Search

Hypermethylation of DAPK1promoter gene was found to be a frequent epigenetic alteration in follicular lymphoma (FL). We evaluated whether the quantification of DAPK1methylation in the bone marrow (BM) and peripheral blood of FL patients at diagnosis and during follow-up provides important prognostic information. DAPK1methylation was quantitated by real-time MethyLight PCR in 107 patients at diagnosis, at end of therapy, and during follow-up. Information on BCL2-IGHrearrangement and clinical characteristics were available for all patients. Aberrant DAPK1methylation was found in 22 of 26 (85%) lymph node biopsy samples, 62 of 107 (58%) BM specimens, and 25 of 63 (40%) peripheral blood samples at diagnosis. DAPK1methylation was greater in patients with BM infiltration and a higher Follicular Lymphoma International Prognostic Index score. The presence of aberrant DAPK1methylation in BM significantly reduced progression-free survival following immunochemotherapy, independent of Follicular Lymphoma International Prognostic Index score. Residual or increased methylation after treatment was associated with an increased risk for relapse. With watchful waiting, greater DAPK1methylation at diagnosis was associated with a shorter time to antilymphoma treatment. Our study indicates that quantification of DAPK1methylation represents a prognostically relevant FL biomarker, with promising implications for risk assessment.

No relevant conflicts of interest to declare.

Tumor tissue is known to release cell-free DNA, that can be detected in peripheral blood. Cell-free DNA (cfDNA) is significantly higher and more fragmented in plasma of cancer patients and levels have been reported to associate with prognosis. The aim of our study was to further explore cfDNA as a biomarker in Hodgkin lymphoma (HL) at diagnosis, during treatment and at follow-up. We studied cfDNA levels in pre-treatment plasma samples of 163 consecutive HL patients diagnosed at our Institution and 64 healthy volunteers. cfDNA was analyzed by quantitative PCR for the beta-globin gene. cfDNA at diagnosis was significantly increased in patients with HL when compared to controls (median, 21.83 vs 12.18 ng/ml; p<0.0001). Using the cfDNA levels of normal volunteers to determine the cut-off point for pathologically elevated cfDNA levels, cfDNA levels were elevated in 62 (38%) patients. Increased cfDNA at diagnosis was associated to parameters indicating an unfavorable prognosis as an IPS score>2 (p=0.0008). In a group of 31 patients and 19 controls, we also evaluated cfDNA fragmentation, amplifying fragments of different length. DNA integrity indices were calculated according to Mouliere et al, PLoS One 2011. At diagnosis, cfDNA was more fragmented in HL patients than in controls resulting in a lower DNA integrity index (median: 0.15 vs 0.22; p=0.016). However, cfDNA in HL patients resulted less fragmented when compared to cfDNA form patients with aggressive tumors like Burkitt lymphoma or solid tumors. Patients received treatment with ABVD (101 patients), BEACOPP (42 young patients with advanced HL) and other types of chemotherapy (17 patients). Seventy patients were also evaluated after two cycles of chemotherapy. cfDNA levels significantly decreased after 2 cycles of chemotherapy in patients with increased pre-treatment levels, while they remained in the normal range in patients with low cfDNA at diagnosis (ratio 0.32 and 1.44, respectively; p<0.0001). Patients with elevated cfDNA levels at diagnosis had an inferior outcome compared to patients with cfDNA levels within the normal range both in univariate and in a multivariate analysis that included IPS and was adjusted for the type of chemotherapy (EFS, p<0.03). In conclusion, our data suggest that cfDNA could serve as a biomarker in HL, that it is frequently elevated at diagnosis of HL, tends to normalize during treatment, and is associated to an inferior outcome. The study of DNA fragmentation suggests that most of the increase of cfDNA in HL is probably due to the tumor microenvironment rather than to the neoplastic cells.

No relevant conflicts of interest to declare.

Abstract 2678

Tjalsma: hepcidinanalysis.com: Managing Director, Membership on an entity's Board of Directors or advisory committees. Swinkels:hepcidinanalysis.com: Medical Director, Membership on an entity's Board of Directors or advisory committees.

Abstract 3656

It is thought that the production of a variety of cytokines by the malignant Reed-Sternberg cells and the surrounding tissue contributes to the abnormal immune response which is a clinical and pathological feature in Hodgkin’s lymphoma. Single nucleotide polymorphism in the 5′-promoter region of cytokine genes are key factors for cytokine production and may modify the biology of the disease. Recently, differences in the prognosis according to the Interleukin-10 (IL-10) genotype have been shown in patients with diffuse large B cell lymphomas (Lech-Maranda et al, Blood2004; 103:3529). We assessed the distribution of frequencies of polymorphic allele variants in the genes of IL-10 (A-1082G and C-592A), TNF-alpha (C-863A and G-308A) and IL-6 (G-174C) in 218 patients with Hodgkin’s lymphoma and analysed for associations with patient charcteristics and prognosis. The polymorphism were analyzed using a multiplex amplification and mismatched polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) technique as described by Tseng et al (Tissue Antigens2002; 59:280). DNA was extracted either from peripheral blood or paraffin-embedded lymph node biopsies from 218 patients with Hodgkin’s lymphoma (median age 32 years, range 13–71 years; 104 females and 114 males). 206 patients were treated with standard chemotherapy regimens: 123 patients received ABVD, 33 pts a modified Stanford V regimen (substituting 6 mg/m2 metchloramine with 650 mg/m2 cyclophosphamide), 29 pts MOPP (±ABVD), 21 pts BEACOPP. The distribution of allele frequencies in Hodgkin’s lymphoma at position -592 of the IL-10 gene was as follows: 43 % were homozygous for the CC genotype, 41 % were heterozygous and 16 % were homozygous for the AA genotype. The IL10 -592AA genotype was associated with a decreased progression-free survival (p=0.0057). The probability of progression-free survival at median time of observation of 4 years for patients homozygous for the IL-10 -592 AA genotype was 33 % (95 % C.I, 14–54 %), while for heterozygous patients and for patients homozygous for the −592 C allele it was 74% (95% C.I., 61–83). When the analysis was restricted to 123 patients treated with ABVD chemotherapy, essentially the same differences in progression-free survival were observed. No associations between genotype distributions of TNF alpha at position −308 and −863, IL-6 at position −174, and IL-10 at position −1082 and clinical variables and prognosis were found. In univariate analysis of established prognostic factors, stage proved to be of prognostic value in our patient group (limited disease in stage I–IIA vs advanced disease in stage IIB–IV, p=0.012) The Cox multivariate analysis showed that IL-10 -592AA genotype and stage were independent prognostic factors (p=0.01 and 0.018, respectively). Our study indicates that the cytokine genotype can predict clinical outcome in patients with Hodgkin’s lymphoma and points to the importance of the genetic background of the host.

It is thought that the production of a variety of cytokines by the malignant Reed-Sternberg cells and the surrounding tissue contributes to the abnormal immune response which is a clinical and pathological feature in Hodgkin's lymphoma. Single nucleotide polymorphism in the 5′-promoter region of cytokine genes are key factors for cytokine production and may modify the biology of the disease. Recently, differences in the prognosis according to the Interleukin-10 (IL-10) genotype have been shown in patients with diffuse large B cell lymphomas (Lech-Maranda et al, Blood 2004; 103:3529). We assessed the distribution of frequencies of polymorphic allele variants in the genes of IL-10 (A-1082G and C-592A), TNF-alpha (C-863A and G-308A) and IL-6 (G-174C) in 218 patients with Hodgkin's lymphoma and analysed for associations with patient charcteristics and prognosis. The polymorphism were analyzed using a multiplex amplification and mismatched polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) technique as described by Tseng et al (Tissue Antigens 2002; 59:280). DNA was extracted either from peripheral blood or paraffin-embedded lymph node biopsies from 218 patients with Hodgkin's lymphoma (median age 32 years, range 13–71 years; 104 females and 114 males). 206 patients were treated with standard chemotherapy regimens: 123 patients received ABVD, 33 pts a modified Stanford V regimen (substituting 6 mg/m2metchloramine with 650 mg/m2cyclophosphamide), 29 pts MOPP (±ABVD), 21 pts BEACOPP.