Purpose/Objective(s)Statistically, a disease is considered curable when the mortality rate of survivors returns to the same level as that of the general population. We aimed to assess the curability of extranodal nasal-type NK/T-cell lymphoma (ENKTCL) in the modern treatment era.Materials/MethodsThe data of 1995 patients treated between 2000 and 2016 and registered in China Lymphoma Collaborative Group (CLCG) database were analyzed. We estimated cure fractions, median survival times, and the time point of cure using a non-mixture cure model.ResultsThe relative survival curves attained plateau for the entire cohort and most subsets, indicating that the notion of cure was robust. The overall cure fraction was 72.1%. Median survival was 1.10 years in uncured patients. The cure time was 4.5 years, i.e., beyond this time, mortality in ENKTCL patients was equivalent to that in the general population. In multivariate analysis, cure probability was associated with stage, performance status, primary tumor invasion, and lactate dehydrogenase level. Elderly patients (>60 years) had cure fraction similar to that of younger patients. Five-year overall survival rate correlated well with the cure fraction across risk-stratified groups. Thus, statistical cure is possible in ENKTCL patients receiving modern treatment strategies.ConclusionOverall probability of cure is good, though it is affected by presence of risk factors. These findings provide the rationale for treatment and follow-up of patients with different cure probabilities.

Extranodal NK/T-cell lymphoma (ENKTCL) remains a high unmet clinical need for improving outcome. This trial aimed to explore the safety, pharmacokinetics (PK) and efficacy of mitoxantrone hydrochloride liposome (PLM60) plus pegaspargase in patients (pts) with ENKTCL. Adult pts with histologically confirmed treatment-naïve or relapsed/refractory ENKTCL were recruited. Phase I was 3+3 dose-escalation design with four dose levels of PLM60 (12, 16, 20 and 24 mg/m2) plus pegaspargase 2000 IU/m2 administered on day 1 of every cycle (21 days) for 4-6 cycles. Phase II was dose expansion at the recommended phase 2 dose (RP2D) in pts with treatment-naïve ENKTCL. The primary endpoints were safety and PK. The secondary endpoint was efficacy including complete response (CR) rate, objective response rate (ORR) as per Lugano 2014. At the cut-off data of February 15, 2022, 31 eligible pts were enrolled (phase I, n = 21 and phase II, n = 10). Phase I included 9 relapsed/refractory pts and 12 treatment-naïve pts. Two dose-limiting toxicities (grade-4 neutropenia in 20 mg/m2; grade-3 abdominal pain in 24 mg/m2) occurred. RP2D was PLM60 24 mg/m2 plus pegaspargase 2000 IU/m2. Treatment-related adverse events (TRAEs) of any grade occurred in all 31 pts, in which 27 (87.1%) were ≥ grade 3. The most common ≥ grade 3 TRAEs was neutropenia (77.4%), leucopenia (74.2%), anemia (54.8%), thrombocytopenia (45.2%), hypertriglyceridemia (22.6%), infectious pneumonia (16.1%), lymphocytopenia (16.1%), decreased fibrinogen level (16.1%), hypoglycemia (12.9%) and elevated bilirubin (12.9%). 31 pts were evaluable for response. CR rate and ORR were 61.3% (19/31, 95% CI 42.2%-78.2%) and 87.1% (27/31, 95% CI 70.2%-96.4%), respectively. Median PFS was not reached. Among 22 treatment-naïve pts (13 males) with a median age of 40.5 (range, 23-70), 9 pts (40.9%) had the presence of B symptoms and 6 pts (27.3%) were at the stage III or IV (Lugano classification). The CR rate and ORR of this cohort was 68.2% (15/22, 95% CI 45.1%-86.1%) and 90.9% (20/22, 95% CI 70.8%-98.9%). PLM60 plus pegaspargase had an encouraging efficacy especially in treatment-naïve ENKTCL pts with manageable safety profiles.

Purpose/Objective(s) This study investigated the failure pattern and prognosis of distant metastasis for early-stage extranodal nasal-type natural killer/T-cell lymphoma (ENKTCL). Materials/Methods 1619 patients with complete information on distant metastatic sites were identified. The distribution of distant metastatic sites was reported. The distant metastatic sites were categorized into 4 groups. The overall survival (OS) and subsequent OS of the 4 groups were compared. The effect of distant metastatic number and time on subsequent OS were evaluated. Results DM was the primary mode of failure in this cohort, with a 5-year cumulative of 26.2%. The most frequent distant metastatic site was skin and soft tissue (SST, 32.4%), followed by distant lymph node (LN, 31.1%), lung (17.3%), liver (14.6%), lymphoma-associated hemophagocytic lymphohistiocytosis (LAHS, 10.7%) and gastrointestinal system (10.4%). While central nervous system, spleen, urogenital system, adrenal glands, bone, bone marrow and pancreas involvements were less common. The distant metastatic sites were categorized in to 4 groups: distant LN, SST, other sites and LAHS. The 4 categories demonstrated distinct OS and subsequent OS, with decreased mortality for distant LN and SST compared with other sites and LAHS. Besides, solitary DM and late DM (> 7.5 months) were associated with favorable 3-year OS compared with multiple metastases (45.7% versus 8.7%, P < 0.001) and early DM (≤ 7.5 months; 25.8% versus 15.8%, P = 0.001). Conclusion The findings were helpful understanding the variation in biology of different sites of metastases in patients with early-stage ENKTCL and provided important information for future therapeutic decisions, metastatic surveillance and translational studies.

The recommended non-anthracycline (non-ANT) chemotherapy on treating Extranodal NK/T-cell lymphoma, nasal type (ENKL) is based on evidences of single-arm phase I/II studies and retrospective series with limited cases and short-term follow-ups. The current study was to investigate treatment benefit of non-ANT chemotherapy in comparison to anthracycline (ANT) chemotherapy in a large-scaled cohort. Patients consecutively diagnosed with ENKL and treated with chemotherapy with/without radiotherapy between 2000 and 2015 from 20 Chinese institutes were retrospectively analyzed. Association of short-term response and long-term survival with chemotherapy categories were evaluated. A total of 2560 cases were enrolled, 87% had stage I-II disease, and median age was 43. The proportion of non-ANT chemotherapy increased notably through 2005 to 2012. Demographic and disease characteristics showed great similarities across chemotherapy subgroups. Non-ANT chemotherapy associated with increased response comparing to ANT regimen (CR, 40 vs. 28%; PR, 43 vs.37 %, P<0.05). With a median follow-up of 4 years, overall survival (OS) and progression free survival (PFS) were significantly better favoring non-ANT chemotherapy in the entire cohort (5-year OS, 68.9% vs 57.5%, P < 0.001; 5-year PFS, 59.5% vs 44.5%, P < 0.001), localized disease (5-year OS, 73.3% vs 60.9%, P < 0.001; 5-year PFS, 64.0% vs 47.6%, P < 0.001), advanced disease (OS, 39.8% vs 29.9%, P = 0.013; PFS, 30.1% vs 18.8%, P = 0.003), and each risk subgroups, respectively. The survival benefit remained consistent after adjustments with multivariate analysis and propensity score matching analysis. Gemcitabine + L-Asparaginase combination showed promising treatment outcomes, especially for advanced disease. Non-ANT chemotherapy constituted the mainstay of chemotherapy in ENKL treatment. The application of non-ANT chemotherapy associated improved response and survival comparing to ANT chemotherapy.

Introduction: The recommended non-anthracycline (non-ANT) chemotherapy on treating Extranodal NK/T-cell lymphoma, nasal type (ENKL) is based on evidences of single-arm phase I/II studies and retrospective series with limited cases and short-term follow-ups. The current study was to investigate treatment benefit of non-ANT chemotherapy in comparison to anthracycline (ANT) chemotherapy in a large-scaled cohort. Methods: Patients consecutively diagnosed with ENKL and treated with chemotherapy with/without radiotherapy between 2000 and 2015 from 20 Chinese institutes were retrospectively analyzed. Association of short-term response and long-term survival with chemotherapy categories were evaluated. Results: 2560 cases were enrolled, 87% had stage I-II disease, and median age was 43. The proportion of non-ANT chemotherapy increased notably through 2005 to 2012. Demographic and disease characteristics showed great similarities across chemotherapy subgroups. Non-ANT chemotherapy associated with increased response comparing to ANT regimen (CR, 40 vs. 28%; PR, 43 vs.37 %, P<0.05). With a median follow-up of 4 years, overall survival (OS) and progression free survival (PFS) were significantly better favoring non-ANT chemotherapy in the entire cohort (5-year OS, 68.9% vs 57.5%, P < 0.001; 5-year PFS, 59.5% vs 44.5%, P < 0.001), localized disease (5-year OS, 73.3% vs 60.9%, P < 0.001; 5-year PFS, 64.0% vs 47.6%, P < 0.001), advanced disease (OS, 39.8% vs 29.9%, P = 0.013; PFS, 30.1% vs 18.8%, P = 0.003), and each risk subgroups, respectively. The survival benefit remained consistent after adjustments with multivariate analysis and propensity score matching analysis. Gemcitabine + L-Asparaginase combination showed promising treatment outcomes, especially for advanced disease. Keywords: anthracycline; L-asparaginase; T-cell lymphoma (TCL).

The value of advanced radiotherapy (RT) techniques in early-stage extranodal nasal-type NK/T-cell lymphoma (NKTCL) is not known, and it is unclear which patients are the most appropriate candidates for intensity-modulated radiation therapy (IMRT).To evaluate the survival benefit of IMRT compared with 3-dimension conformal RT (3D-CRT) in a large national cohort of patients with early-stage NKTCL. This observational study reviewed patients with early-stage NKTCL treated with high-dose RT (≥ 45 Gy) at 16 Chinese institutions between 2000 and 2015. Patients were stratified into one of four risk groups by summing their number of risk factors (age > 60 years; ECOG [Eastern Cooperative Oncology Group] score ≥ 2; stage II; elevated lactate dehydrogenase [LDH]; and primary tumor invasion [PTI]) to generate low- (0 factor), intermediate-low- (1), intermediate-high- (2), and high- (3-5) risk groups. Survival was compared using Kaplan-Meier analysis and log-rank tests, propensity score match (PSM) and multivariable Cox regression. IMRT has been widely adopted for early-stage NKTCL in the last ten years across China. Of the 1691 patients, 981 (58%) received IMRT and 710 (42%) received 3D-CRT. Unadjusted 5-year OS and PFS were 75.9% and 67.6% for IMRT compared with 68.9% (p = 0.004) and 58.2% (p < 0.001) for 3D-CRT. After PSM and multivariable analyses to account for confounding factors, IMRT remained significantly associated with improved OS and PFS. The OS and PFS benefits of IMRT persisted in patients treated with modern chemotherapy regimens. Compared with 3D-CRT, IMRT significantly improved OS and PFS for high-risk and intermediate-high-risk patients, but provided limited benefits for low-risk or intermediate-low-risk patients. In this real-world multicenter study, IMRT resulted in improved survival compared to 3D-CRT in early-stage NKTCL. We provide a risk-adapted survival benefit profile that reflects the magnitude of the survival benefit offered by IMRT, which can be used to select patients and make treatment decisions.

This study aimed to evaluate the role of extended involved-field intensity modulated radiation therapy (IMRT) for patients with early stage extranodal nasal-type NK/T-cell lymphoma (NKTCL) who received new regimens chemotherapy. Between 2007 and 2016, 165 patients with early stage NKTCL underwent definitive high-dose and extended involved-field IMRT with (n = 158, 95.8%) or without chemotherapy (n = 7, 4.2%). One hundred forty patients (84.8%) received radiation dose more than 50 Gy to the primary tumor, whereas only 25 patients (15.2%) received less than 50 Gy. The majority of patients (n = 157, 89.1%) were treated with L-asparaginase–based regimens CT，whereas only 11 (6.7%) patients with doxorubicin-based CHOP/CHOP-like regimens. One hundred and nine patients (66.1%) received more than four cycles of chemotherapy. The locoregional control (LRC), overall survival (OS), and progression-free survival (PFS) were calculated using the Kaplan-Meier method. The 5-year OS, LRC, and PFS of all patients were 74.2%, 84.4%, and 72.5%. Patients who received ≥ 50 Gy had significantly higher LRC than those with <50 Gy, with 5-year LRC of 91.8% and 39.7% (P<0.001). The 5-year OS for patients without any risk factors (age >60, Elevated LDH, ECOG ≥2, primary tumor invasion [PTI] and stage II disease, defined as low-risk group) were 94.2%, whereas it was only 68.1% (P = 0.002) for patients with any risk factors (high-risk group). For high-risk early stage group, patients who received more than 4 cycles of chemotherapy significantly improved outcomes. The 5-year OS and PFS rates were 71.3% and 70.4% for patients with ≥4 cycles chemotherapy, compared with 59.5% (P = 0.032) and 54.4% (P = 0.009) for those with <4 cycles chemotherapy respectively. In multivariate analysis, ECOG ≥2, PTI, and Ann Arbor stage II were associated with poor OS. ECOG ≥2 and PTI were associated with increased risk of locoregional recurrence; whereas ECOG ≥2, PTI, primary site outside nasal cavity were associated with increased risk of PFS. In the modern era of IMRT and L-asparaginase–based chemotherapy, High-dose and extended-involved field IMRT for patients with early stage NKTCL achieved favorable outcomes. High-risk early stage patients who received more than 4 cycles chemotherapy had significantly improved OS and PFS.

Clinical prognostic factors for predicting outcomes and guiding treatments in extranodal NK/T-cell lymphoma, nasal type (NKTCL) are suboptimal. The current study sought to define the distribution features and prognostic role of primary tumor invasion (PTI) in a large patient cohort. A cohort of 1383 patients was recruited, including 947 (68.5%) stage I patients, 326 (23.6%) stage II patients, and 110 (8.0%) stage III-IV patients. Seven hundred and fifty-one (54.3%) patients presented with PTI. The presence of PTI was associated with disease aggressiveness, including a higher frequency of B symptoms, advanced-stage disease, regional lymph node involvement, lactate dehydrogenase elevation, a higher Eastern Cooperative Oncology Group performance status, and International prognostic index score. Overall survival (OS) was significantly or borderline higher in the PTI negative group than the PTI positive group in the entire cohort (5-year OS, 72.1% vs. 50.2%, p < 0.001) as well as in different Ann Arbor stage group (stage I-II, 75.9% vs. 53.0%, p < 0.001; stage III-IV, 20.3% vs. 16.8%, p = 0.065). PTI was associated with significantly higher locoregional failure (LRF) in early-stage patients, with a 5-year LRF of 15.1% in the PTI negative group and 28.1% in the PTI positive group (p< 0.001). PTI was associated with multiple adverse clinical features in NKTCL, and was an independent indicator for inferior prognosis.

The survival estimates made at diagnosis might lose accuracy over time. Conditional survival (CS) and annual hazard rate (anHR) dynamically provide more powerful prognostic information for long-term survivors. This study, to our knowledge, presents the first investigation to determine the CS and anHR in early-stage extranodal nasal-type NK/T-cell lymphoma (NKTCL) patients treated with radiation therapy (RT). We further analyzed if established prognostic factors and nomogram score at diagnosis remain relevant in survivorship over time. We identified 1179 early-stage NKTCL patients treated with curative radiotherapy from 10 institutions between 2000 and 2014. Most patients (75.1%) received RT and chemotherapy. The 5-year CS, defined as the probability of surviving an additional 5 years from a given year since treatment, was calculated using the Kaplan-Meier method. Prognostic factors at different time points were identified using multivariable Cox regression model (MVA). Based on the nomogram, we categorized patients into three groups: low-risk (score of 0, n=288) versus intermediate-risk (score of 1-70, n=643) versus high-risk (score of 71-187, n=248), and calculated the CS and anHR for each group. Given a 1-, 3- and 5-year survivorship, the 5-year CS increased from 67.6% at baseline to 76.8% (+9.2), 86.1% (+18.5%) and 91.4% (+23.8%), respectively. The anHR of mortality peaked at 15.2% per year in the first year, and decreased to <3% per year after 5 years of survivorship. At baseline, stage, age, Eastern Cooperative Oncology Group score, lactate dehydrogenase (LDH), and primary tumor invasion were independently associated with overall survival (OS) on MVA. However, after 5 years of survivorship, only stage (hazard ratio =2.28, P = .014) and LDH (hazard ratio =2.20, P = .023) were independently associated with OS. The 5-year CS increased from 85.8% to 93.8% in the low-risk group, 69.4% to 92.5% in the intermediate-risk group, and 52.5% to 80.5% in the high-risk group. The anHR of mortality remained very low (0–5.4%) in the low-risk groups, and was initially high but remained decreasing (0–15.1%) in the intermediate-risk group. In the high-risk group, it was initially markedly high (up to 28.4%) and had a late elevation through 7 to 9 years. These data from our large multicenter database provide dynamic information on awareness of prognosis and counselling of subsequent treatment decisions or care plans for early-stage NKTCL patients treated with radiotherapy. According to risk stratification by nomogram, patterns of CS and anHR between different risk groups distinctly differ, which may help to devise a risk-adjusted surveillance protocol.

The optimal radiation therapy (RT) dose in the definitive treatment of stage I and II extranodal nasal-type NK/T-cell lymphoma (NKTCL) remains controversial. The aim of this study was to optimize RT dose, and determine the survival benefit of RT on the basis of improved locoregional control (LRC) in the largest cohort of early-stage patients. A total of 1332 patients with early-stage NKTCL from ten institutions were retrospectively reviewed. Patients received chemotherapy (CT) alone (n = 172, as defined as 0 Gy), RT alone (n = 293), RT followed by CT (n = 214), or CT followed by RT (n = 653). For patients treated with RT, the majority of patients received ≥50 Gy (n = 996, 85.9%); only 14.1% received 10-49 Gy (n = 164). RT dose was entered into the Cox regression as a continuous variable to allow for a relationship between RT dose and mortality. Regression analysis was used to assess whether a linear relationship exists between LRC and progression-free survival (PFS) or overall survival (OS). The risk of locoregional recurrence, disease progression/relapse, and mortality was lowest at 50 Gy, in a dose-dependent manner. After adjustment for all covariates in a multivariable Cox model, a maximal risk reduction is still observed at 50 Gy. According to different RT dose groups (0 Gy, 10-39 Gy, 40-49 Gy, 50-59 Gy, and 60-70 Gy), regression analysis of the data revealed a linear relationship between the 5-year LRC and 5-year PFS (determination coefficient, R2 = 0.99, P < 0.001) or 5-year OS (R2 = 0.97, P < 0.001). The 5-year PFS rate (%) was equal to 0.988 × 5-year LRC (%) -19.59, while the 5-year OS rate (%) was equal to 0.97 × 5-year LRC (%) - 4.084. For patients who received RT with or without chemotherapy, the 5-year LRC, PFS and OS rates were 85.0%, 60.9%, and 69.5% for ≥50 Gy compared with 72.8% (P < 0.001), 50.2% (P = 0.004), and 58.4% (P = 0.037) for 10-49 Gy. RT dose (10-49 Gy vs. ≥50 Gy) was also an independent prognostic factor for LRC, PFS, and OS in multivariate analysis. For patients who achieved complete response (CR) after initial chemotherapy, RT dose of ≥50 Gy resulted in significantly better LRC than RT dose of 10-49 Gy (87.7% versus 73.3%, P = 0.040). Based on the data from this study and other studies in the literature, we demonstrated a linear relationship between LRC and PFS (R2 = 0.735) or OS (R2 = 0.829). The 50 Gy is the optimal RT dose for early stage patients, and improvement in LRC can translated into improved PFS and OS. The reduced RT dose lead to inferior LRC for patients achieved CR after induction chemotherapy. Clinicians should be aware of the importance of controlling locoregional disease and maintaining a high long-term survival. This finding may help clinicians to assist in defining appropriate standard care and decision making for early-stage NKTCL.

The prognosis and optimal therapeutic strategy for patients with stage II extranodal nasal-type NK/T-cell lymphoma (NKTCL) are not well defined. The aim of this study was to evaluate the prognostic factors and treatment outcomes of patients with stage II NKTCL. A total of 345 patients with stage II NKTCL from 10 Chinese institutions were reviewed. Fifty-two patients were treated with radiation therapy (RT) alone, 63 patients with chemotherapy (CT) alone, and 230 patients with a combination of CT and RT (CMT). Patients were subclassified as low-, intermediate-, and high-risk groups according to prognostic nomogram score, as described in our previous study (Leukemia, 2015). A comprehensive comparative study was performed using multivariable and propensity score-matched (PSM) analyses. The 5-year overall survival (OS) and progression-free survival (PFS) for all patients were 52.0% and 44.8%, respectively. On univariate analysis, only ECOG score ≥ 2 was associated with poor OS and PFS, whereas on multivariate analyses, ECOG score ≥ 2 and presence of primary tumor invasion (PTI) were independent prognostic factors. Patients at low-risk (48), intermediate-risk (49-100), and high-risk (>100) groups, as defined by nomogram score, had significantly different OS (P = 0.002) and PFS (P = 0.015). There was no significant difference in 5-year OS (45.2% versus 26.2%, P = 0.211) or PFS (40.0% versus 15.0%, P = 0.062) between RT alone and CT alone groups. However, compared with single modality therapy, CMT significantly improved survivals. The 5-year OS and PFS rates were 59.4% and 53.6% for CMT, compared with 35.1% (P < 0.001) and 25.8% (P < 0.001) for single modality therapy. After adjustment with PSM, similar significant survival differences were still observed between CMT and single modality therapy. One hundred and eighty-four patients experienced disease progression or relapse. The 5-year cumulative incidences of locoregional and systemic failure were 22.4% and 47.5%, respectively. Patients with stage II NKTCL had an unfavorable prognosis. CMT were proved to be more effective in terms of OS and PFS than CT or RT alone. Further study needs to focus on innovative systematic therapy and optimizing sequence of RT and CT.

In patients with diffuse large B-Cell lymphoma (DLBCL), complete response (CR) to systemic therapy alone can translate into long-term survival. Prior studies have demonstrated that radiation therapy (RT) for early-stage NK/T-cell lymphoma (NKTCL) can improve tumor local control and prolong survival. However, the role of RT in patients with CR after chemotherapy (CT) for early-stage NKTCL remains controversial. This study was conducted to determine whether RT could be safely omitted in patients with early-stage NKTCL who achieved a CR after CT. A total of 844 patients with early-stage NKTCL from 10 institutions received initial CT. Of these patients, 217 achieved a CR after CT. Of the 217 complete responders, 160 patients received sequential planned RT (CT+RT), whereas 57 patients did not (CT alone). A comprehensive comparative study was performed using multivariable and propensity score matching (PSM) method. Compared with CT alone, CT+RT showed significantly better overall survival (OS) (5-year OS: 79.1% versus 59.0%; P = 0.011) and progression-free survival (PFS) (5-year PFS: 70.0% versus 38.0%; P < 0.001). After adjustment with PSM to balance the clinical features, significant differences in OS and PFS were still observed between two treatment groups. Subgroup analysis revealed that, for patients treated with new CT regimens (L-asparaginase–based or gemcitabine-based), adding RT to new regimen chemotherapy significantly improved both OS (P = 0.012) and PFS (P = 0.030). On multivariate analysis, additional RT remained an independent prognostic factor for OS and PFS. Twenty-four of 160 patients (15.0%) treated with CT+RT developed locoregional recurrences, compared with 25 (43.9%) patients treated with CT alone. Systemic failures were found in 33 (20.6%) patients treated with CT+RT and 19 (33.3%) patients treated with CT alone. Consolidation RT was associated with higher locoregional control (LRC) (P < 0.001) and distant metastasis free survival (DMFS) (P= 0.030) compared with CT alone. Addition of RT to CT as an upfront treatment in patients with early-stage NKTCL who achieved a CR was associated with improved LRC, DMFS, PFS, and OS. Omission of RT should not be considered because of the high risk of locoregional recurrence and systemic failure in CR patients received CT alone.

Subtypes defined by primary sites have not been well studied in extranodal nasal-type NK/T-cell lymphoma (NKTCL). This study aimed to determine the clinical diversity and treatment outcome of NKTCL arising in the upper aerodigestive tract (UADT-NKTCL) in a large cohort of multicenter study from an endemic area. A nationwide survey was performed. A total of 1147 patients with a newly diagnosed UADT-NKTCL from 9 large centers in China between 2000 and 2010 were retrospectively reviewed. The primary sites of UADT-NKTCL were localized in the nasal cavity and paranasal sinus (n = 860), nasopharynx (n = 156), tonsil (n = 55), oropharynx (n = 22), and larynx (n = 14). The hard and soft palate (n = 9), oral cavity (n = 7), base of the tongue (n = 6) and hypopharynx (n = 1) are rarely involved (< 1%). In total, 860 patients (75%) had primary nasal-UADT-NKTCL, and 287 patients (25.0%) had primary extranasal-UADT-NKTCL. Among the latter 287 patients, 256 patients (22.3%) had primary Waldeyer's ring NKTCL (nasopharynx, tonsil, oropharynx and base of the tongue, WR-NKTCL) and 156 patients (13.5%) had nasopharyngeal-NKTCL. Compared with patients with nasal-UADT-NKTCL, patients with extranasal-UADT-NKTCL had more adverse clinical features including stage II-IV disease (61.3% vs. 22.9%, p < 0.001), regional lymph node involvement (57.8% vs. 20.0%, p < 0.001), B symptoms (45.6% vs. 38.5%, p = 0.032), poor performance status (ECOG ≥ 1, 75.3% vs. 64.3%, p = 0.007), and high risk scores of the international prognostic index (IPI ≥ 2, 17.4% vs. 9.2%, p = 0.001) or Korean prognostic index (KPI ≥2, 48.4% vs. 27.0%, p < 0.001). Patients with extranasal-UADT-NKTCL showed a similar overall survival (5-year OS, 56.9% vs. 62.2%, P = 0.183), but worse progression-free survival (5-year PFS, 42.5% vs. 55.2%, P = 0.003) than those with nasal-UADT-NKTCL. The prognosis for extranasal-UADT-NKTCL was similar or superior to that of nasal-UADT-NKTCL, when stratified by stage I disease (5-year OS, 64.1% vs. 67.9%, P = 0.429), stage II disease (63.9% vs. 45.9%, P = 0.002), and stage III/IV disease (22.1% vs. 13.0%, P = 0.019). Similarly, clinical and prognostic differences were also observed between nasal-UADT-NKTCL and extranasal-UADT-NKTCL, nasopharyngeal-NKTCL or WR-NKTCL. For early stage disease, primary radiation therapy with or without chemotherapy showed better OS for entire group of UADT-NKTCL (67.9% vs. 34.4%, p < 0.001), and even for nasal-UADT-NKTCL (67.8% vs. 28.1%, p < 0.001), or extranasal-UADT-NKTCL (67.7% vs. 45.4%, p = 0.001) than chemotherapy alone. Patients with nasal and extranasal UADT-NKTCL present with significantly different clinical features and prognosis.

To evaluate the risk of locoregional recurrence(LRR) associated with locoregional treatment of women with T1-2N1M0 breast cancer negative for estrogen receptor, progesterone receptor and human epidermal growth factor receptor 2 (Triple-Negative breast cancer [TNBC]) . Two hundreds and fifteen patients diagnosed with T1-2N1M0 TNBC were retrospectively analyzed. All patients were treated with modified radical mastectomy (MRM). Of them, 66 patients received postmastectomy radiation therapy and 146 patients did not. Locoregional recurrence-free survival (LRRFS) and overall survival (OS) were compared between two groups with or without propensity-score matching methods. With a median follow-up of 55.7 months, 36 patients developed locoregional recurrence. The 5-year LRR-free survival and OS rates were 92.6% and 82.8% for MRM compared with 76.6 % (P = 0.01) and 84.7 (P = 0.499) for postmastectomy radiation therapy, respectively. In multivariate analysis, MRM (compared with MRM +RT) and T2 were associated with increased LRR. In matched-pair of radiation therapy and no radiation therapy using propensity-score matching methods, the 5-year LRR-free survival was 92.6% for MRM plus RT compared with 74.5% for MRM (P = 0.008). Multivariate analysis indicated that no radiation therapy was the only independent prognostic factor associated with increased LRR (hazard ratio, 3.536; 95% CI = 1.153 to 10.844; P = 0.027. Patients with T1-2N1M0 TNBC treated with MRM without RT appear to be at a significantly increased risk for LRR compared with those treated with MRM and RT. Prospective studies are warranted to investigate the benefit of postmastectomy radiation therapy to improve the outcome of patients in T1-2N1 TNBC.