Non ‐ invasive prostate cancer detection by measuring expression level of miR-21 and miR-214 in urine

Background: Prostate cancer is the most prevalent cancer among men worldwide. Diagnosis in this cancer is primarily done using aggressive methods, such as biopsy. Laboratory methods, such as measurement of prostate specic antigen (PSA) in the blood, do not have high sensitivity and specicity. MicroRNAs, a group of diagnostic biomarkers, can be used to diagnose diseases such as cancer. MicroRNA is small, non-coding, single-stranded RNA with a length of 21-23 nucleotides. The present study was undertaken to investigate changes in the expression level of miR-21 and miR-214 in the urine to detect prostate cancer. Methods: Testing was done on 70 urine samples from prostate cancer patients (32 metastatic and 38 non-metastatic) and 30 from healthy individuals with negative biopsy reports as the control group. Changes in the expression level of miR-21 and miR-214 in the urine were investigated by using qRT-PCR. Results: miR-21 showed a signicant increase in expression (p = 0.003) and miR-214 showed a signicant decrease in expression (p = 0.000) over the results of the control group. The specicity, sensitivity and AUC for combined panels of both microRNAs were 100%, 72.14% and 0.721 and for PSA were 63.33%, 61.43% and 0.620, respectively. Conclusions: The results show that miR-21 and miR-214 show signicant changes in expression in patients with prostate cancer compared to the control group. A combined panel of miR-21 and miR-214 can be used as a non-invasive method for detecting prostate cancer with higher sensitivity and specicity than the PSA test.


Background
Prostate cancer is the most common cancer in men and usually occurs in men over 50 years of age (1).
Men have a 10% risk of developing this cancer, of which about 3% of cases are fatal (2). The primary symptoms are frequent urination, intermittent and poor urine ow, urinary incontinence, blood in the urine, semen excretion with pain, persistent pain in the lower back and impotence (3). Digital rectal examination (DRE) and biopsy are more common than laboratory tests for measuring prostate speci c antigen (PSA) in the blood.
PSA is a glycoprotein produced both in cancerous and healthy epithelial cells that helps break down collagen in the semen and make it clot. In the case of prostate disease, over 4 ng/ml (recognized as normal) leaks into the bloodstream. PSA testing for diagnosis of prostate cancer has shown low sensitivity and speci city. For prostatitis and benign prostatic hyperplasia (BPH), the level of PSA may increase but, even with prostate cancer, it may be normal (1,4).
A DRE shows changes in the shape and size of the prostate, but is not sensitive (5). If the results of PSA and DRE are not normal, prostate cancer is generally con rmed by biopsy. It is the gold standard for detection of prostate cancer, but is highly invasive, costly and painful (6,7). One of the most effective factors in increasing patient longevity to improve treatment and reduce the nancial and mental costs to the patient is the quick and early diagnosis of cancer. Researchers continue to search for such a method and have identi ed numerous biomarkers for the detection and tracking of cancer. One of the most important is microRNA.
MicroRNA is small, non-coding single-stranded RNA with a length of 21-23 nucleotides that is involved in the regulation of post-transcriptional expression of about 60% of human genes. They also regulate processes such as the proliferation, apoptosis, evolution and differentiation of cancers and increase and decrease gene expression. This means that they can be used as biomarkers in the prognosis and diagnosis of diseases, including cancer, in the blood, urine and tissue (8). They are found in plants, animals, and viruses, but they do not exist in bacteria (9). MicroRNA can directly regulate oncogenes and tumor osteoporosis can play an oncogenic or tumor-assisted role, depending on the type of mRNA function they perform (10).
Early evidence of the association between microRNA and cancer was presented by Calin in 2002. He showed that, in many patients with chronic lymphocytic leukemia, a piece of the genome had been removed, including the miR-15a and miR-16 gene cluster (11). The pro le of the expression of the microRNA involved in prostate cancer was published in 2007 (12).
One type of microRNA that plays a role in prostate cancer is miR-21. This microRNA can inhibit expression of tumors by PTEN gene suppressor and increases the aggressive proliferation of cancer cells (13). This microRNA is classi ed as an incomer. The most important target mRNAs are PDCD4, TPM1, TIMP3, MARCK and PTEN (14,15). It plays an important role in bone metastasis in the advanced stages of prostate cancer (16) and is located on chromosome 17 (17). The miR-214 plays a role in prostate cancer, and the most important target mRNAs are EZH2 and CTNNB1 (14). It is located on chromosome 1 (17).
The aim of the present study was to determine changes in the expression of miR-21 and miR-214 in urine samples of patients with prostate cancer (metastatic and non-metastatic) compared to healthy control subjects, to carry out ROC curve analysis for each microRNA and compare them with the results of the PSA test, to investigate a non-invasive method for diagnosing prostate cancer.

Collection of urine samples
Urine samples were collected from suspected prostate cancer patients who referred to Hashemi-Nejad Hospital in Tehran, Iran for biopsies from April 2017 to March 2018. The urine samples were transferred to the Molecular Genetics Lab of the Cancer Institute of Imam Khomeini Hospital (Tehran, Iran) and stored at 4°C for up to 3 hours and then were centrifuged at 3500 rpm for 5 min at 4°C. Urine samples were stored at -80°C until completed. Based on the pathology report, 70 urine samples were diagnosed with prostate cancer (32 metastatic and 38 non-metastatic). 30 urine samples were from healthy participants who had received negative biopsy results.

Exclusion and inclusion criteria
Inclusion criteria: Patients diagnosed with prostate cancer (according to pathology reports) who have no other related diseases. All participants have been informed and completed a consent form and a questionnaire before undergoing a biopsy. This study has been approved by the ethics committee of Islamic Azad Tehran Medical Sciences University-Pharmacy and Pharmaceutical Branches Faculty.
Exclusion criteria: Patients who were sampled after a biopsy, patients received any treatment such as chemotherapy and radiotherapy and patients with related diseases such as bladder and kidney diseases were excluded.

RNA extraction from urine
RNA was extracted by using 500 μL of Trizol (Invitrogen; USA) in 2 mL of urine samples according to manufacturer protocol. A NanoDrop 2000 spectrophotometer (Thermo Scienti c; USA) was used to detect concentration and purity of extracted RNA (ng/μL) and RNA with OD260/OD280 ratio between 1.8-2 was taken. The extracted RNA was stored at -80°C until cDNA synthesis.

cDNA synthesis
RNA was converted to cDNA by using a Pars Genome MiR-Amp kit (Iran) according to manufacturer protocol. The cDNA was stored at -20°C until qRT-PCR was performed.

Quantitative reverse transcription PCR (qRT-PCR)
The expression level of miR-21 and miR-214 were measured for 70 prostate cancer patients (32 metastatic and 38 non-metastatic) and 30 healthy participants using Sybr Green Master mix (Takara; Japan) by using Real-time PCR system (Bioneer; South Korea). Urine samples were normalized to internal standard control U6snRNA. This method was chosen for its precision, high sensitivity and low cost. The forward and reverse primer sequences for miR-21, miR-214 and internal reference (U6SnRNA) were designed (Table1).
The reaction was carried out in a volume of 14 μL which included 7 μL of Sybr Green Master mix, 1 μL primer (10 pmol), 3μL cDNA diluted 1 to 4 and 3 μL of distilled water.
The Real-time PCR conditions was as follows: Polymerase activation at 95°C for 12 min, denaturation at 95°C for 15 sec, annealing 60°C for 30 sec and extension at 72°C for 15 sec for 42 cycles. The experiment was repeated three times. For each microRNA and U6snRNA, no template control (NTC) was provided which lacked a template (cDNA) for reproduction. PCR performance was calculated using the standard curve and serial dilutions of 107%. Finally, 2% agarose gel electrophoresis was used to ensure the reproduction and speci city of the components.

Results
The clinical and pathological data of the patients with prostate cancer and healthy controls with negative biopsy results are based on the pathology results shown ( Table 2). The presence of miR-21 and miR-214 was identi ed in the prostate cancer (PCa) samples and miR-21 and miR-214 showed signi cant change in the expression of PCa patients compared to the control group.

Expression of miR-21 in the urine samples of people with prostate cancer
The miR-21 increased signi cantly (p = 0.003) in the prostate cancer group compared with the control group with a fold change of 3.493. The expression of miR-21 in the metastatic and non-metastatic prostate cancer subgroups was examined and was signi cantly different. The metastatic subgroup showed a signi cant increase in expression (p = 0.042) compared to the non-metastatic subgroup (p = 0.036). The rate of expression (fold change) miR-21 in the metastatic subgroup was 3.911 and in the nonmetastatic subgroup was 3.176 ( Fig. 1).

Expression of miR-214 in the urine samples of people with prostate cancer
The miR-214 showed a signi cant decrease in expression (p = 0.000) in patients with prostate cancer with a fold change of 0.008. Expression of miR-214 in the metastatic and non-metastatic prostate cancer subgroups was investigated. In the metastatic subgroup, the expression level (fold change) was 0.007 and, in the non-metastatic subgroup, the expression level (fold change) was 0.011. The metastatic subgroup showed a signi cant decrease in expression (p = 0.000) compared to non-metastatic subgroup (p = 0.000) (Fig. 2).
The expression of miR-214 and miR-21 in the prostate cancer and control groups was calculated separately by using the formula 2 −∆Ct and is plotted in Fig. 3.

The ROC curve analysis of miR-21, miR-214 and PSA
The ROC curve was used to determine the speci city and sensitivity of miRNA for identi cation of prostate cancer. This statistical analysis shows the e cacy of miRNA as diagnostic biomarker. The speci city index provides the ability to diagnose a cancer sample from among non-cancerous samples. The sensitivity index con rms its ability to diagnose the disease for the patient. The area under the curve (AUC) indicates the biomarker's prediction power.

Discussion
Changes in lifestyle and eating habits, alcohol and tobacco use, electromagnetic radiation, environmental pollution have increased the incidence of cancer year by year. Most diagnostic methods are invasive or do not have high speci city and sensitivity.
In recent years, a large number of diagnostic biomarkers in the biological samples of patients have been introduced for the diagnosis of cancer. These biomarkers are often proteins or nucleic acid, but they have low speci city and sensitivity, are unstable and expensive. Problems with laboratory review mean that most biomarkers are seldom used or approved by the World Health Organization. miRNA is a new generation of diagnostic biomarkers, including in tumor tissue, show different expression patterns for cancer. The pattern of expression of miRNA in human cancers is tissue-dependent and miRNA is signi cantly more stable under severe environmental factors such as salt, alkali, proteases and nucleases (18,19). Because microRNA circulating in the blood and other biological uids are more stable with a longer half-lives than proteins, their pro le can be used to differentiate cancer from other diseases (20). Biological uids such as blood and urine are convenient and non-invasive vehicles for collection and provide good samples for laboratory and diagnostic research. Finding miRNA biomarkers in such uids that can accurately detect cancer is a major goal of researchers.
Stuopelyte et al. in Lithuania studied the expression of miR-21, miR-19a and miR-19b in tissue and urine specimens. The expression of miRNA was evaluated by RT-qPCR. The results of this study showed that the miRNAs were expressed in the tissue and in the urine sedimentation of the prostate cancer group. Urinary miR-21 has a signi cant potential for differentiation of prostate cancer patients and individuals with BPH. The combined panels of the three urine miRNA types has more power to detect prostate cancer compared to the PSA (21). The results of the present research showed that miR-21 in patients with prostate cancer increased signi cantly (p = 0.003) in comparison with the healthy subjects. It was shown that the expression of this miRNA was signi cantly different between the metastatic and non-metastatic subgroups. In the metastatic subgroup (p = 0.036), the expression was higher than in the non-metastatic subgroup (p = 0.042). The miR-214 showed a signi cant decrease in expression in the prostate cancer group (p = 0.000) compared with the healthy control group. The results between the metastatic and non-metastatic subgroups indicated that expression of miR-214 was signi cantly different between groups and was lower in the metastatic subgroup (p = 0.000) than in the non-metastatic subgroup (p = 0.000).
The results of ROC curve analysis showed that miR-214 had greater sensitivity and diagnostic power than miR-21 and the combination panel of the two miRNA types had greater diagnostic ability and higher speci city and sensitivity than the PSA test. Compared to other methods, such as biopsy, it is superior.
The results of this study were consistent with the results of other studies. Figure 1 Expression of miR-21 in Prostate cancer, metastatic and Non-metastatic groups. Expression of miR-214 in Prostate cancer, metastatic and Non-metastatic groups.