Neurologic Toxicity with Immune Checkpoint Inhibitors

By Rajiv S. Magge, MD

Synopsis: In a nationwide, multicenter, retrospective study of patients with neurologic immune-related adverse events related to immune checkpoint inhibitors, more than half of patients developed a chronic condition associated with a higher rate of severe neurologic disability and mortality.

Source: Rossi S, Farina A, Malvaso A, et al. Clinical course of neurologic adverse events associated with immune checkpoint inhibitors. Neurol Neuroimmunol Neuroinflamm. 2024;11(6):e200314.

Immune checkpoint inhibitors (ICIs) have revolutionized cancer treatment over the past decade, contributing to substantially improved survival outcomes in several systemic malignancies. These agents take advantage of “immune checkpoints” on normal T cells — specific membrane receptors that, when bound, diminish activation and help prevent autoimmunity. Unfortunately, cancer cells can highjack these pathways and trigger them to avoid destruction by the immune system.

ICIs include antibodies that block cytotoxic T-lymphocyte antigen 4 (CTLA-4), programmed death protein (PD-1), or the PD-1 ligand (PD-L1), and prevent immune system deactivation and enhance antitumor responses with cancer cell death. However, as they release the “brake” on the body’s immune system, they also may cause unwanted nonspecific inflammation and a variety of toxicities.

Neurologic immune-related adverse events (n-irAEs) are relatively rare, occurring in only 1% to 3% of patients treated with ICIs. The most frequent side effects include headache, dizziness, and peripheral sensory neuropathy, but more serious presentations, such as Guillain-Barré syndrome, myasthenia gravis, myositis, encephalitis, meningitis, and transverse myelitis, have been described. These neurologic sequelae can be quite debilitating and potentially life-threatening and, as uniquely illustrated by this study, may remain chronic and refractory issues.

In this multicenter, retrospective cohort study, the investigators enrolled consecutive patients with n-irAEs at seven tertiary centers in Italy over almost eight years. Diagnosis of the n-irAE was based on the local workup as well as the temporal relationship of these neurologic events to the ICI administration (within 12 months of the last ICI infusion).

The clinical course of the n-irAEs was classified as fulminant (death from the event within 12 weeks), monophasic (resolution within 12 weeks), or chronic (process persisted beyond 12 weeks). Patients with chronic toxicity were further categorized as active (presumed if indirect evidence of continued inflammation) or inactive (no signs of continued inflammation, even with ongoing neurologic symptoms). Statistical analyses included comparisons between the groups as well as a focus on time to death.

Sixty-six patients were included in the analysis, most of whom experienced n-irAEs affecting the peripheral nervous system (48 patients; 73%). Twelve patients (18%) exhibited a fulminant course and died within 12 weeks. Concurrent myocarditis significantly increased the risk of early death (odds ratio [OR], 5.4; 95% confidence interval [CI], 1.02-28.31). Twenty-three patients (43%) were classified as having monophasic n-irAEs with relatively rapid resolution. Of the 31 (57%) with chronic neurologic toxicity, 15 (52%) exhibited signs of active inflammation (based on the need for ongoing immunosuppression, symptomatic relapse after corticosteroid tapering/discontinuation, or neurologic disease progression).

Neurologic symptoms began after a median of 8.7 weeks and three cycles from the first ICI administration. Serum antibody testing was positive in several patients. Eleven patients (17%) fulfilled criteria for a definite or probable paraneoplastic syndrome. In terms of treatment, steroids were administered (after ICI discontinuation in all patients) as first-line treatment in 92% of patients. Twenty-three patients (35%) received additional nonsteroidal treatment, such as intravenous immunoglobulin, plasma exchange, rituximab, or other immunosuppressive drugs. Patients with chronic n-irAEs had a higher rate of severe neurologic disability and were associated with shorter survival and increased mortality.

Commentary

Since n-irAEs fortunately remain a rare occurrence with ICI use, it is quite difficult to perform a prospective study following the neurologic sequelae in affected patients over an extended period. As neurologic symptoms often present rapidly, requiring emergent management, most of the attention regarding immune-related toxicities is devoted to their acute presentation, while the chronic course and management of these complications remains opaque.

This multicenter retrospective study admirably assesses the clinical course of these patients past the acute onset. Interestingly, more than half of the patients in this cohort with n-irAEs developed a chronic condition, often requiring ongoing immunomodulatory treatment. Also notable was the high percentage of patients with positive serum antibodies (when tested), as well as potential paraneoplastic syndromes, which may have been catalyzed by immune activation in the setting of checkpoint inhibition.

As noted by the investigators, a significant limitation remains the retrospective nature of the study, although a prospective study would be especially difficult with such a rare complication with quite diverse presenting symptoms. Another significant challenge remains the diagnosis of the n-irAE itself — after other potential etiologies were excluded, toxicities were presumed secondary to ICI use if they presented within 12 months (quite a large window).

Patients with cancer already carry a higher risk of neurologic complications, both directly and indirectly related to the malignancy, so it would not be surprising to see neurologic changes over a year-long period. Many unanswered questions remain, including why immunologic toxicity may have such a delayed onset and only present after multiple ICI infusions. The study investigators should be applauded for highlighting the often-chronic nature of immune-related toxicities and the need for further study and long-term follow-up.

Rajiv S. Magge, MD, is Associate Professor of Clinical Neurology, Weill Cornell Medicine, Weill Cornell Brain Tumor Center.