By Claire Henchcliffe, MD, PhD

Associate Professor of Neurology and Neuroscience, Weill Cornell Medical College

Dr. Henchcliffe reports she is on the speakers bureau and advisory boards for Teva, IMPAX, and ACADIA, and receives grant/research support from Biogen and Kaneka.

SYNOPSIS: This retrospective study of 121 patients with stiff-person spectrum disorder extensively examined antibody correlates of clinical features. Anti-GAD65 antibodies were highly associated with typical stiff-person syndrome, and anti-GlyR antibodies with SPS-plus. However, presence of anti-GAD antibodies predicted worse outcome than presence of antiGlyR antibodies, independent of clinical subtype.

SOURCE: Martinez-Hernandez E, Arino H, McKeon A, et al. Clinical and immunologic investigations in patients with stiff-person spectrum disorder. JAMA Neurol 2016;73:714-720.

Stiff-person spectrum disorder (SPSD) was first reported by Moersch and Woltman in 1956 as stiff man syndrome, but it is now understood to present as a spectrum of conditions, ranging from segmental (stiff-limb syndrome, SLS) through to severe (progressive encephalomyelitis with rigidity and myoclonus, PERM) forms. Martinez-Hernandez and colleagues describe 121 patients with SPSD together with results of comprehensive antibody testing in this retrospective study, with the aims of dissecting out clinic-immunological associations and suggesting prognostic factors. Patients were classified into four groups: 1) stiff-person syndrome (SPS) with classic truncal rigidity and spasms (n = 50); 2) SLS with distal limb rigidity causing abnormal hand or foot postures (n = 24); 3) SPS-plus, with symptoms including myoclonus, seizures, brainstem dysfunction, and others (n = 37); and 4) overlap syndromes that included ataxia (n = 6), limbic encephalitis (n = 1), and epilepsy (n = 3) in combination with SPS or SLS. Antibodies to eight targets on inhibitory synapses were examined in paired serum-CSF (n = 65), serum-only (n = 50), or CSF only (n = 6), with targets as follows: GAD65, glycine receptor alpha1 subunit (GlyR), amphiphysin, gephyrin, GABAa receptor (GABAaR), dipeptidyl peptidase protein 6 (DPPX), and glycine transporter 1 and 2. Patients had a median age of symptom onset of 51 (interquartile range 40-61) years, and 62% were women. Antibodies to inhibitory synapses were commonly detected (67% total), including GAD65 (43%); GlyR (20%); and GABAaR, amphiphysin, or DPPX (4% combined). A minority of patients had more than one antibody detected. Interestingly, of the 33% without identified antibodies, three had serum antibodies to unidentified epitopes in live neuronal cultures. A major finding was the strong association of certain clinical features with specific antibodies. Classic SPS was much more likely in GAD65-positive or antibody negative cases, whereas SPS-plus (with a more aggressive presentation) was more likely in those with GlyR antibodies. However, those with GlyR antibodies had superior clinical outcomes than those with GAD65 antibodies or those who were antibody-negative.


This is a remarkable study that provides comprehensive antibody profiling in a rare disorder: one estimate of incidence is that it affects one in a million. It highlights the variable nature of SPSD, making it sometimes difficult to diagnose, and of the patients included in this study, less than half had typical SPS. The major finding of this study is that the types of antibodies present may predict outcome. In particular the presence of GAD65 and GlyR has different implications. There was preferential association of GAD65 with typical SPS, and of GlyR with SPS-plus, but presence of either antibody had associations with prognosis independent of the type of syndrome. Specifically, although those with GlyR antibodies had more severe symptoms at diagnosis, they had better outcomes than those with GAD65 antibodies. Although the major finding focuses on the two most common antibodies, anti-GAD76 to anti-GlyR, it is helpful in the clinic to know that other antibodies to inhibitory synapses were rarely detected (4% for GABAaR, amphiphysin, or DPPX), and none of the samples tested had antibodies to gephyrin, GlyT1 or GlyT2). This suggests that GAD65 and GlyR testing is sufficient in the majority of cases, along with amphiphysin in given its association with breast and lung cancer. The study, of course, is subject to referral bias and given its retrospective nature there could be other factors at play in the antibody associations determined. Therefore, it will be necessary to dig deeper. Nonetheless, if presence of anti-GAD vs. anti-GlyR antibodies turns out to be a true predictor of prognosis, it will greatly help clinical management and patient counseling.