疲劳是放射治疗（RT）患者常见的问题，会显著影响生活质量。褪黑素作为一种安全、廉价且天然的补充剂，可能改善症状并减轻RT的副作用。本研究旨在进行一项随机双盲安慰剂对照的第三期试验，评估褪黑素对接受乳腺癌放疗的疲劳和其他症状的影响。对象为乳腺癌早期或导管原位癌患者，年龄≥18岁，Eastern Cooperative Oncology Group（ECOG）表现状态<3，血红蛋白≥9g/dL，计划进行有治愈意图的门诊放疗治疗者。以褪黑素20mg和安慰剂为1：1随机分组，并从放疗开始前夜开始口服，持续到放疗后两周。根据治疗持续时间（<3周，≥3周）和先前化疗情况分层随机分组。主要终点为慢性疾病治疗功能评估-疲劳量表（FACIT-Fatigue scale），次要终点为FACIT-F亚量表，埃德蒙顿症状评估量表（ESAS）和患者报告结果测量信息系统（PROMIS）得分，分别在基线、放疗后2周和8周测定。使用双侧ANOVA F检验，显著性水平设为4.5%进行主要终点分析。次要分析使用双侧ANOVA F检验，显著性水平设为5%。计划招募约140名患者，并计划在招募中期进行中期分析。共有85名患者接受了符合条件的筛选，其中79名患者随机分组，其中40名接受褪黑素，39名接受安慰剂；78名患者在中期分析时纳入统计。两组患者的年龄、种族和ECOG表现状态等基线特征相似。采用纵向混合效应模型以治疗与时间（治疗×时间）的效应作为主要结果参数分析治疗效应。褪黑素组与安慰剂组相比，FACIT-Fatigue的治疗×时间未显示统计学意义（P值为.83）。此外，FACIT身体、社交、情绪和功能福祉得分的次要分析也未显示统计学意义（分别为P值.35、.06、.62和.71）。根据患者报告的次要结果PROMIS总分在时间上也未显示统计学显著改变（P值为.34）。对于每个单独项目进行的次要分析ESAS也未显示统计学显著性，焦虑（P = .56），福祉（.82），嗜睡（.83），食欲不振（.35），恶心（.79），疼痛（.50），呼吸急促（.77），睡眠（.45）和疲劳（.56）。抑郁是唯一显示统计学显著性的项目，安慰剂组下降了0.01单位，但这种变化不被视为临床上显著的。褪黑素耐受性良好，未报告3或4级不良事件。最常见的副作用是头痛、嗜睡和腹痛。在本研究中，没有患者因参与该研究而死亡。两名患者在研究完成后一年内因乳腺癌复发而死亡。在各种原因（包括不良事件、与研究药物无关的住院治疗、放疗中断和COVID-19预防措施）导致的研究中途退出的患者共有16名。在这项双盲安慰剂对照的第三期试验中，褪黑素对接受乳腺癌早期放疗的疲劳和其他症状并没有预防或显著改善的作用。分析结果显示在中期招募时几乎没有证据支持其效果，从而有望提前终止试验。©作者2023年发表于牛津大学出版社。

Fatigue is common in patients undergoing radiotherapy (RT) and can significantly impact quality of life. Melatonin, a safe inexpensive natural supplement, may improve symptoms and attenuate the side effects of RT. The purpose of this randomized double-blind placebo-controlled phase III trial was to assess the effects of melatonin for preventing fatigue and other symptoms in patients with breast cancer undergoing RT.Female early stage or Ductal carcinoma in situ patients with breast cancer ≥18 years of age with Eastern Cooperative Oncology Group (ECOG) performance status <3, hemoglobin ≥9 g/dL, planned for outpatient RT treatment with curative intent, were randomized 1:1 to melatonin 20 mg or placebo, orally, starting the night before RT initiation until 2 weeks post-RT. Randomization was stratified according to treatment duration (<3 weeks, ≥3 weeks) and prior chemotherapy. The primary endpoint was the Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-Fatigue scale), and secondary endpoints were FACIT-F subscales, Edmonton Symptom Assessment Scale (ESAS), and Patient-Reported Outcomes Measurement Information System (PROMIS) scores obtained at baseline, and 2 and 8 weeks post-RT. A 2-sided ANOVA F-test at a 4.5% significance level for the primary endpoint was used. Secondary analyses were reported using an F-test at a 5% significance level. The goal was to recruit approximately 140 patients with interim analysis planned mid-recruitment.Eighty-five patients were screened for eligibility; 79 patients were randomized: 40 to melatonin and 39 to placebo; 78 patients were treated and included in the interim analysis at the mid-recruitment point. Baseline patient characteristics of age, race, and ECOG performance status were similar in both arms. The treatment effect was studied using a longitudinal mixed effects model with the effect of treatment over time (treatment × time) as the primary outcome parameter. The treatment × time for FACIT-Fatigue did not demonstrate statistical significance (P-value .83) in the melatonin group compared to placebo. In addition, secondary analyses of FACIT physical, social, emotional, and functional well-being scores did not demonstrate statistical significance (P-values of .35, .06, .62, and .71, respectively). Total PROMIS scores, collected as secondary outcome reported by patients, did not demonstrate statistically significant change over time either (P-value is .34). The other secondary scale, ESAS, was analyzed for each individual item and found to be nonsignificant, anxiety (P = .56), well-being (.82), drowsiness (.83), lack of appetite (.35), nausea (.79), pain (.50), shortness of breath (.77), sleep (.45), and tiredness (.56). Depression was the only item demonstrating statistical significance with a decrease of 0.01 unit in the placebo group, a change not considered clinically significant. Melatonin was well-tolerated with no grade 3 or 4 adverse events reported. The most common side effects were headache, somnolence, and abdominal pain. No patients died while participating in this study. Two patients died within a year of study completion from breast cancer recurrence. Sixteen patients withdrew prior to study completion for various reasons including adverse events, hospitalizations unrelated to study drug, RT discontinuation, and COVID-19 precautions.In this double-blind placebo-controlled phase III trial, melatonin did not prevent or significantly improve fatigue and other symptoms in patients with early stage breast cancer undergoing RT. The analysis, showing little evidence of an effect, at mid-recruitment, assured early termination of the trial.© The Author(s) 2023. Published by Oxford University Press.