PCSK9 and breast cancer survival: a Mendelian Randomization study.

BACKGROUND: Proprotein convertase subtilisin/kexin type 9 (PCSK9) affects the lipid metabolism. A recent study identified an association between rs562556 within PCSK9 gene and breast cancer survival (BCS), suggesting PCSK9 inhibition as an early intervention strategy to prevent metastasizing breast cancer. We attempt to replicate these findings using genome-wide association study (GWAS) data and Mendelian Randomization (MR). METHODS: We used GWAS data from the Breast Cancer Association Consortium (BCAC, N=91,686), and performed analyses in the FinnGen study (N=4,648, additive and recessive model). First, we tested the association of single variant rs562556, then genetically proxied PCSK9 inhibition using multiple variants, and finally adjusted for indirect effects of the lipid metabolism using multivariable MR. Coronary artery disease was chosen as a positive control outcome. RESULTS: Estimates are scaled as log-hazard ratios (HR) per 1 standard deviation higher PCSK9 levels. We found no significant association between PCSK9 variants and BCS in any MR approach with any replication data: single variant (BCAC: log(HR)=2.37, p=0.066; FinnGen additive: log(HR)=0.58, p=0.91; FinnGen recessive: log(HR)=-2.75, p=0.87), multiple variants (BCAC: log(HR)=-0.22, p=0.30; FinnGen: log(HR)=-0.40, p=0.54), or multivariable (BCAC: log(HR)=0.42, p=0.65; FinnGen: log(HR)=1.23, p=0.75). Positive control analyses were significant throughout. CONCLUSION: A significant association of PCSK9 variants with BCS could only be reproduced using outcome data from the original study, but not when using independent, larger GWASs. IMPACT: Potential reasons for the discrepant results are different genetic models, sample selection criteria, and the time-variability of HR estimates. They should be explored before considering PCSK9 a therapeutic target in BC patients.