Specialist Neurology Involvement and Impact in Immune Checkpoint Inhibitor-Related Neurotoxicity: Experience in a Unified Healthcare System.

BACKGROUND

Neurological immune related adverse events (N-irAEs) following immune checkpoint inhibitor (ICI) therapy are associated with significant morbidity and mortality. The early involvement of neurological services is therefore recommended to assist diagnosis and guide management. However, the practical experience of specialist neurology involvement is poorly understood.

METHODS

A multi-centre, retrospective case note review was performed in a unified healthcare setting in the United Kingdom via predetermined proforma to investigate the involvement and impact of neurology services in this setting.

RESULTS

One hundred and nine patients with N-irAE were identified with a median time from ICI treatment to symptom onset of 52 days. Neurology service models, reasons for referral and referral rates varied by centre. Overall, eighty-seven (79.8%) patients (range 52.9-100% by centre) had neurology involvement. Neurology input was associated with younger age (median 67.2 vs. 72.8 years), anatomical location (Central > Peripheral) and severity of neurotoxicity (p < 0.001, q < 0.004). Patients with neurology involvement were more likely to undergo specialist investigations: MR imaging (p = 0.041, q = 0.043), lumbar puncture (p < 0.001, q < 0.004), and neurophysiology (p = 0.005, q = 0.007) resulting in a broader range of specific N-irAE diagnoses. Steroids were appropriately prescribed, with second line treatment (Intravenous immunoglobulins/Plasma exchange) associated with neurology involvement. At lower grades (CTCAE ≤ 2), resolution rates were similar in those with or without neurology involvement. At grades 3-4, one-third of patients with neurology involvement had resolution. In a centre with a model of early neurology involvement for all possible N-irAEs the aetiology of the neurological presentation was changed in 63.7%.

CONCLUSIONS

This study highlights the potential to improve diagnosis and treatment algorithms and therefore patient outcomes through development of uniform N-irAE models of care to support this area of growing clinical need.