Onset timing of statin‐induced musculoskeletal adverse events and concomitant drug‐associated shift in onset timing of MAEs

Abstract To evaluate the onset timing of musculoskeletal adverse events (MAEs) that develop during statin monotherapy and to determine whether concomitant drugs used concurrently with statin therapy shifts the onset timing of MAEs. Cases in which statins (atorvastatin, rosuvastatin, simvastatin, lovastatin, fluvastatin, pitavastatin, and pravastatin) were prescribed were extracted from the US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) Data Files. The onset timing of MAEs during statin monotherapy was evaluated by determining the difference between statin start date and MAE onset date. The use of concomitant drugs with statin therapy was included in the analysis. Statins used in combination with concomitant drugs were compared with statin monotherapy to determine if the use of concomitant drugs shifted the onset timing of MAEs. The onset of MAEs was significantly faster with atorvastatin and rosuvastatin than with simvastatin. A difference in onset timing was not detected with other statins because the number of cases was too small for analysis. When evaluating concomitant drug use, the concomitant drugs that shifted the onset timing of MAEs could not be detected. Statins with strong low‐density lipoprotein cholesterol‐lowering effects (atorvastatin and rosuvastatin) contributed not only to a high risk of MAE onset, but also to a shorter time‐to‐onset. No concomitant drug significantly shifted the onset timing of MAEs when used concurrently with statins.

of atherosclerotic cardiovascular disease (ASCVD). 1,2 However, if musculoskeletal adverse events (MAEs) such as myalgia, myopathy, and rhabdomyolysis develop during the statin use period, these cholesterol-lowering treatments may need to be temporarily or permanently discontinued. 2 There are a few reports detailing the onset timing of drug-induced adverse events. 3,4 In 2004, Chang et al reported the onset timing of statin-induced rhabdomyolysis. 5 However, the difference in onset timing of rhabdomyolysis between each statin was not detected owing to few number of cases.
It is difficult to detect drug-drug interactions (DDIs) that may cause severe adverse events at the stage of drug approval examination. 6 DDIs are usually discovered during postmarketing surveillance. 7,8 It is already known that the concomitant use of statins and specific nonstatin drugs increases the risk of rhabdomyolysis.
For example, the concomitant use of statins with fibrates [9][10][11] or cytochrome P450 (CYP) inhibitors, such as clarithromycin (CYP3A4 inhibitor), cyclosporine (CYP3A4 inhibitor), and clopidogrel (CYP2C8 inhibitor), increases the risk of rhabdomyolysis. [12][13][14] It has been reported that the increased risk of statin-induced rhabdomyolysis may be due to the pharmacokinetic changes caused by concomitant drugs. 15 If DDIs cause changes in the time-course of blood concentration of statins, it not only changes the onset risk, but may also affect the onset timing. There are limited studies evaluating the risk of concomitant drugs on the onset timing of statin adverse events.
To reduce and prevent the risk of adverse events in a clinical setting, it is important to acquire information on both the risk and onset timing of drug-induced adverse events. We have already evaluated the onset timing of adverse events as well as the risk of these events. 3,16 Although many drugs have the potential to cause the same adverse event, especially those within the same medication class, the onset timing of these events for individual drugs may differ; thus, it is important to evaluate the onset timing of side effects associated with each drug.

The Food and Drug Administration (FDA) Adverse Event
Reporting System (FAERS) is often used to detect DDIs. Risk evaluation using disproportionality is performed to determine these DDIs. 17,18 Although the incidence of statin-induced MAEs differs in literature, statins are well-tolerated and rarely cause MAEs. Therefore, the incidence of statin-induced MAEs is very low, 11,19,20 and it is difficult to evaluate the onset timing of statin-induced MAEs through clinical trials as these adverse events may occur within 12 months of starting statin therapy or after many years. 10

| Standardisation of names of drugs reported to FAERS
The drugs reported to FAERS can be registered by arbitrary names, including trade names and typographical errors. 22 Therefore, we used DRUGBANK (version 5.0.11) to standardise the names of drugs, including statins and other concomitant drugs. 23

| Cases of statin monotherapy
Some of the cases reported to the FAERS were the same cases that were reported by different reporters (duplicate cases). Thus, to exclude duplicate cases from our analysis, among the cases of statin monotherapy reported, cases in which all the 4 items of age, sex, adverse event onset date (EVENT_DT), and start date of statin use (START_DT), were the same were regarded as duplicate cases and eliminated. The differences between EVENT_DT and START_DT of cases of statin monotherapy were regarding the time-to-onset for MAEs. As the onset period (time-to-onset) of MAEs during statin use was mostly within 1 year, 10 cases of statin monotherapy in which MAEs had developed within 365 days were selected for analysis in this study. A statistical analysis was performed to determine whether the time-to-onset of MAEs differed with statin type. In addition, statins for which less than 30 cases were reported for analysis were not included in the subsequent study in which the impact of concomitant drugs on the time-toonset of MAEs was investigated (described below).

| Statistical analysis
Cases of statin monotherapy were collected and a nonparametric method, the Steel-Dwass test, was used to determine if the time-toonset of MAEs differed by statin type. This nonparametric analysis method was adopted as it was assumed that the time-to-onset would not be normally distributed. To determine if concomitant drugs affected the time-to-onset of statin-induced MAEs, paired comparisons between cases of each concomitant drug use and cases of statin monotherapy were also performed using the Steel test.
All statistical analyses were performed using R software (version 3.2.2, R Foundation for Statistical Computing, Vienna, Austria) for Windows ® . The significance level (P) was set at 0.05.    The magnitude of HMG-CoA reductase 50% inhibitory concentration (IC 50 ) for each statin was in the order of rosuvastatin < atorvastatin < simvastatin < fluvastatin < pravastatin. 24 Atorvastatin or rosuvastatin, which possesses a high HMG-CoA reductase inhibitory activity, is considered high-intensity statin therapy that reduces LDL cholesterol by more than 50%. The use of the other statins is considered either moderate or low-intensity statin therapy. 1    The Steel test was performed for cases of the simvastatin monotherapy group (n = 409) as the control group. IQR, interquartile range; MAEs, musculoskeletal adverse events.