Ocular Motor and Vestibular Characteristics of Antiglutamic Acid Decarboxylase-Associated Neurologic Disorders

Antiglutamic acid decarboxylase (GAD)- associated neurologic disorders are rare, with varied presentations, including stiff-person syndrome (SPS) and cerebellar ataxia (CA). Vestibular and ocular motor (VOM) dysfunction can be the main presentation in a subset of patients

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Ocular Motor and Vestibular Characteristics of Antiglutamic Acid Decarboxylase–Associated Neurologic Disorders Yujie Wang, MD, Roksolyana Tourkevich, MD, Justin Bosley, Daniel R. Gold, DO, Scott D. Newsome, DO Background: Antiglutamic acid decarboxylase (GAD)associated neurologic disorders are rare, with varied presentations, including stiff-person syndrome (SPS) and cerebellar ataxia (CA). Vestibular and ocular motor (VOM) dysfunction can be the main presentation in a subset of patients. Methods: Retrospective review of the Johns Hopkins Hospital medical records from 1997 to 2018 identified a total of 22 patients with a diagnosis of anti-GAD–associated SPS or CA who had detailed VOM assessments. Eight had prominent VOM dysfunction at the initial symptom onset and were referred to neurology from ophthalmology or otolaryngology (“early dominant”). Fourteen patients had VOM dysfunction that was not their dominant presentation and were referred later in their disease course from neurology to neuro-ophthalmology (“nondominant”). We reviewed clinical history, immunological profiles, and VOM findings, including available video-oculography. Results: In the 8 patients with early dominant VOM dysfunction, the average age of symptom onset was 53 years, and 5 were men. The most common symptom was dizziness, followed by diplopia. Seven had features of CA, and 4 had additional features of SPS. None had a structural lesion on brain MRI accounting for their symptoms. The most common VOM abnormalities were downbeating and Department of Neurology, Johns Hopkins University, Baltimore, Maryland. Y. Wang is supported by the National Multiple Sclerosis Society. S. D. Newsome has received consultant fees for scientific advisory boards from Biogen, Genentech, Celgene, and EMD Serono, is an advisor for Gerson Lehrman Group and BioIncept, a clinical adjudication committee member for a MedDay Pharmaceuticals’ clinical trial and has received research funding (paid directly to institution) from Biogen, Novartis, Genentech, National Multiple Sclerosis Society, Department of Defense, and Patient-Centered Outcomes Institute. The remaining authors report no conflicts of interest. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www. jneuro-ophthalmology.com). Address correspondence to Scott D. Newsome, DO, Division of Neuroimmunology and Neurological Infections, Johns Hopkins Hospital, 600 North Wolfe Street, Pathology 627, Baltimore, MD 21287; E-mail: [e-mail] Wang et al: J Neuro-Ophthalmol 2021; 41: e665-e671 gaze-evoked nystagmus and saccadic pursuit. All received immune therapy and most received symptomatic therapy. Most experienced improvement in clinical outcome measures (modified Rankin scale and/or timed 25-foot walk test) or VOM function. By contrast, in the 14 patients in whom VOM dysfunction was nondominant, most had an SPS phenotype and were women. VOM abnormalities, when present, were more subtle, although mostly still consistent with cerebellar and/or brainstem dysfunction. Conclusions: Individuals with anti-GAD–associated neurologic disorders may present with prominent VOM abnormalities at the initial symptom onset that localize to the cerebellum and/ or brainstem. In our cohort, immune and symptomatic therapies improved clinical outcomes and symptomatology. Journal of Neuro-Ophthalmology 2021;41:e665–e671 doi: 10.1097/WNO.0000000000001084 © 2020 by North American Neuro-Ophthalmology Society G lutamic acid decarboxylase (GAD) is the rate-limiting enzyme for the synthesis of gamma-aminobutyric acid (GABA), the major inhibitory neurotransmitter in the central nervous system. There are 2 isoforms of GAD, one with a molecular size of 65 kDa (GAD65) and other, of 67 kDa (GAD67) (1,2). GAD65 seems to be the isoform that is the main target of autoantibodies in anti-GAD–associated neurologic disorders (2). Anti-GAD antibodies have been identified at high levels in the serum and cerebrospinal fluid (CSF) of a number of rare neurologic syndromes; most well-known being stiff-person syndrome (SPS) but includes others, such as cerebellar ataxia (CA) and autoimmune encephalitis (2). Some of these individuals, in particular those with CA, have been found to have an array of vestibular and ocular motor (VOM) abnormalities (2,3). This is likely explained by involvement of the flocculus/paraflocculus (contributes to gaze holding, smooth pursuit, and modulation of vestibulo-ocular reflex [VOR]) and nodulus/uvula (contributes to velocity storage), which together makeup the vestibulocerebellum. e665 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Given the rarity of anti-GAD–associated neurologic disorders, there is a paucity of literature regarding individuals who present with predominantly VOM presentations especially at the initial symptom onset. Hence, we sought to identify individuals who were referred for neurologic evaluation by ophthalmology or otolaryngology and characterize their VOM function, with the aim of demonstrating the clinical profile of individuals who have predominantly VOM symptoms, including unique VOM abnormalities not previously reported in this population. To the best of our knowledge, specific characterization of the VOM function of these individuals has only been reported in single case reports and not a case series as described in this article. METHODS We performed a retrospective review of the Johns Hopkins Hospital medical records from January 1, 1997, to June 1, 2018, and identified 170 individuals with a diagnosis of SPS or CA with elevated anti-GAD antibodies (SPS spectrum disorder). We used previously published criteria for diagnosis of SPS (4,5). CA was determined by the presence of either appendicular, truncal, or gait ataxia, based on clinical notes. This study was approved by the Institutional Review Board at Johns Hopkins. The main inclusion criteria for the study included having a diagnosis of SPS spectrum disorder based on above-mentioned criteria and presence of anti-GAD antibody in serum and/or CSF. For the primary group of interest (those with early and predominant VOM dysfunction), we identified patients who were referred from ophthalmology or otolaryngology (comparator group as discussed below was referred from neurology to neuro-ophthalmology); primary referral reason was vestibular or ocular motor in nature, and a brain MRI was performed. The main exclusion criteria included having seronegative SPS spectrum disorder, presence of another explanation for vestibular or ocular motor symptoms, or insufficient medical records for review. None of the patients in this case series had an underlying paraneoplastic process. We identified 8 individuals with prominent VOM dysfunction at the initial symptom onset (early dominant group) who were referred to the neurology clinic for evaluation by either ophthalmology (all of which were neuro-ophthalmologists) or otolaryngology. We extracted clinical data from the medical records, which included: 1) demographics (age at symptom onset, sex, and race), 2) clinical profile (referral information, neurologic examination, concurrent autoimmune disorders, treatments implemented, and patient response to them), 3) immunologic profiles (serum and CSF anti-GAD65 autoantibodies, oligoclonal banding, and immunoglobulin G index in the CSF), 4) brain MRI results, and 5) ocular motor and vestibular examination findings, including any available quantitative assessments. e666 From the patients’ neurologic examination, we determined their clinical phenotype by reviewing clinic notes from the neuroimmunology specialist (S.D.N.). A rater blinded to the patient’s VOM profile and treatments calculated the modified Rankin Scale (mRS) for the first and last clinic visits (Y.W.). Timed 25-foot walk (T25FW) assessments were available for a subset of patients. We extracted VOM examination findings from the neuro-vestibular clinic notes and from the review of available video-oculography recordings, which were interpreted by the neuro-vestibular specialists (R.T. and D.G.). In addition, as a comparator group, we performed chart review of 14 individuals with SPS spectrum disorders who had VOM dysfunction that was not prominent at the symptom onset (nondominant group) and who were referred directly to neuro-ophthalmology from our neuroimmunology clinic, to determine their clinical characterization and VOM function. RESULTS Clinical and Immunological Profiles of the Cohort With Early Dominant Vestibular and Ocular Motor Dysfunction Table 1 outlines the clinical and immunological profiles of the 8 patients with prominent VOM dysfunction at the initial symptom onset who are included in this case series. The average age of symptom onset was 53 years, 5 were men, and 5 were Caucasian. Six had features of CA at initial neurologic evaluation, and 4 had additional features of SPS (2 reported symptoms of back stiffness and spasms, one reported both back and leg stiffness and spasms, and one was found to have axial rigidity on examination). Three had concurrent autoimmune disorders, and none had diabetes mellitus or thyroid disease. None of the patients had an underlying paraneoplastic process. The anti-GAD antibody levels were elevated in the serum of all patients and present in the CSF of 5 of the 6 who underwent lumbar puncture. Three individuals had CSF-restricted oligoclonal bands, and one individual who also had rheumatoid arthritis had oligoclonal bands in both the serum and CSF (mirror bands). The brain MRI did not demonstrate abnormalities that would account for the patients’ symptoms or signs. Table 2 outlines the referral data. Neuro-ophthalmology referred 3 patients and otolaryngology referred 5 patients for further evaluation of most commonly dizziness and diplopia. After suspicion for an anti-GAD–associated disorder, referrals were made to neuroimmunology clinic for confirmation and treatment. Only 3 patients had previously seen a neurologist; most were previously evaluated by multiple non-neurological subspecialists. The average time to proper diagnosis was 5.6 years, ranging from one to 18 years. Wang et al: J Neuro-Ophthalmol 2021; 41: e665-e671 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. Clinical and immunological profiles of the cohort with early dominant vestibular and ocular motor dysfunction ID Age of Symptom Onset 1 Clinical Phenotype(s) Concurrent Autoimmune Disorders Raynaud Sex Race 55 M C SPS+, CA 2 3 4 46 34 70 M F F AA AA C SPS+, CA SPS+, CA CA 5 6 7 63 47 57 M M M C C C CA SPS+ CA 8 51 F AA CA None None Rheumatoid arthritis None None Vitiligo and pernicious anemia None Anti-GAD Ab (Serum, IU/mL) Anti-GAD Ab (CSF, Nmol/L) CSF OCB/IgG Index Elevated* Absent/0.5 256,000† 256,000† 79,616 18.60 N/A 0.94 Restricted/0.7 Restricted/0.5 Mirror/0.6 328 32,085 84,736 0 N/A 3.69 Absent/0.5 N/A Absent/0.6 256,000† 72.4 Restricted/0.6 896 Brain MRI Mild prominence of lateral ventricles Normal Diffuse volume loss Left schwannoma‡ Normal Normal Mild nonspecific subcortical white matter lesions in supratentorial regions Normal *The laboratory did not provide a titer. † The laboratory did not provide titer beyond 256,000 international units/mL. ‡ This participant did not experience any symptoms or exhibit signs localizing to this lesion, specifically no left-sided hearing loss or tinnitus and no evidence of fixation removed right-beating nystagmus. AA, African American/Black; Ab, Antibody; C, Caucasian/White; CA, cerebellar ataxia; CSF, cerebrospinal fluid; GAD, glutamic acid decarboxylase; F, female; IU, International Units; IgG, immunoglobulin G; M, male; N/A, not available; OCB, Oligoclonal band; SPS, stiff-person syndrome; SPS+, classic SPS with cerebellar and/or brainstem findings. Treatment Response and Clinical Outcomes of the Cohort With Early Dominant Vestibular and Ocular Motor Dysfunction All individuals received immune therapy after diagnosis, with 6 receiving intravenous immunoglobulin (IVIg) monotherapy; 5 of the 6 reported symptomatic improvement. Two individuals received trials of multiple medications (one with IVIg, azathioprine, and rituximab, and the other with IVIg, plasmapheresis, and rituximab), and both derived symptomatic improvement after starting rituximab (Table 3). Five individuals received symptomatic therapy in addition to immune therapy, with agents used being baclofen (2), clonazepam (2), and 4-aminopyridine (one), with all but one experiencing benefit— either symptomatically or with improved or stable VOM function when follow-up testing was available (Table 3). Participants were followed for a mean of 56 months (range: 13–152). The mean initial visit mRS was 2.9, and the mean last visit mRS was 2.1. Initial visit and final visit T25FW assessments were available in 6 participants, with a mean of 13.4 seconds, with 2 individuals requiring a walking aid (one cane and one walker), and 8.7 seconds, with one individual requiring a walking aid (one walker), respectively (Table 3). Interpretation of Vestibular and Ocular Motor Function of the Cohort With Early Dominant Vestibular and Ocular Motor Dysfunction Table 4 summarizes the VOM examination findings of the patients. Two key illustrative patient examples are given below: Wang et al: J Neuro-Ophthalmol 2021; 41: e665-e671 ID #3 This individual was a 54-year-old woman whose symptoms initially started when she was 34-year-old. She reported episodic dizziness that increased in frequency over the years. A few years after the onset of her dizziness, she began to experience persistent diplopia. On examination, she had spontaneous downbeating nystagmus (DBN) (See Supplemental Digital Content 1, Video E1, http://links.lww. com/WNO/A439), which increased with hyperventilation and straight head hanging (provocative maneuvers that commonly aggravate or provoke DBN in patients with cerebellar dysfunction). In lateral gaze, DBN increased, and gaze-evoked nystagmus (GEN) was seen, causing the appearance of so-called “side pocket nystagmus” due to summation of downward and horizontal nystagmus vectors. Downward corrective saccades were seen with horizontal head impulse testing (so-called “cross-coupling,” which localizes to the vestibulocerebellum), alternating skew deviation was present with alignment testing, and smooth pursuit and VOR suppression were saccadic. Saccadic accuracy and velocity were normal. Taken together, this patient demonstrated a variety of VOM features that localized well to the flocculus/paraflocculus (6). In addition, she demonstrated features of SPS, including lumbar hyperlordosis and axial rigidity. She was treated with IVIg and experienced improvement in her axial rigidity, although her VOM dysfunction remained stable. ID #4 This individual was a 70-year-old woman who presented with subacute onset of dizziness and imbalance, for which e667 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 2. Referral information of the cohort with early dominant vestibular and ocular motor dysfunction ID Referral Source 1 2 3 Neuro-ophthalmology Neuro-ophthalmology Neuro-ophthalmology Oscillopsia Diplopia and vertigo Dizziness 4 5 6 7 Otolaryngology Otolaryngology Otolaryngology Otolaryngology Dizziness Dizziness Dizziness Dizziness and diplopia 8 Otolaryngology Dizziness No. of Relevant Specialists Seen Before Referral Time to Diagnosis (yrs) 2 (ophthalmology) 4 (1 ophthalmology and 3 neurology) 5 (3 otolaryngology, 1 ophthalmology, and 1 neuro-ophthalmology) Unknown 2 (1 otolaryngology and 1 neurology) Unknown 3 (1 neurology, 1 otolaryngology, and 1 ophthalmology) Unknown 2 1 18 Reason for Referral she was hospitalized for expedited workup and treatment. Her examination demonstrated significant GEN, bilateral vestibular loss (with bedside and video head impulse testing), and saccadic smooth pursuit (the NeuroOphthalmology Virtual Education Library [NOVEL]— Daniel Gold Collection: https://collections.lib.utah.edu/ ark:/87278/s6wx1bgp). She was treated with IVIg, with improvement in her symptoms and VOM function. 5 2 11 1 5 phenotype, and one had SPS-plus (classic SPS with cerebellar and/or brainstem findings). Supplemental Digital Content 2 (see Table E1, http://links.lww.com/WNO/ A438) outlines their clinical profile and VOM function. The average age at the SPS symptom onset of this cohort was 46 years, and 11 were women. The primary reasons for referral were diplopia (9), dizziness (3), and oscillopsia (2). In the patients with an SPS phenotype, 3 had normal examinations; 4 had mild abnormalities (including slow saccades, gaze-evoked nystagmus, and saccadic pursuit); 2 had clear features of cerebellar pathology (alternating skew deviation and DBN without fixation); and one had bilateral, fatigable ptosis. In the patients with CA, all had abnormalities, including spontaneous downbeating or upbeating nystagmus, hypermetric saccades, and alternating skew deviation. Vestibular and Ocular Motor Function of the Cohort With Nondominant Vestibular and Ocular Motor Dysfunction Fourteen individuals with SPS spectrum disorders were referred from neuroimmunology clinic to neuroophthalmology; 10 had an SPS phenotype, 3 had a CA TABLE 3. Immune and symptomatic therapies and clinical outcomes of the cohort with early dominant vestibular and ocular motor dysfunction First Visit T25FW (seconds) ID Follow-up Duration (mo) First Visit mRS 1 106 1 7.8 2 152 2 3 76 4 Last Visit T25FW (seconds) Immune Therapy Symptomatic Therapy 4-aminopyridine and clonazepam None 3 15.0 Improved nystagmus N/A IVIg, azathioprine, and rituximab IVIg 1 6.1 3 6.5 IVIg and rituximab Clonazepam 2 4.5 26 4 8.1 IVIg None 2 7.6 5 29 3 N/A IVIg and steroids Diazepam 1 N/A 6 7 21 13 3 4 5.0 26.9 IVIg IVIg and rituximab Clonazepam Baclofen and clonazepam 3 3 5.4 N/A 8 25 3 25.9 IVIg Baclofen and clonazepam 2 13.5 Improved dizziness and ataxia Improved rigidity and stable VOM dysfunction Improved GEN, VOR suppression, and smooth pursuit Improved dizziness and ataxia Stable VOM dysfunction Improved ataxia and stable VOM dysfunction Improved dizziness and ataxia. Stable VOM dysfunction Last Visit mRS Symptom Evolution GEN, gaze-evoked nystagmus; IVIg, intravenous immunoglobulin; mRS, modified Rankin scale; T25FW, timed 25-foot walk test; VOM, vestibular and ocular motor; VOR, vestibulo-ocular reflex. e668 Wang et al: J Neuro-Ophthalmol 2021; 41: e665-e671 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 4. Vestibular and ocular motor examination findings of the cohort with early dominant vestibular and ocular motor dysfunction ID Ocular Alignment 1 Divergence insufficiency 2 Normal 3 Alternating skew deviation 4 Normal 5 Normal 6 Normal 7 Divergence insufficiency 8 Exodeviation (longstanding and unrelated) Spontaneous Nystagmus Eccentric Gaze Saccades Smooth Pursuit and VOR Suppression Hyperventilation or Head Shaking (HS)– Induced Nystagmus Positional Testing DB with and without fixation None Gaze evoked with rebound Normal Saccadic None Normal Hypermetric Normal None DB with and without fixation; LB without fixation DB without fixation DB without fixation Gaze evoked with rebound Normal Saccadic Hypervenitilation— increased DB Increased DB with straight head hanging Gaze evoked with rebound Normal Hypometric Saccadic† None Normal Normal Normal Hyperventilation and HS—increased DB DB without fixation DB without fixation Gaze evoked with rebound Gaze evoked with rebound Normal Saccadic Hypermetric Saccadic Hyperventilation— increased DB HS—increased DB Increased DB with straight head hanging and Dix-Hallpike Normal DB without fixation Gaze evoked with rebound Slow Saccadic Hyperventilation— increased DB LB with left DixHallpike (nonlocalizing) Normal Apogeotropic nystagmus with supine roll and DixHallpike LB with supine roll to the left (nonlocalizing) Video Head Impulse Test (HIT) N/A (bedside HIT normal) N/A (bedside HIT normal) Abnormal to the right* Abnormal bilaterally Normal Normal Abnormal to the left* Normal *Unclear if this finding is related to anti-GAD. † Although the smooth pursuit was saccadic, VOR suppression was normal. DB, downbeating; LB, left beating; VOR, vestibulo-ocular reflex. The one patient who had an SPS-plus phenotype had an abnormal examination with alternating skew deviation and fixation removed DBN. DISCUSSION In our study, we demonstrate that a subset of individuals with anti-GAD–associated neurologic disorders present with predominantly ocular motor–related (oscillopsia and/ or diplopia) or vestibular (dizziness/vertigo and/or imbalance) symptomatology. These individuals may present initially to ophthalmology or otolaryngology and can go several years before receiving the proper diagnosis. The rarity of these disorders makes diagnosis challenging because it requires a high index of suspicion to perform the appropriate testing. In reviewing the clinical characteristics, including VOM examination, we show that many have features of cerebellar involvement, most commonly DBN and saccadic smooth pursuit. In the literature, there are case reports of abnormal eye movements in individuals with elevated anti-GAD antibodies. These include nystagmus (2,3,7–18), saccadic disturbances (7–9,14,16), and skew deviation (11,15,17). These abnormalities are most often seen in conjunction with CA, although may occur independently (10,11). Most Wang et al: J Neuro-Ophthalmol 2021; 41: e665-e671 of the case reports and series do not describe detailed VOM examinations for these individuals. For example, the largest series to date of CA with anti-GAD antibodies were 2 studies from the same center in Europe, with some overlapping patients, and in these studies, most patients presented with nystagmus (2,3). However, further details regarding these individuals’ vestibular findings are not presented. Characterizing such abnormalities with targeted examination may provide us with a better understanding of the full spectrum of VOM presentations as well as provide a window into the pathogenesis and mechanism of the disease. Moreover, earlier recognition of these underrecognized presentations could lead to earlier diagnoses and implementation of immune therapies, which may help minimize long-term accrual of disability. Dysfunction of GABAergic pathways may lead to the VOM abnormalities seen in these individuals. Involvement of the flocculus/paraflocculus commonly results in one or more of the following: impaired (saccadic or choppy) smooth pursuit and VOR suppression, gaze-evoked nystagmus (usually with rebound nystagmus), positional nystagmus (e.g., positional DBN), spontaneous DBN, central patterns of head-shaking nystagmus (DBN with horizontal HS), VOR disturbances (abnormal direction of VOR head impulse response), or alternating skew deviation (6). e669 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Lesions of the nodulus/uvula can theoretically cause each of the VOM abnormalities as above (with the exception of gaze-evoked nystagmus because nodulus/uvula do not contribute to gaze holding), although periodic alternating nystagmus is unique to this localization (6). Additional cerebellar structures that play a role in normal ocular motor function (especially saccadic accuracy) include the fastigial nucleus (FN) and ocular motor vermis. A unilateral or bilateral FN lesion will result in bilaterally hypermetric saccades (4,5). A unilateral ocular motor vermis lesion will result in hypometric ipsiversive and hypermetric contraversive saccades, whereas bilateral lesions will result in bilaterally hypometric saccades (6,19). Horizontal diplopia from divergence insufficiency (esodeviation that increases and becomes symptomatic at distance) can be due to bilateral vermis or flocculus/paraflocculus lesions (6). Spontaneous DBN (with or without fixation) was seen in 7 of the 8 early dominant VOM dysfunction patients and 4 of the 14 nondominant VOM dysfunction patients. DBN may manifest when floccular Purkinje cells fail to inhibit the anterior semicircular canal (upward or antigravity) VOR pathways at the level of the vestibular nuclei. If the anterior canal pathways are not inhibited by the Purkinje cells, this creates a slow upward ocular drift (slow phase of nystagmus due to anterior canal VOR hyperactivity) that is followed by a fast downward position reset mechanism (fast phase of DBN) (20). This is one mechanism by which impairment of the normal inhibitory effects of GAD could result in VOM abnormalities in anti-GAD syndromes (19,20). One of the unique findings in our study is that in most patients in whom DBN was seen, it was with removal of fixation and amplified by provoking maneuvers. In progressive cerebellar pathologies, DBN without fixation may be a subtle finding of early disease (6). Therefore, to detect such DBN earlier on, detailed examination, with removal of fixation or with use of provoking maneuvers, is important. In addition, such DBN becoming amplified with removal of fixation may imply that it is coming from the anterior commissure–posterior commissure asymmetry (centrally at the level of the flocculus), rather than from another mechanism (such as pursuit or utricle asymmetry or vertical gaze-holding impairment) (6). Head impulse testing (either bedside or by video) demonstrated 2 patients in the early dominant group with unilateral positivity and one with bilateral positivity. No convincing peripheral etiologies were identified with audiometry or caloric testing; however, whether these findings are related to the anti-GAD–associated neurologic disorder is uncertain. One possible mechanism through which a central process may mediate a head impulse test (HIT) is through the flocculus, which mediates the high-velocity VOR, with dysfunction in this system potentially impairing a HIT (6,21). In comparison with the 8 patients who presented with early dominant VOM dysfunction, some patients that e670 presented with non-VOM symptomatology at the onset also had later identified VOM symptoms. In the latter group, as expected, those with a CA phenotype had similar cerebellar patterns of VOM dysfunction as described above. VOM findings in those with an SPS phenotype were more varied. Over half demonstrated either normal VOM function or mild dysfunction. The main abnormalities, when present, were similar (albeit more subtle) to what were seen in the early dominant group, which may point toward underlying cerebellar dysfunction even in patients with a predominant SPS phenotype and supports that these conditions present along a wide spectrum (2,3). Slowed saccades were seen in one patient in the early dominant group and 4 in the nondominant group. In addition, one patient with a predominant SPS phenotype had fatigable ptosis. These findings may be seen in association with myasthenia gravis, and there are a few case reports in the literature describing patients with cooccurrence of SPS and myasthenia gravis (22). The underlying mechanisms for the VOM dysfunction in these patients—whether it may be related to the anti-GAD– related condition or perhaps due to myasthenia gravis instead—remain unclear. The finding of slow saccades is an interesting phenomenon, and there exists a variety of putative mechanisms. Slowed saccades may be explained by the impairment of burst neurons (23). Burst neurons, located in the rostral interstitial nucleus of the medial longitudinal fasciculus (for vertical saccades) and the paramedian pontine reticular formation (for horizontal saccades), project to the ocular motor neurons and control saccades. They receive excitatory input from the cerebellum and inhibitory input from omnipause neurons. Impairment of inhibitory neurons may compromise the synchronicity of the firing of these burst neurons, resulting in impaired saccade generation, although the exact mechanism(s) have not been elucidated (9,16,24). An additional hypothesis is that multiple hypometric saccades (seen in cerebellar pathology) may also produce an appearance of slowed saccades (24). This study does have limitations. It is a retrospective study, relying on some chart-documented information, although for most participants, the detailed VOM findings were identified and characterized by our neuro-vestibular specialists (R.T. and D.G.), and in the early dominant group, this was conducted before implementation of immune and symptomatic therapies. Future prospective studies would provide further insights into the characteristic of these individuals and allow for additional longitudinal evaluation of changes in VOM function with time and treatment, for example. In addition, prospective studies with a systematic assessment of VOM function in antiGAD–associated neurologic disorders would be of interest in determining the full spectrum of VOM abnormalities and prevalence of such in this population. This study included a small number of individuals, although this Wang et al: J Neuro-Ophthalmol 2021; 41: e665-e671 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution disorder is quite rare and our case series is the largest to date that provides detailed characterization of the possible VOM dysfunction that can be seen in anti-GAD–associated neurologic disorders. Our study demonstrates that although rare, anti-GAD– associated disorders should be considered in individuals who present with VOM abnormalities that localize to the cerebellum and/or brainstem (with no clear distinguishing pattern), particularly if brain imaging fails to identify causative lesion(s). Referral to neurology for evaluation aides in the diagnosis and management of these individuals, and in our cohort, the institution of immune and symptomatic therapies indeed resulted in improved clinic outcomes and symptomatology. In addition, we demonstrate that some patients with a predominant SPS phenotype may present with VOM dysfunction localizable to the cerebellum and/or brainstem, supporting that these disorders present across a wide spectrum. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: S. D. Newsome and Y. Wang; b. Acquisition of data: Y. Wang, R. Tourkevich, J. Bosley, D. Gold, and S. D. Newsome; c. Analysis and interpretation of data: Y. Wang, R. Tourkevich, D. Gold, and S. D. Newsome. Category 2: a. Drafting the manuscript: Y. Wang and S. D. Newsome; b. Revising it for intellectual content: Y. Wang, R. Tourkevich, D. Gold, and S. D. Newsome. Category 3: a. Final approval of the completed manuscript: Y. Wang and S. D. Newsome. REFERENCES 1. Jun HS, Khil LY, Yoon JW. Role of glutamic acid decarboxylase in the pathogenesis of type 1 diabetes. Cell Mol Life Sci CMLS. 2002;59:1892–1901. 2. Saiz A, Blanco Y, Sabater L, González F, Bataller L, Casamitjana R, Ramió-Torrentà L, Graus F. Spectrum of neurological syndromes associated with glutamic acid decarboxylase antibodies: diagnostic clues for this association. Brain J Neurol. 2008;131:2553–2563. 3. Honnorat J, Saiz A, Giometto B, Vincent A, Brieva L, de Andres C, Maestre J, Fabien F, Vighetto A, Casamitjana R, Thivolet C, Tavolato B, Antoine J, Trouillas P, Graus F. Cerebellar ataxia with anti-glutamic acid decarboxylase antibodies: study of 14 patients. Arch Neurol. 2001;58:225–230. 4. Levy LM, Dalakas MC, Floeter MK. The stiff-person syndrome: an autoimmune disorder affecting neurotransmission of g-aminobutyric acid. Ann Intern Med. 1999;131:522. 5. Dalakas MC, Fujii M, Li M, McElroy B. The clinical spectrum of anti-GAD antibody-positive patients with stiff-person syndrome. Neurology 2000;55:1531–1535. 6. Leigh RJ, Zee DS. The Neurology of Eye Movements. 5th edition. Oxford, NY: Oxford University Press, 2015. Wang et al: J Neuro-Ophthalmol 2021; 41: e665-e671 7. Rakocevic G, Raju R, Semino-Mora C, Dalakas MC. Stiff person syndrome with cerebellar disease and high-titer antiGAD antibodies. Neurology. 2006;67:1068–1070. 8. Tilikete C, Vighetto A, Trouillas P, Honnorat J. Anti-GAD antibodies and periodic alternating nystagmus. Arch Neurol. 2005;62:1300–1303. 9. Zivotofsky AZ, Siman-Tov T, Gadoth N, Gordon CR. A rare saccade velocity profile in stiff-person syndrome with cerebellar degeneration. Brain Res. 2006;1093:135–140. 10. Antonini G, Nemni R, Giubilei F, Gragnani F, Ceschin V, Morino S, Bucci E, Accornero N. Autoantibodies to glutamic acid decarboxylase in downbeat nystagmus. J Neurol Neurosurg Psychiatry. 2003;74:998–999. 11. Martins AI, Carvalho JN, Amorim AM, Geraldo A, Eggenberger E, Lemos J. Disabling central paroxysmal positioning upbeat nystagmus and vertigo associated with the presence of antiglutamic acid decarboxylase antibodies. J Neuroophthalmol. 2018;38:32–35. 12. Baizabal-Carvallo JF. The neurological syndromes associated with glutamic acid decarboxylase antibodies. J Autoimmun. 2019;101:35–47. 13. Vale TC, Pedroso JL, Alquéres RA, Dutra LA, Barsottini OGP. Spontaneous downbeat nystagmus as a clue for the diagnosis of ataxia associated with anti-GAD antibodies. J Neurol Sci. 2015;359:21–23. 14. Shaikh AG, Wilmot G. Opsoclonus in a patient with increased titers of anti-GAD antibody provides proof for the conductancebased model of saccadic oscillations. J Neurol Sci. 2016;362:169–173. 15. Ances BM, Dalmau JO, Tsai J, Hasbani MJ, Galetta SL. Downbeating nystagmus and muscle spasms in a patient with glutamic-acid decarboxylase antibodies. Am J Ophthalmol. 2005;140:142–144. 16. Tilikete C, Vighetto A, Trouillas P, Honnorat J. Potential role of anti-GAD antibodies in abnormal eye movements. Ann N Y Acad Sci. 2005;1039:446–454. 17. Farooq AV, Soin K, Moss HE. Alternating skew deviation in association with anti-glutamic acid decarboxylase antibodies. Neuro-ophthalmol Aeolus Press. 2015;39:140–143. 18. Brokalaki C, Kararizou E, Dimitrakopoulos A, Evdokimidis I, Anagnostou E. Square-wave ocular oscillation and ataxia in an anti-GAD-positive individual with hypothyroidism. J Neuroophthalmol. 2015;35:390–395. 19. Kheradmand A, Zee DS. Cerebellum and ocular motor control. Front Neurol. 2011;2:53. 20. Glasauer S, Stephan T, Kalla R, Marti S, Straumann D. Up– down asymmetry of cerebellar activation during vertical pursuit eye movements. Cerebellum. 2009;8:385–388. 21. Baek SH, Choi JY, Jung JM, Kwon DY, Park MH, Choi J, Kim JS. Abnormal head impulse test in a unilateral cerebellar lesion. J Clin Neurol. 2015;11:279–282. 22. Thomas S, Critchley P, Lawden M, Farooq S, Thomas A, Proudlock FA, Constantinescu CS, Gottlob I. Stiff person syndrome with eye movement abnormality, myasthenia gravis, and thymoma. J Neurol Neurosurg Psychiatry. 2005;76:141– 142. 23. Bhidayasiri R, Riley DE, Somers JT, Lerner AJ, Buttner-Ennever JA, Leigh RJ. Pathophysiology of slow vertical saccades in progressive supranuclear palsy. Neurology. 2001;57:2070– 2077. 24. Jensen K, Beylergil SB, Shaikh AG. Slow saccades in cerebellar disease. Cerebellum Ataxias. 2019;6:1. e671 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.