Introduction Pediatric autoimmune neuropsychiatric disorder associated with streptococcal infections (PANDAS) is characterized by the abrupt onset of obsessive-compulsive disorder and/or tics, along with other neuropsychiatric symptoms in children, in the setting of Group A beta-hemolytic streptococcus infection and presumed immune dysregulation. Current diagnostic testing has limited utility in diagnosis of PANDAS. Methods The Nucleic Acid Protein Programmable Array (NAPPA) is a protein expression system that enables the study of protein-autoantibody interactions. We employed this system for unbiased screening of potential autoimmune targets in PANDAS. Results Initially, we compared children with PANDAS to patients with autism spectrum disorder (ASD) and identified 1,235 non-overlapping protein targets. These targets were subsequently used to re-screen samples from PANDAS, ASD and healthy controls (HC). In the secondary screen, 117 autoantibody targets were exclusively detected in PANDAS and absent from both ASD and HC groups. However, autoantibody profiles were highly heterogeneous: 67.5% of PANDAS-positive targets were reactive in only a single patient, and no individual protein or multi-protein panel classifier achieved reliable diagnostic discrimination under proper cross-validation. Targets that discriminated PANDAS from HC largely failed to discriminate PANDAS from ASD, and vice versa, indicating that autoantibody-based diagnosis is confounded by reactivity shared across neuropsychiatric conditions. Despite poor diagnostic performance, functional enrichment analysis revealed that PANDAS-specific targets were significantly overrepresented among transcription factors, apoptosis regulators, chromatin/epigenetic modifiers, neural development proteins, and immune-regulatory molecules, with the majority being intracellular proteins enriched for intrinsically disordered regions. Discussion These findings support the hypothesis of autoimmunity and epitope spreading as features of PANDAS but indicate that the heterogeneity of the humoral response represents a fundamental challenge for serological diagnosis. Future studies with larger cohorts and complementary approaches are necessary to determine whether autoantibody signatures can be leveraged for clinical utility.
Eremija et al. (Wed,) studied this question.