【作者】 张璐；

【导师】 张海军；

【作者基本信息】 东南大学 ， 临床医学（专业学位）， 2022， 硕士

【摘要】 目的:本研究旨在探讨术前全身免疫炎症指数(Systemic immune-inflammation index,SII)对接受根治性手术的结肠癌(Colon cancer)患者术后1、2、3年无病生存期(Disease-free survival,DFS)的预测价值,同时构建列线图预测结肠癌患者术后1、2、3年DFS的概率,提供网页版无病生存概率计算器。方法:回顾性分析2013年9月1日至2018年8月31日于东南大学附属中大医院住院行根治性切除术的初诊结肠癌患者共计188例,并收集相应的临床病理资料。1.根据术前血常规结果计算SII值(SII=血小板(P)×中性粒细胞(N)/淋巴细胞(L)),依据SII的中位数将纳入研究的对象分为高SII组(n=94例)和低SII组(n=94例),比较两组患者临床特征上的差异性。2.应用生存分析,比较SII水平高低对患者术后DFS的影响,并对可能影响术后DFS的因素进行单因素及多因素Cox分析。3.将多因素Cox分析得出的独立危险因素引入R软件(R4.1.1),构建患者根治术后1、2、3年DFS概率的列线图预测模型,并生成网页版DFS概率计算工具。4.在出现术后复发的患者中,按照复发方式,将患者分为发生肝转移组(n=36例)和非肝转移组(n=27例),分析影响肝转移的可能因素。结果:1.所有188例结肠癌患者术后中位随访时间1294(206-2766)天,随访期间,63例出现术后复发(其中局部复发11例,52例出现远处转移。在52例出现远处转移病人中,出现肝转移36例,多发转移及其他部位单发转移共16例),死亡16例(其中因复发转移死亡共13例,包括因肝转移死亡6例及因多发转移及其他部位单发转移死亡6例及局部复发死亡1例)。术后复发率33.5%,死亡率8.5%。2.SII的中位数为514.13,以此将188例患者分为低SII组(SII<514.13,94例)和高SII组(SII≥514.13,94例)。两组除了在肿瘤直径上有统计学差异外(c~2=8.489,P=0.004),其余临床特征均无统计学差异(P>0.05)。单因素分析提示术前SII水平、患者年龄、术前癌胚抗原(Carcinoembryonic antigen,CEA)水平、原发肿瘤部位、脉管侵犯、神经侵犯以及病理学N、TNM分期是影响结肠癌患者术后复发的相关因素(P<0.1),对上述指标进行多因素COX分析提示术前SII水平、患者年龄、原发肿瘤部位、病理学N分期(P<0.05)是结肠癌患者术后复发的独立危险因素。同时Kaplan-Meier生存分析提示高低两组患者术后平均DFS分别为1833.57±117.644天及2091.91±106.308天,两者统计学差异(P=0.048)。3.将上述所述的独立危险因素纳入R软件,构建列线图预测模型预测结肠癌患者根治术后1、2、3年DFS的概率(每个独立危险因素对应相应分值,具体如下:0分:术前SII<514.13,30分:SII≥514.13;0分:年龄<65周岁,52分:年龄≥65周岁;0分:原发肿瘤位于左半结肠,50分:原发肿瘤位于右半结肠;0分:肿瘤分期N1,60分:N2,100分:N3)。受试者工作特征曲线(Receiver operating characteristic curve,ROC)及校准曲线表明列线图的预测结果与实际的观测结果一致性较好。同时通过“Dyn Nom”和“shiny”软件包成功构建网页版的1、2、3年DFS概率的动态计算器(https://medicalclimbers.shinyapps.io/Dyn Nomapp/)。4.将术后肿瘤复发的63例患者(包括因复发转移死亡的13例患者)按是否发生肝转移分为肝转移组36例,非肝转移组27例,两组在原发肿瘤部位上有统计学差异(c2=5.452,P=0.020),其余相关临床特征上未见明显统计学差异。结论:1.术前SII水平可以作为预测结肠癌患者根治性手术后DFS的指标。2.年龄≥65周岁,术前SII升高,原发肿瘤位于右半结肠,术后病理N分期为N2的结肠癌患者具有更差的DFS。3.根据独立危险因素构建的列线图有助于预测患者行结肠癌根治术1、2、3年的DFS概率,且预测价值良好。由此生成的网页版动态计算器可便捷计算术后3年内DFS的概率。4.原发肿瘤位于右半结肠较左半结肠更容易发生肝转移。更多还原

【Abstract】 Objective:The study aimed to investigate the predictive value of preoperative systemic immune-inflammatory index(SII)of 1-,2-and 3-year disease-free survival(DFS)in patients with colon adenocarcinoma who underwent radical surgery.At the same time,a nomogram was constructed to predict the probability of disease-free survival in patients with colon adenocarcinoma at 1,2,and 3 years after surgery,and on the basis of this,a web-based probability calculator was established.Methods:A total of 188 newly diagnosed patients with colon cancer were retrospectively analyzed after receiving radical surgery in Zhongda Hospital affiliated to Southeast University from September 1,2013 to August 31,2018.The corresponding clinicopathological data were collected at the same time.1.The SII values were calculated according to the results of preoperative blood routine(SII=platelet(P)×neutrophil(N)/lymphocyte(L)).On the basis of the median of SII,the enrolled patients were divided into the high SII group(94 cases)and the low SII group(94cases).The prevalence of clinical characteristics between the two groups were compared.2.Survival analysis was used to compare the influence of various SII level on DFS of patients.The univariate and multivariate Cox regression analysis were performed on the risk factors that might affect postoperative DFS.3.The independent risk factors obtained via multivariate Cox regression analysis were introduced into R software(R4.1.1)to construct a nomogram prediction model of DFS in 1,2 and 3 years after radical surgery.According to the nomogram above,a web-based calculation tool was built.4.Patients with postoperative recurrence were divided into liver metastasis group(36 cases)and non-liver metastasis group(27 cases)according to the way of relapse to analyze the possible factors affecting liver metastasis.Results:1.The median follow-up time after surgery of all patients with colon cancer was 1294(206-2766)days.During the follow-up period,63 of 188 cases had postoperative recurrence including 11 local recurrence and 52 distant metastasis.Among 52 patients with distant metastasis,36 had liver metastasis while other 16 patients had multiple metastasis or single metastasis in other sites.16 patients died including 6 cases who died of liver metastasis and other 7 cases who died of multiple metastasis and single metastasis in other locations.Postoperative recurrence rate and mortality was 33.5%and 8.5%respectively.2.The median of SII was 514.13,which divided 188 patients into the low SII group(SII<514.13,94 cases)and the high SII group(SII≥514.13,94 cases).There were no significant differences in clinical features between the two groups except for the tumor diameter(c~2=8.489,P=0.004).Univariate analysis suggested that preoperative SII level,the age of patients,preoperative carcinoembryonic antigen(CEA)level,primary tumor location,presence of vascular and nerve invasion,pathological N and TNM stage were the related factors affecting postoperative DFS of colon cancer patients(P<0.1).At the same time,multivariate Cox analysis of the above indicators showed that preoperative SII level,the age of patients,primary tumor location and pathological N stage(P<0.05)were the independent risk factors for postoperative DFS of patients with colon cancer.Besides,there was a statistical difference via Kaplan-Meier survival analysis in the length of DFS between the two groups which were1833.57±117.644 days and 2091.91±106.308 days,respectively(P=0.048).3.The independent risk factors mentioned above were incorporated into R software,and a nomogram prediction model was constructed to predict the probability of DFS in 1,2 and 3years after radical resection of colon cancer.Each independent risk factor had a corresponding score as follows:0 point:preoperative SII<514.13,30 points:SII≥514.13;0 points:age<65years old,52 points:age≥65 years old;0 point:primary tumor of left colon,50 points:primary tumor of right colon;0:tumor N1 stage,60 points:N2,100 points:N3).The receiver operating characteristic curve(ROC)and calibration curves showed that the prediction results of the nomogram were in good agreement with the actual observation results.Meanwhile,a dynamic calculatorfor1,2,and3-yearDFSprobability(https://medicalclimbers.shinyapps.io/Dyn Nomapp/)was successfully constructed through the"Dyn Nom"and"shiny"package.4.The 63 patients with postoperative tumor recurrence including 13 patients who died of recurrence and metastasis were divided into liver metastasis group(36 cases)and non-liver metastasis group(27 cases).There was a statistically significant difference in the location of primary tumor between the two groups(c~2=5.452,P=0.020).Unfortunately,it failed to find the significant difference in other clinical features.Conclusion:1.Preoperative SII level can be used as an indicator to predict DFS of patients with colon cancer after radical surgery.2.Those patients with colon cancer whose age were 65 years or older,with elevated preoperative SII,primary tumor located in the right colon,and postoperative pathological staging N2 had worse DFS.3.The nomogram constructed according to the independent risk factors is helpful to predict the 1-,2-and 3-year DFS probability of patients after radical surgery and has good predictive value.The web-based dynamic calculator can conveniently calculate the probability of DFS within 3 years after surgery.4.The primary tumor located in the right colon is more likely to have liver metastases than the left colon.更多还原