Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD, 21287, USA.
Department of Neurosurgery, Mayo Clinic, Rochester, MN, 55905, USA.
J Neurooncol. 2022 May;158(1):117-127. doi: 10.1007/s11060-022-04017-4. Epub 2022 May 11.
Surgical resection is considered standard of care for primary intramedullary astrocytomas, but the infiltrative nature of these lesions often precludes complete resection without causing new post-operative neurologic deficits. Radiotherapy and chemotherapy serve as potential adjuvants, but high-quality data evaluating their efficacy are limited. Here we analyze the experience at a single comprehensive cancer center to identify independent predictors of postoperative overall and progression-free survival.
Data was collected on patient demographics, tumor characteristics, pre-operative presentation, resection extent, long-term survival, and tumor progression/recurrence. Kaplan-Meier curves modeled overall and progression-free survival. Univariable and multivariable accelerated failure time regressions were used to compute time ratios (TR) to determine predictors of survival.
94 patients were included, of which 58 (62%) were alive at last follow-up. On multivariable analysis, older age (TR = 0.98; p = 0.03), higher tumor grade (TR = 0.12; p < 0.01), preoperative back pain (TR = 0.45; p < 0.01), biopsy [vs GTR] (TR = 0.18; p = 0.02), and chemotherapy (TR = 0.34; p = 0.02) were significantly associated with poorer survival. Higher tumor grade (TR = 0.34; p = 0.02) and preoperative bowel dysfunction (TR = 0.31; p = 0.02) were significant predictors of shorter time to detection of tumor growth.
Tumor grade and chemotherapy were associated with poorer survival and progression-free survival. Chemotherapy regimens were highly heterogeneous, and randomized trials are needed to determine if any optimal regimens exist. Additionally, GTR was associated with improved survival, and patients should be counseled about the benefits and risks of resection extent.
手术切除被认为是原发性脊髓星形细胞瘤的标准治疗方法，但这些病变的浸润性特征常常导致无法在不引起新的术后神经功能缺损的情况下进行完全切除。放疗和化疗可作为潜在的辅助治疗方法，但评估其疗效的高质量数据有限。在这里，我们分析了单一综合癌症中心的经验，以确定术后总生存期和无进展生存期的独立预测因素。
收集患者的人口统计学、肿瘤特征、术前表现、切除范围、长期生存和肿瘤进展/复发的数据。使用 Kaplan-Meier 曲线对总生存期和无进展生存期进行建模。使用单变量和多变量加速失效时间回归计算时间比 (TR)，以确定生存的预测因素。
共纳入 94 例患者，其中 58 例（62%）在最后一次随访时仍存活。多变量分析显示，年龄较大（TR=0.98；p=0.03）、肿瘤级别较高（TR=0.12；p<0.01）、术前背痛（TR=0.45；p<0.01）、活检[与 GTR]（TR=0.18；p=0.02）和化疗（TR=0.34；p=0.02）与生存率较差显著相关。肿瘤分级较高（TR=0.34；p=0.02）和术前肠道功能障碍（TR=0.31；p=0.02）是肿瘤生长检测时间较短的显著预测因素。
肿瘤分级和化疗与生存率和无进展生存率较差相关。化疗方案高度异质，需要进行随机试验以确定是否存在任何最佳方案。此外，GTR 与生存率的提高相关，应向患者提供关于切除范围的益处和风险的咨询。
