Mendelian randomization and observational studies find no consistent association between Lp(a) and prostate cancer (Int J Mol Sci 2026)

Cancer is a major cause of morbidity worldwide. Prostate cancer (PCa) is one of the most commonly diagnosed malignancies in men and is clinically heterogeneous, highlighting the need for additional biomarkers that may improve risk stratification. Lipoprotein(a) [Lp(a)] is a genetically determined and increasingly available biomarker, but its relationship with PCa remains uncertain. This systematic review aimed to synthesize observational, genetic, and Mendelian randomization evidence on the associations of circulating Lp(a), LPA-related genetic variants, and genetically predicted Lp(a) with prostate cancer incidence, mortality, and clinicopathological features, and to assess the potential clinical implications of the available evidence. We conducted this review in accordance with PRISMA guidelines. PubMed, Scopus, and Web of Science databases were searched from inception to May 2026. Eligible studies were original, peer-reviewed studies in English that assessed circulating Lp(a), LPA-related genetic variants, or genetically predicted Lp(a) in relation to prostate cancer incidence, mortality, or clinicopathological features. Certainty of evidence was assessed using GRADE. Due to heterogeneity in study designs, exposure assessments, endpoints, and effect measures, results were synthesized narratively; no meta-analysis was performed. The search identified 489 records. After duplicate removal, 253 records were screened, 15 full-text reports were assessed, and 12 studies were included. These comprised prospective cohort or biobank-observational studies, observational studies, one genetic cohort study, and Mendelian randomization studies. Studies of circulating Lp(a) showed inconsistent findings for prostate cancer incidence and mortality. One cohort reported a positive association in a categorical analysis, whereas other studies found no association or a non-significant inverse association. The genetic cohort study found no robust association between LPAL2-LPA variants and prostate cancer incidence after correction for multiple testing. Mendelian randomization findings were mixed. Although several analyses yielded directionally positive or statistically significant estimates, several primary analyses were null or borderline, and estimates varied across methods and datasets. Overall, the Mendelian randomization evidence was of low certainty and does not establish a causal association between genetically predicted Lp(a) and PCa risk. Evidence for clinicopathological features was limited and inconsistent. Certainty of evidence was low for Mendelian randomization studies and very low for circulating Lp(a), genetic cohort evidence, and clinicopathological outcomes. Current evidence regarding Lp(a), LPA genetic variants, and genetically predicted Lp(a) in PCa remains heterogeneous and clinically inconclusive. Although some studies reported positive associations, the overall evidence linking Lp(a) to prostate cancer is inconsistent and of low certainty, and does not currently support a clear association.