1. Introduction
Clinical trials and real-life studies have extensively evaluated the SARS-CoV-2 immune responses both after natural infection and after vaccination, demonstrating preserved protection against hospitalization and deaths, with diminishing efficacy against symptomatic infections caused by the continuously evolving, immune-evasive viral variants [
1]. Nevertheless, there are several vulnerable populations who are still at risk for severe forms of COVID-19, including immunosuppressed persons. An analysis of the data gathered in the WHO Global Clinical Platform on COVID-19, revealed an independent correlation between HIV infection and severe outcomes of patients hospitalized with SARS-CoV-2 infection [
2]. Consequently, the European AIDS Clinical Society strongly recommends vaccination of people living with HIV (PLWH), irrespective of CD4 count and HIV viral load. Variant adapted booster doses are also recommended with high priority for PLWH [
3,
4].
Passive immunization using anti-SARS-CoV-2 monoclonal antibodies, used as a possible pre-exposure prophylaxis in PLWH (especially in those unvaccinated, or with advanced immunosuppression), is no longer used due to the diminishing efficacy against the currently circulating SARS-CoV-2 variants [
3].
In PLWH with controlled HIV infection, who are effectively treated, immunocompetent and with suppressed viral replication, SARS-CoV-2 infection has similar incidence rates and comparable evolution with the general population [
5]. Nevertheless, PLWH with advanced HIV disease and severe immunodeficiency are prone to a more severe evolution of COVID-19, with higher risk of hospitalization and increased mortality [
6,
7]. After the emergence of the Omicron variant and its continuous antigenic drift, numerous breakthrough SARS-CoV-2 infections have been reported in all PLWH, including in vaccinated persons, with a higher frequency compared to the general population [
8]. In addition, persistent SARS-CoV-2 infections, with continuous viral replication and high variability have been reported in immune-suppressed HIV infected persons, representing a potential source of new SARS-CoV-2 viral variants [
1,
9]. Repeated breakthrough infections could also potentially increase the HIV reservoirs in PLWH, even in those with undetectable HIV RNA in plasma on combination antiretroviral therapy (cART) [
10].
Data on anti-SARS-CoV-2 immune responses efficiency in PLWH are controversial. While some studies indicate similar humoral and cellular immune responses to those developed in HIV-negative subjects [
11,
12,
13,
14,
15], others suggest a decreased immune response efficiency in PLWH [
16,
17,
18]. Severe immunosuppression decreases the ability to mount specific immune responses, both after infection and vaccination [
5,
19], and more severe breakthrough infections were diagnosed in PLWH even during the early COVID-19 waves [
7,
20]. As such, data regarding the SARS- CoV-2 immunity in HIV infected patients are still needed and are continuously gathered across diverse global regions with variable HIV demographics and healthcare standards [
6].
The objective of the present study was to assess the seroprevalence of SARS-CoV-2 infection in non-hospitalized PLWH under antiretroviral therapy, and to analyze the persistence of specific SARS-CoV-2 humoral and cellular immune responses after vaccination and/or infection.
3. Results
3.1. General Characteristics of the Enrolled Subjects
The study included 104 PLWH (65.4% males, mean age 38.4 ± 0.98 years, mean duration of HIV infection 127.3 ± 10.55 months). HIV infection was acquired by sexual transmission (58.7% of cases, either hetero or homosexual), parenteral transmission (32.7% by injecting drug use, 4.8% parenterally infected during early childhood), and by vertical transmission (3.8% of cases). The mean CD4 count at the time of the study was 606.6 ± 32.62 cells/mm
3, 27.9% of the patients had mild immunosuppression (CD4 200 - 500 cells/mm
3) and 12.5% had severe immunosuppression (CD4 < 200 cells/mm
3). The general characteristics of the study patients divided by the degree of immunosuppression are presented in
Table 1.
All patients were treated with cART at the time of the study, with a mean length of antiretroviral therapy of 120.5 ± 9.9 months. All regimens included a backbone of 2 nucleoside analogues reverstranscriptase inhibitors (NRTI), combined with an integrase inhibitor (INSTI) in 74% of cases, a protease inhibitor (PI) in 18.3% of cases, and a non-nucleoside reverstranscriptase inhibitor (NNRTI) in 7.7% of cases. Only 60.6% of PLWH included in the study (63/104) had undetectable viral load, with significant differences according to the degree of immunosuppression, as shown in
Table 1. Patients who were severely immunosuppressed have been diagnosed more recently, had lower CD4 nadir counts and higher HIV viral loads.
Out of 104 PLWH, only 40 (38.5%) were vaccinated against SARS-CoV-2 infection, either with an mRNA vaccine (Pfizer-BioNTech vaccine, BNT162b2 - 16 patients), or with a viral-vector based vaccine (Astra-Zeneca vaccine, ChAdOx1-S - 11 patients, or Janssen, Ad26.COV2.S - 13 patients). Just eight patients had received a booster dose; all boosters were monovalent mRNA vaccines, with the ancestral SARS-CoV-2 strain. The mean duration of time between the last vaccine dose and enrollment was 8.76 ± 4.1 months.
Only 11 subjects (10.6%) had been previously diagnosed with a SARS-CoV-2 infection, all with a mild clinical form. The mean time length between infection and enrollment was 9.36 ± 1.6 months.
3.2. SARS-CoV-2 Seroprevalence
Seventy-three patients (70.2%) had anti-Spike IgG antibodies (mean titer 477.3 ± 232 BAU/mL), 58 (55.8%) had also neutralizing antibody against the Omicron variant (an inhibition value > 30% in the surrogate SARS-CoV-2 virus neutralization test). Only 24 patients (23%) had anti-NCP antibodies (mean reactivity 2.3 ± 0.7) associated with anti-S IgG antibodies, indicating a prior SARS-CoV-2 infection.
Anti-S IgA antibodies were present in 62 patients (59.6%, with a mean reactivity 4.2 ± 1.1), always associated with anti-S IgG antibodies.
All vaccinated patients had anti-S IgG antibodies, 12 of them had also anti-NCP antibodies, signaling a prior SARS-CoV-2 infection and thus, hybrid immunity. Out of 40 vaccinated patients, 30 (75%) had detectable neutralizing antibodies against the Omicron variant (neutralizing activity in the surrogate SARS-CoV-2 virus neutralization test higher than 30%) and anti-S IgA antibodies.
The serologic profile of the previously SARS-CoV-2 infected patients was more heterogenous. Interestingly, 4 out of the 11 patients who declared past COVID-19, had no SARS-CoV-2 antibodies (negative for anti-S IgG and IgA, and for anti-NCP IgG), and 2 patients had only anti-S IgG antibodies (mean titer 636.2 ± 124 BAU/mL), with no detectable anti-NCP antibodies. Nevertheless, other 33 patients without a prior history of SARS-CoV-2 infection or vaccination, displayed anti-S IgG antibodies (with or without anti-NCP antibodies), indicating asymptomatic, undiagnosed infections.
The patients’ distribution according to their SARS-CoV-2 serologic status and vaccination/infection history is shown in
Figure 1.
As seen in
Figure 1, out of all enrolled patients, 31 have
no SARS-CoV-2 serologic markers; they have significantly decreased CD4 counts (mean 443.8 ± 64.49
vs. 670.4 ± 35.3;
p = 0.0015) and significantly lower percentages of undetectable HIV viral load (19.4%
vs. 78.1%;
p = 0.021) compared to those with detectable SARS-CoV-2 antibodies.
3.3. Antibody Response according to SARS-CoV-2 Vaccination/Infection Status
PLWH with hybrid immunity against SARS-CoV-2 (infection plus vaccination) have significantly higher levels of anti-S IgG antibodies (mean titer 653.3 BAU/mL
vs. 412.4 BAU/mL in those infected only, and 464.3 BAU/mL in those vaccinated only, and higher neutralization capacity (82.96%
vs. 60.97% in those infected only, and 57.06% in those vaccinated only), as observed in
Figure 2. There is a significant correlation between the anti S IgG titers and the neutralizing activity against the Omicron variant. There are no significant differences in the antibodies titers according to the time since vaccination (vaccinated < 6 months
vs. vaccinated 6 - 12 months ago
vs. vaccinated > 12 months) (results not shown). The reactivity of anti-spike IgA antibodies is similar in all patients (vaccinated, infected or with hybrid immunity).
3.4. SARS-CoV-2 Humoral Immunity in Severe Immunosuppressed PLWH
As shown in
Table 2, severe immunosuppression (CD4 count < 200 cells/mm
3) is associated with significantly lower percentages of patients with detectable anti-S IgG (
p = 0.0003), anti-S IgA antibodies (
p < 0.0001), and with lack of neutralizing activity against Omicron variant (
p < 0.0001). Patients with advanced immunosuppression have lower anti-S IgG antibodies mean titers (
p = 0.005) and anti-S IgA antibodies reactivities (
p = 0.002) compared to patients with mild or no immunosuppression; moreover, there is a significant correlation between the CD4/CD8 ratio and the titers of anti-S IgG and IgA antibodies (r = 0.353;
p =0.0003 and r = 0.4208;
p <0.0001, respectively, using non-parametric Spearman test).
3.5. SARS-CoV-2 Cellular Immunity
Cellular immunity was assessed in 15 PLWH similarly distributed in three groups: without immunosuppression
- CD4 count > 500 cells/mm
3, mild immunosuppression - CD4 count 200-500 cells/mm
3, and severe immunosuppression - CD4 count < 200 cells/mm
3. PBMCs from the patients with CD4 count lower than 200 cells/mm
3 have inferior responses when stimulated with a recombinant Omicron BA.4/BA.5 S1 protein, as compared to patients with mild or absent immunosuppression (
Figure 3A).
Detectable IFN-γ producing T cell responses were present in 13 out of 15 evaluated patients (86.7%), including two patients with CD4 < 200 cells/mm
3 who did not display any serological markers of SARS-CoV-2 infection. There were no detectable IL-2 secreting T cells. Subjects without immunosuppression (> 500 CD4 cells/mm
3) respond better to re-stimulation, although the cellular immune response displays individual heterogeneity. The non-parametric Spearman test did not detect any correlation between anti-SARS-CoV-2 specific cellular response and the CD4/CD8 ratio (r = 0.2151;
p = 0.4384) (
Figure 3B).
4. Discussion
We report that the majority of PLWH have preserved SARS-CoV-2 humoral and cellular immunity post-vaccination and post-infection, including anti-S IgA antibodies, neutralizing antibodies against Omicron variant, and IFN-γ secreting T cell responses. An important proportion of PLWH who have not been diagnosed with a prior SARS-CoV-2 infection and have not been SARS-CoV-2 vaccinated display serologic markers of SARS-CoV-2 infection, suggesting rather frequent asymptomatic infections, with preserved subsequent immunity.
Still, our results show important differences according to the degree of immunosuppression, with lower levels of seropositivity and lower antibodies titers in PLWH who are severely immunosuppressed (CD4 count < 200 cells/mm
3), irrespective of their vaccination/infection status. A lower immunogenicity of SARS-CoV-2 vaccines has been demonstrated in numerous studies for PLWH who are severely immunocompromised, regardless of their HIV viral load [
23,
24,
25,
26]. On the contrary, PLWH on cART, with undetectable plasma HIV viral loads (less than 50 copies/mL), and CD4 counts of more than 350 cells/mm
3, develop strong anti-SARS-CoV-2 humoral immune responses after vaccination, comparable in magnitude and persistence with those of people without HIV [
12,
23,
24]. More recent studies have shown lower neutralization activity against all Omicron SARS-CoV-2 variants (SARS-CoV-2 B.1, B.1.617, BA.1, and BA.2) in PLWH with severe immunosuppression [
27]. Nevertheless, the sub-variant specific neutralizing antibodies increase after a booster dose, reaching high levels in PLWH with high CD4 nadir and lower levels in PLWH with a low CD4 nadir; no substantial difference was detected in the rate of specific antibodies drop several months after variant specific booster vaccination in PLWH and HIV negative individuals [
28,
29].
A third of the PLWH enrolled in this study displayed no serologic markers for SARS-CoV-2, all had significantly decreased CD4 levels and tend to have detectable viral load under antiretroviral treatment. This might indicate either an (improbable) lack of exposure to SARS-CoV-2 infection, or previous asymptomatic infections, with a weaker immune response stimulation, followed by a complete waning of serologic markers. Data on decreased and volatile SARS-CoV-2 immune responses post-infection and post-vaccination have been also reported for patients with other causes of primary or secondary immunodeficiency (solid organ or stem cell transplantation recipients, patients on chimeric antigen receptor T cell therapy, or patients with chronic leukemias) [
30,
31,
32]. The magnitude of the immune response in PLWH can be further influenced by the infecting viral variant, as some reports have demonstrated a higher neutralization activity against the Delta SARS-CoV-2 variant, but a very decreased one for the Beta variant [
33]. We did not find any correlation between the presence and magnitude of the SARS-CoV-2 specific immune responses in PLHW and their age, gender or body mass index, although such associations were previously reported in the general population [
34], probably due to the fact that most HIV infected persons included in this study are young, without associated comorbidities.
The majority of the HIV-positive individuals enrolled in the present study had detectable SARS-CoV-2 specific cellular immune responses following re-stimulation with a BA.4/BA.5 Spike recombinant protein. The cellular immune responses were higher in immunocompetent patients (CD4 > 500 cells/mm
3). Nevertheless, SARS-CoV-2 specific cellular immune responses were also detected in two immunosuppressed PLWH, who had no serological markers of SARS-CoV-2 infection. This suggests a durable presence of cellular immunity irrespective of the individual's level of immunosuppression. Earlier studies have already demonstrated that PLWH, virally suppressed and on ART, present robust T cell response able to cross-recognize SARS-CoV-2 variants [
12,
17,
35,
36] even when the humoral immune response showed low or absent neutralization of SARS-CoV-2 [
17].
Still, in PLWH with CD4 < 200 cells/mm
3, the cellular immune response induced by mRNA vaccines was described as inferior to the one developed in PLWH without immunosuppression or in healthy controls [
24]. Follow-up studies on the persistence of the cellular immune responses in PLWH, showed a decline 6 months after vaccination, but with no significant difference compared to healthy controls. In immunocompetent individuals, studies on the long-term durability of the cellular immune responses against SARS-CoV-2 suggest that memory T-cell responses initiated by the first viral infection retain significant cross-reactivity even after a span of 2 years [
37,
38].
In our study, PLWH with hybrid immunity (acquired after vaccination and infection) display the strongest immune responses against SARS-CoV-2, similar to the data available from immunocompetent persons [
39,
40]. This is in accordance with previous reports, showing a significant increase in antibody titers and neutralizing capacity after breakthrough SARS-CoV-2 infections in PLWH [
41]. Noteworthy, in the present study, SARS-CoV-2 anti-S IgA antibodies, a surrogate marker for local immunity, were still present in most PLWH at 8 - 9 months after infection or vaccination; similar to the data reported in longitudinal follow-up studies in the general population, showing persistent adaptative systemic and local immune responses up to two years, with more durable and organized immune responses after recovery from COVID-19 [
37]. Recent studies have also demonstrated that breakthrough infections with the currently circulating Omicron variants enhance both the breadth and persistence of all types of specific immune responses, including nasal spike-specific IgA levels [
42]. Longitudinal studies have also demonstrated a rapid activation of SARS-CoV-2 specific cellular immune responses during breakthrough infections, explained both by the recall of CD4+ and CD8+ spike-specific memory cells, and
de novo T cell responses [
43]. Increased anti-SARS-CoV-2 IgA titers have been associated with better viral neutralization and protection against breakthrough infections [
44,
45]. Taken together, these results support prioritization of PLWH with severe immunosuppression for SARS-CoV-2 vaccination and variant adapted booster administration. A meta-analysis including 22 studies and 6522 PLWH reported that a second dose of SARS- CoV-2 vaccine is associated with an improved seroconversion, although with lower titres of antibodies in PLWH than in immunocompetent subjects [
46]. Another meta-analysis based on 50 studies reported increases in the seroconversion rate achieved by PLWH after each new dose of vaccine, that remains however inferior to the one acquired in the general population [
47]. In our study only 8 individuals received a vaccine booster dose, this low percentage reflects the general low level of booster uptake in Romania (< 10%) of the general population) [
48]. Administration of additional vaccine doses (up to two variant adapted boosters) in PLWH, have been associated with an expansion of the specific humoral and cellular immune responses that are maintained up to 6 months [
14]. Moreover, even in PLWH with CD4 < 200 cells/mm
3, there is an increase of the humoral response after a third dose of SARS-CoV-2 vaccine, especially in the case of heterologous vaccination, while the cellular immune response appears to remain more stable [
49]. Our results show that PLWH with preserved cellular immunity (> 500 CD4 cells/mm
3) display improved cellular immune responses, with higher numbers of IFN-γ secreting T cells after stimulation with a SARS-CoV-2 Omicron S1 protein. The absence of IL-2 secreting cells might be explained by the mild or asymptomatic SARS-CoV-2 infections in our study participants. Previous reports have signaled that IL-2 secreting cells are less frequent in those with mild disease compared to those with severe or moderate forms of COVID-19 [
50]. In addition, in vaccinated PLWH, SARS-CoV-2 specific T cells predominantly produce IFN-γ and TNF-α, while cells producing IL-2 are very rare [
51].
The main limitation of this study is the relatively small number of participants who have been evaluated for the cellular immune response. Nevertheless, the data are still valuable, as they show persistence of the cellular immunity even after asymptomatic infections, and in persons without any serological markers of SARS-CoV-2 infection.
Author Contributions
Conceptualization, S.R and C.S., methodology, S.R.; software, L.M.; validation, C.G., L.M. and A.P.; formal analysis, A.P.; investigation, C.P.; resources, C.P.; data curation, C.O.; writing—original draft preparation, C.S., writing—review and editing, C.S. and S.R.; visualization, L.M.; supervision, S.R.; project administration, C.P.; funding acquisition, S.R. All authors have read and agreed to the published version of the manuscript.