In this randomized, placebo-controlled trial, we demonstrated that oral administration of cilostazol, a PDE-3 inhibitor, can elicit migraine-like headache in persons with PPTH, despite them having no pre-mTBI history of migraine. The spontaneous episodes of migraine-like headache experienced by the participants, closely resembled the characteristics of the migraine-like headache induced by cilostazol. Taken together, our results support the involvement of PDE-3 and cAMP-dependent signaling pathways in the pathogenesis of PPTH. Targeting these mechanisms might present novel avenues for drug development and fulfilling unmet treatment needs of persons with PPTH.

PDE-3 inhibition, cAMP-Dependent signaling pathways, and Cephalic Pain

The involvement of PDE-3 inhibition in cephalic pain has been well-documented in previous human experimental studies [8,9,10,11]. Oral administration of cilostazol has proven to elicit migraine attacks in persons with migraine [9,10,11], while causing mild headache in healthy volunteers [8]. It is also worth noting that cilostazol-induced migraine attacks are reproducible in persons with migraine, who were administered cilostazol on two separate experiment days [11]. Considering the evidence above and our own findings, it seems apparent that the pathogenesis of cephalic pain involves, at least in part, increased intracellular cAMP levels (Fig. 4).

Fig. 4

Proposed Mechanisms and Sites of Action of Cilostazol-Induced Migraine-Like Headache in Persons with PPTH. The illustration delineates a hypothesized mechanism and site of action whereby diverse pharmacological triggers contribute to the genesis of migraine-like headache in people with persistent post-traumatic headache (PPTH). In this proposed model, the neuropeptides calcitonin gene-related peptide (CGRP) and pituitary adenylate cyclase-activating polypeptide (PACAP-38) engage with their respective G protein-coupled receptors situated on vascular smooth muscle cells within the walls of meningeal arteries. Both peptides instigate adenylate cyclase (AC) activation via their transmembrane receptors, leading to an augmented intracellular cyclic adenosine monophosphate (cAMP) production. Cilostazol, a selective inhibitor of phosphodiesterase type 3 (PDE-3), impedes cAMP breakdown, resulting in its accumulation. The cAMP signaling pathway is postulated to activate and open ATP-sensitive potassium channels (KATP) and large conductance calcium-activated potassium channels (BKCa) channels. These subsequent events lead to potassium efflux, accompanied by the vasodilation of meningeal arteries. Modified from Al-Khazali et al., 2023.[19, 23]

To further substantiate the importance of cAMP-dependent signaling pathways, previous experimental studies have used calcitonin gene-related peptide (CGRP) and pituitary adenylate cyclase-activating polypeptide (PACAP) (Fig. 4) [13,14,15,16]. Both of these endogenous signaling molecules have been shown to induce migraine-like headache in persons with PPTH and migraine attacks in persons with migraine [15, 17,18,19]. Furthermore, the downstream effects of both CGRP and PACAP signaling converge on the upregulation of intracellular cAMP levels [13, 20]. This convergence provides yet another layer of evidence to support the assertion that targeting cAMP-dependent signaling pathways holds promise for developing mechanism-based drugs for PPTH.

Possible sites and mechanisms of Action

The evidence discussed above has established that upregulating intracellular cAMP using PDE-3 inhibitors, such as cilostazol, elicits cephalic pain. However, important questions remain unanswered regarding the exact site(s) and mechanism(s) of action underlying this pro-nociceptive effect. One possible explanation involves a cascade of events, focusing on activation of perivascular meningeal nociceptors via mechanical and/or chemical stimulation [13, 20]. This hypothesis posits that the site of action for PDE-3 inhibitors is the vascular smooth muscle cells (VSMCs) within the walls of meningeal arteries [13, 20]. These arteries, located in the meninges that envelops the brain, regulate the cerebral blood flow. PDE-3 inhibitors exert their action by blocking the activity of PDE-3, an enzyme responsible for the breakdown of cAMP [6, 7]. This blockade leads to an accumulation of cAMP within the VSMCs, which facilitates the opening of two types of potassium channels– large conductance calcium-activated potassium (BKCa) channels and ATP-sensitive potassium (KATP) channels [21, 22]. Of note, recent experimental findings have demonstrated that pharmacologic opening of either potassium channel can elicit migraine-like headache in persons with PPTH [23]. Upon opening, both BKCa channels and KATP channels allow positively charged potassium ions to flow out of the VSMCs, causing them to hyperpolarize [13, 20]. This, in turn, leads to dilation of the meningeal arteries and possible mechanical stimuli, activating perivascular nociceptors [13, 20]. Furthermore, the increased levels of extracellular potassium might also contribute with chemical stimulation of the nociceptors. However, it merits emphasis that there is no firm evidence to confirm this sequence of events. It is, nonetheless, intriguing that all inducers of migraine-like headache in persons with PPTH are known to dilate meningeal arteries [18, 19, 23]. Findings from magnetic resonance angiography have further revealed that the middle meningeal artery is dilated exclusively on the pain-side at the onset time of cilostazol-induced migraine attacks [24]. This also aligns well with the reporting of cephalic pain after intraluminal balloon inflation of the middle cerebral artery [25]. Taken together, it seems timely for animal studies to ascertain whether dilation of meningeal arteries and potassium efflux from VSMCs can activate and sensitize meningeal nociceptors.

Another plausible site of action is the meningeal nociceptors themselves or their cell bodies located in the trigeminal ganglia and upper cervical dorsal root ganglia [26]. Preclinical studies have indicated that increased intracellular levels of cAMP can elicit activation of these ganglia neurons [27]. In support, application of prostaglandin E2 sensitizes cell culture with trigeminal ganglia neurons [28], as well as those with dorsal root ganglia neurons [29]. This finding is particularly relevant because prostaglandin E2 exerts its downstream effects through the upregulation of intracellular cAMP levels, and administration of prostaglandin E2 is known to trigger migraine attacks in persons with migraine [30]. This line of reasoning suggests that administration of cilostazol increases intracellular cAMP levels within the primary afferent nociceptive neuron. If true, the accompanying cilostazol-induced dilation of meningeal arteries would be unrelated to the pathogenesis of cephalic pain per se. The increased blood flow via vasodilation would then serve to meet the nociceptors’ increased metabolic demands.

In addition to potential peripheral sites of action, it is pertinent to consider the possibility that cilostazol might induce migraine-like headache a central site of action. There is indeed some evidence to support that cilostazol can impact the blood-brain barrier (BBB) [31]. Preclinical studies have shown that cilostazol can protect against BBB dysfunction in certain pathologic conditions, such as hemorrhagic stroke [31, 32]. This interaction with the BBB implies that cilostazol might exert a direct effect within the CNS. Nonetheless, it remains to be determined whether cilostazol’s pro-nociceptive effects are mediated at a central site of action, or if any central effects are consequential to its peripheral actions.

Therapeutic implications and future directions

The therapeutic potential of PDE-3 activators for the treatment of PPTH is an emerging frontier in neuropharmacology. Our understanding of the role of PDE-3 and cAMP-dependent signaling pathways in the pathogenesis of headaches has evolved immensely in recent years. Experimental evidence, including the induction of cephalic pain in persons with PPTH after oral administration of cilostazol, underscores the pivotal role of PDE-3 and cAMP-dependent pathways in PPTH pathogenesis. Given this mechanistic insight, it is plausible to hypothesize that PDE-3 activation, leading to a decrease in intracellular cAMP levels, could potentially mitigate the symptoms of PPTH. This innovative approach could herald a novel therapeutic strategy for PPTH, a condition for which current treatment options are limited.

A key consideration in the development of PDE-3 activators is the specific targeting of VSMCs within the walls of meningeal arteries. This specificity is predicated on the understanding that the dilation of these arteries, a phenomenon associated with the activation of perivascular nociceptors and the onset of headaches, is modulated by the activity of PDE-3 in these cells [13, 20]. However, the path to developing PDE-3 activators as a viable treatment for PPTH is fraught with challenges [33]. Paramount among these is the need for selective targeting of PDE-3 in the affected tissues without disrupting cAMP signaling in other physiological systems [33]. This necessitates a granular understanding of the precise location and function of PDE-3 across different cell and tissue types [33]. Moreover, the development of PDE-3 activators must account for potential side effects, particularly in cells with high basal or stimulated cyclic nucleotide levels. It is also crucial to consider potential variations in the subcellular compartmentalization of PDE-3, which might differ across species, age, tissue types, and disease status.

Despite these challenges, the development of PDE-3 activators for the treatment of PPTH is an exciting and promising area of research. With continued studies to elucidate the precise role and regulation of PDE-3 in PPTH, coupled with advances in drug delivery systems, this approach could potentially revolutionize the therapeutic landscape for individuals suffering from this debilitating condition.

Cilostazol-Induced Headache: a proof-of-Concept Model for Acute Headache medications

A recent innovative proof-of-concept model has involved using cilostazol-induced headache to test new drugs for treating acute headache. This concept rests on the assertion that if a drug can alleviate cilostazol-induced headache, it holds promise as an effective treatment for acute headache and merit further investigation in phase II/III trials. To validate this proof-of-concept model, two experimental studies used oral administration of sumatriptan, a well-documented acute headache medication, to treat cilostazol-induced headache [34, 35]. The first study used a double-blind, two-way crossover design, in which 30 healthy volunteers were randomly assigned to receive pre-treatment with sumatriptan or placebo on separate experiment days, followed by cilostazol [35]. The results showed no significant differences in headache intensity scores between sumatriptan and placebo groups at both 2- and 4-hours post-treatment. A similar experimental design was used in the second study, in which 30 adult participants with migraine were enrolled [34]. All participants received administration of oral cilostazol and were then randomly assigned to oral sumatriptan or placebo at 6 h post-cilostazol intake or upon the onset of moderate headache. The authors found no significant difference in headache intensity scores at 2 h post-treatment, but a significant difference was observed at 4 h, suggesting a delayed effect of sumatriptan. The collective findings from these studies suggest that cilostazol-induced headache is not a straightforward model for screening new acute headache medications. However, it is plausible that these studies were not adequately powered to detect small, yet clinically meaningful differences, between sumatriptan and placebo. In support, sumatriptan had a delayed effect at 4 h post-treatment in participants with migraine [34]. This highlights the need for further research with more robust experimental designs to ascertain whether cilostazol-induced headache is useful in a proof-of-concept model screening new acute headache medications.

Limitations

Our study has some limitations that warrant some discussion. First, we limited the in-hospital observation window to 1 h due to logistic constrains. Environmental factors, such as dietary intake and stress, might therefore influence our findings. Second, we cannot exclude the impact of rescue medication use on the reported headache intensity scores. Third, we enrolled nine participants who reported using preventive headache medication. It is possible that this might influence the onset, duration, and characteristics of migraine-like headache induced by cilostazol. Fourth, our study population was predominantly female, and it cannot be excluded that the results might have been somewhat different in males. Lastly, epidemiologic data have raised intriguing questions about the relationship between migraine, head injuries, and PPTH. The available evidence suggests that head injuries are a risk factor for migraine chronification [36, 37], whilst pre-injury migraine is a risk factor for PPTH after mTBI [2, 38]. Thus, further research is needed to determine whether PPTH actually represents a migraine unmasked by trauma.