Epidemiological and Clinical Features of SARS-CoV-2 Variants Circulating between April–December 2021 in Italy

SARS-CoV-2 is constantly evolving, leading to new variants. We analysed data from 4400 SARS-CoV-2-positive samples in order to pursue epidemiological variant surveillance and to evaluate their impact on public health in Italy in the period of April–December 2021. The main circulating strain (76.2%) was the Delta variant, followed by the Alpha (13.3%), the Omicron (5.3%), and the Gamma variants (2.9%). The B.1.1 lineages, Eta, Beta, Iota, Mu, and Kappa variants, represented around 1% of cases. There were 48.2% of subjects who had not been vaccinated, and they had a lower median age compared to the vaccinated subjects (47 vs. 61 years). An increasing number of infections in the vaccinated subjects were observed over time, with the highest proportion in November (85.2%). The variants correlated with clinical status; the largest proportion of symptomatic patients (59.6%) was observed with the Delta variant, while subjects harbouring the Gamma variant showed the highest proportion of asymptomatic infection (21.6%), albeit also deaths (5.4%). The Omicron variant was only found in the vaccinated subjects, of which 47% had been hospitalised. The diffusivity and pathogenicity associated with the different SARS-CoV-2 variants are likely to have relevant public health implications, both at the national and international levels. Our study provides data on the rapid changes in the epidemiological landscape of the SARS-CoV-2 variants in Italy.


Introduction
The ongoing global coronavirus Disease 2019 (COVID-19) pandemic has caused significant mortality and morbidity [1], requiring unprecedented efforts to develop novel vaccines and strategies for fighting COVID-19. Despite its limited intrinsic genetic variability, the huge number of infections has led to the evolution of SARS-CoV-2 in an expanding array of new variants. While the driver of this evolution is certainly immune escape, the new emerging variants may differ from those previously circulating also in terms of increased transmissibility and virulence. These variations can affect diagnostic testing reliability, available treatment effectiveness, and vaccine efficacy [2,3]. Currently, the spike (S) protein is the region most affected by mutations, and the circulating SARS-CoV-2 variants are classified on the basis of variations in the S protein compared to the ancestral strain [4]. Among the growing number of SARS-CoV-2 variants documented worldwide during the pandemic, previously identified variants of concern (VOCs), such as Alpha (B.1.1.7/20I), Beta (B.1.351/20H), Gamma (B.1.1.28/P1), and Delta (B.1.617. 2), have been associated with increased transmissibility, mortality, and decreased susceptibility to neutralising antibodies induced by vaccination or previous infection [2,3,[5][6][7][8].

Statistical Analysis
Descriptive analyses of the demographic and clinical data are presented as a median and an inter-quartile range (IQR) when continuous and as a frequency and a proportion (%) when categorical. To compare normally distributed, non-normally distributed continuous, and categorical variables, parametric tests (t-test and ANOVA), nonparametric tests (Mann-Whitney and Kruskal-Wallis), and the Pearson χ 2 test (or Fisher exact test, when necessary) were used, respectively. Significance was established at p < 0.05. Data analysis was performed using the IBM SPSS Statistics version 25.
At the time of testing, mild infections were the most frequent (2330/2965, 78.6%), followed by moderate/severe infections requiring hospitalisation (474/2965, 16%). Only 4.7% (138/2965) of patients were asymptomatic. Deaths were reported in 23 patients among the subjects with known outcomes (0.8%). Statistically significant differences in clinical status were observed in patients <60 y.o. and ≥60 y.o. Globally, the resulting proportions of hospitalised subjects (44.2% vs. 55.7%; p < 0.001) and deaths (14.4% vs. 85.7%; p < 0.001) were higher in older compared to younger patients. Of note, asymptomatic status was more prevalent in subjects under the age of 60; on the contrary, deaths were observed only in patients over 60 y.o., with the exception of two unvaccinated patients in September and one patient vaccinated with only a single dose in November.

Lineages and Clades over Time
Over the study period, the prevalence of the Alpha variant (B.1.1.7/20I) significantly (p < 0.001) decreased from 77.8% (n = 273) in April 2021 to 0.7% (n = 3) in August 2021 until its disappearance. Only one single case was reported in November 2021 in Lombardy.
The Gamma variant (P.1 and its descendant/20J) remained stable in the first three months with a prevalence of 12.5% (n = 44), 17.1% (n = 45), and 19% (n = 35), in April, May, and June, respectively. The last cases were observed in July (n = 3); one of these (lineage P.1.15) was identified in a 25-year-old hospitalised patient travelling from Argentina to Apulia.
The previously circulating lineage, B.1.1, and its descendants were present at low prevalence only until August 2021, decreasing from 6.8% in April (n = 24) to 0.7% (n = 2) in July. One case of B.1.177 (20E.EU1) was observed in August in Lombardy.
The last case of the Beta variant (B.1.351/20H) was identified in August, never reaching a prevalence of above 2%.
The first cases of the Omicron variant (B.1.1.529/21K) were detected in Lombardy (n = 4), with the first sequence reported in the city of Cremona (Lombardy) on 5 November, 2021. Its prevalence reached 16.8% of cases in December 2021.
Five cases of XF recombinants were identified in December in the Marche Region, predominantly in vaccinated and hospitalised subjects (80%). Figure 1 shows the main viral variants and clades over time.
Five cases of XF recombinants were identified in December in the Marche predominantly in vaccinated and hospitalised subjects (80%). Figure 1 shows viral variants and clades over time.    Figure 2 illustrates the distribution of lineages in different Italian regions.

Vaccinated vs. Unvaccinated
We observed a significant increase in the proportion of vaccinated subjects receiving at least one dose of the COVID-19 vaccine, rising from 15.8% to 73.2% (p < 0.001) over the study period, with the highest percentage reached in November 2021 (85.2%, 115/135) ( Figure 3).
The median age of patients who had received three doses of a vaccine was significantly (p = 0.018) higher compared to that of those who had received one or two doses (70, IQR: 57.5-82 vs. 60, IQR: 45-77 and 59 years, IQR: 39-69).
Twelve patients experienced a documented reinfection; 11 were diagnosed among the unvaccinated subjects, while the remaining one occurred in a subject who had received a single dose of a vaccine. They had a median age of 54 years (IQR: 41-63), and three of them were hospitalised. Reinfections were distributed as follows: 3, 3, 1, 2, 1, 1, and 1 in April, May, June, July, August, September, and December, respectively. Until June, all patients were re-infected with the Alpha variant, while from August, the Delta variant (21J) was reported among reinfections.
Twelve patients experienced a documented reinfection; 11 were diagnosed among the unvaccinated subjects, while the remaining one occurred in a subject who had received a single dose of a vaccine. They had a median age of 54 years (IQR: 41-63), and three of them were hospitalised. Reinfections were distributed as follows: 3, 3, 1, 2, 1, 1, and 1 in April, May, June, July, August, September, and December, respectively. Until June, all patients were re-infected with the Alpha variant, while from August, the Delta variant (21J) was reported among reinfections.
Among subjects with a known vaccination status (n = 881), no differences were observed between the vaccinated and the unvaccinated concerning clinical status (12.7% vs. 11.8%, 53.3% vs. 49.7%, 31.6% vs. 35.9%, and 0.2% vs. 0.3%, in regard to asymptomatic, symptomatic, hospitalised, and dead subjects, respectively) ( Supplementary Table S1). Moreover, no differences were observed considering single-dose vaccinated or fully vaccinated and unvaccinated subjects (Figure 4). When stratifying patients according to age, a statistically significant higher proportion of hospitalisations and deaths were observed in the vaccinated subjects over 60 y.o. compared to the vaccinated subjects <60 y.o. (76.5% and 88.9%, p < 0.001); importantly, deaths were significantly higher in the unvaccinated versus the vaccinated individuals (81.8%, p < 0.001) (Supplementary Table S2). Of note, at the end of the data collection (April 2022), 91% (n = 429) of the unvaccinated subjects had not received any doses yet.

Clinical Status and Vaccination According to Variants
The viral variants also significantly correlated to clinical status in the subset of patients with known clinical status and viral variants (n = 2899, p = 0.001). Symptomatic patients were observed more frequently (59.5%) amongst those infected by the Delta variant compared to the other variants. Subjects harbouring the Gamma variant (n = 80) showed the highest frequency of asymptomatic status (21.6%), but oddly, also of deaths (5.4%).   Table S2). Of note, at the end of the data collection (April 2022), 91% (n = 429) of the unvaccinated subjects had not received any doses yet.

Clinical Status and Vaccination According to Variants
The viral variants also significantly correlated to clinical status in the subset of patients with known clinical status and viral variants (n = 2899, p = 0.001). Symptomatic patients were observed more frequently (59.5%) amongst those infected by the Delta variant compared to the other variants. Subjects harbouring the Gamma variant (n = 80) showed the highest frequency of asymptomatic status (21.6%), but oddly, also of deaths (5.4%).
In the subset of patients with known vaccination and clinical status (n = 772), significant differences according to viral variant were observed only among the vaccinated subjects (n = 393, p < 0.001). The vaccinated symptomatic patients were more frequent (61.1%) among those with the Delta variant infection, while the hospitalised subjects were more frequent in those infected with the Alpha variant (55.6%). As already observed in the whole cohort, the vaccinated subjects with the Gamma variant presented the highest proportion of asymptomatic status (41.7%) but also of deaths (8.3%). No death was observed in the vaccinated subjects carrying the Alpha and the Omicron variants ( Figure 5). In the subset of patients with known vaccination and clinical status (n = 772), significant differences according to viral variant were observed only among the vaccinated subjects (n = 393, p < 0.001). The vaccinated symptomatic patients were more frequent (61.1%) among those with the Delta variant infection, while the hospitalised subjects were more frequent in those infected with the Alpha variant (55.6%). As already observed in the whole cohort, the vaccinated subjects with the Gamma variant presented the highest proportion of asymptomatic status (41.7%) but also of deaths (8.3%). No death was observed in the vaccinated subjects carrying the Alpha and the Omicron variants ( Figure 5). The Omicron variant was only observed in vaccinated subjects, of which 47% had been hospitalised.
Concerning vaccination status, the majority of infected subjects (63%) with the Alpha The Omicron variant was only observed in vaccinated subjects, of which 47% had been hospitalised.
Concerning vaccination status, the majority of infected subjects (63%) with the Alpha variant had been vaccinated with one dose, while 58.3%, 76.5%, and 83.5% of those with the Gamma, Omicron, and Delta variants had received two doses. Only 11.8% and 3.7% of subjects carrying the Omicron and Delta variants, respectively, reported a booster dose. A similar proportion of deaths was associated with the Alpha and Gamma variants (4.9% and 4%, respectively).

Discussion
In this study, based on the characterisation of the SARS-CoV-2 strains circulating in Italy, complemented by demographics and partially by clinical data, we provide a clear picture of the spread of the SARS-CoV-2 variants in the Italian territories, highlighting how the replacement of the previously dominant variants of concern has dictated the subsequent epidemic waves in the country.
The timing of this study mostly overlaps with the emergence and the spread of the Delta variant and its descendants throughout Italy at the beginning of summer 2021 and offers a unique and well-characterised cohort of hospitalised and non-hospitalised patients covering all the Italian territories, with the exclusion of the islands.
All of these variants were clearly characterised by increased transmissibility compared with the preceding lineages, starting with the original lineage, which was detected at very low levels (<1% of total sequences) in the study period. As a consequence, B.1.617.2 + AY was the most prevalent variant in our population, followed by B.1.1.7 (76.2% and 13.3%). Most of the epidemiological success of these new variants can be attributed to their ability to overcome previous natural or vaccine-induced herd immunity, in addition to an intrinsic increase in their transmissibility.
The SARS-CoV-2's most recent VOC, Omicron, has rapidly replaced the SARS-CoV-2 Delta variant in most European countries, including Italy, starting from December 2021 [22]. Omicron, with its potential to evade the host's immune response induced by previous infections or vaccination, is considered more adaptable and transmissible than its predecessors. Immunity acquired after previous infection or vaccination is less effective against Omicron than against other variants, but the risk of severe COVID-19 is low [23].
Our study was interrupted at the end of December when only the first sublineage of Omicron (BA.1) was present; however, such a lineage was observed only in vaccinated subjects, mostly with asymptomatic or mild disease (53%).
The first detection of the Omicron variant in our country was dated at the beginning of November 2021 in Lombardy, followed by its rapid and wide diffusion in December (from 0.4% to 61.7% in 4 weeks).
On 20 December 2021, the institutional "flash survey" conducted in Italy showed that the Delta variant was still predominant, and the prevalence of the Omicron variant was 21.0% (18), in line with our data, indicating a prevalence of 16.8%.
Our data confirm that older age is associated with an increased risk of severe COVID-19 clinical course and death, also in vaccinated individuals, as reported in several studies [24,25].
Surprisingly, our results did not reveal differences between vaccinated and unvaccinated subjects concerning their clinical status of the disease. This is certainly due to the fact that the population was not randomly selected among infected subjects, but most subjects were included as patients independently referred to hospitals (and therefore, affected by more severe symptoms), therefore missing most mild/asymptomatic infections. In fact, the study population showed a low proportion of mild/asymptomatic infections (12% in our data). By contrast, a significant association was found between SARS-CoV-2 lineages and COVID-19 presentation, even if, for some variants, a low number of cases was considered.
Significant associations were found only in vaccinated subjects; specifically, hospitalisation and death showed a higher incidence in those over 60 y.o. and in subjects with the Alpha and Gamma variants, respectively.
This could possibly be explained by the fact that when these variants circulated, most of the vaccinated patients had only received a single dose. The observed median time from vaccination to infection was similar to that reported in recent work [26] and possibly due to the early production of anti-SARS-CoV-2 facilitating antibodies.
This study presents some limitations. Our study does not analyse the course and the outcome of the infections and does not consider the Ct (cycle threshold of real-time PCR) levels at diagnosis. Albeit a large number of samples was included in the study, these are not proportionally distributed among regions in relation to their total population and the related number of confirmed cases. Moreover, data were missing for many subjects, particularly concerning vaccination and data related to the first three months of 2021, characterised by the Alpha variant circulation, were not evaluated in the present analysis because they have already been discussed in previous work [16].
The COVID-19 pandemic continues to represent a global health crisis. Analyses of the epidemiology and clinical outcomes associated with the SARS-CoV-2 variants have important public health implications, both nationally and internationally. Our report complements previous analyses by providing further data on the rapid change in the epidemiological landscape of SARS-CoV-2 variants in Italy, revealing that in addition to the known variants of concern, other minor variants circulate and contribute to the epidemic.
Our study provides strong evidence consolidating the notion that since the beginning of the epidemic, local lineage replacements are associated with new local epidemic waves [16,27]. This is in line with recent work indicating that the re-increase in SARS-CoV-2 incidence could be due to the emergence of new variants rather than a rebound of previous viral genotypes [28].
Our data also reveal that during each main variant epidemic wave, the expansion of minor variants (descendant lineages) contributes significantly to the epidemics, creating "sub-waves" that likely extend the epidemic course of their parent lineage. Furthermore, independent of where these subvariants were generated, they may spread to other geographical areas with the potential to reverberate globally. In this context, the role of continuous SARS-CoV-2 genomic monitoring to follow local viral evolution in real time appears of the utmost importance.
In addition, the association between the genomic and clinical data allowed us to evaluate the role of viral variants in vaccinated people with respect to the severity of the disease.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/v14112508/s1, Supplementary information. Table S1: Number of individuals with known clinical status/outcome stratified according to vaccination data.   Institutional Review Board Statement: This study was approved by the Sacco Hospital ethics committee (protocol n. 47866, 9 September 2020).