Background Prior systematic reviews addressing the impact of diet on cancer outcomes have focused on specific dietary interventions. In this systematic review, we assessed all RCTs investigating dietary interventions for cancer patients, examining the range of interventions, endpoints, patient populations, and results. Methods This systematic review identified all RCTs conducted prior to January 2023 testing dietary interventions in patients with cancer. Assessed outcomes included quality of life, functional outcomes, clinical cancer measurements (eg, progression-free survival, response rates), overall survival, and translational endpoints (eg, inflammatory markers). Results 252 RCTs were identified involving 31,067 patients. The median sample size was 71 (interquartile range 41 to 118), and 80 (32%) studies had a sample size greater than 100. Most trials (n = 184, 73%) were conducted in the adjuvant setting. Weight/body composition and translational endpoints were the most common primary endpoints (n = 64, 25%; n = 52, 21%, respectively). Direct cancer measurements and overall survival were a primary endpoint in 20 (8%) and seven (3%) studies respectively. Eight trials with a primary endpoint of cancer measurement (40%) met their endpoint. Large trials in colon (n = 1429), breast (n = 3088), and prostate cancer (n = 478) each showed no effect of dietary interventions on endpoints measuring cancer. Conclusion Most RCTs of dietary interventions in cancer are small and measure non-clinical endpoints. Although only a small number of large RCTs have been conducted to date, these trials have not shown an improvement in cancer outcomes. Currently, there is limited evidence to support dietary interventions as a therapeutic tool in cancer care.

中文翻译：

---

癌症的饮食干预：所有随机对照试验的系统评价

背景 先前关于饮食对癌症结果影响的系统评价主要集中在特定的饮食干预措施上。在这次系统评价中，我们评估了所有研究癌症患者饮食干预的随机对照试验，检查干预措施的范围、终点、患者群体和结果。方法 这项系统评价确定了 2023 年 1 月之前进行的所有随机对照试验，测试癌症患者的饮食干预措施。评估的结果包括生活质量、功能结果、临床癌症测量（例如无进展生存期、缓解率）、总生存期和转化终点（例如炎症标志物）。结果 确定了 252 项随机对照试验，涉及 31,067 名患者。中位样本量为 71（四分位数范围为 41 至 118），80 项研究（32%）的样本量大于 100。大多数试验（n = 184，73%）是在辅助环境中进行的。体重/身体成分和转化终点是最常见的主要终点（分别为 n = 64、25%；n = 52、21%）。直接癌症测量和总生存率分别是 20 项 (8%) 和 7 项 (3%) 研究的主要终点。以癌症测量为主要终点的八项试验 (40%) 达到了终点。针对结肠癌 (n = 1429)、乳腺癌 (n = 3088) 和前列腺癌 (n = 478) 的大型试验均显示饮食干预对测量癌症的终点没有影响。结论 大多数癌症饮食干预的随机对照试验规模较小，且测量的是非临床终点。尽管迄今为止只进行了少数大型随机对照试验，但这些试验并未显示出癌症结果的改善。目前，支持饮食干预作为癌症护理的治疗工具的证据有限。