Lipoprotein(a) Lowering—From Lipoprotein Apheresis to Antisense Oligonucleotide Approach

Round 1

Reviewer 1 Report

This is a well written review article on approaches to lower Lp(a).  The authors have performed a thorough search of literature and critically reviewed relevant references.  The readers will have clear understanding of the recent developments in the field and contributes to their knowledge.

 

Minor comment: Reference 1: Emerging Risk Factor, C is not a name of author for this reference.  Authors should go through thoroughly and correct such mistakes.

 

Author Response

We thank the reviewer. All the references have been checked.

Reviewer 2 Report

This paper thoroughly describes the current knowledge about Lp(a) and especially about the effects of interventions like LA and PCSK9 inhibitors.

Unfortunately, the pdf-file which the reviewer had downloaded did not contain the Figures.

Some sentences to drugs which clearly are not accepted for reducing Lp(a) could be eliminated (i. e. Aspirin).

Tredaptive had been withdrawn from the market after it was observed in 2 studies that the reduction of Lp(a) was not inducing any effect on cardiovascular outcome – these studies should be cited. Moreover, adverse effects were also observed.

The chapter about PCSK9 inhibitors and their effect on Lp(a) is written quite well. But the authors should stress that nevertheless an elevation of Lp(a) is no accepted indication for initiating these drugs.

With respect to the antisense oligonucleotide against apo(a) it should be written that the Phase III study will start in 2020. The fate of this new drug will depend on two factors: 1. Outcome data for cardiovascular events, 2. Costs of the drug. In the experience of the reviewer the current therapeutic approach (optimization of all other risk factors like LDL-C, hypertension, no smoking etc.) in patients with high Lp(a) is rather effective with respect to preventing new events, it will not be easy to demonstrate a superiority of the new drug.

Minor comments

Page 2 Line 105: of LDL-C should be roughly 100 mg/dL (2.6 mmol/L) – this number evidently pertains to Lp(a) which is never given in mmol/l but in nmol/l

Page 4 Line 155: with possible probable or – possibly

Page 4 Line 168: Convincing evidence support – supports

Page 5 Line 199: available LA techniques are categorized as selective (immune adsorption, dextran sulfate adsorption, heparin precipitation, cascade filtration and polyacrylamide adsorption) – with all LA methods other proteins are eliminated like fibrinogen, immunoglobulins and others, but in a different degree; immune adsorption may be the most specific, but even with this methods other proteins get lost

The data listed in Table 1 are to be regarded as preliminary – the acute reductions rates very much depend on the volume treated which can be easily modified

Page 5 Line 213: Thus, despite resulting in dramatic immediate Lp(a) reductions (70-213 75%), weekly or biweekly apheresis result in much lower mean interval concentrations (20-40%) – this statement is not correct with respect to mean interval concentrations: they are lowered less effectively when comparing with acute reduction rates

Page 6 Line 229: The possible application of selective Lp(a) apheresis also appears to offer a promising approach to the prevention of Lp(a) associated CV risk. – unfortunately this statement has up to now not been proven in a scientific study comparing usual LA methods which reduce Lp(a) levels in a similar degree but also have an effect on LDL-C concentrations. It could be an advantage to avoid this LDL-C effect in patients whose LDL-C levels are rather low (i. e. by PCSK9 inhibitor therapy). But again this should be the subject of a prospective study.

Page 6 Line 261: DNA-antisense mipomersen – is this correct?

Page 6 Line 260: A recent kinetic finding relative to no effect in Lp(a) production with the DNA-antisense mipomersen suggests that the hepatic availability of apoB is unlikely to be rate- limiting for assembly and production of Lp(a) particles – as far as the reviewer knows mipomersen is decreasing Lp(a)

Page 7 Line 292: Table 2 – rosuvastatin should be added

Page 8 Line 346: also that in acute syndrome patients – acute coronary syndrome

Page 11 Line 401: Aspirin. In studies on the use of aspirin for CV prevention, an Lp(a) reduction of 26-55% was 401 found. [134]. This effect was possibly linked to a suppressed Lp(a) liver gene transcription [135]. – the authors should add a comment on the clinical significance of these findings – in the practice nobody accepts this indication for Aspirin

Page 12 Line 456: The DE LAVAL study showed.. – This study did not look at Lp(a) at all. The ODYSSEY ESCAPE study should be discussed – in this study an increase of Lp(a) concentration under alirocumab was observed in LA patients.

Page 13 Line 480: LDL-C concentration of  100 mg/dL and an Lp(a) concentration of 100 mg/dL – usually it is recommended to measure the Lp(a) particle concentration (instead of mass), the unit will be nmol/l. The authors should discuss this fact and explain what to do when you want to calculate corrected LDL-C – the first step would be to convert nmol/L into mg/dl for Lp(a), for several reasons this is a doubtful approach

Page 13 Line 486: the care of patients require further studies – requires

Author Response

This paper thoroughly describes the current knowledge about Lp(a) and especially about the effects of interventions like LA and PCSK9 inhibitors.

 

Unfortunately, the pdf-file which the reviewer had downloaded did not contain the Figures.

Figure 1 has been now uploaded as a graphical abstract

 

Some sentences to drugs which clearly are not accepted for reducing Lp(a) could be eliminated (i. e. Aspirin).

The sections pertaining aspirin and N-acetylcysteine have been deleted

 

Tredaptive had been withdrawn from the market after it was observed in 2 studies that the reduction of Lp(a) was not inducing any effect on cardiovascular outcome – these studies should be cited. Moreover, adverse effects were also observed.

We thank the reviewer for raising this issue. The impact of niacin-laropiprant on Lp(a) leves has been described in the text and findings reported in the new table 1. The text now reads as follows “A sub-analysis of the HPS2-THRIVE study showed that, at 1 year, allocation to niacin-laropiprant resulted in an overall Lp(a) mean reduction of 12.2 nmol/L that became 33.8 nmol/L in the group with Lp(a) baseline levels ³ 128 nmol/L. Overall, the percentage reduction was 31 (95%CI 28%-33%), attenuating to 36% and 18% across quintiles by increasing baseline [58].”

 

The chapter about PCSK9 inhibitors and their effect on Lp(a) is written quite well. But the authors should stress that nevertheless an elevation of Lp(a) is no accepted indication for initiating these drugs.

A sentence has been added in the conclusion which now reads as follows “While the use of PCSK9 inhibitors has led to a mean fairly modest reduction in Lp(a) levels (20-25%), there is no currently accepted indication for the use of these biosynthetic compounds to treat hyperlipoproteinemia(a) [150]. However, in combination with LA, the use of PCSK9 inhibitors decreases the need for apheresis as most patients with HeHF and other forms of hypercholesterolemia respond very well to this therapy.”

 

With respect to the antisense oligonucleotide against apo(a) it should be written that the Phase III study will start in 2020. The fate of this new drug will depend on two factors: 1. Outcome data for cardiovascular events, 2. Costs of the drug. In the experience of the reviewer the current therapeutic approach (optimization of all other risk factors like LDL-C, hypertension, no smoking etc.) in patients with high Lp(a) is rather effective with respect to preventing new events, it will not be easy to demonstrate a superiority of the new drug.

This concept has been addressed in the conclusion section which now reads as follows “The estimated completion date is March 2024 (NCT04023552). However, future safety and cost-effectiveness studies are required to establish the role of these new agents in clinical practice.”

Minor comments

 

Page 2 Line 105: of LDL-C should be roughly 100 mg/dL (2.6 mmol/L) – this number evidently pertains to Lp(a) which is never given in mmol/l but in nmol/l

We thank the reviewer. This mistake has been amended accordingly.

 

Page 4 Line 155: with possible probable or – possibly

We thank the reviewer. This mistake has been amended.

 

Page 4 Line 168: Convincing evidence support – supports

We thank the reviewer. This mistake has been amended.

 

Page 5 Line 199: available LA techniques are categorized as selective (immune adsorption, dextran sulfate adsorption, heparin precipitation, cascade filtration and polyacrylamide adsorption) – with all LA methods other proteins are eliminated like fibrinogen, immunoglobulins and others, but in a different degree; immune adsorption may be the most specific, but even with this methods other proteins get lost

The following lines have been added in the section pertaining the lipoprotein apheresis “LA can modify also a number of pathological processes associated with CVD: improves markers of vascular inflammation, and decreases fibrinogen, E-selectin, vascular cellular adhesion molecule-1, intercellular adhesion molecule-1, monocyte chemoattractant protein-1, lipopolysaccharide binding protein, matrix metalloproteinase and tissue inhibitor of metalloproteinase [133-136].”

 

The data listed in Table 1 are to be regarded as preliminary – the acute reductions rates very much depend on the volume treated which can be easily modified

The following sentence has been added “However, the percentage changes reported in table 1 reflect not only the intrinsic efficiency of each method to remove plasma lipoproteins but also differences in the volume of blood or plasma treated and the extent of haemodilution caused by the anticoagulant used [121]. Typically, from 4 to 6 L exchanges of plasma are carried out for 2–4 h weekly or biweekly [122].”

 

Page 5 Line 213: Thus, despite resulting in dramatic immediate Lp(a) reductions (70-213 75%), weekly or biweekly apheresis result in much lower mean interval concentrations (20-40%) – this statement is not correct with respect to mean interval concentrations: they are lowered less effectively when comparing with acute reduction rates

The sentence has been rewritten and now reads as follows “Lp(a) rebounds at a slower rate than LDL-C but with a similar monoexponential function [102]. Thus, despite an acute decrement of 70-75%, regular apheresis can translate into a significant interval mean Lp(a) reduction between 25 and 40% [124]. Specifically, depending on the Lp(a) baseline and the selected interval, a biweekly apheresis generally results in a much lower interval mean reduction (20%) compared to a weekly procedure (36%) [105,125]. This bulk of reduction persisted also in patients undergoing long-term LA: mean pre-apheresis levels of Lp(a) were reduced by 22% after 1 year and by 19% after three years [126,127]”.

 

Page 6 Line 229: The possible application of selective Lp(a) apheresis also appears to offer a promising approach to the prevention of Lp(a) associated CV risk. – unfortunately this statement has up to now not been proven in a scientific study comparing usual LA methods which reduce Lp(a) levels in a similar degree but also have an effect on LDL-C concentrations. It could be an advantage to avoid this LDL-C effect in patients whose LDL-C levels are rather low (i. e. by PCSK9 inhibitor therapy). But again this should be the subject of a prospective study.

This concept has been discussed in the conclusion as follows “However, it should be taken into consideration that these studies all suffer from potential confounding such as the lack of ability to rule out the effect of LA on other drivers of events such as LDL or fibrinogen. Moreover, to determine the contribution of Lp(a) apheresis, a study with a controlled prospective randomized design, enrolling only patients with elevated Lp(a), should be planned [123].”

 

Page 6 Line 261: DNA-antisense mipomersen – is this correct?

It has been amended

 

Page 6 Line 260: A recent kinetic finding relative to no effect in Lp(a) production with the DNA-antisense mipomersen suggests that the hepatic availability of apoB is unlikely to be rate- limiting for assembly and production of Lp(a) particles – as far as the reviewer knows mipomersen is decreasing Lp(a)

This information was reported in the conclusion. The text reads “Relative to mipomersen, although the potential benefit in reducing Lp(a) was between 20% and 50% [152]”

 

Page 7 Line 292: Table 2 – rosuvastatin should be added

Data on the JUPITER study have been added

 

Page 8 Line 346: also that in acute syndrome patients – acute coronary syndrome

Amended

 

Page 11 Line 401: Aspirin. In studies on the use of aspirin for CV prevention, an Lp(a) reduction of 26-55% was 401 found. [134]. This effect was possibly linked to a suppressed Lp(a) liver gene transcription [135]. – the authors should add a comment on the clinical significance of these findings – in the practice nobody accepts this indication for Aspirin

This section has been deleted

 

Page 12 Line 456: The DE LAVAL study showed.. – This study did not look at Lp(a) at all. The ODYSSEY ESCAPE study should be discussed – in this study an increase of Lp(a) concentration under alirocumab was observed in LA patients.

The ODYSSEY ESCAPE study has been described. The text now reads as follows “The effect of PCSK9 inhibition on the frequency of standard LA treatments was the aim of the ODYSSEY ESCAPE study [127]. Although LA was discontinued in 63.4% of patients on alirocumab and the rate was reduced at least 50% in 92.7% of patients, any additive effect of alirocumab on top of LA on Lp(a) levels was not found [128].”

 

Page 13 Line 480: LDL-C concentration of  100 mg/dL and an Lp(a) concentration of 100 mg/dL – usually it is recommended to measure the Lp(a) particle concentration (instead of mass), the unit will be nmol/l. The authors should discuss this fact and explain what to do when you want to calculate corrected LDL-C – the first step would be to convert nmol/L into mg/dl for Lp(a), for several reasons this is a doubtful approach

The text now reads as follows “[14]. These observations are to be combined with the difficulties in establishing a generally accepted mode of measurement [15]. First of all, available assays report results in mass (mg/dL) instead of concentration in nmol/L and a direct conversion may not be possible because of the variable number of repeated units in different apo(a) isoforms. Secondly, absolute differences in Lp(a) measurements for single samples have been reported to almost 80 mg/dL [16]. Third, Lp(a) values vary if samples used are fresh or have been frozen for prolonged periods of time.”

 

Page 13 Line 486: the care of patients require further studies – requires

Thank you. It has been corrected.

Reviewer 3 Report

In this review article MF Greco et al. describe some features of lipoprotein(a) epidemiology and treatment of patients with elevated plasma concentrations. There are several points that need attention by the authors.

 

GENERAL COMMENTS

 

Lp(a) is currently in the focus of interest in numerous lipid laboratories and thus a great number of review articles have been published in recent years. The Journal of Lipid Research even published a thematic issue dealing entirely with Lp(a). The present manuscript summarizes several aspects of Lp(a) that certainly are by far not complete and it is not evident on what basis the highlights were chosen. In addition it appears that the cited references were chosen rather arbitrarily. 

 

SPECIFIC COMMENTS

 

Given the fact that numerous reviews on Lp(a) have been published recently the authors should concentrate here more on features related to reduce plasma Lp(a). Thus the content of pages 1 until the middle of 4 may be condensed and focused on points that are pertinent for treatment.

 

Also the sequence of therapeutic measures should probably be changed. I suggest to start with fibrates, followed by nicotinic acid, PCSK9 inhibitors, LA and anti-sense therapy. At the end all the other medications that may be of little effect may be mentioned briefly.

 

Literature: Whenever the authors fail to cite the original literature they should mention: ”cited in…”  For example, Ref. 89 is not the original work on the action of FXR on Lp(a) concentrations.

 

Finally, the manuscript would gain significantly if a native English speaking person would revi

Author Response

Lp(a) is currently in the focus of interest in numerous lipid laboratories and thus a great number of review articles have been published in recent years. The Journal of Lipid Research even published a thematic issue dealing entirely with Lp(a). The present manuscript summarizes several aspects of Lp(a) that certainly are by far not complete and it is not evident on what basis the highlights were chosen. In addition it appears that the cited references were chosen rather arbitrarily.

The topic has been chosen according to a pre-inquiry.

 

SPECIFIC COMMENTS

 

Given the fact that numerous reviews on Lp(a) have been published recently the authors should concentrate here more on features related to reduce plasma Lp(a). Thus the content of pages 1 until the middle of 4 may be condensed and focused on points that are pertinent for treatment.

We agree with the reviewer and we have tried to cut these sections. Chapter 1.1 has been deleted and the others revised.

 

Also the sequence of therapeutic measures should probably be changed. I suggest to start with fibrates, followed by nicotinic acid, PCSK9 inhibitors, LA and anti-sense therapy. At the end all the other medications that may be of little effect may be mentioned briefly.

About the sequence, one of us (CRS), a specialist in lipids who has visited over 15,000 patients, did not feel at ease mentioning fibrates first. So, we started with statins (albeit not effective!), then the historical nicotinic acid (with which we have some experience), PCSK9 antagonists, miscellaneous agents (fibrates, vitamin C, hormones), then lipoprotein apheresis and finally the upcoming antisense products.

 

Literature: Whenever the authors fail to cite the original literature they should mention: ”cited in…” For example, Ref. 89 is not the original work on the action of FXR on Lp(a) concentrations.

We thank the reviewer for pointing this out. The original article by Chennamsetty (J Clin Invest 2011, 121, 3724-3734) was quoted. Moreover, in the section pertaining to Nicotonic acid, the study by Chennamsetty on transgenic APOA mice was added.

 

Finally, the manuscript would gain significantly if a native English speaking person would revi

The text has been edited by Cinzia Nasuelli, a Cambridge English C2 Proficiency (CPE) person.

Round 2

Reviewer 3 Report

The manuscript has improved significantly. I now understand the rational of the sequence of drugs chosen