Abstract

Background

Clinical features and outcomes of extranodal natural killer/T-cell lymphoma (ENKL) arising from extranasal sites are not fully understood. The purpose of this study was to study the prognosis and treatment outcome of skin/soft tissue primary ENKL.

Patients and methods

This multicenter retrospective study included 48 patients with skin/soft tissue primary ENKL diagnosed from 1993 to 2010.

Results

Patients with Ann Arbor stage I, T1–2N0M0 by International Society for Cutaneous Lymphomas–European Organization of Research and Treatment of Cancer TNM (tumour–node–metastasis) stage, International prognostic index score of 0–1, and a Korean prognostic index (KPI) score of 0–1 were associated with better survival. Four of five patients with T1–2N0M0 disease achieved complete response with radiation alone. In disseminated disease, only 6 of 13 patients responded to anthracycline-containing chemotherapy, and all the two patients receiving SMILE showed response.

Conclusion

In conclusion, we identified the prognostic value of KPI, and we suggest a treatment recommendation according to the TNM (tumour–node–metastasis) stage. Radiotherapy with/without chemotherapy seemed to be optimal in localized disease. In advanced stages, a more aggressive treatment regimen with newer agents should be sought.

introduction

According to the 2008 World Health Organization (WHO) classification of hematopoietic and lymphoid tumors [1], extranodal natural killer/T-cell lymphoma (ENKL) is a distinctive clinicopathological disease entity with aggressive clinical feature and is characterized by its strong association with Epstein-Barr virus.

Around 60%–90% of cases of ENKL originate from the aerodigestive tract including the nasal cavity, oropharynx, oral cavity, and tonsils. Extranasal ENKL commonly occurs in skin/soft tissues and the gastrointestinal tract, followed in decreasing order of frequency by lung, adrenal glands, testis, and central nervous system [2–4]. Compared with nasal ENKL, extranasal ENKL exhibits more aggressive clinical features in terms of advanced stage, higher lactate dehydrogenase (LDH) level, higher International prognostic index (IPI), poorer performance status, inferior response to chemotherapy, and worse survival [2, 4]. However, previous studies combined all extranasal ENKL cases without considering the primary site. Even though skin and soft tissue comprise the most common primary site, the clinical features and treatment outcome are not well known. The purpose of the present study was to clarify the characteristic clinical features, prognostic factors, and treatment outcome of primary skin/soft tissue ENKL.

patients and methods

Forty-eight patients who were diagnosed with ENKL of the skin or soft tissue as the primary site were included in the analysis. All cases were confirmed histologically as ENKL according to the WHO classification. Patients with involvement of the nasal area were excluded. The following clinical data at the time of diagnosis were collected retrospectively: patient demographics, extent of disease involvement, LDH level, Ann Arbor stage, IPI, bone marrow findings, presence of B symptoms, performance status, date of diagnosis, type of treatment, treatment response, date of last follow-up, vital status, and cause of death. The International Society for Cutaneous Lymphomas and the Cutaneous Lymphoma Task Force of the European Organization of Research and Treatment of Cancer (ISCL-EORTC) TNM (tumour–node–metastasis) stage for cutaneous lymphoma other than mycosis fungoides and Sezary syndrome (MF/SS) [5] were applied to the collected data and were used to evaluate each case and to categorize it into stage I–IV. We defined localized cutaneous disease (T1–2N0M0) as stage I, localized disease with regional lymph node (LN) involvement (T1–2N1M0) as stage II, disseminated disease without visceral organ involvement (T3 NX M0 or TX N2-3M0) as stage III, and disease with visceral organ involvement (TX NX M1) as stage IV. The Korean prognostic index (KPI) was calculated using the TNM stage instead of the Ann Arbor stage.

Treatment was categorized according to the initial treatment modality: anthracycline-based chemotherapy, etoposide-based chemotherapy with or without radiation including electron beam therapy, CVP (cyclophosphamide, vincristine, prednisone), SMILE (dexamethasone, methotrexate, ifosfaimide, L-asparaginase, etoposide), and radiation including electron beam therapy alone. The anthracycline-based chemotherapy regimens used were CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone), CHOP with bortezomib [6], LSG9 [7], and mLSG4 [7]. The etoposide-based chemotherapy regimens were ESHAP (etoposide, methylprednisolone, cisplatin, cytarabine) [8, 9] and IMEP (ifosfamide, methotrexate, etoposide, prednisone) [10].

Overall survival (OS) was measured from the date of diagnosis to the date of death or the last follow-up visit. Survival was estimated using the Kaplan–Meier method and compared using log-rank analysis. Multivariate survival analysis using Cox's proportional hazard model was carried out to identify the prognostic factors for OS. The following variables were included in the final model: LDH level (categorical variable using the upper normal limit as the cut-off), presence of B symptoms, disease extent (categorical variable as TNM stage I–II versus stage III–IV), regional LN involvement, age >60 years, and Eastern Cooperative Oncology Group performance status (ECOG PS) two or more. P values <0.05 were considered significant, and all P values correspond to two-sided tests. Of the 48 patients, 44 patients had complete information and were included in the prognostication.

results

clinical characteristics of skin/soft tissue primary NK/T-cell lymphoma

Among all 48 patients, 37 patients (77%) had localized skin disease involving fewer than two contiguous sites of skin and 11 patients (23%) had disseminated skin involvement. The common presenting cutaneous symptoms were subcutaneous nodules with erythematous skin rash, erythematous papules, and ulcerative lesions. Seventeen of 48 (35%) patients had regional LN involvement and 15 (31%) had visceral organ involvement. The median age was 57 years. About half of the patients were male (52%) and had Ann Arbor stage I/II (50%). Most of the patients (90%) had a good PS of ECOG 0–1. B symptoms were present in 31%. Twenty-one of 48 (44%) patients had TNM stage I (T1–2N0M0) localized disease without regional LN involvement. Patients with primary skin/soft tissue ENKL showed a trend toward more advanced Ann Arbor staged diseases than those with nasal NK/T-cell lymphoma, but the other clinical characteristics were similar (Table 1).

Table 1.

Patients' baseline characteristics and comparison of patients' characteristics according to primary sites

 Present study, skin/soft tissue % (No. of patients) N = 48 Au et al. [2], nasal %, N = 92 Lee et al. [4], UNKTL %, N = 70 Kim et al. [3], UNKTL %, N = 221 Suzuki et al. [11], nasal %, N = 123 
Age      
 Mean, years (range) 57 (14–89) 52 (21–89) 46 (21–68) — 52 (14–89) 
Male 52% (25) 64% 61% — 66% 
Ann Arbor stage III/IV 50% (24) 27% 21% 15.4% 32% 
TNM stage      
 I (T1–2N0M0) 44% (21)     
 II (T1–2N1M0) 8% (4)     
 III (T3 or N2-3M0) 17% (8)     
 IV (TX NX M1) 31% (15)     
No. of extranodal sites ≥2 33% (16) 16% — 16.7% — 
B symptom 31% (15) 39% 26% 38% 46% 
ECOG PS ≥2 10% (5)  3% 17.6% 20% 
Elevated LDH 56% (27) 45% 41% 38% — 
BM involvement 10% (5) 10% — — 7% 
IPI score      
 0–1 (low) 40% (19) — 67% — 47% 
 2 (low–intermediate) 17% (8) — 11% — 28% 
 3 (high–intermediate) 25% (12) — 14% — 11% 
 4–5 (high) 10% (5) — 7% — 15% 
KPI score      
 0 23% (11) — — — 30% 
 1 21% (10) — — — 24% 
 2 15% (7) — — — 24% 
 3–4 33% (16) — — — 22% 
5-year OS 23.5%  54% 41%  
Median OS, years 1.0 1.6 — —  
 Present study, skin/soft tissue % (No. of patients) N = 48 Au et al. [2], nasal %, N = 92 Lee et al. [4], UNKTL %, N = 70 Kim et al. [3], UNKTL %, N = 221 Suzuki et al. [11], nasal %, N = 123 
Age      
 Mean, years (range) 57 (14–89) 52 (21–89) 46 (21–68) — 52 (14–89) 
Male 52% (25) 64% 61% — 66% 
Ann Arbor stage III/IV 50% (24) 27% 21% 15.4% 32% 
TNM stage      
 I (T1–2N0M0) 44% (21)     
 II (T1–2N1M0) 8% (4)     
 III (T3 or N2-3M0) 17% (8)     
 IV (TX NX M1) 31% (15)     
No. of extranodal sites ≥2 33% (16) 16% — 16.7% — 
B symptom 31% (15) 39% 26% 38% 46% 
ECOG PS ≥2 10% (5)  3% 17.6% 20% 
Elevated LDH 56% (27) 45% 41% 38% — 
BM involvement 10% (5) 10% — — 7% 
IPI score      
 0–1 (low) 40% (19) — 67% — 47% 
 2 (low–intermediate) 17% (8) — 11% — 28% 
 3 (high–intermediate) 25% (12) — 14% — 11% 
 4–5 (high) 10% (5) — 7% — 15% 
KPI score      
 0 23% (11) — — — 30% 
 1 21% (10) — — — 24% 
 2 15% (7) — — — 24% 
 3–4 33% (16) — — — 22% 
5-year OS 23.5%  54% 41%  
Median OS, years 1.0 1.6 — —  

BM, bone marrow; ECOG PS, Eastern Cooperative Oncology Group performance status; IPI, international prognostic index; LDH, lactate dehydrogenase; KPI, Korean prognostic index; OS, overall survival; TNM, tumour–node–metastasis; UNKTL, upper aerodigestive tract NK/T-cell lymphoma.

treatment and response

The initial treatment of the 20 patients with TNM stage I localized disease comprised chemotherapy in 15 patients and radiation alone including electron beam therapy in 5 patients.

Complete remission (CR) was observed in four of the nine (44%) patients who received anthracycline-based chemotherapy and in two of the five (40%) patients who received etoposide-based chemotherapy. The treatment outcome for T1–2N0M0 disease is displayed in supplemental Figure S1A (available at Annals of Oncology online). Four of five patients attained CR with radiotherapy including electron beam therapy alone, and these four patients achieved long-term survival. Three patients who received radiotherapy after CR were attained with anthracycline-containing chemotherapy also achieved long-term survival. Overall, addition of radiation or radiation alone produced significantly better survival in patients with TNM stage I disease (median OS not reached versus 12 months, P = 0.015; Figure 1). Only four patients had TNM stage II diseases, and all were treated with anthracycline-based chemotherapy. Two of these patients, including one given additional radiotherapy, achieved CR (supplemental Figure S1B, available at Annals of Oncology online). This patient was the only one to have achieved a long-term survival (17 months), although this patient eventually died of disease relapse. In the 21 patients with TNM stage III–IV disease who received chemotherapy, only 4 of the 13 (31%) patients treated with anthracycline-containing chemotherapy and one of two patients receiving SMILE achieved CR (supplemental Figure S1C, available at Annals of Oncology online).

Figure 1.

OS according to radiotherapy/electron beam therapy in 20 localized TNM stage I (T1–2N0M0) patients (CTx, chemotherapy; IF-RT, involved-field radiotherapy; OS, overall survival).

Figure 1.

OS according to radiotherapy/electron beam therapy in 20 localized TNM stage I (T1–2N0M0) patients (CTx, chemotherapy; IF-RT, involved-field radiotherapy; OS, overall survival).

staging and prognostic factors for survival

The median OS of the 48 patients was 12 months (range 0–128 months). We first assessed the prognostic value of preexisting indexes such as the Ann Arbor stage, IPI, and KPI. Ann Arbor stage I, TNM stage I (T1–2N0M0), and IPI 0–1 were significantly associated with longer survival (supplemental Figure S2A–C, available at Annals of Oncology online). OS was significantly longer in those with KPI score 0–1 (median OS 43 versus 5 months in KPI score 0–1 versus 2–4, P < 0.001) (supplemental Figure S2D, available at Annals of Oncology online). In multivariate analysis, elevated LDH level (P = 0.022; relative risk 2.783; 95% confidence interval (CI) 1.155–6.703), presence of B symptoms (P = 0.005; relative risk 3.351; 95% CI 1.443–7.778), and TNM stage III–IV (P = 0.009, relative risk 3.692; 95% CI 1.379–9.883) were significantly associated with worse survival (Table 2).

Table 2.

Multivariate analysis of prognostic factors

 HR (95% CI) P value 
B symptom 3.4 (1.4–7.8) 0.005 
LDH elevation 2.8 (1.2–6.7) 0.022 
ECOG PS ≥2 2.0 (0.4–10.4) 0.434 
Age ≥60 1.1 (0.5–2.5) 0.771 
Regional LN involvement 0.4 (0.1–1.2) 0.108 
TNM stage III, IV 3.7 (1.4–9.9) 0.009 
 HR (95% CI) P value 
B symptom 3.4 (1.4–7.8) 0.005 
LDH elevation 2.8 (1.2–6.7) 0.022 
ECOG PS ≥2 2.0 (0.4–10.4) 0.434 
Age ≥60 1.1 (0.5–2.5) 0.771 
Regional LN involvement 0.4 (0.1–1.2) 0.108 
TNM stage III, IV 3.7 (1.4–9.9) 0.009 

CI, confidence interval; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, hazard ratio; LDH, lactate dehydrogenase; LN, lymph node; TNM, tumour–node–metastasis.

discussion

Extranasal ENKL has more adverse clinical features and worse survival than nasal ENKL, and a few studies focused on the primary sites. In the present study of skin/soft tissue primary ENKL, male predominance was not apparent (male 52%), but there was a tendency toward a more advanced stage (Ann Arbor stage III–IV in 50% and disseminated disease with TNM stage III–IV in 48%) than in the previously reported nasal ENKL (Table 1). The better survival for patients with localized disease is consistent with previous reports [11, 12], and one might speculate that skin lesions can be easily identified and enable an early diagnosis. Preexisting prognostic scores including the Ann Arbor stage and IPI did not discriminate survival well in each risk subgroups; however, IPI score of 0–1 and Ann Arbor stage I were significantly associated with better survival. This finding is similar to that of previous reports [2, 3, 13], highlighting the need for a more discriminating staging or prognostic system. Ann Arbor staging may not be appropriate for estimating disease extent in skin or soft tissue, placing high number of patients in the highest stage disproportionately or inappropriately [5]. The ISCL-EORTC proposed TNM classification system for primary cutaneous lymphomas other than MF/SS; however, it was based purely on the anatomical extent of disease and the prognostic relevance was unclear. Here, we first validated the TNM classification for the primary cutaneous ENKL and found that TNM stage I (T1–2N0M0) localized disease was an independent prognostic factor for better survival. However, survival in the remaining patients was not well discriminated using the TNM classification. Survival of stage II localized disease with regional LN involvement was poorer than expected, which might relate to the observation that three of these four patients had a high KPI score ≥2. This suggests that factors other than anatomical disease extent are important for estimating the prognosis. In addition to the presence of distant metastases, B symptoms and elevated LDH level were independent prognostic factors for both cutaneous ENKL and nasal ENKL. The KPI was also useful in predicting survival. The best prognostic group (KPI score of 0–1) had a median survival of 43.0 months, which was shorter than the best survival reported previously [14] in a group of patients with nasal ENKL with the same KPI score.

Chemotherapy regimens for ENKL do not differ according to the primary site. In our series, the CR rate was 44% in TNM stage I localized disease and 31% in TNM stage III–IV disseminated disease treated with an anthracycline-based regimen, suggesting that more aggressive treatment strategies are needed, especially for disseminated disease. SMILE was administered to only two patients and responses were observed in both patients (one CR, one partial response), which is similar to its phase I study result [15]. In nasal ENKL, anthracycline-based therapy has an unsatisfactory efficacy [16] and other nonanthracycline agents such as methotrexate, etoposide, and L-asparaginase are under investigation [10, 15, 17]. New treatment regimens for extranasal disease should also be sought considering the disappointing result of anthracycline-based treatment. A clinical trial of a more efficient regimen such as SMILE is needed [17]. Although 16 patients in the phase II study presented with cutaneous involvement, 12 showed nasal involvement. One of the four cutaneous ENKL patients attained CR, and two attained partial response with the SMILE regimen.

The role of radiotherapy has been increasingly recognized, especially for localized nasal ENKL patients. The importance of radiotherapy in treating localized nasal disease has been shown [2, 18, 19], but its role in treating extranasal diseases is unclear. Skin/soft tissue is a feasible area for radiotherapy when the lesion is localized. In our study, four of five patients with localized disease achieved a CR with radiation alone including electron beam therapy, suggesting a role of radiotherapy in treating localized disease. However, historically, there is a substantial systemic failure rate of 25% in radiotherapy in patients with TNM stage I–II disease [20], highlighting the need for multimodal treatment. A phase II study of concurrent chemoradiotherapy for localized nasal ENKL recently showed a promising efficacy with a CR rate of >70% [21, 22]. In the present study, in patients with TNM stage I localized disease, those who received radiotherapy had a significant longer survival compared with those treated with chemotherapy alone.

In conclusion, we identified prognostic factors in skin/soft tissue primary ENKL. The KPI is useful for predicting the prognosis in the rare entity of extranasal ENKL, and localized disease is associated with a better survival. Early recognition and diagnosis of the skin/soft tissue lesion are important, and both disease extent and prognostic factors should be considered when planning the treatment. The addition of radiotherapy seemed to have a role in treating TNM stage I localized disease and should be considered in this patient population. More aggressive regimens than conventional anthracycline-containing treatment should be pursued, especially in patients with a poor prognosis and with a high KPI score or disseminated disease.

funding

JW Pharmaceutical, Seoul, Korea (#PHO1116085).

disclosure

The authors have declared no conflicts of interest.

references

1
Swerdlow
SH
Campo
E
Harris
NL
, et al.  . 
World Health Organization classification of tumors of haematopoietic and lymphoid tissues
lyon:IARC
 , 
2008
2
Au
WY
Weisenburger
DD
Intragumtornchai
T
, et al.  . 
Clinical differences between nasal and extranasal natural killer/T-cell lymphoma: a study of 136 cases from the International Peripheral T-Cell Lymphoma Project
Blood
 , 
2009
, vol. 
113
 (pg. 
3931
-
3937
)
3
Kim
TM
Lee
SY
Jeon
YK
, et al.  . 
Clinical heterogeneity of extranodal NK/T-cell lymphoma, nasal type: a national survey of the Korean Cancer Study Group
Ann Oncol
 , 
2008
, vol. 
19
 (pg. 
1477
-
1484
)
4
Lee
J
Park
YH
Kim
WS
, et al.  . 
Extranodal nasal type NK/T-cell lymphoma: elucidating clinical prognostic factors for risk-based stratification of therapy
Eur J Cancer
 , 
2005
, vol. 
41
 (pg. 
1402
-
1408
)
5
Kim
YH
Willemze
R
Pimpinelli
N
, et al.  . 
TNM classification system for primary cutaneous lymphomas other than mycosis fungoides and Sezary syndrome: a proposal of the International Society for Cutaneous Lymphomas (ISCL) and the Cutaneous Lymphoma Task Force of the European Organization of Research and Treatment of Cancer (EORTC)
Blood
 , 
2007
, vol. 
110
 (pg. 
479
-
484
)
6
Lee
J
Suh
C
Kang
HJ
, et al.  . 
Phase I study of proteasome inhibitor bortezomib plus CHOP in patients with advanced, aggressive T-cell or NK/T-cell lymphoma
Ann Oncol
 , 
2008
, vol. 
19
 (pg. 
2079
-
2083
)
7
Kinoshita
T
Hotta
T
Tobinai
K
, et al.  . 
A randomized controlled trial investigating the survival benefit of dose-intensified multidrug combination chemotherapy (LSG9) for intermediate- or high-grade non-Hodgkin's lymphoma: Japan Clinical Oncology Group Study 9002
Int J Hematol
 , 
2004
, vol. 
80
 (pg. 
341
-
350
)
8
Ezzat
AA
Khalifa
F
Berry
J
, et al.  . 
E-SHAP: an effective treatment in selected patients with relapsed non-Hodgkin's lymphoma
Ann Oncol
 , 
1994
, vol. 
5
 (pg. 
453
-
456
)
9
Velasquez
WS
McLaughlin
P
Tucker
S
, et al.  . 
ESHAP—an effective chemotherapy regimen in refractory and relapsing lymphoma: a 4-year follow-up study
J Clin Oncol
 , 
1994
, vol. 
12
 (pg. 
1169
-
1176
)
10
Lee
KW
Yun
T
Kim
DW
, et al.  . 
First-line ifosfamide, methotrexate, etoposide and prednisolone chemotherapy +/− radiotherapy is active in stage I/II extranodal NK/T-cell lymphoma
Leuk Lymphoma
 , 
2006
, vol. 
47
 (pg. 
1274
-
1282
)
11
Suzuki
R
Suzumiya
J
Yamaguchi
M
, et al.  . 
Prognostic factors for mature natural killer (NK) cell neoplasms: aggressive NK cell leukemia and extranodal NK cell lymphoma, nasal type
Ann Oncol
 , 
2010
, vol. 
21
 (pg. 
1032
-
1040
)
12
Choi
YL
Park
JH
Namkung
JH
, et al.  . 
Extranodal NK/T-cell lymphoma with cutaneous involvement: ‘nasal’ vs. ‘nasal-type’ subgroups—a retrospective study of 18 patients
Br J Dermatol
 , 
2009
, vol. 
160
 (pg. 
333
-
337
)
13
Chim
CS
Ma
SY
Au
WY
, et al.  . 
Primary nasal natural killer cell lymphoma: long-term treatment outcome and relationship with the International Prognostic Index
Blood
 , 
2004
, vol. 
103
 (pg. 
216
-
221
)
14
Lee
J
Suh
C
Park
YH
, et al.  . 
Extranodal natural killer T-cell lymphoma, nasal-type: a prognostic model from a retrospective multicenter study
J Clin Oncol
 , 
2006
, vol. 
24
 (pg. 
612
-
618
)
15
Yamaguchi
M
Suzuki
R
Kwong
YL
, et al.  . 
Phase I study of dexamethasone, methotrexate, ifosfamide, L-asparaginase, and etoposide (SMILE) chemotherapy for advanced-stage, relapsed or refractory extranodal natural killer (NK)/T-cell lymphoma and leukemia
Cancer Sci
 , 
2008
, vol. 
99
 (pg. 
1016
-
1020
)
16
Kim
WS
Song
SY
Ahn
YC
, et al.  . 
CHOP followed by involved field radiation: is it optimal for localized nasal natural killer/T-cell lymphoma?
Ann Oncol
 , 
2001
, vol. 
12
 (pg. 
349
-
352
)
17
Yamaguchi
M
Kwong
YL
Kim
WS
, et al.  . 
Phase II study of SMILE chemotherapy for newly diagnosed stage IV, relapsed, or refractory extranodal natural killer (NK)/T-cell lymphoma, nasal type: the NK-Cell Tumor Study Group Study
J Clin Oncol
 , 
2011
, vol. 
29(33)
 (pg. 
4410
-
4416
)
18
Isobe
K
Uno
T
Tamaru
J
, et al.  . 
Extranodal natural killer/T-cell lymphoma, nasal type: the significance of radiotherapeutic parameters
Cancer
 , 
2006
, vol. 
106
 (pg. 
609
-
615
)
19
You
JY
Chi
KH
Yang
MH
, et al.  . 
Radiation therapy versus chemotherapy as initial treatment for localized nasal natural killer (NK)/T-cell lymphoma: a single institute survey in Taiwan
Ann Oncol
 , 
2004
, vol. 
15
 (pg. 
618
-
625
)
20
Kim
GE
Cho
JH
Yang
WI
, et al.  . 
Angiocentric lymphoma of the head and neck: patterns of systemic failure after radiation treatment
J Clin Oncol
 , 
2000
, vol. 
18
 (pg. 
54
-
63
)
21
Kim
SJ
Kim
K
Kim
BS
, et al.  . 
Phase II trial of concurrent radiation and weekly cisplatin followed by VIPD chemotherapy in newly diagnosed, stage IE to IIE, nasal, extranodal NK/T-Cell Lymphoma: Consortium for Improving Survival of Lymphoma study
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
6027
-
6032
)
22
Yamaguchi
M
Tobinai
K
Oguchi
M
, et al.  . 
Phase I/II study of concurrent chemoradiotherapy for localized nasal natural killer/T-cell lymphoma: Japan Clinical Oncology Group Study JCOG0211
J Clin Oncol
 , 
2009
, vol. 
27
 (pg. 
5594
-
5600
)