This systematic review was conducted following the PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) [15] and was registered in PROSPERO. A comprehensive literature search was performed to identify studies evaluating oral PCSK9 inhibitors published up to September 12, 2025. Two independent reviewers screened multiple databases, including PubMed/MEDLINE, Scielo, Latindex, Lilacs, and the Cochrane Library using both Medical Subject Headings (MeSH) and relevant keywords, such as “oral PCSK9 inhibitors,” “AZD0780,” “Laroprovstat,” “MK-0616,” “Enlicitide,” “CVI-LM001,” “NNC0385-0434,” and “DC371739.” Boolean operators were applied to combine search terms: “PCSK9 inhibitors”[MeSH Terms] OR “AZD0780”[tiab] OR “Laroprovstat”[tiab] OR “MK-0616”[tiab] OR “Enlicitide”[tiab] OR “CVI-LM001”[tiab] OR “NNC0385-0434”[tiab] OR “DC371739”[tiab]. In Pubmed, the search was further refined by applying the built-in filter for Clinical Trials. In Scielo, Latindex, and LILACS, free-text searches were conducted using equivalent keywords in English, Spanish, and Portuguese, with manual screening to identify clinical trials. In the Cochrane Library, searches were performed in the CENTRAL database, which specifically indexes randomized and controlled trials. Gray literature was also examined, including clinicaltrials.gov, conference presentations, and information available on the respective laboratories’ websites. Additionally, a snowballing approach was employed to identify further relevant studies. Only human studies were considered, and no language restrictions were applied during the search.

Eligible studies were experimental human trials, specifically randomized clinical trials assessing the lipid-lowering efficacy and safety of oral PCSK9 inhibitors compared with placebo. Excluded studies comprised expert opinions, narrative or systematic reviews, and observational designs.

The percentage change in lipid parameters and the frequency of clinically relevant adverse events were determined. The lipid markers assessed in this study were LDL-C, non–high-density lipoprotein cholesterol (non-HDL-C), apolipoprotein B (ApoB), triglycerides, and lipoprotein(a) [Lp(a)]. Safety was evaluated by analyzing all reported adverse events (AEs). Serious adverse events (SAEs) were also considered, defined as any AE resulting in death, being life-threatening, requiring hospitalization or prolonging existing hospitalization, causing persistent or significant disability, or considered medically important by the investigator.

Potential risks of bias were assessed using the Cochrane RoB-2 tool [16], which evaluates five domains: bias arising from the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Each domain was rated as “low risk,” “some concerns,” or “high risk,” and an overall risk rating was assigned to each study. Two reviewers independently conducted the assessment, and discrepancies were resolved by consensus.

Methodological quality was evaluated in parallel to the risk of bias assessment to provide an overall judgement of study robustness. As no additional standardized quality tool was applied, overall study quality was inferred from the RoB-2 results: studies with low risk of bias across domains were considered high-quality, whereas those with “some concerns” or “high risk” in one or more domains were classified as moderate or low quality, respectively. This approach ensured consistency in the interpretation of both bias and overall methodological soundness.

Initially, a qualitative synthesis was performed for all studies that met the predefined inclusion and exclusion criteria. Subsequently, a quantitative synthesis (meta-analysis) was conducted on studies reporting sufficient data to enable statistical pooling. Regarding the doses evaluated in the meta-analysis, for NNC0385-0434 and AZD0780, the maximum reported doses were included in the quantitative lipid analysis, since the optimal dosing for upcoming advanced-phase studies has not yet been established and both compounds demonstrated an acceptable safety profile. For MK-0616, an 18-mg dose was selected in the study by Ballantyne et al. [8], which closely corresponds to the regimen used in later-phase trials (20 mg; CorReef program), while a 20-mg dose was selected in the study conducted by Johns et al. [14]. For each study, the percentage change in lipid parameters and the frequency of clinically relevant AEs were determined. Effect sizes were reported as mean differences (MDs) or risk ratios [along with their corresponding 95% confidence intervals (95% CIs)]. If the data were presented using measures other than mean and standard deviation, they were converted using approaches previously described in the literature [17]. Additionally, the I2 statistic was computed to assess heterogeneity and inconsistency across studies. A fixed- or random-effects model was used depending on the observed heterogeneity. To compare the mean effects between subgroups, a Z-test was employed. Statistical significance was defined as a p value of 0.05 (two-tailed). The analysis was performed using R statistical software (v. 3.5.1) [18].

Sensitivity analyses were performed only for lipid markers that were reported in three or more studies. Since triglycerides and non–HDL-C were each available in only two studies, excluding one study for the sensitivity analysis would have left a single study; therefore, these markers were not included in this analysis.

No formal assessment of publication bias was performed due to the limited number of included studies.

This article is based on previously conducted studies and does not contain any studies with human participants or animals performed by any of the authors.