BACKGROUND Intravenous daratumumab for treatment of patients with multiple myeloma involves a lengthy infusion that affects quality of life, and infusion-related reactions are common. Subcutaneous daratumumab is thought to be easier to administer and to cause fewer administration-related reactions. In this study (COLUMBA), we tested the non-inferiority of subcutaneous daratumumab to intravenous daratumumab. METHODS In this ongoing, multicentre (147 sites in 18 countries), open-label, non-inferiority, randomised, phase 3 trial, we recruited adult patients (age ≥18 years) if they had confirmed relapsed or refractory multiple myeloma according to International Myeloma Working Group criteria; received at least three previous lines of therapy, including a proteasome inhibitor and immunomodulatory drug, or were double refractory to both a proteasome inhibitor and immunomodulatory drug; and had an Eastern Cooperative Oncology Group performance status score of 2 or lower. Patients were randomly assigned (1:1) by a computer-generated randomisation schedule and balanced using randomly permuted blocks to receive daratumumab subcutaneously (subcutaneous group) or intravenously (intravenous group). Randomisation was stratified on the basis of baseline bodyweight (≤65 kg, 66-85 kg, >85 kg), previous therapy lines (≤four vs >four), and myeloma type (IgG vs non-IgG). Patients received 1800 mg of subcutaneous daratumumab co-formulated with 2000 U/mL recombinant human hyaluronidase PH20 or 16 mg/kg of intravenous daratumumab once weekly (cycles 1-2), every 2 weeks (cycles 3-6), and every 4 weeks thereafter (28-day cycles) until progressive disease or toxicity. The co-primary endpoints were overall response and maximum trough concentration (Ctrough; cycle 3, day 1 pre-dose). The non-inferiority margin for overall response was defined using a 60% retention of the lower bound (20·8%) of the 95% CI of the SIRIUS trial. Efficacy analyses were done by intention-to-treat population. The pharmacokinetic-evaluable population included all patients who received all eight weekly daratumumab doses in cycles 1 and 2 and provided a pre-dose pharmacokinetics blood sample on day 1 of cycle 3. The safety population included all patients who received at least one daratumumab dose. This trial is registered with ClinicalTrials.gov, NCT03277105. FINDINGS Between Oct 31, 2017, and Dec 27, 2018, 655 patients were screened, of whom 522 were recruited and randomly assigned (subcutaneous group n=263; intravenous group n=259). Three patients in the subcutaneous group and one in the intravenous group did not receive treatment and were not evaluable for safety. At a median follow-up of 7·5 months (IQR 6·5-9·3), overall response and Ctrough met the predefined non-inferiority criteria. An overall response was seen in 108 (41%) of 263 patients in the subcutaneous group and 96 (37%) of 259 in the intravenous group (relative risk 1·11, 95% CI 0·89-1·37). The geometric means ratio for Ctrough was 107·93% (90% CI 95·74-121·67), and the maximum Ctrough was 593 μg/mL (SD 306) in the subcutaneous group and 522 μg/mL (226) in the intravenous group. The most common grade 3 and 4 adverse events were anaemia (34 [13%] of 260 patients evaluable for safety in the subcutaneous group and 36 [14%] of 258 patients in the intravenous group), neutropenia (34 [13%] and 20 [8%]), and thrombocytopenia (36 [14%] and 35 [14%]). Pneumonia was the only serious adverse event in more than 2% of patients (seven [3%] in the subcutaneous group and 11 [4%] in the intravenous group). There was one death resulting from a treatment-related adverse event in the subcutaneous daratumumab group (febrile neutropenia) and four in the intravenous group (sepsis [n=2], hepatitis B reactivation [n=1], and Pneumocystis jirovecii pneumonia [n=1]). INTERPRETATION Subcutaneous daratumumab was non-inferior to intravenous daratumumab in terms of efficacy and pharmacokinetics and had an improved safety profile in patients with relapsed or refractory multiple myeloma. These data could contribute to the approval of the subcutaneous daratumumab formulation by regulatory bodies. FUNDING Janssen Research & Development.

中文翻译：

---

复发性或难治性多发性骨髓瘤（COLUMBA）患者的皮下注射vs静脉注射daratumumab：一项多中心，开放标签，非劣效，随机，3期试验。

背景技术用于治疗多发性骨髓瘤的患者的静脉内达拉他珠单抗涉及长效输注，这影响生活质量，并且与输注相关的反应是常见的。皮下达拉妥单抗被认为更易于给药并且引起较少的与给药相关的反应。在这项研究（COLUMBA）中，我们测试了皮下daratumumab对静脉注射daratumumab的非劣效性。方法在这项正在进行的，多中心（18个国家/地区的147个地点），开放标签，非劣效，随机，3期临床试验中，我们根据国际标准招募了已确诊复发或难治性多发性骨髓瘤的成年患者（≥18岁）。骨髓瘤工作组标准；至少接受了之前的三种疗法，包括蛋白酶体抑制剂和免疫调节药物，或对蛋白酶体抑制剂和免疫调节药物均双重耐药；并且东部合作肿瘤小组的表现状态得分为2或更低。通过计算机生成的随机分组表对患者进行随机分配（1：1），并使用随机排列的区块进行平衡，以通过皮下注射（皮下注射组）或静脉注射（静脉注射组）接受daratumumab。根据基线体重（≤65kg，66-85 kg，> 85 kg），既往治疗线（≤4对> 4）和骨髓瘤类型（IgG对非IgG）进行分层。患者每周两次（第1-2个周期），每2周（第1-2个周期）接受1800 mg皮下注射daratumumab与2000 U / mL重组人透明质酸酶PH20或16 mg / kg静脉注射daratumumab联合配制，然后此后每4周一次（28天周期），直到疾病或毒性逐渐恶化。共同主要终点是总体反应和最大谷浓度（Ctrough；第3周期，给药前1天）。使用SIRIUS试验的95％CI的下限保留60％（20·8％）来定义总体反应的非劣效性。疗效分析是按意向治疗人群进行的。药代动力学可评估的人群包括所有在第1和第2周期每周接受所有八次达拉妥单抗剂量并在第3轮第1天提供给药前药代动力学血样的患者。安全人群包括所有接受至少一次daratumumab剂量的患者。该试验已在ClinicalTrials.gov（NCT03277105）上注册。结果在2017年10月31日至2018年12月27日之间，对655例患者进行了筛查，其中522名被招募并随机分配（皮下注射组n = 263；静脉注射组n = 259）。皮下组中有3例患者，静脉内组中有1例患者未接受治疗，安全性无法评估。中位随访7·5个月（IQR 6·5-9·3），总体反应和Ctrough符合预定的非自卑标准。皮下注射组263例患者中有108例（41％），静脉注射组259例中96例（37％）有总体反应（相对危险度1·11，95％CI 0·89-1·37）。Ctrough的几何均数比为107·93％（90％CI 95·74-121·67），皮下注射组的最大Ctrough为593μg/ mL（SD 306），皮下注射组的最大Ctrough为522μg/ mL（226）。静脉注射组。最常见的3级和4级不良事件是贫血（皮下注射组中可评估安全性的260名患者中的34 [13％]，静脉注射组中258名患者中的36 [14％]），中性粒细胞减少（34 [13％]和20 [8％]）和血小板减少症（36 [14％]和35 [14％]）。在超过2％的患者中，肺炎是唯一的严重不良事件（皮下注射组为7 [3％]，静脉注射组为11 [4％]）。皮下注射daratumumab组（发热性中性粒细胞减少症）与治疗相关的不良事件导致死亡，静脉注射组（败血症[n = 2]，乙型肝炎再激活[n = 1]和吉氏肺炎肺炎[n]造成4例死亡）。 = 1]）。解释在疗效和药代动力学方面，皮下达拉妥单抗不次于静脉注射达拉妥单抗，在复发或难治性多发性骨髓瘤患者中安全性得到改善。这些数据可能有助于监管机构批准皮下达拉妥单抗制剂。资金Janssen研究与开发。