Comment
Are statins to be STASHed in subarachnoid haemorrhage? In The Lancet Neurology, Peter Kirkpatrick and colleagues report the results of an important randomised trial1 of simvastatin treatment for aneurysmal subarachnoid hemorrhage (SAH). Unfortunately, simvastatin 40 mg per day, started within 96 h of subarachnoid haemorrhage and given for up to 21 days, had no effect on outcome measured on the modified Rankin Scale at 6 months. The only secondary endpoint improved by statin use was a reduction in need for extended hypervolaemic treatment, which is of unknown clinical importance because it emerges alone among other measures that should be associated with delayed cerebral ischaemia. Does this disappointing trial result in the end of investigation of statins in subarachnoid haemorrhage? Why did simvastatin not work, in view of the excellent safety profile of statins, promising early data, and seemingly perfect list of protective effects for prevention of neurovascular damage after subarachnoid haemorrhage? Statins were isolated about 40 years ago in a search for molecules that would inhibit 3-hydroxy-3methylglutaryl coenzyme A (HMG CoA) reductase, the rate-limiting step in cholesterol synthesis, and thereby, reduce serum cholesterol concentrations.2 Bacteriaderived statins that decreased serum cholesterol and reduced heart disease and stroke were identified.3 Inhibition of HMG-CoA reductase has several effects. In addition to the reduction of cholesterol synthesis, decreased synthesis of farnesyl-pyrophosphate and geranylgeranyl-pyrophosphate might also occur. These compounds bind to proteins such as heterotrimeric G proteins (Ras, Rho, and others) in a process called protein prenylation.4,5 Prenylation is necessary for membrane localisation and activation of these enzymes. Thus, statins might affect many cellular signal transduction pathways. The importance of these cholesterolindependent effects of statins to their clinical efficacy is unknown but, interestingly, statin-induced inhibition of protein prenylation in some experimental systems causes vascular fragility and intracranial haemorrhage.4 Low serum cholesterol concentrations are associated with intracerebral haemorrhage, and some clinical trials of statins have reported an increased risk of intracerebral haemorrhage with statin treatment.5 This finding raises the question of whether there are potential, unrecognised detrimental effects of statins.
However, the pharmacological profile of statins, which includes pleiotrophic vascular and neuroprotective effects, such as increased expression of endothelial nitric oxide synthase, reduced inflammation, decreased platelet activation, and fibrinolytic and antioxidant effects, seems to be the perfect combination for treatment of subarachnoid haemorrhage.6 Experimental studies supported the efficacy of statins in subarachnoid haemorrhage, but those studies were of poor quality and small numbers and would not meet Stroke Academic Industry Roundtable (STAIR) criteria.2,7 Nevertheless, these findings led to various observational and randomised clinical studies of statins in subarachnoid haemorrhage. All of these studies were single-centre, phase 2 studies with fewer than 100 patients. The absence of consistent definitions and trial designs, and the use of different statins and doses (studies used simvastatin 20–80 mg or pravastatin 40 mg) limited comparison between studies, and no definitive results emerged. Now we have one arguably definitive answer about statins in subarachnoid haemorrhage. It is reasonable to ask whether this question could have been answered sooner if a more coordinated clinical trials network existed, as it is being developed in the USA.8 At the very least, these findings support the need for common data elements and definitions for trials in subarachnoid haemorrhage. The STASH trial1 was well designed by a competent, experienced group and the centres were also entirely appropriate for the study; so why did simvastatin not improve outcome? The authors have addressed some of the reasons, but other causes may exist. I agree that compliance with drug administration is unlikely to account for the absence of effect. Analysis of the subgroup of patients who had high drug compliance still showed absolutely no evidence for benefit. Patients were included within 96 h of subarachnoid haemorrhage, but was this timing too late? The mean time to treatment was about 2 days, which should be adequate in view of the time it takes for statin-induced effects to occur. It is possible that statins might be effective in a selected subgroup, but no effects in subgroups emerge here. Was the subarachnoid haemorrhage population biased to not benefit? This possibility seems unlikely because 17% of patients were randomised at sites that kept screening
www.thelancet.com/neurology Published online May 16, 2014 http://dx.doi.org/10.1016/S1474-4422(14)70095-X
Lancet Neurol 2014 Published Online May 16, 2014 http://dx.doi.org/10.1016/ S1474-4422(14)70095-X See Online/Articles http://dx.doi.org/10.1016/ S1474-4422(14)70084-5
1
Comment
logs, which is good for a clinical trial. Did simvastatin improve outcome but the effect go undetected on the modified Rankin Scale? Examination of the figures shows no evidence for a general shift in outcomes and the other clinical measures such as the SF-36 were not affected. The argument that reduction in angiographic vasospasm may not be associated with improved clinical outcome6 has less weight here, since the investigators assessed delayed cerebral ischaemia, which has a closer relation to outcome, and simvastatin had no effect. The authors note that the absence of effect of simvastatin could be due to the low incidence of delayed cerebral ischaemia in the study (about 16%, compared with 30% in general).5 However, not even a trend towards reduced delayed cerebral ischaemia or use of rescue therapy was detected, other than the single hypervolaemic therapy endpoint that was probably statistical chance. Nor was there any consistent effect in the subgroups at more risk for delayed cerebral ischaemia, such as those with poor clinical grade, higher Fisher grade, or who underwent aneurysm repair by clipping as opposed to coiling. Since statins might have an effect on neurogenesis and angiogenesis, it is intriging to ask if stopping simvastatin 21 days after subarachnoid haemorrhage caused some negative rebound effect.5 Does the type and dose of statin matter? Previous studies used simvastatin (20–80 mg per day) or pravastatin (40 mg per day).2 STASH used simvastatin, which is more lipophilic and might have higher blood– brain barrier permeability than pravastatin, and simvastatin 40 mg has twice the cholesterol-lowering effect of pravastatin 40 mg. A particularly interesting question is why the higher pharmacological dose of a more brain penetrating statin did not work in STASH, whereas the lower dose of pravastatin showed promise previously.9 Almost 200 patients in STASH and over twice as many in the previous Cambridge study9 came from the same centre as the study of Tseng and colleagues. A critical subgroup analysis would be the outcomes in patients from the Cambridge centre. If that subgroup did better with simvastatin in STASH, then critical insights into the treatment of subarachnoid haemorrhage might be gleaned. If they did not do better, then again, a detailed analysis of the possible reasons might be useful. If no reasons can be identified, then, as the authors suggest, we are stuck with the well known 2
issue of positive findings in small, single-centre studies that are not borne out in large, multicentre trials.10 What are the implications of the STASH results? I believe they should change practice. It is difficult to justify starting a statin on a patient in the acute phase after subarachnoid haemorrhage for the purpose of trying to improve their outcome. If the patient is already on a statin when they have subarachnoid haemorrhage, the treatment probably should be continued. The most logical next step is already underway: a study comparing 40 mg and 80 mg simvastatin.11 Fortunately, the design is very similar to STASH; hopefully, it will show that highdose simvastatin is efficacious. If not, then the statin story in subarachnoid haemorrhage might be over. R Loch Macdonald Division of Neurosurgery, St Michael’s Hospital, Labatt Family Centre of Excellence in Brain Injury and Trauma Research, Keenan Research Centre for Biomedical Science and the Li Ka Shing Knowledge Institute of St. Michael’s Hospital, Department of Surgery, University of Toronto, ON M5B 1W8, Canada [email protected] I receive grant support from the Physicians Services Incorporated Foundation, Brain Aneurysm Foundation, Canadian Institutes for Health Research, and the Heart and Stroke Foundation of Canada; and I am Chief Scientific Officer of Edge Therapeutics Inc. 1
2
3
4
5 6 7 8 9
10 11
Kirkpatrick PJ, Turner CL, Smith C, Hutchinson PJ, Murray GD, for the STASH Collaborators. Simvastatin in aneurysmal subarachnoid haemorrhage (STASH): a multicentre randomised phase 3 trial. Lancet Neurol 2014; published online May 16. http://dx.doi.org/10.1016/ S1474-4422(14)70084-5. Sabri M, Macdonald RL. Statins: a potential therapeutic addition to treatment for aneurysmal subarachnoid hemorrhage? World Neurosurg 2010; 73: 646–53. Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 2005; 366: 1267–78. Eisa-Beygi S, Hatch G, Noble S, Ekker M, Moon TW. The 3-hydroxy-3methylglutaryl-CoA reductase (HMGCR) pathway regulates developmental cerebral-vascular stability via prenylation-dependent signalling pathway. Dev Biol 2013; 373: 258–66. Flaster M, Morales-Vidal S, Schneck MJ, Biller J. Statins in hemorrhagic stroke. Expert Rev Neurother 2011; 11: 1141–49. Macdonald RL. Delayed neurological deterioration after subarachnoid haemorrhage. Nat Rev Neurol 2014; 10: 44–58. Macleod MR, Fisher M, O’Collins V, et al. Good laboratory practice: preventing introduction of bias at the bench. Stroke 2009; 40: e50–52. Landis S, Fisher M. Why the United States needs a network for stroke clinical trials. Stroke 2013; 44: 1217–18. Tseng MY, Hutchinson PJ, Czosnyka M, Richards H, Pickard JD, Kirkpatrick PJ. Effects of acute pravastatin treatment on intensity of rescue therapy, length of inpatient stay, and 6-month outcome in patients after aneurysmal subarachnoid hemorrhage. Stroke 2007; 38: 1545–50. Ioannidis JP. Why most published research findings are false. PLoS Med 2005; 2: e124. Wong GK, Liang M, Tan H, et al. High-dose simvastatin for aneurysmal subarachnoid hemorrhage: a multicenter, randomized, controlled, double-blind clinical trial protocol. Neurosurgery 2013; 72: 840–44.
www.thelancet.com/neurology Published online May 16, 2014 http://dx.doi.org/10.1016/S1474-4422(14)70095-X
Are statins to be STASHed in subarachnoid haemorrhage? In The Lancet Neurology, Peter Kirkpatrick and colleagues report the results of an important randomised trial1 of simvastatin treatment for aneurysmal subarachnoid hemorrhage (SAH). Unfortunately, simvastatin 40 mg per day, started within 96 h of subarachnoid haemorrhage and given for up to 21 days, had no effect on outcome measured on the modified Rankin Scale at 6 months. The only secondary endpoint improved by statin use was a reduction in need for extended hypervolaemic treatment, which is of unknown clinical importance because it emerges alone among other measures that should be associated with delayed cerebral ischaemia. Does this disappointing trial result in the end of investigation of statins in subarachnoid haemorrhage? Why did simvastatin not work, in view of the excellent safety profile of statins, promising early data, and seemingly perfect list of protective effects for prevention of neurovascular damage after subarachnoid haemorrhage? Statins were isolated about 40 years ago in a search for molecules that would inhibit 3-hydroxy-3methylglutaryl coenzyme A (HMG CoA) reductase, the rate-limiting step in cholesterol synthesis, and thereby, reduce serum cholesterol concentrations.2 Bacteriaderived statins that decreased serum cholesterol and reduced heart disease and stroke were identified.3 Inhibition of HMG-CoA reductase has several effects. In addition to the reduction of cholesterol synthesis, decreased synthesis of farnesyl-pyrophosphate and geranylgeranyl-pyrophosphate might also occur. These compounds bind to proteins such as heterotrimeric G proteins (Ras, Rho, and others) in a process called protein prenylation.4,5 Prenylation is necessary for membrane localisation and activation of these enzymes. Thus, statins might affect many cellular signal transduction pathways. The importance of these cholesterolindependent effects of statins to their clinical efficacy is unknown but, interestingly, statin-induced inhibition of protein prenylation in some experimental systems causes vascular fragility and intracranial haemorrhage.4 Low serum cholesterol concentrations are associated with intracerebral haemorrhage, and some clinical trials of statins have reported an increased risk of intracerebral haemorrhage with statin treatment.5 This finding raises the question of whether there are potential, unrecognised detrimental effects of statins.
However, the pharmacological profile of statins, which includes pleiotrophic vascular and neuroprotective effects, such as increased expression of endothelial nitric oxide synthase, reduced inflammation, decreased platelet activation, and fibrinolytic and antioxidant effects, seems to be the perfect combination for treatment of subarachnoid haemorrhage.6 Experimental studies supported the efficacy of statins in subarachnoid haemorrhage, but those studies were of poor quality and small numbers and would not meet Stroke Academic Industry Roundtable (STAIR) criteria.2,7 Nevertheless, these findings led to various observational and randomised clinical studies of statins in subarachnoid haemorrhage. All of these studies were single-centre, phase 2 studies with fewer than 100 patients. The absence of consistent definitions and trial designs, and the use of different statins and doses (studies used simvastatin 20–80 mg or pravastatin 40 mg) limited comparison between studies, and no definitive results emerged. Now we have one arguably definitive answer about statins in subarachnoid haemorrhage. It is reasonable to ask whether this question could have been answered sooner if a more coordinated clinical trials network existed, as it is being developed in the USA.8 At the very least, these findings support the need for common data elements and definitions for trials in subarachnoid haemorrhage. The STASH trial1 was well designed by a competent, experienced group and the centres were also entirely appropriate for the study; so why did simvastatin not improve outcome? The authors have addressed some of the reasons, but other causes may exist. I agree that compliance with drug administration is unlikely to account for the absence of effect. Analysis of the subgroup of patients who had high drug compliance still showed absolutely no evidence for benefit. Patients were included within 96 h of subarachnoid haemorrhage, but was this timing too late? The mean time to treatment was about 2 days, which should be adequate in view of the time it takes for statin-induced effects to occur. It is possible that statins might be effective in a selected subgroup, but no effects in subgroups emerge here. Was the subarachnoid haemorrhage population biased to not benefit? This possibility seems unlikely because 17% of patients were randomised at sites that kept screening
www.thelancet.com/neurology Published online May 16, 2014 http://dx.doi.org/10.1016/S1474-4422(14)70095-X
Lancet Neurol 2014 Published Online May 16, 2014 http://dx.doi.org/10.1016/ S1474-4422(14)70095-X See Online/Articles http://dx.doi.org/10.1016/ S1474-4422(14)70084-5
1
Comment
logs, which is good for a clinical trial. Did simvastatin improve outcome but the effect go undetected on the modified Rankin Scale? Examination of the figures shows no evidence for a general shift in outcomes and the other clinical measures such as the SF-36 were not affected. The argument that reduction in angiographic vasospasm may not be associated with improved clinical outcome6 has less weight here, since the investigators assessed delayed cerebral ischaemia, which has a closer relation to outcome, and simvastatin had no effect. The authors note that the absence of effect of simvastatin could be due to the low incidence of delayed cerebral ischaemia in the study (about 16%, compared with 30% in general).5 However, not even a trend towards reduced delayed cerebral ischaemia or use of rescue therapy was detected, other than the single hypervolaemic therapy endpoint that was probably statistical chance. Nor was there any consistent effect in the subgroups at more risk for delayed cerebral ischaemia, such as those with poor clinical grade, higher Fisher grade, or who underwent aneurysm repair by clipping as opposed to coiling. Since statins might have an effect on neurogenesis and angiogenesis, it is intriging to ask if stopping simvastatin 21 days after subarachnoid haemorrhage caused some negative rebound effect.5 Does the type and dose of statin matter? Previous studies used simvastatin (20–80 mg per day) or pravastatin (40 mg per day).2 STASH used simvastatin, which is more lipophilic and might have higher blood– brain barrier permeability than pravastatin, and simvastatin 40 mg has twice the cholesterol-lowering effect of pravastatin 40 mg. A particularly interesting question is why the higher pharmacological dose of a more brain penetrating statin did not work in STASH, whereas the lower dose of pravastatin showed promise previously.9 Almost 200 patients in STASH and over twice as many in the previous Cambridge study9 came from the same centre as the study of Tseng and colleagues. A critical subgroup analysis would be the outcomes in patients from the Cambridge centre. If that subgroup did better with simvastatin in STASH, then critical insights into the treatment of subarachnoid haemorrhage might be gleaned. If they did not do better, then again, a detailed analysis of the possible reasons might be useful. If no reasons can be identified, then, as the authors suggest, we are stuck with the well known 2
issue of positive findings in small, single-centre studies that are not borne out in large, multicentre trials.10 What are the implications of the STASH results? I believe they should change practice. It is difficult to justify starting a statin on a patient in the acute phase after subarachnoid haemorrhage for the purpose of trying to improve their outcome. If the patient is already on a statin when they have subarachnoid haemorrhage, the treatment probably should be continued. The most logical next step is already underway: a study comparing 40 mg and 80 mg simvastatin.11 Fortunately, the design is very similar to STASH; hopefully, it will show that highdose simvastatin is efficacious. If not, then the statin story in subarachnoid haemorrhage might be over. R Loch Macdonald Division of Neurosurgery, St Michael’s Hospital, Labatt Family Centre of Excellence in Brain Injury and Trauma Research, Keenan Research Centre for Biomedical Science and the Li Ka Shing Knowledge Institute of St. Michael’s Hospital, Department of Surgery, University of Toronto, ON M5B 1W8, Canada [email protected] I receive grant support from the Physicians Services Incorporated Foundation, Brain Aneurysm Foundation, Canadian Institutes for Health Research, and the Heart and Stroke Foundation of Canada; and I am Chief Scientific Officer of Edge Therapeutics Inc. 1
2
3
4
5 6 7 8 9
10 11
Kirkpatrick PJ, Turner CL, Smith C, Hutchinson PJ, Murray GD, for the STASH Collaborators. Simvastatin in aneurysmal subarachnoid haemorrhage (STASH): a multicentre randomised phase 3 trial. Lancet Neurol 2014; published online May 16. http://dx.doi.org/10.1016/ S1474-4422(14)70084-5. Sabri M, Macdonald RL. Statins: a potential therapeutic addition to treatment for aneurysmal subarachnoid hemorrhage? World Neurosurg 2010; 73: 646–53. Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 2005; 366: 1267–78. Eisa-Beygi S, Hatch G, Noble S, Ekker M, Moon TW. The 3-hydroxy-3methylglutaryl-CoA reductase (HMGCR) pathway regulates developmental cerebral-vascular stability via prenylation-dependent signalling pathway. Dev Biol 2013; 373: 258–66. Flaster M, Morales-Vidal S, Schneck MJ, Biller J. Statins in hemorrhagic stroke. Expert Rev Neurother 2011; 11: 1141–49. Macdonald RL. Delayed neurological deterioration after subarachnoid haemorrhage. Nat Rev Neurol 2014; 10: 44–58. Macleod MR, Fisher M, O’Collins V, et al. Good laboratory practice: preventing introduction of bias at the bench. Stroke 2009; 40: e50–52. Landis S, Fisher M. Why the United States needs a network for stroke clinical trials. Stroke 2013; 44: 1217–18. Tseng MY, Hutchinson PJ, Czosnyka M, Richards H, Pickard JD, Kirkpatrick PJ. Effects of acute pravastatin treatment on intensity of rescue therapy, length of inpatient stay, and 6-month outcome in patients after aneurysmal subarachnoid hemorrhage. Stroke 2007; 38: 1545–50. Ioannidis JP. Why most published research findings are false. PLoS Med 2005; 2: e124. Wong GK, Liang M, Tan H, et al. High-dose simvastatin for aneurysmal subarachnoid hemorrhage: a multicenter, randomized, controlled, double-blind clinical trial protocol. Neurosurgery 2013; 72: 840–44.
www.thelancet.com/neurology Published online May 16, 2014 http://dx.doi.org/10.1016/S1474-4422(14)70095-X
