Real-World Insights into Evolocumab Use in Patients with Hyperlipidemia Across Five Countries: Analysis from the ZERBINI Study

This analysis of the full ZERBINI study population is the first systematic intercontinental evaluation of real-world evolocumab use and outcomes for dyslipidemia management. The current findings from this international study are largely consistent with the approved indications [21], international guideline recommendations [4‐6] and local reimbursement criteria (Table S2) for evolocumab, in that evolocumab was predominantly used in patients with ASCVD, FH or both (representing 83.7% of patients), with a significant proportion having multiple ASCVD and/or comorbid conditions. Further, 75% of patients had a LDL-C ≥ 2.7 mmol/L (≥ 104.4 mg/dL), and 82.4% were receiving a background statin and/or ezetimibe at the time of evolocumab initiation. The patient profile initiated on evolocumab in North America, South America and the Middle East reported herein and in recent continental sub-analyses [18], Roncancio et al. (submitted for publication), and [22] provides important additional diversified and global perspectives to that observed in the HEYMANS study, the largest real-world European study of evolocumab use to date [16]. Within the 1952 patients across the 12 countries studied in HEYMANS, 97% had ASCVD, FH or both, with 75% having a LDL-C of ≥ 3.17 mmol/L (≥ 122.6 mg/dL), yet only 59% were receiving a background statin and/or ezetimibe at evolocumab initiation [16], perhaps reflecting differences in local dyslipidemia management practices and/or evolocumab reimbursement criteria across countries and continents. Notably, the prevalence of MI/acute coronary syndrome was consistent between the current ZERBINI (46.9%) and HEYMANS (42%) studies [16]. Accordingly, the clinical characteristics of most patients in this study cohort are aligned with the approved indication [21] and guideline recommendations for evolocumab usage [4‐6], with it being primarily prescribed for patients at high and very high CV risk whose LDL-C remains elevated despite maximally tolerated statin and/or ezetimibe therapy.

Differences in local practice patterns and evolocumab reimbursement criteria (Table S2) may underscore differences in the specific evolocumab patient profile observed in each country herein. For instance, Canada had a higher proportion of patients with FH (55.0%) compared to the other countries (≤ 16.1%) even though evolocumab is publicly reimbursed for FH in all of them, suggesting Canadian patients with FH may be prioritized for evolocumab initiation. Also noteworthy, the prevalence of patients with ≥ 2 ASCVD conditions as defined in the current study was higher in the Middle East (Saudi Arabia, 71.0% and Kuwait, 54.3%) and Colombia (71.1%) compared to North America (≤ 42.0%). In the Middle East in particular, public access to evolocumab for ASCVD is restricted to patients with a LDL-C well above the ESC/EAS-recommended target for LLT intensification with a PCSK9 inhibitor in patients with ASCVD (> 1.4 mmol/L) [5], being > 2.6 mmol/L (> 100 mg/dL) in Saudi Arabia and > 1.8 mmol/L (> 70 mg/dL) in Kuwait. Further, the prevalence of diabetes was higher in the Middle East (≥ 58.6%) than the other countries (≤ 32.4%), resulting in a higher prevalence of diabetes in the full ZERBINI study cohort (39.1%) compared to that in HEYMANS (19%) [16]. This is consistent with estimates showing the Middle East has the highest prevalence of diabetes worldwide [23] and collectively suggests patients at very high CV risk are prioritized for evolocumab use in this region. Finally, reported SI was highest in Canada (61.8%) compared to the other countries studied (7.7–44.4%), yet in line with that reported in the HEYMANS study (60%) [16], which may reflect the requirement to document SI for evolocumab reimbursement in Canada and many European countries, and/or a lack of consensus on the definition and diagnosis of SI across countries.

Upon evolocumab initiation, at least half of all patients in each of the studied countries had LDL-C concentrations at least double those recommended for LLT intensification in international guidelines [4‐6], yet almost 25% were not receiving recommended first-line statin therapy. While these data are reassuring compared to the 57% of European patients not receiving a background statin at evolocumab initiation in the HEYMANS study [16], it points to significant worldwide care gaps in dyslipidemia management. One contributing factor may be infrequent LDL-C monitoring, despite the ESC/EAS recommendation to measure lipids within 1–3 months following LLT initiation [5]. In the current study, while only 6.7% of patients did not have a LDL-C measurement within 6 months prior to evolocumab initiation, 23.0% did not have a follow-up LDL-C measurement within 12 months post evolocumab initiation. Whether this reflects therapeutic complacency or a “fire and forget” practice pattern is not known, but it highlights the continued need for guidance and implementation of routine LDL-C measurements in vulnerable patients to identify candidates for LLT intensification and optimization. Reassuringly, in patients with ≥ 1 follow-up LDL-C measurement, most (60.9%) had ≥ 2 measurements over 12 months and half had their first measurement within 3 months, which may reflect local reimbursement requirements and/or suggest lipid monitoring is prompt and thorough when prioritized in clinical practice.

The 70.2% reduction in LDL-C observed over 12 months post evolocumab initiation in this real-world analysis validates the 65.0–67.6% reduction averaged over the dosing interval as reported in randomized clinical trials [24]. In the current study, a clinically significant reduction in LDL-C was observed during months 1–6 post evolocumab initiation, and maintained during months 7–12. Likewise, in the HEYMANS study, LDL-C was reduced by 58.1% within 3 months post evolocumab initiation and maintained for 2 years [16]. A similar LDL-C reduction of 58.4% was maintained for a median of 7 years post evolocumab initiation in 3355 European and US patients in the Open Label Extension of the Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects With Elevated Risk trial (FOURIER-OLE) [25]. Interestingly, patients originally randomized to receive evolocumab during the parent FOURIER trial had a 15–20% lower risk of major adverse CV events and 23% lower risk of CV death compared to those originally randomized to placebo, despite both groups receiving evolocumab during the OLE [25]. Hence, the current findings reinforce the real-world effectiveness of evolocumab for prompt, significant and sustained LDL-C clearance in a heterogenous patient population. Indeed, LDL-C was reduced to some extent in almost all (97.4%) patients and by ≥ 30% in 89.2% and by ≥ 50% in 76.0% of patients over 12 months. These data reassure clinically significant LDL-C reductions in the real-world setting consistent with that in the FOURIER trial wherein LDL-C was reduced by ≥ 50% in 94.7% of patients within the same duration [26]. The biological basis for suboptimal response to PCSK9 inhibition in the remaining 2.6% of patients in the current real-world study warrants further investigation but may be explained by factors relevant to routine clinical practice, including a lack of adherence to LLT, dietary or lifestyle changes, and/or human error or technical faults during sample handling or analyses, among others.

LDL-C reductions post evolocumab initiation were consistent across patients at high and very high CV risk. Particularly among patients with ASCVD without FH in the current study, 81.2% achieved a LDL-C < 1.8 mmol/L. These real-world data are comparable to the FOURIER trial wherein 87% of patients achieved a LDL-C < 1.8 mmol/L versus only 18% of patients in the placebo arm on statin monotherapy [27], reinforcing the clinical benefit of LLT intensification with evolocumab. Among patients with FH in the current study, 67.6% achieved a LDL-C < 1.8, and 75% achieved a ≥ 50% LDL-C reduction as recommended as an alternative in these patients with severe dyslipidemia [4‐6]. The ESC/EAS also recommend an alternative LDL-C goal of < 1.0 mmol/L to help prevent subsequent CV outcomes in very high-risk patients [5], which almost half of the full cohort achieved, including 56.3% of very high-risk patients with ≥ 2 ASCVD conditions. These impressively low LDL-C levels in very high-risk patients in the real-world setting were associated with a 19–21% reduction in subsequent CV event risk in evolocumab patients with higher baseline CV risk in the FOURIER trial, without increased AEs [28, 29]. Collectively, these findings accentuate CCS recommendations on the patient types expected to derive the greatest benefit from LLT intensification with a PCSK9 inhibitor [6].

While evolocumab is indicated in addition to maximally tolerated background LLT, patients initiated on evolocumab monotherapy achieved a 62.5% LDL-C reduction. This is consistent with a trial in patients with hyperlipidemia and SI wherein evolocumab monotherapy reduced LDL-C by 52.8% after 6 months, which was 36.1% more than that achieved by ezetimibe monotherapy [30]. Hence, the current real-world data are reassuring for the significant SI patient population (61.8% of evolocumab patients in Canada [18]) and 60% in Europe [16] in that international LDL-C guideline recommendations are achievable with evolocumab alone when indicated and clinically appropriate. Still, the difference in LDL-C reductions post evolocumab initiation between the full cohort (75.6% of whom were on a background statin) and patients on monotherapy is consistent with the science showing the impact of PCSK9 inhibition on LDL-C clearance is enhanced when used in combination with a statin [31]. Aligned with current international guidelines [4‐6], a low background LLT discontinuation rate for patients on evolocumab was observed in the current study, with stable statin use and low rates of ezetimibe discontinuation over 12 months. Hence, when tolerable, the clinical benefit of statin therapy and LLT intensification with a PCSK9 inhibitor and ezetimibe, being the gold standard of care in high- and very high-risk patients with LDL-C above recommended values, appears to be prioritized across the five intercontinental countries studied.

The current findings also add to the growing body of real-world evidence that patients persist on evolocumab, recognizing potential differences between studies in the methodology used [16, 17, 32]. The observed 90.2% evolocumab persistence rate over 12 months is consistent with that reported over 2 years in the European HEYMANS study (92–98%) [16] and in the US GOULD-2 study (92%) [14]. If patients continue to take their medication, as indicated in the current and other real-world studies, the substantial reductions in LDL-C observed over the course of the study are more likely to result in reduced clinical manifestations in routine care [16]. Nevertheless, international real-world evidence shows a lack of persistence to statin therapy, even among most vulnerable patients following an ASCVD event [33‐35], which may be attributed to intolerance and fear of known side effects [36, 37]. Meanwhile, underlying persistence to evolocumab may be owing to its favourable safety profile, which was observed in the current real-world study to be consistent with that in the evolocumab clinical trial program [21], FOURIER-OLE [25] and other real-world studies [14, 16‐18]. Notably, only 1 (0.2%) puncture site ecchymosis was reported in the current study, which is lower than the approximately 3–4% of injection site reactions consistently reported in past randomized and OLE trials. This may be reflective of improved patient counselling and administration skills over time. The low incidence of myalgia (0.5%) in the current study is also reassuring, especially considering 35.6% of patients had reported SI, suggesting evolocumab did not exacerbate myalgia.

This retrospective and prospective chart review study is the first intercontinental real-world evaluation of the patient profile initiated on evolocumab, as well as its effectiveness and safety. The study cohort included a representative sample of patients from five countries with varying disease states and indications for PCSK9 inhibition, resulting in a heterogenous population reflective of the real-world dyslipidemia population. A diversity of races was represented with 62.8% of patients being non-White, thus providing data on evolocumab effectiveness and safety in patients who were previously underrepresented in randomized clinical trials [27]. This analysis also included consideration of factors known to impact the effectiveness of evolocumab (e.g. background LLT, persistence), with a robust follow-up of 12 months and 60.9% of patients having ≥ 2 LDL-C measurements post evolocumab initiation. However, important limitations must be addressed, many due to the nature of the chart review study design and the fact data analysis was limited to that collected in the CRF. For one, as the sites selected in the study were anticipated to be specialist sites, they were more used to seeing high/very high CV risk and were more likely to have clinical experience with evolocumab. Hence, our results may be more indicative of clinical experience with evolocumab than is true for sites that do not see as many patients with high/very high CV risk. Furthermore, the possibility of bias resulting from inaccurate chart entry by physicians or staff at the clinical site cannot be ruled out as well. Additionally, the heterogeneity of the population of the five countries and the variability of risk factors on a small sample of patients could be considered a bias. Moreover, though patients were stratified on the basis of CV event risk, an important consideration when understanding the effectiveness of evolocumab in vulnerable patient populations, the CRF did not capture the complete definition of ASCVD according to international guidelines [4‐6], nor the timing of patients’ past MI or other CV events; hence, some patients may not have been appropriately stratified. Related to this, the limited baseline demographic and clinical data available for analysis may affect the generalizability of the findings. Further, clinical outcomes were not captured as part of this study, which rather focused on the LDL-C response. Still, causal inferences cannot be made, especially considering the frequency of LDL-C monitoring post evolocumab initiation was not structured as in clinical trials. In fact, the COVID-19 pandemic may have affected access and availability of laboratory testing for a small number of patients. Finally, persistence was self-reported by patients, which may limit its validity but the results were compatible with documented persistence of measured LDL-C lowering. Future studies should aim to overcome these data collection limitations to advance understanding of the real-world use and impact of evolocumab on dyslipidemia management in high- and very high-risk patients.