Patients

To be eligible for the registry, patients had to be at least 18 years of age and have established ASCVD, defined as having any one of following clinical conditions: prior history of MI, coronary artery disease (CAD), coronary or other arterial revascularization, ischemic stroke or transient ischemic attack, carotid artery stenosis, or documented peripheral arterial disease secondary to atherosclerosis (aortic aneurysm, ankle brachial index <0.9, imaging evidence of >50% stenosis in any peripheral artery, or intermittent claudication). Patients were enrolled into one of three cohorts: (1) currently receiving a PCSK9i, (2) receiving no PCSK9i with LDL-C 70–99 mg/dL, and (3) receiving no PCSK9i with LDL-C ≥100 mg/dL. All patients had to be receiving some type of stable LLT for at least 4 weeks prior to enrollment, including the PCSK9i cohort, at the discretion of the treating physician. Of note, patients with LDL-C <70 mg/dL were not eligible unless they were receiving a PCSK9i.

Patients were not eligible for inclusion in the registry if unable or unwilling to provide informed consent including but not limited to cognitive or language barriers, were currently participating or planned to participate in an interventional clinical study involving any investigational medical device or drug treatment at the time of enrollment, had a life expectancy <12 months, or were currently pregnant, breast feeding, or planning to become pregnant.

Study procedures

Sites screened potentially eligible patients, and those who were eligible were asked to provide informed consent prior to enrollment. Baseline data collection and a 1-year retrospective chart review were carried out by research personnel at the enrolling site. Patient charts are prospectively reviewed and data will be abstracted every 6 months for 2 years from the time of enrollment. Laboratory values and medication use recorded in the electronic care report forms were the most recent values or medication use to the time of the visit. In this study, the collection and reporting of safety information is required only for patients exposed to evolocumab. Patients were approached to complete interactive phone surveys at baseline and every 6 months throughout the study to determine their perceptions of LLT. At each center, one physician was asked to fill out a questionnaire at the time of enrollment of their first patient, and at 1 and 2 years following this date regarding their perceptions and patterns of use of LLT. (See Fig. 1 .)

Study schema.

Patients were asked about their understanding of why they were taking LLT, and reasons for any changes in their therapy in the past, as well as detailed questions about their experience with statins, ezetimibe and PCSK9 inhibitors if they had taken these, including any barriers to being able to take these agents such as issues with insurance and co-pays. Physicians were asked about their goals of LLT, the frequency of lipid testing they use, and their perception of most reasons for non-adherence in their patients. They were asked about their approach to specific patient types, including those with statin intolerance.

Outcome variables

The primary outcome variable is change in LLT. This outcome includes initiating or discontinuing statin therapy, PCSK9i, or ezetimibe, increasing or decreasing the dose of a statin or PCSK9i, switching to a different type of statin or PCSK9i, changes in other LLT (including fish oil/omega-3 preparations, bile acid sequestrants, mipomersen, lomitapide, apheresis, and any new LLT therapy that enters the market after study initiation) or no changes in LLT. A secondary outcome variable is whether or not LDL-C and other lipid values are measured, and if so, when and the values. A key outcome variable of interest from the physician questionnaires is physicians’ lipid treatment objective in patients with ASCVD. During patient phone surveys, we collect patient-reported outcomes, and patient perceptions and attitudes toward their ASCVD risk and LLT.

Statistical considerations

The patient population cohorts had planned sizes (±10%) as follows: approximately 500 patients receiving a PCSK9i at baseline; approximately 2500 patients with LDL-C between 70 and 99 mg/dL; and approximately 2000 patients with LDL-C ≥100 mg/dL.

No formal hypothesis will be tested in this observational study. Point estimates of outcomes (eg, the percent of patients intensifying nonstatin therapy) will be generated along with 95% confidence intervals (CIs). Continuous variables will be summarized by the evaluable sample size, mean, median, standard deviation, quartiles, minimum and maximum. With 5000 patients, the percentage of patients with a given LLT pattern (eg, adding a nonstatin therapy, reducing a statin dose, or switching statins) would be estimated with a 95% CI no wider than ±0.69%. For analyses limited to a 2000 patient cohort, the maximum 95% CI would be ±1.10%.