Paraneoplastic neurological disorders target diverse areas of the nervous system. Commercially available panels of antibody tests are comprised of tests for clinically distinct disorders, but physicians may be tempted to order comprehensive panels with the thought that this will improve diagnostic sensitivity.
In the current issue, Callaghan and colleagues studied 500 consecutive patients tested with the Mayo paraneoplastic panel at their institution, using clinical information to study the specificity and sensitivity of the panel tests.1 Overall, false positive were more than twice as common as true positive results. Positive results were usually informative in patients with a classic paraneoplastic syndrome, but in patients without a clear paraneoplastic clinical phenotype, positive results were almost always false positives.
The striational muscle antibody test, gAchR antibody test, AchR, N- and P/Q type calcium channel antibodies, and VGKC antibodies generated most of the false positive results. To explore one example, the gAChR test may positive in patients with autoimmune autonomic ganglionopathy, an extremely rare disease that manifests with pan-dysautonomia. Patients with low titer “positive” results most often have neither an autoimmune nor autonomic disorder.2 A test with occasional false positives, that is designed to diagnose an ultra-rare disease, being run on patients with the wrong phenotype creates the perfect storm of false positive results. In contrast, the paper determined high specificity for antibody testing focused on particular clinical syndromes, such as AchR antibodies in the setting of myasthenia gravis.
This study suggests one important way forward is to re-focus testing on specific phenotypes. A single panel that includes such diverse tests as gAchR, AchR, VGKC, anti-Tr (DNER), and ANNA-2 is unlikely to ever be entirely appropriate. Rather, there should be separate panels optimized for encephalitis, cerebellar syndromes, brainstem syndromes, sensory neuropathy, autonomic dysfunction, myasthenia, etc. This more focused approach would be cheaper to perform and also prevent further costly investigations and even treatments triggered by the inappropriate tests.
Physicians may erroneously interpret any positive test as diagnostic of an autoimmune disorder, and not recognize the potential for false positives. While the defenders of these panels may, correctly, note that better physician education could allow spurious results to be dealt with more expeditiously, it is unrealistic to expect all neurologists to completely master this arcane area of neurology. The panels and their reports should be designed for the average users, who in aggregate order the great majority of the panels, and not the expert. This means not only proper organization of panels by phenotype but also clear reporting that highlights the risk of false positives, especially when the phenotype does not fit the antibody. Panels should nudge users toward focused testing and appropriate interpretation rather than maximal test utilization.
- Ebright MJ, Li S-H, Reynolds E, Burke JF, Claytor BR, Grisold A, Banerjee M, and Callaghan BC. Unintended consequences of Mayo paraneoplastic evaluations. Neurology Epub 2018 October 26
- McKeon A, Lennon VA, Lachance DH, Fealey RD, Pittock SJ. Ganglionic Acetylcholine Receptor Autoantibody: Oncological, Neurological, and Serological Accompaniments. Archives of Neurology 2009;66:735-741