DNA损伤修复基因列线图模型预测头颈部肿瘤患者放疗后甲状腺功能减退发生风险的价值

Value of DNA damage repair gene nomogram model to the prediction of hypothyroidism after radiotherapy for head and neck cancer

目的 分析头颈部肿瘤患者XRCC1、XRCC3、ERCC2基因单核苷酸多态性,观察放疗后甲状腺功能减退(甲减)发生情况,探讨构建的列线图模型预测头颈部肿瘤患者放疗后发生甲减的价值。方法 120例头颈部肿瘤患者均给予调强放疗,随访6个月记录甲减发生情况。120例患者中40例发生甲减者为甲减组,80例甲状腺功能正常者为对照组。比较2组年龄、性别比例、肿瘤部位及放疗前红细胞计数、白细胞计数、白蛋白、血肌酐水平及放疗总处方剂量等临床资料;采用聚合酶链反应-限制性片段长度多态性分析法于放疗前检测XRCC1、XRCC3、ERCC2基因单核苷酸多态性;多因素logistic回归分析头颈部肿瘤患者放疗后发生甲减的影响因素;构建列线图模型,采用ROC曲线、校准曲线、内部决策曲线预测头颈部肿瘤患者放疗后发生甲减的价值。结果 甲减组女性比率(62.5%)高于对照组(40.0%)(χ~2=5.413,P=0.019),年龄,体质量指数,肿瘤部位,合并糖尿病、高血压病、冠心病、慢性肾脏病比率,红细胞计数,白细胞计数,白蛋白,血肌酐水平及放疗总处方剂量与对照组比较差异均无统计学意义(P>0.05)。甲减组XRCC1基因280位点AA型(57.5%)、399位点GG型(55.0%)比率均高于对照组(32.5%、31.3%)(P<0.05),ERCC2基因rs1799793位点GG型(25.0%)比率低于对照组(48.8%)(P<0.05),GA型比率(70.0%)高于对照组(45.0%)(P<0.05);2组XRCC1基因280位点AT、TT型,194位点CC、CG、GG型,399位点GC、CC型;XRCC3基因241位点AA、AT、TT型;ERCC2基因rs1799793位点AA型,rs13181位点AA、AC、CC型比率比较差异均无统计学意义(P>0.05)。女性(OR=2.500,95%CI:1.145~5.459,P=0.021)、XRCC1基因280位点AA型(OR=7.733,95%CI:1.969~30.377,P=0.003)、XRCC1基因399位点GG型(OR=5.976,95%CI:1.398~25.542,P=0.016)是头颈部肿瘤患者放疗后发生甲减的危险因素;ERCC2基因rs1799793位点GG型(OR=0.058,95%CI:0.006~0.539,P=0.012)是头颈部肿瘤患者放疗后发生甲减的保护因素。列线图模型预测头颈部肿瘤患者放疗后发生甲减的AUC为0.935(95%CI:0.891~0.973,P<0.001),灵敏度为87.34%,特异度为81.22%,模型校准曲线与理想模型接近;决策曲线显示该模型曲线远离横、纵坐标,预测价值较高。结论 女性、XRCC1基因280位点AA型、XRCC1基因399位点GG型的头颈部肿瘤患者放疗后易发生甲减,构建的列线图模型对头颈部肿瘤患者放疗后发生甲减有较高预测价值。

Objective To analyze the single nucleotide polymorphisms of XRCC1,XRCC3 and ERCC2 genes in patients with head and neck cancer,to observe the occurrence of hypothyroidism after radiotherapy,and to investigate the value of model construction to the prediction of hypothyroidism after radiotherapy.Methods A total of 120patients with head and neck cancer received intensity-modulated radiotherapy,and were divided into 40 cases of hypothyroidism(hypothyroidism group)and 80cases of normal thyroid function (control group)according to the occurrence of hypothyroidism after 6-month follow-up.The age,gender ratio,cancer location,red blood cell count,white blood cell count,albumin,serum creatinine and total prescription dose were compared between two groups.PCR-restriction fragment length polymorphism analysis was used to detect the single nucleotide polymorphisms of XRCC1,XRCC3 and ERCC2 genes.Multivariate logistic regression was used to analyze the influencing factors of hypothyroidism after radiotherapy in patients with head and neck cancer.Nomogram model was constructed,and ROC curve,calibration curve and internal decision curve were used to predict hypothyroidism after radiotherapy in patients with head and neck cancer.Results The female ratio was higher in hypothyroidism group(62.5%)than that in control group(40.0%)(χ~2=5.413,P=0.019).There were no significant differences in the age,body mass index,cancer location,percentages of patients with diabetes mellitus,hypertension,coronary heart disease and chronic kidney disease,red blood cell count,white blood cell count,albumin,serum creatinine and total prescription dose between two groups (P>0.05).The percentages of XRCC1 gene 280AA genotype and 399GG genotype were higher in hypothyroidism group(57.5%,55.0%)than those in control group(32.5%,31.3%)(P<0.05).The percentage of ERCC2 gene rs1799793GG genotype was lower in hypothyroidism group(25.0%)than that in control group (48.8%)(P<0.05),and the percentage of ERCC2 gene rs1799793GA genotype was higher in hypothyroidism group(70.0%)than that in control group(45.0%)(P<0.05).There were no significant differences in XRCC1 gene 280AT and TT genotypes,194CC,CG and GG genotypes,399GC and CC genotypes,XRCC3 gene 241AA,AT and TT genotypes,rs1799793AA genotypes and rs13181AA,AC and CC genotypes between two groups (P>0.05).Female (OR=2.500,95%CI:1.145-5.459,P=0.021),XRCC1 gene280AA genotype(OR=7.733,95%CI:1.969-30.377,P=0.003),and XRCC1 gene 399GG genotype(OR=5.976,95%CI:1.398-25.542,P=0.016)were the risk factors of hypothyroidism after radiotherapy in patients with head and neck cancer,and ERCC2 gene rs1799793GG genotype(OR=0.058,95%CI:0.006-0.539,P=0.012)was a protective factor.The AUC of the nomogram model for predicting hypothyroidism in patients with head and neck cancer after radiotherapy was 0.935(95%CI:0.891-0.973,P<0.001),with a sensitivity of 87.34% and a specificity of 81.22%.The model calibration curve is close to the ideal model.The decision curve showed that the model curve was far from the abscissa and ordinate,and the predictive value was higher.Conclusion The female head and neck cancer patients with XRCC1 gene 280 AA genotype and XRCC1 gene 399 GG genotype are prone to develop hypothyroidism after radiotherapy,and the model has a high predictive value.