前列腺癌新型尿液诊断指标—长链非编码RNA 383及融合基因USP9Y-TTTY15

发布时间：2018-07-16 20:53

【摘要】：研究目的近年来，经典的前列腺癌早期诊断及筛查方式——血清学PSA遭到了众多的质疑。主要是因为良性/正常前列腺腺体细胞均产生和分泌PSA，从而导致血清学PSA诊断的假阳性问题。大量研究证实，单纯血清PSA指标指导前列腺癌早期诊断的优势明显不足，对于PSA处于灰区的患者，前列腺穿刺活检阳性率只有17-32%，PSA假阳性率高达65%之多，有超过2/3的对象接受了不必要的穿刺[4,5]。然而，阴性的穿刺结果也并不能完全排除肿瘤的存在，反复穿刺阳性率也仅有10%左右[6]。伴随着2012年2月美国FDA正式批准首个尿液中长链非编码RNA诊断试剂盒——PROGENSA PCA3用于临床[7,8]，前列腺癌分子诊断进入了尿沉渣RNA无创检测时代。据报道尿液中PCA3的诊断敏感度67%、特异度83%[9]。而同期发现的TMPRSS2-ERG融合基因不仅也能在尿沉渣中被检测到，而且仅在前列腺肿瘤中发生融合，，欧美人群中的发生频率50%左右。因此我们亟需开发种新型的高特异性且兼顾敏感性的分子诊断标记物，提高现如今“尴尬的”穿刺诊断阳性诊断率。本研究的目的，是针对课题组前期发现的新型非编码RNA-lncRNA383及中国人群样本中有效表达的融合基因USP9Y-TTTY15作为前列腺癌早期尿液诊断可行性的研究。研究方法本课题组前期应用第二代基因转录本测序技术对14对来源于前列腺癌根治手术的肿瘤及癌旁正常组织样本进行测序，筛选在组织学层面明显异常表达的非编码RNA和融合基因，并扩大组织学样本含量，运用qRT-PCR验证测序结果。回顾性选取肿瘤患者及穿刺阴性患者尿沉渣，并应用qRT-PCR的方式探索上述差异性表达的非编码RNA及融合基因是否具有诊断价值。前瞻性连续性入组因PSA异常计划行前列腺穿刺活检患者的尿沉渣，应用根据测序结果入选的诊断指标进行前瞻性连续性验证，进步明确其诊断能力，并应用logistic回归分析建立诊断模型，以期提高前列腺癌早期诊断敏感性及特异性。结果经过前期14对前列腺肿瘤及癌旁正常组织的转录本测序，我们发现137个差异性表达的长链非编码RNA,我们按照以下2个标准进行筛选差异性表达最为明显的目的基因进行后续研究。标准：RNA-Seq结果中所有符合FDR≤0.001，且倍数差异≥2的差异性表达lncRNA；标准二：目的long non-coding RNA应至少在10对以上的配对组织样本中出现致的变化趋势。最终，在筛选出的lncRNA中，我们发现FR0348383（lncRNA383）表达量在前列腺肿瘤组织中的表达量很高。在后续的组织学水平验证中，lncRNA383在肿瘤组织中的表达量明显高于癌旁正常组织。在回顾性分析中我们发现，前列腺穿刺患者尿沉渣中均存在lncRNA383，且表达趋势与组织学结果类似，具有良好的穿刺阳性结果预测能力（AUC=0.858，cut-off=0.877，敏感性=68.8%，特异性=97.8%，阳性预测值=98.19%，阴性预测值=64.17%）。在后续的前瞻性连续性研究中，虽然其诊断效能较回顾性分析的结果有所下降（AUC=0.631，cut-off=0.776，敏感性85.4%，特异性43.5%），但仍与尿液中PCA3的诊断效率相近，且敏感性优于文献报道的尿沉渣中TMPRSS2-ERG的诊断敏感性，为将来进步建立多诊断指标综合应用模型提供新的选择。此外我们还发现，lncRNA383的表达水平与前列腺癌患者血清PSA，Gleason评分及DRE结果无明显相关性，这在定程度上可能会限制lncRNA383在临床上的应用价值。与此同时，我们还在前期的肿瘤测序样本中发现了个已知的融合基因（TMPRSS2-ERG）和37个首次报道的仅在前列腺肿瘤组织中表达的肿瘤特异性融合基因，其中发生频率最高的是TMPRSS2-ERG和USP9Y-TTTY15，均达到21.4%。USP9Y-TTTY15基因的融合发生在USP9Y基因的exon3和TTTY15基因的exon4上。在后续的54对癌及癌旁组织验证中，我们发现USP9Y-TTTY15基因融合的发生频率高达35.2%。有趣的是，融合后的USP9Y-TTTY15不具有明显的开放阅读区，而是可能作为个非编码RNA发挥生物学作用。在涵盖63例前列腺癌患者及90例穿刺阴性（-）患者的回顾分析中，我们发现尿沉渣中均存在融合基因USP9Y-TTTY15的表达，且表达趋势与组织学结果类似，并且具有良好的诊断能力（AUC=0.802，cut-off=0.894，敏感性=85.7%，特异性=73.3%，阳性预测值=69.20%，阴性预测值=87.98%）。在后续涵盖172例患者的前瞻性连续性研究当中，融合基因USP9Y-TTTY15继续保持了良好的诊断学效能（AUC=0.783， cut-off=0.872，敏感性=66.7%，特异性=81.6%）。在PSA处于4-10ng/ml灰区内，融合基因USP9Y-TTTY15具有良好的预测前列腺穿刺阳性结果的能力（AUC=0.774，cut-off值=0.845，敏感性=81.8%，特异性=62.2%，阳性预测值=35.59%，阴性预测值=93.32%），显著的阴性预测值有利于提高PSA灰区内前列腺穿刺的阳性率，具有很强的临床转化意义。此外，其还可以联合PSA和fPSA形成多指标综合诊断模型Y=71.354（XUSP9Y-TTTY15）+0.213（XPSA）-1.754（XfPSA）-34.286，其诊断敏感性和特异性更分别高达66.7%和91.9%，明显高于文献记录的PCA3与TMPRSS2-ERG联合诊断效率，是种无创的，理想的早期前列腺癌诊断指标，对于提高前列腺穿刺诊断阳性率具有积极的临床意义。结论非编码RNA-lncRNA383和融合基因USP9Y-TTT15Y均可以作为独立的预测因子应用于早期前列腺癌的组织学、尿沉渣的筛查与诊断中。USP9Y-TTT15Y还能有效预测PSA处于灰区内的前列腺穿刺阳性结果。此外，USP9Y-TTT15Y还对于Gleason评分≤6的前列腺癌具有良好的预测效能，为临床早期治疗手段的选择及治疗后随访方案的制定提供了重要的诊断依据。在试图联合lncRNA383与融合基因USP9Y-TTT15Y及其他众多诊断指标的综合诊断模型过程中，很可惜lncRNA383被剔除，入选指标仅有USP9Y-TTT15Y、PSA和fPSA，最终形成多指标综合诊断模型Y=71.354（XUSP9Y-TTTY15）+0.213（XPSA）-1.754（XfPSA）-34.286，其诊断敏感性和特异性更是分别高达66.7%和91.9%，明显高于文献记录的PCA3与TMPRSS2-ERG联合诊断效率，是种无创的、理想的早期前列腺癌诊断指标，对于提高前列腺穿刺诊断阳性率具有积极的临床转化意义。
[Abstract]:research objective
In recent years, the classic diagnosis and screening of prostate cancer, serological PSA, has been questioned. It is mainly due to the production and secretion of PSA in benign / normal prostate gland cells, which leads to the false positive problems of serological PSA diagnosis. A large number of studies have proved that the simple blood purity PSA index guides the early diagnosis of prostate cancer. The positive rate of PSA in the grey area was only 17-32%, the false positive rate of PSA was as high as 65%, and those with more than 2/3 received the unnecessary puncture [4,5].. However, the negative puncture results could not completely exclude the existence of the tumor, and the positive rate of repeated puncture was only about 10% [6]. accompanied by 2012. In February, FDA officially approved the first long chain non coded RNA diagnostic kit in urine - PROGENSA PCA3 for clinical [7,8]. The molecular diagnosis of prostate cancer entered the era of RNA noninvasive detection of urine sediment. It is reported that the diagnostic sensitivity of PCA3 in urine is 67%, the specificity 83%[9]. and the current TMPRSS2-ERG fusion gene can not only be in the urine. The slag was detected and fused only in the prostate tumor, and in the European and American population, the frequency was about 50%. Therefore, we urgently need to develop a new highly specific and sensitive molecular diagnostic marker to improve the positive diagnostic rate of "awkward" puncture. The present new non coded RNA-lncRNA383 and the fusion gene USP9Y-TTTY15, which are effectively expressed in the Chinese population samples, are used as a study of the feasibility of early urine diagnosis for prostate cancer.
research method
The second generation gene transcriptional sequencing technology was used to sequence the 14 pairs of tumor and normal tissue samples from radical prostatic cancer surgery in the earlier period, to screen out the uncoded RNA and fusion genes that were clearly expressed at the histological level, and to enlarge the content of histology samples, and to use qRT-PCR to verify the sequencing results. The diagnostic value of the non coded RNA and fusion gene of the above differential expression was explored by using qRT-PCR to explore the diagnostic value of the non coded RNA and fusion gene of the above differential expression. The prospective continuous entry group was planned for the urine sediment of the patients with prostate biopsy, and the diagnostic indexes selected according to the sequencing results were prospectively evaluated. In order to improve the sensitivity and specificity of the early diagnosis of prostate cancer, we developed a diagnostic model based on logistic regression analysis.
Result
After 14 pairs of prostatic tumors and the transcriptional transcripts of normal tissues adjacent to the cancer, we found 137 differentially expressed long chain noncoding RNA. We performed the following 2 criteria for screening the most distinct target genes for differential expression. Standard: all RNA-Seq results were consistent with FDR less than 0.001, and the multiple difference was more than 2. Differential expression of lncRNA; standard two: target long non-coding RNA should be at least in 10 pairs of paired tissue samples. Finally, in the screened lncRNA, we found that the expression of FR0348383 (lncRNA383) expression in the prostate tumor tissue is very high. In subsequent histological examination, lncRNA383 The expression in the tumor tissue was significantly higher than that in the normal paracancerous tissue. In the retrospective analysis, we found that lncRNA383 was present in the urine sediment of the patients with prostatic puncture, and the expression trend was similar to that of the histology. It had good prediction ability of the positive results of puncture (AUC=0.858, cut-off=0.877, sensitivity =68.8%, specific =97.8%, positive preconditioning. =98.19%, negative predictive value =64.17%). In subsequent prospective continuity studies, although their diagnostic effectiveness was lower than that of retrospective analysis (AUC=0.631, cut-off=0.776, sensitivity 85.4%, specificity 43.5%), the diagnostic efficiency of PCA3 was similar to that in urine, and the sensitivity was better than that of TMPRSS2-ERG in the reported urine sediment. Diagnostic sensitivity can provide a new choice for future progress in the establishment of multiple diagnostic indicators. In addition, we also found that there is no significant correlation between the expression level of lncRNA383 and the serum PSA, Gleason score and DRE results of the prostate cancer patients, which may limit the clinical application of lncRNA383 to a certain extent.
At the same time, we found a known fusion gene (TMPRSS2-ERG) and 37 first reported tumor specific fusion genes expressed in the prostate tumor tissue for the first time, with the highest frequency of TMPRSS2-ERG and USP9Y-TTTY15, which all reached the fusion of 21.4%.USP9Y-TTTY15 gene. USP9Y gene exon3 and TTTY15 gene Exon4. In the subsequent 54 pairs of cancer and para cancerous tissue validation, we found that the frequency of USP9Y-TTTY15 gene fusion is as high as 35.2%., and the fusion USP9Y-TTTY15 does not have a clear open reading area, but may play a biological role as a non coded RNA. It covers 63 cases before it is covered. In the retrospective analysis of patients with adenocarcinoma and 90 cases of puncture negative (-), we found that the expression of fusion gene USP9Y-TTTY15 was found in urine sediment, and the expression trend was similar to that of histology, and had good diagnostic ability (AUC=0.802, cut-off=0.894, sensitivity = 85.7%, specific =73.3%, positive predictive value =69.20%, negative predictive value = = 87.98%. The fusion gene USP9Y-TTTY15 continued to maintain a good diagnostic efficiency (AUC=0.783, cut-off=0.872, sensitivity =66.7%, specific =81.6%) in a prospective continuous study covering 172 patients. The fusion gene USP9Y-TTTY15 had a good prediction of prostatic positive results in the 4-10ng/ml grey area of PSA. Ability (AUC=0.774, cut-off value =0.845, sensitivity =81.8%, specific =62.2%, positive predictive value =35.59%, negative predictive value =93.32%), significant negative predictive value is beneficial to improve the positive rate of prostate puncture in PSA gray area, and has a strong clinical significance. Furthermore, it can also combine PSA and fPSA to form a multi index comprehensive diagnostic model Y=71.. The diagnostic sensitivity and specificity of 354 (XUSP9Y-TTTY15) +0.213 (XPSA) -1.754 (XfPSA) -34.286 are up to 66.7% and 91.9% respectively. It is obviously higher than the combined diagnostic efficiency of PCA3 and TMPRSS2-ERG recorded in the literature. It is a noninvasive and ideal diagnostic index for early prostate cancer. It has positive clinical significance in improving the positive rate of prostate biopsy.
conclusion
Both non coded RNA-lncRNA383 and fusion gene USP9Y-TTT15Y can be used as independent predictors for early histology of prostate cancer. In the screening and diagnosis of urine sediment,.USP9Y-TTT15Y can also effectively predict the positive results of prostate biopsy in the gray area of PSA. In addition, USP9Y-TTT15Y is also used for prostate cancer with a Gleason score of less than 6. Good predictive effectiveness provides an important diagnostic basis for the selection of early clinical treatments and the formulation of follow-up follow-up programs. In the process of combining lncRNA383 with fusion gene USP9Y-TTT15Y and many other diagnostic indicators, it is a pity that lncRNA383 is eliminated, and only USP9Y-TTT15Y, PSA, and PSA are selected. FPSA, finally formed a multi index comprehensive diagnostic model, Y=71.354 (XUSP9Y-TTTY15) +0.213 (XPSA) -1.754 (XfPSA) -34.286, and its diagnostic sensitivity and specificity are up to 66.7% and 91.9% respectively. It is significantly higher than the combined diagnostic efficiency of PCA3 and TMPRSS2-ERG recorded in the literature. It is a noninvasive, ideal early diagnostic index for prostate cancer. The positive rate of diagnosis by row biopsy is of positive clinical significance.
【学位授予单位】：第二军医大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：R737.25

【参考文献】