Effectiveness of proprotein convertase subtilisin/kexin‐9 monoclonal antibody treatment on plasma lipoprotein(a) concentrations in patients with elevated lipoprotein(a) attending a clinic

Abstract Background Lipoprotein(a) (Lp[a]) is a causal risk factor for atherosclerotic cardiovascular disease (ASCVD). Proprotein convertase subtilisin/kexin‐9 monoclonal antibodies (PCSK9mAbs) can lower Lp(a) levels in clinical trials, but their effects in patients with elevated Lp(a) in clinical practice remain unclear. Aims To investigate the effectiveness and safety of PCSK9mAbs in lowering plasma Lp(a) in patients with elevated Lp(a) concentrations in a lipid clinic. Methods This was an open‐label study of 53 adult patients with elevated Lp(a) concentration (≥0.5 g/L). Clinical, biochemical, and safety data were collected before and on treatment with evolocumab or alirocumab over a mean period of 11 months. Results Treatment with a PCSK9mAb resulted in a significant reduction of 0.29 g/L (−22%) in plasma Lp(a) concentration (p<.001). There were also significant reductions in low‐density lipoprotein‐cholesterol (LDL‐C) (−53%), remnant‐cholesterol (−12%) and apolipoprotein B (−43%) concentrations. The change in Lp(a) concentration was significantly different from a comparable group of 35 patients with elevated Lp(a) who were not treated with a PCSK9mAb (−22% vs. −2%, p<.001). The reduction in Lp(a) concentration was not associated with the corresponding changes in LDL‐C, remnant‐cholesterol, and apolipoprotein B (p>.05 in all). 7.5% and 47% of the patients attained a target concentration of Lp(a) <0.5 g/L and LDL‐C <1.8 mmol/L, respectively. PCSK9mAbs were well tolerated, the common adverse effects being pharyngitis (9.4%), nasal congestion (7.6%), myalgia (9.4%), diarrhoea (7.6%), arthralgia (9.4%) and injection site reactions (11%). Conclusion PCSK9mAbs can effectively and safely lower plasma Lp(a) concentrations in patients with elevated Lp(a) in clinical practice; the impact of the fall in Lp(a) on ASCVD outcomes requires further investigation.

and Mendelian randomization studies consistently demonstrate that elevated Lp(a) is an independent and causal risk factor for atherosclerotic cardiovascular disease (ASCVD), especially in women; [1][2][3][4]6,7 this also applies to high-risk patients treated with statins. [8][9][10] However, a lack of effective treatments for lowering Lp(a) remains a major challenge in the management of patients with elevated Lp(a). [2][3][4]11 Proprotein convertase subtilisin/kexin-9 (PCSK9) is a key regulator of hepatic receptors involved in the metabolism of apoB-containing lipoproteins. 12 Inhibition of PCSK9 by a monoclonal antibody (mAb), such as evolocumab and alirocumab, can effectively reduce plasma concentrations of LDL-cholesterol (LDL-C) by up to 60%. 13 Hence, PCSK9mAbs are used for the treatment of elevated LDL-C in patients with familial hypercholesterolaemia (FH) and/or ASCVD on background maximum tolerated dose of statin, or in such patients with statin intolerance. 14 Recent clinical trials and meta-analyses also demonstrate that PCSK9mAbs can lower Lp(a) by up to 30% and this may partly account for the benefits of these agents in reducing ASCVD events. [15][16][17][18] However, current guidelines do not recommend the use of a PCSK9mAb to specifically lower elevated Lp(a). [2][3][4]14 While inhibition of PCSK9 is efficacious in lowering Lp(a) in clinical trials, [15][16][17][18] its effectiveness and safety have not yet been formally evaluated in real-world clinical practice particularly in patients with elevated Lp(a). [19][20][21] In the present study, we assessed the effectiveness and safety of PCSK9mAbs in lowering Lp(a) in patients with elevated Lp(a) attending a lipid clinic. We further investigated whether the change in plasma concentration of Lp(a) was associated with the corresponding changes in plasma LDL-C and other lipids and lipoproteins.

| Selection of patients
Fifty-three patients with elevated Lp(a) from the Lipid Disorders Clinics, Department of Cardiology, Royal Perth Hospital were included in the study. Patients were selected for having a plasma Lp(a) concentrations ≥0.5 g/L. 2 Fifty patients were on evolocumab according to pharmaceutical benefits scheme (PBS) criteria and three patients were on alirocumab in a product familiarization program. 22 These patients were selected from a large pool of clinic patients who were receiving treatment with these agents.

| Study design
The study was a sequential, open-label study of the effect of PCSK9mAbs in patients with elevated Lp(a), carried out over a period from 2015 to 2020. Forty-seven patients were on 140 mg fortnightly (n = 44), monthly (n = 2) or six-weekly (n = 1) and three on 420 mg monthly subcutaneous injections of evolocumab; two received 150 mg fortnightly and one 75 mg monthly injections of alirocumab.
Patients attended regular clinic visits, where they were reviewed by a nurse specialist and/or a consultant physician and received best standard of care; clinical and biochemical data were collected and compared before and on treatment with evolocumab or alirocumab. For comparison, 35 patients with elevated Lp(a), attending the clinic and did not receive evolocumab or alirocumab, were also studied as a control group. At the time of recruitment, the 35 patients had plasma LDL-cholesterol levels above guideline recommended target of <1.8 mmol/L but were not eligible for receiving a PCSK9mAb according to the PBS criteria in Australia. Clinical audit approval was granted by the Royal Perth Hospital (Quality Activity 29009).

| Clinical data
The following clinical data were recorded: Age, gender, ethnicity, body mass index (BMI), blood pressure; personal history of coronary artery disease (CAD), cerebrovascular event (a stroke or transient ischemic attack), peripheral artery disease (PAD), hypertension, diabetes, or kidney disease; family history of ASCVD; current medications. CAD was defined as a history of either; (i) myocardial infarction, (ii) angioplasty, or (iii) coronary artery bypass grafting. 23 Hypertension was defined as systolic and diastolic blood pressure >140 or >90 mmHg, respectively or on anti-hypertensive medication. Diabetes was defined as a fasting glucose concentration of ≥7.0 mmol/L, random glucose of ≥11.1 mmol/L, glycated hemoglobin >6.5% or use of anti-diabetic medication (insulin or oral hypoglycaemic agents), and obesity as BMI ≥30 kg/m 2 . Renal function was determined by the estimated glomerular filtration rate (eGFR) using the modification of diet in renal disease (MDRD) equation. Similar information to the above was collected in the 35 clinic patients with elevated Lp(a) who were not treated with PCSK9mAb. Perth Hospital, as previously described. 23

| Definition of statin intolerance
Statin intolerance was defined as the inability (complete or partial) to tolerate at least two different statins (one statin at the lowest starting daily dose and another statin at any daily dose) due to clinically important adverse effects, such as musculoskeletal symptoms and markedly elevated plasma creatinine kinase and/or hepatic aminotransferase. 24

| Definition of treatment adverse events
Adverse events (AEs) during PCSK9mAb therapy were defined as general systemic symptoms and injection site reactions. General systemic symptoms were flu-like (pharyngitis, nasal congestion), musculoskeletal (myalgia, back pain, arthralgia), gastrointestinal (diarrhea, constipation, nausea, abdominal discomfort/pain), and other (headache and fatigue). Injection site reactions were defined as a cluster of erythema, pain, bruising, swelling, induration, rash, or pruritus around the injection sites. The severity of the above was assessed by the numerical rating scale on a 0-10 scale, any score of >4 defined as the presence of a clinically significant AE. 25

| Clinical review of PCSK9mAb therapy
Patients first attended a nurse-managed clinic for education and training on the administration, dosage, and frequency of the subcutaneous injections, as well as on the storage requirements of the injectate and potential AEs. Typically, follow-up reviews were done via telephone, or with intermittent clinic visits as required. Patients were telephoned by the specialist nurse every fortnightly or monthly, based on the injection schedules to elicit and discuss clinical progress and AEs. The first post-treatment fasting blood test for lipids, lipoproteins, and other biochemical analytes were carried out 6 weeks after the first injection, followed by blood tests at 3-4 monthly intervals; blood tests were not done during acute illness. significantly correlated with a liquid chromatography-mass spectrometry method. 27 Interassay coefficient of variation of all measurements were <6%.

| Statistical analyses
Statistical analyses were carried out using the SPSS Statistics (Version 25; Armonk, New York; IBM Corp). Data were presented as mean ± SD unless otherwise specified. The Shapiro-Wilk test was used to determine whether variables were normally distributed.   Table 1 shows the clinical and biochemical characteristics of the patients studied. They were mostly late middle-aged, Caucasians and overweight, with normal blood pressure, glucose, and renal function.

| Patient characteristics
The majority of the patients had a family history of CAD; half had a history of CAD and one-fifth a history of cerebral event or PAD.

| Adverse effects of PCSK9mAb therapy
The common adverse effects reported by the patients treated with    29 We also reported that in statin-treated patients with elevated Lp(a), PCSK9mAb treatment resulted in an absolute reduction of plasma Lp(a) concentration by −0.23 g/L (−23%), which is mediated by increased catabolism of Lp(a) particles. 30 This is consistent with the notion that PCSK9 inhibition could enhance Lp(a) removal by upregulating hepatic LDL receptor activity. 12,30,31 Previous data also suggest that PCSK9 could directly bind to Lp(a) particles in the F I G U R E 2 Associations between the absolute changes in plasma Lp(a) concentration with PCSK9mAb treatment and corresponding changes in plasma concentrations of low-density lipoproteincholesterol (LDL-C) (A), REM-C (B) and apoB (C), respectively circulation. 32,33 Hence, another possible mechanism of action of PCSK9mAb treatment is the formation of a circulating triple immune complex of PCSK9mAb with PCSK9-bound Lp(a), followed by removal of this complex from the circulation. However, it may only represent a very small percentage of total Lp(a) that is complexed. Despite clinical trials usually having very restricted inclusion and exclusion criteria, the magnitude of Lp(a) reduction with PCSK9mAb treatment in the present real-world study was comparable to those reported in clinical trial settings. [15][16][17][18]30 Moreover, we also found a significant trend to an increase in absolute reductions of plasma Lp(a) levels in patients from the lowest quartile to the highest quartile (Table 3). This is consistent with our correlational analysis that the absolute reduction, but not relative reduction, in Lp(a) with PCSK9mAb treatment was directly dependent on the pre-treatment Lp(a) concentration. This observation could be due to a regression-to-the-mean effect, or to individuals with higher baseline Lp(a) concentration having a truly greater absolute Lp(a) reduction, the lack of association with proportional reduction in Lp(a) being expected on mathematical grounds. 34

| Association with LDL-cholesterol, remnantcholesterol and apoB
PCSK9 is a key regulator of LDL and other hepatic receptor involved in the metabolism of apoB-100 containing lipoproteins. 12 However, we found that the reduction of Lp(a) was not significantly associated with the fall in LDL-C with PCSK9mAb treatment, consistent with previous studies showing a discordant association between the reductions in Lp(a) and LDL-C with PCSK9mAb treatment. [35][36][37][38] This implies that the LDLR may not play a major role in the catabolism of Lp(a) particles. However, the lack of association between the reductions in Lp(a) and LDL-cholesterol may be due to the limited number of patients in the study. Hence, our inferences need to be interpreted with caution and confirmed in a study with a larger sample size. Previous experimental data suggest that Lp(a) can be preferably catabolized via non-LDLR pathways, such as VLDL receptor, LDLR-related protein 1, megalin/gp330, scavenger receptor class B type 1, and plasminogen receptors. 39

| Safety and adverse effects
Recent randomized controlled trials (RCTs) and meta-analyses have favorable safety and good tolerability profile of PCSK9mAbs. 13,[16][17][18][19]28,30,34,35 These findings may not be fully applicable to clinical practice which encompasses patients with diverse comorbidities and multiple therapies. There are also limited data on the safety of PCSK9mAb in real-world settings. [19][20][21]40 In the present study, the most common reported adverse effects were nasopharyngitis, diarrhea, myalgia, arthralgia, as well as injection-site reactions. These findings are compatible with data from clinical trials. 13,17,18,28,30 Although 40% of patients reported at least one AE, the severity of those AEs was insufficient to stop treatment.

| Study strengths and limitations
Our sample size was relatively small, but the findings were comparable to previous studies in real-world settings of PCSK9mAb. 9,19,20 We also focused on patients with elevated Lp(a). We did not employ a formal placebo-controlled design to study the effects of

| Clinical implications
Several clinical practice guidelines recommend measuring plasma Lp(a) concentration in patients with high-risk of ASCVD. [2][3][4]14 However, current guidelines do not specify the degree to which Lp(a) needs to be lowered to prevent ASCVD events. The recent FOURIER and ODYSSEY Outcomes trials demonstrate that the reduction in Lp(a) could partly mediate the benefit of PCSK9mAb treatment in reducing ASCVD risk in high-risk patients, particularly in those with elevated Lp(a). 17,18,28 In the ODYSSEY trial, each 5 mg/dl reduction in Lp(a) with alirocumab was associated with a 2.5% relative reduction in total ASCVD event. 28 This is generally in agreement with a population study showing that a reduction of Lp(a) concentration by 50 mg/dl lowered ASCVD risk by 20% in a secondary prevention setting. 43 The effect of Lp(a) on the risk of ASCVD events appears to be proportional to the absolute reduction in Lp(a), not the percent reduction. 28 The mean reduction in Lp(a) of 0.29 g/L with a PCSK9mAb could be approximately translated into 12%-15% reduction of ASCVD. 28,43 However, this speculation remains to be demonstrated.
Patients with elevated Lp(a) who also have CAD and FH, as in our sample population, remain at high residual risk of ASCVD despite statin treatment. [2][3][4]8 Recent studies suggest that lowering Lp(a) by approximately 1 g/L could reduce major ASCVD events by 22%. 44 As demonstrated in the present study, PCSK9mAbs cannot achieve this degree of lowering of Lp(a) concentrations. Attaining such an effective reduction in Lp(a) requires the use of RNA-based therapies that target the apo(a) gene transcripts. 45

| CONCLUSION
In clinic patients with elevated Lp(a), most of whom had CAD and FH, PCSK9mAb treatment safely and significantly lowered the plasma concentration of Lp(a), with a greater reduction in LDL-C and apoB.
Whether the incremental reduction in Lp(a) translates into a decrease in ASCVD events remains unclear. The aggregate fall in atherogenic apoB-containing lipoproteins, including Lp(a), with PCSK9mAb, is likely due to increased hepatic clearance of these lipoproteins 46,47 and probably mediates the principal cardiovascular benefits of this treatment. 18,19,28,48,49 ACKNOWLEDGMENT Anindita Chakraborty acknowledges the support of the Australian