第一部分：目的：肺鳞状细胞癌是肺癌主要的病理亚型之一，全球每年约40万人死于肺鳞癌。既往已有研究报道肺鳞癌基因编码区的突变特征，然而对于基因启动子区域测序因为缺少生物信息学分析算法和测序准确性而较少受到关注，因此我们设计了基于捕获启动子和外显子区域的测序方案对肺鳞癌同时进行启动子及全外显子测序，旨在系统分析肺鳞癌启动子及外显子突变谱，及其与肿瘤突变负荷之间的关系。探索肺鳞癌启动子区域潜在的分子治疗靶点及能够替代肿瘤突变负荷的免疫治疗标志物。方法：本研究共纳入我院2017年12月至2019年12月接受手术治疗的肺鳞癌患者共计109例。收集手术切除的肿瘤原发灶、癌旁正常肺组织及外周血标本，使用Illumina HiSeq 2500测序平台进行启动子测序、全外显子测序及转录组测序，通过生物信息学分析肺鳞癌基因启动子及外显子突变谱，并计算肿瘤突变负荷，分析肿瘤突变谱与肿瘤突变负荷之间的关系。结果：在109例肺鳞癌患者的全外显子测序结果中，我们发现绝大多数外显子的突变为错义突变（missense mutation）。在所有的错义突变中，最主要的变异为C＞A（腺嘌呤突变为胞嘧啶）。经过MutSigCV算法筛选分析后发现突变频率较高的前10个基因分别为TP53（82%）、KMT2D（23%）、NFE2L2（20%）、NOTCH1（17%）、ANKS1A（13%）、CDKN2A（10%）、SIX3（8%）、RB1（7%）、KRT76（7%）、PTH2（6%），其中首次发现ANKS1A基因在我国肺鳞癌患者中存在高频突变并且在肿瘤组织中显著低表达。在拷贝数变异中，我们发现AMDHD1与GTF2IRD2在肺鳞癌中发生高频融合（12.8%）。在所有检测到的启动子的突变中，共有13879个基因的启动子区域发生突变，绝大多数突变为单核苷酸变异（single nucleotide variant，SNV）,在所有的单核苷酸变异中，最主要的变异为C＞A。经MutSigCV算法筛选分析后发现只有40个基因启动子突变有意义。在40个启动子突变的基因中，只有WDR74基因启动子突变后对WDR74的表达产生显著影响，转录表达显著增加（P=0.048）。109例肺鳞癌患者中，有16例患者出现WDR74基因启动子突变，突变频率为（14.7%），其中14例突变为SNV,其余2例分别为插入突变和缺失突变。我们根据109例肺鳞癌标本全外显子测序的结果计算得到TMB值，发现109例肺鳞癌患者中最高TMB值为42.19 mu/Mb，最低TMB值为0.05 mu/Mb，平均TMB值为9.67 mu/Mb，中位TMB值为8.30 mu/Mb。我们研究了TMB和肺鳞状细胞癌患者的临床资料之间的关联，结果表明只有患者年龄与TMB显著相关，高龄患者的TMB值越高（P=0.006）。其他临床病理特征与TMB值均不相关。我们利用启动子区域突变基因ZNF595、DUSP22、GLYAT、MYCT1、OR1C1、OR4C3、OR4D11、OR4M2、OR4N4、TMEM132C和WDR74等11个基因构建计算模型，发现该11基因与TMB预测值具有高度相关性（R²=0.8017，P＜10^-16），我们将此基因PNAEL确定为Promoter-TMB PANEL（PT-PANEL）。我们利用二分法将TMB的截断值确定为10，发现PT-PANEL预测TMB的AUC值为0.899 ± 0.065。结论：本研究首次对我国较大样本肺鳞癌患者同时进行了全外显子、启动子及RNA测序。在外显子突变中，我们发现ANKS1A基因在肺鳞癌中存在高频突变并且在肿瘤组织中显著低表达。在融合基因变异中，我们发现AMDHD1与GTF2IRD2在肺鳞癌中发生高频融合，AMDHD1-GTF2IRD2基因融合变异在肺鳞癌的发生发展发挥的作用有待进一步研究。在启动子突变中，我们发现启动子突变基因ZNF595、DUSP22、GLYAT、MYCT1、OR1C1、OR4C3、OR4D11、OR4M2、OR4N4、TMEM132C和WDR74等11个基因与TMB具有高度相关性，期望通过进一步临床验证证实启动子突变预测免疫治疗的可行性和准确性。

第二部分：目的：非小细胞肺癌是全球最常见的恶性肿瘤之一，也是导致癌症相关死亡的首要原因。TP53突变是在非小细胞肺癌中发现的最常见的突变，并且与患者预后密切相关。间变性淋巴瘤激酶（ALK）基因重排的晚期非小细胞肺癌患者对克唑替尼治疗敏感，但并非所有ALK阳性的患者都同样受益于克唑替尼的治疗。本文旨在研究伴随有TP53突变对克唑替尼治疗ALK重排的晚期非小细胞肺癌患者疗效的影响。方法：回顾性分析我院2011年1月至2016年12月64例行克唑替尼治疗的ALK重排的非小细胞肺癌患者的临床病理资料。对这64例患者的肿瘤石蜡样本进行二代测序以确定TP53的突变状态，采用SPSS.20统计软件统计分析TP53突变与患者临床病理特征、克唑替尼治疗疗效以及患者预后之间的关系。结果：64例ALK重排的非小细胞肺癌患者中，15例（23.4%）患者伴随有TP53突变。15例患者中，6例为破坏性TP53突变，9例为非破坏性TP53突变。伴有TP53突变患者的客观缓解率和疾病控制率均显著低于TP53野生型患者(P=0.003和P=0.023)。TP53突变型患者的无进展生存期（Progression-free survival, PFS）明显短于TP53野生型患者(P=0.045)。在TP53突变的患者中，伴有非破坏性TP53突变的患者与伴有破坏性TP53突变的患者相比具有更短的PFS（P=0.069）。当伴有非破坏性TP53突变的患者与TP53野生型患者相比，前者具有显著缩短的PFS（P=0.003）。结论：ALK重排的非小细胞肺癌患者中，伴随有TP53突变者，特别是非破坏性突变者，对克唑替尼的治疗反应较差，并且无进展生存期缩短。

Part1：Background: Lung squamous cell carcinoma is one of the most common pathological subtypes of lung cancer, causing approximately 400,000 deaths per year worldwide. Previous studies have reported the mutational profile of the gene coding region in lung squamous cell carcinoma, but little attention has been paid to the mutational characteristics of gene regulatory region because of the lack of bioinformatics analysis algorithm and sequencing accuracy. Therefore, we designed a sequencing scheme based on capturing promoter and exon region to sequence both promoters and exons in lung squamous cell carcinoma, in order to systematically analyze the mutation profile of promoter and exon in lung squamous cell carcinoma and its relationship with tumor mutation load. This study aims to explore the potential molecular therapeutic targets in the promoter region of lung squamous cell carcinoma and the immunotherapy markers that can replace the tumor mutation burden. Methods: This study included 109 patients with lung squamous cell carcinoma who underwent surgery in our hospital from December 2017 to May 2019 and the samples of lung squamous cell carcinoma tissues, paired normal lung tissue and peripheral blood were collected. Promoter sequencing, whole-exome sequencing, and transcriptome sequencing were performed on Illumina HiSeq 2500 sequencing platform following the Illumina protocols. The mutational characterization of lung squamous cell carcinoma was analyzed by bioinformatics, and the tumor mutational burden was calculated to explore the relationship between tumor mutational characterization and tumor mutation burden. Results: In the whole-exome sequencing results of 109 patients with lung squamous cell carcinoma, we found that 96 patients (88.07%) had at least one mutation site in their genomes. Of all the mutations detected in whole exons, the vast majority were missense mutations. Among all missense mutations, the main mutation was C > A (adenine mutation to cytosine). After analysis by MutSigCV algorithm, it was found that the top 10 genes with high mutation frequency were TP53 (82%), KMT2D (23%), NFE2L2 (20%), NOTCH1 (17%), ANKS1A (13%), CDKN2A (10%), SIX3 (8%), RB1 (7%), KRT76 (7%) and PTH2 (6%). We found for the first time that ANKS1A gene had high frequency mutation in Chinese patients with lung squamous cell carcinoma and significantly low expression in tumor tissues. In copy number variations, we found that high frequent fusion of AMDHD1 and GTF2IRD2 occurred in lung squamous cell carcinoma (12.8%). Among all the mutations detected, there were 13879 mutations in the promoter region, most of which were single nucleotide variants (SNV). Among all the single nucleotide variants, the main mutation was C > A. After analysis by MutSigCV algorithm, it was found that only promoter mutations of 40 genes were significant. Among the 40 promoter mutated genes, only the promoter mutation of WDR74 gene had a significant effect on the expression of WDR74, and the transcriptional expression of WDR74 was significantly increased (P = 0.048). Among the 109 patients with lung squamous cell carcinoma, 16 patients (14.7%) had promoter mutations of WDR74 gene: 14 cases were SNV and the other 2 cases were insertion mutation and deletion mutation. The TMB value was calculated based on the sum of nonsynonymous single nucleotide and indel variants. It was found that the highest TMB value was 42.19 mu/Mb, the lowest TMB value was 0.05 mu/Mb, the average TMB value was 9.67 mu/Mb, and the median TMB value was 8.30 mu/Mb. We analyzed the relationship between TMB and clinicopathological features of patients with lung squamous cell carcinoma and the results showed that only advanced age was significantly correlated with high TMB (P = 0.006). We used 11 genes with promoter mutations, ZNF595, DUSP22, GLYAT, MYCT1, OR1C1, OR4C3, OR4D11, OR4M2, OR4N4, TMEM132C, WDR74 to construct the TMB prediction model, and found that the 11 genes were highly correlated with TMB value (R² =0.8017, P < 10^-16). We identified the PNAEL of the 11 genes as Promoter-TMB PANEL (PT-PANEL). We used 10 mu/Mb as the cutoff value of TMB, and found that the AUC value of TMB predicted by PT-PANEL is 0.899 ±0.065. Conclusions：In this study, whole exon, whole promoter and RNA sequencing were performed simultaneously in a large sample of Chinese patients with lung squamous cell carcinoma for the first time. Among exon mutations, we found that the ANKS1A gene frequently mutated in lung squamous cell carcinoma and ANKS1A expression was significantly decreased in tumor tissues. In copy number variations, we found that high frequent fusion of AMDHD1 and GTF2IRD2 occurred in lung squamous cell carcinoma, and the role of AMDHD1-GTF2IRD2 fusion in lung squamous cell carcinoma needs to be further studied. Among the promoter mutations, we found that 11 genes with promoter mutations, ZNF595, DUSP22, GLYAT, MYCT1, OR1C1, OR4C3, OR4D11, OR4M2, OR4N4, TMEM132C, WDR74, were highly correlated with TMB. It is expected to confirm the feasibility and accuracy of promoter mutations in predicting immunotherapy through further clinical validation.

Part 2：Background: Non-small cell lung cancer (NSCLC) is one of most common malignancies and the leading cause of cancer deaths worldwide. TP53 mutations are the most prevalent mutations detected in NSCLC and have been reported to be associated with outcome. ALK-rearranged NSCLC patients are sensitive to crizotinib, but not all ALK-rearranged patients benefit equally from crizotinib. The aim of study was to investigate the impact of concomitant TP53 mutations in efficiency of crizotinib treatment in ALK-rearranged NSCLC. Methods: Tumor samples from 64 ALK-rearranged NSCLC patients receiving crizotinib treatment were subjected to next-generation sequencing (NGS) to identify TP53 mutational status. The clinicopathologic features of the TP53 mutations and its impact on the effect of crizotinib treatment were analyzed. Results: Among the 64 ALK-rearranged patients, 15 (23.4%) patients showed a TP53 mutation. Of these, 6 cases had disruptive mutations and 9 with nondisruptive mutations. The objective response rate (ORR) and disease control rate (DCR) for TP53 mutated patients were both significantly lower compared with those for TP53 wild-type patients (P = 0.003 and P = 0.023, respectively). A significantly shorter progression-free survival (PFS) was found in TP53 mutated patients compared with TP53 wild-type patients (P = 0.045). Nondisruptive TP53 mutations were associated with a shorter PFS in comparison with disruptive TP53 mutations in ALK-rearranged patients (P = 0.069). When nondisruptive TP53 mutated patients were in comparison with TP53 wild-type patients, nondisruptive TP53 mutations were associated with a significant reduced PFS (P = 0.003). Conclusions: TP53 mutations, especially nondisruptive mutations, negatively affected the response to crizotinib and correlated with shorter PFS in ALK-rearranged NSCLC patients.