Statin Use and the Risk of Myeloproliferative Neoplasms in a Population Based Cohort

INTRODUCTION Several large-scale epidemiologic studies has associated the use of statins (STs) to a reduced risk of pancreatic, hepatic, lung, and gastrointestinal cancers(Wang et al, 2022; Khurana et al, 2007; Kim et al, 2017). Preclinical studies of STs have suggested antileukemic activity in myeloid neoplasms including myeloproliferative neoplasms (MPNs) (Jang et al, 2019; Li et al, 2003; Burke & Kukoly, 2008) and several effects of STs including antiproliferative and proapoptotic effects, decreasing release of several growth factors, and antiangiogenetic effects have led to the perception that STs could prevent the development of fibrotic diseases(Hasselbalch & Riley, 2006; Clendening & Penn, 2012). In this study we aimed to investigate the association between ST use and risk of MPNs.

METHODS We conducted a population-based nested case-control study comparing ST use among patients diagnosed with MPN between 2010 and 2018 and a matched population (without MPN) to estimate odds ratios (ORs) for MPN in association to ST use. We utilized and linked five nationwide Danish health registers to collect exposure data and covariates for adjusted analyses. Cases were identified using the National Chronic Myeloid Neoplasia Registry covering >90% of MPN cases in Denmark (Bak et al, 2016) and controls were selected by incidence density sampling (five to one) from the general Danish population. The index date was defined as the first date of the MPN diagnosis. The exposure assessment of ST was identified using Danish National Prescription Registry. Ever-use of ST was defined as having redeemed at least one prescription of ST prior to the index date, never-use was defined as no filled prescription of ST prior to the index date, and long-term use was defined as ≥5 years treatment with ST, assuming a daily intake of one tablet of the prescribed dose and added 25% additional days to account for variations and minor incompliance. Conditional logistic regression was used to obtain ORs provided with 95% confidence intervals (CIs) for 1) analysis adjusted for the matching variables by study design (age, sex, and index date) and 2) fully adjusted ORs (aORs) (see footnote in Table 2 for adjusted variables). Additionally, we used a lag-time of 12 months before index data to disregard information on covariates. The same lag-time was applied to the exposure assessment for STs to avoid reverse causation.

RESULTS The study population comprised 3,816 cases and 19,080 controls and the characteristics and exposure patterns are given in Table 1. Among cases 34.9% were categorized as ever users of STs compared to 33.5% among controls resulting in a OR of 1.07 (95% CI, 0.99-1.16) and an aOR for MPN of 0.87 (0.80-0.96). Overall, 17.2% of cases and 19.0% of controls were long-term users yielding a crude OR of 0.90 (0.81-1.00) and an aOR for MPN of 0.72 (0.64-0.81). Analyses using cumulative time of exposure to STs revealed a clear dose-dependent response with decreasing aORs with longer treatment duration (<1 years: aOR 1.15 (0.97-1.36); 1-4.99 years: aOR 0.96 [0.85-1.08]; 5-9.99 years: aOR 0.83 [0.72-0.95]; ≥10 years: aOR 0.63 [0.54-0.74]). Analyses of the association between long-term use of STs and MPN for age subgroups revealed a null estimate for age <60 years with an aOR of 0.80 (0.52-1.25), whereas age 60-75 years and >75 years revealed a significant negative association with an aOR of 0.77 (0.65-0.91) and 0.65 (0.54-0.78), respectively. Subgroup analysis for male sex was associated with decreased odds of MPN after long-term ST use with an aOR of 0.57 (0.47-0.68) whereas this was not seen for female with an aOR of 0.88 (0.75-1.04). Furthermore, long-term use of ST was associated with decreased odds of myelofibrosis (aOR 0.60 [0.44-0.82]), polycythemia vera (PV) (aOR 0.73 [0.59-0.89]) and MPN unclassifiable (MPN-U) (aOR 0.52 [0.39-0.71]), whereas a neutral association with essential thrombocythemia (ET) were observed (aOR 0.89 [0.72-1.10]).

CONCLUSION To the best of our knowledge, this is the first study to examine the association between ST use and risk of MPNs. A protective effect of STs on the risk of developing MPN with an aOR of 0.87 (0.80-0.96) was demonstrated, primarily driven by long-term use and a strong interaction by gender, where the association was particularly strong in males. Subgroup analyses revealed a protective effect for PV, MF and MPN-U, however not for ET. Further preclinical and clinical studies on this subject is warranted.

Jakobsen:Roche: Honoraria. El-Galaly:Roche: Ended employment in the past 24 months; Abbvie: Other: Teaching in 2021. Roug:Jazz Pharmaceuticals: Other: Travel grant.