-
替莫唑胺(temozolomide, TMZ)是初诊或复发性胶质瘤患者的重要化疗药物[1]。O6-甲基鸟嘌呤-DNA甲基转移酶(O6-methylguanine-DNA methyl-transferase, MGMT)基因甲基化可使胶质瘤患者对TMZ治疗产生更好的临床反应,特别是能显著延长患者的生存期[2]。因此,MGMT基因甲基化状态的评估对于胶质瘤患者治疗方案的选择至关重要。
18F-FDG PET/CT已被广泛应用于神经胶质瘤的诊断。葡萄糖代谢可以直接反映胶质瘤的代谢特征,并进一步指示胶质瘤的生物学特征,如肿瘤细胞的增殖状态、胶质瘤的分级和内在的血管分布等[3-4]。尽管术中组织病理学检查仍是诊断胶质瘤的“金标准”,但利用分子探针来探索显像剂与胶质瘤分子特征之间的相关性逐渐成为研究焦点。已有研究结果发现,通过分析18F-FDG的代谢指标[肿瘤与正常组织摄取率(tumor-to-normal-tissue uptake ratio,TNR)]能够预测MGMT基因的甲基化状态[5]。
影像组学通过提取ROI高通量的图像特征,能将肉眼不可见的影像信息转化为多维可挖掘的空间特征数据[6]。相关研究结果表明,影像组学的研究方法相较于传统的手动勾画ROI,能够从图像中获取更多有用的信息,此类信息对于提高确诊率和预测预后至关重要[7-8],特别是近年来,因其可无创、重复且整体分析肿瘤内部异质性而愈发受到重视[9-10]。
据笔者所知,迄今尚未有已发表的研究通过分析18F-FDG PET/CT图像的纹理特征来探索胶质瘤的特征与MGMT基因甲基化状态之间的关系。本研究尝试以此来综合评估MGMT基因甲基化状态,以预测胶质瘤患者对TMZ治疗的反应,并探索胶质瘤的多种生物学特性,以服务于临床诊疗。
-
17例患者中,2例(11.8%)经组织病理学结果证实为WHO国际肿瘤组织学分类Ⅰ级、1例(5.9%)为Ⅲ级、14例(82.4%)为Ⅳ级胶质瘤。9例(52.9%)患者的MGMT基因甲基化(甲基化组)、8例(47.1%)患者的MGMT基因未甲基化(未甲基化组)。甲基化组与未甲基化组比较,患者的年龄和肿瘤分级的差异无统计学意义(t=−0.251、−0.016, P=0.806、0.198);而性别之间的差异有统计学意义(t=−1.426,P=0.031)(表1)。MGMT基因甲基化组的SUVmax高于MGMT基因未甲基化组的SUVmax,且差异有统计学意义(t=−3.095,P=0.007);MGMT基因甲基化组的TNRmax也高于MGMT基因未甲基化组,差异有统计学意义(t=−3.402,P=0.004)。18F-FDG PET/CT图像的常规半定量指标、直方图特征及纹理特征间的差异均无统计学意义(表1)。典型病例的PET/CT图像见图1,MGMT基因甲基化组与未甲基化组胶质瘤患者18F-FDG PET图像的SUVmax和TNRmax比较见图2。
MGMT基因未甲基化组(n=8) MGMT基因甲基化组(n=9) t值 P值 临床指标 年龄(岁) 45.7±19.6 47.9±16.3 −0.251 0.806 性别(男∶女,例) 8∶0 5∶4 −1.426 0.031 肿瘤级别(Ⅰ∶Ⅱ∶Ⅲ∶Ⅳ,例) 2∶0∶0∶6 0∶0∶1∶8 −0.016 0.198 图像的常规特征 最大标准化摄取值 9.66±4.13 18.83±7.77 −3.095 0.007 平均标准化摄取值 4.83±2.19 6.20±2.98 −1.088 0.294 代谢肿瘤体积 39.71±62.87 49.34±77.73 −0.282 0.782 糖酵解总量 123.41±179.85 216.06±250.27 −0.884 0.390 TNRmax 1.20±0.52 2.37±0.87 −3.402 0.004 TNRmean 0.61±0.27 0.77±0.34 −1.044 0.313 图像的直方图特征 中位标准化摄取值 4.66±2.11 6.03±2.98 −1.098 0.289 Percentile 5th 3.00±1.73 2.57±1.10 0.599 0.558 Percentile 95th 7.20±3.12 10.41±5.29 −1.548 0.142 Skewness 0.52±0.32 0.50±0.63 0.081 0.937 Kurtosis 0.25±1.27 0.49±2.91 −0.224 0.826 图像的纹理特征 DiffEntropy 0.93±0.37 0.96±0.50 −0.161 0.874 DiffVariance 0.27±0.16 0.42±0.46 −0.879 0.404 Contrast 0.84±0.51 1.58±2.12 −0.956 0.368 Entropy 1.76±0.72 1.81±0.73 −0.159 0.876 注:表中,FDG:氟脱氧葡萄糖;PET/CT:正电子发射断层显像计算机体层摄影术;MGMT:O6-甲基鸟嘌呤-DNA甲基转移酶;TNRmax:最大肿瘤与正常组织摄取率;TNRmean:平均肿瘤与正常组织摄取率; Percentile 5t、Percentile 95th、Skewness、Kurtosis为图像的直方图特征参数;DiffEntropy、DiffVariance、Contrast、Entropy为图像的纹理特征参数 表 1 17例胶质瘤患者的临床特征及18F-FDG PET/CT图像特征(
±s)$\bar x $ Table 1. Clinical and 18F-FDG PET/CT imaging characteristics of 17 glioma patients (
±s)$\bar x $
18F-FDG PET/CT图像的影像组学分析在胶质瘤MGMT基因甲基化状态评估中的初步应用
Radiomic analysis of 18F-FDG PET/CT images in the evaluation of the MGMT methylation status in gliomas
-
摘要:
目的 探讨18F-氟脱氧葡萄糖(FDG) PET/CT图像的影像组学分析综合评估O6-甲基鸟嘌呤-DNA甲基转移酶(MGMT)基因甲基化状态,预测胶质瘤患者对替莫唑胺(TMZ)治疗的反应。 方法 回顾性分析2016年1月至2018年9月在中国人民解放军总医院第一医学中心经组织病理学结果证实的17例胶质瘤患者,其中男性13例、女性4例,患者均于术前进行了18F-FDG PET/CT检查,手动勾画肿瘤的感兴趣体积(VOI)并进行纹理分析。通过焦磷酸测序法检测、分析MGMT基因甲基化状态。根据MGMT基因甲基化状态分将患者为甲基化组和未甲基化组,采用两独立样本t检验分析两组数据之间各个影像组学参数的差异。 结果 17例胶质瘤患者中,9例(52.9%)MGMT基因未甲基化、8例(47.1%)MGMT基因甲基化。甲基化组与未甲基化组比较,患者的年龄和肿瘤分级的差异无统计学意义(t=−0.251、−0.016,P=0.806、0.198);而性别之间的差异有统计学意义(t=−1.426,P=0.031)。MGMT基因甲基化组的最大标准化摄取值(SUVmax)、最大肿瘤与正常组织摄取率(TNRmax)显著高于MGMT基因未甲基化组(SUVmax:18.83±7.77 对 9.66±4.13,t=−3.095,P=0.007;TNRmax:2.37±0.87对1.20±0.52,t=−3.402,P=0.004)。 结论 18F-FDG PET/CT图像的SUVmax和TNRmax可能是评估胶质瘤MGMT基因甲基化状态的关键指标,或许可用于预测TMZ化疗患者的临床反应。 -
关键词:
- 正电子发射断层显像术 /
- 体层摄影术,X线计算机 /
- 胶质瘤 /
- MGMT基因甲基化状态 /
- 影像组学分析
Abstract:Objective To explore the relationship between comprehensive information of gliomas and O6-methylguanylmethyltransferase (MGMT) promoter methylation status non-invasively by analyzing radiomic features of multi-modality 18F-fluorodeoxyglucose (FDG) PET/CT images. The response to temozolomide (TMZ) was determined through the abovementioned method for the clinical management of glioma patients. Methods A retrospective study of 17 patients (13 males and 4 females) with glioma confirmed by histopathological results in the First Medical Center of General Hospital of Chinese PLA from January 2016 to September 2018 was conducted. Preoperative 18F-FDG PET/CT scanning was performed. Radiomic texture analysis was performed after manually delineating the volume of interest. MGMT promoter methylation was examined by pyrosequencing analysis. MGMT data were categorized according to the methylation status, i.e., methylated and unmethylated groups. Two independent sample t-tests were used to analyze the differences in imaging omics parameters between the two groups of data. Results Among the 17 patients with glioma, 9 (52.9%) had MGMT unmethylation and 8 (47.1%) had MGMT methylation. Between the methylated group and the unmethylated group, there was no significant difference in patient age or tumor grade (t=−0.251, −0.016, P=0.806, 0.198); The difference between genders was statistically significant Meaning (t=−1.426, P=0.031). Both the SUVmax and TNRmax values of the MGMT methylated group were significantly higher than those of the MGMT unmethylated group (SUVmax: 18.83±7.77 vs. 9.66±4.13; t=−3.095, P=0.007; TNRmax: 2.37±0.87 vs. 1.20±0.52; t=−3.402, P=0.004). Conclusion The features (SUVmax and TNRmax) of 18F-FDG PET/CT images are two key indicators in the detection of MGMT methylation status in gliomas and are valuable predictors of the clinical responses of patients scheduled to receive TMZ chemotherapeutics. -
表 1 17例胶质瘤患者的临床特征及18F-FDG PET/CT图像特征(
±s)$\bar x $ Table 1. Clinical and 18F-FDG PET/CT imaging characteristics of 17 glioma patients (
±s)$\bar x $ MGMT基因未甲基化组(n=8) MGMT基因甲基化组(n=9) t值 P值 临床指标 年龄(岁) 45.7±19.6 47.9±16.3 −0.251 0.806 性别(男∶女,例) 8∶0 5∶4 −1.426 0.031 肿瘤级别(Ⅰ∶Ⅱ∶Ⅲ∶Ⅳ,例) 2∶0∶0∶6 0∶0∶1∶8 −0.016 0.198 图像的常规特征 最大标准化摄取值 9.66±4.13 18.83±7.77 −3.095 0.007 平均标准化摄取值 4.83±2.19 6.20±2.98 −1.088 0.294 代谢肿瘤体积 39.71±62.87 49.34±77.73 −0.282 0.782 糖酵解总量 123.41±179.85 216.06±250.27 −0.884 0.390 TNRmax 1.20±0.52 2.37±0.87 −3.402 0.004 TNRmean 0.61±0.27 0.77±0.34 −1.044 0.313 图像的直方图特征 中位标准化摄取值 4.66±2.11 6.03±2.98 −1.098 0.289 Percentile 5th 3.00±1.73 2.57±1.10 0.599 0.558 Percentile 95th 7.20±3.12 10.41±5.29 −1.548 0.142 Skewness 0.52±0.32 0.50±0.63 0.081 0.937 Kurtosis 0.25±1.27 0.49±2.91 −0.224 0.826 图像的纹理特征 DiffEntropy 0.93±0.37 0.96±0.50 −0.161 0.874 DiffVariance 0.27±0.16 0.42±0.46 −0.879 0.404 Contrast 0.84±0.51 1.58±2.12 −0.956 0.368 Entropy 1.76±0.72 1.81±0.73 −0.159 0.876 注:表中,FDG:氟脱氧葡萄糖;PET/CT:正电子发射断层显像计算机体层摄影术;MGMT:O6-甲基鸟嘌呤-DNA甲基转移酶;TNRmax:最大肿瘤与正常组织摄取率;TNRmean:平均肿瘤与正常组织摄取率; Percentile 5t、Percentile 95th、Skewness、Kurtosis为图像的直方图特征参数;DiffEntropy、DiffVariance、Contrast、Entropy为图像的纹理特征参数 -
[1] Weller M, Tabatabai G, Kastner B, et al. MGMT promoter methylation is a strong prognostic biomarker for benefit from dose-intensified temozolomide rechallenge in progressive glioblastoma: the DIRECTOR trial[J]. Clin Cancer Res, 2015, 21(9): 2057−2064. DOI: 10.1158/1078-0432.CCR-14-2737. [2] Weller M, Stupp R, Hegi ME, et al. Personalized care in neuro-oncology coming of age: why we need MGMT and 1p/19q testing for malignant glioma patients in clinical practice[J]. Neuro Oncol, 2012, 14(Suppl 4): iv100−iv108. DOI: 10.1093/neuonc/nos206. [3] Kim S, Chung JK, Im SH, et al. 11C-methionine PET as a prognostic marker in patients with glioma: comparison with 18F-FDG PET[J]. Eur J Nucl Med Mol Imaging, 2005, 32(1): 52−59. DOI: 10.1007/s00259-004-1598-6. [4] Okita Y, Kinoshita M, Goto T, et al. 11C-methionine uptake correlates with tumor cell density rather than with microvessel density in glioma: a stereotactic image-histology comparison[J]. Neuroimage, 2010, 49(4): 2977−2982. DOI: 10.1016/j.neuroimage.2009.11.024. [5] Choi H, Bang J-I, Cheon GJ, et al. 18F-Fluorodeoxyglucose and 11C-methionine positron emission tomography in relation to methyl-guanine methyltransferase promoter methylation in high-grade gliomas[J]. Nucl Med Commun, 2015, 36(3): 211−218. DOI: 10.1097/MNM.0000000000000236. [6] Zhou M, Hall L, Goldgof D, et al. Radiologically defined ecological dynamics and clinical outcomes in glioblastoma multiforne: preliminary results[J]. Transl Oncol, 2014, 7(1): 5−13. DOI: 10.1593/tlo.13730. [7] Zhou Y, He L, Huang YQ, et al. CT-based radiomics signature: a potential biomarker for preoperative prediction of early recurrence in hepatocellular carcinoma[J]. Abdom Radiol (NY), 2017, 42(6): 1695−1704. DOI: 10.1007/s00261-017-1072-0. [8] Zhang Q, Xiao Y, Suo JF, et al. Sonoelastomics for breast tumor classification: a radiomics approach with clustering-based feature selection on sonoelastography[J]. Ultrasound Med Biol, 2017, 43(5): 1058−1069. DOI: 10.1016/j.ultrasmedbio.2016.12.016. [9] Padma MV, Said S, Jacobs M, et al. Prediction of pathology and survival by FDG PET in gliomas[J]. J Neurooncol, 2003, 64(3): 227−237. DOI: 10.1023/a:1025665820001. [10] Summers RM. Texture analysis in radiology: Does the emperor have no clothes?[J]. Abdom Radiol (NY), 2017, 42(2): 342−345. DOI: 10.1007/s00261-016-0950-1. [11] Xie H, Tubbs R, Yang B. Detection of MGMT promoter methylation in glioblastoma using pyrosequencing[J]. Int J Clin Exp Pathol, 2015, 8(2): 1790−1796. [12] Eoli M, Menghi F, Bruzzone MG, et al. Methylation of O6-methylguanine DNA methyltransferase and loss of heterozygosity on 19q and/or 17p are overlapping features of secondary glioblastomas with prolonged survival[J]. Clin Cancer Res, 2007, 13(9): 2606−2613. DOI: 10.1158/1078-0432.CCR-06-2184. [13] Aerts HJ, Velazquez ER, Leijenaar RT, et al. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach[J]. Nat Commun, 2014, 5: 4006. DOI: 10.1038/ncomms5006. [14] Eichinger P, Alberts E, Delbridge C, et al. Diffusion tensor image features predict IDH genotype in newly diagnosed WHO grade Ⅱ/Ⅲ gliomas[J]. Sci Rep, 2017, 7(1): 13396. DOI: 10.1038/s41598-017-13679-4. [15] Yu JH, Shi ZF, Lian YX, et al. Noninvasive IDH1 mutation estimation based on a quantitative radiomics approach for grade Ⅱ glioma[J]. Eur Radiol, 2017, 27(8): 3509−3522. DOI: 10.1007/s00330-016-4653-3. [16] Lohmann P, Stoffels G, Ceccon G, et al. Radiation injury vs. recurrent brain metastasis: combining textural feature radiomics analysis and standard parameters may increase 18F-FET PET accuracy without dynamic scans[J]. Eur Radiol, 2017, 27(7): 2916−2927. DOI: 10.1007/s00330-016-4638-2. [17] Colavolpe C, Metellus P, Mancini J, et al. Independent prognostic value of pre-treatment 18-FDG-PET in high-grade gliomas[J]. J Neurooncol, 2012, 107(3): 527−535. DOI: 10.1007/s11060-011-0771-6. [18] Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma[J]. N Engl J Med, 2005, 352(10): 987−996. DOI: 10.1056/NEJMoa043330. [19] Seo S, Hatano E, Higashi T, et al. Fluorine-18 fluorodeoxyglucose positron emission tomography predicts lymph node metastasis, P-glycoprotein expression, and recurrence after resection in mass-forming intrahepatic cholangiocarcinoma[J]. Surgery, 2008, 143(6): 769−777. DOI: 10.1016/j.surg.2008.01.010. [20] Yu CJ, Wan WX, Zhang B, et al. Evaluation of the relationship between [18F] FDG and P-glycoprotein expression: an experimental study[J]. Nucl Med Biol, 2012, 39(5): 671−678. DOI: 10.1016/j.nucmedbio.2011.12.007.