s73 Clinical Neurology and Neurosurgery, 94 (Suppl.) (1992) S73 -S77 Q 1992 Elsevier Science Publishers B.V. All rights reserved 0303~8467/92/$05.00 CNN 00113
Clinical effects and mechanism of action of sumatriptan in migraine Michel D. Ferrarilt3 and Pramod
R.S. Saxena2’3
’ Depanment of Neurologv Universily Hospital, Leiakn (The Netherlands), ’ Depanment of Pharmacology, Erasmus Universiy, Rouerdum (The Netherlands), and 3 The Durch Migraine Research Group (The Netherlands)
Key words: Sumatriptan;
Clinical
efficacy; 5-HT receptors;
Vasoconstriction;
Migraine;
Cluster
headache
Summary Sumatriptan is a novel, highly effective drug against migraine and cluster headache attacks. It shows a remarkably selective pharmacological profile in animals. Determination of its mechanism of action in human should further the understanding of the pathophysiology of migraine and cluster headache. We, therefore, review current knowledge on the clinical and pharmacological effects of sumatriptan. Important pharmacological actions of sumatriptan are (i) poor penetration of the blood-brain barrier suggesting a peripheral point of action; (ii) 5-HT,-like/S-HT1,, receptor-mediated vasoconstriction of large cerebral arteries and dural vessels; and (iii) blockade of neurogenic dural inflammation via 5-HTld autoreceptor-mediated inhibition of vasoactive neuropeptides within the trigeminovascular system. Future research will tell which mechanism is involved in the pathophysiology of migraine and cluster headache.
Introduction Recently, the selective 5-HT,-like/S-HT,, receptor agonist sumatriptan has been introduced as a novel, highly effective treatment of migraine and cluster headache attacks [l--S]. Because of the selective pharmacological profile of sumatriptan, studying its mechanism of action may help to unravel the pathophysiology of both headache syndromes. Here, we review and correlate the clinical and pharmacological actions of sumatriptan. Clinical
effects
Sumatriptan is highly effective and rapidly acting drug against all features of the headuclte please of migraine attacks (headache, nausea, vomiting, photo- and phonophobia), both in attacks which are preceded by an aura and which are not [2,6]. Its effects on the aura symptoms are as yet unknown, but probably nil (vide i&a). Following subcutaneous (s.c.) administration the bioavailability of sumatriptan is about 96% [9]. After one S.C. injection with 6 mg sumatriptan, about 50% of the patients significantly improved within 30 min, 72% at 1 h and nearly 90% at 2 h [2,8]. A 6-mg dose appears optimal [2]. Those Correspondence to: Michel D. Ferrari, MD, Department of Neurology, University Hospital, P.O. Box 9600, 2300 RC Leiden, The Netherlands. Tel.: (71) 261641 / 262134; Fax: (71) 154537.
patients who do not improve sufficiently with a 6 mg dose, will also not improve when a second 6 mg dose is given [2]. Following oral administration, the bioavailability of sumatriptan is only about 14% [9]. The response rates are also lower (about 67% at 2 h), but still considerable and clearly better than placebo (27%) [4] or ergotamine (48%) [lo]. The optimal oral dose of sumatriptan is one tablet of 100 mg [4]. Most remarkably and in sharp contrast to crgotamine, sumatriptan is equally effective when given early or late in the migraine attack, even after scvera1 hours [2]. Sumatriptan is also highly and remarkably rapidly effective in the acute treatment of cluster headache attacks [7]. About 50% improves within 10 min and 74% within 15 min after one S.C. injection with 6 mg sumatriptan. The effect of sumatriptan on the aura symptoms of a migraine attack is as yet unknown. Aura symptoms are believed to be due to either vasoconstriction [ll] or spreading depression [12]. The vasoconstriction may be either at the arterial [11,13] or at the urteriolur level [14]. Sumatriptan, when administered systemically, causes primarily constriction of large cerebral conductance vessels and probably not of microvessels and is not known to cause significant rCBF changes (vide infra). Therefore, sumatriptan is cxpccted not to interfere with the aura symptoms and hence should bc given safely already during the aura phase. Anecdotal reports of 3 patients with
s74 familial hemiplegic migraine, who have used the drug during the hemiplegic phase of an attack, indeed confirm this notion (personal observation). The safety of sumatriptan given during the aura phase is currently being investigated in a controlled trial. Until these data are available, sumatriptan should not be given during the aura phase of a migraine attack. Its administration should rather be delayed until after the aura symptoms have disappeared. The results of this trial should also contribute to the further understanding of the mechanism of the aura. The “acute” side effects of sumatriptan are usually early after its administration, short lasting and mild [2,15]. Apical systemic symptoms are tingling, warm or hot feelings, feelings of heaviness or pressure and flushing. Only limited information is available with respect to long-term tolerability. A conspicuous feature of treatment of migraine attacks with sumatriptan is that in at least l/3 of the cases the headache recurs within 24 h after initial rapid abolishing [2,3,8]. The median time to recurrence is about 10-14 h [2]. Interestingly, headache recurrences also occurred in 18% of the placebo-treated patients after a median time of about 9 h [2] and have also been described in 5-HT-treated spontaneous as well as reserpine-induced attacks [X-18] and in ergotamine-treated spontaneous attacks [lo]. The headache recurrences following treatment with sumatriptan may be severe, but can usually, be treated with a repeated dose of sumatriptan (personal observation). Whether this indeed is the case, is currently being investigated in a formal clinical trial. In some patients sumatriptan-treated attacks seem to have a more protracted course. Whereas the usual attack would last about 2 days, the sumatriptan-treated attack may be spread over 5 days, during which headache-free hours are alternating with headache-recurrences, which are each time treated successfully by a repeated dose of sumatriptan (personal observation). The exact mechanism of the recurrences after sumatriptan is unknown. It appears that, due to the short plasma half-life of sumatriptan of about 2 h [9], the original attack breaks through again, after initial suppression of the migraine symptoms for several hours. The notion that it is the original attack which breaks through again (and not a new attack) is based on the following: (i) the symptoms of the headache-recurrence, are usually of the same quality as before treatment, (ii) the headache and associated symptoms usually recur within the usual duration of a for that particular patient typical attack ([19]; personal observation), and (iii) some patients may tell that although the headache and associated features are disappeared after the sumatriptan, they still have some ill-defined feeling in their head as if “the migraine is suppressed though still working”. Lastly, it should be em-
phasized that the headache and associated symptoms recur but never the aura or premonitory symptoms. This would suggest that the recurrence takes place at a point in the migraine cascade beyond the actual onset of the attack, e.g. a suppressed but still ongoing headache generator. If the short plasma half-life of sumatriptan is indeed important, the administration of an additional dose of sumatriptan given some hours after the first dose, should be effective in preventing or delaying at least some of these recurrences. This is currently under investigation. General pharmacological
properties
In animal studies, sumatriptan lacks anti-nociceptive effects [20] and does not cross the blood-brain barrier readily [21-231. However, no human data are available and we must consider the possibility that the functional properties of the blood-brain barrier may alter during a migraine attack [24]. Rare, but sometimes prominent side-effects such as transient drowsiness, sedation, diiness, vertigo and fatigue may point to snme central effects of sumatriptan during an attack [2,1.5]. Although these symptoms may also be part of the recovery phase of a migraine attacks it should be emphasized that these putative central effects are not seen in the limited experience with sumatriptan in healthy volunteers [25] or migraine patients outside an attack (personal observation). Vascular effects
Sumatriptan is a potent vasoconstrictor of primarily cerebral vessels in animal and man [26]. The drug contracts isolated basilar arteries of various species 127-291 and man [30] and bloodvessels within human isolated perfused dura mater [31]. In vivo animal studies sumatriptan caused a dose-dependent decrease in carotid arterial blood flow and increase of carotid arterial vascular resistance [32]. This was shown to be entirely due to a selective vasoconstriction of arteriovenous anastornoses (AVAs) within the carotid vascular bed, without a noticeable effect on the cerebral or extracerebral circulation [33,34]. Perivascular application of sumatriptan caused concentration-related constriction of pial vessels, while iv administration of sumatriptan caused con&i&on of the carotid vascular bed but not of pial vessels [22]. Thus following systemic administration sumatriptan does not seem to penetrate the vessel wall intima and does not affect rCBE Limited information is available on the vascular actions of sumatriptan in human. Sumatriptan increases blood flow velocity (BFV) in the internal carotid artery
S75 (ICA) and middle cerebral artery (MCA) [35-371, but not in the external and common carotid artery [36,37], in migraine patients both during [35-371 and between (Caekebeke et al., in preparation) migraine attacks. During attacks, the increase of BFV appears related to the clinical response and dose [36,37]. Six mg was found to be optimal both in terms of clinical improvement [2] and increase of BFV [36,37]. Because increase in BFV probably reflects vasoconstriction [35,36] it was concluded that sumatriptan constricts large cerebral conductance vessels such as the ICA and MCA. Diener et al. [38] investigated BFV in the MCA and basilar artery (BA) of 6 patients during the withdrawal phase from ergotamine-induced headache, before and after 4 mg S.C. sumatriptan. Despite (transient) clinical improvement, they failed to demonstrate a significant change in BFV The lack of change in BFV may have been related either to the small number of patients or to some counteracting action in ergotamine-dependent patients. This would suggest that the clinical effect of sumatriptan in treating ergotamine-withdrawal symptoms may be independent of its vasoconstrictor action on large conductance vessels. The effects of sumatriptan on rCBF, i.e. at the microvasculature level, are complex. When considering cortical (carotid) flow only, no consistent uniform changes could be demonstrated after sumatriptan [35,39]. However, by directly comparing the basilar and carotid blood flow, a differential opposing effect on both vascular areas was found during migraine attacks, normalizing the balance of the carotid/basilar flow towards attack-free values [39]. Relevance of vasoconstrictor
action
Vasoconstriction appears important for anti-migraine and anti-cluster headache action of drugs, because efficacious drugs all share the ability to cause vasoconstriction within the carotid circulation [40]. However, recently it has been shown that the vasoconstrictor actions of sumatriptan and ergotamine are probably mediated via different receptor subtypes [41]. Further, clinical improvement from ergotamine withdrawal symptoms by sumatriptan may be independent of its vasoconstrictor action on the large conductance vessels [38]. However, the possibility that constriction of for instance dural vessels is involved remains open. Neuronal
effects
Neurogenic inflammation (vasodilation and plasma protein extravasation) in dura mater following stimulation of the trigeminovascular system, is suggested to be important to the mechanism of the headache in migraine
and cluster headache [42,43]. Neurogenic inflammation is, at least in part, mediated by release of vasoactive neuropeptides such as substance P, neurokinin A and calcitonin gene-related peptide (CGRP) [42&l+. In the rat model, systemic infusion of capsaicin may cause plasma extravasation within dura mater via an axonal reflex, whereas treatment with substance P or neurokinin A may cause plasma extravasation via a direct post-synaptic mechanism [47]. Electrical trigeminal ganglion stimulation causes, in addition to plasma extravasation within dura mater [47], also ultrastructural changes in bloodvessels and mast cells within dura mater [48], plasma extravasation within extracranial cephalic tissues including conjunctiva, lip and eyelid [44,47] and increase of CGRP in the superior sagittal sinus [49]. Moskowitz and colleagues [44,45,50,51] demonstrated that pre-treatment with clinical relevant doses of antimigraine drugs such as ergotamine tartrate, dihydergot (DHE) and sumatriptan or pre-treatment with 5-HT,,,,, receptor agonists selectively block plasma extravasation from bloodvessels in dura mater following electrical trigeminal ganglion stimulation, but not in the extracranial tissues. Sumatriptan, ergotamine tartrate and DHE also blocked plasma extravasation following systemic capsaicin, but not following systemic treatment with substance P or ncurokinin A. The blocking effect of sumatriptan on plasma extravasation was partially antagonized by pre-treatment with the 5-HT,,,, receptor antagonist metergoline, but not by pre-treatment with the 5-HT,-like receptor antagonist methiothepin or 5-HT, or 5-HT, antagonists. Pre-treatment with sumatriptan or DHE also attenuated CGRP increase in superior sagittal sinus and ultrastructural changes in bloodvessels and mast cells within dura mater following electrical trigeminal ganglion stimulation [49]. It was concluded that the blocking effect of sumatriptan and ergot alkaloids on neurogenic plasma extravasation is not related to their vasoconstrictor action, nor to a post-synaptic cffcct, but rather to a pre-synaptic action, presumably by inhibition of the release of vasoactive neuropcptidcs mediated by activation of 5-HTlbild autoreceptors on sensory fibers [44,45]. Human evidence in support of a significant role of the trigeminovascular system and CGRP in migraine can be derived from the elegant studies of Goadsby et al., showing that (i) thermocoagulation of the trigeminal ganglion causes a marked elevation of plasma levels of substance P and CGRP in the ipsilateral external jugular vein [52], (ii) during migraine attacks there is a selective increase in plasma levels of CGRP [53], and (iii) sumatriptan normalizes thcsc elevated CGRP levels [54].
S76 Interactions with neurotransmitter
receptor sites
Several lines of experiments suggest that in human sumatriptan acts via a 5-HTi,, receptor-mediated mechanism. First, sumatriptan extremely selectively binds to 5-HT,, receptors and activates functional biochemical models of 5-HT,, receptors [SS-581. Second, the rank order of effective plasma extravasation-blocking doses of sumatriptan in Moskowitz’s rat model was most consistent with a 5-HT,, receptor mediated response (vide supra [44]). Finally, sumatriptan is essentially equipotent to other clinically effective abortive anti-migraine agents with respect to 5-HT,, agonism [55-581. The 5-HT,, receptor is the most common type of 5-HT receptor subtype in human brain [59-61] and functions as an auto-receptor, controlling the release of 5-HT [21] and other neurotransmitters such as NE and ACh [62,63]. The 5-HTid receptor is also believed to closely resemble the vascular 5-HT,-like receptor [27,30,32,647]. It thus seems that sumatriptan is a highly selective agonist at a subpopulation of 5-HT receptors [55-581, designated “5-HT,-like” in the vasculature [68,69] and 5-HTld in nervous tissue [27,30,32,64,65]. Conclusion Sumatriptan is a novel, most effective treatment of migraine and cluster headache attacks. Acute side effects appear only mild. Long-term experience, however, is limited. Main disadvantage in the clinical use are the headache recurrences. These appear to be due to short plasma half-life of the drug, causing too short of a suppression of a still ongoing “headache generator”. Current clinical research focusses on how to prevent and treat these recurrences. Three pharmacological actions of sumatriptan following systemic administration appear pathophysiologically informative. First, poor penetration of the blood-brain barrier suggesting a peripheral point of action. Second, 5-HTi-like/S-HT,, receptor-mediated ability to cause vasoconstriction, primarily of large cerebral conductance arteries and dural vessels. Third, blocking effect of neurogenic dural inflammation, possibly via a pre-synaptic 5-HT,, autoreceptor-mediated inhibition of vasoactive neuropeptides within the trigeminovascular system. Which of these mechanisms is relevant to the clinical mechanism of action of sumatriptan is the focus of intensive scientific research. The results should increase the understanding of the pathophysiology of migraine and cluster headache.
References 1 Ferrari MD, Bayliss EM, Ludlow S, Pilgrim AJ. Cephalalgia 1989; 9(S.10): 348-349. 2 The subcutaneous sumatriptan international study group (Ferrari MD, Melamed E, Gawel M et al). New Engl J Med 1991; 325: 316-321. 3 Goadsby PJ, Zagami AS, Donnan GA, Symington G, Anthony M, Bladin PF, Lance JW. Lancet 1991; 338: 782-783. 4 Sumatriptan - an oral dose-defining study. Eur Neural 1991; 31: 300-30s. 5 The sumatriptan auto-injector study group. Eur Neurol 1991; 31: 323-331. 6 The oral sumatriptan international multiple-dose study group.Eur Neurol 1991; 31: 306-313. 7 Ekbom K et al. New Engl J Med 1991; 32.5: 322-326. 8 Cady RK, Wendt JK, Kirchner JR, Sargent JD, Rothrock JF, Skaggs I I. JAMA 1991; 265: 2831-2835. 9 Fowler PA, Lacey LF, Thomas M, Keene ON, Tanner RJN, Baber NS. Eur Neurol 1991; 31: 291-294. 10 The multinational oral sumatriptan and cafergot comparative study group. Eur Neurol 1991; 31: 314-322. 11 Wolff IIG. New York; Oxford University Press, 1963. 12 Lauritzen M. Trends Neurosci 1987; 10: 8-13. 13 Solomon S, Lipton RB, Harris PY. Headache 1990; 30: 52-61. 14 Fribcrg L, Olsen TS, Roland PE, Lassen NA. Brain 1987; 110~ 917-934. 15 Brown EG, Endersby CA, Smith RN, Talbot JCC. Eur Neural 1991; 31: 339-344. 16 Anthony MA, Hinterberger H, Lance JW. Arch Neural 1%7; 16: 544-554. 17 Carroll JD, Hilton BP. Headache 1974; 14: 149-156. 18 Kimball RW, Friedman AP, Vallejo E. Neurology196o; 10: 107-111. 19 Tfelt-I Hansen P, Brand J, Dano P, et al. Cephalalgia 1989; 9 (S.9): 73-77. 20 Skingle M, Birch PJ, Leighton GE, Humphrey PPA. Cephalalgia 1990; 10: 207-212. 21 Sleight AJ, Cervenka A, Peroutka SJ. Neuropharmacology 1990; 29: 551-513. 22 1Iumphrey PPA, Connor IIE, Stubbs CM, Feniuk W. In: Jes Olesen, editor. Migraine and other headaches: the vascular mechanisms. New York-Raven Press Ltd, 1991: 335-339. 23 tlumphrey PPA, Feniuk W, Perren MJ, Beresford IJM, Skingle M. Ann NY Acad Sci 1990; 600: 587-598. 24 Parsons AA. Trends Pharmacol Sci 1991; 12: 310-315. 25 Fowler PA, Thomas M, Lacey LF, Andrew P, Dallas FAA. Gephalalgia 1989; 9(S. 9): 59-62. 26 Harper AM, MacKenzie ET. J Physiol 19n, 271: 721-733. 27 Connor IIE, Feniuk W, Humphrey PPA. Br J Pharmaeol1989; 96: 379-387. 28 Parsons AA, Whalley ET. Int J Microcirc Clin Exp 1990; 9: 227. 29 Parsons AA, Whalley ET. Eur J Pharmacol1989; 174: 189-l%. 30 Parsons AA, Whalley ET, Feniuk W, Connor HE, Humphrey PPA. Br J Pharmacol 1989; 96: 434-449. 31 IIumphrey PPA, Feniuk W, Motevalian M, Parsons AA, Whalley ET. In: Fozard JR, Saxena PR, eds. Serotonin: molecular biology, receptors and functional effects. Base]: BirkhZuser Verlag, 1991: 2&t-274. 32 Fcniuk W, Humphrey PPA, Perren MJ. Br J Pharmacol 1989; 96: 83-90. 1989; 9(S 9): 33 Perren MJ, Feniuk W, Humphrey PPA. Cephalalgia 41116. 34 Boer MO den, VillaIon CM, Heiligers JPC, Humphrey PPA, Saxena PR. Br J Pharmacoll991; 102: 323-330. 35 Fribcrg L, Olesen J, Iversen HK, Sperling B. Lancet 1991; 338: 13-17.
s77 36 Caekebeke JFV, Zwetsloot CP, Jansen J, Saxena PRS, Ferrari MD. In: Olesen .I (ed), Frontiers in Hedacahe Research. Volume 1: Migraine and other headaches: the vascular mechanisms. Raven Press: New York, 1991: 331-334. 37 Caekebeke JFV, Ferrari MD, Zwetsloot CP, Jansen J, Saxena PRS. Neurology 1992, in press. 38 Diener HC, Peters C, Rudzio M, et al. J Neurol1991; 238: 245-250. 39 Ferrari MD, Haan J, Blokland JAK, Minnee P, Zwinderman KH, Saxena PR In: Jes Olesen, editor. Migraine and other headaches: the vascular mechanism. NewYork: Raven Press Ltd, 1991: 245-248. 40 Saxena PR and Ferrari MD. Trends Pharmacol Sci 1989; 10: 200204. 41 Boer MO den, Helligers JPC, Ferrari MD, Saxena PR. Cephalalgia 1991; ll(S 11): 203-204. 42 Moskowitz MA, Henrikson BM, Markowitz S, Saito K. In: Olesen J and Edvinsson L, eds. Basic Mechanism of Headache. Amsterdam: Elsevier Science Publ, 1988: 429-437. 43 Moskowitz MA. Headache 1988; 28: 584-586. 44 Buzzi MG, Moskowitz MA, Peroutka SJ, Buyn B. Br J Pharmacol 1991; 103: 1421-1428. 45 Buzzi MG, Moskowitz MA. Cephalalgia 1991; 11: 165-168. 46 McCulloch J, Uddman R, Kingman TA, Edvinsson L. Proc Natl Acad Sci USA 1986; 83: 5731-5735. 47 Markowitz S, Saito K, Moskowitz MA. J Neurosci 1987; 7: 41294136. 48 Drummond PD, Lance JW. Ann Neurol1983; 13: 32-37. 49 Buzzi MG, Carter WB, Shimizu T, Heath HIII, Moskowitz MA. Neuropharmacology 1991; 30: 1193-1200.
50 Saito K, Markowitz 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
67 68 69
S, Moskowitz
MA. Ann Neurol1988;
24: 732-
737. Buzzi MG, Moskowitz MA. Br J Pharmacoll990; 99: 202-206. Goadsby PJ, Edvinsson L, Ekman R. Ann Neurol1988; 23: 193-196. Goadsby PJ, Edvinsson L, Ekman R. Ann Neurol1990,28: 183-187. Goadsby PJ, Edvinsson L. Cephalalgia 1991; ll(S 11): 3-4. Deliganis AV, Peroutka ST. Headache 1991; 31: 22&231. Peroutka ST, McCarthy BG. Eur J Pharmacol1989; 163: 133-136. Waeber C, Iloyer D, Palacios JM. Synapse 1989; 4: 168-170. Schoeffter P, IIoyer D. Arch Pharmacol 1989; 340: 135-138. Waeber C, Schoeffter P, Palacios JM, Hoyer D. Arch Pharmacol 1988; 337: 595-601. Wacber C, Diet1 MM, IIoyer D, Probst A, Palacios JM. Neurosci Lett 1988; 88: 11-16. Peroutka SJ, Switzer JA, Hamik A. Synapse 1989; 3: 61-66. Schlicker E, Fink K, Gothert M, et al. Arch Pharmacol 1989; 340: 45-51. Hoyer D, Middlemiss DN. TIPS 1989; 10: 130-132. Humphrey PPA, Feniuk W, Perren MJ, et al. Br J Pharmacol1988; 94: 1123-1132. Schoeffter P, Iloyer D. J Pharmacol Exp Ther 1990; 252: 387-396. Ilamel E, Bouchard D. Saxena PR, Eds. Serotonin: molecular biology, receptors and functional effects. Basel: Birkhauser Verlag, 1991: 264-274. Ilamel E, Bouchard D. Br J Pharmacoll991; 102: 227-233. Ilumphrey PPA, Feniuk W, Perren MJ, Connor HE, Oxford AW. Cephalalgia 1989; 9(S 9): 23-33. Sumner MJ, IIumphrey PPA. Br J Pharmacol 1990; 99: 219-220.
Clinical effects and mechanism of action of sumatriptan in migraine Michel D. Ferrarilt3 and Pramod
R.S. Saxena2’3
’ Depanment of Neurologv Universily Hospital, Leiakn (The Netherlands), ’ Depanment of Pharmacology, Erasmus Universiy, Rouerdum (The Netherlands), and 3 The Durch Migraine Research Group (The Netherlands)
Key words: Sumatriptan;
Clinical
efficacy; 5-HT receptors;
Vasoconstriction;
Migraine;
Cluster
headache
Summary Sumatriptan is a novel, highly effective drug against migraine and cluster headache attacks. It shows a remarkably selective pharmacological profile in animals. Determination of its mechanism of action in human should further the understanding of the pathophysiology of migraine and cluster headache. We, therefore, review current knowledge on the clinical and pharmacological effects of sumatriptan. Important pharmacological actions of sumatriptan are (i) poor penetration of the blood-brain barrier suggesting a peripheral point of action; (ii) 5-HT,-like/S-HT1,, receptor-mediated vasoconstriction of large cerebral arteries and dural vessels; and (iii) blockade of neurogenic dural inflammation via 5-HTld autoreceptor-mediated inhibition of vasoactive neuropeptides within the trigeminovascular system. Future research will tell which mechanism is involved in the pathophysiology of migraine and cluster headache.
Introduction Recently, the selective 5-HT,-like/S-HT,, receptor agonist sumatriptan has been introduced as a novel, highly effective treatment of migraine and cluster headache attacks [l--S]. Because of the selective pharmacological profile of sumatriptan, studying its mechanism of action may help to unravel the pathophysiology of both headache syndromes. Here, we review and correlate the clinical and pharmacological actions of sumatriptan. Clinical
effects
Sumatriptan is highly effective and rapidly acting drug against all features of the headuclte please of migraine attacks (headache, nausea, vomiting, photo- and phonophobia), both in attacks which are preceded by an aura and which are not [2,6]. Its effects on the aura symptoms are as yet unknown, but probably nil (vide i&a). Following subcutaneous (s.c.) administration the bioavailability of sumatriptan is about 96% [9]. After one S.C. injection with 6 mg sumatriptan, about 50% of the patients significantly improved within 30 min, 72% at 1 h and nearly 90% at 2 h [2,8]. A 6-mg dose appears optimal [2]. Those Correspondence to: Michel D. Ferrari, MD, Department of Neurology, University Hospital, P.O. Box 9600, 2300 RC Leiden, The Netherlands. Tel.: (71) 261641 / 262134; Fax: (71) 154537.
patients who do not improve sufficiently with a 6 mg dose, will also not improve when a second 6 mg dose is given [2]. Following oral administration, the bioavailability of sumatriptan is only about 14% [9]. The response rates are also lower (about 67% at 2 h), but still considerable and clearly better than placebo (27%) [4] or ergotamine (48%) [lo]. The optimal oral dose of sumatriptan is one tablet of 100 mg [4]. Most remarkably and in sharp contrast to crgotamine, sumatriptan is equally effective when given early or late in the migraine attack, even after scvera1 hours [2]. Sumatriptan is also highly and remarkably rapidly effective in the acute treatment of cluster headache attacks [7]. About 50% improves within 10 min and 74% within 15 min after one S.C. injection with 6 mg sumatriptan. The effect of sumatriptan on the aura symptoms of a migraine attack is as yet unknown. Aura symptoms are believed to be due to either vasoconstriction [ll] or spreading depression [12]. The vasoconstriction may be either at the arterial [11,13] or at the urteriolur level [14]. Sumatriptan, when administered systemically, causes primarily constriction of large cerebral conductance vessels and probably not of microvessels and is not known to cause significant rCBF changes (vide infra). Therefore, sumatriptan is cxpccted not to interfere with the aura symptoms and hence should bc given safely already during the aura phase. Anecdotal reports of 3 patients with
s74 familial hemiplegic migraine, who have used the drug during the hemiplegic phase of an attack, indeed confirm this notion (personal observation). The safety of sumatriptan given during the aura phase is currently being investigated in a controlled trial. Until these data are available, sumatriptan should not be given during the aura phase of a migraine attack. Its administration should rather be delayed until after the aura symptoms have disappeared. The results of this trial should also contribute to the further understanding of the mechanism of the aura. The “acute” side effects of sumatriptan are usually early after its administration, short lasting and mild [2,15]. Apical systemic symptoms are tingling, warm or hot feelings, feelings of heaviness or pressure and flushing. Only limited information is available with respect to long-term tolerability. A conspicuous feature of treatment of migraine attacks with sumatriptan is that in at least l/3 of the cases the headache recurs within 24 h after initial rapid abolishing [2,3,8]. The median time to recurrence is about 10-14 h [2]. Interestingly, headache recurrences also occurred in 18% of the placebo-treated patients after a median time of about 9 h [2] and have also been described in 5-HT-treated spontaneous as well as reserpine-induced attacks [X-18] and in ergotamine-treated spontaneous attacks [lo]. The headache recurrences following treatment with sumatriptan may be severe, but can usually, be treated with a repeated dose of sumatriptan (personal observation). Whether this indeed is the case, is currently being investigated in a formal clinical trial. In some patients sumatriptan-treated attacks seem to have a more protracted course. Whereas the usual attack would last about 2 days, the sumatriptan-treated attack may be spread over 5 days, during which headache-free hours are alternating with headache-recurrences, which are each time treated successfully by a repeated dose of sumatriptan (personal observation). The exact mechanism of the recurrences after sumatriptan is unknown. It appears that, due to the short plasma half-life of sumatriptan of about 2 h [9], the original attack breaks through again, after initial suppression of the migraine symptoms for several hours. The notion that it is the original attack which breaks through again (and not a new attack) is based on the following: (i) the symptoms of the headache-recurrence, are usually of the same quality as before treatment, (ii) the headache and associated symptoms usually recur within the usual duration of a for that particular patient typical attack ([19]; personal observation), and (iii) some patients may tell that although the headache and associated features are disappeared after the sumatriptan, they still have some ill-defined feeling in their head as if “the migraine is suppressed though still working”. Lastly, it should be em-
phasized that the headache and associated symptoms recur but never the aura or premonitory symptoms. This would suggest that the recurrence takes place at a point in the migraine cascade beyond the actual onset of the attack, e.g. a suppressed but still ongoing headache generator. If the short plasma half-life of sumatriptan is indeed important, the administration of an additional dose of sumatriptan given some hours after the first dose, should be effective in preventing or delaying at least some of these recurrences. This is currently under investigation. General pharmacological
properties
In animal studies, sumatriptan lacks anti-nociceptive effects [20] and does not cross the blood-brain barrier readily [21-231. However, no human data are available and we must consider the possibility that the functional properties of the blood-brain barrier may alter during a migraine attack [24]. Rare, but sometimes prominent side-effects such as transient drowsiness, sedation, diiness, vertigo and fatigue may point to snme central effects of sumatriptan during an attack [2,1.5]. Although these symptoms may also be part of the recovery phase of a migraine attacks it should be emphasized that these putative central effects are not seen in the limited experience with sumatriptan in healthy volunteers [25] or migraine patients outside an attack (personal observation). Vascular effects
Sumatriptan is a potent vasoconstrictor of primarily cerebral vessels in animal and man [26]. The drug contracts isolated basilar arteries of various species 127-291 and man [30] and bloodvessels within human isolated perfused dura mater [31]. In vivo animal studies sumatriptan caused a dose-dependent decrease in carotid arterial blood flow and increase of carotid arterial vascular resistance [32]. This was shown to be entirely due to a selective vasoconstriction of arteriovenous anastornoses (AVAs) within the carotid vascular bed, without a noticeable effect on the cerebral or extracerebral circulation [33,34]. Perivascular application of sumatriptan caused concentration-related constriction of pial vessels, while iv administration of sumatriptan caused con&i&on of the carotid vascular bed but not of pial vessels [22]. Thus following systemic administration sumatriptan does not seem to penetrate the vessel wall intima and does not affect rCBE Limited information is available on the vascular actions of sumatriptan in human. Sumatriptan increases blood flow velocity (BFV) in the internal carotid artery
S75 (ICA) and middle cerebral artery (MCA) [35-371, but not in the external and common carotid artery [36,37], in migraine patients both during [35-371 and between (Caekebeke et al., in preparation) migraine attacks. During attacks, the increase of BFV appears related to the clinical response and dose [36,37]. Six mg was found to be optimal both in terms of clinical improvement [2] and increase of BFV [36,37]. Because increase in BFV probably reflects vasoconstriction [35,36] it was concluded that sumatriptan constricts large cerebral conductance vessels such as the ICA and MCA. Diener et al. [38] investigated BFV in the MCA and basilar artery (BA) of 6 patients during the withdrawal phase from ergotamine-induced headache, before and after 4 mg S.C. sumatriptan. Despite (transient) clinical improvement, they failed to demonstrate a significant change in BFV The lack of change in BFV may have been related either to the small number of patients or to some counteracting action in ergotamine-dependent patients. This would suggest that the clinical effect of sumatriptan in treating ergotamine-withdrawal symptoms may be independent of its vasoconstrictor action on large conductance vessels. The effects of sumatriptan on rCBF, i.e. at the microvasculature level, are complex. When considering cortical (carotid) flow only, no consistent uniform changes could be demonstrated after sumatriptan [35,39]. However, by directly comparing the basilar and carotid blood flow, a differential opposing effect on both vascular areas was found during migraine attacks, normalizing the balance of the carotid/basilar flow towards attack-free values [39]. Relevance of vasoconstrictor
action
Vasoconstriction appears important for anti-migraine and anti-cluster headache action of drugs, because efficacious drugs all share the ability to cause vasoconstriction within the carotid circulation [40]. However, recently it has been shown that the vasoconstrictor actions of sumatriptan and ergotamine are probably mediated via different receptor subtypes [41]. Further, clinical improvement from ergotamine withdrawal symptoms by sumatriptan may be independent of its vasoconstrictor action on the large conductance vessels [38]. However, the possibility that constriction of for instance dural vessels is involved remains open. Neuronal
effects
Neurogenic inflammation (vasodilation and plasma protein extravasation) in dura mater following stimulation of the trigeminovascular system, is suggested to be important to the mechanism of the headache in migraine
and cluster headache [42,43]. Neurogenic inflammation is, at least in part, mediated by release of vasoactive neuropeptides such as substance P, neurokinin A and calcitonin gene-related peptide (CGRP) [42&l+. In the rat model, systemic infusion of capsaicin may cause plasma extravasation within dura mater via an axonal reflex, whereas treatment with substance P or neurokinin A may cause plasma extravasation via a direct post-synaptic mechanism [47]. Electrical trigeminal ganglion stimulation causes, in addition to plasma extravasation within dura mater [47], also ultrastructural changes in bloodvessels and mast cells within dura mater [48], plasma extravasation within extracranial cephalic tissues including conjunctiva, lip and eyelid [44,47] and increase of CGRP in the superior sagittal sinus [49]. Moskowitz and colleagues [44,45,50,51] demonstrated that pre-treatment with clinical relevant doses of antimigraine drugs such as ergotamine tartrate, dihydergot (DHE) and sumatriptan or pre-treatment with 5-HT,,,,, receptor agonists selectively block plasma extravasation from bloodvessels in dura mater following electrical trigeminal ganglion stimulation, but not in the extracranial tissues. Sumatriptan, ergotamine tartrate and DHE also blocked plasma extravasation following systemic capsaicin, but not following systemic treatment with substance P or ncurokinin A. The blocking effect of sumatriptan on plasma extravasation was partially antagonized by pre-treatment with the 5-HT,,,, receptor antagonist metergoline, but not by pre-treatment with the 5-HT,-like receptor antagonist methiothepin or 5-HT, or 5-HT, antagonists. Pre-treatment with sumatriptan or DHE also attenuated CGRP increase in superior sagittal sinus and ultrastructural changes in bloodvessels and mast cells within dura mater following electrical trigeminal ganglion stimulation [49]. It was concluded that the blocking effect of sumatriptan and ergot alkaloids on neurogenic plasma extravasation is not related to their vasoconstrictor action, nor to a post-synaptic cffcct, but rather to a pre-synaptic action, presumably by inhibition of the release of vasoactive neuropcptidcs mediated by activation of 5-HTlbild autoreceptors on sensory fibers [44,45]. Human evidence in support of a significant role of the trigeminovascular system and CGRP in migraine can be derived from the elegant studies of Goadsby et al., showing that (i) thermocoagulation of the trigeminal ganglion causes a marked elevation of plasma levels of substance P and CGRP in the ipsilateral external jugular vein [52], (ii) during migraine attacks there is a selective increase in plasma levels of CGRP [53], and (iii) sumatriptan normalizes thcsc elevated CGRP levels [54].
S76 Interactions with neurotransmitter
receptor sites
Several lines of experiments suggest that in human sumatriptan acts via a 5-HTi,, receptor-mediated mechanism. First, sumatriptan extremely selectively binds to 5-HT,, receptors and activates functional biochemical models of 5-HT,, receptors [SS-581. Second, the rank order of effective plasma extravasation-blocking doses of sumatriptan in Moskowitz’s rat model was most consistent with a 5-HT,, receptor mediated response (vide supra [44]). Finally, sumatriptan is essentially equipotent to other clinically effective abortive anti-migraine agents with respect to 5-HT,, agonism [55-581. The 5-HT,, receptor is the most common type of 5-HT receptor subtype in human brain [59-61] and functions as an auto-receptor, controlling the release of 5-HT [21] and other neurotransmitters such as NE and ACh [62,63]. The 5-HTid receptor is also believed to closely resemble the vascular 5-HT,-like receptor [27,30,32,647]. It thus seems that sumatriptan is a highly selective agonist at a subpopulation of 5-HT receptors [55-581, designated “5-HT,-like” in the vasculature [68,69] and 5-HTld in nervous tissue [27,30,32,64,65]. Conclusion Sumatriptan is a novel, most effective treatment of migraine and cluster headache attacks. Acute side effects appear only mild. Long-term experience, however, is limited. Main disadvantage in the clinical use are the headache recurrences. These appear to be due to short plasma half-life of the drug, causing too short of a suppression of a still ongoing “headache generator”. Current clinical research focusses on how to prevent and treat these recurrences. Three pharmacological actions of sumatriptan following systemic administration appear pathophysiologically informative. First, poor penetration of the blood-brain barrier suggesting a peripheral point of action. Second, 5-HTi-like/S-HT,, receptor-mediated ability to cause vasoconstriction, primarily of large cerebral conductance arteries and dural vessels. Third, blocking effect of neurogenic dural inflammation, possibly via a pre-synaptic 5-HT,, autoreceptor-mediated inhibition of vasoactive neuropeptides within the trigeminovascular system. Which of these mechanisms is relevant to the clinical mechanism of action of sumatriptan is the focus of intensive scientific research. The results should increase the understanding of the pathophysiology of migraine and cluster headache.
References 1 Ferrari MD, Bayliss EM, Ludlow S, Pilgrim AJ. Cephalalgia 1989; 9(S.10): 348-349. 2 The subcutaneous sumatriptan international study group (Ferrari MD, Melamed E, Gawel M et al). New Engl J Med 1991; 325: 316-321. 3 Goadsby PJ, Zagami AS, Donnan GA, Symington G, Anthony M, Bladin PF, Lance JW. Lancet 1991; 338: 782-783. 4 Sumatriptan - an oral dose-defining study. Eur Neural 1991; 31: 300-30s. 5 The sumatriptan auto-injector study group. Eur Neurol 1991; 31: 323-331. 6 The oral sumatriptan international multiple-dose study group.Eur Neurol 1991; 31: 306-313. 7 Ekbom K et al. New Engl J Med 1991; 32.5: 322-326. 8 Cady RK, Wendt JK, Kirchner JR, Sargent JD, Rothrock JF, Skaggs I I. JAMA 1991; 265: 2831-2835. 9 Fowler PA, Lacey LF, Thomas M, Keene ON, Tanner RJN, Baber NS. Eur Neurol 1991; 31: 291-294. 10 The multinational oral sumatriptan and cafergot comparative study group. Eur Neurol 1991; 31: 314-322. 11 Wolff IIG. New York; Oxford University Press, 1963. 12 Lauritzen M. Trends Neurosci 1987; 10: 8-13. 13 Solomon S, Lipton RB, Harris PY. Headache 1990; 30: 52-61. 14 Fribcrg L, Olsen TS, Roland PE, Lassen NA. Brain 1987; 110~ 917-934. 15 Brown EG, Endersby CA, Smith RN, Talbot JCC. Eur Neural 1991; 31: 339-344. 16 Anthony MA, Hinterberger H, Lance JW. Arch Neural 1%7; 16: 544-554. 17 Carroll JD, Hilton BP. Headache 1974; 14: 149-156. 18 Kimball RW, Friedman AP, Vallejo E. Neurology196o; 10: 107-111. 19 Tfelt-I Hansen P, Brand J, Dano P, et al. Cephalalgia 1989; 9 (S.9): 73-77. 20 Skingle M, Birch PJ, Leighton GE, Humphrey PPA. Cephalalgia 1990; 10: 207-212. 21 Sleight AJ, Cervenka A, Peroutka SJ. Neuropharmacology 1990; 29: 551-513. 22 1Iumphrey PPA, Connor IIE, Stubbs CM, Feniuk W. In: Jes Olesen, editor. Migraine and other headaches: the vascular mechanisms. New York-Raven Press Ltd, 1991: 335-339. 23 tlumphrey PPA, Feniuk W, Perren MJ, Beresford IJM, Skingle M. Ann NY Acad Sci 1990; 600: 587-598. 24 Parsons AA. Trends Pharmacol Sci 1991; 12: 310-315. 25 Fowler PA, Thomas M, Lacey LF, Andrew P, Dallas FAA. Gephalalgia 1989; 9(S. 9): 59-62. 26 Harper AM, MacKenzie ET. J Physiol 19n, 271: 721-733. 27 Connor IIE, Feniuk W, Humphrey PPA. Br J Pharmaeol1989; 96: 379-387. 28 Parsons AA, Whalley ET. Int J Microcirc Clin Exp 1990; 9: 227. 29 Parsons AA, Whalley ET. Eur J Pharmacol1989; 174: 189-l%. 30 Parsons AA, Whalley ET, Feniuk W, Connor HE, Humphrey PPA. Br J Pharmacol 1989; 96: 434-449. 31 IIumphrey PPA, Feniuk W, Motevalian M, Parsons AA, Whalley ET. In: Fozard JR, Saxena PR, eds. Serotonin: molecular biology, receptors and functional effects. Base]: BirkhZuser Verlag, 1991: 2&t-274. 32 Fcniuk W, Humphrey PPA, Perren MJ. Br J Pharmacol 1989; 96: 83-90. 1989; 9(S 9): 33 Perren MJ, Feniuk W, Humphrey PPA. Cephalalgia 41116. 34 Boer MO den, VillaIon CM, Heiligers JPC, Humphrey PPA, Saxena PR. Br J Pharmacoll991; 102: 323-330. 35 Fribcrg L, Olesen J, Iversen HK, Sperling B. Lancet 1991; 338: 13-17.
s77 36 Caekebeke JFV, Zwetsloot CP, Jansen J, Saxena PRS, Ferrari MD. In: Olesen .I (ed), Frontiers in Hedacahe Research. Volume 1: Migraine and other headaches: the vascular mechanisms. Raven Press: New York, 1991: 331-334. 37 Caekebeke JFV, Ferrari MD, Zwetsloot CP, Jansen J, Saxena PRS. Neurology 1992, in press. 38 Diener HC, Peters C, Rudzio M, et al. J Neurol1991; 238: 245-250. 39 Ferrari MD, Haan J, Blokland JAK, Minnee P, Zwinderman KH, Saxena PR In: Jes Olesen, editor. Migraine and other headaches: the vascular mechanism. NewYork: Raven Press Ltd, 1991: 245-248. 40 Saxena PR and Ferrari MD. Trends Pharmacol Sci 1989; 10: 200204. 41 Boer MO den, Helligers JPC, Ferrari MD, Saxena PR. Cephalalgia 1991; ll(S 11): 203-204. 42 Moskowitz MA, Henrikson BM, Markowitz S, Saito K. In: Olesen J and Edvinsson L, eds. Basic Mechanism of Headache. Amsterdam: Elsevier Science Publ, 1988: 429-437. 43 Moskowitz MA. Headache 1988; 28: 584-586. 44 Buzzi MG, Moskowitz MA, Peroutka SJ, Buyn B. Br J Pharmacol 1991; 103: 1421-1428. 45 Buzzi MG, Moskowitz MA. Cephalalgia 1991; 11: 165-168. 46 McCulloch J, Uddman R, Kingman TA, Edvinsson L. Proc Natl Acad Sci USA 1986; 83: 5731-5735. 47 Markowitz S, Saito K, Moskowitz MA. J Neurosci 1987; 7: 41294136. 48 Drummond PD, Lance JW. Ann Neurol1983; 13: 32-37. 49 Buzzi MG, Carter WB, Shimizu T, Heath HIII, Moskowitz MA. Neuropharmacology 1991; 30: 1193-1200.
50 Saito K, Markowitz 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66
67 68 69
S, Moskowitz
MA. Ann Neurol1988;
24: 732-
737. Buzzi MG, Moskowitz MA. Br J Pharmacoll990; 99: 202-206. Goadsby PJ, Edvinsson L, Ekman R. Ann Neurol1988; 23: 193-196. Goadsby PJ, Edvinsson L, Ekman R. Ann Neurol1990,28: 183-187. Goadsby PJ, Edvinsson L. Cephalalgia 1991; ll(S 11): 3-4. Deliganis AV, Peroutka ST. Headache 1991; 31: 22&231. Peroutka ST, McCarthy BG. Eur J Pharmacol1989; 163: 133-136. Waeber C, Iloyer D, Palacios JM. Synapse 1989; 4: 168-170. Schoeffter P, IIoyer D. Arch Pharmacol 1989; 340: 135-138. Waeber C, Schoeffter P, Palacios JM, Hoyer D. Arch Pharmacol 1988; 337: 595-601. Wacber C, Diet1 MM, IIoyer D, Probst A, Palacios JM. Neurosci Lett 1988; 88: 11-16. Peroutka SJ, Switzer JA, Hamik A. Synapse 1989; 3: 61-66. Schlicker E, Fink K, Gothert M, et al. Arch Pharmacol 1989; 340: 45-51. Hoyer D, Middlemiss DN. TIPS 1989; 10: 130-132. Humphrey PPA, Feniuk W, Perren MJ, et al. Br J Pharmacol1988; 94: 1123-1132. Schoeffter P, Iloyer D. J Pharmacol Exp Ther 1990; 252: 387-396. Ilamel E, Bouchard D. Saxena PR, Eds. Serotonin: molecular biology, receptors and functional effects. Basel: Birkhauser Verlag, 1991: 264-274. Ilamel E, Bouchard D. Br J Pharmacoll991; 102: 227-233. Ilumphrey PPA, Feniuk W, Perren MJ, Connor HE, Oxford AW. Cephalalgia 1989; 9(S 9): 23-33. Sumner MJ, IIumphrey PPA. Br J Pharmacol 1990; 99: 219-220.
