Search

Introduction: Aberrant methylation of the 5′ gene promoter regions is an epigenetic phenomenon that is the one of the major mechanism of inactivation of tumor supressor genes. DNA methylation of the promoter region has been described for several genes in various malignant diseases, and each tumour type may have its own pattern of methylation. To determine the methylation status and expression of cell cycle inhibitors genes (p14, p15, p16), repair genes (MGMT and hMLH1) and the apoptosis regulator gene (DAPKinase) and the possible role of this epigenetic phenomenon in tumour progression of plasma cell disorders, we analyzed the methylation profile of MM, MGUS, and plasmacytomas comparing them with their protein expression. Patients and Methods: A total of 51 cases: 30 MM, 13 MGUS, and 8 plasmacytomas (3 Solitary Bone Plasmacytoma, 5 Extramedullary Plasmacytoma) were included in the study. Bone marrow plasma cells were purified using magnetic microbeads labeled with CD138 in samples with MM and MGUS. Methylation-specific polymerase chain reaction (MSP) for p14, p15, p16, MGMT, hMLH1 and DAPKinase was performed. MSP results were matched to protein expression studies by immunohistochemistry for p15, p16, MGMT and hMLH1. Results: The frequency of aberrant methylation among the MM samples was: 50% for p16, 16.7% for p15, 10% for hMLH1, 23.3% for MGMT, 30% for DAPK. In MGUS samples we found 38.5% for p16, 15.4% for p15, 0% for hMLH1, 7.7% for MGMT and 15.4% for DAPK methylation. The frequency of methylation in plasmacytomas was 62.5% for p16, 25% for p15, 0% for hMLH1, 62.5% for MGMT and 50% for DAPK. p14 was unmethylated in all cases (n=51). The correlation between gene methylation status and protein expression was assessed in 17 MM, 11 MGUS, and 8 plasmacytomas and we found that promoter methylation was strongly associated with gene silencing. All the samples methylated had lost the protein expression. In summary these findings demonstrate that aberrant methylation is an important mechanism of gene silencing in plasma cell disorders: 83.3% of MM, 46.1 of MGUS, and 75% of plasmacytomas have at least one hypermethylated gene (figure 1). We also show, that hypermethylation of p16 is a common phenomenon in plasma cell discrasias while there was no methylation of p14. Although the size of sample is small, we found that hMLH1 hypermethylation was found only in MM, while in plasmacytomas hypermethylation of MGMT and DAPK were frequent events. Moreover in survival studies of MM patients, a trend was observed between simultaneous aberrant methylation of hMLH1 and MGMT and poorer survival, but the number of cases studied limits the statistical analysis and the clinical implications of these results. To better define the clinical impact of methylation markers in plasma cell discrasias, it is therefore necessary to analyze a large number of patient samples. Figure 1. Gene methylation profiling of MM(A), MGUS(B), and plasmacytomas(C) using MSPCR Figure 1. Gene methylation profiling of MM(A), MGUS(B), and plasmacytomas(C) using MSPCR

Purpose: to compare the response rate and survival between patients diagnosed of PTLD and treated with front-line rituximab and those not treated with rituximab. Patients and Methods: 108 patients with PTLD have been studied from January 1996 to January 2004. Survival curves were expressed as Kaplan-Meier plots and were compared by the Log-rank test. A multivariate Cox regression analysis was performed to asses the effect of prognostic factors on survival. Results: median age was 55 years (limits: 18–73). 70% were males. The transplanted organ was: kidney 46%, liver 28%, heart 16%. Median time between transplant and PTLD was 59 months, 25% were diagnosed during the first year after the transplant. The most frequent histological subtypes were: large B-cell lymphoma 53% and polymorphic SLPT 13%. 70% were EBV +. Clinical characteristics at diagnosis were: disseminated disease: 52%, extra-nodal disease: 81%, ECOG 3 2: 43%, LDH >N: 60%, IPI 3 3: 40%. Treatments used were: reduction of immunosuppression 91%, chemotherapy 59%, rituximab 33%, antiviral 13%. Response to treatment was: CR 46%, PR 13% failure 11%, not evaluable (early deceased): 29%. With a median follow-up of 15,2 months, survival was: OS 21% and EFS 15%. Forty-six (43%) patients died. The causes of death were: lymphoma progression 15 (33%), infection 15 (33%), toxicity 16 (34%). Survival of patients treated with rituximab was significantly better than the general group: OS 76% (p=0.007) and EFS 70% (p=0.02). Among patients treated with rituximab, 8 (23%) patients died. The significant bad prognostic factors for EFS in the multivariate analysis were: disseminated disease (RR: 2, 95% IC:1,02–3,8; p=0,04), ECOG 3 2 (RR: 5, 95% IC:2,6–9,8; p=0,0001), not been treated with rituximab (RR: 3,8, 95% IC: 1,7–10; p=0,001). IPI did not have prognostic impact. Conclusions: survival of patients with PTLD is low with conventional therapy, and the main causes of death are toxicity and infections. Treatment with Rituximab significantly improves their survival. Patients with disseminated disease and bad performance status have worse prognosis. IPI is not a useful index of prognosis in patients with PTLD.

The authors describe five white patients with peripheral T-cell lymphoma. Four patients were older than 65 years. All cases presented with a short clinical course and advanced stage at the time of diagnosis. Clinical manifestations included asthenia, weight loss, peripheral and abdominal lymphadenopathy. One case showed tonsillar involvement and subcutaneous lymph node enlargement; hepatomegaly was present in four cases, two of them with splenomegaly. Only one case presented peripheral lymphocytosis and antibodies to human T-leukemia virus. Although three cases were classified as diffuse mixed lymphomas and two as poorly differentiated lymphocytic lymphomas, there were some common characteristics: diffuse infiltration by different proportions of small lymphoid cells and large immunoblasts, some of them multinucleated and similar to Reed-Sternberg cells; accumulation of histiocytes, plasmacytosis, eosinophilia, venular proliferation and compartmentalization were also found. Bone marrow infiltration was observed in two patients. Results of monoclonal markers showed four cases to be OKT4+ and the other OKT8+. The morphologic and immunologic characteristics of these patients were typical and similar to those reported from other geographical areas.

Summary We describe the case of a woman aged 34 years infected with the human immunodeficiency virus and whose illness was complicated by visceral leishmaniasis that ultimately led to her death.

Visceral leishmaniasis is infrequently reported in immunocompromised hosts; hence, the clinical manifestations and outcome of the disease in these patients are unknown. In a series of 10 patients with visceral leishmaniasis complicating renal transplantation (three), hematologic neoplasms (two), systemic lupus erythematosus (two), or infection with human immunodeficiency virus (three), typical hallmarks of kalaazar such as enlargement of spleen or hyperglobulinemia were absent in three and six patients, respectively. Extensive visceral involvement was noted by biopsies or autopsies in four patients. Diagnosis was made during evaluation for fever of unknown origin. Myriads of amastigotes were seen in bone marrow smears. Measurement of antibodies against Leishmania donovani was positive in each patient tested. Ultimately, three patients died, and chronic infections refractory to treatment developed in two other patients. Visceral leishmaniasis is a potentially fatal infection in immunocompromised hosts. Current antiparasitic therapy frequently fails to eradicate L. donovani from infected tissues.

Objectives: To identify a clonal population by molecular techniques in patients treated with Fluradabine, Cyclophosphamide, Rituximab (CF+R) and included in the LNF-03 clinical trial. To relate these alterations to the clinical variables at diagnosis and response to treatment. Patients and Methods: Patients: 71 patients from the LNF-03 clinical trial. Their treatment consisted of CF+ Rx6 and posterior maintenance with Rituximab. Peripheral Blood (PB) and Bone Marrow (BM) were studied at diagnosis. Methods: bcl2/IgH rearrangements were studied by real-time PCR (Olsson et al, Mod Pathol 1999). Ig rearrangement of VDJ IgH, Igs IgL KVJ, Kdel and LVJ was studied according to BIOMED II protocols. Results: 43 of the 71 patients (60.56%) presented some molecular alteration in PB or BM. 33% of the patients in the study presented t(14;18) in some of its majority forms, bcl2/IgH mayor 19 (26.4%) and minor 3 (4.22%). 13 patients were positive for bcl-2/IgH, both in BM and in PB. The median amount of blc2/IgH in BM was 8.315% (62.9–0.04), Whereas in pb it was 4.775% (92.6–0.06), with no significant differences observed. Nevertheless, monoclonality was detected in BM in eight cases but not in PB, and to the inverse, one it happened in three patients. Of 71 patients, 38 (53.52%), presented displayed monoclonality in at least one of the Ig genes. Of these patients, 38% presented the clonal rearrangement in IG-L KVJ, 26% in IG-H CDR1, 23% in IG-L KDEL, and 13% rearranged in IG-L LAMBDA. There were 57 patients whose BM was available. 28 of them presented pathological infiltration, and monoclonality was detected in 23 cases; in 29 patients who were not pathologically infiltrated, monoclonality was detected in 12 cases. 83% (6/37) of the patients presented a complete response (CR), 8.1% (3/37) uncertain CR (uCR) and 8.1% (3/37) partial response (PR), after the induction treatment. No relation was found between the presence of molecular alterations in BM/PB and the achievement of a CR. Conclusions: The study of the rearrangements of Ig and bcl2/IgH is a good strategy for identifying lymphomatosus disease at diagnosis; bone marrow is more informative than peripheral blood. The presence of molecular alterations in PB and/or BM is not related to a poorer answer to induction treatment. Partially funded by Roche Farma SA.

Summary Background Gastric MALT lymphoma can be histologically classified into two groups, low-grade (LG) and high-grade (HG); however, their natural history is poorly understood. We have studied a large retrospective series aiming to confirm whether the histological groups confer different clinical features and behavior and to analyze the prognostic factors in these patients. Patients and methods A series of 143 gastric B-cell MALT lymphomas is reported. Eighty-four were low-grade lymphomas (LG) and 59 were high-grade lymphomas (HG). Median follow-up was 36 months. The clinical and analytical parameters of the 84 LG patients were compared with those of the 59 HG patients. In the patients who had been operated on, the pathological features (macroscopical patterns, tumor size, involvement of resection margins, degree of parietal invasion and involvement of abdominal lymph nodes and adjacent viscera) of the LG patients were compared with those of the HG patients. The sites of relapses were studied. In the 132 treated and followed-up patients the influence of the treatment and that of clinical, analytical and pathological features on survival were investigated with the Kaplan and Meier and log-rank tests. To identify the factors with independent influence on survival, a Cox model was fitted for the whole series and separately for 53 HG patients. Results HG group differed from the LG group by a significantly higher frequency of weight loss at presentation, palpable abdominal mass, hepatomegaly, peripheral lymphadenopathy, elevated serum LDH, higher incidence of stage m-IV and tumor/mass patterns in the endoscopy and in the gastrectomy specimen. The tumor was significantly larger in the HG group than in the LG and the deeper invasion of the gastric wall, the higher frequency of infiltration of the abdominal lymph nodes and the visceral extension were also significant in the HG group. Complete remission (CR) was achieved in 91 of the patients of the LG group, but was significantly lower, 70%, in the HG group. Relapses occurred in the stomach and also in non-MALT sites. In 132 treated and followed-up patients, elevated serum LDH, absence of CR, HG group and stage in-IV were associated with a worse survival. In the Cox multivariate model, stage was the only variable influencing survival, although stage was related to the histological grade. In the HG group, stage was also an independent significant risk factor, whereas treatment with surgery, chemotherapy or both was not. In the 103 patients treated with surgery, a worse survival was associated with the involvement of the resection borders, depth of the infiltration of the gastric wall, dissemination to distant abdominal nodes and adjacent organs, but not with the addition of chemotherapy. Conclusions Histological classification into LG and HG separates distinctive groups of gastric MALT lymphoma that show striking clinical and prognostic differences. Besides histological grade, stage is the most important prognostic feature.

The International Prognostic Index identifies four risk groups with different survival rates in aggressive non-Hodgkin's lymphoma. We have studied whether a slight modification of this index has prognostic significance in high grade gastric B-cell MALT lymphoma. In 53 patients with high grade gastric B-cell MALT lymphoma the following survival factors were investigated: age over or under 60 years, sex, B symptoms, more than one extranodal site of involvement other than the stomach, serum LDH levels, performance status, stage I/IIE1/IIE2 v.s. stage III/IV, treatment with surgery, chemotherapy or both modalities together and the four risk groups as defined by the Modified International Prognostic Index (MIPI). A multivariate Cox's test was used to evaluate the independent prognostic significance on survival of all the above variables. Advanced stage (III/IV) and involvement of more than one extranodal site not including stomach were the only variables influencing survival. The MIPI was not sufficient to separate groups with significant differences in survival or to stratify prognostic groups. In this series, the MIPI did not show prognostic significance in high grade gastric B-cell MALT lymphoma.

Objectives: To evaluate the efficacy and toxicity of a response-adapted therapy with rituximab/bendamustine/mitoxantrone/dexamethasone (RBMD), followed by rituximab (R) maintenance therapy in patients with relapsed or refractory follicular lymphoma (R/R FL) to first-line treatment with R-chemotherapy (R-ChemoT). Material and methods: Multicenter phase II trial including 60 patients with R/R FL, after a first R-ChemoT line. Induction therapy: R 375 mg/m2 IV, day 1; bendamustine 90 mg/m2 IV, days 1 and 2; mitoxantrone 6 mg/m2 IV, day 1; dexamethasone 20 mg/day, PO, days 1 to 5. Cycles of 28 days. Evaluation of response after third cycle. If stable (SD) or progression disease (PD): patient withdrawn from the study. If complete response (CR) or unconfirmed complete response (CRu): administration of fourth cycle. If partial response (PR): administration up to 6 cycles. If CR, CRu, or PR after induction: patient received maintenance therapy with R (375 mg/m2/day every 12 weeks for 2 years). Results: N=60 patients: 50% female, age 63 (32-76) years. Ann Arbor stage III-IV 70% (42/60). FLIPI intermediate-high risk: 50% (30/60). Refractory to R-ChemoT: 18% (11/60). Received RCHOP as first line therapy: 77% (43/60). R maintenance after first line therapy: 43% (26/60). Number of administered RBMD cycles: 4 (1-6). Efectiveness: CR/CRu after 3 cycles (CR-3), 27 patients. Response after induction therapy (4-6 cycles): Overall response (OR), 88.5% (53/60); CR, 58.5% (35/60); CRu, 12% (7/60); PR, 18% (11/60); SD, 1.5% (1/60); PD, 10% (6/60). Median follow-up: 24.41 mo. (15.58-30.15). Progression free survival, median not reached (NR) (28.28-NR). Overall survival, median NR (NR-NR). OR after RBMD in patients who received R maintenance after first line therapy, 96% (25/26; CR + CRu, 65%). OR in patients who did not receive R maintenance after first line therapy, 94% (32/34; CR + CRu, 82%). Safety: Grade 3/4 hematologic toxicity: neutropenia, 60% (n=36; 34 patients received G-CSF); anemia, 2% (n=1); thrombopenia, 4% (n=2). Grade 3/4 infections: 8% (n=4); febrile neutropenia 8%, (n=4). Exitus, 2/60 (PD, cycle 1; influenza A, cycle 5). Grade 3/4 non-hematologic toxicity: infusion reaction, 4% (n=2). No skin reactions. Patient withdrawals before R maintenance therapy: no inclusion criteria (NIC), 2/60; protocol delay of over 4 weeks, 7/60. Forty-four patients began R maintenance therapy (CR-3, 24/44). During R maintenance therapy: PD, 8/44 (18%; 5/8 in CR-3 patients); SD, 1/44 (2.5%); NIC, 3/44 (7%, all CR); protocol delay, 1/44 (2.5%); toxicity, 1/44 (2.5%, myelodysplastic syndrome). Response to R maintenance therapy: OR, 68% (30/44); CR, 61% (27/44; in CR-3 patients, 65%, 17/26); CRu, 4.5% (2/44); PR, 2.5% (1/44). Conclusions: RBMD is an effective and safe alternative for patients with R/R FL who have received first line treatment with R-ChemoT and R maintenance therapy. This response-adapted treatment strategy achieved results similar to the scheme with 6 cycles and may allow reduction of the intensity and duration of induction therapy and minimize toxicity. No skin reactions were reported, possibly due to the inclusion of dexamethasone in the treatment scheme. Disclosures No relevant conflicts of interest to declare.

Abstract 3660 Poster Board III-596 Introduction Despite the major advances in the treatment of classical Hodgkin Lymphoma (cHL) patients, around 30% to 40% of cases in advanced stages may relapse or die as result of the disease. Current predictive systems, based on clinical and analytical parameters, fail to identify accurately this significant fraction of patients with short failure-free survival (FFS). Transcriptional analysis has identified genes and pathways associated with clinical failure, but the biological relevance and clinical applicability of these data await further development. Robust molecular techniques for the identification of biological processes associated with treatment response are necessary for developing new predictive tools. Patients and Methods We used a multistep approach to design a quantitative RT-PCR-based assay to be applied to routine formalin-fixed, paraffin-embedded samples (FFPEs), integrating genes known to be expressed either by the tumor cells and their reactive microenvironment, and related with clinical response to adriamycin-based chemotherapy. First, analysis of 29 patient samples allowed the identification of gene expression signatures related to treatment response and outcome and the design of an initial RT-PCR assay tested in 52 patient samples. This initial model included 60 genes from pathways related to cHL outcome that had been previously identified using Gene Set Enrichment Analysis (GSEA). Second, we selected the best candidate genes from the initial assay based on amplification efficiency, biological significance and treatment response correlation to set up a novel assay of 30 genes that was applied to a large series of 282 samples that were randomly split and assigned to either estimation (194) or validation series (88). The results of this assay were used to design an algorithm, based on the expression levels of the best predictive genes grouped in pathways, and a molecular risk score was calculated for each tumor sample. Results Adequate RT-PCR profiles were obtained in 264 of 282 (93,6%) cases. Normalized expression levels (DCt) of individual genes vary considerably among samples. The strongest predictor genes were selected and included in a multivariate 10-gene model integrating four gene expression pathway signatures, termed CellCycle, Apoptosis, NF-KB and Monocyte, which are able to predict treatment response with an overall accuracy of 68.5% and 73.4% in the estimation and validation sets, respectively. Patients were stratified by their molecular risk score and predicted probabilities identified two distinct risk groups associated with clinical outcome in the estimation (5-year FFS probabilities 75.6% vs. 45.9%, log rank statistic p≈0.000) and validation sets (5-year FFS probabilities 71.4% vs. 43.5%, log rank statistic p Conclusions We have developed a molecular risk algorithm that includes genes expressed by tumoral cells and their reactive microenvironment. This makes it possible to classify advanced cHL patients with different risk of treatment failure using a method that could be applied to routinely prepared tumor blocks. These results could pave the way for more individualized and risk-adapted treatment strategies of cHL patients, enabling subsets of patients to be identified who might benefit from alternative approaches Disclosures: No relevant conflicts of interest to declare.

Background and aims: Family history of cancer in lymphoma patients has largely been described. On the other hand, genetic susceptibility associated to lymphoma risk is being investigated at the present time. However, single nucleotide polymorphisms (SNPs) associated to the entity among those reporting family history of cancer have yet not been identified. In a previous study of our group, we identified that the probability of having a first-degree relative with cancer was significantly higher among the patients with lymphoid neoplasm than among the control subjects (43% vs 35%, p Material and methods: Newly diagnosed cases of lymphoma were recruited between 1998–2002 in 4 Spanish centers. Controls were hospitalized patients matched to the cases by age, gender and study center. Personal interviews were conducted in order to collect data on demographics, environmental exposure, medical and family history, including cancer. The site of cancer, age at diagnosis, and status of any affected relatives with cancer were requested. In this analysis, we only included those subjects who reported a family history of cancer. 72 SNPs in 47 genes were included in the analysis. DNA of 503 cases and 569 controls were genotyped using the TaqmanTM platform. Odds Ratios (OR) and 95% confidence intervals (CI) for the association of the variants with the risk of lymphoma were calculated using unconditional logistic regression analysis, under the log additive inheritance model. Results: Family history of cancer was reported by 421 subjects (196 controls and 225 cases, p-value=0.001). Among those, 41 had an hematological origin. The two SNP genotyped in IL8RB gene were associated to an increase in risk of lymphoma for every extra variant allele (IL8RB +1235C>T OR=1.68, 95%CI=1.25–2.27 and IL8RB -1010A>G OR=1.59, 95%CI=1.17–2.17). The OR associated to an additional variant allele of Ggh IVS7-3001 polymorphism was 1.82 (95%CI=1.06–3.14). IL-1082 A>G was observed to decrease the risk of lymphoma (OR=0.72, 95%CI=0.54–0.96).Restricting the analysis to those with a familial cancer of hematological origin under a log-additive inheritance model, a statistical increase in the risk was found for every extra variant allele MTHFR +429A>C in the genotype (OR=21.52, 95%CI=2.42–191.08), whereas presenting at least one variant allele of IL10 -1082 A>G decreased the risk of lymphoma (OR=0.05, 95%CI=0.01–0.38). In order to evaluate the association between SNPs and lymphoma subtypes in subjects with family history of cancer, we explored the relation between all statistically significant SNPs reported above and non-Hodgkin lymphomas (NHL), Chronic lymphocytic leukemia (CLL), multiple myeloma (MM), and Hodgkin lymphoma (HL). We observe that the association of SNPs and all lymphoma is similar after stratifying by subtypes, although these results are based on few subjects. Interestingly, those SNPs related to low risk of lymphoma, were observed to have the lowest risk among HL cases (IL10 -1082 A>G OR=0.28, 95%CI=0.09– 0.82), whereas HL and MM seems to be at higher risk for those SNPs related to increased risk of lymphoma (IL8RB+1235 OR=2.29, 95%CI=1.29–4.05 and IL8RB-1010 OR=2.45, 95%CI=1.33–4.51 for MM and Ggh IVS7-3001 OR=5.47, 95%CI=1.33–22.52 for HL). Conclusions: These results suggest a potential role of IL8RB, GGH IVS7 and IL10 SNPs in the risk of lymphoid neoplasms among subjects with family history of cancer. HL and MM seem to be those entities in which these associations appear to be stronger, although the number of cases included is small.

In an area endemic for visceral leishmaniasis, 16patients with human immunodeficiency virus (HIV) infection developed the disease. All belonged to populations at risk for AIDS (15 were intravenous drug abusers). Five patients fulfilled the criteria for full-blown AIDS, and two more fulfilled them after diagnosis of leishmaniasis. All presented with the classic manifestations of visceral leishmaniasis, but leishmania serology was negative in 15 patients (93%). Leishmania donovani amastigotes were identified in the bone marrow in all cases. Most patients responded initially to treatment with pentavalent antimonial drugs, but seven (43%) followed a chronic course, with multiple relapses in five, despite alternative treatments. Visceral leishmaniasis occurred in patients with different levels of depression of the CD4 to CD8 lymphocyte ratio. Mortality was 37% (six patients) and was independent of the chronic-relapsing course of the disease. In no case was leishmaniasis the primary cause of death. Our data establish that visceral leishmaniasis is an opportunistic infection in HIV-infected patients, and we suggest that in endemic areas it should be considered an indicator disease for the diagnosis of AIDS. © 1989 by The University of Chicago.

The effect of molecular factors in the outcome of Hodgkin's Disease (HD) is being currently studied. In a previous series of HD, including patients treated only with radiotherapy and patients treated with chemotherapy (with or without radiotherapy), we found that a high proliferation index had an adverse influence in overall survival (OS) and in the achievement of a complete remission (CR). Loss of Rb expression also had an adverse prognostic influence in achievement of CR. On the other hand LMP1-EBV expression had a favorable influence for OS. The expression of other molecular factors, p53, bcl2 and CD15 did not show prognostic influence. In the present paper we have studied the effect of these molecular variables in 110 patients, of the previous series who had been treated with chemotherapy. A retrospective study was performed in these 110 patients with HD treated with chemotherapy (ABVD or variants, 62%, or regimes not containing adriamycin, 38%) with or without adjutant radiotherapy, collected at the 11 centers belonging to the Spanish Collaborative Group for the Study of Hodgkin's Disease. The prognostic value of clinical variables and the expression of p53, bcl2, CD15, Rb, LMP 1-EBV and proliferative fraction demonstrated with sensitive immunohistochemical methods were studied. Cox's multivariate analysis was performed to assess their influence in failure-free survival (FFS) and OS. A multivariate logistic regression analysis was performed for studying the effect of the variables in the achievement of a CR. Of the clinical variables, only advanced stage (III/IV) had a significant independent adverse influence in FFS, in OS and in the achievement of CR and advanced age in OS. Of the molecular variables, LMP1-EBV had an independent and strong favorable influence in FFS, in OS and in the achievement of CR. Rb expression had a modest favorable influence in CR. The rest of the molecular variables had no independent influence on the outcome of the disease. In conclusion these results confirm the favorable prognostic value of LMP1-EBV expression in the subset of patients with HD treated with chemotherapy.