Diagnostic Performance of Three Serological Assays in Myasthenia Gravis: A Prospective Multicentre Study.

Anti-acetylcholine receptor (AChR) and anti-muscle-specific tyrosine kinase (MuSK) autoantibody detection is crucial in the diagnosis and choice of treatment of myasthenia gravis (MG). We conducted a multicentre prospective study comparing radioimmunoprecipitation assay (RIPA), enzyme-linked immunosorbent assay (ELISA) and fixed cell-based assay (F-CBA) for anti-AChR and anti-MuSK antibody detection in 78 patients with suspected MG with at least 6 months of clinical follow-up. In the diagnosis of seropositive MG (anti-AChR+anti-MuSK abs), RIPA was most sensitive (82.8%) compared to ELISA (81.0%) and F-CBA (70.7%). F-CBA exhibited the highest specificity overall (95.0%). For anti-AChR detection, F-CBA demonstrated a sensitivity of 73.6% and specificity of 95.0%; ELISA showed sensitivity and specificity of 81.1% and 85.0%, respectively; and RIPA yielded sensitivity and specificity of 81.1% and 95.0%. Sensitivity of F-CBA improved when sera were tested at lower dilution (1:5) versus the manufacturer's recommended 1:10, without compromising specificity. Agreement among methods was almost perfect for anti-AChR detection (Cohen's Kappa > 0.81). For anti-MuSK detection, agreement was substantial between ELISA and RIPA, moderate between ELISA and F-CBA, and fair between RIPA and F-CBA. Higher anti-AChR antibody levels were found in generalised versus ocular MG by both RIPA and ELISA. F-CBA confirmed its optimal specificity while the sensitivity seems to be influenced by sample dilution. In conclusion, given the radioactive nature of RIPA and consequent limitations, F-CBA may represent a valid alternative in anti-AChR and anti-MuSK antibody detection in MG diagnosis. We suggest that the use of live-CBA or RIPA could be reserved for inconclusive cases.