Radiographic changes might not fully capture the treatment effects of immune checkpoint inhibitors (ICIs). We aimed to assess correlations of overall response rate and progression-free survival with overall survival in trials of ICIs for metastatic non-small-cell lung cancer (NSCLC). To assess trial-level and patient-level correlations of overall response rate and progression-free survival with overall survival, we conducted a pooled analysis of first-line randomised trials (including patients aged ≥18 years with metastatic squamous and non-squamous NSCLC and an Eastern Cooperative Oncology Group performance status of 0–1) submitted to the US Food and Drug Administration from June 24, 2016, to March 16, 2021. Eligible trials evaluated at least one ICI in the experimental group versus chemotherapy in the control group. At the trial level, we used weighted linear regression to derive coefficients of determination (). At the patient level, we used Cox proportional hazards models to compare overall survival between responders versus non-responders per Response Evaluation Criteria in Solid Tumours (version 1.1). A total of 13 trials including 9285 patients evaluated ICIs alone or in combination with chemotherapy versus chemotherapy alone. At the trial level, the was 0·61 (95% CI 0·32–0·84) for correlation of overall response rate with overall survival and 0·70 (0·40–0·89) for correlation of progression-free survival with overall survival. Correlations ranged from weak to moderate when evaluating subgroups by PD-L1 expression and were consistent across trials evaluating ICIs alone or in combination with chemotherapy. At the patient level, responders had longer overall survival than non-responders (hazard ratio [HR] 0·28 [95% CI 0·26–0·30]). Among responders, overall survival was longer in patients enrolled in experimental groups than in control groups (HR 0·54 [95% CI 0·48–0·61]). Correlations of overall response rate and progression-free survival with overall survival were generally moderate in this pooled analysis. The findings support routine analysis of mature overall survival data, where feasible, in first-line randomised trials of ICIs for metastatic NSCLC. US Food and Drug Administration.

中文翻译：

---

转移性非小细胞肺癌免疫治疗试验中缓解率和无进展生存期与总生存期的相关性：FDA 汇总分析

放射学变化可能无法完全反映免疫检查点抑制剂 (ICIs) 的治疗效果。我们的目的是评估 ICI 治疗转移性非小细胞肺癌 (NSCLC) 试验中总体缓解率和无进展生存期与总体生存期的相关性。为了评估试验水平和患者水平的总体缓解率和无进展生存期与总体生存期的相关性，我们对一线随机试验（包括年龄≥18岁的转移性鳞状和非鳞状非小细胞肺癌患者和2016 年 6 月 24 日至 2021 年 3 月 16 日期间向美国食品和药物管理局提交了东部肿瘤合作组表现状态为 0-1 的结果。符合条件的试验评估了实验组中至少一种 ICI 与对照组化疗的比较。在试验层面，我们使用加权线性回归来推导决定系数 ()。在患者层面，我们使用 Cox 比例风险模型根据实体瘤疗效评估标准（1.1 版）比较有反应者与无反应者之间的总生存期。总共 13 项试验（包括 9285 名患者）评估了单独使用 ICI 或联合化疗与单独化疗的比较。在试验水平，总体缓解率与总生存期的相关性为 0·61 (95% CI 0·32–0·84)，无进展生存期的相关性为 0·70 (0·40–0·89)。生存与总体生存。当通过 PD-L1 表达评估亚组时，相关性范围从弱到中度，并且在单独评估 ICI 或与化疗联合评估的试验中保持一致。在患者层面，有反应者比无反应者有更长的总生存期（风险比 [HR] 0·28 [95% CI 0·26–0·30]）。在应答者中，实验组患者的总生存期比对照组患者更长（HR 0·54 [95% CI 0·48–0·61]）。在该汇总分析中，总体缓解率和无进展生存期与总体生存期的相关性一般为中等。这些发现支持在可行的情况下，在 ICI 治疗转移性 NSCLC 的一线随机试验中对成熟的总体生存数据进行常规分析。美国食品和药物管理局。