Prognostic significance of body mass index and diabetes in patients with malignant glioma

Aim: We aimed to determine whether there is a relationship between body mass index (BMI) and diabetes (DM) before treatment and survival with this study. Material and Methods: The results of patients who received radiotherapy between 2010 2018 were evaluated with this retrospective study. BMI was categorized into 3 groups: normal (18.5-24.9 kg/m2), overweight (25-29.9 kg/m2), obese (≥30 kg/m2). Presence of diabetes was evaluated by considering oral antidiabetic use and file information before treatment. Patient, treatment and tumor characteristics were evaluated with descriptive statistics. Kaplan-Meirer, log-rank and coxregression analyzes were performed. P <0.05 was considered statistically significant. Results: The results of 174 cases were evaluated. Diabetes was present in 22 patients (12.6%). In univariate analyzes, being over the age of 65 (p <0.001), Karnofsky performance score (KPS) below 80 (p <0.001), diabetes (p = 0.017), having grad 4 pathology (p <0.001), performing subtotal excision / biopsy (p <0.001), hypofractioned / whole brain radiotherapy (p <0.001), and not receiving adjuvant chemotherapy (CT) (p <0.001) had a negative effect on overall survival (OS). In multivariate analyzes, being over 65 years old, having grad 4 pathology, performing subtotal excision / biopsy and not taking adjuvant CT were found to be effective on OS. Median overall survival in diabetics was 9.65 months and 17.74 months in non-diabetics (p = 0.017). No statistically significant relationship was found between BMI and OS. Conclusion: Pre-existing diabetes in malignant glioma patients is a risk factor for poor outcomes. It is important to control diabetes and related conditions.


INTRODUCTION
Malignant gliomas account for about half of all brain tumors in adults [1]. The World Health Organization (WHO) classifies grad 3 (anaplastic gliomas) and grad 4 (glioblastoma) gliomas as malign gliomas [1]. Currently the standard treatment applied to high grade glioma involves a maximally-safe resection, concurrent radiotherapy (RT) and temozolamide treatment followed by adjuvant temozolamide [2]. Despite all treatments, survival is very low in malignant gliomas. 5-year survival is 18% for WHO grad 3 tumors and <5% for glioblastomas (GBM) [3]. In addition to known bad prognostic factors such as age, O-6 methylguanine DNA methyltransferase (MGMT) status, Karnofsky performance score (KPS), it is important to identify prognostic factors before treatment [4].
Obesity and diabetes (DM) are among the most important health problems in the world [5]. Cancer incidence is expected to increase due to increased risk factors such as obesity, DM and lifestyle [6,7]. While 9.3% of the global adult population is diabetes in the world [8], this rate was found to be 9-15.7% in studies conducted with glioma patients [3,9,10]. World health organization defines obesity as body mass index (BMI) ≥30kg / m 2 and overweight as BMI≥25 kg / m 2 [5]. The prevalence of obesity in the Turkey was reported to be 28.5% [11]. In studies on gliomas, the frequency of obesity was found between 20-30 [3,9,12]. Increased BMI has been associated with various types of cancer: colorectal cancers, breast, endometrium, ovarian, kidney, pancreas, esophageal cancers [13,14]. It was stated that DM and high BMI contributed to 5.7% of all incidental cancer cases in 2012 in a study conducted in 2018 [10]. There is evidence that DM is associated with endometrium, bladder, pancreas, liver, colorectal and breast cancer [3,9,15]. The underlying cause of the relationship between BMI, DM and cancer is hyperglycemia, hyperinsulinemia, chronic inflammation, and irregularity in sex hormone activity [10]. However, the relationship between malignant gliomas and high BMI and DM is contradictory in studies. Therefore, with this study, we aimed to determine the relationship between malignant gliomas and DM, high BMI.

MATERIAL AND METHODS
With this study, the results of 174 patients who received radiotherapy in our clinic between January 2010 and December 2018 were retrospectively analyzed. The study included grade 3-4 glioma patients according to WHO criteria, over the age of 18, whose diagnosis was confirmed by histopathologically, weight, height and DM story can be reached. Type 1 diabetes patients were not excluded from the study. Ethics committee approval was obtained before starting the study. Due to the nature of the study, informed Ortadoğu Tıp Dergisi / Ortadogu Medical Journal consent forms were not obtained from the patients. Patient data and treatment characteristics were obtained from medical records and hospital system. Body mass index was calculated according to the kg / m2 formula, considering weight and height measurements before treatment. It was categorized into 3 groups: normal weight (18.5-24.9 kg / m 2 ), overweight (25-29.9 kg / m 2 ), obese (≥30 kg / m 2 ).
The presence of diabetes was defined by considering disease history, antidiabetic use, blood glucose levels at least 6 months before the operation.
Groups were categorized as gross total excision (GTR) and subtotal excision/biopsy considering the resection width operation notes and Magnetic Resonance Imaging (MR) images taken after surgery.

Radiotherapy
All patients were treated with 3D conformal radiotherapy. Conventional radiotherapy was considered as a 1.8-2 Gy fraction dose and 50 Gy and above dose. 30-42.5 Gy RT was applied in 10-16 fractions in hypofraction and whole brain radiotherapy. RT was started within 5-6 week after the operation. MRI was performed in each patient before RT. CT simulation was performed in supine position for planning purposes. By fixing with a thermoplastic mask, a tomography of 3-5 mm section thickness was taken. Computed tomography (CT) images were fused with preop and postop MR images. Gross tumor volume (GTV) was defined as the volume and operation bed enhanced by MR image. This volume was created by giving the clinical target volume (CTV) with a margin of 1.5-2 cm and the planned tumor volume (PTV) with a margin of 0.5 cm to CTV. All brain RT was applied with 3D conformal radiotherapy, hypofractionated or conventional treatments were applied with intensity-adjusted RT (IMRT). All patients were treated with eclipse planning system.

Follow-up
During the treatment, a complete blood count, biochemistry tests and a physical examination were performed once a week. Potential side effects were evaluated. After RT, he/she was followed up with a physical examination and MR every 3 months.

Endpoint
The primary endpoint of the study was overall survival (OS). The second endpoint was to determine the effect of DM and BMI on OS.

Statistical Analysis
Descriptive statistics were applied to determine patient and treatment characteristics. Average, median, and standard deviations were calculated in order. Overall survival was defined as the time from the time of diagnosis to death or final control. Progression-free survival was determined as the time from diagnosis until relapse or progression, or until death. Chi-square test was carried out to compare the categorical variables. Kaplan Meier Analysis was carried out for survival analysis. In univariate analysis, the survival curves of the subgroups were evaluated with a log-rank test. In univariate analysis, all variables with p <0.10 were included in the multivariate analysis. Cox regression analysis was performed. P <0.05 was considered statistically significant. Version 13.0 of Statistical Package for Social Sciences Software (SPSS Inc.; Chicago, IL, USA) was utilized in the whole statistical analysis.

Patient and Treatment Characteristics
The results of 174 cases were evaluated retrospectively. Median follow-up was 16.11 months. The median age was 57 . 37 (21.3%) patients had grad 3, 137 (78.7%) patients had grad 4 pathology. DM was present in 22 (12.6%) of the cases. The median BMI was 27.14. 36.8% of the cases were normal weight, 37.4% were overweight and 25.9% were obese. Treatment and patient characteristics are summarized in Table 1.

Survival Analysis
Median survival was 17.28 months. Overall survival for 1, 2, 5 and 10 years was 64.4%, 34%, 17.7% and 10.3%, respectively (Figure 1). During the follow-up period, 138 (79.3%) patients died. Being over 65 years of age, KPS <80, presence of diabetes, grad 4 pathology, subtotal excision / biopsy, hypofractionated / whole brain radiotherapy and absence of adjuvant chemotherapy (CT) had negative effects on overall survival ( Table 2) in univariate analysis. Being over 65 years old, having grad 4 pathology, subtotal excision and not taking adjuvant CT were found to be effective on OS ( Table 3) in multivariate analysis. Median overall survival in diabetics was 9.65 months, while in non-diabetics 17.74 months (p=0.017) (Figure 2). It could not be shown statistically significant relationship between BMI and OS.

DISCUSSION
Diabetes is a metabolic condition that increases the risk of many types of cancer [6]. Many studies have evaluated the relationship between cancer types and DM and BMI. This is not clear in gliomas. Pearson-stuttard et al. showed that alone DM contributes to 2.1% and high BMI contributes to     [10]. As obesity and DM continue to increase all over the world, it is predicted that there will be an increase in cancer mortality and incidence in the coming period. In the case-control study conducted by Barami et al.,15.7% diabetes and 27.7% obesity were observed in the glioma patient group, while this ratio was 16.8% and 32.1%, respectively in the control group. There was no relationship between DM and obesity and GBM risk. However, when the relationship between DM and survival was evaluated, it was associated with worse survival in univariate analyzes, but could not be demonstrated in multivariate analyzes [9]. In the metaanalysis performed by Tong et al., no increase was observed in the risk of brain tumors in diabetic and nondiabetic patients [16]. In our study, although the risk between glioma patients and diabetes and obesity was not examined, pre-existing DM has been shown to negatively affect survival in malignant glioma patients. With all these data, it can be said that diabetes has prognostic value, not predisposing.
It has been stated that hyperglycemia is a poor prognostic factor on survival in metaanalysis performed on glioblastoma [17]. Many mechanisms of action of hyperglycemia are emphasized. The level of insulin rises due to acquired insulin resistance. Insulin resistance, increase of various cytokines, increase of insulin -like growth factor-1 (IGF-1), increase of adipokine balance play a role in cancer development [18]. When Derr et al. classified glioblastoma patients according to their blood glucose levels in their study, they found 14.5 months of survival in patients with glucose <94 mg/dl and 9.1 months in patients with >137 mg/dl (p=0.041) [19].
The increase in obesity causes comorbidities such as cardiovascular disease, DM, cancer [13]. When the relationship between survival and BMI is examined, there are contradictions in the literature. Patharaju et al. showed that the increase in survival occurred when BMI was elevated [20]. However, Chambless et al. showed that DM and high BMI in high-grade gliomas are independent risk factors for poor results in a retrospective study on 171 patients [3]. These results are contradict with Jones et al. study with 1259 patients. In this study, no relationship was found between BMI and survival [12]. Similarly, in a prospective cohort study for brain tumors, could no relationship was found between obesity and glioma risk [21]. In our study, could no relation was found between BMI and survival. This may be due to the fact that weight and height data were evaluated at the first examination before radiotherapy. Patients may also have experienced weight change due to steroid use before and after the operation.
The limitation of our study was that it was retrospective. HbA1c levels and MGMT status of patients could not be determined before treatment due to the retrospective nature. Clear information on hyperglycemic control adequacy could not be obtained.

CONCLUSION
As a result, diabetes is a poor prognostic factor for the survival of malignant glioma patients. It is important to control diabetes and related conditions. In the future, prospective studies should be conducted to investigate the relationship and mechanism of action between both DM and high BMI and the risk and mortality of malignant glioma.