The COVID-19 pandemic has become one of the biggest challenges of the 21st century. Suddenly appearing on December 31, 2019 in Wuhan (PRC), it covered most of the world within a matter of weeks. The lack of any effective antiviral drugs and specific vaccines prompted most countries globally to respond in the form of unprecedented restrictive measures. On the other hand, the lack of effective therapeutic and prophylactic agents contributed to the unprecedented activity of the scientific community. This resulted in the rapid development and implementation of a whole series of vaccines on various platforms, with no parallels in the history of medicine [33, 53, 54]. Four of them are the most widespread: messenger RNA-based vaccines (designated as group V1 in this work) [55, 56]; vector vaccines (V2) [57, 58]; peptide and protein vaccines (V3); and inactivated, whole-virion vaccines (V4) [42, 43, 44].
It should be noted that most vaccines were created on the basis of the progenitor Wuhan strain [59]. Meanwhile already in 2020, new genetic variants of the virus began to appear, characterized by more pathogenic properties. In result, data appeared on the ability of new viral variants to overcome the adaptive immunity created by vaccines [60, 61]. In addition, it has been shown that various vaccines form post-vaccination immunity differently; they can differ both in duration and in conferred resistance to new viral genetic variants [62]. All of the above may have contributed to the varying COVID-19 prevalence around the world.
In this regard, it seemed relevant to assess the relationship between COVID-19 prevalence in the population and SARS-CoV-2 vaccine types used in different countries globally, taking into account demographic (population size, density) and immunobiological (testing and vaccination coverage) factors. The work was sequentially performed, according to algorithm, in several stages (Fig. 1). The main results are presented in Fig. 2. To this end, the necessary information was collected from available sources for 104 countries. The main inclusion criterion was availability of information on the vaccines used, including their proportional contributions to overall vaccination.
The collected information was subsequently summarized in a general table (Table 1S), and initial analysis showed high heterogeneity in all indicators. Heterogeneity in terms of demographic indicators was seen by the presence of all country groups in the initial data set (according to UN classification); only dwarf states were subsequently excluded. For the goals set in the research, vaccination coverage was a critical indicator, initially ranging from 0.1% in Burundi to 96.6% in Chile and South Africa (Table 1S). For valid analysis, countries with vaccination coverage below 50% were further excluded. The People's Republic of China (PRC) was included in the data set despite local peculiarities of controlling the COVID-19 epidemic, specifically the PRC's overall 'zero COVID-19 strategy' in the event of outbreaks. This strategy is essentially unique and is not applied in other countries of the world.
Following preparation of a final data set of 53 countries, reported COVID-19 case numbers were found to be most strongly correlated with testing coverage and proportion of vaccine types used. Due to the fact that an inverse correlation was found between 'reported COVID-19 case numbers' with V2, V3 and V4, we considered it possible to combine these three types of vaccines into one group (non-mRNA group vaccines, Vnmg) for further analysis. Moreover, to offset the impact of SARS-CoV-2 testing on reported COVID-19 case numbers, countries with low testing rates (< 1 test/person) and countries with extremely high testing rates were excluded. The resulting analyzed sample represented 33 countries. When analyzing reported COVID-19 case numbers within this sample, it was shown that the highest numbers occur in countries with V1 vaccine (mRNA) dominance. The lowest was seen in countries with Vnmg (vector, peptide/protein, whole-virion/inactivated) vaccine dominance (Fig. 12).
For the most appropriate comparison of Russia with other countries, a data set was formed with a specific median level of 'testing coverage' (2001 per 1,000 people). Analysis showed that, like analysis of the overall sample, there were significant differences between countries with V1 vaccine (mRNA) dominance and countries with Vnmg dominance.
Moreover, it should be noted that, despite exactly the same testing levels, 'reported COVID-19 case numbers' for Russia (with Vnmg usage exclusively) were 1.6-fold lower than, for example, in Finland. Russia's low numbers are remarkable given the fact that Finland itself already represents with the lowest case numbers among 10 countries with the dominant use of mRNA vaccines (Fig. 13). The results show that in countries where non-mRNA vaccines have been used, 'reported COVID-19 case numbers' are significantly lower than in countries with dominant use of mRNA vaccines.
A chart summarizing information for all 53 countries regarding reported COVID-19 case numbers, and their dependence on vaccine types used, is presented in Fig. 14. In addition to polar groups (nations dominated by V1 or Vnmg), the group of countries with equal proportions of mRNA and non-mRNA vaccines is of particular interest. As expected, reported COVID-19 case numbers for these countries are middle values.
The South Asian country of Bhutan stands apart because unique experience was gained through combined use of different vaccine types. The first immunization was carried out with an mRNA vaccine; the second immunization used a vector vaccine [63]. At the same time, high vaccination coverage and the lowest 'reported COVID-19 case numbers' were achieved among countries in this group. However, case numbers in Bhutan were higher than in the neighboring regional countries of India and Nepal. To be fair, in the latter two countries, testing coverage was less than 1 test per person.
A similar situation has developed in East Asian countries. The maximum prevalence was registered in South Korea, which used mainly mRNA vaccines. The minimum was registered in China with the dominant use of whole-virion vaccines. Hong Kong, with an equal proportion of mRNA and non-mRNA vaccines, featured an intermediate prevalence. This is despite the fact that: the minimum testing coverage was in South Korea; and in China and Hong Kong the level of testing was equally high. Interestingly, in Japan, the prevalence was the lowest among countries in the region; this was combined with a very low level of testing for highly developed countries. On the other hand, in Israel, where testing coverage was even slightly lower than in Hong Kong and China, 'reported COVID-19 case numbers' were 2.8-fold higher than in Hong Kong and three orders of magnitude higher than in China.
In a number of European countries in which mRNA vaccines dominated (France, Portugal, Luxembourg, Estonia, Slovenia, Switzerland), prevalence was 18–45% higher than in the UK. The UK had an even vaccine-type ratio (mRNA vs. non-mRNA), and testing coverage was equal to, or higher (1.5 to 3-fold) than, the listed countries.
In another group of four European countries with the same testing level, maximum prevalence was registered in Germany and Croatia, which mainly used mRNA vaccines. The minimum was in Belarus with dominant use of vector and whole-virion vaccines (2.6 to 3-fold less than Germany and Croatia). An intermediate level was seen in Hungary with an equal ratio of mRNA and non-mRNA vaccines (1.4 to 1.6-fold less than in Germany and Croatia).
The presented statistical calculations inevitably raise the question of what is the underlying cause of the difference noted. Comparative studies have convincingly shown the high efficacy of all currently approved vaccines [61, 64]. However, it is impossible not to notice some differences that can affect prevalence. Regarding mRNA vaccines, one can agree with the opinion of a number of researchers who have shown that they are able to form rapid immunity already in the early stages after immunization, persisting for 3 months, following which, use of a booster dose may be required [21]. As for vector vaccines (assigned to the Vnmg group), full-fledged immunity is formed by the 14th day after the second dose, but it exists for at least 6 months [28, 30, 65, 66]. Regarding whole-virion vaccines, it is worth noting their lowest immunogenicity, alongside their longest elicited immunity [67, 68], which is closest to a post-infectious response.
The key parameter is probably the formation of stable herd immunity. Unfortunately, there are currently no published studies that have been conducted according to a single methodology with different countries globally when examining epidemic process dynamics as was done here. In on our experience, assessment of SARS-CoV-2 collective immunity, carried out according to a single methodology [69] at different stages of the epidemic in Russia [8, 70], Belarus [71], and Kyrgyzstan [72], showed that there is successful formation of herd immunity in those countries. Those countries were discussed in our earlier work, and usage of vector and whole-virion vaccines usage dominates in them.
On the other hand, in the work of Morens et al. [73], it was suggested that it is impossible to achieve long-term herd immunity with COVID-19 and therefore regular booster immunization is necessary. This is probably true primarily for mRNA vaccines, whose distribution currently dominates globally. In our opinion, the main reasons for this are: breadth (spectrum) of the immune response; and constant viral variability, through which new genetic variants appear regularly [74].
Vector vaccines, and even more so whole-virion vaccines, induce a significantly wider range of post-vaccination antibodies than mRNA vaccines [33, 75]. Minimal diversity and maximal specificity were features embraced in the initially ideology of mRNA vaccine development. Undoubtedly, it is a very progressive technology that makes it possible to induce a narrow spectrum of post-vaccination antibodies, resulting in a decrease in the share of post-vaccination adverse reactions, including those of an autoimmune nature [76]. In conditions of high viral variability, however, the technology likely has a number of limitations, and the formation of a narrow Ab spectrum is more of a disadvantage than an advantage. In result, the level of post-vaccination immunity persists for a short time, and its restoration is impossible without the introduction of a second vaccine dose.
In contrast, an immune response is formed to a much wider range of antigens and their epitopes with the use of vector and whole-virion vaccines. In result, when post-vaccination immunity is attenuated, a repeat encounter with the virus, even a new genetic variant, leads to activation of the secondary immune response. As such, regular booster immunization is not required. A similar situation is observed with healthcare workers. Often, they have been ill only once, but regular restoration of post-infectious immunity ensues through periodic contact with infected carriers.
When analyzing the assembled country database, we did not have data on the timing of primary or booster vaccinations of course. Therefore, we could not in any way assess post-vaccination immunity level at the time of information collection. Another factor is the uneven spread across countries of new genetic variants of the virus, which may feature different abilities to evade the immune response. Answering these and other questions would likely permit more accurate determination of the nature of post-vaccination morbidity. Perhaps this will be the subject of a separate study someday.
Thus, we have shown for the first time that 'reported COVID-19 case numbers' (per million population) depend not only on SARS-CoV-2 testing coverage and vaccination coverage, which is quite logical, but also on the vaccine types used. With the same level of vaccination and testing coverage, countries using predominantly vector and whole-virion vaccines experienced significantly lower prevalence than countries predominantly using mRNA vaccines.