Objective: To investigate the efficacy and safety of 36 000 IU recombinant human erythropoietin (rhEPO) in the treatment of cancer-related anemia (CRA) and to evaluate whether 36 000 IU rhEPO can serve as a rational "reduced-dose alternative" to the 40 000 IU rhEPO regimen. Methods: The multicenter, open-label, non-inferiority, randomized controlled trial was conducted from March 2023 to July 2024 across 12 hospitals in China, including Liaoning Cancer Hospital. A total of 119 patients with CRA were enrolled and randomly assigned to receive the 36 000 IU rhEPO (n=61) or 40 000 IU rhEPO (n=58). The primary efficacy endpoint was the change in hemoglobin (Hb) levels from baseline at weeks 9-13. Secondary efficacy endpoints included hematologic and other biochemical parameters, transfusion requirements, quality of life (QOL), which was assessed by QOL scores and Karnofsky performance status (KPS) scores, and overall survival. Safety was evaluated by the incidence of treatment-emergent adverse events (TEAEs). Results: The least-squares mean changes in Hb from baseline to weeks 9-13 were (12.9±2.3) g/L in the 36 000 IU group and (13.4±2.4) g/L in the 40 000 IU group. Analysis of covariance showed no statistically significant difference between groups (F=-0.21, P=0.836), with a between-group difference of (-0.5±2.5) g/L (95% CI: -5.4 g/L, 4.4 g/L). The lower limit of the 95% CI (-5.4 g/L) exceeded the predefined non-inferiority margin of -10 g/L, indicating non-inferiority of the 36 000 IU dose compared to the 40 000 IU. At week 13, the proportions of patients with Hb increase ≥10 g/L were 82.0% (50/61) in the 36 000 IU group and 86.2% (50/58) in the 40 000 IU group, with no significant difference between groups (Qmh=0.40, P=0.527). The average weekly transfusion rate was 2.0% in both groups. No significantly significant differences were observed between the two groups in terms of changes in hematocrit, reticulocyte percentage, folate, vitamin B12, albumin, iron metabolism markers, QOL scores, KPS scores, or overall survival (all P>0.05). Regarding safety, the incidence of TEAE was 80.3% (49/61) in the 36 000 IU group and 84.5% (49/58) in the 40 000 IU group, with nausea, fever, and fatigue being the most common symptoms (incidence>5%). No drug-related serious adverse events were reported, and there were no significant differences between the groups (P>0.05). Conclusions: The 36 000 IU dose of rhEPO is non-inferior to the 40 000 IU dose in terms of efficacy and has a favorable safety profile for the treatment of CRA. These findings support the use of 36 000 IU rhEPO as a reasonable clinical option for managing CRA. 目的: 探索36 000 IU重组人促红细胞生成素(rhEPO)治疗肿瘤相关性贫血(CRA)的有效性和安全性,评估36 000 IU rhEPO能否成为40 000 IU rhEPO的合理“减量替代”。 方法: 采用多中心、开放标签、非劣效、随机对照试验设计,于2023年3月至2024年7月在辽宁省肿瘤医院等中国12家医院开展。共纳入119例CRA受试者,根据rhEPO给药剂量规格随机分为36 000 IU组(n=61)和40 000 IU组(n=58)。主要疗效指标为第9~13周血红蛋白(Hb)水平较基线的变化值,次要疗效指标包括血细胞比容等血液生化指标、输血需求、生活质量[生活质量(QOL)评分和卡氏功能状态(KPS)评分]及总生存时间。安全性指标为治疗期不良事件(TEAE)发生率。 结果: 36 000 IU组和40 000 IU组受试者在第9~13周Hb较基线变化值的最小二乘均值分别为(12.9±2.3)g/L与(13.4±2.4)g/L,协方差分析显示组间差异无统计学意义(F=-0.21,P=0.836),组间差值为(-0.5±2.5)g/L(95% CI:-5.4~4.4 g/L),95% CI下限(-5.4 g/L)高于预设的非劣效界值(-10 g/L),表明36 000 IU组相对于40 000 IU组具有非劣效性。在第13周时,36 000 IU组和40 000 IU组Hb升高≥10 g/L的受试者比例分别为82.0%(50/61)和86.2%(50/58),差异无统计学意义(Qmh=0.40,P=0.527),且两组平均周输血率均为2.0%。两组受试者治疗前后的血细胞比容、网织红细胞百分比、叶酸、维生素B12、白蛋白、铁代谢指标、QOL评分、KPS评分以及总生存时间差异均无统计学意义(均P>0.05)。在安全性方面,36 000 IU组和40 000 IU组TEAE发生率分别为80.3%(49/61)与84.5%(49/58),以恶心、发热和乏力症状为主(发生比例高于5%),均未报告任何药物相关严重不良事件,组间差异无统计学意义(均P>0.05)。 结论: 36 000 IU rhEPO在治疗CRA方面对比40 000 IU rhEPO具有非劣效性,且安全性良好,36 000 IU剂量可作为临床治疗CRA的合理选择。.