Evaluating the Effect of Atorvastatin Exposure and Vitamin D Levels on Lipid Outcomes in Persons with HIV-1 with Suppressed HIV-1 RNA and LDL cholesterol <130 mg/dL

Abstract

Introduction:

Cardiovascular disease (CVD) has become a leading cause of morbidity and mortality among persons with HIV (PWH). Atorvastatin is known to reduce cardiovascular risk. Here, we (1) compared atorvastatin concentrations between different boosted protease inhibitors (PI) and with lipid outcomes, and (2) compared pre-atorvastatin 25-OH Vitamin D levels with atorvastatin concentrations and with lipid outcomes, in PWH with suppressed HIV-1 RNA and LDL-C<130.

Methods:

A5275 was a randomized, double-blind, placebo-controlled crossover study of atorvastatin in virally suppressed PWH with fasting low-density lipoprotein cholesterol < 130 mg/dL. We analyzed results over the 20 weeks of active atorvastatin treatment. Atorvastatin was initiated at 10 mg daily and increased to 20 mg daily after 4 weeks if no findings of toxicity. Atorvastatin trough concentrations were measured at week 20. Participants took combination antiretroviral therapy (ART) that included a boosted PI throughout.

Results:

Overall (n=67), 70% were male; median age was 51 years. There was no apparent association of atorvastatin trough concentrations with pre-atorvastatin Vitamin D levels (r=0.01, p=0.9), or by boosted PI (p=0.20). Median pre- to post- atorvastatin change was −39.0 mg/dL in fasting total cholesterol, −40.4 ng/mL in lipoprotein-associated phospholipase A2 (LP-PLA2) and −13.8 U/L in oxidized low-density lipoprotein (oxLDL), with all changes negatively correlated with atorvastatin trough concentrations (r= −0.19, −0.09, −0.21; p≥0.096).

Conclusions:

No apparent associations between pre-atorvastatin Vitamin D levels and outcomes were observed (all p>0.70). In virologically suppressed PWH, higher atorvastatin concentrations were marginally associated with greater decreases in lipid outcomes.

Introduction

Cardiovascular disease (CVD) has become a leading cause of morbidity and mortality among persons with HIV (PWH).^1,2 Statins reduces cardiovascular risk by decreasing low-density lipoprotein cholesterol (LDL-C) and inducing plaque regression.³AIDS Clinical Trial Group A5275 reported that atorvastatin did not significantly decrease levels of soluble or cellular biomarkers of immune activation and inflammation in PWH with virologic suppression on ART (antiretroviral therapy) but resulted in robust reductions in LDL-C, oxidized low-density lipoprotein (oxLDL), and lipoprotein-associated phospholipase A2 (LP-PLA2), biomarkers associated with cardiovascular risk.⁴ This analysis of the A5275 study examined correlates of atorvastatin concentrations, as well as the relationship between atorvastatin concentrations, vitamin D levels and study outcomes.

Some studies have reported low levels of vitamin D have been associated with worse prognosis in CVD. Vitamin D has been suggested for use of prevention and/or treatment, but no conclusive evidence as been found.⁵ Vitamin D levels may also influence the lipid-lowering effects of atorvastatin.⁵ Studies have shown that those patients who are on statin therapy have higher serum vitamin D concentrations (specifically 25(OH)D).⁶ One study showed that adequate vitamin D levels may be required for atorvastatin to reduce lipids as those with vitamin D deficiency did not respond to low or high dose atorvastatin while patients with higher levels showed expected reductions in cholesterol.⁷ Another study showed that patients who took supplemental vitamin D had an enhanced effect of atorvastatin therapy on total and LDL-C.⁸

Another consideration in PWH is that some boosted protease inhibitors (PI) increase serum concentrations of statins through inhibition of cytochrome P450 3A4 isoenzyme.⁹ Atorvastatin can increase Vitamin D levels which has led to speculation that the improvement in endothelial function and regression of atherosclerosis associated with statins may be related to both the increase in Vitamin D as well as lipid-lowering properties.¹⁰Here, we examined atorvastatin concentrations between different boosted PIs, and associations between lipid outcomes and both atorvastatin concentrations and pre-atorvastatin 25-OH Vitamin D levels in PWH with suppressed HIV-1 RNA and LDL-C <130 mg/dL.

Materials and Methods

A5275 was a randomized, double-blind, placebo-controlled study of atorvastatin in virally suppressed participants with fasting LDL-C < 130 and ≥ 70 mg/dL. We analyzed the 20 weeks of active atorvastatin treatment. Atorvastatin was initiated at 10mg daily. After 4 weeks, if no findings of toxicity, the dose was increased to 20 mg daily. Atorvastatin trough concentrations were measured at week 20. All participants took combination antiretroviral therapy (ART) that included a boosted PI throughout the entire study period.

There were two treatment arms. Arm A had atorvastatin from week 0 to week 20, then placebo from week 24 to week 44; arm B had placebo from week 0 to week 20, then atorvastatin from week 24 to week 44. This analysis focused on the 20 weeks of atorvastatin (study weeks 0 to 20 for Arm A; study weeks 24 to 44 for Arm B).

Atorvastatin trough concentrations (therapeutic range for trough 1.0 +/− 0.7 ng/mL after 40mg dose), for the active parent drug, were obtained 20 weeks after initiation of active atorvastatin treatment.¹¹ Participants were to hold their daily atorvastatin dose on this visit until after the blood draw, approximately 24 hours after the prior dose, and reported the time of their last atorvastatin dose. Atorvastatin concentration were analyzed using liquid chromatography/tandem mass spectrometry.⁴

Atorvastatin trough concentrations at 24 hours post-dose were estimated; these were calculated based on the observed concentrations obtained between 17–28 hours post-dose and then applying a first-order elimination rate corresponding to a half-life of 14 hours to correct concentrations to represent exposure at 24 hours.¹² Concentrations below assay lower limit (0.1 ng/mL) were analyzed as the lowest rank. Week 20 trough concentrations were compared between groups with nonparametric Wilcoxon or Kruskal-Wallis tests. Spearman correlations assessed associations of atorvastatin trough concentrations with pre- to post-atorvastatin changes in lipid outcomes, and pre-atorvastatin 25-OH Vitamin D levels (therapeutic range for Vitamin D 30.0–100.0 ng/ml).⁵ All statistical tests are two-sided at the 0.05 nominal level of significance, without adjustments for multiple testing.

Results

Among the 72 participants who completed the active treatment period and had lipid outcome data, 1 had lab error and 4 had atorvastatin trough blood sample drawn outside of target time from last dose window, thus these five participants were excluded. Henceforth, there was a total of 67 participants included in the analysis.

The analysis population was 70% male, median (Q1, Q3) age 51 (42, 55) years, 51% Black Non-Hispanic, 28% Hispanic (regardless of race), and 21% White Non-Hispanic. Almost all (99%) had HIV-1 RNA below the assay lower limit of 40 cp/mL at study entry; 1 participant on LPV/r had entry HIV-1 RNA of 58 copies/mL. Median (Q1, Q3) baseline CD4 cell count was 545 (340, 679) cells/mm³.

Table 1 describes atorvastatin trough concentrations at 20 weeks of treatment. Overall, median (Q1, Q3) 24-hour estimated atorvastatin trough concentration was 2.57 (1.20, 4.90) ng/mL. There were no apparent differences among the boosted-PIs in atorvastatin trough concentrations [median (Q1, Q3) = 2.39 (1.25, 4.58) for ATV/r, 3.85 (1.24, 8.40) for DRV/r, and 2.15 (0.90, 3.80) for LPV/r; p=0.20]. No differences in atorvastatin trough concentrations were evident by age, sex or race/ethnicity (details not shown).

Table 1:

Atorvastatin Trough Levels by Boosted PI Regimen at Week 20 post treatment start, in participants receiving 20mg daily atorvastatin

Trough levels		Boosted PI Regimen
		ATV/r (N=35)	DRV/r (N=19)	LPV/r (N=10)	P- value
24h Estimated Atorvastatin (ng/mL)	Median (Q1, Q3)	2.39 (1.25, 4.58)	3.85 (1.24, 8.40)	2.15 (0.90, 3.80)	0.20
	Min, Max	0.10, 21.32	0.10, 64.16	0.10, 4.90
	< Assay Lower Limit (0.10 ng/mL)	1 (3%)	1 (5%)	1 (10%)

Note: ATV/r: ritonavir-boosted atazanavir. DRV/r: ritonavir-boosted darunavir. LPV/r: ritonavir-boosted lopinavir. Not included in analysis were one participant on fosamprenavir/ritonavir, one on both ATV/r and DRV/r, and one participant who stopped taking ART drugs two weeks prior to the atorvastatin trough blood draw.

Table 2 summarizes pre- to post-atorvastatin changes in lipid outcomes, showing 30–40% decreases for fasting total cholesterol, LP-PLA2 and oxidized LDL. The 24-hour estimated atorvastatin trough concentration was marginally associated with pre- to post-atorvastatin change in oxLDL (r = −0.21, p=0.096), change in fasting total cholesterol (r = −0.19, p=0.13) and change in LP-PLA2(r = −0.09, p=0.47). These negative correlations correspond to higher drug levels being associated with greater decreases in lipid outcomes.

Table 2:

Summary of Lipid Outcomes

		Pre-atorvastatin		Pre-to post- atorvastatin change
Lipid Outcomes	N	Median	95% CI for Median	Median (Q1, Q3)	95% CI for Median
Fasting Total Cholesterol (mg/dL)	66	101.5	(96.0, 112.0)	−39.0 (−58.0, −25.0)	(−46.0, −32.0)
LP-PLA2 (ng/mL)	67	135.7	(123.6, 150.9)	−40.4 (−68.6, −16,4)	(−51.4, −26.0)
Oxidized LDL (u/L)	67	44.9	(41.8, 48.1)	−13.8 (−18.7, −8.7)	(−15.8, −12.3)

Note: LP-PLA2: Lipoprotein-associated phospholipase A2. LDL: Low-density lipoprotein.

Overall, median (Q1, Q3) pre-atorvastatin 25-OH Vitamin D levels were 26 (16, 35) ng/mL. Median (Q1, Q3) 25-OH Vitamin D was 29 (15, 37) ng/mL among those on ATV/r, 24 (16, 31) ng/mL among those on DRV/r, and 27 (17, 32) ng/mL among those on LPV/r. No apparent trends between pre-atorvastatin Vitamin D levels and 24-hour estimated atorvastatin trough concentration (r=0.01, p=0.9), or pre- to post-atorvastatin changes in fasting total cholesterol, LP-PLA2, or oxLDL were observed (all p>0.70).

Discussion

This trial, in virologically suppressed PWH with LDL-C <130 mg/dL on a boosted-PI, showed that higher atorvastatin concentrations were marginally associated with greater decreases in lipid outcomes. Associations with other parameters, including pre-atorvastatin Vitamin D levels and the specific boosted-PI, were not apparent.

Vitamin D is a known inducer of CYP3A4, which results in heightened enzyme activity and metabolism of certain statins, and possible reduction of atorvastatin toxicity after vitamin D supplementation.⁵ Vitamin D deficiency can lead to insufficient CYP enzyme activity and thus can increase toxicity of CYP-metabolized statins.⁵

Vitamin D has a possible inverse relationship with lipid biomarkers, as vitamin D deficiency has been associated with increased blood levels of inflammatory markers.^{,13,14,15, 16} Some observational studies have shown an inverse relationship between vitamin D levels and CVD.^13,14 However, our study did not find an association between statin trough concentration and vitamin D level.

Regarding non-boosted PI administration of atorvastatin, a study showed that 40mg of atorvastatin lowered total cholesterol by 38% and 80mg of atorvastatin lowered total cholesterol by 46%.¹² Our study showed a 39% decrease in total cholesterol with a boosted PI regimen and 20mg of atorvastatin as the expected concentration of atorvastatin is equivalent of 40mg due to the drug interaction. Although integrase inhibitors have become the preferred class for initial ART, PIs remain an integral component of the 2021 initial ART recommendations for specific clinical scenarios including pregnancy and are a mainstay in those treatment experienced.¹⁷

Statin-mediated decreases in lipoprotein biomarkers suggest possible reduction in CVD risk in virologically suppressed individuals, which is an ongoing strategy being tested.¹⁴ Our study contributes to that growing literature. Nonetheless, the effect of decreasing lipoprotein biomarkers on cardiovascular events in this population should be investigated further.

Data Availability:

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.