Multiple Extranodal Involvements; Significant Risk Factor of Relapse In Peripheral T-Cell Lymphoma, Not Otherwise Specified

Peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS) comprises clinically and histopathologically a heterogeneous group of predominantly nodal diseases which do not fit into definition of any other identified subtype of PTCLs. Most of cases are advanced diseases and characterized by aggressive behavior and very poor prognosis. Although several risk factors for response and survival in PTCL-NOS have been reported, little information is available for the incidence and risk factors of relapse after complete response (CR) achievement. Further analyses of the relapse will give valuable information for effective therapeutic strategies for PTCL-NOS. The aim of this study is to evaluate the incidence and risk factors of the relapse in PTCL-NOS patients.

We retrospectively reviewed 107 patients with PTCL-NOS diagnosed by The 4^th edition of the WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues in Hokkaido Hematology Study Group, which includes 30 hematology/oncology or pediatrics departments of 23 institutes, from January 2002 to December 2009. We further investigated the incidence, clinical backgrounds, and risk factors of relapse of the PTCL-NOS patients. Response to treatment was assessed by international workshop to standardize response criteria for non-Hodgkin's lymphomas (Cheson BD et al. J Clin Oncol. 1999). Overall survival (OS) curves were estimated using Kaplan-Meier method and compared by log-rank test. The association between clinical factors and relapse rate was evaluated in univariate analysis by the chi-squared test. Factors independently associated with relapse or OS were identified in multivariate analysis by logistic regression model or Cox proportional hazards model, respectively.

The median follow-up of the patients was 24 months (range 1–95). The sample population included 70 males and 37 females with a median age of 67 years (range 9–94). Chemotherapy (ChT) was selected in 90% (96/107) patients as the primary treatment. CHOP-lile regemens were chosen in 91% (86/96) patients as the primary ChT. The estimated 5 year-OS of all the patients was 35%. 48 (52%) patients achieved CR after primary treatment in the evaluable 92 patients, and were futher analyzed. Relapse was occured in 46% (22/48) patients with a median remission duration of 9 months (range 2–51). Multiple or diffuse extranodal relapses were presented in 64% (14/22) patients. Only 5 (24%) paitients achived 2nd CR in the 21 relapsed patients recived salvage ChT. Survaval of patients with relapse was significantly shorter than that without relapse (5 year-estimated OS 32% vs 100%, p<0.0001). Factors of high relapse rates at diagnosis were as follows; central nervous system involvement (100%), extranodal involvement sites>1 (87%), gastrointestinal involvement (80%), bulky disease (67%), bone marrow (BM) involvement (67%), high (H) and high-intermediate (HI) risk groups of International Prognositic Index (IPI) (63%). In addition, relapse rate of the patients showing FDG-uptake by PET after primary treatment was 100%. In univariate analysis, extranodal involvement sites at diagnosis >1 (p=0.022), H and HI risk groups of IPI at diagnosis (p=0.021), FDG-uptake by PET after primary treatment (p=0.043) were identified as risk factors of relapse. Age, clinical stage, B-sympton, perfomance status, BM imvolvement, bulky disease, high LDH level, risk groups of prognositic index for PTCL-U (PIT), and regemen or dose-reduction of primary treatment were not significant indicators for relapse in univariate analysis. In multivariate analysis, extranodal involvement sites at diagnosis >1 was identified as an independent risk factor not only for relapse (relative risk 26.0; 95% CI 1.115–605.9; p=0.043) but also for OS (relative risk 10.8; 95% CI 1.1466–80.13; p=0.02).

Almost half of PTCL-NOS patients who achieved CR after primary treatment will relapse and the prognosis is poor. Multiple extranodal involvements at diagnosis will be a predictive factor of relapse and survival. These findings warrant further studies regarding specific treatment approaches for PTCL-NOS patients with multiple extranodal involvements to inhibit the relapse.

No relevant conflicts of interest to declare.