Chromosomal Aberrations In Peripheral T-cell Lymphoma , Not Otherwise Specified

Background and Objective Peripheral T-cell lymphoma,not otherwise specified(PTCL-NOS)comprises a heterogeneous group of haematological tumours, which originates from mature T-cells, and constitutes near 20% of all non-Hodgkin's lymphomas (NHLs) in adults in China, and is biologically and clinically heterogeneous. And it comprises pathologically and clinically heterogeneous groups with aggressive behaviors, poor response to therapy and dismal prognosis. Althrough a number of cytogenetic studies had been conducted, the genetic basis of this lymphoma entity has still being unclear. The purpose of this study was to identify the genetic changes in PTCL-NOS cases and then find the key alterations in the cancer pathogenesis and prognosis, and meanwhile, to try to find the subgroups of the tumor with distinct genetic, histopathologic and prognostic features.Methods Comparative genomic hybridization (CGH) is a molecular technology, which uses different dyes (Cy3: blue; Cy5: red) labeling on tumor DNA and normal DNA hybridizing on the metaphase chromosome. After calculating the ratio of the dyes, we can get the abnormal copy numbers and locations of the tumor DNA. The array-CGH with hybridized DNA on the array makes the method much more sensitive and accurate. Totally 37 cases of PTCL-NOS were investigated by 1Mb resolution array-CGH in this study, in which some genetic imbalances were further studied by using a Tile path array-CGH and fluorescence in situ hybridization (FISH). Based on the clonality results, morphologic changes and genetic imbalances, we finally selected the number of the cases used in this study. The brief procedures were as follows:1. Paraffin-embedded tissue samples of 47 cases of PTCL-NOS (from Oct. 2001 to Dec. 2008) were retrieved from the 306 Hospital of PLA, Shanxi Tumor Hospital and Shanghai Changhai Hospital. Crude microdissection was performed on 5 m paraffin-embedded sections, based on the immunohistochemical staining with antibody CD3 and haematoxylin-stained tissue sections, to avoid necrotic areas and connective tissues, and to isolate various numbers of tumor cells.2. DNA was extracted and purified using a DNA micro kit. The integrity of DNA samples was assessed by using the method of multiplex PCR, to identify variable size of DNA fragments (100bp, 200bp, 300bp and 400bp).3. Assessment of lymphocyte clonality was analyzed by BIOMED-2 multiplex PCR of T-cell receptor (TCR) gene rearrangements.4. 37 cases of PTCL-NOS were investigated by 1Mb resolution Array-CGH. Then some imbalances on chromosomes 1, 6, 7 and 14 were further studied by a Tile path array-CGH.5. Finally, FISH was used to confirm or exclude some unconfirmed imbalances in the array-CGH findings. Meanwhile, copy number changes of chromosomes could be identified with this method.Results1. 47 cases of PTCL-NOS were selected for DNA extraction after immunohistochemical staining and crude microdissection. The multiplex PCR analysis of the extracted DNA showed that 400 bp fragment was found in 17 cases, 300 bp in 10 cases,200 bp+ in 10 cases, 200bp in 8 cases and 100 bp in 2 cases. The clonality analysis demonstrated that 26 of the total 37 cases had clonal TCRγgene rearrangement and 10 cases were polyclonal or germline configuration of TCRγgene, and 1 case showed an equivocal result. Array-CGH analysis showed that various DNA gains and losses were detected in 18 of the 37 PTCL-NOS samples. Based on the polyclonal results, morphologic changes and genetic imbalances, 6 cases were excluded and therefore 31 cases were finally used in this study.The clinical data analysis showed that there were 21 males and 10 females, with mean age of 52 years (ranged from 12 to 75 years). The follow-up data including survival and overall outcome were obtained in 25 cases out of 31 cases, in which 20 cases died and 5 cases were still alive. There were 11 cases occurred in lymph node and 18 cases in the outside of lymph node. When the diagnosis was established, 3 cases were in the stageⅢ, 9 cases in stageⅣand others unclear.2. There were genetic alterations in 17 cases (55%). The regions of frequently-occured genomic gains (≥3 cases) were 1p36.13-1p36.32, 1q21.1-1q23.3, 3p24.2-3p24.3, 3p13-3p14.3, 7q22.1, 7p22.1-7p22.3, 7q36.1-7q36.3, 7q32.1-7q32.3, 7q22.1-7q34, 8q11.21-8q22.3, 8q24.3, 9p11.2-9q12, 9q33.3-9q34.3, and 11p15.5, in which 1p36.13-1p36.32, 7q22.1, 7q36.1-7q36.3, 7q32.1-7q32.3, 7q22.1-7q34, 9p11.2-9q12 and 9q33.3-9q34.3 are the most frequent gains (≥4 cases). The regions of frequent genomic losses (≥3 cases) were 1p21.2 -1p12), 5q21.7-5q23.2, 6q24.2-6q25.3, 6q27, 6q22.1, 9p21.1-9p21.3, 10q25.3-10q26.3, 10p11.23-10q11.21, 10q11.23-10q25.2, 12p13.1-12p13.2, 12q24.11-12q24.3, 13q14.11-13q14.3, 13q21.3-13q22.2 and 13q13.3-13q14.3, in which 1p12-1p21.1 and 13q14.11-13q14.3 are the most frequent loss regions (≥4 cases). 3. Using the Kaplan-Meier, Pearson chi-square test and Cox regression analyses, we found that changes of 1p36.13-1p36.32 (including genes FGR and TP73),and losses of 10q, 12p and 13q were negative factors for the evaluation of prognosis.4. Tile path array-CGH and FISH confirmed some of the above genetic imbalances, meanwhile confirmed and excluded some equivocal findings in 1Mb array-CGH analysis. We also found that there were losses of single copies in 7q34 and 14q11 by FISH.Conclusion1. CGH is a useful tool for the identification of gene alterations and locations in cases of tumors and that the aforementioned genes might represent potential candidate genes involved in the pathogenesis. The subsequent array-CGH developed in resolution and accuracy. Our results contribute to a complete description of genomic structural aberrations in relation to lymphoma, provide a valuable basis with understanding the characteristics of the tumor, predict patient's outcomes and discover novel therapeutic targets for lymphoma.2. Most of the PTCL-NOS cases with genomic imbalance histologically show pleomorphic nuclear atypia. The use of array-CGH enables us to identify frequently altered genomic regions, which may play a significant role in development of this tumor. However the significance of these genetic imbalances remains poorly understood. Our study shows many frequently-occured genomic imbalance regions in PTCL-NOS.3. Our study also shows that among the morphologically and immunohistochemically heterogeneous group of PTCL-NOS, genetic features may segregate at least one particular group with change of 1p36.32-36.13 (involving an oncogene FGR and tumor suppressor genes NBL1 and TP73) or loss of 1p21.1-12 (involving RBM15).4. The follow-up data indicate significant difference between groups with genetic imbalances in overall survival. The overall survival of the high-grade group with dramatic genetic imbalances (5 cases,≥6 regions/case) is shorter than that of the low-grade group without dramatic genetic imbalances (20 cases, <6 regions/case). Therefore, the high-grade group has a worse prognosis. The group of change 1p36 has a independently worse prognosis than group without change of 1p36. The losses of 10q, 12p and 13q are three independent poor prognostic factors as well. There is a tendency that the overall survival with loss of 1p12-21.1 is shorter than that without. We could confirm the correlation between loss of 1p12-21.1 and poor prognosis by further analyses of more cases. The above parameters are all negative factors for the evaluation of prognosis.5. FISH could confirm the recurrent genomic gains and losses and show changes of copy numbers, which is useful for finding the most important changes in PTCL-NOS. More genetic imbalances in PTCL-NOS cases will be found with their clinical implications by further studies.