GQ1b-Seronegative Miller Fisher Syndrome Associated With Pembrolizumab.

Abstract

A 62-year-old man with a history of metastatic squamous cell carcinoma (SCC) of the scalp, who recently completed 5 cycles of pembrolizumab, presented to the emergency department with 2 weeks of progressive gait ataxia, dysarthria, dysphagia, and diplopia. On physical examination, he had profound areflexia, lower-extremity ataxia, facial diplegia, and ophthalmoplegia. Brain MRI was normal. Lumbar puncture showed albuminocytologic dissociation of 7 white blood cells/µL and protein of 236 mg/dL (normal: 15–45 mg/dL). Additional cerebrospinal fluid (CSF) studies including cytology and flow cytometry were normal. A tentative diagnosis of Miller Fisher syndrome (MFS) was made, and a 5-day course of intravenous immunoglobulin (IVIg) therapy was initiated. Within 2 days, there was marked improvement in facial weakness and ophthalmoplegia. Ganglioside (Asialo-GM1, GM1, GM2, GD1a, GD1b, and GQ1b), acetylcholine receptor (blocking, binding, and modulating), and paraneoplastic (antineuronal nuclear type I/II/III, antiglial nuclear type I, purkinje cell cytoplasmic types I/II/Tr, amphiphysin, CRMP-5, striational, P-/Q-type calcium channel, N-type calcium channel, AChR ganglionic neuronal, and neuronal [V-G] potassium channel) antibodies sent before starting IVIg were within the normal range. Outpatient follow-up 2 months later showed that ophthalmoplegia had resolved, and there was improvement in gait and residual mild left facial palsy. No further treatment with pembrolizumab has been administered. Miller Fisher syndrome is a rare form of Guillain–Barre syndrome characterized by a classic triad of ataxia, areflexia, and ophthalmoplegia (1). The disorder primarily is one of peripheral nerve dysfunction; however, rare variants manifest as central dysfunction—usually limited to the rhombencephalon (2). The pathogenesis is believed to be a postinfectious autoimmune disorder due to molecular mimicry (3). Anti-GQ1b antibody is most frequently detected in up to 80% of cases (2). The GQ1b epitope is found both centrally and peripherally, with significant abundance around nodes of Ranvier of the third, fourth, and sixth cranial nerves (4,5). Defects in the nerve terminals inside muscle spindles are rich in GQ1b epitopes, which correlate with ataxia and areflexia (6). Other less-abundant ganglioside antibodies (GT1a and GD1b) that are present in extraocular muscles (7), partly account for a minority of patients with MFS that are GQ1b antibody–negative. The majority are positive for a single and/or a complexed form of other ganglioside antibodies (8,9). A small subset of the GQ1b-seronegative cases is associated with other antibodies. Anti–glutamic acid decarboxylase (10) and voltage-gated potassium channel (11) antibodies have been reported in the literature. There is an even smaller group of truly seronegative cases, where no antibodies are detected (8). Our patient with metastatic SCC developed seronegative MFS, shortly after completing 5 rounds with a novel PD-1 inhibitor, pembrolizumab. He had the classic clinical phenotype of MFS and the typical CSF findings indicative of peripheral demyelination. The possibility of a paraneoplastic disease and/or leptomeningeal metastasis was excluded with serologic markers, CSF testing, and neuroimaging. The typical antecedent infection and ganglioside antibody seropositivity were not present. As this is atypical for MFS, we questioned an association with pembrolizumab, given the temporal association between therapy and onset of symptoms. The immune checkpoint inhibitor pembrolizumab has been linked to several neurological immune-related events (12). The drug modulates the host's ability to protect itself from autoimmunity by activating T-cell functionality (13)—a response that is beneficial in oncotherapy yet the source of potentially fatal immune-mediated adverse events. There are reports of Guillain–Barre syndrome associated with pembrolizumab, all negative for ganglioside antibodies (Table 1). Of the reported cases, only those treated with IVIg after discontinuation of the drug had clinical improvement (12,14,15).TABLE 1.: Reported cases with pembrolizumab-associated Guillain–Barrè–like syndromesThe pathophysiology of the peripheral nerve autoimmunity related to pembrolizumab is not fully understood. All 3 of the reported patients had varying degrees of facial nerve involvement; perhaps, pembrolizumab has an underlying predilection for ganglioside epitopes clustering on the facial nerve. This case adds to the myriad of adverse autoimmune reactions of neurological importance that complicates therapy with immune checkpoint inhibitors. STATEMENT OF AUTHORSHIP Category 1: a. conception and design: K. E. Green; b. acquisition of data: A. M. Levine; c. analysis and interpretation of data: K. E. Green and A. M. Levine. Category 2: a. drafting the manuscript: K. E. Green; b. revising it for intellectual content: J. Ward and K. E. Green. Category 3: a. final approval of the completed manuscript: D. I. Kaufman.