Patients with HIV infection have an overall increased risk of developing lymphomas . The most common type of HIV-related lymphoma is diffuse large B-cell lymphoma (DLBCL). Burkitt lymphoma, plasmablastic lymphoma (PBL), and primary effusion lymphoma (PEL) usually occur at lower frequencies; T-cell lymphomas and small B-cell lymphomas, including follicular lymphoma, are rarely associated with HIV infection .
PBL is a clinically aggressive lymphoma that often has a diffuse histologic pattern and is composed of large neoplastic cells that have a plasmablastic immunophenotype, that is, negative for CD20 and PAX5, and positive for CD138, MUM1/IRF4, and other plasma cell-associated markers . PBL was initially described by Delecluse et al . All cases in that study occurred in HIV+ patients and the oral cavity was the most common site. Subsequent studies demonstrated that PBL is not exclusive to the HIV+ population, and it can arise in other settings of immunodeficiency or in apparently immunocompetent, usually elderly patients, and can involve a wide variety of anatomic sites . The prognosis of PBL is generally poor with a median survival of less than 12 months . However, in earlier studies, many were HIV patients who were not treated with HAART. Thus, it is uncertain whether the poor prognosis of PBL patients in the HIV setting is due to the aggressive behavior of lymphoma or lack of effective control of HIV.
The aim of this study is to characterize the clinicopathological features of PBL occurring in HIV+ patients in the era of HAART from a single health center. We also describe the immunophenotypic features of these neoplasms and sought to assess for any prognostic parameters in HIV+ patients with PBL.
Materials and methods
We searched our pathology archives for cases with a diagnosis of HIV infection and B-cell lymphoma over an interval from 1 January 2008 to 31 December 2018. We then focused on cases of PBL and collected clinical information including laboratory parameters, Ann Arbor stage, treatment regimens, disease progression, and follow-up data. In addition, we reviewed pathologic, immunophenotypic, and cytogenetic findings. Lumbar puncture and evaluation of the cerebrospinal fluid was not routinely performed. Of note, all patients in this study were seen at one hospital in our health center network; this hospital cares for low-income communities. This study was approved by the Institutional Review Board of The University of Texas Health Center.
Histological and immunohistochemical studies
Smears were prepared from fine needle aspirate specimens of lymph nodes or soft tissue masses and stained with Diff-Quick or Papanicolaou or both (most often). Bone marrow aspirate smears were stained with Wright Giemsa. Bone marrow and tissue biopsy specimens were fixed in formalin, paraffin-embedded, and 4-μm-thick tissue sections were stained with hematoxylin-eosin.
Immunohistochemical studies and in-situ hybridization for Epstein–Barr virus (EBV)-encoded RNA (EBER) and immunoglobulin kappa and lambda light chains were performed using formalin-fixed paraffin-embedded tissue sections with appropriate controls. The antibodies used for immunohistochemistry included: CD3, CD20, CD30, CD34, CD79a, CD138, PAX5, MUM-1/IRF4, CD79a, TDT, ALK-1, EMA, and HHV-8. All antibodies and reagents were from Ventana (Roche, Basel, Switzerland). The staining dilutions and protocols recommended by the manufacturer were followed.
The diagnostic criteria of the current WHO classification of lymphoid neoplasms were used to refine the subtypes of aggressive non-Hodgkin lymphomas in this case series. Cases with plasmablastic morphology, but with strong expression of B-cell markers (such as CD20 and PAX5), were classified as DLBCL-NOS, not PBL .
Flow cytometric analysis
Flow cytometry immunophenotypic analysis was performed on lymph node biopsy samples and bone marrow aspirate specimens when available. These studies were performed using standard multicolor analysis. The panels usually included antibodies specific to: CD2, CD3, CD4, CD5, CD7, CD10, CD11c, CD19, CD20, CD23, CD25, CD30, CD38, CD43, CD45, CD56, CD57, FMC-7, HLA-DR, CD138, kappa, lambda, cytoKappa, and cytoLambda.
Conventional karyotyping analysis was performed on G-banded metaphase cells prepared from unstimulated tissue samples or stimulated bone marrow aspirate cultures (72 h) using standard methods. Twenty metaphase cells were analyzed for conventional cytogenetics studies. Karyotypes were reported following the current International System for Human Cytogenetic Nomenclature. Fluorescent in situ hybridization (FISH) analysis was performed on nondecalcified bone marrow specimens or tissue sections of formalin-fixed paraffin-embedded lymph node or soft tissue specimens according to standard protocols. DNA break apart probes for BCL2, BCL6 and MYC (Leica; Kreatech Inc, Buffalo Grove, Illinois, USA) were employed. Two hundred interphase nuclei were examined for each probe.
The statistical software of GraphPad Prism 7 (GraphPad Software, San Diego, California, USA) was used. Univariate survival analysis was performed to determine statistical significance. Data for continuous variables were described as median and range. Results were statistically significant when the P value was less than 0.05 for all analyses.
In our archives, we identified 95 cases of HIV-associated aggressive non-Hodgkin lymphoma over the study interval. These neoplasms were classified as DLBCL (n = 40; 42%), PBL (n = 21; 22%), high-grade B-cell lymphoma (n = 17; 18%), Burkitt lymphoma (n = 10; 11%), PEL (n = 4; 4%), B lymphoblastic lymphoma (n = 2; 2%), and angioimmunoblastic T-cell lymphoma (n = 1; 1%). Among the PBL patients, their median age at diagnosis was 45 years (range, 28–60) with 19 men and two women. The median HIV viral load, CD4+ cell count, and CD4+ : CD8+ ratio at the time of PBL diagnosis were 90 000 copies/ml (range, 2–4530 000), 115/μl (range, 1–301), and 0.19 (range, 0.01–0.40), respectively (Table 1). There were two (9.5%) patients with a viral load less than 50 copies/ml.
Ten (48%) patients with PBL had lymphadenopathy, including five involving the head and neck, three inguinal, and two retroperitoneal regions (Table 1). A mass in the oral cavity or sinonasal region was the second common site, in six (29%) patients, followed by rectal masses in five (24%). Two (9.5%) of these patients also had disseminated disease involving the brain, kidneys, liver, anus, spinal cord, other regional lymph nodes, and one (4.8%) patient had cutaneous involvement. Stage based on clinical and radiologic studies was available for 19 patients: seven (37%) patients had stage I/II disease and 12 (63%) patients had stage III/IV disease (Tables 1 and 2). Fifteen patients underwent bone marrow examination; five (33%) patients had disease involving the bone marrow and blood with a leukemic presentation (Table 2).
Seven (33%) patients had PBL at the time of the initial HIV diagnosis. These patients were not on HAART and their median viral load, CD4+ cell count, and CD4+ : CD8+ ratio were 972 746 copies/ml (range, 20–4530 000), 143 cells/μl (range, 17–271), and 0.25 (range, 0.05–0.40), respectively. Fourteen (67%) patients developed PBL after their HIV diagnosis with a median interval of 7.7 months. In this subgroup, two did not receive HAART therapy and 12 were on HAART therapy; however, only two of these patients were evaluated to be compliant with HAART therapy. The median viral load, CD4+ cell count, and CD4+ : CD8+ ratio were 429,163 copies/ml (range, 2–3200 000), 138 cells/μl (range, 1–301), and 0.25 (range, 0.03–0.33), respectively, in these 14 patients. There is no significant difference between these two groups in terms of viral load, CD4+ cell count, and CD4+/CD8+ ratio. After the diagnosis of PBL, all the patients received HAART, except for one patient who refused treatment.
Lymph nodes involved by PBL were enlarged with complete and diffuse effacement of the nodal architecture. All cases showed intermediate to large lymphoma cells with immunoblastic or plasmablastic morphology; specifically, the lymphoma cells had large central or eccentrically located nuclei with fine to vesicular chromatin and prominent nucleoli, and abundant basophilic cytoplasm; some tumors cells had cytoplasmic vacuoles (Figs. 1 and 2). Mitotic figures were numerous along with frequent apoptotic bodies, karyorrhexis, and foci of necrosis (Fig. 1). In some cases, the presence of tangible body macrophages created a ‘starry sky’ appearance. Similar findings were observed in most extranodal sites. In cases with bone marrow involvement, lymphoma cells showed an interstitial growth pattern; no sheets of lymphoma cells or confluent necrosis were observed in the bone marrow (Fig. 2).
Immunohistochemical studies showed that the lymphoma cells were positive for MUM-1/IRF4 (19/19; 100%), CD138 (18/20; 90%), CD45 (10/16; 63%), CD79a (8/17; 47%), EMA (4/9, 44%), CD4 (4/12; 33%), CD10 (5/16; 31%), CD30 (3/12; 25%), BCL2 (2/9; 22%), BCL6 (2/10; 20%), and CD56 (3/18; 17%). CD20 and PAX5 stains were focal and weakly positive in two cases individually. Proliferation rate as assessed by Ki67 stain was at least 90% in 18 of 20 (90%) and the remaining two (2/20) showed ∼80%. The lymphoma cells were negative for CD3 (n = 20), cyclin D1 (n = 6), ALK (n = 15), and HHV8 (n = 15). In-situ hybridization analysis showed that EBER was positive in all 19 (100%) cases assessed. Kappa or lambda light chain restriction was seen in 10/18 (56%) cases (five kappa, five lambda).
Conventional karyotyping analysis was performed on the bone marrow aspirates of 11 patients. Nine (82%) cases demonstrated a normal karyotype and in two cases there was no growth in culture. FISH analysis was performed on the diagnostic tissue in four PBL cases; MYC rearrangement was positive in two out of four (50%), inconclusive in one out of four (25%), and one had gains of bcl-6/3q27, MYC/8q24, and bcl-6/18q21.
Treatment and outcome
Eighteen of 21 (86%) patients were treated with chemotherapy, including 13 patients with EPOCH (etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin); and two with CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone); one with cyclophosphamide, vincristine, and prednisone; one with methotrexate, cytarabine, thiotepa, rituximab; and one with hyper CVAD (cyclophosphamide, vincristine, doxorubicin, and dexamethasone). The remaining three patients did not receive chemotherapy. Twelve (67%) patients had opportunistic infections including parasites, viruses, or bacteria during and after chemotherapy. With a median follow-up of 19 months (range, <1–112), nine patients died and 12 were alive at the time of the last clinical follow-up (Table 2).
Although a small patient cohort, we identified parameters associated with a poorer prognosis. Patients with bone marrow involvement had a median overall survival (OS) of 4.7 months, compared with than patients without bone marrow disease (median OS not reached) (P = 0.015) (Fig. 3). Patients with an elevated serum lactate dehydratase (LDH)-level more than 300 also showed a poorer OS (P = 0.0301). We also observed two trends that were not statistically significant. Patients with stage III/IV disease appear to have a poorer survival (P = 0.2182). Patients with a CD4+ T-cell count of more than 200 cells/μl have a seemingly unfavorable survival versus patients with a CD4+ T-cell count of less than 200 cells/μl at the time of diagnosis, P = 0.1863. An important negative result regarding prognosis is that there was no difference in survival for patients with PBL who were diagnosed concurrently with the initial diagnosis of HIV infection versus patients who developed PBL after their HIV diagnosis (P = 0.8280).
In patients with HIV-associated lymphomas, PBL is the second most common type in the era of HAART therapy at our institution. PBL represented ∼22% of HIV-associated aggressive B-cell lymphomas. Most patients were men with a median age of 45 years, similar to the findings in earlier studies . Recent reports by the Centers for Disease Control have shown that gay and bisexual men account for more than 65% of all HIV+ patients as well as more than 80% of all new HIV+ patients each year . These factors likely explain the predilection for the men in this study.
The incidence of HIV-associated non-Hodgkin lymphomas has decreased substantially following the introduction of HAART therapy . Previous studies showed that the most frequent B-cell lymphoma in HIV+ patients was DLBCL, followed by Burkitt lymphoma [9,10]. In the current study, we show a higher frequency of PBL in the era of HAART therapy, second only to DLBCL. This increased frequency of PBL possibly may be attributable to the relatively recent recognition of PBL, initially described in 1997 and included as a new entity in the WHO classification in 2008. Prior to this time point, PBL cases were included in the DLBCL category .
In this case series, one-third of patients were diagnosed with PBL and HIV simultaneously, and the remaining two-thirds of patients developed PBL after their HIV diagnoses. Only two of 12 (17%) patients adhered to their HAART regimen. Patients with PBL had a much lower compliance rate with HAART therapy compared with patients with other types of aggressive B-cell lymphoma in our experience, about 50%. Furthermore, our health system cares for low-income communities and the patients have a lower HAART compliance rate than that shown in other studies, such as 62% described by Ortego et al., and far lower than the required adherence rate about 95% for an optimal outcome for patients with HIV infection . Most patients in our case series were poor and uninsured, which might be the reason why there is a high proportion of patients with a late HIV diagnosis and their noncompliance to HAART therapy. Therefore, the high incidence of PBL may also be associated with lack of optimal HIV control in this patient cohort.
The pathogenesis of PBL is not fully understood. EBV is strongly linked with HIV-associated PBL, as demonstrated by the findings in this study in which all HIV-associated PBL cases were positive. EBV encodes several proteins including LMP-1, LMP-2A, BHRF-1, EBNA-3A, and EBNA-3C that are implicated in the development of non-Hodgkin lymphomas . These proteins are produced during the latent or active phases of infection. In immunocompetent patients, EBV reactivation can be eradicated by healthy cytotoxic T cells, whereas HIV+ patients have a compromised immune system and reactivation of EBV can lead to a lymphoproliferative neoplasm . LMP-1 induces the expression of Flice-like inhibitory protein (FLIP) and activates the nuclear factor κB pathway to prevent neoplastic cells from undergoing FAS-mediated apoptosis, thereby resulting in uncontrolled proliferation of neoplastic cells . LMP-2A mimics the B-cell receptor (BCR) and associates with Syk and Src to protect B cells from BCR-mediated apoptosis . BHRF-1 binds to BAX/BAK activators, preventing TGF-β-mediated apoptosis . Lastly, EBNA-3A and EBNA-3C prevent MYC-induced apoptosis . All these factors can potentially cooperate to induce the development of PBL in HIV+ patients. Furthermore, MYC is dysregulated in PBL secondary to MYC translocations or amplification; these genetic changes allow MYC to overcome the repression of B lymphocyte-induced maturation protein-1, a transcription factor important for plasma cell differentiation . MYC expression also provides the neoplastic cells with a proliferative and survival advantage.
PBL generally lacks expression of B-cell markers such as CD20 and PAX5 whereas these B-cell markers can be weakly positive in about ∼10% of cases . Similar to other studies , PBL in this cohort showed variable expression of CD30, CD45, CD79a, and EMA. The diagnosis of PBL can be challenging as it has overlapping features with other lymphomas with immunoblastic/plasmablastic features including plasmacytoma/plasma cell myeloma, ALK+ large B-cell lymphoma, and HHV8+ large B-cell lymphomas, the latter including extracavitary variant of PEL and HHV8+ large B-cell lymphoma, not otherwise specified. The morphological features of PBL also can closely resemble anaplastic plasmacytoma. PBL in the oral cavity and lymph nodes can be confused with extraosseous/extramedullary plasmacytoma. Clinical features that favor anaplastic plasmacytoma/myeloma over PBL include a serum monoclonal paraprotein, hypercalcemia, lytic bone lesions, and renal failure . In contrast, PBL is more often positive for EBV infection. Immunohistochemistry is essential to recognize ALK+ large B-cell lymphoma. These neoplasms are often associated with t (2;17) and less often t (2;5). Extracavitary or solid variant of PEL is always positive for HHV8 as is HHV8+ DLBCL not otherwise specified. In contrast, HHV8 is negative in PBL.
PBL is an aggressive and refractory lymphoma that poses a therapeutic challenge for clinicians. Due to its rarity and limited clinical trials, there is no consensus as to the standard of care for PBL. CHOP, dose adjusted (DA)-EPOCH, hyper-CVAD and CODOX-M/IVAC (cyclophosphamide, vincristine, doxorubicin, methotrexate/ifosfamide, etoposide, and cytarabine) are potential options and previously used to treat PBL patients [24,25]. In this study, about 60% of patients received the EPOCH regimen as National Comprehensive Cancer Network guidelines have deemed the traditional CHOP regimen as inadequate therapy for patients with PBL. Several reports showed some success in treating PBL patients with the proteasome inhibitor bortezomib, a common chemotherapeutic agent for patients with plasma cell myeloma . Additional studies shown efficacy using bortezomib with other chemotherapeutic agents, such as EPOCH, dexamethasone, gemcitabine, oxaliplatin, and cytarabine [27,28]. Autologous stem-cell transplantation also is a potential treatment choice . In summary, the optimal management for PBL patients is uncertain and needs to be further explored and developed.
Prognostic stratification is not well established in PBL patients. Loghavi et al. demonstrated that age, stage, and EBV status were significant prognostic parameters in a cohort study of 61 patients with PBL. By reviewing 13 patients with PBL in India, Rudresha et al. showed that PBL has an unfavorable prognosis and they suggested that the most important prognostic factors were Ann Arbor stage, Ki-67 proliferation index, and Eastern Cooperative Oncology Group (ECOG) performance status at the time of presentation. In our study, we determined that bone marrow involvement by PBL and an elevated serum LDH-level more than 300 were associated with poorer OS.
In conclusion, in the HAART era PBL is the second most frequent type of HIV-associated aggressive B-cell lymphoma in underserved patients, representing about 20% of all cases. Lymph nodes are the most common site of involvement and EBV infection of the lymphoma cells is very common, if not universal. Bone marrow involvement and high-serum LDH levels were significantly associated with a poorer prognosis in this cohort.
We acknowledge that Dr Karen W. Eldin of The University of Texas Health Science Center at Houston kindly helped us edit the article. No research funding was obtained for this study.
Author contributions: B.M., L.J.M., and Z.H. wrote the article. All authors provided data or contributed to the discussion and revision of the article.
Conflicts of interest
There are no conflicts of interest.
1. Grulich AE, van Leeuwen MT, Falster MO, Vajdic CM. Incidence of cancers in people with HIV/AIDS compared with immunosuppressed transplant recipients: a meta-analysis
2. Grogg KL, Miller RF, Dogan A. HIV infection and lymphoma
. J Clin Pathol
3. Campo E, Stein H, Harris NL. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, et al. Plasmablastic lymphoma
. WHO classification of tumours of haematopoietic and lymphoid tissues
4th ed.Lyon, France:IARC Press; 2008. 321322.
4. Delecluse HJ, Anagnostopoulos I, Dallenbach F, Hummel M, Marafioti T, Schneider U, et al. Plasmablastic lymphomas of the oral cavity: a new entity associated with the human immunodeficiency virus infection
5. Liu JJ, Zhang L, Ayala E, Field T, Ochoa-Bayona JL, Perez L, et al. Human immunodeficiency virus (HIV)-negative plasmablastic lymphoma: a single institutional experience and literature review
. Leuk Res
6. Castillo JJ, Furman M, Beltran BE, Bibas M, Bower M, Chen W, et al. Human immunodeficiency virus-associated plasmablastic lymphoma: poor prognosis in the era of highly active antiretroviral therapy
7. Castillo JJ, Bibas M, Miranda RN. The biology and treatment of plasmablastic lymphoma
8. CDCMonitoring selected national HIV prevention and care objectives by using HIV surveillance data – United States and 6 dependent areas, 2016
. HIV, surveillance supplemental report
2018; Atlanta, Georgia:CDC, 23–24.
9. Said J, Cesarman E, Rosenwald A, Harris NL. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, et al. Lymphomas associated with HIV infection
. WHO classification of tumours of haematopoietic and lymphoid tissues
. Lyon, France:IARC Press; 2008. 449452.
10. Levine AM. AIDS-related malignancies: the emerging epidemic
. J Natl Cancer Inst
11. Kumar S, Kimlinger T, Morice W. Immunophenotyping in multiple myeloma and related plasma cell disorders
. Best Pract Res Clin Haematol
12. Ortego C, Huedo-Medina TB, Vejo J, Llorca FJ. Adherence to highly active antiretroviral therapy in Spain. A meta-analysis
. Gac Sanit
13. Paterson DL, Swindells S, Mohr J, Brester M, Vergis EN, Squier C, et al. Adherence to protease inhibitor therapy and outcomes in patients with HIV infection
. Ann Intern Med
14. Castillo JJ, Reagan JL. Plasmablastic lymphoma: a systematic review
15. Maurmann S, Fricke L, Wagner HJ, Schlenke P, Hennig H, Steinhoff J, et al. Molecular parameters for precise diagnosis of asymptomatic Epstein–Barr virus reactivation in healthy carriers
. J Clin Microbiol
16. Kilger E, Kieser A, Baumann M, Hammerschmidt W. Epstein–Barr virus-mediated B-cell proliferation is dependent upon latent membrane protein 1, which simulates an activated CD40 receptor
. EMBO J
17. Bultema R, Longnecker R, Swanson-Mungerson M. Epstein–Barr virus LMP2A accelerates MYC-induced lymphomagenesis
18. Khanim F, Dawson C, Meseda CA, Dawson J, Mackett M, Young LS. BHRF1, a viral homologue of the Bcl-2 oncogene, is conserved at both the sequence and functional level in different Epstein–Barr virus isolates
. J Gen Virol
1997; 78 (Pt 11):29872999.
19. Cooper A, Johannsen E, Maruo S, Cahir-McFarland E, Illanes D, Davidson D, et al. EBNA3A association with RBP-Jkappa down-regulates c-myc and Epstein–Barr virus-transformed lymphoblast growth
. J Virol
20. Castillo JJ, Beltran BE, Miranda RN, Paydas S, Winer ES, Butera JN. Epstein–barr virus-positive diffuse large B-cell lymphoma of the elderly: what we know so far
21. Montes-Moreno S, Gonzalez-Medina AR, Rodriguez-Pinilla SM, Maestre L, Sanchez-Verde L, Roncador G, et al. Aggressive large B-cell lymphoma with plasma cell differentiation: immunohistochemical characterization of plasmablastic lymphoma and diffuse large B-cell lymphoma with partial plasmablastic phenotype
22. Lopez A, Abrisqueta P. Plasmablastic lymphoma: current perspectives
. Blood Lymphat Cancer
23. McKenna RW, Kyle RA, Kuehl WM, Harris NL, Coupland RW, Fend F. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, et al. Plasma cell neoplasms
. WHO classification of tumours of haematopoietic and lymphoid tissues
. Lyon, France:IARC Press; 2017. 241258.
24. Loghavi S, Alayed K, Aladily TN, Zuo Z, Ng SB, Tang G, et al. Stage, age, and EBV status impact outcomes of plasmablastic lymphoma patients: a clinicopathologic analysis of 61 patients
. J Hematol Oncol
25. Pretscher D, Kalisch A, Wilhelm M, Birkmann J. Refractory plasmablastic lymphoma-a review of treatment options beyond standard therapy
. Ann Hematol
26. Saba NS, Dang D, Saba J, Cao C, Janbain M, Maalouf B, et al. Bortezomib in plasmablastic lymphoma: a case report and review of the literature
27. Bibas M, Grisetti S, Alba L, Picchi G, Del Nonno F, Antinori A. Patient with HIV-associated plasmablastic lymphoma responding to bortezomib alone and in combination with dexamethasone, gemcitabine, oxaliplatin, cytarabine, and pegfilgrastim chemotherapy and lenalidomide alone
. J Clin Oncol
28. Spina M, Chimienti E, Martellotta F, Vaccher E, Berretta M, Zanet E, et al. Phase 2 study of intrathecal, long-acting liposomal cytarabine in the prophylaxis of lymphomatous meningitis in human immunodeficiency virus-related non-Hodgkin lymphoma
29. Al-Malki MM, Castillo JJ, Sloan JM, Re A. Hematopoietic cell transplantation for plasmablastic lymphoma: a review
. Biol Blood Marrow Transplant
30. Rudresha AH, Lakshmaiah KC, Agarwal A, Babu KG, Loknatha D, Jacob LA, et al. Plasmablastic lymphoma in immunocompetent and in immunocompromised patients: experience at a regional cancer centre in India
. South Asian J Cancer