Objective To determine the immunogenicity of the third dose of CoronaVac vaccine in a large population of patients with autoimmune rheumatic diseases (ARD) and the factors associated with impaired response.
Methods Adult patients with ARD and age-balanced/sex-balanced controls (control group, CG) previously vaccinated with two doses of CoronaVac received the third dose at D210 (6 months after the second dose). The presence of anti-SARS-CoV-2 S1/S2 IgG and neutralising antibodies (NAb) was evaluated previously to vaccination (D210) and 30 days later (D240). Patients with controlled disease suspended mycophenolate mofetil (MMF) for 7 days or methotrexate (MTX) for 2 weekly doses after vaccination.
Results ARD (n=597) and CG (n=199) had comparable age (p=0.943). Anti-S1/S2 IgG seropositivity rates significantly increased from D210 (60%) to D240 (93%) (p<0.0001) in patients with ARD. NAb positivity also increased: 38% (D210) vs 81.4% (D240) (p<0.0001). The same pattern was observed for CG, with significantly higher frequencies for both parameters at D240 (p<0.05). Multivariate logistic regression analyses in the ARD group revealed that older age (OR=0.98, 95% CI 0.96 to 1.0, p=0.024), vasculitis diagnosis (OR=0.24, 95% CI 0.11 to 0.53, p<0.001), prednisone ≥5 mg/day (OR=0.46, 95% CI 0.27 to 0.77, p=0.003), MMF (OR=0.30, 95% CI 0.15 to 0.61, p<0.001) and biologics (OR=0.27, 95% CI 0.16 to 0.46, p<0.001) were associated with reduced anti-S1/S2 IgG positivity. Similar analyses demonstrated that prednisone ≥5 mg/day (OR=0.63, 95% CI 0.44 to 0.90, p=0.011), abatacept (OR=0.39, 95% CI 0.20 to 0.74, p=0.004), belimumab (OR=0.29, 95% CI 0.13 to 0.67, p=0.004) and rituximab (OR=0.11, 95% CI 0.04 to 0.30, p<0.001) were negatively associated with NAb positivity. Further evaluation of COVID-19 seronegative ARD at D210 demonstrated prominent increases in positivity rates at D240 for anti-S1/S2 IgG (80.5%) and NAb (59.1%) (p<0.0001).
Conclusions We provide novel data on a robust response to the third dose of CoronaVac in patients with ARD, even in those with prevaccination COVID-19 seronegative status. Drugs implicated in reducing immunogenicity after the regular two-dose regimen were associated with non-responsiveness after the third dose, except for MTX.
Trial registration number NCT04754698.
- autoimmune diseases
- biological therapy
Data availability statement
All data relevant to the study are included in the article.
This article is made freely available for personal use in accordance with BMJ’s website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.https://bmj.com/coronavirus/usage
Statistics from Altmetric.com
What is already known about this subject?
The waning of immunity elicited by vaccines was reported to be associated with breakthrough cases in different countries, driven predominantly by the Delta variant and now the threat of Omicron variant of SARS-CoV-2.
Patients with autoimmune rheumatic diseases (ARD) are at high risk of severe COVID-19 and are known to have reduced primary vaccination response.
There is evidence on the efficacy of a third dose in increasing humoral response and protective effect in the general population.
What does this study add?
The third dose of COVID-19 vaccine results in a robust immunogenicity response for patients with ARD overall and for those who were COVID-19 seronegative at 6 months post primary full vaccination.
How might this impact on clinical practice or future developments?
The third dose of anti-SARS-CoV-2 vaccine should be strongly recommended for patients with ARD 6 months after primary vaccination as an excellent strategy to improve waning of vaccine-induced COVID-19 immunogenicity over time.
The COVID-19 pandemic is still a global problem and mass vaccination has been crucial for the waning of the epidemic worldwide. Until now, almost one-third of the Brazilian population has been vaccinated with CoronaVac. There is real-world evidence on the high effectiveness of CoronaVac in reducing hospitalisations and deaths related to SARS-CoV-2 infection in 10.2 million people in Chile.1
However, even in countries with high vaccination rates, breakthrough cases in both vaccinated and unvaccinated persons are increasingly being reported,2–4 driven predominantly by the SARS-CoV-2 Delta variant5–7 and more recently by the Omicron variant. The Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention (CDC) recommended a third dose of COVID-19 vaccine to high-risk groups, including immunocompromised individuals, to address potential waning immunity against the SARS-CoV-2 variants, with acceptable safety profile.8–10
Immunosuppressed patients with autoimmune rheumatic diseases (ARD) are generally under increased risk of severe COVID-19.11 12 We have previously demonstrated that two doses of CoronaVac elicited moderate humoral response in patients with naïve ARD, with a parallel major decrease in incident COVID-19 cases post immunisation.10 A more robust response was observed in individuals with ARD pre-exposed to COVID-19, with a high plateau of response after a single CoronaVac dose.13 Neutralising antibodies’ (NAb) response dynamics in patients who have recovered from COVID-19 may vary greatly.14 We further demonstrated a substantial decline of anti-SARS-CoV-2 antibodies in patients with ARD 6 months after a two-dose schedule of CoronaVac.15
There is increasing evidence on the efficacy of a third dose of vaccine in enhancing protective effect.16–19 However, data on CoronaVac are limited. A study in healthy adults demonstrated a strong humoral booster following an additional dose administered 8 months after the primary schedule, indicating an efficient recalling of SARS-CoV-2-specific immune memory.20 The efficacy of a third dose of messenger RNA (mRNA) vaccine was also recently reported in the general population in Israel, with more than 90% reduction in COVID-19-related hospitalisations and deaths.21 In subjects ≥60 years, a third dose of BNT162b2 reduced severe infection rate by a factor of almost 20.19 Furthermore, a third dose of the viral vector ChAdOx1 nCoV-19 vaccine could boost antibody and T cell responses in healthy volunteers.22 Among immunocompromised populations, a significant proportion of patients (30%–50%) with inadequate response to two mRNA vaccine doses seroconverted after an additional dose,23–30 and insufficient response was mainly associated with higher degree of immunosuppression.23 24
Regarding ARD population, there is one case series reporting that a third dose of mRNA vaccine induced seroconversion in almost 90% of 17 patients with rheumatoid arthritis with minimal response to primary vaccination. Most of them discontinued disease modifying anti-rheumatic drugs (DMARD) temporarily to receive the third dose17 and more data are necessary.
We therefore performed a prospective analysis of the immunogenicity and safety of a third CoronaVac dose, administered 6 months after the standard two-dose homologous schedule, in a large ARD population compared with an age-balanced and sex-balanced control group (CG). Possible factors associated with lack of humoral immune response after the third dose were secondarily assessed.
Study design and population
We conducted a phase 4 prospective longitudinal study (CoronavRheum) at a large academic hospital in Sao Paulo, Brazil. All participants signed the written informed consent. Patients with ARD were diagnosed according to the international classification criteria for each disease31–40 and were regularly followed at the outpatient rheumatology clinics. Subsequently, we invited subjects without ARD or immunosuppressive therapy as the CG. All subjects were ≥18 years old. CG was sex-balanced and age-balanced with patients with ARD (±5 years) at entry (1 control to 3 patients). All participants had previously received two doses of the inactivated vaccine CoronaVac (batch #20200412; Sinovac Life Sciences, Beijing, China) 28 days apart (first dose (D0): 9–17 February 2021; second dose (D28): 9–17 March 2021) and were recruited to receive the third dose from 13 September to 18 September 2021 (D210: 6 months later). The electronic charts of all patients under mycophenolate mofetil (MMF) or methotrexate (MTX) were reviewed for disease activity, 10 days before vaccination. Those with low disease activity/inactive disease at last visit (up to 2 months) were interviewed at D210 by a physician to confirm clinical status and to guide medication withdrawal after vaccination (1 week for MMF and 2 weekly doses for MTX). The exclusion criteria were third dose vaccination with any other SARS-CoV-2 vaccine, prior anaphylactic events to vaccines, immunisation with live virus in the last 4 weeks or inactivated vaccine in the last 2 weeks, Guillain-Barré syndrome, decompensated heart failure, demyelinating disease, COVID-19-related symptoms, acute febrile illness, or hospitalisation at vaccination day (figure 1). All participants had their first blood sample collected at D210 and the second sample collected at D240 (1 month after the third dose; from 14 October to 4 November 2021). Patients or the public were not involved in the design, or conduct, or reporting or dissemination plans of the study.
Primary and secondary outcomes
The primary outcome was humoral immunogenicity assessed by the presence of anti-SARS-CoV-2 S1/S2 IgG at D240 in all participants. The secondary outcomes were factor increase (FI) in anti-S1/S2 geometric mean titre (GMT) from D210 to D240, presence of NAb, NAb activity, and the influence of demographic data, ARD diagnosis and current therapy on anti-SARS-CoV-2 seropositivity at D240.
Serological assay consisted of total anti-SARS-CoV-2 S1/S2 IgG (chemiluminescent immunoassay by indirect ELISA; ETI-MAX 3000 equipment, LIAISON SARS-CoV-2 S1/S2 IgG Kit, DiaSorin, Italy) and circulating NAb against SARS-CoV-2 using the SARS-CoV-2 sVNT Kit (GenScript, Piscataway, New Jersey, USA), following the manufacturer’s instructions. Samples with 15.0 AU/mL or more for total IgG and with 30% or more inhibition in the neutralising assay were considered seropositive according to the manufacturer.41 Quantitative results were reported, attributing the value of 1.9 AU/mL (half of the lower limit of quantification 3.8 AU/mL) to undetectable levels (<3.8 AU/mL) of IgG. NAb activity was calculated as median (IQR) only considering positive samples at D210 and D240.
Data and statistical analysis
Data were presented as number (percentage) for categorical variables and as mean±SD or median (IQR) for continuous variables. Comparisons were performed by χ2 or Fisher’s exact tests, as appropriate, for categorical variables, and by Student’s t-test or Mann-Whitney test for continuous variables. Seropositivity rates of anti-S1/S2 IgG and NAb were presented as number (percentage) and were compared between groups (ARD and CG) and between timepoints (D210 vs D240) using repeated measures analysis of variance with two factors followed by Bonferroni’s multiple comparisons in Napierian logarithm-transformed data for IgG. IgG titres were expressed as geometric mean with 95% CI. Multivariate logistic regression analyses were performed using as dependent variables seroconversion (SC)/NAb positivity at D240 and as independent variables those with p<0.2 in each univariate analysis. Statistical significance was defined as p<0.05. Most of the statistical analyses were performed using Statistical Package for the Social Sciences V.20.0.
A total of 956 patients with ARD and 451 controls were recruited in this protocol. After applying the exclusion criteria and random sampling (3 ARD to 1 CG), the final study groups consisted of 597 patients with ARD and 199 controls who collected blood sample and received the third dose of CoronaVac (D210) and returned after 30 days for blood collection (D240) (figure 1). Patients with ARD and CG were comparable with regard to median age and female sex (p>0.05) (table 1).
As part of standard of care, based on American College of Rheumatology guidance for COVID-19 vaccination,42 269 patients under MTX and 87 patients under MMF on low disease activity/inactivity were instructed to withhold these drugs for 2 weeks and 1 week, respectively.
Anti-S1/S2 IgG and NAb seropositivity rates in patients with ARD and in CG before (D210) and after (D240) the third dose of vaccine are presented in table 2. From D210 to D240, significant increase in anti-S1/S2 IgG and NAb positivity rates was observed for both ARD and CG (p<0.0001).
Anti-S1/S2 IgG GMT increased significantly from D210 to D240 in ARD (25.3 AU/mL vs 140.5 AU/mL, p<0.001) and in CG (47.9 AU/mL vs 253.8 AU/mL, p<0.001) (table 3). Expressive increments in NAb activity were also observed after the third dose of vaccine for both groups (p<0.0010).
The factors associated with IgG and NAb positivity after the third dose of vaccine (D240) in the ARD group are presented in table 4. The number of patients included in table 4 (n=875) comprised the total number of patients with ARD who were initially recruited and attended all study visits (received the third dose of CoronaVac at D210 and returned for blood collection at D240), before the random sampling (3 ARD to 1 CG) for immunogenicity analysis. The frequencies of patients with systemic vasculitis, prednisone, immunosuppressive drugs, MMF and biologic drug use, particularly abatacept, belimumab and rituximab, were significantly lower in IgG seropositive patients (p<0.05). NAb-positive patients at D240 presented lower frequencies of female sex, rheumatoid arthritis diagnosis, prednisone, immunosuppressive drugs and biologic drugs use, especially abatacept, belimumab and rituximab (p<0.05). On the other hand, spondyloarthritis, systemic sclerosis diagnosis and secukinumab use were associated with NAb positivity (p<0.05).
Multiple logistic regression analysis using IgG positivity at D240 as the dependent variable revealed that older age (OR=0.98, 95% CI 0.96 to 1.0, p=0.024), vasculitis (OR=0.24, 95% CI 0.11 to 0.53, p<0.001), prednisone ≥5 mg/day (OR=0.46, 95% CI 0.27 to 0.77, p=0.003), MMF use (OR=0.30, 95% CI 0.15 to 0.61, p<0.001) and biologic drug use (OR=0.27, 95% CI 0.16 to 0.46, p<0.001) were independently associated with anti-S1/S2 IgG response after the third dose (D240) in patients with ARD. For NAb analysis, multiple logistic regression revealed that prednisone ≥5 mg/day (OR=0.63, 95% CI 0.44 to 0.90, p=0.011), abatacept (OR=0.39, 95% CI 0.20 to 0.74, p=0.004), belimumab (OR=0.29, 95% CI 0.13 to 0.67, p=0.004) and rituximab use (OR=0.11, 95% CI 0.04 to 0.30, p<0.001) were independently associated with NAb negativity after the third vaccine dose.
Further analysis of immune response to the third vaccine dose was performed considering COVID-19 seronegative (negative IgG and NAb at D210; n=215) and seropositive (positive IgG and/or NAb at D210; n=316) patients with ARD and the age-balanced and sex-balanced controls for each ARD group (figure 1). COVID-19 seronegative ARD and controls at D210 demonstrated prominent increase in IgG and NAb positivity rates as well as in GMT at D240 (p<0.0001). At D210, 120 patients with ARD were positive only for anti-S1/S2 IgG, 8 patients only for NAb and 187 patients for both. In COVID-19 seropositive ARD, significant GMT increase was observed from D210 to D240 (p<0.001), with 30.4% of the patients reaching the ceiling of the assay (>400 AU/mL). NAb activity also increased significantly at D240 (p<0.001) (table 5). Comparison of the FI-GMT from D210 to D240 between COVID-19 seronegative and seropositive ARD demonstrated a higher increment in the former group (11.3 (95% CI 9.5 to 13.4) vs 3.9 (95% CI 3.4 to 4.4), p<0.001).
Regarding safety, there were no differences between ARD and CG for any of the reported adverse events. The most frequently reported adverse events were local pain (23.8% vs 19.4%, p=0.238) and headache (13.1% vs 11.7%, p=0.712). No serious adverse event was reported.
Among patients with ARD on MTX treatment, those who withdrew the drug for 2 weeks (n=269) after third vaccine dose presented higher frequency of positive anti-SARS-CoV-2 S1/S2 IgG as well as higher GMT (183.5 (95% CI 142.7 to 236.0) vs 101.7 (95% CI 76.5 to 135.2), p=0.002) compared with those who maintained medication (55.8% vs 44.2%, p=0.029), with no statistically significant differences in NAb positivity (54.8% vs 45.7%, p=0.682) and NAb activity (83.9% (58.8%–95.9%) vs 79.0% (56.1%–91.3%), p=0.186). Comparison of patients who withheld MMF (n=109) and those who maintained the drug after the third dose showed no statistically significant differences in IgG (64.9% vs 35.1%, p=0.894) and NAb positivity (65.5% vs 34.5%, p=0.869).
This study provides novel evidence of a substantial increase in immune response with an additional dose administered 6 months after two doses of SARS-CoV-2 inactivated vaccine, in a large prospective controlled cohort of patients with ARD. We identified a distinct pattern of response to the third dose characterised by an expressive seroconversion of the COVID-19 seronegative ARD of 81% for IgG and 59% for NAb, with a parallel 11-fold increment of IgG GMT. For the COVID-19 seropositive ARD, the magnitude of IgG booster response reached almost a fourfold rise.
Immunocompromised individuals should receive a third dose of COVID-19 vaccine as proposed by the CDC43 and WHO. This recommendation is in line with the 6-month immunogenicity waning observed for mRNA vaccines in healthcare workers, including a small proportion of individuals under immunosuppression,44 and in a large ARD population immunised with CoronaVac.14 Of note, no parallel increase in incident cases in the 6 months post vaccination was observed in this latter population, in contrast to the reported upsurge of cases in vaccinees with mRNA vaccine in Israel.21
The overall analysis of patients with ARD revealed that the additional dose resulted in global 93% IgG positivity and 33% rise. This finding is superior to the 47%–68% rate of positive antibodies reported for transplant recipients after mRNA additional dose25 28 29 45 and may be explained in part by the distinct intensity of immunosuppression, with a high frequency of multiple drug therapy in transplanted population. In addition, the temporary discontinuation of MTX and MMF in more than half of patients under these therapies might also have contributed to improved immune response, as reported recently by our group in rheumatoid arthritis population.46 The higher rates of anti-SARS-CoV-2 seropositivity in patients who withheld MMF herein were statistically not significant, probably due to a limited power for this analysis.
The specific analysis of immunogenicity to the third dose in non-responsive ARD 6 months after the second dose revealed a remarkable increase in IgG levels, suggesting that the three-dose strategy seems to be effective in recalling SARS-CoV-2 immune memory. The seroconversion rate of non-responsive transplant recipients was inferior, ranging from 25% to 49%, after a third dose,24 29 47 reinforcing response differences among immunocompromised subgroups.48 Alternatively, this observation may be related to other factors known to influence vaccine response.44 In this regard, we have identified that prednisone, immunosuppressive drugs and biological therapy, mainly abatacept, belimumab and rituximab, were associated with decreased antibody production after the additional dose. Of note, we demonstrated herein that prednisone doses from 5 mg/day considerably reduced vaccine immune response. These same factors negatively influenced response to the primary immunisation with CoronaVac in anti-SARS-CoV-2-negative patients with ARD.10
COVID-19 seropositive patients with ARD had a distinct pattern of response to the third dose, with a lower but still significant increase in IgG levels, with a fourfold increase, and one-third reached the assay ceiling value (>400 AU/mL). The same was reported for solid organ transplant recipients and dialysis patients.23 25 27 29 There are scarce data on NAb after third shot in immunocompromised individuals.28 45 49 50 The overall NAb positivity in ARD reached 81% after the third dose with high median activity. These figures are similar to those reported for transplant recipients with mRNA vaccine.28 45 We further demonstrated different immunogenicity in primary non-responsive and responsive patients with ARD. The former group achieved 59% positivity and a moderate activity, whereas COVID-19 seropositive patients with ARD had a robust response for both NAb positivity and activity. Robert et al49 observed 66% response to the third dose in nine haemodialysis patients. This finding suggests that the additional dose is efficient in boosting the immune response in both ARD groups. In the context of the reported immunity waning observed for all SARS-CoV-2 vaccines at 6 months for the general population and patients with ARD,15 44 our findings strengthen the relevance of the third dose in ARD.
The large number of patients with ARD and the inclusion of an age-balanced and sex-balanced CG are relevant strengths of the present study and provided a sizeable sample to evaluate humoral response to additional vaccine dose and the impact of the drugs. In fact, the few studies focusing on the third dose for immunocompromised individuals were small-sized17 50 or did not have a control group.24 25 28–30 44 46
This study has some limitations, such as the non-assessment of cellular immunity, which may play an important role of protection during SARS-CoV-2 infection. However, this limitation was partially mitigated by NAb assessed herein, which is highly predictive of immune protection.51 The assessment of the effectiveness of the third dose was hampered by the short-term follow-up and the very low incidence of COVID-19 cases during the study period. The global analysis of the influence of some drugs that are used only for specific ARD in the immunogenicity evaluation performed herein probably underestimated the effect of these medications. However, these therapies remained as relevant independent factors that negatively impacted immunogenicity in the multivariate evaluation. The lack of disease activity assessment, especially for those who withdrew MTX or MMF, is also an important limitation.
In summary, to our knowledge this is the first demonstration of a robust response to the third dose of an inactivated vaccine in patients with ARD, with greater benefit for those who are COVID-19 seronegative before the third dose. We further identified drugs as unfavourable for response to additional vaccine doses. These findings may be generalised to other platform vaccines, and heterologous fourth-dose boosting could be an alternative strategy for the minority of persistently non-responsive patients with ARD.
Data availability statement
All data relevant to the study are included in the article.
Patient consent for publication
This study involves human participants. The protocol was approved by the national and institutional ethical committee (Conep, Comissão Nacional De Ética Em Pesquisa; reference number: CAAE: 42566621.0.0000.0068). Participants gave informed consent to participate in the study before taking part.
We thank the volunteers for participating in all inperson visits and for handling the biological material, as well as those responsible for the follow-up of all participants. We also thank the contribution of the undergraduate and postgraduate students for collecting epidemiological and clinical data, as well as the Central Laboratory Division, Registry Division, Security Division, IT Division, Superintendence, Pharmacy Division and Vaccination Center for their technical support. Instituto Butantan supplied the study product and had no other role in the trial.
Handling editor Josef S Smolen
CAS and EB contributed equally.
Contributors NEA, CAS, LdVKK, ACM-R, CGSS, EFNY, SGP and EB conceived and designed the study, participated in data collection and analysis, supervised the clinical data management, and wrote and revised the manuscript. EB is responsible for the overall content as the guarantor. EB and PR organised and supervised the blood collection and vaccination. SGP, VAOM and VLNB supervised the serum processing, SARS-CoV-2-specific antibody ELISA/neutralisation assays and SARS-CoV-2 RT-PCR. NEA, LdVKK, ACM-R, CGSS, EFNY, SGP, PTR, RMRP, SKS, PDS-B, DA, ASRH, RF, FHCS, LKNG, APLA, JCBdM, MRUL, VAOM, LB, CTR, LPS, IMB, RLPM, GGMB, LA, AMCS, CAS and EB collected epidemiological and clinical data and assisted with the identification of SARS-CoV-2 infection and follow-up of patients. PR organised and supervised the vaccination protocol. All authors helped to edit the manuscript.
Funding This study was sponsored by grants from Fundação de Amparo à Pesquisa do Estado de São Paulo (#2015/03756-4 to CAS, SGP, NEA and EB; #2019/17272-0 to LdVKK; #2020/09367-8; #2018/09937-9 to VAOM), Conselho Nacional de Desenvolvimento Científico e Tecnológico (#304984/2020-5 to CAS; #305556/2017-7 to RMRP; #303379/2018-9 to SKS; #305242/2019-9 to EB), B3-Bolsa de Valores do Brasil and Chamada Instituto Todos pela Saúde (ITPS 01/2021, proposta C1313 to EB, CAS, NEA and SGP).
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.