Jae Eun Ahn, Kevin R Sweeney, Candace Bramson, Caroline Bezzubik, Phillip Saccone, David A DeMicco, Jing Liu
Pfizer, Inc
Objectives: Model-based meta-analysis (MBMA) was conducted to quantify the changes in lipid endpoints including triglycerides (TG) achieved by interventions in patients with hypertriglyceridemia.
Methods: A literature search was performed according to the following PICOS criteria [1]. Study population (P) was hypertriglyceridemia with baseline TG ≥500 mg/dL (or reported baseline range included 500 mg/dL or greater). Interventions (I) included but were not limited to fibrates, statins, omega-3 fatty acids (including fish oil and icosapent ethyl), niacin, or any treatment evaluated for hypertriglyceridemia. Placebo, active control, or lifestyle modification (typically diet and exercise) served as comparators (C). The outcomes (O) of interest included TG, non-HDL cholesterol, LDL cholesterol, and apolipoprotein B. Placebo and active controlled studies (S) were included. Observational studies, and studies with patients with HIV, familial chylomicronemia syndrome, or renal failure, and whose publication date was prior to 1990 were excluded. The searched databases included OVID MEDLINE®, OVID MEDLINE® In-process and Epub Ahead of Print, BIOSIS Previews, Embase, Scopus®, and ClinicalTrials.gov (https://clinicaltrials.gov/) up to August 24, 2020. Each publication returned from the literature search was reviewed by two reviewers and only articles that met PICOS eligibility were digitized into a MBMA data file. R 3.6.1 (https://www.R-project.org/) was used for data cleaning, modification, and plotting; NONMEM 7.5 (ICON Development Solution, Gaithersburg, MD) was used for estimating the average TG % change from baseline measures (% CFB, as reported or imputed using the reported baseline and change from baseline measures) by treatment group at the primary analysis time points. Each observation was weighted according to the sample size in each arm. The effects of baseline TG were explored graphically. Other than omega-3, different dosages of each treatment were not differentiated but pooled together in current analysis.
Results: A total of 165 TG % CFB at primary time points from 48 studies, representing 11821 patients, were included in the analysis. The median TG baseline (assessed at arm level) in the analysis data set was 405 mg/dL ranging from 163 to 1095 mg/dL (11 of them were missing). The median primary time point was 12 weeks (from 4 to 157 weeks). The average placebo-corrected % CFB (% relative standard error, RSE) on TG reduction for fibrates, omega-3 high dose (4 g/day), statin, lifestyle, and omega-3 lower doses (<4 g/day) were estimated to be 45.1% (3.41), 29.7% (7.47), 26.6% (7.73), 25.7% (26.9), and 22.6% (7.99), respectively. Placebo effects, estimated only from the placebo-controlled studies, were negligible and very uncertain (1.91% reduction with 86.7% RSE). Most studies did not report longitudinal profiles of the endpoints and the effects of study duration on the placebo or treatment effects were not evaluated. There was no apparent association between the baseline TG and % CFB in TG. Due to heterogeneity and limitation in the literature data, the assessment of covariate effects such as statin background therapy or baseline effects were not tested in the model but explored graphically (results not shown).
Conclusions: A MBMA was performed in patients with hypertriglyceridemia and the mean treatment effects were quantified to be 22.6% to 45.1% for commonly used TG lowering treatment classes.
References:
[1] Liberati et al, PLoS Med 6: e1000100 (2009)
Reference: PAGE 29 (2021) Abstr 9853 [www.page-meeting.org/?abstract=9853]
Poster: Drug/Disease Modelling - Endocrine