In Reply: Clinical Prognostic Factors and Molecular Characteristics of Spinal Cord Diffuse Midline Gliomas

To the Editor: Thank you for your correspondence. In this response, we address the concerns raised by Yu et al1 and elaborate the considerations underlying our study.2 We hope this provides a more comprehensive explanation of our research. First: This study initially enrolled 48 patients with diffuse midline glioma (DMG). Two patients were excluded due to loss to follow-up—either unreachable or refusal of further evaluation—resulting in missing survival time and treatment response data. Therefore, the final cohort consisted of 46 patients included in the analysis. No other exclusions were applied. Second: To achieve maximal safe resection, all surgeries were conducted with intraoperative neurophysiological monitoring, as detailed in the Methods section. Alarm criteria for somatosensory evoked potentials were defined as an amplitude reduction greater than 50% or a latency increase of 10%. For electromyography, any spontaneous muscle electrical activity was considered indicative of nerve irritation or injury. Motor evoked potentials used the complete disappearance of waveforms as the alarm threshold. Third: Univariate analyses were performed for all variables shown in Figure 1 of the published article,2 as detailed in the Methods section. To avoid excluding relevant covariates, variables with a univariate P-value <.10 were included in the multivariate analysis. Subsequent multivariate Cox regression revealed no significant prognostic effect of preoperative Modified McCormick Scale grading or age. Wang et al3 reported that patients with lower histological grades tended to be older, associating advanced age with longer overall survival. Given that spinal DMG predominantly affects adults, dichotomizing age into adult and pediatric groups may be insufficient to reflect its prognostic value. A more granular age stratification may be warranted. In our cohort, preoperative Modified McCormick Scale grading showed no significant prognostic impact. Fourth: Our study primarily examines the influence of histology and the prognostic effects of surgical intervention across distinct histological subtypes. None of the patients received adjuvant therapy before the initial surgery. Postoperative adjuvant treatments varied among patients, including radiotherapy, chemotherapy, and targeted therapy, reflecting the absence of standardized guidelines for DMG management, especially in spinal DMG. For analysis, all adjuvant therapies were grouped into 2 categories—patients who received adjuvant treatment and those who did not—as shown in Figure 1, and this classification was incorporated into the survival analysis. But no statistically significant differences were detected. If a more detailed investigation of adjuvant therapy is warranted, we propose stratifying spinal DMG patients into multiple groups based on histological grade to evaluate the prognostic effects of radiotherapy and chemotherapy within each subtype. This approach may provide clearer insights. However, despite being one of the larger spinal DMG centers, our sample size for analyses of radiotherapy and chemotherapy stratified by histological grade remains limited. Therefore, future multicenter studies are needed to validate these findings.