MGMT promoter gene methylation and neurological scale improvement in glioma : a cohort study [ version 2 ; peer review : 1 approved , 1 not approved ]

Background: Glioma is one of the most common primary brain tumours and conveys a dismal prognosis despite aggressive treatment. Several biomarkers have been studied in the hope of yielding better diagnostic accuracy and improving patient management. Besides survival, functional and neurological disability are concerns that have no lesser importance. In 2017, a disease-specific assessment tool – the Neurologic Assessment in Neuro-Oncology (NANO) scale – was developed to measure neurologic function in neuro-oncology cases. We sought to determine biomarkers that might be associated with neurological scale improvement in glioma patients. Methods: Glioma grade II-IV patients were recruited from three major hospitals in Jakarta-Tangerang. Isocitrate dehydrogenase (IDH) mutation and O6-methylguanineDNA methyltransferase (MGMT) promoter gene methylation were tested, as well as patients’ neurological function before surgery and three months after. Improvement in neurological scale (NANO scale) was considered positive if there was a decrement of ≥1 of the scale. Results: There were 54 patients included in the study. Mean age was 43.63 (14.723) years old, and 61.1% were male. As much as 16 (29.6%) carried a mutation in codon 132 of the IDH1 gene, and 33 (61.1%) were MGMT methylated. Median NANO scale score before and three months after surgery was 4 (012) and 3 (0-12), respectively. Neurological improvement was found in 44 (81.5%) of the patients. Among patients with MGMT promoter gene Open Peer Review


Amendments from Version 1
The new version contains a few changes as listed below.
1. In Methods section under the 'Histopathological and molecular analysis' heading: The histopathological criteria included in this study and how the pathological examination was performed, were added.

Introduction
Glioma is one of the most common primary brain tumours, in which 80% of the cases are found to be malignant ( Grade II glioma has a median survival ranging from four to more than ten years (Bell et al., 2018;Claus & Black, 2006). However, glioblastoma -a grade IV glioma -accounts for the majority of gliomas and carries a poor prognosis with median survival of six to 14 months despite aggressive treatment with radiotherapy plus chemotherapy, and less than 12 months with radiotherapy alone (Stupp et al., 2005).
Molecular characteristics of glioma have undergone extensive research in the last decade, and some biomarkers have even been integrated into the 2016 glioma classification. Isocitrate dehydrogenase (IDH) mutation and O6-Methylguanine-DNA methyltransferase (MGMT) promoter gene methylation has been associated with better survival. Nonetheless, patients' quality of life and neurological deficit have not been considerably measured and correlated with these prognostic biomarkers.
In 2017, the Neurological Assessment of Neuro-Oncology (NANO) scale was developed to objectively measure neurological deficit in patients with brain tumours (Nayak et al., 2017). Previous studies have shown that a good initial NANO scale score was significantly associated with improvement in neurological deficit two months after surgery (Gunawan et al., 2020) and a more powerful method to predict prognosis during initial diagnosis and disease progression (Lee et al., 2018). In this study, we aimed to elucidate biomarkers that might be associated with neurological scale improvement three months after surgery.

Patient collections and ethics statement
This is a cohort study, with a purposive sampling method. There were 72 suspected brain tumour patients, aged more than ten years old, who presented at three major hospitals in Jakarta and Tangerang over a period of one year (July 2019-July 2020). After a brain MRI, 5 patients were excluded due to tumour location (brainstem or infratentorial). Written consent was obtained from every patient or their caregiver (for patients who are unable to give consent). Karnofsky Performance Scale (KPS) as well as neurological scale were assessed before surgery. Following tumour resection, 62 patients were histologically confirmed as having glioma grade II-IV. Tumour tissue was tested for IDH mutation as well as MGMT promoter gene methylation. Patients underwent standardised therapy, and neurological function was re-assessed three months after surgery. Six patients were deceased before the second neurological scale assessment, and two patients were lost to follow-up, hence were excluded from the study. The study was approved by Medical Ethical Research Committee, Universitas Hasanuddin, No: 1232/UN4.6.4.5.31/PP36/2019.

Histopathological and molecular analysis
Histopathological and molecular analysis was carried out in Mochtar Riady Institute Tangerang and Kalgen Innolab Clinical Laboratory, Jakarta. Tissue sample dissected from the tumor was immediately put in formaldehyde 10% buffer. Six to 24 hours later, the desired tumor tissue was stored into cassettes and processed in an automatic tissue processor. After an overnight processing, tissue was put into a mold with paraffin wax. Paraffin block with the embedded tissue was cut using a microtome and placed on glass slides stained with hematoxylin and eosin to determine the diagnosis and grade of malignancy. Histopathological examination was conducted by two neuropathologists and classified according to the 2016 WHO Classification of Tumours of the Central Nervous System (Louis et al., 2016). Diffuse astrocytic and oligodendroglial tumours and other astrocytic tumours WHO grade II-IV were included in the study. Area with the most tumour accumulation along with tumour cell percentage was marked. This area was then scraped for further analysis.

DNA isolation
Genomic DNA was extracted from paraffin-embedded tumor tissues using the QIAamp ® DNAMicroKit (QIAGEN, catalogue number 56304) following the manufacturers' protocol.
HRM analysis consists of 3 steps: PCR reaction, DNA melting, and data analysis. These analyses were performed using the LightCycler ® 480 High-Resolution Melting Master Kit (Roche) according to the manufacturer's instructions. PCR amplification and HRM analysis were carried out in a LightCycler ® 480 real-time PCR system (Roche Diagnostics). PCR amplification started with an initial denaturation of 95°C for 10 minutes (which included activation of FastStart TM Taq DNA polymerase and denaturation of DNA), followed by 45 cycles of amplification (denaturation at 95°C for 10 seconds, annealing at 60°C for 30 seconds, and extension at 72°C for 12 seconds). Melting was performed with a denaturation step at 95°C for 1 minute, followed by an annealing step at 40°C for 1 minute and a melt from 70 to 97°C at a ramp rate of 0.03°C/second with 18 acquisitions per degree Celsius. These results were analyzed using LightCycler ® 480 software version 1.5.0.
Sanger sequencing was conducted using Applied Biosystems 3500 Genetic Analyzer. For each reaction, 200 ng of genomic DNA was amplified with the following PCR conditions: an initial 10-minute denaturation at 95°C followed by 40 cycles of 30 seconds at 95°C; 30 seconds at 60°C; 30 seconds at 72°C, and a final extension of 7 minutes at 72°C. Sequencing were aligned and edited with the BioEdit Software version 7.2.5. Samples with conflicting results by HRM and sequencing were re-tested and only HRM-positive samples confirmed by sequencing were considered mutated.
MGMT promoter gene methylation DNA extracted first underwent bisulphite treatment to convert all unmethylated cytosine to uracil, leaving 5-methylcytosine unaltered. After being eluted in DNase-free water, methylation analysis was commenced using real-time methylation-specific PCR (MSP). PCR amplification was carried out in a LightCycler ® 480 real-time PCR system (Roche Diagnostics). QuantiFast ® Multiplex PCR Kit (QIAGEN) was used for PCR and DNA was amplified utilizing HotStarTaq Plus DNA polymerase which was included in the kit. Primers used were: MGMT Methylatedforward 5'-TTTCGACGTTCGTAGGTTTTCGC-3', MGMT Methylated reverse 5'-GCACTCTTCCGAAAACGAAACG-3' and MGMT Unmethylated-forward 5' TTTGTGTTTTGATGTT TGTAGGTTTTTGT-3', MGMT Unmethylated-reverse 5'-AA CTCCACACTCTTCCAAAAACAAAACA-3'. Cycling conditions were 95°C for 5 min, followed by 42 cycles of amplification (denaturation at 95°C for 30 seconds, annealing at 59°C for 30 seconds, and extension at 72°C for 30 seconds) and 72°C for 5 min. PCR reactions (15 μl) were analyzed on a 2% agarose gel stained with SYBR Safe. Commercial methylated DNA and unmethylated DNA (Zymo Research) served as positive controls. Outcomes were classified as either methylated or unmethylated.
Neurological scale assessment and follow-up Neurological function was assessed using the NANO scale. NANO scale is a simple neurological assessment evaluating patients in nine domains: gait, strength, upper extremity ataxia, sensation, visual fields, facial strength, language, level of consciousness, and behaviour. Each domain was scored from 0 to 2 or 3, with higher score indicating worse neurologic function. Neurological scale was assessed by two individual physicians, and subjective complaints not included in the scale were also noted. Any discrepancies between investigators were discussed and optimal scale were determined. NANO scale was first assessed before surgery and then re-assessed three months later during their routine clinical follow up. Patients who were deceased before the second assessment were excluded from the study. Neurological improvement was defined as decrement in NANO scale score by 1 or more, which was calculated from difference of initial NANO and NANO three months after surgery.

Statistical analysis
The association between baseline characteristics and biomarkers were analysed. Categorical data are presented as proportions and interpreted using chi-square or Pearson Fisher's exact test. All calculations were performed using IBM SPSS Statistics version 24. The reported p values are two-sided, and a probability value of <0.05 was considered statistically significant.

Results
In total, 54 patients completed the follow up and were analysed. Half of the cases (50%) were glioma grade IV, followed by grade II (37%) and grade III (13%). Overall mean age was 43.63 (14.723) years old for all gliomas, and mean age for glioblastoma was 50 (12.7) years old. Male to female ratio was 1.57. Median Karnofsky Performance Scale (KPS) score was 55 (30-80). Most of the patients underwent surgery followed by radiation and chemotherapy (79.6%) ( Table 1).
In 16 patients (29.6%) an IDH1 mutation was found and no IDH2 mutation was found. Younger age and male gender were significantly associated with having an IDH mutation (Table 2).
There were 33 patients (61.1%) that were MGMT methylated (Table 3). Older patients had a higher tendency to have methylated MGMT than younger patients. Patients with IDH mutations were more likely to harbour MGMT methylation (p=0.049; OR=1.54; 95%CI 1.05-2.26).
The number of patients with MGMT methylation who underwent surgery plus chemoradiation, surgery plus radiation only, and surgery only were 26 patients (78.8%), one patient (3%) and six patients (18.2%), respectively.
Median NANO scale score before surgery was 4 (0-12) and three months after surgery was 3 (0-12). Improvement in neurological function, measured using the NANO scale, was found in 44 (81.5%) of the patients. Age, gender, initial KPS, tumour location, grade of glioma and IDH mutation were not associated with improvement in neurological function. Among patients with MGMT promoter gene methylation, 90.9% showed improvement in neurological function. A multivariate analysis using logistic regression confirmed MGMT methylation as an independent factor towards NANO scale improvement (p=0.035; OR=5; 95%CI 1.122-22.272) ( Table 4).  Further analysis shows that coexistence of IDH mutation and MGMT methylation were mostly found in grade II patients (53.8%) and 100% of patients with coexistence of both biomarkers showed improvement in neurological scale (p=0.032) ( Table 5).
We did a subgroup analysis of patients with glioblastoma. Baseline characteristics of patients with glioblastoma were depicted in Table 6. IDH mutation and MGMT methylation did not show any relation towards neurological scale improvement (Table 7).

Discussion
Glioma remains to be a challenging tumour, with diverse clinical presentation, phenotype and molecular parameters (Louis et al., 2016). It is not unusual for clinicians to encounter cases of higher-grade glioma with longer survival than  lower-grade glioma during conventional treatment. Despite survival, clinical and functional status is another concern in treatment initiation and disease progression. KPS has been generally used to evaluate brain tumours' performance status. However, like tumour grade, there have also been cases in which those with a low KPS score survived longer than those with a high KPS score (Lee et al., 2018).  The NANO scale is a relatively new scale that serve as an objective and quantifiable metric of neurologic function in brain tumour patients (Nayak et al., 2017). A previous study found that performance status estimated by the NANO scale was significantly associated with overall survival, and was a more powerful method to predict the prognosis of GBM than the KPS during both initial diagnosis and disease progression (Lee et al., 2018). It was also found that initial NANO scale possesses a stronger correlation neurological scale improvement  ., 2017). Taking into account that this is a short follow-up study, a decrement of ≥2 will be unlikely, especially if the mean progression of NANO scale is 4.1 points in 18.9 months, which means a progression of 1 point in 4.7 months. Thus, decrement of ≥1 score of NANO scale was considered as seemingly relevant.
From 54 patients included in the study, most of the cases were glioma grade IV, glioblastoma, IDH-wildtype (44. Tumours were mostly located in the frontal lobe, followed by temporal, parietal, and occipital. The most common presenting symptom was headache, and 58.8% of tumours located in the frontal lobe presented with this symptom. Surgery, radiation, and chemotherapy were underwent by 79.6% of patients.

References
It was found that 29.6% of patients were positive for IDH1 mutation. Being of a younger age and male were factors associated with having an IDH mutant. This is similar to previous studies that found younger age to be associated with IDH mutation in glioblastoma (Molenaar et al., 2014; Songtao et al., 2012). However, a study in Indonesia showed no differences in age and gender towards IDH mutation (Malueka et al.,  2020). Dissimilarity may be caused by differences in characteristics of patients included in their study (which also included grade I glioma). Grade I glioma is frequently found in children, with different characteristics, therapeutic interventions, and prognosis. Therefore, in this study, we did not include grade I glioma and excluded patients less than 10 years old. Although IDH mutation alone was not significantly related to scale improvement, the combination of IDH mutation and MGMT methylation was associated with improvement in neurological function three months later. This might be due to the significant association of MGMT methylation towards NANO scale improvement, which influenced the relation between its combination with IDH mutation towards scale improvement.
MGMT promoter gene methylation has been commonly accepted to confer survival advantage regardless of therapy (Reifenberger et al., 2014). It carries both prognostic and predictive value. In this study, it was found that gliomas with MGMT gene promoter methylation are five times likely to show neurological improvement three months after surgery. However, a subgroup analysis of patients with glioblastoma, dis not show significant association between MGMT methylation status and neurological scale improvement. This might be apparently due to the small sample size of glioblastoma patients in this study.
IDH mutation is one of the molecular tests well known for its prognostic and predictive implications in high-grade as well as some low-grade gliomas. Patients with IDH mutation have been shown to convey a longer overall survival and progression-free survival compared to those in the wild-type group ( Our study found no association between IDH mutation and neurological improvement. This might be due to the difference in study outcome, which is improvement in neurological function rather than survival or progression-free survival. Secondly, since we used the difference between initial NANO scale score and score three months later as a measure for neurological improvement, it is possible that patients with low initial scale scores had a persistent score three months later, which is then assessed as not having improvement after three months. Third, the NANO scale as a measure for neurological improvement does not integrate headaches into the scale, which is the most common presenting symptom in this study.
There are several limitations of this study. First, we did not analyse other predictive biomarkers such as 1p19q deletion and G-CIMP phenotype. Second, the follow up period of three months after surgery might not reveal changes in neurological improvement, since some patients might still be in chemotherapy and radiation therapy. Hopefully, future studies could integrate other biomarkers as well and conduct a longer period of follow up.

Conclusions
MGMT promoter gene methylation as well as coexistence of IDH mutation and MGMT methylation shows a significant link towards improvement in NANO scale score three months after surgery. Glioma patients with MGMT gene promoter methylation are five times more likely to show neurological improvement three months after surgery.

Open Peer Review © 2021 Dwianingsih E.
This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Ery Kus Dwianingsih
Department of Anatomical Pathology, Dr Sardjito Hospital, Yogyakarta, Indonesia Thank you for providing the revised manuscript. However as I mentioned before, the length of follow up in the patients (only 3 months post-surgery) is not long enough to determine the patients' outcome. As this issue cannot be addressed by this revision I could not recommend for this paper to be indexed at this time. Thank you.
In this study the patients were followed for only 3 months post-surgery, which is not long enough to determine the patients' outcome. Longer follow up is needed to give better insight of the MGMT profile related to its prognostic evaluation as seen in previous study conducted by lee et al, in 2018, which was also cited in the manuscript.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Partly

Are sufficient details of methods and analysis provided to allow replication by others? Partly
If applicable, is the statistical analysis and its interpretation appropriate? I cannot comment. A qualified statistician is required.
Are all the source data underlying the results available to ensure full reproducibility? Partly Are the conclusions drawn adequately supported by the results? Yes 'Histopathological and molecular analysis' heading. 3. We agree that the length of follow-up was of short term, hence further studies with longer period of follow-up may give a better insight. For that reason, we have stated this matter under the Discussion section, at the last paragraph.
Thank you for your kind and valuable suggestions.
Best Regards, Pricilla et al