Prognostic Factors and Outcome of Human Herpesvirus 8–Associated Primary Effusion Lymphoma in Patients With AIDS
http://www.100md.com
《临床肿瘤学》
the Departments of Clinical Immunology, Infectious Diseases, and Biostatistics, Hpital Saint-Louis, Assistance Publique-Hpitaux de Paris (AP-HP)
Department of Clinical Hematology, Hpital La Pitie Salpêtrière, AP-HP
Department of Pneumology, Hpital Tenon, AP-HP, Paris
Department of Infectious Diseases, Hpital d'Instruction des Armees Begin, Saint-Mande
Department of Infectious Diseases, Hpital Eugenie, Ajaccio, France
ABSTRACT
PURPOSE: Primary effusion lymphoma (PEL) is a rare high-grade B-cell non-Hodgkin's lymphoma associated with Kaposi sarcoma–associated herpesvirus/human herpesvirus 8 (KSHV/HHV-8) infection, and is mostly observed in the course of HIV infection. The prognosis is poor, with reported median survival time shorter than 6 months. To date, no prognostic factor has been identified in this subset of lymphoma.
PATIENTS AND METHODS: We describe here a large series of HIV-infected patients with PEL, including 28 cases diagnosed in six centers during an 11-year time period. Prognosis analysis was performed using a Cox proportional hazard regression model. Statistically significant covariates were further analyzed in a forward, stepwise multivariate model.
RESULTS: After a median follow-up of 3.8 years (range, 10 months to 10.8 years), nine patients (32%) were still alive, and eight of them remained progression free. The median survival was 6.2 months, and the 1-year overall survival rate was 39.3%. Fourteen patients (50%) achieved complete remission, with a 1-year disease-free survival rate at 78.6%. In a multivariate analysis, only a performance status more than 2 (hazard ratio, 5.84; 95% CI, 1.76 to 19.33) and the absence of highly active antiretroviral therapy (HAART) before PEL diagnosis (hazard ratio, 3.26; 95% CI, 1.14 to 9.34) were found to be independent predictors for shorter survival.
CONCLUSION: Based on a retrospective series of 28 patients, two prognostic factors were identified as being independently associated with impaired clinical outcome in HIV-related PEL—(1) a poor performance status and (2) the absence of HAART before PEL diagnosis.
INTRODUCTION
Kaposi sarcoma (KS) and non-Hodgkin's lymphoma (NHL) represent the most common malignancies associated with HIV infection. Most HIV-related NHL are of B-cell origin and high-grade histology. They are characterized by clinical aggressiveness, advanced stage, and high frequency of extranodal involvement.1 Viral agents, in particular human gamma-herpesviruses, might play a role in the development of some AIDS-related malignancies. Among virus-associated lymphoid disorders, primary effusion lymphoma (PEL) is a KS-associated herpesvirus/human herpesvirus 8 (KSHV/HHV-8) –associated NHL, which predominantly develops in serous body cavities resulting in recurrent lymphomatous effusions. PEL is an uncommon disease that accounts for approximately 3% of all HIV-related NHL.1 PEL tumor cells display large heterogeneous size and pleomorphic aspect and are coinfected with Epstein-Barr virus (EBV) in most cases. Although of B-cell origin, these cells frequently present a CD45+ CD30+ non-B non-T phenotype.2
Since the widespread use of highly active antiretroviral therapy (HAART), morbidity and mortality associated with HIV infection have shown impressive declines.3 Overall outcome of NHL has improved since the introduction of HAART.4,5 Nevertheless, despite extensive use of HAART and improvement of chemotherapy management in HIV-infected patients, the prognosis of patients with HIV-related PEL remains poor. The median survival does not exceed 6 months even in the most recent series. Moreover, only short series have been published,6-13 the largest including 15 patients. To our knowledge, no prognostic factor has been identified and the influence of HAART on survival has not been determined. To address this issue, we present data from the follow-up of a cohort series of 28 HIV-infected patients diagnosed with PEL during a 11-year period.
PATIENTS AND METHODS
Patients
The study included all consecutive HIV-infected patients with PEL diagnosed between January 1, 1993, and September 30, 2003 at the following hospitals in France : Hpital Saint-Louis, Paris (20 cases); Hpital La Pitie-Salpêtrière, Paris (four cases); Htel-Dieu, Paris (one case); Hpital Tenon, Paris (one case); Hpital Begin, Saint-Mande (one case); and Hpital Eugenie, Ajaccio (one case). All patient records were reviewed by the same examiner (E.B.). Since May 1, 1999, data have been collected prospectively. Among these 28 cases, 17 have previously been reported.14-17
Data recorded at diagnosis were demographic features (age, sex, and geographic origin), previous AIDS-defining illnesses including history of KS, pre-existing multicentric Castleman disease, previous treatment with low-dose chemotherapy for KS and/or multicentric Castleman disease, HAART use at the time of PEL diagnosis (defined as a combination of at least three antiretroviral drugs given for at least 2 months), CD4+ and CD8+ lymphocyte counts, plasma HIV RNA, performance status based on the Eastern Cooperative Oncology Group scale,18 hemoglobin level, platelet count, serum lactate dehydrogenase levels, serum albumin and ferritin levels, hemophagocytic syndrome diagnosed in accordance with the Histiocyte Society criteria,19 EBV and CD30 status of lymphomatous cells. Disease staging included computerized tomodensitometry scan of chest, abdomen, and pelvis; echocardiogram; and bone marrow biopsy and/or aspirate. As the patients had received heterogeneous treatments for lymphoma, only two therapy-related factors were analyzed: first-line treatment with interferon alfa, and use of systemic chemotherapy regimen containing high-dose methotrexate. These two modalities of treatment were chosen because they were the most frequently used in the present series, and because both have been reported to induce complete remission in PEL.16,17 The prognostic value of HAART given before PEL diagnosis and/or started within 1 year from the date of PEL diagnosis ("on HAART") was also assessed.
Three criteria were required to establish the diagnosis of PEL: (1) the occurrence of lymphomatous effusions in at least one serous cavity, (2) the morphological features of May-Grünwald-Giemsa–stained lymphoma cells collected from the effusion fluid, and (3) the evidence for a KSHV/HHV-8 infection of lymphoma cells provided either by molecular methods (quantitative polymerase chain reaction [PCR] analysis or Southern blot), or by immunohistochemistry with a monoclonal antibody directed against the latency-associated nuclear antigen of KSHV/HHV-8 (LNA/orf73, clone 13B10; Novocastra Laboratories Ltd, Newcastle, UK).
Coinfection with EBV was assessed in all cases by PCR analysis or in situ hybridization for Epstein-Barr virus–encoded RNA detection. Expression of CD30 antigen was determined by immunohistochemistry on cytospin preparations or by flow cytometry.
Statistical Analysis
Complete remission (CR) was defined as the resolution of all evidence of disease as determined by clinical and radiologic evaluation, persisting for at least 2 months after the end of treatment. Overall survival (OS) was calculated from the date of PEL diagnosis until the date of death due to any cause. Disease-free survival (DFS) was calculated only for patients who achieved CR, from the date of the first documentation of CR until the date of progression or death due to any cause. Data on patients who had survived until the end of observation time period, were censored at the date of last contact. OS and DFS were estimated using the Kaplan-Meier method.20 All parameters were analyzed as categoric variables. The cutoff points chosen were based on clinical relevance. A Cox proportional hazard regression model was used to identify statistically significant differences in survival and to estimate the hazard ratio and 95% CIs.21 Covariates identified with P < .10 by univariate analysis were included in a multivariate model. A forward stepwise selection procedure was used to assess the relative role of prognostic factors. The level of significance was .05. Statistical analysis was performed using the software package SAS (Version 8.2 for Windows; SAS Institute, Cary, NC).
RESULTS
Description of Patients
Twenty-eight HIV-infected patients were included in the study, 27 men and one woman, ranging in age from 33 to 78 years (median, 44 years). The main clinical and biologic features at the time of diagnosis are presented in Table 1. Ten men came from geographic areas characterized by a KSHV/HHV-8 seroprevalence ranging from 5% to 35% in healthy adults, mostly in Mediterranean countries. The woman was originated from an African country with a KSHV/HHV-8 seroprevalence higher than 35%. Sixteen patients were originated from Western European countries with seroprevalence lower than 5%.22-25
At the time of diagnosis, pleural effusions were present in 24 cases, ascitis, in 17 cases, and pericardial effusion, in nine cases. A simultaneous involvement of all serous cavities was found in seven cases. Twelve patients had extracavitary PEL localizations involving bone marrow (n = 1), CNS (n = 1), maxillary sinus (n = 1), mediastinum (n = 2), pancreatic gland (n = 1), mesenteric (n = 4), and peripheral lymph nodes (n = 3; Table 1). The median hemoglobin level was 9.5 g/dL (range, 6 to 12.7 g/dL), the median platelet count, 146 x 109/L (range, 14 to 854 x 109/L), and the median serum albumin level, 26 g/L (range, 14 to 47 g/L). Twenty PEL (71.5%) were EBV+.
Before lymphoma diagnosis, 12 patients had received low-dose sequential chemotherapy for pre-existing KS and/or multicentric Castleman disease, mainly consisting of bleomycin, liposomal daunorubicin, etoposide, or vinblastine. At the time of PEL diagnosis, one patient was being treated with interferon alfa for chronic hepatitis C. Severe immunosuppression was a common finding, as the median CD4+ cell count was 133 x 106/L (range, 5 to 756 x 106/L), and the median CD8+ cell count, 652 x 106/L (range, 137 to 3,075 x 106/L). Among the 12 patients previously treated with low-dose chemotherapy, the median CD4+ cell count was 148 x 106/L (range, 5 to 756 x 106/L), compared with 95 x 106/L (range, 17 to 557 x 106/L) in the 16 others. The median CD4+ cell count was 136 x 106/L (range, 33 to 557 x 106/L) in the pre-HAART period and 130 x 106/L (range, 5 to 756x 106/L) in the post-HAART period. The plasma HIV RNA load was available for 26 patients diagnosed after 1996, and was below the limit of detection in 11 of them (Table 1). In France, HAART became available as early as in June 1996. Of five patients diagnosed before this date, only one received HAART during the year following the date of PEL diagnosis (patient 5). Of the 23 patients diagnosed in the post-HAART period, 17 have been receiving HAART for a median of 19 months (range, 2 to 63 months), when PEL was diagnosed. All but two (patients 19 and 26) remained on HAART during and after lymphoma treatment. In the six previously untreated patients of the post-HAART period, HAART was started at the onset of lymphoma treatment.
Treatment and Outcome
One patient rapidly achieved CR after antiretroviral therapy initiation.14 All other patients received specific treatment for PEL (Table 2). As first-line treatment, five patients received interferon alfa (3 to 5 x 106 U x 3/wk), either alone (n = 2) or associated with cidofovir (n = 3), and 12 received intensive chemotherapy regimens consisting of ACVBP (doxorubicin, cyclophosphamide, vindesine, bleomycin, prednisone; n = 2) or high-dose methotrexate (2.5 to 3 g/m2) combined with a CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) –derived regimen (n = 10), as previously published.17 Standard or reduced-dosage anthracyclin- and/or cyclophosphamide-based regimens (CHOP, mini-CHOP, CDE [cyclophosphamide, doxorubicin, etoposide], or ABVP [doxorubicin, bleomycin, etoposide]) were given to the others. All lines of treatment considered, 13 patients received systemic chemotherapy containing high-dose methotrexate. This drug was used either in consolidation after ACVBP induction (n = 2) or in combination with CHOP-derived regimen (n = 11).
Treatment-related toxicities are detailed in Tables 3 and 4. Among the 11 patients treated with CHOP-derived chemotherapy containing high-dose methotrexate, all had grade 3/4 neutropenia, and eight had febrile neutropenia, despite granulocyte colony-stimulating factor use in the intercourse. Pseudomonas aeruginosa (n = 4) and herpesvirus (n = 6) infections were the most frequent infectious events. High-dose methotrexate also induced renal (n = 2) and liver (n = 3) toxicities. In three cases, this chemotherapy regimen was discontinued due to severe adverse events. Interferon alfa had a lower toxicity, although grade 3/4 neutropenia occurred in three patients.
By May 1, 2004, after a median follow-up of 3.8 years (range, 10 months to 10.8 years), 14 patients achieved CR, and nine patients (32%) were still alive, including eight in CR. No patient was lost to follow-up. Relapses occurred in six cases (5, 7.5, 8, 24, 25, and 97 months, respectively, after PEL diagnosis) and led to death in 5 cases (Table 2). At the time of PEL relapse, 3 patients presented with extraserous disease sites involving lymph nodes (n = 3), spleen (n = 1) and liver (n = 1) whereas serous effusions were minor or absent. One patient relapsed twice, and the second relapse was localized to maxillary sinus and orbit in the absence of serous cavity involvement.
Throughout the study period, 19 patients (68%) died. The causes of death were progressive lymphoma in 14 cases, infectious events and/or toxicity of chemotherapy in four cases, and disseminated visceral KS in one case. The overall median survival was 6.2 months. The 1-year OS rate was 39.3% (95% CI, 20.9 to 57.7; Fig 1A). For 14 patients who achieved CR, the 1-year DFS rate was 78.6% (95% CI, 56.6 to 100.6) and the median DFS was 94.8 months (Fig 1B).
Prognostic Analysis
In univariate analyses (Table 5), several factors were found to be associated with poor clinical outcome (P < .10): age less than 45 years, absence of HAART before lymphoma, performance status greater than 2, thrombocytopenia, hypoalbuminemia, and treatment regimen not including high-dose methotrexate.
In the multivariate analysis including the previous factors, only a performance status greater than 2 (P = .004) and the absence of HAART before PEL diagnosis (P = .028) remained independent predictors for shorter survival. The increased hazard ratios were 5.84 (95% CI, 1.76 to 19.33) and 3.26 (95% CI, 1.14 to 9.34), respectively.
DISCUSSION
The present HIV-associated PEL cases shared several clinical and epidemiological features with the series previously published by American, Italian, and Spanish groups. Nearly all patients were homosexual white men with a median age between 37 and 44 years.6-13 Women were exceptionally affected.26,27 The frequent association with multicentric Castleman disease observed in the present series (32%) might be explained by the large cohort of HIV-infected patients treated for multicentric Castleman disease at Hpital Saint-Louis. Extracavitary PEL localizations were commonly found at the time of diagnosis (43%) and even more frequently observed at the time of relapse.
We report here the first prognostic study in HIV-associated PEL. In previous series, the median survival ranged from 1.5 to 6 months. Our data confirmed this poor clinical outcome, with a median survival at 6.2 months and a 1-year OS rate at 39.3%. In a multivariate analysis, only the impaired performance status and the absence of pre-existing HAART were found to be independent prognostic indicators of shorter survival. The prognostic value of performance status has previously been established in HIV-associated NHL,5,28 but its negative influence on AIDS-related PEL survival has never been demonstrated.
KSHV/HHV-8 primary infection induces the development of specific CD8+ cytotoxic T-lymphocyte (CTL) responses targeted to lytic and latent viral proteins.29-31 In HIV-infected patients, the development of KS is associated with impaired CTL responses against the virus. During HAART-induced immune reconstitution, the control of KSHV/HHV-8 replication is apparently better in patients without KS than in patients with KS.32 CTL responses might then be preserved by early initiation of HAART. An improved survival associated with the previous use of HAART has also been demonstrated in HIV-associated systemic NHL.5
In the present study, we were not able to demonstrate any negative predictive value for several factors generally accepted as being associated with poor prognosis in HIV-related NHL.28 Although some authors reported that patients with severe thrombocytopenia at the time of PEL diagnosis deteriorated more rapidly than the others,9 the prognostic value of platelet count was not confirmed in the present study. Due to its particular tropism for serous cavities, standard lymphoma staging according to the Ann Arbor classification might not apply to PEL. For this reason, staging was not included in the prognostic analysis. The clinical outcome of patients concomitantly affected with other KSHV/HHV-8–related diseases was similar to those of patients with PEL only. Despite its usual severity, hemophagocytic syndrome was not found to influence PEL outcome. However, its common association with PEL prompted us to often include Etoposide or Vinblastine in the chemotherapy regimens. As observed in the present series, around 70% of PEL are coinfected with EBV. Infection of KSHV/HHV8+ EBV-negative PEL cell lines with a recombinant EBV strain led to an increased tumorigenicity in severe combined immunodeficiency (SCID) mice compared to the parental cell lines, suggesting a growth advantage for PEL cells infected with both viruses.33 However, the present study did not show any significant difference in survival regarding the EBV status of PEL.
To date, no specific treatment regimen has been recommended for PEL. Apart from exceptional reports of antiretroviral therapy–induced response, only few patients achieved CR.13-14 Impaired clinical condition and severe immunodeficiency enhanced the chemotherapy toxicity and increased the risk of treatment-related mortality. Only rare cases of CHOP-induced remission have been reported in patients simultaneously treated with HAART.11-13 The potential for methotrexate to diffuse in serous cavities could suggest the use of high-dose methotrexate in association with a CHOP-derived regimen.17 However, it has been reported that patients with serous effusions might be at increased risk of toxicity following high-dose methotrexate chemotherapy.34 The effusions might behave as a reservoir slowly releasing the drug, resulting in increased half-life and delayed excretion of methotrexate.35 Several precautions are then required to avoid severe toxicities in patients affected with PEL, especially in a context of severe hypoalbuminemia and abundant effusions. Abundant alkaline intravenous hydratation, transient discontinuation of trimethoprim/sulfamethoxazole prophylaxis, sustained leucovorin rescue, tight monitoring of weight, diuresis, renal function, and serum methotrexate levels are recommended, as well as the use of granulocyte colony-stimulating factor in the intercourse. When clinical condition or visceral failure hamper chemotherapy use, interferon alfa might represent a good alternative for PEL treatment. As observed for other KSHV/HHV-8-associated disorders and for AIDS-related NHL,4,5,32 the control of HIV replication is likely essential in PEL management.
Although we present here a large series of PEL cases, the statistical analysis might be hampered by the relatively small population size. Other limits are the retrospective collection of data and the lack of standardization of treatment regimens. However, based on a retrospective series of 28 HIV-infected patients with PEL, we identified two prognostic factors independently associated with shorter survival: (1) the poor performance status and (2) the absence of pre-existing HAART use. Prospective studies and therapeutic trials including more patients could give a better insight into the prognostic factors in PEL, and help for the further establishment of therapeutic guidelines.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
E.B. was the recipient of a fellowship from the Agence Nationale de Recherche sur le Sida (ANRS). We thank Drs M.T. Daniel for cytological analyses; F. Agbalika for virologic analyses; S. Legac, P. de Truchis, J. Julien, S. Neuville, D. Salmon, and A. Simon for clinical information about patients; and P. Boulanger for his critical reading of the manuscript.
NOTES
Supported by the Agence Nationale de Recherche sur le SIDA (ANRS)
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Knowles DM: Etiology and pathogenesis of AIDS-related non-Hodgkin's lymphoma. Hematol Oncol Clin N Am 17:785-820, 2003
Ansari MQ, Dawson DB, Nador R, et al: Primary body cavity-based AIDS-related lymphomas. Am J Clin Path 105:221-229, 1996
Palella FJ Jr, Delaney KM, Moorman AC, et al: Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection: HIV Outpatient Study Investigators. N Engl J Med 338:853-860, 1998
Gerard L, Galicier L, Maillard A, et al: Systemic non-Hodgkin lymphoma in HIV-infected patients with effective suppression of HIV replication: Persistent occurrence but improved survival. J Acquir Immune Defic Syndr 30:478-484, 2002
Vaccher E, Spina M, Talamini R, et al: Improvement of systemic human immunodeficiency virus-related non-Hodgkin lymphoma outcome in the era of highly active antiretroviral therapy. Clin Infect Dis 37:1556-1564, 2003
Nador RG, Cesarman E, Chadburn A, et al: Primary effusion lymphoma: A distinct clinicopathologic entity associated with the Kaposi's sarcoma-associated herpes virus. Blood 88:645-656, 1996
Carbone A, Gloghini A, Vaccher E, et al: Kaposi's sarcoma-associated herpesvirus DNA sequences in AIDS-related and AIDS-unrelated lymphomatous effusions. Br J Haematol 94:533-543, 1996
Otsuki T, Kumar S, Ensoli B, et al: Detection of HHV-8/KSHV DNA sequences in AIDS-associated extranodal lymphoid malignancies. Leukemia 10:1358-1362, 1996
Komanduri KV, Luce JA, McGrath MS, et al: The natural history and molecular heterogeneity of HIV-associated primary malignant lymphomatous effusions. J AIDS Hum Retrovirol 13:215-226, 1996
Karcher DS, Alkan S: Human herpesvirus-8-associated body cavity-based lymphoma in human immunodeficiency virus-infected patients: A unique B-cell neoplasm. Hum Pathol 28:801-808, 1997
Valencia ME, Martinez P, Moreno V, et al: AIDS-related body cavity-based lymphomas, herpesvirus-8 and HIV infection: A study of seven cases. AIDS 13:2603-2605, 1999
Ascoli V, Signoretti S, Onetti-Muda A, et al: Primary effusion lymphoma in HIV-infected patients with multicentric Castleman's disease. J Pathol 193:200-209, 2001
Simonelli C, Spina M, Cinelli R, et al: Clinical features and outcome of primary effusion lymphoma in HIV-infected patients: A single-institution study. J Clin Oncol 21:3948-3954, 2003
Oksenhendler E, Clauvel JP, Jouveshomme S, et al: Complete remission of a primary effusion lymphoma with antiretroviral therapy. Am J Hematol 57:266, 1998
Boulanger E, Agbalika F, Maarek O, et al: A clinical, molecular and cytogenetic study of twelve cases of human herpesvirus 8 associated primary effusion lymphoma in HIV-infected patients. Hematol J 2:172-179, 2001
Hocqueloux L, Agbalika F, Oksenhendler E, et al: Long-term remission of an AIDS-related primary effusion lymphoma with antiviral therapy. AIDS 15:280-282, 2001
Boulanger E, Daniel MT, Agbalika F, et al: Combined chemotherapy including high-dose methotrexate in KSHV/HHV8-associated primary effusion lymphoma. Am J Hematol 73:143-148, 2003
Oken MM, Creech RH, Tormey DC, et al: Toxicity and response criteria to the Eastern Cooperative Oncology Group. Am J Clin Oncol 5:649-655, 1982
Henter JI, Elinder G, Ost A: Diagnostic guidelines for hemophagocytic lymphohistiocytosis: The FHL Study Group of the Histiocyte Society. Semin Oncol 18:29-33, 1991
Kaplan EL, Meier P: Non parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Cox DR: Regression models for life tables. JR Stat Soc 34:187-220, 1977
Gao SJ, Kingsley L, Li M, et al: KSHV antibodies among Americans, Italians and Ugandans with and without Kaposi's sarcoma. Nat Med 2:925-928, 1996
Simpson GR, Schulz TF, Whitby D, et al: Prevalence of Kaposi's sarcoma-associated herpesvirus infection measured by antibodies to recombinant capsid protein and latent immunofluorescence antigen. Lancet 348:1133-1138, 1996
Calabro ML, Sheldon J, Favero A, et al: Seroprevalence of Kaposi's sarcoma-associated herpesvirus/human herpesvirus 8 in several regions of Italy. J Hum Virol 1:207-213, 1998
Gambus G, Bourboulia D, Esteve A, et al: Prevalence and distribution of HHV-8 in different subpopulations, with and without HIV infection, in Spain. AIDS 15:1167-1174, 2001
Said JW, Tasaka T, Takeuchi S, et al: Primary effusion lymphoma in women: Report of two cases of Kaposi's sarcoma herpes virus-associated effusion-based lymphoma in human immunodeficiency virus-negative women. Blood 88:3124-3128, 1996
Boulanger E, Hermine O, Fermand JP, et al: Human herpesvirus 8 (HHV-8)-associated peritoneal primary effusion lymphoma (PEL) in two HIV-negative elderly patients. Am J Hematol 76:88-91, 2004
Gabarre J, Raphael M, Lepage E, et al: Human immunodeficiency virus-related lymphoma: Relation between clinical features and histological subtypes. Am J Med 111:704-711, 2001
Wang QJ, Jenkins FJ, Jacobson LP, et al: Primary human herpesvirus 8 infection generates a broadly specific CD8+ T-cell response to viral lytic cycle proteins. Blood 97:2366-2373, 2001
Wang QJ, Jenkins FJ, Jacobson LP, et al: CD8+ cytotoxic T lymphocyte responses to lytic proteins of human herpes virus 8 in human immunodeficiency virus type 1-infected and -uninfected individuals. J Infect Dis 182:928-932, 2000
Osman M, Kubo T, Gill J, et al: Identification of human herpesvirus 8-specific cytotoxic T-cell responses. J Virol 73:6136-6140, 1999
Wilkinson J, Cope A, Gill J, et al: Identification of Kaposi's sarcoma-associated herpesvirus (KSHV)-specific cytotoxic T-lymphocyte epitopes and evaluation of reconstitution of KSHV-specific responses in human immunodeficiency virus type 1-infected patients receiving highly active antiretroviral therapy. J Virol 76:2634-2640, 2002
Trivedi P, Takazawa K, Zompetta C, et al: Infection of HHV-8+ primary effusion lymphoma cells with a recombinant Epstein-Barr virus leads to restricted EBV latency, altered phenotype, and increased tumorigenicity without affecting TCL1 expression. Blood 103:313-316, 2004
Pauley JL, Panetta JC, Schmidt J, et al: Late-onset delayed excretion of methotrexate. Cancer Chemother Pharmacol 54:146-152, 2004
Li J, Gwilt P: The effect of malignant effusions on methotrexate disposition. Cancer Chemother Pharmacol 50:373-382, 2002(Emmanuelle Boulanger, Lau)
Department of Clinical Hematology, Hpital La Pitie Salpêtrière, AP-HP
Department of Pneumology, Hpital Tenon, AP-HP, Paris
Department of Infectious Diseases, Hpital d'Instruction des Armees Begin, Saint-Mande
Department of Infectious Diseases, Hpital Eugenie, Ajaccio, France
ABSTRACT
PURPOSE: Primary effusion lymphoma (PEL) is a rare high-grade B-cell non-Hodgkin's lymphoma associated with Kaposi sarcoma–associated herpesvirus/human herpesvirus 8 (KSHV/HHV-8) infection, and is mostly observed in the course of HIV infection. The prognosis is poor, with reported median survival time shorter than 6 months. To date, no prognostic factor has been identified in this subset of lymphoma.
PATIENTS AND METHODS: We describe here a large series of HIV-infected patients with PEL, including 28 cases diagnosed in six centers during an 11-year time period. Prognosis analysis was performed using a Cox proportional hazard regression model. Statistically significant covariates were further analyzed in a forward, stepwise multivariate model.
RESULTS: After a median follow-up of 3.8 years (range, 10 months to 10.8 years), nine patients (32%) were still alive, and eight of them remained progression free. The median survival was 6.2 months, and the 1-year overall survival rate was 39.3%. Fourteen patients (50%) achieved complete remission, with a 1-year disease-free survival rate at 78.6%. In a multivariate analysis, only a performance status more than 2 (hazard ratio, 5.84; 95% CI, 1.76 to 19.33) and the absence of highly active antiretroviral therapy (HAART) before PEL diagnosis (hazard ratio, 3.26; 95% CI, 1.14 to 9.34) were found to be independent predictors for shorter survival.
CONCLUSION: Based on a retrospective series of 28 patients, two prognostic factors were identified as being independently associated with impaired clinical outcome in HIV-related PEL—(1) a poor performance status and (2) the absence of HAART before PEL diagnosis.
INTRODUCTION
Kaposi sarcoma (KS) and non-Hodgkin's lymphoma (NHL) represent the most common malignancies associated with HIV infection. Most HIV-related NHL are of B-cell origin and high-grade histology. They are characterized by clinical aggressiveness, advanced stage, and high frequency of extranodal involvement.1 Viral agents, in particular human gamma-herpesviruses, might play a role in the development of some AIDS-related malignancies. Among virus-associated lymphoid disorders, primary effusion lymphoma (PEL) is a KS-associated herpesvirus/human herpesvirus 8 (KSHV/HHV-8) –associated NHL, which predominantly develops in serous body cavities resulting in recurrent lymphomatous effusions. PEL is an uncommon disease that accounts for approximately 3% of all HIV-related NHL.1 PEL tumor cells display large heterogeneous size and pleomorphic aspect and are coinfected with Epstein-Barr virus (EBV) in most cases. Although of B-cell origin, these cells frequently present a CD45+ CD30+ non-B non-T phenotype.2
Since the widespread use of highly active antiretroviral therapy (HAART), morbidity and mortality associated with HIV infection have shown impressive declines.3 Overall outcome of NHL has improved since the introduction of HAART.4,5 Nevertheless, despite extensive use of HAART and improvement of chemotherapy management in HIV-infected patients, the prognosis of patients with HIV-related PEL remains poor. The median survival does not exceed 6 months even in the most recent series. Moreover, only short series have been published,6-13 the largest including 15 patients. To our knowledge, no prognostic factor has been identified and the influence of HAART on survival has not been determined. To address this issue, we present data from the follow-up of a cohort series of 28 HIV-infected patients diagnosed with PEL during a 11-year period.
PATIENTS AND METHODS
Patients
The study included all consecutive HIV-infected patients with PEL diagnosed between January 1, 1993, and September 30, 2003 at the following hospitals in France : Hpital Saint-Louis, Paris (20 cases); Hpital La Pitie-Salpêtrière, Paris (four cases); Htel-Dieu, Paris (one case); Hpital Tenon, Paris (one case); Hpital Begin, Saint-Mande (one case); and Hpital Eugenie, Ajaccio (one case). All patient records were reviewed by the same examiner (E.B.). Since May 1, 1999, data have been collected prospectively. Among these 28 cases, 17 have previously been reported.14-17
Data recorded at diagnosis were demographic features (age, sex, and geographic origin), previous AIDS-defining illnesses including history of KS, pre-existing multicentric Castleman disease, previous treatment with low-dose chemotherapy for KS and/or multicentric Castleman disease, HAART use at the time of PEL diagnosis (defined as a combination of at least three antiretroviral drugs given for at least 2 months), CD4+ and CD8+ lymphocyte counts, plasma HIV RNA, performance status based on the Eastern Cooperative Oncology Group scale,18 hemoglobin level, platelet count, serum lactate dehydrogenase levels, serum albumin and ferritin levels, hemophagocytic syndrome diagnosed in accordance with the Histiocyte Society criteria,19 EBV and CD30 status of lymphomatous cells. Disease staging included computerized tomodensitometry scan of chest, abdomen, and pelvis; echocardiogram; and bone marrow biopsy and/or aspirate. As the patients had received heterogeneous treatments for lymphoma, only two therapy-related factors were analyzed: first-line treatment with interferon alfa, and use of systemic chemotherapy regimen containing high-dose methotrexate. These two modalities of treatment were chosen because they were the most frequently used in the present series, and because both have been reported to induce complete remission in PEL.16,17 The prognostic value of HAART given before PEL diagnosis and/or started within 1 year from the date of PEL diagnosis ("on HAART") was also assessed.
Three criteria were required to establish the diagnosis of PEL: (1) the occurrence of lymphomatous effusions in at least one serous cavity, (2) the morphological features of May-Grünwald-Giemsa–stained lymphoma cells collected from the effusion fluid, and (3) the evidence for a KSHV/HHV-8 infection of lymphoma cells provided either by molecular methods (quantitative polymerase chain reaction [PCR] analysis or Southern blot), or by immunohistochemistry with a monoclonal antibody directed against the latency-associated nuclear antigen of KSHV/HHV-8 (LNA/orf73, clone 13B10; Novocastra Laboratories Ltd, Newcastle, UK).
Coinfection with EBV was assessed in all cases by PCR analysis or in situ hybridization for Epstein-Barr virus–encoded RNA detection. Expression of CD30 antigen was determined by immunohistochemistry on cytospin preparations or by flow cytometry.
Statistical Analysis
Complete remission (CR) was defined as the resolution of all evidence of disease as determined by clinical and radiologic evaluation, persisting for at least 2 months after the end of treatment. Overall survival (OS) was calculated from the date of PEL diagnosis until the date of death due to any cause. Disease-free survival (DFS) was calculated only for patients who achieved CR, from the date of the first documentation of CR until the date of progression or death due to any cause. Data on patients who had survived until the end of observation time period, were censored at the date of last contact. OS and DFS were estimated using the Kaplan-Meier method.20 All parameters were analyzed as categoric variables. The cutoff points chosen were based on clinical relevance. A Cox proportional hazard regression model was used to identify statistically significant differences in survival and to estimate the hazard ratio and 95% CIs.21 Covariates identified with P < .10 by univariate analysis were included in a multivariate model. A forward stepwise selection procedure was used to assess the relative role of prognostic factors. The level of significance was .05. Statistical analysis was performed using the software package SAS (Version 8.2 for Windows; SAS Institute, Cary, NC).
RESULTS
Description of Patients
Twenty-eight HIV-infected patients were included in the study, 27 men and one woman, ranging in age from 33 to 78 years (median, 44 years). The main clinical and biologic features at the time of diagnosis are presented in Table 1. Ten men came from geographic areas characterized by a KSHV/HHV-8 seroprevalence ranging from 5% to 35% in healthy adults, mostly in Mediterranean countries. The woman was originated from an African country with a KSHV/HHV-8 seroprevalence higher than 35%. Sixteen patients were originated from Western European countries with seroprevalence lower than 5%.22-25
At the time of diagnosis, pleural effusions were present in 24 cases, ascitis, in 17 cases, and pericardial effusion, in nine cases. A simultaneous involvement of all serous cavities was found in seven cases. Twelve patients had extracavitary PEL localizations involving bone marrow (n = 1), CNS (n = 1), maxillary sinus (n = 1), mediastinum (n = 2), pancreatic gland (n = 1), mesenteric (n = 4), and peripheral lymph nodes (n = 3; Table 1). The median hemoglobin level was 9.5 g/dL (range, 6 to 12.7 g/dL), the median platelet count, 146 x 109/L (range, 14 to 854 x 109/L), and the median serum albumin level, 26 g/L (range, 14 to 47 g/L). Twenty PEL (71.5%) were EBV+.
Before lymphoma diagnosis, 12 patients had received low-dose sequential chemotherapy for pre-existing KS and/or multicentric Castleman disease, mainly consisting of bleomycin, liposomal daunorubicin, etoposide, or vinblastine. At the time of PEL diagnosis, one patient was being treated with interferon alfa for chronic hepatitis C. Severe immunosuppression was a common finding, as the median CD4+ cell count was 133 x 106/L (range, 5 to 756 x 106/L), and the median CD8+ cell count, 652 x 106/L (range, 137 to 3,075 x 106/L). Among the 12 patients previously treated with low-dose chemotherapy, the median CD4+ cell count was 148 x 106/L (range, 5 to 756 x 106/L), compared with 95 x 106/L (range, 17 to 557 x 106/L) in the 16 others. The median CD4+ cell count was 136 x 106/L (range, 33 to 557 x 106/L) in the pre-HAART period and 130 x 106/L (range, 5 to 756x 106/L) in the post-HAART period. The plasma HIV RNA load was available for 26 patients diagnosed after 1996, and was below the limit of detection in 11 of them (Table 1). In France, HAART became available as early as in June 1996. Of five patients diagnosed before this date, only one received HAART during the year following the date of PEL diagnosis (patient 5). Of the 23 patients diagnosed in the post-HAART period, 17 have been receiving HAART for a median of 19 months (range, 2 to 63 months), when PEL was diagnosed. All but two (patients 19 and 26) remained on HAART during and after lymphoma treatment. In the six previously untreated patients of the post-HAART period, HAART was started at the onset of lymphoma treatment.
Treatment and Outcome
One patient rapidly achieved CR after antiretroviral therapy initiation.14 All other patients received specific treatment for PEL (Table 2). As first-line treatment, five patients received interferon alfa (3 to 5 x 106 U x 3/wk), either alone (n = 2) or associated with cidofovir (n = 3), and 12 received intensive chemotherapy regimens consisting of ACVBP (doxorubicin, cyclophosphamide, vindesine, bleomycin, prednisone; n = 2) or high-dose methotrexate (2.5 to 3 g/m2) combined with a CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) –derived regimen (n = 10), as previously published.17 Standard or reduced-dosage anthracyclin- and/or cyclophosphamide-based regimens (CHOP, mini-CHOP, CDE [cyclophosphamide, doxorubicin, etoposide], or ABVP [doxorubicin, bleomycin, etoposide]) were given to the others. All lines of treatment considered, 13 patients received systemic chemotherapy containing high-dose methotrexate. This drug was used either in consolidation after ACVBP induction (n = 2) or in combination with CHOP-derived regimen (n = 11).
Treatment-related toxicities are detailed in Tables 3 and 4. Among the 11 patients treated with CHOP-derived chemotherapy containing high-dose methotrexate, all had grade 3/4 neutropenia, and eight had febrile neutropenia, despite granulocyte colony-stimulating factor use in the intercourse. Pseudomonas aeruginosa (n = 4) and herpesvirus (n = 6) infections were the most frequent infectious events. High-dose methotrexate also induced renal (n = 2) and liver (n = 3) toxicities. In three cases, this chemotherapy regimen was discontinued due to severe adverse events. Interferon alfa had a lower toxicity, although grade 3/4 neutropenia occurred in three patients.
By May 1, 2004, after a median follow-up of 3.8 years (range, 10 months to 10.8 years), 14 patients achieved CR, and nine patients (32%) were still alive, including eight in CR. No patient was lost to follow-up. Relapses occurred in six cases (5, 7.5, 8, 24, 25, and 97 months, respectively, after PEL diagnosis) and led to death in 5 cases (Table 2). At the time of PEL relapse, 3 patients presented with extraserous disease sites involving lymph nodes (n = 3), spleen (n = 1) and liver (n = 1) whereas serous effusions were minor or absent. One patient relapsed twice, and the second relapse was localized to maxillary sinus and orbit in the absence of serous cavity involvement.
Throughout the study period, 19 patients (68%) died. The causes of death were progressive lymphoma in 14 cases, infectious events and/or toxicity of chemotherapy in four cases, and disseminated visceral KS in one case. The overall median survival was 6.2 months. The 1-year OS rate was 39.3% (95% CI, 20.9 to 57.7; Fig 1A). For 14 patients who achieved CR, the 1-year DFS rate was 78.6% (95% CI, 56.6 to 100.6) and the median DFS was 94.8 months (Fig 1B).
Prognostic Analysis
In univariate analyses (Table 5), several factors were found to be associated with poor clinical outcome (P < .10): age less than 45 years, absence of HAART before lymphoma, performance status greater than 2, thrombocytopenia, hypoalbuminemia, and treatment regimen not including high-dose methotrexate.
In the multivariate analysis including the previous factors, only a performance status greater than 2 (P = .004) and the absence of HAART before PEL diagnosis (P = .028) remained independent predictors for shorter survival. The increased hazard ratios were 5.84 (95% CI, 1.76 to 19.33) and 3.26 (95% CI, 1.14 to 9.34), respectively.
DISCUSSION
The present HIV-associated PEL cases shared several clinical and epidemiological features with the series previously published by American, Italian, and Spanish groups. Nearly all patients were homosexual white men with a median age between 37 and 44 years.6-13 Women were exceptionally affected.26,27 The frequent association with multicentric Castleman disease observed in the present series (32%) might be explained by the large cohort of HIV-infected patients treated for multicentric Castleman disease at Hpital Saint-Louis. Extracavitary PEL localizations were commonly found at the time of diagnosis (43%) and even more frequently observed at the time of relapse.
We report here the first prognostic study in HIV-associated PEL. In previous series, the median survival ranged from 1.5 to 6 months. Our data confirmed this poor clinical outcome, with a median survival at 6.2 months and a 1-year OS rate at 39.3%. In a multivariate analysis, only the impaired performance status and the absence of pre-existing HAART were found to be independent prognostic indicators of shorter survival. The prognostic value of performance status has previously been established in HIV-associated NHL,5,28 but its negative influence on AIDS-related PEL survival has never been demonstrated.
KSHV/HHV-8 primary infection induces the development of specific CD8+ cytotoxic T-lymphocyte (CTL) responses targeted to lytic and latent viral proteins.29-31 In HIV-infected patients, the development of KS is associated with impaired CTL responses against the virus. During HAART-induced immune reconstitution, the control of KSHV/HHV-8 replication is apparently better in patients without KS than in patients with KS.32 CTL responses might then be preserved by early initiation of HAART. An improved survival associated with the previous use of HAART has also been demonstrated in HIV-associated systemic NHL.5
In the present study, we were not able to demonstrate any negative predictive value for several factors generally accepted as being associated with poor prognosis in HIV-related NHL.28 Although some authors reported that patients with severe thrombocytopenia at the time of PEL diagnosis deteriorated more rapidly than the others,9 the prognostic value of platelet count was not confirmed in the present study. Due to its particular tropism for serous cavities, standard lymphoma staging according to the Ann Arbor classification might not apply to PEL. For this reason, staging was not included in the prognostic analysis. The clinical outcome of patients concomitantly affected with other KSHV/HHV-8–related diseases was similar to those of patients with PEL only. Despite its usual severity, hemophagocytic syndrome was not found to influence PEL outcome. However, its common association with PEL prompted us to often include Etoposide or Vinblastine in the chemotherapy regimens. As observed in the present series, around 70% of PEL are coinfected with EBV. Infection of KSHV/HHV8+ EBV-negative PEL cell lines with a recombinant EBV strain led to an increased tumorigenicity in severe combined immunodeficiency (SCID) mice compared to the parental cell lines, suggesting a growth advantage for PEL cells infected with both viruses.33 However, the present study did not show any significant difference in survival regarding the EBV status of PEL.
To date, no specific treatment regimen has been recommended for PEL. Apart from exceptional reports of antiretroviral therapy–induced response, only few patients achieved CR.13-14 Impaired clinical condition and severe immunodeficiency enhanced the chemotherapy toxicity and increased the risk of treatment-related mortality. Only rare cases of CHOP-induced remission have been reported in patients simultaneously treated with HAART.11-13 The potential for methotrexate to diffuse in serous cavities could suggest the use of high-dose methotrexate in association with a CHOP-derived regimen.17 However, it has been reported that patients with serous effusions might be at increased risk of toxicity following high-dose methotrexate chemotherapy.34 The effusions might behave as a reservoir slowly releasing the drug, resulting in increased half-life and delayed excretion of methotrexate.35 Several precautions are then required to avoid severe toxicities in patients affected with PEL, especially in a context of severe hypoalbuminemia and abundant effusions. Abundant alkaline intravenous hydratation, transient discontinuation of trimethoprim/sulfamethoxazole prophylaxis, sustained leucovorin rescue, tight monitoring of weight, diuresis, renal function, and serum methotrexate levels are recommended, as well as the use of granulocyte colony-stimulating factor in the intercourse. When clinical condition or visceral failure hamper chemotherapy use, interferon alfa might represent a good alternative for PEL treatment. As observed for other KSHV/HHV-8-associated disorders and for AIDS-related NHL,4,5,32 the control of HIV replication is likely essential in PEL management.
Although we present here a large series of PEL cases, the statistical analysis might be hampered by the relatively small population size. Other limits are the retrospective collection of data and the lack of standardization of treatment regimens. However, based on a retrospective series of 28 HIV-infected patients with PEL, we identified two prognostic factors independently associated with shorter survival: (1) the poor performance status and (2) the absence of pre-existing HAART use. Prospective studies and therapeutic trials including more patients could give a better insight into the prognostic factors in PEL, and help for the further establishment of therapeutic guidelines.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
E.B. was the recipient of a fellowship from the Agence Nationale de Recherche sur le Sida (ANRS). We thank Drs M.T. Daniel for cytological analyses; F. Agbalika for virologic analyses; S. Legac, P. de Truchis, J. Julien, S. Neuville, D. Salmon, and A. Simon for clinical information about patients; and P. Boulanger for his critical reading of the manuscript.
NOTES
Supported by the Agence Nationale de Recherche sur le SIDA (ANRS)
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Knowles DM: Etiology and pathogenesis of AIDS-related non-Hodgkin's lymphoma. Hematol Oncol Clin N Am 17:785-820, 2003
Ansari MQ, Dawson DB, Nador R, et al: Primary body cavity-based AIDS-related lymphomas. Am J Clin Path 105:221-229, 1996
Palella FJ Jr, Delaney KM, Moorman AC, et al: Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection: HIV Outpatient Study Investigators. N Engl J Med 338:853-860, 1998
Gerard L, Galicier L, Maillard A, et al: Systemic non-Hodgkin lymphoma in HIV-infected patients with effective suppression of HIV replication: Persistent occurrence but improved survival. J Acquir Immune Defic Syndr 30:478-484, 2002
Vaccher E, Spina M, Talamini R, et al: Improvement of systemic human immunodeficiency virus-related non-Hodgkin lymphoma outcome in the era of highly active antiretroviral therapy. Clin Infect Dis 37:1556-1564, 2003
Nador RG, Cesarman E, Chadburn A, et al: Primary effusion lymphoma: A distinct clinicopathologic entity associated with the Kaposi's sarcoma-associated herpes virus. Blood 88:645-656, 1996
Carbone A, Gloghini A, Vaccher E, et al: Kaposi's sarcoma-associated herpesvirus DNA sequences in AIDS-related and AIDS-unrelated lymphomatous effusions. Br J Haematol 94:533-543, 1996
Otsuki T, Kumar S, Ensoli B, et al: Detection of HHV-8/KSHV DNA sequences in AIDS-associated extranodal lymphoid malignancies. Leukemia 10:1358-1362, 1996
Komanduri KV, Luce JA, McGrath MS, et al: The natural history and molecular heterogeneity of HIV-associated primary malignant lymphomatous effusions. J AIDS Hum Retrovirol 13:215-226, 1996
Karcher DS, Alkan S: Human herpesvirus-8-associated body cavity-based lymphoma in human immunodeficiency virus-infected patients: A unique B-cell neoplasm. Hum Pathol 28:801-808, 1997
Valencia ME, Martinez P, Moreno V, et al: AIDS-related body cavity-based lymphomas, herpesvirus-8 and HIV infection: A study of seven cases. AIDS 13:2603-2605, 1999
Ascoli V, Signoretti S, Onetti-Muda A, et al: Primary effusion lymphoma in HIV-infected patients with multicentric Castleman's disease. J Pathol 193:200-209, 2001
Simonelli C, Spina M, Cinelli R, et al: Clinical features and outcome of primary effusion lymphoma in HIV-infected patients: A single-institution study. J Clin Oncol 21:3948-3954, 2003
Oksenhendler E, Clauvel JP, Jouveshomme S, et al: Complete remission of a primary effusion lymphoma with antiretroviral therapy. Am J Hematol 57:266, 1998
Boulanger E, Agbalika F, Maarek O, et al: A clinical, molecular and cytogenetic study of twelve cases of human herpesvirus 8 associated primary effusion lymphoma in HIV-infected patients. Hematol J 2:172-179, 2001
Hocqueloux L, Agbalika F, Oksenhendler E, et al: Long-term remission of an AIDS-related primary effusion lymphoma with antiviral therapy. AIDS 15:280-282, 2001
Boulanger E, Daniel MT, Agbalika F, et al: Combined chemotherapy including high-dose methotrexate in KSHV/HHV8-associated primary effusion lymphoma. Am J Hematol 73:143-148, 2003
Oken MM, Creech RH, Tormey DC, et al: Toxicity and response criteria to the Eastern Cooperative Oncology Group. Am J Clin Oncol 5:649-655, 1982
Henter JI, Elinder G, Ost A: Diagnostic guidelines for hemophagocytic lymphohistiocytosis: The FHL Study Group of the Histiocyte Society. Semin Oncol 18:29-33, 1991
Kaplan EL, Meier P: Non parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Cox DR: Regression models for life tables. JR Stat Soc 34:187-220, 1977
Gao SJ, Kingsley L, Li M, et al: KSHV antibodies among Americans, Italians and Ugandans with and without Kaposi's sarcoma. Nat Med 2:925-928, 1996
Simpson GR, Schulz TF, Whitby D, et al: Prevalence of Kaposi's sarcoma-associated herpesvirus infection measured by antibodies to recombinant capsid protein and latent immunofluorescence antigen. Lancet 348:1133-1138, 1996
Calabro ML, Sheldon J, Favero A, et al: Seroprevalence of Kaposi's sarcoma-associated herpesvirus/human herpesvirus 8 in several regions of Italy. J Hum Virol 1:207-213, 1998
Gambus G, Bourboulia D, Esteve A, et al: Prevalence and distribution of HHV-8 in different subpopulations, with and without HIV infection, in Spain. AIDS 15:1167-1174, 2001
Said JW, Tasaka T, Takeuchi S, et al: Primary effusion lymphoma in women: Report of two cases of Kaposi's sarcoma herpes virus-associated effusion-based lymphoma in human immunodeficiency virus-negative women. Blood 88:3124-3128, 1996
Boulanger E, Hermine O, Fermand JP, et al: Human herpesvirus 8 (HHV-8)-associated peritoneal primary effusion lymphoma (PEL) in two HIV-negative elderly patients. Am J Hematol 76:88-91, 2004
Gabarre J, Raphael M, Lepage E, et al: Human immunodeficiency virus-related lymphoma: Relation between clinical features and histological subtypes. Am J Med 111:704-711, 2001
Wang QJ, Jenkins FJ, Jacobson LP, et al: Primary human herpesvirus 8 infection generates a broadly specific CD8+ T-cell response to viral lytic cycle proteins. Blood 97:2366-2373, 2001
Wang QJ, Jenkins FJ, Jacobson LP, et al: CD8+ cytotoxic T lymphocyte responses to lytic proteins of human herpes virus 8 in human immunodeficiency virus type 1-infected and -uninfected individuals. J Infect Dis 182:928-932, 2000
Osman M, Kubo T, Gill J, et al: Identification of human herpesvirus 8-specific cytotoxic T-cell responses. J Virol 73:6136-6140, 1999
Wilkinson J, Cope A, Gill J, et al: Identification of Kaposi's sarcoma-associated herpesvirus (KSHV)-specific cytotoxic T-lymphocyte epitopes and evaluation of reconstitution of KSHV-specific responses in human immunodeficiency virus type 1-infected patients receiving highly active antiretroviral therapy. J Virol 76:2634-2640, 2002
Trivedi P, Takazawa K, Zompetta C, et al: Infection of HHV-8+ primary effusion lymphoma cells with a recombinant Epstein-Barr virus leads to restricted EBV latency, altered phenotype, and increased tumorigenicity without affecting TCL1 expression. Blood 103:313-316, 2004
Pauley JL, Panetta JC, Schmidt J, et al: Late-onset delayed excretion of methotrexate. Cancer Chemother Pharmacol 54:146-152, 2004
Li J, Gwilt P: The effect of malignant effusions on methotrexate disposition. Cancer Chemother Pharmacol 50:373-382, 2002(Emmanuelle Boulanger, Lau)