Using Mendelian randomization to investigate etiologic heterogeneity across renal cell carcinoma subtypes.

Authors

Winter TD, Jahagirdar O, Renal Cancer Genetics Consortium, , Johansson M, Brennan P, Machiela MJ, Chanock SJ, Purdue MP, Dutta D

Affiliations

• Laboratory of Genetic Susceptibility, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, MD, United States.
• Integrative Tumor Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, MD, United States.
• Genomic Epidemiology, International Agency for Research on Cancer, Lyon, France.
• Occupational and Environmental Epidemiology Branch, Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, MD, United States.

Abstract

BACKGROUND: Renal cell carcinoma (RCC) histological subtypes clear cell RCC (ccRCC; >75% of cases) and papillary RCC (papRCC; ∼15%) exhibit distinct molecular and genetic profiles, patient demographics, and prognoses. Previous epidemiologic studies have identified several risk factors for overall RCC, although few have explored differences in etiology across subtypes.

METHODS: For this study, we applied two-sample Mendelian randomization (MR) to findings from a genome-wide association study of RCC (27 213 cases, 488 019 controls) to investigate the effects of RCC risk factors with ccRCC (15 507 cases) and papRCC (2103 cases). We also conducted case-only MR analyses contrasting ccRCC and papRCC cases to test for heterogeneity in risk factor effects across subtypes.

RESULTS: MR for overall RCC confirmed associations with obesity, blood pressure, smoking, and several other suspected risk factors. In subtype-specific analyses, we observed stronger associations with ccRCC than for papRCC for anthropometric measures such as body mass index [ccRCC odds ratio (ORccRCC) = 1.58 per standard deviation increase, 95% confidence interval (CI) = 1.50-1.68; papRCC odds ratio (ORpapRCC) = 1.24, 95% CI = 1.07-1.42; Pheterogeneity = 2.7 × 10-4], while stronger associations with papRCC were observed for chronic kidney disease (ORccRCC = 1.07, 95% CI = 0.99-1.15; ORpapRCC = 1.39, 95% CI = 1.16-1.66; Pheterogeneity = 5.42 × 10-5), creatinine-based estimated glomerular filtration rate (ORccRCC = 0.96, 95% CI = 0.92-1.01; ORpapRCC = 0.71, 95% CI = 0.64-0.79; Pheterogeneity = 7.76 × 10-5), and telomere length (ORccRCC = 1.98, 95% CI = 1.93-2.06; ORpapRCC = 2.50, 95% CI = 2.28-2.72; Pheterogeneity = 6.2 × 10-3). Further analysis identified the colocalization of significant RCC risk loci and 20 risk factors along with potential target genes through transcriptomic analysis.

CONCLUSION: These results highlight the heterogeneous nature of RCC etiology and the importance of considering histologic subtypes in etiologic and genetic studies.