Two-Sample Network Mendelian Randomization and Single-Cell Analysis Reveal the Causal Associations and Underlying Mechanisms Between Antihypertensive Drugs and Kidney Cancer

Abstract

Background: Antihypertensive drugs represent the most widely used drugs worldwide. However, the association between antihypertensive drugs and the risk of kidney cancer remains unclear. This study innovatively integrates multi-omics and causal inference approaches to investigate the long-term effects and potential mechanisms of 12 antihypertensive drug classes on kidney cancer risk. Methods: In this study, novel approaches including two-sample mendelian randomization (MR), summary-data-based mendelian randomization (SMR), two-step network MR, and single-cell transcriptomic analysis were employed. Single nucleotide polymorphisms (SNPs) were obtained from genome-wide association studies (GWASs) to proxy exposures and outcomes. The cis-expression quantitative trait loci (cis-eQTL) as the proxies of exposure were also obtained. MR estimates were generated using the inverse-variance weighted method or Wald ratio method. Sensitivity analyses were undertaken to interrogate the robustness of the main findings. Two-step network MR and single-cell analysis were specifically designed to dissect pathway-level mediation and expression patterns of identified targets. Results: In the main analysis, genetically proxied calcium-channel blockers (odds ratio [OR]: 0.95, 95% confidence interval [CI]: 0.91-0.99, p=0.021) and vasodilator antihypertensives (OR: 0.86, 95% CI: 0.76-0.97, p=0.018) were suggestively associated with decreased risk of kidney cancer, whereas genetically proxied angiotensin-converting enzyme inhibitors (OR: 1.13, 95% CI: 1.00-1.27, p=0.043) was suggestively associated with increased risk of kidney cancer. Genetically proxied antiadrenergic agents (OR=0.94, 95% CI: 0.90-0.99, p=0.021) and centrally acting antihypertensives (OR=0.93, 95% CI: 0.88-0.98, p=0.010) were suggestively associated with a decreased risk of clear cell renal cell carcinoma. SMR analysis revealed that these suggestively significant associations might be driven by CACNA1C, CALM1, ACE, and LTA4H. Upon two-step network MR analyses, 10 pathways with directional consistency were identified, and the mediation proportion ranged from 3.22% to 7.12%. The influence of antihypertensive drugs on kidney cancer risk might be associated with their regulation of levels of blood cells and lipids. Single-cell analysis further revealed the expression patterns of the four identified targets in peripheral blood and tumor infiltrating immune cells. Conclusion: This study pioneers the integration of causal inference and single-cell omics to demonstrate that antihypertensive drugs modulate kidney cancer risk through target-specific mechanisms involving blood cell and lipid pathways. Our findings provide actionable targets (CACNA1C, CALM1, ACE, and LTA4H) for drug repurposing trials and underscore the clinical importance of personalized antihypertensive therapy in cancer prevention.

Keywords: Mendelian randomization; antihypertensive drugs; drug target; kidney cancer; single-cell analysis.

Figure 1

Study design. Abbreviations: CAD, coronary artery disease; ccRCC, clear cell renal cell carcinoma; chRCC, chromophobe renal cell carcinoma; CI, confidence interval; GWASs, genome-wide association studies; IV, instrumental variable; IVW, inverse-variance weighted; MR, mendelian randomization; pRCC, papillary renal cell carcinoma; SNP, single nucleotide polymorphisms; SMR, Summer-data-based Mendelian Randomization.

Figure 2

MR association of genetically proxies for antihypertensive drugs with (A) the risk of hypertension, and (B) CAD. Abbreviations: CAD, coronary artery disease; MR, Mendelian randomization. For the specific drug type, if the analysis result was suggestively significant, the name of this specific drug type would be presented in the form of red font. And if the analysis result was strongly significant, the name of this specific drug type would be presented in the form of red bold font.

Figure 3

MR association of genetically proxies for antihypertensive drugs with (A) the risk of kidney cancer, (B) ccRCC, (C) pRCC, and (D) chRCC. Abbreviations: ccRCC, clear cell renal cell carcinoma; chRCC, chromophobe renal cell carcinoma; MR, Mendelian randomization; pRCC, papillary renal cell carcinoma. For the specific drug type, if the analysis result was suggestively significant, the name of this specific drug type would be presented in the form of red font. And if the analysis result was strongly significant, the name of this specific drug type would be presented in the form of red bold font.

Figure 4

SMR analysis of antihypertensive drug targets and kidney cancer risk. Abbreviations: SMR, summary-data-based Mendelian randomization; HEIDI, heterogeneity in dependent instruments.

Figure 5

Two-step network mediation analysis connecting genetically proxies for antihypertensive drugs to kidney cancer through potential mediators. (A) The overview of pathways linking antihypertensive drugs to kidney cancer; (B) The proportion of association between genetically proxies for antihypertensive drugs and kidney cancer mediated by potential mediators. The bar chart is labeled as "mediator_ outcome (mediating proportion)". Abbreviations: ACEI, Angiotensin converting enzyme inhibitors; AG, Antiadrenergic agents; CAA, centrally acting antihypertensives; ccRCC, clear cell renal cell carcinoma; HDL, High density lipoprotein; IDL, Intermediate density lipoprotein; LDL, Low density lipoprotein.

Figure 6

Single-cell analysis revealed the expression patterns of four identified antihypertensive drug targets causally associated with kidney cancer. (A) UMAP plot of the identified cell clusters in PBIC and tumor from renal cell carcinoma patients. (B) The composition of each cell type. (C) Heatmap distribution of marker genes in each cell type. (D) Bubble plot of the average and percent expression of marker genes in each cell type. (E) Cell-cell communications among cell types by Cellchat analysis. (F) and (G) show the expression patterns of the four identified antihypertensive drug targets in each cell cluster. (H) Violin plots of the expression of the four identified targets in different cell types and tissue types. Red represents peripheral blood immune cells and cyan represents tumor-infiltrating immune cells. Abbreviations: UMAP, Uniform manifold approximation and projection.