PERIPHERAL T-CELL LYMPHOMA, UNSPECIFIED – THE ANALYSIS OF THE DATA FROM THE CZECH LYMPHOMA STUDY GROUP (CLSG) REGISTRY

BACKGROUND
Peripheral T-cell lymphoma, unspecified (PTCL-US) is one of the entities from the infrequent family of nodal mature T-cell lymphomas. The clinical course is aggressive, and despite multiagent chemotherapy, the median survival is about 2 years. Published data are limited to retrospective, mostly single-center studies or reviews and usually include more lymphoma subtypes.


AIM
To evaluate the current treatment modalities, clinical outcome and prognostic factors in unselected, new diagnosed patients with PTCL-US in the population of the central european region (Czech Republic).


METHOD
Czech Lymphoma Study Group is a national scientific organization which provides an on-line database registry which collects a data about almost all new diagnosed lymphoma patients since year 2000. All diagnostic biopsies were reviewed by a reference pathologist.


RESULTS
We analyzed 63 patients with new diagnosis of PTCL-US. The median age was 59 years (25-81), chemotherapy (CHT) was administered in 56 of the 63 patients: anthracyclin-based CHT in 51%, intensive CHT in 21% and non-anthracyclin regimen was applied in 13% of the patients. The overall response rate was 74.4%, (CR in 57.4%). After a median follow-up of 19.6 months, 41% of the patients were in CR, 3.4% in PR or stable disease and 55% of the patients died. The estimated survival probability in 3 years was 36%. Clinical stage (IV) and CR achievement were found to be independent survival predictors in a multivariate analysis.


CONCLUSIONS
Although the current treatment modalities are mostly ineffective in PTCL-US, appropriate intensive treatment may lead to prolonged remission but not survival.


INTRODUCTION
Peripheral T-cell lymphomas (PTCLs) are infrequent types of non-Hodgkin's lymphomas (NHLs).In Western countries they represent about 7 % of NHLs 1 , the frequency varied from 1.5 % to 18.3 % in diff erent regions.Highest incidence of PTCLs have been documented in Hong Kong and Taiwan 2 , this fact may refl ect increased exposure to oncogenic factors such as human T-cell leukemia virus-1 (HTLV-1) and Epstein Barr virus (EBV) in Asian nations.Published data from the Surveillance, Epidemiology, and End Results (SEER) registries in the USA, which included 114 548 lymphoid neoplasms diagnosed during 1992-2001 show evident increase of T-cell lymphoma incidence (up to 6.4 % per year)(ref. 3).
Recently published World Health Organization (WHO) Classifi cation of Tumours 4 subdivide T-cell lym-phomas into predominantly leukemic, extranodal and nodal types.There are three groups of primary nodal PTCLs: anaplastic large cell lymphoma, angioimmunoblastic lymphoma and a broad category of peripheral Tcell lymphoma-unspecifi ed (PTCL-US).PTCLs usually presents with disseminated, often extranodal disease and outcome of the patients is much worse than in B-cell counterpart lymphomas [5][6][7] .Current treatment modalities did not dramatically improve outcome of the patients and 5-year overall survival still remains between 30 %-35 % using standard chemotherapy with second and third generation regimens 7 .The role and timing of high dose therapy with autologous stem cell support (ASCT) remains unclear -some studies proved survival advantage of ASCT 8,9 , others brought inconsistent results 10,12 .Nonanaplastic T-cell phenotype was found as negative independent prognostic factor both in patients treated with 10,13 104 V. Prochazka, M. Trneny, R. Pytlik, I. Vasova, Z. Kral, D. Belada, T. Kozak, K. Kubackova, H. Siff nerova, M. Matuska, M. Lysy, I. Bolomska, K. Petrakova, B. Otavova, J. Pribylova, J. Svecova, T. Papajik, M. Hamouzova, M. Petrova, J. Zapletalova, K. Langova or without 7 high dose therapy with ASCT.Promising longterm results were obtained only in younger patients after allogeneic stem cell transplantation 11 .To predict survival in PTCL patients is widely used International progostic index (IPI) 14 , whose prognostic power was tested in patients with PTCLs with satisfactory results 15 .Novel prognostic scores -"T-cell lymphoma specifi c" were published in last years.The most predictive ones are: PIT score which include 4 factors (age > 60 years, PS ≥ 2, elevated LDH level, bone marrow attainment) 16 and Novel clinical-pathologic progostic score (age > 60 years, elevated LDH evel, PS ≥ 2, Ki-67 ≥ 80 %) 17 .
Objective of our retrospective multicenter study was to analyze the epidemiology, clinical features, prognostic factors, treatment modalities and outcome in the unselected patients with PTCL-US.

PATIENTS AND METHODS
Czech Lymphoma Study Group (CLSG) is a national scientifi c organization which provides a background for collaboration of hematologists, oncologists and hematopathologists.The Lymphoma registry is an on-line database founded and operated by CLSG to collect a data of all new diagnosed lymphoma patients since year 2000.We analyzed a data of patients with PTCL-US, who were enrolled into our clinical registry between January 2000 and May 2006.No patient selection was done.A data from 3043 patients with non-Hodgkin's lymphomas was obtained during last 6 years.PTCL-US group was represented by 63 cases (2.07 %).The clinical data included age, sex, lactate dehydrogenase (LDH) level, Ann Arbor stage, IPI, extranodal involvement, systemic symptoms, performance status (PS; ambulatory Eastern Cooperative Oncology PS 0-1 v nonambulatory PS ≥2), presence of systemic (B) symptoms, date of last follow-up, status (alive or dead), type of treatment and chemotherapy protocol.
All biopsies were reviewed by a reference pathologist and fi nal diagnosis was provided in compliance with published World Health Organization (WHO) Classifi cation of Lymphoid Tumours.

Treatment strategies
Most of the patients were treated with CHOP or CHOP-like anthracyclin-based chemotherapy 51 % (32 of 63), 21 % (13 of 63) patients recived intensive protocols (MegaCHOP-ESHAP, PROMACE-CytaBOM, Hyper-C-VAD) -intensive protocol was terminated with ASCT in 6 of them.No allogeneic stem cell transplant was indicated.Non-anthracyclin regimen was applied in 13 % (8 of 63) of the cases.No therapy was given in 3 patients due to very poor performance status. 1 patient undergone only radiotherapy and 1 patient only surgical treatment.We have no data about 6 % (4 of 63) of the patients.
The treatment responses: complete response (CR), unconfi rmed CR (uCR), partial response (PR) stable disease and progressive disease were classifi ed acording the International Workshop NHL Response Criteria published by Cheson et al 18 .
Statistical methods.Our data were analyzed with Statistical Package for the Social Sciences (SPSS) 19 .Overall survival (OS) was defined as the time from fi rst treatment to the date of last follow-up examination (censored) or the date of death (event) from any cause.Progression free survival was defi ned as the date of fi rst treatment to the date of documented disease progression or death (event) or the date of last follow-up examina-  20 was used to derive survival probabilities.The log-rank test was used to compare diff erences in survival times between patient subgroups.Pearson Chi-Square test and the Fisher's Exact Test was used to test the associations of bone marrow involvement, clinical stage, splenic involvement, extranodal involvement, performace status (≥ 2), elevated lactate dehydrogenase level, IPI score, achievement of complete remission, intensive chemotherapy, relapse and CR achievement or exitus probability.Cox proportional hazards regression analysis (Forward Stepwise method) 21 was used to model the association of prognostic factors and survival time.The p value signifi cance limit was assessed as .05,2-sided tests were used in all calculations.
The EMBASE and PubMed databases were searched for literature review.

RESULTS
We analyzed a data of the 63 patients, baseline clinical parameters are summarized in Tab. 1.The median age was 59 years (range, 25-81 years), there was a male preponderance (male-to-female ratio 1.4 : 1).Most of the patients had advanced disease clinical stage III and IV: 48 of 63 (76.2 %) with extranodal lymphoma involvement in 70.9 % of the cases.Most frequently was involved bone marrow (30.2 %), followed by spleen (25.4 %).B-symptoms were present in more than half of the patients (53 %).More than two thirds of the patients had ambulatory performance status (PS 0-1, 43 of 63, 68 %).Lactate dehydrogenase (LDH) levels were elevated in more than half of the cases (>1 X normal value, 34 of 60, 56.7 %).IPI score was aviable in 61 (97 %) of 63 of the patients; 33 patients (54 %) were classifi ed as low risk (0-1), 18 (29.5 %) as intermediate-low (IPI 2) and 10 (16.4 %) as intermediate-high (IPI 3).No patients with high risk IPI score were enrolled into the study.

Analysis of response
In 4 (6.6 %) of the patients the treatment modality is unknown, 3 (4.9%) patients did not recived any treatment.In 9 (14.8 %) of the patients was not the treatment response analyzed.Finally, the treatment response was recorded in 83.9 % (47 of 56) of the patients.Overall CR rate was 57.4 % (27 of 47), PR rate 17 % (8 of 47) and 25.6 % (12 of 47) of the patients had stable or progressive disease after fi rst line treatment.Follow-up data were aviable in 94.5 % (58 of 63) of the patients.After a median follow-up of 19.6 months 41 % (24 of 58) of the patients were in CR, 3.4 % (2 of 58) in PR or SD and 55 % (32 of 58) of the patients died.

DISCUSSION AND CONCLUSION
Last years have brought an explosion of a new information about tumor cell origin 22 in PTCLs.Expression profi les both on DNA 23 or protein 17,24 level were widely studied.The prognostic impact of Epstein -Barr virus postitivity of the tumor cells using In Situ Hybridization (EBER-ISH) was confi rmed 28 .New subclassifi cations of heterogenous group of PTCL-US were provided by a novel "T-cell lymphoma specifi c" prognostic scores 16,17 .Unfortunately, none of these advances on the fi eld of preclinical research were not translated into better outcome of PTCL-US patients.PTCL-US still represent a heterogenous group of aggressive lymphomas with disapponting treatment results.Gisselbrecht et al report a 5-year OS for PTCLs 35 % 7 , Escalón et al. published even worse results both in patients treated intensively or with CHOP regimen (3-years OS 49 % v 43 %)(ref. 25).The appropriate dose intensity of the chemotherapy and role of high dose therapy with ASCT are still unclear due to heterogenity of studied populations and frequent inclusion of anaplastic PTCL or B-cell lymphoma subtypes.
Here we have reported a data of unselected group of patients with PTCL-US from the CLSG registry.The overall low incidence of PTCL-US (2.07 %) and also baseline clinical characteristic of the patients correspond with previously published data [4][5][6][7] .
First line treatment was based mostly on CHOP or CHOP-like regimens, less frequently was applied intensive chemotherapy with or without ASCT or non-anthracyclin combinations.Overall CR rate was 57.4 %, PR rate 17 % (8 of 47) and 25.6 % (12 of 47) of the patients had stable or progressive disease after fi rst line treatment.Achievement of CR was more freguently achieved using intensive chemotherapy and was associated with lower probability of death.Patients in CR had signifi cantly longer PFS, but not OS.The estimated survival probability of all patients in 3 years was 36 %.Half of the patients died during the fi rst 24 months.Patients with low IPI have longer survival in contrast to those with intermedi-ate low/intermediate high IPI (3-year OS 43 % v 12 %, p = 0.0052).Overall poor survival may be aff ected by enrolment of the patients without any treatment or with up-front paliative treatment.
We conclude that patients with PTCL-US have one of the worst prognoses among all non-Hodgkin's lymphoma subtypes patients.The current treatment modalities are mostly ineff ective or lead only to response with limited duration.The therapy is still based on protocols primary designed for B-cell lymphomas.Total dose intensity studies were not performed 26 .No "big step forward" as observed in the rituximab era in B-cell lymphomas was acquired.The future progress in management of PTCL-US patients must include wide collaboration of all institutions, application of new tissue microarray immunohistochemistry, novel prognostic scores, design of new treatment strategies including chemo-immunotherapy 5,27 and allogeneic stem cell transplantation 11 .

Table 2 .
Clinical characteristics infl uencing overall survival in multivariate anaysis.Peripheral T-cell lymphoma, unspecifi ed -the analysis of the data from the Czech Lymphoma Study Group (CLSG) registry tion (censored).The Kaplan-Meier method