| Purpose:This study aims to determine failure pattern, distributing characterics and risk factors among patients with Stage IE and HE upper aerodigestive tract NK/T-cell lymphoma (UADT-NKTCL) after treatment,.Patients and Methods:Between December1979and December2012,468patients with Stage IE and HE UADT-NKTCL were treated. Radiotherapy was the primary treatment for most of patients.170patients were treated with radiotherapy alone,278patients with radiotherapy combined with chemotherapy,19patients with chemotherapy alone and one patient with anti-inflammation therapy. Cumulative probabilities for developing a failure were estimated using the Kaplan-Meier method. Cox regression analysis was used for multivariable analysis.Results:The median follow-up was35months (range,1-401months).162patients developed failures, and most were occurred within2years (130/162,80.2%). The2-year cumulative probability of total failure was30.7%. Distant extranodal failure was primay failure pattern. The2-year cumulative probability of distant extranodal, local, distant node and regional node failures were19.7%,9.3%,7.0%and2.9%, respectively. The common failure locations were skin, lung, liver and mesenteric nodal area. In risk factors analysis, local extensive disease, stage HE and ECOG≥2at initial diagnosis were independent risk factors to incidence of total failure with adjusted hazard ratios (HR) of1.882(1.312-2.501, p<0.001),1.812(1.312-2.501, p<.001) and1.922(1.143-3.230, p=0.014). Of these factors, local extensive disease were independent risk factors to incidence of local and distant extranodal failures with adjusted HR of2.233(1.189-4.196, p=0.013) and1.767(1.142-2.734, p=0.011). Stage HE werer independent risk factors to incidence of distant extranodal and distant lymph node failures with adjusted HR of2.178(1.444-3.286, p<0.001) and3.853(1.978-7.507, p<0.001). ECOG≥2were independent risk factors to distant extranodal failure with adjusted HR of1.940(1.003-3.749, p=0.049). In therapeutic factors analysis, the patients with chemotherapy alone had higher incidence of local and regional lymph node failures than radiotherapy alone and combined modality therapy. The2-year cumulative probability of local failures were69.1%,4.0%and9.8%(p<0.001), and the2-year cumulative probability of lymph node failures were29.9%,2.2%and2.1%, respectively (p<0.001). Combined modality therapy didn’t benefit patients with lower local and regional lymph node failures than radiotherapy alone. But it benefit stage HE patients with lower distant extranodal and distant lymph node failures than radiotherapy alone. The2-year cumulative probability of distant extranodal failures were29.7%and52.6%(p=0.007), and the2-year cumulative probability of distant lymph node failures were11.8%and37.5%, respectively (p=0.004). CR (complete response) after treatment had lower distant extranodal and distant lymph node failures than non-CR (p<0.001). Radiotherapy might decrease the incidence of distant failure by achieving a higher CR rate.Conclusion:Distant extranodal failure was primary failure pattern of stage IE and HE UADT-NKTCL after treatment. Radiotherapy benefited patients with lower incidence of local and regional lymph node failures, and with lower incidence of distant extranodal and distant lymph node failures at the same time. Patients with stage HE should be treated by combined modality therapy to decrease incidence of distant extranodal and distant lymph node failures. Part Ⅱ:Survival and prognostic model for patients with upper aerodigestive tract NK/T-cell lymphoma after failureAbstractPurpose:This study aims to evaluate survival of patients with upper aerodigestive tract NK/T-Cell Lymphoma after failure and to distinguish different risk groups using identified prognostic factors.Patients and Methods:From468patients with stage ⅠE and stage ⅡE upper aerodigestive tract Natural Killer/T-Cell Lymphoma (UADT-NKTCL) were treated between December1979and December2012,162patients with relapse or progression after initiall treatment were identified. Of those162patients,145patients who had full and detailed clinical data about their relpase or progression were enrolled this analysis. Survivals were estimated using the Kaplan-Meier method. Cox regression analysis was used for multivariable analysis.Results:The median follow-up after failure was8.1months (range,1-222months). Median overall survival (OS) and progression-free survival (PFS) after failure were9.6and6.1months respectively, and2-year OS and PFS were35.0%and26.5%, respectively. In multivariate analysis, independent risk factors for OS were relapse or progression within6months after the start of original treatment, ECOG≥2and more than one lesion at first relapse or progression. Hazard ratios were2.151(1.409-3.283, p<0.001),5.179(3.313-8.096, p<0.001) and1.920(1.254-2.941, p=0.003), respectively. Using three risk factors, we established prognostic index for UADT-NKTCL at first relapse or progresson (rNKTPI). The patients were identified four different rNKTPI groups:rNKTPI0, no risk factor; rNKTPI1, one risk factor; rNKTPI2, two risk factors; and rNKTPI3, three risk factors. Four subgroups with significantly different OS after first relapse or progression were identified (p<0.001). The patients with rNKTPI0-1had better OS. Median OS was37.8months, and2-year and5-year OS were57.5%and49.7%, respectively. In addition, relapse or progressive disease was still sensitive to radiotherapy, and refractory to conventional chemotherapy. The overall response rate was88.9%to radiation, but only38.6%to chemotherapy. The patients with salvage radiotherapy and combined modality therapy had better OS than the patients with salvage chemotherapy alone. The2-year OS were72.3%,64.1%and18.6%, respectively (p<0.001).Conclusion:Most of patients with relapsed or progressive UADT-NKTCL have poor survival outcome. Relapse or progressive disease was still sensitive to radiotherapy, and refractory to conventional chemotherapy. The prognostic modal of rNKTPI could identified patients with failure disease to four groups with significantely different OS. Patients with rNKTPI0-1had chance to have long-term survival by combined-modality therapy. Part Ⅲ:Late relapse in early stage extranodal natural killer/T-cell lymphoma, nasal typeAbstractPurpose:The purpose of this study was to determine the incidence of late relapse, which were defined as those occurring more than5years after initial treatment, among patients with Stage IE and HE extranodal nasal-type Natural Killer/T-Cell Lymphoma (NKTCL).Patients and Methods:Between January1983and October2007,114patients underwent treatment for Stage IE and HE NKTCL and were clinically without evidence of disease at5years of follow-up. All except two patients received radiotherapy (RT) alone (32patients) or RT combined with chemotherapy (80patients). Cumulative probabilities for developing a subsequent late relapse were estimated using the Kaplan-Meier method. Cox regression analysis was used for multivariable analysis.Results:Twelve patients had late relapse. The median time to late relapse was8.2years (range,5.1-23.1years) from the date of initial therapy.50%(6/12) patients with late relapse developed first locoregional relapse without systemic failure. The10-year cancer-specific survival and cumulative probabilities of late relapse for entire patients were92.2%and8.4%, respectively. In univariate analysis, B symptoms, ECOG≥2, chemotherapy alone and the dose of RT<50Gy were risk factors to incidence of late relapse, In the final multivariate Cox regression model, B symptom was identified as independent correlative factor to incidence of late relapse with an adjusted hazard ratio of4.74(95%confidence interval1.33～16.93, p=0.017).Conclusion:Late disease relapse occurred in a small proportion of early stage patients who lived more than5years after initial treatment. B symptoms at the initial diagnosis was an independent risk factor for late relapse. |
PDF Full Text Request