基于HOXA9 DNA甲基化的两位点联合预测模型可用于脑膜瘤进展风险的早期筛查

doi:10.12122/j.issn.1673-4254.2024.11.07

摘要/Abstract

摘要：

目的探讨同源异形盒基因（HOXs）在脑膜瘤中的甲基化模式，筛选对脑膜瘤复发风险分层有临床指导意义的HOXs基因，建立预测模型并评估其预测效能。方法利用GEO数据库下载脑膜瘤相关数据集。通过甲基化差异分析和ROC曲线分析筛选有预后评估价值的HOXs基因，再通过Cox回归分析、分子特征分析对特征基因的临床应用价值进行验证。进一步筛选差异CpG位点并评估其预测效能，通过Lasso-cox回归分析建立预测模型，根据cut off值将患者分成高、低风险组并进行分析。最后通过甲基化特异性PCR（MS-PCR）在细胞和组织水平验证差异CpG位点的甲基化水平，并纳入脑膜瘤组织样本验证该模型的预测效能。结果 HOXA9甲基化水平在脑膜瘤中显著上调（P<0.001），且具有较高诊断效能（AUC=0.884）。验证分析表明HOXA9甲基化是影响脑膜瘤患者总生存期的独立危险因素（P<0.01），与脑膜瘤恶性程度和不良预后正相关（P<0.05），且基于HOXA9甲基化水平的分组方法在预测患者复发和生存时间时比WHO分级精度更高。筛选出的CpG位点cg03217995和cg21001184对脑膜瘤诊断的AUC均大于0.8，预测脑膜瘤患者复发的AUC均大于0.6。构建的两位点联合预测模型cut off值为1.226，以此分组的患者临床特征均有显著性差异（P<0.001），并且该模型预测评分是脑膜瘤的独立预后因素（P<0.05）。MS-PCR结果显示，位点cg03217995和cg21001184甲基化水平在脑膜瘤细胞中升高（P<0.0001），在不同WHO分级患者间无统计学差异。临床样本分析表明联合模型有较高预测效能（AUC=0.857），预测状态与患者真实临床进展结果高度一致。结论 HOXA9甲基化是脑膜瘤预后不良的有效预测指标，基于其CpG位点的联合预测模型有望成为恶性进展风险病例早筛的新方法。

关键词: 脑膜瘤, HOXA9, DNA甲基化, CpG位点, 诊断, 预后

Abstract:

Objective To establish a recurrence risk prediction model for meningioma based on HOXA9 DNA methylation. Methods Meningioma-related datasets were downloaded from GEO database for screening homeobox genes (HOXs) with prognostic values using differential methylation and ROC curve analysis and Cox regression analysis. The differentially methylated CpG sites with high predictive efficacy were selected to establish the risk prediction model using Lasso-Cox regression analysis, based on which the patients were divided into high- and low-risk groups by the cutoff value. The methylation levels of CpG sites were verified at the cell and tissue levels using methylation-specific PCR (MS-PCR). Clinical meningioma tissue samples were used to validate the predictive efficacy of the model. Results HOXA9 methylation level was significantly up-regulated in meningiomas (P<0.001) and showed a high diagnostic efficiency (AUC=0.884) as an independent risk factor for overall survival (P<0.01) positively correlated with the degree of malignancy and poor prognosis of meningioma (P<0.05). Risk stratification by HOXA9 methylation was more accurate than WHO grading for predicting recurrence and patient survival time. The AUCs of the sites cg03217995 and cg21001184 were both above 0.8 for meningioma diagnosis and above 0.6 for predicting recurrence. The patients' clinical characteristics differed significantly between the high- and low-risk groups (P<0.001), and the prediction score of the model was an independent prognostic factor for meningioma (P<0.05). MS-PCR results showed that the methylation levels of the two sites increased significantly in meningioma cells. In clinical samples, the combined model showed a high prediction efficiency (AUC=0.857), and the predicted risk of progression was highly consistent with the patients' actual condition. Conclusion High HOXA9 methylation level is a predictor for poor prognosis of meningiomas, and the combined prediction model based on its CpG sites provides a new approach to early screening of meningioma patients at risk of progression.

Key words: meningioma, HOXA9, DNA methylation, CpG sites, diagnosis, prognosis

谭茹雪, 包晓樟, 韩亮, 李朝晖, 田男. 基于HOXA9 DNA甲基化的两位点联合预测模型可用于脑膜瘤进展风险的早期筛查[J]. 南方医科大学学报, 2024, 44(11): 2110-2120.

Ruxue TAN, Xiaozhang BAO, Liang HAN, Zhaohui LI, Nan TIAN. A two-site combined prediction model based on HOXA9 DNA methylation for early screening of risks of meningioma progression[J]. Journal of Southern Medical University, 2024, 44(11): 2110-2120.

图/表 7

表1 MS-PCR引物序列

Tab.1 Primer sequences for MS-PCR

Site	Primer sequence (5'-3')		Product length (bp)
cg03217995	M	F:TGGTGTTTTGTATAGGGGTATCG	95
	M	R:AACCCAATATTTCTCTTCCCCG	95
	U	F:AGGGTTTGGTGTTTTGTATAGGGGTATTG	107
	U	R:CTCCTAAACCCAAGATTTCTCTTCCCCA	107
cg21001184	M	F:GGTCGTGCGCGTTACGTGTTCGT	92
	M	R:AAATTATAACTACAAAACATCGA	92
	U	F:TTTATTGGTTGTGTGTGTTATGTGTTTGT	104
	U	R:CTATAAAAATTATAACTACAAAACATCAA	104

图2 HOXA9甲基化与脑膜瘤临床特征相关性分析

Fig.2 Correlation analysis of HOXA9 methylation and clinical characteristics of meningioma. A: Univariate Cox analysis and multivariate Cox analysis of HOXA9 methylation and clinical characteristics with overall survival (OS) of meningioma patients. B: HOXA9 methylation difference in meningioma patients with CDKN2A/B deletion and TRAF7/AKT1 mutation. C: Proportion of meningioma patients with NF2-SSV, NF2-CNV, Chr22q Loss, and Chr1q Amp in hypomethylation (Hypo) and hypermethylation (Hyper) groups and box plot of the differences in gene instability. D: Kaplan-Meier survival curves for recurrence-free survival (RFS) and OS in Hypo and Hyper groups. E: Alluvial chart of Choudhury classification and WHO grade in Hypo and Hyper groups. F: Box plots comparing activation of molecular signatures of proliferation between HOXA9 groups. G: Brier prediction analysis of RFS and OS in HOXA9 groups and WHO grade in meningiomas. *P<0.05, **P<0.01, ****P<0.0001.

图3 CpG位点的筛选

Fig.3 Screening of CpG sites. A: Volcano map of differential methylation of CpG sites. B: The location of CpG sited. C: Heat map of CpG sites. D: Differential methylation between cg03217995 and cg21001184 in meningiomas and meningeal tissues. E: ROC analysis of cg03217995 and cg21001184. ****P<0.0001.

图4 两位点联合预测模型的构建

Fig.4 Construction of the two-site combined prediction model. A: Heat map of CpG sites arranged by risk score. The relationship between the risk scores and patients' clinic pathological characteristics was evaluated (a, Wilcoxon test; b, One-way ANOVA test). B: RFS and OS survival curves grouped by risk score.

表2 两 CpG位点RFS建模风险评分的单因素和多因素分析

Tab.2 Univariate and multivariate analysis of two CpG sites RFS modeling risk scores

Characteristics	Total (n)	Multi score	Univariate analysis		Multivariate analysis
Characteristics	Total (n)	Multi score	HR (95% CI)	P	HR (95% CI)	P
Risk score	243	1.341±0.422	3.654 (1.586-8.420)	0.002	2.725 (1.184-2.269)	0.018
WHO grade
1	212	1.317±0.416	4.391 (2.504-7.700)	<0.0001	3.203 (1.257-8.161)	0.015
2	29	1.505±0.450
3	2	1.548±0.222
RT
Yes	29	1.499±0.390	4.382 (2.071-9.072)	<0.0001	1.282 (0.380-2.322)	0.689
No	214	1.320±0.423	4.382 (2.071-9.072)	<0.0001	1.282 (0.380-2.322)	0.689
Age (year)
≤60	139	1.304±0.393	0.934 (0.484-1.802)	0.838
>60	104	1.390±0.456	0.934 (0.484-1.802)	0.838

图5 两位点联合预测模型效能验证

Fig.5 Validation of the combined prediction model of the two sites. A: MS-PCR results of CpG sites cg03217995 and cg21001184 in meningioma cells and meningioma tissues. The density of each strip was quantified by imaging analysis, and the relative band density value was calculated as the ratio of methylation to methylation plus unmethylation (M/U+M). U: Unmethylated; M: Methylated; MM: molecular marker. B: ROC of CpG sites cg03217995 and cg21001184 for diagnosis of meningioma patients. C: Prediction of meningioma patients at CpG sites cg03217995 and cg21001184. P: Patient. ***P<0.001, ****P<0.0001.

表3 脑膜瘤组织样本临床信息和实验结果

Tab.3 Clinical information and experimental results of meningioma tissue samples

Case No.	WHO grade	Progression	β value		Risk score
Case No.	WHO grade	Progression	cg03217995	cg21001184	Risk score
1	1	no	0.389±0.018	0.233±0.022	0.708
2	1	no	0.428±0.016	0.445±0.008	1.008
3	1	yes (Invasion)	0.668±0.037	0.393±0.006	1.228
4	2	no	0.330±0.039	0.344±0.036	0.779
5	2	no	0.437±0.015	0.157±0.016	0.668
6	2	yes (Multiple recurrence)	0.588±0.019	0.639±0.029	1.420
7	3	no	0.492±0.034	0.169±0.011	0.741
8	3	no	0.696±0.016	0.393±0.008	1.238
9	3	no	0.614±0.024	0.731±0.013	1.561

参考文献 31

1	Ostrom QT, Gittleman H, Truitt G, et al. CBTRUS statistical report: primary brain and other central nervous system tumors diagnosed in the United States in 2011-2015[J]. Neuro Oncol, 2018, 20(): iv1-86.
2	Louis DN, Perry A, Wesseling P, et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary[J]. Neuro Oncol, 2021, 23(8): 1231-51.
3	Roehrkasse AM, Peterson JEG, Fung KM, et al. The discrepancy between standard histologic WHO grading of meningioma and molecular profile: a single institution series[J]. Front Oncol, 2022, 12: 846232.
4	Robert SM, Vetsa S, Nadar A, et al. The integrated multiomic diagnosis of sporadic meningiomas: a review of its clinical implications[J]. J Neurooncol, 2022, 156(2): 205-14.
5	Pellerino A, Bruno F, Palmiero R, et al. Clinical significance of molecular alterations and systemic therapy for meningiomas: where do we stand?[J]. Cancers, 2022, 14(9): 2256.
6	Maas SLN, Stichel D, Hielscher T, et al. Integrated molecular-morphologic meningioma classification: a multicenter retrospective analysis, retrospectively and prospectively validated[J]. J Clin Oncol, 2021, 39(34): 3839-52.
7	Nassiri F, Liu J, Patil V, et al. A clinically applicable integrative molecular classification of meningiomas[J]. Nature, 2021, 597(7874): 119-25.
8	Driver J, Hoffman SE, Tavakol S, et al. A molecularly integrated grade for meningioma[J]. Neuro Oncol, 2022, 24(5): 796-808.
9	Choudhury A, Magill ST, Eaton CD, et al. Meningioma DNA methylation groups identify biological drivers and therapeutic vulnerabilities[J]. Nat Genet, 2022, 54(5): 649-59.
10	Gehring WJ, Hiromi Y. Homeotic genes and the homeobox[J]. Annu Rev Genet, 1986, 20: 147-73.
11	Hu X, Wang Y, Zhang XY, et al. DNA methylation of HOX genes and its clinical implications in cancer[J]. Exp Mol Pathol, 2023, 134: 104871.
12	Di Vinci A, Brigati C, Casciano I, et al. HOXA7, 9, and 10 are methylation targets associated with aggressive behavior in meningiomas[J]. Transl Res, 2012, 160(5): 355-62.
13	Mamatjan Y, Voisin MR, Nassiri F, et al. Integrated molecular analysis reveals hypermethylation and overexpression of HOX genes to be poor prognosticators in isocitrate dehydrogenase mutant glioma[J]. Neuro Oncol, 2023, 25(11): 2028-41.
14	Ghisai SA, van Hijfte L, Vallentgoed WR, et al. Epigenetic landscape reorganization and reactivation of embryonic development genes are associated with malignancy in IDH-mutant astrocytoma[J]. bioRxiv, 2024: 2024.03.19.585212.
15	Kurscheid S, Bady P, Sciuscio D, et al. Chromosome 7 gain and DNA hypermethylation at the HOXA10 locus are associated with expression of a stem cell related HOX-signature in glioblastoma[J]. Genome Biol, 2015, 16(1): 16.
16	Schmitz RJ, Lewis ZA, Goll MG. DNA methylation: shared and divergent features across eukaryotes[J]. Trends Genet, 2019, 35(11): 818-27.
17	Constâncio V, Nunes SP, Henrique R, et al. DNA methylation-based testing in liquid biopsies as detection and prognostic biomarkers for the four major cancer types[J]. Cells, 2020, 9(3): 624.
18	Palanca-Ballester C, Rodriguez-Casanova A, Torres S, et al. Cancer epigenetic biomarkers in liquid biopsy for high incidence malignancies[J]. Cancers, 2021, 13(12): 3016.
19	Ibrahim J, Peeters M, van Camp G, et al. Methylation biomarkers for early cancer detection and diagnosis: current and future perspectives[J]. Eur J Cancer, 2023, 178: 91-113.
20	房明浩, 刘文励, 孟凡凯, 等. 白血病细胞HOX A基因亚群启动子区域异常甲基化的研究[J]. 中华血液学杂志, 2009, 30(7): 468-72.
21	Li SY, Wu HC, Mai HF, et al. Microarray-based analysis of whole-genome DNA methylation profiling in early detection of breast cancer[J]. J Cell Biochem, 2019, 120(1): 658-70.
22	Cheng WJ, Jiang Y, Liu CX, et al. Identification of aberrant promoter hypomethylation of HOXA10 in ovarian cancer[J]. J Cancer Res Clin Oncol, 2010, 136(8): 1221-7.
23	Na FF, Pan XY, Chen JY, et al. KMT2C deficiency promotes small cell lung cancer metastasis through DNMT3A-mediated epigenetic reprogramming[J]. Nat Cancer, 2022, 3(6): 753-67.
24	Lin ZH, Luo MZ, Chen XQ, et al. Combined detection of plasma ZIC1, HOXD10 and RUNX3 methylation is a promising strategy for early detection of gastric cancer and precancerous lesions[J]. J Cancer, 2017, 8(6): 1038-44.
25	Guo YL, Peng YJ, Gao D, et al. Silencing HOXD10 by promoter region hypermethylation activates ERK signaling in hepatocellular carcinoma[J]. Clin Epigenetics, 2017, 9: 116.
26	Han Y, Tu WW, Wen YG, et al. Identification and validation that up-expression of HOXA13 is a novel independent prognostic marker of a worse outcome in gastric cancer based on immunohistochemistry[J]. Med Oncol, 2013, 30(2): 564.
27	Wu YS, Jiang GL, Zhang N, et al. HOXA9, PCDH17, POU4F2, and ONECUT2 as a urinary biomarker combination for the detection of bladder cancer in Chinese patients with hematuria[J]. Eur Urol Focus, 2020, 6(2): 284-91.
28	Kim YJ, Yoon HY, Kim JS, et al. HOXA9, ISL1 and ALDH1A3 methylation patterns as prognostic markers for nonmuscle invasive bladder cancer: array-based DNA methylation and expression profiling[J]. Int J Cancer, 2013, 133(5): 1135-42.
29	Clark VE, Erson-Omay EZ, Serin A, et al. Genomic analysis of non-NF2 meningiomas reveals mutations in TRAF7, KLF4, AKT1, and SMO[J]. Science, 2013, 339(6123): 1077-80.
30	Sahin B, Katar S, Şahin SA, et al. Influence of human telomerase reverse transcriptase mutation on the aggressiveness and recurrence in meningiomas[J]. Cureus, 2021, 13(5): e15342.
31	彭怡琛, 郑义, 李生兰, 等. 高级别脑膜瘤中异常甲基化调控相关基因识别研究[J]. 首都医科大学学报, 2023, 44(5): 747-52.

[1]	李英博, 包寒, 张森, 孟靖昕. 循环肿瘤细胞的响应性分离、释放和临床应用研究进展[J]. 南方医科大学学报, 2024, 44(9): 1637-1644.
[2]	陈孝华, 鲁辉, 王子良, 王炼, 夏勇生, 耿志军, 张小凤, 宋雪, 王月月, 李静, 胡建国, 左芦根. ABI2在胃癌进展和预后中的作用及其调控机制[J]. 南方医科大学学报, 2024, 44(9): 1653-1661.
[3]	叶梦楠, 武鸿美, 梅琰, 张庆玲. CREM在胃癌中高表达并与患者的不良预后相关[J]. 南方医科大学学报, 2024, 44(9): 1776-1782.
[4]	纪凯, 于冠宇, 周乐其, 张天帅, 凌潜龙, 满文江, 朱冰, 张卫. HNRNPA1基因在结直肠癌组织中高表达及其潜在的诊断和治疗价值[J]. 南方医科大学学报, 2024, 44(9): 1685-1695.
[5]	刘鹏程, 娄丽娟, 刘霞, 王建, 姜颖. M2巨噬细胞特征基因风险评分能准确预测HBV相关肝细胞癌患者的预后[J]. 南方医科大学学报, 2024, 44(5): 827-840.
[6]	杨晶晶, 殷丽霞, 段婷, 牛民主, 何震东, 陈心蕊, 张小凤, 李静, 耿志军, 左芦根. 胃癌组织中高表达ATP5A1与患者术后的不良预后和肿瘤细胞的糖代谢有关[J]. 南方医科大学学报, 2024, 44(5): 974-980.
[7]	高志强, 林洁, 洪鹏, 胡再宏, 董军君, 石秦林, 田小毛, 刘丰, 魏光辉. 基于高通量 RNA 测序分析 Wilms 瘤中关键基因对预后及免疫应答的影响[J]. 南方医科大学学报, 2024, 44(4): 727-738.
[8]	沈梦迪, 赵娜, 邓晓晶, 邓敏. COX6B2在胃癌组织中高表达并影响患者的远期预后：基于抑制p53信号调控胃癌细胞的增殖及细胞周期[J]. 南方医科大学学报, 2024, 44(2): 289-297.
[9]	张诺, 张震, 张雨路, 宋雪, 张小凤, 李静, 左芦根, 胡建国. PCID2在胃癌组织中高表达并通过调控细胞周期进程和增殖影响患者预后[J]. 南方医科大学学报, 2024, 44(2): 324-332.
[10]	张文静, 张诺, 杨子, 张小凤, 孙奥飞, 王炼, 宋雪, 耿志军, 李静, 胡建国. BZW1 高表达促进胃癌细胞的侵袭和转移：基于调控Wnt//β-catenin通路和促进上皮间质转化[J]. 南方医科大学学报, 2024, 44(2): 354-362.
[11]	黄晓茵, 陈凤莲, 张煜, 梁淑君. 多参数多区域MRI影像组学特征与临床信息联合模型可有效预测脑胶质瘤患者生存期[J]. 南方医科大学学报, 2024, 44(10): 2004-2014.
[12]	段婷, 张震, 施金冉, 肖林雨, 杨晶晶, 殷丽霞, 张小凤, 耿志军, 陆国玉. CPNE3在胃癌中高表达并与患者的预后不良相关[J]. 南方医科大学学报, 2024, 44(1): 129-137.
[13]	姚倚钠, 刘佳, 周想军, 刘泽宇, 邱士珍, 何颖政, 周雪琼. TTC9A在泛癌中的表达水平与多种癌症的预后和免疫微环境相关[J]. 南方医科大学学报, 2024, 44(1): 70-82.
[14]	李莹, 王倩, 陈小鸟, 席悦, 杨建, 刘晓敏, 王远大, 张利, 蔡广研, 陈香美, 董哲毅. 基于糖尿病视网膜病变的诊断模型对糖尿病肾病有较好诊断效能[J]. 南方医科大学学报, 2023, 43(9): 1585-1590.
[15]	王炼, 夏勇生, 张震, 刘馨悦, 施金冉, 王月月, 李静, 张小凤, 耿志军, 宋雪, 左芦根. 高表达MRPL13促进胃癌细胞增殖并影响患者预后：基于抑制p53信号[J]. 南方医科大学学报, 2023, 43(9): 1558-1566.