This retrospective study investigated the medical records of the patients with TAMG who visited the Severance Hospital between January 2010 and January 2024. Diagnosis of MG was made based on clinical, serological, and electrodiagnostic evaluations. Of the patients with TAMG, the patients 1) who underwent thymectomy at Severance Hospital and 2) whose thymoma was histologically confirmed after thymectomy were included. The patients whose medical records were insufficient to assess the clinical course of MG was excluded. Patients who were finally included were classified into the patients who experienced recurrence of thymoma (recurrence group) and whose who did not (non-recurrence group). The requirement for obtaining informed consent from study subjects was waived by the Institutional Review Board (IRB) of Severance Hospital, Yonsei University College of Medicine, Seoul, Korea (IRB No. 4-2024-1322).

In our institution, patients with thymoma undergo regular follow-up by thoracic surgeons to monitor the recurrence of thymoma. Although the scheduling of follow-up visits is personalized for each patient according to patient symptoms and physician preferences, most of the patients were recommended to conduct chest computed tomography (CT) every 6 months for first 3 years. After the initial 3 years, annual chest CT scan is recommended until 10 years postoperatively. When recurrence of thymoma was suspected in CT imaging, histopathological confirmation of recurrence was made on surgical specimens.7

Medical records were reviewed to collect data on age, sex, age at symptom onset of MG, clinical manifestations of MG, and treatments for MG. Myasthenia Gravis foundation of America (MGA) clinical classification, MGA post-intervention status (MGA PIS) and myasthenia gravis activities of daily living (MG ADL) scores were also recorded, which had been routinely assessed by the treating neurologist at every visit. The titer of AChR Ab was also recorded. or the patients with TAMG who experienced recurrence of thymoma, the date when the recurrence of thymoma were detected on chest CT (recurrence date) and the date of surgery for the recurrent thymoma (second thymectomy) were recorded. Symptoms of MG and MG ADL score during the period 1 year before and 1 year after the recurrence date was recorded to assess the worsening of MG during this period (Supplementary Fig. 1). MG aggravation was defined when the patient reported worsening of symptoms associated with MG and the MG-specific treatments were modified to manage these symptoms. In order to compare the frequency of MG aggravation, the baseline recurrence rate was calculated by dividing the number of MG aggravation instances that occurred from the time of MG onset until the time of the study, excluding the 2-year period around the point of MG recurrence, by the duration of the corresponding period (Supplementary Fig. 2). Based on the symptoms, MG ADL score, and change in MG-specific treatments during the 2-year period before and after thy thymoma recurrence, patients were classified into three categories: improved, unchanged, and worse.

Continuous variables were expressed as mean±standard deviation or median with interquartile range. Categorical variables were presented as percentages. Differences in baseline characteristics between patients with recurrence and those without recurrence were evaluated using the chi-squared test test for categorical variables and either the two-sided t-test or Wilcoxon test for continuous variables. Statistical analyses were conducted using SPSS software version 29 for Windows (SPSS Inc., Chicago, IL, USA).

A total of 178 patients with TAMG who were histologically diagnosed with thymoma in Severance Hospital were included. Of these patients, 18 patients (10.1%) experienced recurrence of thymoma and were classified as recurrence group. Mean age at MG onset of the patients in recurrence group was 43.7±9.7 years. Mean age at initial thymectomy and at thymoma recurrence was 44.1±9.4 years and 48.8±9.1 years, respectively. Duration from initial thymectomy to recurrence of thymoma was 56.3±35.3 months. At the point of initial thymectomy, two patients (11.1%) were classified as ocular and 16 patients (88.9%) as generalized MG. Worst MGA classification before initial thymectomy was class I in two patients (11.1%), class II in six (33.3%), class III in six (33.3%), and class V in four patients (22.2%). Median MG ADL score at the point of initial thymectomy was 4.0 (interquartile range [Q1-Q3], 2.3-8.8). The initial World Health Organization (WHO) histology classification was B1 in 16.7%, B2 in 61.1%, and B3 in 22.2%. Masaoka stage II was the most common (44.4%), followed by stage III (22.2%), stage I (16.7%) and stage IV (16.7%). When clinical and histological characteristic were compared between the recurrence and non-recurrence groups (Table 1), no significant differences were observed between groups in the aspect of age of MG onset, age at thymectomy, sex, MGA classification at initial thymectomy, WHO histology classification, and titer of AChR Ab. However, the proportion of patients with invasive thymoma were significantly higher in recurrence group (38.9%) compared to non-recurrence group (5.6%; p<0.01).

Table 2 presents the clinical and treatment status of the patients with TAMG at the point that thymoma recurrence was radiologically observed. MGA PIS of the patient was complete stable remission in four (22.2%), pharmacologic remission in five (27.8%), minimal manifestation in five (27.8%), improved in two (11.1%), and unchanged in two patients (11.1%). None of the patients were experiencing worsening or exacerbation of MG at this point. In terms of treatment, 10 patients (55.6%) were taking prednisolone and four (22.2%) were on oral immunosuppressants. Five patients (27.8%) were not receiving MG-specific therapy. Median MG ADL score was 0.0 (Q1-Q3, 0.0-1.8). Table 3 presents the detailed clinical characteristics of the 18 patients in recurrence group. In patient 17 and patient 18, recurrence of thymoma was confirmed exclusively on CT imaging without surgical intervention. Of the 14 patients whose serum level of AChR Ab was assessed both before and after the first thymectomy, level of AChR Ab increased in eight patients (57.1%) and decreased in six patients (42.9%).

Table 4 displays the baseline recurrence rate of MG and changes in MG ADL score before and after thymoma recurrence. Baseline recurrence rate, calculated by dividing the number of total events of worsening by the total follow-up duration, was 0.29 per years. A total of seven events of worsening occurred during the 2-year period around the recurrence of thymoma, with the worsening rate of 0.19 per year. MG ADL scores were recorded at four time points: 1 year prior to thymoma recurrence, at the time when thymoma recurrence was first detected, following secondary thymectomy, and 1 year after thymoma recurrence. Of the 18 patients in recurrence group, 12 patients remained unchanged during the 2-year period around the thymoma recurrence. Three patients experienced worsening of MG during the 1-year period prior to the recurrence date. In detail, symptoms of MG in patient 14 worsened along with pulmonary tuberculosis infection. Patients 15 and 16 experienced worsening of MG after the self-discontinuation of prednisolone treatment. our patients (patients 11, 12, 13, and 14) exhibited worsening of MG symptoms during the 1-year period after the recurrence date. In these four patients, symptoms of MG worsened 2-4 months after second thymectomy.