ONCOLOGY LETTERS 5: 783-786, 2013 Prevalence of hepatitis B and C in patients with meningiomas and glioblastoma multiforme 1 1 2 2 MARC B. CABANNE , QUANG D. MA , LILLIAN MECUM , RAHUL JANDIAL , 1 2 JAVED SIDDIQI and MIKE Y. CHEN 1 Division of Neurosurgery, Arrowhead Regional Medical Center, Colton, CA 92324; 2 Division of Neurosurgery, City of Hope National Medical Center, MOB 2001J, Duarte, CA 91010, USA Received September 4, 2012; Accepted November 29, 2012 DOI: 10.3892/ol.2013.1126 Abstract. The prevalence of hepatitis B and C in patients primary brain tumors. However, substantial gains have been with glioblastoma multiforme or meningiomas has not been slow and the prognosis of patients with glioblastoma multi- described. These infections are known to modulate the forme (GBM), the most common type of primary malignant activity of the immune system, which potentially influences brain tumor in adults, remains poor. The median survival time the development and course of cancer. We hypothesized following diagnosis is 14 months. A greater understanding of that chronic hepatitis infection, which activates the immune the epidemiology, in particular with regard to immunological system, decreases the risk of brain tumors, particularly those factors, may increase insights into more effective therapeutic that are highly malignant. We performed a retrospective study strategies. to examine the prevalence of hepatitis B and C in patients with According to the 2000-2004 data of the Central Brain meningiomas and glioblastomas. The combined prevalence of Tumor Registry of the United States (1), GBM accounts for 19% hepatitis B and C in the USA from 1999-2008 was 5.7%. The of all primary intraparenchymal brain tumors. Although GBM prevalence of hepatitis B and C in patients with meningiomas is relatively rare in younger patients, its incidence increases was 2.4%; while among glioblastoma patients, the prevalence with age; individuals between 75 and 84 years of age have the of hepatitis B and C was 1.38%. The odds ratio of having highest incidence rates. The median affected age is 64 years hepatitis B or C with glioblastoma versus meningiomas was and the overall annual incidence rate is 3.1/100,000 indi- 0.56, with a confidence interval of 0.19-1.6 and a P-value of viduals. The prognosis remains poor, with a 5-year survival 0.29. Compared with historical controls, the prevalence of rate of <4% from the time of diagnosis. hepatitis B and C in meningioma and glioblastoma patients The highest incidence of primary brain tumors is due to was decreased. However, this difference may be attributed to GBM, second only to meningioma. Similar to with GBM, the the retrospective nature of our data and the natural history incidence of meningioma increases with age. The total number of hepatitis B and C infections. The prevalence of these of meningioma cases in the USA between 2004 and 2006 was viral infections was not statistically different in patients with 53,455, compared with 27,040 cases of GBM. The occurrence of meningiomas and glioblastomas. This suggests that hepatitis B meningioma is 7-fold more common in Caucasians compared and C primarily influence slow-growing, benign tumors and with African-Americans, and its frequency of occurrence in more aggressive cancers equally, if at all. To definitively test females is 2.7-fold higher that in males (1). In contrast to GBM, our hypothesis, future studies in which data are prospectively the overall 5-year survival rate of patients with benign meni- gathered are likely to be required. giomas is 70%; skull base meningioma patients and younger patients have an even higher survival rate (2). Introduction Given the poor prognosis associated with malignant brain tumors, intense research of the epidemiological and molecular Advances in resection techniques and adjuvant therapies profile of this disease is underway. Established risk factors for continue to improve the outcome of patients with malignant the development of primary brain tumors include exposure to ionizing radiation, increasing age, male gender, Caucasian ethnicity and familial tumor syndromes. The hepatitis B (HBV) and C (HCV) viruses are well characterized. A correlation between hepatitis infection and Correspondence to: Dr Mike Yue Chen, Division of Neurosurgery neoplasia has also been identified, and is reflected in the - MOB 2001J, City of Hope National Medical Center, 1500 East increased risk of hepatocellular carcinoma observed in patients Duarte Road, Duarte, CA 91010, USA E-mail: [email protected] with HBV/HCV (3-6). It is estimated that ~4% of individuals in the USA are affected by HBV. The annual incidence repre- Key words: glioblastoma, meningioma, hepatitis, epidemiology, sents 200-300,000 new cases in the USA, or an incidence rate prevalence of 1/1,359 (0.07%). By comparison, 2-5 million people in the USA are proposed to be affected by HCV, often as asymptom- 784 CABANNE et al: HEPATITIS B/C IN MENINGIOMA AND GLIOBLASTOMA PATIENTS atic carriers. The annual incidence is 150,000 new cases, with tissue diagnosis of GBM were considered in the study. When a rate of 1/1,813 (0.06%) (7). a patient was identified as having GBM, the remainder of the A correlation between brain tumors and hepatitis infection patient's medical record was explored, including past labora- is yet to be reported. An impact of co-morbidities, including tory studies, serological testing and the patient's response to the effect of HBV/HCV infection on the occurrence and/or medical history questionnaires included within the medical progression of GBM, has not yet been described. One line of record. As our comparison group, HBV/HCV infection was evidence supporting the hypothesis that a chronic hepatitis also measured in patients with meningiomas, in the same infection may decrease the risk of tumorigenesis is epidemio- manner that patients with GBM were evaluated. Given the logical data suggesting a correlation between immune system mostly benign pathology of meningiomas, patients may status and the risk of gliomagenesis. Population-based studies elect not to undergo biopsy or resection. Thus, patients with have focused on the presence of allergic and inflammatory a radiographic diagnosis of meningioma in addition to tissue states in patients diagnosed with glioma. Certain studies diagnosis were included in the study. indicate that the risk of developing glioma is decreased by heightened immune states. Schoemaker et al identified a nega- Statistical analysis. Given the study design, which included tive correlation between the risk of developing glioma and a cases and controls, we deemed an odds ratio to be an appro- history of asthma and hay fever (8). Likewise, Wiemels et al priate measure of relative risk. The incidence of meningiomas demonstrated that patients with allergies were 50% less likely with or without a history of HBV/HCV served as a control to develop gliomas (9). Additionally, certain infections have in order to compare and calculate an odds ratio, and to been demonstrated to induce a similar effect. For example, estimate the relative risk of GBM with a prior hepatitis infec- patients with gliomas are less likely to have had a herpes virus tion. The odds ratio was calculated by dividing the number infection (10). Furthermore, it is also possible that hepatitis of positive HBV/HCV patients in the GBM population (GP) infection, by way of immunomodulation, decreases the risk by the number of negative HBV/HCV patients (GN); this of glioblastoma tumorigenesis independently of the shortened number was then divided by the number of positive HBV/ lifespan associated with the viral illness. HCV patients in the meningioma population (MP) divided The mechanisms underlying these epidemiological obser- by the number of negative HBV/HCV patients (MN). Thus, vations remain unclear. However, animal studies have clearly odds ratio = (GP/GN)/(MP/MN). An odds ratio of 1 implies indicated that natural killer cell activity, macrophage activa- there is an equal probability of HBV/HCV in GBM and tion and interferon production following viral infection are meningioma. An odds ratio of >1 suggests a higher probability increased in the setting of chronic irritation of the immune of having HBV/HCV and GBM compared with meningioma. system, as observed in murine hepatitis infections (11-14). Conversely, an odds ratio of <1 suggests a lower probability of Notably, mice previously infected with murine hepatitis virus having HBV/HCV and GBM versus meningioma. P<0.05 was demonstrated augmented responses to antitumor chemo- considered to indicate a statistically significant result. therapy, resulting in longer survival times (15). Human studies have also revealed that interferon levels are significantly higher Results in patients with HCV compared with healthy controls (16). It is well-known that macrophage activity and endogenous Between 2001 and 2010, 76 ARMC patients were identified interferon levels are key elements in the cancer immunosur- as having GBM, 3 of whom also had positive anti-HCV anti- veillance response (17). Thus, it is reasonable to expect that body serologies and no positive HBV serologies. At the COH, a heightened immune state is able to alter glioma formation between 1998 and 2010, 359 patients had GBM and again and/or progression. only 3 were found to have HBV or HCV. Retrospectively, the Given the epidemiological findings relevant to other overall prevalence of any type of hepatitis virus within this viruses and the molecular basis of the immune response, we combined GBM population was 1.38%, or a rate of 0.0138. hypothesize that patients with GBM have a lower incidence of As a control, the rates of hepatitis infection in patients with prior hepatitis B/C virus infection compared with the overall meningioma were also examined at both institutions. Of the population. Conversely, this finding may be interpreted as 332 meningioma patients, 8 had either active infection or a evidence that patients with a history of HBV/HCV are less history of hepatitis infection, producing a rate of 0.0241, or likely to be diagnosed with GBM. 2.4%, of the meningioma population (Figs. 1 and 2). The odds ratio of having HBV or HCV with GBM compared with HBV Materials and methods or HCV with meningiomas was 0.566 with a 95% confidence interval of 0.194-1.648 (Table I). A χ2 test of association Patients. Patient data were obtained by performing a retro- revealed a Pearson's value of 1.115 (P=0.29) and a one-tailed spective chart review at two different institutions in Southern Fisher's exact probability of P=0.21. California, following IRB approval. The first institution was the City of Hope National Medical Center (COH), a Discussion comprehensive cancer center located in Duarte (CA, USA), and the second was the Arrowhead Regional Medical Center To the best of our knowledge, this is the first investigation of (ARMC) in Colton (CA, USA). At both institutions, a search the correlation between GBM and hepatitis virus infection. of the electronic medical record was performed for all patients The main limitations of our study are two-fold. The first diagnosed with GBM. Data were available from 1998-2009 at limitation is that this was a retrospective analysis. Therefore, the COH and 2001-2010 at the ARMC. Only patients with a patients in this study were not prospectively tested or specifi- ONCOLOGY LETTERS 5: 783-786, 2013 785 Table I. Statistical analysis. Condition and population Rate (%) Test Result 95% confidence interval Hepatitis in GBM population 1.38 Odds ratio 0.566 0.195-1.649 Hepatitis in meningioma 2.41 Fisher's exact probability P=0.21 population (control) In addition to showing the rates or prevalence of hepatitis infection in the two brain tumor populations, Table I demonstrates the risk ratio and odds ratio for having GBM with a previous hepatitis infection, using the meningioma population as a control. The odds ratio is used to estimate relative risk. With an odds ratio of 0.566, this suggests there may be a lower probability of having a previous hepatitis infection with GBM than having a previous hepatitis infection with meningioma, although this difference was not statistically significant. GBM, glioblastoma multiforme. Figure 2. Percentage prevalence of hepatitis infection. The prevalence of hepatitis infection in both patient populations in this study compared with the Figure 1. Patient data. The number of patients in each population of the study, prevalence of hepatitis infection in the US population (22). The GBM popu- and whether or not they were found to have a history of, or an active, hepatitis lation demonstrates the lowest rate of hepatitis infection. Both brain tumor infection. The GBM population was larger than the control meningioma populations have substantially lower hepatitis infection rates when compared population, however fewer patients in the GBM population were found to with the US population during the same time period. GBM, glioblastoma have a hepatitis infection. GBM, glioblastoma multiforme. multiforme. cally asked about hepatitis status at the time of diagnosis of prospective analysis of hepatitis B and C populations. Also, it is GBM or meningioma, and thus the incidence of infection may noteworthy that relative risk in this instance is only applicable be underreported. The second limitation was the small sample if hepatitis infection occurred prior to the diagnosis of GBM size relative to the number of patients who were expected or meningioma. Although not all patient information relevant to have GBM and hepatitis B/C if the two diseases did not to the timing of hepatitis infection was available, it is almost preclude each other. The discordance between the co-diagnosis certain that hepatitis infection occurred prior to GBM, due to of GBM and hepatitis is that patients with hepatitis often have the rapid progression and lethality of the disease. However, this associated medical or social factors that predispose them to assumption may be invalid for meningioma patients whose shorter life spans and thus are less likely to be diagnosed with disease typically occurs over a chronic course. Nonetheless, GBM, a condition associated with increasing age. This may although our findings are not statistically significant, we further explain the lower-than-expected prevalence of HBV/ propose that further, statistically powerful prospective studies HCV in our study populations. are required to investigate the correlation between hepatitis Our study limitations allows us to only calculate an odds infection status and GBM. ratio, which is not necessarily a relative risk, which is the most It may be speculated that currently unidentified factors accurate value for correlational statistics. However, given our underlie the lower incidence of HBV/HCV observed in GBM sample size, our limitation is that we are assuming odds ratio patients and that infection with one of these two viruses is approaching a relative risk if it was able to be determined. provides a form of immunoprotection against the develop- In other words, we are assuming the two values are similar for ment of GBM. One possible mechanism to support this idea this study. Our original hypothesis was that a prior hepatitis B is that patients with a chronic hepatitis infection demon- or C infection would confer resistance to GBM, and thus strated elevated levels of interleukin 13 (IL13). IL13, a potent patients with hepatitis B and C would have a lower relative risk mediator of apoptosis in tumor cells, has been demonstrated of contracting the disease. However, calculation of the relative to be significantly increased in patients with chronic hepatitis risk is flawed by the retrospective nature of the study in which infection and allergies (18,19). High grade gliomas frequently we evaluated GBM populations rather than undertaking a overexpress an IL13 receptor subtype, designated IL13Rα2. 786 CABANNE et al: HEPATITIS B/C IN MENINGIOMA AND GLIOBLASTOMA PATIENTS IL13Rα2 functions as a ‘decoy receptor’ and is ineffective 5. Hsia CC, Scudamore CH, Di Bisceglie AM, et al: Molecular and serological aspects of HBsAg-negative hepatitis B virus in activating the apoptotic cascade upon ligand binding. We infections in North America. J Med Virol 70: 20-26, 2003. speculate that hepatitis infections may increase IL13 levels, 6. Bréchot C: Pathogenesis of hepatitis B virus-related hepatocellular dampening the competitive effects of IL13Rα2 and thereby carcinoma: old and new paradigms. Gastroenterology 127: S56-S61, 2004. inhibiting gliomagenesis. 7. Everhart JE (ed): Digestive Diseases in the United States: Glioma progression is also associated with immuno- Epidemiology and Impact. US Department of Health and Human suppression in both murine models and patients. This state of Services, Public Health Service, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases. immunosuppression is characterized by a diminished capacity US Government Printing Office, Washington, DC, 1994. of macrophages and lymphocytes to secrete interleukins and 8. Schoemaker MJ, Swerdlow AJ, Hepworth SJ, et al: History of tumor necrosis factor α (TNFα) (20,21). Therefore, increased allergies and risk of glioma in adults. Int J Cancer 119: 2165-2172, 2006. circulating levels of TNFα released as a consequence of hepa- 9. Wiemels JL, Wiencke JK, Sison JD, et al: History of allergies titis-induced stimulation of the immune system may diminish among adults with glioma and controls. Int J Cancer 98: 609-615, tumor viability. 2002. 10. Wrensch M, Weinberg A, Wiencke J, et al: Does prior infection In summary, there is limited epidemiological evidence in with varicella-zoster virus influence risk of adult glioma? Am J support of a role for HBV/HCV in GBM tumorigenesis and/or Epidemiol 145: 594-597, 1997. progression. In our study, we observed a lower incidence of 11. Dempsey WL, Smith AL and Morahan PS: Effect of inapparent murine hepatitis virus infections on macrophages and host HBV/HCV in GBM patients compared with the general popu- resistance. J Leukoc Biol 39: 559-565, 1986. lation, suggesting that HCV/HBV may interfere with GBM 12. Schindler L, Engler H and Kirchner H: Activation of natural development. However, there was no statistically significant killer cells and induction of interferon after injection of mouse hepatitis virus type 3 in mice. Infect Immun 35: 869-873, 1982. difference when compared with the incidence of hepatitis in 13. Tamura T, Sakaguchi A, Kai C, et al: Enhanced phagocytic meningioma patients that served as a control group. activity of macrohpages in mouse hepatitis virus-infected nude Identifying factors that are able to influence gliomagenesis mice. Microbiol Immunol 24: 243-247, 1980. 14. Virelizier JL, Virelizier AM and Allison AC: The role of circu- may substantially increase our understanding of GBM biology lating interferon in the modifications of immune responsiveness and enhance the treatment of patients affected by this invariably by mouse hepatitis virus (MHV-3). J Immunol 117: 748-753, 1976. life-threatening disease. We suggest that a future prospective 15. Li LH, DeKoning TF, Nicholas JA, et al: Effect of mouse hepatitis virus infection on combination therapy of P388 leukemia with study following patients with HCV and HBV to assess their cyclophosphamide and pyrimidinones. Lab Anim Sci 37: 41-44, rates of acquired GBM compared with non-hepatitis patients, 1987. in addition to prospectively screening patients with GBM for 16. Kawakami Y, Nabeshima S, Furusyo N, et al: Increased frequency of interferon-gamma-producing peripheral blood CD4+ T cells HCV and HBV serological markers, is required. in chronic hepatitis C virus infection. Am J Gastroenterol 95: 227-232, 2000. Acknowledgements 17. Morahan PS, Edelson PJ and Gass K: Changes in macrophage ectoenzymes associated with anti-tumor activity. J Immunol 125: 1312-1317, 1980. The authors would like to thank Dr Rebecca Nelson of the 18. Inoue M, Kanto T, Miyatake H, et al: Enhanced ability of COH for her assistance with this study, in addition to the staff peripheral invariant natural killer T cells to produce IL-13 in chronic hepatitis C virus infection. J Hepatol 45: 190-196, 2006. and administration of the COH and the ARMC. 19. Li L, Xia Y, Nguyen A, et al: Effects of Th2 cytokines on chemokine expression in the lung: IL-13 potently induces eotaxin References expression by airway epithelial cells. J Immunol 162: 2477-2487, 1999. 20. Kennedy BC, Maier LM, D'Amico R, et al: Dynamics of central 1. Dolecek TA, Propp JM, Stroup NE, et al: CBTRUS Statistical and peripheral immunomodulation in a murine glioma model. Report: Primary Brain and central nervous system tumors BMC Immunol 10: 11, 2009. diagnosed in the United States in 2005-2009. Neuro Oncol 14 21. Hao C, Parney IF, Roa WH, et al: Cytokine and cytokine (suppl 5) 2012. receptor mRNA expression in human glioblastomas: evidence 2. McCarthy BJ, Davis FG, Freels S, et al: Factors associated with of Th1, Th2 and Th3 cytokine dysregulation. Acta Neuropathol survival in patients with meningioma. J Neurosurg 88: 831-839, 103: 171-178, 2002. 1998. 22. McQuillian GM, Kruszon-Moran D, Denniston MM, et al: Viral 3. Kasai Y, Takeda S and Takagi H: Pathogenesis of hepatocellular hepatitis. NCHS Data Brief 1-8, 2010. carcinoma: a review from the viewpoint of molecular analysis. Semin Surg Oncol 12: 155-159, 1996. 4. Zondervan PE, Wink J, Alers JC, et al: Molecular cytogenetic evaluation of virus-associated and non-viral hepatocellular carcinoma: analysis of 26 carcinomas and 12 concurrent dysplasias. J Pathol 192: 207-215, 2000.