Thrombotic microangiopathy (TMA) presents with microangiopathic hemolytic anemia, thrombocytopenia and acute kidney injury. The authors report a systematic analysis linking biallelic pathogenic variants in the RNA exosome component EXOSC3 to a severe, early-onset, often infection-triggered renal TMA in children with pontocerebellar hypoplasia 1b (PCH1b). Among 34 children with EXOSC3 variants, 13 developed TMA. Clinical and laboratory data indicate that these episodes commonly follow infection and that terminal complement blockade with eculizumab is not effective in this entity.
Patients with genetically confirmed PCH1b due to rare EXOSC3 variants (n = 34) were identified from pan-European registries and the UK national atypical HUS service and screened for TMA. Where available, samples were tested for known causes of complement-mediated atypical HUS, including genetic causes and acquired factor H autoantibodies. To investigate mechanisms, a tamoxifen-inducible whole-body Exosc3 conditional knockout mouse (Exosc3KO) was generated. Recombination was confirmed by GFP expression in splenocytes. Haematology, biochemistry, cell-cycle, and apoptosis assays were performed on mouse tissues using standard flow cytometry and laboratory platforms described in the methods.
TMA was identified in 13 of 34 children (8 male, 5 female). The median age of first TMA presentation was 6 months (range 2–20 months). Across the cohort there were 17 TMA episodes, and every reported episode was triggered within 10 days of a preceding infection; respiratory infections were most common. Diagnostic criteria included the triad of microangiopathic hemolytic anemia, thrombocytopenia and acute kidney injury. Hypertension occurred in 10 of 13 patients and five required dialysis at presentation. One post-mortem kidney examination confirmed widespread platelet- and fibrin-rich thrombi within glomerular capillaries and arterioles consistent with TMA.
No laboratory evidence of complement dysregulation was detected in the tested patients, supporting a non–complement-driven mechanism in these cases.
Management strategies reported included supportive care, fresh frozen plasma (FFP, n = 7) and eculizumab (n = 5). Eculizumab was administered at initial presentation in four cases and for relapse in one additional case. Three children received prolonged eculizumab therapy; among those, one child had no response and two experienced relapses while on treatment. Two patients progressed to end-stage kidney disease requiring long-term renal replacement. Eleven children with TMA died during follow-up; the mean age at death was 10 months (range 2–43 months). Eight of 13 patients showed evidence of renal recovery after their TMA episodes.
Adult Exosc3fl mice treated with tamoxifen to induce recombination (Exosc3KO) developed rapid clinical decline with weight loss and gastrointestinal symptoms, necessitating euthanasia at a median of eight days. Post-mortem haematology showed aplastic anemia with reduced haemoglobin, leukopenia, thrombocytopenia, markedly reduced marrow cell counts and low reticulocyte percentage compared with controls. Bone marrow analysis demonstrated cell-cycle arrest at G0/G1 and increased early and late apoptosis.
Histology revealed hypocellularity of bone marrow, thymic atrophy and pathology of the large intestine consistent with loss of rapidly dividing cells. In the acute interval examined in this model there was no evidence of renal TMA or overt kidney pathology.
The RNA exosome complex (EXOSC1–9) trims precursor ribosomal RNAs during ribosome biogenesis. Pathogenic EXOSC3 variants impair RNA processing and ribosome assembly, leading to nucleolar stress, upregulation of cell-cycle regulators, cell-cycle arrest and apoptosis in actively dividing tissues. The authors link this mechanism to vascular injury in the kidney and propose a concept of a “vascular ribosomopathy,” highlighting mechanistic overlap with Shiga-toxin–mediated HUS in which ribosomal inactivation is central.
The absence of renal pathology in the acute Exosc3KO mouse model may reflect model differences: complete knockout in adult mice causes rapid global failure of proliferative compartments and death before a kidney phenotype emerges, whereas human disease arises from hypomorphic missense alleles, often with an infectious trigger in early life.
This report is based on a cohort assembled from registries and national referral services; some data items were not available for all patients. A single post-mortem kidney sample provided histological confirmation of TMA; additional renal pathology data were limited. The Exosc3KO model used complete conditional disruption in adult mice and did not reproduce the renal TMA seen in humans, likely because human disease typically involves hypomorphic alleles and early-life infection triggers. The authors note that patient-specific or hypomorphic animal models and inclusion of infectious stressors will be required to better recapitulate renal TMA and to delineate the precise endothelial mechanisms that lead to thrombosis. Further work is needed to clarify whether targeted therapies other than complement inhibition could modify disease course.
Biallelic pathogenic variants in EXOSC3 cause a severe, early-onset, infection-associated renal TMA in children with PCH1b. The phenotype is not driven by terminal complement dysregulation and is resistant to eculizumab in this series. Clinicians should consider TMA surveillance in PCH1b and genetic testing for EXOSC3 in eculizumab-resistant pediatric TMA, particularly when neurodevelopmental disease is present.