The Value of miR-296 and miR-517c in Evaluating the Prognosis of Patients with Glioma after Radiotherapy and Chemotherapy.

OBJECTIVE: To explore the value of miR-296 and miR-517c in evaluating the prognosis of patients with glioma after radiotherapy and chemotherapy.
METHODS: 732 patients with glioma were selected from January 2012 to January 2018. According to the effect of postoperative chemotherapy, the patients were divided into two groups: the effective group and the ineffective group. The serum miR-296, miR-517c, and clinicopathological parameters of the two groups before chemotherapy were compared. The factors affecting the sensitivity of radiotherapy and chemotherapy and the predictive efficacy of miR-296 and miR-517c on the prognosis of patients were analyzed.
RESULTS: The expression level of miR-296 in glioma tissue was significantly correlated with tumor pathological grade and depth of invasion (P < 0.05), and the expression level of miR-296 in glioma tissue was significantly correlated with tumor pathological grade (P < 0.05). Logistic regression analysis showed that tumor size, WHO grade, and serum miR-296 and miR-517c levels were all factors affecting chemosensitivity (P < 0.05). The sensitivity, specificity, accuracy, and AUC of serum miR-296 prediction were 76.95%, 89.64%, 85.35%, and 0.891, respectively. The sensitivity, specificity, accuracy, and AUC of serum miR-517c prediction were 72.81%, 86.50%, 82.19%, and 0.739, respectively.
CONCLUSION: miR-296 and miR-517c are closely related to the chemosensitivity and prognosis of glioma patients. High levels of miR-296 and miR-517c can enhance chemosensitivity and serve as reliable indexes to predict the prognosis of patients with glioma.

1. Introduction

Glioma is the most common malignant tumor of the central nervous system [1, 2], accounting for more than 80% of malignant brain tumors in the world [3]. Glioma, as the most common primary malignant intracranial brain tumor, is a serious threat to the life and health of women and men [4]. Because of its high recurrence and high invasiveness, glioma has become the most difficult type of primary tumor in the adult central nervous system and has strong lethality [5]. The pathogenesis of glioma involves multigenes and multisteps, but the pathogenesis has not been elucidated so far [6, 7]. The clinical treatment mainly adopts the comprehensive treatment mode of operation, radiotherapy, chemotherapy, and biological immunotherapy [8, 9]. With the deepening of gene-level research, the molecular mechanism of glioma has been deeply explored. O6-methylguanine-DNA methyltransferase (MGMT)-initiated methylation, loss of heterozygosity in 1p/19q, isocitrate dehydrogenase-1 (IDH1) gene mutation, and loss of ATRX expression have become important molecular biomarkers in the pathological diagnosis of gliomas and enrich people's understanding of tumor gene variation [10, 11].

miRNA is a kind of endogenous noncoding RNA. Its abnormal expression is not only closely related to the occurrence and development of glioma but also provides a new idea for exploring new diagnoses and treatments for glioma [12]. Upregulating the expression of miR-124 can downregulate the expression of CMTM6, thus inhibiting the invasion and migration of glioma cells. miR-199a-5p can downregulate the expression of DDR1 and inhibit the proliferation and migration of human glioma cells. The decrease of miR-145 expression in serum exocrine and tissue of glioma patients is related to tumor diameter, WHO pathological grade, and KPS score, while the expression of SMAD2 protein in glioma tissue increases accordingly. The miR-145/SMAD2 axis is expected to become a potential target for the diagnosis and treatment of glioma. Therefore, the exploration of biomarkers related to the occurrence and progression of glioma can provide a new direction for its accurate treatment [13]. Microribonucleic acid miR-296and miR-517c are a class of endogenous noncoding single-stranded small molecules RNA, which can bind to mRNA to block the expression of protein-coding genes and participate in the regulation of a variety of physiological and pathological functions [14]. Studies have shown that serum miR-296 and miR-517c are abnormally expressed in breast cancer, colorectal cancer, prostate cancer, and other malignant tumors and are closely related to the occurrence and development of these malignant tumors. However, there are few studies on the expression of miR-296 and miR-517c in gliomas [15, 16]. The purpose of this study was to analyze the expression of miR-296 and miR-517c in gliomas and their correlation with clinicopathological features and survival after radiotherapy and chemotherapy, in order to explore the expression of miR-296 and miR-517c in gliomas and their value in the prognosis of glioma patients after radiotherapy and chemotherapy.

2. Materials and Methods

2.1. Study Design and Participants

732 patients with glioma were selected from January 2012 to January 2018 in our hospital. All patients were randomly divided into two groups: the effective group (n = 366) and the ineffective group (n = 366). There were 190 males and 176 females in the observation group, aged from 18 to 66 years, with an average age of 36.1 ± 5.5 years and a BMI of 21.96 ± 2.57. In the control group, there were 198 males and 168 females, aged from 20 to 64 years old, with an average age of 35.9 ± 6.1 years and a BMI of 22.19 ± 1.98. There was no difference in the condition, course, and general data of the selected patients (P > 0.05), which was comparable. All the subjects obtained informed consent. The Ethics Committee of Zhejiang Provincial People's Hospital approved this research plan. All participants underwent a complete medical history examination and a clinical examination.

2.2. Reagents and Instruments

The miRNA Extraction kit, miRNA cDNA First chain synthesis kit, and miRNA fluorescence quantitative PCR detection kit were purchased from Beijing Tiangen Co., Ltd. RIPA Lytic fluid, and BCA protein quantitative kit was purchased from Shanghai Biyuntian Co., Ltd. The first antibodies of Foxo3a, p21, p27, and Bim were purchased from Abcam Company. The fluorescence quantitative PCR instrument was purchased from ABI Company, and the chemical developer was Shanghai Tianneng Company.

2.3. Detection of Serum miR-296 and miR-517c Expression

For the detection of serum miR-296 and miR-517c levels to collect fasting venous blood samples of 5 mL before chemotherapy, the Trizol Kit (American Invitrogen Company, item number 15596-018) is used. Extraction of total RNA and quantitative reverse transcription to obtain cDNA is carried out using the miRNA fluorescence quantitative PCR detection kit to configure the PCR reaction system. According to the two-step reaction procedure, repeat 40 cycles at 95°C for 5 s and 60°C for 15 s, miR-296, miR-517c, and U6 were amplified, respectively. According to the cycle curve and cycle threshold, U6 was used as the housekeeper gene to calculate the expression of miR-296 and miR-517c.

2.4. Evaluation of Curative Effect

Chemotherapy was given after the operation. Evaluation standard of curative effect are as follows: (1) complete remission: the focus of the tumor disappeared completely, and the physical signs disappeared completely, and continuous observation for 4 weeks still has a good clinical effect; (2) partial remission: after continuous observation for 4 weeks, the volume of tumor lesions decreased by more than 50%; (3) stable: after continuous observation for 4 weeks, there was no significant change in the disease, and the tumor focus volume decreased by more than 25% or increased by less than 25%; and (5) progress: the volume of the original tumor was enlarged by more than 25% or new tumor lesions were found. The evaluation time of all patients is the same. If there is progress, the chemotherapy regimen will be changed in time, and the clinical benefit (complete remission, partial remission, or stability) will continue throughout the course of chemotherapy. Patients with complete remission and partial remission were included in the effective group, and the rest were included in the ineffective group.

2.5. Follow-Up

The deadline for follow-up was May 2021. The methods of follow-up include outpatient reexamination, telephone or SMS follow-up, and so on. The total survival time was the time from definite diagnosis to death or the last follow-up, and the progression-free survival time was the time from definite diagnosis to imaging examination confirmed recurrence or last follow-up.

2.6. Statistical Method

The data are in accordance with the normal distribution. The measurement data are expressed by mean ± standard deviation ( ± S), the frequency is expressed by percentage, the counting data are compared by the chi-square test, the measurement data are compared by the t-test, and the multivariate analysis was carried out by stepwise regression unconditional logistic regression analysis. All statistical tests were bilateral tests, P < 0.05 was regarded as statistically significant, and SPSS25.0 software was used for statistical analysis.

3. Result

3.1. Comparison of miR-296 Expression in Gliomas with Different Pathological Features

The expression level of miR-296 in glioma was significantly correlated with tumor pathological grade and depth of invasion (P < 0.05). It had nothing to do with the sex, age, and degree of differentiation of the patients, and the difference was not statistically significant (P > 0.05) (Table 1).

Table 1

Expression comparison of miR-296 in gliomas with different pathological features.

Clinicopathological features	n	miR-296 expression quantity	t	P
Gender
Male	388	3.62 ± 0.89	0.601	0.548
Female	344	3.58 ± 0.91
Age (years)
<50	386	2.91 ± 0.73	1.246	0.213
≥50	346	2.84 ± 0.79
Degree of differentiation
Low	331	2.44 ± 0.16	1.109	0.268
High	401	2.48 ± 0.64
Pathological grading
I–II	376	2.92 ± 0.25	12.121	<0.001
III–IV	356	2.74 ± 0.13
Infiltration depth
≤1/2	421	1.81 ± 0.67	5.920	<0.001
>1/2	311	2.05 ± 0.29

3.2. Comparison of miR-517c Expression in Gliomas with Different Pathological Features

The expression level of miR-296 in brain glioma was significantly correlated with tumor pathological grade (P < 0.05). However, it had nothing to do with the sex, age, degree of differentiation, and depth of infiltration of the patients, and the difference was not statistically significant (P > 0.05), as shown in Table 2.

Table 2

Expression comparison of miR-517c in gliomas with different pathological features.

Clinicopathological features	n	miR-517c expression quantity	t	P
Gender
Male	388	2.20 ± 0.43	0.635	0.526
Female	344	2.18 ± 0.42
Age (years)
<50	386	2.37 ± 0.35	2.135	0.033
≥50	346	2.31 ± 0.41
Degree of differentiation
Low	331	2.83 ± 0.30	2.528	0.012
High	401	2.69 ± 0.97
Pathological grading
I–II	376	1.69 ± 0.40	28.054	<0.001
III–IV	356	2.53 ± 0.41
Infiltration depth
≤1/2	421	2.39 ± 0.11	1.150	0.251
>1/2	311	2.42 ± 0.52

3.3. Curative Effect and Follow-Up

The total effective rate of 1464 patients was 36.75% (538/1464). Among them, the complete remission rate was 18.72% (274/1464). The partial remission rate was 19.26% (282/1464), the stability rate was 38.25% (560/1464), and the rate of progress was 17.42% (255/1464). The overall survival rate of the patient was 38.79% (568/1464), the progression-free survival rate was 25.54% (374/1464), and the progressive survival rate was 6.42% (94/1464), as shown in Table 3.

Table 3

Curative effect and follow-up.

Total efficiency	Complete remission rate	Partial remission rate	Stable rate	Progress rate	Overall survival rate	Progression-free survival rate	Progressive survival rate
269/732 (36.75%)	137/732 (18.72%)	141/732 (19.26%)	280/732 (38.25%)	128/732 (17.49%)	284/732 (38.79%)	187/732 (25.54%)	47/732 (6.42%)

3.4. Comparison of Serum miR-296 and miR-517c Levels before Chemotherapy between the Effective Group and the Ineffective Group

The relative expressions of serum miR-296 and miR-517c in the effective group were higher than those in the ineffective group, and the difference was statistically significant (P < 0.01). See Table 4 for details.

Table 4

Comparison of serum miR-296 and miR-517c levels before chemotherapy between the effective group and the ineffective group ( ± S).

Groups	n	Serum miR-296	Serum miR-517c
Effective group	366	4.62 ± 0.53	4.44 ± 2.15
Ineffective group	366	2.94 ± 1.88	2.66 ± 1.39
t		16.455	13.301
P		<0.001	<0.001

3.5. Comparison of Clinical Data between the Effective Group and the Ineffective Group

There were significant differences in age, history of epilepsy, WHO grade, and mode of operation between the effective group and the ineffective group (P < 0.05). See Table 5 for details.

Table 5

Comparison of the clinical data of the effective group and the ineffective group.

	Effective group	Ineffective group	χ 2	P
Gender
Male (n = 388)	200	188	0.790	0.374
Female (n = 344)	166	178
Age (years)
<50 (n = 386)	228	158	26.856	<0.001
≥50 (n = 346)	138	208
Have a history of epilepsy
Yes	206	155	14.216	<0.001
No	160	211
WHO classification
I	125	107	12.748	0.005
II	108	97
III	65	50
IV	68	112
Complication
Yes	130	137	0.289	0.591
No	236	229
Mode of operation
Full cut	122	93	9.527	0.009
Subtotal cutting	150	144
Part of the cut	94	129
Site of lesion involvement
1 cerebral leaf	179	181	0.022	0.882
2 and above brain leaves	187	185

3.6. Logistic Regression Analysis of Factors Affecting Chemosensitivity

Logistic regression analysis showed that tumor size, WHO grade, and serum miR-296 and miR-517c levels were all factors affecting chemosensitivity (P < 0.05). See Table 6 for details.

Table 6

Logistic regression analysis of factors affecting chemosensitivity.

Variables	B	Standard error	Wald.	Significance	OR
Tumor size	1.302	0.607	6.919	0.031	(0.068–0.732)
WHO grading	0.995	0.344	7.862	0.002	(1.329–5.017)
Serum miR-296	1.221	0.387	10.518	<0.001	(1.777–7.558)
Serum miR-517c	1.090	0.365	9.736	<0.001	(1.638–6.220)

3.7. The Efficacy of Serum miR-296 and miR-517c in Predicting the Prognosis of Patients with Glioma

The best cutoff point for predicting the 1-year death of glioma patients by analyzing the relative expression of serum miR-296 and miR-517c is 6.33. The sensitivity, specificity, accuracy, and AUC of serum miR-296 prediction were 76.95%, 89.64%, 85.35%, and 0.891, respectively. The sensitivity, specificity, accuracy, and AUC of serum miR-517c prediction were 72.81%, 86.50%, 82.19%, and 0.739, respectively. See Figure 1 for details.

Figure 1

Serum miR-296 and miR-517c ROC curves to predict the prognosis of patients with glioma.

4. Discussion

Glioma is one of the most fatal malignant tumors of the central system. Although it has been studied for many years, the overall prognosis of malignant glioma is not optimistic [17]. At present, it is believed that there is a kind of brain tumor stem cell in brain tumors. This theory holds that tumor stem cells are the determinants of tumor occurrence, development, invasion, metastasis, and drug resistance [18]. Tumor stem cells are more tolerant to radiotherapy and chemotherapy than ordinary tumor cells [19, 20]. The survival of tumor stem cells is the root cause of malignant tumor recurrence [21]. It shows diffuse growth and that the invasiveness is strong. It also shows that the 5-year survival rate of the patients is less than 5%, and that the prognosis is poor [22]. Because of the fine structure of brain tissue, the unclear boundary between tumor and normal brain tissue, and its characteristics of easy local dissemination, strong invasiveness and easy recurrence, the therapeutic effect is often poor, and most of the patients have a poor prognosis [23]. In recent years, it has become one of the five most difficult to cure tumors in the field of tumor treatment [24].

Recent studies have shown that miRNA plays a certain regulatory role in the self-renewal and differentiation of human glioma stem cells, and some miRNAs affect the expression of important genes in human glioma stem cells through negative regulation [25]. miRNA is a kind of noncoding small RNA with a size of 21–23 nt, which exists widely in eukaryotes [26]. Recent studies have shown that miRNA is closely related to the occurrence and development of human gliomas [27]. Some miRNA are specifically expressed in different tissue types of gliomas. Apoptosis or proliferation of human glioma cells can be induced or inhibited by upregulating or downregulating the corresponding miRNA. [28]. The study of miRNAs in gliomas began with an article published by Kong et al. in BBRC in 2005. They screened the expression levels of 245 miRNAs in gliomas by miRNA chip technology, and found that some miRNAs were abnormally expressed in gliomas (compared with normal brain tissues), such as increased expression of miR-21 and miR-123, decreased expression of miR-181, miR-128, and so on [29]. In recent years, the roles of several miRNAs in glioma have been found one after another. Aiqin Qi et al. confirmed that miR-21 can inhibit the apoptosis of glioma cells. In addition, miR-146b can inhibit the migration and infiltration of glioma cells by acting on MMPs [30], and high expression of miR-221 in highly malignant gliomas. miR-296a and miR-296b play a role similar to tumor suppressors in gliomas. MiR-296b is lowly expressed in highly malignant gliomas, which may affect genes related to malignant degree; miR-296 and miR-517cA can inhibit the proliferation of glioma cells, and the expression of miR-7 in glioma is decreased, which inhibits the expression of epidermal growth factor receptor and the Akt pathway [31]. These studies have revealed that miRNA is closely related to glioma and may play a positive role in molecular therapy and drug resistance of glioma. Studies have found that several miRNAs, including miR-296 and miR-517c, are abnormally expressed in human gliomas, among which miR-296 is least expressed in glioma cells. It is also found that miR-517c significantly induces apoptosis and inhibits cell growth in glioma cells.

The results of this study confirmed that the relative expressions of serum miR-296 and miR-517c were factors affecting chemosensitivity, suggesting that serum miR-296 and miR-517c were closely related to the chemosensitivity of gliomas [32]. High levels of serum miR-296 and miR-517c could enhance chemosensitivity. The possible mechanism may be to regulate the expression of damage repair factors in cancer cells and enhance chemosensitivity [33]. Invasion and metastasis of cancer cells are the main causes of disease recurrence and poor prognosis. Zhang et al. [31] found that the expression levels of serum miR-296 and miR-517c in gastric cancer tissues were lower than those in paracancerous tissues and that it is closely related to lymph node metastasis and staging. The overall survival time of patients with high expression of serum miR-296 and miR-517c was longer than that of patients with low expression, which was similar to the results of this study. Shi et al. [34] found that the low expression level of serum miR-296 and miR-517c was closely related to the differentiation, clinical stage, and metastasis of colorectal cancer by analyzing the expression levels of serum miR-296 and miR-517c in 50 patients with colorectal cancer, which can be used as a reference index to predict the prognosis of colorectal cancer, which is consistent with the results of Liu et al. [35]. The results of this study showed that the relative expressions of serum miR-296 and serum miR-517c were highly sensitive, specific, accurate, and AUC in predicting 1-year death in patients with glioma, suggesting that serum miR-296 and miR-517c were closely related to the prognosis of glioma. It is further suggested that the levels of serum miR-296 and miR-517c can be used as reliable indexes to predict the prognosis of chemotherapy. Previous studies have not explored whether serum miR-145-5p is closely related to chemosensitivity and prognosis of patients with glioma. A high level of miR-145-5p can enhance chemosensitivity and can be used as a reliable index to predict the prognosis of patients with chemotherapy. PAMAM can effectively coload as-miR-21 and 5-FU and more effectively inhibit the growth of U251 glioma cells in vitro and increase the chemosensitivity to 5-Fu. The level of miR-126 in high-risk LGG tissue is negatively correlated with the level of promoter methylation. The hypermethylation rate of the miR-126 gene promoter is an independent risk factor for the adverse outcome of simultaneous radiotherapy and chemotherapy in high risk LGG. The weakness of this study is that all the subjects in this study are admitted to our hospital, and the sample size is limited, which may lead to a certain bias in the results. In the future, it is necessary to further expand the sample size and carry out multicenter research to confirm the accuracy of the research results.

To sum up, serum miR-296 and miR-517c are closely related to chemosensitivity and prognosis of patients with glioma. High levels of serum miR-296 and miR-517c can enhance the sensitivity to chemotherapy and serve as reliable indicators for predicting the prognosis of chemotherapy, which can provide a reference for improving the efficacy of postoperative chemotherapy, predicting the prognosis of glioma, and adjusting the treatment strategy in time to strive for more chances of survival for patients.