
浏览全部资源
扫码关注微信
1.陆军军医大学(第三军医大学):军事预防医学系军队卫生统计学教研室,重庆
2.陆军军医大学(第三军医大学):第一附属医院血液病中心,重庆
伍亚舟,E-mail:asiawu@tmmu.edu.cn
收稿日期:2024-10-23,
修回日期:2024-12-29,
纸质出版日期:2025-02-15
移动端阅览
黄 侦,伍亚舟.基于随机生存森林算法的NK/T细胞淋巴瘤患者预后模型的构建与验证[J].陆军军医大学学报,2025,47(03):275-284.DOI:10.16016/j.2097-0927.202410090.
HUANG Z,WU Y Z.Construction and validation of a prognostic model for NK/T-cell lymphoma based on random survival forest algorithm[J].J Army Med Univ,2025,47(03):275-284.DOI:10.16016/j.2097-0927.202410090.
黄 侦,伍亚舟.基于随机生存森林算法的NK/T细胞淋巴瘤患者预后模型的构建与验证[J].陆军军医大学学报,2025,47(03):275-284.DOI:10.16016/j.2097-0927.202410090. DOI:
HUANG Z,WU Y Z.Construction and validation of a prognostic model for NK/T-cell lymphoma based on random survival forest algorithm[J].J Army Med Univ,2025,47(03):275-284.DOI:10.16016/j.2097-0927.202410090. DOI:
目的
2
探讨NK/T细胞淋巴瘤(natural killer T-cell lymphoma,NKTL)患者的生存预后影响因素。基于随机生存森林(random survival forest,RSF)算法,构建预测NKTL患者总生存期(overall survival,OS)的预后模型。
方法
2
从SEER数据库收集2000-2020年的NKTL患者的人口统计学和临床病理资料。按照7∶3的比例将患者划分为训练队列(
n=
471)和验证队列(
n
=203)。通过Cox回归分析确定影响患者OS的预后因素,并基于分析结果构建列线图模型。同时,使用RSF算法确定影响患者OS的预后因素,并构建RSF模型。采用ROC曲线、校准曲线和决策曲线、净重新分类指数(net reclassification improvement,NRI)和综合判别改善指数(integrated discrimination improvement,IDI)对模型的预测能力进行评估,并比较2种模型的预测效果。通过2种模型计算每位患者的风险得分,根据风险得分的中位数,将患者分为高风险组和低风险组,并绘制生存曲线进行比较。
结果
2
Ann Arbor分期、年龄、放疗、联合治疗和疾病类型是与生存显著相关的预后变量。在验证队列中,列线图模型的1、3和5年ROC曲线下面积(area under curve, AUC)分别为0.745、0.771和0.748,而RSF模型的AUC为0.764、0.792和0.761。ROC曲线显示2种模型在预测OS方面具有良好的准确性和区分性。校准曲线显示2种模型预测的生存与实际生存之间具有良好的一致性。两种模型均能有效地将患者分为预后差组和预后好组,并且预后差组患者的OS显著低于预后好组(
P
<
0.000 1)。决策曲线显示RSF模型的净获益优于列线图模型。与列线图模型相比,RSF模型的NRI为0.184 (95%
CI
:0.098~0.267,
P
<
0.01),IDI为0.300 (95%
CI
:0.241~0.359,
P
<
0.01),RSF模型的预测能力优于列线图模型。
结论
2
Ann Arbor分期、年龄、放疗、联合治疗和疾病类型是NKTL患者预后的影响因素,本研究据此建立的RSF模型对NKTL患者预后具有很好的预测能力,可有效评估患者预后。
Objective
2
To investigate the prognostic factors affecting survival in patients with natural killer T-cell lymphoma (NKTL)
and then develop a prognostic model for predicting their overall survival (OS) based on random survival forest (RSF) algorithm.
Methods
2
Demographic and clinical pathological data of NKTL patients were collected from the SEER database during 2000 and 2020. The patients were divided into a training cohort (n=471) and a validation cohort (n=203) in a 7∶3 ratio. Cox regression analysis was performed to identify prognostic factors affecting OS
and a nomogram model was constructed based on the obtained factors. Meanwhile
RSF algorithm was used to determine prognostic factors affecting OS to build the RSF model. The models were evaluated using receiver operating characteristic (ROC) curve
calibration curve
decision curve
net reclassification improvement (NRI)
and integrated discrimination improvement (IDI)
and the predictive performances of the 2 models were compared. Risk scores for each patient were calculated using the 2 models. Then the patients were divided into high- and low-risk groups based on the median risk score
and survival curve was plotted for comparison.
Results
2
Ann Arbor stage
age
radiotherapy
combined treatment
and type of disease were identified as significant prognostic variables associated with OS. In the validation cohort
the area under the ROC curve (AUC) for the nomogram model at 1
3
and 5 years was 0.745
0.771
and 0.748
respectively
while the AUC for the RSF model was 0.764
0.792
and 0.761 at the same time points. ROC curve analysis indicated that both models demonstrated good accuracy and discrimination in predicting OS. Calibration curve analysis showed a strong consistency between the predicted and actual OS for both models. Both models effectively stratified the patients into poor and favorable prognosis groups
with the OS of patients in the poor prognosis group being significantly shorter than that of the favorable prognosis group (P<0.000 1). Decision curve analysis revealed that the net benefit of the RSF model was superior to that of the nomogram model. Compared to the nomogram model
the NRI for the RSF model was 0.184 (95%CI: 0.098~0.267
P<0.01)
and the IDI was 0.300 (95%CI: 0.241~0.359
P<0.01). Overall
the RSF model demonstrated superior predictive capability than the nomogram model.
Conclusion
2
Ann Arbor stage
age
radiotherapy
combined treatment
and type of disease are prognostic factors affecting the prognosis of NKTL patients. Our RSF model demonstrates strong predictive capability for the prognosis of NKTL patients and can effectively assess patient outcomes.
WANG H , FU B B , GALE R P , et al . NK-/ T-cell lymphomas [J]. Leukemia , 2021 , 35 ( 9 ): 2460 - 2468 . DOI: 10.1038/s41375-021-01313-2 http://dx.doi.org/10.1038/s41375-021-01313-2 .
TSE E , ZHAO W L , XIONG J , et al . How we treat NK/T-cell lymphomas [J]. J Hematol Oncol , 2022 , 15 ( 1 ): 74 . DOI: 10.1186/s13045-022-01293-5 http://dx.doi.org/10.1186/s13045-022-01293-5 .
KIM S J , YOON S E , KIM W S . Treatment of localized extranodal NK/T cell lymphoma, nasal type: a systematic review [J]. J Hematol Oncol , 2018 , 11 ( 1 ): 140 . DOI: 10.1186/s13045-018-0687-0 http://dx.doi.org/10.1186/s13045-018-0687-0 .
WEI Y C , QI F , ZHENG B M , et al . Intensive therapy can improve long-term survival in newly diagnosed, advanced-stage extranodal NK/T-cell lymphoma: a multi-institutional, real-world study [J]. Int J Cancer , 2023 , 153 ( 9 ): 1643 - 1657 . DOI: 10.1002/ijc.34672 http://dx.doi.org/10.1002/ijc.34672 .
MCINNIS M C , MA J , KARUR G R , et al . Chronic lung allograft dysfunction phenotype and prognosis by machine learning CT analysis [J]. Eur Respir J , 2022 , 60 ( 1 ): 2101652 . DOI: 10.1183/13993003.01652-2021 http://dx.doi.org/10.1183/13993003.01652-2021 .
YIN H F , XIE J Y , XING S , et al . Machine learning-based analysis identifies and validates serum exosomal proteomic signatures for the diagnosis of colorectal cancer [J]. Cell Rep Med , 2024 , 5 ( 8 ): 101689 . DOI: 10.1016/j.xcrm. 2024.101689 http://dx.doi.org/10.1016/j.xcrm.2024.101689 .
YAO F S , ZHANG R Q , LIN Q , et al . Plasma immune profiling combined with machine learning contributes to diagnosis and prognosis of active pulmonary tuberculosis [J]. Emerg Microbes Infect , 2024 , 13 ( 1 ): 2370399 . DOI: 10.1080/22221751.2024.2370399 http://dx.doi.org/10.1080/22221751.2024.2370399 .
BOEHM K M , AHERNE E A , ELLENSON L , et al . Multimodal data integration using machine learning improves risk stratification of high-grade serous ovarian cancer [J]. Nat Cancer , 2022 , 3 ( 6 ): 723 - 733 . DOI: 10.1038/s43018-022-00388-9 http://dx.doi.org/10.1038/s43018-022-00388-9 .
ZHANG T , AUER P , DONG J , et al . Whole-genome sequencing identifies novel predictors for hematopoietic cell transplant outcomes for patients with myelodysplastic syndrome: a CIBMTR study [J]. J Hematol Oncol , 2023 , 16 ( 1 ): 37 . DOI: 10.1186/s13045-023-01431-7 http://dx.doi.org/10.1186/s13045-023-01431-7 .
ZHANG Y Y , XU J , HUA J , et al . A PD-L2-based immune marker signature helps to predict survival in resected pancreatic ductal adenocarcinoma [J]. J Immunother Cancer , 2019 , 7 ( 1 ): 233 . DOI: 10.1186/s40425-019-0703-0 http://dx.doi.org/10.1186/s40425-019-0703-0 .
JAMET B , MORVAN L , NANNI C , et al . Random survival forest to predict transplant-eligible newly diagnosed multiple myeloma outcome including FDG-PET radiomics: a combined analysis of two independent prospective European trials [J]. Eur J Nucl Med Mol Imaging , 2021 , 48 ( 4 ): 1005 - 1015 . DOI: 10.1007/s00259-020-05049-6 http://dx.doi.org/10.1007/s00259-020-05049-6 .
ZHANG Z Q , HE G , LUO Z W , et al . Individual mortality risk predictive system of patients with acute-on-chronic liver failure based on a random survival forest model [J]. Chin Med J , 2021 , 134 ( 14 ): 1701 - 1708 . DOI: 10.1097/CM9.0000000000001539 http://dx.doi.org/10.1097/CM9.0000000000001539 .
CHE W Q , LI Y J , TSANG C K , et al . How to use the Surveillance, Epidemiology, and End Results (SEER) data: research design and methodology [J]. Mil Med Res , 2023 , 10 ( 1 ): 50 . DOI: 10.1186/s40779-023-00488-2 http://dx.doi.org/10.1186/s40779-023-00488-2 .
WU R Y , LI Y X , WANG W H , et al . Clinical disparity and favorable prognoses for patients with Waldeyer ring extranodal nasal-type NK/T-cell lymphoma and diffuse large B-cell lymphoma [J]. Am J Clin Oncol , 2014 , 37 ( 1 ): 41 - 46 . DOI: 10.1097/COC.0b013e318261084b http://dx.doi.org/10.1097/COC.0b013e318261084b .
CHEN S Y , YANG Y , QI S N , et al . Validation of nomogram-revised risk index and comparison with other models for extranodal nasal-type NK/T-cell lymphoma in the modern chemotherapy era: indication for prognostication and clinical decision-making [J]. Leukemia , 2021 , 35 ( 1 ): 130 - 142 . DOI: 10.1038/s41375-020-0791-3 http://dx.doi.org/10.1038/s41375-020-0791-3 .
WANG H , FU B B , WUXIAO Z J , et al . A prognostic survival nomogram for persons with extra-nodal natural killer-/ T-cell lymphoma [J]. Leukemia , 2022 , 36 ( 11 ): 2724 - 2728 . DOI: 10.1038/s41375-022-01679-x http://dx.doi.org/10.1038/s41375-022-01679-x .
CAO J Z , LAN S M , SHEN L H , et al . Hemoglobin level, a prognostic factor for nasal extranodal natural killer/T-cell lymphoma patients from stage I to IV: a validated prognostic nomogram [J]. Sci Rep , 2017 , 7 ( 1 ): 10982 . DOI: 10.1038/s41598-017-11137-9 http://dx.doi.org/10.1038/s41598-017-11137-9 .
TIAN D , YAN H J , HUANG H , et al . Machine learning-based prognostic model for patients after lung transplantation [J]. JAMA Netw Open , 2023 , 6 ( 5 ): e2312022 . DOI: 10.1001/jamanetworkopen.2023.12022 http://dx.doi.org/10.1001/jamanetworkopen.2023.12022 .
MINAMI T , SATO M , TOYODA H , et al . Machine learning for individualized prediction of hepatocellular carcinoma development after the eradication of hepatitis C virus with antivirals [J]. J Hepatol , 2023 : S0168-S8278(23)00424 - 5 . DOI: 10.1016/j.jhep.2023.05.042 http://dx.doi.org/10.1016/j.jhep.2023.05.042 .
HUANG B , SOLLEE J , LUO Y H , et al . Prediction of lung malignancy progression and survival with machine learning based on pre-treatment FDG-PET/CT [J]. EBioMedicine , 2022 , 82 : 104127 . DOI: 10.1016/j.ebiom.2022.104127 http://dx.doi.org/10.1016/j.ebiom.2022.104127 .
WANG Z Y , LI Y X , WANG H , et al . Unfavorable prognosis of elderly patients with early-stage extranodal nasal-type NK/T-cell lymphoma [J]. Ann Oncol , 2011 , 22 ( 2 ): 390 - 396 . DOI: 10.1093/annonc/mdq347 http://dx.doi.org/10.1093/annonc/mdq347 .
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 [J]. Ann Oncol , 2010 , 21 ( 5 ): 1032 - 1040 . DOI: 10.1093/annonc/mdp418 http://dx.doi.org/10.1093/annonc/mdp418 .
YANG Y , ZHU Y , CAO J Z , et al . Risk-adapted therapy for early-stage extranodal nasal-type NK/T-cell lymphoma: analysis from a multicenter study [J]. Blood , 2015 , 126 ( 12 ): 1424 - 1432 ;quiz1517. DOI: 10.1182/blood-2015-04-639336 http://dx.doi.org/10.1182/blood-2015-04-639336 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621