Epstein-Barr Virus-Positive Diffuse Large B-Cell Lymphoma: is it different between Over and Under 50 Years of Age?

Lymphoid neoplasm which are derived from B, T or NK cell are among the most common malignancies in the world (Hanif et al., 2009) . Epstein Bar virus (EBV) is a member of the human herpes virus family, which is a DNA virus that was initially detected in a Burkitt’s lymphoma (BL) cell culture during 1964 (Dunleavy et al., 2012). Most people get infected with EBV during childhood or adolescence and it can cause infectious mononucleosis in 35% to 69% of the cases (Adam et al., 2011) .Epstein Barr virus (EBV) infection is associated with many lymphoproliferative disorders, such as Burkitt lymphoma, Plasmablastic lymphoma, NK/Tcell lymphoma, Angioimmunoblastic T cell lymphoma, Hodgkin lymphoma and immunodeficiency-associated lymphoproliferative disorder (Anagnostopoulos et al., 1992; Kanavaros et al., 1996; Hecht et al., 2000; Barrionuevo et al., 2002, Castillo et al., 2008). Diffuse large B cell lymphomas (DLBCLs) are the most common type of lymphomas, accounting for 30Abstract


Introduction
Lymphoid neoplasm which are derived from B, T or NK cell are among the most common malignancies in the world (Hanif et al., 2009) .Epstein Bar virus (EBV) is a member of the human herpes virus family, which is a DNA virus that was initially detected in a Burkitt's lymphoma (BL) cell culture during 1964 (Dunleavy et al., 2012).Most people get infected with EBV during childhood or adolescence and it can cause infectious mononucleosis in 35% to 69% of the cases (Adam et al., 2011) .Epstein Barr virus (EBV) infection is associated with many lymphoproliferative disorders, such as Burkitt lymphoma, Plasmablastic lymphoma, NK/Tcell lymphoma, Angioimmunoblastic T cell lymphoma, Hodgkin lymphoma and immunodeficiency-associated lymphoproliferative disorder (Anagnostopoulos et al., 1992;Kanavaros et al., 1996;Hecht et al., 2000;Barrionuevo et al., 2002, Castillo et al., 2008).
Diffuse large B cell lymphomas (DLBCLs) are the most common type of lymphomas, accounting for 30-
In reports from Asia, patients with EBV+ DLBCL have older age, more advanced stage and more extranodal involvement in comparison with EBV-DLBCL (Park et al., 2007) .Their overall survival rate was worse than EBV-DLBCLs (Park et al., 2007;Dojcinov et al., 2011;Ok et al., 2015).Few studies about lymphomas have been done previously in Yazd, center of Iran in a limited number of cases (Binesh et al., 2013).Only few studies have focused on the EBV status in DLBCL of young adults (Cohen et al., 2014;Hong et al., 2015;Ok et al., 2015;Uccini et al., 2015).In this study we investigate the frequency of EBV infection, immunohistochemical subclasses and clinical data in DLBCL of patients younger and older than 50 years of age.Also we compared the phenotype characteristics and clinical parameters in those who were EBV positive in young and old groups.We attempted to answer two questions: 1-Is EBV+ DLBCL, exclusive to older than 50 years?2-If not, are EBV + DLBCL under 50 phenotypically and clinically different from who are older than 50?

Materials and Methods
In this retrospective study, the medical records of all patients who were diagnosed with DLBCL between 2012-2014, were retrieved from archives of Hematopathology ward in Shahid Fagihi hospital affiliated to Shiraz University of Medical Sciences.Our archive is one of the largest referral centers for hematopathology consultation in Iran which covers mainly the Southern part of the country.During those years all the diagnosed DLBCL had been made according to the latest WHO classification .Diagnosis were based on morphology and immunohistochemistry (IHC) panel (CD3, CD5, CD20, Pax5, CD10, CD30, BCL2, BCL6, MUM1, Ki67) (Figure 1).Cases with prior diagnosis of lymphoma, immunosuppression, and transplantation or inadequate samples as well as clinical data were excluded in this study.From 120 cases of DLBCL, 25 cases were excluded and 95 enrolled in this study.To investigate EBV frequency, IHC for identification of LMP-1 (latent membrane protein, monoclonal antibodies clones CS. 1-4, DAKO) and in situ hybridization for detection of EBV-encoded RNA (EBER: Zyto Vision Kit) was performed (Figure 1E, 1F).EBER is considered positive, when nuclear staining is seen in more than 20% of tumor cells (Park et al., 2007).To classify DLBCLs as germinal center (GC) versus non germinal center (N-GC) types, Hans algorithm by the aid of CD10, BCL6 and MUM1 markers was used.Cases with positive CD10 reaction were considered as GC type (Figure 1D).Also cases with negative CD10, positive BCL6 and negative MUM1 were classified as GC type (Hans et al., 2004).Any other profile was regarded as N-GC.For each patient, EBV status, IHC results (GC vs. Non-GC and CD30 status) and clinical data (age, sex, nodal vs. extranodal involvement) were analyzed.This study is the first report about EBV prevalence in youth DLBCL from Iran.The study was approved by the Ethic Committee of Shiraz University of Medical Sciences.All statistical analysis was performed using the SPSS version 20.Comparison of parameters between DLBCL of young and old adults was performed by Chi square test and a p<0.05 was considered significant.

Results
A total of 95 cases were enrolled in this study.Fifty five of them were older than 50 years and 40 were younger.Mean age was 53.9 for all patients, 37.2 and 66 years for young and old group respectively.Of all 95 cases, 52 were males and 43 were females, 53.6% had extranodal involvement and 27.3% were GC type.CD30 positivity was seen in 15.7% of all.Total numbers of 11 cases (11.6%) were EBV positive (positive EBER reaction).Among them, 8 (72.7%) were LMP positive.
In comparison of young and old group, the prevalence of EBV positivity was 7.5% (3/40) and 14.5% (8/55) respectively (p=0.289,Chi square test).There was no gender predominance (1.12 and 1.29 M/F ratio).Extranodal involvement were seen in 51 cases (53.6%), 18 of them were younger than 50 years and 33 of them were older (p=0.148,Chi square test).Of total 95 cases, 26 (27.3%) were GC type, 11 belonged to the young and 15 to the old group, without any statistical difference (P>0.05,Chi square test).CD30 positivity were seen in 15 cases (15.7%), 7 of them were younger than 50 years and 8 were older (p=0.697,Chi square test) (Table 1).
In EBV+ DLBCL, mean age of young and old group were 41 and 66.25 years respectively.We did not find any significant differences in clinical data, GC type prevalence (33.3% vs. 25%) and CD30 positivity (33.3% vs. 25%) between EBV+ DLBCL of young and old age patients (Table 1).LMP positive reaction was seen more in young patients (100% vs. 62.5%) (P>0.05,Chi square test).

Discussion
DLBCL comprises of different entities with diverse pathogenesis and clinical behavior.EBV-positive DLBCL of the elderly is a new entity in the current WHO classification.It is seen in patients older than 50 years without prior lymphoma and immunodeficiency (Sabattini et al., 2010).It is associated with more advanced stage and extranodal involvement (Park et al., 2007).Mechanism of lymphomagenesis is related to latent membrane protein-1 (LMP-1) and decreased cytotoxic T-cell effector responses (immunosenescence) (Aw et al., 2007).Ishtiaq et al, stated that frequency of EBV in NHL is 12.7% and is mostly seen in diffuse large B cell lymphoma (Ishtiaq et al., 2013).There are a few EBV+DLBCL reports in young adults (Beltran et al., 2011;Ao et al., 2014;Cohen et al., 2014;Hong et al., 2015;Uccini et al., 2015).Hong et al., stated that EBV + DLBCL in young adults is a different entity which has distinct pathogenesis and behavior in compare to EBV+DLBCL of the elderly (Hong et al., 2015).Cohen et al., found that this entity is not restricted to elderly patient and suggested to revise the cutoff age in WHO classification (Cohen et al., 2014).Recently OK et al., in a comprehensive gene expression study concluded that EBV+DLBCL in <50 years versus >50 have similar clinicopathologic, immunophenotypic and genetic features (Ok et al., 2015).Hence, the prevalence of EBV+ DLBCL in our study was 11.6%, which is comparable with the results of Asian and American studies (Kuze et al., 2000;Oyama et al., 2007;Park et al., 2007;Cohen et al., 2014;Hong et al., 2015).In this research, the prevalence of EBV+ DLBCL in cases older than 50 years was higher than younger (14.5% vs. 7.5%), which is close to results of the study from Asia (9.3% vs. 6.7%)(Hong et al., 2015), and another study from Peru (15% vs. 2.2%) (Shimoyama et al., 2009) ,on the contrary to a report from Argentina (8% vs.13%) (Cohen et al., 2014) .Mean age of EBV+ DLBCL in young and old group (41, 66.25 years) is similar to other reports (Beltran et al., 2011;Cohen et al., 2014;Hong et al., 2015).The reason for increased EBV positivity in DLBCL of old patients may be due to defective immune surveillance for EBV which is associated with immunological deterioration as a result of aging process (Kuze et al., 2000;Shimoyama et al., 2008).Most EBV+ DLBCL had extranodal presentation (63.6%) like 2 reports from Far East Asia (Oyama et al., 2007;Shimoyama et al., 2009) but in contrast to some reports (Gibson et al., 2009;Quintanilla-Martinez et al., 2009;Hoeller et al., 2010;Hofscheier et al., 2011;Cohen et al., 2014;Uccini et al., 2015).Surprisingly, the prevalence of GC type DLBCL was very low in our cases (27.3%) and we did not find any significant differences between young and old as well as EBV positive and negative groups.
There has been only one study in this geographical region in which Uccini et al. ( 2015) reported 7 Iraqi children with EBV+ DLBCL, which all of them were Non-GC type.In order to recognize the significance of this difference in incidence of GC vs. Non-GC subtypes, further evaluation is required.We also evaluated the clinical data and IHC findings in EBV+ DLBCL and compare them between young and old group.Clinical data (excluding age) and IHC subclasses were not statistically different between these groups.Our study showed that EBV + DLBCL, is not restricted to older patients, even though the prevalence was higher in this group.Also we did not find any significant differences in IHC subclasses of young and old DLBCL.CD 30 expression was also another issue that we had focused on.CD30 expression was frequently seen in EBV+ DLBCL in comparison to EBV-DLBCL (27.2% vs.14.2%) which is similar two other reports (42% vs.14% and 28% vs. 6.4%)(Hu et al., 2013;Lu et al., 2015).Strong CD30 positivity in large mature B cells has been associated with higher frequency of EBV infection in tumor cells (Hu et al., 2013) .Our study showed that even though CD30 reaction is predictor of EBV infection, it is neither specific nor sensitive.Also morphology is not very useful in leading us toward a survey for EBV infection.So it is still a matter of debate which DLBCL should be more assessed by EBER.
In conclusion, EBV+DLBCL are not restricted to old age.Also we did not find any significant difference between EBV+DLBCL of young and old age patient.We suggest that in the next revision of WHO classification, age cutoff should be removed.

Figure 1 .
Figure 1.Different immunohistochemical reactions of CD20, pax5, CD30, CD10, EBER and LMP.A) Strong and diffuse positive reaction of CD20 in an EBV+DLBCL of 62 year old man.B) Strong nuclear staining of pax5 in an EBV+DLBCL of 43 year old man.C) Scattered expression of CD30 in large cells of 57 year old man with EBV+DLBCL.D) Diffuse expression of CD10, indicative of GC type subclasses in a 37 year old woman.E) Diffuse and strong nuclear staining for EBER in the same case as B. F) LMP in a 37 year old man with EBV+DLBCL