Abstract：
OBJECTIVE: Controversy surrounds the prognostic value of contrast-enhanced T1-weighted (T1CE) imaging-based subventricular zone (SVZ) classification in isocitrate dehydrogenase (IDH)-wildtype glioblastomas (GBMs). In this study, the authors aimed to assess the potential of incorporating FLAIR imaging into T1CE imaging-based classification for improving prognostic accuracy.
METHODS: A retrospective analysis was conducted on 281 patients with IDH-wildtype GBM. T1CE imaging-based classification was performed, and T2-weighted/FLAIR imaging was integrated to evaluate its prognostic estimation ability. Based on the relationship between the tumors and SVZ, patients were categorized into SVZ+ and SVZ- cohorts based on T1CE and T2-weighted/FLAIR imaging findings. Kaplan-Meier and Cox proportional hazards regression analyses were used to assess progression-free survival (PFS) and overall survival (OS), respectively. Patients were then categorized into three subgroups based on their combined classifications: group 1 (SVZ+ on T1CE and T2-weighted/FLAIR imaging), group 2 (SVZ- on T1CE but SVZ+ on T2-weighted/FLAIR imaging), and group 3 (SVZ- on T1CE and T2-weighted/FLAIR imaging). Subgroup analysis was used to evaluate differences in clinical and molecular factors as well as in prognoses.
RESULTS: The T1CE imaging-based classification failed to stratify OS between SVZ+ and SVZ- cohorts (16.0 vs 20.0 months, p = 0.36). Survival analysis revealed similar prognoses for patients in groups 1 and 2, and patients in group 2 exhibited worse OS compared with those in group 3 (19.0 vs 23.5 months, p = 0.024). Logistic regression identified lower Karnofsky Performance Status (KPS) (p = 0.011), tumor diameter (p = 0.002), and telomerase reverse transcriptase (TERT) promoter mutation (p = 0.003) to be associated with a higher incidence of group 2 GBMs. Additionally, T2-weighted/FLAIR imaging-based classification provided significant prognostic value (17.0 vs 23.5 months p = 0.021) and was found to be an independent prognostic factor in the Cox multivariate analysis (HR 1.79, 95% CI 1.08-2.96; p = 0.024).
CONCLUSIONS: This study underscores the limitations of T1CE imaging-based SVZ-associated classification in predicting prognosis for IDH-wildtype GBMs. The authors therefore propose an integrated approach that involves T2-weighted/FLAIR imaging that can provide improved prognostic ability. Notably, the presence of TERT promoter mutation was identified as a critical factor in nonenhancing tumor infiltration into the SVZ. Further validation through extensive cohort studies is recommended to confirm these findings.
METHODS: A retrospective analysis was conducted on 281 patients with IDH-wildtype GBM. T1CE imaging-based classification was performed, and T2-weighted/FLAIR imaging was integrated to evaluate its prognostic estimation ability. Based on the relationship between the tumors and SVZ, patients were categorized into SVZ+ and SVZ- cohorts based on T1CE and T2-weighted/FLAIR imaging findings. Kaplan-Meier and Cox proportional hazards regression analyses were used to assess progression-free survival (PFS) and overall survival (OS), respectively. Patients were then categorized into three subgroups based on their combined classifications: group 1 (SVZ+ on T1CE and T2-weighted/FLAIR imaging), group 2 (SVZ- on T1CE but SVZ+ on T2-weighted/FLAIR imaging), and group 3 (SVZ- on T1CE and T2-weighted/FLAIR imaging). Subgroup analysis was used to evaluate differences in clinical and molecular factors as well as in prognoses.
RESULTS: The T1CE imaging-based classification failed to stratify OS between SVZ+ and SVZ- cohorts (16.0 vs 20.0 months, p = 0.36). Survival analysis revealed similar prognoses for patients in groups 1 and 2, and patients in group 2 exhibited worse OS compared with those in group 3 (19.0 vs 23.5 months, p = 0.024). Logistic regression identified lower Karnofsky Performance Status (KPS) (p = 0.011), tumor diameter (p = 0.002), and telomerase reverse transcriptase (TERT) promoter mutation (p = 0.003) to be associated with a higher incidence of group 2 GBMs. Additionally, T2-weighted/FLAIR imaging-based classification provided significant prognostic value (17.0 vs 23.5 months p = 0.021) and was found to be an independent prognostic factor in the Cox multivariate analysis (HR 1.79, 95% CI 1.08-2.96; p = 0.024).
CONCLUSIONS: This study underscores the limitations of T1CE imaging-based SVZ-associated classification in predicting prognosis for IDH-wildtype GBMs. The authors therefore propose an integrated approach that involves T2-weighted/FLAIR imaging that can provide improved prognostic ability. Notably, the presence of TERT promoter mutation was identified as a critical factor in nonenhancing tumor infiltration into the SVZ. Further validation through extensive cohort studies is recommended to confirm these findings.
摘要：
目的：在异柠檬酸脱氢酶（IDH）-野生型胶质母细胞瘤（GBMs）中，基于对比增强T1加权（T1CE）成像的脑室下区（SVZ）分类的预后价值存在争议。在这项研究中,作者旨在评估将FLAIR成像纳入基于T1CE成像的分类以提高预后准确性的潜力.
方法：对281例IDH-野生型GBM患者进行回顾性分析。进行了基于T1CE成像的分类，并结合T2加权/FLAIR成像评估其预后评估能力。根据肿瘤与SVZ的关系,根据T1CE和T2加权/FLAIR成像结果,将患者分为SVZ+和SVZ-组.Kaplan-Meier和Cox比例风险回归分析用于评估无进展生存期（PFS）和总生存期（OS）。分别。然后根据其组合分类将患者分为三个亚组：第1组（T1CE和T2加权/FLAIR成像上的SVZ），第2组（T1CE上的SVZ-但T2加权/FLAIR成像上的SVZ+），和第3组（T1CE和T2加权/FLAIR成像上的SVZ）。亚组分析用于评估临床和分子因素以及预后的差异。
结果：基于T1CE成像的分类未能在SVZ和SVZ-队列之间对OS进行分层（16.0对20.0个月，p=0.36）。生存分析显示，第1组和第2组患者的预后相似，与第3组相比，第2组患者的OS较差（19.0vs23.5个月，p=0.024)。Logistic回归确定较低的Karnofsky绩效状态（KPS）（p=0.011），肿瘤直径（p=0.002），和端粒酶逆转录酶（TERT）启动子突变（p=0.003）与组2GBM的较高发生率相关。此外，基于T2加权/FLAIR成像的分类提供了显着的预后价值（17.0vs23.5个月，p=0.021），并且在Cox多变量分析中被发现是独立的预后因素（HR1.79，95％CI1.08-2.96；p=0.024）。
结论：这项研究强调了基于T1CE成像的SVZ相关分类在预测IDH野生型GBM预后方面的局限性。因此，作者提出了一种包括T2加权/FLAIR成像的综合方法，可以提供改善的预后能力。值得注意的是,TERT启动子突变的存在被确定为不增强肿瘤浸润进入SVZ的关键因素.建议通过广泛的队列研究进一步验证以证实这些发现。
方法：对281例IDH-野生型GBM患者进行回顾性分析。进行了基于T1CE成像的分类，并结合T2加权/FLAIR成像评估其预后评估能力。根据肿瘤与SVZ的关系,根据T1CE和T2加权/FLAIR成像结果,将患者分为SVZ+和SVZ-组.Kaplan-Meier和Cox比例风险回归分析用于评估无进展生存期（PFS）和总生存期（OS）。分别。然后根据其组合分类将患者分为三个亚组：第1组（T1CE和T2加权/FLAIR成像上的SVZ），第2组（T1CE上的SVZ-但T2加权/FLAIR成像上的SVZ+），和第3组（T1CE和T2加权/FLAIR成像上的SVZ）。亚组分析用于评估临床和分子因素以及预后的差异。
结果：基于T1CE成像的分类未能在SVZ和SVZ-队列之间对OS进行分层（16.0对20.0个月，p=0.36）。生存分析显示，第1组和第2组患者的预后相似，与第3组相比，第2组患者的OS较差（19.0vs23.5个月，p=0.024)。Logistic回归确定较低的Karnofsky绩效状态（KPS）（p=0.011），肿瘤直径（p=0.002），和端粒酶逆转录酶（TERT）启动子突变（p=0.003）与组2GBM的较高发生率相关。此外，基于T2加权/FLAIR成像的分类提供了显着的预后价值（17.0vs23.5个月，p=0.021），并且在Cox多变量分析中被发现是独立的预后因素（HR1.79，95％CI1.08-2.96；p=0.024）。
结论：这项研究强调了基于T1CE成像的SVZ相关分类在预测IDH野生型GBM预后方面的局限性。因此，作者提出了一种包括T2加权/FLAIR成像的综合方法，可以提供改善的预后能力。值得注意的是,TERT启动子突变的存在被确定为不增强肿瘤浸润进入SVZ的关键因素.建议通过广泛的队列研究进一步验证以证实这些发现。
