| Background: Prostate cancer(PCa) in China underwent a dramatically increase of its morbidity and mortality, marked the highest growth rate among malignancies. Meanwhile, the low specificity of biomarker for early diagnosis as well as straitened strategy to effectively control tumor progression hangs over clinicians and scientist. Therefore, novel non-invisive biomarkers or strategies that are more sensitive and specific to diagnose PCa are urgetly needed as a substitute or complement to serum PSA. The crucial molecular mechanism is badly in need of elucidation in order to develop more effective drugs to treat castration-resistant prostate cancer(CRPC).Purpose:To study the diagnostic performance of urinary long non-coding RNA MALAT1 and PCA3 in Chinese patients underwent first prostate biopsy and establish a comprehensive model based on MALAT1 or PCA3 score combined with clinical risk factors of PCa; To explore the role of novel mi R-n5 in PCa progression and try to find the potential therapeutic target for treating CRPC.Methods:Total RNA which was extracted from urine sediment was amplified using the Trans Plex Complete Whole Transcriptome Amplification Kit according to the manufacturer’s instructions. The expression levels of MALAT1 and PCA3 were detected by q RT-PCR. Univariate logistic regression was used to identify independent predictors of PCa upon biopsy and multivariate logistic regression was used to establish PCa diagnostic models. Co-relationships between MALAT1 or PCA3 score and the clinical variables were assessed by the Spearman rank correlation test. The area under the Receiver operating characteristic(ROC) curve(AUC) was used to assess the predictive power and calculate cut off value and its corresponding sensitivity and specificity. Decision curve analysis(DCA) was used to evaluate the clinical effects of MALAT1 and PCA3 score. All the novel mi RNA sequences were identified by Northern blot. Using proliferation assay, transwell assay(for migration and invasion) and colony formation to evaluated the biological function of mi R-n5 in aggressive PCa cell lines. The target genes of mi R-n5 were searched by bioinformatical analysis. And then the target genes were validated by dual luciferase report gene assay. Graph Pad Prism5 software was used to display data and calculate the IC50 of indicated small molecular drug. CRPC xenograft models were established subcutaneously or intratibially. All of the statistical calculations were performed using SPSS v.17.0(SPSS Inc., Chicago, IL, USA), Med Calc v.10.4.7.0(Med Calc Software bvba, Mariakerke, Belgium) and R software v.3.1.1(The R Foundation for Statistical Computing). Hem I software was used to generate heatmap for indicated data. p<0.05 was considered to be statistically significant.Results:Urinary MALAT1 or PCA3 score sufficed to discriminate positive from negative prostate biopsy results and strongly correlated with PCa detection rate. However, MALAT1 or PCA3 score did not correlate with other risk factors or Gleason score based on correlation analysis. The univariable logistic regression demonstrated MALAT1 score, age, t PSA, volume, %f PSA and DRE were independent risk factors in the overall group, while t PSA was not included in the PSA “grey zone” group. There was a trend in the overall cohort that MALAT1 score was superior to t PSA, but it was not meet statistically significant(p=0.510), while it got a better diagnostic performance compared to t PSA in PSA “grey zone” group(p=0.034). Meanwhile, in PSA “grey zone” group, PCA3 score was superior to both t PSA(p=0.0407) and %f PSA(p=0.046). Further multivariable logistic regression demonstrated that MALAT1 score based comprehensive model showed a higher AUC of 0.853 and predictive accuracy(PA) of 79.79% to predict PCa and got an increased AUC of 0.0318 and increased PA of 5.32% in the PSA “grey zone” discovery cohort. The same parameters derived from the discovery phase were assessed in an independent validation cohort and got the same results. PCA3 score based comprehensive model also displayed a better diagnostic performance compared to base model in the PSA “grey zone” group.DCA showed that MALAT1 score based comprehensive model was superior to the base model in both discovery and validation phase in the PSA “grey zone” cohort. At a threshold of 25%, MALAT1 score based comprehensive model detected more cancers than the base model(13.97% vs. 11.47%) and also prevented more unnecessary biopsies than the base model(47.32% vs. 39.78%) in the “grey zone” cohort of the discovery phase. Applying the same threshold in validation phase, the results showed that MALAT1 score based comprehensive model also detect more cancers(15.73% vs. 13.11%) and prevented more unnecessary biopsies than the base model(30.33% vs. 22.47%). More importantly, MALAT1 score based comprehensive model did not miss any high grade cancer patients. Meanwhile, PCA3 score based comprehensive model was superior to the base model with a higher net benefit for most of threshold probabilities, especially in 25%–40%. At a threshold of 30%, PCA3 score based comprehensive model could spare 60.3% unnecessary biopsies at a cost of missing 4 PCa patients included 2 high grade PCa.In addition, more than 300 novel mi RNAs were identified by RNA-seq which analyzed 65 PCa with their adjacent normal tissues. 8 novel mi RNAs, which were selected based on their high expression levels, were then validated by northern blot. After evaluating the expression levels of these mi RNAs in aggressive PCa cell lines compared to indolent ones, we selected 4 progression associated novel mi RNAs for further study. Our study finally chose novel mi R-n5 as it also related with clinical results. In vitro functional studies demonstrated that mi R-n5 significantly inhibited cell growth, migration and invasion in aggressive PCa cell lines. And intratumor injection of mi R-n5 mimics minimized tumor growth in mice. Furthermore, KDM6 B were found by bioinformatical analysis and validated as a direct target of mi R-n5 by 3’UTR dual luciferase reporter assay. Futher study showed that GSK-j4, a small molecular drug which inhibit the activity of KDM6 B, could abolish PCa growth in vitro and in vivo.Conclusions:Urine-based MALAT1 and PCA3 were independent risk factors of PCa and sufficed to discriminate positive from negative prostate biopsy results. MALAT1 or PCA3 score based comprehensive model could significantly improve the diagnostic performance of PCa. Expecially in PSA “grey zone” population, both diagnostic power of MALAT1 and PCA3 were superior to that of t PSA. Either MALAT1 or PCA3 score based comprehensive model showed a better performance to predict PCa than base model which was established by clinical risk factors, and could spare more unnecessary biopsies with no more high grade cancer patients missed. In addition, mi R-n5 significantly inhibited aggressive PCa in vitro and in vivo through attenuating the expression of KDM6 B, which suggests a novel therapeutic target for CRPC therapy. Meanwhile, GSK-j4, a small molecular drug targets KDM6 B, could abolish PCa growth in vitro and in vivo, represents a potential targeting drug for CRPC treatment. |
PDF Full Text Request