陳立堅

氫質(zhì)子磁共振波對顱內(nèi)轉(zhuǎn)移瘤與膠質(zhì)瘤鑒別及生存分析的意義

陳立堅*

目的 探討氫質(zhì)子磁共振波譜(1H magnetic resonance spectroscopy,1H-MRS)鑒別高級別腦膠質(zhì)瘤與肺癌腦轉(zhuǎn)移瘤的臨床鑒別診斷及其對生存預測的應用價值。材料與方法 回顧性分析經(jīng)手術(shù)病理確診及經(jīng)lH-MRS檢查的高級別腦膠質(zhì)瘤患者36例和肺癌腦轉(zhuǎn)移瘤患者43例，并計算1H-MRS檢查的N-乙酰天門冬氨酸(N-acetylaspartate, NAA)/膽堿(Choline, Cho)、NAA/肌酸(Creatine, Cr)和Cho/Cr的比值。根據(jù)回訪記錄的生存時間，采用Cox風險模型進行多因素分析篩選出對肺癌腦轉(zhuǎn)移瘤患者生存具有獨立預測作用的因素。結(jié)果 高級別腦膠質(zhì)瘤和肺癌腦轉(zhuǎn)移瘤的瘤周組織的Cho/NAA和Cho/Cr比值差異存在顯著性(P＜0.01)，經(jīng)Cox多因素回歸分析肺癌腦轉(zhuǎn)移瘤外周Cho/Cr是生存時間的獨立預測因素(P＜0.05)。結(jié)論 肺癌腦轉(zhuǎn)移瘤和高級別腦膠質(zhì)瘤瘤周組織間Cho/NAA和Cho/Cr 比值的差異可作為1H-MRS鑒別高級別腦膠質(zhì)瘤和肺癌腦轉(zhuǎn)移瘤的特征，肺癌腦轉(zhuǎn)移瘤外周Cho/Cr可能對評估患者生存時間有一定意義。

氫質(zhì)子磁共振波譜；腫瘤轉(zhuǎn)移；腦轉(zhuǎn)移瘤；高級別腦膠質(zhì)瘤；磁共振成像；N-乙酰天門冬氨；膽堿；肌酸

Received 21 Sep 2015, Accepted 30 Dec 2015

ACKNOW LEDGMENTSThis work was part of Qinzhou Municipal Science and Technology Bureau of self funding research project of Guangxi (No. 20135801).

近年來國內(nèi)外臨床研究就磁共振波譜(magneticresonance spectroscopy, MRS)鑒別孤立腦轉(zhuǎn)移瘤及高級別腦膠質(zhì)瘤多側(cè)重于腫瘤實質(zhì)的研究。氫質(zhì)子磁共振波譜(1H m agnetic resonance spectroscopy,1H-MRS)分析可反映出腦內(nèi)主要代謝物的水平，并且為無創(chuàng)性研究腦腫瘤物質(zhì)代謝特征的方法；再者由于膠質(zhì)瘤與腦轉(zhuǎn)移瘤的組織起源及瘤周病理的不同，本文通過回顧性分析探討瘤周1H-MRS特征在孤立腦轉(zhuǎn)移瘤與腦膠質(zhì)瘤的鑒別及預后評估的意義。

1 材料與方法

1.1 一般資料

回顧性收集2012年1月至2015年4月經(jīng)手術(shù)取得病理診斷為肺癌腦轉(zhuǎn)移瘤并術(shù)后進行相應處理及隨訪患者43例，腦膠質(zhì)瘤患者36例。所有患者MRS檢查前均未接受化療和/或放療，沒有腦外傷和/或腦手術(shù)史。MRS檢測采用德國西門子Sonata 1.5 T超導型磁共振儀和環(huán)形極化頭線圈。

1.2 檢查方法

使用1.5 T超導磁共振機(西門子MAGNETOM Essenza，德國)，以磁共振成像(magnetic resonance im aging, MRI)常規(guī)影像為基礎(chǔ)，在增強掃描后選擇掃描層面進行1H-MRS檢查，采用多體素氫質(zhì)子波譜腦檢查(p roton brain exam/m uitiple voxel, PROBE/SI)。波譜采樣體素定位盡量避開腦脊液、顱骨及液化壞死區(qū)，感興趣區(qū)(region o f interest, ROI)包括病灶區(qū)、周邊水腫帶和對側(cè)(或同側(cè))正常腦組織，層厚10 mm，掃描野(field of view, FOV)24 cm × 24 cm，激勵次數(shù)(number of excitation, NEX): l，相位矩陣18 × 18，抑水率＞96%，線寬(FMNH)值＜8 Hz。

1.3 圖像分析和數(shù)據(jù)測量方法

用波譜儀自帶軟件包FuneTool對2D1H-MRS檢查獲得信號進行后處理分析，由我科兩位經(jīng)驗豐富的MRI診斷醫(yī)師采取雙盲法分別進行圖像分析，選擇如下圖像定義特征：(1)瘤灶組織：在T1W I增強時強化的區(qū)域，代表腫瘤組織的實體部分。(2)瘤周水腫區(qū)：平掃T1W I呈低信號，T2W I呈高信號，無強化效應的緊鄰瘤灶區(qū)域邊緣20 mm以內(nèi)的區(qū)域。(3)對照區(qū)：對側(cè)相應位置的正常腦實質(zhì)。以掃描時ROI內(nèi)的體素作為單位，分別觀察和測量計算出瘤灶組織區(qū)、瘤周組織區(qū)和對側(cè)正常腦組織三處ROI體素內(nèi)代謝物N-乙酰天門冬氨酸(N-acety laspartate, NAA)、膽堿(Choline, Cho)及肌酸(Creatine, Cr)等物質(zhì)的波峰下面積積分，并計算出ROI體素內(nèi)NAA/Cho、NAA/Cr及Cho/Cr的比值。

1.4 統(tǒng)計學處理

采用SPSS 16.0軟件進行統(tǒng)計分析。對計數(shù)資料采用行×列χ2檢驗，對1H-MRS所得數(shù)據(jù)等資料兩樣本間的比較采用t檢驗。采用Cox風險模型進行多因素分析，篩選出對腦轉(zhuǎn)移生存具有獨立預測作用的因素。本研究所有的檢驗均采用雙側(cè)檢驗，P＜0.05為差異有統(tǒng)計學意義。

2 結(jié)果

2.1 患者一般資料比較

高級膠質(zhì)瘤組和腦轉(zhuǎn)移瘤組的年齡、性別、民族、病理類型及KPS評分比較差異無統(tǒng)計學意義(P＞0.05)，見表1。

2.2 腦轉(zhuǎn)移瘤和高級腦膠質(zhì)瘤的瘤灶代謝物比值

高級腦膠質(zhì)瘤和腦轉(zhuǎn)移瘤的瘤灶組織的Cho/NAA、Cho/Cr和NAA/Cr比值差異無統(tǒng)計學意義(P＞0.05)，見表2。

表1 高級膠質(zhì)瘤組和腦轉(zhuǎn)移瘤組患者一般臨床資料比較Tab. 1 General comparison of the clinical data in brain metastasis group and high-grade gliomas group

表2 腦轉(zhuǎn)移瘤組和高級膠質(zhì)瘤組的瘤灶代謝物比值比較(± s)Tab. 2 The comparison for the ratio of metabolites in brain metastasis group and high-grade gliomas group± s)

表2 腦轉(zhuǎn)移瘤組和高級膠質(zhì)瘤組的瘤灶代謝物比值比較(± s)Tab. 2 The comparison for the ratio of metabolites in brain metastasis group and high-grade gliomas group± s)

Group Cho/NAA NAA/Cr Cho/Cr High-grade gliomas 3.75 ± 0.55 1.08 ± 0.43 2.99 ± 0.75 Brain metastasis 3.62 ± 0.44 1.00 ± 0.37 3.22 ± 0.76 P value 0.2500 0.3799 0.1841

表3 高級別腦膠質(zhì)瘤組與腦轉(zhuǎn)移瘤組的瘤周代謝物的比值比較(± s)Tab. 3 The comparison for the ratio of metabolites in peritumoral area of brain metastasis group and high-grade gliomas group (± s)

表3 高級別腦膠質(zhì)瘤組與腦轉(zhuǎn)移瘤組的瘤周代謝物的比值比較(± s)Tab. 3 The comparison for the ratio of metabolites in peritumoral area of brain metastasis group and high-grade gliomas group (± s)

Group Cho/NAA NAA/Cr Cho/Cr High-grade gliomas 3.77 ± 0.58 1.19 ± 0.46 3.29 ± 0.80 Brain metastasis 3.26 ± 0.48 1.10 ± 0.35 2.62 ± 0.66 P value 0.0001 0.3303 0.0001

2.3 高級別腦膠質(zhì)瘤組與腦轉(zhuǎn)移瘤組的瘤周組織各代謝物比值的比較

高級別腦膠質(zhì)瘤和腦轉(zhuǎn)移瘤的瘤周組織的Cho/NAA和Cho/Cr比值差異存在顯著性(P＜0.01), NAA/Cr比值差異無統(tǒng)計學意義(P＞0.05)，見表3及圖1～3。

2.4 Cox多因素回歸分析

對包括年齡、性別、民族、病理類型、治療情況、KPS評分、外周Cho/NAA和外周Cho/Cr內(nèi)的8個變量進行Cox比例風險模型分析(Enter法)，結(jié)果顯示，KPS評分和外周Cho/Cr是肺癌腦轉(zhuǎn)移生存時間的獨立預測因素(P＜0.05)。而其他因素包括顯示年齡、性別、民族、病理類型、治療情況和外周Cho/NAA均不是肺癌腦轉(zhuǎn)移生存時間(P＞0.05)，見表4。

表4 Cox回歸分析結(jié)果Tab. 4 The results of analysis by Cox regression

圖1 肺癌腦轉(zhuǎn)移瘤健側(cè)波譜圖 圖2 肺癌腦轉(zhuǎn)移瘤瘤灶區(qū)波譜圖 圖3 肺癌腦轉(zhuǎn)移瘤瘤周區(qū)波譜圖(Cho的波峰無升高，Cho和NAA的波峰比健側(cè)低)Fig. 1 The wave spectrum in normal side of brain metastasis from lung cancer. Fig.2 The wave spectrum in tumor region of brain metastasis from lung cancer. Fig. 3 The wave spectrum in peritumoral area of brain metastasis from lung cancer (The rise of peak for Cho was not found, the peak of Cho and NAA was lower than those in normal side).

3 討論

近年來病理學研究發(fā)現(xiàn)腦轉(zhuǎn)移瘤為膨脹性生長，腫瘤組織血管內(nèi)皮與其腦外起源組織相似，無血腦屏障，通透性高，導致瘤周水腫明顯，瘤周的高T2信號主要是由血管源性水腫引起。臨床中也發(fā)現(xiàn)當腦轉(zhuǎn)移瘤被切除后，其原瘤周高T2信號可完全消失，證實了腦轉(zhuǎn)移瘤瘤周高T2信號由血管源性水腫引起，而非腫瘤浸潤。高級別膠質(zhì)瘤為浸潤性生長或兩種方式兼有，腫瘤組織因血腦屏障破壞而強化明顯，但瘤周的中樞神經(jīng)血管的通透性相對較低，瘤周高T2信號至少部分是由于腫瘤浸潤引起。高級別膠質(zhì)瘤腫瘤組織被切除后，原瘤周高T2信號區(qū)多不能消失，且常在該區(qū)域出現(xiàn)復發(fā)病灶，也證實瘤周有腫瘤細胞浸潤圓。Cho反映了腫瘤組織內(nèi)細胞膜轉(zhuǎn)換和細胞增殖，所以高級別膠質(zhì)瘤瘤周腫瘤細胞浸潤及腫瘤細胞增殖，導致出現(xiàn)瘤周Cho升高，而腦轉(zhuǎn)移瘤瘤周主要是血管源性水腫，從而導致Cho增高不明顯。本研究顯示高級別腦膠質(zhì)瘤和孤立腦轉(zhuǎn)移瘤的瘤灶組織的Cho/NAA、Cho/Cr和NAA/Cr比值差異無統(tǒng)計學意義，高級別腦膠質(zhì)瘤和孤立腦轉(zhuǎn)移瘤的瘤周組織的Cho/NAA和Cho/Cr比值差異存在顯著性，前者明顯高于后者。目前國內(nèi)外研究也證實高級別腦膠質(zhì)瘤的瘤周組織的Cho/NAA和Cho/C r比值高于孤立腦轉(zhuǎn)移瘤的瘤周組織的Cho/NAA和Cho/Cr比值[1-5]。由此可見，瘤周1HMRS將提高孤立腦轉(zhuǎn)移瘤和高級別腦膠質(zhì)瘤術(shù)前診斷的準確性。

惡性腫瘤一旦發(fā)生腦轉(zhuǎn)移，則預后極差，惡性腫瘤發(fā)生腦轉(zhuǎn)移的預后評估對臨床治療有積極意義。實際上，影響腦轉(zhuǎn)移瘤患者的預后的因素較為復雜，不僅受治療方法的影響，還與病人自身的因素(如：年齡、性別及全身狀態(tài))和腫瘤因素(如原發(fā)腫瘤的病理類型、腦轉(zhuǎn)移瘤情況、有無顱外轉(zhuǎn)移等)有關(guān)。在本研究中發(fā)現(xiàn)Cho/Cr比值對肺癌腦轉(zhuǎn)移患者的中位生存時間影響不同，其差別具有統(tǒng)計學意義(P=0.011)，Cho/Cr比值低的患者生存期較短。在惡性腫瘤患者的腦轉(zhuǎn)移瘤瘤周血管源性水腫，從而導致Cho增高不明顯[1-2]，Zakaria等[6]研究發(fā)現(xiàn)應用彌散加權(quán)成像(diffusion w eighted imaging, DW I)檢查腦轉(zhuǎn)移瘤周圍組織表面擴散系數(shù)與患者的預后有關(guān)。故Cho/Cr比值可能是腦轉(zhuǎn)移獨立的預后因素。在研究中研究病例數(shù)仍較少，Cho/Cr比值是否可以作為腦轉(zhuǎn)移獨立的預后評估因素仍有待增加樣本量研究。

本研究資料統(tǒng)計發(fā)現(xiàn)，在確診腦轉(zhuǎn)移時身體狀況各不相同、治療方法也有差異的狀況下，單因素分析顯示KPS評分≥70分的腦轉(zhuǎn)移瘤患者的中位生存時間(10個月)長于KPS評分＜70分的患者(7個月)，其差異有顯著統(tǒng)計學意義(P=0.022)。KPS評分是目前評估惡性腫瘤患者行為狀態(tài)常用的方法，可以用于反映患者的機體功能狀態(tài)。在多個學者的研究中[7-8]，KPS評分是影響腦轉(zhuǎn)移瘤患者生存期的重要因素，它代表了患者一定的腫瘤負荷情況及對治療的耐受能力。Rodrigus等人[9]在其研究中將KPS評分分為兩組(≥70分和＜70分)，結(jié)果顯示不同的KPS評分對肺癌腦轉(zhuǎn)移患者的中位生存時間影響不同，其差別具有統(tǒng)計學意義，KPS評分高的患者生存期較長，故得出KPS評分是腦轉(zhuǎn)移獨立的預后因素[10-11]。目前對于眾多可能影響腦轉(zhuǎn)移瘤患者預后的因素尚無統(tǒng)一的意見，各研究報道仍存有分歧[12-13]。本組資料顯示，60歲以上患者與60歲以下患者的生存期相似，女性與男性患者的生存期相仿，漢族和少數(shù)民族腦轉(zhuǎn)移瘤患者的生存期無明顯差異，病理類型與腦轉(zhuǎn)移瘤患者的預后無明顯差異。本研究對接受手術(shù)后行化療和分子靶向藥物治療進行了比較分析差異無統(tǒng)計學意義。但該結(jié)果存在的問題是，合并分子靶向治療病例數(shù)較少，且為回顧性資料，有其內(nèi)在的局限性，確切的效果有待于大量的隨機研究。

通過本研究表明，腦轉(zhuǎn)移瘤的瘤周組織1HMRS的Cho/NAA和Cho/Cr比值對在臨床中與高級別腦膠質(zhì)瘤鑒別有重要意義，另在本研究中發(fā)現(xiàn)Cho/Cr比值與患者的預后有關(guān)，是否可以作為腦轉(zhuǎn)移獨立的預后評估因素仍有待增加樣本量研究。

[References]

[1]Yu K, Liu Q, Hao LW, et al. Differential diagnosis of glioblastoma m ultiforme from m etastatic brain tum or using proton and diffusion-weighted image. J Med Imaging, 2014, 24(7): 1085-1088.

于柯, 劉強, 郝利緯, 等.1H-MRS和DW I在膠質(zhì)母細胞瘤與單發(fā)轉(zhuǎn)移瘤中的鑒別診斷價值. 醫(yī)學影像學雜志, 2014, 24(7): 1085-1088.

[2]B lanchet L, K rooshof PW, Postma GJ, et al. D iscrim ination between metastasis and glioblastoma multiforme based on morphometric analysis of MR images. AJNR Am J Neuroradiol,2011, 32(1): 67-73.

[3]Wang WW, Niu TL, M iao YW, et al. SWI study on astrocytoma grading and differential diagnosis of astrocytoma and solitary metastases. Chin J Magn Reson Imaging, 2015, 6(4): 246-252.

王微微, 牛田力, 苗延巍, 等. SW I對星形細胞瘤分級及與單發(fā)轉(zhuǎn)移瘤鑒別診斷的價值. 磁共振成像, 2015, 6(4): 246-252.

[4]Cheng JL, Yang T. Application and development of magnetic resonance im aging in glioma. Chin J M agn Reson Im aging, 2014, 5(S1): 62-67.

程敬亮, 楊濤. 磁共振成像在腦膠質(zhì)瘤中的應用及進展. 磁共振成像, 2014, 5(S1): 62-67.

[5]He JB, He GW, Geng DY, et al. Diagnostic value of MRS in singular brain metastasis from lung cancer. Journal of Medical Imaging, 2008, 18(1): 8-10.

何江波, 何光武, 耿道穎, 等. MRS在單發(fā)肺癌腦轉(zhuǎn)移瘤中的診斷價值. 醫(yī)學影像學雜志, 2008, 18(1): 8-10.

[6]Zakaria R, Das K, Radon M, et al．Diffusion-weighted MRI characteristics of the cerebral metastasis to brain boundary predicts patient outcomes. BMC Med Imaging, 2014, 3(14): 26.

[7]Caballero JA, Sneed PK, Lamborn KR, et al. Prognostic factors for survival in patients treated w ith stereotactic radiosurgery for recurrent brain metastases after prior whole brain radiotherapy. Int J Radiat Oncol Biol Phys, 2012, 83(1): 303-309.

[8]Jezierska D, Adamska K, Liebert W. Evaluation of results of linac-based radiosurgery for brain metastases from primary lung cancer. Rep Pract Oncol Radiother, 2013, 19(1): 19-29.

[9]Rodrigus P, de Brouwer P, Raaymakers E. Brain metastases and non-small cell lung cancer. Prognostic factors and correlation w ith survival after irradiation. Lung Cancer, 2001, 32(2): 129-136.

[10]K ress MA, Ramakrishna N, Makgoeng SB, et al. Physician selfreported treatment of brain metastases according to patients' clinical and demographic factors and physician practice setting. Radiat Oncol, 2012, 8(7): 188.

[11]Barnholtz-Sloan JS, Yu C, Sloan AE, et al. A nomogram for individualized estimation of survival among patients w ith brain metastasis. Neuro Oncol, 2012, 14(7): 910-918.

[12]Cui SH, Bai H, Dong LL, et al. Analysis of survival predictors in patients w ith lung cancer and brain metastases. Chinese Journal of Lung Cancer, 2015, 18(7): 436-442.

崔少華, 白皓, 董莉莉, 等. 肺癌腦轉(zhuǎn)移生存預測因素分析. 中國肺癌雜志, 2015, 18(7): 436-442.

[13]Tang WH, Alip A, Saad M, et al. Prognostic factors in patients w ith non-small cell lung carcinoma and brain metastases: a Malaysian perspective. Asian Pac J Cancer Prev, 2015, 16(5): 1901-1906.

The discrim ination between brain metastases from lung cancer and high-grade glioma with proton MR spectroscopy and its value for survival predictorsand

CHEN Li-jian*
Department of radiology, tenth A ffiliated Hospital of Guangxi Medical University, Qinzhou 535099, China

Objective: To investigate the clinical discrim ination value of proton MR spectroscopy(1H-MRS) between brain metastasis from lung cancer and high-grade glioma, and access its prognostic value at the same time. M aterials and M ethods: Thirty-six patients of high-grade gliomas and forty-three patients of brain metastasis from lung cancer were studied. Thirty-six patients of high-grade gliomas and fortythree patients of brain metastasis from lung cancer were received1H-MRS. The variety of metabolite peaks including n-acety laspartate (NAA), choline (Cho) and creatine (Cr) were detected in different areas including tumor, peritumoral areas and the normal brain, at the same time the ratio of NAA/Cho, NAA/Cr and Cho/Cr were computed. Cox regression were performed for univariate and multivariate analyses to explore independent predictors influencing the survival of patients w ith lung cancer and brain metastases respectively. Resu lts: Significant differences between peritumoral area of high-grade gliomas and brain metastasis from lung cancer were found in the ratios of Cho/Cr and Cho/NAA (P＜0.05). The ratios of NAA/Cr, Cho/Cr and NAA/Cho were independent survival predictors in patients w ith lung cancer and brain metastasis as indicated by Cox proportional hazard model(P＜0.05). Conclusion: The ratios of Cho/Cr and Cho/NAA in peritumoral area can be found in brain metastases from lung cancer w ith1H-MRS, which is helpful in discrim ination between lung cancer and highgrade gliomas. The ratios of Cho/Cr may be an independent survival predictors in patients with lung cancer and brain metastasis.

1H magnetic resonance spectroscopy; Neoplasm metastasis; Brain metastases; High-grade glioma; M agnetic resonance imaging; N-acetylaspartate; Choline; Creatine

廣西欽州市科技局自籌經(jīng)費科研課題(編號：20135801)

廣西醫(yī)科大學第十附屬醫(yī)院(欽州市第一人民醫(yī)院)放射科，欽州 535099

陳立堅，E-mail: clj0066@163.com

2015-09-21接受日期：2015-12-30

R445.2；R739.41

A

10.12015/issn.1674-8034.2016.04.007

陳立堅. 氫質(zhì)子磁共振波對顱內(nèi)轉(zhuǎn)移瘤與膠質(zhì)瘤鑒別及生存分析的意義.磁共振成像, 2016, 7(4): 277–281.

*Correspondence to: Chen LJ, E-mail: clj0066@163.com