基于SEER数据库的急性淋巴细胞白血病患者继发第二肿瘤的预测模型构建-现代医学

基于SEER数据库的急性淋巴细胞白血病患者继发第二肿瘤的预测模型构建

单位：新乡医学院第一附属医院血液内科, 河南新乡 453100

分类号：R181.2

出版年·卷·期（页码）：2025·44·第六期（941-947）

摘要：

目的： 构建基于监测、流行病学和结果计划(SEER)数据库的急性淋巴细胞白血病(ALL)患者继发第二肿瘤的预测模型。方法： 在SEER数据库中筛选出19 683例ALL患者临床资料，将收集到的受试者资料按照7∶3的比例随机分为训练集(n=13 778)和测试集(n=5 905)。根据是否继发第二肿瘤将训练集分为继发组(n=268)和未继发组(n=13 510)，采用Logistic回归分析影响ALL患者继发第二肿瘤的相关因素。将Logistic回归分析结果纳入R软件，构建ALL患者继发第二肿瘤风险列线图预测模型，并对构建的模型进行验证。结果： 本研究ALL患者的队列中，19 683例患者中有383例患者继发第二肿瘤，发生率为1.95%。与普通人群相比，ALL患者继发第二肿瘤的风险更高[相对风险(SIR)=2.354，95%CI2.125~2.687]，ALL患者继发第二肿瘤的绝对超额率(AER)为16.852。继发组年龄、免疫分型、化疗、放疗、白细胞计数(WBC)与未继发组比较，差异有统计学意义(P<0.05)。Logistic回归分析结果显示，年龄<60岁、放疗、化疗、WBC≥100×10⁹ L^-1与ALL患者继发第二肿瘤关系密切(P＜0.05)。列线图风险模型显示，训练集和测试集预测ALL患者继发第二肿瘤的ROC曲线下面积(AUC)分别为0.876(95%CI0.689～0.993)、0.865(95%CI 0.672~0.989)，敏感度分别为96.70%、94.30%，特异度分别为79.50%、78.70%(P<0.001)，表明该模型预测效能良好。列线图预测模型C指数为0.889(95%CI0.812~0.994)。训练集和测试集校准曲线结果显示，训练集和测试集的校准曲线均在理想曲线附近，且Hosmer-Lemeshow拟合优度曲线检验χ²=2.843，P=0.673。结论： 年龄<60岁、化疗、放疗、WBC≥100×10⁹ L^-1与ALL患者继发第二肿瘤关系密切。基于上述因素构建的ALL患者继发第二肿瘤风险列线图模型具有良好的预测效果，具有临床指导作用。

Objective: To construct a predictive model for secondary tumors in patients with acute lymphoblastic leukemia(ALL) using the Surveillance, Epidemiology, and End Results(SEER) database. Methods: We systematically screened clinical data of 19 683 ALL patients from the SEER database. Eligible subjects were randomly allocated into a training set(n=13 778) and testing set(n=5 905) at a 7∶3 ratio. Patients in the training set were stratified into secondary tumor and non-secondary tumor groups. Logistic regression identified factors influencing secondary tumor development. The analysis results were incorporated into R software to construct and validate a nomogram prediction model. Results: Among 19 683 ALL patients, 383(1.95%) developed secondary tumors. Compared with the general population, ALL patients showed significantly elevated secondary tumor risk [standardized incidence ratio(SIR)=2.354, 95%CI 2.125-2.687], with an absolute excess risk(AER) of 16.852. There were significant differences in age, immunotyping, chemotherapy, radiotherapy and WBC between the secondary group and the non-secondary group(P<0.05).Logistic regression revealed age, radiotherapy, chemotherapy, and WBC ≥100×10⁹ L^-1 as significant predictors(P<0.05). The nomogram model demonstrated good predictive performance with AUCs of 0.876(training set) and 0.865(testing set), sensitivities of 96.70% and 94.30%, specificities of 79.50% and 78.70%(P<0.001), respectively. The C-index reached 0.889(95%CI 0.812-0.994). Calibration curves showed close alignment with ideal predictions, supported by Hosmer-Lemeshow test(χ²=2.843, P=0.673). Conclusion: Age, chemotherapy, radiotherapy, and WBC ≥100×10 L^-1 are significant risk factors for secondary tumors in ALL patients. The developed nomogram model shows excellent predictive capacity and clinical utility.

参考文献：

[1] JABBOUR E，SHORT N J，SENAPATI J，et al.Mini-hyper-CVD plus inotuzumab ozogamicin，with or without blinatumomab，in the subgroup of older patients with newly diagnosed Philadelphia chromosome-negative B-cell acute lymphocytic leukaemia:long-term results of an open-label phase 2 trial[J].Lancet Haematol，2023，10(6):433-444.
[2] 陈俐，傅彩红.急性白血病化疗患儿医院感染的影响因素分析及防控策略研究[J].现代医学，2024，52(2):209-214.
[3] SI LIM S J，FORD J B，HERMISTON M L.How I treat newly diagnosed and refractory T-cell acute lymphoblastic lymphoma in children and young adults[J].Blood，2023，141(25):3019-3030.
[4] ALDOSS I，GU Z，AFKHAMI M，et al.Ph-like acute lymphoblastic leukemia in adults:understanding pathogenesis，improving outcomes，and future directions for therapy[J].Leuk Lymphoma，2023，64(6):1092-1101.
[5] CANAANI J，FRISCH A，POLLYEA D A，et al.Venetoclax-based salvage therapy for adult patients with relapsed/refractory acute lymphoblastic leukemia[J].Eur J Haematol，2023，111(3):365-372.
[6] FIRTINA S，ERBILGIN Y，HATIRNAZ N G O，et al.Impact of TP53 gene variants on prognosis and survival of childhood acute lymphoblastic leukemia[J].Scand J Clin Lab Invest，2023，83(3):187-193.
[7] DASUKIL S，LAHOTI A，JENA A K，et al.A rare maxillofacial manifestation of acute lymphoblastic leukemia in a 9-year-old child[J].J Clin Pediatr Dent，2022，46(5):98-101.
[8] AN K，HE Y，TANG Y，et al.Histiocytic sarcoma following CAR T-cell therapy:a case report[J].Int J Hematol，2024，119(3):338-341.
[9] WANG W，WANG X，CHE G，et al.The Establishment and verification of a nomogram model for predicting the overall survival of medullary thyroid carcinoma:an analysis based on the SEER database[J].Curr Oncol，2023，31(1):84-96.
[10] WANG C，WU B，SHEN D，et al.Establishment and validation of a prognostic nomogram for patients with renal cell carcinoma based on SEER and TCGA database[J].Transl Cancer Res，2023，12(6):1411-1421.
[11] 金运钦，姜海虹，喻静,等.急性淋巴性白血病患者发生第二原发癌症的影响因素分析[J].公共卫生与预防医学，2023，34(1):25-30.
[12] LUU Y，HAN J，AGARWAL A，et al.Risk for second primary cancers among pediatric and young adult melanoma survivors[J].Pediatr Dermatol，2024，41(1):12-15.
[13] TEUFEL A，LI M，GERKEN M，et al.Second cancer after additive chemotherapy in patients with colon cancer[J].Clin Colorectal Cancer，2022，21(4):354-361.
[14] DOTE S，INOSE R，GOTO R，et al.Risk of a second cancer and infection in patients with indolent B-cell lymphoma exposed to first-line bendamustine plus rituximab:a retrospective analysis of an administrative claims database[J].Hematol Oncol，2023，41(3):354-362.
[15] 马李洁，赵建强，刘卉,等.急性髓系白血病患者外周血FOXO1,PBX3的表达及对预后的预测价值[J].现代生物医学进展，2023，23(9):1766-1770.
[16] WU H，DING P，WU J，et al.A new online dynamic nomogram:construction and validation of a predictive model for distant metastasis risk and prognosis in patients with gastrointestinal stromal tumors[J].J Gastrointest Surg，2023，27(7):1429-1444.
[17] LIU Z，ZHANG L，LI G，et al.A nomogram model for prediction of mortality risk of patients with dangerous upper gastrointestinal bleeding:a two-center retrospective study[J].Curr Med Sci，2023，43(4):723-732.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】

提示：您还未登录，请登录！点此登录