Mirna based identification of prostate cancer by investigation of urinary exosomes

Abstract

Prostate cancer is one of the most incident cancer subtypes with high mortality rate. Currently, diagnosis of prostate cancer is based on rectal examination, Prostate Specific Antigen (PSA) testing and biopsy. Normal range of PSA is defined as 0-4 ng/ml and individuals with higher PSA levels are considered as potential prostate cancer patients. However, PSA fluctuates as a result of many factors and it is shown to increase with age. Thus, PSA testing causes significantly high false positive results and many healthy men have biopsy unnecessarily due to high PSA levels or they even get overtreated. This situation has huge psychological as well as financial effects on these people. Herein, we investigated the diagnostic potential of urinary exosomal microRNAs (miRNA) in prostate cancer. Rather than investigation of cellular miRNAs, we focused on exosomal miRNAs because of high integrity of exosomes and their abundance in many biofluids including urine. Development of a sensitive diagnostic method from urine would be advantageous because accurate diagnosis of prostate cancer would be possible by a non-invasive procedure. miRNAs are the small non-coding RNAs and they suppress expression of target genes via degradation of mRNA or post-translational regulation. miRNAs have high potential as cancer biomarkers because they can act as tumor suppressor and repress oncogenic gene expression or function as oncogenic miRNA and suppress tumor suppressor gene expression. For this reason, we identified several candidate exosomal tumor suppressor and oncogenic miRNAs and continued our study with the most potent ones. At the beginning of the study, we validated that we efficiently isolated exosomes via several techniques such as flow cytometry, Dynamic Light Scattering (DLS), Nanoparticle Tracking Analysis (NTA) and Transmission Electron Microscopy (TEM). Then, we studied differential expression of candidate miRNAs in prostate cancer PC-3 cell line. Similar to our literature search findings, we found that expression levels of miR-107, miR-139, miR-145 and miR-204 were significantly lower in PC-3 exosomes in comparison to healthy urinary exosomes. On the other hand, oncomiRs; miR-21-5p, miR-375-5p and miR-574 3p were upregulated in PC-3 cell line exosomes. Of note, expression of another candidate miRNA; miR-30a was almost the same in PC-3 exosomes with healthy controls. We used these results as preliminary data and collected urine specimens from 17 prostate cancer patients. We firstly analyzed their PSA level-age and PSA Level- Gleason Score correlations. Our analyses revealed that PSA level was increasing with age and PSA was not correlated with Gleason Score. We concluded that PSA was being affected by several reasons in addition to tumor formation and it was not correlated with disease progression. Again, this was a finding which supported that a more sensitive diagnosis method than PSA testing was necessary for prostate cancer. When we detected expression levels of candidate miRNAs in urinary exosomes of prostate cancer patients and healthy controls, we observed that miR-107, miR-139, miR-145 and miR-204 were downregulated whereas miR-375-5p was upregulated in patients’ urinary exosomes. For miR-21, there was a very slight upregulation and miR-30a-5p and miR-574-3p levels were almost the same with healthy controls. Further, we wondered the diagnostic potential of these miRNAs in our patient cohort and we performed Receiver Operator Characteristic (ROC) Curve analysis for them. When they were used as combination, our miRNAs had 77% (AUC=0.7731) accuracy in distinguishing patients from healthy controls. After that, we examined miRNA dysregulations in patients with different PSA levels with the hypothesis that they may have different miRNA expression profiles. We saw that expression of miRNAs in exosomes patients with PSA<10 ng/ml and PSA=10-15 ng/ml were very similar with our expectations; downregulation of tumor suppressor and upregulation in oncogenic miRNAs whereas patients with PSA>15 ng/ml had a unique profile and all miRNAs were downregulated. Due to sample size limitations, we only tested diagnostic potential of candidate miRNAs in exosomes of patients with PSA<10 ng/ml and observed that AUC was 0.8500 so we could discriminate patients and healthy controls with 85% accuracy by using our candidate miRNA panel in testing. Taken together, our findings indicated that candidate urinary exosomal tumor suppressor and oncogenic miRNAs which we suggested in thesis are very potent in diagnosis of prostate cancer with differential expressions in prostate cancer patients with different PSA levels and this study opens the way for development of a noninvasive prostate cancer diagnosis method.