Complete remission after first-line radio-chemotherapy as predictor of survival in extranodal NK/T cell lymphoma

Our study included 36 patients with a median age of 49 years (range: 22 to 80), comprising 24 males and 11 females (ratio: 2.18:1). The majority of patients (75 %) were younger than 60 years. Among the 36 patients, 29 (80 %) were classified as upper aerodigestive tract NK/T cell lymphoma (UNKTL) versus 7 (20 %) patients as extra-upper aerodigestive tract NK/T cell lymphoma (EUNKTL). In addition, 72 % of UNKTL patients had local invasiveness greater than T3 with TNM classification. According to the Ann Arbor staging system, 10 (28 %) patients were categorized as stage I, six (17 %) as stage II, and 20 (55 %) as stage IV. All EUNKTL patients (n = 7) were considered to be stage IV. Furthermore, 75 % of patients had a good performance status (0–1). The most frequently involved sites were the nose (69 %), paranasal sinus (58 %), palate-pharyngeal (25 %), and bones (22 %). Regional lymphadenopathies were involved in 44 % of cases, bone marrow and central nervous system in 22 % and 11 %, respectively. The staging was based on computer tomography and MRI for most of the patients. Positron emission tomography/computed tomography (PET/SCAN) was not used to stage or follow evolution of the disease exept in recent diagnosis of T/NK lymphoma. The most frequent symptoms at diagnosis were obstructive in nature (purulent rhinorrhea, nasal obstruction, sinusitis, and dysphagia) in addition to epistaxis and cervical lymphadenopathy. Hematophagocytosis was observed in three patients. B symptoms were present in 39 % of patients. Overall, 14 (39 %) patients presented lactate dehydrogenase levels above the normal limit, 18 (50 %) high CRP levels, and 21 (58 %) increased beta2-microglobulin levels. Lymphopenia was found in 26 (61 %) patients, anemia in 13 (36 %), and low serum albumin level in 12 (33 %). According to IPI scoring, 23 (64 %) patients were classified as low-risk (0–2) and 13 (36 %) as high-risk. As a bone marrow biopsy was not performed in one patient, only 35 patients were assessed using PIT scores, revealing 24 (68 %) patients to be low-risk versus 11 (32 %) high-risk. According to Korean Prognostic Index (KPI) scoring, 19 (53 %) patients were considered low-risk compared to 17 (47 %) high-risk. Patient characteristics are listed in Table 1.

The different first-line CT regimens are presented in Table 2. CT alone was the primary treatment for advanced-stage disease (84 % for CT vs. 10 % for CT + RT), with 82 % of patients receiving anthracycline-based regimens. For the 35 patients receiving CT, forty different protocols were used: anthracycline based regimen, high dose aracytine plus high dose methotrexate, high dose methotrexate plus L-asparaginase, cisplatin based regimen. For early-stage disease, an equal number of patients received CT + RT or CT alone (50 %-50 %). All seven patients with EUNKTL were administered CT alone. For patients with UNKTL, CT alone was given to 17 (61 %) patients compared to 10 (36 %) receiving the combination of CT + RT. Patients were treated with a median of 2 CT lines (range: 1 to 4). One patient was lost of view after 6 cycles of chemotherapy without evaluation of the disease response.

Ten patients received CT + RT. The RT was given with a median dose of 40 gray (40 to 46). Only one patient received RT alone (30 gray) plus corticosteroids because of advaned age and comorbidities.

For UNKTL patients, CR rates were only 50 % for CT alone versus 90 % for RT + CT.. However, UNKTL patients with CT + RT have more favorable prognostic parameters than CT alone patients: 20 % stage IV versus 61 %, 10 % of elevated LDH versus 44 %; but the same percentage of local invasiveness and regional lymphadenopathy involved (70 % vs 67 % and 40 % vs 44 %, respectively). One patient received RT alone, but experienced disease progression on this regimen.

Among the low stage group (stage I-II), the distribution between the two treatment modalities is balanced confirming the better outcome with the combination CT + RT. However in the stage IV patients, the difference in the distribution between CT and CT + RT could create a significant bias in favor of the CT + RT group.

For EUNKTL patients, all were treated with anthracycline-based regimens, resulting in a CR rate of 43 %. Responses to the different modalities are provided in Table 3.

The 5-year OS and PFS rates were 39 % and 33 %, respectively (Figure 1A and Figure 1B). Eighty-two percent of patients died from disease progression, one from abdominal infection, and one from secondary acute leukemia 6 years after autologous transplantation. In univariate analysis, male sex, B symptoms, disease stage, lactate dehydrogenase level, prognostic indexes (IPI, PIT, and KPI), and quality of response (CR versus no response (or PD)) were found to be significantly associated with OS and PFS. Extensive disease stage, local invasiveness, high lactate dehydrogenase levels, and high-risk scores according to the IPI, PIT, and KPI classifications were related to poor survival. Quality of response was significantly associated with survival, with 5-year OS rates being 80 % for patients in CR versus 0 % for those with progressive disease (p < 0.01). When we focused on the UNKTL population, the best results were obtained when treatment consisted of CT + RT, with a statistically significant difference in terms of 5-year OS and PFS. The 5-year OS rate was 75 % for patients receiving CT + RT compared with 35 % for those receiving CT alone (p = 0.041) (Figure 2A). Accordingly, PFS was significantly higher in patients treated with RT + CT versus CT alone (p = 0.0063) (Figure 2B). Lymphopenia, low serum albumin level, high CRP level, and the classification into UNKTL or EUNKTL were statistically linked to OS or PFS.

Between January 1989 and September 2010, 36 patients were recruited in seven hematological centers in France. All the patients in this series been entirely of European descent (no patients were of Asia or south America descent). Patients were required to fulfill the WHO criteria for clinicopathologic diagnosis and classification of NK/T-cell lymphoma [5]. All biopsied tissues were critically reviewed. The following data was collected: gender, age, clinical characteristics (performance status and symptoms), biological parameters (lactate dehydrogenase, C-reactive protein [CRP], and serum albumin), radiological presentation, and histopathological reports describing angiocentricity, angioinvasion, necrosis zones, and polymorphism of individual cells. Immunohistochemical studies had to be positive for NK/T-cell markers, including CD2, CD3, or CD56 cytotoxic molecule (TIA-1 or granzyme B), and for EBV-encoded small RNA, and negative for B-cell markers, such as CD20 or CD79. Patients with blastic NK-cell lymphoma/leukemia, aggressive NK-cell lymphoma/leukemia, and unspecified peripheral T-cell lymphoma were excluded from the analysis. Scores for the International Prognostic Index [51] (IPI), Korean Prognostic Index (KPI) [9], and Prognostic Index for PTCL/NOS [52] (PIT) were calculated for all patients. We separately analyzed the survival of patients with upper aerodigestive tract NK/T-cell lymphoma (UNKTL) and extra-upper aerodigestive tract NK/T-cell lymphoma (EUNKTL) [1, 27]. UNKTL included all lymphomas confined to the nasal cavity, nasopharynx, and upper aerodigestive tract, whereas lymphomas at all other sites were considered to be EUNKTL. Patients with primary lesions within the nasal cavity and secondary lesions in other organs were classified as UNKTL. Local invasiveness was defined in accordance with the 2002 TNM classification of the American Joint Committee on Cancer [9].

CT schedules, RT dosages, and chronological sequence of treatments were analyzed for each patient. In terms of CT, treatments were based on anthracycline, aracytine or cisplatin regimens (doxorubicin, cyclophosphamide, vindesine, bleomycin, and prednisone [ACVBP]) for 17 (49 %) patients; cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) for 9 (25 %) patients; cyclophosphamide, vincristine, prednisone, doxorubicin, and methotrexate (COPADM) followed by cyclophosphamide, cytosine arabinoside, and etoposide (CYVE) for four (11 %) patients; etoposide, methylprednisolone, cytosine arabinoside, and cisplatin (ESHAP) and dexamethasone (DHAP) for 3 (8.3 %) patients; high-dose methotrexate plus L-asparaginase for 2 (6 %) patients. Autologous bone marrow transplantation was performed as first-line consolidation on one patient. The different treatment schedules are listed in Table 2. Studied outcomes were treatment response, progression-free survival (PFS), and OS, with treatment response being evaluated at the end of treatment according to the standardized response criteria of Cheson et al.[53].

Regarding RT treatment, the median dose was 40 grays (Gy). All patients received RT from a linear accelerator with 4 megavolt (MV), 6 MV or 10 MV photons to achieve dose homogeneity. Generally, the planning target volume included all macroscopic lesions, the paranasalsinuses, the nasopharynx, the upper gum, and the palate with adequate margins. Regardless of primary tumor localization, elective cervical lymph node irradiation was not delivered unless the neck was involved clinically. The most common field arrangement was two lateral opposing photons fields with supplementation between the medial canthus by appropriate energy of electron.

OS and PFS were estimated using the Kaplan-Meier product-limit method. OS was measured from the date of diagnosis to death or last follow-up visit. Unadjusted Cox proportional hazards models were employed to make group comparisons for baseline and treatment characteristics and Kaplan –Meier curves were used to quantify the percentage of patients who were free of recurrence or those who stay alive over time. A p-value <0.05 was considered statistically significant, with two-sided significance tests used for all p-values.