摘要目的:在淋巴管肌瘤病（LAM）患者中，评价更低谷浓度的西罗莫司治疗是否不劣于常规目标浓度（≥5 μg/L）的肺功能保护作用。方法:回顾性研究基于2017年4月至2022年4月在北京协和医院LAM队列注册的183例患者数据，构建分析数据集，根据时间加权的西罗莫司平均谷浓度划分为常规浓度组（≥5 μg/L）和低浓度组（<5 μg/L），其中西罗莫司常规浓度LAM患者103例（平均年龄40.1岁），低浓度组患者80例（平均年龄38.8岁），均为女性患者。对第一秒用力呼气量（FEV 1）较基线变化量（简称FEV 1变化量）进行混合效应模型建模，计算西罗莫司治疗后不同时间点的边际均值FEV 1变化量。主要比较结局为西罗莫司治疗1年时FEV 1变化量，若该变化量差异无统计学意义，则比较西罗莫司治疗3年时FEV 1变化量。同一时间点的组间FEV 1变化量估计边际均值比较采用Welch- t检验或Welch-ANOVA检验。 结果:西罗莫司治疗1年时低浓度组的FEV 1变化量的估计边际均值为45.2（95% CI：13.9~76.5）ml，与常规浓度组（估计边际均值33.7 ml，95% CI：6.6~60.8 ml）相比差异无统计学意义（ t=0.55， P=0.585）。治疗3年时，两个浓度组间的FEV 1变化量估计边际均值差异仍无统计学意义（低浓度组-24.9 ml，95% CI：-115.7~65.9 ml；常规浓度组-4.1 ml，95% CI：-78.9~70.8 ml； t=0.35， P=0.728）。 结论:低浓度的西罗莫司治疗对LAM患者FEV 1保护作用与常规目标浓度类似，西罗莫司治疗LAM的目标浓度或可进一步下调。

abstractsObjective:To determine whether sirolimus therapy at lower trough concentrations is non-inferior to the conventional target concentration (≥5 μg/L) with respect to pulmonary protective effects and lung function preservation in patients with lymphangioleiomyomatosis (LAM).Methods:A retrospective analytical dataset was constructed using patient-level data from the Peking Union Medical College Hospital LAM cohort registry between April 2017 and April 2022. Patients were stratified into a conventional-concentration group (≥5 μg/L) and a low-concentration group (<5 μg/L) according to the time-weighted average sirolimus trough concentration. The conventional concentration group comprised 103 LAM patients (mean age 40.1 years), and the low concentration group included 80 patients (mean age 38.8 years). All patients were female. Changes in FEV 1 were analyzed with a linear mixed-effects model, and marginal mean FEV 1 changes from baseline were estimated at prespecified time points after treatment initiation. The primary endpoint was the change in FEV 1 after 1 year of sirolimus therapy. If no statistically significant difference was observed at one year, the change at three years was compared. The estimated marginal means of FEV1 change between groups at the same time point were compared using Welch's t-test or Welch's analysis of variance (ANOVA). Results:After one year, the estimated marginal mean FEV 1 increase was 45.2 ml (95% CI: 13.9-76.5) in the low-concentration group versus 33.7 (95% CI: 6.6-60.8) ml in the conventional-concentration group ( t=0.55, P=0.585). At three years, the between-group difference in the estimated marginal mean change in FEV 1 from baseline remained statistically non-significant (low-concentration group:-24.9 ml, 95% CI:-115.7-65.9 ml; conventional-concentration group:-4.1 ml, 95% CI:-78.9-70.8 ml; t=0.35, P=0.728). Conclusions:Maintaining sirolimus at lower trough concentrations (median 4.1 μg/L, 3.5-4.5 μg/L) provides similar FEV? preservation to that achieved with the conventional target concentration. This suggested that the therapeutic target for sirolimus in LAM could be safely reduced.