Lack of Immunotherapy as the Only Predictor of Secondary Generalization in Very-Late-Onset Myasthenia Gravis With Pure Ocular Onset

During the past two decades, an increasing number of patients with very-late-onset myasthenia gravis (v-LOMG) with an onset age of 65 years or older have been identified. However, few studies explore the predictors of secondary generalization in patients with v-LOMG with pure ocular onset. In this retrospective cohort study, 69 patients with v-LOMG were divided into ocular MG (OMG) and generalized MG (GMG), and the clinical characteristics and outcomes were compared. Cox regression analysis was performed to explore the predictors of generalization. The average onset age of the study population was 73.1 +/- 4.2 years and the median disease duration was 48.0 months (interquartile range, 32.5-64.5 months). Serum acetylcholine receptor (AChR) antibody was detected in up to 86% of patients and concomitant diseases in approximately half of the patients. Male predominance was seen in OMG group while female predominance in GMG group (p = 0.043). Patients with OMG showed a lower positive rate of repetitive nerve stimulation (RNS) than those with GMG (p = 0.014), and favorable outcomes were obtained in more patients with OMG than those with GMG (p < 0.001). Of the 51 patients with pure ocular onset, 25 (49.0%) underwent secondary generalization. A higher probability of generalization was found in patients with positive RNS results and without immunotherapy (p = 0.018 and <0.001). Upon Cox regression analysis, immunotherapy was negatively associated with secondary generalization [HR (hazard ratio) 0.077, 95%CI [0.024-0.247], p < 0.001]. Altogether, compared to the patients with very-late-onset GMG, the counterparts with OMG exhibit a significantly higher female predominance and a lower positive rate of RNS tests, especially on facial and accessory nerves. Lack of immunotherapy is the only predictor of secondary generalization in those with pure ocular onset.