Abstract. Hematopathology / HIGH KI-67 INDEX AND STARRY-SKY PATTERN IN BL AND DLBCL

Transcription

1 Hematopathology / HIGH KI-67 INDEX AND STARRY-SKY PATTERN IN BL AND DLBCL Histopathology and Immunohistochemistry in Distinguishing Burkitt Lymphoma From Diffuse Large B-Cell Lymphoma With Very High Proliferation Index and With or Without a Starry-Sky Pattern A Comparative Study With EBER and FISH Shih-Sung Chuang, MD, 1,2 Hongtao Ye, MD, PhD, 3 Ming-Qing Du, MD, PhD, 3 Chin-Li Lu, MS, 4 Ahmet Dogan, MD, PhD, 5 Pin-Pen Hsieh, MD, 6 Wan-Ting Huang, MD, 7 and Yun-Chih Jung, MD 8 Key Words: Burkitt lymphoma; Diffuse large B-cell lymphoma; Epstein-Barr virus; Fluorescence in situ hybridization; Taiwan Abstract Burkitt lymphoma (BL) is characterized by c-myc translocation and CD10+/bcl-2 /bcl-6+ with a very high Ki-67 proliferation index (PI). Occasional diffuse large B-cell lymphomas may exhibit a very high PI with or without a starry-sky pattern (DLBCL-HPSS). We compared 28 consecutive BL and 16 DLBCL-HPSS cases in immunocompetent Taiwanese diagnosed by histopathologic examination and immunophenotyping and compared the results with results for Epstein-Barr virus encoded messenger RNA (EBER) and fluorescence in situ hybridization (FISH). There were statistically significant differences in the expression of CD10 (28/28 vs 1/16), bcl-2 (3/28 vs 11/16), MUM1 (5/28 vs 15/16), a PI of 95.0% or more (27/28 vs 2/16), and combined CD10+/bcl-2 /bcl-6+ (24/28 vs 1/16) between BLs and DLBCL-HPSSs. Of the BLs, 7 (25%) of 28 and 26 (96%) of 27 were positive for EBER and c- myc rearrangement as compared with 0 of 16 and 1 (7%) of 15 DLBCL-HPSSs, respectively. We can confidently distinguish BL from DLBCL-HPSS by using histopathologic and immunohistochemical (CD10, bcl- 2, bcl-6, Ki-67) methods without the aid of EBER and FISH in the great majority of cases. Burkitt lymphoma (BL) is defined by the new World Health Organization (WHO) classification as a highly aggressive non-hodgkin lymphoma (NHL) composed of monomorphic medium-sized B cells with basophilic cytoplasm and numerous mitotic figures. 1 The tumor cells exhibit a high mitotic rate and a high degree of apoptosis with diffuse sheets and starry-sky pattern imparted by numerous benign histiocytes with phagocytosis of apoptotic cellular debris. Morphologically, in addition to classical BL, there are 2 cytologic variants BL with plasmacytoid differentiation and atypical Burkitt/Burkitt-like lymphoma (abl/bll), with the latter showing greater pleomorphism in nuclear size and shape than the classical cases. Immunophenotypically, the tumors of all cytologic variants express membranous IgM, CD10, and bcl-6 but not IgD, bcl-2, or terminal deoxynucleotidyl transferase (TdT). Cytogenetically, all cases have a translocation involving the myc gene at 8q24 with the immunoglobulin heavy chain gene (IGH) on 14q32 or less commonly with the κ light chain locus (IGK) at 2q11 or λ light chain locus (IGL) at 22q11. The immunophenotypic prototype of BL is IgM+/CD10+/bcl-2 /bcl-6+ with the Ki-67 proliferation index (PI) nearly at 100%; however, cases with an aberrant immunophenotype such as bcl-2 expression exist. 2 Some diffuse large B-cell lymphomas (DLBCLs) may exhibit a very high PI and starry-sky growth pattern or have medium-sized tumor cells showing slight nuclear pleomorphism mimicking those of classical BL or abl/bll. In addition to morphologic features, DLBCL may share immunophenotypic features of BL. 3 Furthermore, c-myc rearrangement is not unique in BL and may occur in DLBCLs and impart Burkitt-like morphologic features to these tumors. 2,4-7 These gray-zone cases pose 558 Am J Clin Pathol 2007;128: Downloaded 558 from

2 Hematopathology / ORIGINAL ARTICLE a great diagnostic challenge, while the distinction is of great clinical significance because the treatment strategies for BL and DLBCL are different In this study, we compared the results of immunohistochemical analysis with in situ hybridization for Epstein-Barr virus (EBV)-encoded messenger RNA (EBER) and fluorescence in situ hybridization (FISH) among BLs and DLBCLs with very high PI and with or without starry-sky growth pattern (HPSS). Materials and Methods A total of 28 consecutive cases of BL were obtained retrospectively from 4 hospitals in southern Taiwan and the consultation files of one of us (S.-S.C.). A cohort of 16 DLBCL- HPSS tumors with a PI of 90% or more and with (14 tumors) or without (2 tumors) a starry-sky pattern were randomly selected from the pathology files of Chi-Mei Medical Center, Tainan, Taiwan. Of the DLBCL-HPSS tumors, 14 were de novo, and 2 were transformed tumors. The Chi-Mei Institutional Review Board approved the study. The specimens from all cases were fixed in 10% formalin, processed by routine methods, and embedded in paraffin. All H&E-stained sections were reviewed by 2 of us (S.-S.C. and A.D.). An immunohistochemical study was performed for each case using the labeled streptavidin-biotin peroxidase method (LSAB kit, DAKO, Carpinteria, CA), and an antigenretrieval technique was applied as needed for each specific antibody. The antibodies used were CD3, CD20, CD45, bcl-2, bcl-6, Ki-67 (MIB-1), IgD, IgM, MUM1, and TdT (DAKO); and HHV-8 latency-associated nuclear antigen (HHV8- LANA) and CD10 (Novocastra, Newcastle upon Tyne, England). For the Ki-67 PI, 500 cells were counted in each case under high-power ( 400) with the aid of a grid. EBER was performed using polymerase chain reaction derived, digoxigenin-labeled specific DNA probes, as previously described, with nasopharyngeal carcinoma as the positive control samples in each experiment. 12 Locus-specific interphase FISH was performed on paraffin-embedded tissue sections without knowledge of the histopathologic diagnosis. Briefly, deparaffinized sections were pretreated by pressure cooking (3 to 3.5 minutes) in EDTA buffer (1 mmol/l; ph 8.0) and subsequent incubation in pepsin solution (25 minutes) at 37 C to increase DNA accessibility. Sections were then fixed in 1% paraformaldehyde for 2 minutes, dehydrated through increasing ethanol series, and air dried. The appropriate probe mix (1.5 µl) was applied to the tissue section and covered with a round, 10-mm coverslip. The probe and target DNA were simultaneously denatured at 80 C for 25 minutes and incubated up to 2 days at 45 C. Posthybridization washes were performed according to the rapid-wash protocol (Vysis, Downers Grove, IL). Sections were mounted in Vectashield antifade solution with 4,6- diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA). Image acquisition and processing were performed as previously described. 13 Five probes, including C-MYC, IGH, and BCL-6 dual-color break-apart rearrangement probes and IGH/C-MYC and IGH/BCL-2 dual-fusion translocation probes (Vysis/Abbott Laboratories, Maidenhead, England) were used. Break-apart rearrangement probes for IGK and IGL, as previously described, were applied in selected cases. 14 Categorical data were compared by using the Fisher exact test. A P value less than 5% was considered statistically significant. All statistical analyses were performed using SPSS software for Windows, version 12.0 (SPSS, Chicago, IL). Results Clinical Features The demographic data for patients with BL are summarized in Table 1. There were 16 pediatric (57%) and 12 adult patients with median ages of 6 years (range, <1-16 years) and 45 years (18-71 years), respectively. In the pediatric group, there were 10 boys and 6 girls; in the adult group, there were 6 men and 6 women. Of the 28 tumors, 7 were nodal and 21 were extranodal, with the gastrointestinal tract the most common organ of involvement (8/28 [29%]). Of the 16 patients with DLBCL-HPSS, there was an equal number of both sexes with a median age of 66 years (range, years); 5 tumors were nodal. All patients with BL and DLBCL-HPSS were immunocompetent Taiwanese. Morphologic Features Under low-power examination, all BLs showed a diffuse lymphoid infiltrate and all but 1 case exhibited a starry-sky pattern. Under high-power, the cell size in most tumors was uniformly medium with round nuclei and a few small nucleoli, typical features of BL Image 1. However, a mild degree of irregular nuclear contours was noted in some tumors, and it was difficult to reach a consensus diagnosis among authors as BL or abl/bll. With a typical immunophenotype of BL (CD10+/bcl-2 /bcl-6+ and a high PI), we chose to classify them as BL. The 16 DLBCL-HPSSs were randomly selected based on a very high PI. All tumors exhibited a diffuse infiltrate. In addition, a starry-sky pattern was present in all but 2 tumors (DLBCL-HPSS cases 4 and 9) Image 2. Of the DLBCLs, 14 were de novo; the remaining 2 cases were transformed from a primary gastric grade 3b follicular lymphoma (DLBCL-HPSS case 2) and a mucosa-associated lymphoid tissue lymphoma of the salivary gland (DLBCL-HPSS case 14). Downloaded from Am J Clin Pathol 2007;128:

3 Chuang et al / HIGH KI-67 INDEX AND STARRY-SKY PATTERN IN BL AND DLBCL Table 1 Immunohistochemical, EBER, and FISH Results in Burkitt Lymphomas Immunohistochemical Analysis ISH FISH Case No./ Ki-67 IGH/C- IGH/ Trans- Sex/Age (y) Organ CD10 bcl-2 bcl-6 MUM1 IgD IgM (%) EBER C-MYC IGH MYC BCL-2 BCL-6 location 1/M/5 Lymph node Unknown * 2/M/12 Ileum (8;14) 3/F/12 Lymph node (8;14) 4/F/42 Uterine cervix (8;14) 5/F/11 Spine (8;14) 6/M/18 Cecum (8;22) 7/F/14 Salivary gland (8;14) 8/F/30 Lymph node (8;14) 9/F/4 Lymph node (8;14) 10/M/3 Duodenum (8;14) 11/F/<1 Retroperitoneum (8;14) 12/M/3 Paranasal sinus (8;14) 13/F/6 Parapharynx (8;22) 14/M/3 Cecum (8;14) 15/M/29 Stomach (8;14) 16/M/6 Soft palate (8;14) 17/M/9 Nasopharynx (8;14) 18/F/49 Bone marrow Failed Failed ND Failed Failed Failed 19/M/4 Lymph node (8;14) 20/M/3 Ileocecum (8;14) 21/F/71 Lymph node (8;14) 22/M/65 Perineum (8;14) 23/F/67 Lymph node ND ND (8;14) 24/M/21 Retroperitoneum ND + (8;14) 25/F/64 Thyroid ND + (8;14) 26/M/28 Ileum ND ND (8;14) 27/M/16 Ileocecum ND ND (8;14) 28/M/48 Tonsil Unknown Positive/ 28/28 3/28 27/28 5/28 4/28 25/28 7/28 26/27 22/25 23/27 0/24 0/24 total (%) (100) (11) (96) (18) (14) (89) (25) (96) (88) (85) (0) (0) BCL-6, LSI BCL-6 dual-color, break-apart rearrangement probe; C-MYC, C-MYC break-apart rearrangement probe; EBER, Epstein-Barr virus encoded messenger RNA; IGH, IGH break-apart rearrangement probe; IGH/BCL-2, IGH/BCL-2 dual-fusion translocation probe; IGH/C-MYC, IGH/C-MYC dual-fusion translocation probe; FISH, fluorescence ISH; ISH, in situ hybridization; ND, not done. * Positive for a c-myc rearrangement but negative for breakage in immunoglobulin κ (IGK) and λ (IGL) gene loci using specific break-apart rearrangement probes. Positive for c-myc and IGL but not IGK rearrangement, suggesting translocation involving t(8;22)(q24;q11). Immunohistochemical Analysis The immunohistochemical results are summarized in Table 1. All 28 BL cases expressed CD10 and CD20. The majority of the tumors expressed bcl-6 (27/28 [96%]) and IgM (25/28 [89%]). MUM1, IgD, and bcl-2 were expressed by 5 (18%), 4 (14%), and 3 (11%) tumors, respectively. The DLBCL-HPSS tumors expressed CD20 (16/16 [100%]), MUM1 (15/16 [94%]), bcl-6 (14/16 [88%]), bcl-2 (11/16 [69%]), IgM (7/16 [44%]), IgD (1/15 [7%]), and CD10 (1/16 [6%]). The DLBCL-HPSS case 2 tumor was the only one expressing CD10. In addition, this tumor expressed bcl-6 and MUM1 but not bcl-2 and had a PI of 92.0%. The other case of germinal center cell origin, as defined by Hans et al, 15 was DLBCL-HPSS case 10 with bcl-6 expression but not CD10, bcl-2, or MUM1. All BL and DLBCL-HPSS tumors were negative for CD3, TdT, and HHV8-LANA. In Situ Hybridization for EBER Of the BL tumors, 7 (25%) were positive for EBER with the proportion of positive tumor cells ranging from 10% to 70% (mean, 46%). All 16 DLBCL-HPSS tumors were negative for EBER, with almost no single positive cells. Fluorescence In Situ Hybridization The FISH study was successful in all but 1 case each of BL and DLBCL-HPSS tumors, and the results for BL cases are given in Table 1. The c-myc translocation was identified in 26 (96%) of 27 BL compared with 1 (7%) of 15 DLBCL- HPSS tumors. Of the 26 BLs showing signal split with C- MYC break-apart FISH probes, 23 (88%) were positive for IGH/C-MYC fusion FISH signals, indicating t(8;14)(q24;q32); 2 (8%) (BL cases 6 and 13) were positive for C-MYC and IGL break-apart probe signals, suggesting t(8;22)(q24;q11); the remaining tumor (BL case 1) was negative for breakage in the IGH, IGK, IGL, and BCL-6 loci and negative for the IGH/BCL-2 dual-fusion translocation probes. We were unable to identify the partner chromosome associated with c-myc rearrangement in the last tumor. IGH/BCL-2 and BCL-6 rearrangements were negative in all 560 Am J Clin Pathol 2007;128: Downloaded 560 from

4 Hematopathology / ORIGINAL ARTICLE A C B CD10 bcl-2 bcl-6 Ki-67 D Image 1 (Case 22) Burkitt lymphoma shows a starry-sky growth pattern (A, 400) with expression of CD10 and bcl-6 and a proliferation index by Ki-67 immunostaining of 97.8%; the tumor cells do not express bcl-2 (B, 400). C, The neoplastic cells are positive for Epstein-Barr virus encoded messenger RNA ( 400). D, Fluorescence in situ hybridization study using a C-MYC break-apart rearrangement probe (Vysis/Abbott, Maidenhead, England) shows dissociation of the red and green signals indicating the presence of a chromosomal breakpoint in the c-myc locus ( 1,000). 24 successfully tested cases of BL, including BL case 28 that was negative for c-myc rearrangement but positive for IGH rearrangement. The 1 DLBCL-HPSS tumor (Case 13) with c-myc rearrangement was positive for IGH, IGH/C-MYC, and BCL-6 rearrangement probes but negative for IGH/BCL2, indicating t(8;14) with additional BCL-6 rearrangement. Of the successfully studied cases, IGH and BCL-6 rearrangements were identified in 5 (42%) of 12 and 2 (13%) of 15 DLBCL-HPSS tumors. Not a single tumor in the 15 successfully tested cases of DLBCL-HPSS was positive for IGH/BCL-2. Statistics The comparison of immunohistochemical findings between BL and DLBCL-HPSS is summarized in Table 2. There were significant differences in the expression of CD10, bcl-2, MUM1, and IgM between BL and DLBCL-HPSS tumors. The difference in expression of bcl-6 and IgD was not statistically significant. The typical immunophenotype of CD10+/bcl-2 /bcl-6+ was present in 24 (86%) of 28 BLs compared with 1 (6%) of 16 DLBCL-HPSS tumors (P <.001). The median PIs were 97.0% and 92.2% for BL and DLBCL-HPSS, respectively, with a significant difference in a PI of 95.0% or more (Table 2). Am J Clin Pathol 2007;128: Downloaded from

5 Chuang et al / HIGH KI-67 INDEX AND STARRY-SKY PATTERN IN BL AND DLBCL A B CD10 bcl-2 bcl-6 Ki-67 Image 2 (Case 13) Diffuse large B-cell lymphoma with high proliferation fraction shows a starry-sky growth pattern (A, 400) with expression of bcl-2 but not CD10 or bcl-6; the proliferation index by Ki-67 immunostaining is 93.0% (B, 400). This tumor is negative for c-myc translocation (data not shown). Table 2 Comparison of Immunohistochemical Findings in BL and DLBCL-HPSS* BL (n = 28) DLBCL-HPSS (n = 16) P CD10 bcl-2 bcl-6 CD10+/ bcl-2 /bcl-6+ MUM1 IgD IgM Ki % 28 (100) 1 (6) 3 (11) 11 (69) 27 (96) 14 (88) (86) 1 (6) 5 (18) 15 (94) 4 (14) 1 (7) (89) 7 (44) (96) 2 (13) BL, Burkitt lymphoma; DLBCL-HPSS, diffuse large B-cell lymphoma with high proliferation index and with or without starry-sky growth pattern. * Data are given as number (percentage). n = 15. Fisher exact test. The comparison of EBER and FISH results between BL and DLBCL-HPSS is summarized in Table 3. Of 28 cases of BL, 7 (25%) were positive for EBER, whereas all 16 DLBCLHPSS tumors were negative (P =.037). The c-myc translocation was identified in 26 (96%) of 27 BLs in contrast with 1 (7%) of 15 DLBCL-HPSS tumors (P <.001). Discussion We have shown that with morphologic and immunohistochemical (CD10, bcl-2, bcl-6, and Ki-67) studies, and with an additional benefit if the EBER result was positive, we were able to distinguish BL from DLBCL-HPSS without doing FISH in the majority of cases. This is of great importance to practicing pathologists because FISH is not available in most diagnostic laboratories. Furthermore, some of the BL tumors in our series showed a mild degree of nuclear pleomorphism and slightly irregular nuclear contours, corresponding to abl/bll by the WHO definition; however, their cytogenetic 562 Am J Clin Pathol 2007;128: Downloaded 562 from Table 3 Comparison of EBER and c-myc Translocation of BL and DLBCL-HPSS* BL DLBCL-HPSS P EBER c-myc 7/28 (25) 0/16 (0) /27 (96) 1/15 (7) BL, Burkitt lymphoma; DLBCL-HPSS, diffuse large B-cell lymphoma with high proliferation index and with or without starry-sky growth pattern; EBER, EpsteinBarr virus encoded messenger RNA. * Data are given as number/total (percentage). Fisher exact test. features were homogeneous with c-myc translocation and absence of IGH/BCL-2 or BCL-6 rearrangement. Our results indicate that immunophenotyping is superior to morphologic examination for separating BL and abl/bll from their morphologic mimic, DLBCL-HPSS. Haralambieva et al3 reported findings similar to ours. In an analysis of 10 reference pediatric BLs and 74 adult lymphomas with morphologic

6 Hematopathology / ORIGINAL ARTICLE features of BL, they found that BL, as defined by biologic markers (Ki-67 PI of 90%, CD10+, bcl-2, bcl-6+, myc breakpoint+, and BCL-2 and BCL-6 breakpoint ) was more homogeneous than as defined by a 2-step review by 4 hematopathologists. They suggested that immunohistochemical analysis for Ki-67, CD10, and bcl-2 with analysis of myc and preferably also BCL-2 and BCL-6 may be advised as a marker panel for this diagnostic dilemma. However, some DLBCLs may have a very high PI, such as DLBCL-HPSS with a PI of 90% or more in our study. We demonstrated that a PI of 95.0% or more was a powerful parameter to distinguish BL (including abl/bll) from DLBCL-HPSS. abl/bll is defined by the 2001 WHO classification as a variant of BL with features of BL but exhibits greater pleomorphism in nuclear size and shape, and the nucleoli are more prominent and fewer. 1 Braziel et al 16 studied the phenotypic distinction in 13 patients with BLL and 55 patients with DLBCL by immunohistochemical analysis using an extensive antibody panel. They found that compared with DLBCL, BLL had a higher PI (mean Ki-67, 88% vs 53%), greater expression of CD10 and p53 antigens, decreased expression of bcl-2, and a shorter median survival (1.2 vs 2.5 years). They concluded that BLL could be recognized by its combined morphologic and phenotypic features and that it represented a high-grade lymphoma much closer to BL than to DLBCL. Retention of the BLL category or inclusion of BLL as a variant of BL is biologically and clinically more appropriate than absorbing the category of BLL into DLBCL. In a recent study by McClure et al 17 on a series of 31 BLLs, they found that even after independent and joint review by 4 pathologists, it was extremely difficult to separate many adult lymphomas into the morphologic categories defined by the WHO for BL and abl/bll because of interobserver variability, tumor heterogeneity, and tissue fixation variability. It seems to be rational to group BL and abl/bll together as we had done in our study because these tumors exhibit homogeneous immunophenotypic and cytogenetic features, although there is a mild degree of morphologic variation. EBV infects the vast majority of adults worldwide and establishes nonproductive (latent) and productive (lytic) infections. Latent EBV infection is associated with various hematologic malignancies in immunocompetent patients, including BL, nasal natural killer/t-cell lymphoma in Asia and Central and South America, and immunodeficiency-associated lymphoproliferative disorders. By using the Kiel classification in a series of European B-cell NHLs in immunocompetent patients, Hummel et al 18 identified the EBER transcript mainly in high-grade lymphomas and occasionally in low-grade lymphomas but not in any centroblastic (39 cases) or immunoblastic (28 cases) lymphomas. In immunocompetent patients with B-cell NHL in Taiwan, EBER was identified mainly in BL (7/28 [25%]) and occasionally in DLBCL (2/39 [5%]) (S.-S.C., unpublished data, 2007). With immunoglobulin gene analysis, Bellan et al 19 recently revealed 2 distinct cells of origin for EBV+ and EBV BL. The EBV BLs exhibited a lower average mutation frequency, indicating their origin of early centroblasts for sporadic BL, whereas EBV+ cases had a much higher mutation frequency, corresponding to late germinal center B cells that have begun the differentiation process into memory B cells for endemic and AIDS-related BL. In developed countries such as Japan, the association of EBV with BL is low, ranging from 0% to 8%, 2,20 whereas the association in developing countries is intermediate between the sporadic and endemic types. For example, 47% of cases from Argentina and 80% of cases from India were EBER+. 21 In Israel, 11 (34%) of 32 childhood sporadic BL cases were positive for EBER with no prognostic significance. 22 In Taiwan, Chao et al 23 reported the presence of EBER+ signals in 10 (56%) of 18 BLs, and the EBER+ tumors were frequently associated with extra-abdominal but rarely with intraabdominal localization. Although the EBER+ rate (7/28 [25%]) in BLs in our study was lower than that reported by Chao et al, 23 5 of our 7 EBER+ tumors were located in peripheral lymph nodes, confirming their finding of frequent extraabdominal localization of EBER+ BL in Taiwan. The myc gene rearrangement has been reported in 6% to 10% of DLBCL cases with conflicting prognostic significance. 4,6,7 In the study by Kramer et al, 4 DLBCLs with myc rearrangement were more prevalent in extranodal than nodal lymphomas (16% vs 2%), particularly in the gastrointestinal tract, and carried a trend for better disease-free survival. On the contrary, other studies revealed that DLBCLs with myc rearrangement were aggressive and carried a poor prognosis. 6,7,24 One of our 15 cases of DLBCL-HPSS exhibited c-myc rearrangement and c-myc/igh translocation in addition to BCL- 6 rearrangement. This patient had extensive intra-abdominal and retroperitoneal lymphomatosis and died of sepsis shortly after diagnosis. Although the number of cases of DLBCL-HPSS, a morphologic mimic of BL, in our study was small, it seemed that the relative frequency of cases with c-myc rearrangement was not higher than ordinary or usual DLBCL. We are planning to do a large cohort of study for c-myc rearrangement in DLBCL- HPSS in Taiwan to see if there is a higher frequency of such genetic alteration in this group of tumors. In the present study, we identified t(8;14) by FISH in 23 (85%) of 27 BLs and t(8;22) in 2 tumors (7%); the partner chromosome in 1 tumor with c-myc rearrangement was unidentified, and the remaining case was negative for c-myc rearrangement. The relative frequency of various partner chromosomes for c-myc rearrangement in our series was similar to previous studies. Namiki et al 20 found that of 40 Japanese patients with BL examined by cytogenetic and/or FISH or Southern blot analysis using MYC probes, 35 tumors (88%) had t(8;14) translocations and 5 (13%) had t(8;22) without any Downloaded from Am J Clin Pathol 2007;128:

7 Chuang et al / HIGH KI-67 INDEX AND STARRY-SKY PATTERN IN BL AND DLBCL case of t(2;8). By using a long polymerase chain reaction method, Burmeister et al 25 recently revealed molecular heterogeneity of 56 sporadic BL cases with 29 tumors (52%) positive for t(8;14) and rare cases with variant myc translocations, including t(8;22), t(2;8), and t(8;9) and 1 with an unidentified partner chromosome. Our results suggest that we can confidently distinguish BL from DLBCL-HPSS by using histopathologic and immunohistochemical (CD10, bcl-2, bcl-6, Ki-67 PI 95.0%) studies without the aid of EBER and FISH in the great majority of cases. From the Departments of Pathology, 1 Chi-Mei Medical Center, Tainan, Taiwan; 2 Taipei Medical University, Taipei; 6 Veterans General Hospital-Kaohsiung, Kaohsiung, Taiwan; 7 Chang Gung Memorial Hospital Kaohsiung Medical Center, Chang Gung University College of Medicine, Kaohsiung; 8 Sin-Lau Christian Hospital, Tainan; 4 Department of Medical Research, Chi-Mei Medical Center, Tainan; 3 Department of Pathology, Cambridge University, Cambridge, England; and 5 Department of Pathology and Laboratory Medicine, Mayo Clinic, Rochester, MN. Supported by research grant NSC b from the National Science Council, Taipei, Taiwan. Presented at the XIII meeting of the European Association for Haematopathology; October 7-12, 2006; Vienna, Austria. Address reprint requests to Dr Chuang: Dept of Pathology, Chi-Mei Medical Center, 901 Chung-Hwa Rd, Yung-Kang City, Tainan County 710, Taiwan. Acknowledgment: We thank P.G. Isaacson, MD, PhD, for consultation of some study cases. References 1. Diebold J, Jaffe ES, Raphael M, et al. Burkitt lymphoma. In: Jaffe ES, Harris NL, Stein H, et al, eds. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2001: Nakamura N, Nakamine H, Tamaru J, et al. The distinction between Burkitt lymphoma and diffuse large B-cell lymphoma with c-myc rearrangement. Mod Pathol. 2002;15: Haralambieva E, Boerma EJ, van Imhoff GW, et al. Clinical, immunophenotypic, and genetic analysis of adult lymphomas with morphologic features of Burkitt lymphoma. Am J Surg Pathol. 2005;29: Kramer MHH, Hermans J, Wijburg E, et al. Clinical relevance of BCL2, BCL6, and MYC rearrangements in diffuse large B- cell lymphoma. Blood. 1998;92: Barth TF, Muller S, Pawlita M, et al. Homogeneous immunophenotype and paucity of secondary genomic aberrations are distinctive features of endemic but not of sporadic Burkitt s lymphoma and diffuse large B-cell lymphoma with MYC rearrangement. J Pathol. 2004;203: Vitolo U, Gaidano G, Botto B, et al. Rearrangements of bcl-6, bcl-2, c-myc and 6q deletion in B-diffuse large-cell lymphoma: clinical relevance in 71 patients. Ann Oncol. 1998;9: Kawasaki C, Ohshim K, Suzumiya J, et al. Rearrangements of bcl-1, bcl-2, bcl-6, and c-myc in diffuse large B-cell lymphomas. Leuk Lymphoma. 2001;42: Schiffer CA. Treatment of high-grade lymphoid malignancies in adults. Semin Hematol. 2001;38(4 suppl 10): Rosenwald A, Wright G, Chan WC, et al. The use of molecular profiling to predict survival after chemotherapy for diffuse large- B-cell lymphoma. N Engl J Med. 2002;346: Shipp MA, Ross KN, Tamayo P, et al. Diffuse large B-cell lymphoma outcome prediction by gene-expression profiling and supervised machine learning. Nat Med. 2002;8: Blum KA, Lozanski G, Byrd JC. Adult Burkitt leukemia and lymphoma. Blood. 2004;104: Tsai ST, Jin YT, Wu TC. Synthesis of PCR-derived, digoxigenin-labeled DNA probes for in situ detection of Epstein-Barr early RNAs in Epstein-Barr virus infected cells. J Virol Methods. 1995;54: Martin-Subero JI, Chudoba I, Harder L, et al. Multicolor- FICTION: expanding the possibilities of combined morphologic, immunophenotypic, and genetic single cell analyses. Am J Pathol. 2002;161: Martin-Subero JI, Harder L, Gesk S, et al. Interphase FISH assays for the detection of translocations with breakpoints in immunoglobulin light chain loci. Int J Cancer. 2002;98: Hans CP, Weisenburger DD, Greiner TC, et al. Confirmation of the molecular classification of diffuse large B-cell lymphoma by immunohistochemistry using a tissue microarray. Blood. 2004;103: Braziel RM, Arber DA, Slovak ML, et al. The Burkitt-like lymphomas: a Southwest Oncology Group study delineating phenotypic, genotypic, and clinical features. Blood. 2001;97: McClure RF, Remstein ED, Macon WR, et al. Adult B-cell lymphomas with Burkitt-like morphology are phenotypically and genotypically heterogeneous with aggressive clinical behavior. Am J Surg Pathol. 2005;29: Hummel M, Anagnostopoulos I, Korbjuhn P, et al. Epstein- Barr virus in B-cell non-hodgkin s lymphomas: unexpected infection patterns and different infection incidence in lowand high-grade types. J Pathol. 1995;175: Bellan C, Lazzi S, Hummel M, et al. Immunoglobulin gene analysis reveals 2 distinct cells of origin for EBV-positive and EBV-negative Burkitt lymphomas. Blood. 2005;106: Namiki T, Sakashita A, Kobayashi H, et al. Clinical and genetic characteristics of Japanese Burkitt lymphomas with or without leukemic presentation. Int J Hematol. 2003;77: Rao CR, Gutierrez MI, Bhatia K, et al. Association of Burkitt s lymphoma with the Epstein-Barr virus in two developing countries. Leuk Lymphoma. 2000;39: Peylan-Ramu N, Diment J, Krichevsky S, et al. Expression of EBV encoded nuclear small non-polyadenylated RNA (EBER) molecules in 32 cases of childhood Burkitt s lymphoma from Israel. Leuk Lymphoma. 2001;40: Chao TY, Wang TY, Lee WH. Association between Epstein- Barr virus and Burkitt s lymphoma in Taiwan. Cancer. 1997;80: Akasaka T, Akasaka H, Ueda C, et al. Molecular and clinical features of non-burkitt s, diffuse large-cell lymphoma of B-cell type associated with the c-myc/immunoglobulin heavy-chain fusion gene. J Clin Oncol. 2000;18: Burmeister T, Schwartz S, Horst HA, et al. Molecular heterogeneity of sporadic adult Burkitt-type leukemia/lymphoma as revealed by PCR and cytogenetics: correlation with morphology, immunology and clinical features. Leukemia. 2005;19: Am J Clin Pathol 2007;128: Downloaded 564 from