Associations of miR-181a with Health-Related Quality of Life, Cognitive Functioning, and Clinical Data of Patients with Different Grade Glioma Tumors

Gliomas are central nervous system tumors with a lethal prognosis. Small micro-RNA molecules participate in various biological processes, are tissue-specific, and, therefore, could be promising targets for cancer treatment. Thus, this study aims to examine miR-181a as a potent biomarker for the diagnosis and prognosis of glioma patients and, for the first time, to find associations between the expression level of miR-181a and patient quality of life (QoL) and cognitive functioning. The expression level of miR-181a was analyzed in 78 post-operative II-IV grade gliomas by quantitative real-time polymerase chain reaction. The expression profile was compared with patient clinical data (age, survival time after the operation, tumor grade and location, mutation status of isocitrate dehydrogenase 1 (IDH1), and promoter methylation of O-6-methylguanine methyltransferase). Furthermore, the health-related QoL was assessed using the Karnofsky performance scale and the quality of life questionnaires; while cognitive assessment was assessed by the Hopkins verbal learning test-revised, trail-making test, and phonemic fluency tasks. The expression of miR-181a was significantly lower in tumors of grade III and IV and was associated with IDH1 wild-type gliomas and a worse prognosis of patient overall survival. Additionally, a positive correlation was observed between miR-181a levels and functional status and QoL of glioma patients. Therefore, miR-181a is a unique molecule that plays an important role in gliomagenesis, and is also associated with changes in patients’ quality of life.


Introduction
Astrocytic origin gliomas (astrocytomas) are central nervous system (CNS) tumors. According to the World Health Organization's (WHO) 2016 classification of CNS tumors, these tumors are graded as grade II-diffuse astrocytoma, grade III-anaplastic astrocytoma, and grade IV-glioblastoma (GBM), of which GBM is the most common and malignant primary brain tumor [1]. The treatment of GBM remains a challenge. Despite available treatment options (surgical resection, adjuvant radiotherapy, and chemotherapy), the average life expectancy of patients diagnosed with GBM is only slightly above one year [2,3]. The dismal patient outcome is due to a tumor's ability to regrow (~90%) and molecular heterogeneity [4]. Genomic and transcriptome studies revealed that histologically identical GBM forms may belong to different molecular subtypes, leading to different responses to treatment and patient life expectancy [5][6][7][8]. Therefore, more precise molecular identification

Association of miR-181a Expression with Patient Clinicopathological Data
To reveal the importance of miR-181a in the pathogenesis of astrocytomas, the mRNA expression of miR-181a was analyzed in 78 different malignancy grade tumors. In Figure 1, it was demonstrated that the gene expression was diversified. The majority of GBM patients with lower than average (<−1.26) mRNA expression died within 2 years, while patients with lower-grade tumors had a higher expression of miR-181a and survived 2 to 6 years.
As follows, to determine whether expression changes of miR-181a were significantly associated with patient clinicopathological characteristics, miR-181a expression was divided into "low" (<mean of miR-181a mRNA expression) and "high" (≥mean) gene expression groups. The analysis reveals that a higher expression of miR-181a is significantly associated with younger age of patients (<54-year, p = 0.006), lower tumor malignancy grade (p = 0.036), and gliomas with a mutant-type of IDH1 (p = 0.002) (see Table 1; Figure 2a-c). According to the Kaplan-Meier analysis, patients with a higher expression of miR-181a have a significantly higher chance of longer survival, compared with patients with low gene expression values (Log-rank test, χ 2 = 4.465, df = 1, p = 0.035) (see Figure 2d). The median survival time was 8.9 months longer in the patient group with higher expression of miR-181a.    The univariate cox regression analysis reveals that patients' clinical characteristics such as age, tumor malignancy grade, and IDH1 status, as well as the expression of miR-181a, are significantly associated with patient overall survival (OS). However, according to the multivariate cox regression analysis, the tumor stage and IDH1 status are the only covariates significantly associated with the OS of glioma patients (see Table 2). The univariate cox regression analysis reveals that patients' clinical characteristics such as age, tumor malignancy grade, and IDH1 status, as well as the expression of miR-181a, are significantly associated with patient overall survival (OS). However, according to the multivariate cox regression analysis, the tumor stage and IDH1 status are the only covariates significantly associated with the OS of glioma patients (see Table 2).

Associations of miR-181a Expression with IDH1 Status of GBM Tumors
In addition, it was noted that all IDH1 mutant GBM tumors were detected in the higher miR-181a mRNA expression group, and the majority of IDH1 wild-type GBM tumors (62%, 32/52) were detected in the lower gene expression group (see Figures 1 and 2b). According to Student's t-test, the noted difference was statistically significant (p = 0.005). To find out whether miR-181a expression was associated with patient survival, only IDH1 wild-type GBM tumors were selected and divided into two groups, according to the median of miR-181a expression. Kaplan-Meier analysis shows the tendency that patients with a higher expression of miR-181a have a significantly higher chance of longer survival, compared with patients with low gene expression values (see Figure 3). According to the log-rank test, the difference was not statistically significant (χ 2 = 2.64, df = 1, p = 0.104); however, the Gehan-Breslow-Wilcoxon method, which gives more weight to deaths at early time points, shows a statistically significant difference (χ 2 = 5.83, df = 1, p = 0.016).
HR-hazard ratio. Significant associations p < 0.05 indicated in bold numbers.

Associations of miR-181a Expression with IDH1 Status of GBM Tumors
In addition, it was noted that all IDH1 mutant GBM tumors were detect higher miR-181a mRNA expression group, and the majority of IDH1 wild-ty tumors (62%, 32/52) were detected in the lower gene expression group (see Figu 2b). According to Student's t-test, the noted difference was statistically signif 0.005). To find out whether miR-181a expression was associated with patient surv IDH1 wild-type GBM tumors were selected and divided into two groups, accord median of miR-181a expression. Kaplan-Meier analysis shows the tendency tha with a higher expression of miR-181a have a significantly higher chance of longer compared with patients with low gene expression values (see Figure 3). Accord log-rank test, the difference was not statistically significant (χ 2 = 2.64, df = 1, p however, the Gehan-Breslow-Wilcoxon method, which gives more weight to early time points, shows a statistically significant difference (χ 2 = 5.83, df = 1, p =

Associations of mir181a Expression with Functional Status, QoL, and Cognitive Functioning of Patients
The correlation analysis was performed to reveal miR-181a expression associations with variables related to the quality of functioning (see Table 3). The results reveal that the expression of miR-181a positively correlates with general quality of life (EORTC QLQ-C30; r = 0.310, p = 0.010) and functional status, evaluated by a clinician (KPS; r = 0.237, p = 0.049) in all glioma patients. Interestingly, a statistically significant positive correlation between miR-181a and a better quality of life was reflected in the group of men (p = 0.009) rather than women (p = 0.469). In addition, as patients with GBM experience more severe symptoms, it was decided to analyze this subgroup separately. The expression levels of miR-181a show a significantly positive correlation with patient quality of life (EORTC QLQ-C30; r = 0.290, p = 0.041), but an inverse correlation with patient memory (HVLT-R; r = −0.291, p = 0.040). However, no statistically significant correlations of miR-181a with tumor-related symptoms (EORTC QLQ-BN20), depression (PHQ-9), cumulative learning (HVLT-R), psychomotor speed (TMT-A), executive functioning (TMT-B), and/or verbal fluency were determined.

Discussion
Numerous studies have shown that miRNAs, which regulate biological processes such as cell proliferation, apoptosis, metabolism, and/or differentiation, are thought to have clinical potential in cancer prognosis and treatment [21]. Among the so-far-characterized miRNAs, miR-181a is involved in several types of cancer [18]. A significant upregulation of miR-181a level has been found in breast cancer [22], ovarian cancer [23], pancreatic cancer [24], hepatocellular carcinoma [25], and oral squamous cell carcinoma [26], whereas evident downregulation of miR-181a has been detected in non-small cell lung cancer [27] and prostate cancer [28], as compared with healthy controls. Our study also reveals that the expression levels of miR-181a decrease during astrocytoma progression. Comparable results were observed by other researcher groups, which demonstrates the downregulation of miR-181a in all grade glioma tumors (WHO II-IV), GBM cell lines, and glioma stem cells (GSCs), as compared with normal brain tissue, astrocytes, and differentiated GBM cells, respectively [29][30][31][32]. Importantly, the overexpression of miR-181a inhibited proliferation, migration, invasion, epithelial-mesenchymal transition (EMT), and induced apoptosis of GBM cells [30,33,34]. The process of apoptosis was modulated by targeting the apoptosisrelated genes (p53, Bax, Bcl-2, Bim, etc.) [19], while proliferation was modulated by the downregulation of the MAPK pathway [35]. In addition, the upregulation of miR-181a sensitized the GBM cells to temozolomide (TMZ) and radiation treatment [34,36] also suppressed the formation of GSCs and inhibited GBM tumorigenesis [31]. More importantly, the study by Wu et al. showed the clinical significance of circulating miR-181a in patients with glioma tumors. Before the operation, circulating miR-181a was found to be downreg-ulated in the plasma of GBM patients as compared with lower-grade tumors [37]. After 10 days, the levels of miR-181a increased more than 10-fold. In addition, a lower expression of circulating miR-181a was significantly associated with poor OS [37], as was demonstrated in our study with astrocytoma tumors. Therefore, the aberrantly downregulated miR-181a could be a critical factor that contributes to the malignant appearance of astrocytoma.
Our study also demonstrates a significant association between the expression levels of miR-181a and patient age, as well as IDH1 status. However, according to the multivariate Cox regression analysis, only the tumor grade and IDH1 mutation were the best predictors of patient OS in our study cohort. Importantly, we noted that the mutant form of IDH1, which is the factor of a good astrocytoma patient survival prognosis, was found in GBM tumors with an increased expression of miR-181a. All these GBM patients with IDH1 mutation survived more than 14 months, and one of them survived even more than 57 months. We hypothesize that the expression of mirR-181a may influence the activity of the IDH1 protein which affects several major metabolic processes of the cells [38] and, therefore, has an impact on the OS of GBM patients. This relation between IDH1 and miR-181a in GBM patients was also observed by Sippl et al. [39]. They determined the inverse correlation between the expression of miRNA-181a2 and mRNA expression of IDH1 (p = 0.06, r = −0.55). Nevertheless, studies on a larger sample would be needed to confirm our observations. Next, we wanted to find out whether the survival of patients with IDH1 wild-type GBM tumors depends on miR-181a expression level. Since the survival of GBM patients after surgery is generally short, we additionally used the Gehan-Breslow-Wilcoxon method, which gives more weight to deaths at early time points. The analysis demonstrates that patients with a lower expression of miR-181a have significantly worse OS as compared with those with a higher gene expression level. However, the opposite effect was noted by another research group. They demonstrated that in patients with IDH1 wild-type GBMs, low miR-181a2 expression correlated with a prolonged OS (p = 0.019), [39]. The discrepancy may have occurred because different miR-181a isoforms were used in the analyses, which might have distinct biological functions. In our study, we analyzed the miR-181a1 isoform, which is located on chromosome 1 (37.p5), while miR-181a2 is situated on chromosome 9 (37.p5) [18]. Both isoforms produce almost identical mature miR-181a, but could be regulated by distinct molecules and, therefore, be differently expressed. For example, in human blood natural-killer cells, pri-miR-181ab-2 levels were higher than pri-miR-181ab-1. The immunosuppressive cytokine, TGF-β, suppressed pri-miR-181ab-1 expression while elevating pri-miR-181ab-2 expression. On the contrary, interleukins −2, −15, and −12/−18 increased the expression of pri-miR-181ab-1, but inhibited pri-miR-181ab-2 [40].
To the best of our knowledge, this was the first study comparing expression levels of miR-181a with patient quality of life. The analysis reveals that miR-181a positively correlates with the general quality of life subjective reported by patients themselves, as well as functioning status, evaluated by the treating clinician. However, no statistically significant correlations were found between miR-181a expression and more specific tumorrelated symptoms, levels of depression, or cognitive functioning. A relatively small sample size, especially of tumor grades II and III, could affect the statistical power of our analysis. Furthermore, cognitive impairment and the profile of tumor-related symptoms are highly dependent on the localization of the tumor and other clinical factors, such as edema, tumorinduced compression to nearby tissues, or the frequency of seizures. Thus, it might be that relationships were lost due to the heterogeneity of the sample regarding tumor locations and other clinical characteristics. However, previous findings by our team group showed that miR-34a or miR-181b/d expression levels were related to patients' functioning and tumor-related symptoms [41,42]. Therefore, we believe that current findings support the idea that levels of miR-181a expression might be valuable not only in predicting longer survival but also in better general functioning. Still, further studies are needed to clarify the mixed findings in male and female subgroups.

Samples and Patient Clinicopathological Data
Seventy-eight samples from patients with the diagnosis of II-IV grade glioma tumors were analyzed in the study. All patients underwent neurosurgery at the Department of Neurosurgery, Hospital of Lithuanian University of Health Sciences, from 2015 to 2018. After surgical resection, tumor samples were frozen in liquid nitrogen. The diagnosis was confirmed by the pathologists. The study was approved by the Kaunas Regional Biomedical Research Ethics Committee, and written patient consent was taken from each patient before inclusion in the study.
The clinical data, such as gender, age at the time of surgery, tumor grade, isocitrate dehydrogenase 1 (IDH1) status (the R132H mutation in the IDH1 gene), and methylation of O-6-methylguanine methyltransferase (MGMT) were collected from medical records. According to the WHO classification 2016 [1], there were 14 tumor samples of grade II, 6 of grade III, and 58 of grade IV (GBM). There were 35 women and 43 men, with a mean age of 54 years (range: 24-80 years). The overall survival of the patient was calculated from the date of tumor resection to the date of patient death or database closure (5 October 2021).

Functional Status
Patient functional status was assessed by a treating neurosurgeon during the hospital stay using the Karnofsky performance scale (KPS) [43]. The KPS measures a patient's ability to carry on his/her normal daily activities and dependence on help and nursing care using an 11-point rating scale. The total KPS score ranges from 100 (normal functioning) to 0 (death), with higher scores indicating better daily functioning and higher functional independence. Data on functional status was available for 70 (88.6%) of patients.

Quality of Life Assessment
Quality of life (QoL) assessment was performed 2-3 days before the neurosurgery.
Patients were asked to fill out self-report questionnaires addressing their symptoms and quality of life. If assistance was needed due to reading, motor, or visual problems, questionnaires were filed with the help of a medical psychologist. Data on QoL was available for 68 (87.2%) of the patients.
The European Organization for Research and Treatment of Cancer quality of life questionnaires QLQ-30 [44] and QLQ-BN20 [45] for brain tumor-related symptoms were used in this study. The QLQ-C30 contains 30 items that were designed to assess global health status, subjectively reported functional status, role functioning, emotional functioning, cognitive functioning, social functioning, and various cancer-related symptoms. Raw scores were linearly transformed to 0-100 scales, with higher scores indicating better quality of life.
The QLQ-BN20 is a 20-item self-rating scale specifically developed for the assessment of health complaints in brain tumor patients. The questionnaire contains many common BTrelated symptoms, including future uncertainty, visual disorder, cognitive impairment, etc. The QLQ-BN20 scores were linearly transformed to a 0-100 scale, with a higher score indicating greater BT-related symptom severity.
In addition to QoL instruments, we also included a measure of depression, as depression is significantly related to decreased QoL in many patient populations. The patient health questionnaire-9 (PHQ-9) [46] was chosen for the assessment of current depressive symptoms. The PHQ-9 is based on the diagnostic statistical manual-IV depression diagnostic criteria, and it is recognized as a valid and reliable tool for depression screening in glioma patients [47,48].

Assessment of Cognitive Functioning
A cognitive assessment was performed by a medical psychologist 2-3 days before the neurosurgery at the Department of Neurosurgery of the Hospital of LUHS. A set of neuropsychological tests, recommended for the assessment of treatment outcomes in glioma studies, was used [49].
The Hopkins verbal learning test-revised (HVLT-R) was used for verbal memory assessment [50]. The test consists of 12 words that are read aloud for three trials, each trial followed by a patient's free recall. After an approximately 20 min. delay, during which other tests are administered, the patient is asked to recall the list of words. Two scores: cumulative learning (total number of words recalled in trials 1, 2, and 3) and delayed recall (number of words recalled after a delay) were analyzed in this study.
The Trail-making test (TMT, parts A and B) was used for the assessment of psychomotor speed and executive functioning [51]. During the task, a patient is asked to connect a sequence of 25 targets (numbers 1, 2, 3, etc. in Part A, and numbers and letters in Part B) on a sheet of paper. The time of completion (in seconds) is considered as an indicator of psychomotor speed and executive functions.
Verbal fluency was measured using phonemic fluency tasks [52]. Patients were asked to produce as many words as possible beginning with a specific letter within the one-minute interval. Three trials using the letters K, A, and S were performed. The total number of words produced during three trials is used as a verbal fluency indicator.
As cognitive performance is sensitive to aging processes, all primary scores of neuropsychological tests were transformed to age-adjusted T scores (Mean 50, SD 10) using available norms for the Lithuanian population, with higher scores indicating better cognitive function. Memory and verbal fluency data were available for 68 (87%) of the patients, and data on psychomotor speed and executive function was available for 62 (79.5%) of the patients, as some patients were not able to perform these tasks due to visual or motor impairment.

Gene Expression Analysis
Small RNAs were extracted from frozen tumor tissue using a mirVana™ miRNA Isolation Kit (Life Technologies, Carlsbad, CA, USA, cat. no. AM1560), according to the manufacturer's instructions. The quality and concentration were determined by NanoDrop 2000 (Thermo Fisher Scientific, Wilmington, DE, USA). Following this, 10 ng of purified micro RNAs were synthesized to cDNA using the "TaqMan Advanced miRNA cDNA Synthesis Kit" (Thermo Fisher Scientific, Pleasanton, CA, USA, cat. no. A25576).
The RT-PCR with TaqMan™ Fast Advanced Master Mix (Life Technologies, Carlsbad, CA, USA, cat. no. 4444965) was performed to analyze miR-181a expression changes in II-IV grade glioma tumors and healthy human brain RNA sample "FirstChoice Human Brain Reference RNA" (RHB; Ambion, Austin, TX, USA, cat. no. AM7962). The reaction consisted of 6 µL of TaqMan ® Fast Advanced Master Mix, 0.6 µL of hsa-miR-181a-3p probe (Applied Biosystems, Foster City, CA, USA, Assay ID: 479405_mir), 3 µL of cDNA sample, and nuclease-free water to a total volume of 12 µL. The reactions were performed in the RT-PCR System "Applied Biosystems 7500 Fast" (Applied Biosystems, Foster City, CA, USA) using a fast-cycling program. In addition, the housekeeping genes has-miR191-5p (Assay ID: 477952_mir), has-miR361-5p (Assay ID: 478056_mir), has-miR345-5p (Assay ID: 478366_mir), and has-miR103a-3p (Assay ID: 478253_mir) were measured to normalize the data. The average of the housekeeping genes was used for the comparative 2 −∆∆Ct method, in which gene expression in tumor samples was compared to healthy brain tissue.

Statistical Analysis
Statistical programs SPSS (version 25.0, IBM SPSS, Armonk, NY, USA) and GraphPad Prism (version 7.0, Graph-Pad Software, San Diego, CA, USA) were used for data analysis. The normality was confirmed by the Kolmogorov-Smirnov test. The Student's independent t-test was applied to evaluate the difference in miR-181a expression between the two groups. Meanwhile, one-way ANOVA with Tukey's correction was used for the three groups. The Chi-square test was used for the comparison of categorical variables and the Pearson correlation was used for the quantitative variables. Patient survival was evaluated according to the Kaplan-Meier curves with log-rank or Breslow tests. Univariate and multivariate (with the backward conditional method) Cox regression analyses were used to evaluate the relationships between clinical, psychological, and molecular variables and patient overall survival. The significance level was defined as p < 0.05.

Conclusions
The study suggests that miR-181a could be a promising biomarker for glioma patients, since the expression of miR-181a decreases during tumor progression and the downregulation of the gene is significantly associated with the worst patient survival prognosis. More importantly, lower expression levels of miR-181a are also related to wild-type IDH1, older age, as well as worse patient quality of life and functioning (see Figure 4).  However, the findings of this study should be seen in light of some limitations. One of them is a small number of tumor samples and incomplete sample information. Unfortunately, not all of the patients had information about IDH1, MGMT status, and functional data. Furthermore, patients' genetic and medical histories were not considered during this study, which might affect the evaluation of patients' health-related quality of life. This may have affected the reliability of the results. Therefore, follow-up studies are needed to confirm the prognostic value of miR-181a in glioma patients. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data is available from the corresponding author on reasonable request.