providing a rational explanation of the increased anti-SARS-CoV-2 spike IgG responses and the increased vaccine efficacy produced by delayed boosting.

| POSSIBLE INFLUENCE OF UK SARS-COV-2 VARIANT(S) ON EFFICACY OF CHADOX1 NCOV-19 VACCINE
Of note, comparison of vaccine efficacy data between the two different cut-off dates for the participants in the LD/SD cohort (the only group that remained constant in numbers and therefore comparable in longitudinal analyses), the values may suggest a slight decrease of protection (Voysey et al., 2021a). Indeed, the relative risk of infection in LD/SD group at the first data cut-off date (4 November 2020) was 0.10 (95% CI 0.03-0.33) and 0.41 (95% CI 0.17-0.99) for symptomatic and asymptomatic disease, respectively (Voysey et al., 2021b), whereas they were subsequently estimated to be, respectively, 0.19 (95% CI 0.10-0.38) and 0.51 (95% CI 0.28-0.93) at the second data cut-off (7 December 2020) (Voysey et al., 2021a), possibly suggesting a slight decrease of vaccine efficacy during the last period of about a month. In this regard, at the first data cut-off date (4 November 2020), within the LD/SD group, the symptomatic infected individuals were three in 1367 participants (0.2%) in the vaccinated group and 30 in 1374 participants (2.2%) in the control group.
Instead, at the second data cut-off date (7 December 2020), within the same LD/SD group, the symptomatic infected individuals were (1.53%) in the control group, which correspond to 0.33 (95% CI 0.14-0.77) of relative risk of symptomatic infection that is about three times higher than that of the first time period. Notably, the LD/SD cohort was enrolled between 31 May and 10 June 2020 (Voysey et al., 2021b), and most of them had a booster dose about 3 months later (median 84 days, interquartile range 77-91) (Voysey et al., 2021a), that means that booster doses occurred between the end of August and the beginning of September (Voysey et al., 2021b). Therefore, at the two data cut-off dates (4 November and 7 December 2020), LD/SD cohort was, respectively, analysed about 2 (September to 4 November) and 3 months (5 November to 7 December) after the booster dose (considering that full vaccine protection was assessed to be achieved 22 days after vaccination). During the longitudinal study, the frequency of infected individuals in control group was 2.2% in the first time period (data cut-off date 4 November) and 1.53% in the second time period (between the two data cut-off dates). Because in United Kingdom the frequency of infected individuals increased during the month of November, it suggests that the second time period was shorter than the first. Nevertheless, the frequencies of spontaneous infection were somehow similar (and comparable) between the two groups. In addition, during the time between the two data cut-off dates, a similar trend of the relative risk of infection was observable  . In this regard, a recent sequencing of the Alpha variant (Wise, 2021) revealed the presence of the E484K mutation (first identified in South Africa) and several studies showed reduced neutralising activity of monoclonal antibodies from convalescent or vaccinated individuals against virus mutants containing the E484K mutation (Chen et al., 2021;Madhi et al., 2021;Wibmer et al., 2021;Zhou et al., 2021). Moreover, the presence of the N439K mutation, which has emerged independently in several variant lineages, has been shown to increase both spike binding affinity for human ACE2 and resistance to several anti-SARS-CoV-2 neutralising antibodies, which give SARS-CoV-2 variants carrying N439K a selective advantage (Thomson et al., 2021). Altogether these observations suggest that neutralising antibodies elicited by ancestral spike vaccines induce cross-protection from the Alpha variant, although they may not be able to fully protect against the UK variant and, in particular, against its transmission. In this regard, a post hoc analysis of the efficacy of ChAdOx1 nCoV-19 vaccine against the Alpha variant has shown that clinical efficacy against symptomatic infection was 70.4% (95% CI 43.6-84.5) (whereas it was 81.5% [95% CI 67.9-89.4] for non-Alpha lineages) and it was 28.9% (95% CI À77.1 to 71.4) against asymptomatic infection (Emary et al., 2021); values that are very similar to those calculated in the present report for the LD/SD group between 5 November and 7 December, which were, respectively, 67.2% (95% CI 23.0-86.0) and 35.9% (95% CI À47.6 to 72.2). In line with these observations, the report showed that the neutralisation activity of vaccine-induced antibodies in a live-virus neutralisation assay against the Alpha variant was about nine times lower than against the ancestral lineage (GMR 8.9 [95% CI 7.2-11.0]) (Emary et al., 2021). Notably, participants of the study were recruited between 31 May and 13 November 2020 (Emary et al., 2021) which was before the Alpha variant expanded and evolved acquiring new immune escape mutations, thus suggesting that current vaccine protection could be even lower, in particular that of the single dose (see Figure 1). Altogether, these observations suggest that with potential consequences for the future generations (see Gonzalez et al., 2021). Therefore, in order to reduce current pressure on healthcare systems, vaccination should be focused on protecting the most vulnerable (minority) part of the population for which the risk/ benefit balance of vaccination is more favourable.  (Challen et al., 2021). Results are in agreement with those of another recent report   After consideration of all these possibilities, it is clear that, although vaccine strategies may temporarily reduce both disease severity and spread, they are unlikely to prevent the appearance of new variants and to be effective in quickly solving the pandemic crisis.
It is instead likely that mild endemic disease will be slowly achievable by mass vaccination and neutralising antibody strategies, whereas global herd immunity (as for the influenza virus) is unlikely to be attained. On the other hand, it is not clear how mass vaccination (never undertaken against highly mutable RNA viruses) may influence the course of the SARS-CoV-2 mutant evolution, although this information will rapidly be available in highly vaccinated countries. Therefore, more caution should be taken with vaccination strategies in order to avoid both the development of more aggressive mutants and health risks in young people, potentially inducible by repeated vaccination using the new type of vaccines. Again, vaccination strategy for influenza virus, which is limited to the most vulnerable people, has been shown to successfully control the severe infection and to not induce more aggressive variants. Finally, there is the need to keep searching for new pharmacological therapies, and more scientific efforts should be directed towards pharmacological approaches that, working downstream in the infection pathways, are independent of the virus variant and allow the development of a natural and lasting immunity. In this regard, clinical trials employing new safe pharmacological treatments for COVID-19 with a potentially effective mechanism of action that are not tested in clinical trials yet, such as inhibitors of ACE2/ADAM17 zinc-metalloprotease activity, are urgently needed (see Zamai, 2020Zamai, , 2021Yuan et al., 2020).

| Nomenclature of Targets and Ligands
Key