Clinical Trial Design
The VULCANO study was a phase III, international, multicentre, prospective, randomised, open-label, parallel-group trial of treatment with CNIC-polypill for 16 weeks compared with standard care in people at high or very high CVD risk but without established disease. The study took place between the 5th June 2017 and the 11th December 2019.
Data in this paper were reported in compliance with the Consolidated Standards of Reporting Trials (CONSORT) guidelines extension for randomised trials.
Participants were recruited from 47 hospital clinics or primary care centres, 41 in Spain, five in Portugal, and one in Mexico.
Eligible individuals were men and women aged 18 years or older at high or very high risk of CVD, defined as a) evidence of subclinical atherosclerosis diagnosed through invasive or non-invasive techniques, including aortic aneurism, significant carotid or coronary atheroma plaque (intima-media thickness [IMT] ≥1.5 mm), severe coronary artery calcification (coronary calcium >300 Agatston units) or ankle-brachial index [ABI] <0.9; b) subjects with diabetes mellitus (DM) with concomitant hypertension (HT) or albuminuria; or c) patients with HT with the following associated conditions: left ventricular hypertrophy (LVH), albuminuria, or renal impairment (defined as estimated glomerular filtration rate [eGFR] 30-60 mL/min/1.73 m2).
According to the investigator’s criteria, each participant was required to have adequate control of BP and LDL-c while on stable lipid-lowering treatment (with at least one statin) and BP-lowering treatment (with an angiotensin-converting-enzyme inhibitor [ACEi] or an angiotensin II receptor blocker [ARB]) during three months before randomisation. Finally, according to the investigator’s view, it should not be expected that participants would need to change their lipid-lowering and BP-lowering treatments for 16 weeks after randomisation.
Exclusion criteria were contraindications to the CNIC-polypill monocomponents (i.e., atorvastatin, ramipril, or aspirin) to control LDL-c and BP levels. These included patients with recurrent peptic ulcer and/or gastric intestinal bleeding, cerebrovascular haemorrhage, history of dyspepsia, and worsening of haemoglobin levels or anaemia that suggested active bleeding according to the investigator’s view. Moreover, we excluded subjects with a history of prior CV events (e.g., myocardial infarction, revascularisation procedures; ischaemic or haemorrhagic stroke or peripheral artery disease); grade 3 hypertension (i.e., systolic BP [SBP] >180 mm Hg and diastolic BP [DBP] >110 mm Hg); familiar hypercholesterolemia; triglycerides level ≥400 mg/dL; or congestive heart failure functional class III-IV (New York Heart Association classification [NYHA]).
Randomisation and Treatment
After obtaining written informed consent, individuals entered a 10-days screening phase during which eligibility for the study was assessed. If eligible, a central randomisation website randomly assigned (1:1) participants to the CNIC-polypill treatment or conventional treatment (usual care).
Participants assigned to the CNIC-polypill arm received one of the six available formulations, namely 100 mg of aspirin with either 20 or 40 mg atorvastatin and either 2.5, 5.0, or 10 mg ramipril (Additional file; Table S1) on a daily regimen for 16 consecutive weeks. The investigators selected the dose based on the prior doses of statins and ACEi/ARB medications taken by the patients and according to equipotency tables provided to the clinicians built according to comparative clinical trials and expert opinion (Additional file; Tables S2 and S3). In patients previously treated with a single-pill combination (SPC) of ACEi/ARB plus a diuretic or a calcium channel blocker (CCB), the SPC was substituted by the CNIC-polypill (at equipotent doses), and they were also prescribed a diuretic or a CCB. Those allocated to usual care continued to receive the same separate medications and doses before inclusion in the study. After randomisation, no modifications of lipid-lowering or antihypertensive drugs, doses, or regimens were allowed in any treatment group for 16 weeks. In the case of a clinically significant increase in BP or the need to modify these treatments, the patient was withdrawn from the study.
After randomisation, a first visit took place to obtain baseline data and a second visit was scheduled after 16 weeks of treatment. During the visits, SBP, DBP, and heart rate were measured, and a fasting blood sample was obtained to assess the lipid profile and other routine haematology and biochemical parameters. Safety was evaluated at Week 8 (via telephone), Week 16, and Week 20 (via telephone) to allow for the reporting of adverse events (AEs). Moreover, the use of concomitant medication was assessed at baseline, Week 8, and Week 16 to discard the usage of the following concomitant drugs during the trial: triple antihypertensive fix-dose combination, oral anticoagulants or antiaggregants (except aspirin), non-steroidal anti-inflammatory drugs (NSAIDs), gemfibrozil, rifampicin, cyclosporine, aliskiren, methotrexate, telaprevir, tipranavir/ritonavir, oral fusidic acid, or drugs able to modify the lipid profile (e.g., antidepressants, antipsychotics, corticoids, or thyroid hormones).
Participants were followed until the end of the follow-up, unacceptable toxicity, or premature withdrawal.
The two primary efficacy outcomes were absolute values of plasma LDL-c level (Friedewald formula) and SBP after 16 weeks of treatment. Secondary outcomes were differences in SBP and DBP values, plasma levels of total cholesterol, high-density lipoprotein cholesterol (HDL-c), LDL-c, and non-HDL-c. Finally, baseline and Week-16 10-year predicted CVD risk was estimated using the US-derived Pooled Cohort Risk Equations (PCE) and the European Systematic Coronary Risk Evaluation (SCORE). Incidence of AEs up to 20 weeks of follow-up was also assessed.
Calculation of sample size
We estimated the power for the trial based on assumptions about differences between the baseline and Week 16 for the two primary outcomes, plasma LDL-c level and SBP values. The required sample size was determined using the confidence interval approach, considering where the confidence interval for the treatment effect lies with respect to the margin of non-inferiority and assuming a null effect (treatments are equal).
In terms of LDL-c, the enrolment of 250 patients (125 per treatment group) provided >90% power (α =0.025) to detect non-inferiority, defined as an absolute between-group difference of 10 mg/dL (SD = 30 mg/dL). In terms of SBP, we determined that a recruitment target of 424 patients (212 per treatment group) provided 80% power (α =0.025) to detect absolute between-group differences of 3 mm Hg (standard deviation [SD] = 11 mm Hg). Accounting for a maximum of 15% loss to follow-up, approximately 250 participants per group were necessary to claim for non-inferiority of the polypill. Based on these calculations, the recruitment target was revised to up to 500 randomised participants (250 per treatment group).
Analysis of efficacy variables
All randomised participants who received at least one dose of medication and had available measures for the main covariables at baseline and Week 16 were included in the analysis (N=439; modified intent-to-treat population [mITT]).
Univariate descriptive statistics and frequency distributions were calculated at baseline for all biological variables in each treatment group. For the primary efficacy endpoint (comparison of absolute LDL-c and SBP values at Week 16), we used analysis of covariance (ANCOVA) to calculate adjusted means and two-sided 95% CIs, with treatment as fixed explanatory variable and baseline values as covariates. Noninferiority of the CNIC-polypill to usual care could be claimed if the upper 95% CI limit for the difference in LDL-c and SBP after 16-week treatment, calculated in terms of baseline-adjusted means (least-squares mean, LSM), was entirely below the prespecified non-inferiority margin of 10 mg/dL for LDL-c and 3 mmHg for SBP [21, 22]. If the 95% CI for the treatment benefit excluded not only the non-inferiority margin but also zero, the P-value to prove the superiority of the CNIC-polypill was calculated. As a sensitivity analysis, non-inferiority of the CNIC-polypill was also evaluated in the per-protocol population (PP), defined as all randomised patients who complied with at least one dose of the assigned treatment, had baseline and Week 16 data on LDL-c and SBP measures, and had no significant protocol deviations (N=403).
For secondary efficacy endpoints (absolute values and LSM differences with respect to baseline in vital signs and lipid profile), the same analysis of covariance (ANCOVA) was performed to calculate the two-sided 95% CIs using baseline values as covariates. Intergroup differences between treatments were assessed using Fisher’s exact test for categorical variables and the Mann-Whitney U test for continuous variables.
The proportion of patients with adequate LDL-c and BP control, based on target values from the 2016 European Society of Cardiology (ESC) guidelines [23, 24], was provided by descriptive statistics and the comparison between treatments at 16 weeks by the exact Fisher test.
SCORE and PCE scores were analysed as continuous variables (risk percentage), by the corresponding risk categories, and as the reduction in the risk category at 16 weeks compared to baseline as binary variable. The comparison between treatments after 16 weeks (absolute values and differences with respect to baseline) was assessed by ANCOVA using the Mann-Whitney U test or the exact Fisher test.
Two-sided P values of less than 0.05 were considered to indicate statistical significance. All statistical analyses were performed using SAS 9.4 software (SAS Institute, Inc., Cary, NC).
Analysis of safety variables
The number of patients who experienced one or more AEs was analysed using descriptive statistics and reported as the percentage of the subjects enrolled in the trial who received at least one dose of the study treatments (N=492).