Statin use in patients with non‐HMGCR idiopathic inflammatory myopathies: A retrospective study

Abstract Background Statins are the most widely used lipid lowering therapies which reduce cardiovascular risk, but are associated with muscular adverse events (AEs). Idiopathic inflammatory myopathies (IIM) are autoimmune diseases of the muscle with higher risk of cardiovascular disease. More data is needed regarding statin safety in patients with intrinsic muscle disease such as IIM. Hypothesis Statins are tolerated in patients with IIM without leading to significant increase in muscular AEs. Methods Statin use was retrospectively examined in a longitudinal IIM cohort. Safety analysis included assessment of muscular and nonmuscular AEs by chart review. IIM patients receiving a statin during the cohort follow‐up period were matched to IIM patients not receiving a statin for comparative analysis of longitudinal outcomes. Results 33/214 patients had a history of statin use. 63% started for primary prevention, while others were started for clinical ASCVD events, vascular surgery, IIM related heart failure, and cardiac transplantation. A high intensity statin was used in nine patients with non‐HMGCR myositis, and tolerated in 8/9 patients. Statin related muscular AE was noted in three patients. There were no cases of rhabdomyolysis, or statin related nonmuscular AEs in a median observation period of 5 years. In patients newly started on statins during cohort follow‐up (n = 7) there was no change in disease activity after statin initiation. Long term outcomes were not different between statin and nonstatin IIM control groups. Conclusion Statins were well tolerated in patients with non‐HMGCR positive IIM. Given the accelerated atherosclerotic risk in IIM patients, further prospective studies of statin safety in IIM patients are warranted.


| INTRODUCTION
Idiopathic inflammatory myopathies (IIM) are a group of autoinflammatory muscle diseases characterized by debilitating muscle weakness and increased morbidity and mortality. Patients with IIM have a significantly higher risk of cardiovascular (CV) disease and dyslipidemia compared to the general population. 1-3 HMG CoA reductase inhibitors (statins) are the first line pharmacologic intervention for primary and secondary prevention of atherosclerotic cardiovascular disease (ASCVD). 4,5 Studies have consistently demonstrated the efficacy of statins on LDL cholesterol (LDL-C) and reducing the risk of ischemic heart disease, stroke and CVD associated mortality. [6][7][8] Data also suggests potential anti-inflammatory benefits of statins in autoimmune diseases. 9 Muscle adverse effects (AEs) including muscle pain, weakness and cramps are reported in 5% to 20% of patients on statins, and the majority of these resolve within weeks to months after drug cessation. 10,11 Rhabdomyolysis occurs with an incidence of approximately 0.4 per 10 000 patient years 12 and has been reported mostly in patients with pre-existing comorbid conditions or on multiple medications. 13 Recently, a unique entity of statin induced immune mediated necrotizing myopathy (IMNM) has been described with autoantibodies targeting the HMG CoA reductase (HMGCR) protein. 14,15 Patients with underlying muscle disease including muscular dystrophies or metabolic myopathies are frequently hesitant to comply with statin therapy due to the widely publicized muscular AEs and small studies suggesting possible increased risk in these populations. 16,17 Such reports raise the concern of whether statins can be used safely for prevention of CV disease in these patients. To date, there is limited data on the safety and tolerability of statin use in IIM patients and further evaluation is warranted. Here we describe our experience of statin use in a longitudinal cohort of patients with IIM from a single tertiary academic center.

| Patients
A retrospective chart review was conducted of the UCLA IIM cohort, a longitudinal observational cohort including 214 adult patients with IIM. Patients fulfilled the EULAR/ACR classification criteria for IIM meeting the definition for at least "probable IIM." 18 All subjects gave written informed consent for the study under a protocol approved by the UCLA IRB (#10-001833). All patients who reported ever taking a statin as part of their daily medications were identified.
Patients were also analyzed longitudinally if they were taking a daily statin at any point during their follow-up period for >2 consecutive months and had available data regarding disease activity measures.
Each patient receiving a statin during the cohort follow-up period was matched to a control subject by (a) age ± 5 years, (b) gender, and (c) baseline physician global disease activity score by 100 mm visual analog scale (VAS) ±10 mm. 19 Baseline visit was defined as the first visit on a statin for the statin group, and cohort enrollment visit for the control group.
All patients had baseline lipid profiles and repeat lipid profiles were routinely assessed during longitudinal follow up. The history of prior CV events was identified by questionnaires and chart review.

| Atherosclerotic cardiovascular disease (ASCVD) risk assessment
We report 10 year ASCVD risk scores using the pooled cohort equations (PCE) risk calculator. 20,21 Outputs of the PCE risk calculator include the 10-year and lifetime risk for developing a first CV event (nonfatal myocardial infarction, nonfatal stroke, fatal coronary heart disease, or fatal stroke). Patients with 10 year ASCVD risk of greater than 7.5% are recommended to initiate high to moderate intensity statin for primary prevention.

| IIM disease assessments
Baseline disease characteristics were assessed including IIM type, myositis specific antibodies, and disease duration. Disease activity was assessed using physician global myositis disease activity by 100 mm VAS and 5 point Likert scales. 19 Laboratory measures included creatine phosphokinase (CPK), aldolase, estimated sedimentation rate (ESR), and C-reactive protein (CRP).
Disease activity was assessed at multiple time points. For patients that were already on a statin at time of cohort enrollment, disease activity was assessed at the baseline visit and the consecutive follow-up visit. For patients that were started on a statin during the cohort follow-up period, disease activity measures were collected at the visit before statin initiation and the first visit after statin initiation. Data from the most recent clinic visit was also reviewed to assess long term follow up. Clinically quiescent myositis was determined as no evidence of muscular or extra-muscular myositis disease activity by subjective report, on physical exam and muscle enzymes.

| Safety assessments
Patients in the cohort were followed every 2 to 3 months in clinic.
Data from all visits were reviewed for the following prespecified muscular and nonmuscular AEs: (a) myalgias, (b) CPK elevations (>25% increase compared to prior visit, for two consecutive visits),

| Statistical analysis
Baseline demographics and disease activity measures between groups were compared using chi-square test for categorical variables, and student's t-test or Wilcoxon Rank sum test for continuous variables.
In comparing changes in disease activity measures, paired student's t-test and paired Wilcoxon signed rank test were used. Statistical significance was defined as a two-sided P value of <.05. Statistical analysis was performed on JMP Pro version 13.0.0 (SAS Institute Inc., Cary, North Carolina).

| Statin use in the IIM cohort
Past or present statin use was identified in 33 patients in the IIM cohort ( Figure 1). Seven patients reported statin use in the past but had discontinued the statin prior to cohort enrollment. Twenty-three patients were actively receiving a statin during the cohort follow-up period with disease activity measures available for review (statin group, Table 1). These patients were matched to IIM controls by age, gender and myositis disease activity (control group, see Section 2 for details).

| Type of statin therapy
The most common type of statin used was atorvastatin 5 to 40 mg (n = 22) followed by rosuvastatin 5 to 20 mg (n = 8) ( Table 2). Simvastatin was used in two patients, and one reported related myalgias.
Simvastatin has been associated with a higher risk of muscular AEs compared to other statins. 11 A high intensity statin was used in nine patients with non-HMGCR myositis, and tolerated in 8/9 patients.
The majority of these patients were started after a clinical ASCVD event.
F I G U R E 1 Flowchart of patient groups. *Patients that discontinued statin prior to cohort enrolment. **Control group: matched to each patient in statin group by (a) age ± 5 years, (b) gender, and (c) baseline physician global disease activity score by 100 mm visual analog scale (VAS) ±10 mm

| Statin safety
AEs during statin therapy are outlined in Table 2. Seven patients were previously on statins but discontinued prior to myositis diagnosis (prior statin group in Figure 1 The most common laboratory abnormality was elevation in liver enzymes (n = 5), followed by increased creatinine (n = 2), none of which were statin-related (Table 3)

| Longitudinal analysis: baseline characteristics of statin and comparator groups
Comparison of statin (n = 23) and control groups (n = 23) is outlined in Baseline lipid profiles were similar between the two groups. Bold values are high intensity statin.

| longitudinal analysis: disease activity assessments
To assess whether statin use was associated with worsened myositis activity, changes in disease activity between baseline and consecutive follow-up visits were compared between statin and control groups. Consecutive follow-up visit was chosen for repeat disease activity assessment to minimize potential confounding by changes in immunomodulatory medications. Changes in disease activity measures over time were not significantly different between patients on statins and IIM controls (p = NS for all, Table 3).
Subgroup analysis of patients who were newly started on a statin during cohort follow-up (n = 7) and matched controls showed no differences in disease activity measures after statin initiation.  24 Histopathologic studies also implicate direct involvement of the microvasculature in the pathophysiology of DM. 25,26 Taken together, this data suggests that the study of lipids and lipid lowering agents is particularly relevant to patients with inflammatory myopathies.
Statins are the first line lipid lowering agent and make up over 80% of all lipid lowering medications used in clinical practice. 27 Muscular AEs are reported in 5% to 20% of the general population using statins 28 and in the current study occurred in 10% (3/29) of IIM patients without HMGCR antibody-associated disease, which is within the range of AEs in the general population. This data is also consistent with the survey study in which IIM specialists reported worsening muscle symptoms in~10% of their IIM patients using statins, with the majority improving after discontinuation of statin therapy. 29 In contrast to the widely publicized concerns of statin related muscular AEs, a systematic review of clinical trials reported that muscular AEs were minimally higher in statin patients when compared to placebo controls, 30 suggesting that statin related muscular complaints may be overemphasized in clinical practice. The purpose of this study was to further extend the understanding of statin related muscular AEs, by examining outcomes in a group of patients with intrinsic auto inflammatory muscle diseases.
The current work reported statin use in IIM patients from a longitudinal cohort at a tertiary academic center, which included high risk, complex patients with a history of clinical ASCVD events, vascular surgery, IIM related heart failure and cardiac transplantation. To assess the impact of statins on myositis disease activity, we compared IIM patients on statins to a matched nonstatin exposed IIM group with median follow-up of over 5 years. Recent work by Borges and colleagues also reported a retrospective analysis of statin use in 24 patients with IIM on either atorvastatin or simvastatin with a slightly shorter median follow-up of 22.5 months. 31 While this previous study did not include an IIM comparator group, or high risk CVD patients (such as patients with prior ASCVD event), a similarly good tolerability and safety of statins was observed in IIM patients with stable disease.
The ASCVD risk assessment tool was developed as a strategy to personalize the estimation of ASCVD risk in order to help target CV preventative strategies, including statin use. 32 The ACC/AHA PCE risk calculator is widely used to personalize the estimation of benefits from risk reducing therapies. However, there are caveats to the application of the risk calculator that are noteworthy when considering statin therapy in patients with IIM as presented in our study.   48 Interestingly, the current work demonstrated a modest trend for greater decreases in CRP levels in the statin group compared to the control group. This has previously been reported with statin therapy in the general population. 49 Additional work to examine the molecular effects of statins on disease pathogenesis in IIM may be warranted.
Our study has several limitations. IIM are rare diseases with a prevalence of 2 to 58 per 100 000. 50 Therefore, the total number of patients reported in the current work is small. However, all patients were part of the same single center cohort of over 200 IIM patients and statin patients were compared to matched, nonstatin exposed IIM controls from the same cohort. The majority of patients presented in this work had DM, which is consistent with the predominance of DM in our center's longitudinal cohort. Additional study of statin tolerance in larger numbers of patients with other IIM is warranted. In addition, limitations to a retrospective review include other selection bias of the patients included in the initial cohort and the possibility of data gaps including lack of proper adverse event (AE) recording and medication compliance.
Many patients initiated statin use under the care of an outside physician prior to referral to our center, and thus missing data was inevitable. Also, with the concern of muscular AEs, many patients with high disease activity may not have been treated with a statin which may have introduced a potential bias to the statin cohort reported. However, all patients were followed routinely in our clinic, and chart data was carefully reviewed in order to limit these known caveats. Lastly, a median follow-up period of 5 years may be inadequate to determine long term safety. Longer term follow-up of this cohort is ongoing.
In conclusion, statins were well tolerated in a single center retrospective study of IIM patients. Use of statins may be considered in IIM patients without HMGCR antibody-associated IMNM when clinically indicated for CV risk reduction. Further prospective studies with larger patient groups are warranted to assess the safety of statins in IIM patients.
Dr. Charles-Schoeman has served as a consultant to Pfizer, Abbvie, and Octapharma, Gilead and Regeneron-Sanofi and has received research grants from Pfizer, Bristol Myers Squibb, Abbvie, and Octapharma.