- Top of page
- What this paper adds
- Molecular analyses
Aim To describe a spectrum of intracerebral large artery disease in Aicardi–Goutières syndrome (AGS) associated with mutations in the AGS5 gene SAMHD1.
Method We used clinical and radiological description and molecular analysis.
Results Five individuals (three males, two females) were identified as having biallelic mutations in SAMHD1 and a cerebral arteriopathy in association with peripheral vessel involvement resulting in chilblains and ischaemic ulceration. The cerebral vasculopathy was primarily occlusive in three patients (with terminal carotid occlusion and basal collaterals reminiscent of moyamoya syndrome) and aneurysmal in two. Three of the five patients experienced intracerebral haemorrhage, which was fatal in two individuals. Post-mortem examination of one patient suggested that the arteriopathy was inflammatory in origin.
Interpretation Mutations in SAMHD1 are associated with a cerebral vasculopathy which is likely to have an inflammatory aetiology. A similar disease has not been observed in patients with mutations in AGS1 to AGS4, suggesting a particular role for SAMHD1 in vascular homeostasis. Our report raises important questions about the management of patients with mutations in SAMHD1.
Aicardi–Goutières syndrome (AGS) is a genetically determined encephalopathy, which, in its classical presentation, shows phenotypic overlap with the sequelae of congenital infection.1 It is a genetically heterogeneous disorder caused by mutations in any of the genes encoding the 3′ to 5′ exonuclease TREX1 (AGS1),2 the three non-allelic components of the RNASEH2 endonuclease complex (AGS2, 3, and 4),3 and the uncharacterized SAMHD1 protein (AGS5).4 With identification of the genetic basis of AGS, the clinical phenotype has expanded to include a milder disease with later onset in many cases.5,6 Additionally, mutation screening of AGS-related genes has shown an association with a wider spectrum of apparently distinct neurological and non-neurological phenotypes, including retinal vasculopathy with cerebral leukodystrophy,7 familial chilblain lupus,8 systemic lupus erythematosus,9 and non-specific inflammatory arthropathy.10
Neuropathological data,11 the sometimes patchy white-matter changes seen on brain imaging, and the observation of a vasculitis12 on biopsy of the chilblain lesions experienced by 40% of patients, suggest that an inflammatory disturbance of vascular homeostasis may be central to the pathogenesis of AGS. Here we report five patients with a clinical diagnosis of AGS who developed intracranial large artery disease that was primarily occlusive in three patients, and associated with aneurysm formation in two further patients. All five affected individuals were shown to harbour biallelic mutations in SAMHD1.
- Top of page
- What this paper adds
- Molecular analyses
Here we describe five patients, four with confirmed mutations and a fifth likely to harbour a homozyguous intragenic deletion in the AGS5 gene, SAMHD1, all of whom had intracerebral large artery disease in association with a cutaneous vasculopathy. Three patients showed multiple and extensive stenoses of intracerebral large arteries with, in two patients, evidence of basal collaterals consistent with a radiological diagnosis of moyamoya syndrome. Two further patients showed intracranial aneurysms without evidence of occlusive disease. One child, for whom imaging was not available, presented at 7 months of age with acute hemiparesis, reportedly due to thrombosis of the right internal carotid artery. Three patients experienced intracerebral haemorrhage before the age of 14 years, which was fatal in two.
We note a previous report of a patient demonstrating an occlusive vasculopathy with moyamoya collaterals and peripheral vascular occlusive disease, presenting at 13 years of age with acute hemiplegia.13 This male, resident in Australia, had a history of chilblains of the hands and feet from the age of 6 months which were more severe in cold weather, and loss of tissue from the tips of his ears. Carotid angiography revealed narrowing of the supra-clinoid portions of both internal carotid arteries, occlusion of the anterior and middle cerebral vessels just distal to their origins from the internal carotid arteries, and a vast network of fine collaterals in the region of the basal ganglia. Of note, arteriography of the left forearm and hand showed occlusion of the ulnar artery and very slow circulation in the hand, fingers, and thumb. We speculate that this patient’s disease was due to mutations in SAMHD1.
Two of our patients were considered to have moyamoya syndrome on radiological grounds, with occlusive disease of the large vessels of the circle of Willis, and the compensatory development of a collateral circulation. Although most cases of moyamoya disease have no definable explanation,14 rare monogenic associations are recognized, including sickle cell disease, neurofibromatosis type 1, homocystinuria, microcephalic osteodysplastic dwarfism, ACTA2-related disease, and Schimke immunosseous dysplasia. Our findings suggest that disease due to mutations in SAMHD1 should now be added to this list. The spectrum of associated conditions reinforces the importance of appreciating that moyamoya remains a radiological label, and is likely to represent the end result of more than one disease process.
All five of the patients we describe experienced significant chilblain-like lesions, a feature reported in association with mutations in all five known AGS-associated genes. Two patients demonstrated a leukocytoclastic vasculitis on biopsy, characterized by the presence of fibrinoid necrosis, red-cell extravasation, and neutrophil infiltrate. This well-defined histopathological designation is considered indicative of a true vasculitis. Taken together with the reported findings of an intracranial vasculitis on post mortem of patient 3, it would seem likely that the cerebral arteriopathy we describe has an inflammatory basis. Akin to ACTA2-related vascular disease,15 a non-inflammatory vasculopathy, the skin of patients with SAMHD1 mutations may reflect a more diffuse involvement of multiple arterial beds, predisposing to both occlusions and aneurysm formation.
We previously described glaucoma in three children with AGS.16 In this report, one patient and a sibling of another (patient 5), showed congenital glaucoma, possibly caused by an acquired abnormality of the microvasculature of the trabecular network at the angle of the anterior chamber.
Previous, limited, neuropathological data have suggested that AGS may represent a primary microangiopathy, a contention supported by the pattern of changes seen on brain imaging, and the frequently reported finding of a vasculitis on biopsy of the chilblain lesions experienced by 40% of patients with AGS. These observations suggest that an inflammatory disturbance of vascular homeostasis may be central to the pathogenesis of AGS. Interestingly, heterozygous mutations in the AGS1 gene TREX1 cause the adult-onset disorder retinal vasculopathy with cerebral leukodystrophy, previously known as hereditary endotheliopathy, retinopathy, and nephropathy.7 The neurological manifestations of retinal vasculopathy with cerebral leukodystrophy include transient ischaemic attacks and strokes, and the mass lesions observed on brain imaging demonstrate a coagulative necrosis secondary to an obliterative vasculopathy with minimal inflammatory infiltrate on biopsy. Additionally, the ophthalmological findings seen in retinal vasculopathy with cerebral leukodystrophy, which include capillary dropouts with prominent juxta-foveolar capillary obliteration and telangiectasias, are also consistent with a primary vasculopathy.17 Although the function of SAMHD1 remains unknown, evidence suggests that, like TREX1, it plays an important role in innate immunity and inflammation.18
Clinical and laboratory data have previously indicated that, after an initial subacute encephalopathic period, the disease process in AGS is most frequently non-progressive.5 However, in the patients described here there was definite evidence of ongoing pathology, and our report raises important questions about the management of patients with mutations in SAMHD1. In practical terms both the occlusive and aneurysmal arteriopathies described here might be amenable to treatment (revascularization for the former, and coiling or clipping for the latter). Moreover, the likely inflammatory basis of the arteriopathy suggests that immunosuppression may have a role in management. A key question is whether inflammatory disease is active at the time of clinical presentation, or whether the arterial abnormalities observed represent the end result of a now-quiescent inflammatory process. Given the potential for intervention, it could be argued that individuals with mutations in SAMHD1 should be actively screened for intracranial arteriopathy. Cognisance would need to be taken of the overall clinical presentation when making decisions about screening and intervention.
Interestingly, we have never observed large artery disease in association with mutations in AGS1-4, perhaps indicating a particular role for SAMHD1 in blood vessel integrity and homeostasis. More generally, following our recent description of progressive arthropathy with distal joint contractures and painful mouth ulcers in association with bialleic SAMHD1 mutations,10 this report suggests the need to consider mutation analysis of SAMHD1 in overlapping inflammatory phenotypes.