中晚期宫颈癌患者同步放化疗后缓解效果的影响因素及预测模型构建

Influencing factors in remission effect of patients with advanced cervical cancer after concurrent chemoradiotherapy and prediction model construction

目的探讨中晚期宫颈癌患者同步放化疗后缓解效果的影响因素并构建预测模型,旨在为临床疗效预测及治疗方案制定提供更多的参考。方法回顾性纳入2020年1月至2023年1月在江苏省海安市人民医院行同步放化疗治疗中晚期宫颈癌患者共131例,根据同步放化疗后是否达到缓解分为缓解组(90例)和未缓解组(41例);采用单因素和多因素法评估中晚期宫颈癌患者同步放化疗后缓解效果的独立影响因素,同时基于上述独立危险因素构建偏回归系数模型方程式,分析该模型用于中晚期宫颈癌患者同步放化疗后的缓解效果预测临床效能。结果单因素分析结果显示,肿瘤最大径、宫旁浸润、血小板/淋巴细胞比值、中性粒细胞/淋巴细胞比值及单核细胞/淋巴细胞比值均可能与中晚期宫颈癌患者同步放化疗后的缓解效果有关(P<0.05);Logistic回归模型多因素分析结果显示,肿瘤最大径、宫旁浸润、血小板/淋巴细胞比值、中性粒细胞/淋巴细胞比值及单核细胞/淋巴细胞比值均是中晚期宫颈癌患者同步放化疗后缓解效果的独立影响因素(P<0.05);根据Logistic回归模型证实中晚期宫颈癌患者同步放化疗后缓解效果的独立影响因素,构建自变量偏回归系数模型方程式:中晚期宫颈癌患者同步放化疗后达到缓解可能性Logit(Y)=1.84-1.16×肿瘤最大径(cm)-2.72×宫旁浸润+0.08×血小板/淋巴细胞比值-1.09×中性粒细胞/淋巴细胞比值-1.13×单核细胞/淋巴细胞比值;ROC曲线分析结果显示,上述临床模型用于中晚期宫颈癌患者同步放化疗后的缓解效果预测曲线下面积为0.88(95%CI:0.83~0.94),特异度为75.90%,灵敏度为87.36%,最佳cut-off值为0.65。结论中晚期宫颈癌患者同步放化疗后的缓解效果与多种因素有关,如肿瘤最大径、宫旁浸润及免疫炎症指标等;而基于以上指标构建预测模型有助于指导临床医生早期识别同步放化疗后疗效不佳的人群。

Objective To explore the factors influencing the remission effect of concurrent chemoradiotherapy in patients with advanced cervical cancer,and constructing a predictive model,in order to provide more reference basis for clinical efficacy prediction and treatment plan.Methods A total of 131 patients with advanced cervical cancer who underwent concurrent chemoradiotherapy in our hospital from January 2020 to January 2023 were retrospectively included.They were divided into a remission group(90 cases)and an unrelieved group(41 cases)based on whether they achieved remission after concurrent chemoradiotherapy.Using univariate and multivariate methods to evaluate the independent influencing factors of remission effect in patients with advanced cervical cancer after concurrent chemoradiotherapy,and based on the independent risk factors mentioned above,constructing a partial regression coefficient model equation,the clinical efficacy of the model in predicting remission effect in patients with advanced cervical cancer after concurrent chemoradiotherapy is analyzed.Results The results obtained from univariate analysis indicate that the maximum tumor diameter,parametrial infiltration,platelet/lymphocyte ratio,neutrophil/lymphocyte ratio,and monocyte/lymphocyte ratio may all be related to the remission effect of concurrent chemoradiotherapy in patients with advanced cervical cancer(P<0.05).The results of multivariate analysis using logistic regression model showed that tumor maximum diameter,parametrial infiltration,platelet/lymphocyte ratio,neutrophil/lymphocyte ratio,and monocyte/lymphocyte ratio were all independent influencing factors(P<0.05)on the effect of concurrent chemoradiotherapy in patients with advanced cervical cancer.According to the logistic regression model,it was confirmed that there are independent influencing factors on the remission effect of concurrent chemoradiotherapy in patients with advanced cervical cancer.An independent variable partial regression coefficient model equation was constructed,and the likelihood of achieving remission in patients with advanced cervical cancer after concurrent chemoradiotherapy was obtained as Logit(Y)=1.84-1.16×maximum tumor diameter(cm)-2.72×parametrial infiltration+0.08×platelet/lymphocyte ratio-1.09×neutrophil/lymphocyte ratio-1.13×monocyte/lymphocyte ratio.The predictive area under the curve for the remission effect of concurrent chemoradiotherapy in patients with advanced cervical cancer using this clinical model is 0.88(95%CI:0.83~0.94),with a specificity of 75.90%,sensitivity of 87.36%,and an optimal cut-off value of 0.65.Conclusions The remission effect of concurrent chemoradiotherapy in patients with advanced cervical cancer is related to many factors,such as tumor maximum diameter,paracentral infiltration and immunoinflammatory indexes.The prediction model based on the above indicators is helpful to guide clinicians to early identify the poor clinical effect after concurrent chemoradiotherapy.