Abstract
Patients with grade 2‐3 essential hypertension and postplacebo mean clinic systolic blood pressure (SBP) 160‐190 mm Hg and 24‐hour SBP 140‐175 mm Hg by ambulatory blood pressure monitoring (ABPM) received 40 mg azilsartan medoxomil (AZL‐M) monotherapy for 4 weeks. “Nonresponders” were then randomized to 8 weeks of double‐blind treatment with AZL‐M 40 mg, AZL‐M/chlortalidone (CLD) 40/25, or AZL‐M/CLD 40/12.5 mg. After 8 weeks, mean clinic SBP change was −21.1 (±1.04) mm Hg for AZL‐M/CLD 40/25 mg, −15.8 (±1.08) mm Hg for AZL‐M/CLD 40/12.5 mg, and −6.4 (±1.05) mm Hg for AZL‐M 40 mg (P < 0.001 for both AZL‐M/CLD vs AZL‐M, ANCOVA). Drug discontinuation rates were 8.9% (AZL‐M/CLD 40/25 mg), 7.5% (AZL‐M 40 mg), and 3.9% (AZL‐M/CLD 40/12.5 mg). Creatinine increased in 8.1% (AZL‐M/CLD 40/25), 3.1% (AZL‐M/CLD 40/12.5 mg), and 3.0% (AZL‐M 40 mg) of patients. AZL‐M/CLD was effective and well tolerated in patients not achieving blood pressure targets with AZL‐M.
Keywords: antihypertensive therapy, clinical management of high blood pressure (HBP), clinical trials, combination therapy, Hypertension—general
1. INTRODUCTION
While in some countries, for example, the United Kingdom (UK), there is evidence for improving rates of blood pressure (BP) control in patients treated for hypertension, only 37% of patients with survey‐defined high BP in 2011 had achieved target.1 Poor attainment of BP goals remains a global problem.
Many patients require ≥2 drugs to achieve BP control targets, but international guidelines widely diverge in advice on regimen choice.2, 3
Azilsartan medoxomil (AZL‐M) is the most potent angiotensin receptor blocker (ARB),4, 5 but even this, given alone to patients with grade 2 or 3 hypertension, does not control BP in the majority of such patients. Combination preparations of ARBs with diuretics have usually used true thiazides, most commonly hydrochlorothiazide (HCTZ). National Institute for Health and Clinical Excellence (NICE) guidance from the UK (clinical guideline CG 127), in 2011, advised switching to thiazide‐like diuretics indapamide and chlortalidone (CLD).6 While controversial,7, 8 evidence of benefit from huge outcome trials such as ALLHAT, SHEP, and HYVET favored CLD/indapamide.9, 10
AZL‐M is more potent in combination with CLD than HCTZ, as demonstrated in a forced titration study.11
The present study addressed the effectiveness of adding CLD to AZL‐M monotherapy in patients who have not reached BP target on AZL‐M alone.
2. METHODS
2.1. Study design
Patients with grade 2 or 3 hypertension who did not get to BP target on 4 weeks’ treatment with AZL‐M alone were randomized to a fixed‐dose combination (FDC) with 12.5 mg, 25 mg, or placebo CLD. After informed consent, patients stopped prestudy antihypertensive medication, washed out over a 2‐ to 4‐week placebo run‐in period. Eligible patients entered the 4‐week AZL‐M 40‐mg single‐blind monotherapy treatment phase. Patients who did not achieve target BP (trough clinic (c) systolic blood pressure [SBP] ≥140 mm Hg (https://www.nice.org.uk/guidance/cg127)) after single‐blind monotherapy were randomly assigned via an automated Interactive Voice and Web Response System (IxRS) 1:1:1 to double‐blind (DB) treatment with AZL‐M 40 mg, AZL‐M/CLD FDC 40/12.5 mg, or 40/25 mg for 8 weeks. Initiation of combination treatment in these uncontrolled patients is consistent with the 2007 European Society of Hypertension (ESH) and European Society of Cardiology (ESC) hypertension management guidelines, which were current during protocol development and conduct of the study. Patients were evaluated for efficacy and safety end points every 2 weeks during the single and DB treatment phases and followed up for safety 2 weeks after the last dose of study drug Figure 1.
Figure 1.
Scheme of study and patient flow
2.2. Patients
Patients aged ≥18 years with grade 2 or 3 primary hypertension uncontrolled at screening defined as follows:
SBP ≥160 to ≤180 mm Hg and not receiving any antihypertensive medication ≤14 days prior to screening; or
SBP ≥150 to ≤170 mm Hg on one antihypertensive medication; or
SBP ≥140 to ≤160 mm Hg on two antihypertensive medications.
Each patient was required to have a postplacebo run‐in, 24‐hour mean SBP by ambulatory blood pressure monitoring (ABPM) of 140‐175 mm Hg (inclusive) that met ABPM quality criteria and a cSBP of 160‐190 mm Hg (inclusive) for entry into the 4‐week, single‐blind, monotherapy treatment period. Patients who did not achieve target BP (trough cSBP ≥140 mm Hg) following 4 weeks of single‐blind treatment with AZL‐M 40 mg were randomized to DB treatment. Those who achieved target BP were discontinued and resumed standard of care management at physician discretion.
2.3. BP assessments
cBP was measured using a validated semiautomated device (Omron 705CP‐II, 705‐IT, and M6 models) provided by the sponsor. BP was determined by the mean of the three readings ≈24 hours after the previous dose of study medication. Three cBP measurements were obtained ≥2 minutes apart after the patient was seated quietly for ≥5 minutes with feet on the floor and arm supported at the heart level. In addition, a single BP measurement was obtained after the patient remained standing for 2 minutes to evaluate for orthostatic hypotension.
The 24‐hour ABPM was recorded with a portable, automated device (Spacelabs 90207; Issaquah, WA, USA) supplied by the sponsor immediately after administration of study drug in the clinic. If a patient remained eligible for the next stage of the study based on their cSBP measurements, ABPM was performed at the following three visits: postplacebo run‐in (for eligibility), prerandomization/DB baseline (after 4 weeks of single‐blind AZL‐M monotherapy treatment), and the final visit (week 8/early termination). For patients’ safety, daily home BP measurements were taken using a BP measuring device supplied by the sponsor, as for the clinic measurements. Patients were instructed to contact their investigator if their SBP/diastolic BP (DBP) were >190/115 mm Hg during the washout/placebo run‐in or >180/110 mm Hg during the single and DB treatment phases.
2.4. Safety and tolerability assessment
Safety monitoring procedures included recording of adverse events (AEs), clinical laboratory test results, vital sign measurements, electrocardiogram (ECG), and physical examination findings.
Investigators were instructed to report serum creatinine elevations ≥30% from pretreatment values and more than the upper limit of normal (ULN) as an AE of special interest.
2.5. Statistical analysis
2.5.1. End points
The primary end point was change from DB baseline to week 8 in trough, sitting, cSBP. Secondary end points included changes from baseline to week 8 in cDBP, 24‐hour mean SBP and DBP measured by ABPM, and other ABPM parameters, including trough mean BP (22‐24 hours postdosing). The proportion of patients who achieved various BP targets was also calculated.
2.5.2. Analysis of end points
BP changes from baseline to end of the study were tested for differences between treatment with AZL‐M/CLDs vs AZL‐M monotherapy using an analysis of covariance adjusted for baseline BP. Response rates were evaluated using the logistic regression adjusted for baseline BP, expressed as odds ratios and associated 95% confidence intervals (CIs) for achieving target BP with the AZL‐M/CLDs vs the AZL‐M monotherapy.
2.5.3. Sample size
Sample size calculation, based on the findings of a recent factorial study with AZL‐M/CLDs, showed that a sample size of 130 patients per treatment group would deliver ≥90% power to detect an 8‐mm Hg difference between each of AZL‐M/CLDs and the AZL‐M monotherapy in the reduction from baseline in trough sitting SBP with a 5% significance level in a two‐sided t test.12 This planned enrollment was based on an anticipated standard deviation (SD) of 17 mm Hg and a 20% dropout rate.
3. RESULTS
3.1. Patient disposition and demographics
In all, 1754 patients were screened; of these, 1295 (74%) entered the washout/placebo run‐in, yielding 507 (39%) patients who entered the AZL‐M 40‐mg single‐blind monotherapy phase. Main reasons for screen failure and placebo run‐in failure were failure to meet entrance criteria and voluntary withdrawal (Figure 2).
Figure 2.
Patient disposition. Data are n (%). AZL‐M, azilsartan medoxomil; CLD, chlortalidone; D/C, discontinuation of study drug. aSeventeen patients were not eligible to continue in the monotherapy phase or to meet the criteria for the next double‐blind phase, seven patients due to randomization closure, three patients due to scheduling conflict, and one patient due to discontinuation unspecified. bSix patients who did not meet the nonresponder criteria were randomized by error; these patients were excluded from the PPS analyses
Of the 507 patients who entered the single‐blind monotherapy phase, 395 (78%) remained eligible to enter the DB treatment phase and 112 were not randomized due to either premature discontinuation during monotherapy or achievement of target SBP at day −1. Overall, 368 of the 395 randomized patients (93.2%) completed the study and 27 (6.8%) prematurely withdrew. The lowest withdrawal rate was in the AZL‐M/CLD 40/12.5 mg group (3.9%), with slightly higher withdrawal rates in the AZL‐M 40‐mg (7.5%) and AZL‐M/CLD 40/25‐mg (8.9%) groups. Discontinuations that resulted from AEs (primary reason) were low across treatments, with the highest rate in the AZL‐M/CLD 40/25‐mg group (5.2%), followed by AZL‐M/CLD 40/12.5 mg (1.6%) and AZL‐M 40 mg (0.8%).
Patients entering both monotherapy and DB treatment phases were well matched by baseline characteristics (Table 1). In addition to hypertension, hypercholesterolemia (22.8%), type 2 diabetes mellitus (8.4%), or diabetes mellitus (7.6%), obesity (12.7%), dyslipidemia (11.9%), and osteoarthritis (10.4%) were the most common concurrent medical conditions. As expected in this mostly middle‐age/elderly hypertensive population, the most commonly used concomitant medications were statins, drugs used in diabetes (metformin, glimepiride, and gliclazide), and acetylsalicylic acid.
Table 1.
The characteristics of study patients at the start of the single‐blind monotherapy treatment period and the start of the double‐blind treatment period of the study
| Variable | Single‐blind Treatment Perioda | Double‐blind Treatment Periodb | ||
|---|---|---|---|---|
|
AZL‐M 40 mg N = 507 |
AZL‐M 40 mg N = 133 |
AZL‐M/CLD 40/12.5 mg N = 127 |
AZL‐M/CLD 40/25 mg N = 135 |
|
| Age (y) | ||||
| N | 507 | 133 | 127 | 135 |
| Mean (SD) | 58.3 (10.4) | 57.9 (10.2) | 59.2 (10.7) | 57.7 (10.5) |
| Age categories, n (%) | ||||
| <45 y | 45 (8.9) | 13 (9.8) | 10 (7.9) | 12 (8.9) |
| 45 to <65 y | 323 (63.7) | 87 (65.4) | 79 (62.2) | 88 (65.2) |
| ≥65 y | 139 (27.4) | 33 (24.8) | 38 (29.9) | 35 (25.9) |
| Sex, n (%) | ||||
| Male | 317 (62.5) | 87 (65.4) | 73 (57.5) | 91 (67.4) |
| Female | 190 (37.5) | 46 (34.6) | 54 (42.5) | 44 (32.6) |
| Race, n (%) | ||||
| American Indian/Alaska Native | 3 (0.6) | 1 (0.8) | 0 (0.0) | 1 (0.7) |
| Asian | 2 (0.4) | 0 (0.0) | 1 (0.8) | 0 (0.0) |
| Blackc | 2 (0.4) | 1 (0.8) | 0 (0.0) | 0 (0.0) |
| Native Hawaiian/Other Pacific Islander | 1 (0.2) | 0 (0.0) | 0 (0.0) | 1 (0.7) |
| White | 502 (99.0) | 132 (99.2) | 126 (99.2) | 134 (99.3) |
| Multiracial | 3 (0.6) | 1 (0.8) | 0 (0.0) | 1 (0.7) |
| Weight (kg) | ||||
| N | 507 | 133 | 127 | 135 |
| Mean (SD) | 86.4 (15.6) | 87.6 (15.8) | 86.7 (17.0) | 87.4 (14.9) |
| Height (cm) | ||||
| N | 506 | 133 | 126 | 135 |
| Mean (SD) | 171.2 (9.4) | 171.8 (8.9) | 170.6 (10.0) | 172.1 (9.7) |
| BMI (kg/m2) | ||||
| N | 506 | 133 | 126 | 135 |
| Mean (SD) | 29.5 (4.7) | 29.6 (5.1) | 29.8 (5.2) | 29.5 (4.6) |
| Smoking classification, n (%) | ||||
| Never smoked | 305 (60.2) | 77 (57.9) | 83 (65.4) | 77 (57.0) |
| Current smoker | 104 (20.5) | 31 (23.3) | 25 (19.7) | 30 (22.2) |
| Ex‐smoker | 98 (19.3) | 25 (18.8) | 19 (15.0) | 28 (20.7) |
| Diabetes status, n (%) | ||||
| Yes | 80 (15.8) | 20 (15.0) | 27 (21.3) | 20 (14.8) |
| No | 427 (84.2) | 113 (85.0) | 100 (78.7) | 115 (85.2) |
| eGFR (mL/min/1.73 m2)d | ||||
| N | 490 | 133 | 125 | 134 |
| Mean (SD) | 83.1 (17.48) | 84.8 (16.41) | 81.5 (16.25) | 82.4 (16.51) |
| Baseline eGFR categories (mL/min/1.73 m2), n (%) | ||||
| <30 | 0 (0.00) | 0 (0.00) | 0 (0.00) | 0 (0.00) |
| 30 to <60 | 37 (7.3) | 11 (8.3) | 10 (7.9) | 11 (8.1) |
| 60 to <90 | 309 (60.9) | 78 (58.6) | 85 (66.9) | 80 (59.3) |
| ≥90 | 161 (31.8) | 44 (33.1) | 32 (25.2) | 44 (32.6) |
AZL‐M, azilsartan medoxomil; BMI, body mass index; CLD, chlortalidone; eGFR, estimated glomerular filtration rate; SD, standard deviation.
Data are based on monotherapy baseline and include randomized and nonrandomized patients.
Data are based on double‐blind baseline.
Full category = Black or African American.
For the single‐blind treatment period, only patients with both a monotherapy baseline value and at least one postmonotherapy baseline value were included. For the double‐blind treatment period, only patients with both a double‐blind baseline value and at least one post–double‐blind baseline value were included.
3.2. Monotherapy phase
All patients (N = 507) who entered the monotherapy phase started from a relatively high baseline BP (mean cSBP/cDBP 164/94 mm Hg) because they had been selected for poorly controlled grade 2 or 3 hypertension, on ≤2 BP‐lowering agents at screening. Despite this high baseline BP, at the end of 4 weeks of AZL‐M 40‐mg monotherapy, 37 patients achieved the <140 mm Hg target and cSBP (mean [SD] fell by 17.8 [16.3] mm Hg and DBP by 8.5 [9.9] mm Hg). These patients did not need additional treatment, so were not randomized to start DB therapy (except six patients per Figure 2, footnote b).
In patients who did not hit BP target (SBP <140 mm Hg) on AZL‐M 40‐mg monotherapy for 4 weeks, cSBP fell by 12.2 (±12.2) mm Hg and DBP by 5.1 (±7.7 mm Hg, mean ± SD; Figure 3, left panel, mean cSBP only). These patients needed additional therapy and were randomized to the DB addition of CLD 12.5 mg, 25 mg, or placebo, and those randomized to each of the three treatment groups of the DB phase had similar BP reductions in response to AZL‐M 40‐mg monotherapy.
Figure 3.
Clinic SBP through the monotherapy and then double‐blind phases of the study. Patients later randomized to CLD add‐on are denoted by matching symbols during the monotherapy phase on the left of the figure. AZL‐M, azilsartan medoxomil; CLD, chlortalidone; SBP, systolic blood pressure
3.3. Double‐blind phase
BP generally declined over the next 8 weeks, even in patients randomized to “placebo,” that is, continuing on AZL‐M monotherapy, by 6.4 (±1.05) mm Hg SBP and 3.2 (±0.65) mm Hg DBP (LS mean ± SE; Figure 3). Patients switched to AZL‐M/CLD 40/25‐mg had greater mean BP reductions at all time points after entering the DB phase; SBP fell by 21.1 (±1.04) mm Hg and DBP by 10.3 (±0.65) mm Hg at 8 weeks. These reductions again were substantially greater than the mean reductions achieved by continuing on monotherapy, by 14.7 mm Hg (95% CI, 17.6‐11.8) for SBP, and by 7.1 (95% CI, 8.9‐5.3) mm Hg for DBP (P < 0.001 for both comparisons; Figure 3, right panel for SBP, and Figure 4). Those patients switching to AZL‐M/CLD 40/12.5 had reduced cSBP by 15.8 (±1.08) mm Hg and DBP by 7.7 (±0.67) mm Hg at 8 weeks. Reductions were substantially greater than that achieved by continuing on monotherapy for SBP by 9.5 mm Hg (95% CI, 6.5‐12.4) and for DBP by 4.5 mm Hg (95% CI, 2.7‐6.4,P < 0.001 for both comparisons; Figure 4).
Figure 4.
Change from baseline to week 8 in clinic SBP and DBP. DBP, diastolic blood pressure; LS, least squares; SBP, systolic blood pressure; SE, standard error
Results for 24‐hour mean SBP by ABPM were consistent with the primary end point of trough, sitting, and cSBP (Table 2). Mean (±SE) reductions of 18.1 (1.03) mm Hg and 14.7 (1.07) mm Hg were observed with AZL‐M/CLD 40/25 mg and 40/12.5, respectively, from baseline to week 8. These effects were, on average, 15.8 (95% CI, 12.9‐18.6) mm Hg and 12.4 (95% CI, 9.5‐15.3) mm Hg greater than the 2.3 (1.02) mm Hg mean reduction by continuing AZL‐M 40 mg (P < 0.001 for both comparisons,Figure 5A). Results of the analyses between the AZL‐M/CLD 40/25‐mg and 40/12.5‐mg groups and AZL‐M 40‐mg monotherapy for daytime and nighttime SBP also showed large and significant reductions (P < 0.001 for all comparisons), consistent with the primary end point.
Table 2.
Ambulatory blood pressure recording data at the start of the double‐blind treatment period of the study, and after 8 weeks of monotherapy with AZL‐M or combination therapy with AZL‐M/CLD 12.5 mg or 25 mg
| SBP by ABPMa |
AZL‐M 40 mg N = 133 |
AZL‐M/CLD 40/12.5 mg N = 127 |
AZL‐M/CLD 40/25 mg N = 135 |
|||
|---|---|---|---|---|---|---|
| N | 107 | 98 | 104 | |||
| SBP | DBP | SBP | DBP | SBP | DBP | |
| 24‐h mean | ||||||
| Double‐blind baseline LS mean (SE) | 138.0 (1.21) | 81.9 (0.89) | 137.0 (1.26) | 80.3 (0.93) | 138.2 (1.22) | 82.5 (0.91) |
| Change from double‐blind baseline to week 8 | ||||||
| Group mean | 135.5 (−2.5) | 80.1 (−1.8) | 122.6 (−14.3) | 72.4 (−7.9) | 119.8 (−18.3) | 72.0 (−10.5) |
| LS mean (SE) | −2.3 (1.02) | −1.6 (0.66) | −14.7 (1.07) | −8.5 (0.69) | −18.1 (1.03) | −10.1 (0.67) |
| P‐value vs AZL‐M 40 mgb | <0.001* | <0.001* | <0.001* | <0.001* | ||
| LS mean differencec | −12.4 | −6.9 | −15.8 | −8.5 | ||
| 95% CI for difference | (−15.3, −9.5) | (−8.7, −5.0) | (−18.6, −12.9) | (−10.3, −6.6) | ||
| Daytime mean—6 am−10 pm | ||||||
| Double‐blind baseline LS mean (SE) | 141.5 (1.24) | 84.9 (0.96) | 141.1 (1.30) | 83.6 (1.01) | 141.9 (1.26) | 85.7 (0.98) |
| Change from double‐blind baseline to week 8 | ||||||
| Group mean | 139.1 (−2.4) | 83.1 (−1.7) | 126.0 (−15.1) | 75.3 (−8.3) | 123.5 (−18.4) | 75.1 (−10.6) |
| LS mean (SE) | −2.4 (1.09) | −1.7 (0.72) | −15.3 (1.14) | −8.9 (0.75) | −18.2 (1.11) | −10.1 (0.73) |
| P‐value vs AZL‐M 40 mgb | <0.001* | <0.001* | <0.001* | <0.001* | ||
| LS mean differencec | −12.9 | −7.2 | −15.8 | −8.5 | ||
| 95% CI for difference | (−16.0, −9.8) | (−9.2, −5.1) | (−18.8, −12.7) | (−10.5, −6.5) | ||
| Nighttime mean—12 am−6 am | ||||||
| Double‐blind baseline LS mean (SE) | 127.3 (1.50) | 73.2 (0.95) | 124.2 (1.57) | 70.6 (0.99) | 126.3 (1.52) | 72.9 (0.97) |
| Change from double‐blind baseline to week 8 | ||||||
| Group mean | 124.5 (−2.9) | 71.1 (−2.1) | 112.4 (−11.8) | 64.3 (−6.3) | 108.8 (−17.5) | 62.7 (−10.2) |
| LS mean (SE) | −2.2 (1.18) | −1.6 (0.78) | −12.7 (1.24) | −7.2 (0.82) | −17.3 (1.20) | −9.8 (0.79) |
| P‐value vs AZL‐M 40 mgb | <0.001* | <0.001* | <0.001* | <0.001* | ||
| LS mean differencec | −10.5 | −5.7 | −15.2 | −8.3 | ||
| 95% CI for difference | (−13.9, −7.2) | (−7.9, −3.5) | (−18.5, −11.9) | (−10.5, −6.1) | ||
ABPM, ambulatory blood pressure monitor; AZL‐M, azilsartan medoxomil; CI, confidence interval; CLD, chlortalidone; DBP, diastolic blood pressure; LS, least squares; SBP, systolic blood pressure; SE, standard error.
Only patients with a baseline and at least one postbaseline value of acceptable quality were included.
Baseline P‐values were obtained from an ANOVA model with treatment as a factor. Postbaseline P‐values were obtained from an ANCOVA model with treatment as a fixed effect and baseline as a covariate. Baseline value was the last observation before the first dose of double‐blind study drug.
LS mean difference = LS mean change of each TAK‐491 CLD group—LS mean change of TAK‐491 40 mg group.
Statistically significant at 0.05.
Figure 5.
Change from double‐blind baseline to week 8 in mean BP by 24‐hour ABPM. (A) SBP; (B) DBP. ABPM, ambulatory blood pressure monitor; AZL‐M, azilsartan medoxomil; BP, blood pressure; CLD, chlortalidone; DBP, diastolic blood pressure; SBP, systolic blood pressure
Results for all DBP parameters by ABPM were also consistent with the primary end point: in the AZL‐M/CLD 40/25‐mg and 40/12.5‐groups, mean reductions of 10.1 (±0.67) mm Hg and 8.5 (±0.69), respectively, in 24‐hour mean DBP by ABPM were observed from baseline to week 8. These effects were 8.5 (95% CI, 6.6‐10.3) and 6.9 (95% CI, 5.0‐8.7) mm Hg greater than the mean 1.6 (±0.66) mm Hg reduction observed with AZL‐M 40‐mg monotherapy (P < 0.001 for both comparisons; Figure 5B). Daytime and nighttime DBP also showed large and statistically significant reductions (P < 0.001 for all comparisons;Table 2).
The significant mean reductions in BP measured by both clinic and ABPM were paralleled by statistically significant odds of reaching target BP with either AZL‐M/CLD dose over monotherapy. Target SBP (SBP <140 or <130 mm Hg for patients with diabetes or chronic kidney disease [CKD]) was achieved in 77.8% and 62.7% of patients in the AZL‐M/CLD 40/25‐mg and 40/12.5‐groups, respectively, vs 35.3% in the AZL‐M 40‐mg monotherapy group. For DBP target (<90 or <80 mm Hg for patients with diabetes or CKD), these same figures were 85.9% and 81.0% in the AZL‐M/CLD 40/25‐mg and 40/12.5‐mg groups, respectively, vs 60.2% in the AZL‐M 40‐mg monotherapy group. Both target SBP and DBP were achieved in 74.1% and 59.5% of patients in the AZL‐M/CLD 40/25‐mg and 40/12.5‐mg groups, respectively, vs 30.8% in the AZL‐M 40‐mg monotherapy group (Figure S1).
No differences in efficacy (for the primary end point—change from DB baseline in cSBP) by age, sex, body mass index (BMI), baseline trough cSBP (above/below median), estimated glomerular filtration rate (eGFR), diabetes status, and baseline hypertension severity were observed in subgroup analyses. Within each of these subgroups, treatment with AZL‐M/CLD 40/25 and 40/12.5 mg resulted in numerically greater reductions in trough sitting cSBP compared with AZL‐M 40‐mg monotherapy. Differences were statistically significant (at the 0.05 level) for all comparisons with two exceptions: AZL‐M/CLD 40/12.5 and AZL‐M 40‐mg monotherapy in patients with moderate renal impairment (eGFR 30 to <60 mL/minute/1.73 m2) at baseline, and in patients with diabetes at baseline (for both of these cases, the numbers of patient in these categories were relatively small, n < 30 in any treatment group). Similar observations by subgroups were made for cDBP.
3.4. Safety
3.4.1. Adverse events
During the monotherapy phase, treatment‐emergent AEs (TEAEs) were reported by 57 (11.2%) patients; most AEs were mild to moderate in intensity. Three (0.6%) patients reported serious AEs (SAEs)—one had a cerebrovascular accident on day 1 of monotherapy, another had ventricular tachycardia, acute myocardial infarction, and increased BP on day 5 of monotherapy that led to permanent discontinuation of study drug, and the third reported a renal cancer on day 22 of monotherapy that led to permanent discontinuation of study drug; none of these events were fatal. Five (1.0%) patients had a TEAE that led to permanent discontinuation of study drug—in addition to the two SAE cases described earlier, the other three cases were hypertension‐related TEAEs.
During the DB treatment phase, 27.8%, 31.9%, and 23.6%, of patients reported TEAEs in the AZL‐M 40‐mg, AZL‐M/CLD 40/25‐mg, and 40/12.5‐mg groups, respectively. The majority of TEAEs were of mild or moderate intensity; the most frequent events were creatinine elevations, nasopharyngitis, headache, and dizziness. Only two SAEs were reported during the DB treatment phase; both occurred in the AZL‐M 40‐mg group (n = 1 myocardial infarction on day 42 that led to permanent discontinuation of study drug; n = 1 cerebrovascular accident on day 69 [12 days after the last dose of study drug]). No deaths occurred during the study (Table S1).
3.4.2. Increased blood creatinine
TEAEs of increased blood creatinine occurred in a higher percentage of patients in the AZL‐M/CLD 40/25‐mg group than in the AZL‐M 40‐mg and AZL‐M/CLD 40/12.5‐mg groups (8.1% vs 3.0% and 3.1%, respectively). Only one patient (AZL‐M/CLD 40/25‐mg group) discontinued study drug due to creatinine elevations. One case of renal failure was reported in the AZL‐M/CLD 40/25‐mg group. A 63‐year‐old male with cholecystectomy reported a TEAE of increased creatinine (122 µmol/L) on day 48, followed by nonserious renal failure on day 56 (final visit, last dose of study drug) with eGFR 34.3 mL/minute/1.73 m2 and creatinine 175 µmol/L. The event was followed up, and creatinine values returned to baseline/normal range (119 µmol/L on day 70; 108 µmol/L on day 85).
The transient nature of creatinine elevations is demonstrated by the rather rare occurrence of creatinine elevations observed on ≥2 consecutive visits (representing persistent elevations) vs elevations at any visits (representing both persistent and transient elevations; Table S2).
3.4.3. Hypotension and dizziness
Ten patients reported dizziness or dizziness postural events during the DB period; all patients were in AZL‐M/CLD arms with a dose‐response, none in the in AZL‐M arm. Events in 8 of these patients were considered treatment‐related. Similarly, 9 patients reported hypotension or orthostatic hypotension events during the DB study; all but one were in AZL‐M/CLD arms when these events occurred; and in 8 patients, these events were reported as related to treatment. Three hypotension events and 1 orthostatic hypotension event on AZL‐M/CLD 40/25 arm led to premature discontinuation, and none were serious.
3.4.4. Sodium, potassium, and metabolic parameters
Other laboratory parameters deserving special attention in connection to ARB‐diuretic treatments are serum electrolytes (especially potassium and sodium), uric acid, glucose, and lipid parameters. Mean changes from baseline to final visit in electrolyte parameters were small and similar among treatment groups. Importantly, even at the AZL‐M/CLD 40/25‐mg dose, mean reductions in potassium were small and not considered clinically important. For potassium and all other electrolyte parameters, shifts from normal at baseline to high or low at final visit were generally low in frequency and did not vary with the increased dose of the diuretic component.
Larger mean increases in uric acid were observed in the combination treatment groups than in the monotherapy group, with the largest changes observed in the AZL‐M/CLD 40/25‐mg group (73.2 for AZL‐M/CLD 40/25 vs 45.2 and 9.5 for AZL‐M/CLD 40/12.5 and AZL‐M 40 mg, respectively). Uric acid increase did not result in an increased incidence of gout‐related events. Mean changes in glucose, triglyceride, total‐, low‐density lipoprotein‐, and high‐density lipoprotein cholesterol were small, not considered clinically meaningful, and were not CLD dose‐dependent (Table S2). No clinically meaningful changes in liver function tests were observed in any treatment group, and no patient met Hy's law criteria (alanine aminotransferase [ALT] or aspartate aminotransferase [AST] >3 × ULN with a total bilirubin >2 × ULN).
4. DISCUSSION
This study shows that addition of the thiazide‐like diuretic CLD, on top of AZL‐M monotherapy, to patients not reaching BP target on AZL‐M alone was well tolerated and effective. Almost 80% of patients achieved BP target after 8 weeks of combination therapy with AZL‐M/CLD 40/25 mg, compared with 35% who stayed on monotherapy.
ABPM improvements were concordant with those seen in sitting cBP. The smoothness of the averaged ABPM plots, with BP separation for the combined therapy almost perfectly maintained for the full 24‐hour dosing interval, seemed to fit with the long duration of action of both AZL‐M and CLD.13, 14
This study was limited in duration, but a 1‐year, open‐label, treat‐to‐target study,15 with escalating combinations of either olmesartan/HCTZ or AZL‐M/CLD, showed lower BPs in the AZL‐M/CLD group of around 5/3 mm Hg (systolic/diastolic) consistently during the trial, despite more olmesartan/HCTZ patients receiving dose uptitrations (49% vs 32% first titration, 29% vs 17% second titration, and 11% vs 8% third titration; olmesartan/HCTZ vs AZL‐M/CLD). cSBP/cDBP reductions from baseline to final visit averaged −41.5/−17.8 mm Hg in the olmesartan/HCTZ vs −47.7/−20.9 mm Hg for AZL‐M/CLD. Such discrepancies of achieved BP in treat‐to‐target titration studies are typical of more powerful BP drug combinations, despite greater uptitrations for the weaker combinations, as happened in the Anglo‐Scandinavian Cardiac Outcomes Trial (ASCOT) atenolol ± bendroflumethiazide arm versus the amlodipine ± perindopril arm.16
While multiple‐drug combination approaches like the polypill concept17 advocate use of multiple BP‐reducing drugs at low doses and point out the robustness of such a strategy against nonresponse to one of the components, these ideas are still to be widely adopted. The study of Mahmud and Feely18 showed that a four‐drug, quarter‐dose combination could be effective in hypertension, but the dose‐response relationships for unwanted effects of antihypertensive drugs are not the same.19 Angiotensin‐converting enzyme (ACE) inhibitors block the breakdown of bradykinin maximally, even at the starting doses used in man for hypertension, so increasing dose does not increase the rate of cough.20 ARBs have so few unwanted effects at standard doses that choosing maximum dosing levels has not been easy—the maximum launch dose of losartan (50 mg) was increased to 100 mg after its use at the higher dose in the LIFE study.21 For a number of reasons, ARBs have been frequently underdosed for hypertension22 and are safe and well tolerated to ≥3 times their maximum licensed doses when used for effects beyond BP lowering, such as candesartan to 128 mg in the SMART trial in patients with proteinuria.23 In contrast, calcium channel–blocking drugs like amlodipine show much less separation of the wanted and unwanted dose‐response curves, such that individual dose titration feels much more appropriate.
Potent ARBs like AZL‐M, and their combination with CLD, can increase creatinine and uric acid, but these effects are consistent with other powerful ARBs and diuretics and are reversible on stopping the drug(s). AZL‐M tends to “tame” the potassium, lipid, and glucose effects of CLD when used in combination, giving little or no net impact on these, as observed in this study.
There is no doubt, however, that simpler, once‐daily medication regimens improve compliance/concordance. This is even more important in an asymptomatic condition such as hypertension. Urine drug assays now including high‐performance liquid chromatography‐tandem mass spectrometry have revealed increasing nonconcordance rates in patients considered for renal denervation (23.5%), and the more the drugs in the regimen, the worse the fidelity of therapy.24
Getting good BP control can be achieved in the setting of large, randomized trials.16 In ASCOT, large numbers of hypertensives with inadequate BP control were treated in a variety of centers in primary (Scandinavia) and mostly secondary care (UK; Table S3). Coincidentally, the starting SBP in ASCOT was almost exactly the same as the baseline value in this study (164 mm Hg). BP achieved in the whole ASCOT cohort by year 7 (137.7 mm Hg [atenolol arm], 136.1 mm Hg [amlodipine arm]) was close to that achieved by 8 weeks of DB therapy with AZL‐M/CLD [SBP mean 134.1 (±12.8) mm Hg with 40/12.5 and 128.7 (±11.9) mm Hg with 40/25 mg].
In the Barts center for ASCOT, the largest with 1157 patients, the BP reduction at 1 year (−21/−11 mm Hg SBP/DBP) was as large as that achieved by 8 weeks of AZL‐M/CLD 40/25 (−21/−10 mm Hg SBP/DBP) in this study. Over more comparable timescales, at 6 weeks in ASCOT Barts patients’ BP fell by −13/−9 (ASCOT overall −10/−7), 3 months BP fell by −17/−11 (ASCOT overall −14/−9), but this was achieved by a multidisciplinary team in a hypertension center. In the ASCOT, choice of drugs was restricted by the need to compare newer and older medication (avoiding crossover and dilution between drug treatment arms), but the stepped care protocol had six steps, and step seven mandated spironolactone or moxonidine as add‐ons. ASCOT patients were a little older (mean 63 [±8.5] years) than the patients (mean 58.2 [±10.45] years) in the present study, although BMI was similar (ASCOT 28.7 [±4.6] kg/m2 vs 29.6 [±4.94] kg/m2 present study).
If you can get comparable levels of BP reduction with AZL‐M/CLD in 8 weeks of fixed dose, as was achieved at Barts in ASCOT in 1 year, or in the whole ASCOT cohort by 7 years, it seems inevitable that faster BP control will likely reduce events with the FDC.
The recently reported Systolic Blood Pressure Intervention Trial (SPRINT)25 demonstrated a dramatic benefit of intensive (SBP <120 mm Hg) versus standard (SBP <140 mm Hg) BP control, which reduced both the primary outcome by 25% (MI, ACS, stroke, decompensated heart failure, or death from cardiovascular cause) and all‐cause mortality by 27%.25 SPRINT showed a SBP target of 120 mm Hg (mean achieved 121.4 mm Hg in the intensive group at 1 year vs 136.2 mm Hg in the standard treatment group) improved major outcomes, although with an increased rate of treatment‐related AEs—hypotension, syncope, renal injury, or failure and electrolyte abnormalities though not injury‐causing falls). The difference in BP achieved between the two groups in SPRINT was large, given a mean of only one BP drug extra per patient. From Table S2, ARB use in SPRINT increased by 13%, from 27% to 40% in the intensive group, compared with 22% (from 33% to 55%) for thiazide‐type diuretics (the biggest absolute increase in usage). Notably, CLD was the recommended thiazide‐type diuretic in the SPRINT trial.
While in the era of relatively cheap generic drugs, “cheap drugs used expensively” may work well, as in ASCOT and in preliminary data from the community‐based HiLo study which uses extensive education and case management to improve BP control,26, 27 the convenience of relatively “expensive drugs used simply” may be more effective. A population‐based trial comparing the best ways of using inexpensive generics, pitted against the most potent FDCs available such as AZL‐M/CLD, with very neutral impacts on potassium level, glucose, and lipids, would be important and revealing. Squeezing the best BP control out of multiple inexpensive therapies requires training, time, and staff. Potent FDC such as AZL‐M/CLD may provide viable alternatives if we can design cost‐matched studies which balance inputs.
CONFLICT OF INTEREST
This work was supported by Takeda Pharmaceuticals; Collier & Caulfield are partly supported by the NIHR cardiovascular Biomedical Research Unit at Barts, Queen Mary University of London.
This study was sponsored by Takeda Development Centre Europe, Ltd., London, UK. Support for tables and figures was provided by Sam Schmitt of Extension Media Group, LLC, and Blue Momentum, an Ashfield Company, and was funded by Takeda. MJC and DJC are investigators on this trial, and their university received reimbursement of trial costs from Takeda. DJC received a consulting fee from Takeda during the time of the study. EA‐R is an investigator on this trial, and his university hospital received a reimbursement grant from Takeda, according to EC rules. EL and MH are employees of Takeda. AJ, SK, and LZ are no longer employees of Takeda, though they were employees of Takeda at the time the research was conducted and the manuscript was developed. DJC and MJC were partly funded by the National Institute of Health Research (NIHR) Biomedical Research Centre at Barts.
Supporting information
Collier DJ, Juhasz A, Agabiti‐Rosei E, et al. Efficacy and safety of azilsartan medoxomil/chlortalidone fixed‐dose combination in hypertensive patients uncontrolled on azilsartan medoxomil alone: A randomized trial. J Clin Hypertens. 2018;20:1473–1484. 10.1111/jch.13376
REFERENCES
- 1. Falaschetti E, Mindell J, Knott C, Poulter NR. Hypertension management in England: a serial cross‐sectional study from 1994 to 2011. Lancet. 2014;383(9932):1912‐1919. [DOI] [PubMed] [Google Scholar]
- 2. Mancia G, Fagard R, Narkiewicz K et al.: Task Force Members . 2013 ESH/ESC guidelines for the management of arterial hypertension: the Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). J Hypertens. 2013; 31(10):1281‐1357. [DOI] [PubMed] [Google Scholar]
- 3. James PA, Oparil S, Carter BL, et al. 2014 Evidence‐based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA. 2014;311(5):507‐520. [DOI] [PubMed] [Google Scholar]
- 4. White WB, Weber MA, Sica D, et al. Effects of the angiotensin receptor blocker azilsartan medoxomil versus olmesartan and valsartan on ambulatory and clinic blood pressure in patients with stages 1 and 2 hypertension. Hypertension. 2011;57(3):413‐420. [DOI] [PubMed] [Google Scholar]
- 5. Bakris GL, Sica D, Weber MA, et al. The comparative effects of azilsartan medoxomil and olmesartan on ambulatory and clinic blood pressure. J Clin Hypertens (Greenwich). 2011;13(2):81‐88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. The clinical management of primary hypertension in adults: clinical guideline 127. NICE 2011. https://www.nice.org.uk/guidance/cg127. Accessed January 23, 2018.
- 7. Brown MR, Cruickshank JK, MacDonald TM. Navigating the shoals of hypertension: discovery and guidance. BMJ. 2012;344:23‐26. [DOI] [PubMed] [Google Scholar]
- 8. Sofat R, Casas JP, Grosso AM, et al. Could NICE guidance on the choice of blood pressure lowering drugs be simplified? BMJ. 2012;344:d8218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. McManus RJ, Caulfield M, Williams B,National Institute for Health and Clinical Excellence . NICE hypertension guideline 2011: evidence based evolution. BMJ. 2012; 344:e181. [DOI] [PubMed] [Google Scholar]
- 10. Olde Engberink RH, Wijnanda J, Frenkel WJ, et al. Effects of thiazide‐type and thiazide‐like diuretics on cardiovascular events and mortality systematic review and meta‐analysis. Hypertension. 2015;65(5):1033‐1040. [DOI] [PubMed] [Google Scholar]
- 11. Bakris GL, Sica D, White WB, et al. Antihypertensive efficacy of hydrochlorothiazide vs chlorthalidone combined with azilsartan medoxomil. Am J Med. 2012;125(12):1229.e1‐1229.e10. [DOI] [PubMed] [Google Scholar]
- 12. Sica D, Bakris GL, White WB, et al. Blood pressure–lowering efficacy of the fixed‐dose combination of azilsartan medoxomil and chlorthalidone: a factorial study. J Clin Hypertens. 2012;14(5):284‐292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Ojima M, Igata H, Tanaka M, et al. In vitro antagonistic properties of a new angiotensin type 1 receptor blocker, azilsartan, in receptor binding and function studies. J Pharmacol Exp Ther. 2011;336(3):801‐808. [DOI] [PubMed] [Google Scholar]
- 14. Carter BL, Ernst ME, Cohen JD. Hydrochlorothiazide versus chlorthalidone: evidence supporting their interchangeability. Hypertension. 2004;43(1):4‐9. [DOI] [PubMed] [Google Scholar]
- 15. Neutel JM, Cushman WC, Lloyd E, Barger B, Handley A. Comparison of long‐term safety of fixed‐dose combinations azilsartan medoxomil/chlorthalidone versus olmesartan medoxomil/hydrochlorothiazide. J Clin Hypertens. 2017;19(9):874‐883. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Dahlöf B, Sever PS, Poulter NR, et al. Prevention of cardiovascular events with an antihypertensive regimen of amlodipine adding perindopril as required versus atenolol adding bendroflumethiazide as required, in the Anglo‐Scandinavian Cardiac Outcomes Trial‐Blood Pressure Lowering Arm (ASCOT‐BPLA): a multicentre randomised controlled trial. Lancet. 2005;366(9489):895‐906. [DOI] [PubMed] [Google Scholar]
- 17. Wald NJ, Law MR. A strategy to reduce cardiovascular disease by more than 80%. BMJ. 2003;326(7404):1419‐1423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Mahmud A, Feely J. Low‐dose quadruple antihypertensive combination: more efficacious than individual agents – a preliminary report. Hypertension. 2007;49(2):272‐275. [DOI] [PubMed] [Google Scholar]
- 19. Law MR, Wald NJ, Morris JK, Jordan RE. Value of low dose combination treatment with blood pressure lowering drugs: analysis of 354 randomised trials. BMJ. 2003;326(7404):1427‐1431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Dahlöf B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity and mortality in the losartan intervention for endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet. 2002;359(9311):995‐1003. [DOI] [PubMed] [Google Scholar]
- 21. Weinberg AJ, Zappe DH, Ramadugu R, Weinberg MS. Long‐term safety of high‐dose angiotensin receptor blocker therapy in hypertensive patients with chronic kidney disease. J Hypertens. 2006;24(1):S95‐S99. [DOI] [PubMed] [Google Scholar]
- 22. Izzo J, Sica DA Antihypertensive drugs: pharmacological principles and dosing effects. In: Izzo JL, Sica DA, Black HR eds. Hypertension Primer: The Essentials of High Blood Pressure. Philadelphia: PA: Lippincott, Williams & Wilkins; 2003:407. [Google Scholar]
- 23. Burgess E, Muirhead N, de Cotret PR, et al. Supramaximal dose of candesartan in proteinuric renal disease. J Am Soc Nephrol. 2009;20(4):893‐900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Tomaszewski M, White C, Patel P, et al. High rates of non‐adherence to antihypertensive treatment revealed by high‐performance liquid chromatography‐tandem mass spectrometry (HP LC‐MS/MS) urine analysis. Heart. 2014;100(11):855‐861. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Wright JT, Williamson JD, Whelton PK, et al. A randomised trial of intensive versus standard blood‐pressure control. N Engl J Med. 2015;373(22):2103‐2116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Collier DJ, Robson J, Scott R, et al. HiLo: a health services intervention study: methods and interim results on the first 2000 patients at 1 year. J Hum Hypertens. 2010;24(10):708‐709. [Google Scholar]
- 27. UK Clinical Trials Gateway . High blood pressure control and lipid lowering in patients at high cardiovascular risk: a cluster randomised factorial comparison of educational and protocol‐based interventions to help primary care teams improve control of patients with poor risk factor control. https://www.ukctg.nihr.ac.uk/trials/trial-details/trial-details?trialId=13829. Accessed January 23, 2018.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials