为了量化过去二十年美国食品和药物管理局 (FDA) 批准的新型抗癌药物的随机对照试验 (RCT) 的统计证据的强度。我们使用了总生存期 (OS)、无进展生存期 (PFS) 的数据）和 FDA 在 2000 年 1 月至 2020 年 12 月期间首次批准的新型抗癌药物的肿瘤反应（TR）。我们通过计算所有可用终点的贝叶斯因子（BF）来评估统计证据的强度，并使用以下方法汇总证据对基于两项随机对照试验批准的适应症进行贝叶斯固定效应荟萃分析。对终点、批准途径、治疗方案和癌症类型之间的统计证据强度进行了比较。我们分析了 82 个 RCT 的可用数据，对应于单个 RCT 支持的 68 个适应症和两个 RCT 支持的 7 个适应症。 OS（BF = 1.9；IQR 0.5-14.5）的统计证据中位强度不明确，PFS（BF = 24,767.8；IQR 109.0-7.3*106）和 TR（BF = 113.9；IQR 3.0-547,100）的统计证据中位强度较强。总体而言，44 种适应症 (58.7%) 获得批准，但没有明确的 OS 改善统计证据；7 种适应症 (9.3%) 获得批准，但没有任何终点改善的统计证据。与所有三个终点的非加速批准相比，加速批准的统计证据强度较低。治疗线和癌症类型没有观察到有意义的差异。该分析仅限于统计证据。我们没有考虑非统计因素（例如偏倚风险、证据质量）。BF 为癌症药物批准背后的统计证据的强度提供了新的见解。大多数新型抗癌药物缺乏强有力的统计证据证明它们可以改善 OS，并且有一些药物缺乏疗效的统计证据。这些案件需要透明、清晰的解释。当证据不明确时，额外的上市后试验可以减少不确定性。版权所有 © 2024。由 Elsevier Inc. 出版。

To quantify the strength of statistical evidence of randomised controlled trials (RCTs) for novel cancer drugs approved by the Food and Drug Administration (FDA) in the last two decades.We used data on overall survival (OS), progression-free survival (PFS), and tumour response (TR) for novel cancer drugs approved for the first time by the FDA between January 2000 and December 2020. We assessed strength of statistical evidence by calculating Bayes Factors (BFs) for all available endpoints, and we pooled evidence using Bayesian fixed-effect meta-analysis for indications approved based on two RCTs. Strength of statistical evidence was compared between endpoints, approval pathways, lines of treatment, and types of cancer.We analysed the available data from 82 RCTs corresponding to 68 indications supported by a single RCT and seven indications supported by two RCTs. Median strength of statistical evidence was ambiguous for OS (BF = 1.9; IQR 0.5-14.5), and strong for PFS (BF = 24,767.8; IQR 109.0-7.3*106) and TR (BF = 113.9; IQR 3.0-547,100). Overall, 44 indications (58.7%) were approved without clear statistical evidence for OS improvements and seven indications (9.3%) were approved without statistical evidence for improvements on any endpoint. Strength of statistical evidence was lower for accelerated approval compared to non-accelerated approval across all three endpoints. No meaningful differences were observed for line of treatment and cancer type.This analysis is limited to statistical evidence. We did not consider non-statistical factors (e.g., risk of bias, quality of the evidence).BFs offer novel insights into the strength of statistical evidence underlying cancer drug approvals. Most novel cancer drugs lack strong statistical evidence that they improve OS, and a few lack statistical evidence for efficacy altogether. These cases require a transparent and clear explanation. When evidence is ambiguous, additional post-marketing trials could reduce uncertainty.Copyright © 2024. Published by Elsevier Inc.