Psychosis associated with expansions in the C9orf72 gene: the influence of a 10 base pair gene deletion

Introduction

We previously reported1 that expansions in the C9orf72 gene, identified by Southern blotting or using a repeat primed PCR assay,2 may go undetected using an alternative, standard PCR assay.3 We attributed this discrepancy to a 10 base pair deletion adjacent to the expansion which was presumed to interfere with genotyping.1 To determine the possible clinical and pathological relevance of the deletion, we compared patients in whom the C9orf72 expansion was detected using the Renton3 assay (C9 reference group) with those carrying an expansion not detected by this method (deletion group).
Methods

The criteria for inclusion in the study were: (A) patients had been investigated and diagnosed with dementia within a single, specialist dementia clinic, (B) they had donated DNA as part of an ethically approved study into the molecular genetics of dementia and (C) expansions in the C9orf72 gene were identified using one or more standard procedures.1–3 An additional patient, in whom pathological examination showed p62 positive neuronal inclusions immunoreactive for the dipeptide repeat proteins, poly-GA, poly-GP and poly-GR, was also included since such histological findings are considered pathognomonic of the presence of an expansion in C9orf72.4 Patients were divided into two groups (deletion vs C9 reference) on the basis of whether the expansion was detected by the Renton assay.3 Clinical, neurological and demographic characteristics were examined and coded from patient records without reference to genetic grouping to avoid bias. Neuropathological characterisation of autopsied cases was also independent of genetic analysis. Group comparisons used χ2 or, where appropriate, Fisher's exact tests for categorical data. t Tests were used to compare onset age and illness duration.
Results

Fifty patients were identified, 14 in the deletion group and 36 in the C9 reference group. In deletion cases, the C9orf72 expansion was detected using the assay of De Jesus-Hernandez et al,2 and confirmed in 4/4 cases by Southern blotting. In most respects, the deletion and C9 reference groups were indistinguishable (table 1). Apart from two patients in whom neurological signs were confined to bulbar palsy, all patients with motor neuron disease (MND) showed both bulbar and limb involvement and upper and lower motor neuron signs consistent with amyotrophic lateral sclerosis. Despite the overall similarities, there was a highly significant group difference in the frequency of psychosis (table 1), psychotic symptoms being common in the C9 reference group but very rare in the deletion group. Psychosis was characterised by persecutory or somatoform delusions and accompanied by bizarre and irrational behaviours as described previously.