Effects of Evolocumab on Vitamin E and Steroid Hormone LevelsNovelty and Significance
Results From the 52-Week, Phase 3, Double-Blind, Randomized, Placebo-Controlled DESCARTES Study
Rationale: Vitamin E transport and steroidogenesis are closely associated with low-density lipoproteins (LDLs) metabolism, and evolocumab can lower LDL cholesterol (LDL-C) to low levels.
Objective: To determine the effects of evolocumab on vitamin E and steroid hormone levels.
Methods and Results: After titration of background lipid-lowering therapy per cardiovascular risk, 901 patients with an LDL-C ≥2.0 mmol/L were randomized to 52 weeks of monthly, subcutaneous evolocumab, or placebo. Vitamin E, cortisol, adrenocorticotropic hormone, and gonadal hormones were analyzed at baseline and week 52. In a substudy (n=100), vitamin E levels were also measured in serum, LDL, high-density lipoprotein, and red blood cell membranes at baseline and week 52. Absolute vitamin E decreased in evolocumab-treated patients from baseline to week 52 by 16% but increased by 19% when normalized for cholesterol. In the substudy, vitamin E level changes from baseline to week 52 mirrored the changes in the lipid fraction, and red blood cell membrane vitamin E levels did not change. Cortisol in evolocumab-treated patients increased slightly from baseline to week 52, but adrenocorticotropic hormone and the cortisol:adrenocorticotropic hormone ratio did not change. No patient had a cortisol:adrenocorticotropic hormone ratio <3.0 (nmol/pmol). Among evolocumab-treated patients, gonadal hormones did not change from baseline to week 52. Vitamin E and steroid changes were consistent across subgroups by minimum postbaseline LDL-C <0.4 and <0.6 mmol/L.
Conclusions: As expected, vitamin E levels changed similarly to lipids among patients treated for 52 weeks with evolocumab. No adverse effects were observed in steroid or gonadal hormones, even at very low LDL-C levels.
- adrenal steroid hormones
- cardiovascular disease
- gonadal hormones
- low-density lipoprotein cholesterol
- PCSK9 protein
- risk factors
- vitamin E
Apolipoprotein B (apoB)-containing lipoproteins are causal in the pathogenesis of atherosclerotic cardiovascular disease. Lowering low-density lipoprotein cholesterol (LDL-C) by 1.0 mmol/L (38.7 mg/dL) with statins reduces atherosclerotic cardiovascular disease event rates by ≈20%.1 Ezetimibe also lowers LDL-C and, in patients with a recent acute coronary syndrome event, cardiovascular risk.2 However, not all patients are able to achieve optimal LDL-C levels with current therapies. Failure to reach the goal is most commonly because of high LDL-C at baseline or an inability to tolerate statin therapy of sufficient intensity.
In This Issue, see p 655
Editorial, see p 662
Evolocumab (AMG 145) is a fully human monoclonal antibody to proprotein convertase subtilisin kexin type 9 (PCSK9).3 PCSK9 plays an important role in regulating LDL-C by binding to extracellular LDL receptors and targeting them for intracellular degradation.4,5 Evolocumab reduces LDL-C by 55% to 75% either alone or in combination with other lipid-lowering therapies, including statins, ezetimibe, or in combination.6,7 Because of the marked lipid-lowering efficacy of evolocumab, especially when used in combination with other lipid-lowering therapies, very low LDL-C levels rarely seen previously can be achieved.
Vitamin E is the collective term for a group of 8 lipid-soluble antioxidant compounds, of which α- and γ-tocopherol are the 2 major forms. Vitamin E protects long-chain polyunsaturated fatty acids from oxidative damage. Vitamin E deficiency is associated with neurological dysfunction, retinopathy, myopathies, and diminished erythrocyte life span.8,9 After initial absorption by enterocytes, dietary vitamin E is incorporated into chylomicrons and transported to the liver. In the liver, tocopherol transfer protein selectively recognizes and binds α-tocopherol, whereas the other vitamin E isoforms are preferentially excreted in bile. Subsequently, α-tocopherol is transferred to the plasma membrane and secreted by an ABCA1 (adenosine triphosphate-binding cassette transporter A-1)-mediated process and incorporated into nascent very low-density lipoproteins (VLDLs). Small amounts of vitamin E are transferred to high-density lipoprotein (HDL) via surface remnants during VLDL lipolysis, but the majority of vitamin E remains with the VLDL remnant.8 Approximately 50% of VLDL remnants undergo receptor-mediated endocytosis, with the remaining remnants undergoing further remodeling to form LDL. In addition to surface remnant–mediated transfer, vitamin E can also be exchanged directly from apoB-containing lipoproteins (VLDL and LDL) into HDL. HDL likely plays an important role in delivering vitamin E to tissues by scavenger receptor B1–mediated uptake of HDL particles.10 The final distribution of vitamin E into different lipoprotein classes is approximately proportional to the lipid content of each fraction.11,12 The importance of apoB lipoprotein–mediated vitamin E transport is illustrated by the severe vitamin E deficiency that is characteristic of abetalipoproteinemia.9,10
Steroid hormones are synthesized by enzymatic modification of cholesterol in steroidogenic cells, which require an adequate supply of free cholesterol. Potential sources of cholesterol include endogenous synthesis, but can also include hydrolysis of stored intracellular cholesterol esters and lipoprotein-mediated cholesterol uptake.13
Because vitamin E transport and steroidogenesis are closely associated with LDL metabolism and may potentially be affected by very low levels of serum LDL-C, we studied the effects of evolocumab on both vitamin E and steroid hormone levels.
We analyzed data from the Durable Effect of PCSK9 antibody CompARed wiTh placEbo Study (DESCARTES; NCT01516879), a 52-week study of evolocumab. The design and main results of this study have been published previously.6 Men or women aged 18 to 75 years with an LDL-C ≥2.0 mmol/L (75 mg/dL) and fasting triglycerides ≤4.5 mmol/L (400 mg/dL) after a lipid stabilization period were eligible to participate. Eligible patients were randomized in a 2:1 ratio to 52 weeks of evolocumab 420 mg or placebo given subcutaneously once every 4 weeks.
Patients with LDL-C ≥2.0 mmol/L (75 mg/dL) at screening were eligible to enter the lipid stabilization period during which they were assigned 1 of 4 possible lipid-lowering therapies (diet alone [Therapeutic Lifestyle Changes diet] or diet in combination with either atorvastatin 10 mg daily, atorvastatin 80 mg daily, or atorvastatin 80 mg daily plus ezetimibe 10 mg daily) based on National Cholesterol Education Program Adult Treatment Panel (NCEP ATP III) risk, baseline LDL-C, and statin therapy at enrollment. Patients were seen at 4-week intervals during the lipid stabilization period and were eligible for randomization if their LDL-C was at or below NCEP ATP III goal (<2.6 mmol/L [100 mg/dL] for those with coronary heart disease or coronary heart disease risk equivalents or <3.4 mmol/L [130 mg/dL] for those without coronary heart disease or coronary heart disease risk equivalents) but ≥2.0 mmol/L (75 mg/dL). Patients not at NCEP ATP III goal had their therapy increased to the next level of therapy for a further 4 weeks. This process could be repeated one further time if the LDL-C level remained above the goal. Patients receiving maximum lipid-lowering therapy (atorvastatin 80 mg daily plus ezetimibe 10 mg daily) could be randomized despite LDL-C levels above the goal. Patients with LDL-C <2.0 mmol/L (75 mg/dL) at any time during the lipid stabilization period were excluded from the study except for patients initially allocated to combination therapy with atorvastatin 80 mg daily plus ezetimibe 10 mg daily who could be downtitrated to atorvastatin 80 mg daily and randomized, provided that their LDL-C was ≥2.0 mmol/L (75 mg/dL) while maintaining NCEP ATP III goal. Figure 1 depicts the study design.
Vitamin E and Steroid Hormone Design and Analysis Populations
Vitamin E (α+γ tocopherol) was measured in the full-analysis set of the DESCARTES study population; serum, LDL, HDL, and red blood cell membrane (RCM) vitamin E were measured in a substudy (n=100) with patients randomized via the interactive voice randomization system. Patients signed a separate informed consent form to participate in the substudy. In both the full-analysis set and the substudy, patients taking vitamin E supplements at any time were excluded; all other patients with at least 1 recorded vitamin E level were included in the analysis population. Steroid hormone measurements were added to the study protocol as an amendment, and thus, baseline (day 1) levels were not available for all patients. We measured cortisol and adrenocorticotropic hormone (ACTH) in all patients. We evaluated gonadal steroid metabolism in female patients by measuring estradiol, follicle-stimulating hormone (FSH), and luteinizing hormone (LH). For male patients, we measured testosterone as well as FSH and LH. Patients receiving systemic corticosteroids were not eligible to participate in DESCARTES. In an effort to minimize confounding by menopause or exogenous estrogens, we excluded female participants from our analysis if at baseline, they were receiving hormone replacement therapy, were ≥50 years, or had baseline FSH levels ≥25 IU/L. We excluded male patients from our analyses if they were receiving testosterone supplementation or had LH levels ≥15 IU/L at baseline.
LDL-C was measured after preparative ultracentrifugation (β-quantification) in 2 central laboratories (Medpace Reference Laboratories, Cincinnati, OH and Leuven, Belgium). Both laboratories maintained Centers for Disease Control and Prevention Lipid Standardization Program Part III certification throughout the study. Vitamin E was measured at baseline and weeks 12, 24, 36, and 52 in the same central laboratories and is expressed as the sum of α+γ tocopherol. All steroids were measured by electrochemiluminescence immunoassay at baseline and weeks 12, 24, 36, and 52 in the same central laboratories. Further details are given in the Online Appendix I.
We categorized patients randomized to evolocumab according to the minimum postbaseline LDL-C recorded during the 52-week treatment period. We used the following LDL-C thresholds: <0.4 mmol/L (15 mg/dL), <0.6 mmol/L (25 mg/dL), <1.0 mmol/L (40 mg/dL), and ≥1.0 mmol/L (40 mg/dL). Patients could be counted in several categories, for example, a patient with an LDL-C of 0.5 mmol/L (20 mg/dL) would be included in the <0.6 mmol/L (25 mg/dL) and <1.0 mmol/L (40 mg/dL) groups. We analyzed patients randomized to placebo as a single group because few patients achieved LDL-C levels <1.0 mmol/L (40 mg/dL).
Within each category of patients described above, summary statistics and the change from baseline were calculated. Paired t tests were used to assess for statistically significant changes from baseline within the categories.
We compared the decrease in LDL-C from baseline to week 52 with changes in steroid hormones for the corresponding period using scatterplots and correlation coefficients. We calculated cortisol:ACTH for all patients, because a ratio of <3.0 (nmol/pmol) may suggest a diagnosis of primary hypoadrenalism.14
The primary end point of the study (percent change from baseline in ultracentrifugation LDL-C at week 52 with evolocumab when compared with the change in those receiving placebo) was calculated for the vitamin E substudy as described previously.6
Of the 905 patients randomized, 901 received at least 1 dose of study drug (full-analysis set; evolocumab, n=599; placebo, n=302) and 800 patients completed 52 weeks of therapy. Background lipid-lowering therapy was optimized based on NCEP risk to diet alone (n=111), atorvastatin 10 mg daily (n=383), atorvastatin 80 mg daily (n=218), and atorvastatin 80 mg plus ezetimibe 10 mg daily (n=189). Baseline demographic information is shown in Table 1.
Low-Density Lipoprotein Cholesterol
The placebo-corrected reduction in LDL-C with evolocumab at week 52 for all patients who received at least 1 dose of study drug was 57.0±2.1% (least square mean±SE). Evolocumab treatment was associated with an anytime postbaseline minimum LDL-C of <0.4 mmol/L (15 mg/dL), <0.6 mmol/L (25 mg/dL), <1.0 mmol/L (40 mg/dL), and ≥1.0 mmol/L (40 mg/dL) in 240 (41%), 398 (66%), 522 (87%), and 65 (11%) patients, respectively. For patients allocated to placebo, the corresponding numbers were 1 (<1%), 3 (1%), 6 (2%), and 291 (96%) patients. LDL-C at baseline and week 52 is shown in Online Table I for the full-analysis set and Online Table II for the vitamin E substudy.
Vitamin E Levels
There were 738 patients (evolocumab, n=499; placebo, n=239) in the vitamin E analysis population for the full-analysis population and 80 patients (evolocumab, n=44; placebo, n=36) in the vitamin E substudy. We excluded patients from our analyses if they received vitamin E supplementation at any time.
Vitamin E levels for the full-analysis set are shown in Figure 2. In evolocumab-treated patients, absolute vitamin E levels decreased from baseline to 52 weeks by 16±41% (6.4 μmol/L; P<0.0001), whereas normalized vitamin E levels increased significantly by a mean±SD of 19±63% (1.0 μmol/L per mmol/L of cholesterol; P<0.0001). A similar pattern was observed in evolocumab-treated patients by minimum postbaseline LDL-C (all P<0.005 for changes in absolute levels). There were no significant changes in either absolute or normalized vitamin E levels in placebo-treated patients.
Vitamin E levels among patients enrolled in the substudy are shown in Table 2 and in Online Tables III and IV. In evolocumab-treated patients, serum and LDL vitamin E levels decreased by a mean±SD of 26±18% and 61±31%, respectively (10.1 μmol/L and 8.8 μmol/L; both P<0.0001), whereas the level of vitamin E in HDL increased by 31±47% (2.1 μmol/L; P<0.005). These changes parallel the lipid changes observed with evolocumab treatment. Serum and lipoprotein vitamin E normalized for cholesterol did not change significantly. Vitamin E in RCM remained unchanged from baseline to week 52 (both absolute and normalized). We observed a similar pattern of changes when we analyzed patients according to minimum postbaseline LDL-C (Online Tables III and IV).
Absolute serum, LDL, and normalized serum and LDL vitamin E levels in placebo-treated patients enrolled in the vitamin E substudy decreased significantly from baseline to week 52 (P<0.05), whereas absolute and normalized vitamin E levels increased in HDL (P<0.005). Absolute RCM and normalized RCM changes were not significant.
In the full-analysis set, only 2 patients, both treated with evolocumab, had vitamin E levels below the reference range of 11.84 μmol/L at week 52. Vitamin E levels in 1 patient showed a persistent decline from baseline. The patient had 2 adverse events of gastroenteritis, the second of which was reported as ongoing for most of the study and may have possibly led to malabsorption. The other patient had vitamin E levels from 12.07 to 13.00 μmol/L throughout the study and 11.38 μmol/L at week 52. One additional patient, treated with evolocumab, had normal vitamin E levels until week 36 (9.1 μmol/L) but no result at week 52.
One patient, treated with evolocumab, had >1 low vitamin E level. This patient had vitamin E levels above the lower limit at baseline and week 52; the levels in between were 9.06 and 11.61 μmol/L.
In evolocumab-treated patients, cortisol increased from 377.2±156.73 nmol/L at baseline to 404.1±146.89 nmol/L at week 52 (P<0.005). The increase in cortisol was statistically significant among patients with minimum postbaseline LDL-C <1.0 mmol/L (40 mg/dL) and <0.6 mmol/L (25 mg/dL); there were no significant differences from baseline to week 52 among patients with minimum postbaseline LDL-C of <0.4 mmol/L (15 mg/dL) or ≥1.0 mmol/L (≥40 mg/dL; Figure 3; Online Table V). There were no significant changes in ACTH or the cortisol:ACTH ratio (Figure 3; Online Table V) in the evolocumab-treated patients. In particular, no patient had a ratio <3.0 suggestive of primary hypoadrenalism.14 We found no correlation between the change in LDL-C from baseline to week 52 and the corresponding change in cortisol (Online Figure I). Cortisol, ACTH, and the cortisol:ACTH ratio also showed no significant changes in placebo-treated patients.
Female gonadal hormones (estradiol, FSH, and LH) were highly variable (Figure 4; Online Table VI). There was, however, no pattern of decreasing estradiol with increasing FSH or LH in evolocumab-treated patients, even in those with the lowest postbaseline minimum LDL-C. The only statistically significant changes in female gonadal hormones were increases in LH and FSH for all analyzed evolocumab patients, which increased by 4.0±7.8 IU/mL and 5.7±15.5 IU/L, respectively, from baseline to week 52 (P<0.005). The analysis of female gonadal hormones by minimum LDL-C was supportive of that of all evolocumab-treated patients. The marked increases in FSH and LH that we observed in a few individual patients can be attributed to the onset of menopause considering that the mean±SD age was 57.2±10.4 years for women in DESCARTES, overall, and 40.8±6.6 years for those included in this analysis. There was no significant correlation between change in LDL-C and change in estradiol for evolocumab-treated patients (Online Figure II).
In male patients, testosterone levels among evolocumab-treated patients decreased nonsignificantly from baseline, irrespective of minimum postbaseline LDL-C (Figure 5; Online Table VII). Testosterone levels among placebo-treated patients increased nonsignificantly from baseline. FSH decreased by 0.2±1.36 IU/L (P<0.05), whereas LH did not change significantly (Figure 5), nor was there a correlation between change in LDL-C and change in testosterone for evolocumab-treated patients (Online Figure III).
In this study of vitamin E and steroid hormones in DESCARTES, we show that the substantial LDL-lowering observed in evolocumab-treated patients lowers absolute serum and LDL vitamin E levels but increases vitamin E levels normalized for cholesterol and does not alter tissue vitamin E levels as assessed by RCM vitamin E. No evidence of impairment of adrenal or gonadal steroid hormone synthesis was found, even in patients with extremely low (<0.4 mmol/L [15 mg/dL]) LDL-C.
Abetalipoproteinemia is characterized by low levels of apoB-containing lipoproteins and is a cause of vitamin E deficiency in humans.10 This raises the theoretical concern that the low levels of LDL-C seen in some evolocumab-treated patients may also result in vitamin E deficiency. However, the mechanisms of hypocholesterolemia in abetalipoproteinemia and evolocumab-treated patients differ fundamentally. The former is characterized by an inability to form chylomicrons and VLDL, which are essential for the absorption and distribution of vitamin E, whereas the latter is secondary to increased catabolism of apoB-containing lipoproteins, primarily LDL. Increased catabolism of apoB-containing lipoproteins is not likely to impair vitamin E absorption or distribution to tissues.
In this study, we found that absolute serum and LDL vitamin E levels decreased from baseline in evolocumab-treated patients, reflecting the substantial reduction in the lipid content of these fractions. However, serum vitamin E levels increased when normalized for cholesterol and normalized LDL vitamin E levels showed a mild increase, which did not reach statistical significance. These results confirm that the lower absolute vitamin E levels are secondary to a reduction in the number and lipid content of apoB carrier lipoproteins rather than a lowered ratio of vitamin E to cholesterol in lipoproteins. The slightly higher ratio of vitamin E to cholesterol in LDL seen in the substudy could potentially lead to enhanced protection of LDL from oxidation and additional studies to examine this should be considered. Although vitamin E levels in HDL increased in evolocumab-treated patients, this change was no longer significant when vitamin E was normalized for cholesterol. In placebo-treated patients, the results for vitamin E, both absolute and normalized, for the full-analysis population and the substudy were disparate. In the full-analysis population, we observed no clinically meaningful change in either absolute or normalized vitamin E levels. In the substudy, absolute serum vitamin E and absolute and normalized vitamin E in the LDL fraction decreased, whereas vitamin E in the HDL fraction increased. The decrease in serum and LDL vitamin E levels in placebo-treated substudy patients is unexpected as LDL-C increased by 5.9±5.1% from baseline in these patients. Possible explanations include cessation of undeclared vitamin E intake in some patients or natural variation in vitamin E levels because the substudy cohort was much smaller than the full cohort, where we observed no change in serum vitamin E in placebo-treated patients.
Previous studies of vitamin E in statin-treated patients have also documented reduced vitamin E levels secondary to statin-induced lipid reductions, but unchanged lipid normalized levels.15–17 Importantly, there was no change in the vitamin E content of RCM, which integrates vitamin E exposure and is more representative of tissue vitamin E levels.18,19 The average erythrocyte lifespan is around 115 days,20 and thus, in case of a clinically relevant reduction in the availability of vitamin E, the vitamin E content of RCM would be reduced at week 52. These observations are supported by the fact that patients with biallelic PCSK9 loss-of-function mutations have very low LDL-C but no clinical evidence of vitamin E deficiency.21,22
The question whether lowering LDL-C pharmacologically may impair the ability of the adrenal glands to synthesize cortisol, was a prominent concern especially during the early development of statins.23–32 Some of this concern was based on studies in isolated patients with abetalipoproteinemia and homozygous hypobetaliproteinemia showing mildly raised ACTH levels, normal cortisol levels, and a blunted response to prolonged ACTH stimulation.33,34 Further studies of abetalipoproteinemic patients have not always confirmed these early observations, although the endocrine evaluations undertaken have varied substantially.35,36
The impact of statins on adrenal function has been studied in multiple populations with a wide variety of statins.23,25–29,31,32,37–45 None of these studies showed clinically relevant impairment of adrenal function, secondary to statin therapy. Patients with heterozygous or homozygous familial hypercholesterolemia could also theoretically be at increased risk of adrenal dysfunction because of their reduced capacity to use LDL receptor–mediated cholesterol uptake. However, in people with homozygous familial hypercholesterolemia or with homozygous hypobetalipoproteinemia, basal levels of steroids are normal.46 In addition, studies involving both heterozygous and homozygous patients with familial hypercholesterolemia did not find any evidence of impaired adrenal function as a consequence of statin therapy.26,28 Animal studies have also showed that LDL-C or LDL receptor activity are not required for steroid hormone production47 and that absence of receptor-mediated endocytosis of cholesterol from LDL particles does not alter the cholesterol balance in extrahepatic organs, further showing that the role of cholesterol derived from LDL particles is minimal in the production of steroid hormones.48
The adrenal safety of pharmacologically lowering LDL-C to levels <1.0 mmol/L (40 mg/dL) has not been studied in detail previously because currently available lipid-lowering therapy is unable to achieve such low levels in a significant proportion of patients. Rosuvastatin is currently the most potent statin in the market.49 Two studies of adrenal function in patients treated with rosuvastatin did not find any impairment in adrenal function despite lowering LDL-C to a mean of 1.5 mmol/L (59 mg/dL) and 1.6 mmol/L (61 mg/dL), respectively.27,31 In another study of patients selected on the basis of LDL-C <1.8 mmol/L (70 mg/dL), with a mean LDL-C of 1.5 mmol/L (58 mg/dL), adrenal function was also normal.50 In our study, we observed no evidence of hypoadrenalism (decreased cortisol, increased ACTH, or a decreased cortisol:ACTH ratio) in evolocumab-treated patients, even in those achieving very low LDL-C. The statistically significant increase in cortisol among evolocumab-treated patients was an unexpected finding but is likely not of clinical significance given the small absolute increase from a mean of 377.2±156.73 nmol/L at baseline to 404.1±146.89 nmol/L at week 52. The timing of sample collection and season did not differ between the 2 time points, and this change remains unexplained. Our study shows that the adrenal gland remains functional even at very low levels of LDL-C. In keeping with this observation, patients with loss-of-function mutations in both the alleles of PCSK9 are clinically well, despite lifelong extremely low levels of LDL-C.21,22 These findings suggest that adrenal steroidogenesis is not critically dependent on LDL-mediated cholesterol delivery and that alternative pathways such as endogenous synthesis and HDL-mediated delivery can supply adequate cholesterol for steroid synthesis.13,51
Gonadal steroidogenesis in statin-treated patients has not been studied as extensively as adrenal steroidogenesis. Most studies have focused on male patients because of the lack of confounding by the menstrual cycle or hormonal therapies. Although some studies have reported minor reductions of serum testosterone in statin-treated patients, the vast majority of studies showed no change in testosterone, gonadotropins, or semen quality.24,27,29,30,37–40,45,52–54 Studies that included female patients found no change in ovarian steroidogenesis.26,38,40,54 Our study confirms these findings and extends them to patients with very low levels of LDL-C.
This is the first publication describing vitamin E and steroid hormone levels in patients treated with monoclonal antibodies to PCSK9. Strengths of our study include a double-blind and placebo-controlled design, the large number of participants, and the widely varying intensity of background lipid-lowering therapy. The study duration of 52 weeks is longer than most other studies of PCSK9 monoclonal antibodies, and should be adequate to evaluate for changes in vitamin E, particularly RCM vitamin E given the red blood cell lifespan of 115 days, and steroid hormone synthesis. The duration of the trial may not be long enough to evaluate for long-term adverse effects; moreover, the results may not apply to a young patient with familial hypercholesterolemia begun on therapy at an early age. Important limitations of our study include incomplete steroid hormone baseline data for some patients and a limited endocrine evaluation. We confined our evaluation to measuring cortisol and ACTH as well as testosterone or estradiol together with FSH and LH. We did not, for instance, evaluate urinary excretion of steroid metabolites, progesterone levels, or sex hormone–binding globulin levels. In addition, we did not assess adrenal or gonadal function dynamically with cosyntropin or human chorionic gonadotropin stimulation testing. As a consequence, we are unable to completely exclude subtle deficiencies in steroid hormone synthesis or the possibility that prolonged severe stress, as simulated with a prolonged infusion of cosyntropin, may provoke adrenal insufficiency in some patients. The fact that the cortisol:ACTH ratio was >3.0 in all patients is reassuring because a cortisol:ACTH ratio of <3.0 was recently shown to be 100% sensitive for the diagnosis of primary hypoadrenalism in 349 patients undergoing testing because of suspected adrenal or pituitary disease.14
In conclusion, we show that adding evolocumab to lipid-lowering therapies ranging from diet alone to the combination of maximal dose of atorvastatin with ezetimibe does not affect vitamin E or steroid hormone metabolism in a clinically significant fashion, despite LDL-C being lowered to previously unprecedented levels.
We acknowledge editorial assistance from Tim Peoples, MA, ELS, CMPP, of Amgen Inc.
D.J. Blom received honoraria from Amgen Inc, Sanofi-Aventis, Merck Sharp & Dohme, Pharma Dynamics, AstraZeneca, Pfizer, and Aegerion; he is a consultant/advisory board to Merck Sharp & Dohme, Sanofi-Aventis, Gemphire, and Aegerion. C.S. Djedjos, M.L. Monsalvo, I. Bridges, S.M. Wasserman, R. Scott are employees and stockholders of Amgen Inc. E. Roth received research funding from Amgen Inc; he is a speaker for Merck, AstraZeneca, and Amarin; and he is a consultant to Regeneron and Sanofi.
In June 2015, the average time from submission to first decision for all original research papers submitted to Circulation Research was 12.31 days.
The online-only Data Supplement is available with this article at http://circres.ahajournals.org/lookup/suppl/doi:10.1161/CIRCRESAHA.115.307071/-/DC1.
- Nonstandard Abbreviations and Acronyms
- adrenocorticotropic hormone
- apolipoprotein B
- Durable Effect of PCSK9 antibody CompARed wiTh placEbo Study
- follicle-stimulating hormone
- high-density lipoprotein
- low-density lipoprotein-cholesterol
- luteinizing hormone
- NCEP ATP III
- National Cholesterol Education Program Adult Treatment Panel
- proprotein convertase subtilisin kexin type 9
- red blood cell membrane
- very low-density lipoprotein
- Received June 25, 2015.
- Revision received July 15, 2015.
- Accepted July 30, 2015.
- © 2015 American Heart Association, Inc.
- Cholesterol Treatment Trialists Collaboration,
- Baigent C,
- Blackwell L,
- Emberson J,
- Holland LE,
- Reith C,
- Bhala N,
- Peto R,
- Barnes EH,
- Keech A,
- Simes J,
- Collins R.
- Lambert G,
- Sjouke B,
- Choque B,
- Kastelein JJ,
- Hovingh GK.
- Koren MJ,
- Giugliano RP,
- Raal FJ,
- Sullivan D,
- Bolognese M,
- Langslet G,
- Civeira F,
- Somaratne R,
- Nelson P,
- Liu T,
- Scott R,
- Wasserman SM,
- Sabatine MS
- Thurnham DI,
- Davies JA,
- Crump BJ,
- Situnayake RD,
- Davis M.
- Lee MK,
- Vasikaran S,
- Doery JC,
- Wijeratne N,
- Prentice D.
- Tousoulis D,
- Antoniades C,
- Vassiliadou C,
- Toutouza M,
- Pitsavos C,
- Tentolouris C,
- Trikas A,
- Stefanadis C.
- Lehmann J,
- Rao DD,
- Canary JJ,
- Judd JT.
- Illingworth DR,
- Corbin D.
- Mastroberardino G,
- Costa C,
- Gavelli MS,
- Vitaliano E,
- Rossi F,
- Catalano A,
- Barletta R,
- Guarini G.
- Krysiak R,
- Okopie B.
- Prihoda JS,
- Pappu AS,
- Smith FE,
- Illingworth DR.
- Kraemer FB,
- Shen WJ,
- Patel S,
- Osuga J,
- Ishibashi S,
- Azhar S.
- Bochem AE,
- Holleboom AG,
- Romijn JA,
- Hoekstra M,
- Dallinga-Thie GM,
- Motazacker MM,
- Hovingh GK,
- Kuivenhoven JA,
- Stroes ES.
- Kanat M,
- Serin E,
- Tunckale A,
- Yildiz O,
- Sahin S,
- Bolayirli M,
- Arinc H,
- Dirican A,
- Karagoz Y,
- Altuntas Y,
- Celebi H,
- Oguz A.
Novelty and Significance
What Is Known?
Evolocumab, an investigational PCSK9 inhibitor, markedly lowers serum levels of low-density lipoprotein cholesterol (LDL-C) and results in favorable changes in other lipids.
Evolocumab is well tolerated in short-term clinical studies.
Vitamin E transport and steroid hormone synthesis are associated with LDL metabolism.
What New Information Does This Article Contribute?
Vitamin E and steroid hormone serum levels were assessed at study enrollment and after 1 year of evolocumab treatment in a randomized study (evolocumab, n=599; placebo, n=302).
Evolocumab treatment was associated with reduced absolute vitamin E levels, but with increased levels when normalized for cholesterol levels.
Levels of cortisol, follicle-stimulating hormone, and luteinizing hormone were increased modestly in the evolocumab treatment group during the follow-up.
Thirty-one patients in the evolocumab and 2 patients in the placebo groups had low vitamin E values at any visit.
No patients had a cortisol:adrenocorticotropic hormone ratio <3.0 (nmol/pmol), which would be suggestive of primary hypoadrenalism.
Evolocumab is an investigational drug that has been demonstrated to reduce serum LDL-C to very low levels. We determined the effect of LDL lowering with evolocumab on vitamin E and steroid hormone levels in a 1-year randomized, placebo-controlled study of evolocumab plus risk-specific background lipid-lowering therapy. Among evolocumab-treated patients, vitamin E changes were reduced as were the levels of lipids, but were increased when corrected for the cholesterol levels. There were modest increases in serum levels of cortisol, follicle-stimulating hormone, and luteinizing hormone in the evolocumab group. Achievement of very low levels of LDL-C at any postbaseline visit (<0.4 or <0.6 mmol/L) did not result in substantially lower vitamin E or steroid hormone levels compared with patients achieving higher LDL-C levels. The observations were consistent across patient subgroups, suggesting that evolocumab has a consistent effect regardless of the cardiovascular risk profile or the type of background therapy.