ABO blood group and risk of glioma: a case control study from Serbia studija slu č aj-kontrola iz Srbije

Backgraund/Aim. Gliomas are the most common primary brain tumors and the etiology is unknown. The aim of this study was to investigate possible association between incidence in relation to glioma and certain blood groups. Methods. The case-control study included 100 pathologi-cally confirmed cases of glioma at the Clinical centar of Kragujevac, Serbia, between 2014 and 2015, and 200 age-and sex-matched controls without malignant diseases in personal and family history at the same institution. After signing the informed consent, all patients filled out an epi-demiological questionnaire. Results. In the analysis comparing the glioma patients with the control group, a significant association ( p < 0.0005) was observed in relation to the blood group AB. Furthermore, it was not observed a significant association in relation to the blood group A ( p = 0.070), blood group B ( p = 0.256), blood group O ( p = 0.768) among the compared groups. Also, in the analysis comparing glioma patients with the control group, a significant association was observed in relation to the years spent in hometown ( p = 0.035), changing the place of residence ( p = 0.007), the body weight ( p = 0.002) and the body mass index ( p < 0.0005). Univariate binary logistic regression showed that higher number of years spent in the hometown [odds ratio (OR) 1.011 95% confidence interval (Cl) 1.000– 1.023; p = 0.043], increased body weight (OR 0.976 95% Cl 0.959–0.993, p = 0.006) and increased body mass index (OR 0.898 95% Cl 0.839–0.961; p = 0.002) increase the risk of glioma. However, a change of the residence decreases the risk of glioma (OR 0.327 95% Cl 0.147–0.727; p = 0.006). Univariate binary logistic regression analysis revealed that the individuals with group AB were at 3.5-fold increased risk of developing glioma compared to the individuals with other ABO blood groups (OR 3,429 95% Cl 1,83–6,41; p < 0.0005). Also, there was more male patients with glioma with the blood group AB ( p = 0.001). Conclusion. In a present study, we demonstrate that individuals with the group AB have an increased risk of developing glioma. platelets


Introduction
Gliomas are the most common primary brain tumors 1 . They accounts for approximately 27% of all brain and central nervous system tumors and about 80% of malignant brain tumors. Gliomas are tumors arising from glial or precursor cells and include astrocytoma, glioblastoma, oligodendroglioma, ependymoma, mixed glioma, malignant glioma, not otherwise specified (NOS) and few rare histologies 2 .
The etiology of glioma is unknown. Several possible risk factors for glioma were suggested in the past, but with the exception of ionizing radiation, none was consistently confirmed 3,4 . Several studies investigated the relationship between the blood groups ABO and glioma risk and demonstrated conflicting results [5][6][7][8][9][10] . The scientist Landsteiner found the blood group ABO system in 1901 at the University of Vienna. Locus ABO is localized on chromosome 9 and gene ABO is inherited in an autosomal dominant 11 . The A and B antigens are complex carbohydrate molecule and they represent extracellular domains on the surface membrane of red blood cells. The A and B alleles of the ABO locus encode different enzyme glycosyltransferases, adding N-acetylgalactosamine and Dgalactose on a common precursor of the side chain, the H determinant, which is then converted into the A or B antigen. The group O individuals lack such functional enzymes, and express unchanged H determinant. In addition to the surface of red blood cells, A and B antigens are found on a variety of human cells and tissues, including epithelial cells, neurons, platelets and vascular endothelium 12,13 .
Distribution of the four different blood groups varies between countries. Generally, blood group O has highest prevalence, while blood group AB has the lowest 14 . In Serbia, the proportion of ABO blood groups is 41.5% for A, 37.5% for O, 16% for B and about 5.5% for AB 15 .
The aim of this study was to examine incidence of glioma in the patients with different blood groups ABO.

Methods
The study group included a series of 100 consecutive patients (59 males, mean age 59.76 ± 10.76 years, and 41 females, mean age 58.36 ± 8.95 years) with histopathologically verified diagnosis of glioma according to the WHO criteria (ICD-O-3). Of these 100 patients, the blood groups were available in 96 patients. The study was realized according to the Declaration of Helsinki and performed in accordance with the Ethics Committee of Clinical Center Kragujevac. The criteria for inclusion into the study were the patients who had a consultative decision evaluated on the basis of stages of the disease and the general condition of the patient. Also, the criteria for inclusion into the study were primary, previously untreated glioma. The criteria for exclusion from the study were previously diagnosed and treated glioma and glioma recidivans. The patients were treated surgically, followed by radio and chemotherapy at the Center for Oncology, Clinical Center Kragujevac. Data were collected between April 2015 and May 2016. The control group included 200 patients matched by sex and age (109 males, mean age 59.32 ± 10.71 years, and 91 females, mean age 58.09 ± 9.12 years), who were hospitalized in the Clinical Centre Kragujevac, without malignant diseases in personal and family history. Of these 200 patients, the blood group were available in 172 patients. After obtaining the patient consent, a questionnaire based on the reviewed literature data and personal considerations was to collect detailed information. It included questions about different factors supposed to be a possible risk or protective factors.
The first part of the questionnaire referred to demographic features of the patients [sex, age, birth and residency places, blood type, Rh (D) antigen, school education, occupation, number of family members, the size of the home space]. The second part of the questionnaire included the family history of chronic diseases, including brain tumors and other malignant tumors. The third part of the questionnaire referred to the personal history of the disease including reproductive risk factors. The fourth part deals with information on the risks of exposure to external factors, the fifth included data on habits (smoking, drinking coffee, tea, alcohol) and the sixth part referred to information about nutrition.

Statistical analysis
Statistical analysis was performed using the SPSS software version 19.0. The chi-square (χ 2 ) test was used for the comparison of qualitative variables between the patients with glioma and the control group. The categorical variables were presented as frequencies and percentages. The univariate analysis using the logistic regression techniques, including the odds ratio (OR), was performed to determine the effects of blood groups on the dependent variable (glioma). The P value of less than 0.05 was considered statistically significant. Table 1 shows the demographic and personal characteristic of patients and controls. Difference in years spent in hometown between the study and the control group was statistically significant (p = 0.035). The control group patients were more frequently changing the residence than patients with glioma (p = 0.007). Also, the statistically significant difference was found between the compared groups in relation to the body weight (p = 0.002) and to the body mass index (p < 0.0005). The patients with glioma had a higher body weight and higher body mass index in comparison to the patients of the control group. In comparing the patients with glioma with the control group patients, a significant association (p < 0.005) was observed in relation to the blood group AB. Furthermore, no significant association was observed in relation to the blood group A (p = 0.070), blood group B (p = 0.256), blood group O (p = 0.768) among the patients with glioma and the control group (Table 2).

Results
In the group of patients with glioma, 32.3% of patients had the blood type AB, 25.0% of patients had the blood group A, 10.4% the blood group B and 32.3% the blood group O. The ratios of blood groups AB, A, B and O were 12.2% v.s. 32.6% v.s. 18.5% v.s. 34.9% in the control group, respectively ( Table 2).
Comparing the male patients with glioma and the control group as the blood groups ABO, a significant association was observed in relation to the blood group AB (p = 0.001). There was no significant difference regarding the blood group A (p = 0.726), blood group B (p = 0.668) and the blood group O (p = 0.726), (Table 3).
Comparing the female patients with glioma and the control group as the blood group AB, differences did not reach significance there (p = 0.057). Also, there was no significant difference regarding the blood group A (p = 0.288), blood group B (p = 0.549) and the blood group O (p = 0.522) ( Table 4).

Groups
Patients with glioma n (%)   Univariate binary logistic regression showed that higher number of years spent in the hometown (OR 1.011 95% confidence interval (Cl) 1.000-1.023; p = 0.043) increased the body weight (0.976 95% Cl 0.959-0.993) and increased the body mass index (OR 0.898 95% Cl 0.839-0.961) increased the risk of glioma. However, a change of the residence decreased the risk of glioma about three times (OR 0.327 95% Cl 0.147 -0.727; p = 0.006) ( Table 5).

Discussion
In our study, we demonstrated that the distribution of blood groups in the patients with glioma differ significantly than in the control group, with a higher incidence of glioma reported in the individuals with the blood group AB. Previous studies showed different results about the ratios of blood groups in the patients with glioma. The study of Yates and Pearce 5 , which was conducted on 473 cases with astrocytoma excluding glioblastoma multiforme, showed that in 164 patients, who were diagnosed before 1945, a distribution of the blood groups was normal. In 305 patients analyzed after that year, a statistically significantly lower number of glioma in the patients with the blood group O was noticed. Selvestrone and Cooper 6 also examined the relationship between the blood groups and astrocytic brain tumors, including glioblastoma multiforme, and they found a statistically significantly lower number of patients with the B and O blood types. The study of Jates 7 recruited 160 patients with glioma and the same number of controls and observed no significant difference in the distribution of blood groups be-tween the glioma patients and controls from the Oxford region in the United Kingdom. Strang et al. 8 did not detect significant difference in the distribution of blood group between 900 astrocytoma patients and the control population. However, once adjusted for sex, there was significantly more male cerebral astrocytoma patients with group A (p < 0.05). In the meta-analysis by Zhang et al. 11 which examined the relationship between the blood groups and cancer risk were included 89 studies (82 case control studies and 7 cohort studies). The study by Akhtar et al. 9 analyzed 1674 patients with glioma and results suggested that the blood groups were not significantly associated with a glioma risk. In a recent study, Allouh et al. 12 reported a significant association between the distribution of blood group ABO antigens and glioblastoma multiforme in Jordanians, with a higher incidence reported in the persons with the blood group A.
In the last five decades, relationship between the blood groups ABO and carcinoma has been extensively investigated.
Mechanisms that explain the relationship between the blood groups ABO and a cancer risk are unclear. Several hypotheses have been proposed, including a modulatory role of blood group ABO antigens. The blood group antigens (A, B) are expressed on the surface of red blood cells and numerous other tissues. For a variety of tumor types, the blood group antigens expressed on the surface of malignant cells was found to be different from the antigens expressed on normal cells [15][16][17] . Modified expression of blood group antigens on the surface of cancer cells may alter cell motility, sensitivity to apoptosis and immune escape, and thus influence the initiation and spread of cancer 18 . Also, the blood group ABO system regulate the level of circulating proinflammatory and adhesion molecules (such as E-selectin, P-selectin and intracellular adhesion molecule-1), which are important for tumourigenesis. Chronic inflammation was extensively linked with a cancer development and provides a potential mechanisms by which the ABO antigens may influence a risk of cancer. One of the possible explanations about the role of the blood group ABO in the tumourigenesis process is the recent discovery of vWF that is an important modulator of angiogenesis and apoptosis. Individuals with other blood groups have significantly higher levels of plasma concentrations of vWF compared to individuals with blood group O. On the other hand, vWF may not have a significant role in cancer progression, and the elevated vWF levels can simply be an indication of the extent of endothelial dysfunction caused by tumor growth 19 . Some research have shown that the structure of certain tumor antigens is similar to the structure of antigens of the ABO blood group system. For example, Prieto and Smith 20 presented the Forssman antigen. This antigen is synthesized predominantly in stomach and colon cancer and structurally it is identical to the A antigen determinant. One study found neoexpression of ABO blood group antigens in hepatocellular carcinoma tissues. These mechanisms are biologically plausible to explain the association between the ABO blood group and a risk of cancer; the underlying mechanisms for the discrepancy among different cancer sites still remains a challenge 19 . Morever, our results suggest that a change of place of living decreases a risk of glioma and that higher number of years spent in hometown increases the risk of glioma. Until now, there have been no studies that have examined the relationship between the change of residence and a number of years spent in hometown and a risk of glioma. In this case control study, increased body weight and increased body mass index increase a risk of glioma.

Conclusion
To our knowledge, this is the first study that examined the frequency of blood group in patients with glioma among the Serbian population. Also, for the first time our study results suggest that blood group AB increases the risk of glioma.
The results of our study suggest that the blood group AB can be the one of hereditary factors which have an influence on the occurrence of glioma. The further research on a much larger sample is needed to confirm these findings as well as potential mechanisms by which the blood group ABO system affects the occurrence of glioma.