Advances in circulating microRNAs as diagnostic and prognostic markers for ovarian cancer.

Chim et al. 9 investigated circulating fetal nucleic acids in maternal plasma in 2007. They discovered that four abundant placental miRs (miR-141, miR-149, miR-299-5p, and miR-135b) could be detected in maternal plasma during pregnancy, and showed a reduction in post-delivery plasma. Plasma miR-141 increases in the third-trimester of pregnancy. To some degree, this finding may indicate an increase of the placenta size or the miR-141 level in the placenta. Their findings may indicate a new class of biological markers for pregnancy monitoring. Lawrie et al. 10 also found that miR-155, miR-210, and miR-21 are higher in serum from diffused large B-cell lymphoma (DLBCL) patients compared with normal controls, and high miR-21 expression is associated with relapse-free survival. This first description of serum miR in cancer patients suggests that miR has the potential as a non-invasive diagnostic marker for DLBCL and possibly for other cancers. Mitchell et al. 6 isolated 18 to 24 nt RNA fractions from the human plasma sample of a healthy donor. Compared with real-time polymerase chain reaction amplification-generated small RNA cDNA library, they found that 27 of the 125 clones sequenced from this library corresponded to spiked-in size marker oligos or linker-linker dimers. Ninety-one of the other 98 sequences (93%) corresponded to known miRs provide direct confirmation that mature miR is presented in human plasma. Further studies showed that endogenous plasma miR exists in a remarkably stable form and is resistant to plasma RNase activity. After incubation at room temperature for 24 h or eight freeze-thaw cycles, plasma miR did not significantly change, and measurements obtained from serum and plasma were strongly correlated. They established a mouse prostate cancer xenograft model system. Plasma miR-629 and miR-660 were difficult to detect in the control mice (no known mouse homologs), whereas they were easily detected in all of the xenograft mice. The levels of these miRs were moderately correlated with tumor mass, indicating that the plasma miR may be derived from the tumor. Subsequent research in miR-141 showed the greatest differential expression in prostate cancer plasma compared with normal control, and was moderately correlated with the prostate specific antigen levels. The results extend the concept that peripheral circulating miR can serve as a marker for human cancer detection. Compared with former studies, Mitchell’s team comprehensively explained the SP and stability of plasma miR. Researchers have gradually detected plasma/serum-specific miR expression profiling in liver, lung, colorectal, ovarian, and other cancers.

A variety of effective miR detection methods exist. We selected the most appropriate method based on the research purpose and sample type. Cloning and sequencing remain the primary choices in discovering new miR 11 , 12 . Northern blot is an important tool to verify and validate miR, but the method does not apply to high-throughput detection of clinical samples because it is complicated and has low SN 13 . Microarray chip technology is quick and has high-throughput 14 . A variety of miR chips are available, but their reproducibility and accuracy are relatively low. Thus, they are generally used for screening, and their results need to be validated by RT-quantitative PCR 
(RT-qPCR) 15 . RT-qPCR analysis can accurately quantify the plasma/serum miR expression. Various PCR-based miR detection methods exist, such as stem-loop RT-PCR and polyA tailed RT-PCR 16 . RT-PCR is the most common and effective method in quantitative detection of plasma/serum miR. Driskell et al. 17 established a surface-enhanced Raman scattering platform to detect and classify miR. Kato et al. 18 developed a novel fluorescent DNA probe to detect mature miRs with high SP. The miR detection limit has currently reached fmol level. With further research, miR detection methods will continue to be improved and standardized. A circulating miR detection method with high SN and accuracy will eventually be established. In summary, a large number of studies have found that plasma/serum miR has an important function in tumorigenesis, development, invasion, and metastasis, and has potential as a new class of diagnostic markers and therapeutic targets. A study on the relationship between plasma/serum miR and ovarian cancer will serve as an important guide for early diagnosis, clinical prognosis, and personalized ovarian cancer treatment.

The cause of ovarian cancer is not fully understood to date, which could be attributed to several factors, including fertility, ovulation drugs, environmental factors and genetics. Although ultrasound, laparoscopy, cytology, and serum CA125 remain the major approaches for ovarian cancer detection, their invasiveness and low SN and SP hinder most of their clinical use. Studies show that the use of plasma/serum miR as a non-invasive marker of ovarian cancer is expected for early diagnosis and prognosis monitoring, and also has great potential in individualized treatment. Attention should be focused in the clinical use of plasma/serum miR. (1) Plasma/serum miR expression profiling varies significantly with changed physiological or pathological condition, such as pregnancy, heart failure or sepsis 9 , 46 , 47 . (2) In experimental phase, the control group generally consists of healthy people without family history of related diseases. The situation is often more complicated in population-wide clinical screening, and whether plasma/serum miR can exhibit good SN and SP still need to be investigated. (3) The screened out tumor-associated plasma/serum miRs lack further experimental demonstration, and the reference range is not determined. (4) Ruptured erythrocytes release inhibitors of RT-PCR reaction. To control other factors, selecting the most sensitive economical detection methods and the most specific miR combination is the major challenge in clinical practice. Further studies will gradually solve these problems. Plasma/serum miR will become a novel molecular detection and treatment biomarker because of its broad application prospects in future clinical diagnosis and prognosis of ovarian cancer.

Ovarian cancer is one of the most lethal malignant gynecological tumors. Its incidence rate ranks the second among malignant tumors of the genital system following uterine corpus cancer, but its mortality rate was the highest 1 . Given that ovarian cancer is located deep within the pelvis and is difficult to touch, as well as the lack of typical early symptoms and effective diagnostic methods, more than 70% of patients are diagnosed at advanced stage. At this stage, the tumor has metastasized into the peritoneal cavity or to distant locations 2 . Although surgical treatment and chemotherapy of ovarian cancer have improved in recent years, the prognosis of ovarian cancer remains poor. The 5-year survival in patients with advanced ovarian cancer is less than 30% 2 . Statistics from the Tianjin Medical University Cancer Institute and Hospital showed that the incidence of ovarian cancer is 9.71/100,000. It ranks the sixth in female incidence rate of malignant tumors. Ovarian cancer ranks the eighth in female cancer mortality, with a rate of 2.59/100,000 (2004, Tianjin) 3 . Early diagnosis and treatment to prevent and manage ovarian cancer is significant. Current diagnostic methods for ovarian cancer mainly include pelvic examination, transvaginal ultrasound, and serum CA125. However, these methods are not sensitive or sufficiently specific to diagnose ovarian cancer at an early stage. Consequently, finding a diagnostic marker with high sensitivity (SN) and specificity (SP) for early detection remains a major clinical challenge. MicroRNA (miR) is a class of evolutionarily conserved 22-nucleotide noncoding RNAs. These small molecules bind to the 3’untranslated regions of their target mRNAs, mediating translational repression and/or mRNA degradation. Thus, they function as negative regulators of gene expression 4 . MiR was first discovered in Caenorhabditis elegans. Thousands of miRs are involved in multiple biological processes, including cell differentiation, proliferation, apoptosis and metabolism. Certain miRs are involved in the pathogenesis of tumors and function as oncogenes (oncomiRs) or tumor suppressors 5 . A comprehensive and systematic understanding of ovarian cancer-specific miR profiling for early diagnosis, prognosis, and personalized treatment of ovarian cancer would be very beneficial. Early studies on cancer-specific miR expression profiling were only confined to tumor tissue samples. A number of miRs aberrantly expressed in ovarian cancer tissues have important functions in cancer occurrence and development. Thus, miRs may represent a new approach for the detection, diagnosis, and therapy of this deadly disease. However, tissue samples are not easily accessible, which hinders the application of miR in ovarian cancer diagnosis and prognosis. Latest research shows that miR can pass between cells or tissues and organs through blood circulation. Circulatory miRs are mainly from monocytes, plasma, and exosomes 6 , 7 ; they are resistant to degradation of RNase enzyme and are stable in the blood 8 . These characteristics make miRs potentially valuable as novel biomarkers for the detection of early cancer. Given that blood samples can be easily obtained and have the advantages of minimally invasive continuous in vitro testing and high reproducibility, determining disease-specific miRs in peripheral blood to predict and diagnose cancer has become the focus of many studies. This article reviews the development of circulating miRs in ovarian cancer.

Plasma/serum miR research is still in its early stage, and few references can be found in prognosis and personalized treatment of ovarian cancer. Many studies on miR in ovarian cancer tissue have been carried out, and are mainly related to miR-200 and let-7 families 33 . The function of the miR-200 family in ovarian carcinoma is not clearly elucidated. On one hand, miR-200 family members are believed to be metastasis suppressors, and most of the studies performed on the family are about their overexpression in ovarian cancer. On the other hand, some studies demonstrated that miR-200 family members are downregulated 34 or even unchanged 35 . These diverse results may be due to the inclusion of ovarian stromal cells lacking miR-200 expression or differences in normal controls 36 . Most of the let-7 family members are confirmed to be tumor suppressors 37 . The detected abnormal expression of plasma/serum miR is mainly from the miR-200 family, including miR-200a, miR-200b, miR-200c, miR-141 and miR-429. Kan et al. 38 extracted highly expressed miR-200a, miR-200b, and miR-200c in the plasma of serous ovarian cancer. Taylor et al. 8 found that miR-141, miR-200a, miR-200c, and miR-200b are overexpressed in blood-derived exosomes. However, these researches 8 , 38 lacked early-stage disease plasma/serum samples and long-term follow-up data. Thus, no consensus has been reached on the relationship of peripheral tumor-associated miR with metastasis, recurrence, and survival of ovarian cancer. We believe that the application of plasma/serum miR-200 family in clinical prognosis, therapeutic effect, and tumor recurrence monitoring of ovarian cancer patients will be further developed. The standard treatment for advanced ovarian cancer is surgical tumor debulking, followed by platinum-based chemotherapy 39 . To date, few effective treatments for advanced ovarian cancer patients are available mainly because of the molecular heterogeneity of ovarian tumor tissue, which has also led to different clinical effects 32 . Studies are needed to find predictive and prognostic markers to help optimize and personalize treatment of ovarian cancer and to improve the therapeutic effect. MiR functions as an oncogene or tumor suppressor. For miR with oncogenic character, anti-miR oligonucleotides, “miR-sponges”, or “miR making” can be used to silence their oncogenic activity 40 . The expression of tumor suppressor miR can be restored with engineered viral vector approach for anti-cancer therapy 41 . Considering the risk of insertional mutagenesis and the toxicity of viral vector application, the application of miR mimics is a promising alternative for the therapeutic restoration of candidate miR in cancer cells. MiR mimics constitute double-stranded and chemically modified miR molecules, which can be transiently transfected into target cells in precursor form, where they can resume functioning as tumor suppressors 42 . Studies show that a great potential of tumor tissue miR exists in drug resistance and anti-cancer therapy prediction 43 - 45 . Given that miR are also present in the blood, we propose that plasma/serum miR expression profiling can provide personalized treatment information, such as monitoring of treatment effects and predicting drug resistance, in addition to diagnosis and prognosis of ovarian cancer. Plasma/serum miR seems more reliable and more sensitive compared with traditional markers, such as mRNA and protein, because conventional markers are often disproportionate and easily degraded in the blood. However, the relationship between plasma/serum miR and individualized treatment of ovarian cancer is not clearly elucidated. Given that not all patients can tolerate surgery-derived gene analysis, plasma/serum miR has the potential for molecular prediction of therapeutic effects as a non-invasive marker. It also helps in determining treatment processes and therapeutic targets of ovarian cancer patients in this new field. This new method will be used to identify different subtypes of ovarian cancer patients who may be sensitive or resistant to certain drugs. More importantly, we will be able to effectively prevent the occurrence of secondary drug resistance by detecting changes in plasma/serum miR expression level in ovarian cancer. A large number of controlled clinical trials is needed to confirm whether plasma/serum miR is a novel marker for adjustment of dosage regimen, personalized treatment, and therapeutic targets.