Kahana Noam, Boaz Elad, Horesh Nir, Emile Sameh Hany, Dourado Justin, Aeschbacher Pauline, Rogers Pete, Gefen Rachel, Lo Menzo Emanuele, Rosenthal Raul J
Department of General Surgery, Cleveland Clinic Florida, 2950 Cleveland Clinic Blvd., Weston, FL, 33331, USA.
Department of General Surgery Shaare Zedek Medical Center, Faculty of Medicine, Hebrew University of Jerusalem, Jerusalem, Israel.
Surg Endosc. 2024 Dec;38(12):7037-7044. doi: 10.1007/s00464-024-11343-3. Epub 2024 Oct 23.
Esophageal cancer remains a significant global health challenge. Several treatment modalities were explored in randomized controlled trials (RCTs) in recent decades. This study evaluates the robustness of RCTs focusing on esophageal cancer treatment using the fragility index (FI) and reverse fragility index (RFI).
A systematic review of RCTs studying different treatment modalities for esophageal cancer from 2000 to 2023 was conducted. The FI and RFI were utilized to gauge the robustness of statistically significant and non-significant outcomes, respectively. The FI represents the minimal number of patient outcomes that would need to alter to overturn a trial's statistical significance, while RFI indicates the minimal changes required to achieve significance in non-significant results.
Out of 4028 studies retrieved, 21 RCTs were included for final analysis. The studies spanned 2001 to 2023 with a mean followup of 66 months (range, 29-108 months) and median number of patients of 194 (range, 45-802). The most common treatment modalities examined in these studies were neoadjuvant chemoradiotherapy (n = 7, 33.3%), neoadjuvant chemotherapy (n = 4, 19.0%), and neoadjuvant immunotherapy (n = 2, 9.5%). Only 5 studies (23.8%) had a statistically significant primary outcome result with a median FI of 6 (IQR, 2.5-8.5). Non-significant primary outcomes were seen in 16 studies (76.2%) with a median RFI of 4 (IQR 1-11) and lost to followup of 0 (IQR 0-4). In the study with the highest FI (10), the FI was lower than the number of patients lost to followup (13).
Our findings demonstrate that most RCTs on esophageal cancer treatments did not report significant primary outcomes. The few studies that reported significant results had a low fragility index, suggesting a vulnerability in their findings.
食管癌仍然是一项重大的全球健康挑战。近几十年来，随机对照试验（RCT）探索了多种治疗方式。本研究使用脆弱性指数（FI）和反向脆弱性指数（RFI）评估聚焦于食管癌治疗的RCT的稳健性。
对2000年至2023年研究食管癌不同治疗方式的RCT进行系统评价。FI和RFI分别用于衡量具有统计学显著性和非显著性结果的稳健性。FI代表要推翻试验的统计学显著性所需改变的最少患者结局数量，而RFI表示使非显著性结果达到显著性所需的最少变化。
在检索到的4028项研究中，纳入21项RCT进行最终分析。这些研究涵盖2001年至2023年，平均随访66个月（范围29 - 108个月），患者中位数为194例（范围45 - 802例）。这些研究中研究的最常见治疗方式是新辅助放化疗（n = 7，33.3%）、新辅助化疗（n = 4，19.0%）和新辅助免疫治疗（n = 2，9.5%）。只有5项研究（23.8%）有统计学显著性的主要结局结果，中位数FI为6（IQR，2.5 - 8.5）。16项研究（76.2%）出现非显著性主要结局，中位数RFI为4（IQR 1 - 11），失访率为0（IQR 0 - 4）。在FI最高的研究（10）中，FI低于失访患者数量（13）。
我们的研究结果表明，大多数关于食管癌治疗的RCT未报告显著性主要结局。少数报告显著性结果的研究脆弱性指数较低，表明其研究结果存在脆弱性。
