Abstract
Objective
Seizures are a common clinical presentation in patients with glioma and substantially impact patients' quality of life. Hyperhomocysteinemia is defined as abnormally high serum levels of homocysteine (Hcy) and is reportedly linked to susceptibility to various nervous system diseases. However, it remains unclear whether and how hyperhomocysteinemia and its associated genetic polymorphisms promote seizures in glioma patients.
Methods
We retrospectively reviewed all medical data from 127 patients with malignant gliomas, who underwent initial tumor resection by our team between July 2019 and June 2021 and had preoperative measurements of serum Hcy levels. According to whether they had at least one seizure before surgery, they were divided into the seizure and nonseizure groups. We also detected polymorphisms in the methylenetetrahydrofolate reductase (MTHFR) gene and measured intratumoral Hcy levels in these patients.
Results
Hyperhomocysteinemia was a susceptibility factor for preoperative seizures in glioma patients according to both univariate analyses (P < 0.001) and multivariate logistic regression analyses (OR 1.239, 95% CI 1.062–1.445, P = 0.007). Patients with the MTHFR C677T variant exhibited elevated serum Hcy levels (P = 0.027) and an increased prevalence of preoperative seizures (P = 0.019). Intratumoral Hcy levels were positively correlated with serum Hcy levels (R = 0.231, P = 0.046) and were elevated in patients with hyperhomocysteinemia (P = 0.031), the MTHFR C677T variant (P = 0.002) and preoperative seizures (P = 0.003). High intratumoral Hcy levels, rather than hyperhomocysteinemia or the MTHFR C677T variant, emerged as an independent risk factor for preoperative seizures (OR 1.303, 95% CI 1.015–1.673, P = 0.038). Furthermore, the effects of hyperhomocysteinemia on epileptic susceptibility were reduced to nonsignificance when intratumoral Hcy was controlled to the same level between groups.
Significance
Glioma patients with hyperhomocysteinemia and the MTHFR C677T variant were susceptible to preoperative seizures, suggesting their potential as biomarkers for the management of seizures in glioma patients. The elevation of intratumoral Hcy is a possible mechanism underlying this susceptibility.
Keywords: glioma, hyperhomocysteinemia, intratumoral homocysteine, MTHFR, preoperative seizure
Key points.
Hyperhomocysteinemia is a susceptibility factor for preoperative seizures in glioma patients.
The MTHFR C677T variant is a genetic susceptibility factor for seizure in glioma patients.
Increased intratumoral Hcy levels could be the mechanism underlying the susceptibility effect of hyperhomocysteinemia and the MTHFR C677T variant.
1. INTRODUCTION
Seizures are a common clinical presentation in patients with glioma 1 , 2 that substantially impacts patients' quality of life. 3 The identification of risk factors for seizure susceptibility will facilitate optimal clinical decision‐making in the treatment of seizures as well as glioma. 4 To date, several variables, including younger age, 1 lower‐grade and smaller tumors, 1 , 4 oligodendroglioma pathology, 1 , 4 , 5 temporal lobe tumors, 6 and mutations in isocitrate dehydrogenases (IDHs), 7 have been reported as susceptibility factors correlated with the incidence of seizures in glioma patients and thus applied as biomarkers for its diagnosis and perioperative management. 4 , 5
Homocysteine (Hcy) is a sulfur‐containing amino acid generated as a byproduct during the metabolism of methionine (Met) to cysteine (Cys), 8 in which methylenetetrahydrofolate reductase (MTHFR) is a key enzyme with the role of supplying 5‐methyltetrahydrofolate for the remethylation of Hcy to methionine. 9 , 10 A deficiency of this enzyme leads to an abnormally elevated Hcy level in serum, usually defined as hyperhomocysteinemia when it exceeds 15 μmol/L. 11 , 12 Several polymorphisms in the MTHFR gene have been associated with enzyme deficiency, 13 among which C677T (rs1801133) and A1298C (rs1801131), are the two most commonly reported polymorphisms. 9 The heterozygous (CT) and homozygous (TT) genotypes of the C677T variant lead to enzyme activity declines of 34% and 75%, respectively, while the homozygous A1298C variant leads to an enzyme activity level of 61% of that of the wild‐type enzyme. 14 The serum Hcy levels of MTHFR gene knockout (MTHFR−/−) mice were tenfold higher than those of wild‐type mice. 15
Recent studies have closely associated hyperhomocysteinemia and its related MTHFR gene polymorphisms with susceptibility to structural or unknown epilepsies. 16 , 17 , 18 , 19 In terms of possible mechanisms, several studies have proposed that high Hcy concentrations could stimulate glutamate receptors, the major class of receptors in neuroexcitatory pathways in the brain, and thus contribute to epilepsy. 18 , 20 This mechanism has been confirmed by animal model studies. 21 , 22 Although MTHFR gene polymorphisms and hyperhomocysteinemia have been reported to be involved in pathogenesis or impact clinical outcomes of several malignancies, their roles in glioma remain controversial. 23 , 24 , 25 No studies on their effects on seizures in glioma patients have been reported to our knowledge.
Compared to other seizure types, glioma‐related seizures have unique pathogenetic mechanisms, for example, tumor‐induced changes in the glutamatergic and c‐aminobutyric acid (GABA)ergic systems or tumor‐released oncometabolite D‐2‐hydroxyglutarate also involved in epileptogenicity. 26 , 27 Therefore, further confirmation on whether hyperhomocysteinemia and associated genetic variants would play a similar role in seizures in glioma patients as in those in other epilepsy types is warranted. More importantly, clinical evidence of the mechanism linking them to seizure susceptibility is still lacking.
Herein, we retrospectively examined the effects of hyperhomocysteinemia and related MTHFR gene polymorphisms on preoperative seizure prevalence in a group of glioma patients who underwent surgery in our center. We discovered that hyperhomocysteinemia and the MTHFR C677T variant increased susceptibility to seizures in glioma patients. In addition, we measured intratumoral Hcy concentrations and identified elevation of intratumoral Hcy as a possible mechanism underlying the seizure susceptibility of patients with hyperhomocysteinemia and the MTHFR C677T variant.
2. MATERIALS AND METHODS
2.1. Patients
The medical records of patients whose initial resection was performed by our team at Beijing Tiantan Hospital between July 2019 and June 2021 were reviewed. Patients were included in this study according to the following criteria: (1) available preoperative serum Hcy level measurement; (2) pathological diagnosis of WHO grade 2–4 glioma; (3) complete tumor genetic information for the 2021 WHO classification of central nervous system (CNS) tumors. Patients taking anti‐seizure medications (ASMs) other than levetiracetam, such as carbamazepine and valproic acid, were excluded from the study, since these drugs can disturb the metabolism of Hcy, 28 while levetiracetam has no such effect. 29 This study was supported by the Neurosurgical Clinical Information and Biobanking Project of Beijing Tiantan Hospital (Brain Tumor Section) and was approved by the ethics committee of Beijing Tiantan Hospital (KY2014‐021‐02). Informed consent was obtained from all subjects involved in the study.
2.2. Collection of seizure‐related variables
All possible seizure‐related variables were collected from patients' medical records. These variables included age, sex, history of preoperative seizures, serum Hcy concentration, focal neurological deficit (motor or sensory impairment, various types of aphasia, visual field defect or vision loss), cognitive symptoms (memory loss, personality/behavior changes, cognitive impairment, mental disorder), presence of headache, comorbidity (coronary heart disease, hypertension, or diabetes requiring medication) and Karnofsky Performance Status (KPS) score (100–0). The seizure history was further reviewed by an experienced neurologist based on symptomatic analysis according to the ILAE 2017 Classification of Seizure Types. 30 The serum Hcy levels were measured by Hcy test kits (Changli Biotech Co., Ltd, Wuhan, China) using a colorimetric enzymatic assay. Hyperhomocysteinemia was defined as a serum Hcy level above 15 μmol/L. 12
Pathological diagnoses were also reviewed and revised by a senior neuropathologist according to the 2021 WHO Classification of Tumors of the Central Nervous System. 31 All patients underwent MRI scans within 2 weeks before surgery and the following MRI parameters were collected: tumor volume (calculated by multiplying the longest tumor diameters in three dimensions divided by 2), side, location, with/without contrast enhancement, and peritumoral edema.
2.3. Intratumoral Hcy‐level measurement
We used the corresponding tumor tissue samples of the included patients for analysis of intratumoral Hcy levels. A total of 50 mg tissue sample from each tumor was lysed with 1.2 mL lysis buffer (80% methanol and 20% water) on ice in an OMNI Bead Ruptor (OMNI International Inc., USA). The lysate was centrifuged at 14 000 × g at 4°C for 15 min, and the supernatant was collected and evaporated in a vacuum. The dried samples were resuspended in 200 μL of 2% ACN and filtered with a 10 kDa filter membrane (Pall, USA).
LC–MS/MS was performed using a Waters H‐Class HPLC system and a 6500 Q‐TRAP mass spectrometer (AB Sciex, USA). LC was carried out on a Waters HSS T3 column (1.8 μm 3.0 mm × 100 mm) at 50°C. Mobile phase A consisted of 0.1% formic acid and 99.9% H2O, and mobile phase B consisted of 0.1% formic acid and 99.9% ACN. A linear gradient elution was applied at a flow rate of 350 μL/min.
Selected reaction monitoring (SRM) was performed in positive mode, with the following main parameters: spray voltage, 4500 V; temperature, 450°C; cone, 100; collision energy, 18. For Hcy detection, we selected m/z 136 in Q1 and m/z 90 in Q3.
The standard curve (range 0.25 μM ~ 250 μM) was created using an external standard, and the concentration of each sample was calculated using the external standard curve. To estimate the data quality, the mixed sample was used as a quality control before and after analysis for every 10 samples and for all samples. The MS coefficient of variation was calculated for the estimation.
2.4. MTHFR genotyping
We used the corresponding blood samples of the included patients for the analysis of C677T and A1298C MTHFR polymorphisms. Genotyping of blood samples was performed at Sangon Biotech Co., Ltd. (Shanghai, China) utilizing the Sanger sequencing method. Briefly, genomic DNA was extracted from blood cells using an Ezup column blood genomic DNA extraction kit (Sangon Biotech, China). After DNA quality assessment, polymerase chain reaction (PCR) was performed to amplify target DNA fragments. Primer sequences and PCR conditions are presented in Tables S1 and S2. The PCR products were further verified by electrophoresis and then retrieved using a SanPrep column DNA gel recovery kit (Sangon Biotech, China). Finally, the sequence of the PCR products was obtained with an ABI3730XL DNA Analyzer (Applied Biosystems, USA) and analyzed using Chromas (version 2.0) software.
2.5. Statistical analysis
A descriptive analysis was used to characterize the studied population. Receiver operating characteristic (ROC) curves were generated to identify the optimal cutoff values for continuous variables (age and tumor volume) to convert them to binary variables. Univariate analyses were performed using the chi‐squared (χ2) test or Fisher's exact test to compare binary variables. Continuous data were compared using Wilcoxon nonparametric tests. Variables with a significance level less than 0.10 were included in the subsequent multivariate logistic regression to identify risk factors for seizures. A two‐tailed P value < 0.05 was considered significant. All statistical analyses were performed using R (version 4.1.3) statistical software. Our study implemented the STROBE reporting guidelines. All datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.
3. RESULTS
3.1. Baseline characteristics of patients included in the study
We retrospectively reviewed the cases of all patients who were pathologically diagnosed with malignant gliomas (WHO grades 2–4) and whose initial resection was performed by our team between July 2019 and June 2021. A total of 127 patients with preoperative serum Hcy measurements were included in this study (Figure 1). Among them, 62 (48.8%) patients had at least one seizure before the operation, while 65 (51.2%) patients did not have preoperative seizures. The median serum concentration of Hcy was 13.65 μmol/L, ranging from 6.90 to 88.30 μmol/L; 51 (40.2%) patients had a serum Hcy level above 15 μmol/L and were diagnosed with hyperhomocysteinemia (Table 1).
FIGURE 1.
Flowchart of patient inclusion.
TABLE 1.
Risk factors for preoperative seizures in glioma patients.
| Variables | Univariate analysis | Multivariate analysis | ||||
|---|---|---|---|---|---|---|
| Preoperative seizure | P value | Odds ratio | 95% confidence interval | P value | ||
| Yes (%) | No (%) | |||||
| n | 62 (48.8) | 65 (51.2) | – | – | – | – |
| Sex | ||||||
| Male | 40 (64.5) | 33 (50.8) | 0.166 | – | – | – |
| Female | 22 (35.5) | 32 (49.2) | – | – | – | |
| Age | ||||||
| ≥50 y | 15 (24.2) | 45 (69.2) | <0.001*** | 0.818 | 0.696–0.962 | 0.015* |
| <50 y | 47 (75.8) | 20 (30.8) | – | – | – | |
| Serum Hcy concentration | ||||||
| ≥15 μmol/L | 35 (56.5) | 16 (24.6) | <0.001*** | 1.239 | 1.062–1.445 | 0.007** |
| <15 μmol/L | 27 (43.5) | 49 (75.4) | – | – | – | |
| Neurological symptoms | ||||||
| Yes | 9 (14.5) | 39 (60.0) | <0.001*** | 0.887 | 0.735–1.070 | 0.207 |
| No | 53 (85.5) | 26 (40.0) | – | – | – | |
| Cognitive symptoms | ||||||
| Yes | 1 (1.6) | 14 (21.5) | <0.001*** | 0.800 | 0.636–1.006 | 0.056 |
| No | 61 (98.4) | 51 (78.5) | – | – | – | |
| Headache | ||||||
| Yes | 25 (40.3) | 35 (53.8) | 0.178 | – | – | – |
| No | 37 (59.7) | 30 (46.2) | – | – | – | |
| Comorbidity | ||||||
| Coronary heart disease | 0 (0) | 2 (3.1) | 0.496 | – | – | – |
| Hypertension | 6 (9.7) | 18 (27.7) | 0.018* | 0.884 | 0.733–1.067 | 0.197 |
| Diabetes | 3 (4.8) | 7 (10.8) | 0.325 | – | – | – |
| KPS | ||||||
| >70 | 57 (91.9) | 52 (80.0) | 0.094 | 0.910 | 0.738–1.121 | 0.373 |
| ≤70 | 5 (8.1) | 13 (20.0) | – | – | – | |
| Tumor volume | ||||||
| ≥40.13 cm3 | 26 (41.9) | 33 (50.8) | 0.412 | – | – | – |
| <40.13 cm3 | 36 (58.1) | 62 (49.2) | – | – | – | |
| Location | ||||||
| Frontal lobe | 30 (48.4) | 31 (47.7) | Ref = 1 | – | – | – |
| Parietal lobe | 1 (1.6) | 17 (26.2) | <0.001*** | 0.873 | 0.703–1.084 | 0.215 |
| Temporal lobe | 26 (39.4) | 7 (10.3) | 0.017* | 1.263 | 1.067–1.494 | 0.007** |
| Occipital lobe | 2 (3.2) | 3 (4.6) | 1.000 | – | – | – |
| Insula lobe | 5 (8.1) | 7 (10.8) | 1.000 | – | – | – |
| Side | ||||||
| Left | 34 (54.8) | 25 (38.5) | 0.095 | 0.904 | 0.788–1.037 | 0.149 |
| Right | 28 (45.2) | 40 (61.5) | – | – | – | |
| Contrast enhancement | ||||||
| Yes | 37 (59.7) | 55 (84.6) | 0.003** | 0.953 | 0.802–1.131 | 0.578 |
| No | 25 (40.3) | 10 (15.4) | – | – | – | |
| Peritumoral edema | ||||||
| Yes | 35 (56.5) | 49 (75.4) | 0.039* | 0.990 | 0.850–1.152 | 0.893 |
| No | 27 (43.5) | 16 (24.6) | – | – | – | |
| Pathology | ||||||
| Astrocytoma IDH‐M WHO Grade 2–3 | 19 (30.6) | 9 (13.8) | Ref = 1 | – | – | – |
| Oligodendroglioma | 22 (35.5) | 9 (13.8) | 1.000 | – | – | – |
| Astrocytoma IDH‐M WHO Grade 4 | 10 (16.1) | 4 (6.2) | 1.000 | – | – | – |
| Glioblastoma | 11 (17.7) | 43 (66.2) | <0.001*** | – | – | – |
| IDH1 mutation | ||||||
| Mutate | 51 (82.3) | 21 (32.3) | <0.001*** | 1.245 | 1.020–1.520 | 0.032* |
| Wild type | 11 (17.7) | 44 (67.7) | ||||
| Chromosomes 1p and 19q co‐deletion | ||||||
| Co‐deleted | 23 (37.1) | 10 (15.4) | 0.010** | 1.046 | 0.877–1.247 | 0.615 |
| Non‐co‐deleted | 39 (62.9) | 55 (84.6) | ||||
Abbreviation: KPS, Karnofsky performance score.
0.01 ≤ P < 0.05
0.001 ≤ P < 0.01
P < 0.001.
Among these included cases, a total of 75 and 82 patients with available tumor tissues and blood samples, respectively, were selected for intratumoral Hcy measurement and MTHFR genotyping. The baseline characteristics of these selected patients were comparable to those of all the patients included in the study (Table S3).
3.2. The association of hyperhomocysteinemia with preoperative seizure prevalence
We utilized univariate analyses to screen for variables associated with the prevalence of preoperative seizures. We identified 11 variables that were significantly associated with seizure prevalence (Figure S2 and Table 1). Among these variables, age, presence of neurological/cognitive symptoms, hypertension, tumor location, glioblastoma pathology, tumors with IDH‐1 mutations, and tumors with co‐deletion of chromosomes 1p and 19q have all been previously reported as seizure‐related factors in glioma. 6 , 16 , 32 Contrast enhancement and peritumoral edema are classic radiological features of glioblastoma 33 ; thus, this feature was also associated with decreased prevalence of preoperative seizures. We observed that 56.5% of patients with preoperative seizures also had hyperhomocysteinemia, which was much higher than the 24.6% proportion of hyperhomocysteinemia in nonseizure patients (P < 0.001), suggesting a close association between this anomaly and preoperative seizures in glioma patients.
To explore whether hyperhomocysteinemia is also an independent clinical variable correlated with preoperative seizures, we fitted a multivariate logistic regression model including all the above variables significantly related to seizure, except the variable of “glioblastoma pathology” due to its complete overlap with the variable of “IDH‐wildtype glioma”, as well as another two variables (the KPS score and tumor side) that showed marginally significant associations with preoperative seizures (P < 0.1). As shown in Figure 2, hyperhomocysteinemia was still a statistically significant factor correlated with preoperative seizures (OR 1.239, 95% CI 1.062–1.445, P = 0.007), as were age ≥ 50 years (OR 0.818, 95% CI 0.696–0.962, P = 0.015), temporal lobe tumors (OR 1.263, 95% CI 1.067–1.494, P = 0.007) and tumors with IDH1 mutations (OR 1.245, 95% CI 1.020–1.520, P = 0.032) (Figure 2 and Table 1). Therefore, hyperhomocysteinemia was revealed as an independent clinical variable associated with preoperative seizure prevalence in glioma patients.
FIGURE 2.
Independent factors associated with preoperative seizures in glioma patients. Multivariate logistic regression analysis revealed four variables as independent factors associated with preoperative seizure susceptibility. *0.01 < P < 0.05. **0.001 < P < 0.01.
3.3. The association of the MTHFR C677T polymorphism with preoperative seizure prevalence
Recent studies have identified a C677T polymorphism in the MTHFR gene as a risk factor for epilepsy in patients with temporal mesial sclerosis or traumatic brain injury 18 , 19 and the authors speculated that the elevation of Hcy levels in serum would mediate this genetic susceptibility. 18 Herein, we examined whether the gene variants play a similar role in seizures in glioma patients by detecting two of the most common MTHFR polymorphisms (C677T and A1298C), both of which are associated with hyperhomocysteinemia, 9 via Sanger sequencing. We observed that the C677T variant, but not the A1298C variant, was a risk factor for preoperative seizure prevalence (Figure 3A,B). Moreover, the C677T variant was associated with increased serum Hcy levels (Figure 3C, P = 0.027), whereas no significant difference was observed in the analyses for the A1298C variant (Figure 3D). This finding indicates that the C677T variant increases both serum Hcy levels and the susceptibility to seizures in glioma patients, thus supporting the speculation of previous study. 18
FIGURE 3.
Association of MTHFR polymorphisms with serum Hcy levels and preoperative seizures. (A,B) Chi‐squared (χ2) tests were used to determine the difference in seizure proportions between the MTHFR C677T(A)/A1298C(B) variants and their corresponding wild‐type alleles. (C,D) Wilcoxon nonparametric tests were used to determine the differences in serum Hcy levels between the MTHFR C677T(C)/A1298C(D) variants and their corresponding wild‐type alleles. CC, wild‐type MTHFR C677T variant. TT/TC, homozygous (TT), or heterozygous (TC) genotypes of the MTHFR C677T variant. AA, wild‐type MTHFR A1298C variant. CC/AC, homozygous (CC), or heterozygous (AC) genotypes of the MTHFR A1298C variant.
3.4. Susceptibility to preoperative seizure depends on intratumoral Hcy levels
Previous animal studies have revealed that local delivery of Hcy and its sulfur‐containing derivatives (ie, homocysteic acid and homocysteine thiolactone) into the brain can induce seizures by stimulating NMDA receptors. 34 , 35 Therefore, we hypothesized that the elevation of intratumoral Hcy levels would be a direct trigger for preoperative seizures in glioma, and thus mediate the contribution of hyperhomocysteinemia and the MTHFR C677T variant to seizure susceptibility. To test this hypothesis, we utilized LC–MS/MS to quantitate Hcy concentrations in glioma tissues from patients with available tumor tissue samples. As shown in Figure 4A, the intratumoral and serum levels of Hcy were significantly correlated (R = 0.231, P = 0.046). Patients with hyperhomocysteinemia or harboring the MTHFR C677T variant showed significantly increased intratumoral levels of Hcy (P = 0.031, Figure 4B; P = 0.002, Figure 4D). Meanwhile, patients with preoperative seizures accordingly exhibited significantly higher intratumoral levels of Hcy than the nonseizure patients (Figure 4C) (P = 0.003). Therefore, elevated intratumoral Hcy levels are highly correlated with hyperhomocysteinemia, the MTHFR C677T variant, and preoperative seizure prevalence.
FIGURE 4.
Association of intratumoral Hcy levels with serum Hcy levels, preoperative seizures and MTHFR polymorphisms. (A) Pearson correlation tests were performed between serum and intratumoral Hcy levels. (B‐E) Wilcoxon nonparametric tests were used to determine the differences in intratumoral Hcy levels between (B) the with/without hyperhomocysteinemia groups, (C) the seizure/nonseizure groups and (D‐E) the MTHFR variants and their wild‐type alleles. CC, wild‐type MTHFR C677T variant. TT/TC, homozygous (TT), or heterozygous (TC) genotypes of the MTHFR C677T variant. AA, wild‐type MTHFR A1298C variant. CC/AC, homozygous (CC), or heterozygous (AC) genotypes of the MTHFR A1298C variant.
We next fitted another multivariate logistic regression analysis that comprised all the variables that previously showed independent association with preoperative seizures (Figure 2), as well as elevated intratumoral Hcy level and the MTHFR C677T variant. An ROC analysis was used to set an optimal cutoff value for defining the elevated intratumoral Hcy level (Figure S3). As a result, elevated intratumoral Hcy levels, but not hyperhomocysteinemia or the MTHFR C677T polymorphism, emerged as a factor independently correlated with preoperative seizure prevalence (OR 1.303, 95% CI 1.015–1.673, P = 0.038). This result indicates a more profound role of elevated intratumoral Hcy levels in affecting preoperative seizure prevalence compared to hyperhomocysteinemia and the MTHFR C677T variant.
Finally, we examined whether elevated intratumoral Hcy levels are the underlying variable that mediates the contribution of hyperhomocysteinemia to patients' susceptibility to preoperative seizures. We paired the patients from the normal Hcy (<15 μmol/L) and hyperhomocysteinemia (≥15 μmol/L) groups who had similar tissue Hcy levels (the between‐group difference in intratumoral Hcy levels was controlled to less than 20% of the level of the normal Hcy group), rendering the difference in intratumoral Hcy levels statistically insignificant between the two groups (Figure S3A). After the adjustment, the difference in seizure prevalence between the two groups became nonsignificant (Figure S3B) (P = 0.099), and this change was not due to the smaller sample sizes (Figure S3C), indicating that the contribution of hyperhomocysteinemia to seizure susceptibility relies on Hcy levels in tumor tissue. Concerning the well‐established link of the MTHFR C677T variant with hyperhomocysteinemia 9 , 14 , 15 , 29 and patients' susceptibility to seizures likely relying on intratumoral Hcy levels (Figure 5), this finding also suggests that the contribution of the MTHFR C677T variant to seizure susceptibility depends on elevated intratumoral Hcy levels.
FIGURE 5.
High intratumoral Hcy levels were independently associated with preoperative seizure prevalence. Multivariate logistic regression analysis revealed intratumoral Hcy level as an independent factor associated with preoperative seizure prevalence. *0.01 < P < 0.05; **0.001 < P < 0.01.
Taken together, these findings indicate that elevated intratumoral Hcy levels are a profound factor that is independently associated with preoperative seizure susceptibility and mediates the susceptibility of hyperhomocysteinemia and the MTHFR C677T variant to seizures.
4. DISCUSSION
Seizures are a major clinical presentation for glioma patients 1 , 4 , 5 that significantly impact patients' quality of life 3 and correlate with their survival. 36 Recent developments in nonenzyme‐inducing ASMs (eg, levetiracetam) have enabled effective control of most seizures in glioma patients while minimizing drug‐related toxicities. 37 However, side effects, including fatigue, drowsiness, headache, and behavioral disturbances, are still common with these new‐generation ASMs, especially in long‐term therapy. 38 Although prophylactic administration of ASMs in brain tumor patients without seizures is not recommended, it is still a debated issue in patients undergoing neurosurgery. 37 Therefore, the recognition of glioma patients who are susceptible to seizures is paramount, since it would facilitate the precise selection of patients for long‐term therapy as well as prophylactic administration of ASMs during the perioperative period.
Numerous clinical studies have established a close link between MTHFR gene polymorphisms and hyperhomocysteinemia. 11 Both have also been revealed as risk factors for a variety of structural/metabolic and unknown epilepsies. 16 , 17 , 18 , 19 However, their impact on seizures in glioma patients has not been elucidated. In this study, we explored the correlations of MTHFR gene polymorphisms and hyperhomocysteinemia with preoperative seizure prevalence in a cohort of glioma patients. Consistent with their roles in other epilepsies, 16 , 17 , 18 , 19 we also observed that hyperhomocysteinemia was independently correlated with an increased prevalence of preoperative seizures in glioma patients (Figure 2), and patients harboring the C677T polymorphism in the MTHFR gene showed increased susceptibility to preoperative seizures (Figure 3A). All these findings indicate that both factors are potential biomarkers for the early recognition of seizure susceptibility, suggesting that serum Hcy level measurement and/or MTHFR polymorphism detection could be recommended for better management of seizures in glioma patients.
Based on the direct proconvulsive potential of Hcy or its derivatives in animal models, 21 , 22 , 34 , 35 several researchers have speculated that hyperhomocysteinemia would mediate the susceptibility of patients with MTHFR gene polymorphisms to epilepsies. 18 , 20 In this study, we observed that the C677T variant increases both serum Hcy levels and the susceptibility to seizures in glioma patients (Figure 3), thus supporting the speculation. We further hypothesized that the mechanism of seizure susceptibility induced by increased serum Hcy levels is that serum Hcy enters the brain and exerts epileptogenic effects locally, as suggested by the findings of animal studies. 21 , 22 Therefore, we simultaneously examined the levels of Hcy in tumor tissue samples from this group of patients and observed a positive correlation between serum and intratumoral Hcy levels (Figure 4A) and elevated intratumoral Hcy levels in the hyperhomocysteinemia group (Figure 4B). We next confirmed that seizure susceptibility induced by hyperhomocysteinemia relied on increased intratumoral Hcy levels (Figure S4). In addition, multivariate logistic regression analysis identified intratumoral Hcy level elevation, but not hyperhomocysteinemia or the MTHFR C677T polymorphism, as an independent risk factor for preoperative seizure (Figure 5), further showing a more fundamental role of intratumoral Hcy level elevation in contributing to seizure susceptibility. Together, these findings suggest a causal chain of epileptogenicity by Hcy disorders, that is the MTHFR C677T variant leads to elevated levels of Hcy in serum as well as in tumor tissues, which contribute to seizure susceptibility in glioma patients.
Regarding the molecular mechanism of the proconvulsive effects of Hcy, the triggering of NMDAR‐mediated neurotoxicity is a widely accepted theory. In this theory, Hcy acts as a high affinity agonist of NMDARs, thus persistently igniting the GluN2A subunit‐dependent ERK/p38/Cas‐9/Cas‐3 pathway, causing neuroexcitatory toxicity and leading to epilepsy through intracellular calcium accumulation and mitochondrial membrane potential changes. 39 , 40 Conversely, Hcy‐induced neurotoxicity and related seizures can be effectively ameliorated by using NMDA receptor antagonists, 22 , 41 , 42 further supporting this view. Notably, all the above evidence was obtained from in vitro or in vivo animal studies. To our knowledge, our study is the first to provide clinical evidence that supports this theory.
Other than hyperhomocysteinemia, younger age (<50 years), temporal lobe tumors and tumors with IDH‐1 mutations were also independent risk factors for preoperative seizures in multivariate regression analysis. Many studies have observed an inverse correlation between age and seizures in glioma patients 6 , 32 , 36 , 43 due to older patients' predisposition to glioblastomas, which were associated with a lower seizure prevalence than lower‐grade gliomas. 44 Our study also revealed a significant association of older age with glioblastoma pathology (Figure S5). Previous studies have reported that seizures are particularly common in patients with lesions of the temporal lobe. 45 The temporal area has been proposed to feature some of the lowest thresholds for seizure induction in humans, 46 and the inherent epileptogenicity of structures in the mesial temporal lobe may contribute to seizure generation. 47 A few studies also reported seizure susceptibility in patients with IDH‐mutant gliomas. 7 The oncometabolite D‐2‐hydroxyglutarate, which is specifically increased in this type of tumor, 48 was proposed to be the culprit for inducing seizures by mimicking the activity of glutamate on the NMDA receptor, leading to neuroexcitatory toxicity. 27 , 49
There are some limitations of the study. First, given the limited sample size and retrospective design of this study, the current findings should be further confirmed in a larger and more carefully designed study. Second, accurate measurement of actual Hcy levels in tumor tissues may be limited by the high intratumoral heterogeneity exhibited by gliomas. 50 Third, nearly all seizure diagnoses were made based on symptomatic analysis and lacked objective evidence from electroencephalogram (EEG) or magnetoencephalography (MEG) to confirm whether epileptogenic focus localization is consistent with tumor localization, which limits the extension of our results to glioma‐related epilepsy. 2
However, on the basis of all our findings, we propose that elevated intratumoral Hcy levels could be a profound risk factor for preoperative seizures in glioma patients, and preoperative elevation of serum Hcy levels and the MTHFR C677T variant could be valuable indicators for managing perioperative seizures in glioma patients, thus requiring further investigation.
5. CONCLUSION
In conclusion, our study is the first to reveal increased susceptibility to seizures in glioma patients with hyperhomocysteinemia and its associated genetic susceptibility factor, the MTHFR C677T variant. The elevation of intratumoral Hcy is a possible mechanism underlying this susceptibility.
AUTHOR CONTRIBUTIONS
Conceptualization, writing—review and editing, and funding acquisition: Y.Z. and N.J. Methodology: X.C., Z.J., and J.L. Software: X.C. and Z.J. Formal analysis: X.C. and T.S. Investigation and resources: Z.G. and Y.W. Pathological diagnoses: J.W. Data curation: X.C., G.L., and Q.Z. Writing—original draft preparation: X.C. and J.L. Visualization: X.C. and Z.G. Supervision and project administration: Y.Z., J.L., G.L., and N.J. All authors have read and agreed to the published version of the manuscript.
FUNDING INFORMATION
This research was funded by the National Natural Science Foundation of China (81702451, 81930048), the Capital Characteristic Clinical Application Project (Z181100001718196), and a Capital Health Research and Development Special Grant (2022–2‐2047).
CONFLICT OF INTEREST STATEMENT
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
ETHICS STATEMENT
We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.
CONSENT TO PARTICIPATE
Informed consent was obtained from all subjects involved in the study.
Supporting information
Data S1
Chi X, Lu J, Guo Z, Wang J, Liu G, Jin Z, et al. Susceptibility to preoperative seizures in glioma patients with elevated homocysteine levels. Epilepsia Open. 2023;8:1350–1361. 10.1002/epi4.12797
Xiaohan Chi and Jingjing Lu contributed equally to this work.
Nan Ji and Yang Zhang shared senior authorship.
DATA AVAILABILITY STATEMENT
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.
REFERENCES
- 1. Goldstein ED, Feyissa AM. Brain tumor related‐epilepsy. Neurol Neurochir pol. 2018;52:436–447. [DOI] [PubMed] [Google Scholar]
- 2. Liang S, Fan X, Zhao M, Shan X, Li W, Ding P, et al. Clinical practice guidelines for the diagnosis and treatment of adult diffuse glioma‐related epilepsy. Cancer Med. 2019;8:4527–4535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Maschio M, Zarabla A, Maialetti A, Fabi A, Vidiri A, Villani V, et al. Quality of life, mood and seizure control in patients with brain tumor related epilepsy treated with lacosamide as add‐on therapy: a prospective explorative study with a historical control group. Epilepsy Behav. 2017;73:83–89. [DOI] [PubMed] [Google Scholar]
- 4. Chen DY, Chen CC, Crawford JR, Wang SG. Tumor‐related epilepsy: epidemiology, pathogenesis and management. J Neurooncol. 2018;139:13–21. [DOI] [PubMed] [Google Scholar]
- 5. You G, Sha Z, Jiang T. The pathogenesis of tumor‐related epilepsy and its implications for clinical treatment. Seizure. 2012;21:153–159. [DOI] [PubMed] [Google Scholar]
- 6. Bech KT, Seyedi JF, Schulz M, Poulsen FR, Pedersen CB. The risk of developing seizures before and after primary brain surgery of low‐ and high‐grade gliomas. Clin Neurol Neurosurg. 2018;169:185–191. [DOI] [PubMed] [Google Scholar]
- 7. Liang R, Fan Y, Wang X, Mao Q, Liu Y. The significance of IDH1 mutations in tumor‐associated seizure in 60 Chinese patients with low‐grade gliomas. ScientificWorldJournal. 2013;2013:403942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Kumar A, Palfrey HA, Pathak R, Kadowitz PJ, Gettys TW, Murthy SN. The metabolism and significance of homocysteine in nutrition and health. Nutr Metab (Lond). 2017;14:78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Wan L, Li Y, Zhang Z, Sun Z, He Y, Li R. Methylenetetrahydrofolate reductase and psychiatric diseases. Transl Psychiatry. 2018;8:242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Raghubeer S, Matsha TE. Methylenetetrahydrofolate (MTHFR), the one‐carbon cycle, and cardiovascular risks. Nutrients. 2021;13:4562. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Zaric BL, Obradovic M, Bajic V, Haidara MA, Jovanovic M, Isenovic ER. Homocysteine and Hyperhomocysteinaemia. Curr Med Chem. 2019;26:2948–2961. [DOI] [PubMed] [Google Scholar]
- 12. Maron BA, Loscalzo J. The treatment of hyperhomocysteinemia. Annu Rev Med. 2009;60:39–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Liew SC, Gupta ED. Methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism: epidemiology, metabolism and the associated diseases. Eur J Med Genet. 2015;58:1–10. [DOI] [PubMed] [Google Scholar]
- 14. Weisberg I, Tran P, Christensen B, Sibani S, Rozen R. A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab. 1998;64:169–172. [DOI] [PubMed] [Google Scholar]
- 15. Schwahn BC, Laryea MD, Chen Z, Melnyk S, Pogribny I, Garrow T, et al. Betaine rescue of an animal model with methylenetetrahydrofolate reductase deficiency. Biochem J. 2004;382:831–840. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Tao H, Gong Y, Yu Q, Zhou H, Liu Y. Elevated serum matrix metalloproteinase‐9, interleukin‐6, hypersensitive C‐reactive protein, and homocysteine levels in patients with epilepsy. J Interferon Cytokine Res. 2020;40:152–158. [DOI] [PubMed] [Google Scholar]
- 17. Ono H, Sakamoto A, Mizoguchi N, Sakura N. Methylenetetrahydrofolate reductase 677C > T mutation and epilepsy. J Inherit Metab Dis. 2000;23:525–526. [DOI] [PubMed] [Google Scholar]
- 18. Scher AI, Wu H, Tsao JW, Blom HJ, Feit P, Nevin RL, et al. MTHFR C677T genotype as a risk factor for epilepsy including post‐traumatic epilepsy in a representative military cohort. J Neurotrauma. 2011;28:1739–1745. [DOI] [PubMed] [Google Scholar]
- 19. Caccamo D, Condello S, Gorgone G, Crisafulli G, Belcastro V, Gennaro S, et al. Screening for C677T and A1298C MTHFR polymorphisms in patients with epilepsy and risk of hyperhomocysteinemia. Neuromolecular Med. 2004;6:117–126. [DOI] [PubMed] [Google Scholar]
- 20. Rai V, Kumar P. Methylenetetrahydrofolate reductase C677T polymorphism and susceptibility to epilepsy. Neurol Sci. 2018;39:2033–2041. [DOI] [PubMed] [Google Scholar]
- 21. Stanojlovic O, Rasic‐Markovic A, Hrncic D, Susic V, Macut D, Radosavljevic T, et al. Two types of seizures in homocysteine thiolactone‐treated adult rats, behavioral and electroencephalographic study. Cell Mol Neurobiol. 2009;29:329–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Rasic‐Markovic A, Hrncic D, Djuric D, Macut D, Loncar‐Stevanovic H, Stanojlovic O. The effect of N‐methyl‐D‐aspartate receptor antagonists on D,L‐homocysteine thiolactone induced seizures in adult rats. Acta Physiol Hung. 2011;98:17–26. [DOI] [PubMed] [Google Scholar]
- 23. Kumawat R, Gowda SH, Debnath E, Rashid S, Niwas R, Gupta Y, et al. Association of Single Nucleotide Polymorphisms (SNPs) in genes encoding for folate Metabolising enzymes with glioma and meningioma in Indian population Asian Pac. J Cancer Prev. 2018;19:3415–3425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Petrone I, Bernardo PS, Dos Santos EC, Abdelhay E. MTHFR C677T and A1298C polymorphisms in breast cancer, gliomas and gastric cancer: a review. Genes (Basel). 2021;12:587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Pandith AA, Qasim I, Zahoor W, Shah P, Bhat AR. ACE I/D sequence variants but not MTHFR C677T, is strongly linked to malignant glioma risk and its variant DD genotype may act as a promising predictive biomarker for overall survival of glioma patients. Gene. 2018;639:62–68. [DOI] [PubMed] [Google Scholar]
- 26. Pallud J, Capelle L, Huberfeld G. Tumoral epileptogenicity: how does it happen? Epilepsia. 2013;54(Suppl 9):30–34. [DOI] [PubMed] [Google Scholar]
- 27. Chen H, Judkins J, Thomas C, Wu M, Khoury L, Benjamin CG, et al. Mutant IDH1 and seizures in patients with glioma. Neurology. 2017;88:1805–1813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Bochynska A, Lipczynska‐Lojkowska W, Gugala‐Iwaniuk M, Lechowicz W, Restel M, Graban A, et al. The effect of vitamin B supplementation on homocysteine metabolism and clinical state of patients with chronic epilepsy treated with carbamazepine and valproic acid. Seizure. 2012;21:276–281. [DOI] [PubMed] [Google Scholar]
- 29. Zhu S, Ni G, Sui L, Zhao Y, Zhang X, Dai Q, et al. Genetic polymorphisms in enzymes involved in one‐carbon metabolism and anti‐epileptic drug monotherapy on homocysteine metabolism in patients with epilepsy. Front Neurol. 2021;12:683275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Fisher RS, Cross JH, French JA, Higurashi N, Hirsch E, Jansen FE, et al. Operational classification of seizure types by the international league against epilepsy: position paper of the ILAE commission for classification and terminology. Epilepsia. 2017;58:522–530. [DOI] [PubMed] [Google Scholar]
- 31. Louis DN, Perry A, Wesseling P, Brat DJ, Cree IA, Figarella‐Branger D, et al. The 2021 WHO classification of tumors of the central nervous system: a summary. Neuro Oncol. 2021;23:1231–1251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Yu Z, Zhang N, Hameed NUF, Qiu T, Zhuang D, Lu J, et al. The analysis of risk factors and survival outcome for Chinese patients with epilepsy with high‐grade glioma. World Neurosurg. 2019;125:e947–e957. [DOI] [PubMed] [Google Scholar]
- 33. Maugeri R, Schiera G, Di Liegro CM, Fricano A, Iacopino DG, Di Liegro I. Aquaporins and brain tumors. Int J Mol Sci. 2016;17:1029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Folbergrova J, Haugvicova R, Mares P. Behavioral and metabolic changes in immature rats during seizures induced by homocysteic acid: the protective effect of NMDA and non‐NMDA receptor antagonists. Exp Neurol. 2000;161:336–345. [DOI] [PubMed] [Google Scholar]
- 35. Flott‐Rahmel B, Schurmann M, Schluff P, Fingerhut R, Musshoff U, Fowler B, et al. Homocysteic and homocysteine sulphinic acid exhibit excitotoxicity in organotypic cultures from rat brain. Eur J Pediatr. 1998;157(Suppl 2):S112–S117. [DOI] [PubMed] [Google Scholar]
- 36. Pallud J, Audureau E, Blonski M, Sanai N, Bauchet L, Fontaine D, et al. Epileptic seizures in diffuse low‐grade gliomas in adults. Brain. 2014;137:449–462. [DOI] [PubMed] [Google Scholar]
- 37. Walbert T, Harrison RA, Schiff D, Avila EK, Chen M, Kandula P, et al. SNO and EANO practice guideline update: anticonvulsant prophylaxis in patients with newly diagnosed brain tumors. Neuro Oncol. 2021;23:1835–1844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Howard P, Remi J, Remi C, Charlesworth S, Whalley H, Bhatia R, et al. Levetiracetam. J Pain Symptom Manage. 2018;56:645–649. [DOI] [PubMed] [Google Scholar]
- 39. Ivanova MA, Kokorina AD, Timofeeva PD, Karelina TV, Abushik PA, Stepanenko JD, et al. Calcium export from neurons and multi‐kinase signaling cascades contribute to ouabain neuroprotection in hyperhomocysteinemia biomolecules. 2020;10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Sitnikova LS, Ivanova MA, Stepanenko YD, Karelina TV, Giniatullin R, Sibarov DA, et al. Collapse of neuronal energy balance As a basis of L‐homocysteine neurotoxicity. Biochemistry (Moscow). 2018;12:360–368. [Google Scholar]
- 41. Poddar R, Paul S. Homocysteine‐NMDA receptor‐mediated activation of extracellular signal‐regulated kinase leads to neuronal cell death. J Neurochem. 2009;110:1095–1106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Poddar R, Paul S. Novel crosstalk between ERK MAPK and p38 MAPK leads to homocysteine‐NMDA receptor‐mediated neuronal cell death. J Neurochem. 2013;124:558–570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. You G, Sha ZY, Yan W, Zhang W, Wang YZ, Li SW, et al. Seizure characteristics and outcomes in 508 Chinese adult patients undergoing primary resection of low‐grade gliomas: a clinicopathological study. Neuro Oncol. 2012;14:230–241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Ostrom QT, Gittleman H, Liao P, Rouse C, Chen Y, Dowling J, et al. CBTRUS statistical report: primary brain and central nervous system tumors diagnosed in the United States in 2007‐2011. Neuro Oncol. 2014;16(Suppl 4):iv1–iv63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Englot DJ, Han SJ, Berger MS, Barbaro NM, Chang EF. Extent of surgical resection predicts seizure freedom in low‐grade temporal lobe brain tumors. Neurosurgery. 2012;70:921–928. discussion 928. [DOI] [PubMed] [Google Scholar]
- 46. Mahaley MS Jr, Dudka L. The role of anticonvulsant medications in the management of patients with anaplastic gliomas. Surg Neurol. 1981;16:399–401. [DOI] [PubMed] [Google Scholar]
- 47. Engel J Jr, Bragin A, Staba R, Mody I. High‐frequency oscillations: what is normal and what is not? Epilepsia. 2009;50:598–604. [DOI] [PubMed] [Google Scholar]
- 48. Dang L, White DW, Gross S, Bennett BD, Bittinger MA, Driggers EM, et al. Cancer‐associated IDH1 mutations produce 2‐hydroxyglutarate. Nature. 2009;462:739–744. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Kolker S, Pawlak V, Ahlemeyer B, Okun JG, Horster F, Mayatepek E, et al. NMDA receptor activation and respiratory chain complex V inhibition contribute to neurodegeneration in d‐2‐hydroxyglutaric aciduria. Eur J Neurosci. 2002;16:21–28. [DOI] [PubMed] [Google Scholar]
- 50. Jacob F, Salinas RD, Zhang DY, Nguyen PTT, Schnoll JG, Wong SZH, et al. A patient‐derived glioblastoma organoid model and biobank recapitulates inter‐ and intra‐tumoral heterogeneity. Cell. 2020;180:188–204 e122. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1
Data Availability Statement
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.