血清细胞因子和microRNA检测在宫颈疾病中的临床应用价值

发布时间：2018-08-16 07:36

【摘要】：研究背景：宫颈癌是全球女性第三大恶性肿瘤，早期诊断宫颈癌及癌前病变是降低宫颈癌发生率和死亡率的重要方法。循环炎症标志物和microRNA有可能成为宫颈癌相关的生物标志物用于肿瘤的早期检测或预后。研究目的：研究宫颈癌、宫颈疾病和健康对照的血清细胞因子和miRNA谱的差异，筛选新的标记物，并建立具有高灵敏度和特异度的宫颈癌辅助诊断模型。方法：从2010年1月至2010年12月收集原发性宫颈癌95例、宫颈病变80例及健康对照组85例血清标本，用Luminex200检测10项血清细胞因子水平，生物信息学分析三组间肿瘤标志物和细胞因子的表达差异，建立辅助诊断模型，并用验证集进行验证；对其中60名宫颈癌患者进行为期2年的随访，统计分析她们的2年总生存率及影响预后的因素，建立预后模型；2012年3月至2013年5月收集133例宫颈癌患者、123例宫颈病变患者和106例健康对照的血清标本，每组各由21例血清标本组成3个血清池，共组成9个血清池进行Solexa测序，筛选并预测新的miRNA，对预测的新miRNA用实时定量反转录聚合酶链反应（Quantitative real-time polymerase chain reaction）进行验证，并用Receiveroperating characteristic（ROC）曲线分析新miRNA对宫颈疾病的辅助诊断能力。结果：（1）在用单一炎性细胞因子和肿瘤标志物区分宫颈疾病组和健康对照组时，interleukin(IL)-8的Area under curve (AUC)最大为0.921(95%：0.885-0.958)，在24.04pg/ml的临界值时，其敏感度为84.8％，特异度为88.1％，高于SCC；用二元logistic回归建立的多参数诊断模型模(包括Squamous cell carcinoma associatedantigen （SCC）、IL-8和Monocyte chemoattractant protein1（MCP-1)）的AUC为0.939。鉴别诊断宫颈癌和宫颈病变时，单个肿瘤标志物和细胞因子的价值均较低，运用分类树多参数分析建立的模型（包括SCC和IL-10）的ROC曲线下为0.729，较临床上常用的SCC有了一定提高。（2）对60例I-II期宫颈癌患者的2年随访表明，术前血清Carbohydrate antigen153（CA153）、SCC、IL-10和TNF-α（tumornecrosis factor-α）水平与宫颈癌患者的两年总生存率相关。COX多参数分析显示CA153≥17.60U/ml、SCC≥1.60ng/ml和TNF-α≥10.60pg/ml可能为宫颈癌I-II期患者的独立预后危险因素。依据血清CA153水平和TNF α水平可以对宫颈癌患者进行危险分层，分为三层：①血清CA153≥17.60μg/L的水平；②血清CA15317.60μg/L和TNFα≥10.60pg/ml；③血清CA15317.60μg/L和TNF α10.60pg/ml。这三组间的两年总生存率分别为33.3％、60.0％和93.9％。（3）通过对宫颈癌患者血清池进行Solexa测序，，我们预测、筛选和验证了5个新的miRNA。在区分宫颈疾病和健康对照时，PmiR-3的AUC为0.924(95％CI:0.888-0.959)，当约登指数最大时，其敏感度为92.7％，特异性为78.3％；鉴别诊断宫颈癌与宫颈病变患者时，PmiR-1的AUC达到了0.893(95％CI：0.842-0.943)，约登指数最大时的敏感度为85.9％，特异性为83.9％。文章已经提交，尚未接收，所以以上新miRNA为暂命名，待文章接收后，miRbase会对新miRNA进行命名。结论：运用多元统计分析方法，将肿瘤标志物和炎性细胞因子联合运用可以提高其辅助诊断宫颈疾病的能力，IL-8、IL-10和MCP-1可以为宫颈癌的早期预警提供重要参考价值。炎性细胞因子TNF α在宫颈癌的预后中也发挥重要作用。循环血清miRNAs有望成为宫颈癌相关的生物标志物的用于疾病的早期诊断，并为宫颈癌的机理研究提供新的方向。
[Abstract]:Background: Cervical cancer is the third largest malignant tumor in the world. Early diagnosis of cervical cancer and precancerous lesions is an important method to reduce the incidence and mortality of cervical cancer. Methods: From January 2010 to December 2010, 95 cases of primary cervical cancer, 80 cases of cervical lesions and 85 cases of healthy control group were collected and tested by Luminex 200. The levels of 10 serum cytokines were measured, and the expression of tumor markers and cytokines were analyzed by bioinformatics. The adjuvant diagnosis model was established and validated by validation set. Sixty patients with cervical cancer were followed up for 2 years. Their 2-year overall survival rate and prognostic factors were analyzed and prognostic models were established. From March 2012 to May 2013, serum samples from 133 patients with cervical cancer, 123 patients with cervical lesions and 106 healthy controls were collected. Each group consisted of 21 serum samples and consisted of 3 serum pools. Solexa sequencing was performed in 9 serum pools to screen and predict new microRNAs. The predicted new microRNAs were identified by real-time quantitative reverse transcription polymerase chain reaction (Qu-PCR). Antitative real-time polymerase chain reaction was used to validate the method and the Receiver operating characteristic (ROC) curve was used to analyze the auxiliary diagnostic ability of new microRNAs for cervical diseases.
Results: (1) The area under curve (AUC) of interleukin (IL) - 8 was 0.921 (95%:0.885-0.958) when using a single inflammatory cytokine and tumor marker to distinguish cervical disease group from healthy control group, and the sensitivity and specificity were 84.8% and 88.1% at the critical value of 24.04 pg/ml, which were higher than those of SCC. The AUC of the parametric diagnostic model (including Squamous cell carcinoma associatedantigen (SCC), IL-8 and Monocyte chemoattractant protein 1 (MCP-1)) was 0.939. In the differential diagnosis of cervical cancer and cervical lesions, the value of single tumor marker and cytokines was low. The model established by using classification tree multiparametric analysis (including SCC and IL-10) The two-year follow-up of 60 patients with stage I-II cervical cancer showed that preoperative serum levels of Carbohydrate antigen 153 (CA153), SCC, IL-10 and TNF-alpha were correlated with the two-year overall survival rate of patients with cervical cancer. SCC (> 1.60ng/ml) and TNF-a (> 10.60pg/ml) may be independent prognostic factors for patients with stage I-II cervical cancer. According to serum CA153 level and TNF-a level, risk stratification of patients with cervical cancer can be divided into three layers: serum CA153 (> 17.60ug/L); serum CA15317.60ug/L and TNF-a (> 10.60pg/ml); serum CA15317.60 pg/ml; and TNF-a (> 10.60pg/ml The two-year overall survival rates of the three groups were 33.3%, 60.0% and 93.9%. (3) By sequencing the serum pools of cervical cancer patients with Solexa, we predicted that five new microRNAs were screened and validated. The sensitivity and specificity were 92.7% and 78.3% respectively, and the AUC of PmiR-1 was 0.893 (95% CI: 0.842-0.943) and 85.9% and 83.9% respectively in the differential diagnosis of cervical cancer and cervical lesions. E will name the new miRNA.
Conclusion: Combined use of tumor markers and inflammatory cytokines can improve the ability of assistant diagnosis of cervical diseases. IL-8, IL-10 and MCP-1 can provide important reference value for early warning of cervical cancer. Inflammatory cytokine TNFalpha also plays an important role in the prognosis of cervical cancer. IRNAs are expected to become biomarkers related to cervical cancer for the early diagnosis of diseases, and provide a new direction for the study of the mechanism of cervical cancer.
【学位授予单位】：中国人民解放军医学院
【学位级别】：博士
【学位授予年份】：2014
【分类号】：R737.33

【共引文献】