Differentiating Burkitt Lymphoma from Burkitt-like Lymphoma in Pediatrics: Report from the Japanese Pediatric Leukemia/Lymphoma Study Group B-NHL03 Study

Introduction: Pediatric mature B-cell non-Hodgkin lymphoma (B-NHL) is mostly classified as Burkitt lymphoma (BL) and diffuse large B-cell lymphoma (DLBCL). It has a predominantly germinal center origin. In pediatric BL, some cases are difficult to distinguish from DLBCL, which has been referred to as atypical BL or Burkitt-like lymphoma (BLL). In the recent World Health Organization classification of 2008, B-cell lymphoma, unclassifiable, with features intermediate between DLBCL and BL (IDB) was defined as a new but heterogeneous category. This group has morphological and genetic features of both DLBCL and BL. Therefore, IDB includes atypical BL or BLL in children. To the best of our knowledge, little is known about the exact difference between pediatric BL and IDB, and a distinction needs to be made between the clinical aspects of these two groups. In this study, we performed a clinicopathological study using the cohort from Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG) B-NHL03 study, which was the nation-wide prospective study for pediatric B-NHL in Japan, to understand the meaning of the classification of BL in children.

Methods: From November 2004 to January 2011, 346 cases of newly diagnosed pediatric B-NHL were enrolled in the B-NHL03 study. Of these, 208 cases were diagnosed in the central pathology review system with reference laboratory results. Seventy-eight cases were subclassified as BL (37.5%), 42 as IDB (20.2%), 57 as DLBCL (27.4%), and 31 as not subclassified (14.9%). Finally, 120 cases of BL or IDB were included in our study cohort. The pathological features and diagnostic and prognostic factors of those cases were analyzed.

Results: The median age was 9.0 years (range, 1.9–15.5) in BL patients and 9.1 years (3.0–15.8) in IDB patients. Male (M) predominance was observed in both groups and the M/female (F) sex ratio was 5.5 (M/F 66/12) in BL and 9.5 (38/4) in IDB cases. According to the Murphy staging system, 7.7%/24.4%/39.7%/28.2% of BL patients and 14.3%/40.5%/26.2%/19.0% IDB patients were at stage I/II/III/IV, respectively. The frequency of the cases with central nervous system diseases and cerebrospinal fluid (CSF) blasts were 10.3% and 5.1% in BL and 7.1% and 2.4% in IDB, respectively. In addition, the number of the cases with elevated lactate dehydrogenase (LDH) levels (more than twice the institutional upper limit) was 43.6% for BL and 33.3% for IBD. 4-year overall survival and event-free survival (EFS) were 93.6% and 91.0% for BL and 95.2% and 85.7% for IBD, respectively. In general, there is little statistical difference in clinical presentation between both groups. Pathological examination showed that BL has a significantly high frequency of C-myc translocation (57.1% vs. 33.3%, p = 0.03). Immunohistochemistry demonstrated that most cases in both groups were CD10 positive (BL 97.4%, IBD 95.1%), Bcl-6 positive (both groups 100%), and overexpressed Ki67 (≥90%; BL 93.6%, IBD 92.9%). Moreover, 11.1% of BL and 5.7% of IBD patients were Bcl-2 positive. Because a large difference was not observed in the clinicopathological presentation of the two patient groups, we combined both to analyze prognostic factors. To determine the effects of various risk factors of EFS, log-rank testing with stratification for age (≥10 or <10), sex (M or F), stage (I & II or III & IV), CSF blasts level (±), LDH level (more than or less than twice the institutional upper limit), C-myc translocation(±), immunohistochemical staining of CD10 (±), Bcl-2 (±), Bcl-6 (±), and Ki67 (≥ or <90%) were examined. High staging (83.3% vs. 97.9%, p = 0.012), CSF blast+ levels (40.0% vs. 91.3%, p = 0.001), and high LDH levels (77.1% vs. 97.2%, p = 0.001) significantly correlated with inferior outcome.

Conclusions: We conclude that in our cohort, the distinction between BL and IDB does not have much meaning in B-NHL in children. Furthermore, our study shows that the prognostic factors in childhood B-NHL identified in clinical trials in Western countries are identified in patients with the broad BL in Japan.