New Onset and Exacerbation of Autoimmune Bullous Dermatosis Following COVID-19 Vaccination: A Systematic Review

Background: Cases of autoimmune bullous dermatosis (AIBD) have been reported following COVID-19 vaccination. Objective: We aimed to provide an overview of clinical characteristics, treatments, and outcomes of AIBDs following COVID-19 vaccination. Methods: We conducted a systematic review and searched the Embase, Cochrane Library, and Medline databases from their inception to 27 March 2024. We included all studies reporting ≥ 1 patient who developed new-onset AIBD or experienced flare of AIBD following at least one dose of any COVID-19 vaccine. Results: We included 98 studies with 229 patients in the new-onset group and 216 in the flare group. Among the new-onset cases, bullous pemphigoid (BP) was the most frequently reported subtype. Notably, mRNA vaccines were commonly associated with the development of AIBD. Regarding the flare group, pemphigus was the most frequently reported subtype, with the mRNA vaccines being the predominant vaccine type. The onset of AIBD ranged from 1 to 123 days post-vaccination, with most patients displaying favorable outcomes and showing improvement or resolution from 1 week to 8 months after treatment initiation. Conclusions: Both new-onset AIBD and exacerbation of pre-existing AIBD may occur following COVID-19 vaccination. Healthcare practitioners should be alert, and post-vaccination monitoring may be essential.


Methods
This systematic review was registered with PROSPERO (CRD42023390478), and it was performed in accordance with the updated Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [20][21][22].Comprehensive searches were performed in the Embase, Cochrane Library, and Medline databases from their inception to 27 March 2024 using relevant terms, including 'COVID-19', 'vaccine', 'autoimmune bullous dermatosis', 'vesiculobullous skin diseases', 'pemphigus', 'pemphigus vulgaris', 'pemphigus foliaceus', 'pemphigus erythematosus', 'bullous pemphigoid', 'mucous membrane pemphigoid', and 'linear IgA bullous dermatosis'.These terms were applied as free text, medical subject headings (MeSH in PubMed and Emtree in Embase), and abbreviations in the literature search.Boolean operators were used to combine keywords, and a primary search strategy was developed without language or publication data limitations (Table S1).Additionally, the reference lists of all identified articles were screened to identify further relevant studies.
We included studies reporting at least one patient who developed new-onset AIBD or experienced an exacerbation of AIBD following administration of at least one dose of any COVID-19 vaccine.Exacerbation was defined as the presence of increased body surface area involvement, the presence of vesiculobullous lesions or skin erythema, subjective worsening reported by the patient, worsening described in physical examination findings, or clinician assessment or plan indicating exacerbation, rebound, or worsening of AIBD compared to previous examination.Review articles, conference abstracts, and in vitro or animal model studies were excluded.Two experienced authors (Wu and Wang) independently conducted the literature search, data extraction, and quality assessments.Any discrepancies between the reviewers were resolved by a third author (Huang).The quality of case reports and series was assessed using the appraisal tool developed by Murad et al. [23], while observational studies were evaluated using the National Institute of Health quality assessment tool (Tables S2 and S3) [24].
Data extraction was performed independently by two authors (Wu and Wang) and included the following information from the included studies: author, year of publication, country, demographic information of patients (age and sex), blister sites, COVID-19 vaccination details (vaccine type and dose), onset time, classification of cases as new-onsets or exacerbations, AIBD subtype, other potential triggers, pathology examinations (Hematoxylin and Eosin stains and immunofluorescence study), enzyme-linked immunoassay (ELISA) results (such as BP180, BP 230, desmoglein [dsg] 1, and desmoglein 3), prior and post-exacerbation treatments, outcomes, and reactions to subsequent COVID-19 vaccination.The patient groups were further categorized based on the occurrence of new AIBD onset or exacerbation of AIBD, and all patients were classified according to AIBD subtypes.

Literature Search
As shown in Figure 1, 333 studies were identified after searching three major databases and performing a manual search of the reference lists of identified studies.We excluded 91 studies as duplicates, and 75 studies were excluded for being unrelated to the study question after assessing the title or abstract.The full texts of the remaining 167 studies were reviewed, and 98 studies were identified as meeting the inclusion criteria for qualitative Vaccines 2024, 12, 465 3 of 26 synthesis.A total of 74 studies reporting new-onset AIBD, 15 studies reporting exacerbation of AIBD, and 9 studies reporting both new onset and exacerbation of AIBD were included in this study (Tables 1 and 2).The quality assessments of case reports and series consistently received scores ranging from five to seven according to the methodology proposed by Murad et al. [23].For observational studies, all of the assessments were rated as 'fair' using the National Institute of Health quality assessment tool [24].

Literature Search
As shown in Figure 1, 333 studies were identified after searching three major databases and performing a manual search of the reference lists of identified studies.We excluded 91 studies as duplicates, and 75 studies were excluded for being unrelated to the study question after assessing the title or abstract.The full texts of the remaining 167 studies were reviewed, and 98 studies were identified as meeting the inclusion criteria for qualitative synthesis.A total of 74 studies reporting new-onset AIBD, 15 studies reporting exacerbation of AIBD, and 9 studies reporting both new onset and exacerbation of AIBD were included in this study (Tables 1 and 2).The quality assessments of case reports and series consistently received scores ranging from five to seven according to the methodology proposed by Murad et al. [23].For observational studies, all of the assessments were rated as 'fair' using the National Institute of Health quality assessment tool [24].

Patient Characteristics
Detailed patient information is presented in Tables 1 and 2. The characteristics of the included studies are summarized in Table 3.The new-onset group comprised 229 patients, mostly from America, with ages ranging from 11 to 97 years.Although most studies did not report patients' sex, a slight male predominance was noted among those that The most frequently encountered diagnosis in the group was BP in 174 patients, followed by PV in 23 and PF in 16.The flare group included 216 patients, with ages ranging from 20 to 88 years, who primarily had pemphigus (specific subtype unspecified).Most patients were from Asia (44%) and America (41%).Similarly to the new-onset group, most studies did not provide information on patients' sex, but a slight male predominance existed among those that did.

Vaccine Type, Vaccine Dose, and Time to AIBD Onset Following Vaccination
In the new-onset group, 55% of patients received the BioNTech/Pfizer vaccine, followed by the Moderna vaccine (16%) and the Oxford-AstraZeneca vaccine (13%).However, it is noteworthy that the vaccine type was not reported for a large number of patients.Most cases of new-onset AIBD occurred after the second (39%) or first vaccine dose (34%), while 15% of AIBD patients experienced onset following both doses.The onset times varied widely, ranging from 1 to 123 days after vaccination.
In the flare group, most patients were administered the BioNTech/Pfizer vaccine (56%), followed by the Sinovac vaccine (18%) and the Moderna vaccine (16%).Flares were most frequently reported after the third vaccine dose (63%), followed by the first dose (24%) and the second dose (10%).The onset of AIBD symptoms ranged from 1 day to 92 days following vaccination.

Other Potential Non-Vaccine Triggers
In the new-onset group, most studies did not provide information on other potential non-vaccine triggers.However, some BP patients had pre-existing neurological or psychiatric disorders, such as dementia, depression, or Alzheimer's disease, which are known to be associated with the development of BP [28,114,115].Additionally, dipeptidyl peptidase 4 (DPP-4) inhibitors, a well-established risk factor for BP [116], were used by some patients [34,44,45,48].In the majority of cases, patients denied any new medication use.
In the flare group, the information regarding other potential triggers was unavailable in most studies.Nevertheless, two patients had a history of COVID-19 infection prior to receiving the COVID-19 vaccines, and subsequently experienced a BP eruption [99,103].
In the flare group, the predominant treatment approach involved topical or systemic corticosteroids supplemented by immunomodulators, such as doxycycline, nicotinamide, methotrexate, azathioprine, or mycophenolate mofetil, in refractory cases [6,18,30,105,106].Additional corticosteroid therapy was used in most patients experiencing a flare of AIBD, with further immunosuppressants utilized for treatment-resistant cases [14,18,32,106].Rituximab was administered in six cases, resulting in four cases experiencing disease improvement; one case died 15 days after the administration of COVID-19 vaccination due to sepsis, and one case had ongoing treatment and no final outcome was reported [51,99,105,[108][109][110].The majority of cases showed improvement (65%) or resolution (22%) after treatment, with resolution times ranging from 1 to 10 weeks.Only three of the reported cases (20%) experienced a similar flare following their initial COVID-19 vaccination and exhibited disease exacerbation after the second dose [6,18,30,105].

Discussion
In this systematic review, we have compiled all available reports of new-onset AIBD or AIBD flares following COVID-19 vaccination.Our analysis included 98 studies, encompassing 229 patients in the new-onset group and 216 patients in the flare group.Among the new-onset cases, BP was the most frequently reported subtype, while pemphigus was the most commonly reported subtype in the flare group.As we know, clinical relapse is commonly seen in pemphigus, with a relapse rate as high as 82% [118].The chronic and relapsing features of pemphigus may contribute to the larger number of flare cases relative to BP. Notably, both new onset and exacerbation of AIBDs were frequently observed following the administration of mRNA vaccines.However, we should recognize that mRNA vaccines were the most frequently administered vaccine worldwide.Onset time varied widely among both new-onset and flare groups, ranging from 1 to 123 days.Most patients achieved favorable outcomes, with improvement or resolution occurring within 1 week to 8 months after treatment initiation.
The potential association between vaccination and AIBD has been investigated in the previous research [119].Various vaccines, including influenza, tetanus and diphtheria, hepatitis B, herpes zoster, and quadrivalent human papillomavirus, have been reported to be associated with AIBD development [120].With the substantial increase in COVID-19 vaccinations, the link between newly developed vaccines and AIBD has been reexamined.The theory of molecular mimicry between specific basement membrane proteins and the spike protein of SARS-CoV-2 has been proposed as a potential cause [121].Additionally, mRNA vaccines are suggested to activate pro-inflammatory pathways by interacting with toll-like receptors, potentially leading to increased production of interleukin (IL) -4, IL-17, interferon-γ, and tumor necrosis factor-α cytokines [71,79,122].Because autoreactive T cells and the dysregulation of T helper (Th)1 and Th2 responses play a crucial role in both pemphigus and pemphigoid [123], the vaccine trigger and cytokine modulation may promote an imbalance between Th2 responses against cutaneous antigens, fostering the generation of autoreactive B cells and contributing to AIBD development [122].Vaccineinduced inflammation may also disrupt the basement membrane, leading to the production of anti-basement membrane antibodies [121].Furthermore, human leukocyte antigen (HLA) molecules, including alleles HLA-DQB1*0503 and HLA-DRB1*0402 in pemphigus, as well as HLA-DQB1*0301 in pemphigoid, may represent key predisposing factors for druginduced AIBDs [124].However, none of the included cases underwent HLA examinations, necessitating further investigations.
On the contrary, Birabaharan et al. conducted a cohort study involving over 1.5 million individuals who received mRNA COVID-19 vaccinations, which revealed no difference in the risk of new-onset BP within a 6-month period between vaccinated patients and those who remained unvaccinated [33].Another investigation by Kasperkiewicz et al. demonstrated that circulating anti-SARS-CoV-2 antibodies did not cross-react with the main AIBD autoantigens, including dsg 1, dsg 3, envoplakin, BP180, BP230, and type VII collagen [125].This perspective is consistent with the findings of previous systematic reviews, which posited that the hypothesized causal relationship is likely to be a relatively rare occurrence [126,127].In our study, we not only included a substantially larger sample size compared to previous studies, but we also employed the Naranjo score to investigate causality.Patients with severe or extensive AIBD are usually advised against re-exposure to the same vaccine.However, in our study, 23 patients who experienced new onset or exacerbation of AIBD were re-exposed to the same vaccine, leading to recurrence.This implicates COVID-19 vaccines as the likely causative agents, supported by the high Naranjo rating score of 7. Our research provides evidence suggesting a potential association between COVID-19 vaccination and the development of AIBD to some extent, as indicated by the short onset interval and the absence of other triggers in most cases.These findings are in accordance with the previous literature, underscoring that mRNA vaccines were the most commonly reported vaccine type in both new onset and exacerbation of AIBD cases, followed by inactivated and viral-vectored vaccines [127].
It is worth noting that some studies reported potential non-vaccine triggers, such as neurological or psychiatric disorders, use of DPP-4 inhibitor, polypharmacy, or a history of COVID-19 infection [28,34,43,44,46,49,50,92,99].The etiology and pathogenesis of AIBD remain largely elusive.However, the occurrence of exacerbation of AIBD has been reported in association with specific triggering factors, including medications, physical stimuli, infections, and organ transplantations [128].We outlined these cases and assigned lower scores on the Naranjo score, which consequently decreased the overall rating.Only 13% of the new-onset AIBD patients and 8% of AIBD flare cases were rated as probable according to the Naranjo score.Nevertheless, it is essential to acknowledge that most studies did not report such triggers, limiting the calculation of the Naranjo score.Given that the existing data predominantly consist of anecdotal, single-case reports with a low level of evidence, real-world, population-based studies are warranted to elucidate a definitive link between COVID-19 vaccinations and risk of AIBD.However, this should not dissuade the current vaccination recommendations for patients with AIBD, given the favorable risk-benefit ratio.
Our study has certain limitations.Firstly, most of the included studies were case reports, case series, and retrospective observational studies from database collections.Some studies lacked comprehensive documentation of patients' clinical conditions, while others were deficient in critical information, including vaccine dosage, additional triggers, laboratory findings, treatment modalities, and disease outcomes.Secondly, not all studies presented results of skin biopsies, immunofluorescence studies, or ELISA tests, thereby raising questions about the accuracy of disease diagnoses in some cases.Thirdly, essential parameters for assessing disease severity in AIBD patients, such as the bullous pemphigoid disease area index (BPDAI), the pemphigus area and activity score (PAAS), and the percentage of body surface area affected, were not reported among all studies.These parameters are pivotal for evaluating disease severity before vaccination, after vaccination, and following treatment.Fourthly, only a limited number of cases provided information regarding whether patients received subsequent vaccine doses, and the duration of follow-up was relatively short.In our analysis, most patients in the new-onset and flare groups showed improvement or resolution.However, given the chronic and relapsing nature of AIBD, future long-term follow-up studies are imperative to establish a stronger evidence base, and ongoing monitoring is essential for these patients [129].

Conclusions
In conclusion, both new-onset AIBD and exacerbation of pre-existing AIBD may occur following COVID-19 vaccination.Healthcare practitioners should raise concerns for AIBD when administering COVID-19 vaccines, and post-vaccination monitoring may be essential.Current evidence continues to favor COVID-19 vaccination in individuals with AIBD, owing to its significant protective benefits against SARS-CoV-2.More studies are imperative to elucidate the underlying mechanisms of the association between COVID-19 vaccines and the development of AIBD.

Figure 1 .
Figure 1.PRISMA flowchart of the selection of studies.Figure 1. PRISMA flowchart of the selection of studies.

Figure 1 .
Figure 1.PRISMA flowchart of the selection of studies.Figure 1. PRISMA flowchart of the selection of studies.

Table 1 .
Characteristics of the included studies reporting new onset of autoimmune bullous dermatosis.

Table 2 .
Characteristics of the included studies reporting exacerbation of autoimmune bullous dermatosis.

Table 3 .
Summary of characteristics of the included studies.