【作者】 孙海基；

【导师】 安利国；

【作者基本信息】 山东师范大学 ， 动物学， 2009， 博士

【摘要】 目的肺癌是一种常见的肺部恶性肿瘤,其死亡率已占癌症死亡率之首。其中非小细胞肺癌(Non-small cell lung cancer, NSCLC)占肺癌85%。早期NSCLC患者一般手术切除,但30%-60%的患者仍会复发并可能死亡。临床研究已验证,顺铂(cisplatin)辅助化疗可以提高术后NSCLC的生存期。对NSCLC患者术后行辅助化疗目前已成为标准,但总生存率仅能提高4.1%左右,可见多数患者并未得益于术后化疗。筛选出可能从中获益者以使多数未获益者免受化疗带来的副作用,这将是当今肺癌研究的热点。随着分子生物技术的发展,肿瘤研究已进入分子水平时代。现阶段对于肿瘤的基础及临床研究分为分子机制、分子诊断、分子预后等几个方面。沿用至今的恶性肿瘤TNM分期在预后的估计上已证明有巨大的价值,同时也深刻地影响着癌症治疗决策,但TNM分期本身并不能从分子层次预测和揭示抗癌治疗的效果。因此迫切需要应用分子诊断技术,对肿瘤发生发展的病理学机制及生物学行为从分子水平上加以认识,在解决肿瘤异质性、分期的合理性、治疗方案的设计和预后估计的准确性上能提供更好的帮助。当前,由于一系列新技术的开发及应用(如基因芯片、组织芯片及蛋白芯片等),发现了许多肺癌新的分子标记或分子标记组合方式,可在TNM分期基础上,把分子生物学的最新研究成果结合到TNM分期中,为NSCLC的预后和治疗决策提供更有利的证据。因此,肿瘤的分子预后研究愈加受到临床的重视。一般认为,顺铂药物的毒性来自于cisplatin与DNA的结合。Cisplatin-DNA共价复合物影响细胞的基本过程,包括DNA修复、细胞周期阻滞及细胞凋亡。参与这些过程的基因突变是肿瘤发生关键因素,其中,TP53、KRAS和EGFR基因突变是肺癌中最常见的突变。TP53突变至少在50%的II-III期肺癌患者中发生,多数突变为错义突变;KRAS密码子12和13的突变在肺腺癌中的突变率为20%-25%,在其它NSCLC中突变率较低;EGFR在40%-80%的NSCLC中过表达,尤其在非吸烟女性肺腺癌中高发。本文,我们系统探讨了国际肺癌研究组(IALT)体系中,TP53、KRAS和EGFR基因突变是否可以做为NSCLC顺铂辅助化疗的分子标记物,以筛选受益于顺铂辅助化疗的NSCLC患者。方法样本783例样本为国际辅助肺癌研究组(International Adjuvant Lung Cancer Trial, IALT)的石蜡包埋组织,来自于14国家的28个中心。所设计标本已获伦理委员会使用许可,并按WHO 2004年分类标准进行分类。TP53基因突变检测及分析QIAgen DNA提取试剂盒提取783例石蜡包埋切片的基因组DNA。TP53外显子5至8经PCR扩增后,变性高压液相质谱仪(DHPLC)和测序法检测TP53外显子5至8的突变。初筛结果为阳性的标本,二次PCR扩增及测序以确定TP53突变结果。TP53突变密码子的分布及碱基变化特点与IARC TP53数据库已知数据进行比较分析。KRAS基因突变检测及分析PCR扩增KRAS外显子1 (含密码子12和13),产物经测序筛查KRAS突变。初筛结果为阳性的标本,二次PCR扩增及测序以确定KRAS的突变。KRAS的突变特点与COSMIC数据库相关资料进行对比分析。EGFR基因突变检测及分析直接测序法筛查EGFR基因外显子18至21存在的突变。初筛结果为阳性的标本,二次PCR扩增及测序以确定EGFR的突变。EGFR的突变特点与COSMIC数据库相关资料进行对比。统计学分析Kaplan--Meier法计算生存率,Log-Rank检验比较生存差别,利用多因素Cox回归模型分析分析TP53,KRAS和EGFR基因突变与其它因素的相关性,以确定TP53,KRAS和EGFR基因突变的预后和预测价值。运用SAS软件,版本8.2。结果1. TP53突变实验结果TP53突变率为46%(240/524),其密码子突变热点分布与IARC TP53数据报道的基本一致。IALT数据中,碱基变化和功能变化分布特点与以往报道中NSCLC的突变特点相一致。其中,G>T碱基变换(29.97%)为典型的吸烟患者的印迹突变。TP53基因突变与NSCLC临床病理特征相关性分析表明,TP53突变与年龄、性别、TNM分期和切片的质量具有相关性。五年生存期的分析资料显示,TP53突变对NSCLC整体无预后和预测意义。但是,存在一种趋势:顺铂辅助化疗有利于TP53野生型患者的生存,而有害于TP53突变型患者的生存(P=0.18总生存期和P=0.13无病生存期)。在5年生存期分析中,非腺癌非小细胞肺癌中TP53突变状态和顺铂治疗有临界值显著相关性(P=0.05,无病生存期和P=0.25,总生存期)。8年生存期分析结果与5年生存期的分析结果基本一致,并且TP53突变对非小细胞肺癌整体无病生存期组有预测意义(P=0.04),非腺癌非小细胞肺癌中TP53突变状态和顺铂治疗有显著相关性(P=0.02,无病生存期和P=0.09,总生存期)。2. KRAS突变实验结果KRAS的突变率为14%(98/718),其中90例突变发生在密码子12,8例突变发生在密码子13。KRAS在IALT肺癌中的突变模式与Cosmic数据库所报道的相一致,突变主要为G>T转换。KRAS基因突变与NSCLC临床病理特征相关因素相关性分析表明,KRAS突变与TNM分期、病理组织分型和切片的质量具相关性。五年生存期的分析资料显示,KRAS突变在NSCLC整体患者中有临界值预后意义(P=0.08总生存期和P=0.02无病生存期);在NSCLC整体中,KRAS突变无预测意义。腺癌中KRAS基因突变率明显高于非腺癌非小细胞肺癌(P<0.0001),KRAS基因突变状态与非腺癌患者患者预后明显相关(P=0.006,5年总生存期;P=0.0009,5年无疾病生存期)。3. EGFR突变实验分析结果在208例肺腺癌标本中,共检测到EGFR基因突变为22例(11%,22/208)。主要分布于外显子19和外显子21。EGFR基因突变特点,与COSMIC数据库报道一致。EGFR基因突变与NSCLC临床病理特征相关性分析表明,EGFR突变与临床病理隔膜浸入有相关性(P=0.01)。五年总生存期和无病生存期的分析资料显示,在肺腺癌整体中,EGFR突变无预后和预测意义(P>0.05)。IALT中腺癌样本量少(200例),突变率低(11%),因此需要进一步扩大实验数据进行研究。结论1. 5年生存期和8年生存期分析结果表明,TP53基因突变对NSCLC无预后意义。2. 8年无病生存期分析结果表明,TP53突变对NSCLC整体组有预测意义(P=0.04),顺铂辅助化疗有利于TP53野生型患者的生存,而有害于TP53突变型患者的生存。3. 5年生存期和8年生存期分析结果一致表明,非腺癌非小细胞肺癌中TP53突变状态和顺铂辅助化疗有显著相关性。4. KRAS突变对顺铂化疗NSCLC无预测意义,但在无病生存期分析中具有预后意义。5. KRAS在不同组织分型中具有预后意义,在非腺癌非小细胞肺癌中,预后最差。6. EGFR突变对顺铂化疗NSCLC既无预后作用,又无预测作用。由上结论我们可见,热点基因突变的检测可能有助于分选病人,筛选出可能从顺铂化疗中获益患者,以使多数未获益患者免受化疗带来的副作用。本文研究结果,将为TP53、KRAS和EGFR基因突变是否可以成为顺铂辅助化疗NSCLC分子标记物提供重要的理论依据。更多还原

【Abstract】 ObjectiveThe most common lung malignancy worldwide is the non-small cell lung cancer (NSCLC) occupied 85% of lung cancer, which is also one of the leading causes of cancer related deaths today. Although early-stage NSCLC patients are treated with complete resection, 30% to 60% will develop recurrence and die as a result of their disease. Clinical trials have tested the ability of adjuvant cisplatin-based chemotherapy to improve survival after complete resection of NSCLC. To date, adjuvant cisplatin-based chemotherapy for completely resected NSCLC patients is becoming a criterion. While an absolute benefit of 4.1% in 5-year overall survival is quite low, the majority couldn’t benefit from this chemotherapy. To assign the patients to get benefits from chemotherapy is becoming hot spots in lung cancer study.With the development of molecular bio-technique, the research on cancer has already studied on molecular level. The basic and clinical researchs on cancer include molecular mechanism, molecular diagnosis and molecular prognosis so on. Up to date, the only prospectively validated prognosis and predictive factors that can be used as a guide in the clinics is pathological stage (such as TNM stage). Patients with higher stage tumors with poor prognosis have the greatest potential to benefit from adjuvant cisplatin-based therapy. However, there is no accepted biomarker of prognosis or prediction of outcome. Evaluation of further molecular events associated with carcinogenesis is required to determine who among patients stands the best chances to benefit from this therapeutic approach. As a series of new techniques developing and putting into practices (such as gene array, tissue array and protein array, etc), many molecular biomarker alone or combination are found in lung cancer prognosis analysis. To combine TNM stage and molecular markers to provide more powerful data for NSCLC appropriate treatment protocols. Now more and more attentions are focus on the prognosis and predictive value of bio-marker on lung cancer chemotherapy.It is generally accepted that cytotoxicity of cisplatin is mediated through induction of apoptosis and arrest of cell cycle resulting from its interaction with DNA, such as the formations of cisplatin-DNA adducts, which activates multiple signaling pathways, including those involving p53, etc. Increased expression of anti-apoptotic genes and mutations in the intrinsic apoptotic pathway may contribute to the inability of cells to detect DNA damage or to induce apoptosis. Of the genes that are frequently mutated in lung cancers, TP53, KRAS and EGFR are among the most common. TP53 mutations occur in at least 50% of stage II-III lung cancers. Most mutations are missense, turning the p53 protein into a transcriptionally inactive form that often accumulates within cells. In contrast, KRAS mutations at codon 12 and 13 occur in about 20-25% of adenocarcinomas and in a low proportion on other NSCLC. And EGFR mutations at Exon18 and Exon21 were found in around 40-80% of NSCLC, especially in adenocarcinoma with no smoking in women. In this study, we have investigated whether mutations in TP53, KRAS or EGFR could provide prognostic indications or predict a survival benefit from cisplatin-based adjuvant chemotherapy in completely resected non–small-cell lung cancer in international adjuvant lung cancer trial (IALT).MethodsSpecimens 783 paraffin-embedded tumor samples had participated in the IALT, which compared adjuvant cisplatin-based chemotherapy with observation among patients with non–small cell lung cancer. Twenty-eight centers in 14 countries contributed specimens. Approval was obtained from the local institutional review boards, according to the legal regulations in each participating country. All tumors were reviewed according to the histopathological classification system adopted by the World Health Organization (WHO) in 2004.Detection and analysis of TP53 Mutations Mutation analyses were conducted using genomic DNA isolated from 783 paraffin-embedded archived tissue sections. DNA was extracted by standard QIAamp DNA extraction Kit (QIAGEN S.A.) and exons5 to 8 of TP53 were amplified by polymerase chain reaction (PCR) using their respective primer sequences. PCR products were directly DHPLC or sequenced. PCR products with mutation were reanalyzed by independent PCR and sequencing to confirm the presence and nature of mutations for TP53. The codon distribution and pattern of TP53 mutations were compared with those of the IARC TP53 database, which compiles all mutations described in TP53 in the literature (IARC TP53 database).Detection and analysis of KRAS Mutations Exon1 of KRAS were amplified by polymerase chain reaction (PCR) using its primer sequences. KRAS mutations at codon 12 and 13 were analyzed by directly sequencing and confirmed by sequencing or Restriction Fragment length Polymorphism (RFLP). The prevalence and patterns of KRAS mutations were compared with those of the COSMIC database, which compiles all mutations described in KRAS in the literature (COSMIC database).Detection and analysis of EGFR Mutations Exon18 to 21 of EGFR were amplified by polymerase chain reaction (PCR) using their respective primer sequences. EGFR mutations at exons18 to 21 were analyzed by sequencing and confirmed by a separately sequencing.The prevalence and patterns of EGFR mutations were compared with those of the COSMIC database, which compiles all mutations described in EGFR in the literature (COSMIC database).Statistical Analyses The prognostic values of TP53,KRAS and EGFR status and chemotherapy for survival were studied using a Cox model in order to incorporate factors related to survival.Survival rate was studied by Kaplan—Meier method. All other factors that were statistically related to the TP53,KRAS and EGFR status in the multivariate Log-Rank (P <0 .05) were added to the survival Cox model. All analyses were performed using SAS software, version 8.2.Results1. Results of TP53The prevalence of TP53 mutation was 46% (240/524). The overall prevalence of TP53 mutations in this series is compatible with data in NSCLC from the IARC TP53 database. The codon distribution and mutation patterns of TP53 mutations (in terms of base changes) were consistent with those reported for lung cancers in patients with tobacco smoking history. Specifically, G to T transversion represented 29.97% of all detected mutations, an observation that has been shown to represent a typical mutagen signature of components of tobacco smoke. When all significant variables are put together in a logistic model stratified by centre, only age, sex, pathological TNM stage, T of TNM and quality after final H&E are retained to explain TP53 status.For 5-year survival analysis, TP53 mutation status had no prognostic or predictive value for survival in all NSCLC grouped together. However, cisplatin-based therapy showed a trend to benefit in TP53 wild type and to be harmful in TP53 mutated patients (P =0.18 for Overall Survival (OS) and P =0.13 for Disease Free Survival (DFS). A borderline significant interaction between TP53 and treatment was observed in non-adenocarcinoma patients (test for interaction, P =0.05 for DFS and P =0.25 for OS).Eight-year survival results support those of the 5-year analysis while TP53 mutation has a predictive value at P =0.04 for disease free survival on the whole group. Notably, a significant interaction between TP53 and treatment was observed in non-adenocarcinoma patients (test for interaction, P =0.02 for DFS and P =0.09 for OS).2. Results of KRASThe prevalence of KRAS mutation was 14% (98/718, including 90 mutations at codon 12 and 8 mutations at codon 13). The overall mutation pattern was similar to the one already reported in Cosmic database in lung cancer, with an excess of G to T transversion. More TP53 mutations were found in KRAS wild-type samples than in KRAS mutated samples (48% vs 27%, P =0.002). When all significant variables are put together in a logistic model stratified by centre, only T of TNM, histological type and final H&E are retained to explain KRAS status.For 5-year survival analysis, KRAS mutation has a borderline prognostic effect on overall survival (P =0.08) and disease-free survival (P =0.02) on the whole group of patients ;mutation of KRAS was not predictive of the effect of chemotherapy. The percentage of KRAS mutation in lung adenocarcinoma was clearly higher than that in non-adenocarcinoma NSCLC(P<0.0001). KRAS mutation has prognostic role for survival in non-adenocarcinoma patients(P=0.006,for 5-year OS;P=0.0009,for 5-year DFS).3. Results of EGFRIn 208 adenocarcinoma lung cancers, the prevalence of EGFR mutation was 11%(22/208). The overall EGFR mutation pattern was similar to the one already reported in Cosmic database in adenocarcinoma lung cancer. When all variables are put together in a logistic model stratified by centre stratum, only Pleural invasion (P=0.01) is retained to explain EGFR status.EGFR mutation has neither prognostic nor predictive role for survival in IALT adenocarcinoma patients(P >0.05). Since the rate of mutation is low (11%), the power for showing prognostic or predictive effects is low in this series of about 200 patients. It will be very important to perform a pooled analysis in order to study EGFR mutation with enough power.Conclusions1. TP53 mutation status had no prognostic for both 5 year survival and 8 year survival in NSCLC.2. TP53 mutation has a predictive value at P =0.04 for disease free survival on the whole group for 8 year analysis with a trend to benefit in TP53 wild type and to be harmful in TP53 mutated patients.3. The analyses of 5 year survival and 8 year survival accordantly suggested that a significant interaction between TP53 and treatment was observed in non-adenocarcinoma patients on disease free survival.4. KRAS is not predictive of the effect of chemotherapy and has a borderline prognostic effect on disease-free survival on the whole group of patients.5. Prognostic effect of KRAS mutation was different among different histology groups, with the worst prognostic effect in non-adenocarcinoma NSCLC.6. EGFR mutation has neither prognostic nor predictive role for survival in IALT adenocarcinoma patients.Above all, mutation detection of hot genes in NSCLC may help in assigning patients to get benefits from adjuvant cisplatin-based chemotherapy. The results of this study, TP53, KRAS and EGFR mutation analysis, will provide theory basis for finding molecular markers for adjuvant chemotherapy for resected NSCLC.更多还原