AUTHORS
Sergiu Pasca, Yun Soo Hong2, Wen Shi3, Daniela Puiu, PhD4, Dan Arking, PhD5, Lukasz P. Gondek, MD, PhD6
AFFILIATIONS
1 Johns Hopkins University, Baltimore, Maryland, United States
2 New York University, New York, United States
3 Johns Hopkins University, Baltimore, United States
4 Johns Hopkins University, Baltimore, United States
5 Johns Hopkins University, Baltimore, Maryland, United States
6 Johns Hopkins University, Baltimore, Maryland, United States
ABSTRACT
Background: Clonal hematopoiesis of indeterminate potential (CHIP) is a common condition associated with aging and chronic inflammation that increases the risk of myeloid neoplasms (MN). In addition to CHIP, the presence of mitochondrial DNA (mtDNA) mutations has also been associated with MN risk, suggesting that mitochondrial dysfunction represents an additional mechanism for the development of MN. Currently, there are no safe therapies proven to prevent CHIP or its evolution to MN. Statins are widely prescribed, safe medications with established anti-inflammatory properties and emerging effects on mitochondrial function. Supporting the role of statins in MN risk, population studies demonstrated an inverse association between statin use and the risk of myeloproliferative neoplasms (MPN). We therefore investigated whether statin use is associated with the prevalence of CHIP and the subsequent risk of MN in the UK Biobank (UKB) cohort.
Methods: CHIP variants were called from whole exome sequencing and mtDNA mutations were assessed from whole genome sequencing using MitoHPC. Since more than one mtDNA mutations can be present in an individual, the cumulative effect of mtDNA mutations was calculated as mMLC score sum (mMSS). Clonal hematopoiesis risk score mitochondria (CHRS-M) was used to summarize the overall effect of both CHIP and heteroplasmy. Statin use was determined from verbal interview by a trained nurse in the UKB. The associations between statins and incident MN were estimated using Cox proportional hazards models adjusting for sex, age, body mass index, smoking status, type 2 diabetes, prevalent cardiovascular disease (CVD), aspirin, blood pressure medications and low-density lipoprotein (LDL) cholesterol.
Results: We included 475,966 individuals from the UKB. The mean age was 56.5 years and 216,709 (45.5%) individuals were male. Statins were used by 77,593 (16.7%) individuals at baseline.
We found that statin use was associated with a lower prevalence of CHIP (P = 0.01). Moreover, statin use was inversely associated with high-risk CHIP features, including VAF (P < 0.01), number of CHIP mutations (P = 0.02) and high-risk mutations (P < 0.0001). Interestingly, there was no association between statins use and a single DNMT3A mutation, which rarely progresses to MN (P = 0.75). These effects were consistent across different statin types (simvastatin, atorvastatin, rosuvastatin, fluvastatin, and pravastatin). Conversely, ezetimibe, a different class of cholesterol lowering medication, was not associated with a lower prevalence of CHIP. This suggests that the effect of statins on CHIP and MN risk may be independent of their LDL cholesterol reduction effect.
We further assessed if statin use is also associated with the MN risk. We found that statin use was significantly associated with a lower risk of developing MN (HR = 0.6; 95% CI 0.5–0.8; P < 0.0001). Similar to our findings in CHIP, this effect was consistent across different statins and was not observed with ezetimibe.
Given the effect of statins on mitochondria including the inhibition the electron transport chain, increasing oxidative stress and decreasing cell energetics, it is possible that they imposed a different effect on cells with and without deleterious mtDNA mutations (high mMSS). Indeed, we observed a significant difference in MN risk reduction based on mMSS score. The effect was more pronounced in individuals with high mMSS, suggesting that persons with likely deleterious mtDNA mutation were more likely to benefit from statin use (P = 0.04).
Even though statins were associated with an overall reduction in the risk of MN, this effect might be different between different risk groups. We have recently developed a modified CHRS risk score that includes heteroplasmy (CHRS-M). The significant effect of statins on MN risk reduction was observed in CHRS-M intermediate (HR = 0.6; 95% CI 0.4–1.0; P = 0.08) and not in CHRS-M high (HR = 1.0; 95% CI 0.6–1.7; P = 0.99) category. It would suggest that statins may be more effective as primary prevention or at the early stages of clonal evolution.
Conclusions: Our results suggest that statins decrease the prevalence of CHIP and subsequent MN through non-LDL cholesterol-mediated mechanism. Due to the pleiotropic effects of statins, several mechanisms could explain these associations including mevalonate pathway inhibition, anti-inflammatory effects, and inhibition of cell growth and division.
DISCLOSURES
Sergiu Pasca: No relevant conflicts of interest to declare.