Can neural autoantibodies disrupt CNS signaling without proving autoimmune disease?

Neural autoantibodies are biologically capable of causing CNS dysfunction, but a laboratory result is not equivalent to a diagnosis. Their clinical significance depends on the specific antigen, specimen type, assay, neurologic phenotype, and objective evidence of CNS inflammation or dysfunction.

Mechanistic and clinical evidence

• Pathogenic neuronal-surface antibodies can directly impair signaling. Anti-NMDA-receptor patient CSF IgG binds extracellular GluN1, cross-links receptors, and reversibly internalizes synaptic and surface NMDA receptors, reducing NMDA-mediated currents and plasticity.
• AMPAR antibodies similarly promote receptor internalization and reduce AMPAR-mediated miniature excitatory currents. GABA(_B)-receptor antibodies can block receptor-mediated inhibition without lowering surface receptor abundance.
• Ion-channel-associated effects are more target-specific: LGI1 patient IgG is associated with reduced Kv1.1, altered hippocampal excitability and presynaptic release, impaired long-term potentiation, and reduced synaptic AMPARs. CASPR2 antibodies can disrupt potassium-channel-associated organization, though downstream current effects are less firmly established.

Diagnostic interpretation

• Serum positivity alone does not establish pathogenic CNS autoimmunity. In a retrospective cohort, only 46/148 (31%) serum-only positive patients met probable or definite autoimmune encephalitis/paraneoplastic neurologic syndrome criteria, versus 29/32 (91%) with CSF-only positivity.
• A compatible syndrome, supportive MRI, EEG and/or CSF findings, exclusion of alternative causes, and antibody-appropriate confirmation are essential. Paired serum/CSF testing is particularly important: CSF is especially informative for NMDAR and GFAP antibodies, while serum may be more sensitive for LGI1 and CASPR2. Suspected anti-NMDAR encephalitis requires CSF IgG anti-GluN1 confirmation.

Bottom line

• Neural antibodies can be directly pathogenic, but an isolated serum elevation—especially with nonspecific or phenotype-discordant symptoms—should be regarded as a finding requiring clinical and laboratory corroboration, not proof of CNS autoimmune disease.