Effects of metreleptin in patients with lipodystrophy with and without baseline concomitant medication use

To evaluate the effects of metreleptin in distinct subgroups of patients with generalized lipodystrophy (GL) and partial lipodystrophy (PL), using multivariate linear regression modeling to account for the role of patients’ baseline usage of concomitant glucose and lipid-lowering medications and other covariates on their outcomes. A <i>post-hoc</i> statistical analysis of two published single-arm, interventional, phase 2 clinical trials at NIH was conducted. Concomitant medication use was assessed for the clinical trial population using prescription fill data, measured at baseline and the post-one year following metreleptin initiation. Pre-specified co-primary efficacy endpoints measured were change from baseline in HbA1c at month 12, and the percent change from baseline in fasting serum triglycerides (TG) at month 12. Descriptive and statistical analyses were conducted for the overall population, the separate populations with GL and PL, and additional PL subgroups defined by baseline metabolic markers of elevated HbA1c and elevated fasting TG. As previously reported, improvement in HbA1c and fasting TG from baseline to 12 months on metreleptin were observed in the overall population (mean change −1.57 percentage points and median change −37.9%, respectively) and subgroups. For both HbA1c and TG, baseline levels were significant predictors of changes after metreleptin. After considering baseline characteristics such as disease type, age, sex, and baseline HbA1c, baseline insulin use was not found to be a significant predictor of HbA1c improvement following metreleptin initiation. Similar results were seen for TG levels, with the use of any lipid-lowering medications at baseline not found to be a significant predictor of reductions in fasting TG levels. Patients treated with metreleptin experienced statistically significant improvement in metabolic markers of glycemic and hypertriglyceridemic control—e.g. HbA1c and triglyceride levels—across various subgroups after controlling for baseline characteristics and concomitant medication usage.

为评估美曲普汀（metreleptin）对全身性脂肪营养不良（generalized lipodystrophy, GL）与部分性脂肪营养不良（partial lipodystrophy, PL）不同亚组患者的治疗效果，本研究采用多元线性回归模型，以校正患者基线时伴随使用的降糖、调脂药物及其他协变量对结局指标的影响。本研究针对美国国立卫生研究院（National Institutes of Health, NIH）开展的两项已发表单臂干预性Ⅱ期临床试验开展了事后（post-hoc）统计分析。研究通过处方配药数据评估临床试验人群的伴随用药情况，数据采集时点为基线期及美曲普汀启动治疗后1年。预设的共同主要疗效终点为：第12个月时糖化血红蛋白（HbA1c）较基线的变化值，以及第12个月时空腹血清甘油三酯（triglycerides, TG）较基线的百分比变化。研究对整体受试人群、GL与PL单独人群，以及由基线高HbA1c和高空腹TG代谢标志物定义的额外PL亚组分别开展了描述性分析与统计分析。如既往报道所示，在整体人群及各亚组中，接受美曲普汀治疗12个月后，患者的HbA1c与空腹TG水平较基线均获得改善（平均变化值分别为-1.57个百分点，中位变化值分别为-37.9%）。对于HbA1c与TG而言，基线水平均为美曲普汀治疗后结局变化的显著预测因子。在校正疾病类型、年龄、性别及基线HbA1c等基线特征后，基线胰岛素使用并未被证实为美曲普汀启动治疗后HbA1c改善的显著预测因子。针对TG水平也得到了相似的结果：基线时使用任何调脂药物并未被证实为空腹TG水平降低的显著预测因子。在校正基线特征与伴随用药情况后，接受美曲普汀治疗的患者在多个亚组中，其血糖控制与高甘油三酯血症控制的代谢标志物（如HbA1c与TG水平）均出现了具有统计学意义的改善。