Abstract
Background/Aim: MicroRNA (miR) is implicated in the development of ovarian cancer (OC), the emergence of therapy resistance, and metastatic spread. In the past decade, a variety of relevant miRs have been identified in the context of OC, including miR-1 and miR-21. miR-21 plays a crucial role in OC carcinogenesis via activation of phosphatidylinositol-4,5-bisphosphate 3-kinase signaling and development of platinum resistance, and has prognostic value. miR-1 has been identified as a tumor suppressor across various cancer entities, with reduced expression in malignant tissue compared to benign. This evidence highlights the potential of miR-1 and miR-21 to serve as diagnostic and prognostic biomarkers in OC. Patients and Methods: We studied the diagnostic potential of serum expression of miR-1 and miR-21 in a cohort comprising patients with malignant and benign ovarian tumors. Furthermore, levels of miR expression were analyzed with regard to clinical outcomes in patients with malignant ovarian tumors to evaluate their prognostic value. Results: We identified significant differential expression of miR-1 (p=0.0397) and miR-21 (p=0.0154) in malignant and benign ovarian tumors: Higher expression of miR-1 was found in patients with benign ovarian tumors, whereas expression of miR-21 was higher in patients with malignant ovarian tumors. In receiver operating characteristic analyses, the diagnostic potential of both miRs was confirmed, however, diagnostic significance was inferior to that of cancer antigen 125. No further value, in particular no prognostic value, was obtained for miR-1 and miR-21 in patients with malignant ovarian tumors. Conclusion: Serum levels of miR-1 and miR-21 might serve as promising biomarkers in the diagnosis of ovarian cancer.
Ovarian cancer (OC) represents the most lethal gynecological malignancy, accounting for >150,000 cancer-related deaths per year worldwide. Whereas early-stage disease exhibits an excellent prognosis with a 5-year survival rate of >90%, advanced tumor disease confers a substantially limited prognosis, with a 5-year survival rate of under 30%. However, the dilemma of this malignancy is that the majority (>70%) of cases are diagnosed at an advanced stage of the disease (1, 2). This is in part due to the fact that early-stage tumor is usually asymptomatic and, compared to other gynecological cancers, such as cervical or breast cancer, there are no effective screening programs. Previous studies that evaluated screening programs based on regular transvaginal sonography or determination of the tumor marker cancer antigen 125 (CA-125) have not led to a reduction in OC-specific mortality. Of note, this includes the high-risk population with familial predisposition for OC (3-5). Hence, screening for OC is not covered by state health insurance. Another diagnostic difficulty is differentiating a benign from a malignant ovarian tumor in the presence of conspicuous ovarian findings on imaging. Even though the determination of the serum tumor markers CA-125 and human epididymis protein 4 (HE4) can be helpful in this regard (6), definitive diagnosis remains surgical removal with histopathological workup of the resected tissue. If a more precise diagnosis could be made preoperatively, a different surgical procedure might be chosen (minimally invasive versus open surgery) and the patient would be spared subsequent operations to complete the surgical staging. In this context, there is an unmet need for molecular diagnostic biomarkers that can be considered for preoperative diagnosis.
MicroRNAs (miRs) are small, non-coding RNA molecules that are involved in almost every biological process by regulating gene expression of target messenger RNAs. Research has shown that a single miR has the capacity to modulate the expression of several hundred different genes thereby exerting an impact on a multitude of different metabolic processes and pathways [reviewed in (7). In the oncological context, there are increasing data demonstrating that altered expression of certain miRs or the presence of certain miRs plays a major role in the development of cancer, the emergence of therapy resistance, and metastatic spread. Furthermore, there is scientific evidence that miRs might also function as tumor suppressors or oncogenes (7-9). This crucial role of miRs in the context of carcinogenesis highlights their great potential to serve as both diagnostic tools and as targets for novel antineoplastic systemic therapies. From a diagnostic point of view, circulating miRs are protected from RNase degradation in the peripheral blood, rendering them potential non-invasive biomarkers (10). It has been shown that miR-21 plays a role in OC carcinogenesis via activation of phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) signaling and development of platinum resistance, and has prognostic value in clinical outcomes (11-13). miR-1 has been identified as a tumor suppressor across various cancer entities, with reduced expression in malignant tissue compared to benign tissue (14, 15).
Here we evaluated whether quantification of miR-1 and miR-21 in serum could be used as a non-invasive diagnostic marker to distinguish a malignant ovarian finding from a benign ovarian finding.
Patients and Methods
Patient cohort. We investigated a well-characterized retrospective cohort comprising patients with malignant and benign ovarian tumors treated at the Department of Obstetrics and Gynecology of the Greifswald University Hospital. Biobanking of tumor tissue and blood serum was performed within the framework of the ‘TOC’ Tumor Registry. The collection of biomaterials was performed at the time of surgery. The study was approved by the Ethical Committee of the University of Greifswald (registration no. III SV 05/04) and each patient provided written informed consent in accordance with the declaration of Helsinki principles. Baseline patient characteristics are provided in Table I.
Clinicopathological characteristics of the malignant (N=48) and benign (N=28) ovarian tumor study population. Comparisons were performed applying Man–Whitney test. Significance threshold p<0.05.
Tissue collection was performed from within the ‘TOC’ Tumor Registry. All patients provided written informed consent prior to the collection of biomaterials. The study was approved by the Ethical Committee of the University of Greifswald (registration no. III SV 05/04) and conducted in accordance with the Declaration of Helsinki principles.
miR preparation from blood. To isolate small RNA species including miRs, a MirVana PARIS kit (Ambion, Carlsbad, CA, USA) was used according to the manufacturer’s instructions. For normalization, 499 μl of serum were spiked with 1.0 μl of 25 nM synthetic Caenorhabditis elegans miR-238 (5′-UUUGUACUCCGAUGCCAUUCAGA-3′). miRs were isolated by phase separation and total RNA concentration was determined using a Nano-Drop 2000c spectrophotometer (Thermo Scientific, Wilmington, NC, USA). miR samples were stored at −80°C until quantification.
Real-time miR quantification. For miR-specific cDNA synthesis, 100 ng of the miR preparation was transcribed with moloney murine leukemia virus reverse transcriptase (Promega, Mannheim, Germany) according to the protocol of Chen et al. (12). cDNA was stored at −20°C.
Quantification of miRNA expression was determined by quantitative polymerase chain reaction using a SensiMix SYBR hi-ROX Kit (Bioline, Luckenwalde, Germany) and a CFX96 Real-Time System (Bio-Rad, Munich, Germany). Expression data was determined applying the 2−ΔΔCT algorithm. Data were analyzed with CFX Manager software (Bio-Rad). Oligonucleotides used are listed in Table II.
miR-specific oligonucleotides (sequence 5′→3′) used in polymerase chain reaction.
Statistic analysis. Descriptive statistics (mean, median and range) and frequencies are presented for the distributions of continuous variables. Comparisons between two groups were statistically tested with Welch’s t-test and a value of p<0.05 was considered statistically significant. Kaplan–Meier estimates with the endpoint overall survival (defined as time from date of biospecimen collection/accession to date of last follow-up or death) and Kruskal–Wallis test were performed with the Statistical Package for the Social Sciences (SPSS) version 28 (IBM, Armonk, NY, USA) and GraphPad Prism software (GraphPad Software, San Diego, CA, USA). Statistical significance was approved at p<0.05.
Results
To evaluate the diagnostic potential of miR-1 and miR-21 in OC, their expression in serum was evaluated in a cohort comprising malignant (N=48) and benign ovarian tumors (N=28). A detailed presentation of the clinicopathological parameters of both cohorts is presented in Table I.
Both miRs studied were significantly differently expressed in malignant and benign ovarian tumors (miR-1: p=0.0397, Figure 1A; and miR-21: p=0.0154, Figure 1B). For miR-21, expression was significantly higher in the group with malignant ovarian tumor group compared to that with benign ovarian tumor. In contrast, for miR-1, higher expression was observed in the group with benign histology compared to the group with malignant histology. Within the group of malignant ovarian tumors, expression levels of both miR did not correlate with the disease stage according to the International Federation of Obstetrics and Gynecology classification based on Kruskal–Wallis test (p>0.05; Table III). Furthermore, there was no significant association between overall survival and miR expression levels (Kaplan–Meier survival analysis, log-rank test; Figure 2A and B).
Analysis of expression of miR-1 (A) and miR-21 (B) in serum obtained from patients with malignant ovarian tumors compared to benign ovarian tumors. Data are expressed as fold change and are presented as the mean±standard deviation.
Correlation analysis of relative miR-1 and miR-21 expression level with International Federation of Obstetrics and Gynecology tumor stage. Kruskal–Wallis test, all p>0.05.
To further evaluate the diagnostic potential of both miRs, receiver operating characteristic (ROC) analyses were performed. ROC analyses showed an area under the curve (AUC) of 0.531 for miR-1 and for miR-21 an AUC of 0.548, respectively. Hence, the diagnostic potential for both miRs was confirmed; however, in comparison, the AUC for CA-125 determined preoperatively was 0.836 (Figure 2C-E). Overall, the determination of the tumor marker CA125 was clearly superior to the determination of both miRs.
Kaplan–Meier estimates showed no significant difference in overall survival in patents with malignant ovarian tumors according to serum expression of miR-1 (A) and miR-21 (B), based on log-rank test (p>0.05). Receiver operating characteristic curves for miR-1 (C), miR-20 (D), and cancer antigen 125 (CA-125) (E), respectively, with the corresponding area under the curve (AUC).
Discussion
In the present study, we analyzed the diagnostic potential of the two miR species miR-1 and miR-21 to discriminate between malignant and benign ovarian tumors. In our study cohort, we observed differential serum expression levels of both miRs in patients with benign and malignant ovarian tumors: Higher serum expression levels of miR-21 (p=0.0154) were observed in patients with malignant ovarian tumor compared to patients with benign ovarian tumors. For miR-1, significantly higher expression levels were obtained in patients with benign ovarian histology (p=0.0397).
There is broad evidence for the important tumorigenic role of miR-21 in cellular growth and carcinogenesis (16, 17). Research has shown a significant up-regulation of miR-21 in malignant tissue compared to normal tissue. In mice, induction of miR-21 led to development of a malignant pre-B-cell lymphoid-like phenotype (18). In the KRASLA2 mouse model of non-small cell lung cancer, deletion of mir-21 resulted in reduced tumorigenesis, whereas its overexpression led to increased tumor growth (19). The miR-21 locus was mapped to chromosomal region 17q23.2, a region that is found to be frequently amplified in breast, lung, hepatocellular, ovarian and prostate cancer (20, 21). In human ovarian cancer cell culture models, suppression of miR-21 led to reduced tumor cell proliferation and enhanced apoptosis that was linked to decreased PI3K signaling and increased expression of phosphatase and tensin homolog (11). A recent study investigated the prognostic potential of miR-21 in a cohort of 80 patients with OC and 50 age-matched healthy individuals. In line with our data from the present study, miR-21 was differentially expressed in serum of patients with OC compared to healthy controls. In particular, higher levels of miR-21 were detected in patients with malignant histology. Of note, the degree of expression correlated positively with the disease stage (22). Comparable data were compiled in another study (12). In our study, the differential expression of miR-21 in patients with OC was confirmed. We found significantly higher serum expression levels of miR-21 in patients with malignant ovarian tumors than in patients with benign ovarian tumors. However, within the group of patients with malignant ovarian tumors, levels of miR-21 expression were not associated with the disease stage. With regard to prognostic value, research has shown that miR-21 expression is correlated with overall survival in patients with gastrointestinal malignant tumors, pancreatic, lung, breast, and liver cancer (13). Another study linked increased miR-21 levels to platinum-based chemotherapy resistance in non-small cell lung cancer (23). In the treatment of OC, platinum resistance is challenging. However, within our study cohort, there was no association of miR-21 expression levels and clinical outcomes. This might be attributable to the small cohort size or a tumor cell-specific phenomenon.
Besides its role in response to platinum-containing chemotherapy, high miR-21 expression levels were observed in patients who displayed poor response to gemcitabine (13). Alongside the diagnostic, prognostic and predictive potential of miR-21, there have been attempts to exploit miR-21 therapeutically. In breast cancer cells, combinatory treatment of paclitaxel and a miR-21 inhibitor demonstrated enhanced cytotoxic effects (24). Currently, there are early clinical trials ongoing to evaluate the clinical efficacy of miR-21 inhibition across various cancer entities with results pending (e.g. NCT02466113). Our data on miR-21 expression levels in OC support the potential of miR-21 to serve as a target in antineoplastic therapy of OC.
There is strong evidence for miR-1 suppression in various cancer types, including gastric, lung, and bladder cancer, compared to non-malignant adjacent tissue (14, 15, 25, 26). Our data further support these findings showing enhanced expression of miR-1 in benign ovarian tumors compared to malignant ovarian tumors. Of note, higher expression levels of miR-1 were associated with favorable outcome in our study cohort of patients with malignant ovarian tumors without reaching statistical significance. This is consistent with data published by Kim et al. correlating high miR-1 levels with sensitivity to cisplatin/fluorouracil in gastric cancer (27). Likewise, miR-1 was found to mediate radiosensitivity in a colorectal cancer cell line model (25). However, in the context of OC, overexpression of miR-1 in human OC cell lines did not result in cell growth inhibition compared to wild-type cells (28). Comparable data was also found in uterine leiomyosarcoma (29). However, in the clinical context, significant differences in expression were observed in benign and malignant ovarian tumors. Compared to the established tumor marker CA-125, ROC analyses showed a significantly inferior performance of the two miRs, miR-1 and miR-21, in distinguishing between malignant and benign ovarian tumors. However, further studies in larger collectives are needed to draw a definitive conclusion regarding their diagnostic potential. An integration of different markers into a diagnostic algorithm might be advantageous in this regard.
Overall, our study highlights the important role of miR-1 and miR-21 in OC tumorigenesis. The extent to which these two miRs can provide decision support in the clinical context needs to be studied in prospective studies evaluating biomarkers in OC.
Acknowledgements
D.J.R. is supported by the BONFOR program of the Medical Faculty of the University of Bonn (grant ID 2021-1A-14).
Footnotes
Authors’ Contributions
D.J.R. and M.B.S were involved in the study design and concept. D.J.R. and M.B.S. drafted the article. D.J.R. and M.C. performed the experimental and statistical analyses. E.E., D.K. and A.M. revised the article for critical intellectual content. All Authors read and approved the final version of the article.
Conflicts of Interest
The Authors declare no conflicts of interest exist in relation to this study.
- Received July 6, 2022.
- Revision received October 5, 2022.
- Accepted October 10, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
