Evidence is mounting that statin therapy can prevent the progression of coronary atherosclerosis (hardening and narrowing of the arteries supplying the heart) in people living with HIV, according to the results of two randomised clinical trials reported last week at the Conference on Retroviruses and Opportunistic Infections (CROI 2015) in Seattle, USA.
In the first study, the use of atorvastatin (Lipitor) reduced the overall volume of plaque deposits in the coronary arteries in people with HIV, including the 'high risk morphology' plaques that are particularly vulnerable to acute rupture leading to heart attacks, stroke and sudden cardiac death. The second study of statin therapy in people on antiretroviral therapy found that rosuvastatin (Crestor) halted progression of carotid intima-media thickening, a surrogate marker for vascular and atherosclerotic disease.
While it has yet to be demonstrated whether these changes on statin therapy will translate into fewer cardiac events and less death among people living with HIV, statin use has been shown to reduce cardiovascular disease (CVD) and provide a survival benefit in the HIV-negative general population. However, the risk factors for CVD are somewhat different for people living with HIV, so the question is being addressed by the REPRIEVE study, a very large multicentre clinical trial which is now getting underway.
CVD in people with HIV
"Even as rates of death and mortality related to HIV have decreased with use of more potent antiretroviral therapy, rates of cardiovascular disease, including myocardial infarctions, stroke and sudden cardiac death remain increased among HIV patients – stubbornly so – and are a leading cause of morbidity and mortality," Steven Grinspoon of Harvard Medical School told conference attendees during a plenary talk on the final morning of the conference. "But unfortunately there is only a limited understanding of the mechanisms and treatment strategies for cardiovascular disease in HIV."
Several cohort studies have suggested that the rates of heart attack and coronary heart disease are up to two times greater in HIV-positive people than in the HIV-negative population. Both traditional and non-traditional risk factors contribute to the increased risk. For instance, smoking is more common among people living with HIV than in the general population. According to recently published analysis of 17,995 people with HIV in the ART Cohort Collaboration, those who smoke had a greater than six-fold higher rate of CVD-related mortality than those who do not smoke.
However, even when traditional risk factors are excluded, there is clear evidence of an increased risk of CVD in people living with HIV. An analysis of the Veterans Aging Cohort Study (VACS) found that the hazard ratio (HR) of acute myocardial infarction (MI) was almost 50% higher in HIV-positive compared to HIV-negative veterans, even after adjusting for traditional risk factors, comorbidities and substance use.
Earlier in the antiretroviral therapy (ART) era, much of the blame for the increase in heart attacks among people living with HIV was pinned on specific antiretroviral drugs, such as abacavir (Ziagen, also in Kivexa), although this association has been controversial due to conflicting evidence. More recent data suggest that effective ART has a positive effect on CVD risk, most likely by improving immune function and reducing immune activation.
According to Grinspoon, recent studies support an emerging mechanistic paradigm in which persistent immune activation and inflammation contribute to a unique presentation of atherosclerotic disease in people living with HIV. Studies using coronary computed tomography angiography (CTA) have shown that people with HIV have an increased prevalence of non-calcified coronary plaque deposits in their arteries, with more high-risk plaque morphology features that have been associated with heart attacks, stroke and sudden death.
Chronic inflammation at the arterial surface, and perhaps altered cholesterol metabolism, facilitate the formation of these plaque deposits. The inflammation is believed to be the result of T-cell and monocyte activation as a consequence of ongoing viral replication or microbial translocation (the escape of gut bacteria into the systemic circulation). Studies have found evidence of alterations in gut flora and damage to the gut and impaired mucosal integrity even among people with undetectable viral load on ART. Indeed, much of the damage to the gut may have taken place before people went on treatment, as it begins soon after infection.
Preventing and treating heart disease in people living with HIV will thus require an intervention that addresses both traditional and immune-related risk factors – and Grinspoon believes statins may do just that. Statins lower low-density lipoprotein (LDL) 'bad' cholesterol to a similar extent in HIV-positive and HIV-negative people, but Grinspoon noted that studies in the general population have demonstrated a greater effect on lowering CVD events than could be explained by the reduction in LDL alone. Some of the benefit may be due to statins' anti-inflammatory effects, as statins have been shown to reduce arterial inflammation and reduce monocyte activation and trafficking.
Although statins have been used to lower LDL cholesterol in people on ART, studies are only now beginning to evaluate the effects of statin therapy on the course of CVD in people living with HIV. The results of two of the first of these studies were reported shortly after Grinspoon’s talk.
In addition to reducing CVD events, intravascular ultrasound studies have shown that statin therapy can lead to regression of plaques in the general population. However, "no studies have yet directly assessed the effects of statins on coronary plaque in people living with HIV," said Janet Lo, a colleague of Grinspoon’s at Massachusetts General Hospital and Harvard Medical School.
So she and her colleagues conducted a randomised, double blind, placebo-controlled trial to see whether atorvastatin would prevent coronary plaque progression and decrease vascular inflammation in HIV-positive people without signs or symptoms of clinical CVD and optimal or near optimal LDL cholesterol levels (< 3.36 mmol/L /130mg/dl). However, participants were required to have subclinicial, coronary atherosclerosis, with one or more plaques as assessed by coronary computed tomography angiography. Eligible participants then underwent fluorodeoxyglucose positron emission tomography (FDG-PET) scanning and those with arterial inflammation, demonstrated by aortic target-to-background ratio greater than 1.6, were included in the study.
A total of 40 qualifying participants were then randomised to one year of treatment with atorvastatin or placebo to determine the effects on coronary atherosclerotic plaque. Most (about 80%) were men, the average age was around 50 years and nearly 30% were smokers
There were no statistically significant differences in baseline characteristics of participants in the two arms. Framingham risk scores (which predict the risk of a cardiac event in the next ten years based on the presence of traditional CVD risk factors) were quite low and the mean LDL level was around 3.23 mmol/L (125mg/dL). The majority of patients had well-controlled HIV disease and C-reactive protein (a marker of systemic inflammation) was not elevated at baseline.
During the first three months, atorvastatin was dosed at 20mg per day, and if the participant tolerated atorvastatin with no laboratory safety abnormalities, the dose was increased to 40mg for the remaining nine months of the study. Both placebo and active drug groups also received lifestyle and nutritional counselling according to National Cholesterol Education Program guidelines.
The analysis of the CT images was performed by a single experienced cardiac imaging specialist who was blinded to treatment group and clinical status of the patients. The main endpoint was plaque volume and the plaque was also analysed for high-risk morphology features such as low attenuation, spotty calcification and positive remodelling index.
As expected, LDL cholesterol decreased in the atorvastatin group by 0.98 mmol/L (-38 (+/- 29) mg/dl) while increasing in the placebo group by +0.28 mmol/L (+11 (+/- 21) mg/dl). Overall plaque volume measured on coronary CTA decreased by 4.7% with atorvastatin. Notably, there was an 18% increase in plaque volume within only one year in the placebo arm. Also, atorvastatin reduced non-calcified plaque volume by 19.4% but worryingly, in the placebo arm, non-calcified plaque volume increased over just one year by 20.4%.
Similarly, while 80% of participants in the placebo arm experienced progression of plaque over the one-year period, 65% of those taking atorvastatin experienced plaque regression. Three patients in the placebo group developed clinically significant progression of stenosis (valve narrowing). There were no changes in calcium scores (used to predict the risk of future cardiac events). However, there was an increase in high-risk morphology features of plaques in the placebo arm compared with a slight regression of those features in the atorvastatin arm.
Due to technical limitations, statin effects on arterial inflammation by FDG-PET could not be adequately assessed. However, the researchers also looked at vascular inflammation using another technique: levels of lipoprotein-associated phospholipase A2, an enzyme secreted by inflammatory cells that has been implicated in atherosclerosis. They found that levels significantly decreased with atorvastatin, consistent with findings in other trials, including the SATURN-HIV study.
Glucose and haemoglobin A1C levels did not worsen in the atorvastatin group. Although the study was not powered to specifically assess inflammatory endpoints, a decrease was observed in several markers of inflammation of immune activation – for instance C-reactive protein and soluble CD14. Estimated GFR (a marker of kidney function) worsened within the placebo group but not in the atorvastatin group, although the between-group comparisons were not statistically significant.
Rates of adverse events were low and similar between the atorvastatin and placebo arms.
The second study, presented by Grace McComsey of Case Western Reserve University in Cleveland, reported findings from the SATURN-HIV study, showing that rosuvastatin therapy could be similarly beneficial using somewhat different measures of CVD health and risk.
SATURN-HIV was a 96-week double-blind, randomised clinical trial of 10mg daily rosuvastatin versus placebo among 147 people with HIV on stable ART with LDL cholesterol <3.36 mmol/L (130mg/dl). Baseline characteristics were well balanced between the two arms. Most (80%) were men and about two-thirds were smokers. Half were taking protease inhibitors at study entry and the median CD4 count was over 600 cells/mm3. At baseline, carotid IMT was similar in both arms. At least a third had carotid plaque (33% on rosuvastatin vs 43% on placebo) or detectable coronary artery calcification (33% vs 40%, respectively).
To be eligible, participants had to have evidence of heightened T-cell activation (CD8+CD38+HLA-DR+ >19%) or increased inflammation (high sensitivity C-reactive protein >2mg/l). Randomisation was stratified by protease inhibitor use (given the potential for drug-drug interactions) and by coronary artery calcium (CAC) score.
Ultrasound was used to assess intima-media thickness (IMT) of the common carotid artery and presence of carotid plaque. Gated-cardiac CT scans were used to measure the CAC score.
As expected, there were significant reductions in LDL cholesterol in the rosuvastatin compared with the placebo group. Progression of mean carotid IMT was slower than expected, but progressed significantly in the placebo group over 96 weeks while remaining unchanged in the rosuvastatin group. The differences between the two arms were more marked among participants with baseline coronary artery calcification. There was no change in the prevalence of carotid plaques by study arm. The overall prevalence of detectable coronary artery calcification seemed to increase somewhat in the rosuvastatin arm over the course of the study, but the change in CAC score in the subset of patients with coronary artery calcification at baseline decreased significantly on rosuvastatin.
Changes in carotid IMT occurred independently of protease inhibitor use, LDL level, HOMA-IR (a measure of insulin resistance), CD4 count, C-reactive protein or T-cell activation. Greater declines in carotid IMT were seen among participants with higher IMT at baseline and with higher IL-6 and a greater percentage of CD14dimCD16+ patrolling monocytes (both markers of inflammation and immune activation).
A total of 28 participants – nine in the rosuvastatin arm, the rest in the placebo arm – prematurely discontinued the study, but only three, including two rosuvastatin recipients, did so for reasons possibly having to do with the study drug. These include one individual with myalgia and high creatinine phosphokinase, and one possible exacerbation of neuropathy symptoms.
Moving forward: more accurate risk prediction needed
The effects of atorvastatin and rosuvastatin in people living with HIV seem to be of a similar magnitude to what has been seen in the general population, but both Lo and McComsey concluded that further study is needed to see whether statins and other anti-inflammatory agents reduce the number of cardiac events in people with HIV.
The REPRIEVE study, which uses pitavastatin (Livalo) – which is differently metabolised and has no known drug-drug interactions with antiretrovirals – should be able to provide this answer, and furthermore may also shed light on the mechanisms by which statins work in people with near optimal LDL levels.
But the study will take over 7000 participants and several years to reach a conclusion. It is not clear that everyone living with HIV has time to wait. In the meantime, it will be critical to better identify which people with HIV are truly at risk of cardiac events – and to treat them if necessary.
Grinspoon S Cardiovascular disease in HIV patients: an emerging paradigm and call to action. 2015 Conference on Retroviruses and Opportunistic Infections (CROI), Seattle, abstract 134, 2015.
Lo J et al. Statin therapy reduces coronary noncalcified plaque volume in HIV patients: a randomized controlled trial. 2015 Conference on Retroviruses and Opportunistic Infections (CROI), Seattle, abstract 136, 2015.
Longenecker CT et al. (McComsey G presenting) Rosuvastatin arrests progression of carotid intima-media thickness in treated HIV. 2015 Conference on Retroviruses and Opportunistic Infections (CROI), Seattle, abstract 137, 2015.