Differences in treatment effect size between progression-free survival and overall survival in anti-PD-1/PD-L1 inhibitors-based trials in advanced NSCLC: a systematic review and meta-analysis.
Differences in treatment effect size between progression-free survival and overall survival in anti-PD-1/PD-L1 inhibitors-based trials in advanced NSCLC: a systematic review and meta-analysis.
复制标题
基于抗 PD-1/PD-L1 抑制剂的晚期 NSCLC 试验中无进展生存期和总生存期之间治疗效果大小的差异：系统评价和荟萃分析。
DOI：
10.21037/tlcr-21-199
复制
发表时间：
2021-06
影响因子：
4
通讯作者：
Jiang T
中科院分区：
文献类型：
--
作者：
Zhou Z;Ren S;Chen L;Zhou C;Jiang T
目的探讨程序性细胞死亡1(PD-1)及其配体(PD-L1)阻断治疗晚期非小细胞肺癌(NSCLC)的无进展生存期(PFS)和总生存期(OS)的疗效大小差异。用危险比(RHR)：PFS和OS的危险比(HR)来评估PFS和OS之间的治疗效果大小的差异。对所有试验进行随机效应荟萃分析，以生成合并的RHR。我们还利用logHRPFS和logHROS之间的Spearman相关系数(R)来评估采用PFS作为操作系统代理的可行性。共纳入27项随机对照试验(RCT)，涉及15,590名患者。平均而言，OS的治疗效果大小与PFS相似(合并RHR，0.98；95%CI，0.91至1.08)。亚组分析显示，在以免疫治疗为二线或以上治疗的试验中，OS的治疗效果大小大于PFS(RHR，1.17；95%CI，1.06至1.29)，而在以免疫治疗为一线设置的试验中，PFS的差异大于OS(RHR，0.91；95%CI，0.84至0.99；P交互作用；0.01)。LogHRPFS与logHROS之间的决定系数为40%，相关系数为0.63。亚组分析显示，以免疫治疗为一线治疗的试验的决定系数和相关系数分别为62%和0.79，以免疫治疗为二线或以上治疗的试验的决定系数和相关系数分别为22%和0.47。在不同的基于抗PD-(L)1抑制剂的治疗试验中，PFS和OS之间的治疗效果大小大致一致。在以免疫治疗为一线治疗的试验中，PFS可能是OS的潜在替代终点，但对于以免疫治疗为第二或更高一线治疗的试验，PFS应谨慎地解释为没有OS数据。
To investigate the differences in treatment effect sizes between progression-free survival (PFS) and overall survival (OS) in advanced non-small cell lung cancer (NSCLC) treated with programmed cell death 1 (PD-1) and its ligand (PD-L1) blockade-based treatments. The differences in treatment effect sizes between PFS and OS were assessed by using a ratio of hazard ratio (rHR): the HR for PFS to that for OS. A random effects meta-analysis across trials was conducted to generate the combined rHR. We also evaluated the feasibility of adopting PFS as the surrogate of OS by using Spearman correlation coefficient (R) between logHRPFS and logHROS. A total of 27 randomized controlled trials (RCTs) with 15,590 patients were included. Treatment effect sizes were comparable, on average, for OS than for PFS (pooled rHR, 0.98; 95% CI, 0.91 to 1.08). Subgroup analysis revealed that treatment effect sizes were greater for OS than for PFS for trials with immunotherapy as second or above line treatment (rHR, 1.17; 95% CI, 1.06 to 1.29), while the differences were greater for PFS than for OS for trials with immunotherapy as first-line setting (rHR, 0.91; 95% CI, 0.84 to 0.99; Pinteraction<0.01). The coefficient of determination was 40% and R was 0.63 between logHRPFS and logHROS. Subgroup analysis showed that coefficient of determination and R were 62% and 0.79 in trials with immunotherapy as first-line setting, 22% and 0.47 in trials with immunotherapy as second or above line treatment, respectively. Treatment effect sizes between PFS and OS were roughly consistent in trials with different anti-PD-(L)1 inhibitor-based therapies. PFS could be a potential alternative endpoint for OS in trials with immunotherapy as first-line setting, but PFS should be cautiously interpreted without OS data for trials with immunotherapy as second or above line treatment.