Author + information
- Robert S. Rosenson, MD∗ ( and )
- Eva Hurt-Camejo, PhD
- ↵∗Icahn School of Medicine at Mount Sinai, 1425 Madison Avenue, MC Level, New York, New York 10029
Holmes et al. (1) investigated the association between secretory phospholipase A2-IIA (sPLA2-IIA) as a potential therapeutic target for prevention of cardiovascular disease, using observational studies between the PLA2G2A rs11573156 variant and cardiovascular events, and deductions from published data with the pan-sPLA2 inhibitor varespladib methyl. The validity of the analysis by Holmes et al. remains unclear due to inaccurate summary of data, incorrect assumptions related to the biology of sPLA2-IIA, and the pharmacological effects of varespladib (2). For brevity, the major issues include the following:
1. No reporting of the absolute values for sPLA2-IIA levels and activity, which is important because of variable results with different analytical methods and marked differences in sPLA2 levels that result from the acute phase reaction in patients with acute coronary syndrome versus patients with stable coronary heart disease;
2. Use of total sPLA2 activity as a surrogate for sPLA2-IIA activity;
3. Investigation of the PLA2G2A rs11573156 variant with messenger RNA expression even though none of the variants illustrated in Figure 2 included the variant as a genetic tool in the current analysis;
4. A marginally significant correlation between PLA2G2A rs11573156 variant and sPLA2 activity despite higher correlations between sPLA2-IIA mass and sPLA2 activity reported in Online Figure 2;
5. Reporting of percentage changes in sPLA2 levels and sPLA2 activity when absolute changes may be more important, particularly because of wide differences in baseline sPLA2-IIA levels in different cohorts;
6. A lack of clear presentation from the original studies of the reported effects of varespladib on sPLA2-IIA mass, and in fact, the table understates the effect of varespladib 500 mg/day on sPLA2-IIA mass by more than 50%, whereas the reduction is 80%; the authors concluded incorrectly that the homozygous rs11573156C allele resulted in a reduction in sPLA2-IIA mass “similar” to the effect of varespladib;
7. Biomarker effects of sPLA2 in randomized clinical trials with >500 subjects should have included only the FRANCIS (Fewer Recurrent Acute Coronary Events With Near-Term Cardiovascular Inflammation Suppression)-ACS trial on the basis of the sample size requirement, but 3 trials are reported;
8. Biomarker effects should have been reported from the PLASMA I (Phospholipase Levels And Serological Markers of Atherosclerosis) and PLASMA II (Phospholipase Levels And Serological Markers of Atherosclerosis II) trials because these trials reported results from clinically stable patients, whereas the FRANCIS-ACS trial mandated a change in statin therapy for all patients to atorvastatin 80 mg daily regardless of their prior statin regimen;
9. Varespladib is a pan-sPLA2 inhibitor with similar efficacy in lowering groups IIA and X sPLA2 with somehow lower potency against group V, despite the incorrect data cited in this report (3).
Moreover, the use of Mendelian randomization studies to deduce pharmacological effects does not account for the properties of the specific inhibitor. Specifically, varespladib is hydrophilic and may not penetrate into vascular tissues with sufficient potency to reduce intracellular effects versus the consistent effects on plasma biomarkers. Also, because varespladib methyl inhibits sPLA2-X and a recent report by Ait-Oufella et al. (4) demonstrated that overexpression of sPLA2-X is atheroprotective, nonspecific effects of varespladib as a pan inhibitor may have positive and negative effects. In addition, preliminary results from the VISTA-16 trial (Evaluation of Safety and Efficacy of Short-term A-002 Treatment in Subjects With Acute Coronary Syndrome) indicate an increase in myocardial infarctions, so the effect of varespladib was harmful (5). As discussed previously, proinflammatory pathways are redundant, and multiple anti-inflammatory pathways modulate inflammatory responses.
In conclusion, careful review of primary data and cautious conclusions must be considered in these pharmacogenetic analyses (6).
Please note: Dr. Rosenson has served as an advisory board member for Aegerion, Amgen, F. Hoffman LaRoche, GlaxoSmithKline, Janssen, LipoScience, Novartis, Regeneron, Sanofi; served as a consultant for Novartis and Sanofi; has received research support from Amgen, Boehringer Ingelheim, F. Hoffman LaRoche, and Sanofi; and is a stockholder in LipoScience and The Medicines Company. Dr. Hurt-Camejo is an employee of AstraZeneca.
- American College of Cardiology Foundation
- Holmes M.V.,
- Simon T.,
- Exeter H.J.,
- et al.
- Rosenson R.S.,
- Hurt-Camejo E.
- Rosenson R.S.,
- Fraser H.,
- Hislop H.,
- Trias J.
- Ait-Oufella H.,
- Herbin O.,
- Lahoute C.,
- et al.