Cannabinoids modulate pain. The system determines whether that matters.

Chris Emerson, PhD Founder, Chief Scientist, and CEO, LEVEL

About the Author

Chris Emerson, PhD, holds a doctorate in small molecule chemistry and is the founder and chief scientist of LEVEL. He has conducted IRB-approved clinical research on cannabinoids in human populations, holds US patents in activated cannabinoid-controlled release formulation technology, and has published peer-reviewed research in cannabinoid science and organic chemistry. The positions taken in this guide reflect his scientific assessment of current evidence. A full scientific biography, publications, and patents are available.

SECTION I

What Pain Actually Is

What is pain, and why does that definition matter for understanding cannabinoids?

The brain does not passively receive pain. It constructs it. Ascending signals from peripheral nociceptors, sensory neurons that detect potentially damaging stimuli, travel through the spinal cord and are processed, filtered, amplified, or suppressed by a brain already shaped by prior experience, current physiological state, expectation, and context before pain is consciously experienced [1.2]. Two people with identical tissue states can experience dramatically different pain. The same person can experience dramatically different pain from the same tissue state on different days. This is not a measurement artifact or a question of pain tolerance. It is the system behaving exactly as it is designed to behave.

Pain is not a single problem. It is a category of problems with different underlying mechanisms. Any pharmacological approach that treats it as a single target will produce predictable failures in the subtypes it was not designed for.

This reframe has direct consequences for understanding what cannabinoid modulation can and cannot do. If pain were a signal with a fixed intensity traveling a fixed route to a passive receiver, the pharmacological goal would be to block that signal at the strongest available point. Classical analgesics, including opioids and NSAIDs (non-steroidal antiinflammatory drugs), operate largely within that model. Cannabinoids do not operate primarily by blocking nociceptive input. They interact with the regulatory systems that determine how nociceptive input is processed, weighted, and ultimately expressed as pain experience. That is a different pharmacological target, with different implications for who benefits, under what conditions, and at what stage of a pain problem’s development. Readers interested in the full scientific reconceptualization of pain as a constructed rather than transmitted experience will find Moseley’s foundational work in pain neuroscience a productive starting point [3].

This guide develops a single thesis: pain relief from cannabinoids is not primarily molecule-dependent. The system the molecule enters determines whether the modulation matters. The sections that follow build the mechanistic case for that claim and develop its operational consequences for matching cannabinoid intervention to pain system condition.

What is the difference among acute, subacute, and chronic pain?

Acute pain is the appropriate output of a correctly functioning system responding to actual or threatened tissue damage[4]. It is protective. In a correctly calibrated system, it is proportional to the severity of the input. It resolves as the underlying tissue state resolves. The pharmacological goal in acute pain is to reduce unnecessary suffering while the tissue heals, not to eliminate the signal entirely, because that signal carries information the system is using.

Subacute pain occupies the transition window between acute and chronic, roughly six weeks to three months following initial injury or onset, though this boundary is not universally standardized in the clinical literature and should be understood as a functional description rather than a precise threshold[4]. In this window, the pain system is at a decision point. If the underlying tissue damage resolves and the nervous system recalibrates accordingly, the pain resolves with it. If resolution does not occur, or if the nervous system fails to recalibrate despite tissue healing, the pain system begins reorganizing around the persistent input in ways that progressively change the character of the problem.

Chronic pain, defined by the International Association for the Study of Pain as pain persisting or recurring for more than three months[4], is not prolonged acute pain. This distinction is not semantic. It reflects a genuine neurophysiological difference in what is happening in the system. Chronic pain frequently involves structural and functional changes in the spinal cord and brain, altered signaling thresholds, and maintained pain states that no longer depend on the original tissue damage for their persistence[5]. Understanding this distinction is the prerequisite for understanding where cannabinoid modulation is most logically relevant and where it is not.

Why does the mechanism of pain matter more than its severity?

Severity is how pain is most commonly measured and most commonly discussed. It is also, from a pharmacological standpoint, the least useful primary variable. Two people reporting pain at seven out of ten on a standard scale may have entirely different mechanisms driving that experience, require different interventions, and respond to the same intervention in entirely different ways[2, 6].

The mechanism of pain, meaning which biological systems are generating and maintaining it, determines which pharmacological tools are relevant. This is not a novel observation in pain medicine. It is the foundational premise of neuropathic pain pharmacology, which developed precisely because clinicians recognized that the drugs working well for inflammatory and nociceptive pain were producing inconsistent and often inadequate results in patients whose pain was driven by nerve system dysfunction rather than tissue injury[6]. The question is not whether cannabinoids help with pain in general. The question is which pain mechanisms are addressable through ECS (endocannabinoid system) modulation, under what systemic conditions, and at what stage of the problem’s development. This guide builds that case from the mechanism up.

SECTION II

Two Mechanisms and a Third State

What is inflammatory pain and how does it differ from other pain types at the biological level?

Inflammatory pain begins at the site of tissue injury or immune activation. When tissue is damaged or when immune signaling identifies a threat, a cascade of chemical mediators, including prostaglandins, bradykinin, and pro-inflammatory cytokines, is released at the peripheral site[7, 8]. These mediators sensitize nearby nociceptors, lowering their activation threshold so that previously non-painful stimuli become painful and mildly painful stimuli become disproportionately intense. This peripheral sensitization is a normal and protective biological response. It creates a zone of heightened sensitivity around damaged tissue that discourages use of the injured area and supports healing.

The pharmacological tools most effective against inflammatory pain target this cascade directly. NSAIDs inhibit cyclooxygenase enzymes and reduce prostaglandin synthesis at the inflammatory site[7]. Corticosteroids suppress the broader immune signaling that drives mediator release. Opioids modulate the nociceptive signal at multiple points in the ascending pathway, reducing the perceived intensity of the input the inflamed tissue is generating. These tools work because they are targeting the correct mechanism: the peripheral inflammatory cascade and its amplified nociceptive output.

What is neuropathic pain and why does it represent a categorically different problem?

Neuropathic pain arises from a lesion or disease of the somatosensory nervous system itself [6, 9]. It is not the result of peripheral tissue damage sending an amplified signal. It is the result of the signaling infrastructure generating aberrant output in the absence of, or disproportionate to, ongoing peripheral input. The nervous system has become the primary generator of the signal.

The clinical consequences of this distinction are significant and frequently underappreciated. Neuropathic pain is commonly described as burning, electric, or shooting, and is often accompanied by allodynia, a state in which normally non-painful stimuli such as light touch produce pain, and by hyperalgesia, in which normally painful stimuli produce disproportionately severe pain[9]. These are not simply more intense versions of inflammatory pain. They reflect different neural mechanisms: ectopic discharge from damaged afferent fibers, central sensitization at the spinal cord level, and loss of descending inhibitory tone that would normally suppress excessive nociceptive signaling[5, 6].

The pharmacological consequences are equally significant. Opioids, effective against inflammatory and nociceptive pain, demonstrate limited and inconsistent efficacy against neuropathic pain across the published clinical literature. In a 2015 systematic review and meta-analysis in The Lancet Neurology examining pharmacotherapy for neuropathic pain in adults, the authors found that although opioids produced some benefit, the evidence supported them as second- or third-line treatment rather than first-line, and the magnitude of benefit was substantially lower than for conditions driven by nociceptive mechanisms[6]. First-line pharmacological options for neuropathic pain are dominated instead by agents that modulate aberrant neural signaling directly: gabapentinoids, tricyclic antidepressants, and serotonin-norepinephrine reuptake inhibitors[6]. These are not analgesics in the classical sense. They are neural stabilizers addressing a different level of the problem.

If purpose-built analgesics with decades of clinical validation cannot generalize across pain mechanisms, then any claim that a cannabinoid formulation does so must meet the same evidentiary standard. It does not yet do so uniformly.

What is central sensitization and why does it represent a third distinct state?

Central sensitization is not severe inflammatory pain. It is not severe neuropathic pain. It is a distinct neurophysiological state in which the central nervous system itself has undergone functional and sometimes structural changes that maintain pain that is no longer reliably dependent on the original peripheral input that may have triggered it[5] Woolf’s 2011 characterization of central sensitization as an amplification of neural signaling within the central nervous system that causes pain hypersensitivity remains the clinical standard for defining this state[2].

The mechanisms involved include sensitization of neurons in the spinal cord dorsal horn, which reduces the threshold for activation by ascending peripheral signals; altered function of inhibitory interneurons that would normally gate nociceptive transmission; activation of spinal glial cells that sustain a neuroinflammatory environment; and changes in the descending modulatory pathways from the brain that normally provide inhibitory tone to the spinal cord[2, 5]. Taken together, these changes mean the system is no longer responding proportionally to peripheral input. It has reorganized around a pain state that it now maintains through central mechanisms, with peripheral input serving as one of several triggers rather than as the primary driver.

Fibromyalgia is the most extensively characterized central sensitization syndrome in the clinical literature[10]. Rheumatoid arthritis is primarily an inflammatory condition, though central sensitization components develop in a significant proportion of patients with long-standing disease. Osteoarthritis, frequently described as a mechanical and inflammatory joint condition, shows central sensitization features in a clinically meaningful subset of patients, particularly those with widespread pain or pain disproportionate to radiological findings[10]. These are the three conditions enrolled in 2024 Kruger et al.’s study, in which they assessed cannabinoid compositions across all three. The mechanistic diversity of that population is not incidental. It is the context within which the study’s finding of equivalent cannabinoid outcomes across conditions must be interpreted, a point developed in full in the Evidence Review section.

Why do acute, subacute, and chronic pain represent different pharmacological opportunities for cannabinoids?

Acute nociceptive pain is a system functioning correctly. The pharmacological priority is reducing unnecessary suffering while the underlying cause resolves. Cannabinoids are not the most direct pharmacological tools for acute nociceptive pain, where interventions targeting the peripheral inflammatory cascade are more mechanistically aligned. The evidence for cannabinoid efficacy in acute nociceptive pain is substantially thinner than for chronic conditions, and this guide does not argue otherwise[11].

Subacute pain represents the window where the system is deciding whether to resolve or reorganize. This is mechanistically the most consequential period. If central sensitization mechanisms begin establishing themselves during the subacute window, the character of the pain problem changes in ways that make it progressively more difficult to address. Whether cannabinoid modulation during this window could influence the trajectory of that reorganization is a question the current evidence cannot yet answer definitively. This remains a mechanistically coherent hypothesis, not a demonstrated clinical effect, grounded in the ECS’s documented role in modulating neuroinflammatory tone and descending inhibitory signaling, both of which participate in the transition from acute to chronic pain states[12].

Chronic pain, where the system has already reorganized, is where the evidence for cannabinoid relevance is strongest and where this guide’s central argument is most fully grounded. It is also where the mechanism-dependent logic matters most acutely. Chronic inflammatory pain, chronic neuropathic pain, and central sensitization syndromes each present different mechanistic targets and different implications for which cannabinoid interactions are most relevant, a distinction developed in the Molecule-Specific Differentiation section.

SECTION III

The ECS and Pain Regulation

What role does the endocannabinoid system play in pain biology?

The endocannabinoid system is not a pain-blocking system. It is a regulatory system that modulates how nociceptive signals are processed, weighted, and ultimately expressed as pain experience across multiple levels of the nervous system simultaneously[13]. This distinction matters because it determines what cannabinoid interventions can plausibly accomplish and where they sit relative to classical analgesics in the pain pharmacology hierarchy. Opioids occupy the nociceptive relay directly. NSAIDs occupy the peripheral inflammatory cascade directly. The ECS occupies neither of those positions. It operates in parallel to both, adjusting the gain on nociceptive signaling rather than blocking transmission at a single anatomical or biochemical point[13, 14].

The system’s two primary endogenous ligands, anandamide and 2-arachidonoylglycerol, known as 2-AG, are synthesized on demand from membrane phospholipids in response to neural activity and released retrogradely, from postsynaptic to presynaptic neuron, a signaling direction opposite to most classical neurotransmitters[14]. This retrograde architecture means the ECS functions as a feedback regulator: when nociceptive signaling reaches sufficient intensity at a synapse, the postsynaptic neuron synthesizes endocannabinoids that travel backward to suppress further presynaptic release. The system is built for modulation, not transmission. It is the nervous system’s endogenous mechanism for preventing runaway amplification of nociceptive input, which aligns directly with the failure mode observed in central sensitization, where amplification is no longer adequately constrained.

Where are cannabinoid receptors located in the pain signaling pathway?

Cannabinoid receptors are distributed at multiple levels of the pain signaling architecture, and their location at each level determines the functional consequences of activation[13, 15]. CB1 receptors are expressed on peripheral sensory neurons, including small-diameter C fibers and A-delta fibers that carry nociceptive information from the periphery; on neurons throughout the spinal cord dorsal horn where ascending nociceptive signals are first processed centrally; and at supraspinal sites including the periaqueductal gray, the thalamus, and the anterior cingulate cortex, regions that participate in both the sensory and affective dimensions of pain experience[13]. This multilevel distribution means that CB1 activation does not produce analgesia through a single mechanism or pathway. It modulates nociceptive processing simultaneously at the peripheral, spinal, and supraspinal levels, with different functional consequences at each.

CB2 receptors show a distinct distribution pattern with direct relevance to the inflammatory and neuroinflammatory pain mechanisms described in Section II[15]. CB2 expression is concentrated in immune cells and peripheral tissues under baseline conditions, including in macrophages, mast cells, and peripheral sensory terminals, where activation suppresses pro-inflammatory mediator release and reduces peripheral sensitization[15]. Critically, CB2 expression is upregulated in neuroinflammatory states, including in spinal cord microglia during central sensitization, positioning CB2 as a mechanistically relevant target precisely in the conditions where chronic pain is most difficult to address pharmacologically. This upregulation under neuroinflammatory load is not incidental. It indicates that the ECS is recruited as part of the endogenous response to the conditions that drive chronic pain, a point with direct implications for why cannabinoid modulation may be most relevant in chronic rather than acute pain states.

How does the ECS interact with descending pain modulation?

The periaqueductal gray, the midbrain region that serves as a primary hub of descending pain inhibition, expresses high levels of CB1 receptors and is a site of endocannabinoid synthesis and release in response to nociceptive load[16]. Descending inhibitory pathways originating in the PAG (periaqueductal gray) and projecting through the rostral ventromedial medulla to the spinal cord dorsal horn provide tonic suppression of nociceptive signaling under normal conditions. Preclinical evidence establishes that endocannabinoid signaling in the PAG participates in stress-induced analgesia and contributes to the tonic inhibitory tone this system maintains, though human confirmation of the precise contribution of ECS signaling within this pathway remains an active area of investigation[16]. The inference that ECS dysfunction contributes to reduced descending inhibitory tone in central sensitization is mechanistically coherent and consistent with available evidence, though it remains a hypothesis rather than a demonstrated clinical mechanism.

The functional consequence of this architecture is significant for understanding why cannabinoids may be relevant in chronic pain states specifically. Central sensitization involves not only sensitization of ascending pathways but loss of descending inhibitory tone. If ECS signaling contributes to maintaining that descending tone, then states of ECS dysregulation or endocannabinoid deficiency are mechanistically consistent with the loss of inhibition that characterizes central sensitization. Conversely, cannabinoid supplementation that restores or supports descending inhibitory tone represents a mechanistically coherent target in chronic pain, distinct from the peripheral nociception-blocking mechanisms that classical analgesics pursue.

This is a hypothesis grounded in system architecture. It is not a clinical claim.

What does it mean that the ECS modulates pain gain rather than blocking pain transmission?

The distinction between modulating gain and blocking transmission is not rhetorical. It has direct pharmacological consequences that determine where cannabinoids are most and least likely to be effective. A gain-control system does not produce a fixed analgesic response regardless of input. It adjusts the sensitivity of the system to its current input, amplifying or suppressing the response depending on the context and the system’s current state[13, 14]. This means the output of cannabinoid modulation is not determined by the compound alone. It is determined by the interaction between the compound and the state of the pain system it is entering.

In a correctly functioning system with normal descending inhibitory tone and no central sensitization, the ECS is already performing its regulatory function. Introducing exogenous cannabinoids into that system produces modest effects at best, because the system’s gain is already being managed endogenously. In a system where central sensitization has elevated the gain, where descending inhibitory tone has been reduced, and where neuroinflammation is maintaining a state of aberrant amplification, the regulatory target is larger and the potential for meaningful modulation is correspondingly greater.

This is not speculation. It is the expected behavior of a gain-control system.

Direct empirical validation of this specific dynamic in human chronic pain populations remains limited, and the inference is grounded in receptor pharmacology and preclinical evidence rather than large-scale human trials.

Patients with higher baseline symptom burden may show larger absolute improvements under cannabinoid intervention, a pattern consistent with the gain-control model, though direct controlled evidence for this relationship across pain populations specifically requires further investigation.

Cannabinoids are not operating in a static system. They are entering a dynamic system already in motion, one whose response to any input is shaped by its history, its current state, and the other regulatory pressures acting on it simultaneously.

Why does cannabinoid pain modulation not inherently require intoxication?

CB1 agonism in the central nervous system is the mechanism primarily responsible for THC’s psychoactive effects[15]. It is also one of several mechanisms through which cannabinoids modulate pain signaling. The conflation of these two, the assumption that cannabinoid analgesia requires central CB1 agonism and therefore requires intoxication, is not supported by the receptor distribution and mechanistic evidence available.

Peripheral CB1 and CB2 receptors participate in pain modulation through mechanisms that do not require central nervous system penetration. CB2 suppression of peripheral inflammatory mediator release, CB1 modulation of peripheral nociceptor threshold, and the in vitro COX-2 inhibitory activity demonstrated for cannabidiolic acid represent pharmacological interactions at the peripheral level that are mechanistically separable from central psychoactivity[15, 17, 18]. Cannabidiolic acid, CBDa, is consistent with a distribution profile favoring peripheral and spinal compartments, though in vivo relevance depends on pharmacokinetic and tissue concentration factors not fully characterized[19]. The inference that this distribution profile translates to clinically meaningful peripheral pain modulation is mechanistically coherent and consistent with available evidence, though it remains a hypothesis rather than a demonstrated clinical mechanism.

The 2024 Kruger et al. study provides observational data consistent with the hypothesis that this mechanistic separation may be clinically relevant, though the study design does not establish causality or isolate mechanism[20]. A nonintoxicating formulation containing CBD and CBDa produced outcomes the researchers could not distinguish from an intoxicating formulation containing THC and CBD across three chronic pain conditions over 12 weeks. This finding does not establish which mechanisms are responsible for the observed outcomes, nor does it isolate the contribution of any individual compound within the formulation. This does not position nonintoxicating cannabinoids as universal replacements for THC across all pain system conditions. Central CB1 mechanisms remain relevant in specific pain contexts, and multi-cannabinoid formulation logic that includes sub-intoxicating THC doses represents a distinct and evidence-supported approach to those conditions. The full mechanistic case for CBDa’s peripheral activity and its distinction from neutral CBD is examined in depth in LEVEL’s canonical guide to acidic cannabinoids.

The ECS does not determine whether pain exists. It determines how loudly it is expressed.

SECTION IV

Which Cannabinoids Are Most Relevant to Which Pain Systems?

How should cannabinoids be evaluated in the context of pain, and why does molecule selection matter less than system context?

No cannabinoid produces a consistent outcome across all pain system conditions. The mechanism driving the pain, the degree of central sensitization present, the neuroinflammatory load, and the current state of descending inhibitory tone all determine which cannabinoid interactions are most mechanistically relevant and whether any modulation crosses the threshold of meaningful effect[13, 14]. Evaluating cannabinoids in isolation from those system conditions produces exactly the kind of inconsistent and sometimes contradictory findings that have historically made cannabinoid pain research difficult to interpret. The same molecule entering different pain systems produces different outputs. That is not variability in the molecule. It is the system determining the outcome.

Pain relief from cannabinoids is not primarily molecule-dependent. This is the thesis developed mechanistically in the preceding sections and that is now operationalized in this section molecule by molecule. The structure may at first appear to run against the thesis: detailed profiles of individual cannabinoids are presented in sequence, with each molecule’s distinct pharmacology examined in turn. That structure is deliberate. The system-over-molecule argument requires first establishing what each molecule does mechanistically; the closing synthesis at the section’s end returns to why those molecule profiles matter less than the system context they enter.

The molecules reviewed in this section are presented in the context of the pain system conditions they are most mechanistically aligned with, not as stand-alone efficacy claims. In each subsection the system-state condition is identified first, then the mechanistic rationale for that molecule’s relevance, then the evidence available at each tier, and finally the boundary conditions where that molecule’s mechanism is less likely to be the primary driver of outcome. This sequence is deliberate. It operationalizes the guide’s governing thesis at the level of individual molecule assessment.

What is THC’s role in pain modulation and when is central CB1 engagement most relevant?

THC produces analgesic effects primarily through the CB1 receptor agonism distributed across the pain signaling architecture, including peripheral sensory neurons, the spinal cord dorsal horn, and supraspinal regions including the periaqueductal gray and anterior cingulate cortex[15]. This multilevel CB1 engagement means THC’s analgesic mechanism is not localized to a single point in the nociceptive pathway. It modulates pain processing simultaneously at the peripheral, spinal, and supraspinal levels, including the affective and attentional dimensions of pain experience mediated by cortical and limbic CB1 populations[15, 21]. For pain system conditions where central processing is a significant driver of the pain burden, including conditions with substantial central sensitization, significant affective amplification, or impaired descending inhibitory tone, THC’s central CB1 mechanism is most mechanistically aligned with these conditions[21].

THC’s psychoactivity is a dose and context variable, not a binary property. At sub-intoxicating doses within a multi-cannabinoid formulation, CB1 engagement can be achieved at levels relevant to pain modulation without necessarily producing full intoxication[22]. This dose-dependency has direct implications for formulation design, addressed in the multi-cannabinoid subsection to follow. The relevant clinical and regulatory question is not whether THC has a role in pain modulation. The evidence for that role is substantial. The relevant question is which pain system conditions make central CB1 engagement the appropriate primary mechanism, and at what doses that engagement is clinically meaningful without producing intoxication as an unwanted effect in populations where it is problematic.

Cannabis-induced hyperalgesia represents the boundary condition for THC’s pain role and warrants explicit acknowledgment. In a subset of high-frequency, high-dose THC users, paradoxical worsening of pain has been reported, with evidence drawn from case reports and observational studies rather than large controlled trials[23]. This is not a finding that disqualifies THC from a role in pain management. It is a finding that specifies the conditions under which dose escalation becomes counterproductive, and it is mechanistically consistent with the gain-control model established in Section III. Increasing signal intensity in a dysregulated system amplifies noise rather than improving outcomes. This is consistent with receptor desensitization and downregulation dynamics documented in GPCR (G-protein coupled receptor) systems, including CB1 receptor pharmacology under sustained high-dose agonist exposure.

What is THCa’s mechanistic position in pain modulation and what does the evidence currently support?

THCa occupies a distinct and relatively undercharacterized position in pain pharmacology. Unlike THC, THCa does not produce meaningful CB1 agonism at physiologically relevant concentrations in its unheated form, meaning its mechanism of action is not dependent on the central CB1 pathway that drives THC’s psychoactivity[24]. Its documented pharmacological activity runs through a different set of molecular targets, most significantly peroxisome proliferator-activated receptor gamma, PPARgamma, a nuclear receptor with established roles in regulating inflammatory gene expression and neuroinflammatory signaling[24]. PPARgamma agonism by THCa has been demonstrated in preclinical models, where it has demonstrated anti-inflammatory and neuroprotective activity distinct from the COX-enzyme-dependent pathways targeted by NSAIDs[24]. These are preclinical findings representing a mechanistically coherent hypothesis. The translation to human pain populations has not yet been confirmed in a controlled clinical trial, and I want to acknowledge that evidentiary gap explicitly here.

What the formulation-level evidence does provide is directional signal. In 2024 Kruger et al.’s, they enrolled participants to a formulation containing 10mg THCa, 10mg CBDa, 5mg CBG, and 3mg CBC and found that this formulation was indistinguishable from a THC-containing formulation at the study’s power and design, which is consistent with, but does not establish, comparable efficacy, across three chronic pain conditions over 12 weeks, with trends toward differential effects on neuropathic pain that larger studies may clarify[20]. This finding cannot be attributed to THCa specifically because the formulation contained multiple active compounds.

Based on THCa’s documented PPARgamma agonism, its preclinical anti-inflammatory and neuroprotective activity, and observational data across a substantial patient population, my working hypothesis is that THCa may contribute to pain modulation through a pathway that is mechanistically distinct from THC, with potential overlap in outcomes under certain conditions, though comparative efficacy has not been established[24]. I would extend this further: under pain system conditions dominated by neuroinflammatory signaling, where PPARgamma-mediated modulation may be more directly aligned with the underlying mechanism than CB1 agonism, THCa’s activity may be particularly relevant. No controlled human evidence currently establishes comparative efficacy relative to THC. This is the most speculative component of the hypothesis and the one most directly dependent on the forthcoming MoreBetter investigation for empirical grounding. A dedicated real-world investigation examining the THCa, CBDa, CBG, and CBC formulation at the systems level across pain and function outcomes is currently underway on the MoreBetter platform and will provide the structured evidence base needed to evaluate this hypothesis directly. The results of that investigation will be incorporated into this guide as an exception-triggered update at the time of publication.

The system-state conditions where THCa’s mechanism is most plausibly relevant are those involving active neuroinflammatory signaling and inflammatory gene expression dysregulation, conditions present in rheumatoid arthritis, fibromyalgia with neuroinflammatory components, and the neuroinflammatory maintenance phase of central sensitization. Whether THCa’s PPARgamma mechanism provides meaningful modulation in neuropathic pain driven primarily by aberrant neural signaling rather than neuroinflammation is a more open question that the current evidence does not resolve.

What does CBD contribute to pain modulation and where are its limitations most relevant?

CBD does not produce meaningful CB1 agonism at physiologically relevant concentrations and does not engage the central nociceptive modulation pathway that THC and, through a distinct mechanism, THCa address[15]. Its contributions to pain modulation run through several CB1-independent pathways: interaction with transient receptor potential vanilloid 1 channels, TRPV1, which are expressed on peripheral nociceptors and participate in thermal and inflammatory pain signaling; modulation of anxiety and psychological stress that secondarily reduces the affective and attentional amplification of pain experience; and modest anti-inflammatory activity through mechanisms less precisely characterized than those of CBDa or THCa. CBD’s interaction with TRPV1 channels is established in vitro, though the extent to which this interaction contributes to clinical pain modulation at typical human dosing remains uncertain[25].

The honest comparative position, grounded in mechanistic evidence and extensive observational experience, is that CBD’s direct pain modulation mechanism is more limited than is commonly represented in cannabinoid literature. Its anxiolytic and stress-modulatory effects are better established and represent a genuine contribution to pain burden reduction through the co-occurring regulatory failure pathway described in Section V.

This is a secondary mechanism. It is not a primary nociceptive pathway.

CBD’s strongest mechanistic case in the pain domain is as a formulation component that addresses the anxiety, neuroinflammatory, and TRPV1-mediated dimensions of chronic pain burden rather than as a primary nociceptive modulator. This is not a finding that removes CBD from pain formulation logic. It is a finding that specifies where CBD’s contribution lives within that logic.

Explore LEVEL’s CBD Protab.

What is CBDa’s specific mechanistic advantage in inflammatory pain and why does its peripheral distribution matter?

CBDa’s mechanistic position in pain biology is more precisely defined than that of most cannabinoids, and that precision is the basis of its specific relevance rather than a limitation. CBDa demonstrates COX-2 inhibitory activity in vitro, providing an in vitro mechanism relevant to prostaglandin-driven peripheral sensitization that characterizes inflammatory pain, the clinical significance of which has not been established in human pain populations[17]. It also demonstrates 5-HT1A receptor agonism with greater potency than CBD at that receptor, with potential relevance to both pain modulation and the anxiety and stress dimensions that amplify chronic pain burden[18]. And it is consistent with a distribution profile favoring peripheral and spinal compartments, though in vivo relevance depends on pharmacokinetic and tissue concentration factors not fully characterized[19].

The combination of these three properties makes CBDa mechanistically aligned with inflammatory pain system conditions specifically. The inference from this distribution and mechanism profile to clinically meaningful peripheral pain modulation remains a hypothesis consistent with available evidence, not a demonstrated clinical effect, and it is labeled as such[19]. What the Kruger data adds is observational evidence that a formulation containing CBDa as a primary component was associated with large effect sizes across multiple symptom domains in a noncontrolled observational design, in populations that include conditions with significant inflammatory components, including rheumatoid arthritis and osteoarthritis[20]. That finding is consistent with CBDa’s mechanistic position without establishing a CBDa-specific effect. The full mechanistic case for CBDa, including its structural distinction from CBD, its in vivo stability evidence, and its complete receptor interaction profile, is examined in depth in LEVEL’s canonical guide to acidic cannabinoids.

Explore LEVEL’s CBDa Protab.

What does CBG contribute to pain modulation and what does the evidence currently support?

CBG’s mechanistic relevance to pain operates through a pathway distinct from both CB1-mediated modulation and peripheral inflammatory signaling. CBG demonstrates alpha-2 adrenoceptor agonism, engaging the noradrenergic component of the descending pain inhibitory system that is mechanistically relevant to the descending modulation architecture described in Section III, where loss of descending inhibitory tone is a feature of central sensitization and a target for pain system intervention[26]. CBG also demonstrates activity at TRPV1 channels and has shown anti-inflammatory properties in preclinical models, though the clinical relevance of these additional mechanisms in human pain populations is less established than the adrenoceptor interaction[26].

The evidence for CBG in pain is emerging rather than established. The primary mechanistic data are preclinical. The CBG RCT conducted by Emerson et al. and published in Medical Cannabis and Cannabinoids in 2026 was designed around sleep outcomes in a veteran population and was not a pain study, though the population carried significant pain burden and the study’s observational findings on pain-related sleep interference are directionally relevant[27]. No controlled human trial has yet taken place to examine CBG specifically in chronic pain populations. The mechanistic case for CBG’s relevance in pain system conditions involving impaired descending inhibitory tone is grounded in receptor pharmacology and is mechanistically coherent. It is not yet grounded in direct clinical evidence and is presented at that evidence tier.

Why does multi-cannabinoid formulation logic outperform single-molecule thinking in chronic pain?

Chronic and subacute pain are maintained by the convergence of peripheral sensitization, central sensitization, neuroinflammatory tone, impaired descending inhibition, and co-occurring regulatory failures in sleep and stress, operating simultaneously and reinforcing each other. A pharmacological approach that addresses only one of these mechanisms will reach a ceiling determined by the mechanisms it does not address[13, 14].

Multi-cannabinoid formulation logic is not a claim that combining molecules produces additive or synergistic effects in a pharmacological sense, a claim that would require controlled evidence this guide does not have. It is the recognition that different cannabinoids are mechanistically aligned with different components of the pain system condition, and that covering multiple mechanistic targets simultaneously is more consistent with the system’s complexity than optimizing for a single target. No single molecule covers all of those targets. A formulation designed around the system condition being addressed covers more of the mechanistic terrain than any single molecule can.

The 2024 Kruger et al. study provides the clearest available observational evidence for this logic. The formulation containing THCa, CBDa, CBG, and CBC was associated with large effect sizes across multiple symptom domains in a noncontrolled observational design, spanning the peripheral, central, affective, and regulatory dimensions of chronic pain burden simultaneously[20]. That breadth of outcome is more consistent with multi-mechanism coverage than with a single dominant analgesic effect. The finding that this nonintoxicating formulation produced outcomes the researchers could not distinguish from a THC-containing formulation is consistent with the hypothesis that multi-mechanism coverage through nonintoxicating pathways can produce comparable system-level effects. This remains a hypothesis. It is mechanistically grounded and observationally consistent. It is not yet confirmed in a controlled trial.

Sub-intoxicating THC doses within a multi-cannabinoid formulation represent a distinct and legitimate approach where central CB1 engagement is part of the mechanistic target, particularly where central sensitization or affective amplification is significant. The formulation logic is not maximum CB1 engagement. It is the minimum THC dose needed to engage the central mechanism, whereas the other formulation components address the peripheral, neuroinflammatory, and descending modulation dimensions simultaneously.

Explore LEVEL’s RELIEF Protab with multiple cannabinoids.

Closing Synthesis: Matching Mechanism to System State

The molecules reviewed in this section are mechanistically distinct compounds aligned with different components of a pain system that is itself mechanistically heterogeneous. THC’s central CB1 mechanism is most relevant when central processing, affective amplification, and impaired descending tone are the primary drivers. THCa’s PPARgamma mechanism is most relevant when neuroinflammatory signaling is active. CBDa’s peripheral profile is most relevant when peripheral inflammatory sensitization dominates. CBG’s adrenoceptor agonism is most relevant when descending inhibitory tone is the target. CBD’s contributions are most relevant to the anxiety, stress, and secondary inflammatory dimensions that amplify pain burden across mechanism types.

Correctly identifying the system state matters more than selecting a single molecule. Getting the system right, meaning identifying the dominant pain mechanism and the co-occurring regulatory failures, is the prerequisite to everything else. That is the thesis made operational.

SECTION V

Why System State Determines Outcome

Why does the same cannabinoid produce different results in different people with the same pain diagnosis?

The primary driver is not individual variation in the conventional pharmacological sense. It is the system state the cannabinoid is entering. Two people carrying the same diagnostic label are frequently not in the same system state. Their pain systems have different histories, different degrees of central sensitization, different neuroinflammatory loads, different descending inhibitory tone, and different co-occurring regulatory failures in sleep and stress. A cannabinoid entering one of those systems is entering a different system than the same cannabinoid entering the other, even when the diagnosis on the chart is identical. The outcome is determined by the system being entered, not by the molecule alone[2, 13, 14].

This explanation differs meaningfully from the framework that dominated early cannabis legalization discourse, which attributed variable outcomes primarily to individual differences in ECS biology, the observation that some people respond differently to different cultivars, or that personal biochemistry makes cannabinoid response inherently unpredictable. That framework was not wrong about individual variation existing, but it underspecified what that variation consists of and which components of it are most consequential for pain outcomes specifically. As the product landscape has expanded from cultivar preference into a complex market of defined cannabinoid compositions, isolates, acidic forms, and multi-compound formulations, the explanatory limits of everyone’s ECS being different have become increasingly visible. System state is not a refinement of that framework. It is a more mechanistically grounded replacement for it, one that makes variability predictable rather than merely acknowledged.

This heterogeneity is not evidence that cannabinoids are ineffective. It is the predicted output of a gain-control system operating across populations with different system states. A system whose outcome is determined by the state it enters should produce variable results across variable inputs. That variability is the model behaving correctly, not failing.

This is the thesis made operational at the individual level. It explains why cannabinoid outcomes in pain are so heterogeneous across populations, why some individuals report substantial and sustained relief while others report no meaningful effect from the same product at the same dose, and why the research literature produces findings that appear contradictory until the system-state variable is applied as a filter.

The system state is not a confounder to be controlled away. It is the primary explanatory variable.

Why do chronic pain systems resist the interventions that work for acute pain?

Chronic pain reflects a system that has undergone structural and functional reorganization in response to sustained or unresolved input, and that reorganization changes the pharmacological problem in ways not adequately captured by scaling up the interventions that work acutely[2, 5]. The spinal cord dorsal horn has sensitized. Inhibitory interneurons have lost function. Spinal glial cells have activated and are maintaining a neuroinflammatory environment that sustains aberrant signaling independent of peripheral input. Descending inhibitory tone from the periaqueductal gray and rostral ventromedial medulla has diminished[5, 16]. The system is no longer responding to nociceptive input proportionally. It has reorganized around a pain state it is now actively maintaining.

Interventions designed for acute pain are targeting a system that has already moved. An NSAID reducing prostaglandin synthesis at a peripheral site is addressing a mechanism that is no longer the primary driver of the pain. An opioid modulating ascending nociceptive signal intensity is working at a relay the system has learned to work around. The pharmacological target has shifted from the peripheral cascade to the central reorganization[6]. This is not a failure of those drugs. It is a consequence of applying tools designed for one system state to a different system state.

Cannabinoids face the same constraint. The ECS mechanisms most relevant to chronic pain are those that address the central reorganization: modulation of neuroinflammatory tone through CB2 and PPARgamma pathways, support for descending inhibitory function through PAG and noradrenergic mechanisms, and reduction of the co-occurring regulatory failures that maintain the reorganized state[13, 15, 16]. These are not the same mechanisms that would be most relevant to acute nociceptive pain. The system state is the filter through which every pharmacological claim about cannabinoids in pain must pass.

Why does dose escalation fail in a dysregulated pain system?

The intuition behind dose escalation is straightforward: if a lower dose produces partial relief, a higher dose should produce more. In a naive, correctly functioning system responding proportionally to pharmacological input, this logic sometimes holds. In a chronic pain system, this logic frequently fails, and for reasons that are mechanistically predictable rather than merely empirically observed.

A gain-control system that has been chronically overstimulated does not respond to increased input by producing proportionally increased modulation. It responds by downregulating the receptor populations through which that modulation occurs. CB1 receptor desensitization and downregulation under sustained high-dose exposure is a well-characterized pharmacological phenomenon. It is the mechanism through which cannabinoid modulation becomes less effective over time and through which cannabis-induced hyperalgesia has been reported in subsets of high-frequency, high-dose users, with evidence drawn from case reports and observational studies rather than large controlled trials[23]. The system responds to chronic high-intensity input by reducing its own sensitivity to that input, which is the opposite of the therapeutic goal.

The opioid literature provides the most extensively characterized parallel. Opioid-induced hyperalgesia, the paradoxical worsening of pain sensitivity under sustained opioid exposure, is mechanistically grounded in mu-opioid receptor desensitization and in central sensitization mechanisms that opioid escalation can actively worsen rather than relieve[28]. The ECS parallel is less extensively documented in the human pain literature but is grounded in the same receptor pharmacology logic. Sustained high-dose agonist exposure at a G-protein coupled receptor produces desensitization. That is not a speculation about cannabinoids. It is a fundamental property of the receptor class[14, 15].

Increasing signal intensity in a dysregulated system amplifies noise rather than improving outcomes.

This is consistent with receptor desensitization and downregulation dynamics documented in GPCR systems, including CB1 receptor pharmacology under sustained high-dose agonist exposure.

The therapeutic implication is not that more is always worse. It is that dose escalation in the absence of system-state assessment is pharmacologically uninformed. The relevant question before any dose adjustment is not whether more of the same molecule will produce more of the same effect. It is whether the system is currently in a state where increased modulation is the appropriate intervention, or whether the system state itself needs to be addressed first.

Why are sleep disruption, stress, and neuroinflammation inputs to the pain system rather than parallel conditions?

Sleep disruption, psychological stress, and neuroinflammatory tone are not parallel conditions. They are inputs into the pain system.

Their dysregulation actively maintains the reorganized pain state[29, 30].

Sleep disruption has a bidirectional relationship with pain that is not symmetric in effect. Pain disrupts sleep through obvious mechanisms. But sleep disruption also lowers pain thresholds, increases central sensitization, reduces descending inhibitory tone, and elevates pro-inflammatory cytokine levels, all of which sustain and amplify the chronic pain state independently of the peripheral tissue condition. [29] A chronic pain system operating on disrupted sleep is a system with a consistently elevated gain that does not return to baseline during the recovery window that sleep is supposed to provide.

Managing the pain without addressing sleep disruption treats one input. It leaves the others intact.

Psychological stress operates through overlapping neurobiological mechanisms. Stress activates the hypothalamic-pituitary-adrenal axis and the sympathetic nervous system, both of which have downstream effects on peripheral sensitization, neuroinflammatory tone, and the descending modulatory pathways that provide inhibitory control over spinal nociceptive processing[30]. Chronic stress does not simply amplify pain perception psychologically. It sustains the neurobiological conditions under which the pain system remains reorganized around an elevated-gain state.

Neuroinflammation is the mechanism that bridges these inputs to the structural reorganization of chronic pain. Activated spinal microglia, sustained by both peripheral inflammatory signaling and central stress-mediated pathways, maintain the synaptic environment of the dorsal horn in a state of heightened excitability[5]. That maintenance is not passive. Glial activation is an active biological process that sustains central sensitization after the peripheral input that originally triggered it may have resolved.

Addressing neuroinflammation is not a supplementary strategy. It is the mechanism maintaining the pain state.

The 2024 Kruger et al. study provides observational evidence consistent with this architecture. Large effect sizes were reported across multiple domains; however, these values reflect observational change and do not establish treatment effects or comparative efficacy between formulations. The breadth of these effects across pain intensity, anxiety, sleep disturbance, depression, and fatigue simultaneously is consistent with a multisystem modulatory interpretation, though the study design does not allow attribution of these outcomes to specific mechanisms or compounds[20].

Why does consistency of cannabinoid use matter more than dose intensity in chronic pain?

The ECS functions as both a tonic and phasic regulatory system. Its endogenous ligands, anandamide and 2-AG, are synthesized on demand in response to neural activity, but the system’s baseline regulatory capacity, its tonic function, is maintained by a continuous low-level endocannabinoid presence that influences receptor sensitivity, neuroinflammatory tone, and descending inhibitory function across time [14]. A system receiving intermittent high-dose input is not being consistently modulated. It is being pushed between states.

Consistent, appropriately dosed cannabinoid use supports the ECS’s tonic regulatory function rather than attempting to override it episodically. The distinction is between managing the system’s ongoing regulatory state and attempting to acutely suppress the system’s current output. The former is consistent with how the ECS is designed to operate. The latter is consistent with how classical analgesics are designed to operate, and it is less well-matched to the ECS’s modulatory architecture.

LEVEL’s real-world observational data on the Sleep Protab provides proprietary evidence consistent with this argument in the sleep domain. The finding that consistent product use was associated with greater improvement in sleep outcomes than inconsistent use, including a reduction in pain as a sleep interference factor from 11.81% pre-product to 5.53% during product use, a relative reduction of 53.2% (p less than 0.05), supports the inference that consistency of ECS modulation over time produces cumulative system-level effects that intermittent use does not. [Sleep Protab observational study, MoreBetter platform, 2024.] This finding is in the sleep domain. Its direct application to pain outcomes requires pain-specific investigation, which LEVEL’s ongoing real-world investigation, conducted on the MoreBetter platform, is designed to provide. The inference that consistency matters more than dose intensity in chronic pain is grounded in ECS tonic function pharmacology and in the sleep observational evidence, and is presented at that evidence tier.

Managing distance from threshold matters more than managing current performance.

In a chronic pain system, the threshold in question is the sensitization threshold at which the system maintains its reorganized state independently of peripheral input. Consistent cannabinoid modulation that gradually reduces neuroinflammatory tone, supports descending inhibitory function, and addresses the co-occurring regulatory failures that maintain the elevated-gain state is an intervention aimed at that threshold. Dose escalation aimed at acute symptom suppression is an intervention aimed at current performance. The former changes what the system is doing. The latter changes how loudly the system is expressing what it is already doing.

What does the timing of intervention tell us about the system’s capacity for modulation?

By the time nonlinear failure is observable, the window for low-cost intervention is often already closed.

This is not a rhetorical device. It is a description of how chronic pain develops mechanistically. Central sensitization does not announce itself. It develops through progressive changes in synaptic threshold, inhibitory interneuron function, glial activation, and descending modulation that accumulate below the level of clinical detection until the reorganization is already entrenched[2, 5].

The practical consequence is that intervention in the subacute window, before the system has completed its reorganization, is mechanistically a different problem than intervention in established chronic pain, even though the symptom presentation may not yet clearly differentiate between the two. A system reorganized for two years around central sensitization has had two years to establish the glial activation patterns, inhibitory interneuron dysfunction, and descending modulation loss that sustain its current state. The same system at six weeks has had six weeks. The intervention required to meaningfully shift the system’s regulatory state is not the same in both cases, even if the pain intensity score on intake looks similar.

This is where the system’s framing of cannabinoid intervention is most consequential and most different from the conventional analgesic framing. A classical analgesic is evaluated on whether it reduces pain intensity acutely. A cannabinoid intervention aimed at the regulatory architecture of the pain system is better evaluated on whether it shifts the system’s state over time, whether it reduces the neuroinflammatory load that maintains central sensitization, whether it supports the descending inhibitory tone that the system has lost, and whether it addresses the co-occurring sleep and stress inputs that sustain the elevated-gain state. These are different outcome questions, operating on different timescales and requiring different measurement frameworks.

In the evidence review in the following section I examine what the current literature establishes about cannabinoid effects across these dimensions, what the 2024 Kruger et al. findings add to that picture at the population level, and where the evidence remains insufficient to support confident claims.

SECTION VI

What the Evidence Shows

The observational data presented in this section report symptom outcomes observed in a 12-week study. They are presented for scientific and educational purposes only. No claim is made that cannabinoids, or any product containing them, treats, cures, prevents, or mitigates fibromyalgia, rheumatoid arthritis, osteoarthritis, or any other medical condition. Individuals with medical conditions should consult qualified health care providers regarding treatment decisions.

What does the current evidence establish about cannabinoids and chronic pain?

The evidence base for cannabinoids in chronic pain is real, uneven, and frequently misread in both directions. It is stronger than the most skeptical clinical voices acknowledge and weaker than the most enthusiastic consumer-facing narratives imply. Reading it accurately requires the analytical frame established in Section V: sleep disruption, psychological stress, and neuroinflammatory tone are not parallel comorbidities to be separated from pain outcomes. They are inputs into the pain system. Evidence of cannabinoid effects across those domains simultaneously is evidence consistent with multisystem modulation of a system whose components are mechanistically interconnected.

The National Academies of Sciences, Engineering, and Medicine concluded in their 2017 comprehensive review that there is substantial evidence that cannabis is effective for chronic pain in adults, representing the most rigorous systematic evaluation of the broader literature conducted to that point[31]. That conclusion has not been materially overturned by subsequent evidence. What subsequent research has added is greater mechanistic specificity about which pain conditions and which cannabinoid mechanisms are most relevant, and greater appreciation for the limitations of generalizing across pain types. Finnerup et al. (2015), in their systematic review in The Lancet Neurology, established that cannabinoids occupy a third-line position in neuropathic pain management, behind gabapentinoids and antidepressants, reflecting real but limited and inconsistent efficacy in that specific mechanism domain[6].

This is not a contradiction. It is mechanism specificity expressed in the evidence base.

In October 2025, a consensus statement developed through a modified Delphi process involving 23 clinical and academic experts, including Boehnke, was published in JAMA Network Open, identifying six core competencies for medical cannabis education, including describing the evidence base for health conditions commonly managed with cannabis[32]. That publication represents institutional recognition of cannabis pain management as a clinical competency domain, a signal that the medical education establishment has formally acknowledged the gap between patient need and clinician preparedness in this area.

What did Kruger et al. find and how should it be interpreted?

LEVEL and Overcome were the study sponsors and product suppliers, and Chris Emerson, PhD, is an employee, shareholder, and fiduciary officer of Metta Medical dba LEVEL. The University of Michigan IRB review reflects ethical oversight of the study design and does not constitute institutional endorsement of any findings or products.

The Kruger et al. study, presented at the International Cannabinoid Research Society Annual Symposium in 2024 and currently under peer review, is one of the first investigations whose authors systematically assess the effect of different cannabinoid compositions on pain symptoms across multiple types of chronic pain simultaneously[20]. Adult California residents with fibromyalgia (n=59), rheumatoid arthritis (n=26), and osteoarthritis of the knee or hip (n=76), all reporting substantial pain interference at enrollment as defined by a PROMIS Pain Interference score of 11 or higher, were randomly assigned to receive a 12-week supply of one of three formulations: 12.5mg CBD and 12.5mg THC delivered as a tablet (n=46); 10mg THCa, 10mg CBDa, 5mg CBG, and 3mg CBC delivered as a tablet (n=56); or 10mg CBD and 10mg CBDa delivered as a capsule (n=59).

Participants were 57 years old on average, 80% women, ranging from 18–78 years of age. Of 276 recruited participants, 162 completed all survey measures and were included in the analysis. Outcomes were assessed using 12 standard instruments including the PROMIS 29+2 at baseline and 12-week timepoints.

The following effect sizes reflect observed within-group changes in a 12-week observational study without a placebo control and cannot be interpreted as causal effects of any formulation or compound.

OBSERVED WITHIN-GROUP CHANGES: Noncontrolled Observational Design, No Placebo Arm. These values do not establish causal treatment effects or comparative efficacy between formulations or compounds.
Domain	Partial-eta2	Effect Size
Pain Intensity Interference	0.451	Large
Fibromyalgia Symptom Severity	0.325	Large
Pain Interference	0.309	Large
Anxiety	0.307	Large
Sleep Disturbance	0.288	Large
Physical Function	0.250	Large
Depression	0.221	Large
Ability to Participate in Activities	0.214	Large
Pain Catastrophizing	0.210	Large
Fatigue	0.120	Medium
Neuropathic Pain	0.105	Medium
Cognitive Function Abilities	0.046	Small
These effect sizes are derived from a noncontrolled observational design and should not be interpreted as evidence of causal treatment effects, comparative efficacy, or expected outcomes for any specific product. Effect size estimates are drawn from the ICRS 2024 poster presentation; degrees of freedom, confidence intervals, and pairwise sample sizes will be reported in the peer-reviewed manuscript currently under review. The study had no placebo arm; reported improvements cannot be attributed to pharmacological effects of the formulations alone. The study population may have experienced BBB disruption due to chronic inflammatory pain, complicating peripheral-distribution inferences. Generalization to populations differing on age, sex, or geography should be made with appropriate caution. Dosage form differed across arms, with two arms administered as tablets and one as a capsule; tablet and capsule formulations can differ in dissolution kinetics, excipient composition, and unit-dose precision, and cross-arm comparisons should account for this design feature in addition to the absence of a placebo control.

Large effect sizes were reported across multiple domains; however, these values reflect observational change and do not establish treatment effects or comparative efficacy between formulations. The large overall pattern of improvement did not show statistically distinguishable differences in magnitude across product or type of chronic pain. Because the study was not powered or designed to test between-formulation differences, this result does not support a claim of comparable efficacy; it indicates only that the study, given its size and design, could not detect differences between arms. There were trends toward differential effects on neuropathic pain, anxiety, and sleep disturbance that larger studies may clarify[20].

The breadth of reported effects across pain intensity, anxiety, sleep disturbance, depression, and fatigue simultaneously is consistent with a multisystem modulatory interpretation, though the study design does not allow attribution of these outcomes to specific mechanisms or compounds[20]. The finding that a nonintoxicating formulation produced outcomes the researchers could not distinguish from an intoxicating formulation is consistent with the hypothesis that meaningful pain burden modulation does not inherently require central CB1 engagement. This finding does not establish which mechanisms are responsible for the observed outcomes, nor does it isolate the contribution of any individual compound within the formulation.

These are not rhetorical hedges. They define what this evidence class can and cannot establish.

What does the evidence show about cannabinoids in cancer-related pain and palliative care?

The clinical domain where the evidence for cannabinoid utility is most internally consistent, and where the mechanistic argument from this guide is most directly applicable, is the management of pain and related symptoms in patients with serious illness. In this context the goal is not cure. It is reducing the burden of a dysregulated system in a body under sustained biological stress. This is the condition under which ECS modulation is most mechanistically coherent and where the nonintoxication finding from the Kruger data has the most immediate practical implication for clinical access.

In a 2022 clinical review published in JCO Oncology Practice, Worster et al. synthesized the evidence for cannabis use in cancer patients, addressing pain, nausea, anorexia, and the integration of cannabis discussions into oncology care, identifying meaningful clinical utility for cannabinoids in cancer-related symptom management while noting the limitations of available evidence and the need for clinician competency in cannabis counseling[33]. In a 2025 pilot study conducted at the University of Michigan Health System, published in Cancer Medicine, the authors examined pharmaceutical-grade cannabidiol (Epidiolex) for aromatase inhibitor-associated musculoskeletal symptoms in postmenopausal hormone receptor-positive breast cancer patients and found clinically significant improvements in pain in a subset of participants over 15 weeks of hospital-monitored treatment[34]. In this trial they used a pharmaceutical-grade CBD formulation, and its findings are not directly transferable to other cannabinoid products or formulations, but it represents the highest evidence tier currently available for hospital-monitored cannabinoid pain intervention in an oncology population.

The regulatory landscape is moving in alignment with this clinical direction. As of May 2026, four states have enacted legislation creating explicit legal frameworks for cannabis administration in hospital, hospice, or palliative care settings. California’s Ryan’s Law (SB 311), enacted in 2021 and subsequently expanded by SB 988 in 2022 and SB 302 in 2023, requires nearly all health care facilities, including hospitals, to permit terminally ill patients and patients aged 65 or older with chronic disease to use nonsmoked medical cannabis on premises. Virginia signed HB 75 and SB 332 into law in 2026, directing the Department of Health to promulgate regulations permitting hospital staff to store, dispense, and administer cannabis oil to terminally ill patients with valid certification, with implementing regulations to be in place by January 2027. Washington enacted HB 2152 in March 2026, requiring hospitals, nursing homes, and hospice care centers to permit qualifying terminal patients to use medical cannabis beginning January 2027. Oregon signed HB 4142 in April 2026, expanding the definition of debilitating medical condition to include patients requiring hospice, palliative, comfort, or comprehensive pain management care; the law applies to hospice programs, residential facilities, and palliative care settings, but does not extend to hospitals. Active legislation as of this publication is pending in New York, Connecticut, Pennsylvania, Hawaii, and several other states[35].

This does not establish clinical efficacy. It reflects regulatory alignment with clinical use patterns.

What does the evidence not yet establish?

Direct evidence for neuropathic pain remains the most significant gap. As established in Section II, neuropathic pain is mechanistically distinct from inflammatory and central sensitization pain, and the pharmacological tools that work across other pain types frequently fail here. Cannabinoids are not an exception to this pattern. The evidence for cannabinoids in neuropathic pain specifically is limited and inconsistent, and the medium effect size on the neuropathic pain domain in the Kruger study, partial-eta2 0.105 compared to 0.451 for pain intensity interference overall, is consistent with this gap rather than resolving it[6, 20].

Cannabis-induced hyperalgesia in high-frequency, high-dose THC users represents the boundary condition the guide has flagged throughout and states plainly here: escalating cannabinoid dose in a dysregulated system can paradoxically worsen pain sensitivity through CB1 receptor desensitization mechanisms.

This is not a theoretical risk. It is documented in the clinical literature and mechanistically grounded in CB1 receptor desensitization[23].

The early legalization era produced claims about cannabis and pain, and about cannabis and cancer specifically, that were not proportionate to available evidence. Cannabis was positioned in some advocacy contexts as broadly curative rather than modulatory, a framing the mechanistic and clinical evidence does not support. The more defensible and empirically grounded position is narrower: cannabinoids may provide meaningful modulation of pain burden in certain chronic and subacute conditions, with effects that are context-dependent and not uniformly observed across populations. The overclaim era’s residue is a combination of inflated patient expectations in some populations and defensive clinical skepticism in others. Both are obstacles to the measured, mechanism-informed use of cannabinoids in pain management that the current evidence supports.

The forthcoming peer-reviewed publication of the Kruger et al. manuscript will provide degrees of freedom, confidence intervals, and pairwise sample sizes that the ICRS 2024 poster does not report. LEVEL’s ongoing real-world investigation of the THCa, CBDa, CBG, and CBC formulation, conducted on the MoreBetter platform, will provide the first structured evidence base for evaluating the Emerson hypothesis on THCa’s mechanistic position in pain modulation. Both will be incorporated into this guide as exception-triggered updates upon publication.

SECTION VII

Working with the System

Where does your pain break down?

Pain breaks down at the level of the system driving it. Identifying whether that system is primarily inflammatory, centrally sensitized, or neuropathic determines which mechanisms are active and which cannabinoid interactions are relevant. Three questions can orient that application, not as a diagnostic tool, which requires clinical evaluation, but as a framework for thinking about which mechanisms are most likely active and which cannabinoid interactions are most mechanistically relevant as a result.

The first question is where the pain originates and what drives it. Pain that is localized, associated with a specific inflammatory condition, worsened by activity at the affected site, and accompanied by visible or confirmed tissue involvement such as a joint diagnosis or an inflammatory condition with laboratory markers, is more likely to be dominated by peripheral inflammatory mechanisms. The relevant pharmacological targets in this condition are the peripheral sensitization cascade, the prostaglandin and cytokine signaling at the site of inflammation, and the peripheral nociceptor threshold that those signals maintain. Cannabinoid interactions most mechanistically relevant to this condition operate at the peripheral level: CBDa’s in vitro COX-2 inhibitory activity and peripheral distribution profile, CB2 modulation of peripheral immune signaling, and formulations that concentrate activity in the peripheral and spinal compartments rather than in the central nervous system.

The second question is whether the pain has spread beyond its original site, become disproportionate to what imaging or examination findings would predict, or begun to involve hypersensitivity to stimuli that would not normally be painful. These are signals of central sensitization. A pain system reorganized around an elevated-gain state is no longer primarily driven by peripheral input. The relevant targets shift toward neuroinflammatory tone, descending inhibitory function, and the co-occurring regulatory failures in sleep and stress that maintain the reorganized state. Multi-cannabinoid formulation coverage that addresses multiple mechanistic components simultaneously is most logically aligned with this condition, and sub-intoxicating THC doses within that formulation may be appropriate where affective amplification and impaired descending tone are significant contributors.

The third question is whether the pain has features of aberrant neural signaling rather than inflammatory sensitization: burning, electric, or shooting quality; allodynia; or pain in areas without identifiable tissue involvement. These are signals of neuropathic mechanism dominance. As established in Section II and confirmed by the evidence review in Section VI, this is the condition where cannabinoid utility is most limited and most variable. It is also the condition where clinical evaluation and pharmacological guidance from a qualified practitioner is most necessary before introducing any cannabinoid intervention.

How should formulation match the mechanism driving the pain?

The goal is not to identify the most potent cannabinoid. It is to identify the mechanism driving the pain and match it accordingly. The formulation logic developed in Section IV applies directly here, and the principle is the same: match the mechanism, not the symptom. A formulation selected for symptom-suppression properties without reference to the mechanism driving those symptoms applies single-variable thinking to a multi-variable system. The system’s response will be determined by its state, not by the formulation’s potency.

For peripheral inflammatory dominant conditions, a formulation anchored by CBDa’s in vitro peripheral COX-2 activity, 5-HT1A interaction, and limited blood–brain barrier penetration is mechanistically aligned with this condition. CBDa’s peripheral distribution profile concentrates its activity at the site where the inflammatory cascade is most active without requiring central CB1 engagement. The 2024 Kruger et al. finding, in which a CBD and CBDa formulation produced outcomes the authors could not distinguish from a THC-containing formulation across three chronic pain conditions, is consistent with this mechanistic positioning and provides observational support for nonintoxicating formulation approaches in inflammatory-dominant conditions[20].

For conditions with significant central sensitization, neuroinflammatory, or mixed-mechanism components, multi-cannabinoid formulation coverage is more aligned with the system than single-compound approaches. THCa’s PPARgamma agonism addresses neuroinflammatory gene expression dysregulation. CBG’s alpha-2 adrenoceptor agonism supports descending inhibitory tone. CBDa addresses the peripheral inflammatory component. Sub-intoxicating THC doses may be appropriate where central CB1 engagement is part of the mechanistic target, particularly where affective amplification or significant impairment of descending modulation is present.

Why does how you use cannabinoids matter as much as what you use?

The ECS is a tonic regulatory system. Consistent, appropriately dosed cannabinoid use supports that tonic regulatory function over time. Intermittent high-dose use pushes the system between states without producing the sustained regulatory shift that chronic pain requires. The practical implication, established in Section V and grounded in ECS tonic function pharmacology, is that consistency of use is the variable most likely to determine whether cannabinoid modulation produces a meaningful change in the system’s regulatory state or simply modulates the expression of a state that remains fundamentally unchanged.

Improvement in the context of system modulation looks different from improvement in the context of acute symptom suppression. When cannabinoids are working with the system rather than being asked to override it, the signal is typically a gradual reduction in the background pain burden, improved sleep quality, reduced anxiety, and greater ability to participate in daily activities, the domains where the Kruger data showed their largest effect sizes[20].

These are not secondary outcomes. They are inputs into the pain system itself.

Their improvement is pain system improvement, not merely symptomatic relief alongside ongoing pain.

Dose escalation in response to plateau is rarely the correct intervention in a chronic pain system. As established in Section V, a gain-control system that has been chronically overstimulated responds to increased input by downregulating the receptor populations through which modulation occurs. The correct response to plateau is to assess the system state: whether co-occurring sleep disruption, stress load, or neuroinflammatory inputs are maintaining the system at an elevated-gain state that cannabinoid modulation alone cannot shift. Addressing those inputs is more likely to restore the system’s capacity for modulation than increasing the dose of the cannabinoid.

What cannabinoids cannot do in the pain domain states the boundary of everything this guide has argued.

Cannabinoids cannot override an acute nociceptive signal in a functioning system. They cannot reverse structural damage. They cannot substitute for pharmacological tools designed for mechanisms they do not address. They cannot reliably generalize across pain types. And they cannot produce meaningful modulation in a system too dysregulated, or in receptors too desensitized, to respond.

What they may do, under appropriate system conditions, is modulate the gain on a system that has lost its calibration. That is not a small thing. It is precisely the pharmacological target that the chronic pain system presents and that classical analgesics are not designed to reach. The system determines whether that modulation matters.

Getting the system right matters more than getting the molecule right.

References

[1] Tracey I, Mantyh PW. The cerebral signature for pain perception and its modulation. Neuron. 2007;55(3):377-391. DOI: 10.1016/j.neuron.2007.07.012.

[2] Woolf CJ. Central sensitization: implications for the diagnosis and treatment of pain. Pain. 2011;152(3 Suppl):S2-S15. DOI: 10.1016/j.pain.2010.09.030.

[3] Moseley GL. Reconceptualising pain according to modern pain science. Phys Ther Rev. 2007;12(3):169-178. DOI: 10.1179/108331907X223010.

[4] Raja SN, Carr DB, Cohen M, Finnerup NB, Flor H, Gibson S, Keefe FJ, Mogil JS, Ringkamp M, Sluka KA, Song XJ, Stevens B, Sullivan MD, Tutelman PR, Ushida T, Vader K. The revised International Association for the Study of Pain definition of pain: concepts, challenges, and compromises. Pain. 2020;161(9):1976-1982. DOI: 10.1097/j.pain.0000000000001939.

[5] Milligan ED, Watkins LR. Pathological and protective roles of glia in chronic pain. Nat Rev Neurosci. 2009;10(1):23-36. DOI: 10.1038/nrn2533.

[6] Finnerup NB, Attal N, Haroutounian S, McNicol E, Baron R, Dworkin RH, Gilron I, Haanpaa M, Hansson P, Jensen TS, Kamerman PR, Lund K, Moore A, Raja SN, Rice ASC, Rowbotham M, Sena E, Siddall P, Smith BH, Wallace M. Pharmacotherapy for neuropathic pain in adults: a systematic review and meta-analysis. Lancet Neurol. 2015;14(2):162-173. DOI: 10.1016/S1474-4422(14)70251-0.

[7] Julius D, Basbaum AI. Molecular mechanisms of nociception. Nature. 2001;413(6852):203-210. DOI: 10.1038/35093019.

[8] Woolf CJ, Salter MW. Neuronal plasticity: increasing the gain in pain. Science. 2000;288(5472):1765-1769. DOI: 10.1126/science.288.5472.1765.

[9] Jensen TS, Baron R, Haanpaa M, Kalso E, Loeser JD, Rice ASC, Treede RD. A new definition of neuropathic pain. Pain. 2011;152(10):2204-2205. DOI: 10.1016/j.pain.2011.06.017.

[10] Clauw DJ. Fibromyalgia: a clinical review. JAMA. 2014;311(15):1547-1555. DOI: 10.1001/jama.2014.3266.

[11] Stockings E, Campbell G, Hall WD, Nielsen S, Zagic D, Rahman R, Murnion B, Farrell M, Weier M, Degenhardt L. Cannabis and cannabinoids for the treatment of people with chronic non-cancer pain conditions: a systematic review and meta-analysis of controlled and observational studies. Pain. 2018;159(10):1932-1954. DOI: 10.1097/j.pain.0000000000001293.

[12] Latremoliere A, Woolf CJ. Central sensitization: a generator of pain hypersensitivity by central neural plasticity. J Pain. 2009;10(9):895-926. DOI: 10.1016/j.jpain.2009.06.012.

[13] Hohmann AG, Suplita RL. Endocannabinoid mechanisms of pain modulation. AAPS J. 2006;8(4):E693-708. DOI: 10.1208/aapsj080479.

[14] Pacher P, Batkai S, Kunos G. The endocannabinoid system as an emerging target of pharmacotherapy. Pharmacol Rev. 2006;58(3):389-462. DOI: 10.1124/pr.58.3.2.

[15] Pertwee RG, Howlett AC, Abood ME, Alexander SPH, Di Marzo V, Elphick MR, Greasley PJ, Hansen HS, Kunos G, Mackie K, Mechoulam R, Ross RA. International Union of Basic and Clinical Pharmacology. LXXIX. Cannabinoid receptors and their ligands: beyond CB1 and CB2. Pharmacol Rev. 2010;62(4):588-631. DOI: 10.1124/pr.109.002501.

[16] Hohmann AG, Suplita RL, Bolton NM, Neely MH, Fegley D, Mangieri R, Krey JF, Walker JM, Holmes PV, Crystal JD, Duranti A, Tontini A, Mor M, Tarzia G, Piomelli D. An endocannabinoid mechanism for stress-induced analgesia. Nature. 2005;435(7045):1108-1112. DOI: 10.1038/nature03658.

[17] Takeda S, Misawa K, Yamamoto I, Watanabe K. Cannabidiolic acid as a selective cyclooxygenase-2 inhibitory component in cannabis. Drug Metab Dispos. 2008;36(9):1917-1921. DOI: 10.1124/dmd.108.020909.

[18] Bolognini D, Rock EM, Cluny NL, Cascio MG, Limebeer CL, Duncan M, Stott CG, Javid FA, Parker LA, Pertwee RG. Cannabidiolic acid prevents vomiting in Suncus murinus and nausea-induced behaviour in rats by enhancing 5-HT1A receptor activation. Br J Pharmacol. 2013;168(6):1456-1470. DOI: 10.1111/bph.12043.

[19] Anderson LL, Etchart MG, Bahceci D, Golembiewski TA, Arnold JC. Cannabis constituents interact at the drug efflux pump BCRP to markedly increase plasma cannabidiolic acid concentrations. J Nat Prod. 2019;82(11):3047-3055. DOI: 10.1021/acs.jnatprod.9b00600.

[20] Kruger DJ, Dautrich T, Hall B, Keeling K, Provost K, Boehnke K. International Cannabinoid Research Society Annual Symposium 2024. University of Michigan IRB reviewed. [Manuscript under peer review.] LEVEL and Overcome were study sponsors and product suppliers. COI: Chris Emerson, PhD, is an employee, shareholder, and fiduciary officer of Metta Medical dba LEVEL.

[21] Russo EB. Cannabinoids in the management of difficult to treat pain. Ther Clin Risk Manag. 2008;4(1):245-259. DOI: 10.2147/TCRM.S1928.

[22] Wilsey B, Marcotte T, Deutsch R, Gouaux B, Sakai S, Donaghe H. Low-dose vaporized cannabis significantly improves neuropathic pain. J Pain. 2013;14(2):136-148. DOI: 10.1016/j.jpain.2012.10.009.

[23] Hirvonen J, Goodwin RS, Li CT, Terry GE, Zoghbi SS, Morse C, Pike VW, Volkow ND, Huestis MA, Innis RB. Reversible and regionally selective downregulation of brain cannabinoid CB1 receptors in chronic daily cannabis smokers. Mol Psychiatry. 2012;17(6):642-649. DOI: 10.1038/mp.2011.82.

[24] Nadal X, Del Rio C, Casano S, Palomares B, Ferreiro-Vera C, Navarrete C, Sanchez-Carnerero C, Cantarero I, Bellido ML, Meyer S, Morello G, Appendino G, Munoz E. Tetrahydrocannabinolic acid is a potent PPARgamma agonist with neuroprotective activity. Br J Pharmacol. 2017;174(23):4263-4276. DOI: 10.1111/bph.14019.

[25] Bisogno T, Hanus L, De Petrocellis L, Tchilibon S, Ponde DE, Brandi I, Moriello AS, Davis JB, Mechoulam R, Di Marzo V. Molecular targets for cannabidiol and its synthetic analogues: effect on vanilloid VR1 receptors and on the cellular uptake and enzymatic hydrolysis of anandamide. Br J Pharmacol. 2001;134(4):845-852. DOI: 10.1038/sj.bjp.0704327.

[26] Nachnani R, Raup-Konsavage WM, Vrana KE. The pharmacological case for cannabigerol. J Pharmacol Exp Ther. 2021;376(2):204-212. DOI: 10.1124/jpet.120.000340.

[27] Emerson CR, et al. Cannabigerol and Sleep: A Randomized, Double-Blind, Placebo-Controlled Trial in Veterans. Med Cannabis Cannabinoids. 2026;9:1-14. DOI: 10.1159/000549902.

[28] Chu LF, Angst MS, Clark D. Opioid-induced hyperalgesia in humans: molecular mechanisms and clinical considerations. Clin J Pain. 2008;24(6):479-496. DOI: 10.1097/AJP.0b013e31816b2f43.

[29] Finan PH, Goodin BR, Smith MT. The association of sleep and pain: an update and a path forward. J Pain. 2013;14(12):1539-1552. DOI: 10.1016/j.jpain.2013.08.007.

[30] McEwen BS. Physiology and neurobiology of stress and adaptation: central role of the brain. Physiol Rev. 2007;87(3):873-904. DOI: 10.1152/physrev.00041.2006.

[31] National Academies of Sciences, Engineering, and Medicine. The Health Effects of Cannabis and Cannabinoids: The Current State of Evidence and Recommendations for Research. Washington, DC: National Academies Press; 2017. DOI: 10.17226/24625.

[32] Zolotov Y, Boehnke KF, et al. Core competencies for medical cannabis education: a modified Delphi consensus statement. JAMA Netw Open. 2025;8(10):e2535049. DOI: 10.1001/jamanetworkopen.2025.35049.

[33] Worster B, Hajjar ER, Handley N. Cannabis use in patients with cancer: a clinical review. JCO Oncol Pract. 2022;18(11):743-749. DOI: 10.1200/OP.22.00080.

[34] Fleege NMG, Miller EA, Kidwell KM, Zacharias ZR, Houtman J, Scheu K, Kemmer K, Boehnke KF, Henry NL. Pilot study of cannabidiol for treatment of aromatase inhibitor-associated musculoskeletal symptoms in breast cancer. Cancer Med. 2025;14(15):e71117. DOI: 10.1002/cam4.71117.

[35] California Senate Bill 311 (Compassionate Access to Medical Cannabis Act, “Ryan’s Law”). Chapter 384, Statutes of 2021. Subsequently amended by SB 988 (2022) and SB 302 (2023). Virginia House Bill 75 / Senate Bill 332 (2026 session, signed into law 2026). Washington House Bill 2152 (signed March 11, 2026; effective June 11, 2026; implementation January 1, 2027). Oregon House Bill 4142 (signed April 2026; effective January 1, 2027). State legislative landscape current as of May 2026 and subject to ongoing change.

LEVEL Cannabinoids and Pain Guide | v1.4 | May 2026 | Phase 5 Complete