Brief Review Design Considerations for Clinical Trials of Autonomic Modulation Therapies Targeting Hypertension and Heart Failure Faiez Zannad, Wendy Gattis Stough, Felix Mahfoud, George L. Bakris, Sverre E. Kjeldsen, Robert S. Kieval, Hermann Haller, Nadim Yared, Gaetano M. De Ferrari, Ileana L. Piña, Kenneth Stein, Michel Azizi
S
everal device-based approaches to autonomic nervous system modulation are under investigation for the treatment of resistant hypertension and heart failure (Table 1).1 This line of research has evolved from the recognition that these diseases originate or are worsened by excess sympathetic activity and loss of parasympathetic tone.9–13 Drug therapies, including β-blockers, α-blockers, and centrally acting antihypertensive drugs, can modulate these neurohormonal systems, but they are often insufficient to control blood pressure (BP) or are limited by side effects or nonadherence. Technological innovations have produced devices that modulate the autonomic nervous system, including renal denervation, carotid baroreceptor stimulation, vagal nerve stimulation, and spinal cord stimulation. In Europe, several autonomic modulation therapy devices have received the Conformité Européenne mark.14 US Food and Drug Administration evaluation of these devices is ongoing. The need for adequately powered, randomized, controlled studies with longer follow-up to capture definitive evidence of safety and effectiveness has been noted.14–17 The 9th and 10th Global Cardiovascular Clinical Trialists Forum (Paris, France, December 2012 and December 2013) convened a panel of primary investigators of ongoing trials, along with biostatisticians, National Institutes of Health scientists, European, and United States regulators, and medical device and pharmaceutical industry scientists to discuss the strengths and limitations of current clinical trials, optimal designs for future trials, approvability of new devices, and considerations for integrating these technologies into practice. This article summarizes the key discussion points and identifies knowledge gaps in this field that need to be addressed by additional research.
Overview of Autonomic Modulation Therapy The mechanisms of autonomic modulation are complex, and a comprehensive review of these mechanisms is outside the scope
of this article. Briefly, all existing strategies aim to decrease central sympathetic outflow. Renal denervation is hypothesized to achieve this objective through ablation of renal afferent and efferent sympathetic nerves, thereby reducing sympathetic efferent signaling to the kidneys and other organ systems.18 Baroreceptor activation therapy stimulates baroreceptors in the carotid sinus via a lead connected to an implantable pulse generator; these baroreceptors play a major role in sympathetic and parasympathetic autonomic regulation.19 Vagal nerve stimulation activates vagal efferent and afferent fibers. These actions also result in reduced central sympathetic drive.12,20
Autonomic Modulation Therapy Clinical Trials: Methodologic Challenges Several characteristics of autonomic modulation devices pose challenges for clinical trial design. First, the introduction of bias can be problematic because these trials are difficult, but not impossible, to blind. Second, the mechanism of treatment effect is difficult to establish. Although a reduction in central sympathetic activity is the proposed mechanism of action, generating proof that sympathetic outflow is reduced either acutely post procedure or during long-term follow-up can be difficult, especially in a large-scale clinical study.18 Third, the adequacy of surrogate end points (eg, changes in BP or cardiac dimensions) to reflect clinical benefits in trials of autonomic modulation devices has been debated. Finally, safety assessments are of key importance because these devices require invasive procedures, the interventions may be irreversible (renal denervation), or require a repeat invasive procedure when stimulator replacement is necessary. Cumulative data from the Renal Denervation in Patients With Uncontrolled Hypertension (SYMPLICITY HTN-3) study21 and from the SYMPLICITY International Registry22 show a low rate of short-term adverse events with renal denervation delivered by
Received August 21, 2014; first decision September 7, 2014; revision accepted October 1, 2014. From the Department of Cardiology, INSERM, Center d’Investigation Clinique 9501 and Unité 961, Center Hospitalier Universitaire, Nancy University, Université de Lorraine, Nancy, France (F.Z.); Departments of Pharmacy Practice and Clinical Research, Campbell University College of Pharmacy and Health Sciences, Buies Creek, NC (W.G.S.); Klinik für Innere Medizin III, Universtitätsklinikum des Saarlandes, Homburg/Saar, Germany (F.M.); HarvardMIT Biomedical Engineering, Institute of Medical Engineering and Science, Cambridge, MA (F.M.); ASH Comprehensive Hypertension Center, The University of Chicago Medicine, IL (G.L.B.); Oslo University Hospital, Ullevaal, Institute for Clinical Medicine, University of Oslo, Oslo, Norway (S.E.K.); CVRx, Inc, Minneapolis, MN (R.S.K., N.Y.); Department of Nephrology, Medizinische Hochschule Hannover, Hannover, Germany (H.H.); Department of Cardiology, Fondazione IRCCS Policlinico San Matteo, Pavia, Italy (G.M.D.F.); Department of Medicine, Division of Cardiology, Montefiore-Einstein Medical Center, Bronx, NY (I.L.P.); Boston Scientific Corporation, St. Paul, MN (K.S.); Université Paris Descartes, Paris, France (M.A.); Assistance Publique, Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Hypertension Unit, Paris, France (M.A.); and Inserm CIC 9201, Paris, France (M.A.). Correspondence to Faiez Zannad, CIC INSERM CHU, Hôpital Jeanne d’ Arc, 54200 Toul, France. E-mail [email protected] (Hypertension. 2015;65:5-15. DOI: 10.1161/HYPERTENSIONAHA.114.04057.) © 2014 American Heart Association, Inc. Hypertension is available at http://hyper.ahajournals.org
DOI: 10.1161/HYPERTENSIONAHA.114.04057
Downloaded from http://hyper.ahajournals.org/ 5 at Universitaet Giessen on June 8, 2015
6 Hypertension January 2015 Table 1. Select Completed and Ongoing Clinical Trials of Autonomic Modulation Therapies in Hypertension and Heart Failure Trial
No. and Population
Blinding
Control Group
Primary End Point
Length of Follow-Up
Results
Renal denervation SWAN HT NCT01417221
200 Resistant hypertension
Open-label
Medical management
Composite of MI, stroke, HF, and sudden death
3y
Ongoing
EnligHTN-I NCT01438229
47 Resistant hypertension
Open-label
Single arm
Safety and office BP
6 mo
Ongoing
SYMPLICITY HTN-321,65 NCT01418261
535 Resistant hypertension
Single-blind (masked-procedure, randomization occurs at time of renal angiogram, patients and research staff assessing BP blinded)
Medical management plus sham procedure (renal angiogram only)
Change in office-based BP
6 mo
Mean change in SBP renal denervation: −14.13±23.93 mm Hg vs sham control −11.74±25.94 mm Hg (both P160 mm Hg
Open-label
Optimal medical therapy
MI, stroke, hospitalization for heart failure, and death
72 mo
Ongoing
240 Resistant hypertension (≥140/90 office and ≥130/80 24 ABPM on ≥3 antihypertensive drugs; aldosterone antagonist or β-blocker should be attempted unless contraindicated)
Open-label
Optimal medical therapy
Baseline adjusted between group difference in 24 h SBP and in eGFR
36 mo
Ongoing
REACH NCT01639378
100 Chronic heart failure, NYHA class II–IV, LVEF
S
everal device-based approaches to autonomic nervous system modulation are under investigation for the treatment of resistant hypertension and heart failure (Table 1).1 This line of research has evolved from the recognition that these diseases originate or are worsened by excess sympathetic activity and loss of parasympathetic tone.9–13 Drug therapies, including β-blockers, α-blockers, and centrally acting antihypertensive drugs, can modulate these neurohormonal systems, but they are often insufficient to control blood pressure (BP) or are limited by side effects or nonadherence. Technological innovations have produced devices that modulate the autonomic nervous system, including renal denervation, carotid baroreceptor stimulation, vagal nerve stimulation, and spinal cord stimulation. In Europe, several autonomic modulation therapy devices have received the Conformité Européenne mark.14 US Food and Drug Administration evaluation of these devices is ongoing. The need for adequately powered, randomized, controlled studies with longer follow-up to capture definitive evidence of safety and effectiveness has been noted.14–17 The 9th and 10th Global Cardiovascular Clinical Trialists Forum (Paris, France, December 2012 and December 2013) convened a panel of primary investigators of ongoing trials, along with biostatisticians, National Institutes of Health scientists, European, and United States regulators, and medical device and pharmaceutical industry scientists to discuss the strengths and limitations of current clinical trials, optimal designs for future trials, approvability of new devices, and considerations for integrating these technologies into practice. This article summarizes the key discussion points and identifies knowledge gaps in this field that need to be addressed by additional research.
Overview of Autonomic Modulation Therapy The mechanisms of autonomic modulation are complex, and a comprehensive review of these mechanisms is outside the scope
of this article. Briefly, all existing strategies aim to decrease central sympathetic outflow. Renal denervation is hypothesized to achieve this objective through ablation of renal afferent and efferent sympathetic nerves, thereby reducing sympathetic efferent signaling to the kidneys and other organ systems.18 Baroreceptor activation therapy stimulates baroreceptors in the carotid sinus via a lead connected to an implantable pulse generator; these baroreceptors play a major role in sympathetic and parasympathetic autonomic regulation.19 Vagal nerve stimulation activates vagal efferent and afferent fibers. These actions also result in reduced central sympathetic drive.12,20
Autonomic Modulation Therapy Clinical Trials: Methodologic Challenges Several characteristics of autonomic modulation devices pose challenges for clinical trial design. First, the introduction of bias can be problematic because these trials are difficult, but not impossible, to blind. Second, the mechanism of treatment effect is difficult to establish. Although a reduction in central sympathetic activity is the proposed mechanism of action, generating proof that sympathetic outflow is reduced either acutely post procedure or during long-term follow-up can be difficult, especially in a large-scale clinical study.18 Third, the adequacy of surrogate end points (eg, changes in BP or cardiac dimensions) to reflect clinical benefits in trials of autonomic modulation devices has been debated. Finally, safety assessments are of key importance because these devices require invasive procedures, the interventions may be irreversible (renal denervation), or require a repeat invasive procedure when stimulator replacement is necessary. Cumulative data from the Renal Denervation in Patients With Uncontrolled Hypertension (SYMPLICITY HTN-3) study21 and from the SYMPLICITY International Registry22 show a low rate of short-term adverse events with renal denervation delivered by
Received August 21, 2014; first decision September 7, 2014; revision accepted October 1, 2014. From the Department of Cardiology, INSERM, Center d’Investigation Clinique 9501 and Unité 961, Center Hospitalier Universitaire, Nancy University, Université de Lorraine, Nancy, France (F.Z.); Departments of Pharmacy Practice and Clinical Research, Campbell University College of Pharmacy and Health Sciences, Buies Creek, NC (W.G.S.); Klinik für Innere Medizin III, Universtitätsklinikum des Saarlandes, Homburg/Saar, Germany (F.M.); HarvardMIT Biomedical Engineering, Institute of Medical Engineering and Science, Cambridge, MA (F.M.); ASH Comprehensive Hypertension Center, The University of Chicago Medicine, IL (G.L.B.); Oslo University Hospital, Ullevaal, Institute for Clinical Medicine, University of Oslo, Oslo, Norway (S.E.K.); CVRx, Inc, Minneapolis, MN (R.S.K., N.Y.); Department of Nephrology, Medizinische Hochschule Hannover, Hannover, Germany (H.H.); Department of Cardiology, Fondazione IRCCS Policlinico San Matteo, Pavia, Italy (G.M.D.F.); Department of Medicine, Division of Cardiology, Montefiore-Einstein Medical Center, Bronx, NY (I.L.P.); Boston Scientific Corporation, St. Paul, MN (K.S.); Université Paris Descartes, Paris, France (M.A.); Assistance Publique, Hôpitaux de Paris, Hôpital Européen Georges Pompidou, Hypertension Unit, Paris, France (M.A.); and Inserm CIC 9201, Paris, France (M.A.). Correspondence to Faiez Zannad, CIC INSERM CHU, Hôpital Jeanne d’ Arc, 54200 Toul, France. E-mail [email protected] (Hypertension. 2015;65:5-15. DOI: 10.1161/HYPERTENSIONAHA.114.04057.) © 2014 American Heart Association, Inc. Hypertension is available at http://hyper.ahajournals.org
DOI: 10.1161/HYPERTENSIONAHA.114.04057
Downloaded from http://hyper.ahajournals.org/ 5 at Universitaet Giessen on June 8, 2015
6 Hypertension January 2015 Table 1. Select Completed and Ongoing Clinical Trials of Autonomic Modulation Therapies in Hypertension and Heart Failure Trial
No. and Population
Blinding
Control Group
Primary End Point
Length of Follow-Up
Results
Renal denervation SWAN HT NCT01417221
200 Resistant hypertension
Open-label
Medical management
Composite of MI, stroke, HF, and sudden death
3y
Ongoing
EnligHTN-I NCT01438229
47 Resistant hypertension
Open-label
Single arm
Safety and office BP
6 mo
Ongoing
SYMPLICITY HTN-321,65 NCT01418261
535 Resistant hypertension
Single-blind (masked-procedure, randomization occurs at time of renal angiogram, patients and research staff assessing BP blinded)
Medical management plus sham procedure (renal angiogram only)
Change in office-based BP
6 mo
Mean change in SBP renal denervation: −14.13±23.93 mm Hg vs sham control −11.74±25.94 mm Hg (both P160 mm Hg
Open-label
Optimal medical therapy
MI, stroke, hospitalization for heart failure, and death
72 mo
Ongoing
240 Resistant hypertension (≥140/90 office and ≥130/80 24 ABPM on ≥3 antihypertensive drugs; aldosterone antagonist or β-blocker should be attempted unless contraindicated)
Open-label
Optimal medical therapy
Baseline adjusted between group difference in 24 h SBP and in eGFR
36 mo
Ongoing
REACH NCT01639378
100 Chronic heart failure, NYHA class II–IV, LVEF
