Humoral immune responses to COVID-19 vaccination in people living with HIV receiving suppressive antiretroviral therapy

Abstract

Background. Our understanding of COVID-19 vaccine immune responses in people living with HIV (PLWH) remains incomplete. Methods. We measured circulating antibodies against the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein and ACE2 displacement activities after one and two COVID-19 vaccine doses in 100 adult PLWH and 152 controls. Results. All PLWH were receiving suppressive antiretroviral therapy, with median CD4+ T-cell counts of 710 (IQR 525-935) cells/mm3. Median nadir CD4+ T-cell counts were 280 (IQR 120-490) cells/mm3, and ranged as low as 9 cells/mm3. After adjustment for sociodemographic, health and vaccine-related variables, HIV infection was associated with 0.2 log10 lower anti-RBD antibody concentrations (p=0.03) and ~7% lower ACE2 displacement activity (p=0.037) after one vaccine dose. Following two vaccine doses however, the association between HIV and weaker responses no longer remained. Rather, older age, a higher burden of chronic health conditions, and having received two ChAdOx1 doses (as opposed to a heterologous or dual mRNA vaccine regimen) were the most significant correlates of weaker humoral responses. No significant association was observed between the most recent or nadir CD4+ T-cell counts and responses to COVID-19 vaccination in PLWH following two vaccine doses. Conclusions. These results suggest that PLWH whose viral loads are well-controlled on antiretroviral therapy and whose CD4+ T-cell counts are in a healthy range will generally not require a third COVID-19 vaccine dose as part of their initial immunization series, though other factors such as older age, co-morbidities, type of initial vaccine regimen and durability of vaccine responses will influence when this group may benefit from additional doses. Further studies of PLWH who are not receiving antiretroviral treatment and/or who have low CD4+ T-cell counts are needed.

Systematic review and meta-analysis of COVID-19 vaccines safety, tolerability, and efficacy among HIV-infected patients

“…All the medians’ CD4 count was≥ 200 cells/µl (Supplementary material: Table 1). Sixteen studies reported the HIV viral load, among them, six studies included PLWH with viral load < 50 copies/ml [16, 45, 49, 51, 53, 54] and nine other studies included PLWH with viral load 58.1 to 95% suppressed [44, 46, 52, 55, 56, 58, 60, 61, 63, 65] ( Supplementary material: Table 1 ).…”

“…The selected articles reported data from United States of America (n = 4) [48–50, 55], China (n=5) [52, 54, 59, 61, 64], South Africa (n = 3) [45, 46, 57], United Kingdom (n = 2) [16, 53], Canada (n=2) [ 47, 51], Italia (n=1) [58], Brazil (n =1) [56], Germany (n =1) [60], Israel (n = 1) [44], Sweden (n=1) [64], Taiwan (n=1) [65] and Russia (n = 1) [62] ( Figure 2 ). Ten studies used randomized control designs [16, 45, 46, 49, 50, 53, 54, 64, 64, 65], eight studies used prospective cohort designs [44, 47, 48, 51, 52, 56–58], two retrospective cohort studies [62, 63], two cross sectional studies [59, 61] and one case control study [55] ( Supplementary material: Table 1 ).…”

“…The selected articles reported data from United States of America (n = 4) [48–50, 55], China (n=5) [52, 54, 59, 61, 64], South Africa (n = 3) [45, 46, 57], United Kingdom (n = 2) [16, 53], Canada (n=2) [ 47, 51], Italia (n=1) [58], Brazil (n =1) [56], Germany (n =1) [60], Israel (n = 1) [44], Sweden (n=1) [64], Taiwan (n=1) [65] and Russia (n = 1) [62] ( Figure 2 ). Ten studies used randomized control designs [16, 45, 46, 49, 50, 53, 54, 64, 64, 65], eight studies used prospective cohort designs [44, 47, 48, 51, 52, 56–58], two retrospective cohort studies [62, 63], two cross sectional studies [59, 61] and one case control study [55] ( Supplementary material: Table 1 ). Eleven studies included mRNA vaccines among which BNT162b2 mRNA, Vaccine, mRNA-1273 (100 g) administered intramuscularly (IM) as a series of two doses (0.5 μmL each), given 28 days apart [44, 47, 48, 50, 51, 53, 55, 58, 60, 63, 64], six studies included inactivated vaccines μg/0.5 mL) or Sinovac (0.5 μml) given intramuscularly [52, 54, 56, 59, 61, 64], seven studies included viral vector vaccines among which ChAdOx1 nCoV-19 with two doses (5 × 1010 vp) was given 4-6 weeks apart, Ad26.COV2.S [16, 45, 49, 51, 57, 62, 63] and two study included sub unit g (sub unit) of recombinant spike protein with 50μg of Matrix-M1 adjuvant) administered two intramuscular injections, 21 days apart and two standard doses of 15 mcg MVC-COV1901, administered 28 days apart via IM injection [46, 65...…”

“…Regarding the risk of bias of non-randomized studies, sixteen studies were assessed with the ROBIN-I bias tool assessment [26]. Among them, seven were rated as low-moderate for the overall ROBIN-I [44, 51, 52, 55, 57, 58, 60], four studies were low–serious [53, 54, 56, 62], one study was moderate-serious [61], another study was Serious–Critical [48] and two studies were critical [47, 59] ( Supplementary material: Table 3 ).…”

“…Figure 5 reported IgG antibodies level after the first dose of SARS-CoV-2 vaccines. The pooled result including eight studies [44, 45, 51–54, 56, 60] has shown that the mean difference (MD) IgG antibody level (BAU/ml) did not differ between 643 PLWH and 762 HIV negatives after the first dose of SARS-CoV-2 vaccines MD-0.02 (95%CI: -0.09- 0.05, P = 0.58). The test of heterogeneity between studies was high (I 2 = 83%).…”