Background
Heterozygous familial hypercholesterolemia (HeFH) is one of the most common inherited metabolic disorders in humans, with an estimated prevalence of 1:311 worldwide and 1:500 to 1:200 in China in the general population [
1‐
3]. HeFH is primarily caused by autosomal dominant variants in gene encoding low density lipoprotein receptor (
LDLR), apolipoprotein B (
APOB) or proprotein convertase subtilisin/kexin type 9 (
PCSK9), leading to life-long marked elevation in circulating low density lipoprotein cholesterol (LDL-C) levels and increased risk of premature atherosclerotic coronary artery diseases [
2].
Clinical practice guidelines recommend the initiation of lipid-lowering therapy as soon as possible after the diagnosis of HeFH to reduce LDL-C levels and future risk of cardiovascular diseases. Unfortunately, HeFH is largely underdiagnosed and undertreated worldwide [
4,
5]. In China, with the expert consensus on familial hypercholesterolemia (FH) diagnostic criteria formulated in 2018 and the lack of widespread FH awareness, many patients were misdiagnosed or undertreated [
6], with few patients on lipid-lowering medications achieving desired LDL-C levels [
7]. For patients whose treatment goals cannot be achieved with maximum tolerated statins plus ezetimibe, clinical practice guidelines recommend the use of PCSK9 inhibitors. Evolocumab and alirocumab, two PCSK9 monoclonal antibodies approved for the treatment of HeFH in China, demonstrated short-term and long-term reductions in LDL-C levels in both Western and Asian populations, together with positive cardiovascular outcomes [
8‐
11]. Despite the clear clinical evidence and benefit of PCSK9 inhibitors, their clinical use is limited by the high cost of treatment, necessitating the development of more drugs of this class to improve accessibility [
12‐
14].
Tafolecimab is a fully human IgG2 PCSK9 monoclonal antibody developed in China. In the phase 1a study in Chinese healthy volunteers, a single dose of tafolecimab reduced LDL-C levels up to 72%. In the phase 1b study in Chinese patients with hypercholesterolemia, tafolecimab reduced LDL-C levels up to 71.6% and by more than 50% at week 12 for all tafolecimab dose regimens. In both studies, tafolecimab was well tolerated and showed an overall favorable safety profile similar to other PCSK9 antibodies [
15]. We conducted the first randomized, placebo-controlled phase 3 clinical trial, CREDIT-2 (
Clinical
Research of
Developing PCSK9
Inhibitor as Cholesterol-lowering
Therapy in Chinese Patients with Dyslipidemia-
2), to evaluate the efficacy and safety of tafolecimab in Chinese patients with HeFH.
Methods
Study design
This was a randomized, double-blind, placebo-controlled phase 3 study conducted from December 2019 to June 2021 across 22 study centers in China. The study protocol was approved by each center’s institutional review board or independent ethics committee. This study was conducted in accordance with the principles of Declaration of Helsinki and the International Conference on Harmonization Good Clinical Practice guidelines. All participants provided written informed consent prior to screening. The results of the study were reported in adherence to the CONSORT reporting guidelines.
Participants
Patients (aged 18–80 years) diagnosed with definite or possible HeFH by Simon Broome criteria [
16] with a screening fasting LDL-C level of ≥1.8 mmol/L (with a history of arteriosclerotic cardiovascular diseases) or ≥2.6 mmol/L (without a history of arteriosclerotic cardiovascular diseases) and on a stable lipid-lowering therapy (moderate or high intensity statins except for statin intolerance, with or without ezetimibe, niacin, omega-fatty acids) for at least 4 weeks before randomization were eligible. The full inclusion and exclusion criteria are listed in Additional file
1. During the study, participants should remain on the stable dose of moderate or high intensity statins (except for those intolerant for statins) with or without ezetimibe.
Procedures
Eligible participants were randomized in a 2:2:1:1 ratio to receive subcutaneous tafolecimab 150 mg Q2W, tafolecimab 450 mg Q4W, placebo Q2W or placebo Q4W, respectively, in the 12-week double-blind treatment period. Randomization was implemented by an interactive web response system and was stratified by LDL-C levels at screening (≥ or < 4.8 mmol/L), by baseline ezetimibe use (yes/no) and by prior use of PCSK9 inhibitors (yes/no). The participants, investigators and study site personnel involved in treating and assessing participants were masked to treatment allocations.
After the 12-week double-blind treatment period, participants receiving tafolecimab continued to receive open-label tafolecimab with the previous regimens while participants receiving placebo Q2W or Q4W crossed over to receive open-label tafolecimab 150 mg Q2W or 450 mg Q4W, respectively, for 12 weeks, followed by an 8-week safety follow-up.
Fasting LDL-C (OSR6183, Beckman Coulter), HDL cholesterol (HDL-C, OSR6187, Beckman Coulter), total cholesterol (OSR6116, Beckman Coulter) and triglycerides (OSR61118, Beckman Coulter) concentrations were measured by commercial kits on a Beckman Coulter AU600 Chemistry Analyzer. Non-HDL cholesterol (non-HDL-C) concentration was calculated by subtracting HDL-C concentration from total cholesterol concentration. Very low density lipoprotein cholesterol (vLDL-C) concentration was calculated by dividing the triglyceride concentration by 5 (when triglyceride < 4.52 mmol/L) or by subtracting HDL-C and LDL-C concentration from total cholesterol concentration (when triglyceride ≥4.52 mmol/L). Apolipoprotein A1, apolipoprotein B and lipoprotein(a) concentrations were determined using a nephelometric method (OUED, OSAN and OQHL, respectively, Siemens BN ProSpec System). We measured unbound PCSK9 concentrations using an in-house developed enzyme-linked immunosorbent assay. All lipids and PCSK9 were tested in a central laboratory (WuXi AppTec, Shanghai).
DNA extracted from blood samples of all enrolled participants were sequenced by Novogene (
Beijing) for variants in all exons of
LDLR,
APOB,
PCSK9 and
LDLRAP1 gene. Pathogenicity of variants were annotated using clinical classification of the Leiden Open Variation Database (LOVD) [
17].
Endpoints
The primary endpoint was the percent change from baseline to week 12 in LDL-C levels. Secondary endpoints included proportion of participants achieving ≥50% LDL-C reductions and proportion of participants with LDL-C <1.8 mmol/L at week 12 and 24, the change from baseline to week 12 in non-HDL-C, apolipoprotein B and lipoprotein(a) levels, as well as the change from baseline to week 24 in lipid levels.
Safety endpoints were assessed by adverse events, vital signs, electrocardiogram, and laboratory measurements. Adverse events were coded and classified using Medical Dictionary of Regulatory Activities (MedDRA, version 24.0). The severity and causality of adverse events were assessed by investigators based on pre-specified criteria. Immunogenicity was assessed by detecting anti-drug antibodies and neutralizing antibodies.
Statistical analysis
Assuming a common standard deviation of 40% and a dropout rate of 10%, the sample size of 148 participants was sufficient to generate 99% power to detect ≥40% LDL-C reductions [
18] in the tafolecimab 150 mg Q2W or tafolecimab 450 mg Q4W group compared with corresponding placebo group with a 0.025 two-sided significance level (therefore 0.05 for family-wise significant level).
Efficacy endpoints were analyzed in the efficacy analysis population, defined as participants who received at least one dose of tafolecimab or placebo and had at least one post-baseline assessment. Two estimands, the on-treatment estimand and the treatment policy estimand, were used to assess treatment efficacy from different perspectives and accounted for intercurrent events differently [
19].
The on-treatment estimand assesses treatment efficacy prior to the intercurrent events like study drug discontinuation or background statin therapy adjustment. For continuous endpoints, mixed model for repeated measures method was used, with baseline value, baseline-by-visit interaction, visit, treatment, visit-by-treatment interaction and randomization stratification factors as fixed effects and unstructured covariance. For categorical endpoints, Clopper-Pearson method [
20] was used for within-group CI calculation and Mantel-Haenszel method was used for between-group CI calculation and statistical testing, and participants without any available assessment at week 12 were imputed using last observation carried forward method.
The treatment policy estimand assesses treatment efficacy regardless of any intercurrent event. For continuous endpoints, pattern-mixture model method was used for missing data imputation and analysis of covariance (ANCOVA) was used for analysis of imputed data. For categorical endpoints, statistical testing methods were the same as those of the on-treatment estimand except that participants without any available assessment at week 12 were treated as non-responders.
The primary endpoint (the percent change from baseline to week 12 in LDL-C levels) and key secondary endpoints (proportion of participants achieving ≥50% LDL-C reductions and proportion of participants with LDL-C level <1.8 mmol/L at week 12) were assessed by both on-treatment and treatment policy estimands, and controlled for type I error. The percent change from baseline to week 12 in non-HDL-C, apolipoprotein B and lipoprotein (a) levels were assessed by on-treatment estimand. Other efficacy endpoints were reported descriptively.
Safety analyses were done on the safety analysis population, defined as all participants who received at least one dose of tafolecimab or placebo. Adverse events were descriptively summarized.
Exploratory analyses were done to evaluate the effects of LDLR variant status and pathogenicity on 12-week reductions in LDL-C levels. To this end, we used the change in LDL-C levels from baseline to week 12 for the tafolecimab groups and change in LDL-C levels from week 12 to week 24 for the placebo groups.
All statistical analyses were performed with SAS version 9.4.
Discussion
HeFH is significantly underdiagnosed and undertreated in many countries and the situation is even worse in China, posing a great threat to the patients and necessitating earlier and more effective pharmacological intervention on the basis of statins and ezetimibe [
6,
12]. To the best of our knowledge, this study was the first large-scale randomized, placebo-controlled clinical trial in Chinese patients with HeFH. By demonstrating the robust efficacy and overall favorable safety of tafolecimab, this study added important evidence of the applicability of PCSK9 antibodies in Chinese patients with HeFH while picturing a comprehensive genetic profile and clinical feature of the disease population. Moreover, with the upcoming marketing of tafolecimab as the first domestic PCSK9 monoclonal antibody in China, tafolecimab is expected to increase the affordability and accessibility to PCSK9 inhibitors in Chinese patients with FH and satisfy the unmet medical needs thus improving their clinical outcomes.
The average age of HeFH diagnosis in this study (49.6 years) was consistent with those reported for East Asian or Chinese patients with HeFH [
7,
12,
21]. Many participants were previously misdiagnosed and treated as dyslipidemia and the HeFH disease was diagnosed on the study entry, further reflecting that HeFH is significantly underdiagnosed in China. Most participants (87.2%) were on stable statins of moderate intensity, which is in consistency with the fact that high-intensity statins are not well tolerated and seldom prescribed in the clinics in China [
22]. The large gap of the baseline LDL-C level (mean 4.26 mmol/L) from the guideline-recommended target level underlies the necessity of combination with additional lipid-lowering therapies [
7,
12].
In Chinese HeFH participants with higher baseline LDL-C levels and higher proportion of moderate-intensity statin use, the LDL-C-lowering effect of tafolecimab was generally similar to those of other approved PCSK9 antibodies observed in western population. In RUTHERFORD-2 trial, evolocumab 140 mg Q2W and 420 mg monthly achieved placebo-adjusted LDL-C reductions of 59.2% and 61.3% at week 12, respectively, in patients with HeFH [
9]. In ODYSSEY FH trials, alirocumab dosed up to 150 mg Q2W achieved placebo-adjusted LDL-C reductions ranging from 51.4% to 57.9% at week 24 [
8]. In our study, placebo-adjusted mean reductions in LDL-C levels at week 12 were 57.4% with tafolecimab 150 mg Q2W and 61.9% with tafolecimab 450 mg Q4W. In all these studies, the proportion of patients achieving LDL-C <1.8 mmol/L were significantly higher in patients receiving PCSK9 inhibitors than those on statin or ezetimibe alone. Indeed, according to the European Atherosclerosis Society (EAS) Familial Hypercholesterolemia Studies Collaboration (FHSC) global registry, use of PCSK9 inhibitors among patients with HeFH was associated with greater odds of having LDL-C lower than 1.8 mmol/L [
12]. However, only 2.4% of the patients were taking PCSK9 inhibitors in combination with statin or ezetimibe [
12]. Our study provided the first evidence that robust and persistent reductions in LDL-C levels can be achieved with PCSK9 antibodies in Chinese patients with HeFH, demonstrating a potential new option for the great unmet clinical needs.
Another benefit of PCSK9 inhibition is reductions in lipoprotein(a) levels, a LDL-like lipoprotein associated with cardiovascular diseases. Cardiovascular outcome trials of alirocumab and evolocumab both suggested that lipoprotein(a) level lowering with PCSK9 inhibition was associated with a greater cardiovascular risk reduction beyond LDL-C lowering alone, indicating additional benefits from further reduction of lipoprotein(a). Current PCSK9 antibodies lowered lipoprotein(a) levels by approximately 25%, but were least effective in patients with high lipoprotein(a) level (>50 mg/dl), with a 14% reduction [
23]. In this study, 34.1–43.3% placebo-adjusted reductions in lipoprotein(a) levels were achieved in overall population and 32.3–35.7% placebo-adjusted reductions were achieved in participants with baseline lipoprotein(a) levels ≥50 mg/dl. Of note, given the lower lipoprotein(a) levels in Chinese population [
8] and skewed distribution of lipoprotein(a) levels, together with the suboptimal mass units used in this study, the comparison of lipoprotein(a)-lowering effect with other PCSK9 inhibitors is premature and the potential of tafolecimab warrants further investigation.
Consistent with previously reported genetic profiles of HeFH, potential causative variants were most commonly detected in
LDLR gene. By exploratory analysis aggregating 12-week efficacy from the tafolecimab groups in the double-blind treatment period and the placebo groups in the open label period, we observed that participants with single variants in
LDLR responded equally well to those without
LDLR variants. This result, in accordance with data from other PCSK9 inhibitors, indicates that this medication class can effectively reduce lipids levels in patients with only one copy of
LDLR full functional [
9,
24,
25].
Seven compound heterozygotes with two
LDLR variants were enrolled in this study. Among five participants receiving tafolecimab in the double-blind treatment period, four had markedly compromised reductions in LDL-C levels (Additional file
2: Fig. S3). Of note, most
LDLR variants identified in these compound heterozygotes were missense variants and annotated as likely pathogenic according to LOVD. One participant with two likely pathogenic
LDLR variants had a 49.3% LDL-C reduction from baseline at week 12. Obviously, functional annotations of
LDLR variants, especially missense variants, based on which effects of PCSK9 inhibitors were established, remain a significant challenge.
In this study, with 84.5% of participants at very-high cardiovascular risk, tafolecimab demonstrated a favorable risk-benefit profile, with safety profile similar to those identified in previous studies and those of alirocumab and evolocumab [
8,
9,
11,
15,
26]. Serious adverse events were possibly related to underlying medical conditions of the participants, further reflecting late diagnosis of HeFH in China and high risk for cardiovascular diseases.
A major limitation of this study was the short treatment duration. Although improvements on lipid profiles were evident after 24-week treatment with tafolecimab, a future study is warranted to evaluate its long-term efficacy and cardiovascular benefits in patients with HeFH. Another limitation was that all participants were Chinese, limiting the generalizability of the results to patients of other races or ethnicities.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.