Potentially Remediable Shortcomings in the Contemporary Drug Treatment of Migraine

Abstract

Despite the availability over the past decade of a number of new pharmaceutical agents with different mechanisms of action from those of the drugs used previously, the contemporary drug therapy of migraine attacks falls rather short of what would be desirable, while the pharmacological attempt to prevent further attacks appears to prove unsatisfactory about as often as it is successful. The present paper explores reasons for these shortcomings in both the earlier and the current drug therapy of the disorder. Significant major contributory factors appear to be an incomplete understanding of the underlying pathogenic mechanisms of the various stages of the migraine attack, less than optimal pharmacokinetic characteristics of many of the drugs used, and migraine sufferers failing to employ the available drugs to their best advantage. New drugs developed in the light of a more complete understanding of the molecular basis of migraine pathogenesis, together with awareness of pharmacokinetic desiderata in relation to treating and preventing migraine attacks, may go some way towards remedying the situation, but patient decision making may prove more difficult to modify.

1. Introduction

A rather substantial number of pharmacological agents is currently available for the treatment of migraine, and a considerably larger number has been used in the past but subsequently abandoned. Nevertheless, present-day clinicians remain conscious of their relative lack of ability to provide consistently efficacious relief for a significant portion of migraine sufferers. An exploration of possible reasons for this inability, as perceived from a clinician’s standpoint, might provide insights that could lead to ways of bettering the contemporary relative impotency of the drug therapy of the disorder.

The management of migraine is usually considered in relation to two main situations: (i) the attempted relief of the acute attack of the disorder and (ii) the prevention of further attacks when it appears likely that such attacks will continue to occur and significantly interfere with the sufferer’s way of life or otherwise diminish its quality. In clinical practice, the management of these two aspects of migraine involves a good deal more than simply the prescription of appropriate pharmacological agents, but in what follows, the emphasis is very heavily though not entirely given to the pharmacological aspect. Nevertheless, the attitude and behaviour of the intended recipient of these pharmacological agents does need to be taken into some account since, in everyday life, the decision as to whether and when to actually use the agents will nearly always be made by the migraine sufferer and not by the prescriber.

2. The Pathogenesis of Migraine

At a time when the drug therapy of disease is already very substantially based on scientific considerations, it may be useful to touch on some current aspects of the understanding of the pathogenesis of migraine, though not in close detail, before proceeding to consider issues in the pharmacological management and prevention of attacks of the disorder.

At the outset, it seems worth making the point that migraine is a disorder that afflicts humans. There is no convincing proof that an animal counterpart with similar features exists though animal models of aspects of the overall migraine process have been developed [1]. Animals simply do not have the ability to provide the detailed information about their sensory experiences to permit humans to know whether the experiences are those of the whole spectrum of human migraine. In addition, no objective non-verbal measures exist that allow the unequivocal recognition of migraine. Consequently, the application of the results of animal investigations to human migraine necessarily involves some uncertainty.

During at least the past two centuries, there have been two main lines of interpretation of the bodily disturbances that produce the human migraine attack though there were much earlier intimations of the roles of each. The first interpretation and that currently favoured is that a central nervous system disturbance initiates the phenomena of migraine; the alternative line of interpretation holds that the primary events are vascular in nature. As time has passed, there has been increasing overlap between the two lines of interpretation as reflected in the contemporary use of the word “neurovascular” in relation to the nature of the disorder. Leaving aside earlier writings at least in the English language literature, the two different lines of interpretation were set down explicitly but separately in the same year, 1873, by two Cambridge men. Edward Liveing [2] believed the disorder had a neural basis (in his “nerve storms”), while, following advances in understanding of the functions of the sympathetic nervous system, P.W. Latham [3] proposed that the features of the migraine attack were explained by altered cranial vascular behaviour. By the mid-20th century, vascular-based hypotheses had become predominant, principally due to the highly influential investigations and writings of Harold Woolf [4]. Additionally, a biochemical explanation for the vascular alterations was provided by the work of Sicuteri et al. [5], who produced evidence of altered circulating serotonin (5-HT) concentrations during the attacks, while Hanington [6] added a refinement by proposing a role for serotonin released from circulating platelets in the pathogenesis of the attacks. However, by the latter years of the 20th century, and largely as an outcome of the great expansion in neurobiological research that still continues, neurally centred concepts became the favoured explanation for the primary events in the migraine attack. The biochemical evidence underpinning the serotonin-centred vascular hypothesis does not seem to have ever been disproven, with the matter seemingly simply fading from collective medical memory. The various laboratory investigations that have subsequently been carried out in the endeavour to understand the pathogenesis of migraine have revealed an ever-increasing number of molecular and cellular mechanisms connected with nervous system function and its influence on vascular function, whose individual disturbances might suffice to explain particular aspects of migraine. With certain exceptions such as identifying the role of the calcitonin gene-related peptide (CGRP) in vascular biology, a sceptical outsider might be entitled to take the view that it has not been convincingly demonstrated that such disturbances actually occur in human migraine and, if they did occur, in what sequence. Despite such apparent shortcomings, the proposed roles of the changes have been incorporated into a number of increasingly complicated hypotheses concerning the mechanisms underpinning the migraine attack. Such explanations seem to be concerned mainly with the earlier headache phase of the attack and tend to ignore that fact that the site of the head pain may shift during the attack, and its character may be altered. There is a considerable amount of contemporary published material dealing with these matters, with several very useful overviews [7,8] that provide the biochemical background to the material mentioned in the present paper.

Before such contemporary interpretations of the mechanisms of the migraine attack can be accepted as wholly satisfactory, certain other inconvenient but well enough attested clinical observations need to be accommodated or otherwise interpreted. There is, for instance, the ancient finding, often enough subsequently confirmed, that pressure on the arteries in the neck or scalp relieves the headache [9,10]. Galen [11] attributed the description of this phenomenon to the lost writings of his predecessor Archibaldes (circa AD 100). This is not the place to go into such matters in greater depth though it may be of interest that attempts continue to be made to prevent the role of the vascular system in migraine headache pathogenesis from being further relegated [12]. Unfortunately, no explanation supported by convincing evidence is yet available that satisfactorily accounts for all the phenomena that may occur in and before migraine attacks, including ones involving aspects of cerebral and bodily function as well as headache and aura aspects. The experience of history suggests that, as time passes, the current understanding of the pathogenesis of migraine will develop further. As that happens, at least to some extent, the anticipated changes in understanding will influence the evolution of the drug therapy of the disorder, something it has already been doing over more than a century. One of the more successful and long-surviving pharmacological remedies for treating migraine attacks, ergotamine, was very probably a component of an extract of the ergot of rye that Woakes [13], in 1868, advocated as treatment for the migrainous head pain that he believed was a neuralgia. Many of the subsequently available specific remedies (see

Table 1. Drugs used to treat migraine headache attacks.

Simple Analgesics

COX inhibitors, e.g.,

Acetylsalicylic acid; NSAIDs, e.g., naproxen, diclofenac; Paracetamol.

Specifics

5HT1B/1D agonists

Short acting (t1/2 2–4 h): sumatriptan, naratriptan, zolmitriptan, eletriptan, rizatriptan, almatriptan; Intermediate acting: ergotamine (also has other actions); Long acting (t1/2 21 h): frovatriptan.

5HT1F agonists

Lasmiditan.

CGRP antagonists (gepants)

Atogepant; Rimegepant; Ubrogepant.

) have been developed with that same motive of correcting the underlying mechanism of the attacks, however and at whatever level of molecular sophistication that mechanism has been conceived over the years.

3. The Treatment of the Migraine Attack

Both in everyday clinical thinking and in official or semi-official headache classifications such as those of the International Headache Society, migraine attacks have long been sub-divided into those with and those without an aura, though there may sometimes be recognisable premonitory changes in cerebral and bodily functioning. To an appreciable extent, the contemporary headache and migraine literature has insidiously extended this classification of attack categories into a recognition of two major categories of migraine sufferer. The substantial majority of such persons suffer migraine without aura (the old-fashioned “common” migraine), while the remainder have migraine with aura (“classical migraine”). Yet, in actuality, many patients who have attacks of migraine with aura at other times experience attacks without any recognisable aura [14]. Failure to distinguish clearly between migraine with aura and sufferers from migraine with aura can sometimes make the interpretation of published work a little uncertain.

In most instances, migraine sufferers are very much more concerned with obtaining headache relief than in dealing with any migraine aura phenomena that they experience. Even so, for present purposes, there are advantages in considering management of the aura before that of migraine headache itself.

3.1. The Migraine Aura

The usual aura of migraine usually comprises a variety of temporary visual phenomena with typically changing geometrical patterns of visual illusion that involve parts or all of the field of vision, though there may also be areas of lost vision. Soon after the phenomenon of cortical spreading depression was described in experimental animals by Leao in 1944 [15]; this phenomenon, beginning as it does in the occipital lobes, began to be increasingly accepted as responsible for the visual aura of migraine. It is thought that a similar phenomenon accounts for those much less common auras that appear to arise in other brain areas, for instance, the brain stem in the entity that used to be called “basilar artery migraine” and is now termed “migraine with brain stem aura”. In this condition, accompanying the headache, there is episodic vertigo and sometimes temporary deafness and external eye muscle or facial palsy though Yamani et al. [16] believe the entity has been over-diagnosed and have proposed more stringent criteria for its recognition because similar phenomena can originate in the cerebral cortex.

As mentioned above, terminating any aura that is present is rarely a major concern for the patient already suffering or expecting to soon suffer a migraine headache. The aura will almost always prove self-limited. It is headache relief provided by pharmacological or other means that the sufferer seeks. Occasionally, however, dealing with the aura can be important for the sufferer. Rarely, the aura progresses into a more enduring though not necessarily permanent neurological deficit as, for instance, in so-called “hemiplegic migraine”. There does not seem to be much generally available information about pharmacological interruption of the cortical spreading depression process that is thought to underlie the aura though at least one substance, tonabersat, is reported to be able to do this [17]. Little seems to have been published about that substance’s clinical use or efficacy. The migraine auras for which there may be more compelling reasons for providing treatment are often those where there is a clinically appreciable disturbance of function in parts of the brainstem or the cerebral cortex outside the occipital lobes. This is particularly the case if the aura involves temporary expressive dysphasia or limb paraesthesia. There do not seem to have been formal systematic studies or controlled clinical trials of pharmacological agents in the management of these rather uncommon aura situations. However, there are reports based on small numbers of cases that certain measures, such as the use of antiseizure medications or calcium channel antagonists, may be effective in preventing their occurrence or in shortening their durations [18,19].

Perhaps potentially more significant for future possibilities in migraine management than the conventional aura is the very recent and growing interest into the various less definite and less easily perceived disturbances of cerebral function which seem to commence hours or days before any migraine aura or headache begins. Such aura or headache prodromes can comprise a heightened and often widespread bodily sensitivity to pain (allodynia) [20] or altered mood or slowed cognition. The mood changes generally tend towards a degree of mild depression accompanied by a slowing of thinking and reasoning and a hesitancy in making decisions. There may also be alterations or distortions of appetite or in sleep pattern; occasionally, instead, there may be an elevation of mood and a degree of mental hyperactivity, experiences whose implication the migraine sufferer can come to dread. The existence of such symptoms is often not readily volunteered by sufferers. The character of these experiences suggests that they arise from altered activity in the region of the thalamus and hypothalamus. Modern functional imaging studies are reported to show reasonably appropriately situated alterations in brain activity before the onset of recognisable clinical migraine [21]. These diencephalic alterations may be the true primary disturbance that develops into clinical migraine. If so, part of the future prospects for pharmacological relief of migraine may depend on identification of the relevant regional neurotransmitters and receptors that are involved and, on that basis, developing chemical agents that correct their dysfunction. Both in theory and at first sight, correcting or preventing such initiating events would seem highly desirable. However, that approach, if it can be developed, may in practice still fall short of expectations. Migraine sufferers often fail to recognize the presence of such premonitory alterations or appreciate their implication while they are present, though they may do so afterwards. If agents can be found that correct this probable diencephalic migraine-instigating disturbance, they may find their place in migraine prevention rather than in immediate treatment of existing attacks.

3.2. The Headache of Migraine

In recent times, two main lines of pharmacological approach have been utilised to produce relief for migraine headache [22]. Drugs usually considered as analgesics have been employed, with some success, during less severely painful headache episodes. More potent analgesics, some of them opiates, may sometimes have to be employed, albeit then rather reluctantly on the prescriber’s part, to afford relief when headache is severe, and lesser remedies have failed. The provision of such more potent substances for patient self-administration involves the hazard of later undesirable consequences, particularly drug dependence.

Preferably, drugs that act relatively specifically at various receptors thought to be involved in migraine attack pathogenesis are employed.

3.2.1. Analgesics

Disregarding the opiates and other powerful analgesics for whose efficacy in relieving migraine headache only anecdotal evidence is usually available, the pain-relieving agents used in migraine attack treatment are nearly always simple over-the-counter substances usually available in solid dosage form. There appears to be reasonable clinical trial and other literature evidence of efficacy that supports the use of paracetamol, acetylsalicylic acid, and various other cyclo-oxygenase (COX₂) inhibitors in providing migraine relief [23] though there is some evidence that paracetamol may not be as effective in this regard as its fellow mild analgesics [24]. Holland et al. [25] offered an assessment of the strength of the evidence supporting the use of such comparatively readily obtainable substances in the management of migraine attacks.

The co-administration of caffeine can enhance the degree of pain alleviation [26]. Word-of-mouth commendation for a role of caffeine alone in migraine treatment has long existed, perhaps dating back at least a century and a half (and then in the form of a component of guarana), which the very influential London physician Sir Samuel Wilks in 1878 had recommended for his patients whilst confessing that it had failed in his own case [27].

3.2.2. Antimigraine Specific Agents

This line of approach to the pharmacological treatment of migraine headache involves the use of agents that interfere with various biochemical mechanisms whose collective disturbance is currently believed to produce the head pain in migraine attacks. This is a more fundamental approach to treatment than the employment of analgesics and is usually adopted when simple over-the-counter pain-relieving remedies have previously failed to afford adequate relief.

Agents Acting at 5-HT_1B/1D Receptors

The main such specific agents in current use are various triptan derivatives, seven of which are marketed in various countries. All are agonists at 5-HT_1B/1D receptors. They were developed at a time when vascular interpretations of migraine pathogenesis were in the ascendancy, a short-lived rise in circulating serotonin levels being thought to produce the initial intracranial vasoconstriction and cerebral cortical ischaemia that was responsible for the aura manifestations. This initial phase, which might be clinically silent, was followed by a more sustained decrease in circulating serotonin levels that led to intracranial and extracranial vasodilatation that was responsible for the headache.

The initial report in 1988 of the use of the first of these triptans, sumatriptan, then unnamed and referred to as GR43175 [28] suggested that the drug possessed a degree of efficacy well in excess of that of the then usually employed migraine-specific agent, ergotamine, taken by mouth. In this initial publication, the sumatriptan had been given by intravenous injection rather than taken orally, and ergotamine administered parenterally also seemed more consistently effective than when the drug was swallowed. As additional publications appeared reporting the effectiveness of oral sumatriptan in relieving migraine headache, it became increasingly apparent that the drug’s administration route played an appreciable role in determining its success in relieving migraine headache attacks. Some quite good-quality studies showed that self-administered oral sumatriptan was little or no more effective than orally administered acetylsalicylic acid taken in sufficient dosage early in the course of migraine headache [29,30,31,32]. As other triptan derivatives became available for oral administration, their efficacies proved reasonably similar to that of sumatriptan. There were some differences, but they generally were of no great magnitude [33]. It would be hard to exclude the possibility that the differences depended as much on the drug doses that it was practicable to administer using the marketed drug dosage forms as on the intrinsic properties of the drug molecules.

It also became increasingly clear that, after oral intake, incomplete delivery of the swallowed dose of at least some of the triptans into the general circulation in the form of the intact molecule might be denying them their full potential for providing headache relief. The oral bioavailabilities of the marketed triptans, values often enough determined in healthy volunteers rather than during migraine attacks, differ. Sumatriptan has an oral bioavailability of only about 16% because of its extensive pre-systemic elimination. At the other extreme, the value for naratriptan is around 78%. This potential limitation in the effective oral use of the triptans can be overcome by employing higher, perhaps substantially higher, oral than parenteral doses, as shown in the illustration published in Ashina’s paper [7] (p. 1871). Unfortunately, there would still remain possible consequences of the pharmacokinetic truism that drugs with low oral bioavailabilities often have bioavailabilities that vary considerably from person to person. The prescribing uncertainties associated with awareness of this potential issue could, at least in theory, be lessened if data correlating headache relief and circulating drug concentrations were available and were used to determine future dosages in the individual. It is regrettable that very little such information is available.

There is a further potentially complicating circumstance that may impair the efficacy of orally administered triptans in migraine attacks. During the attacks, patients are often nauseated, have decreased gastrointestinal motilities, and sometimes may vomit. Such contingencies are likely to slow the rate and perhaps decrease the completeness of the absorption of swallowed drugs unpredictably even before pre-systemic metabolism may come into play. Tfelt-Hansen’s [30] review of the issue noted the existence of published data suggesting that the early absorptions of sumatriptan and rizatriptan were not delayed during migraine attacks though the time of reaching the peak circulating concentration of the former was. There was evidence of interference in the absorptions of the other triptans during headache. These same issues would probably have applied for oral ergotamine, now disappearing from use in some countries and originally thought to act through catecholaminergic mechanisms but later recognised as a very potent agonist at 5-HT_1B/1D receptors, thus possessing the same mechanism of action as that of the triptans. Actual measured values for ergotamine’s oral bioavailability do not appear to be available, largely because of insufficient assay sensitivity at the time when there was interest in the drug, but ergotamine probably undergoes even more extensive pre-systemic biotransformation than most of the triptans. A similar situation applies for the structurally related molecule dihydroergotamine, usually available only for parenteral use [34].

There have been attempts to overcome this migraine-engendered difficulty by oral co-administration of agents that enhance gut motility and diminish nausea, e.g., metoclopramide or domperidone. In earlier times and occasionally since, prochlorperazine has been used for this purpose [34]. However, taking by mouth these gastrointestinal tract function-altering agents, which themselves act within the central nervous system, exposes them to the same delayed and incomplete absorption possibilities as the triptans may face during migraine attacks.

Measures other than those mentioned above have been tried to overcome the oral drug administration uncertainties. Ergotamine, for instance, was marketed in a rectal suppository preparation. This mode of administration avoided the consequences of impaired gut motility, while the possibility existed that part of the drug dose might be absorbed from the rectum into that part of the rectal venous blood that bypasses the liver on its first circuit around the body, thus avoiding pre-systemic metabolic transformation. There was a clinical impression that such rectal administration probably enhanced the effectiveness of the drug in relieving migraine headache. Apparently, a sumatriptan preparation designed for rectal administration is available in some countries. Otherwise, the rectal administration possibility for the other triptans does not seem to have been pursued.

The possibility of significant absorption of some of these serotoninergic drugs through the buccal mucosa has also been exploited. Blood from the mouth bypasses the liver on its first circuit around the vascular system. The buccal route of administration should also avoid gut motility issues. Unfortunately, the comparatively small surface area of the buccal mucosa does not favour absorption of the complete dosage that enters the mouth before at least part of the drug is swallowed. Further, the local pH of the mouth may not favour the efficient passive absorption of some of the drugs, for example, ergotamine, because there will be an increased proportion of the drug molecules in an ionised and relatively unabsorbable form. In addition, there may be taste-tolerability issues. In the past, an ergotamine preparation intended for buccal administration route was marketed, and currently, rizatriptan is available in a wafer form that allows the drug to be retained in the mouth for long enough to permit a useful degree of absorption.

A rather more extensive absorptive surface is available in the surface membranes of the nasal cavity. Preparations of both dihydroergotamine and sumatriptan designed for absorption via this route have been marketed. However, the taste associated with the sumatriptan preparation, besides being unpleasant enough to sometimes prove unacceptable, suggests that part of the nasal dosage may reach the mouth and tongue. A sumatriptan powder preparation for nasal administration has also been developed and is reported to provide an earlier onset of benefit than contemporary triptans taken orally [35].

In the past, one pharmaceutical firm developed a pressurised spray preparation of ergotamine that was intended to allow part of the drug dose be absorbed through the more extensive surface of the lower respiratory tract and pulmonary alveoli. Whether significant quantities of ergotamine ever reached the lowermost parts of the respiratory tract does not seem to have been established, but there must have been sufficient absorption from the mouth, the mucosa of the upper respiratory tract, and perhaps that of the oesophagus to allow the preparation to display a reasonable degree of efficacy in relieving migraine headache.

The possible absorption and pre-systemic biotransformation limitations that may interfere with the efficacy of the ingested currently available anti-migraine serotoninergic agents would be avoided if the drugs were injected. In headache attacks, self-administered intravenous injection would be impracticable for the great majority of migraine sufferers, but subcutaneous or intramuscular injection might be manageable though the extents and rate of the injected drug´s absorption might then prove an issue. An injection of the ergot derivative dihydroergotamine has been marketed as well as a sumatriptan preparation for subcutaneous administration. The latter does not seem to have achieved any widespread use, possibly largely because of cost considerations, though evidence exists that it possesses a substantial degree of efficacy [36].

Agents Acting at 5-HT_1F Receptors

At the time of writing, at least one 5-HT_1F receptor agonist, lasmiditan, seems to have emerged successfully from preclinical and clinical pre-marketing assessment to become commercially available in some countries for treating migraine attacks. The drug is reported to have an advantage over the 5-HT_1B/1D agonists in that it does not cause vasoconstriction, though clinically unwanted vasoconstrictive manifestations do not seem to often produce serious issues when the triptans are used clinically. This is unlike the situation as regards ergotamine, which, in overdosage, may cause gangrene though, at least in recent times, this catastrophe appears to have occurred mainly when the drug was taken in conjunction with an inhibitor of ergotamine metabolism, e.g., erythromycin or a related macrolide antibiotic [37]. Dihydroergotamine seems to be a less potent vasoconstrictor at therapeutic dosage. Full details of the pharmacokinetic profile of lasmiditan do not yet seem to be generally available in the medical literature though there are indications that the drug has an incomplete oral bioavailability. Nor is its clinical effectiveness at present as well-established as that of the triptans. The size of its oral dosage when used in its clinical trials suggest that, mg for mg, it may be less capable of providing headache relief than most of the 5-HT_1B/1D agonists (data in [7], p. 1871, [38]). It remains to be seen how important an addition to the therapeutic armamentarium of those who treat migraine the drug and forthcoming agents with similar mechanisms of action would prove to be after enough time has passed for the likely initial enthusiasm for a new class of remedy to be replaced by a more sober assessment.

Agents Acting at CGRP Receptors

The role of CGRP in producing the intra- and extra-cranial vasodilatation that occurs in migraine attacks and probably confers on migraine head pain its often pulsatile character has been known for over a decade. Agents have been developed to block the actions of this highly potent molecule and parenteral preparations of monoclonal antibody CGRP receptor antagonists are becoming increasingly available commercially. Their long durations of action make them better suited to providing medium- to longer-term protection against future migraine attacks than in achieving speedy mitigation of existing migraine headache. The latter role is better provided by CGRP receptor antagonists that can be reasonably efficiently absorbed from the alimentary tract. The earlier of these latter so-called gepants proved unacceptable for human use on safety grounds (producing abnormalities of liver function). However, newer derivatives now entering the marketplace, such as ubrogepant, rimegepant, and atogepant [39], appear not to be liable to such toxicity problems though it is perhaps rather too early to be sure of their clinical effectiveness and their general acceptability from the migraine headache sufferer´s standpoint.

Thus, at present, there are available a number of reasonably well-established agents that are useful in relieving migraine headache. Attempts have been made to assess the quality of the evidence supporting the use of many of them [40,41], but no single agent in clinical use has emerged as being distinctly and consistently superior to its fellows. The available substances have the capacity to provide relief of the headache component of the migraine attack through their effects at one or other of three different known classes of receptor, namely the 5-HT_IB/1D, 5-HT_1F, and CGRP ones. This knowledge raises the possibility that the simultaneous intake of two of these existing agents, but ones that individually bind to receptors of different classes, might provide more effective and perhaps more expeditious relief for migraine headache than use of any single agent, though the ergot derivatives have a broader spectrum of relevant receptor action that the triptans and lasmiditan. While such a two-pronged onslaught on relevant receptor mechanisms would be intended to be beneficial in relation to affording headache relief, it must be admitted that experience in practice might show that there were unanticipated and undesirable pharmacodynamic and pharmacokinetic consequences. Yet, there do not seem to be reports of the use of such theoretically rationally receptor-based combination therapy though the sequential use of analgesics followed by migraine-specific receptor active agents has long gone on, and a sumatriptan–naproxen combined product has been studied [7].

Agents Acting at Other Receptor Types

The available literature reports that various drugs believed to act at other receptor types that have already been identified as probably playing a role in the pathogenesis of migraine headache are currently under investigation or are being developed [19,42,43]. The receptors involved in their actions include pituitary adenyl cyclase activating peptide receptors, certain classes of glutamate receptor, orexin receptors, receptors of the endocannabinoid system [44], and perhaps receptors subserving the near final stage headache-producing molecular mechanism by blocking nitric oxide synthase activity or ATP-sensitive potassium channels [45].

Overall, it would seem that potentially quite highly effective specific agents are already available for interrupting some of the mechanisms involved in stages of the probable pathogenesis of migraine headache. In practice, the actual timing of the intake of these agents will nearly always be decided on by headache sufferers themselves and may not always be optimal, usually because of delayed intake. The potential benefits from these drugs in clinical use may be undermined, sometimes to appreciable and probably individually variable extents, by a relative inability to deliver them in consistent and safe quantities to the receptor mechanisms in the body regions where their actions are required. Further, after a single dose, the durations of action of some of them is short enough for an already interrupted headache to return after a few hours of pain relief. The latter limitation applies mainly to those triptans with 2 to 4 h elimination half-lives. It was not a particular problem with ergotamine in the past. Among the triptans, the duration of action of frovatriptan (half-life around 21 h) may be unnecessarily long for dealing with a single migraine attack. The availability of a speedily absorbable triptan derivative with an elimination half-life of around 4 to 6 h and not vulnerable to pre-systemic biotransformation would seem highly desirable. However, at the present time, the pharmaceutical industry might not regard developing such an agent as commercially attractive.

3.2.3. The Later-Stage Headache in Migraine

Patients whose migraine attacks run their natural courses uninfluenced by treatment often notice that, as the hours pass, their headache pain changes in character and may alter in its site of occurrence, perhaps moving from one side of the head to the other. If the earlier stage pain had been pulsatile in character, it may have increasingly taken on the nature of a tight, squeezing scalp discomfort rather like that of tension-type headache. There seems to have been very little recent investigation of the pathogenesis of these later headache stage changes though possible mechanisms have been suggested in the past and may well be proven correct if and when the requisite evidence is obtained. In the late stages of the attack, the migraine specific agents seem to be of little value, nor does the use of simple analgesics. However, women may remark that, at a late stage in their headache, running a hot shower over their heads and the backs of their necks can provide useful relief, suggesting that a degree of neck muscle tightness may be contributing to the pain at that time.

Sometimes, increasingly frequent migraine attacks may merge into a state of continuing headache, nowadays termed chronic migraine, when a migraine component can be recognised in the overall headache on at least 15 days a month, though not all the headache need necessarily be migrainous. The designation “chronic migraine” probably takes in what an older classification of headache termed “mixed headache”, a merging of migraine and tension-type headache. Repeated attempts to manage such chronic migraine, often with a varying and growing array of pharmacological agents, can lead to a state of pharmacological refractoriness and what is currently termed “medication overuse headache”. The underlying mechanisms and management of this entity are not considered further in the present paper. In recent years, chronic migraine and medication overuse headaches have grown into topics warranting consideration in their own rights, for instance [46].

3.3. The Prevention of Migraine

The attempted prevention of migraine is indicated in two somewhat different situations, namely (i) when the migraine sufferer is very likely to soon be exposed to a factor known to reasonably consistently provoke that person´s headache attacks and (ii) when, as is much more often the case, migraine headaches occur unpredictably but frequently enough or are individually severe enough and not responsive to immediately available treatments to warrant continuous use of pharmacological preventative measures to avoid their further occurrence.

3.3.1. Strategically Timed Prevention

Based on their previous experience, a minority of migraine sufferers can identify particular situations or circumstances during or after which they know they are particularly likely to experience a migraine attack. Some of these provoking circumstances may be avoidable, for instance, the intake of a particular foodstuff or a particular form of alcohol, physical exercise particularly in the young, involvement in inescapable and stressful social or employment situations. Other provoking circumstances may be, for various practical considerations, not easily avoidable, e.g., premenstrual migraine attacks. In addition, and particularly if the possibility of successful pharmacological prevention of their attacks exists, some migraine sufferers may elect to enjoy or endure potentially avoidable attack-provoking factors by taking out what they consider is drug-based insurance against the otherwise expected event.

Appropriately timed anticipatory doses of drugs used to treat migraine headache attacks have been employed for the pharmacological prevention of such anticipated attacks. The elimination half-lives of most of the currently available triptans are probably inconveniently brief for this purpose. However, frovatriptan, with its appreciably longer elimination half-life, at least in theory would appear appropriate, as ergotamine sometimes proved to be in the past. While no good data regarding ergotamine´s elimination half-life and its other pharmacokinetic parameters appear to be generally available, clinical experience of the apparent duration of action of a single oral dose of the drug suggests that, whether in the form of the parent substance or its metabolites, it could be effective for around 12 h or longer after intake. The elimination half-life of dihydroergotamine is around 12 h [47].

An alternative pharmacological approach to the prevention of expected attacks is to employ the appropriately timed intake of acetylsalicylic acid or some other NSAID with a reasonably long half-life and for which evidence of efficacy exists [48,49]. The elimination half-life of the intact acetylsalicylic acid molecule is very brief, but the durations of some of its biological effects are much longer, probably because of the drug´s irreversible inhibition of cyclooxygenase. In practice, a single oral dose of acetylsalicylic acid often seems to ensure that a migraine attack expected to occur within the next 12 h or so will not eventuate.

3.3.2. Continuous Prevention

Various non-pharmacological methods and items of apparatus intended for continuous or prolonged intermittent use are available for the prevention of migraine, though evidence for their efficacies often is not readily available. In clinical practice, pharmacological measures comprise a very much greater component of the attempt to prevent migraine attacks when these are occurring frequently enough to warrant such intervention or are individually severe enough when they occur and are not satisfactorily relieved by immediate drug intake.

A rather considerable number of drugs is currently marketed in the hope that they can at least reduce the frequency and severity of future migraine attacks. Over the centuries, there have been numerous others that have subsequently faded from use, probably mainly because of ineffectiveness, unacceptable toxicity, or both. The use of most of the currently available agents is supported by reasonable quality evidence of greater effectiveness than placebo substances. However, a few relics from the past, introduced when such efficacy evidence was not a prerequisite to marketing, continue to survive in use, sustained by clinical impressions of their value. Those who frequently prescribe migraine-preventative agents seem to have an impression that such agents tend to be about twice as effective as the placebo against which they have been compared previously. Comparative clinical trial data, for instance, that illustrated by Ashina [7], seems broadly consonant with that clinical impression. A similar merit assessment regarding the degree of effectiveness seems to apply for the relatively recently available injected anti-CGRP-receptor antagonists [49], which do not seem notably more effective in preventing migraine than their still-available predecessors.

The currently used preventatives do not appear to share any single known mode of pharmacological action. In terms of contemporary clinical categorisation, the preventatives fall into a number of recognised classes (

Table 2. Main drugs currently used in migraine prevention.

Beta-Adrenoceptor Blockers

1. Propranolol, timolol

5-HT2 antagonists and antihistamines

Pizotifen, cyproheptadine

5-HT reuptake inhibitors

Amitriptyline—also a tricyclic antidepressant

Ca2+ channel blockers

Verapamil, flunarizine

Antihypertensives

ACE inhibitor—lisinopril Angiotensin receptor blockers—candesartan

Antiseizure medications

Topiramate, valproate

CGRP antagonists

Injected monoclonal antibodies—eptinezumab, erenumab, fremanezuab, galcanezumab Oral intake—atogepant, rimegepant, ubrogepant

Onabotulinum toxin A

Miscellaneous agents

Riboflavin Baclofen

). Some are beta-adrenergic blockers (that lack any sympatheticomimetic actions); some are antihistamines (which happen to also act on serotonin 5-HT₂ receptors); some are calcium channel antagonists, e.g., verapamil or flunarizine; some are antihypertensives, including certain ACE inhibitors, e.g., lisinopril and angiotensin receptor blockers, e.g., candesartan (although the calcium channel antagonists verapamil could be classed as an antihypertensive agent); some are antiseizure medications (valproate, topiramate); and one or two are antidepressants (principally tricyclic ones, e.g., amitriptyline), while the recently available CGRP receptor antagonist monoclonal antibodies have yet to find a simpler class designation [40,50,51,52]. There also are miscellaneous agents with various accepted modes of action that also have known migraine-preventing capabilities, e.g., baclofen [53], the vitamin riboflavin [54,55], and onabotulinum toxin A. Reasonable-quality evidence for the efficacy of the latter exists but mainly in relation to chronic migraine. Herd et al. [56], in a meta-analysis, could not find good-quality data demonstrating the toxin’s value in preventing episodic migraine. In two subsequent systematic reviews [57,58], the toxin was considered to be less effective for migraine prevention than topiramate or anti-CGRP monoclonal antibodies, though it was not ineffective. Topiramate seems to have recently been accorded the role of something like a gold standard in comparative efficacy studies of migraine prevention [58], but there is now evidence that the drug is a dose-related teratogen [59] though that hazard may not be particularly great at the comparatively low doses usually employed in migraine prophylaxis. The teratogenesis concern applies much more strongly to valproate [60]. If valproate continues to be used in migrainous women of child-bearing potential, it seems inevitable that a pregnancy will occur in someone still taking the drug for migraine prevention, with the outcome of a malformed foetus and subsequent litigation.

Considered at a molecular level, the range of known mechanisms of action of the drugs mentioned in the paragraph immediately above becomes a little less extensive. Beginning with ergotamine, which at one time was marketed for regular daily use in compound preparations, which have long since disappeared, and many of the earlier drugs that remain commercially available act at serotonin-related mechanisms. Among the anti-serotonin antihistamines, pizotifen is an antagonist at 5-HT₂ receptors, and it seems likely that cyproheptadine and methdilazine, with very similar chemical structures to that of pizotifen, would act in the same way. There appears to be some evidence that the beta-blocker propranolol affects platelet serotonin transport [61] though the subsequent study of Shields and Goadsby [62] suggested that the drug´s effects at thalamic level were mediated through its β₁ blocking effects rather than any action on serotoninergic mechanisms. Amitriptyline blocks serotonin reuptake into nerve terminals and thereby increases local extracellular concentrations of this vasoactive neurotransmitter. Other classes of antidepressant that act through the same serotoninergic mechanism in the central nervous system as amitriptyline does do not seem as consistently useful in migraine prevention. It may be that nortriptyline, the metabolite of amitriptyline, which impedes noradrenaline reuptake, also plays a role in the apparent migraine prevention effectiveness of its parent molecule.

It is not beyond possibility that all these available and reasonably adequately established migraine prophylactics share some as-yet unrecognised molecular mechanism of action that decreases the frequency and severity of migraine attacks. However, the very diversity of their currently known mechanisms of action suggests that they may exert their migraine-preventing effects at different stages or on different components of the molecular events that underlie migraine, particularly those in its earlier, often largely clinically silent, pre-headache phase. In practice, a given stable daily dose of some of these preventatives may not seem to produce its full effect for a few weeks after its intake begins, and its benefit seems to persist for more than a month after the drug’s intake ceases. Such seemingly delayed effects raise the possibility that some form of receptor adaptation may develop over some days in response to the initial actions of these drugs. Possibly, this may be the actual mechanism through which some migraine prophylactics act.

While awaiting information that may further clarify the above issues, it seems worth touching on a possibility that arises from existing knowledge of the likely different molecular mechanisms involved in migraine prevention. Might combinations of agents with different but relevant mechanisms of action be employed to try to obtain additive preventative benefits? If that were done, there of course would be the possibility of pharmacokinetic interactions, which might have undesirable consequences, and of pharmacodynamic ones, which might result in any possible dividends from migraine prevention being outweighed by unwanted effects. Nevertheless, it does seem regrettable that more has not already been done in this regard or, if done, reported in the literature. Admittedly, the possibility has been explored occasionally [63], for instance, in combining anti-CGRP receptor antibody therapy with onabotulinum toxin A [64], but that was in relation to managing chronic migraine.

4. Conclusions

The main justification for trying to identify ways in which the contemporary drug treatment of migraine falls short of what might have been wished lies in the hope that the resulting knowledge may point the way to future more effective means of remedy of the disorder.

There is an obvious need for a more complete understanding of the pathogenesis of the migraine headache attack through its successive stages and also of the prodromal cerebral and extra-cerebral events that may progress into the clinical attack. It would be helpful to know whether identified chemical changes that occur during migraine attacks relate to essential events in the pathogenesis of the disorder or to secondary consequences of the essential events. On the basis of such knowledge, additional receptor sites may be recognised where newly devised or perhaps existing drugs may be able to interrupt or reverse stages in the evolving pathogenesis of the headache. Such attempts are currently proposed or already underway, as reported at several places in the recent literature [65].

Rather than attack by chemical means one or more of the sequential intermediate molecular disturbances that produce migraine headache, it might seem better to block the molecular events that initiate the pathogenic processes that underlie the headache. However, to benefit from this more fundamentally based course of action, the migraine sufferer would need to recognize the onset of the manifestations of these events. In reality, it would rarely be possible for the sufferer to be able to do so. Therefore, it seems likely that for the foreseeable future, the pharmacological attempt to relieve migraine headache will continue to be based on reversing or blocking elements of the mechanisms that are already producing recognised head pain. Should new knowledge reveal a way of interrupting the molecular events associated with the initiation of the processes that later culminate in the experience of headache, such knowledge in practice will probably be better applied to the prevention of future migraine attacks in the sufferer than to managing headaches that are already extant.

Certain properties would seem desirable in new drugs developed for migraine headache treatment. Obviously, such drugs should be effective at or below doses at which any unwanted effects cease to be minor or trivial. The drugs should have convenient routes of administration and be available in convenient to use dosage forms. Their absorption from administration sites should be rapid and reasonably complete, and preferably, they should undergo little or no pre-systemic biotransformation. If intended to treat individual migraine attacks, elimination half-lives in the 4 to 6 h range would seem optimal. This would be long enough to make the return of an already suppressed attack unlikely but not so long that if an attack did recur, there would too great an uncertainty about the safety of a further dose. On the other hand, if the forthcoming new drug were intended for the prevention of future migraine attacks, elimination half-lives of 1 or 2 days would make once-daily intake feasible, and there would be a margin of safety of one or two days if doses were missed.

Information about correlations (if they exist) between circulating drug concentrations and degree of pain relief or effectiveness of future attack prevention might provide an invaluable guide to clinical management in individual sufferers.

It would be very useful to have good-quality information regarding the likelihood of obtaining adequate pain relief after a first drug had failed in this regard, if a second drug with the same mechanism of action were prescribed, or if one with a different mechanism was used.

Finding new and intrinsically more effective drugs and fulfilling all these other desiderata should, at least in theory, lead to substantially better outcomes from the pharmacological management of migraine. However, there would remain the problem of overcoming the idiosyncrasies of patients’ behaviour and, in particular, a failure to be organized so that their treatment can be taken at the optimal stage in their migraine episodes rather than deferred until the consequences of the delay become intolerable. Patient education to try to overcome this limitation to the effectiveness of treatment may prove as important for the future of migraine pharmacotherapy or be even more important than the development of new pharmacological entities.