Triptans, 5-HT1B/1D Receptor Agonists in the Acute Treatment of Migraines

Triptans, 5-HT_1B/1D Receptor Agonists in the Acute Treatment of Migraines

Pramod R. Saxena

Peer Tfelt-Hansen

The triptans belong to a class of compounds known as 5-hydroxytryptamine_1B/1D (5-HT_1B/1D, previously 5-HT-like/5-HT_1D [260]) receptor agonists. Undoubtedly, these drugs have significantly advanced the acute treatment of migraine headaches (see Dechant et al. [64], Ferrari et al. [96], Humphrey et al. [148], Plosker and McTavish [238], Saxena and Tfelt-Hansen [262], Tfelt-Hansen [294], Tfelt-Hansen et al. [296], and Wilkinson et al. [322]). The idea that compounds mimicking 5-HT at craniovascular receptors should abort migraine attacks stems from the following observations (261):

1. Urinary excretion of 5-hydroxyindole acetic acid increases, whereas platelet 5-HT decreases during migraine attacks,

2. Migraine-like symptoms can be precipitated by reserpine and alleviated by 5-HT, which causes carotid vasoconstriction via the 5-HT_1B receptor, and

3. Ergotamine and methysergide elicit a selective carotid vasoconstriction (at least partly via the 5-HT_1B receptor), which is confined to cephalic arteriovenous anastomoses that seem to be involved in migraine pathophysiology (259).

Based on this reasoning, tryptamine derivatives were synthesized to achieve selectivity at the craniovascular 5-HT_1B/1D receptors, and this culminated in the design and development of sumatriptan (149). Despite its great utility in migraine treatment, sumatriptan has certain limitations; namely, low oral bioavailability, high headache recurrence possibly owing to a short half-life, and contraindication in patients with coronary artery disease. Therefore, a number of pharmaceutical companies decided to develop newer triptans having agonist activity at 5-HT_1B/1D receptors. Together with sumatriptan, six other such compounds (zolmitriptan, rizatriptan, naratriptan, eletriptan, almotriptan, and frovatriptan) are now available for clinical use; the development of donitriptan (156, 157, 158) seems have been stopped (for chemical structures, see Fig. 51-1). Although avitriptan (253), BMS181885 (330), and the nontriptan alniditan (128), were found effective in the treatment of migraine, these compounds are no longer in clinical development.

In this chapter we review the pharmacology of triptans and rationale for their use in migraine, the randomized clinical trials with triptans demonstrating their efficacy and evaluating the optimum dose, randomized clinical trials comparing triptans, randomized clinical trials comparing sumatriptan with other treatments, long-term studies with triptans, tolerability and safety problems with triptans, and finally the therapeutic use of triptans.

PHARMACOLOGY OF TRIPTANS

Receptor-Binding Profile

All triptans display high affinities at 5-HT₁ receptor subtypes (13,156,232,262,296) (Table 51-1). Among triptans, donitriptan appears to have the highest affinity at both 5-HT_1B and 5-HT_1D receptors and the highest efficacy at the 5-HT_1D receptor. Some triptans also interact with 5-HT_1A and 5-HT_1F receptors, but rizatriptan appears to be more selective for 5-HT_1B/1D receptors. It must be remarked, however, that the nontriptan compound, alniditan, which also proved efficacious in migraine (128), showed little, if any, affinity at the 5-HT_1F receptor (173). Sumatriptan, zolmitriptan, eletriptan, and frovatriptan display a micromolar affinity at the 5-HT₇ receptor, which mediates smooth muscle relaxation (85,260).

Cardiovascular Effects

Systemic Hemodynamics

As described (262,296), human volunteer studies show that the triptans (75,186,265,312) slightly increase
arterial blood pressure. This hypertensive response, which has a limited clinical relevance, is related to peripheral vasoconstriction. Interestingly, in anaesthetized animals, high intravenous doses of triptans (59,225,323) can decrease blood pressure, probably because of a reduction of sympathetic outflow via an agonist action at central 5-HT_1D and/or 5-HT_1A receptors (83,148,225). Stimulation of 5-HT_1B/1D receptor can also result in vasodilatation mediated by nitric oxide release (139).

FIGURE 51-1. Chemical structures of sumatriptan and second-generation triptans.

TABLE 51-1 Affinity and Efficacy of Triptans at Human 5-HT₁ Receptors^a

	*Inhibition of Radioactive Ligand Binding (pK_i)*^b					*Inhibition of cAMP Accumulation (pEC₅₀)*^c		*Enhancement of [³⁵S]GTPγS Binding*
	*Inhibition of Radioactive Ligand Binding (pK_i)*^b					*Inhibition of cAMP Accumulation (pEC₅₀)*^c		pEC₅₀		E_max (% of 5-HT)^d
	5-HT_1A	5-HT_1B	5-HT_1D	5-HT_1E	5-HT_1F	5-HT_1B	5-HT_1D	5-HT_1B	5-HT_1D	5-HT_1B	5-HT_1D
Almotriptan^e	6.3	8.0	8.0			8.8	8.5
Donitriptan^f	7.6	9.4-10.1	9.3-10.2	5.9	5.5	8.9	9.6	8.7	9.1	94	97
Eletriptan	7.4	8.0	8.9	7.3	8.0
Frovatriptan	7.3	8.6	8.4	<6.0	7.0
Naratriptan^g	7.1-7.6	8.1-8.7	8.3-8.4	7.7	8.2	7.8	9.1	7.6	8.4	89	69
Rizatriptan^g	6.4	6.9-8.1	7.9-8.6	6.8	6.8	6.9	8.5	6.6	7.8	94	82
Sumatriptan^g	6.4-6.9	7.8	8.5	5.8	7.9	7.2	8.8	6.6	7.7	97	84
Zolmitriptan^g	6.5	8.3	9.2	7.7	7.2-7.5	7.5	9.6	7.2	8.8	99	85
^aFor references, see Tfelt-Hansen et al. (296). ^bNegative logarithm of the dissociation equilibrium constant. ^cNegative logarithm of the molar concentration that elicits 50% of its maximum effect. ^dMaximum effect compared to 5-HT = 100. ^eData from Bou et al. (14). ^fData from John et al. (156). ^gData from Pauwels et al. (232).

Carotid Hemodynamics and Arteriovenous Anastomotic Blood Flow

Sumatriptan increases internal carotid and middle cerebral artery blood flow velocity in human volunteers (33). This effect is probably caused by constriction of intracranial arteries, consistent with findings in animal models (262,296). Although this is yet to be established, based on similar pharmacologic properties, other triptans are likely to have comparable effects.

Sumatriptan as well as other triptans decrease carotid blood flow in anaesthetized animals (262,296). The apparent rank order of agonist potency (based on intravenous dose needed to decrease canine carotid blood flow by 50%) was: frovatriptan (0.4 μg/kg) (226), greater than zolmitriptan (2.3 μg/kg) (187), greater than eletriptan (12 μg/kg) (137), equal to naratriptan (19 μg/kg) (42), greater than or equal to rizatriptan (30 μg/kg) (270), and equal to sumatriptan (39 μg/kg) (42). Almotriptan also reduces carotid blood flow in the cat (13). Using intracarotid-administered radiolabel microspheres, it also has been shown that the carotid vasoconstriction by triptans (59,63,187,303,323) is confined to arteriovenous anastomoses, which may dilate during migraine headaches (144,259). Similarly, sumatriptan (infused into the brachial artery) can decrease human forearm blood flow by a selective vasoconstrictor action on arteriovenous anastomoses (310). Extracerebral blood flow, in contrast, increases in response to 5-HT_1B/1D receptor agonists (59,63,323), although a decrease has been reported with zolmitriptan (187). Interestingly, cerebral blood flow does not seem to be affected by triptans, as shown not only with sumatriptan (60,63), but also with the lipophilic, brain-penetrating compounds, zolmitriptan (187), eletriptan (323) and rizatriptan (278).

Using SB224289 (selective 5-HT_1B receptor antagonist [138,268]) and BRL15572 (selective 5-HT_1D receptor antagonist [138,240]), it has been shown that sumatriptan constricts porcine carotid arteriovenous anastomoses as well as canine (60,63) external carotid vasculature via 5-HT_1B receptors and not via 5-HT_1D receptors. Similar results have been obtained with naratriptan and alniditan (a nontriptan 5-HT_1B/1D receptor agonist) in anaesthetized cats (114). Considering their similar receptor-binding profile, it is likely that other triptans also exert vascular effects via the 5-HT_1B receptor. The latter is consistent with the expression of 5-HT_1B (high) and 5-HT_1D (low) receptor messenger RNA (mRNA) (15,211,305,307,313) or the corresponding protein (181,211) in blood vessels. Notwithstanding, some lines of evidence suggest that a novel 5-HT receptor, possibly identical to that reported by Castro et al. (35), mediates 5-HT-induced constriction of porcine carotid arteriovenous anastomoses (61). Similarly, ergotamine and dihydroergotamine (DHE) constrict the carotid vascular bed considerably via non-5-HT_1B/1D receptors (314). Significantly, it was recently shown that the ergotinduced carotid vasoconstriction in the anesthetized dog is abolished by a combination of 5-HT_1B/1D receptor and α₂-adrenoceptor antagonists (314). Possibly, these novel 5-HT receptors and α₂-adrenoceptors (324) may be targeted for future antimigraine drugs because constriction of the carotid vasculature and carotid arteriovenous anastomoses is highly predictive for antimigraine potential (259).

Isolated Blood Vessel Contraction and Craniocoronary Selectivity

It is now well known that triptans constrict isolated blood vessels, including the human cranial and coronary arteries; the effect on cranial vessels where, contrary to peripheral arteries, the 5-HT_1B rather than 5-HT₂ receptor dominates, is more marked (262,296). For example, eletriptan (Fig. 51-2) (184,308), sumatriptan (184,308), rizatriptan (180), frovatriptan (41), and almotriptan (14), as well as donitriptan (309), have been demonstrated to be several times more potent in contracting the human middle meningeal artery than the coronary artery. More important, the magnitude of contraction elicited by triptans in cranial vessels is much more than that in the coronary arteries. Therefore, at therapeutic plasma concentrations, triptans contract cranial vessels much more than the coronary artery
(see Fig. 51-2) (14,41,184,308,309). The reason for this is not clear, but it may be related to the higher density of 5-HT_1B receptors in the cranial compared to coronary arteries (180).

FIGURE 51-2. Concentration-response curves of eletriptan on human middle meningeal and coronary arteries. Superimposed on these curves are the free C_max (protein-unbound fraction of the maximum plasma concentration) of eletriptan observed in human subjects after oral ingestion of a single 40- or 80-mg tablet. It may be noted that the drug elicits a substantial contraction of the middle meningeal artery, whereas there is only a minimal effect on the coronary artery (184).

The magnitude of coronary artery contraction by triptans at therapeutic maximum plasma concentrations (C_max), corrected for the fraction of the drug bound to plasma proteins, has been calculated by interpolating concentration-response curves (14,41,184,308,309). As shown in Figure 51-3, the contraction induced by zolmitriptan, eletriptan, and possibly rizatriptan, almotriptan, and frovatriptan may be somewhat smaller than that with sumatriptan. However, both zolmitriptan and rizatriptan have a pharmacologically active N-desmethyl metabolite, which may cause additional contraction. Similarly, a pharmacologically active N-desmethyl metabolite has been described for eletriptan. Although this metabolite is similar to the parent compound with respect to the affinity at 5-HT_1B and 5-HT_1D receptors, its plasma concentration remains 7- to 9-fold lower than that of eletriptan. Therefore, N-desmethyl eletriptan is unlikely to add to the coronary artery contraction induced by eletriptan (182, 184).

In conclusion, human coronary artery contraction elicited by some triptans may be somewhat less than that elicited by sumatriptan, but, clearly, all triptans are cranioselective. These subtle differences may be partly related to the use of relatively low therapeutic doses of newer triptans compared to 100 mg of sumatriptan. Moreover, we cannot be sure if these differences are clinically relevant, but at therapeutic concentrations the coronary artery contraction to all triptans is substantially less than the contraction of middle meningeal artery (see MaassenVan-DenBrink et al. [184]). Therefore, triptans are safe medications, but, despite some differences in the craniocoronary selectivity, this class of drugs remains contraindicated in patients with coronary artery disease. Recently, an expert panel has issued a consensus statement with regard to the cardiovascular safety of triptans (77), where they say that (i) most of the data on triptans are derived from patients without known coronary artery disease, (ii) chest symptoms occurring during use of triptans are generally not serious and are not explained by ischemia, (iii) the incidence of serious cardiovascular events with triptans in both clinical trials and clinical practice appears to be extremely low, and (iv) the cardiovascular risk-benefit profile of triptans favors their use in the absence of contraindications.

FIGURE 51-3. Predicted contraction of the human isolated coronary artery at therapeutic free C_max, (protein-unbound fraction of the maximum therapeutic plasma concentration), expressed as percentage of contraction elicited after 100 mg oral sumatriptan (100%). Contraction to sumatriptan (100 mg orally), obtained as a reference, was derived from the same study as the respective compound: subcutaneous sumatriptan, naratriptan, and zolmitriptan (182), rizatriptan (98,178,182), eletriptan (184), almotriptan (14), and frovatriptan (227). Because in the case of rizatriptan studies from two different laboratories were used to predict the coronary artery contraction, values from both laboratories (closed bar) (98,178) and (open bar) (182) are presented. (Redrawn from MaassenVanDenBrink A, Saxena PR. Coronary vasoconstrictor potential of triptans: a review of in vitro pharmacologic data. Headache. 2004;44 [Suppl 1]:S13-S19.)

Trigeminal Inhibitory Effects

Peripheral Trigeminal Neuronal Inhibition

Triptans inhibit dural plasma protein extravasation following electrical stimulation of the trigeminal nerve (13,42,137,189,325). Because sumatriptan was ineffective in 5-HT_1B receptor knockout mice, this effect seems to involve the 5-HT_1B receptor (332). Similarly, this receptor mediates the effects of sumatriptan in the guinea pig dura mater (331,332), where the 5-HT_1F (160,199) as well as a novel (331,332) receptor also play an important role. In the rat, the inhibition of plasma protein extravasation involves non-5-HT_1B receptors, possibly the 5-HT_1D receptor, and another CP122288-sensitive receptor (271). It should be noted that the inhibition of plasma protein extravasation is not consistent with antimigraine activity because
several drugs with such an action, including the NK₁ receptor antagonist lanepitant (127), the endothelin-receptor antagonist bosentan (198) as well as CP122288 (248), have been found ineffective in migraine. Moreover, the involvement of plasma extravasation itself in migraine has been questioned, as there no changes in retinal permeability during migraine attacks (199). Thus, it is now accepted that the inhibition of dural plasma protein extravasation is not related to the antimigraine action (296). On the other hand, the inhibition of the release of calcitonin gene-related peptide (CGRP) seems to be much more relevant for the antimigraine action, particularly as the novel CGRP receptor antagonist, BIBN4096 (78,161), has been found to be effective in acute migraine (215). Apart from sumatriptan (115) and zolmitriptan (116), donitriptan (174) has been reported to inhibit the release of CGRP during migraine or following electric stimulation of blood vessels or the trigeminal ganglion. However, CGRP release following vanilloid receptor stimulation by capsaicin is not affected by sumatriptan (4).

Eletriptan can prevent the expression of CGRP-immunoreactive neurons in the rat dura mater following electrical stimulation of trigeminal ganglion in a way similar to sumatriptan (169). However, unlike sumatriptan, which does not cross the blood-brain barrier, eletriptan also blocked the stimulation-induced effects in the trigeminal nucleus caudalis as well as in the upper cervical spinal cord (169).

Central Trigeminal Neuronal Inhibition

It has been shown that intravenous administration of zolmitriptan and naratriptan (10,119,120) as well as local iontophoretic application of eletriptan and naratriptan (10,146) inhibit trigeminal nucleus caudalis potentials generated after superior sagittal sinus stimulation in cats. Similarly, in rats, intravenous rizatriptan (46) as well as naratriptan microinjected into the ventrolateral periaqueductal grey inhibited such potentials evoked by dural stimulation, but not to facial cutaneous or corneal stimulation (7). In addition to an inhibitory effect on trigeminal nucleus, the latter results suggest that triptans may also activate descending pain-modulating pathways in the ventrolateral periaqueductal grey. Thus, these drugs exhibit a central inhibitory effect within the trigeminal system, which may contribute in part to their therapeutic effect in migraine. However, because of its poor central penetration, intravenous sumatriptan did not affect c-fos mRNA expression in the trigeminal nucleus caudalis following trigeminal ganglion stimulation in rats (272). This raises the question of whether central trigeminal inhibition is essential for therapeutic activity in migraine. On the other hand, it remains to be clarified whether during migraine headaches the blood-brain barrier is partly disrupted. Indeed, after disruption of the blood-brain barrier by infusion of hyperosmolar mannitol, sumatriptan did inhibit c-fos mRNA expression (272).

The central trigeminal inhibitory effects of eletriptan and naratriptan in the cat, being susceptible to blockade by WAY-100635 (10) and GR127935 (10,120,146), are mediated by 5-HT_1A as well as 5-HT_1B and/or 5-HT_1D receptors. Goadsby and Classey (114) reported that the central inhibitory effects of alniditan (5-HT_1B/1D receptor agonist) and naratriptan (5-HT_1B/1D/1F receptor agonist) were attenuated by both SB224289 (5-HT_1B receptor antagonist) and BRL-15572 (5-HT_1D receptor antagonist) and, although the combination of these two antagonists completely blocked the effects of alniditan, there still remained an inhibitory effect of naratriptan after both 5-HT_1B and 5-HT_1D receptor blockade. This remaining effect of naratriptan may be mediated by 5-HT_1F receptor, because LY344864 (5-HT_1F receptor agonist) also inhibited trigeminal activity evoked by superior sagittal sinus. Thus, it appears that the central trigeminovascular inhibitory activity is mediated by 5-HT_1A, 5-HT_1B, 5-HT_1D as well as 5-HT_1F receptors in the cat. However, in the rat 5-HT_1D receptors, but not 5-HT_1B receptors, seem to mediate the central trigeminal antinociceptive action by zolmitriptan (47).

Central Pain Sensitization

Burstein et al., who have reported that application of an “inflammatory soup” containing histamine, 5-HT, bradykinin, and prostaglandin E2 on rat dura activated trigeminal neurons and enhanced their sensitivity to mechanical stimuli, proposed that this chemosensitivity and sensitization is characteristic of some types of headaches and may contribute to the throbbing pain of migraine (284). Early sumatriptan intervention (together with application of inflammatory soup) effectively blocked the development of all aspects of central sensitization (expansion of dural receptive fields, reduction of neuronal response threshold, spontaneous firing rate, increased neuronal response magnitude to skin brushing, and reduced response threshold to skin heating), but late sumatriptan intervention (2 hours after inflammatory soup application) only counteracted the first two aspects of central sensitization (22). When both peripheral (trigeminal ganglion) and central (medullary dorsal horn) trigeminovascular neuronal potentials were simultaneously recorded in rats, it was found that sumatriptan prevented the induction of sensitization as well as normalized the heightened intracranial mechanical sensitivity following sensitization in central but not in peripheral neurons; however, the drug failed to attenuate the increased spontaneous activity established during sensitization in both neurons (172). The authors conclude that sumatriptan inhibits neither peripheral nor central trigeminovascular neurons directly; it exerts its action via presynaptic 5-HT_1B/1D receptors in the
dorsal horn to block synaptic transmission between axon terminals of the peripheral trigeminovascular neurons and cell bodies of their central counterparts, thus suggesting that the analgesic action of triptans manifests in the absence, but not in the presence, of central sensitization (172).

In concordance with their studies in rats, Burstein et al. reported that over 75% of migraine patients develop cutaneous allodynia during migraine within the referred pain areas, initially on the ipsilateral side of the head and later on the contralateral side and on ipsilateral forearm, and they hypothesized that the cutaneous allodynia can be used to predict the effectiveness of triptans (21,23). Indeed, only 15% of patients with compared to 93% without cutaneous allodynia were rendered pain free by sumatriptan within 2 hours of treatment; similar responses were observed in the two groups whether sumatriptan was administered early or late in migraine attack (20). Interestingly, sumatriptan effectively terminated the throbbing of migraine pain (peripheral sensitization) in the vast majority of patients, even when pain relief was incomplete or allodynia was not suppressed (20).

Gastrointestinal Effects

Sumatriptan relieves not only headache but also the nausea during a migraine attack, but it can occasionally cause nausea as a side effect.

In normal subjects, sumatriptan has been reported to delay gastric emptying (44) and retard the natural postprandial decay in rate of transient lower esophageal relaxations, possibly as a consequence of a prolonged fundus relaxation and retention of meal in the proximal stomach (273). Real-time ultrasonography and computed tomography demonstrated that sumatriptan reduced the proximal and distal transverse area, but increased the sagittal axis of the proximal stomach (188). These gastric effects have the potential to further enhance nausea and epigastric symptoms already occurring in migraineurs (188). The impact of gastric effects of sumatriptan remains difficult to predict (40), because in patients with functional dyspepsia, sumatriptan was reported to improve postprandial gastric accommodation and relieve epigastric symptoms (188). The mechanisms involved in the gastrointestinal effects of sumatriptan are not clear and we do not know if other triptans have similar gastrointestinal effects.

PHARMACOKINETICS

The pharmacokinetic characteristics of triptans in human volunteers and migraine patients are presented in Table 51-2; for references, see (Buchan et al. [19], Dahlöf et al. [56], Ferrari et al. [93], Jansat et al. [154], Millson et al. [204], Saxena et al. [262], and Tfelt-Hansen et al. [296]).

Sumatriptan is quickly absorbed subcutaneously with a t_max of approximately 10 minutes and an average bioavailability of 96%. After oral administration of therapeutic doses (100 mg) of sumatriptan, however, the t_max is longer (1.5 hours) and, more importantly, the bioavailability is low (about 14%). This, at least, theoretically suggests a relatively high intersubject variability. Intranasal or rectal administration of sumatriptan does not seem to improve these parameters much. Low lipid solubility also makes the drug difficult to permeate the blood-brain barrier, but, apparently, this property does not seem to be important for its antimigraine action. Sumatriptan has a short plasma t_1/2 as the drug is subject to extensive hepatic first pass clearance, being metabolized by monoamine oxidase (MAO)-A.

The oral bioavailability of newer triptans, especially naratriptan and almotriptan, is much improved. The latter can be partly attributed to the more lipophilic nature of these drugs. Interestingly, compared to sumatriptan the t_max of oral frovatriptan is worse and those of zolmitriptan, naratriptan, eletriptan, and almotriptan not much different, but rizatriptan seems to reach its peak plasma levels quicker. It may be noted that the unbound C_max values (C_max corrected for plasma protein binding) of newer triptans are lower than that of sumatriptan. This is apparently caused by two main factors: lower therapeutic concentrations are needed as these drugs have a higher affinity at 5-HT_1B/1D receptors (see Table 51-1) and they have been better titrated, thus reducing therapeutic penalty.

With the exception of rizatriptan, the newer triptans are degraded slower than sumatriptan, especially frovatriptan, which has a plasma half-life of more than 24 hours (see Table 51-2). Interestingly, elimination half-life (r = −1.0; P = .0016) and 5-HT_1B receptor potency (r = −0.68, P = .034), but not headache response (r = 0.18; P = .53), therapeutic gain (r = −0.11; P = 0.71), or 5-HT_1D receptor potency (r = −0.20, P = .54) of triptans significantly correlated with headache recurrence (110), suggesting that longer half-life and greater 5-HT_1B receptor potency may be responsible for lower headache recurrence. The view that a long half-life is responsible for low recurrence rate is not universally shared (315).

In contrast to sumatriptan and naratriptan, active metabolites have been reported for zolmitriptan, rizatriptan, and eletriptan (see Table 51-2). It is not known whether and, if so, to what extent, the metabolites contribute to therapeutic efficacy. Interestingly, eletriptan is a substrate for P-glycoprotein blood-brain barrier efflux system (87).

POSSIBLE MECHANISMS OF ACTION OF TRIPTANS IN MIGRAINE

Migraine is a syndrome with characteristic features composed of aura symptoms followed by headache associated with nausea, vomiting, photophobia, and photophobia
(140). Based on the pharmacology of triptans (see above), triptans can abort migraine attack by four different mechanisms (Fig. 51-4): (i) a direct contraction of dilated cranial extracerebral blood vessels (62,296), (ii) suppression of neuropeptide (mainly CGRP) release from peripheral nerve endings around blood vessels (118,121), (iii) inhibition of impulse transmission centrally in the trigeminal nucleus caudalis (118,121), and (iv) presynaptic blockade of synaptic transmission between axon terminals of the peripheral trigeminovascular neurons and cell bodies of their central counterparts (172). Although the case for the involvement of neurogenic mechanisms is compelling, in our view, however, the main action of triptans in migraine is the constriction of dilated cranial extracerebral blood vessels via 5-HT_1B receptors (97,150,151,244,262). First, the possible contribution of the neuronal effects of triptans mediated via 5-HT_1D receptor, which are abundantly present in the trigeminal system (239), has been put in doubt because PNU142633, a selective 5-HT_1D receptor agonist, was found ineffective in the treatment of migraine (132). The intrinsic activity of PNU-142633 (70% of 5-HT) at the human 5-HT_1D receptor is only a little less than that of sumatriptan (84% of 5-HT), but it is equally effective as sumatriptan and a half-log more potent than sumatriptan in preventing plasma protein extravasation and can reduce increases in cat nucleus trigeminal caudalis blood flow elicited by electrical stimulation of the trigeminal ganglion; importantly, in contrast to sumatriptan it does not reduce carotid blood flow (200). Moreover, besides PNU-142633, there are many other selective 5-HT_1D receptor agonists lacking vasoconstrictor activity (179,202), but none has been described as effective in migraine. Another reason that speaks against the involvement of a central action is that sumatriptan (despite a report to the contrary where an extremely high intravenous dose of 3.2 mg/kg was used [159]), does not easily cross the blood-brain barrier (152,162), is comparable in therapeutic effectiveness to centrally penetrative triptans (e.g., rizatriptan and eletriptan); see below. Accordingly, Pascual and Muñoz (229) found no correlation between lipophilicity coefficients and therapeutic efficacy; interestingly there was a significant correlation between lipophilicity and central nervous system (CNS) adverse events. Thus, we submit that the central
mechanisms of triptans may not be necessary for therapeutic efficacy.

TABLE 51-2 Pharmacokinetic Parameters of Triptans in Healthy Volunteers and in Patients with Migraine^a

*Triptan*	*Dose (mg) and Route of Administration*	t_max *(h)*	C_max *(ng Ml⁻¹)***	*Bioavailability (%)*	*t_1/2 (h)*	*AUC (ng h mL⁻¹)*	*CL_R (mL min⁻¹)*	*Main Metabolic/Excretion Route*	*Active Metabolites*	*Plasma Protein Binding (%)*	*Log D_pH7.4*
Almotriptan	12.5 p.o.	2.1/2.6^b	34/31^b	80	3.1	217/216^b	416	MAO-A, Cyp3A4	No	40
	25 p.o.	2.7	64	69	3.6	443		58% renal
Eletriptan	40 p.o.	1.8/2.8^c	82	50	5		597	Cyp3A4	Yes	85	+0.5
	80 p.o.	1.4	246	50	6.3	1661
Frovatriptan	2.5 p.o.	2.3/3.0^c	4.2/7.0^c	21/30^c	27/26^c	43/94^c		Cyp1A2	?	15^d
	40 p.o.	3.0/5.0^c	24.7/53.4^c	10/18^c	25/30^c	300/881^c		50% renal
	0.8 i.v.	0.5/0.5^c	18.6/24.4^c		24/24^c	63/104^c	216/132^c
Naratriptan	2.5 p.o.	2	12.6	74	5.5	98	220	P450, 70% renal	No	20	−0.2
Rizatriptan	10 p.o.	1.0/1.0^c	19.8	40	2.0	50	414	MAO-A, 30% renal	Yes	14	−0.7
Sumatriptan	6 s.c.	0.17	72	96	2	90	220	MAO-A, 60% renal	No	14-21	−1.5
	100 p.o.	1.5	54	14	2	158	260	MAO-A, 60% renal
	20 i.n.	1.5	13	15.8	1.8	48	210
	25 rectal	1.5	27	19.2	1.8	78	200
Zolmitriptan	2.5 p.o.	1.5	3.3/3.8^c	39	2.3/2.6^c	18/21^c	193	P450, MAO	Yes	25	-1.0
	5 p.o.	1.5	10	46	3.0	42	193
^aFor references, see Buchan et al. (19), Dahlöf et al. (56), Ferrari et al. (93), Jansat et al. (154), Millson et al. (204) and Tfelt-Hansen et al. (296) ^bValues outside and during migraine attacks, respectively. ^cValue for men and women, respectively. ^dBesides plasma protein binding, about 60% of frovatriptan is bound to red blood cells.
Abbreviations: AUC, area under curve; MAO-A, monoamine oxidase-A; CL_R, renal clearance; LogD_pH7.4, measure of lipophilicity with increasing numbers indicating greater lipid solubility; s.c., subcutaneous; p.o., oral; i.n., intranasal; i.v., intravenous; P450, Cyp3A4 and Cyp1A2 are hepatic drug metabolizing enzymes.

FIGURE 51-4. The pathophysiologic changes in migraine putatively stem from ion leakage through a genetically predisposed faulty channel (217) in the brainstem (321), leading to a decrease in cerebral blood flow, possibly owing to cortical spreading depression (11) and, subsequently, neuropeptide release and dilatation of cranial extracerebral blood vessels (61). The increased pulsation in these blood vessels stimulates the trigeminovascular system, setting in peripheral and central sensitization (172) and leading to headache and associated symptoms (nausea, vomiting, photophobia, and/or phonophobia). The site of action of triptans in aborting migraine attack include: (i) a direct contraction of dilated cranial extracerebral blood vessels, (ii) suppression of neuropeptide (mainly CGRP) release from peripheral nerve endings around blood vessels; (iii) inhibition of impulse transmission centrally in the trigeminal nucleus caudalis; and (iv) presynaptic blockade of synaptic transmission between axon terminals of the peripheral trigeminovascular neurons and cell bodies of their central counterparts.

It is sometimes suggested (87) that because of the efflux of eletriptan by p-glycoprotein from the brain, one needs a relatively “high” therapeutic dose negating the dose “advantage” by virtue of its high potency at the 5-HT_1B receptor (see Table 51-1). However, it must be pointed out that about 85% of eletriptan is bound to plasma proteins (see Table 51-2), thus reducing its pharmacologically effective (protein-unbound) fraction. The protein-bound fraction of eletriptan may serve as a “reservoir” to prolong its pharmacologic action.

RANDOMIZED CLINICAL TRIALS WITH TRIPTANS

First, it should be demonstrated in randomized, double-blind, placebo-controlled clinical trials that a drug is more effective than placebo. Then the dose-response curve should be established and, taking both efficacy and tolerability into account, the optimum and minimum effective doses should be determined. Subsequently, the optimum dose the drug should be compared with currently established treatment for efficacy and tolerability.

Other questions that have been addressed in randomized clinical trials, mainly with sumatriptan, the first triptan, are listed below:

1. Does a second dose increase efficacy?

2. Can a second dose prevent recurrence of headache?

3. Is a second dose effective in the treatment of recurrences?

4. What is the onset of action compared with placebo/control?

5. How effective are triptans given when the migraine attack is still mild?

6. Can a triptan, being a vasoconstrictor, be given safely during the aura phase?

7. Can a triptan given in the aura phase prevent the headache?

8. Is the efficacy sustained in multiple attacks?

In current randomized clinical trials, headache relief with triptans is defined as a decrease from an initial moderate or severe headache to none or mild (236) after a certain time (1, 2, or 4 hours). In this section, response rates at 1 hour after injection and at 2 hours after other routes of administration are considered the primary responses for active drugs and placebo. These response rates vary considerably in different trials, for example, from 56 to 88% after subcutaneous sumatriptan (295), most likely because of a variable placebo response. Therefore, the results of the trials are given as the therapeutic gain (percentage response for active drug minus percentage response for placebo) with 95% confidence intervals (CIs). The mean therapeutic gains for two triptans or two doses of a triptan
have, in most cases, overlapping 95% CIs (Fig. 51-5), and it should be noted that the relative efficacy of two doses or drugs can only be established definitely in comparative trials (Table 51-3).

FIGURE 51-5. Mean and 95% confidence intervals (CIs) of absolute and placebo-subtracted (therapeutic gain) headache relief response (i.e., proportion of patients whose moderate or severe headache improves to mild or no pain) at 2 hours after different doses of triptans administered orally, except in case of sumatriptan where nasal, rectal, and subcutaneous routes were also employed (for number of patients treated with each dose of drugs and placebo, see text). The 95% CI of sumatriptan 100 mg (dotted rectangle) has been projected over other responses and significant difference from sumatriptan in the therapeutic gain is indicated by an asterisk (*). (Data collated by P. Tfelt-Hanssen.)

Randomized Clinical Trials With Sumatriptan

Subcutaneous Sumatriptan

Subcutaneous sumatriptan has a reasonably well-defined dose-response curve, with 1 mg being the minimum effective dose and 6 mg being the optimum dose with no gain by increasing to 8 mg (294).

Six milligrams subcutaneous sumatriptan has been evaluated against placebo in 12 randomized, double-blind, placebo-controlled clinical trials (16,31,32,88,135,142, 155,193,207,249,285,286), where headache relief was reported after 1 hour. As shown in Figure 51-5, 6 mg subcutaneous sumatriptan (based on 1,994 patients treated with sumatriptan [headache relief in 69%] and 1,265 patients treated with placebo [headache relief in 19%]) has a mean therapeutic gain of 51% (95% CI 48 to 53%) after 1 hour. After 2 hours, in 10 of these trials the therapeutic gain was the same (52% [95% CI 40 to 56%]) as after 1 hour (Tfelt-Hansen, personal observation), indicating that the response to subcutaneous sumatriptan should be evaluated after 1 hour. In two trials with subcutaneous sumatriptan in which headache relief was reported after 1.5 hours (2,28), the mean therapeutic gain was 45% (95% CI 34 to 56%). In 11 of these trials, the total number of adverse events reported was 930 of 1,456 after sumatriptan (64%) versus 290 of 890 after placebo (31%). There were thus 33% (95% CI 29 to 37%) more adverse events after subcutaneous sumatriptan than after placebo. Most of these adverse events were, however, mild to moderate and short term. Subcutaneous sumatriptan also has been shown in one placebo-controlled trial (31) to reduce productivity loss during a migraine attack.

Oral Sumatriptan

The minimum effective dose is 25 mg (131,235) and the optimum dose 50 to 100 mg, with no gain in efficacy, but more adverse events, when the dose is increased to 200 to 300 mg (219,294).

Oral sumatriptan 100 mg has been evaluated against placebo in 19 randomized double-blind, placebo-controlled clinical trials (8,27,37,49,54,72,121,153,208, 210,219,220,235,237,246,257,297,300,318). As shown in Figure 51-5, 100 mg oral sumatriptan (based on 3,874 patients treated with sumatriptan [headache relief in 61%] and 2,153 patients treated with placebo [headache relief in 28%]) has a mean therapeutic gain of 33% (95% CI 31 to 35%) after 2 hours. McCrory et al. (201) performed a systematic Cochrane review of oral sumatriptan 100 mg and found a number needed to treat of 5.1 (95% CI 3.9 to 7.1) for pain free after 2 hours. In 13 of these trials, the total number of adverse events reported was 755 of 1,948 after sumatriptan (39%) versus 284 of 1,236 after placebo (23%). There were thus 16% (95% CI 13 to 19%) more adverse events after 100 mg oral sumatriptan than after placebo. However, most of these adverse events were mild and short term.

The lower doses of oral sumatriptan (25 and 50 mg) have been investigated less, in five and seven clinical trials, respectively. For 50 mg sumatriptan (based on 1,599 patients treated with sumatriptan [headache relief in 59%] and 653 patients treated with placebo [headache relief in 30%]), the mean therapeutic gain was 29% (95% CI 29 to 33%) (49,131,235,237,257,258), a therapeutic gain similar to that of 100 mg sumatriptan (see Fig. 51-5), and in two direct comparative trials the efficacy of the two doses of sumatriptan were comparable (235,255). The total number of adverse events was 357 of 1,034 after 50 mg sumatriptan (35%) versus 164 of 509 after placebo (32%), with no difference between the two (49,131,235,257,258). In the two comparative trials (235,255), 50 mg sumatriptan caused fewer adverse events than 100 mg sumatriptan. The response rate of 50 mg sumatriptan was superior to that of placebo at 30 minutes (131,235,258).

TABLE 51-3 Randomized Clinical Trials Comparing Triptans

*Trial*			*Drug and Dose (mg)*	*Patients (n)*	*Success Rate^a (%)*	*Therapeutic Gain^b (%)*	*Pain Free^c (%)*	*Recurrence^d (%)*
Oral Administration
	Géraud et al. (109)		PL	55	44	—	13	25
			ZO 5	491	59	15	29	26
			SU 100	498	61	17	30	28
		Success rates ZO 5 versus SU 100 −2% (−8 to +4%)
		Pain free ZO 5 versus SU 100 −1% (−7 to +5%)
	Gallagher et al. (105)		ZO 2.5	295	67, 83^x
			ZO 5	305	65, 84^x
			SU 50	306	64, 81^x
		Success rates: Across up to six attacks ZO 2.5 >SU 50 (2 hours)
	Gruffyd-Jones et al. (135)		ZO 2.5	500	63
			ZO 5	514	66
			SU 50	508	66
		Success rates: Across up to six attacks ZO 2.5 >SU 50 (2 hours)
	Dahlöf et al. (54)		PL	91	31, 39^f			36
			NA 1	85	58, 64^f	27		31
			NA 2.5	87	53, 63^f	21		17
			NA 5	93	54, 65^f	23		32
			NA 7.5	93	68, 80^f	37		30
			NA 10	96	69, 80^f	38		29
			SU 100	98	60, 80^f	29		44
		Success rates NA 2.5 versus SU 100 −8% (−23% to +6%)
		Success rates at 4h NA 2.5 versus SU 100 −16% (−29% to −3%)
		Recurrence rates NA 2.5 versus SU 100 −27% (−42% to −7%)
	Göbel et al. (124)^g		NA 2.5	215	76^f			43^h
			SU 100	216	84^f			57
		Success rates at 4h NA 2.5 versus SU 100 −8% (−15% to 0%)
		Recurrence rates NA 2.5 versus SU 100 −17% (−27% to −7%)
	Visser et al. (319)		PL	85	18		3	36
			RI 10	89	52	34	26	41
			RI 20	82	56	38	35	53
			RI 40	120	67	49	49	42
			SU 100	72	46	28	22	43
		Success rates RI 10 versus SU 100 +6% (−10% to +21%) RI 40 versus SU 100 +21% (+7% to + 35%)
		Pain free RI 10 versus SU 100 +4% (−10% to +17%) RI 40 versus SU 100 +27% (+14% to +40%)
	Lines et al. (175)		PL	80	23		3	33
			RI 5	352	63	40	27	38
			SU 50	356	67	44	32	34
		Success rates RI 5 versus SU 50 −4%(−11% to +3%)
		Pain-free RI 5 versus SU 50 −6% (−12% to +2%)
	Goldstein et al. (131)		PL	141	38		9	32
			RI 5	294	68	30	33	33
			RI 10	305	72	34	41	35
			SU 25	297	62	24	27	32
			SU 50	291	68	30	37	31
		Success rates RI 10 versus SU 50 +4% (−3% to +11%) RI 5 versus SU 25 +6% (−2% to +11%)
		Pain-free RI 10 versus SU 50 +4% (−4% to +12%) RI 5 versus SU 25 +6% (−1% to +13%)
	Kolodny et al. (170)		PL	288	41		10
			RI 10	296	66	25	35
			SU 50	285	68	27	35
		Success rates RI 10 versus SU 50 −3% (−10% to + 5%)
		Pain-free RI 10 versus SU 50 0% (−7% to +8%)
	Tfelt-Hansen et al. (300)		PL	159	40		9	20
			RI 5	164	60	20	25	48
			RI 10	385	67	27	40	35
			SU 100	387	62	22	33	32
		Success rates RI 10 versus SU 100 +5% (−2% to +12%)
		Pain-free RI 10 versus SU 100 +7% (+1% to +12%)
	Goadsby et al. (122)		PL	142	24		5	23
			EL 20	144	50	26	17	27
			EL 40	135	56	32	29	33
			EL 80	141	65	41	31	32
			SU 100	129	49	23	19	33
		Success rates EL 20 versus SU 100 +1% (−11% to +12%) EL 40 versus SU 100 +7% (−5% to +19%)
		EL 80 versus SU 1000 +16% (+4% to +27%)
		Pain-free EL 20 versus SU 100 −2% (−13% to +7%) EL 40 versus SU 100 +6% (−4% to +16%)
		EL 80 versus SU 1000 +12% (+2% to +22%)
	Sandrini et al. (256)		PL	84	30		4	25
			EL 40	175	62	33	30	19
			EL 80	164	65	36	36	16
			SU 50	181	49	19	18	26
			SU 100	169	50	33	1	27
		Success rates EL 40 versus SU 50 +13% (+3% to +20%) El 40 versus SU 100 +10% (0% to +22%)
		EL 80 versus SU 50 +17% (+6% to +27%) EL 80 versus SU 100 +15% (+4% to +29%)
		Pain-free EL 40 versus SU 50 +11% (+3% to +20%) El 40 versus SU 100 +13% (4% to +21%)
		EL 80 versus SU 50 +18% (+8% to +27%) EL 80 versus SU 100 +19% (+10% to +29%)
	Pitman et al. (237)		PL	86	40		9	19
			EL 40	175	62	22	19	6
			EL 80	170	70	30	26	8
			SU 25	171	53	13	17	14
			SU 50	175	56	16	18	6
		Success rates EL 40 versus SU 25 +10% (−1% to +20%) EL 40 versus SU 50 +6% (−4% to +17%)
		EL 80 versus SU 25 +17% (+7% to +28%) EL 40 versus SU 50 +14% (+4% to +24%)
		Pain-free EL 40 versus SU 25 +2% (−6% to +11%) EL 40 versus SU 50 +1% (−7% to +10%)
		EL 80 versus SU 25 +10% (+1% to +18%) EL 40 versus SU 50 +8% (−0.5% to +17%)
	Mathew et al. (194)		PL	419	26		5	47
			EL 40	822	64	38	35	31
			SU 100	831	56	30	26	37
		Success rates: EL 40 versus SU 100 + 8% (3% to +12%)
		Pain-free: EL 40 versus SU 100 +9% (+5% to +13%)
	Dowson et al. (27)		PL	99	42		15	20
			AL 12.5	184	57	15	28	18
			AL 25	191	57	15	35	15
			SU 100	194	64	22	34	25
		Success rates AL 12.5 versus SU 100 −7% (−17% to +4%) AL 25 versus SU 100 −7% (−17% to +3%)
		Pain-free Al 12.5 versus SU 100 −6% (−15% to +4%) AL 25 vbs SU 100 +1 (−8% to +11%)
	Spierings et al. (279)		AL 12.5	591	58		18	15
			SU 50	582	57		25	19
		Success rates: AL 12.5 versus SU 50 +2% (−4% to +6%)
		Pain-free: AL 12.5 versus SU 50 −7% (−11% to −2%)
	Pascual et al. (231)		PL	146	29		10	26
			RI 10	292	71	42	43	28
			ZO 2.5	289	67	28	36	29
		Success rates RI 10 versus ZO 2.5 +4% (−4% to + 11%)
		Pain-free RI 10 versus ZO 2.5 +8% (−0 to +15%)
	Bomhof et al. (12)		PL	107	22		8	25
			RI 10	201	68	46	45	33
			NA 2.5	213	48	26	21	21
		Success rates RI 10 versus NA 2.5 +20% (+11% to +26%)
		Pain-free RI 10 versus NA 2.5 +24% (+15% to +33%)
	Steiner et al. (282)		PL	135	22		6	52
			EL 80	360	74	52	26	38
			EL 40	359	64	42	32	29
			ZO 2.5	376	60	38	44	33
		Success rates EL 80 versus ZO 2.5 +14% (+7% to +21%)
		EL 40 versus ZO 2.5 +4% (−3 to +11%)
		Pain-free EL 80 versus ZO 2.5 +17% (+10% to +24%)
		EL 40 versus ZO 2.5 +6% (−1% to +12%)
	Garcia-Ramos et al. (106)		PL	92	31		19	28
			EL 40	192	56	25	35	29
			NA 2.5	199	42	12	18	26
		Success rates EL 40 versus NA 2.5 +14% (+4% to +24%)
		Pain-free EL 40 versus NA 2.5 +17% (+8% to +25%)
Subcutaneous Administration
	Dahlöf et al. (55)		PL	63	41		17	35
			NA 0.5	60	65	24	30	39
			NA 1	55	75	34	44	41
			NA 2.5	42	83	42	60	49
			NA 5	34	94	53	79	22
			NA 10	34	91	50	88	29
			SU 6	47	89	48	55	45
		Success rates NA 5 versus SU 6 +5% (−7% to +17%) NA 10 versus SU 6 +2% (−11% to +15%)
		Pain-free NA 5 versus SU 6 +24% (+4% to +44%) NA 10 versus SU 6 +33% (+15% to +51%)
Percentages in parentheses are 95% CIs.
^aA decrease in headache from severe or moderate to none or mild at 2 hours. ^bPercentage success with active drug minus percentage success with placebo. ^cAt 2 hours. ^dPercentage of patients with an initial success who had an increase in headache to moderate or severe within 24 hours. ^eSecondary efficacy parameter (68). ^fSuccess rate at 4 hours. ^gPatients selected as having frequent recurrences (≥50% of attacks treated with any medication). ^hStatistically significant difference at P < .01. Abbreviations: AL, almotriptan; EL, eletriptan; NA, naratriptan; PL, placebo; RI, rizatriptan; SU, sumatriptan; ZO, zolmitriptan.