The following input is a summary of information provided by clinical specialists with expertise in the diagnosis and management of hypertriglyceridemia and CV risk

Patients with established CVD (defined as the secondary prevention group) and patients with diabetes and one or more CV risk factors (the primary prevention or at high risk group) can remain at high risk for a major adverse CV event (MACE) due to persistent high levels of TGs, even after treatment with statins. Studies of added therapies to reduce these persistent levels of TG, such as fibrates or niacin, have demonstrated little to no effect for decreasing the risk of CV events. Fish oil supplements containing omega-3 fatty acids EPA and DHA have been tested as well, with no difference in the reduction of risk for further MACE. This group of patients, often referred to as patients with residual risk, are a subgroup of interest for whom there is an unmet need for effective and safe treatments. The main goal of treatment in this specific group of patients is the achievement of a reduced risk in CV events, including CV death, MI infarction, stroke, and need for urgent revascularization. Currently available treatments are not meeting the goals and needs of patients. These patients remain at higher risk of CV events when compared to those whose values have been normalized after therapy and lifestyle modifications. Current research has not identified definite subgroups in whom an intervention will have increased efficacy.

Although the exact mechanism of action of icosapent ethyl is not yet fully understood, previous studies have shown a reduction in TG levels, probably involving complex anti-inflammatory, antithrombotic, and TG metabolism effects. Icosapent ethyl could be used when TG levels remain elevated despite stable use of statin therapy (i.e., for more than four weeks). Icosapent ethyl’s main use will be as a complementary intervention. It is not expected to be used as first-line therapy.

Patients meeting the criteria of established CV risk or at high risk for CV, identified by primary care physicians and eventually assessed, when needed, by cardiologists or endocrinologists, will be eligible for treatment with icosapent ethyl. They will have to be receiving treatment with stable statin dosages.

Description of Studies

Two randomized double-blind placebo-controlled studies were included in this review.

The first study is the REDUCE-IT trial, conducted in 11 countries, with a median follow-up of 4.9 years (up to 6.2 years). This trial included 8,179 patients older than 45 years of age with established CV risk, or older than 50 years of age with diabetes in combination with one additional risk factor for CVD. Patients had to have elevated TG levels (≥ 1.7 mmol/L and < 5.6 mmol/L) and to be receiving stable dosages of statins. This study evaluated 4 g daily of icosapent ethyl versus placebo. The primary end point assessed was the time from randomization to the first occurrence of any of the composite outcome events of CV death, non-fatal MI infarction, non-fatal stroke, coronary revascularization, and unstable angina requiring hospitalization. Secondary end points were evaluated in a hierarchical fashion and included a key secondary composite end point of time from randomization to any of CV death, non-fatal MI, or non-fatal stroke; a composite of CV death or non-fatal MI; fatal or non-fatal MI; emergency or urgent revascularization; CV death; hospitalization for unstable angina; fatal or non-fatal stroke; a composite of total mortality, non-fatal MI (including silent MI), or non-fatal stroke; and total mortality.

The second study is the ANCHOR trial, with 12 weeks of follow-up, conducted in 97 centres across the US. The study included 702 patients aged 18 years and older with fasting TG levels 2.3 mmol/L or greater and less than or equal to 5.6 mmol/L, receiving a stable dose of statin therapy (with or without ezetimibe), and at high risk for CVD. This study included three arms: placebo and icosapent ethyl 2 g daily and 4 g daily, of which only the icosapent ethyl 4 g daily dosage arm was included because this is the dosage submitted for the application to Health Canada. The study evaluated the percent change in TG blood levels from baseline to week 12 as the primary outcome; CV events or other clinically important end points were not assessed. Secondary end points included the percent change in non-HDL-c, LDL-c, apolipoprotein B, VLDL-c, and lipoprotein-associated phospholipase A2 from baseline to week 12.

Efficacy Results

Based on data from the REDUCE-IT study (Table 1), 17% of patients treated with icosapent ethyl 4 g daily versus 22% of patients in the placebo group had at least one of the events of the composite outcome of CV death, non-fatal MI, non-fatal stroke, coronary revascularization, and unstable angina (hazard ratio [HR] 0.75; 95% confidence interval [CI], 0.68 to 0.83). Icosapent ethyl 4 g also reduced CV mortality (HR 0.80; 95% CI, 0.65 to 0.98), non-fatal MI (HR 0.69; 95%CI, 0.59 to 0.82), non-fatal stroke (HR 0.70; 95%CI, 0.53 to 0.93), hospitalizations due to unstable angina (HR 0.67; 95% CI, 0.53 to 0.86), and need for coronary revascularization (HR 0.66; 95% CI, 0.58 to 0.75) compared with placebo. Icosapent ethyl did not demonstrate benefit versus placebo on overall mortality (HR 0.87; 95% CI, 0.74 to 1.02), hospitalization due to heart failure (HR 0.97; 95% CI, 0.77 to 1.22), or arrhythmia (HR 1.21; 95% CI, 0.97 to 1.49).

Table 1

Summary of Key Results From Pivotal and Protocol Selected Studies (ITT Population, REDUCE-IT).

Subgroups of interest for the review were baseline CVD risk (established CVD or at high risk for CVD) and baseline diabetes (diabetes or no diabetes). The results of the pre-specified subgroup analyses of the primary outcome in REDUCE-IT were similar to those for the full population: icosapent ethyl reduced the risk of the composite outcome relative to placebo. The subgroup analyses suggested a different magnitude of effect in the subgroups stratified according to the CV risk of patients (at high risk for CVD [primary prevention] or established CVD [secondary prevention]). The results for patients in the secondary prevention subgroup indicated a statistically significant effect on the primary outcome with icosapent ethyl (559/2,892 events [19.3%]) versus placebo (738/2,893 events [25.5%]; HR 0.726; 95% CI, 0.650 to 0.810), similar to that observed in the total population. However, the results in the primary prevention subgroup were not statistically significant because the upper confidence interval for the HR goes beyond 1.00 (146/1,197 [12.2%] events with icosapent ethyl; 163/1,197 [13.6%] events with placebo; HR 0.876; 95% CI, 0.700 to 1.095]). The absolute risk difference between the two groups in this subgroup was 1.4%, which is unlikely to be clinically significant. The test for interaction was statistically significant (P = 0.14; significance level pre-specified at < 0.15), which indicates that the effect of icosapent ethyl differs depending on the CV risk category. This difference is at least partly explained by the imbalance in sample size between the CV risk category subgroups. The results did not find statistically significant differential effects between diabetes subgroup categories.

Data from the REDUCE-IT and ANCHOR trials indicated that icosapent reduced the levels of TGs, LDL-c, HDL-c, and high-sensitivity C-reactive protein (hsCRP) from baseline when compared to placebo.

Harms Results

Adverse events, serious adverse events, and withdrawals due to adverse events occurred at similar frequencies between icosapent ethyl and placebo in both studies. Atrial fibrillation occurred more frequently in the icosapent ethyl arm compared to the placebo arm (5.3% versus 3.9%, respectively) in the REDUCE-IT study but not in the ANCHOR study (0% versus less than 1%, respectively). Peripheral edema occurred more frequently in the icosapent ethyl group than in the placebo group in both REDUCE-IT (6.5% versus 5.0%, respectively) and in the ANCHOR study (1.3% versus 0.9%, respectively). Serious adverse bleeding events occurred in 2.7% in the icosapent ethyl group and 2.1% in the placebo group in REDUCE-IT; there were no fatal bleeding events in either group. There were no differences between the icosapent ethyl group and the placebo group in the proportion of adjudicated hemorrhagic stroke. Also, a higher percentage of patients in the icosapent ethyl group reported constipation compared with those in the placebo group (5.4% versus 3.6%, respectively).

Of all notable harms, only diarrhea was slightly increased in the placebo arm (11%) versus the intervention group (9%), although this was only present in the REDUCE-IT study. The rest of adverse events reported in both studies were rare (less than 3% prevalence) and similar between groups.

Critical Appraisal

Both studies were at low risk of bias regarding randomization schedules, concealment of random allocation sequences, and blinded measurements. Approximately 10% of patients in both studies were lost to follow-up, and 30% had a drug interruption for more than 30 days, but there were no differences between groups in these numbers. Hence, the risk of bias was judged as moderate. Follow-up and management of study discontinuations were appropriate, and both intention-to-treat (ITT) and per-protocol analyses presented consistent results. The ANCHOR trial did not analyze subgroups. The REDUCE-IT study was considered underpowered to obtain appropriate subgroup analyses. Generalizability of the results from REDUCE-IT is a concern, given that the benefits and harms were derived from a single, albeit large, randomized controlled trial, and only 43% of patients who underwent screening were randomized to treatment groups in the study. There is also uncertainty regarding the generalizability of the distribution of statin intensity at baseline in REDUCE-IT, although, regardless of statin intensity distribution, the median baseline LDL-c was within the target range at approximately 1.94 mmol/L (75 mg/dL) across the studied population. ANCHOR was a relatively small, short-term study that focused on evaluating changes in blood lipid profiles instead of clinical outcomes; therefore, this study is supportive in demonstrating the mechanism of action of icosapent ethyl but it does little to elucidate the clinical added value of the drug in the target population.