Prognostic value of microRNA-125a expression status in molecular groups of pediatric medulloblastoma

Purpose Medulloblastoma (MB) is the most common malignant pediatric brain tumor. Current treatment allows decent survival rates but often with life-long morbidity. Molecular classification provides a base for novel therapeutic approaches. However, these groups are heterogeneous. MicroRNA-125a has a tumor suppressor function. It is downregulated in several tumors. The expression of microRNA-125a in MB patients remains unclear. Therefore, this study was designed to evaluate the expression of microRNA-125a in molecular groups of pediatric MB patients in Egyptian population and its clinical significance. Methods Formalin-fixed, paraffin-embedded tissue blocks from 50 pediatric MB patients were retrospectively collected. Immunohistochemistry for β-catenin, GAB1, YAP1, and p53 was done for molecular classification. MicroRNA-125a expression analysis was done using qRT-PCR. Follow-up data were obtained from patients’ records. Results MicroRNA-125a expression was significantly lower in MB patients showing large cell/anaplastic (LC/A) histology and in the non-WNT/non-SHH group. Lower levels of microRNA-125a showed a tendency toward poor survival rates; however, difference was not significant. Infants and larger preoperative tumor size were significantly associated with lower survival rates. On a multivariate analysis, preoperative tumor size was an independent prognostic factor. Conclusion MicroRNA-125a expression was significantly lower in categories of pediatric MB patients with worse prognosis namely LC/A histology and the non-WNT/non-SHH group suggesting a pathogenetic role. MicroRNA-125a expression could represent a promising prognostic factor and a potential therapeutic target in the non-WNT/non-SHH group which represents the most common and the most heterogeneous group of pediatric MBs coupled with the highest rates of disseminated disease. Preoperative tumor size represents an independent prognostic factor.


Introduction
Medulloblastoma (MB) is the most common malignant pediatric brain tumor [1]. Children aged 4 to 9 years old show the highest incidence at 44% [2].
Studies of MB demonstrated that mutations in cell signaling pathways are associated with specific gene expression signatures, which could be detected using relatively simple and widely available techniques [3]. This has led to the introduction of the molecular subtyping in the 2016 World Health Organization (WHO) classification which is still endorsed by the 2021 WHO classification [4,5].
According to the 2021 WHO classification of central nervous system (CNS) tumors, MBs are classified based on histopathological features and molecular characteristics. According to molecular characteristics, MB is classified into four groups: WNT (wingless-related integration site)activated MBs, SHH (sonic hedgehog)-activated MBs which are separated into TP53-mutant and TP53-wild-types, and non-WNT/non-SHH MBs, comprising group 3 and group 4 which are listed as provisional variants [5].
The current treatment of MB allows 5-year survival rates ranging between 50 and 90%. This wide range is multifactorial depending on age at diagnosis, the presence of 1 3 metastasis at diagnosis, histological pattern, and molecular group. Moreover, current therapeutic modalities come at the expense of life-long morbidity [6]. Therefore, the recent WHO classification has provided a base for novel therapeutic approaches that target cell signaling pathways that are activated aberrantly in MB [7]. However, these groups are heterogeneous, and they need more research to improve the molecular understanding and management of MB [8].
MicroRNAs are small noncoding RNAs that regulate gene expression through sequence-specific binding to 3′-untranslated regions of mRNAs, resulting in translational inhibition or mRNA degradation [9]. Some microRNAs show altered expression in many malignancies suggesting a critical role played by them in tumorigenesis [10].
The published literature largely lacks detailed studies of microRNA expression in MBs. A single study, conducted before the release of the 2016 WHO classification, focused on microRNA expression profiling in MB, where specific microRNA signatures distinguished tumors from normal cerebellar tissues [11].
MicroRNA-125a is located at 19q13. It is expressed in differentiated cells and acts as an antiproliferative factor. MicroRNA-125a expression is downregulated in several tumors [12]. MicroRNA-125a gene has two mature micro-RNAs in its precursor structure: hsa-mir-125a-5p and hsamir-125a-3p. MiR-125a-5p has a prognostic role in several tumors, and its mimic alone or in combination with other therapeutic agents could be a novel therapeutic approach against these tumors [13].
To date, the expression profile of microRNA-125a in MB patients remains unclear. Therefore, this study was designed to evaluate the expression of microRNA-125a in different molecular groups of pediatric MB patients in Egyptian population and its clinical significance.

Methods
The current retrospective study included tumor samples from 50 archived formalin-fixed, paraffin-embedded (FFPE) tissue blocks obtained from MB patients up to the age of 18 years. Samples were obtained from the Pathology Department, Faculty of Medicine, Alexandria University, Egypt, during the period from 2009 to 2018 upon the approval of the ethics committee at the Faculty of Medicine, Alexandria University. None of the patients previously received chemotherapy or radiotherapy.

Patient's data
Clinical and follow-up data were obtained by reviewing patients' records at the Clinical Oncology Department, Borg El-Arab Children's Cancer Hospital, Alexandria University, Egypt. The date of assessment was January 2022.
Patients were classified into standard-and high-risk groups based on age at diagnosis, presence/absence of leptomeningeal dissemination (based on magnetic resonance imaging (MRI) and cerebrospinal fluid (CSF) analysis), extent of residual tumor after resection, and presence/absence of large cell/anaplastic (LC/A) histology. Standard-risk patients are children aged > 3 years without leptomeningeal dissemination, postoperative residual tumor < 1.5cm 2 , and non-LC/A histology [6].
Infants received HIT-SKK protocol with intensive chemotherapy followed by craniospinal radiotherapy at the age of 3 years [14].
Standard-risk patients were treated with whole CNS radiotherapy 23.4 Gy and boost up to 54 Gy with weekly vincristine. High-risk patients were treated with whole CNS radiotherapy 36 Gy and boost up to 54 Gy. Both standardand high-risk patients received adjuvant chemotherapy (8 cycles of cisplatin, Endoxan, and vincristine) [15][16][17].

Histopathologic examination
Diagnosis was made conforming to the WHO criteria of 2021 [5].
For cases of LC/A histology, synaptophysin and integrase interactor 1 (INI 1) IHC were performed to confirm the diagnosis.

Histochemical staining for reticulin
It was performed in cases of histologically detected nodules to differentiate between desmoplastic nodular MB (DNMB) and biphasic pattern of classic MB.
Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) 1. Expression profile analysis of microRNA and quantitative real-time PCR 1.1. RNA extraction from FFPE tissue samples Samples were deparaffinized using deparaffinization solution (cat. no. 19093) and protease digested; then, total RNA isolation from samples was performed using QIAGEN® RNeasy FFPE Kit (ID: 73,504) according to the manufacturer's instructions.

Real-time qRT-PCR
Complementary DNA (cDNA) was synthesized using TaqMan ®MicroRNA Reverse Transcription Kit with miRNA specific primers (Applied Biosystems, USA).
RT reactions for target (microRNA-125a) and spiked-in control (RNU6B) were performed. In brief, 10 ng of RNA, specific stem-loop primers for each miRNA, 10 × buffer, dNTPs, reverse transcriptase, and RNase inhibitor were loaded in a thermal cycler (Primus 25 advanced, PEQLAB, UK) following the manufacturer's protocol.
Then real-time PCR was carried out using the miRNA-specific TaqMan probes and TaqMan universal master mix in Applied Biosystems Ste-pOne™ Real-Time PCR System. The expression levels of the miRNA-125a in each sample were normalized to those of RNU6B. The arithmetic formula (2 −ΔCT ) was used to calculate the median and range for each group as individual data points, and the difference in expression between patient groups was assessed [21].

Statistical analysis
Data were analyzed using the Statistical Package for Social Sciences (SPSS ver. 25 Chicago, IL, USA) [22]. Descriptive statistics were summarized for categorical variables as frequencies and percentages and for continuous variables as mean (M) and median (Mdn). Various tests were used including Mann-Whitney (U), Kruskal-Wallis (H), Pearson's correlation (r), and Spearman's rho correlation coefficient (ρ).
Overall survival (OS) was defined as the time from diagnosis until death or last visit. Event-free survival (EFS) was defined as the time from diagnosis until disease recurrence/progression or death. Log rank test was used to compare the patients' outcome between different groups. Univariate Cox regression was used to evaluate the effect of continuous covariates on patients' outcome.
For the multivariate analysis, Cox survival regression analysis was used. For each variable, hazard ratio (HR) was calculated, and its 95% confidence interval (CI) was determined. In all statistical tests, level of significance of 0.05 was used, which the results were considered to be statistically significant.
35 cases (70%) showed classic histology, while 13 cases (26%) were DNMB, and only 2 cases (4%) showed LC/A histology. One case (2%) belonged to the WNT group, the SHH-activated and TP53-wild-type group represented 13 cases (26%), and 36 cases (72%) belonged to the non-WNT/non-SHH group (Figs. 1 and 2). Clinicopathological characteristics of patients based on molecular classification are listed in Table 1. Statistical analysis showed that histology and preoperative size of the tumor (cm 2 ) showed significant relation to molecular groups (p = < 0.001 and 0.03, respectively), where all SHH-activated and TP53wild-type groups were associated with DNMB histology (100%), and the non-WNT/non-SHH group was associated with classic histology in 94.4% of cases. The preoperative size of tumor in the SHH-activated and TP53-wild-type group (Mdn = 22.5) was significantly larger than in the non-WNT/non-SHH group (Mdn = 16).
One case belonged to WNT molecular group; the patient was 15 years old, and the cross-sectional area of the tumor was 13.5cm 2 , high risk with postoperative residual > 1.5 cm 2 , and showed classic histology. For statistical reasons, this case was not included in statistical analysis Fig. 1.

Survival analysis
Median follow-up time was 37 months. The number of deaths in patients included in the study was 23 (46%). The 5-year OS was 60% with a mean of 77 months and a median of 69 months (95% CI, 61.873-92.418). The number of events was 28 (56%). The 5-year EFS was 42% with a mean of 64.5 months and a median of 39 months (95% CI, 48.821-80.318).

Effect of microRNA-125a expression on clinical outcome of MB patients
MicroRNA-125a expression was divided into low-and high-expression levels in relation to median. The 5-year OS tends to be lower in the cases with low microRNA-125a than the cases with high microRNA-125a (52% versus 70%). The cases with low microRNA-125a levels showed a tendency for lower EFS (Mdn = 33 months) than the cases with high microRNA-125a levels (Mdn = 42 months). Neither of the differences in OS and EFS were significant (X 2 = 0.000 and 0.189, p = 1 and 0.664 respectively) ( Table 3, Fig. 4a, b).

Relations of OS and EFS to different clinicopathological parameters (Table 3)
Kaplan-Meier curves revealed that EFS in patients ≤ 3 years of age (Mdn = 6 months) was significantly lower than patients > 3 years of age (Mdn = 56 months), (X 2 = 6.3, p = 0.012). Univariate Cox regression showed that each unit increase in size of the tumor seems to increase the hazard of the event both for OS and EFS; this was significant in EFS (EFS: HR = 1.078, p = 0.031), but it was not significant in OS (OS: HR = 1.062, p = 0.165).

Multivariate Cox model
It was then performed to identify the relationship between prognosis and the following covariates: microRNA-125a gene expression, age of the patients, and preoperative tumor size. Although the increased microRNA-125a gene expression and older age of the patients seemed to be protective against the event in both OS and EFS, this relation was not statistically significant (OS: HR = 0.870 and 0.265, p = 0.607 and 0.266 respectively; EFS: HR = 0.943 and 0.492, p = 0.792 and 0.524 respectively). Each unit increase in size of the tumor seemed to increase the hazard of the event both for OS and EFS; this was statistically significant in EFS (EFS: HR = 1.085, p = 0.040) ( Table 4).

Evaluation of microRNA-125a gene expression within molecular and risk groups
Low levels of microRNA-125a expression showed a probability of poor outcome within the non-WNT/non-SHH group (50% 5-year OS for cases with low microRNA-125a versus 65% 5-year OS for cases with high microRNA-125a). However, the difference was not statistically significant (X 2 = 0.041, p = 0.840). Conversely, within the SHH-activated and TP53-wild-type group, the 5-year OS for cases with low microRNA-125a (67%) was close to those with high microRNA-125a (70%). This result was not statistically significant (X 2 = 0.398, p = 0.528) (Fig. 4c, d).
Low levels of microRNA-125a expression were possibly associated with poor outcome within both standard-risk group (58% 5-year OS for cases with low microRNA-125a versus 90%  5-year OS for cases with high microRNA-125a) and high-risk group (45% 5-year OS for cases with low microRNA-125a versus 55% 5-year OS for cases with high microRNA-125a). However, the differences were not statistically significant (X 2 = 0.784 and 0.236, p = 0.376 and 0.627 respectively) (Fig. 4e, f).

Discussion
MB is the most common malignant pediatric brain tumor [23]. Advances in molecular profiling identified four molecular groups that guided clinical decision [8]. However, in a recent study, seven molecular subgroups were identified. While WNT remained the same, SHH, group 3, and group 4 were split into further groups, hence reflecting heterogeneity at the molecular level which brings challenge to understanding the biology of MB with impact on treatment options [24]. The current treatment regimens show considerable morbidity. Thus, numerous treatment strategies are designed and tailored in light of our molecular understanding of MBs [25].
In the present work, 5-year OS was 60% which is consistent with those reported in the literature (50-70%) [26]. Age showed a prognostic role where children ≤ 3 years old conferred poor survival, in accordance with findings observed by Zeltzer et al. [27].
It was also found in a uni-and multivariate analysis that preoperative tumor size represents an independent prognostic factor, where the larger the tumor, the poorer the prognosis with significant relation to EFS rates. Our findings are similar to those reported by Evans et al. [28].
MicroRNAs play important roles in regulating gene expression acting as either oncogenes or tumor suppressors under certain conditions [29].
MicroRNA-125a expression is downregulated in several tumors and was found to be a prognostic marker and a potential for therapeutic target in some of these tumors [12,13].
In a single study, conducted before the release of the 2016 WHO classification, Ferretti et al. found that most microRNAs displayed overall downregulated expression in MB [11].

3
The current study is the first study to report the expression profile and clinical significance of microRNA-125a in molecular groups of MB and to evaluate its prognostic significance. Fifty MB tumors were collected retrospectively and classified into three molecular groups. The relation between microRNA-125a and different clinicopathological parameters showed statistically significant relation to histology and molecular groups, while there was no significant relation to age or tumor size. MicroRNA-125a was significantly lower in MB cases with LC/A histology followed by classic histology and then DNMB. It was also significantly lower in non-WNT/non-SHH molecular group than SHH-activated and TP53-wild-type. The one WNT MB case could not be included in statistical analysis.
In the present work, it was found that lower levels of microRNA-125a were associated with parameters described in the literature to have worse prognosis namely LC/A histology and non-WNT/non-SHH molecular group suggesting a pathogenetic role for microRNA-125a in these tumors. Huang et al. stated that LC/A histology was one of the most significant factors associated with worse survival rates [30].
In the current study, SHH-activated and TP53-wild-type MB showed 70% 5-year OS, while non-WNT/non-SHH MB showed 55% 5-year OS, consistent with the findings in literature where SHH-activated and TP53-wild-type MB proved to have a favorable outcome [31], while non-WNT/ non-SHH tumors have the worst prognosis overall [32].
The lower levels of microRNA-125a in categories with worse prognosis could be explained by the findings in literature suggesting that it has tumor suppressor function through inhibiting proliferation, invasiveness, and metastasis [33]. It was found to be downregulated in different tumors, where its upregulation by targeted therapy induced apoptosis. In neuroblastoma, it was found to interfere with the expression of proproliferative truncated isoform of the neurotrophin  receptor tropomyosin-related kinase C (t-trkC), where its underexpression was associated with poor prognosis [34]. In glioblastoma, microRNA-125a is downregulated where it targets podoplanin and inhibits cell migration and invasion [35]. According to Ferretti et al., specific microRNA signatures distinguished tumors from normal cerebellar tissues. Most microRNAs displayed overall downregulated expression in MB, which rescued expression promotes MB cell growth arrest and apoptosis while targeting the proproliferative t-TrkC isoform, suggesting a tumor growth-inhibitory function [11]. Therefore, microRNA-125a could represent a potential target for targeted therapy, a feature observed in clinical trials in different tumors [13,36,37]. Non-WNT/non-SHH molecular group which includes group 3 and group 4 remains the genetically most heterogeneous and least understood fraction of MB cases [38]. In contrast to the WNT and SHH groups, no single somatically mutated gene is present in more than 5-10% of either group 3 or group 4, constituting a significant challenge in the development of innovative treatment strategies for these groups [8]. The significantly lower levels of microRNA-125a in the non-WNT/non-SHH group provide a chance for targeted therapy aiming at increasing the levels of micro-RNA-125a in these patients.
In the current study, the cases with lower levels of micro-RNA-125a had lower 5-year OS than the cases with higher levels, and possibly, these cases will have poor outcome, but the difference was not statistically significant. This could be explained by the fact that it acts as tumor suppressor gene and is consistent with the reported poor survival rates of reduced levels of microRNA-125a in different tumors [39,40].
In the present study, whether microRNA-125a expression could be helpful in segregating patients with poor outcome within both risk groups was assessed. Although standard-risk patients with lower expression levels have lower 5-year OS 58% versus 90% for patients with higher expression levels, the difference was not significant. The effect of stratifying patients based on microRNA-125a expression within each molecular group was also examined and showed that patients with lower levels had lower 5-year OS in non-WNT/non-SHH molecular group. This was not the case in SHH molecular group. These differences were not significant. The current study suggests that microRNA-125a might be helpful in predicting patients' outcome; although the differences in survival rates in the present work were not statistically significant which can be explained by the small sample size, the findings are promising in this context and need to be validated in larger groups.

Conclusion
MicroRNA-125a expression was significantly lower in categories of pediatric MB patients with worse prognosis namely LC/A histology and non-WNT/non-SHH molecular group suggesting a pathogenetic role. MicroRNA-125a expression could represent a promising prognostic factor and a potential therapeutic target in non-WNT/non-SHH molecular group which represents the most common and the most heterogeneous group of pediatric MBs coupled with the highest rates of disseminated disease. Preoperative tumor size represents an independent prognostic factor.
Author contribution All authors contributed to the study conception and design. Collection of study material was performed by Soheir M. Hamam, Eman Abdelzaher, and Hend A. Sharafeldin. Collection and revision of clinical data of the patients were performed by Shady H. Fadel and Hend A. Sharafeldin. IHC was performed by Hend A.
Sharafeldin. PCR was performed by Rasha A. Nassra and Hend A. Sharafeldin. IHC and molecular studies were supervised by Soheir M. Hamam and Eman Abdelzaher. All authors contributed to acquisition, analysis, and interpretation of data. Drafting the work was performed by Soheir M. Hamam, Eman Abdelzaher, and Hend A. Sharafeldin. The work was critically revised by Soheir M. Hamam, Eman Abdelzaher, Shady H. Fadel, and Rasha A. Nassra. All authors reviewed and approved the final version of the manuscript.
Funding Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB).

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations
Ethics approval Methodology for this study was approved by the Ethics committee of Faculty of Medicine, Alexandria University, IRB NO. 00012098, FWA NO. 00018699, (Ethics approval number: 0201228).

Conflict of interest
The authors have no competing interests to declare that are relevant to the content of this article.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.