Incidence and Outcome of Neurologic Immune-Related Adverse Events Associated With Immune Checkpoint Inhibitors in Patients With Melanoma

Abstract

Background and objectives: Neurologic immune-related adverse events (n-irAEs) reportedly occur in up to 8% of patients treated with immune checkpoint inhibitors (ICIs) of all age groups. We investigated the association between age and n-irAEs in patients treated with ICIs and examined the effect of n-irAEs on survival outcomes in a large cohort of patients with melanoma.

Methods: We conducted a retrospective analysis of patients with advanced melanoma treated with ICIs at Ella Institute for Immuno-oncology and Melanoma between January 1, 2015, and April 20, 2022. The outcomes of interest were defined as the investigation of age-related frequency and severity of n-irAEs, the need for ICI interruption, the treatment required for n-irAE management, the safety of ICI reintroduction, and n-irAE's effect on survival.

Results: ICI was administered to 937 patients. At least one irAE occurred in 73.5% (n = 689) of them. Among the study population, 8% (n = 76) developed a n-irAE, with a median age of 66 years in female and 68 years in male patients at onset. The median follow-up after n-irAE was 1,147 days (IQR: 1,091.5 range: 3,938). Fewer irAEs occurred in patients older than 70 years (median: 3 events, p = 0.04, CI 2.5-4.7) while specifically colitis and pneumonitis were more common in the 18-60 age group (p = 0.03, 95% CI 0.8-0.38, p = 0.009, 95% CI 0.06-0.2). Grade ≥ 3 toxicity was seen in 35.5% of patients across age groups. The median time from ICI administration to n-irAE development was 48 days across age groups. Common n-irAE phenotypes were myositis (44.7%), encephalitis (10.5%), and neuropathy (10.5%). N-irAE required hospitalization in 40% of patients and steroids treatment in 46% with a median of 4 days from n-irAE diagnosis to steroids treatment initiation. Nine patients needed second-line immunosuppressive treatment. Rechallenge did not cause additional n-irAE in 71% of patients. Developing n-irAE (HR = 0.4, 95% CI 0.32-0.77) or any irAE (HR = 0.7195% CI 0.56-0.88) was associated with longer survival.

Discussion: N-irAEs are a relatively common complication of ICIs (8% of our cohort). Older age was not associated with its development or severity, in contrast with non-n-irAEs which occurred less frequently in the elderly population. Rechallenge did not result in life-threatening AEs. Development of any irAEs was associated with longer survival; this association was stronger with n-irAEs.

Figure 1. Pharmacological Protocols Used in Our Cohort for Patient Who Had Neurological Toxicities

ICIs likely causative protocols leading to neurologic toxicity (A) and protocols likely leading to second neurologic adverse event(B). ICI = immune checkpoint inhibitor.

Figure 2. Survival Curve Among Patients Who Developed Neurologic or Nonneurologic Toxicities After Treatment With Immune Check Point Inhibitors

Survival analysis showed significantly higher overall survival rate in patients who developed n-irAE (yellow line) when compared with patients with nonneurological toxicities (blue and gray lines), p = 0.002. Those who did not develop any adverse events (blue line) had the lowest survival rate. Cox regression analysis comparing the 3 groups found the survival hazard ratio was 0.497 in patients who developed neurologic AEs (highest survival probability), whereas for patient with nonneurological AEs (gray line) it was 0.741 (higher survival probability than those in group A who developed no irAE), p < 0.01.

Figure 3. Survival Curve Among Patients Who Developed Any Toxicity vs No Toxicity

Hazard ratio for survival in the groups developing any irAE after ICI was 0.7 while the inverted hazard ratio for those who did not develop any irAE was 1.407, p < 0.01. ICI = immune checkpoint inhibitor.