PDF(Pubmed)
Abstract：
BACKGROUND: To identify the cut-off values for the number of metastatic lymph nodes (nMLN) and lymph node ratio (LNR) that can predict outcomes in patients with FIGO 2018 IIICp cervical cancer (CC).
METHODS: Patients with CC who underwent radical hysterectomy with pelvic lymphadenectomy were identified for a propensity score-matched (PSM) cohort study. A receiver operating characteristic (ROC) curve analysis was performed to determine the critical nMLN and LNR values. Five-year overall survival (OS) and disease-free survival (DFS) rates were compared using Kaplan-Meier and Cox proportional hazard regression analyses.
RESULTS: This study included 3,135 CC patients with stage FIGO 2018 IIICp from 47 Chinese hospitals between 2004 and 2018. Based on ROC curve analysis, the cut-off values for nMLN and LNR were 3.5 and 0.11, respectively. The final cohort consisted of nMLN ≤ 3 (n = 2,378) and nMLN > 3 (n = 757) groups and LNR ≤ 0.11 (n = 1,748) and LNR > 0.11 (n = 1,387) groups. Significant differences were found in survival between the nMLN ≤ 3 vs the nMLN > 3 (post-PSM, OS: 76.8% vs 67.9%, P = 0.003; hazard ratio [HR]: 1.411, 95% confidence interval [CI]: 1.108-1.798, P = 0.005; DFS: 65.5% vs 55.3%, P < 0.001; HR: 1.428, 95% CI: 1.175-1.735, P < 0.001), and the LNR ≤ 0.11 and LNR > 0.11 (post-PSM, OS: 82.5% vs 76.9%, P = 0.010; HR: 1.407, 95% CI: 1.103-1.794, P = 0.006; DFS: 72.8% vs 65.1%, P = 0.002; HR: 1.347, 95% CI: 1.110-1.633, P = 0.002) groups.
CONCLUSIONS: This study found that nMLN > 3 and LNR > 0.11 were associated with poor prognosis in CC patients.
METHODS: Patients with CC who underwent radical hysterectomy with pelvic lymphadenectomy were identified for a propensity score-matched (PSM) cohort study. A receiver operating characteristic (ROC) curve analysis was performed to determine the critical nMLN and LNR values. Five-year overall survival (OS) and disease-free survival (DFS) rates were compared using Kaplan-Meier and Cox proportional hazard regression analyses.
RESULTS: This study included 3,135 CC patients with stage FIGO 2018 IIICp from 47 Chinese hospitals between 2004 and 2018. Based on ROC curve analysis, the cut-off values for nMLN and LNR were 3.5 and 0.11, respectively. The final cohort consisted of nMLN ≤ 3 (n = 2,378) and nMLN > 3 (n = 757) groups and LNR ≤ 0.11 (n = 1,748) and LNR > 0.11 (n = 1,387) groups. Significant differences were found in survival between the nMLN ≤ 3 vs the nMLN > 3 (post-PSM, OS: 76.8% vs 67.9%, P = 0.003; hazard ratio [HR]: 1.411, 95% confidence interval [CI]: 1.108-1.798, P = 0.005; DFS: 65.5% vs 55.3%, P < 0.001; HR: 1.428, 95% CI: 1.175-1.735, P < 0.001), and the LNR ≤ 0.11 and LNR > 0.11 (post-PSM, OS: 82.5% vs 76.9%, P = 0.010; HR: 1.407, 95% CI: 1.103-1.794, P = 0.006; DFS: 72.8% vs 65.1%, P = 0.002; HR: 1.347, 95% CI: 1.110-1.633, P = 0.002) groups.
CONCLUSIONS: This study found that nMLN > 3 and LNR > 0.11 were associated with poor prognosis in CC patients.
摘要：
背景：为了确定可以预测FIGO2018IIICp宫颈癌（CC）患者预后的转移性淋巴结（nMLN）数量和淋巴结比率（LNR）的临界值。
方法：接受根治性子宫切除术伴盆腔淋巴结清扫术的CC患者被确定为倾向评分匹配（PSM）队列研究。进行受试者工作特征(ROC)曲线分析以确定临界nMLN和LNR值。使用Kaplan-Meier和Cox比例风险回归分析比较了5年总生存率（OS）和无病生存率（DFS）。
结果：本研究包括2004年至2018年间来自47家中国医院的3,135名FIGO2018IIICp期CC患者。基于ROC曲线分析,nMLN和LNR的截止值分别为3.5和0.11。最终队列包括nMLN≤3（n=2,378）和nMLN>3（n=757）组和LNR≤0.11（n=1,748）和LNR>0.11（n=1,387）组。nMLN≤3与nMLN>3之间的生存率存在显着差异（PSM后，操作系统：76.8%vs67.9%，P=0.003；风险比[HR]：1.411，95％置信区间[CI]：1.108-1.798，P=0.005；DFS：65.5％vs55.3％，P<0.001;HR:1.428,95%CI:1.175-1.735,P<0.001),LNR≤0.11且LNR>0.11(PSM后，操作系统：82.5%vs76.9%，P=0.010；HR：1.407，95%CI：1.103-1.794，P=0.006；DFS：72.8%vs65.1%，P=0.002;HR:1.347,95%CI:1.110-1.633,P=0.002)组。
结论：本研究发现nMLN>3和LNR>0.11与CC患者的不良预后相关。
方法：接受根治性子宫切除术伴盆腔淋巴结清扫术的CC患者被确定为倾向评分匹配（PSM）队列研究。进行受试者工作特征(ROC)曲线分析以确定临界nMLN和LNR值。使用Kaplan-Meier和Cox比例风险回归分析比较了5年总生存率（OS）和无病生存率（DFS）。
结果：本研究包括2004年至2018年间来自47家中国医院的3,135名FIGO2018IIICp期CC患者。基于ROC曲线分析,nMLN和LNR的截止值分别为3.5和0.11。最终队列包括nMLN≤3（n=2,378）和nMLN>3（n=757）组和LNR≤0.11（n=1,748）和LNR>0.11（n=1,387）组。nMLN≤3与nMLN>3之间的生存率存在显着差异（PSM后，操作系统：76.8%vs67.9%，P=0.003；风险比[HR]：1.411，95％置信区间[CI]：1.108-1.798，P=0.005；DFS：65.5％vs55.3％，P<0.001;HR:1.428,95%CI:1.175-1.735,P<0.001),LNR≤0.11且LNR>0.11(PSM后，操作系统：82.5%vs76.9%，P=0.010；HR：1.407，95%CI：1.103-1.794，P=0.006；DFS：72.8%vs65.1%，P=0.002;HR:1.347,95%CI:1.110-1.633,P=0.002)组。
结论：本研究发现nMLN>3和LNR>0.11与CC患者的不良预后相关。
