Common Mechanisms in Migraine, Migraine-Related Neck Pain, and Low Back Pain: Implications for Treatment Authors: Dilara Onan, Michalis Kodounis, Lanfranco Pellesi, Marta Waliszewska-Prosół, Paolo Martelletti Published: 2026 ABSTRACT AbstractAimChronic pain conditions, such as migraine, migraine-related neck pain (NP), and low back pain (LBP), are central pain processing disorders that frequently coexist. This narrative review aims to synthesize central sensitization (CS), pain spread and pharmacological and rehabilitation-based interventions for these three conditions and their common underlying mechanisms.MethodsA comprehensive literature search was conducted in PubMed and Web of Science to identify clinical and neurophysiological studies investigating CS, nociceptive processing, pharmacological and non-pharmacological approaches for migraine, migraine-related NP, and LBP.ResultsCurrent evidence suggests that migraine, migraine-related NP, and LBP share cortical hyperexcitability, impaired descending inhibitory control and altered spinal and supraspinal nociceptive processing. Persistent nociceptive input can contribute to multi-site pain phenotypes by facilitating the spread of pain beyond the area of ​​origin. Motor dysfunction, psychological factors, such as fear-avoidance and decreased quality of life, contribute to the maintenance of CS. Pharmacological treatments for migraine-related NP follow migraine treatment principles, although an additional cervical musculoskeletal defect may require independent, non pharmacological management. Physical therapy and rehabilitation approaches can regulate central pain mechanisms and improve functional outcomes in all three conditions. However, the strength of evidence differed for specific interventions; in migraine, the evidence for exercise-based rehabilitation and calcitonin-gene-related-peptide-targeted therapies appeared stronger than for manual therapy approaches.ConclusionsMigraine, migraine-related NP, and LBP can be better understood within the framework of CS and shared pain mechanisms. A mechanism-based, integrative assessment and treatment approach may improve clinical outcomes in these patients.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1007/s40122-026-00871-6.Keywords: Central sensitization, Migraine, Neck pain, Low back pain, Physiotherapy, Pharmacology, Nociplastic pain Key Summary Points Why carry out this study?Migraine, migraine-related neck pain (NP), and low back pain (LBP) frequently occur together, causing significant disability and economic burden globally. However, these conditions are often treated as separate, region-specific diseases, and common pain mechanisms are not adequately addressed in clinical practice.Central-peripheral sensitization, impaired descending pain modulation, and widespread pain spread are thought to play a significant role in the frequent co-occurrence of these three pain conditions.This narrative review aimed to evaluate common neurophysiological mechanisms and mechanism-based treatment approaches in migraine, and migraine-related NP, and LBP.What was learned from the study?The evidence reviewed indicates that migraine, migraine-related NP, and LBP share common features such as cortical hyperexcitability, impaired descending inhibitory control and altered spinal and supraspinal nociceptive processing. Pharmacological and rehabilitation interventions acting on the central nervous system (e.g., exercise, manual therapy) can improve pain, disability and functional outcomes, but the application of the evidence varies depending on the phenotype and intervention.These findings support a mechanism-based, phenotype-focused assessment and treatment model rather than a region-specific approach for patients with overlapping migraine, migraine-related NP, and LBP, and highlight that treatments validated for one condition (e.g., anti-calcitonin gene-related peptide agents in migraine) should not be generalized to pain phenotypes not associated with that condition.Given that the current evidence is largely cross-sectional, these results suggest that comprehensive studies, including those testing whether treating one condition (e.g., migraine) improves outcomes in a concurrent area (e.g., LBP), are needed to clarify causality and refine multidisciplinary, centrally targeted treatment strategies. Introduction Migraine, neck pain (NP), and low back pain (LBP) are chronic pain conditions with high prevalence in the worldwide population, leading to loss of productivity, and impaired quality of life [1–3]. Migraine is one of the leading causes of disability globally, and recent Global Burden of Disease data indicate that the individual and societal burden continues to increase [4]. A large population-based meta-analysis involving 41,614 adults from 17 countries demonstrated that headache disorders remain a significant public health issue associated with substantial functional limitations, decreased productivity and impaired quality of life across diverse populations [5]. In addition, NP represents a highly prevalent musculoskeletal disorder that further contributes to the global disability burden [6]. Similarly, LBP ranks among the leading causes of years spent with disability and is associated with absenteeism from work and long-term functional impairment [7]. The disability burden of migraine, NP, and LBP suggests that these conditions may represent overlapping chronic pain phenotypes with shared biopsychosocial and neurophysiological mechanisms. The clinical features of migraine have a multidimensional framework, including complaints during attacks, sensory sensitivity, autonomic symptoms, cognitive impairment, and functional limitations [1]. Similarly, NP and LBP are influenced by biopsychosocial factors and central nervous system-derived mechanisms [8–11]. In recent years, the description of chronic pain has shifted from models focused on peripheral tissue damage to mechanism-based frameworks emphasizing central sensitization (CS) [12]. CS is characterized by increased excitability of nociceptive neurons, impaired pain inhibition mechanisms, and perception of pain responses beyond normal physiological limits [12, 13]. In migraine, cortical hyperexcitability, trigeminovascular activation, and impaired descending inhibitory control have been demonstrated [14, 15]. Similarly, changes in central pain processing, reduced pain thresholds, widespread tenderness, allodynia, maladaptive motor control patterns, and multi-site pain involvement have been reported in NP and LBP [16, 17]. The relationship between migraine and NP, hereafter referred to as migraine-related NP, is noteworthy in terms of the trigeminocervical complex (TCC) [18]. In addition to pain in the cervical region, musculoskeletal impairments are frequently observed in individuals with migraine [19]. Furthermore, LBP is commonly reported in patients with migraine, and persistent nociceptive input has been suggested to contribute to widespread pain sensitivity extending beyond the original symptomatic region [20]. Pain syndromes arising in anatomically different regions may be explained by shared neurophysiological mechanisms [12, 13]. Clinically, understanding these common mechanisms is important because assessment and treatment approaches focusing solely on local tissues can be insufficient for many chronic diseases. Previous reviews have addressed CS in isolation for a single pain region, focusing, for example, only on its role in LBP [21–25], NP [26–28] and migraine [29–31]. However, no previous review has integrated migraine, migraine-related NP, and LBP within a single framework. This distinction is important because patients often experience more than one of these clinical conditions simultaneously. A region-specific approach may miss a common, centrally directed pain phenotype. This narrative review addresses this gap by (1) synthesizing neuroanatomical pathways that link these three conditions through the TCC and shared descending modulation pathways, (2) proposing a mechanism-based framework applicable to all three conditions, not just a single body region, and (3) presenting a comparative synthesis of centrally acting therapeutic evidence across migraine, NP, and LBP to guide treatment choice when these conditions occur together. Methods Search Strategy A literature search was conducted in PubMed and the Web of Science databases (from inception to 20 May 2026) using the keywords provided in Supplementary Material 1 to identify studies on common mechanisms. A second set of searches combined this with relevant keywords on mechanism-based therapeutic approaches. Only articles in English were considered. Since this study is a narrative review, the Scale for the Assessment of Narrative Review Articles (SANRA) was used for its quality assessment [32] (Supplementary Material 2). The formal Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA)-based procedure was not applied. The following eligibility criteria were used as inclusion criteria to guide study selection: (1) original research articles, systematic reviews, meta-analyses relating to migraine, NP, LBP, central sensitization, or nociplastic pain; (2) studies published in peer-reviewed journals; (3) studies providing epidemiological or evaluative or treatment data relating to common central pain processes discussed in this review; (4) articles in English. Case reports, conference abstracts without full text, and non-English publications were excluded. Reference lists of articles obtained from the searches were selected according to the final reference list, relevance to the topic and study types, with priority given to articles with detailed information and citations and more recent publications where multiple studies addressed the same point. This study was conducted using a narrative review design, therefore ethical approval was not required. Central and Peripheral Mechanisms According to the International Association for the Study of Pain, CS is defined as an increased responsiveness of nociceptive neurons in the central nervous system to their normal or subthreshold afferent input [33]. At the molecular level, this process is driven by the removal of the magnesium block from N-methyl-D-aspartate (NMDA) receptors, following sustained postsynaptic depolarization. This facilitates calcium influx and activates intracellular signaling pathways that contribute to the “unmasking” of silent synapses [34]. This chain triggers a state of long-term potentiation (LTP) in spinal dorsal horn neurons, lowering the threshold for action potential generation and expanding receptive fields. Consequently, the nervous system remains in a maladaptive state of high reactivity, where the physiological, protective role of pain is replaced by a self-sustaining pathological condition [35]. The clinical hallmarks of CS are allodynia and hyperalgesia [36]. Allodynia, the perception of pain from non-noxious stimuli (such as light touch), results from functional “cross-talk” between non-nociceptive Aβ mechanoreceptors and pain-signaling pathways within the dorsal horn. This is often facilitated by disinhibition, which allows harmless sensory input to be misinterpreted as pain [14]. This state of hyperexcitability is further exacerbated by the phenotypic switch of Aβ fibers, which may begin to express substance P, directly contributing to nociceptive transmission. Hyperalgesia involves an exaggerated response to painful stimuli. Primary hyperalgesia occurs at the site of injury owing to peripheral sensitization, whereas CS is characterized by “secondary hyperalgesia,” i.e. the expansion of pain sensitivity into uninjured, anatomically distant tissues [14]. The phenomenon is driven by the expansion of neuronal receptive fields and the "wind-up" mechanism, where repetitive nociceptive stimulation leads to a progressive increase in neuronal firing frequency. This temporal summation is mediated by the release of substance P and calcitonin gene-related peptide (CGRP), which enhance synaptic efficacy and facilitate the recruitment of adjacent neurons [14, 37]. Peripheral processes also play a role in the initiation and maintenance of CS. Peripheral sensitization, resulting from post-translational modification of nociceptor ion channels and receptors, is regulated by intracellular signaling cascades involving mitogen-activated protein kinases (MAPK; ERK, p38 and JNK) that are activated in both peripheral nociceptors and central glial cells and drive transcriptional changes that maintain hyperexcitability [38]. Gap junctions, particularly connexin-43 (Cx43) hemichannels expressed by astrocytes and satellite glial cells, can further exacerbate this sensitized state by facilitating ATP and glutamate release and enhancing neuron-glia cross-interaction both peripherally (dorsal root ganglion) and centrally (spinal dorsal horn) [38]. Genetic and epigenetic factors, such as DNA methylation, histone modifications, and noncoding RNA regulation in sensory neurons and pain-processing brain regions, can further modulate individual predisposition to the development and maintenance of chronic pain states [39]. Finally, autonomic nervous system dysregulation, reflecting an imbalance between sympathetic and parasympathetic activity, has been demonstrated in both migraine, where a complex sympathetic-parasympathetic imbalance pattern underlies cranial autonomic symptoms [40], and in chronic LBP, where decreased parasympathetic tone and sympathetic dominance are observed in heart rate variability measurements [41]. These peripheral, molecular, genetic, and autonomic processes may interact bidirectionally with central mechanisms and should be considered alongside, rather than in place of, the central processes that are the primary focus of this review. Spreading Effect and Development The phenomenon where migraine, NP, and LBP are not usually confined to a single body region but coexist and influence each other is called the “pain spread” or “widespread pain sensitization,” suggesting, according to the International Association for the Study of Pain, that persistent pain in one region can increase sensitivity in other parts of the body even without local tissue damage [42–44]. This is consistent with the clinical structure of chronic overlapping pain conditions (COPCs), recognized by the National Institutes of Health, which frequently co-occur and are thought to share nociplastic mechanisms, including migraine and chronic LBP [45]. In particular, direct evidence of a common etiology between NP and LBP has been previously reported. A genetic study indicated that NP and LBP share a common latent factor of “back pain” with genetic variances in addition to condition-specific environmental contributions [46]. In chronic overlapping pain conditions, a polygenic risk score study indicated that both NP/shoulder pain and LBP risk scores were significantly associated with a central pain score reflecting central sensitization burden [47]. Clinically, co-occurring NP and LBP were associated with a greater impairment in cervical sensorimotor control compared with NP alone [48], suggesting that the simultaneous involvement of multiple spinal regions may reflect central sensitization. A shared mechanistic framework for pain conditions may involve neuroanatomical convergence and neuroimmune activation. At the brainstem level, this is exemplified by the TCC in migraine and NP with nociceptive input from the cervical spine stimulating the trigeminovascular system, and migraine-associated cortical spread depression potentially sensitizing cervical segments [49]. This brainstem-level convergence is anatomically specific to craniocervical structures and does not, by itself, account for the co-occurrence of LBP. The latter is more plausibly linked to the systemic, nonsegmental mechanisms outlined below (neuroimmune activation and descending modulation), which are not anatomically restricted and could in principle extend beyond the head and neck [21]. This framework is consolidated by two pathogenetic factors: neuroinflammation and dysfunctional descending inhibition. In neuroinflammation, astrocyte activation triggers a complex neuroimmune response that releases pro-inflammatory cytokines, including tumor necrosis factor-α (TNF-α) and interleukins (IL-1β and IL-6), which directly modulate synaptic plasticity [50]. These inflammatory mediators enhance excitatory synaptic transmission while simultaneously suppressing inhibitory control [13]. Consequently, this neuroinflammatory environment creates a self-perpetuating cycle that reinforces the persistent hypersensitivity characteristic of migraine and multi-site pain syndromes [38, 51]. In migraine, NP, and LBP, the endogenous pain-control systems fail to provide adequate serotonergic and noradrenergic descending inhibition [52]. This “top-down” dysregulation creates a strong pro-nociceptive environment, leading to an imbalance where descending facilitation outweighs inhibition. This imbalance facilitates the spread of pain to different and anatomically distant body regions [53]. Clinically, the impairment is reflected in reduced conditioned pain modulation efficiency, a hallmark of the transition from localized to widespread pain [54]. This may explain why individuals with migraine also frequently report NP or shoulder pain, and why patients with chronic spinal pain experience headaches [55, 56]. More broadly, functional imaging studies have described overlapping brain regions-including thalamic, insular and cingulate structures-involved in processing and modulation of pain across different body sites, which has been proposed as indirect support for shared central mechanisms across pain conditions [57, 58]. However, these studies were not designed to directly compare headache, NP, and LBP within the same cohort, and this specific tripartite overlap remains to be formally demonstrated. Individuals experiencing pain in multiple areas of the body are thought to have lower pain thresholds and greater disability, suggesting a common central process. Importantly, the development of multicentric pain is not entirely biological [59]. This biological vulnerability is often amplified by biopsychosocial factors, creating a bidirectional link between emotional distress and pain sensitivity [60]. Pain catastrophizing and fear-avoidance behaviors act as potent psychological stressors that trigger the hypothalamic–pituitary–adrenal (HPA) axis, leading to a sustained release of cortisol and pro-inflammatory mediators. This neuroendocrine activation increases systemic excitability and lowers the nociceptive threshold, effectively “priming” the central nervous system for an exaggerated response to even minor stimuli [61]. The proposed migraine-LBP relationship here may reflect two different models. In a sequential model, the sustained nociceptive stimulus from migraine progressively sensitizes larger central networks, potentially causing pain to spread to the lumbal region over time [38, 54]. In a constitutional model, migraine and LBP represent independent but parallel comorbid conditions with a shared genetic and psychological association against pain amplification, without a causal link between them [62]. Since the current evidence is cross-sectional, it is not possible to distinguish which model may be primary. Clinical Implications The occurrence of cutaneous allodynia and cortical stimulation of the somatosensory cortex, insula, and anterior cingulate increases pain perception and emotional weight of pain. Clinically, this mechanism has been associated with increased pain, functional limitations, and decreased quality of life in patients with migraine [30]. Spinal segments innervating body regions with reduced pain thresholds may show corresponding signs of neural hyperexcitability, consistent with segmental CS [20, 63]. With the chronicity of migraine, increased excitability in central pain pathways can lead to a predisposition to widespread pain in different areas of the body. Therefore, provided that any pathology, such as disc herniation or other spinal diseases in the spinal regions are ruled out (because these conditions have their own disease-pain mechanisms), the co-occurrence of NP and LBP in patients with migraine is a common example of a widespread pain phenotype associated with CS [33, 64]. This widespread sensitivity has been associated with decreased pain thresholds, multisite pain involvement, allodynia, kinesiophobia, and activity restriction as a cognitive-emotional consequence and changes in motor control in the muscles [20]. NP frequently accompanies headache in patients with migraine and is explained by the TCC. Nociceptive input originating from cervical structures can trigger or worsen migraine, as there is a decreased pain threshold in the cervical regions owing to impaired pain processing [49, 65, 66]. Cervical range of motions, pressure pain threshold, and forward head posture are important assessments, and these tests can differentiate patients from healthy individuals [67]. Individuals with migraine and accompanying NP exhibit more pronounced musculoskeletal and sensory impairments, characterized by a higher frequency of headaches, greater disability, more pronounced CS, decreased cervical range of motion, decreased cervical flexor and extensor muscle endurance, more myofascial trigger points, and lower pain thresholds, as well as a greater psychological burden, compared with both healthy controls and migraine sufferers without NP [19, 68–70]. Similarly, CS has been identified as a significant mechanism in chronic LBP. In population-based studies, the prevalence of LBP has been reported to be higher in individuals with chronic and episodic headaches compared with individuals without headaches. Furthermore, the association of chronic headache with LBP has been related to increased pericranial muscle sensitivity [20]. Similarly, subtypes of chronic headaches were strongly associated with frequent LBP, and this risk is higher than in episodic headaches [20]. In chronic pain conditions, the fear-avoidance model is another common mechanism. This model was initially developed by Vlaeyen and Linton for chronic musculoskeletal pain [71]. Particularly in chronic LBP, fear-avoidance beliefs have been shown to increase pain and disability through avoidance of physical activity [72]. Similarly, in NP, kinesiophobia and fear-avoidance beliefs are significant predictors of decreased range of motion, reduced functionality, and long-term disability [73]. In the case of migraine, this model indicates that the fear for physical activity triggering migraine attacks and avoidance behaviors develops, which increases the disease burden [74]. However, the extension and adaptation of the model to migraine is more recent, and its applicability is being actively investigated rather than definitively established, given that fear-avoidance behavior in migraine encompasses a wider range of perceived triggers than just physical activity [75]. From a biopsychosocial perspective, recent studies have shown that body image changes in individuals experiencing chronic pain and can be reliably assessed using validated self-report tools [69, 76–78]. Studies on how body regions are perceived in cases of headache, NP, and LBP have revealed that impaired body awareness and proprioceptive changes are associated with pain intensity and disability [77–79]. Similarly, psychological factors play a significant role in chronic pain conditions. Anxiety, depression, and catastrophizing exacerbate the pain experience and negatively affect functional recovery [80, 81]. The concordance between objective measurements and patient-reported results suggests that headache, NP, and LBP may share a common central pain mechanism characterized by widespread hypersensitivity and altered pain processing. Accordingly, considering strategies targeting central pain processes in assessment and treatment approaches may be important. The central mechanisms and pain spread shared between migraine, migraine-related NP, and LBP are presented in Fig. 1. Shared central mechanisms and pain spread in migraine, migraine-related neck pain and low back pain. TCC trigeminocervical complex, MAPK Mitogen-Activated Protein Kinase, Cx43 Connexin-43, HPA Hypothalamic–pituitary–adrenal. The solid line shows the TCC, an anatomically specific pathway linking pains and not inherently extending to LBP. The dotted lines show systemic, non-segmental mechanisms (neuroinflammation, impaired descending inhibition and impaired conditioned pain modulation) linking all three pain states via the central nociplastic sensitization center. The long dashed line extending from the center to the clinical outcomes shows how these central mechanisms manifest as clinically listed outcomes. The dashed shaded line extending from the modulating factors to the round shape shows amplifications. The bidirectional dashed line between modulating factors and clinical outcomes reflects the bidirectional link between emotional distress and pain sensitivity as described in the text. Peripheral sensitization processes, which interact bidirectionally with central sensitization, were also represented. Figure 1 was created with Microsoft PowerPoint Phenotypes: Clinical Triage The most important step in managing patients with concomitant migraine, NP, and LBP is to identify the dominant pain mechanism. The relative contribution of peripheral and central processes can differ significantly among individuals. A mechanism-based classification, particularly in the presence of LBP and chronic musculoskeletal pain, provides a clinically useful framework for guiding treatment decisions [22, 82]. According to this classification, three pain phenotypes can be distinguished: nociceptive, neuropathic, and nociplastic pain [22, 82]. Nociceptive pain results from actual or threatened damage to non-nerve tissue and is usually proportional to identifiable mechanical or inflammatory pathology. Neuropathic pain results from a lesion or disease affecting the somatosensory nervous system and is characterized by neuroanatomically plausible sensory symptoms. Nociplastic pain, which needs to be addressed for the current discussion, is defined by altered nociceptive processing without apparent tissue or nerve damage and is often associated with widespread or multi-site pain, hypersensitivity, impaired conditioned pain modulation, sleep disturbances, anxiety, depression, and kinesiophobia [74, 76, 77, 82, 83]. The clinical features and the therapeutic targets associated with each phenotype are presented in Table 1. Table 1The three pain phenotypes and the treatments each one responds toPhenotypeClinical fingerprintDominant treatmentsNociceptive(local tissue-driven)Localized pain proportionate to identifiable mechanical or inflammatory cause that resolves with healingPressure-pain thresholds are normal at distant sitesNSAIDsActivity modificationTargeted manual therapyNeuropathic(nerve lesion-driven)Burning, electric and/or lancinating quality, with a dermatomal distributionPositive neurological signs, sensory deficits and/or hyperalgesia in a neuroanatomically plausible territoryGabapentinoids in selected patientsSNRIs/TCAsCareful neuro-orthopedic work-upNociplastic / CS-dominant(centrally-driven)Widespread or multi-site pain disproportionate to tissue findings, allodynia, lowered pressure-pain thresholds at distant sites, kinesiophobia, catastrophizing, sleep and mood disturbancesTCAs, SNRIs, anti-CGRP therapies when the nociplastic phenotype occurs in the context of migraine*Aerobic/high-intensity exerciseManual therapy aimed at restoring central processingPain neuroscience educationSleep and psychological co-managementCS central sensitization, CGRP calcitonin gene-related peptide, NSAID nonsteroidal anti-inflammatory drug, SNRI serotonin and norepinephrine reuptake inhibitors, TCA tricyclic antidepressants*Anti-CGRP therapies are supported by migraine-specific evidence and should not be generalized to nociplastic pain phenotypes not associated with migraine (e.g., non-migraine LBP) The three pain phenotypes and the treatments each one responds to Nociceptive (local tissue-driven) Localized pain proportionate to identifiable mechanical or inflammatory cause that resolves with healing Pressure-pain thresholds are normal at distant sites NSAIDs Activity modification Targeted manual therapy Neuropathic (nerve lesion-driven) Burning, electric and/or lancinating quality, with a dermatomal distribution Positive neurological signs, sensory deficits and/or hyperalgesia in a neuroanatomically plausible territory Gabapentinoids in selected patients SNRIs/TCAs Careful neuro-orthopedic work-up Nociplastic / CS-dominant (centrally-driven) TCAs, SNRIs, anti-CGRP therapies when the nociplastic phenotype occurs in the context of migraine* Aerobic/high-intensity exercise Manual therapy aimed at restoring central processing Pain neuroscience education Sleep and psychological co-management CS central sensitization, CGRP calcitonin gene-related peptide, NSAID nonsteroidal anti-inflammatory drug, SNRI serotonin and norepinephrine reuptake inhibitors, TCA tricyclic antidepressants *Anti-CGRP therapies are supported by migraine-specific evidence and should not be generalized to nociplastic pain phenotypes not associated with migraine (e.g., non-migraine LBP) In clinical practice, these phenotypes often overlap. For example, a patient with chronic LBP may present with an acute nociceptive exacerbation superimposed on a nociplastic background, while a patient with neuropathic pain associated with migraine may simultaneously require treatment specific to acute migraine and interventions targeting CS. Treatment of Nociceptive Pain In nociceptive LBP, nonsteroidal anti-inflammatory drugs (NSAIDs) are considered first-line pharmacological treatments, while paracetamol has shown limited efficacy [84, 85]. In resistant cases, short-term opioid therapy may be considered cautiously, but routine use is not recommended [86]. Migraine-associated NP may be a prodromal symptom of migraine or a consequence of TCC sensitivity [28]. In patients in whom NP is associated with an additional cervical musculoskeletal component, targeted manual therapy combined with graded cervical exercise may be beneficial [87–94]. A challenge is to differentiate nociceptive pain from CS-dominant pain presentations. Patients with nociplastic pain often exhibit symptoms disproportionate to imaging findings or tissue pathology and hypersensitivity extending beyond the painful area [22, 82]. Treatment of Neuropathic Pain Neuropathic pain in the lower back is often associated with radiculopathy or nerve damage. Gabapentinoids may provide benefit in selected patients, but evidence is inconsistent and does not support routine use [86, 95]. Serotonin-norepinephrine reuptake inhibitors (SNRIs), particularly duloxetine and venlafaxine, and tricyclic antidepressants such as amitriptyline, may be beneficial owing to their effects on reducing inhibitory pain pathways [83, 96–102]. Migraine itself is not considered a neuropathic disorder, and migraine-associated NP rarely exhibits true neuropathic features. When neuropathic mechanisms are present, treatment should be based on the principles of neuropathic pain [83, 96–102]. Treatment of Nociplastic Pain In this phenotype, mechanism-based management becomes more important than site-specific treatment. Current evidence suggests that pharmacological, manual therapy, exercise-based and educational approaches can all contribute to central pain processing modulation [26, 87–89, 103]. Centrally Acting Pharmacological Treatments Among tricyclic antidepressants, amitriptyline is used in centrally mediated and neuropathic pain conditions. It enhances descending inhibitory pathways via monoaminergic modulation and also affects sodium channels and NMDA receptors [100]. Amitriptyline is effective in migraine prophylaxis [97–99], but its long-term analgesic effects in LBP appear modest [100]. Among SNRIs, duloxetine has the strongest evidence in LBP. Randomized controlled trials have shown a reduction in pain intensity and improvement in function [101]. Duloxetine may also be beneficial in patients with migraine with concomitant chronic pain or CS [97–99]. Venlafaxine is effective in migraine prophylaxis and may be particularly beneficial in neuropathic pain mechanisms [97–99, 102]. Migraine-specific preventive agents targeting the CGRP pathway reduce trigeminovascular activation and neurogenic inflammation [103–105]. Monoclonal antibodies targeting CGRP are effective preventive options in migraine, while gabapentin may be used in neuropathic pain management through modulation of neuronal excitability [103–106]. Topiramate, beta-blockers, and other preventive options, including candesartan, remain appropriate in cases where migraine is the predominant clinical condition [103, 107–109]. Onabotulinumtoxin A is effective in chronic migraine through inhibition of neurotransmitter and neuropeptide release at peripheral trigeminal nerve endings [103]. A randomized placebo-controlled crossover trial reported that vaporized cannabis containing tetrahydrocannabinol and cannabidiol was superior to placebo for the acute treatment of migraine [110]; however, this finding comes from a single trial with a limited sample and needs to be replicated independently before it can guide routine practice. In addition to this trial, further evidence comes from retrospective cohort and survey-based studies reporting a reduction in migraine frequency and associated symptoms with medical cannabinoids [111–113]. Currently, these studies lack blinding and placebo control and are therefore susceptible to significant reporting and selection bias. Recently, the limited evidence regarding LBP has been addressed. A large phase 3, multicenter, randomized, placebo-controlled trial evaluated a chemically standardized, full-spectrum Cannabis sativa extract (VER-01) in chronic LBP (n = 820) and reported improvements in physical function, sleep quality and quality of life, along with a significantly greater reduction in mean Numeric Rating Scale pain intensity (−1.9 versus −1.4 points; mean difference = −0.6, 95% CI −0.9 to −0.3; P < 0.001) compared with placebo at the primary endpoint. VER-01 had a favorable long-term safety profile without signs of dependence or withdrawal [114]. This study provides the first adequately potent, well-defined results for a cannabis-based product in chronic LBP. However, for this to be accepted in standard practice, independent replication and regulatory approval pathways need to be clarified [86, 95]. Gabapentin initially showed promise in migraine prophylaxis, but subsequent analyses have shown inconsistent findings [115]. Riboflavin and magnesium are adjunctive preventive options in migraine. Randomized controlled trials have shown that 400 mg of riboflavin per day is effective [116, 117], whereas magnesium may reduce neuronal excitability and modulate NMDA receptor activity despite heterogeneous evidence [118]. The clinical implications of these treatments are presented in Table 2. Table 2Centrally-acting agents and their evidence across migraine and low back painAgentCentral mechanismMigraine evidenceLBP evidenceAmitriptyline5-HT/NA reuptake inhibition with sodium-channel and NMDA-receptor effects → enhanced descending inhibitionEffective in prophylaxis [97]Modest long-term analgesic benefit [100]Duloxetine5-HT/NA reuptake inhibition → strengthens spinal descending inhibitionBenefits patients with comorbid chronic pain or CS [99]Strongest antidepressant evidence in chronic LBP [119]Venlafaxine5-HT/NA reuptake inhibition (NA effect at higher doses)Effective in prophylaxis [98]Limited, with a potential in neuropathic-leaning phenotypes [98]Anti-CGRP monoclonal antibodiesBlock CGRP ligand or receptor → suppress trigeminovascular driveReduce frequency and severity of attacks [103]No direct LBP indicationGabapentinα2δ-subunit calcium-channel modulation → reduced excitatory transmitter releaseInconsistent evidence [115]Routine use not supportedCannabisEndocannabinoid modulation of central pain pathwaysSuperior to placebo in acute migraine in a single trial [110]A Phase 3 RCT (n = 820) demonstrated a significant reduction in pain intensity, improvement in function/sleep/quality of life and positive long-term safety outcomes [114]5-HT serotonin, CGRP calcitonin gene-related peptide, CS central sensitization, LBP low back pain, NA noradrenaline, NMDA N-methyl-D-aspartate Centrally-acting agents and their evidence across migraine and low back pain 5-HT serotonin, CGRP calcitonin gene-related peptide, CS central sensitization, LBP low back pain, NA noradrenaline, NMDA N-methyl-D-aspartate Manual Therapy Manual therapy can affect local biomechanical structures and central pain mechanisms. Stimulation of cervical myofascial Aδ- and C-fibers can activate descending inhibitory mechanisms, including diffuse noxious inhibitory controls, and thus reduce neuronal excitability associated with CS [92, 93]. In migraine, preliminary findings from small controlled and observational studies suggest that upper cervical manual therapy can affect CS markers and TCC processing [91–93]. These findings should be evaluated with caution given the limited sample sizes and heterogeneous designs of the underlying studies. In LBP, manual therapy approaches have shown improvements in pain intensity, pain thresholds, disability, and pain catastrophizing, but effect sizes are small to moderate [120, 121]. However, evidence regarding manual therapy in both migraine [122, 123] and LBP [124, 125] remains inconsistent, and the need for more rigorous randomized controlled trials is highlighted. A systematic review of manual therapies for migraine revealed that although massage, physiotherapy, and chiropractic spinal manipulation appeared similar to pharmacological prophylaxis in some studies, the included studies had significant methodological limitations, preventing definitive conclusions [122]. Another systematic review and meta-analysis reported that its effectiveness had not yet been proven owing to insufficient sample size, high risk of bias, and methodological and clinical heterogeneity of populations, interventions, and comparators [123]. Similarly, in LBP, comprehensive systematic review and meta-analysis showed that spinal manipulation therapy provided a small and moderate-reliability benefit to other treatments (placebo-independent) in terms of pain and disability in the short term, but evidence for the long term is insufficient [124, 125] and this effect diminished over time [125]. In NP, the evidence appears more consistent regarding improvements [126]. Manual therapy may regulate central pain processing in some patients, but specific treatment effects remain unclear, and expectations regarding its effectiveness should be limited accordingly. Exercise In patients with migraine, rehabilitation approaches including exercise and manual therapy may improve quality of life while reducing headache frequency, pain intensity, disability, and medication overuse [87–89]. Motor control and cervical stabilization exercises may affect pain modulation mechanisms and improve conditioned pain modulation in migraine-related NP [94]. A systematic review with meta-analysis of controlled trials indicates that exercise may also improve CS-related findings in chronic musculoskeletal pain populations [26]. Proposed mechanisms include activation of endogenous pain modulation systems, increased β-endorphin levels, modulation of the endocannabinoid system and exercise-induced increases in brain-derived neurotrophic factor (BDNF) [127, 128]. In addition, exercise can improve sleep, mood, inflammation, autonomic regulation, and vascular function [127]. In LBP, systematic reviews and meta-analyses show that therapeutic exercise can reduce pain intensity and improve functional capacity [129, 130]. Aerobic and high-intensity exercise programs can reduce symptoms associated with CS [129, 130]. Pain Neuroscience Training Pain neuroscience training aims to explain pain as a product of nervous system processing and may help reduce catastrophizing and fear-related behaviors. When combined with manual therapy in LBP, pain neuroscience training has demonstrated improvements in pain catastrophizing and pain thresholds [131]. In addition, awareness of altered body and associated pain regions, increasingly recognized in chronic pain conditions, can be assessed using measurement tools [79–81]. Addressing Treatment in Coexisting Phenotypes When migraine, migraine-related NP, and LBP coexist in the same patient, treatment decisions may consider possibilities together. The mechanism-based classification in Table 1 can be applied to determine whether a nociceptive, neuropathic or nociplastic phenotype is dominant overall. When a nociplastic/CS dominant phenotype is common in two or more regions, centrally acting interventions that can modulate pain processing simultaneously in multiple regions may be prioritized over region-specific treatments because they address a common underlying mechanism, not a single anatomical region. Region- or condition-specific treatments (anti-CGRP therapies for migraine attacks, targeted manual therapy for a cervical problem, NSAIDs for acute nociceptive LBP exacerbations) may then be added on top of this central strategy, addressing the remaining regional features. This sequence, prioritizing systemic/central treatment management followed by targeted regional treatment management, may make the framework proposed in this review functional. Future Perspectives The increasing evidence regarding CS as a common mechanism in migraine-related NP and LBP highlights the need for a paradigm shift from region-specific approaches to integrated, mechanism-based models. This narrative review does not include a risk-bias assessment or a Grading of Recommendations Assessment, Development, and Evaluation (GRADE) evidence evaluation. Because the interventions considered are of different types, the quality of evidence varies significantly. Future systematic reviews with formal grading are needed to reinforce these findings. First, future studies should prioritize comprehensive study designs in terms of pain spread to better capture the dynamic and progressive nature of CS in patients with migraine. Current evidence is largely cross-sectional and limits causal interpretation. Studies incorporating clinical and neurophysiological measurements are needed to clarify how localized pain transforms into multi-zone pain syndromes and to identify early determinants of pain spread [131]. Second, there is a clear need for the development and validation of objective biomarkers of CS in patients with migraine. Patient-reported outcomes, such as the Central Sensitization Inventory [132, 133], are widely used, but emerging approaches, including electrophysiological, neuroimaging, and neurochemical methods (e.g., GABAergic activity) may provide clearer insights into changes in central pain processing and improve patient classification [64]. Thirdly, future research should focus on treatment paradigms targeting multiple areas simultaneously. Interventions addressing pain, sleep disturbances, psychological factors, and cognitive factors may yield clearer outcomes in centrally mediated pain states [132]. Finally, there is a need to address multi-zone pain conditions, such as migraine and LBP, as distinct clinical phenotypes. The frequent co-occurrence of these conditions suggests that they may be different clinical manifestations of a common central dysregulation and CS process. Current studies mostly examine these pain conditions in isolation and do not adequately reflect real-life patient profiles. Future cross-sectional, experimental and randomized controlled trials may evaluate combined pain conditions such as migraine and concomitant LBP within the framework of CS. Furthermore, investigating the impact of treating one condition on the clinical outcomes of the other may provide stronger evidence regarding common underlying mechanisms. For example, it has not yet been tested whether anti-CGRP therapies reduce accompanying LBP beyond their effects on headaches, or whether physical therapy targeting TCC sensitivity improves outcomes simultaneously in migraine, neuropathic pain and LBP. Defining patient subgroups based on CS profiles, psychological characteristics and neurophysiological markers can contribute to the development of more targeted interventions, particularly centrally effective multidisciplinary and education-based approaches. Conclusions This review suggests that migraine, migraine-related NP, and LBP may form a complex pain phenotype interconnected by common central mechanisms. We highlight the importance of holistic assessment and treatment strategies targeting central pain processing mechanisms rather than regional approaches. In the future, developing mechanism-based and personalized interventions for this multi-regional pain structure may lead to more effective improvement outcomes. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (PDF 205 KB)