Postpandemic rebound of adeno‐associated virus type 2 (AAV2) infections temporally associated with an outbreak of unexplained severe acute hepatitis in children in the United Kingdom

Over 1000 cases of unexplained severe acute hepatitis in children have been reported to date worldwide. An association with adeno‐associated virus type 2 (AAV2) infection, a human parvovirus, prompted us to investigate the epidemiology of AAV in the United Kingdom. Three hundred pediatric respiratory samples collected before (April 03, 2009–April 03, 2013) and during (April 03, 2022) the COVID‐19 pandemic were obtained. Wastewater samples were collected from 50 locations in London (August 2021–March 2022). Samples were tested for AAV using real‐time polymerase chain reaction followed by sequencing. Selected adenovirus (AdV)‐positive samples were also sequenced. The detection frequency of AAV2 was a sevenfold higher in 2022 samples compared with 2009–2013 samples (10% vs. 1.4%) and highest in AdV‐positive samples compared with negatives (10/37, 27% vs. 5/94, 5.3%, respectively). AAV2‐positive samples displayed high genetic diversity. AAV2 sequences were either very low or absent in wastewater collected in 2021 but increased in January 2022 and peaked in March 2022. AAV2 was detected in children in association with AdV of species C, with a highest frequency in 2022. Our findings are consistent with the expansion of the population of children unexposed to AAV2, leading to greater spread of the virus once distancing restrictions were lifted.

herpesviruses (e.g., human herpesvirus 6) or adenoviruses (AdVs), or through the effects of cellular stress from physical or chemical carcinogens. 2 In the absence of helper functions, AAV persists as a latent infection as an episome or less frequently through integration into the host's genome. 3,4 The replication, tropism, and normally minimal pathogenicity of AAV are all well characterized since its extensive use as a vector for gene therapy.
AAVs are genetically diverse, with at least 13 serologically distinct types. 5,6 Human infection with AAV is common, with up to 80% of individuals having neutralizing antibodies to one or more serotype, the majority being directed against AAV2, although with considerable variation in prevalence depending on age, sex, and geographical region. 7 Infections are generally believed to be apathogenic, evoking only a minor inflammatory response, although they have been associated with mild hepatitis when used in gene therapy. [8][9][10] However, AAV2 has recently been implicated in the etiology of the 2022 outbreak of unexplained severe acute hepatitis in children (UHC). [11][12][13][14] Infections with human AdV, specifically AdV type F-41, were initially suspected as it was detected in the majority of cases and at rates statistically significantly higher than control groups, 15 although generally with low virus loads. However, two studies in the United Kingdom and one in the United States have found a much more specific association with AAV2 infection. [11][12][13][14] In the United Kingdom studies, high levels of AAV2 DNA were detected in all nine cases requiring liver transplant, and also in 15 out of 16 children with unexplained hepatitis who recovered without needing a transplant. Both UK studies documented very low AAV2 infection frequencies in non-UHC control cohorts, including AdV-F41-positive children without hepatitis. AAV2 DNA was present at much lower levels in these control groups than in UHC cases. In both of these studies, the majority of UHC cases also tested positive for AdV (predominantly AdV-F41) or for HHV-6. These studies suggested that AAV2 was the primary cause of UHC, albeit requiring co-infection with a helper virus such as AdVF-41 and/or HHV-6 for replication. In addition, Ho et al. 11 found an association between disease and the class II human leukocyte antigen (HLA) DRB1*04:01 allele, present in 89% of cases, compared with a background frequency of 15.6% in the general Scottish population. A US study observed similar results, AAV2 was detected in 13 out of the 14 UHC cases studied (93%) whilst it was only detected in 4 of 113 controls (3.5%). 14 Using wastewater samples and clinical samples obtained from children under the age of 5 years, we investigated more broadly whether there had been a systematic change in AAV2 incidence in early 2022 compared with previous years. We chose to study samples from children below the age of 5 years as most cases of UHC were reported in this group, and that is the age when AAV infections are most likely acquired. 16,17 2 | MATERIALS AND METHODS

| Pediatric respiratory samples
Anonymized pediatric respiratory samples were obtained from the Edinburgh Specialist Virology Centre sample archive. Approval was obtained from the Lothian Regional Ethics Committee (approval number 08/S11/02/2) to retain the collected information for epidemiological purposes but to make the information anonymous to protect patient confidentiality. The stored data included age band, partial postcode, any recorded symptoms or clinical information, referral source, month of sample collection, and the results of other virological tests for each sample. For the samples from 2022, the majority were nose and throat swabs (117) with the remainder consisting of nasopharyngeal secretions (14), nose swabs (10), throat swabs (6), and bronchoalveolar lavage (BAL) (2). For the control cohort, the samples predominantly comprised nasopharyngeal aspirates or swabs (130/141) and were routinely screened for the following respiratory viruses by previously described polymerase chain reaction (PCR) assays 18,19 : AdV, influenza A and B viruses, parainfluenza virus type 1 (PIV-1), PIV-2, PIV-3, and respiratory syncytial virus (RSV).
Nucleic acid extraction was carried out from 200 μL of patient sample using the ThermoFisher MagMAX Viral/Pathogen nucleic acid isolation kits with MagMAX magnetic beads using the automated ThermoFisher Kingfisher Apex magnetic particle processor. Extracted nucleic acids were eluted into 30 µL of DNase/RNase-free water.

| Wastewater sample collection processing
Wastewater sample collection, processing, and nucleic acid  Table S1). Potential PCR inhibitors were reduced by a 1.8X NGS Magbind Total-Pure bead (VWR) cleanup, followed by LunaScript (NEB) reverse transcription. 22

| AAV and AdV types and genetic diversity
All samples that were AAV-positive by quantitative polymerase chain reaction (qPCR) were confirmed by nested PCR targeting the REP region and typed by Sanger sequencing; all were AAV2 (Figure 2A AdV types in individuals co-infected with AAV2 were assigned by phylogenetic analysis of hexon gene sequences ( Figure 2B). Of the samples positive for both AAV2 and AdV, which the AdV hexon gene could be amplified and sequenced, six were AdV-C1, and four were AdV-C2. For a further four AdV-positive but AAV2-negative samples, three were serotype AdV-C1 and one was AdV-C5.  Lack of pre-existing immunity through reduced exposure of communicable diseases during the COVID pandemic will be more evident in younger children as they will have spent a greater proportion of their lives under social restrictions. Supporting this, the age range with the highest incidence of AAV2 detection was indeed in the 1-3-year-old age group (Figure 1), a distribution that matches that of children with jaundice and unexplained hepatitis, who were primarily <5 years of age ( Figure 1B). 11,13,14 As proposed by Morfopoulou et al. 13  The association of AAV2 infection with UHC stimulated us to understand the normal patterns of circulation of AAV2, a subject which has been little studied due to the general belief that AAV2 is apathogenic. Our study demonstrated a temporal association between a peak in AAV2 cases and the peak in unexplained hepatitis cases. This data supports its recently proposed etiological link 11,13,14 and the potential of AAV2 to cause severe disease in certain circumstances. These include host genetic background and greater postpandemic host susceptibility. 11 The frequent descriptions of dose-dependent hepatitis in recipients of AAV-vectored vaccines is further consistent with its aetological role in UHC 34