Clinical characteristics and prognostic factors of patients with mature T-cell lymphoid malignancies: a single-institution study of 225 cases

Abstract

Mature T-cell lymphoid malignancies comprise a group of heterogeneous diseases that vary in clinicopathological features, biological behavior, treatment response, and prognosis. Bone marrow (BM) infiltration is more commonly present in mature T-cell lymphoid malignancies compared with their B-cell counterparts and hence important for differential diagnosis. In this study, clinical characteristics and prognostic factors were analyzed in 225 patients with mature T-cell lymphoid malignancies treated in a single institution. These included 29 cases of T-cell lymphoproliferative disorders (T-LPD, all with BM infiltration) and 196 cases of T-/natural-killer-cell lymphoma (T/NKCL, 56 with BM infiltration and 140 without BM infiltration). The estimated 5-year overall survival (OS) rates of T-LPD and T/NKCL were 96.6% and 37.3%, respectively. T-LPD patients were less likely to exhibit poor performance status, advanced disease stage, presence of B symptoms, or abnormal level of serum β-2 microglobulin. With similar pathological characteristics, T/NKCL patients with BM infiltration showed significantly lower response rates and shorter OS than those without BM infiltration (P = 0.0264 and P<0.0001, respectively). Multivariate analysis indicated that poor performance status, advanced disease stage, elevated serum lactate dehydrogenase level, and BM involvement were independent unfavorable prognostic factors. The Glasgow Prognostic Score may be more efficient than the International Prognostic Index in predicting disease outcome in T/NKCL. In conclusion, clinical characteristics may be useful in more effectively stratifying patients with mature T-cell lymphoid malignancies.

Keywords

mature T-cell lymphoid malignancies / clonal T-cell population / bone marrow infiltration / prognostic factors

Introduction

Mature T-cell lymphoid malignancies, which are more common in Asian countries than elsewhere in the world [ 1, 2], comprise a group of heterogeneous diseases that vary in clinicopathological features, biological behavior, treatment response, and prognosis. The clinical presentation of T-cell lymphoid malignancy is somewhat atypical; subsets of patients with T-cell lymphoproliferative disorders (T-LPD) and T-/natural-killer-cell lymphoma (T/NKCL) are easily confused with each other in their early stages. Instead of symptomatic primary lesions, occult clonal T-cell populations in bone marrow (BM) are more commonly seen as the initial presentation of mature T-cell lymphoid malignancies than in their B-cell counterparts [ 3− 5], which merits the need to distinguish T-LPD from T/NKCL.

T/NKCL is often accompanied by extranodal involvement, particularly BM infiltration, thereby leading to poor disease outcome [ 6]. Studying the clinical differences between patients with and without BM involvement, as well as the prognostic factors in these two subgroups, is important. Moreover, clinical prognostic scoring systems are frequently applied in clinical practice. Originally described in aggressive lymphoma, the International Prognostic Index (IPI) successfully predicts disease outcome [ 7− 9]. The Glasgow Prognostic Score (GPS), an inflammation-based prognostic score system, which involves the determination of serum C-reactive protein (CRP) and albumin levels, shows significant prognostic value in several solid tumors [ 10− 12] and diffuse large B-cell lymphoma (DLBCL) [ 13, 14]. However, which system is superior for T/NKCL prognostication remains unknown [ 15].

To better understand the clinical characteristics of T-LPD and T/NKCL with BM infiltration, as well as the role of BM involvement in T/NKCL, we conducted a retrospective study on 225 patients with mature T-cell lymphoid malignancies treated in a single institution.

Patients and methods

Patients

From January 2004 to March 2014, a total of 225 patients diagnosed with T-LPD or T/NKCL in our institution were enrolled in the study. Patients with T-LPD include those with mycosis fungoides (MF, 22 cases), T-cell large granular lymphocytic leukemia (T-LGL, 4 cases), hydroavacciniforme-like lymphoma (HV, 1 case), lymphomatoid papulosis (LyP, 1 case), and EBV-positive T-cell lymphoproliferative disease of childhood (CEBV+ T-LPD, 1 case). The patients with T/NKCL comprise those with peripheral T-cell lymphoma not otherwise specified (PTCL-NOS, 101 cases), nasal-type extranodal NK/T cell lymphoma (nasal NKTCL, 48 cases), angioimmunoblastic T-cell lymphoma (AITL, 21 cases), anaplastic large cell lymphoma (ALCL, 21 cases), and hepatosplenic T-cell lymphoma (HSTCL, 5 cases). The diagnoses were established according to the World Health Organization (WHO) classification [ 16], and all cases were reviewed independently by two pathologists (XC Fei and XY Chen). The study was approved by the Institutional Review Board with informed consent obtained in accordance with the Helsinki Declaration.

Treatment regimens

Among the 196 T/NKCL patients, 167 (85.2%) received cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP)-based chemotherapy, and 17 (8.7%) received Hyper-CVAD (cyclophosphamide, vincristine, doxorubicin, and dexamethasone, followed by high-dose methotrexate and cytarabine) as initial treatment. The remaining 12 patients (6.1%) received only supportive care because of poor performance status or insufficient organ function in these cases.

Among 29 T-LPD patients, 20 MFs received retinoic acid and local UV irradiation, and 9 cases of T-LPD did not receive treatment because of the indolent course of their disease.

Response criteria

Treatment response was evaluated according to the WHO response criteria. Complete response (CR) was defined as no evidence of residual disease, partial response (PR) as at least 50% reduction in tumor burden from the onset of treatment, and no response (NR) as less than 50% reduction in tumor burden or disease progression.

T-cell receptor (TCR) gene rearrangement assay

DNA was extracted from bone marrow mononuclear cells using Promega tissue kits (Wizard Genomic DNA Purification Kit; Madison, Wisconsin, USA) according to the manufacturer’s instructions. TCR-γ and TCR-β gene rearrangement were assessed by multiplex PCR analysis on polyacrylamide gel electrophoresis. Multiplex PCR was performed using BIOMED-2 multiplex kits (YuanQi Biomed, Shanghai, China) according to the manufacturer’s instructions.

Statistical analysis

Overall survival (OS) was the time from initial diagnosis to death or the last follow-up. Progression free survival (PFS) was calculated from the time of initial diagnosis to the time of progression, relapse, second-line therapy, or death from any cause. The inclusive period for the PFS of the patients who did not experience any of these events was terminated at the time of last contact. The OS and PFS were estimated using the Kaplan-Meier method and two-tailed log-rank test. Multivariate survival analysis was performed using a Cox regression model. All data were analyzed using Graph Pad Prism version 6.00 (Graph Pad Software, La Jolla, California, USA; www.graphpad.com). P<0.05 was considered statistically significant.

Results

Clinical characteristics of the 225 patients with mature T-cell lymphoid malignancies

The median age of the 225 patients was 52 years old (in the range 13−83 years old), and 130 (57.8%) were below 60 years old. The male-to-female ratio was 2:1. A total of 196 T/NKCLs (56 cases with BM infiltration and 140 without BM infiltration) and 29 T-LPDs (all with BM infiltration) were included in the study. The clinical characteristics of these patients are shown in Table 1.

Compared with the T/NKCL patients, the T-LPD patients exhibited better performance status (P<0.0001), fewer B symptoms (P<0.0001), and lower levels of serum β-2 microglobulin (P = 0.0192) (Table 1). Hepatitis B Virus and Epstein-Barr virus (EBV) are two important etiological factors of lymphoma, particularly in Asian countries [ 17− 19]. Among the T/NKCL patients, 27 (13.8%) and 54 (27.6%) cases were positive for HBsAg and EBV antibodies, respectively. No statistical difference was observed between T-LPD and T/NKCL (P = 1.000 and P = 0.5087, respectively, Table 1).

The clinicopathological characteristics of T/NKCL according to BM status are listed in Table 2. Significant differences were observed in performance status (P<0.0001), Ann Arbor stage (P<0.0001), extranodal involvement (P<0.0001), serum β-2 microglobulin (P<0.0001), serum CRP (P = 0.0004), white blood cell count (P = 0.0161), hemoglobin count (P = 0.0133), platelet count (P = 0.0286), and IPI stratification (P = 0.0276).

Clinical characteristics of 85 patients with clonal T-cell populations in BM

Clonal T-cell population in BM, under certain circumstances, may be the initial presentation at disease onset, increasing the difficulty in differentiating between T-LPD and T/NKCL. To better understand the clinical characteristics of the two subgroups, we further analyzed the clinical characteristics of 85 patients that presented with clonal T-cell populations in BM, among whom 29 were confirmed with T-LPD by bone marrow investigation and 56 with T/NKCL by biopsy of other lesions.

Compared with the T-LPDs, the T/NKCLs presented with poor performance status (P<0.0001), frequent B symptoms (P = 0.0001), and increased serum β-2 microglobulin (P<0.0001) (Table 3).

Immunophenotype and TCR rearrangement profile of 85 patients with a clonal T-cell population in BM

Flow cytometry allows physicians to distinguish among specific phenotypes of clonal T cells and is crucial in the diagnosis of mature T-lymphoid malignancies with BM infiltration; flow cytometry shows high specificity and sensitivity [ 20]. TCR rearrangement, which occurs during T-cell maturation, is another useful biomarker of T-cell lymphoid malignancies [ 21]. TCR rearrangement detected in the BM is not only a key clue to T-cell clonal proliferation but also an indispensable indicator of minor residual disease.

As shown in Table 4, PTCL-NOS was commonly positive for CD2 (100%), CD4 (72.7%), and TCR-β rearrangement (90.9%). By contrast, CD2 (100%), cytotoxic-granule-associated protein TIA (100%), and CD56 (86.7%) were more frequently expressed in nasal NKTCL, all of which did not involve TCR rearrangement. In AITL, the tumor cells known as follicular helper T cells were 100% positive for CD5, along with high proportions of CD2 (87.5%) and TCR-β rearrangement (100%). Most of the ALCL cases expressed CD2 (83.3%) and CD4 (83.3%), with either TCR-β (50.0%) or TCR-γ (50.0%) rearrangement. CD3 (100%) and CD7 (80%) were highly expressed in HSTCL, with all cases being TCR-γ positive. The majority of T-LPDs, like MF, were positive for CD2 (90.9%), CD5 (86.3%), TCR-β rearrangement (68.2%). Most of the T-LGLs expressed CD3 (75.0%) and CD8 (75.0%) and presented TCR-β rearrangement (100%). Some of the types of T-LPDs included only limited numbers of cases (1 HV, 1 LyP, and 1 CEBV+ T-LPD). Hence, the tumor phenotypes in such diagnoses could not be thoroughly concluded, but most of the cases were positive for CD3, CD4, and CD5, as well as TCR-β rearrangement.

Except for AITL, the T/NKCLs were more likely to lose CD5 (P = 0.002) expression, whereas T-LPD tended not to express CD7 (P = 0.0058), but more frequently expressed CD3 (P = 0.0011). In terms of the TCR-β and TCR-γ rearrangements, the difference between the T/NKCLs and the T-LPDs was not significant (P = 0.6411 for TCR-β and P = 0.1125 for TCR-γ, Table 4).

Treatment and outcome

CR was achieved in 49.0% (96 cases) of 196 T/NKCLs, with an ORR of 63.3% (124 cases). Response rate was significantly lower in patients who suffered from BM infiltration than those who did not (CR: 35.7% vs. 54.3%, P = 0.0264; ORR: 51.8% vs. 67.9%, P = 0.458; Table 2).

By univariate analysis we determined that the factors significantly associated with a lower probability of achieving CR include age older than 60 years old (P = 0.0417), poor performance status (P = 0.0233), BM involvement (P = 0.0138), advanced Ann Arbor stage (P = 0.0467), the presence of B symptoms (P = 0.0249), multiple extranodal involvement (P = 0.0117), and elevated serum lactic dehydrogenase (P = 0.0335). Moreover, IPI proved to be highly related to prognosis (P<0.0001). Patients with high/intermediate- or high-risk IPI exhibited a CR rate of only 19.7%, whereas the rest of the patients exhibited a CR rate of 63.6%. According to GPS scoring, the CR rates of scores 0, 1, and 2 were 67.9%, 37.2%, and 28.5%, respectively (P = 0.0133).

Prognostic factors

The median follow-up duration of the 225 patients was 20 months (1−86 months). Among the 196 T/NKCL patients, the estimated 5–year OS rate was 37.2%, with median OS and PFS of 43 months and 32 months, respectively. No significant difference was observed among histologic subtypes (Fig. 1). Patients who suffered from BM infiltration showed lower OS and PFS rates than those who did not (5– year OS: 26.6% vs. 41.8%, P = 0.0016; 5 year PFS: 16.0% vs. 31.6%, P = 0.0002), with the median OS being 24 months versus 55 months and median PFS being 19 months versus 40 months, respectively. Compared with the T/NKCL patients, T-LPD patients exhibited a longer survival time, with an estimated 5 year OS rate of 96.6% (Fig. 2).

Considering the aggressive behavior of T/NKCL, we further analyzed the prognostic factors of the T/NKCL patients. By univariate analysis, poor PFS and OS rates were correlated with poor performance status (P<0.0010), advanced Ann Arbor stage (P = 0.0172), presence of B symptoms (P = 0.0276), multiple extranodal involvement (P<0.0010), BM involvement (P = 0.0166), elevated lactic dehydrogenase LDH (P = 0.0238), decreased albumin (P = 0.0344), abnormal white blood count (P = 0.0211), decreased platelet counts (P = 0.0037), IPI (P = 0.0021), and GPS (P<0.0010).

All the significant parameters of the univariate analysis, except for IPI and GPS, were inputted into the Cox regression model for multivariate analysis. The results showed that poor performance status (P = 0.0330; relative risk, 1.797; 95% confidence interval [CI] = 1.047−3.084), advanced Ann Arbor stage (P = 0.0030; relative risk, 2.518; 95% CI= 1.369−4.631), elevated LDH (P = 0.0180; relative risk, 1.905; 95% CI= 1.116−3.251), and BM involvement (P = 0.0250; relative risk, 1.808; 95% CI= 1.075−3.039) were independent prognostic factors of OS in 196 patients with T/NKCL.

With regard to IPI, the estimated 5– year OS rates for the low, low/intermediate, high/intermediate, and high-risk groups were 61.7%, 41.3%, 28.7%, and 0% (Fig. 3A). Accordingly, the estimated 5 year OS rates of the GPS scores 0, 1, and 2 were 55.2%, 31.1%, and 15.3%, respectively (Fig. 3B). By further comparing the efficiency of the two evaluation systems using the Kaplan-Meier method, IPI represented a less favorable prognostic values in both T/NKCL patients with BM infiltration and without BM infiltration (Log-rank= 10.12 and 8.078, P = 0.0176 and P = 0.0444, respectively, Fig. 3C and 3E) than that of GPS (Log-rank= 9.292 and 30.48, P = 0.033 and P<0.0001, respectively, Fig. 3D and 3F).

Discussion

In our study, the general condition of T-LPD patients was significantly better than that of the T/NKCL patients. Compared with the T/NKCL patients, the T-LPD patients tended to exhibit better performance statuses, fewer incidences of B symptoms, and normal levels of serum β-2 microglobulin. Meanwhile, T/NKCL patients were more likely to express CD7, but did not express CD3 and CD5. These clinical and immunophenotypical characteristics may be helpful in differentiating between T-LPD and T/NKCL. No significant difference was observed in the incidence of viral infection between T/NKCL and T-LPD, which is consistent with previous reports [ 22− 26]. Viral infection occurs in both groups of diseases, indicating that individual genetic susceptibility may play an important role in the development of abnormal clonal proliferation of T cells in the BM of patients with T-cell lymphoid malignancies [ 27]. Outcomes of our patients were generally in parallel with those in other regions [ 28], and T/NKCL patients presented relatively lower CR, ORR, and inferior survival, as compared with T-LPD.

T/NKCL is an aggressive disease that commonly presents with BM infiltration. Our study showed that the patients with BM infiltration were likely to exhibit poor performance status, advanced disease stage, multiple extranodal involvement, abnormal white blood cell counts, decreased hemoglobin levels, decreased platelet counts, elevated serum β-2 microglobulin, and elevated CRP. Accordingly, poorer performance status, higher IPI score, and worse outcomes were also previously reported in T/NKCL patients with BM infiltration [ 29]. Therefore, patients with BM infiltration showed unfavorable clinical characteristics, adverse disease outcome, and short survival time compared with those without BM infiltration, which agrees with previous reports that suggested the influence of BM infiltration on T/NKCL prognosis [ 30].

Prognostic scoring systems based on clinical and biological features considerably aid risk stratification and treatment determination. In our study, the IPI scoring system could not discriminate the differences in survival rate between risk groups in patients with or without BM infiltration, thereby indicating that a more effective risk-evaluation system should be used. A previous study on the prognostic significance of GPS in 164 extranodal NKTCL showed that GPS might be superior to IPI in discriminating patients with different outcomes in low-risk groups. Inflammatory response plays an important role in the progression and survival of patients with T/NKCL [ 15]. In our study, GPS successfully predicted the survival rate for T/NKCL subgroups and distinguished between the survival rates of T/NKCL patients with or without BM infiltration.

Collectively, our cohort of 225 patients with mature T-cell lymphoid malignances showed that extranodal involvement, β-2 microglobulin and CRP levels, abnormal platelet counts, and immunophenotypes can be used to distinguish T-LPD from T/NKCL. T/NKCL patients with BM infiltration showed less response to treatment and worse disease outcome. Poor performance status, advanced stage, elevated LDH, and BM involvement may be considered as independent prognostic factors in T/NKCL. GPS may be superior to IPI in predicting disease prognosis in patients with T/NKCL.

References

[1] Vose J, Armitage J, Weisenburger D; International T-Cell Lymphoma Project. International peripheral T-cell and natural killer/T-cell lymphoma study: pathology findings and clinical outcomes. J Clin Oncol 2008; 26(25): 4124–4130

[2] Zhao WL. Targeted therapy in T-cell malignancies: dysregulation of the cellular signaling pathways. Leukemia 2010; 24(1): 13–21

[3] Tang T, Tay K, Quek R, Tao M, Tan SY, Tan L, Lim ST. Peripheral T-cell lymphoma: review and updates of current management strategies. Adv Hematol 2010; 2010: 624040

[4] Lima M, Teixeira MA, Queirós ML, Santos AH, Gonçalves C, Correia J, Farinha F, Mendonça F, Soares JM, Almeida J, Orfão A, Justiça B. Immunophenotype and TCR-Vβ repertoire of peripheral blood T-cells in acute infectious mononucleosis. Blood Cells Mol Dis 2003; 30(1): 1–12

[5] Kroft SH. Monoclones, monotypes, and neoplasia pitfalls in lymphoma diagnosis. Am J Clin Pathol 2004; 121(4): 457–459

[6] Gallamini A, Stelitano C, Calvi R, Bellei M, Mattei D, Vitolo U, Morabito F, Martelli M, Brusamolino E, Iannitto E, Zaja F, Cortelazzo S, Rigacci L, Devizzi L, Todeschini G, Santini G, Brugiatelli M, Federico M; Intergruppo Italiano Linfomi. Peripheral T-cell lymphoma unspecified (PTCL-U): a new prognostic model from a retrospective multicentric clinical study. Blood 2004; 103(7): 2474–2479

[7] Weisenburger DD, Savage KJ, Harris NL, Gascoyne RD, Jaffe ES, MacLennan KA, Rüdiger T, Pileri S, Nakamura S, Nathwani B, Campo E, Berger F, Coiffier B, Kim WS, Holte H, Federico M, Au WY, Tobinai K, Armitage JO, Vose JM; International Peripheral T-cell Lymphoma Project. Peripheral T-cell lymphoma, not otherwise specified: a report of 340 cases from the International Peripheral T-cell Lymphoma Project. Blood 2011; 117(12): 3402–3408

[8] Vose JM. International Peripheral T-Cell Lymphoma (PTCL) clinical and pathologic review project: poor outcome by prognostic indices and lack of efficacy with anthracyclines. Blood (ASH Annual Meeting Abstracts) 2005; 106: Abstract 811

[9] Went P, Agostinelli C, Gallamini A, Piccaluga PP, Ascani S, Sabattini E, Bacci F, Falini B, Motta T, Paulli M, Artusi T, Piccioli M, Zinzani PL, Pileri SA. Marker expression in peripheral T-cell lymphoma: a proposed clinical-pathologic prognostic score. J Clin Oncol 2006; 24(16): 2472–2479

[10] Li YJ, Li ZM, Xia Y, Huang JJ, Huang HQ, Xia ZJ, Lin TY, Li S, Cai XY, Wu-Xiao ZJ, Jiang WQ. Serum C-reactive protein (CRP) as a simple and independent prognostic factor in extranodal natural killer/T-cell lymphoma, nasal type. PLoS ONE 2013; 8(5): e64158

[11] Miyazaki T, Yamasaki N, Tsuchiya T, Matsumoto K, Kunizaki M, Taniguchi D, Nagayasu T. Inflammation-based scoring is a useful prognostic predictor of pulmonary resection for elderly patients with clinical stage I non-small-cell lung cancer. Eur J Cardiothorac Surg 2015; 47(4): e140–e145

[12] Ishizuka M, Oyama Y, Abe A, Tago K, Tanaka G, Kubota K. Clinical significance of an inflammation-based prognostic system for gastric cancer patients with a preoperative normal serum level of carcinoembryonic antigen. Anticancer Res 2014; 34(12): 7219–7226

[13] Li X, Zhang Y, Zhao W, Liu Z, Shen Y, Li J, Shen Z. The Glasgow Prognostic Score as a significant predictor of diffuse large B cell lymphoma treated with R-CHOP in China. Ann Hematol 2015; 94(1): 57–63

[14] Kim Y, Kim SJ, Hwang D, Jang J, Hyun SY, Kim YR, Kim JS, Min YH, Cheong JW. The modified Glasgow Prognostic Scores as a predictor in diffuse large B cell lymphoma treated with R-CHOP regimen. Yonsei Med J 2014; 55(6): 1568–1575

[15] Li YJ, Jiang WQ, Huang JJ, Xia ZJ, Huang HQ, Li ZM. The Glasgow Prognostic Score (GPS) as a novel and significant predictor of extranodal natural killer/T-cell lymphoma, nasal type. Am J Hematol 2013; 88(5): 394–399

[16] Vardiman JW, Thiele J, Arber DA, Brunning RD, Borowitz MJ, Porwit A, Harris NL, Le Beau MM, Hellström-Lindberg E, Tefferi A, Bloomfield CD. The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood 2009; 114(5): 937–951

[17] Wang L, Wu-Xiao ZJ, Chen XQ, Zhang YJ, Lu Y, Xia ZJ. Hepatitis B virus infection correlates with poor prognosis of extranodal natural killer/T cell lymphoma. Leuk Lymphoma 2015; 56(4): 936–941

[18] Dupuis J, Emile JF, Mounier N, Gisselbrecht C, Martin-Garcia N, Petrella T, Bouabdallah R, Berger F, Delmer A, Coiffier B, Reyes F, Gaulard P; Groupe d’Etude des Lymphomes de l’Adulte (GELA). Prognostic significance of Epstein-Barr virus in nodal peripheral T-cell lymphoma, unspecified: A Groupe d’Etude des Lymphomes de l’Adulte (GELA) study. Blood 2006; 108(13): 4163–4169

[19] Eminger LA, Hall LD, Hesterman KS, Heymann WR. Epstein-Barr virus: dermatologic associations and implications: part II. Associated lymphoproliferative disorders and solid tumors. J Am Acad Dermatol 2015; 72(1): 21–34, quiz 35−36

[20] de Tute RM. Flow cytometry and its use in the diagnosis and management of mature lymphoid malignancies. Histopathology 2011; 58(1): 90–105

[21] Poopak B, Valeshabad AK, Elahi F, Rezvani H, Khosravipour G, Jahangirpour MA, Bolouri S, Golkar T, Salari F, Shahjahani M, Saki N. PCR analysis of IgH and TCR-γ gene rearrangements as a confirmatory diagnostic tool for lymphoproliferative disorders. Indian J Hematol Blood Transfus 2015; 31(1): 38–45

[22] Huang W, Lv N, Ying J, Qiu T, Feng X. Clinicopathological characteristics of four cases of EBV positive T-cell lymphoproliferative disorders of childhood in China. Int J Clin Exp Pathol 2014; 7(8): 4991–4999

[23] Kimura H, Ito Y, Kawabe S, Gotoh K, Takahashi Y, Kojima S, Naoe T, Esaki S, Kikuta A, Sawada A, Kawa K, Ohshima K, Nakamura S. EBV-associated T/NK-cell lymphoproliferative diseases in nonimmunocompromised hosts: prospective analysis of 108 cases. Blood 2012; 119(3): 673–686

[24] Dearden C. Large granular lymphocytic leukaemia pathogenesis and management. Br J Haematol 2011; 152(3): 273–283

[25] Nakamichi N, Wada N, Kohara M, Fukuhara S, Sugiyama H, Ogawa H, Hino M, Kanamaru A, Kanakura Y, Morii E, Aozasa K. Polymorphous lymphoproliferative disorder: a clinicopathological analysis. Virchows Arch 2010; 456(3): 269–276

[26] Miquel J, Fraitag S, Hamel-Teillac D, Molina T, Brousse N, de Prost Y, Bodemer C. Lymphomatoid papulosis in children: a series of 25 cases. Br J Dermatol 2014; 171(5): 1138–1146

[27] Wang XY, Cen XN, Ren HY. Investigation of lymphoma patients’ EBV infection status. Chinese J ExpHematol (Zhongguo Shi Yan Xue Ye Xue Za Zhi) 2014; 22(6): 1584–1590in Chinese)

[28] Desai M, Liu S, Parker S. Clinical characteristics, prognostic factors, and survival of 393 patients with mycosis fungoides and Sézary syndrome in the southeastern United States: a single-institution cohort. J Am Acad Dermatol 2015; 72(2): 276–285

[29] Boi M, Zucca E, Inghirami G, Bertoni F. Advances in understanding the pathogenesis of systemic anaplastic large cell lymphomas. Br J Haematol 2015; 168(6): 771–783

[30] Wang YQ, Yang Y, Zhuo HY, Zou LQ, Jiang Y, Jiang M. Trial of LVDP regimen (L-asparaginase, etoposide, dexamethasone, and cisplatin, followed by radiotherapy) as first-line treatment for newly diagnosed, stage III/IV extranodal natural killer/T cell lymphoma. Med Oncol 2015; 32(2): 435

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg