13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Histopathology 2022, 81, 826–840. DOI: 10.1111/his.14798 Genetic and phenotypic characterisation of HIV-associated aggressive B-cell non-Hodgkin lymphomas, which do not occur specifically in this population: diagnostic and prognostic implications Maria Joao Baptista,1,† Gustavo Tapia,2,† Ana-Marıa Mu~ noz-Marmol,2 Josep Muncunill,1 Olga Garcia, Silvia Montoto, John G Gribben, Maria Calaminici,3 Antonio Martinez,4 1 3 3 Luis Veloza,4 Alejandra Martınez-Trillos,5 Teresa Aldamiz,6 Javier Menarguez,7 Marıa-Jose Terol,8 Antonio Ferrandez,9 Miguel Alcoceba,10 Javier Briones,11 Eva Gonz alez-Barca,12 Fina Climent,13 Ana Munta~ nola,14 Jose-Marıa Moraleda,15 Mariano Provencio, Pau Abrisqueta, Eugenia Abella,18 Lluis Colomo,19 16 17 Carlos Garcıa-Ballesteros,20 Montserrat Garcia-Caro,21 Juan-Manuel Sancho,1 Josep-Maria Ribera,1 Jose-Luis Mate2 & Jose-Tomas Navarro1 Department of 1Hematology, ICO-Germans Trias i Pujol Hospital, Josep Carreras Leukaemia Research Institute (IJC) and 2Pathology, Hospital Germans Trias i Pujol, IGTP, Universitat Autonoma de Barcelona, 13Department of Pathology, Hospital Universitari de Bellvitge-IDIBELL, L’Hospitalet de Llobregat, Badalona, Department of 4Pathology, Hospital Clinic, IDIBAPS and 5Hematology, Hospital Clinic, IDIBAPS, University of Barcelona, 11Department of Hematology, Hospital de la Santa Creu i Sant Pau, Josep Carreras Leukaemia Research Institute (IJC), 17Department of Hematology, Hospital Vall d’Hebron, Department of 18Hematology, and 19Pathology, Hospital del Mar, 21Department of Hematology, Hospital de Mataro, Barcelona, Department of 6Infectious Diseases and 7Pathology, Hospital Gregorio Mara~ non, Madrid, 10Department of Hematology, Hospital Universitario de Salamanca (HUS/IBSAL), CIBERONC and Centro de Investigacion del Cancer-IBMCC (USAL-CSIC), Salamanca, 12Department of Hematology, ICO-Hospital Duran i Reynals, L’Hospitalet de Llobregat, 14Department of Clinical Hematology, Hospital Universitari Mutua de Terrassa, Terrassa, 15Department of Hematology, Hospital Clinico Universitario Virgen de la Arrixaca, Murcia, 16 Department of Medical Oncology, Hospital Universitario Puerta De Hierro, Majadahonda, 20Department of Hematology, Hospital Arnau de Vilanova, Valencia, Department of 8Hematology and Oncology and 9Pathology, Hospital Clınic Universitari de Valencia, Valencia, Spain and 3Centre for Haemato-Oncology, Barts Cancer Institute, Queen Mary University of London, London, UK Date of submission 21 February 2022 Accepted for publication 12 September 2022 Published online Article Accepted 24 September 2022 Baptista M J, Tapia G, Mu~ noz-Marmol A-M, Muncunill J, Garcia O, Montoto S, Gribben J G, Calaminici M, Martinez A, Veloza L, Martınez-Trillos A, Aldamiz T, Menarguez J, Terol M-J, Ferrandez A, Alcoceba M, Briones J, Gonz alez-Barca E, Climent F, Munta~ nola A, Moraleda J-M, Provencio M, Abrisqueta P, Abella E, Colomo L, Garcıa-Ballesteros C, Garcia-Caro M, Sancho J-M, Ribera J-M, Mate J-L & Navarro J-T (2022) Histopathology 81, 826–840. https://doi.org/10.1111/his.14798 Genetic and phenotypic characterisation of HIV-associated aggressive B-cell non-Hodgkin lym- phomas, which do not occur specifically in this population: diagnostic and prognostic implications The frequency of aggressive subtypes of B-cell non- rearrangement (HGBL-DH/TH) or Burkitt-like lym- Hodgkin lymphoma (B-NHL), such as high-grade B-cell phoma (BL) with 11q aberration, is not well known in lymphomas (HGBL) with MYC and BCL2 and/or BCL6 the HIV setting. We aimed to characterise HIV- Address for correspondence: Jose Tom as Navarro, Department of Hematology, Institut Catala d’Oncologia-Germans Trias i Pujol Hospital, Josep Carreras Leukaemia Research Institute (IJC), Crta. de Canyet, Camı de les Escoles s/n 08916 Badalona, Spain. e-mail: jnavarro@iconcologia.net † These authors contributed equally to this work. Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Characterisation of HIV-associated aggressive NHL 827 associated aggressive B-NHL according to the 2017 BCL2 rearrangements were infrequent (4.3%). MYC WHO criteria, and to identify genotypic and phenotypic expression was identified in 23.4% of DLBCL cases, and features with prognostic impact. Seventy-five HIV- coexpression of MYC and BCL2 in 13.0%, which was associated aggressive B-NHL were studied by immuno- associated with a worse prognosis. As for BL cases, the histochemistry (CD10, BCL2, BCL6, MUM1, MYC, and expression of MUM1 (30.4%) conferred a worse prog- CD30), EBV-encoded RNAs (EBERs), and fluorescence nosis. Finally, the prevalence of HGBL-DH/TH and BL- in situ hybridisation (FISH) to evaluate the status of like with 11q aberration are reported in the HIV set- the MYC, BCL2, and BCL6 genes and chromosome ting. The phenotypic and genotypic characteristics of 11q. The 2017 WHO classification criteria and the HIV-associated aggressive B-NHL are similar to those Hans algorithm, for the cell-of-origin classification of of the general population, except for the low frequency diffuse large B-cell lymphomas (DLBCL), were applied. of BCL2 rearrangements in DLBCL. MYC and BCL2 In DLBCL cases, the frequencies of MYC and BCL6 coexpression in DLBCL, and MUM-1 expression in BL, rearrangements (14.9 and 27.7%, respectively) were have a negative prognostic impact on HIV-infected similar to those described in HIV-negative patients, but individuals. Keywords: genetics, phenotype, HIV, lymphoma Introduction are widely used when gene expression technologies are not available.7 Aggressive B-cell non-Hodgkin lymphomas (B-NHL), Both the 2008 and 2017 WHO classifications of such as diffuse large B-cell lymphoma (DLBCL) and tumours of haematopoietic and lymphoid tissues Burkitt’s lymphoma (BL), have an increased incidence include a category for lymphomas associated with in the HIV-infected population.1 HIV-infected patients HIV infection. This category comprises subtypes are currently treated with the same regimens as those occurring more specifically in HIV-positive patients, given to the noninfected, achieving similar response in other immunodeficient states, and lymphomas also rates.1,2 However, HIV-infected patients have occurring in immunocompetent individuals.7,8 The impaired immunosurveillance. This is due to the most frequent lymphoma subtypes among HIV- direct effect of HIV on CD4+ T cells and the further infected individuals are DLBCL and BL. The 2017 imbalance of cellular populations and levels of cyto- WHO classification introduced new categories in the kines.3 HIV-encoded proteins, like p17, can activate general population, such as Burkitt-like lymphoma PI3K/Akt signalling pathways, leading to the expan- with 11q aberration, high-grade B-cell lymphomas sion of B cells.4 HIV is responsible for the chronic (HGBL) with MYC and BCL2 and/or BCL6 rearrange- activation of B lymphocytes, thus increasing the prob- ment (HGBL-DH/TH), and HGBL NOS, the former ability of genetic alterations promoted by activation- with a dismal prognosis. Moreover, it has been found induced cytidine deaminase (AID), such as chromoso- that coexpression of MYC and BCL2 in the absence of mal translocations involving MYC and BCL6 muta- a double-hit also impacts prognosis. However, most of tions. Moreover, HIV-infected patients are prone to the studies on HIV-associated lymphomas are not coviral infections by Epstein–Barr virus (EBV) and based on the new classifications and data regarding human herpes virus-8 (HHV8), which are known to gene translocations, and the expression of MYC and enhance tumour cell growth and survival, and induce BCL2 in these lymphomas is scarce. We aimed to lymphomagenesis.5 Therefore, it would be expected to ascertain how HIV-associated lymphomas are dis- find phenotypic and genotypic differences between tributed according to the WHO 2017 classification. HIV-associated lymphomas and their corresponding Therefore, we reviewed the diagnosis of a comprehen- immunocompetent counterparts. sive group of HIV-associated aggressive B-cell lym- DLBCL can be divided into two subtypes according phomas applying IHC and fluorescent in situ to the cell of origin (COO): germinal centre B-cell like hybridisation (FISH) studies currently recommended (GCB) and activated B-cell like (ABC), the latter being for the diagnosis of these lymphomas. This study pro- associated with poorer outcomes.6 It is currently rec- vides a useful snapshot of the genotypic and pheno- ommended to study the COO subtype in all DLBCL, typic features of HIV-associated aggressive B-NHL and and immunohistochemistry (IHC)-based algorithms identifies markers with prognostic implications. Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 828 M J Baptista et al. BCL2, and BCL6 rearrangements, according to pub- Materials and methods lished methods.11 MYC rearrangements were evalu- STUDY POPULATION ated with Vysis MYC Break Apart and Vysis IGH/MYC/ CEP8 Tri-Colour Dual Fusion probes (Abbott Molecu- Seventy-five patients previously diagnosed with HIV- lar., Abbot Park, IL, USA). Cases with MYC rearrange- associated aggressive B-NHL including DLBCL, BL, ments detected with the break-apart probe but not and B-cell lymphoma unclassifiable with features with the IGH/MYC/CEP8 were hybridised with a mix- intermediate between DLBCL and classical BL (BCLU) ture of Vysis LSI MYC SpectrumGold probe (Abbott were selected based on the availability of formalin- Molecular) and IGL or IGK FISH DNA Split Signal fixed and paraffin-embedded (FFPE) material. Cases probe (Dako). Cases with IGL or IGK FISH-split pattern with inconclusive FISH analyses were excluded. Other colocalised with a MYC FISH-split signal were consid- aggressive lymphoma subtypes, occurring more ered as IGL-MYC or IGK-MYC rearranged, respectively. specifically in HIV-infected patients, were not Vysis BCL2 and BCL6 dual-colour break-apart probes included. Patients were diagnosed in several hospitals (Abbott Molecular) were used for detection of BCL2 in Spain and the United Kingdom. All cases were ret- and BCL6 translocations. The 11q alteration was stud- rospectively reviewed, and the main clinical-biological ied with ZytoLight SPEC 11q gain/loss Triple Colour variables were retrieved. The study was approved by Probe (ZytoVision, Bremerhaven, Germany). the Institutional Research Board of the Hospital Ger- mans Trias I Pujol HGTIP (Reference: EO-12-072) in accordance with the Declaration of Helsinki and STATISTICAL ANALYSES biomedical research legislation (Law 14/2007). A descriptive study of the variables was performed. Chi-square, Fisher’s exact test, or median test were IMMUNOHISTOCHEMICAL ANALYSES used for comparisons. Overall survival (OS) and progression-free survival (PFS) were defined according Haematoxylin–eosin-stained slides were reviewed and to standard criteria.12 Survival probabilities were esti- representative tumour cell-rich cores were taken in mated according to the Kaplan–Meier method and triplicate from formalin-fixed and paraffin-embedded comparisons were performed with the log-rank test. (FFPE) blocks to build tissue microarrays (TMA). Univariate cox regression analyses for OS and PFS Immunohistochemical stains were performed with were performed to calculate hazard ratios with confi- Dako EnVision system (Dako, Glostrup, Denmark) fol- dence intervals. P ≤ 0.05 was considered statistically lowing the manufacturer’s protocol. The following anti- significant. Analyses were performed using SPSS bodies (clone, concentration; brand) were used: CD10 v24.0 (IBM, Somer, NY, USA). (56C6, 1/20; Novocastra, Newcastle, UK), BCL6 (PG- B6P, 1/30; Dako), MUM1 (clone MUM1p, 1/40; Dako), BCL2 (124, 1/200; Dako) MYC (Y69, 1/100, Epito- Results mics, Burlingame, CA, USA) and CD30 (clone 1G12, 1/20, Novocastra). BCL2 and MYC were scored in 5% DIAGNOSTIC REVISION AND CLINICAL FEATURES increments, and other markers as positive or negative. This study included 75 patients with HIV-associated The cutoff of 30% was applied for CD10, BCL6 and aggressive B-NHL diagnosed between 1998 and MUM1 according to Hans criteria.9 For CD30, the cut- 2013. off was 20% and focal positivity was recorded.10 The clinical features of the 75 patients with com- plete FISH are in Table 1. Information regarding pre- vious combined antiretroviral therapy (cART) was IN SITU HYBRIDISATION ANALYSES available in 72 out of the 75 cases. Analysing the The presence of EBV was assessed in all cases by whole series, patients who were on cART (n = 45), means of EBV-encoded RNA (EBER) detection by when lymphoma was diagnosed, had worse OS than in situ hybridisation with the INFORM EBER PROBE those who were not (n = 27); OS probability (95% assay (Ventana Medical Systems, Tucson, AZ, USA) confidence interval [CI]) 59% (44%, 74%) versus and the ISH IVIEWBLUE detection system (Ventana 78% (62%, 94%) (P: 0.038). No relationship was Medical Systems) in a BenchMark ULTRA automated found between previous cART and any of the clinical instrument (Ventana Medical Systems). or biological variables, including degree of immuno- Fluorescent in situ hybridisation (FISH) analyses suppression, HIV-load, previous AIDS, and year of were performed in all cases for the presence of MYC, lymphoma diagnosis. Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Characterisation of HIV-associated aggressive NHL 829 Table 1. Clinical features of the HIV-associated lymphoma series Clinical features DLBCL N = 47 BL N = 24 HGBL DH/TH N = 3 P-value Males, N (%) 37 (78.7) 21 (87.5) 2 0.287 Age in years, median (min, max) 44 (30, 68) 46 (33, 76) 47 (41, 54) 0.394 ECOG ≥2, N (%) 21/46 (45.7) 10 (41.7) 2 0.750 Ann Arbor stage III-IV, N (%) 32 (68.1) 18 (75.0) 2 0.546 Extranodal sites ≥2, N (%) 12 (25.5) 11 (45.8) 1 0.084 B-symptoms, N (%) 20 (42.6) 18 (75.0) 1 0.010 Bulky disease, N (%) 9 (8.5) 10 (41.7) 1 0.043 Elevated ß2M, N (%) 28/32 (87.5) 10/15 (66.7) 2 0.100 Elevated serum LDH, N (%) 29 (61.7) 24 (100) 2 <0.001 IPI ≥3, N (%) 19/46 (41.3) 15 (60) 2 0.092 Previous OI, N (%) 16/33 (48.5) 3/20 (15) 3 0.014 Previous AIDS, N (%) 21/43 (48.8) 7 (29.2) 2 0.118 Previous cART, N (%) 27/45 (60.0) 14/23 (60.9) 3 0.945 Detectable HIV-load, N (%) 32/41 (78.0) 13/20 (65) 1 0.277 CD4+ T-cell counts/uL, median (min, max) 139 (6, 1198) 199 (12, 697) 499 (353, 600) 0.679 Treatment with R, N (%) 43 (91.5) 20/23 (87) 3 (100) 0.418 ß2M, beta2-microglobulin; BL, Burkitt lymphoma; cART, combination antiretroviral therapy; DLBCL, diffuse large B-cell lymphoma; ECOG, Eastern Cooperative Oncology Group; HGBL DH/TH, high grade B-cell lymphoma with MYC and BCL2 and/or BCL6 rearrangement; IPI, International Prognostic index; LDH, lactate dehydrogenase; OI, opportunistic infection; R, rituximab. P-values were calculated comparing DLBCL and BL groups and statistically significant P-values are highlighted. HGBL DH/TH were not included in the comparison due to the low number of cases. The case of Burkitt-like lymphoma with 11q aberration is not included in the table. In the following sections we describe the pheno- nongerminal centre (non-GC). The patient presented typic and genotypic characteristics of each category with bulky disease, without extranodal involvement, of aggressive B-NHL according to the 2017 WHO and an International Prognostic index (IPI) of 3. classification and their prognostic impact. Of note, no After six cycles of RCHOP (rituximab, cyclophos- cases of HGBL NOS were identified. phamide, doxorubicin, vincristine, and prednisone) with concomitant cART, the patient achieved a com- plete remission that persists, so far, more than HIV-ASSOCIATED HIGH-GRADE B-CELL 10 years after diagnosis. LYMPHOMA WITH MYC, BCL2, AND/OR BCL6 The second case showed an MYC-IGH translocation REARRANGEMENT and a BCL2 rearrangement (Figure 1). The tumour The three HIV-infected patients diagnosed with had a DLBCL-like morphology, and tumour cells HGBL-DH/TH presented all the possible patterns of expressed MUM1, BCL2, and MYC (95%) and were rearrangements of this category. negative for CD10, BCL6, CD30, and EBER; hence, it One case was a triple hit, harbouring MYC-IGH was classified as non-GC. The patient presented with translocation, besides BCL2 and BCL6 rearrange- multiple adenopathies, bone marrow and skin ments. The tumour had a DLBCL-like morphology, involvement, and an IPI of 4. The patient died due to and lymphoma cells expressed BCL6, MUM1, BCL2, lymphoma progression after the 5th cycle of RCHOP. and MYC (40%), whereas they were negative for The third case was a double-hit lymphoma with CD10, CD30, and EBER. Therefore, it was classified as MYC-non-Ig and BCL6 rearrangements (Figure 2). Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 830 M J Baptista et al. Figure 1. High-grade B-cell lymphoma with MYC and BCL2 rearrangement. Diffuse proliferation of large lymphoid cells with centroblastic morphology on H&E (A, 9400). The neoplastic cells were negative for CD10 (B, 9200) and positive for BCL2 (C, 9200). Positivity for MYC in >90% of the neoplastic cells (D, 9200). FISH studies revealed a BCL2 rearrangement (E) and an MYC-IGH fusion (F). Figure 2. High-grade B-cell lymphoma with MYC and BCL6 rearrangements. Diffuse proliferation of medium-sized lymphocytes with blastoid morphology and a starry-sky pattern on H&E (A, 9400). The cells were negative for CD10 (B, 9200) and BCL2 (C, 9200). MYC (D, 9200) was positive in >95% of the neoplastic cells. FISH studies with break apart probes highlighted both BCL6 (E) and MYC rearrangements (F). The tumour had a BL-like morphology, and cells were a GC phenotype. The patient presented with Ann positive for BCL6, MYC (80%), and EBER, and nega- Arbor stage II and IPI of 0. The lymphoma was trea- tive for CD10, MUM1, BCL2, and CD30; hence, it had ted with an immunochemotherapy regimen for BL Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Characterisation of HIV-associated aggressive NHL 831 with concomitant cART and the patient is in com- However, multivariate analyses were not performed plete remission for more than 5 years.13 due to the low number of BL cases (n = 24), espe- cially MUM-1 positive (n = 7), and hence, a lack of statistical power. HIV-ASSOCIATED BURKITT LYMPHOMA Patients with BL were treated with intention to cure as BURKITT-LIKE LYMPHOMA WITH 11Q follows: intensive treatment for BL (Burkimab)13 (cy- ABERRATION clophosphamide, prednisone, rituximab, vincristine, methotrexate, iphosphamide, dexamethasone, teni- One patient with a previous diagnosis of BL without poside, cytarabine, doxorubicine, vindesine, etoposide) demonstrable MYC rearrangement showed losses of (n = 13), RCHOP (n = 6), RHyperCVAD (rituximab, 11q24.1-ter by FISH, being reclassified as Bukitt-like cyclophosphamide, vincristine, doxorubicin, dexam- lymphoma with 11q aberration (Figure 3). He was a ethasone, methotrexate, and cytarabine) (n = 3), and 57-year-old male who presented with extranodal dis- CHOP (cyclophosphamide, doxorubicin, vincristine, ease and adenopathies with an IPI of 0. The lym- and prednisone) (n = 2). All patients received com- phoma cells were positive for CD10 and BCL6, and bined antiretroviral therapy (cART) concomitantly negative for BCL2, MUM1, CD30, and EBER. MYC with immune-chemotherapy or chemotherapy. expression was inconclusive. The patient received An MYC-IGH translocation was detected in 21 out six cycles of RCHOP with concomitant cART, achiev- of 24 HIV-associated BL cases (one of them detected ing a complete remission and died 5 years later of with fusion probes) and an MYC-IGL in 2 out of 24. lung non-small-cell carcinoma. The remaining case was an aggressive lymphoma with morphology and phenotype consistent with BL HIV-ASSOCIATED DIFFUSE LARGE B-CELL (CD10+, BCL6+, BCL2 , MUM1 , EBER ), without LYMPHOMA BCL2 or BCL6 translocation. As MYC rearrangement was not detected, FISH analysis for 11q alterations Patients with DLBCL were treated with intention to was performed and ruled out. Despite a negative cure as follows: RCHOP (n = 40), CHOP (n = 4), MYC rearrangement, the remaining clinicopathologi- intensive treatment for BL (Burkimab) (n = 2), and R- cal features were consistent with BL. EPOCH (etoposide, prednisone, cyclophosphamide, Phenotypic features and gene translocations were vincristine, and doxorubicine) (n = 1). All patients analysed in the 24 BL cases. The GC markers CD10 received cART concomitantly with immune- and BCL6 were expressed in 100% and 95.7% of the chemotherapy or chemotherapy. cases, respectively. Notably, 54.2% of the cases were Rearrangements involving MYC were present in 7 positive for EBER, and all cases were negative for out of 47 (14.9%) of DLBCL cases. Five cases had an CD30. Four HIV-associated BL (16.7%) strongly MYC-IGH rearrangement, one case had an MYC-IGK, expressed BCL2 (90% of the cells in the mean). These and one case was negative for MYC-IGH rearrange- four cases had otherwise typical clinicopathological ment and inconclusive for MYC-IGL and MYC-IGK and phenotypic features of BL, including MYC-IGH (Table 4). Rearrangements of the BCL6 gene were translocation, without concomitant BCL2 or BCL6 found in 13 cases (27.7%), whereas BCL2 rearrange- rearrangement. All four had monomorphic medium- ments were detected only in two (4.3%). sized lymphocytes with fine chromatin and several The immunophenotype of HIV-associated DLBCL is basophilic nucleoli in a starry-sky pattern. Patients described in Table 4. Almost two-thirds were non-GC had extranodal involvement and advanced stage at phenotype according to the Hans algorithm. No dif- presentation in all cases, and bulky mass in three of ferences were found between the two groups regard- them. ing the proportion of patients who were on cART at MUM1 was expressed in 30.4% of the cases (7 out lymphoma diagnosis. Of note, CD30 was positive in of 23). Importantly, HIV-associated BL expressing 10 out of 47 cases (21,3%), four of them only focally, MUM1 had worse OS and PFS (Figure 4A,B). The rest and 41.3% were BCL2-positive. Although there is no of the main clinical and biological variables did not consensual cutoff for BCL2 positivity, the 50% cutoff show any prognostic impact on OS in the univariate was applied, as previously reported.14 analyses (Table 2). On the other hand, PFS was influ- EBER was positive in 21.3% cases, and it was not enced by the level of immunosuppression, having associated with CD30 expression. As expected, BCL6 patients with CD4 + lymphocytes ≤200/ll a signifi- expression was associated with BCL6 rearrangements cantly lower 5-year PFS probability (Table 2). (P = 0.025). Eleven out of 47 cases (23.4%) Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 832 M J Baptista et al. Table 2. Univariate analyses for overall survival and progression-free survival of patients with Burkitt lymphoma Variable 5 year OS (95% CI) P 5 year PFS (95% CI) P ECOG ≤1 70% (45%, 95%) 0.857 57% (31%, 83%) 0.427 >1 70% (42%, 98%) 70% (42%, 98%) Ann Arbor stage I-II 63% (21%, 85%) 0.708 50% (10%, 90%) _ III-IV 72% (51%, 93%) 67% (45%, 89%) Serum LDH Normal _ (n = 0) _ _ (n = 0) _ High 70% (51%, 89%) 63% (44%, 82%) Number of extranodal sites ≤1 84% (64%, 100%) 0.126 69% (44%, 94%) 0.251 >1 55% (26%, 84%) 55% (26%, 84%) Bone marrow involvement No 77% (57%, 97%) 0.271 67% (54%, 80%) 0.590 Yes 50% (10%, 90%) 50% (10%, 90%) IPI score 0–2 76% (47%, 100%) 0.627 67% (36%, 98%) 0.680 3–5 67% (43%, 91%) 60% (35%, 85%) CD4 ≤200 62% (33%, 91%) 0.175 46% (17%, 75%) 0.045 >200 90% (71%, 100%) 90% (71%, 100%) CD4 ≤100 57% (20%, 94%) 0.177 57% (20%, 94%) 0.489 >100 86% (68%, 100%) 71% (47%, 95%) HIV-load Non-detectable 71% (37%, 100%) 0.454 57% (20%, 94%) 0.158 Detectable 85% (65%, 100%) 77% (54%, 100%) Previous cART No 89% (68%, 100%) 0.154 78% (51%, 100%) 0.199 Yes 63% (37%, 100%) 57% (31%, 83%) EBERs Negative 71% (43%, 99%) 0.791 64% (36%, 92%) 0.779 Positive 69% (44%, 94%) 62% (36%, 88%) MUM1 expression No 87.5% (72%, 100%) 0.034 75% (54%, 96%) 0.056 Yes 43% (6%, 80%) 43% (6%, 80%) BL, Burkitt lymphoma; cART, combination antiretroviral therapy; DLBCL, diffuse large B-cell lymphoma; EBER, Epstein–Barr virus–encoded small RNAs; ECOG, Eastern Cooperative Oncology Group; LDH, lactate dehydrogenase. expressed MYC protein, and six cases (12.8%) were strong negative impact on survival, particularly PFS double expressors (DE) of MYC and BCL2. (Figure 5). HIV-associated DLBCL cases with a non- Then we compared genotype and phenotype of GC GC phenotype tended to have a worse OS and PFS and non-GC cases (Table 4). MYC rearrangements than the cases with the GC phenotype (Table 3). were more frequent in HIV-associated DLBCL with a Moreover, cases that had BCL6 rearrangements GC phenotype, whereas BCL6 rearrangements, CD30 tended to have worse OS and PFS (Table 3). All cases and BCL2 expression were almost exclusive of the with a BCL6 rearrangement, except one, had a non- non-GC subtype. GC phenotype. Hence, the impact of BCL6 rearrange- Survival analyses were performed including only ment on outcomes was studied in this phenotype- HIV-associated DLBCL treated with RCHOP (N = 40). subset, but there was none (Supplementary Fig- Concomitant expression of MYC and BCL2 had a ure S1). Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Characterisation of HIV-associated aggressive NHL 833 Figure 3. Burkitt-like lymphoma with 11q alterations. (A, H&E, 9200) diffuse proliferation of medium-sized lymphocytes with a starry-sky pattern, with expression of CD10 (B, 9200) negativity for BCL2 (C, 9200). FISH studies highlighted distal loss on chromosome 11q (D, arrow). A B 1.0 MUM-negative (N=16) 1.0 5y OS 87.5%, 95% CI [72%, 100%] MUM-negative (N=16) 5y PFS 75%, 95% CI [54%, 96%] 0.8 0.8 PFS probability OS probability 0.6 MUM-positive (N=7) 0.6 MUM-positive (N=7) 5y OS 42.9%, 95% CI [6%, 80%] 5y PFS 42.9%, 95% CI [6%, 80%] 0.4 0.4 0.2 0.2 P=0.034 P=0.056 0.0 0.0 0 2 4 6 8 10 12 0 2 4 6 8 10 12 Years elapsed Years elapsed Figure 4. Outcomes of patients with HIV-associated Burkitt lymphoma according to MUM1 expression. (A) Kaplan–Meier curves for overall survival (OS) of HIV-associated Burkitt lymphoma by MUM1 expression. (B) Kaplan–Meier curves of progression-free survival (PFS) of HIV- associated Burkitt lymphoma by MUM1 expression. Of note, the Burkitt-like lymphoma with 11q case was removed from the survival analy- ses. The rest of the important clinical and biological was positive in all three HIV-associated HGBL-DH/TH variables did not show any prognostic impact on OS cases (100%), 23 out of 24 (95.8%) BL, and 11 out and PFS in the univariate analyses (Table 3). Multi- of 47 (23.4%) DLBCL cases (Figure 6). As expected, variate analyses were not performed due to the low MYC rearrangements were associated with MYC number of double-expresser cases (n = 5) and the expression (P < 0.001). However, as reported previ- lack of statistical power. ously in non-HIV-associated lymphomas, the correla- tion was not perfect15; two cases with MYC-Ig rearrangement (one DLBCL and one BL) did not MYC EXPRESSION IN HIV-ASSOCIATED express MYC protein, and six cases without MYC AGGRESSIVE B-CELL LYMPHOMAS rearrangements (five DLBCL and one BL) showed high MYC protein expression. MYC expression studies were successful in 71 out of the 75 HIV-associated HGBL; MYC studies were inconclusive in three HIV-associated BL cases and in the Burkitt-like lymphoma with 11q aberration case. Discussion MYC expression was considered positive when at least The present study included 75 HIV-associated aggres- 40% of the cells were positive for MYC staining. MYC sive B-NHL cases with previous diagnoses of DLBCL, Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 834 M J Baptista et al. A B 1.0 1.0 No-DE (N=35) No-DE (N=35) 0.8 5y OS 68.3%, 95% CI [52.8%, 83.8%] 0.8 5y PFS 68.3%, 95% CI [52.8%, 83.8%] OS probability PFS probability 0.6 0.6 DE (N=5) DE (N=5) 5y PFS 0%, 5y OS 40%, 95% CI [0%, 82.9%] 0.4 0.4 0.2 0.2 0.0 P=0.043 P=0.005 0.0 0 3 5 8 10 13 0 3 5 8 10 13 Years elapsed Years elapsed Figure 5. Phenotypic features with impact on outcomes of patients with HIV-associated diffuse large B-cell lymphoma (DLBCL). (A) Kaplan– Meier curves for overall survival (OS) of HIV-associated DLBCL for double expressers (DE) and no-DE. (B) Kaplan–Meier curves for progression-free survival (PFS) of HIV-associated DLBCL for double expressers (DE) and no-DE. 15 14 13 12 11 10 Number of cases 9 8 7 6 5 4 3 2 1 0 0-5 10 - 15 20 - 25 30 - 35 40 - 45 50 - 55 60 - 65 70 - 75 80 - 85 90 - 100 % of cells expressing MYC DLBCL without MYC rearrangement BL without MYC rearrangement HGBL-DH/TH DLBCL with MYC rearrangement BL with MYC rearrangement Figure 6. Relationship between the percentage of tumour cells expressing MYC and the presence or not of rearrangements involving MYC. BL, Burkitt lymphoma; DLBCL, diffuse large B-cell lymphoma; HGBL-DH/TH, high grade B-cell lymphoma with MYC and BCL2 and/or BCL6 rearrangement. BL, or BCLU (2008 WHO classification) that were patients, to the best of our knowledge this is the first reviewed and reclassified according to the 2017 WHO systematic study by FISH in a large series of HIV- classification. MYC, BCL2, and BCL6 rearrangements associated lymphoma, leading to understand the real were studied in all cases, and 11q alterations were prevalence of HGBL-DH/TH and BL-like with 11q evaluated in cases with BL features without MYC aberration in the HIV-setting.16–19 Importantly, we rearrangements. Although these genetic abnormali- studied the prognostic impact of COO subtypes and of ties have been already reported in HIV-infected some phenotypic markers, including MYC, BCL2 and Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Characterisation of HIV-associated aggressive NHL 835 Table 3. Univariate analyses for overall survival and progression-free survival of the 40 patients with diffuse large B-cell lymphoma treated with RCHOP 5 year OS (95% CI) P 5 year PFS (95% CI) P ECOG ≤1 67% (47%, 87%) 0.658 67% (47%, 87%) 0.488 >1 63% (41%, 85%) 52% (29%, 75%) Ann-Arbor stage I-II 91% (74%, 100%) 0.091 91% (74%, 100%) 0.064 III-IV 55% (37%, 73%) 48% (30%, 66%) Serum LDH Normal 77% (57%, 97%) 0.331 66% (44%, 88%) 0.452 High 55% (34%, 76%) 55% (34%, 76%) N extranodal sites ≤1 66% (49%, 83%) 0.582 63% (46%, 80%) 0.570 >1 60% (30%, 90%) 50% (19%, 81%) Bone marrow involvement No 63% (47%, 79%) 0.999 63% (47%, 79%) 0.505 Yes 80% (45%, 100%) 40% (0, 83%) IPI score 0–2 69% (50%, 88%) 0.420 65% (45%, 85%) 0.518 3–5 59% (36%, 82%) 53% (29%, 77%) CD4 ≤200 69% (50%, 88%) 0.284 69% (50%, 88%) 0.157 >200 63% (39%, 87%) 50% (25%, 75%) CD4 ≤100 58% (34%, 82%) 0.636 58% (34%, 82%) 0.839 >100 73% (54%, 92%) 64% (44%, 84%) HIV-load Non-detectable 86% (60%, 100%) 0.261 62% (27%, 97%) 0.478 Detectable 61% (43%, 79%) 61% (43%, 79%) cART before lymphoma No 71% (49%, 93%) 0.276 71% (49%, 93%) 0.201 Yes 59% (38%, 80%) 50% (29%, 71%) EBERs Negative 67% (50%, 84%) 0.596 61% (44%, 78%) 0.746 Positive 56% (23%, 89%) 56% (23%, 89%) CD30 expression Negative 71% (43%, 99%) 0.791 58% (41%, 75%) 0.965 Positive 69% (44%, 94%) 67% (36%, 98%) MYC expression No 70% (53%, 86%) 0.153 70% (53%, 86%) 0.055 Yes 50% (19%, 81%) 30% (2%, 58%) Double expresser status No 68% (53%, 83%) 0.043 68% (53%, 83%) 0.005 Yes* 40% (0, 83%) 0 COO subtype GC 79% (57%, 100%) 0.082 79 (57%, 100%) 0.061 Non-GC 57% (38%, 76%) 50 (30%, 69%) MYC rearrangement No 65% (49%, 81%) 0.656 62% (46%, 78%) 0.716 Yes 67% (29%, 100%) 50% (10%, 90%) BCL6 rearrangement No 74% (58%, 91%) 0.060 67% (49%, 84%) 0.092 Yes 45% (17%, 72%) 45% (17%, 72%) BL, Burkitt lymphoma; cART, combination antiretroviral therapy; COO, cell of origin; DLBCL, diffuse large B-cell lymphoma; EBER, Epstein–Barr virus–encoded small RNAs; ECOG, Eastern Cooperative Oncology Group; IPI, International Prognostic index; LDH, lactate dehydrogenase. Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 836 M J Baptista et al. Table 4. Genotype and phenotype of HIV-associated diffuse large B-cell lymphoma by cell-of-origin group assessed by Hans algorithm Total N = 47 GC N = 17 (36.2%) Non-GC N = 30 (63.8%) P-value TMYC, N (%) 7 (14.9) 5 (29.4) 2 (6.7) 0.049 TBCL2, N (%) 2 (4.3) 0 2 (6.7) 0.402 TBCL6, N (%) 13 (27.7) 1 (5.9) 12 (40) 0.011 CD10, N (%) 14 (29.8) 14 (82.4) 0 <0.001 BCL6, N (%) 32 (68.1) 17 (100) 15 (50) <0.001 MUM1, N (%) 28 (59.6) 3 (17.6) 25 (83.3) <0.001 CD30, N (%) 10 (21.3) 0 10 (33.3) 0.006 EBER, N (%) 10 (21.3) 2 (11.8) 8 (26.7) 0.206 BCL2, N (%) 19/46 (41.3) 1 (5.9) 18/29 (62.1) <0.001 MYC, N (%) 11 (23.4) 6 (35.3) 5 (16.7) 0.138 DE, N (%) 6/46 (13.0) 1 (5.9) 5/29 (17.2) 0.266 DE, double expressers considering positive cutoffs of BCL2 ≥ 50% and MYC ≥40%; GC, germinal center; GCB, germinal center B-cell; non- GC, nongerminal center; TBCL2, BCL2 rearrangement; TBCL6, BCL6 rearrangement; TMYC, MYC rearrangement. P-values from Chi-square or Fisher’s exact test are disclosed except for the comparison of Hans algorithm results, whose P-value is from McNemar’s test. Statistically significant P-values are highlighted. DE, in the subset of patients with DLBCL treated with expression in the immunocompetent population and RCHOP. a diagnosis of BL can be made after excluding an After a comprehensive study by FISH and immuno- HGBL-DH/TH.23 Second, we observed MUM1 expres- histochemistry we succeeded in reviewing the diagno- sion in a subset of HIV-associated BL (30.4%). It is sis of 75 cases. In our series, 4% of the cases were generally assumed that BL do not express late/post HGBL-DH/TH, a percentage similar to that reported GC markers, thus MUM1 has been scarcely studied in in immunocompetent patients.20,21 The reduced these lymphomas. However, MUM1 expression has number of HGBL-DH/TH prevented survival analyses. been detected in 10% to 41% of BL, including the All three cases were DE and two had a non-GC phe- HIV-associated BL series.24–28 Importantly, we notype. Although the 2017 WHO classification states observed that MUM1 expression negatively impacted that “a double-hit status should be investigated in all outcomes of HIV-associated BL, in line with a recent DLBCLs, NOS, using cytogenetic or molecular cytoge- study on BL in immunocompetent individuals.29 netic studies,” some centres prefer to look for rear- One case previously classified as BL, without detect- rangements only in cases with a GC phenotype and able MYC rearrangements by FISH, was tested for DE, or in cases with a BL-like morphology.22 Our 11q alterations and finally was reclassified as a results would support the need for universal screen- Burkitt-like lymphoma with 11q aberration. This ing for gene rearrangements. lymphoma category is a recently described type of Around 50% of HIV-associated BL were EBER- aggressive lymphoma that shares clinical and pheno- positive. However, we observed BL cases with pecu- typic characteristics with BL, but lacks MYC rear- liarities so far only reported in the immunocompetent rangements and carries gains in 11q23.2–23.3 and/ population. First, 16.7% of the HIV-associated BL or losses in 11q24.1-ter.30 This lymphoma lacks the cases expressed BCL2, even with an intense and dif- genetic profile of BL and is similar to the GCB subtype fuse pattern. BCL2 is usually negative or weak in BL. of DLBCL or high-grade B-cell lymphoma.31,32 This Whereas the high expression of BCL2 in an aggres- lymphoma subtype has been rarely reported in the sive lymphoma with BL morphology suggests a BCL2 HIV setting, and our results show that the full spec- rearrangement consistent with a diagnosis of HGBL- trum of aggressive B-cell lymphoma can be seen in DH/TH, around 20% of BL cases show BCL2 HIV-infected patients. Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Characterisation of HIV-associated aggressive NHL 837 Gene rearrangements involving MYC were detected previously reported (around 30%).40,43,48 EBER was in 15% of HIV-associated DLBCL cases, a frequency expressed slightly more frequently in non-GC cases, that is similar to that reported in the immunocompe- and none of the EBV-positive cases had BCL6 rear- tent population.33–35 Moreover, MYC rearrangements rangements.40,43,48 Although the context of HIV- were more frequently found in cases with a GC phe- associated immunosuppression excludes the diagnosis notype. The impact of MYC rearrangements on of EBV-positive DLBCL NOS, the features above DLBCL prognosis is controversial, but it seems that described are similar to those found in EBV-positive cases with MYC rearrangements treated with RCHOP DLBCL NOS, which rarely show BCL6 rearrange- have a poorer prognosis.33–35 In our series of HIV- ments and are generally non-GC.49 EBER was not associated DLBCL cases, the presence of MYC rear- associated with CD30 expression in the present series rangements was not significantly associated with and, as in other studies including patients treated worse outcomes. The frequency of BCL6 rearrange- with immunochemotherapy, EBV showed no impact ments found in DLBCL of HIV-infected patients (28%) on outcomes of HIV-associated DLBCL.40,43,48 was also similar to that described in non-HIV-infected In DLBCL of the immunocompetent population, patients (around 30%) and were mainly detected in 12% to 19% of the cases are positive for CD30 (≥20% non-GC cases, as reported in the general popula- cutoff) and it is expressed both in GC and non-GC tion.20,34–39 The prognostic impact of BCL6 rear- cases.10,50–52 Interestingly, our results showed that rangements is controversial in immunocompetent around 20% of HIV-associated DLBCL express CD30, DLBCL.7,20,36,37 In this series, cases with BCL6 rear- and this result could drive therapeutic decisions, as rangement showed a trend for shorter PFS and OS. CD30 is effectively targetable.53 In our series, all Due to the BCL6 rearrangements mainly present in CD30-positive cases were of the non-GC phenotype, non-GC cases, we studied the prognostic impact of and in line with the results in non-GC DLBCL of the this genetic aberration in this subgroup of patients general population, it had no prognostic impact. and we did not find any. Interestingly, BCL2 rear- In line with previous studies on HIV-associated rangements were scarcely detected in our series DLBCL, around 40% of the studied cases in our series (4.3%), an incidence substantially lower than that expressed BCL2.40,41,54 BCL2 was mainly expressed reported in immunocompetent patients (20% to in non-GC cases, as described in DLBCL of immuno- 30%).34,35,38,39 competent patients.14 Our survival analyses showed The distribution of HIV-associated DLBCL cases in that BCL2 expression had no impact on outcomes. the GC and non-GC categories is quite inconsistent An MYC expression was detected in 23.4% of HIV- among studies, but the criteria employed for COO associated DLBCL, and 13% were DE, most of them of assignment also.3,40–45 Some authors argue that HIV- the non-GC phenotype. This incidence is slightly associated DLBCL show an immunophenotype inter- lower than that reported in DLBCL series of immuno- mediate between the GC and activated B-cell pheno- competent patients (20–30%).14,55–57 Of note, MYC types of DLBCL of immunocompetent patients, which positivity tended to be associated with worse PFS, could also explain those discrepancies.3,44,45 In line and DE was significantly associated with shorter with this, we have previously showed a disagreement, probability of OS and PFS. These results support the in an HIV-associated case series, between the results consideration, made in general DLBCL, that MYC and of Hans algorithm and a commonly used molecular- BCL2 coexpression, but not MYC expression alone, is based method for COO assignment developed by associated with a worse prognosis.56 The negative Nanostring.38,46 Herein, in line with a previous impact of DE on prognosis of DLBCL affecting report by Chao et al., we identified a higher frequency immunocompetent individuals is well established.10,58 of non-GC types in HIV-associated DLBCL than that However, to our knowledge, only one study has pre- reported in immunocompetent patients.41 Regarding viously assessed MYC expression and DE in HIV- the prognostic impact of COO, two previous studies associated DLBCL, but HGBL-DH/TH cases were not on HIV-associated DLBCL showed that the Hans algo- excluded and patients were not treated with rithm could identify cases with worse outcomes.43,47 immunochemotherapy.41 These authors reported that In the present study, a larger series of HIV-associated 64% of HIV-associated DLBCL expressed MYC and DLBCL treated with RCHOP we observed that non-GC 25% were DE. However, the cutoff employed for MYC cases tended to have a worse OS and PFS than GC positivity was 10% instead of 40% (recommended by cases, although without statistical significance. the WHO classification and used herein). On the Among DLBCL cases, 21% of cases were positive other hand, a recent study showed that patients with for EBER, a percentage slightly lower than that MYC-positive HIV-associated DLBCL, treated with Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 838 M J Baptista et al. REPOCH, had a significantly lower EFS. In this study, from Celgene, Janssen, and AstraZeneca. M.J.T. dis- no differences were found for both OS and PFS closes consultancy in Janssen, Roche, Astra Zeneca, between single MYC+ patients and double expressers Abbvie, and research funding from Janssen and (MYC+/BCL2+).59 Gilead. J.B. discloses honoraria from Roche, Takeda, In summary, this study provides a useful character- Celgene, Novartis, and Gilead; research funding from isation of HIV-associated aggressive B-cell lymphomas Celgene and Roche; consultation or advisory role in according to the 2017 WHO criteria. An important Takeda, Jansen, Celgene, and Gilead/Kyte; and travel message to retrieve from this study is that new enti- and accommodations expenses from Roche, Takeda, ties such as Burkitt-like lymphoma with 11q aberra- Celgene, Jansen, and Gilead. A.M. has received a tion and HGBL-DH/TH can be found in the HIV- speaker honorarium and travel grants from Roche, setting, with similar frequencies to those observed in Abbvie, and Janssen and has participated in advisory the immunocompetent population. Some similarities boards for Roche, Abbvie, and Janssen. M.P. discloses and differences were observed between HIV-associated advisory board BMS, Astra Zeneca, Takeda, MSD, and DLBCL and general DLBCL. MUM1 in BL and MYC/ Roche. P.A. has received honoraria from Janssen, BCL2 coexpression in DLBCL were found to be mark- Roche, Celgene, and Abbvie. J.M.S. discloses hono- ers with prognostic significance in HIV-infected raria from Roche, Janssen, Celgene, Novartis, Gilead, patients. Therefore, the diagnostic criteria and some Takeda, Servier, and Mundipharma; consultation or of the prognostic markers used for the study of advisory role in Roche, Janssen, Celgene, Novartis, aggressive B-NHL in immunocompetent patients Gilead, Celltrion, and Sandoz; speakers’ bureau in should also be taken into account when these lym- Roche and Celgene; and travel and accommodations phomas occur in the setting of HIV infection. expenses from Roche. J.M.R. discloses honoraria and consultation or advisory role in Celgene, Novartis, Takeda, Servier, Incyte, Amgen, and Pfizer; speakers’ Acknowledgements bureau in Amgen and Pfizer; and travel and accom- modations expenses from Amgen, Pfizer, and Incyte. The authors thank the staff from the Department of J.T.N. discloses honoraria from Novartis; consultation Pathology of Hospital Germans Trias i Pujol and from or advisory role in EUSA Pharma; research funding the Biobanking of tumour samples from Institut Ger- from Gilead and EUSA Pharma; and travel and mans Trias i Pujol for their collaboration. This work accommodation expenses from Roche. The remaining was supported in part by Instituto de Salud Carlos III, authors report no conflicts of interest. Ministerio de Economia y Competividad, Spain (PI19/ 01588); Gilead Sciences S.L., Spain (GLD19/00121); 2017 SGR288 (GRE) from AGAUR Programme/Gen- eralitat de Catalunya, and by funds from Josep Car- Data availability statement reras International Foundation and “la Caixa” The data that support the findings of this study are Foundation.” available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Authors contributions M-J.B, G.T. J-L.M., and J-T.N. designed the study; M- References J.B., G.T., and J-T.N. wrote the article; A-M.M-M. and 1. Barta SK, Samuel MS, Xue X et al. Changes in the influence of J.M. performed laboratory methods: M-J.B., J.M., and lymphoma- and HIV-specific factors on outcomes in AIDS- O.G. performed statistical analyses; S.M., J. G., M.C., related non-Hodgkin lymphoma. Ann. Oncol. 2015; 26; 958– A.M., L.V., A.M-T., T. A., J.M., M.J., A.F., M.A., J.B., 966. E.G-B., F.C., A.M., J-M.M., M.P., P.A., E. A., L., C.G-B, 2. Baptista MJ, Garcia O, Morgades M et al. HIV-infection impact M.G-C, J-M.S., and J-M. R. provided patient data and on clinical-biological features and outcome of diffuse large B- cell lymphoma treated with R-CHOP in the combination samples. antiretroviral therapy era. AIDS 2015; 29; 811–818. 3. Dolcetti R, Gloghini A, Caruso A, Carbone A. A lymphoma- genic role for HIV beyond immune suppression? Blood 2016; Conflict of Interest 127; 1403–1409. 4. Giagulli C, Marsico S, Magiera AK et al. Opposite effects of J.G. has received honoraria from Abbvie, Acerta/ HIV-1 p17 variants on PTEN activation and cell growth in B Astra Zeneca, Gilead, and Janssen; and grant funding cells. PLoS ONE 2011; 6; e17831. Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Characterisation of HIV-associated aggressive NHL 839 5. Linke-Serins€ oz E, Fend F, Quintanilla-Martinez L. Human with diffuse large B-cell lymphoma treated with cyclophos- immunodeficiency virus (HIV) and Epstein-Barr virus (EBV) phamide, doxorubicin, vincristine, and prednisone plus ritux- related lymphomas, pathology view point. Semin. Diagn. Pathol. imab. Cancer 2012; 118; 4173–4183. 2017; 34; 352–363. 21. Chen B J, Fend F, Campo E, Quintanilla-Martinez L. Aggressive 6. Wright G, Tan B, Rosenwald A, Hurt EH, Wiestner A, Staudt B-cell lymphomas—From morphology to molecular pathogene- LM. A gene expression-based method to diagnose clinically dis- sis. Annals of Lymphoma; Vol 3 (January 2019): Annals of Lym- tinct subgroups of diffuse large B cell lymphoma. Proc. Natl. phoma 2019; 3. http://aol.amegroups.com/article/view/4949 Acad. Sci. U. S. A. 2003; 100; 9991–9996. (accessed 1 Jan2019), 3. 7. Swerdlow S, Campo E, Harris N et al. eds. WHO classification of 22. Sesques P, Johnson NA. Approach to the diagnosis and treat- tumours of haematopoietic and lymphoid tissues. revised. 4th ed. ment of high-grade B-cell lymphomas with MYC and BCL2 Lyon, France: IARC Press, 2017. and/or BCL6 rearrangements. Blood 2017; 129; 280–288. 8. Swerdlow SH, Campo E, Harris NL et al. eds. WHO classification 23. Masque-Soler N, Szczepanowski M, Kohler CW et al. Clinical of tumours of haematopoietic and lymphoid tissues. Lyon, France: and pathological features of Burkitt lymphoma showing IARC Press, 2008. expression of BCL2--an analysis including gene expression in 9. Hans CP, Weisenburger DD, Greiner TC et al. Confirmation of formalin-fixed paraffin-embedded tissue. Br. J. Haematol. 2015; the molecular classification of diffuse large B-cell lymphoma by 171; 501–508. immunohistochemistry using a tissue microarray. Blood 2004; 24. Chuang S-S, Ye H, Du M-Q et al. Histopathology and immuno- 103; 275–282. histochemistry in distinguishing Burkitt lymphoma from dif- 10. Hu S, Xu-Monette ZY, Balasubramanyam A et al. CD30 fuse large B-cell lymphoma with very high proliferation index expression defines a novel subgroup of diffuse large B-cell lym- and with or without a starry-sky pattern: a comparative study phoma with favorable prognosis and distinct gene expression with EBER and FISH. Am. J. Clin. Pathol. 2007; 128; 558– signature: a report from the international DLBCL rituximab- 564. CHOP consortium program study. Blood 2013; 121; 2715– 25. Nasr MR, Rosenthal N, Syrbu S. Expression profiling of tran- 2724. scription factors in B- or T-acute lymphoblastic leukemia/lym- 11. Mu~ noz-M armol AM, Sanz C, Tapia G, Marginet R, Ariza A, phoma and burkitt lymphoma: usefulness of PAX5 Mate JL. MYC status determination in aggressive B-cell lym- immunostaining as pan-pre-B-cell marker. Am. J. Clin. Pathol. phoma: the impact of FISH probe selection. Histopathology 2010; 133; 41–48. 2013; 63; 418–424. 26. Gualco G, Queiroga EM, Weiss LM, Klumb CEN, Harrington 12. Cheson BD, Pfistner B, Juweid ME et al. Revised response crite- WJ, Bacchi CE. Frequent expression of multiple myeloma 1/in- ria for malignant lymphoma. J. Clin. Oncol. 2007; 25; 579– terferon regulatory factor 4 in Burkitt lymphoma. Hum. Pathol. 586. 2009; 40; 565–571. 13. Ribera J-M, Garcıa O, Grande C et al. Dose-intensive 27. Ye Z, Xiao Y, Shi H et al. Sporadic Burkitt lymphoma in south- chemotherapy including rituximab in Burkitt’s leukemia or ern China: 12 years’ experience in a single institution in lymphoma regardless of human immunodeficiency virus infec- Guangzhou. J. Clin. Pathol. 2011; 64; 1132–1135. tion status: final results of a phase 2 study (Burkimab). Cancer 28. Carbone A, Gloghini A, Larocca LM et al. Expression profile of 2013; 119; 1660–1668. MUM1/IRF4, BCL-6, and CD138/syndecan-1 defines novel his- 14. Johnson NA, Slack GW, Savage KJ et al. Concurrent expression togenetic subsets of human immunodeficiency virus-related of MYC and BCL2 in diffuse large B-cell lymphoma treated lymphomas. Blood 2001; 97; 744–751. with rituximab plus cyclophosphamide, doxorubicin, vin- 29. Satou A, Asano N, Kato S et al. Prognostic impact of MUM1/ cristine, and prednisone. J. Clin. Oncol. 2012; 30; 3452–3459. IRF4 expression in Burkitt lymphoma (BL): a reappraisal of 88 15. Tapia G, Lopez R, Mu~ noz-Marmol AM et al. Immunohisto- BL patients in Japan. Am. J. Surg. Pathol. 2017; 41; 389–395. chemical detection of MYC protein correlates with MYC gene 30. Horn H, Kalmbach S, Wagener R et al. A diagnostic approach status in aggressive B cell lymphomas. Histopathology 2011; to the identification of Burkitt-like lymphoma with 11q aberra- 59; 672–678. tion in aggressive B-cell lymphomas. Am. J. Surg. Pathol. 2021; 16. Hinkle C, Makar GS, Brody JD, Ahmad N, Zhu GG. HIV- 45; 356–364. associated ‘double-hit’ lymphoma of the tonsil: a first reported 31. Gonzalez-Farre B, Ramis-Zaldivar JE, Salmeron-Villalobos J case. Head Neck Pathol. 2020; 14; 1129–1133. et al. Burkitt-like lymphoma with 11q aberration: a germinal 17. Chapman JR, Bouska AC, Zhang W et al. EBV-positive HIV- center-derived lymphoma genetically unrelated to Burkitt lym- associated diffuse large B cell lymphomas are characterized by phoma. Haematologica 2019; 104; 1822–1829. JAK/STAT (STAT3) pathway mutations and unique clinico- 32. Wagener R, Seufert J, Raimondi F et al. The mutational pathologic features. Br. J. Haematol. 2021; 194; 870–878. landscape of Burkitt-like lymphoma with 11q aberration is dis- 18. Ferreri AJM, Cattaneo C, Lleshi A et al. A dose-dense short- tinct from that of Burkitt lymphoma. Blood 2019; 133; 962– term therapy for human immunodeficiency virus/acquired 966. immunodeficiency syndrome patients with high-risk Burkitt 33. Rosenwald A, Bens S, Advani R et al. Prognostic significance lymphoma or high-grade B-cell lymphoma: safety and efficacy of MYC rearrangement and translocation partner in diffuse results of the ‘CARMEN’ phase II trial. Br. J. Haematol. 2021; large B-cell lymphoma: a study by the Lunenburg lymphoma 192; 119–128. biomarker consortium. J. Clin. Oncol. 2019; 37; 3359–3368. 19. Gebauer N, Witte HM, Merz H et al. Aggressive B-cell lym- 34. Tzankov A, Xu-Monette ZY, Gerhard M et al. Rearrangements phoma cases with 11q aberration patterns indicate a spectrum of MYC gene facilitate risk stratification in diffuse large B-cell beyond Burkitt-like lymphoma. Blood Adv. 2021; 5; 5220– lymphoma patients treated with rituximab-CHOP. Mod. Pathol. 5225. 2014; 27; 958–971. 20. Akyurek N, Uner A, Benekli M, Barista I. Prognostic signifi- 35. Copie-Bergman C, Cuilliere-Dartigues P, Baia M et al. MYC-IG cance of MYC, BCL2, and BCL6 rearrangements in patients rearrangements are negative predictors of survival in DLBCL Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840. 13652559, 2022, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/his.14798 by Readcube (Labtiva Inc.), Wiley Online Library on [02/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 840 M J Baptista et al. patients treated with immunochemotherapy: a GELA/LYSA 50. Xu J, Oki Y, Saksena A et al. CD30 expression and prognostic study. Blood 2015; 126; 2466–2474. significance in R-EPOCH-treated patients with diffuse large B- 36. Barrans SL, O’Connor SJM, Evans PAS et al. Rearrangement of cell lymphoma. Hum. Pathol. 2017; 60; 160–166. the BCL6 locus at 3q27 is an independent poor prognostic fac- 51. Slack GW, Steidl C, Sehn LH, Gascoyne RD. CD30 expression tor in nodal diffuse large B-cell lymphoma. Br. J. Haematol. in de novo diffuse large B-cell lymphoma: a population-based 2002; 117; 322–332. study from British Columbia. Br. J. Haematol. 2014; 167; 608– 37. Iqbal J, Greiner TC, Patel K et al. Distinctive patterns of BCL6 617. molecular alterations and their functional consequences in dif- 52. Salas MQ, Climent F, Tapia G et al. Clinicopathologic features ferent subgroups of diffuse large B-cell lymphoma. Leukemia and prognostic significance of CD30 expression in de novo dif- 2007; 21; 2332–2343. fuse large B-cell lymphoma (DLBCL): results in a homogeneous 38. Scott DW, Mottok A, Ennishi D et al. Prognostic significance of series from a single institution. Biomarkers 2020; 25; 69–75. diffuse large B-cell lymphoma cell of origin determined by digi- 53. van der Weyden CA, Pileri SA, Feldman AL, Whisstock J, tal gene expression in formalin-fixed paraffin-embedded tissue Prince HM. Understanding CD30 biology and therapeutic tar- biopsies. J. Clin. Oncol. 2015; 33; 2848–2856. geting: a historical perspective providing insight into future 39. Staiger AM, Ziepert M, Horn H et al. Clinical impact of the directions. Blood Cancer J. 2017; 7; e603. cell-of-origin classification and the MYC/ BCL2 dual expresser 54. Morton LM, Kim CJ, Weiss LM et al. Molecular characteristics status in diffuse large B-cell lymphoma treated within prospec- of diffuse large B-cell lymphoma in human immunodeficiency tive clinical trials of the German high-grade non-Hodgkin’s virus-infected and -uninfected patients in the pre-highly active lymphoma study group. J. Clin. Oncol. 2017; 35; 2515–2526. antiretroviral therapy and pre-rituximab era. Leuk. Lymphoma 40. Chadburn A, Chiu A, Lee JY et al. Immunophenotypic analysis 2014; 55; 551–557. of AIDS-related diffuse large B-cell lymphoma and clinical impli- 55. Horn H, Ziepert M, Becher C et al. MYC status in concert with cations in patients from AIDS malignancies consortium clinical BCL2 and BCL6 expression predicts outcome in diffuse large B- trials 010 and 034. J. Clin. Oncol. 2009; 27; 5039–5048. cell lymphoma. Blood 2013; 121; 2253–2263. 41. Chao C, Silverberg MJ, Xu L et al. A comparative study of 56. Ott G, Rosenwald A, Campo E. Understanding MYC-driven molecular characteristics of diffuse large B-cell lymphoma from aggressive B-cell lymphomas: pathogenesis and classification. patients with and without human immunodeficiency virus Blood 2013; 122; 3884–3891. infection. Clin. Cancer Res. 2015; 21; 1429–1437. 57. Hwang J, Suh C, Kim K et al. The incidence and treatment 42. Xicoy B, Ribera J-M, Mate J-L et al. Immunohistochemical response of double expression of MYC and BCL2 in patients expression profile and prognosis in patients with diffuse large with diffuse large B-cell lymphoma: a systematic review and B-cell lymphoma with or without human immunodeficiency meta-analysis. Cancers (Basel) 2021; 13; 3369. virus infection. Leuk. Lymphoma 2010; 51; 2063–2069. 58. Green TM, Young KH, Visco C et al. Immunohistochemical 43. Dunleavy K, Little RF, Pittaluga S et al. The role of tumor his- double-hit score is a strong predictor of outcome in patients togenesis, FDG-PET, and short-course EPOCH with dose-dense with diffuse large B-cell lymphoma treated with rituximab plus rituximab (SC-EPOCH-RR) in HIV-associated diffuse large B-cell cyclophosphamide, doxorubicin, vincristine, and prednisone. J. lymphoma. Blood 2010; 115; 3017–3024. Clin. Oncol. 2012; 30; 3460–3467. 44. Madan R, Gormley R, Dulau A et al. AIDS and non-AIDS dif- 59. Ramos JC, Sparano JA, Chadburn A et al. Impact of Myc in fuse large B-cell lymphomas express different antigen profiles. HIV-associated non-Hodgkin lymphomas treated with EPOCH Mod. Pathol. 2006; 19; 438–446. and outcomes with vorinostat (AMC-075 trial). Blood 2020; 45. Hoffmann C, Tiemann M, Schrader C et al. AIDS-related B-cell 136; 1284–1297. lymphoma (ARL): correlation of prognosis with differentiation profiles assessed by immunophenotyping. Blood 2005; 106; 1762–1769. Supporting Information 46. Baptista MJ, Tapia G, Morgades M et al. Using the Lymph2Cx assay for assessing cell-of-origin subtypes of HIV-related diffuse Additional Supporting Information may be found in large B-cell lymphoma. Leuk. Lymphoma 2019; 60; 1087–1091. the online version of this article: 47. Philippe L, Lancar R, Laurent C et al. In situ BCL2 expression is an independent prognostic factor in HIV-associated DLBCL, a Figure S1. Impact of BCL6 rearrangements on out- LYMPHOVIR cohort study. Br. J. Haematol. 2020; 188; 413– 423. comes of patients with HIV-associate nongerminal 48. Chao C, Silverberg MJ, Martınez-Maza O et al. Epstein-Barr center diffuse large B-cell lymphoma (DLBCL). (a) virus infection and expression of B-cell oncogenic markers in Kaplan–Meier curves for overall survival of HIV-asso- HIV-related diffuse large B-cell lymphoma. Clin. Cancer Res. ciate nongerminal center (non-GC) DLBCL according 2012; 18; 4702–4712. to the presence of BCL6 rearrangements. (b) Kaplan– 49. Marques-Piubelli ML, Salas YI, Pachas C, Becker-Hecker R, Vega F, Miranda RN. Epstein-Barr virus-associated B-cell lym- Meier curves for progression free survival of HIV-asso- phoproliferative disorders and lymphomas: a review. Pathology ciate nongerminal center (non-GC) DLBCL according 2020; 52; 40–52. to the presence of BCL6 rearrangements. Ó 2022 The Authors. Histopathology published by John Wiley & Sons Ltd., Histopathology, 81, 826–840.