Cardiovascular Drugs and Therapy

, Volume 21, Issue 3, pp 221–225

Design and Rationale of the ARBITER 6 Trial (Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol)-6-HDL and LDL Treatment Strategies in Atherosclerosis (HALTS)


    • Cardiology Service, Department of MedicineWalter Reed Army Medical Center
    • Cardiology ServiceWalter Reed Army Medical Center
  • Mark A. Turco
    • Center for Cardiac and Vascular ResearchWashington Adventist Hospital
  • Allen J. Taylor
    • Cardiology Service, Department of MedicineWalter Reed Army Medical Center

DOI: 10.1007/s10557-007-6020-8

Cite this article as:
Devine, P.J., Turco, M.A. & Taylor, A.J. Cardiovasc Drugs Ther (2007) 21: 221. doi:10.1007/s10557-007-6020-8



Recent evidence on the use of statin therapy indicates the potential for ultra-low levels of LDL-C to provide greater protection from recurrent coronary heart disease events. Guidelines for the treatment of lipid disorders were revised to indicate that an LDL-C treatment goal of 70 mg/dl was optional (NCEP ATPIII). In these same guidelines, low levels of HDL-C are also suggested but not specifically proscribed as a target of therapy. Recently ARBITER 2 (Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol 2) has provided the first evidence of the potential of raising HDL-C with extended release niacin when added to statin monotherapy in secondary prevention. However, whether this approach would be superior to a strategy in which lower concentrations of LDL-C are targeted is unknown.

Materials and methods

ARBITER 6-HALTS ( HDL and LDL Treatment Strategies) will be a randomized, parallel group, open-label study comparing HDL-C and LDL-C focused strategies of lipid treatments for their effects on atherosclerosis. Up to 400 subjects will be assigned to either intensified LDL-C lowering therapy with ezetimibe or HDL-C raising therapy with extended-release niacin. The primary endpoint is the mean change in the intima-media thickness of the common carotid artery after 14 months. Secondary endpoints include the change in lipid values and lipid subfractions, drug discontinuation due to adverse effects, change in quality of life, and a composite endpoint consisting of all major adverse cardiovascular events.


ARBITER 6-HALTS will guide clinicians on whether a lipid treatment strategy of raising HDL-C or further LDL-C reduction is superior in the secondary prevention of coronary heart disease.

Key words

atherosclerosiscarotid intima-media thicknessextended-release niacinezetimibelipid lowering therapy

1 Rationale for the ARBITER 6 trial

Regression or stabilization of atherosclerosis progression is regarded as a surrogate marker for the clinical efficacy of cholesterol-lowering therapies [1]. Treatments to lower cholesterol using combination hypolipidemic therapies (eg, colestipol and niacin) [24] or fixed dosing of monotherapy with HMG-CoA reductase inhibitors can regress atherosclerosis in relation to the degree of cholesterol reduction; however, the dominant effect of these medications is to delay the progression of atherosclerosis [5].

In patients with known CHD, whether a treatment strategy of targeting further LDL-C reduction is superior to a strategy focused on raising HDL-C for retarding atherosclerotic progression remains unknown. The HATS (HDL Atherosclerosis Treatment Study) trial using the combination of simvastatin (10–20 mg/day) and niacin (2–4 g/day) demonstrated stabilization of coronary atherosclerosis [6]. However, this was a placebo controlled study and therefore could not determine the additive effect of niacin to the statin. Recently, ARBITER 2 and ARBITER 3 have extended these data by demonstrating that the addition of niacin to statin monotherapy provides superior effects on atherosclerosis stabilization and regression [7, 8] (see Fig. 1).
Fig. 1

Change in CIMT across 12 months in placebo(statin monotherapy), 12 months of extended release niacin (ERN) + statin monotherapy and 24 months of extended release niacin + statin monotherapy (P < 0.001)

Parallel to these investigations, several studies have shown that an alternative strategy of lowering LDL-C further to ultra-low values could also provide clinical benefit over moderate LDL-C lowering [9, 10]. Recently, ezetimibe, a drug which blocks the intestinal absorption of cholesterol, has become available. At a single, daily dosage of 10 mg/daily, this well-tolerated agent provides up to an additional 15–20% of LDL-C reduction [11, 12].

Thus, two strategies of combination therapy, further LDL-C reductions (such as through the addition of ezetimibe) or the use of agents that increase HDL-C (such as niacin), emerge to extend the treatment of dyslipidemia. To date, there are no clinical data that compare the relative effects of these two strategies. In this study, we will explore this question through the use of a serial study of carotid intima-media thickness (CIMT), a validated surrogate endpoint of atherosclerosis progression.

2 Design of the ARBITER 6 trial

2.1 Objective

ARBITER 6-HALTS compares HDL-C and LDL-C focused strategies of lipid treatments for their effects on atherosclerosis in the carotid arteries.

2.2 Study design and randomization

Arbiter 6-HALTS is a two-center, randomized, parallel group, open-label, blinded endpoint study. This study will randomize 400 subjects with CHD or CHD equivalents who are at goal LDL-C with reduced HDL-C on statin monotherapy in an allocation-concealed fashion to treatment with either ezetimibe 10 mg/day for intensified LDL-C lowering therapy or extended-release niacin (≥1 g/day, titrated to max tolerable dose up to 2 g/day) for HDL-C raising therapy. Niacin will be titrated from an initial dose of 500 mg qhs, in biweekly increments of 500 mg/qhs to the maximum tolerated dose as limited by flushing or other cutaneous effects. The randomization sequence will be generated using a sequence of computer-generated random numbers. The allocation will be concealed from the investigator and the subject prior to the completion of informed-consent for study participation. Treatment will last a total of 14 months and compliance will be assessed using pill counts at each study visit with a research coordinator. This protocol will be approved by the institutional review board at each participating site. This study is an investigator-initiated trial sponsored by Kos Pharmaceuticals, managed under the auspices of the Henry Jackson Foundation. ARBITER 6-HALTS has been registered with (NCT00397657).

2.3 Inclusion and exclusion criteria

Eligible subjects will be age 30 years or older presenting with known coronary heart disease, who are currently treated with a statin. Specifically, this will include subjects with known clinical atherosclerotic coronary or vascular disease or otherwise at high cardiac risk (diabetes mellitus, multiple coronary risk factors with a Framingham Risk Score >2% per year, an elevated coronary calcium score (>400 for men, >200 for women) who are currently stabilized on statin monotherapy (simvastatin 20 mg/day or its equivalent) for dyslipidemia with an LDL-C < 100 mg/dl and HDL-C < 50 mg/dl (men) or < 55 mg/dl (women). Exclusion criteria include current use of or known intolerance to niacin or ezetimibe, a known history of liver disease (cirrhosis, chronic hepatitis) or abnormal liver associated enzymes (LAE), >3× the upper laboratory reference value, and enrollment in another drug or device research protocol. Females who are pregnant, expect to get pregnant during the course of the study, or are breastfeeding will also be ineligible.

2.4 Data collection

Laboratory analyses will be performed at 0, 2, 8, and 14 months for the analysis of liver associated enzymes, blood glucose, C-reactive protein and serum lipids (LDL-C, HDL-C, and triglycerides). Lipid subfractions including lipoprotein (a) and apolipoprotein A and B100 will be analyzed at baseline and at 14 months. At intake, female subjects of child bearing potential will be asked to have a urine pregnancy test to rule out pregnancy for the purposes of this study. Niacin may lead to minor worsening of blood glucose control, which may require titration of hypoglycemic medications. Regular monitoring of blood sugars will be recommended with titration of hypoglycemic medications as necessary. Study drug will be discontinued if the LAE rise to greater than 3× above the upper laboratory reference limit.

Other study measures include quality of life assessment using the EQ5D, and the serial assessment of anthropometric measurements including blood pressure using an automated sphygmomanometer, height, weight, and waist girth. Patients will make four clinic visits for this study—at the beginning of the study, 2, 8, and 14 months (the end of study visit). Table 1 displays the serial measures studied in ARBITER 6-HALTS.
Table 1

Data collection schedule for ARBITER 6



2 months

8 months

14 months

Intake data (urine pregnancy if appropriate)



Blood pressure, height/weight, waist girth





Quality of life questionnaire




Lipid panel, hsCRP, LAE’s, blood glucose





Lipoprotein A/B100, lp(a)




Pill count





B mode ultrasound of carotid arteries





Clinical follow-up



2.5 Carotid intima-media thickness assessment

Carotid B-mode ultrasound will be assessed utilizing a 13 MHz broadband linear array probe (Sonosite, Inc. Bothell, WA). Each ultrasound study (baseline, 8 and 14 months) will be performed in a standard fashion as follows: Subjects will be studied in the supine position. Each common carotid artery will be evaluated with the subject’s head turned slightly to the contralateral side. Attention will be focused on the far wall of the distal common carotid artery and images will be obtained from the anterior and lateral projection. Ultrasound settings such as depth of field, gain, and transmit focus will be optimized using standardized settings to enhance sonographic details of the intima-media thickness. Digitized recordings from a diastolic frame of the cine-loop recording will be electronically stored and transferred to an off-line central work station for quantitation.

All CIMT studies performed at both study sites will be transmitted in electronic format for interpretation at one central core laboratory. An independent observer blinded o treatment group will perform off-line analysis of B-mode ultrasound images for measurement of mean intima-media thickness and maximal intima-media thickness. All measurements will be performed in duplicate, utilizing both automated border detection and traditional manual methods. Study drug allocation will be blinded to both sonographers and to those performing CIMT analysis.

2.6 Endpoints

The primary endpoint will be the 14 month between-group change in mean CIMT as a surrogate marker of clinical efficacy. Comparison of carotid atherosclerosis progression between the two treatment strategies will provide inferences on their relative efficacy for cardiovascular event reduction. CIMT measured using B-mode carotid ultrasound is a validated method of noninvasively assessing plaque burden [13]. The progression of atherosclerosis is a validated surrogate for cardiovascular clinical outcome. Multiple studies including the Cholesterol Lowering Atherosclerosis Study (CLAS) [14], the Program of Surgical Control of the Hyperlipidemias (POSCH) trial [15], the Atherosclerosis Risk in Communities (ARIC) study [16], and the Regression Growth Evaluation Statin Study REGRESS) [17] have established that atherosclerosis progression, measured either in the coronary or carotid arteries, is associated with an increased risk of cardiovascular events, even in patients treated with active cholesterol-lowering therapy. Carefully performed serial CIMT assessment has been used to successfully discriminate between-group differences in atherosclerosis progression over time periods as short as 6–12 months [7, 10], in particular with imaging of the common carotid artery which allows greater reproducibility of CIMT measurement [1822].

Secondary endpoints include (1) the change in lipid values and lipid subfractions, (2) a composite endpoint consisting of all major adverse cardiovascular events (coronary heart disease death, myocardial infarction, myocardial revascularization, admission to the hospital for an acute coronary syndrome) (3) drug discontinuation due to adverse effects and (4) quality of life measured with the EQ-5D questionnaire, a generic questionnaire or describing and valuing subjects’ health-related quality of life that has been studied in cardiovascular subjects [23].

In addition, prespecified subgroup analyses of the primary study endpoint will include (1) type of statin drug, (2) gender, (3) age—above and below the median value, (4) diabetes mellitus and the metabolic syndrome (defined by NCEP criteria), (5) coronary disease risk equivalents, (6) HDL-C subgroups defined by NCEP criteria (HDL < 40 mg/dl), ARBITER 2 criteria (HDL-C < 45mg/dl) and the HDL-C criteria set forth n this study (<50 mg/dl men, <55 mg/dl women) and (7) CRP—above and below the median value.

2.7 Sample size and data analysis

Sample size analysis is for the assessment of the primary outcome variable of the change in mean CIMT. Assuming a modest effect of improved LDL-C on CIMT, the estimated mean CIMT progression in the statin + ezetimibe group is based on data from ARBITER 2 with statin monotherapy (0.03 ± 0.06 mm/year). A sample size of 150 per group will be necessary to detect a difference in mean CIMT progression between the two treatment groups of 0.02 ± 0.06 mm/year (conservative estimate compared to results observed in ARBITER 2) with 80% power. A target enrollment of 200 per group (total N = 400) will allow for dropouts (15%), and to provide additional sample size in the event that variability in the intima-media thickness measurement is greater than anticipated.

The primary endpoint is the change in mean CIMT from baseline to the 14 month follow-up. These data (change in CIMT/year) will be compared for the LDL-C and HDL-C treatment groups using a t-test for independent groups and analyzed on an intention to treat principle. Secondary endpoints as outlined above will be compared using t tests or the chi-square test as appropriate for continuous or categorical variables. Baseline variables will be compared between study groups once the trial is completed to ensure that these characteristics were successfully balanced by the randomization procedure. In the event that a difference among these variables is found to exist (with appropriate statistical corrections for multiple variable testing, e.g., Bonferroni correction), then subsequent analyses (analysis of covariation, using general linear remodeling procedures) will be performed. All statistical tests will be conducted against a two-sided alternative hypothesis using a significance level of 0.05.

The study will conduct a prespecified, statistically-blinded, interim analysis according to the conservative method of O’Brien and Fleming [24]. This will be conducted after 180 subjects (60% of the planned sample size) have completed the study. Should the trial continue (primary endpoint is not met), the final adjusted P value will be set at 0.045 to account for alpha spending as a function of the interim analysis.

3 Conclusions

In patients with coronary heart disease stabilized on statin therapy, it is unknown whether targeting further LDL-C reduction or initiating HDL-C raising therapy is the superior lipid treatment approach. ARBITER 6-HALTS is a randomized clinical trial designed to address the effects of these two dyslipidemia treatment strategies on atherosclerosis progression in the carotid artery by randomizing patients to either ezetimibe or extended-release niacin. These data will guide clinicians on whether a lipid treatment strategy of raising HDL-C or further LDL-C reduction is superior in the secondary prevention of coronary heart disease. Enrollment in ARBITER 6-HALTS will begin in November 2006, with results anticipated in early 2009.

Copyright information

© Springer Science+Business Media, LLC 2007