老年营养风险指数在结外NK/T细胞淋巴瘤中的预后价值评估:一项淮海淋巴瘤工作组的多中心回顾性研究

Prognostic value of GNRI in patients with extranodal NK/T cell lymphoma:a multicenter retrospective study of the Huaihai Lymphoma Working Group

目的:探讨老年营养风险指数(GNRI)及临床特征对结外NK/T细胞淋巴瘤(ENKTL)患者的预后影响。方法:回顾性收集2014年2月—2021年6月在淮海淋巴瘤协作组中9家医疗中心确诊的298例ENKTL患者的临床资料。基于MaxStat算法获取GNRI的最佳截断值,使用Cox比例风险模型进行单因素和多因素分析以确定影响ENKTL患者生存结局的变量。使用Kaplan-Meier分析计算生存率并绘制生存曲线,组间比较采用log-rank检验。基于受试者工作特征的曲线下面积评价GNRI整合国际预后指数(IPI)、韩国预后指数和自然杀伤淋巴瘤预后指数的预测能力。结果:298例患者的中位年龄为49岁,男232例(77.9%),非鼻型患者51例。基于MaxStat统计量计算出GNRI的最佳截断值为97.85。多因素分析结果显示,GNRI<97.85、非鼻型、体重指数(超重)、高淋巴细胞计数、高血肌酐是ENKTL患者的独立危险因素(P<0.05)。亚组分析结果表明,GNRI可以对ECOG PS评分<2分组和ECOG PS评分≥2分组、IPI评分处于低危组和中低危组以及IPI评分处于中高危和高危组的患者进行精准预后分层;GNRI整合3个模型后的曲线下面积可以明显提高。结论:GNRI在评估ENKTL患者预后方面具有一定的判断价值。GNRI≥97.85的患者预后情况较好,且GNRI整合到3个模型后的预后能力更好。

Objective: To investigate the effect of geriatric nutritional risk index(GNRI) and clinical features on the prognosis of patients with extranodal NK/T cell lymphoma based on multicenter data. Methods: Clinical data of 298 patients with ENKTL diagnosed in nine medical centers in the Huaihai Lymphoma Working Group from February 2014 to June 2021 were retrospectively collected. The MaxStat algorithm was used to obtain the best cut-off value of GNRI. Cox proportional hazard model was used for univariate and multivariate analysis of prognosis. Kaplan-Meier analysis was used to calculate the survival rate and draw the survival curve. And the differences between groups were tested by log-rank. The area under the curve based on the receiver operating characteristic curve was used to evaluate the predictive ability of GNRI to integrate IPI, KPI and PINK. Results: The median age of the 298 patients was 49 years, 232 cases(77.9%) were male, and 51 cases were non-nasal. Based on MaxStat statistics, the optimal cut-off value of GNRI was 97.85. Multivariate analysis showed that GNRI<97.85, non-nasal type, BMI(overweight), high lymphocyte count, and high serum creatinine were independent risk factors for ENKTL patients(P<0.05). The results of subgroup analysis showed that GNRI could accurately stratify the prognosis of patients with ECOG PS score<2 and ECOG PS score≥2, IPI(low risk+low and intermediate risk) group and IPI(high and intermediate risk+high risk) group. In addition, the area under the curve of GNRI can be significantly improved after the integration of the three models. Conclusion: GNRI has a certain value in evaluating the prognosis of ENKTL patients. Patients with GNRI ≥ 97.85 indicates a good prognosis and the integration of GNRI into the three models showed better prognostic ability.