[摘要] 目的：探讨NSCLC患者癌组织中Ki-67 与PD-L1 表达的相关性及两者对患者预后的影响。方法：选取符合纳入标准的2012 年1 月至2018 年8 月在海军军医大学附属长海医院，手术确诊为NSCLC并进行免疫组化PD-L1 和Ki-67 检测的患者401 例，收集其临床病理资料，定期生存随访，应用统计学方法分析Ki-67 与PD-L1 表达的相关性及两者对患者术后DFS和化疗后PFS的影响。结果：NSCLC组织中PD-L1 和Ki-67 表达阳性率分别为37.9%（152/401）和96.3%（386/401），单因素分析显示Ki-67 为PDL1表达相关的影响因素（OR=0.33，95%CI=0.28～0.39，P<0.0001），曲线拟合分析显示Ki-67 与PD-L1 表达显著正相关，阈值效应分析、分段多因素Logistic 和ROC曲线分析表明14%是Ki-67 较适宜与PD-L1 联用的阈值。Kaplan-Meier 分析显示，术后DFS，Ki-67 高表达组显著短于Ki-67 低表达组[(21.88±11.25) vs (41.22±16.25)个月，P<0.0001]，PD-L1 阳性组显著短于PD-L1 阴性组[(24.75±14.59) vs (38.27±16.75)个月，P<0.0001]，Ki-67 高表达/PD-L1 阳性组与其余3 组相比术后DFS 最短[(20.57±11.33) vs(24.11±10.79) vs (36.00±16.79) vs (42.91±15.77)个月，P<0.0001]；化疗PFS，Ki-67 高表达组显著长于Ki-67 低表达组[(7.70±3.01) vs(5.80±2.99)个月，P=0.016]，PD-L1 阳性组与阴性组相比差异无统计学意义[(7.04±3.21) vs (6.33±3.06)个月，P=0.22]，Ki-67 与PDL1联合测评，Ki-67 高表达两组的PFS显著长于Ki-67 低表达两组[(7.74±3.25) vs (7.43±2.38) vs(4.91±1.97) vs (6.02±3.19)个月，P=0.041]。结论：NSCLC组织中Ki-67 与PD-L1 表达呈正相关，Ki-67 14%是适宜与PD-L1 联用的阈值，Ki-67 和PD-L1 均为患者预后不良的预测因子，两者联合对预后不良的预测有“叠加效应”，同时Ki-67 高表达患者对化疗的敏感性较好。

[Abstract] Objective: To investigate the relationship between Ki-67 and PD-L1 in patients with non-small cell lung cancer (NSCLC)and their effects on prognosis. Methods: A total of 401patients, who were pathologically diagnosed as NSCLC in Changhai Hospital from January 2012 to August 2018, were enrolled as study subjects; and the patients were immunohistochemically tested for PD-L1 and Ki-67. The clinical and pathological data were collected, and the follow-up was performed regularly. The correlation between Ki-67 and PD-L1 and their effects on postoperative DFS and post-chemotherapy PFS were statistically analyzed. Results: Positive rates of PD-L1 and Ki-67 in NSCLC tissues were 37.9% (152/401) and 96.3% (386/401), respectively. Univariate analysis showed that Ki-67 was an influencing factor for PD-L1 expression (OR=0.33, 95%CI=0.28-0.39, P<0.0001); Curve Fitting analysis showed a positive correlation between Ki-67 and PD-L1; threshold effect analysis, segmentation multivariate logistic and ROC curve analysis showed 14% is a relatively suitable threshold for Ki-67 to be combined with PD-L1. Kaplan-Meier analysis showed that patients in Ki-67 high expression group had a significantly shorter post-operative DFS than those in Ki-67 low expression group ([21.88±11.25] vs [41.22±16.25]m, P<0.0001), patients in PD-L1 positive group had a significantly shorter DFS than those in PD-L1 negative group ([24.75±14.59] vs [38.27±16.75]m, P<0.0001)], and patients in Ki-67 high /PD-L1 positive group had the shortest DFS as compared to the other three groups ([20.57±11.33] vs [24.11±10.79], [36.00±16.79], [42.91±15.77]m, P<0.0001). As for post-chemotherapy PFS, patients in Ki-67 high ex-pression group was significantly longer than those in Ki-67 low expression group [(7.70±3.01) vs (5.80±2.99)m, P=0.016), but there was no significant difference between PD-L1 positive group and PD-L1 negative group [(7.04±3.21) vs (6.33±3.06)m, P=0.22); for combined evaluation with Ki-67 and PD-L1, the PFS of two Ki-67 high expression groups was significantly longer than the other two Ki-67 low expression groups [(7.74±3.25) vs (7.43±2.38) vs (4.91±1.97) vs (6.02±3.19)m, P=0.041). Conclusion: Ki-67 is positively correlated with PD-L1 in NSCLC patients, and Ki-67 14% is a suitable threshold for combined use with PD-L1. Both Ki-67 and PD-L1 are predictors of poor prognosis. The combination of the two has an "additive effect" on the prediction of poor prognosis, and patients with high Ki-67 expression are more sensitive to chemotherapy.

上海市科研计划项目资助（No.19411970600）