Abstract
Objective. To compare the respective effects of tocilizumab (TCZ) monotherapy, etanercept (ETN) monotherapy, and adalimumab (ADA) monotherapy on arterial stiffness in patients with rheumatoid arthritis (RA) in an open-label, randomized controlled trial.
Methods. Patients with RA were eligible if they had active disease (28-joint Disease Activity Score > 3.2) and no prior treatment with methotrexate or biologics. All 64 patients had no history of cardiovascular disease or steroid treatment. Patients were randomly assigned to receive TCZ alone (n = 22), ETN alone (n = 21), or ADA alone (n = 21). Arterial stiffness was assessed with cardio-ankle vascular index (CAVI) and aortic augmentation index normalized to a fixed heart rate of 75 bpm (AIx@75) at baseline and 24 weeks’ followup. Clinical data were collected at regular visits.
Results. The characteristics of each group at baseline were not significantly different. In all groups there was significant attenuation from baseline to 24 weeks in CAVI (Week 0-Week 24, TCZ: 0.85 ± 0.15 m/s, p = 0.02; ETN: 0.81 ± 0.18 m/s, p = 0.03; ADA: 0.90 ± 0.21 m/s, p = 0.02) and in AIx@75. There were no significant differences among the groups in measures of CAVI or AIx@75. The 3 therapies made no difference to carotid intima-media thickness and carotid artery plaque. Only TCZ increased fasting serum total cholesterol from baseline to 24 weeks.
Conclusion. The 3 types of monotherapy limited arterial stiffness in patients with RA to a similar extent.
- ARTERIAL STIFFNESS
- RHEUMATOID ARTHRITIS
- TOCILIZUMAB
- ETANERCEPT
- ADALIMUMAB
We compared the effect of tocilizumab (TCZ) monotherapy with the effect of etanercept monotherapy (ETN) and adalimumab (ADA) monotherapy on arterial stiffness in patients with rheumatoid arthritis (RA).
MATERIALS AND METHODS
Patients with RA were eligible if they had active disease [28-joint disease activity score (DAS)28-erythrocyte sedimentation rate (ESR)1 > 3.2]2 and no prior treatment with methotrexate, steroids, or biologics. The dosage of all disease-modifying antirheumatic drugs had to be stable for at least 8 weeks prior to enrollment.
Patients were assigned randomly to receive TCZ, ETN, or ADA monotherapy. TCZ (8 mg/kg) was administered every 4 weeks subcutaneously, ETN (25 mg) twice a week subcutaneously, and ADA (40 mg) every 2 weeks, subcutaneously. All patients with worsening disease activity as measured by DAS28-ESR at Week 12, defined by change of DAS28-ESR from baseline > 1.2, or DAS 28-ESR > 5.1, were allowed to leave the group (by clinician’s judgment). The primary outcome was change in the cardio-ankle vascular index (CAVI)3. Secondary outcomes were changes in augmentation index4 corrected to a heart rate of 75 bpm (AIx@75), and vascular ultrasound assessments from baseline to 24 weeks. Other outcome measures were cardiovascular risk factors and RA disease activity.
CAVI and AIx@75 were measured at baseline and at 24 weeks by using a CAVI system (Vasera VS-1500N; Fukuda Denshi, Tokyo, Japan).
Brachial ankle pulse wave velocity (PWV) has been used as a therapeutic endpoint in studies of treatments for RA5.
CAVI was calculated by the following formula3:
CAVI measures arterial stiffness independently of blood pressure and is superior to PWV as an index of arterial stiffness3.
Carotid intima-media thickness (CIMT) and carotid artery plaque (CAP) were measured at baseline and 24 weeks. CIMT was examined and measured on both common carotid arteries 1.5 cm proximal to the carotid sinus6. We evaluated the average of the measurements.
CAP was measured from 0 to 4.5 cm proximal to the carotid sinus. A grade of 0 was assigned for no plaque, 1 for minimal, and 2 for extensive. Each individual subject was given a score of 0 to 4, comprising the sum of the scores for both arteries.
Measures of cardiovascular risk included the ankle-brachial index7, the level of fasting serum total cholesterol (TC), and the ratio of TC to high-density lipoprotein cholesterol (HDL; T/H ratio)8. RA disease activity measurements included the Health Assessment Questionnaire (HAQ)9, DAS28-ESR1, and C-reactive protein (CRP). Each measurement was taken every 4 weeks.
Statistical analysis
It was estimated that for a 1:1:1 (TCZ: ETN: ADA) randomized controlled trial, ETN could show an improvement in CAVI in 10 patients with RA, meaning that a sample of 10 patients per group was required (a = 0.05, power = 0.80). Assuming a 30% dropout rate, we aimed to recruit 15 patients for each treatment.
Intention-to-treat analysis
Measures of arterial stiffness, and assessments of others at baseline and at 24 weeks, were compared between subjects within each treatment group by paired t test or chi-squared test.
The effects of treatment (Δ; the change from baseline to 24 weeks) between each group were compared using 1-way ANOVA and a Bonferroni-corrected t test.
Only patients who completed our study at 24 weeks were analyzed. SPSS v15.0 (SPSS, Chicago, IL, USA) was used for all statistical analyses.
RESULTS
Patient characteristics
Sixty-four patients were assigned randomly to receive TCZ (22), ETN (21), or ADA (21). Group characteristics at baseline were not significantly different (Table 1).
One patient taking ADA switched to ETN at 12 weeks. A total of 21 patients in the TCZ, 20 in the ETN, and 19 in the ADA group completed 24 weeks.
Outcome measures
CAVI was attenuated significantly by TCZ (Week 0–Week 24, 0.85 ± 0.15 m/s; p = 0.02), ETN (0.81 ± 0.18 m/s; p = 0.03), and ADA (0.90 ± 0.21 m/s; p = 0.02). The Δ CAVI was not significantly different among TCZ, ETN, and ADA (p > 0.05; Figure 1).
AIx@75 was attenuated significantly by TCZ (Week 0–Week 24, 3.59% ± 0.33%; p = 0.03), ETN (1.03% ± 0.44%; p = 0.03), and ADA (3.54% ± 0.52%; p = 0.02). The Δ AIx@75 was not significantly different among TCZ, ETN, and ADA (p > 0.05).
TCZ, ETN, and ADA did not significantly change CIMT (Week 0–Week 24, TCZ: 0.00 ± 0.13 mm, ETN: 0.00 ± 0.22 mm, ADA: −0.01 ± 0.13 mm; p > 0.05). They also did not produce significant changes in CAP (Week 0–Week 24, numbers of combined grade 0/1/2/3/4, TCZ: −1/1/0/0, ETN: 0/–2/0/0, ADA: −1/0/–1/0; p > 0.05).
There were no significant changes either within or between groups with respect to ankle-brachial index (Week 0–Week 24, TCZ: 0.03 ± 0.01; ETN: 0.09 ± 0.02; ADA: −0.03 ± 0.02) or fasting T/H ratio (Week 0–Week 24, TCZ: 0.09 ± 0.04; ETN: 0.09 ± 0.03; ADA: 0.07 ± 0.03; p > 0.05).
In the TCZ group, fasting serum total cholesterol was significantly increased (Week 0–Week 24, −18.0 ± 5.2 mg/dl; p = 0.03). There were no significant changes within the ETN or ADA groups in TC (ETN: −2.0 ± 0.6 mg/dl; ADA: −5.0 ± 1.8 mg/dl; p > 0.05). Delta TC levels of the TCZ group were significantly higher than those of the other groups (TCZ vs ETN, p = 0.024; TCZ vs ADA, p = 0.032).
HAQ score, DAS28-ESR score, and CRP improved significantly in all groups (Week 0–Week 24, HAQ score, TCZ: 0.70 ± 0.08, ETN: 0.68 ± 0.09, ADA: 0.69 ± 0.11; DAS28-ESR score, TCZ: −2.10 ± 0.35, ETN: −2.84 ± 0.42, ADA: −2.12 ± 0.38; CRP, TCZ: 24.3 ± 3.2 mg/l, ETN: 19.0 ± 2.31 mg/l, ADA: 20.7 ± 2.11 mg/l; p < 0.05). There were no significant differences among the groups.
DISCUSSION
CAVI and AIx@75 improved after 24 weeks of TCZ, ETN, and ADA. The improvement in CAVI measures was remarkable. The observed difference of 1 m/s was equivalent to the difference previously seen between 1- and 2-stenosis vessel disease10. CIMT and CAP did not change over the course of our study, and this may reflect an error, the limited size of the patient groups, or the short followup11.
TCZ could possibly play a role in upregulating levels of serum cholesterol12. Hypercholesterolemia can induce cardiovascular disease13. Our study confirmed that TCZ reduced arterial stiffness but increased levels of TC. This might be because TC rises together with HDL, such that the T/H ratio did not change. T/H ratio is a more important predictor of cardiovascular disease than TC14.
- Accepted for publication May 30, 2011.
REFERENCES
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8.
- 9.
- 10.
- 11.
- 12.
- 13.
- 14.