Comment on “Lipoprotein(a) reduction with inclisiran, alirocumab, evolocumab, enlicitide, and lerodalcibep: A systematic review and meta-analysis of randomized controlled trials” Authors: Vatsala Gupta, Parag J Ratnakar, Kotina Shridevi, Gaurav, Jeffrin Reneus Paul Published: 2026 Dear Editor, Mulligan et al. synthesized 31 randomized trials and concluded that PCSK9-targeting agents provide comparable lipoprotein(a) [Lp(a)] lowering, permitting selection according to dosing, access, cost, and patient preference [1]. The pooled estimate is informative, but the comparative conclusion requires clarification because the reported meta-regression contrasts are internally inconsistent and because non-significance in an indirect analysis does not establish equivalence. First, the inclisiran-versus-evolocumab contrast is +4.9% (95% CI -2.31 to 12.16; P = .18) in the abstract, but +10.64% with the same confidence interval and P value in the Results [1]. Also reports an SE of 6.59 and z = 1.33. These values cannot all arise from one Wald test: +10.64/6.59 gives z = 1.61 and an approximate 95% CI of −2.28 to 23.56. For inclisiran versus alirocumab, the point estimate (+1.59%) lies outside the interval printed in the Results (−8.07 to 1.40), whereas the abstract reports the internally coherent interval −5.38 to 8.55. The alirocumab-versus-evolocumab interval also loses its negative lower bound. A corrected contrast table generated from one stated model, with covariance-based standard errors for every pairwise contrast, would clarify which estimates are authoritative. Second, "no statistically significant difference" should not be translated into therapeutic equivalence. A superiority test with P > .05 indicates insufficient evidence of a difference, not evidence that clinically important differences are absent [2]. Equivalence requires a prespecified clinically acceptable margin and confidence intervals that exclude effects beyond that margin [3]. This distinction is especially important because residual heterogeneity remained high (I2 = 92.41%), and evidence for lerodalcibep and enlicitide comprised only two and one trials, respectively [1]. Moreover, these were study-level meta-regression contrasts rather than randomized head-to-head comparisons. Meta-regression associations are observational and can reflect differences in populations, baseline Lp(a), follow-up, assays, background therapy, or trial design [4]. We therefore suggest reframing the conclusion as "no between-agent difference was detected in an indirect study-level comparison," pending correction of the contrast statistics. Reporting 95% prediction intervals and a sensitivity analysis excluding trials with imputed standard errors would further show how much uncertainty remains across settings. These revisions would preserve the useful pooled estimate while avoiding an equivalence claim that the current analysis does not establish. CRediT authorship contribution statement Vatsala Gupta: Funding acquisition, Investigation, Methodology. Parag J. Ratnakar: Project administration, Resources, Software. Kotina Shridevi: Supervision, Validation. Gaurav: Visualization, Writing – original draft, Writing – review & editing. Jeffrin Reneus Paul: Conceptualization, Data curation, Formal analysis. Author agreement We, the undersigned authors, confirm that we have read and approved the final version of the above manuscript and agree to its submission for consideration for publication. Each author has made a substantial contribution to the conception, critical appraisal, drafting, or revision of the manuscript; has approved the version to be submitted; and agrees to be accountable for the accuracy, integrity, and scientific content of the work. We confirm that the manuscript is original, has not been published previously, and is not currently under consideration by another journal. All relevant sources have been appropriately cited. The authorship list and order have been reviewed and approved by all authors, and no individual who meets the applicable authorship criteria has been omitted. All authors confirm that any funding, conflicts of interest, acknowledgments, use of artificial intelligence, and other required disclosures have been reported accurately. The corresponding author is authorised to communicate with the journal, submit revisions, respond to editorial queries, and complete the publication process on behalf of all authors. Contributor Information Vatsala Gupta, Email: vatsala.gupta@sharda.ac.in. Parag J. Ratnakar, Email: parag.ratnakar@dpu.edu.in. Kotina Shridevi, Email: drshridevi80@gmail.com. Gaurav, Email: gaurav@gehu.ac.in. Jeffrin Reneus Paul, Email: 111701051.smc@saveetha.com. References