In suspected myasthenia, the assay behind the AChR result decides what it is worth

Design

Prospective multicentre diagnostic accuracy study, index tests read blind; Class I evidence

Population

327 consecutive adults and children with suspected myasthenia gravis at four Italian centres; 152 confirmed cases

Primary outcome

Sensitivity and specificity of live cell-based assay, fixed cell-based assay and ELISA against the final clinical diagnosis

Effect

AChR antibody specificity 97.7% (94.2 to 99.3) for both cell-based assays vs 69.1% (61.7 to 75.9) for ELISA; AUC 0.85 vs 0.83 vs 0.72

The ADAM study recruited 327 consecutive adults and children with suspected myasthenia gravis at four Italian centres between July 2023 and December 2025, excluding anyone already on immunotherapy. Every serum was tested blind by an in-house live cell-based assay, a commercial fixed cell-based assay and an indirect ELISA, against a final diagnosis made independently on examination, electrophysiology, sustained treatment response and exclusion of alternatives. Of the 327, 152 had myasthenia gravis.

For acetylcholine receptor antibodies, sensitivity differed modestly — 71.7% (95% CI 63.8 to 78.7) for the live assay, 69% (61.0 to 76.3) fixed, 63.8% (55.6 to 71.4) ELISA. Specificity did not: 97.7% (94.2 to 99.3) for both cell-based assays against 69.1% (61.7 to 75.9) for ELISA, which returned a false positive in 30.9% of patients who turned out not to have myasthenia. Areas under the curve were 0.85, 0.83 and 0.72, and the live-versus-fixed difference, while statistically significant, was small. MuSK antibody performance was the same across all three methods.

The practical consequence is about the report you are reading, not about ordering more tests. A positive ELISA in a patient with fatigable ptosis and a suggestive history means something quite different from a positive cell-based assay: on these numbers roughly a third of ELISA positives in a suspected-myasthenia population are wrong. Laboratories rarely state the method on the report line. Ask, once, what your laboratory runs — and where it is ELISA, treat a positive as a finding to confirm rather than a diagnosis.

• Find out which AChR antibody method your referral laboratory uses, and note it on the request.
• Treat an isolated positive ELISA in an atypical presentation as needing confirmation, not as a diagnosis.
• A fixed cell-based assay is a reasonable routine alternative to a live one — the accuracy gap is small.
• A negative result of any kind still leaves about a quarter to a third of myasthenia unlabelled; seronegative disease needs electrophysiology.
• Testing before starting immunotherapy matters — treated patients were excluded from this cohort for a reason.

The statistics, in plain English

Specificity is what matters in a test used to confirm a suspected diagnosis, because it governs how often a positive is wrong. At 69.1% specificity, ELISA mislabels almost a third of patients who do not have the disease — and in a clinic where half the referrals turn out to be something else, that produces roughly as many false positives as true ones. The cell-based assays' 97.7% changes that arithmetic entirely. Sensitivity differences between the three, by contrast, were modest and their confidence intervals overlap, so no assay here rescues a seronegative patient. This is graded Class I evidence for diagnostic accuracy, meaning a prospective consecutive cohort with blinded index tests and an independent reference standard.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.