Abstract
Professor Goadsby
We move on to an even more challenging problem in some ways. I think when we give medicines to people they assume that they are safe and only really think about tolerability. The next presentation is on sumatriptan safety is by Professor Michael Welch, formerly working in Detroit but now at the University of Kansas. He has also worked in Houston and in Sydney. I wondered why Michael is looking younger and even more together today; but he has just given up the Editorship of Cephalalgia. Having dealt with a mountain of Journal correspondence just before I came here, I can see why he is feeling younger and I am feeling older. Mike did a great service for headache in the way he ran Cephalalgia as Editor-in-Chief in the 1990s and we owe him many thanks for that.
Michael Welch
Thank you Peter; that was a very pleasant introduction. I would like to begin by complimenting the company for allowing us access to their records concerning the adverse events of the triptans, sumatriptan and naratriptan in particular. I also want to acknowledge a number of colleagues who have looked at these data with me, namely Ninian Mathew, Peter Stone and Wayne Rossamund. This review is part of a publication which those of us that I mention share with Jane Sayer and Donna Gutterman. I am told that I was allowed access to these adverse event records because I had the reputation of being the most agreeable and easily influenced neurologist in the community. You laugh?
The adverse event profile of sumatriptan comes from more than 300000 migraine attacks observed during clinical trials in more than 60000 patients and from post-marketing experience up to around 1998. So we are looking at over 9 million patient exposures and more than 271 million migraine attacks. Please bear this in mind as I talk about some special studies that have been done to try to determine the mechanisms of the adverse events, particularly myocardial ischaemia, and then about the data from the post-marketing surveillance, which are the hardest to interpret.
The clinical trials demonstrate that the adverse events associated with the triptans are mild and not clinically serious. They occur early after drug administration, are short-lived (rarely longer than 1 h) and resolve spontaneously, which is a heartening message. You are very familiar with the commonest of these adverse events, the ‘triptan sensations’. Of most concern is the sensation of pressure, particularly pressure in the chest, and it is the question raised by these sensations that led to in vitro studies with isolated coronary artery strips and angiographic studies, especially one of blood flow in the myocardium. Our own studies (Tables 1 and 2) have explored skeletal muscle metabolism using phosphorus spectroscopy in patients given subcutaneous sumatriptan and I will also discuss some of the re-challenge studies done early on by the company.
Special studies of cardiac function employed
Serious cardiovascular adverse events registered
The seminal paper in 1989 by Helen Connor and her associates (see Fig. 1) showed that in coronary artery strips in vitro, there is a very slight, but definite, constrictor effect from sumatriptan, although it is considerably less than that from serotonin. It also demonstrated that the action of serotonin on the coronary arteries is through the 5-HT2 receptor, but that there was some constrictor effect via the 5-HT1-like receptors (as they were called at this time). There have been subsequent studies, particularly that of Massen Vandenbrink, which are very helpful in reviewing the constrictor effects of all of the triptans, presenting the concentration/response curves when the drugs are applied to isolated arterial strips (see Figs 2 and 3). Comparing the concentration response curves of serotonin, rizatriptan, zolmitriptan, sumatriptan and naratriptan, we see that all triptans have a constrictor effect in vitro but it is a minor one. Initially there was little concern about this with sumatriptan, zolmitriptan or rizatriptan, but with time, official caution has crept in.
Effect of sumatriptan on coronary artery. In vitro sumatriptan elicits only 21% of peak effect of 5HT (1). Reprinted from ref. (1) with permission from Elsevier Science.
Effect of 5-HT1 agonists on coronary artery (2). Reprinted from ref. (2) with the publisher’s permission.
Effect of 5-HT1 agonists on coronary artery (3). Reprinted from ref. (3) with the publisher's permission.
I recall a claim (4) that eletriptan has no coronary vasoconstrictor effect and therefore would be the triptan of choice in those patients with coronary risk factors (although, of course, no triptan can be given to people with frank coronary vascular disease). That opinion came from a study of intravenous eletriptan in 10 patients in whom there was no statistically significant coronary vasoconstriction. Nevertheless, one of those patients had very marked segmental constriction of the right coronary artery that persisted for a considerable period and was time-locked with the infusion of eletriptan. Although it was attributed to catheter spasm, that that was an imprudent statement and I encourage you to recall that all triptans have a vasoconstrictor effect on the coronary arteries, although this is mild. This is a class action of this group of drugs and they are all potential culprits when there is any evidence of myocardial ischaemia. It will remain for pharmacogenomics to work out whether or not there is indeed a genetic basis of susceptibility to the triptans in those very, very rare patients that develop myocardial ischaemia as a result of the use of these drugs.
From the angiographic studies conducted (see Tables 3 and 4) it appears that sc. sumatriptan given during coronary angiography compared to placebo does indeed lead to significant (if minor) constriction of the coronary arteries. When sumatriptan was given to patients with more than 50% stenosis (in other words, more marked coronary vascular disease), no statistically significant vasoconstrictor effect was found. Because part of the reason that constriction occurs is thought to be related to the development of atherosclerosis, this result was reassuring but unexpected.
Results of angiographic studies
MacIntyre PD et al. Br. J. Clin. Pharm. 1992; 34:541–546.
MacIntyre PD et al. Circulation 1993; 87:401–405.
Felman et al. Glaxo on file, in preparation.
A myocardial perfusion study was done with positron emission tomography in healthy migraineurs given sc. sumatriptan (see Table 4) and it was found that there was no significant reduction in myocardial blood flow. We all know that you can have constriction of arteries but because of auto-regulation of the distal resistance arteries, flow is balanced and does not change, which is also very reassuring. We have studied a group of patients who complained of muscle aches, pains and chest tightness after sumatriptan because we were interested in the suggestion (5) that mitochondria were abnormal in migraine patients. Knowing that 5-HT can worsen the mitochondrial abnormality, we exercised subjects with migraine and using spectroscopy we serially measured muscle phosphorus during recovery after exercise (6). Recovery of phosphocreatine was slower (although eventually complete) because of diminished oxygen stores in the tissues, in part due to sumatriptan-induced microvascular constriction (Fig. 4). Interestingly, we only see this effect in those patients who have limb aching pain and chest discomfort. Re-challenge studies of 20 patients who took sumatriptan during a migraine attack and experienced chest discomfort showed that three of them reported chest symptoms when re-challenged outside an attack. In no case was there any change in the ECG.
Myocardial perfusion studies
Lewis PJ et al. Neurology 1997; 48:1542–1550.
Sumatriptan effects on skeletal muscle energy metabolism. Kinetic traces of PCr vs. time during recovery from exercise in a patient with side effects to an injection of 6 mg sumatriptan before (circles) and after (squares) injection.
I want to finish by talking about the post-marketing reports of adverse experiences. You must understand that these reports can be given to the company by anybody – primary care doctors, pharmacists, the patients themselves or their family or by the janitor on the fourth floor. So you have to understand that these are hardly controlled reports. Alison Pilgrim and I have studied these reports, particularly the neurological and cardiovascular events of the post-marketing survey. For any deaths within 5 days of sumatriptan treatment, clearly cardiovascular causes are the most prominent, but neurological events also occurred, although very rarely. There have been about 12 cases of stroke – not necessarily leading to death – that might have been associated with the use of sumatriptan. It is well known that there is an increased risk of stroke in migraine patients, but true migraine-induced stroke is rare indeed. In the cases in which stroke occurred in temporal relationship with sumatriptan, other factors were associated that could also have predisposed to stroke, such as coagulopathy and dehydration, but we came to the conclusion that in just a few cases the sumatriptan was associated with stroke.
Spontaneous deaths from all causes within 24 h of sumatriptan are few indeed (Figs 5 and 6). As we all become more aware of the potential for bad effects, we select only patients appropriate to receive any triptan, so the adverse event profile is improving and with it, naturally, there is a better safety profile for this drug.
All spontaneous neurological deaths after sumatriptan (any formulation: any time post-dose).
Spontaneous deaths from all causes after sumatriptan.
The figures for serious cardiovascular events following sumatriptan injection show the same decline in the number of events, and the tablets seem to cause even fewer, reinforcing the message that as we get used to using the drug appropriately, the adverse event profile improves (Fig. 7). One serious cardiovascular adverse event was reported for the sumatriptan nasal spray in 1998; there have been none with naratriptan.
Serious cardiovascular adverse events within 24 h of sumatriptan injection or tablet.
This is the most widely studied acute migraine therapy. Significant cardiovascular events in this group are rare, and are identical among all the 5-HT1B agonists, and if you take home no other message today, I think that would be the one I would like you to take home. All triptans are contraindicated in patients with cardiovascular disease, and I extend that to patients with cerebral vascular disease, although the associations are more difficult to draw. Therefore, be cautious and do not use these drugs in patients with cardiovascular risk factors. Follow the precise and comprehensive labelling, for there is a strongly positive efficacy/safety ratio if you use these drugs in accordance with prescribing information.
Thank you.
Professor Goadsby
Thanks Mike, that is very clear. Let me echo the very strong statement you made. God put 5-HT1B receptors on the human coronaries; if you want to find out when they are going to be taken off, go down to the Vatican and ask His representative, not the representative of a drug company. But they are not going to be taken off the human coronaries for a long time.