Levels of antibodies against the monkeypox virus compared by HIV status and historical smallpox vaccinations: a serological study

ABSTRACT

Men who have sex with men and people living with HIV are disproportionately affected in the 2022 multi-country monkeypox epidemic. The smallpox vaccine can induce cross-reactive antibodies against the monkeypox virus (MPXV) and reduce the risk of infection. Data on antibodies against MPXV induced by historic smallpox vaccination in people with HIV are scarce. In this observational study, plasma samples were collected from people living with and without HIV in Shenzhen, China. We measured antibodies binding to two representative proteins of vaccinia virus (VACV; A27L and A33R) and homologous proteins of MPXV (A29L and A35R) using an enzyme-linked immunosorbent assay. We compared the levels of these antibodies between people living with and without HIV. Stratified analyses were performed based on the year of birth of 1981 when the smallpox vaccination was stopped in China. Plasma samples from 677 people living with HIV and 746 people without HIV were tested. A consistent pattern was identified among the four antibodies, regardless of HIV status. VACV antigen-reactive and MPXV antigen-reactive antibodies induced by historic smallpox vaccination were detectable in the people born before 1981, and antibody levels reached a nadir during or after 1981. The levels of smallpox vaccine-induced antibodies were comparable between people living with HIV and those without HIV. Our findings suggest that the antibody levels against MPXV decreased in both people living with and without HIV due to the cessation of smallpox vaccination.

KEYWORDS:

• Monkeypox
• smallpox vaccination
• antibody
• HIV
• serological study

Introduction

Monkeypox virus (MPXV) is an enveloped double-stranded DNA virus and causes a viral zoonotic disease of monkeypox (recently renamed mpox). MPXV belongs to the Orthopoxvirus genus of the Poxviridae family, which includes variola virus (VARV, also known as smallpox), vaccinia virus (VACV), camelpox virus, and cowpox virus. Monkeypox was endemic in West and Central Africa before the 2022 outbreak. Since May 2022, the first case was diagnosed in the UK, MPXV has been rapidly spreading outside the known endemic zones. On July 23, 2022, the World Health Organization (WHO) declared the monkeypox outbreak a Public Health Emergency of International Concern (PHEIC), and then determined that the event no longer constituted a PHEIC on May 11, 2023. As of June 13, 2023, a total of 87, 979 monkeypox cases, including 147 deaths, have been reported in 111 countries and territories, mostly in the Region of the Americas and the European Region [1].

The smallpox vaccine was assumed to provide cross-protection against monkeypox, as orthopoxviruses can induce cross-reactive antibodies. The current outbreak of monkeypox has been linked to a decline in immunity following the termination of routine smallpox vaccination. Evidence is that the incidence of human monkeypox in the Democratic Republic of Congo has increased dramatically 30 years after the end of the mass smallpox vaccination campaign [2]. A previous study suggested that the smallpox vaccine is about 85% effective in preventing monkeypox infection [3]. Historic smallpox vaccinations have resulted in the milder disease following MPXV infection [4,5]. However, the levels of antibodies against MPXV induced by historic smallpox vaccinations remain largely unknown among both general populations and high-risk populations.

Most of the reported cases had a history of sexual contact, who were mostly found in sexual health or STD clinics and mainly not limited to men who have sex with men (MSM). A total of 41% of these people were living with HIV and the vast majority of these persons with HIV were well-controlled [6]. Studies have suggested that people with HIV had lower immunogenicity to vaccine due to persistent inflammation or low CD4/CD8 T cell ratios [7]. Although our previous study has shown that smallpox vaccine-induced cross-reactivity to MPXV in people without HIV could last for at least 40 years, it is largely unknown whether people with HIV have lower levels of antibodies against MPXV [8]. Population surveillance of antibodies against MPXV in people with HIV is critical to understanding their susceptibility and improving the prevention and control of viral infection.

To assess levels of cross-reactive antibodies against MPXV induced by historic smallpox vaccination, we performed a serological study in Chinese people with and without HIV. Since MPXV belongs to the Orthopoxvirus genus of the Poxviridae family, which contains mature virion (MV) and enveloped virion (EV) that are infectious viral particles produced during replication. These two forms of virions carry different surface antigens. We selected two representative homologous proteins of MPXV (A29L on the MV and A35R on the EV) and VACV (A27L on the MV and A33R on the EV) respectively, to evaluate the binding activity of plasma. We also assessed factors associated with antibody seropositivity in people who were born before the cessation of smallpox vaccination.

Methods

Study design and participants

This serological study used plasma samples from Shenzhen Third People's Hospital in Shenzhen City, Guangdong Province, China. A total of 677 plasma samples from participants with HIV aged at least 18 years old were collected in the study through electronic medical record review. We also included 746 of plasma samples from the donors without HIV as controls. All these plasma samples were provided by the Biobank of Shenzhen Third People's Hospital. The Ethics Committees of the Shenzhen Third People's Hospital approved the protocol (2023-006-02).

Data collection

The demographic (age and sex) and clinical characteristics (HIV status) of participants at sample collection were extracted from the electronic medical record. The participants were divided into the following groups based on HIV status and year of birth: people with HIV born before 1981, people with HIV born during or after 1981, people without HIV born before 1981, and people without HIV born during or after 1981. Participants born before 1981 were expected to have received vaccinia-based vaccines since smallpox vaccination was the standard immunization schedule in China until 1981. The most recent HIV-specific factor at the time of initial sample collection, including the duration of HIV infection (from the first positive test of HIV to initial sample collection), antiretroviral therapy (ART), CD4 T cell count, CD8 T cell count, CD4/CD8 ratio, and HIV RNA load, were collected and analysed in people with HIV born before 1981.

Detection of VACV antigen-reactive and MPXV antigen-reactive antibodies by enzyme-linked immunosorbent assay (ELISA)

We measured antibodies reactive to two VACV antigens (A27L and A33R) and two MPXV antigens A29L (VACV A27L ortholog) and A35R (VACV A33R ortholog) in the plasma by ELISA. Plasma samples was stored at the BioBank of Shenzhen Third People's Hospital. For ELISA, high-binding 96-well plates (Corning) were coated overnight at 4°C with the four antigens respectively (antigens coating concentration: 0.5 μg/mL) in PBS. Coated plates were washed five times with PBST buffer and subsequently blocked for 1 h at RT with blocking buffer (PBS with 5% skim milk and 2% bovine albumin). Diluted plasma samples (1:100 dilution) were transferred to the antigens-coated plates and incubated at 37°C for 1 h. Plates were washed for five times with PBST and incubated for 0.5 h at 37°C with HRP-conjugated goat-anti-human IgG (ZSGB-BIO). Afterwards, plates were again washed five times with PBST and incubated for 5 min with 100 µL TMB peroxidase substrate (Sangon Biotech) after which the reaction was stopped by 2 M H₂SO₄. Absorbance was measured at 450 nm. By setting the antibody levels of the people during or after 1981 as the background, the positive cut-off value for each antibody was at three-folds of the average background level.

Focus reduction neutralization test (FRNT)

Some available plasma with high levels of antibodies by ELISA were selected to test the neutralization activity using FRNT. Vero E6 cells were seeded at 1.5 × 10⁴ cells/well into 96-well plates and used the following day. Plasma samples three-fold serial diluted in D-2 medium with 2% sterile guinea pig complement (Beijing Bersee Science and Technology Co.,Ltd) were mixed with equal volume of diluted live VACV or MPXV (isolate SZTH42, Clade IIb, GISAID accession ID: EPI_ISL_18213374) and then incubated at 37°C for 1 h [9]. Medium from 96-well plates was aspirated, and plasma-virus mixture was added (100 μL/well)，then the plates were incubated at 37°C for about 16 h. Then cells were fixed with 4% paraformaldehyde solution, permeabilized with Perm/Wash buffer (BD Biosciences) containing 0.1% Triton X-100, incubated with the HRP-conjugated anti-VACV polyclonal antibodies (Invitrogen) at 4°C overnight. The reactions were developed with KPL TrueBlue Peroxidase substrates (Seracare Life Sciences). The numbers of infected foci were calculated using an EliSpot reader (Cellular Technology Ltd). Neutralization results were presented by the maximum neutralization effect (E_max) at 1:20 dilution.

Statistical analysis

Continuous variables were expressed as means (SD) or medians (IQR). The normality of antibody levels was assessed by Shapiro–Wilk test. Difference between groups were compared using Student's t test and ANOVA for normally distributed data, and Mann–Whitney U test, Wilcoxon's Signed-Rank test, or Kruskal–Wallis test for non-normally distributed data. Categorical variables were expressed as n (%) and compared by Chi-square test or Fisher's exact test. The correlation of two continuous variables was assessed using the Pearson or Spearman method. To balance the birth decade between groups, each sample was matched on the birth decade with a ratio of 1:1 to compare antibody levels among people living with and without HIV. We used restricted cubic splines to model and visualize the nonlinear relationship between age and antibody levels. A knot was located at the 5th, 27.5th, 50th, 72.5th, and 95th percentiles. We fitted a multivariable logistic regression model to investigate whether the factors of age, sex, HIV infection, and CD4 T cell count were associated with antibody seropositivity in people born before 1981. Statistical analyses were performed, and graphs were drawn with R (version 4.3.0). A p-value less than 0.05 was considered statistically significant unless otherwise specified.

Results

Plasma samples and characteristics of the participants

We assessed antibody reaction to the four viral antigens (A27L, A33R, A29L and A35R) in 1, 423 plasma samples, including 677 samples from people living with HIV and 746 samples from people without HIV (Table 1). Of these participants, 482 (34%) were female, and the median age at initial sample collection was 34 (IQR 28–42) years old. In people living with HIV born before 1981, the median duration of HIV infection was 3 (IQR 0–7) years at initial sample collection. The median CD4 T cell count was 333 (IQR 223–488) cells per μL. Of them, 66% had received ART treatment at the initial sample collection, and 63% was viral suppression (HIV RNA < 500 IU/mL).

Table 1. Baseline characteristics of all participants.

Variable	Overall	People with HIV	People without HIV
n	1423	677	746
Age at sample collection, median (IQR), y	34 (28, 42)	34 (28, 43)	34 (29, 42)
Sex, n (%)
Male	941 (66%)	616 (91%)	325 (44%)
Female	482 (34%)	61 (9%)	421 (56%)
Birth decade, n (%)
<1961	45 (3%)	23 (3%)	22 (3%)
1961–1970	118 (8%)	59 (9%)	59 (8%)
1971–1980	264 (19%)	133 (20%)	131 (18%)
1981–1990	523 (37%)	232 (34%)	291 (39%)
>1990	473 (33%)	230 (34%)	243 (33%)

Antibody against monkeypox virus in the general population

We first analysed the levels of antibodies reactive to four viral antigens (A27L, A33R, A29L, and A35R) in 746 people without HIV. Among people born before 1981, the positive antibody response level of VACV antigens (A27L and A33R) and MPXV antigens (A29L and A35R) were significantly higher than those born during or after 1981 (Figure S1). The seropositivity rates of A27L, A29L, A33R, and A35R were 0–18% for those born before 1961, 0–37% for those born during 1961–1970, 1–15% for those born during 1971–1980, and 0–3% for those born during or after 1981, respectively (Table S1). Our study showed that the levels of antibodies binding to the four antigens ranged from the highest to A33R, followed by A35R and A27L, and the lowest to A29L in people born before 1981 (Figure S2). A comparison of ortholog antibodies suggested a high correlation between antibodies reactive to VACV antigens and antibodies reactive to MPXV antigens (Figure S3).

Antibody against monkeypox virus in people living with HIV

We further detected antibodies against the four viral antigens in 677 samples from people living with HIV. These antibodies were frequently detected in plasma from people born before 1981, but rarely in people born during or after 1981 (Table S2). Using the same cut-off value among people without HIV, the seropositivity rates of antibodies against four viral proteins were 13–35% in people living with HIV born before 1961, 3–32% for those born during 1961–1970, 4–18% for those born during 1971–1980, and 0–3% for those born during or after 1981. The difference in antibodies binding to four viral antigens (Figure S2) and a high correlation between antibodies reactive to VACV antigens and antibodies reactive to MPXV antigens (Figure S3) were also observed in people living with HIV.

Comparison of antibody profiles in people living with and without HIV

To compare levels of antibodies against MPXV infection between people with HIV and those without HIV, we performed a 1:1 matching analysis balancing for the birth decade. 674 pairs of samples were selected and were divided into four groups: people with HIV born before 1981 (n = 212), people with HIV born during or after 1981 (n = 462), people without HIV born before1981 (n = 212), and people without HIV born during or after 1981 (n = 462). In people born before 1981, the mean values of antibodies reactive to four viral antigens were comparable between people without HIV and those with HIV (Figure 1). In people born before 1981, the seropositivity rates of antibodies reactive to the four viral antigens of A27L, A29L, A33R, and A35R were 8%, 5%, 14%, and 24% among people with HIV, respectively. In comparison, the rates were 3%, 1%, 10%, and 21% among people without HIV (Figure S4). Output from spline models based on paired samples showed the dynamic patterns of the four kinds of antibodies. Antibody levels were low in people with and without HIV at younger age, and were then increased with an increase of age (Figure S5).

Figure 1. Detection of antibodies binding to the four viral antigens in people living with and without HIV, stratified by the year of birth. Participants with HIV were 1:1 matched with participants without HIV, balanced for the birth decade. Statistical analyses were performed using the Mann-Whitney U test after Bonferroni correction for multiple comparisons. Black horizontal lines indicate the mean, and the error bars indicate the standard error.

Among participants born before 1981, a total of 33 plasma samples were selected for the assessment of neutralizing activity using FRNT, including 17 from people without HIV and 16 from people with HIV. Consistently, neutralizing activity were comparable between people without HIV and those with HIV (Figure 2). There was a high correlation between anti-VACV and anti-MPXV neutralizing activity (Spearman's correlation; r = 0.7186; p < 0.0001; Figure S6). A gradual decline trend in neutralizing activity was observed with birth decades; however, no significant differences were found among different birth decade groups (Figure S7).

Figure 2. Neutralization activity in people living with and without HIV born before 1981. Neutralization activity against (A) VACV and (B) MPXV were determined in available plasma using FRNT assay. Data were means of two independent experiments. The comparison was performed using the Mann-Whitney U test. Black horizontal lines indicate the mean, and the error bars indicate the standard error. E_max = maximum neutralization effect.

The factors associated with antibody seropositivity

In people with HIV born before 1981, the elder age was positively correlated with the levels of antibodies reactive to the four viral antigens (Table S3). Duration of ART was negatively correlated with levels of A29L and A33R. Antibody levels were similar across different groups of birth decades, stratified by CD4 T cell counts (Figure S8) and HIV RNA viral load (Figure S9). We further assessed whether the factors of age, sex, HIV infection, and CD4 T cell count, were associated with the seropositivity of four antibodies among people born before 1981 (Table 2). Multivariable logistic models showed that elder ages were more likely to be seropositive of A33R and A35R. The female sex was not associated with antibody seropositivity. Compared people with HIV who had CD4 T cell counts of fewer than 500 cells per µL and those who had at least 500 cells per µL to those without HIV, only the association between people with HIV who had CD4 T cell counts of at least 500 cells per µL and the A27L positivity was significant.

Table 2. Factors associated with the seropositivity of antibody induced by historic smallpox vaccination.

	OR (95% CI)	p value*
A27L positivity
Age, per years old	1.01 (0.95–1.06)	0.7153
Female sex	1.30 (0.46–3.39)	0.6079
HIV-negative	1 (ref)
HIV-positive, CD4 T cell count ≥500 cells per µL	3.68 (1.01–12.65)	0.0391
HIV-positive, CD4 T cell count <500 cells per µL	2.11 (0.75–6.30)	0.1629
A29L positivity
Age, per years old	1.06 (0.99–1.12)	0.0712
Female sex	0.27 (0.01–1.50)	0.2207
HIV-negative	1 (ref)
HIV-positive, CD4 T cell count ≥500 cells per µL	5.31 (0.84–42.07)	0.0762
HIV-positive, CD4 T cell count <500 cells per µL	3.38 (0.77–23.48)	0.1409
A33R positivity
Age, per years old	1.07 (1.03–1.11)	0.0002
Female sex	1.24 (0.62–2.40)	0.5347
HIV-negative	1 (ref)
HIV-positive, CD4 T cell count ≥500 cells per µL	1.31 (0.44–3.41)	0.6047
HIV-positive, CD4 T cell count <500 cells per µL	1.60 (0.82–3.16)	0.1695
A35R positivity
Age, per years old	1.05 (1.02–1.08)	0.0008
Female sex	1.20 (0.71–2.01)	0.4931
HIV-negative	1 (ref)
HIV-positive, CD4 T cell count ≥500 cells per µL	1.14 (0.50–2.42)	0.7499
HIV-positive, CD4 T cell count <500 cells per µL	1.25 (0.74–2.11)	0.3961

OR = odds ratio. CI = confidence interval. CD4 T cell counts were not available in seven people with HIV born before 1981. *Multivariable logistic model included age, sex, HIV infection and CD4 T cell count.

Discussion

In the mainland of China, over 1, 400 new cases have been identified between June and September 2023 after the first imported case of MPXV infection, which brings a great challenge to the prevention and control of local transmission regarding a large number of susceptible people [10]. In China, people born during or after 1981 have not received the smallpox vaccine and lack immune protection against MPXV. Therefore, surveillance of antibodies against MPXV has important implications for infection control, contact tracing policies, and prevention strategies. Here, we measured antibodies reactive to two pairs of homologous proteins of VACV (A27L, A33R) and MPXV (A29L, A35R) among people with and without HIV in the large serological study. Our data showed that both antibodies reactive to VACV antigens and MPXV antigens could be detected in individuals receiving historic smallpox vaccination, regardless of their HIV status. Moreover, the levels of virus-specific binding and neutralizing antibodies were comparable between people with and without HIV.

The levels of antibody against MPXV induced by historic smallpox vaccination reflects population susceptibility, which is important for detecting the current global outbreak of monkeypox and preventing virus transmission. However, only a few studies have conducted serological investigations about antibodies against MPXV elicited by smallpox vaccine [11–13]. Consistent with the previous studies, antibodies against MPXV induced by historic smallpox vaccination were detectable in plasma from individuals born before 1960 [14,15]. Several studies have reported that specific antibodies can persist for decades after smallpox vaccination due to long-lived memory B cells [16,17]. Although we did detect antibodies in people born as early as the 1940s, these individuals were more likely to have been vaccinated against smallpox during or after the 1950s as inferred from vaccine coverage in China [18].

Our study also suggested that herd immunity declines over the year of birth, which is consistent with the estimated vaccine coverage of nationwide vaccination campaigns in China [18]. A recent study with 528 monkeypox cases diagnosed during the 2022 outbreak reported that only 9% of those had previously received a smallpox vaccination [6]. Our results demonstrated that the antibody levels against MPXV in participants born after the end of the universal smallpox vaccination programme significantly declined. These data indicated by the lack of cross-protection from smallpox vaccination partly explained why that males aged 18–44 years old account for the vast majority of cases in the current global outbreak [1]. Although smallpox vaccine-induced immunity could not provide full protection against MPXV infection [19,20], the clinical course appears to be milder in vaccinated individuals [21]. In addition, the serious complications and long-term effects of HIV infection were also found to be less common, and mortality was lower in patients with smallpox vaccination [22]. Given the lack of immunity in the general population born after the termination of routine smallpox vaccination and the transmission route of MPXV, the targeted strategies are needed to prevent the spread of MPXV infection to the general population.

A feature of the 2022 outbreak is that homosexual, bisexual, and other MSM account for the majority of cases [6,23]. Of monkeypox cases with known HIV infection status worldwide, 51.9% (16, 604/31, 965) were HIV-positive [1]. The HIV infection may significantly modulate host immunity and cause a diminished serologic immune response to vaccine or virus, as such shown by a reduced immunogenicity and a shorter duration after SARS-CoV-2 vaccine in people living with HIV [24]. Particularly, a previous study showed that the positive proportion for anti-A35 antibodies among people with HIV aged ≥ 42 years was significantly higher than that of controls [25], while the higher levels of anti-vaccinia neutralizing antibody responses were observed in individuals who had not been vaccinated against smallpox (aged ≤ 41) from a MSM cohort [26]. Our data suggested that the levels of smallpox vaccine-induced antibodies were comparable between people with and without HIV, suggesting that two groups may have similar susceptibility to MPXV. Notably, a small proportion of young and unvaccinated people in this study had high levels of antibodies against MPXV, which might be due to that these individuals have been previously exposed other Orthopoxvirus infection. Indeed, several recent studies have also reported the detection of antibodies against MPXV in people with HIV born after the end of smallpox vaccination [25–27]. In addition, it is unclear whether these antibody responses to MPXV is associated with disease severity in people living with HIV. Previous studies have suggested that monkeypox infection was more severe and had higher mortality for people with uncontrolled HIV [4,5,28]. Among people with monkeypox, hospitalization was more common in people with HIV than those in people without HIV [4]. Among 122 patients with monkeypox in Nigeria in 2017–2018, four of seven deaths occurred in individuals with HIV [5]. Therefore, people living with HIV, particularly those without historic smallpox vaccination or with uncontrolled HIV infection, should be given the monkeypox vaccine as a priority.

This study has several limitations. First, as shown in previous observational studies, we did not have historic vaccination records due to retrospective design [14,15]. The history of smallpox vaccination, such as whether people born before 1981 were vaccinated against smallpox, when they were vaccinated, and how many doses they received, could not be obtained. Meanwhile, laboratory data on a small part of people living with HIV born before 1981 were not available in the electronic medical record system. Second, the included participants with HIV were all seeking health care, most of whom were well-controlled after ART. Our study thus lacked data from people with poorly controlled HIV. Third, all plasma samples were collected in a single healthcare institution, which may suffer from selection biases.

In conclusion, this large-scale serological study compared the levels of antibodies reactive to MXPV antigens between people with and without HIV, demonstrating that the antibody levels against MPXV significantly decreased due to the cessation of smallpox vaccination, regardless of their HIV status. Levels of antibodies against MPXV induced by historic smallpox vaccination were comparable between people living with and without HIV, highlighting the need for disease control policies and future responses to MPXV transmission from the high-risk population to general population.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was supported by the National Key R&D Program of China [number 2021YFC2301900]; the Shenzhen Medical Special Project at SMART [number B2302052]; the Guangdong Science and Technology Plan Project-Construction of high-level biosafety laboratories [number 2021B1212030010]; the National Science Fund for Distinguished Young Scholars [number 82025022]; the Science and Technology Research Projects of Shenzhen [number JCYJ20190809115617365; ZDSYS20210623091810030; JSGG20220606141001003]; and the Shenzhen High-level Hospital Construction Fund [number 23250G1002].