Abstract
Background
Chemotherapy-induced peripheral neuropathy (CIPN) is a common complication of neurotoxic chemotherapy, yet effective pharmacologic options for both prevention and treatment of established painful CIPN remain limited. Non-delta-9-tetrahydrocannabinol (THC) cannabinoids and related minor cannabinoids show antinociceptive effects in preclinical CIPN models, but their real-world clinical relevance remains unclear.
Methods
We performed a retrospective secondary analysis of de-identified electronic health record (EHR) data using the TriNetX Research Network to examine whether documented cannabidiol/cannabigerol (CBD/CBG) exposure was associated with subsequent neuropathy-related diagnoses among adults receiving neurotoxic chemotherapy. Patients with neuropathy-related diagnoses on or before chemotherapy were excluded. Two Cox proportional hazards analyses were performed. The post-chemotherapy exposure analysis compared post-chemotherapy CBD/CBG exposure with post-chemotherapy exposure to gabapentin, pregabalin, or duloxetine. The pre-chemotherapy exposure analysis compared patients with documented CBD/CBG exposure within one year before chemotherapy to patients who had no documented CBD/CBG exposure during that pre-chemotherapy window. The outcome was a composite neuropathy-related diagnosis 30–365 days after the index event.
Results
In the adjusted post-chemotherapy exposure model, 83 CBD/CBG-exposed patients were compared with 118,001 active-comparator patients. Post-chemotherapy CBD/CBG exposure was associated with lower hazard of subsequent neuropathy-related diagnosis-code outcomes (Hazard ratio [HR] = 0.162, 95% confidence interval [CI] = 0.041–0.649, p = 0.010). In the adjusted pre-chemotherapy exposure model, 162 CBD/CBG-exposed patients were compared with 633,329 controls. Pre-chemotherapy CBD/CBG exposure was not significantly associated with neuropathy-related diagnosis-code outcomes (HR = 0.859, 95% CI = 0.476–1.551, p = 0.614).
Conclusions
Post-chemotherapy CBD/CBG exposure was associated with lower observed hazard of neuropathy-related diagnosis-code outcomes, whereas pre-chemotherapy exposure was not significant. These exploratory findings require cautious interpretation because of sparse exposed-arm events, EHR exposure limitations, and potential competing-risk bias.
Keywords: cannabidiol, cannabigerol, cannabinoids, chemotherapy-induced peripheral neuropathy, real-world evidence, TriNetX
Introduction
Chemotherapy-induced peripheral neuropathy (CIPN) is a common and clinically significant adverse effect of several widely used anticancer agents, including platinum compounds, taxanes, vinca alkaloids, and proteasome inhibitors (1–3). The prevalence of CIPN varies depending on the type of chemotherapy, with platinum derivatives such as cisplatin and oxaliplatin showing the highest rates, ranging from 70% to 100% (4). CIPN commonly presents with numbness, tingling, paresthesia, dysesthesia, mechanical hypersensitivity, and neuropathic pain, which may persist long after chemotherapy has ended (1, 5). In addition to reducing quality of life, CIPN can interfere with cancer care by necessitating chemotherapy dose reduction or discontinuation, thereby creating a difficult balance between optimal cancer treatment and long-term neurologic toxicity (1). Because cancer survivorship continues to improve, identifying effective strategies to prevent or treat CIPN remains an important unmet clinical need.
Despite the clinical burden of CIPN, available pharmacological options remain limited. No medication is currently recommended as an established preventive therapy, and treatment options for established painful CIPN are modest (6). The American Society of Clinical Oncology guideline identifies duloxetine as the only agent with appropriate evidence supporting its use for established painful CIPN, while also noting that the magnitude of benefit is limited (6). Other medications commonly used for neuropathic pain, including gabapentin and pregabalin, are frequently used empirically in clinical practice but have not shown sufficient evidence to serve as established CIPN-directed therapies (1, 6).
Cannabinoid-based therapies have been used clinically or evaluated for several pain-related conditions, including chronic and neuropathic pain, cancer pain, and multiple sclerosis-associated symptoms, but their analgesic effects are generally modest and evidence for CIPN remains limited and inconclusive (7–9). In addition, delta-9-tetrahydrocannabinol (THC)-containing or THC-dominant approaches remain limited by the psychoactive and euphorigenic effects of THC, as well as acute cognitive and psychomotor effects involving attention, memory, perception, sedation, dizziness, and motor performance (10). Clinical translation is further complicated by product heterogeneity, regulatory barriers, and limited CIPN-specific efficacy data. For example, a small randomized crossover pilot trial of nabiximols, a 1:1 THC:cannabidiol (CBD) oromucosal extract, did not significantly reduce CIPN compared with placebo despite responder signals in a subset of participants (11–14). Notably, the cannabis plant produces numerous phytocannabinoids beyond THC (15). These include non-euphorigenic or minimally euphorigenic compounds often referred to as minor cannabinoids, which in rodent models of CIPN have shown antinociceptive efficacy (16–24). This supports the possibility that non-THC cannabinoids may engage pain-modulatory mechanisms relevant to chemotherapy-associated neuropathy.
Real-world electronic health record datasets provide an opportunity to identify clinical signals that may inform preclinical and translational research priorities. Such analyses cannot establish causality and are limited by coding accuracy, medication capture, confounding by indication, and incomplete information about dose, formulation, and adherence. However, when used carefully, real-world datasets can help determine whether a therapeutic class shows enough clinical signal to justify mechanistic testing under controlled experimental conditions. In the present study, we used the TriNetX Research Network to examine whether documented CBD/cannabigerol (CBG)-related EHR exposure was associated with reduced neuropathy-related diagnoses among adults receiving neurotoxic chemotherapy. We evaluated two complementary questions: first, whether CBD/CBG-related exposure after chemotherapy was associated with reduced subsequent neuropathy-related diagnoses compared with traditionally used neuropathic-pain medications; and second, whether cannabinoid exposure before chemotherapy was associated with reduced later neuropathy-related diagnoses. This approach was designed to determine whether real-world clinical data support continued preclinical investigation of minor, non-THC, cannabinoids as candidate therapeutics for CIPN.
Methods
Data source
This retrospective cohort study was conducted using the TriNetX Research Network, a federated electronic health record database that provides access to de-identified patient-level clinical data from participating healthcare organizations. TriNetX includes structured data on demographics, diagnoses, procedures, medications, laboratory values, and other clinical encounters. At the time of analysis, the Research Network included 115 healthcare organizations (HCOs). The updated Cox proportional hazards analyses were generated in TriNetX on August 24, 2026. No calendar-year restriction was applied. The analyses used all available structured EHR data in the TriNetX Research Network that met the cohort definitions at the time the queries were performed. All analyses were performed within the TriNetX analytics platform using de-identified data. Because the data were de-identified and no direct patient interaction occurred, this study did not constitute human subjects research requiring informed consent.
Study design
Two complementary retrospective cohort analyses were performed to evaluate whether documented CBD/CBG-related EHR exposure was associated with reduced neuropathy-related diagnoses after neurotoxic chemotherapy. The first analysis evaluated a post-chemotherapy exposure design, comparing patients with CBD/CBG exposure after chemotherapy with patients treated with commonly used neuropathic-pain medications after chemotherapy (Figure 1A). The second analysis evaluated a pre-chemotherapy exposure design, comparing patients with CBD/CBG exposure before chemotherapy with patients without documented pre-chemotherapy CBD/CBG exposure (Figure 1B). Both analyses used Cox proportional hazards models to estimate adjusted hazard ratios for subsequent neuropathy-related diagnoses.
Figure 1.
Cohort construction for post-chemotherapy exposure and pre-chemotherapy exposure analyses. Adults receiving neurotoxic chemotherapy were identified in the TriNetX Research Network. Patients with neuropathy-related diagnoses on or before chemotherapy were excluded. (A) In the post-chemotherapy exposure analysis, patients with CBD/CBG exposure within 6 months after chemotherapy were compared with patients receiving gabapentin, pregabalin, or duloxetine within the same post-chemotherapy window. (B) In the pre-chemotherapy exposure analysis, patients with CBD/CBG exposure within 1 year before chemotherapy were compared with patients without documented CBD/CBG exposure in that window. Neuropathy-related outcomes were assessed 30–365 days after the index event using adjusted Cox proportional hazards models.
CBD and CBG were analyzed together because both are non-THC cannabinoids with preclinical relevance to chemotherapy-induced neuropathy and because structured EHR capture of individual CBD-only and CBG-only exposure was too sparse to support stable separate primary models. Therefore, the exposure should be interpreted as documented CBD/CBG-related EHR exposure rather than exposure to a single commercial product, dose, or formulation. Structured EHR medication concepts do not reliably capture whether the documented exposure reflected isolated CBD, isolated CBG, a CBD-rich or CBG-containing product, or a broader cannabinoid formulation. Accordingly, throughout the manuscript, CBD/CBG exposure refers to structured EHR documentation of CBD or CBG concepts and should not be interpreted as confirmation that patients used isolated CBD-only or CBG-only products.
Study population
Eligible patients were adults aged 18 years or older with documented exposure to at least one neurotoxic chemotherapy agent. Neurotoxic chemotherapy exposure was defined using medication and procedure concepts for platinum compounds, taxanes, vinca alkaloids, and proteasome inhibitors. The final chemotherapy concept set included structured medication and procedure concepts for platinum compounds, taxanes, vinca alkaloids, and proteasome inhibitors, specifically cisplatin, oxaliplatin, carboplatin, paclitaxel, protein-bound paclitaxel, docetaxel, vincristine, and bortezomib. Medication and chemotherapy administration concepts were identified using RxNorm, a normalized clinical drug terminology for medications, and Healthcare Common Procedure Coding System (HCPCS) terms available in TriNetX, as listed in Supplementary Table A. These terms were used consistently across both analyses.
To reduce capture of pre-existing neuropathy, patients were excluded if they had a neuropathy-related diagnosis on or before the qualifying chemotherapy exposure. Baseline neuropathy exclusion terms included drug-induced polyneuropathy, paresthesia of skin, other disturbances of skin sensation, unspecified disturbances of skin sensation, polyneuropathy unspecified, and neuralgia/neuritis unspecified. Codes used for exclusion and outcome definitions are listed in Supplementary Table B.
Post-chemotherapy cannabinoid analysis
For the post-chemotherapy exposure analysis, patients were required to have neurotoxic chemotherapy and no neuropathy-related diagnosis on or before chemotherapy. The CBD/CBG-exposed cohort included patients with documented CBD or CBG exposure within 6 months on or after neurotoxic chemotherapy. To create a mutually exclusive cannabinoid cohort, patients in this group were excluded if they had documented gabapentin, pregabalin, or duloxetine exposure in the same post-chemotherapy treatment window. Exposure and comparator concepts are listed in Supplementary Tables C and D.
The active-comparator cohort included patients with documented gabapentin, pregabalin, or duloxetine exposure within 6 months on or after neurotoxic chemotherapy. Patients in the active-comparator cohort were excluded if they had documented CBD or CBG exposure. The purpose of this design was to compare post-chemotherapy cannabinoid exposure with commonly used neuropathic-pain medications rather than with an untreated control group, consistent with active-comparator approaches used to reduce confounding by indication in observational pharmacoepidemiology (25–27). The descriptive characteristics export included 83 patients in the CBD/CBG cohort and 120,807 patients in the active-comparator cohort. After applying Cox model index-date and time-at-risk requirements, 83 and 118,001 patients, respectively, were included in the Cox proportional hazards analysis.
For the Cox proportional hazards model, both cohorts were indexed at initiation of the first qualifying post-chemotherapy treatment. In the cannabinoid cohort, this was the first documented CBD/CBG exposure after neurotoxic chemotherapy. In the active-comparator cohort, this was the first documented gabapentin, pregabalin, or duloxetine exposure after neurotoxic chemotherapy. Outcomes were assessed 30 to 365 days after the cohort-specific treatment index date.
Pre-chemotherapy cannabinoid exposure analysis
For the pre-chemotherapy exposure analysis, patients were required to have neurotoxic chemotherapy and no neuropathy-related diagnosis on or before chemotherapy. The exposed cohort included patients with documented CBD or CBG exposure within 1 year on or before neurotoxic chemotherapy. The control cohort included patients with neurotoxic chemotherapy and no documented CBD or CBG exposure within 1 year on or before chemotherapy. Gabapentin, pregabalin, and duloxetine were not exclusion criteria in this analysis but were handled as covariates as described below. The pre-chemotherapy analysis included 162 patients in the CBD/CBG cohort and 662,672 patients in the control cohort at query definition, with 633,329 control patients entering the Cox model after index-date and time-at-risk requirements.
For the pre-chemotherapy analysis, the index event was defined as the first qualifying neurotoxic chemotherapy exposure. Outcomes were assessed 30 to 365 days after chemotherapy.
Outcome definition
The primary outcome was a composite neuropathy-related diagnosis occurring after the index event. The same outcome definition was used for both analyses. The composite included International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes for drug-induced polyneuropathy, polyneuropathy unspecified, neuralgia and neuritis unspecified, paresthesia of skin, other disturbances of skin sensation, and unspecified disturbances of skin sensation. These corresponded to G62.0, G62.9, M79.2, R20.2, R20.8, and R20.9 (Supplementary Table B). The outcome window was defined as 30 to 365 days after the index event. A 30-day lag was used to reduce capture of diagnoses recorded at the same visit or during the immediate peri-index period and to better identify neuropathy-related diagnoses emerging after chemotherapy or after treatment exposure.
Covariates
Covariates were selected based on clinical relevance to neuropathy risk, treatment selection, and cannabinoid prescribing. Covariates were measured before the index event, ending 1 day before the index date. The covariate window did not include events occurring on the same day as the index event to reduce capture of diagnoses or medications documented as part of the index encounter.
Both Cox models included age at index, sex, diabetes mellitus, seizure-related diagnoses, metastatic cancer, and opioid analgesic exposure. Diabetes was included because it is an established risk factor for peripheral neuropathy and could influence subsequent neuropathy diagnosis. Seizure-related diagnoses were included because prescription CBD is commonly used in seizure disorders, creating potential confounding by indication. Metastatic cancer diagnoses were included as markers of cancer burden and disease severity. Opioid analgesic exposure was included as a marker of baseline pain burden and symptom complexity. Covariate concepts and associated coding are listed in Supplementary Table E.
For the pre-chemotherapy analysis, gabapentin, pregabalin, and duloxetine were additionally included as covariates because pre-index use of these medications may indicate baseline pain, neuropathic symptoms, mood disorders, or other comorbidities that could influence subsequent neuropathy-related diagnosis. For the post-chemotherapy active-comparator analysis, gabapentin, pregabalin, and duloxetine were not included as covariates in the final model because they were used to define the comparator cohort and were therefore exposure-defining rather than independent adjustment variables.
Statistical analysis
Cox proportional hazards regression models were used to estimate adjusted hazard ratios and 95% confidence intervals for neuropathy-related diagnoses. Cohort membership was included as the primary independent variable. Hazard ratios less than 1 indicated lower hazard of neuropathy-related diagnosis in the CBD/CBG cohort relative to the comparator cohort, whereas hazard ratios greater than 1 indicated higher hazard. Models were run within the TriNetX analytics environment using de-identified, aggregated real-world data. Statistical significance was set at p < 0.05.
Because the CBD/CBG-exposed cohorts were substantially smaller than the comparator cohorts, estimates were interpreted with attention to confidence interval width and cohort size. Furthermore, small exposure cohorts may generate less precise estimates. Therefore, analyses were interpreted as exploratory and hypothesis-generating rather than causal.
Descriptive outcomes queries were also performed to summarize the number and proportion of patients with neuropathy-related diagnosis-code outcomes in each cohort using the same composite outcome definition and 30- to 365-day post-index window used for the Cox proportional hazards models. These descriptive event counts were used to contextualize model precision and sparse-event concerns.
As a descriptive post hoc competing-risk check requested during review, death was assessed within 0 to 365 days after the post-chemotherapy exposure index in the CBD/CBG and active-comparator cohorts. This analysis was descriptive only and was not modeled as a formal competing-risk regression. Therefore, mortality findings were used to contextualize interpretation of neuropathy-related diagnosis-code outcomes rather than to generate adjusted competing-risk estimates.
This study was reported in accordance with the STROBE guideline for observational studies. The completed checklist is provided as Additional file 1.
Results
Cohort identification and participant characteristics
The post-chemotherapy exposure analysis compared post-chemotherapy CBD/CBG exposure with active-comparator treatment using gabapentin, pregabalin, or duloxetine. The pre-chemotherapy exposure analysis compared patients with documented CBD/CBG exposure within one year before chemotherapy to patients who had no documented CBD/CBG exposure during that pre-chemotherapy window.
In the post-chemotherapy exposure analysis, the CBD/CBG characteristics cohort included 83 patients from 24 HCOs, while the active-comparator characteristics cohort included 120,807 patients from 112 HCOs. The CBD/CBG cohort had a mean age of 66 years (SD = 16), and 55.42% were female. The active-comparator cohort had a mean age of 65 years (SD = 14), and 55.39% were female.
In the pre-chemotherapy exposure analysis, the CBD/CBG cohort included 162 patients from 26 HCOs, while the control cohort included 662,672 patients from 114 HCOs. The pre-chemotherapy CBD/CBG cohort had a mean age of 66 years (SD = 16), and 59.26% were female. The pre-chemotherapy control cohort had a mean age of 67 years (SD = 15), and 54.90% were female. Race and ethnicity distributions are summarized in Table 1.
Table 1.
Demographic characteristics of post-chemotherapy exposure and pre-chemotherapy exposure cohorts.
| Characteristic | Post-chemo CBD/CBG | Post-chemo comparator | Pre-chemo CBD/CBG | Pre-chemo control |
|---|---|---|---|---|
| Total patients in characteristics export | 83 | 120,807 | 162 | 662,672 |
| HCOs | 24 | 112 | 26 | 114 |
| Mean age, years ± SD | 66 ± 16 | 65 ± 14 | 66 ± 16 | 67 ± 15 |
| Age range | 19–90 | 18–90 | 18–90 | 18–90 |
| Female | 55.42% | 55.39% | 59.26% | 54.90% |
| Male | 44.58% | 44.56% | 40.74% | 45.05% |
| Unknown sex | 0.00% | 0.05% | 0.00% | 0.05% |
| Not Hispanic or Latino | 62.65% | 67.95% | 66.04% | 57.07% |
| Hispanic or Latino | 12.04% | 5.17% | 6.17% | 4.52% |
| Unknown ethnicity | 34.93% | 26.34% | 33.33% | 38.41% |
| White | 69.87% | 61.63% | 67.28% | 56.12% |
| Black or African American | 12.04% | 11.26% | 12.34% | 8.70% |
| Asian | 12.04% | 8.90% | 6.17% | 9.19% |
| American Indian or Alaska Native | 0.00% | 0.44% | 6.17% | 0.33% |
| Native Hawaiian or other Pacific Islander | 0.00% | 0.48% | 0.00% | 0.54% |
| Other race | 12.04% | 4.77% | 6.17% | 8.19% |
| Unknown race | 16.86% | 12.52% | 15.43% | 16.93% |
Descriptive cohort counts reflect TriNetX characteristics exports, whereas Cox model counts reflect patients entering the Cox proportional hazards model after index-event restrictions. Race and ethnicity were extracted from structured electronic health record demographic fields and are reported descriptively to characterize cohort composition. These variables were not interpreted as biological determinants of treatment response.
Post-chemotherapy CBD/CBG exposure was associated with lower observed hazard of subsequent neuropathy-related diagnosis-code outcomes
Before applying the active-comparator medication exclusion, 173 patients had documented post-chemotherapy CBD/CBG exposure. Of these, 90 patients also had documented gabapentin, pregabalin, or duloxetine exposure in the same post-chemotherapy treatment window and were excluded to create the mutually exclusive CBD/CBG cohort. The post-chemotherapy exposure Cox model included 83 CBD/CBG-exposed patients and 118,001 active-comparator patients. In the adjusted Cox proportional hazards model, CBD/CBG cohort membership was associated with a significantly lower hazard of subsequent neuropathy-related diagnosis compared with the active-comparator cohort: HR = 0.162, 95% CI = 0.041–0.649, p = 0.010 (Figure 2). In descriptive outcomes queries, neuropathy-related diagnosis events occurred in 18,290 of 120,067 active-comparator patients. The exact number of neuropathy-related diagnosis events in the post-chemotherapy CBD/CBG cohort could not be displayed because TriNetX suppresses detailed results when outcome counts are between 1 and 9 patients. In a descriptive post hoc mortality comparison, death within 365 days after the post-chemotherapy exposure index occurred in 38 of 83 (45.8%) patients in the CBD/CBG cohort and 33,476 of 116,696 (28.7%) patients in the active-comparator cohort, corresponding to a higher observed 365-day mortality proportion in the CBD/CBG cohort than in the active-comparator cohort (risk ratio = 1.596, 95% CI = 1.263–2.017, p = 0.0006). Together, these results indicate a lower observed hazard of neuropathy-related diagnosis-code outcomes in the post-chemotherapy CBD/CBG cohort. However, this estimate should be interpreted cautiously because the exposed-arm event count was very small and mortality was more frequent in the CBD/CBG cohort.
Figure 2.
Adjusted hazard ratios for neuropathy-related diagnoses associated with CBD/CBG exposure. The post-chemotherapy exposure analysis compared post-chemotherapy CBD/CBG exposure with active-comparator treatment using gabapentin, pregabalin, or duloxetine. The pre-chemotherapy exposure analysis compared patients with documented CBD/CBG exposure within 1 year before chemotherapy with patients who had no documented CBD/CBG exposure during that pre-chemotherapy window. Hazard ratios less than 1 indicate lower hazard of subsequent neuropathy-related diagnosis in the CBD/CBG cohort. CBD, cannabidiol; CBG, cannabigerol; HR, hazard ratio; CI, confidence interval.
In the post-chemotherapy exposure model, age at index, type 1 diabetes, type 2 diabetes, opioid analgesic exposure, and secondary/unspecified malignant neoplasm of lymph nodes were significantly associated with increased hazard of neuropathy-related diagnosis-code outcomes, whereas male sex and secondary malignant neoplasm of respiratory and digestive organs were significantly associated with lower hazard (Table 2).
Table 2.
Adjusted Cox proportional hazards model for the post-chemotherapy exposure analysis.
| Covariate | Hazard ratio | 95% CI | p value |
|---|---|---|---|
| CBD/CBG exposure vs. active comparator | 0.162 | 0.041–0.649 | 0.0101 |
| Male sex | 0.837 | 0.813–0.863 | <0.0001 |
| Age at index | 1.002 | 1.001–1.003 | <0.0001 |
| Type 1 diabetes mellitus | 1.145 | 1.020–1.285 | 0.0222 |
| Type 2 diabetes mellitus | 1.105 | 1.065–1.147 | <0.0001 |
| Other specified diabetes mellitus | 0.965 | 0.857–1.087 | 0.5582 |
| Epilepsy and recurrent seizures | 0.918 | 0.806–1.045 | 0.1937 |
| Convulsions, not elsewhere classified | 1.086 | 0.973–1.213 | 0.1414 |
| Secondary/unspecified malignant neoplasm of lymph nodes | 1.169 | 1.130–1.209 | <0.0001 |
| Secondary malignant neoplasm of respiratory/digestive organs | 0.956 | 0.918–0.995 | 0.0288 |
| Secondary malignant neoplasm of other/unspecified sites | 0.982 | 0.944–1.021 | 0.3638 |
| Opioid analgesic exposure | 1.274 | 1.197–1.355 | <0.0001 |
Cox proportional hazards model for neuropathy-related diagnosis-code outcomes occurring 30–365 days after the post-chemotherapy treatment index. The CBD/CBG cohort was compared with an active-comparator cohort defined by post-chemotherapy exposure to gabapentin, pregabalin, or duloxetine. Gabapentin, pregabalin, and duloxetine were not included as covariates in this model because they were exposure-defining comparator medications. Hazard ratios less than 1 indicate lower hazard of neuropathy-related diagnosis-code outcomes.
Pre-chemotherapy CBD/CBG exposure was not significantly associated with neuropathy-related diagnoses
The pre-chemotherapy Cox model included 162 CBD/CBG-exposed patients and 633,329 control patients after index-event restrictions. In the adjusted Cox model, pre-chemotherapy CBD/CBG exposure was not significantly associated with subsequent neuropathy-related diagnoses (HR = 0.859, 95% CI = 0.476–1.551, p = 0.614) (Figure 2). Although the point estimate was below 1, the confidence interval crossed 1, indicating that pre-chemotherapy CBD/CBG exposure was not significantly associated with reduced neuropathy-related diagnoses. In descriptive outcomes queries, neuropathy-related diagnosis events occurred in 12 of 162 pre-chemotherapy CBD/CBG patients and 40,213 of 646,328 control patients.
In the pre-chemotherapy model, type 1 diabetes, type 2 diabetes, metastatic cancer diagnoses, opioid analgesic exposure, gabapentin, pregabalin, and duloxetine were significantly associated with increased hazard of neuropathy-related diagnosis-code outcomes, whereas male sex was significantly associated with lower hazard (Table 3).
Table 3.
Adjusted Cox proportional hazards model for the pre-chemotherapy exposure analysis.
| Covariate | Hazard ratio | 95% CI | p value |
|---|---|---|---|
| CBD/CBG exposure vs. pre-chemotherapy control | 0.859 | 0.476–1.551 | 0.6137 |
| Male sex | 0.837 | 0.820–0.854 | <0.0001 |
| Age at index | 1.000 | 1.000–1.001 | 0.3342 |
| Type 1 diabetes mellitus | 1.131 | 1.031–1.240 | 0.0092 |
| Type 2 diabetes mellitus | 1.257 | 1.223–1.291 | <0.0001 |
| Other specified diabetes mellitus | 0.977 | 0.886–1.077 | 0.6367 |
| Epilepsy and recurrent seizures | 0.954 | 0.863–1.055 | 0.3627 |
| Convulsions, not elsewhere classified | 1.039 | 0.949–1.138 | 0.4093 |
| Secondary/unspecified malignant neoplasm of lymph nodes | 1.299 | 1.268–1.331 | <0.0001 |
| Secondary malignant neoplasm of respiratory/digestive organs | 1.076 | 1.044–1.108 | <0.0001 |
| Secondary malignant neoplasm of other/unspecified sites | 1.067 | 1.036–1.100 | <0.0001 |
| Opioid analgesic exposure | 1.725 | 1.678–1.774 | <0.0001 |
| Gabapentin | 1.631 | 1.589–1.674 | <0.0001 |
| Pregabalin | 1.356 | 1.291–1.424 | <0.0001 |
| Duloxetine | 1.285 | 1.214–1.360 | <0.0001 |
Cox proportional hazards model for neuropathy-related diagnosis-code outcomes occurring 30–365 days after the first qualifying neurotoxic chemotherapy exposure in the pre-chemotherapy exposure analysis. The exposed cohort included patients with documented CBD/CBG exposure within 1 year on or before chemotherapy, and the control cohort included patients without documented CBD/CBG exposure during that pre-chemotherapy window. Gabapentin, pregabalin, and duloxetine were included as covariates in this model because pre-index use may reflect baseline pain, neuropathic symptoms, mood disorders, or other comorbidities. Hazard ratios less than 1 indicate lower hazard of neuropathy-related diagnosis-code outcomes.
Discussion
In this retrospective real-world cohort study, documented post-chemotherapy CBD/CBG exposure was associated with a lower observed hazard of subsequent neuropathy-related diagnosis-code outcomes compared with active-comparator exposure to gabapentin, pregabalin, or duloxetine. However, this association should be interpreted cautiously because the exposed cohort was small, the exposed-arm event count was suppressed because it was between 1 and 9 patients, and mortality within 365 days was more frequent in the CBD/CBG cohort than in the active-comparator cohort. In contrast, pre-chemotherapy CBD/CBG exposure was not significantly associated with neuropathy-related diagnosis-code outcomes. Together, these findings suggest that documented non-THC cannabinoid exposure after neurotoxic chemotherapy may represent a real-world signal for lower subsequent neuropathy-related diagnosis-code outcomes. However, because patients with neuropathy-related diagnoses on or before chemotherapy were excluded, this analysis should not be interpreted as evidence for treatment of established coded CIPN. Conversely, because post-chemotherapy exposure may have occurred in response to emerging symptoms that had not yet been coded, the finding should also not be interpreted as definitive evidence of prevention.
In this model, patients with post-chemotherapy CBD/CBG exposure had a significantly lower hazard of subsequent neuropathy-related diagnoses compared with patients receiving gabapentin, pregabalin, or duloxetine. This active-comparator design is important because it compares CBD/CBG-exposed patients with individuals receiving medications commonly used in clinical practice for neuropathic pain symptoms, rather than comparing against an untreated population. This approach can reduce confounding by indication by making comparator groups more similar with respect to treatment indication and healthcare-seeking behavior (25–27).
The pre-chemotherapy exposure analysis did not identify a statistically significant protective association. Patients with CBD/CBG exposure within one year before chemotherapy showed a directionally lower hazard of neuropathy-related diagnoses, but the confidence interval crossed 1. This finding should not be interpreted as evidence against a prophylactic effect of minor cannabinoids. Rather, the design captures documented exposure before chemotherapy and does not confirm that the medication was continued during chemotherapy, taken at a therapeutic dose, or used specifically for neuropathy prevention. Therefore, the prophylactic analysis is best interpreted as an exploratory pre-chemotherapy exposure analysis rather than a definitive test of continuous cannabinoid prophylaxis.
The difference between the post-chemotherapy exposure and pre-chemotherapy exposure findings may reflect several factors. First, patients receiving CBD/CBG after chemotherapy may represent a group using cannabinoids in closer temporal proximity to the development or treatment of neuropathy-related symptoms, whereas the pre-chemotherapy group may include patients exposed for unrelated indications that were not continued during chemotherapy. Second, structured EHR data do not capture over-the-counter cannabinoid use, product formulation, dose, duration, adherence, or whether the medication contained other cannabinoids. Third, the pre-chemotherapy exposure cohort was small relative to the control cohort, limiting precision. Finally, it is possible that CBD/CBG-related exposure, or related cannabinoid approaches, may be more relevant for established or emerging neuropathy-related symptoms than for pre-chemotherapy prevention, although this possibility requires controlled experimental testing. Preclinical studies suggest that cannabinoids beyond CBD and CBG, including cannabichromene and combined THC/CBD approaches, may also reduce chemotherapy-associated mechanical hypersensitivity or neuropathic pain-like behavior (17, 22, 28, 29). However, these compounds were not evaluated in the present EHR analysis because the structured exposure definition was limited to CBD and CBG concepts.
Current treatment options for CIPN remain limited, and commonly used neuropathic-pain medications provide incomplete or inconsistent relief for many patients (1, 6). In this context, the observed association between post-chemotherapy CBD/CBG exposure and reduced subsequent neuropathy-related diagnoses supports continued investigation of non-THC cannabinoids as candidate therapeutics. Importantly, this study should not be interpreted as evidence that CBD or CBG is clinically effective for CIPN. Rather, it provides a translational signal that complements preclinical work and supports further mechanistic testing of non-THC cannabinoids in controlled models of chemotherapy-induced neuropathy.
The mechanisms by which CBD/CBG-related exposure might influence chemotherapy-associated neuropathy remain incompletely defined, but the biology of CIPN provides several plausible points of convergence. Across neurotoxic chemotherapy classes, CIPN is thought to involve peripheral sensory-neuron injury, mitochondrial dysfunction and oxidative stress, altered ion-channel function, and neuroimmune/inflammatory signaling, which together increase dorsal root ganglion excitability and promote peripheral sensitization (3, 11, 12, 30, 31). These peripheral changes can drive spinal sensitization and altered sensory processing, including recruitment of normally innocuous mechanosensory input into nociceptive circuits, and may propagate to supraspinal pain-processing networks involved in sensory-discriminative and affective dimensions of pain (3, 30–35).
Within this framework, CBD/CBG and related cannabinoid approaches may influence chemotherapy-associated neuropathic symptoms by modulating several levels of the pain pathway rather than through a single receptor mechanism. Mechanistically, these targets map onto several core features of neuropathic pain biology. TLR4 signaling is linked to neuroimmune activation in CIPN (36–39). Endocannabinoid and PPARγ signaling regulate microglial and inflammatory responses in the nervous system, whereas GPR55 influences neurotransmitter release (37, 40–44). In parallel, 5-hydroxytryptamine 1A (5-HT1A) receptor signaling and alpha-2 (α2)-adrenergic receptor signaling provide monoaminergic inhibitory mechanisms that can reduce nociceptive transmission, including through presynaptic regulation of neurotransmitter release (45–49). α3 glycine receptor potentiation may further strengthen inhibitory control within spinal nociceptive circuits (50–52). Thus, CBD, CBG, and related cannabinoid approaches may reduce neuropathic pain-like behavior by dampening neuroinflammatory signaling, limiting sensory-neuron hyperexcitability, and restoring inhibitory gating of nociceptive input.
For example, CBD has been reported to attenuate paclitaxel-induced mechanical sensitivity through 5-HT1A receptor-dependent mechanisms, and additional work suggests that CBD-related compounds may engage TLR4/endocannabinoid, GPR55, PPARγ-linked, and glycinergic mechanisms relevant to neuropathic pain or chemotherapy neuropathy models (20, 53–57). CBG has been shown to attenuate chemotherapy-induced mechanical hypersensitivity through α2-adrenergic mechanisms, and more recent work implicates modulation of thalamocortical pain signaling and neuroinflammatory markers (16, 18, 19, 23, 24). Because CBD, CBG, and related cannabinoids can also intersect with cannabinoid receptor and endocannabinoid signaling, CB2 receptor-targeting approaches provide additional evidence that cannabinoid-sensitive pathways can suppress chemotherapy-associated neuropathic pain-like behavior, although these findings should not be interpreted as evidence for a CBD- or CBG-specific CB2 mechanism (31, 58).
Other cannabinoid approaches beyond CBD/CBG-coded exposure, including combined THC/CBD exposure, THC-enriched vaporized cannabis, CBN, cannabichromene, and cannabinoid-rich Cannabis sativa extracts, have also shown antinociceptive effects in CIPN or broader nociceptive models (17, 22, 28, 29, 32, 59, 60). Notably, the THC-enriched vaporized cannabis study also showed normalization of raphe nucleus hyperconnectivity in a paclitaxel CIPN model, supporting the broader concept that cannabinoid-sensitive interventions may modulate supraspinal pain-processing circuitry (32). However, the present EHR analysis cannot determine whether the observed association reflects a specific cannabinoid, receptor mechanism, product formulation, dose, timing, THC co-exposure, or pathway. Controlled preclinical and prospective clinical studies are therefore needed to determine whether CBD/CBG-related exposure causally modifies chemotherapy-induced neuropathy and to define the relevant mechanisms.
The covariate results also support the face validity of the models. Diabetes was associated with increased hazard of neuropathy-related diagnoses in both analyses, consistent with diabetes being a major documented risk factor for peripheral neuropathy (61). Opioid analgesic exposure was also associated with increased hazard, likely reflecting greater baseline pain burden, more severe symptoms, or increased clinical surveillance (1, 62, 63). In the pre-chemotherapy model, gabapentin, pregabalin, and duloxetine were associated with increased hazard of neuropathy-related diagnoses. These associations should not be interpreted as evidence that these medications cause neuropathy. More likely, they reflect confounding by indication, where patients receiving these medications have baseline pain, early sensory symptoms, or other neurologic complaints that increase the likelihood of subsequent neuropathy coding.
This study has several strengths. It used a large multi–health-care-organization real-world dataset, included multiple neurotoxic chemotherapy classes, excluded patients with neuropathy-related diagnoses before chemotherapy, and evaluated both post-chemotherapy exposure and pre-chemotherapy exposure designs. The active-comparator treatment analysis also reduced some limitations of comparing CBD/CBG-exposed patients with all non-exposed patients.
Limitations
This observational study cannot establish causality, and residual confounding, confounding by indication, and differences in healthcare utilization may contribute to the observed association. This issue is particularly important for the post-chemotherapy exposure analysis because receipt of gabapentin, pregabalin, or duloxetine after chemotherapy may indicate that patients were already experiencing neuropathic pain, early uncoded CIPN symptoms, or greater baseline pain burden at the time of treatment initiation. Thus, the active-comparator group may have had greater baseline symptom burden or a higher likelihood of neuropathy-related coding than the CBD/CBG group despite covariate adjustment. Excluding patients with both CBD/CBG and gabapentin, pregabalin, or duloxetine exposure created mutually exclusive cohorts but may have selected a distinct CBD/CBG subgroup with different symptom severity, medication preferences, access to cannabinoid products, or clinical documentation patterns.
Additionally, the CBD/CBG-exposed cohorts were small, especially the post-chemotherapy exposure cohort, resulting in wide confidence intervals. Although the Cox models generated finite estimates and confidence intervals, the small CBD/CBG-exposed cohorts may increase vulnerability to sparse-event bias, model instability, and residual confounding. For example, exact event counts for the post-chemotherapy CBD/CBG cohort could not be displayed because TriNetX suppresses detailed results when outcome counts are between 1 and 9 patients. This low exposed-arm event count further underscores the exploratory nature of the estimate and the potential for sparse-event bias and model instability. Therefore, these findings should be interpreted as hypothesis-generating rather than definitive.
Mortality may represent a competing risk for neuropathy-related diagnosis ascertainment in this oncology cohort. Death within 365 days after the post-chemotherapy exposure index was more frequent in the CBD/CBG cohort than in the active-comparator cohort, which may have reduced the opportunity for subsequent neuropathy-related diagnosis coding. Because this analysis did not use a formal competing-risk model, differential mortality between cohorts may have influenced the observed association.
Exposure was limited to structured CBD and CBG medication concepts in the EHR. These records do not reliably capture dose, route, frequency, duration, formulation, indication, adherence, product source, THC content, or whether the documented concept reflected an isolated CBD/CBG product versus a CBD-rich, CBG-containing, or full-spectrum cannabinoid product. Patients with documented THC-containing or other cannabis-related product exposure were not excluded from the primary analyses because structured EHR cannabis-product coding is incomplete and does not reliably distinguish prescription CBD/CBG concepts from over-the-counter, dispensary, CBD-rich, CBG-containing, or full-spectrum products. Excluding only documented THC/cannabis codes would not eliminate unrecorded THC-containing product use and could introduce additional selection bias. Potential co-exposure to THC-containing or full-spectrum cannabinoid products therefore remains an important limitation. Additionally, chemotherapy-agent-specific distributions could not be reliably quantified in the small CBD/CBG-exposed cohorts because chemotherapy exposure was captured using both structured medication and HCPCS procedure concepts, and several agent-specific counts were at or below TriNetX low-count reporting thresholds. Therefore, imbalance in chemotherapy agent, cumulative exposure, dose, cancer type, treatment intent, or concomitant cancer therapy may have influenced the observed associations.
The outcome was based on a diagnosis-code composite rather than clinician-adjudicated CIPN, neuropathy severity scores, patient-reported symptoms, or objective sensory testing. Therefore, the endpoint should be interpreted as a neuropathy-related diagnosis-code outcome rather than confirmed CIPN incidence or symptom severity. The neuropathy-related outcome composite also did not include every ICD-10-CM code potentially related to neuropathy, neuropathic pain, or sensory disturbance. Codes were selected to capture diagnoses plausibly related to chemotherapy-associated neuropathy while avoiding broader pain, neurologic, or disease-specific neuropathy codes that could reduce outcome specificity. As a result, some CIPN-related events may have been missed if they were coded using alternative neuropathy, pain, or sensory-symptom codes, while some captured events may reflect non-CIPN neuropathic or sensory complaints.
Future studies should attempt to replicate this association using alternative datasets, chemotherapy-class-specific analyses, more specific neuropathy outcomes, and designs that better establish cannabinoid exposure timing, persistence, dose, route, formulation, and THC co-exposure. Controlled preclinical studies should directly test whether CBD and CBG reduce chemotherapy-induced mechanical hypersensitivity and alter peripheral or central sensitization, inflammatory signaling, or monoaminergic pain-modulatory pathways. If these findings are replicated, related non-THC cannabinoids could then be prioritized for comparative mechanistic studies.
Conclusions
Post-chemotherapy CBD/CBG exposure was associated with a significantly lower hazard of subsequent neuropathy-related diagnoses compared with active-comparator neuropathic-pain medications in a real-world EHR cohort. Pre-chemotherapy CBD/CBG exposure was not significantly associated with neuropathy-related outcomes. These findings should be interpreted as hypothesis-generating rather than causal and should not be considered evidence for treatment of established CIPN. Instead, they identify a post-chemotherapy exposure signal that supports continued preclinical and translational investigation of CBD/CBG and related non-THC cannabinoids in chemotherapy-associated neuropathy.
Acknowledgments
The authors acknowledge support for TriNetX access through the Penn State Clinical and Translational Science Institute, funded by the National Center for Advancing Translational Sciences of the National Institutes of Health under Clinical and Translational Science Award UL1 TR002014. The funding body had no role in study design, data analysis, interpretation, manuscript preparation, or the decision to submit the work for publication.
Funding Statement
The author(s) declared that financial support was not received for this work and/or its publication.
Footnotes
Edited by: Monique Van Velzen, Erasmus Medical Center, Netherlands
Reviewed by: Ulises Coffeen, National Institute of Psychiatry Ramon de la Fuente Muñiz (INPRFM), Mexico
Lauren Chiec, UH Seidman Cancer Center, United States
Abbreviations CBD, cannabidiol; CBG, cannabigerol; CI, confidence interval; CIPN, chemotherapy-induced peripheral neuropathy; EHR, electronic health record; HCO, healthcare organization; HCPCS, Healthcare Common Procedure Coding System; HR, hazard ratio; ICD-10-CM, International Classification of Diseases, Tenth Revision, Clinical Modification; THC, delta-9-tetrahydrocannabinol.
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
Ethical approval was not required for the study involving humans in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and the institutional requirements.
Author contributions
NG: Writing – review & editing, Writing – original draft, Investigation, Formal analysis, Project administration, Supervision, Data curation, Conceptualization, Visualization, Methodology. AM: Investigation, Methodology, Writing – review & editing. MG: Methodology, Investigation, Writing – review & editing.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was used in the creation of this manuscript. Generative AI was used to assist with language editing. Generative AI was not used to generate study data, conduct the TriNetX queries, perform the statistical analyses, interpret the primary results, or create the scientific conclusions. All AI-assisted edits were reviewed and approved by the authors.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fain.2026.1945720/full#supplementary-material
References
- 1.Colvin LA. Chemotherapy-induced peripheral neuropathy: where are we now? Pain. (2019) 160(Suppl 1):S1–10. 10.1097/j.pain.0000000000001540 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cioroiu C, Weimer LH. Update on chemotherapy-induced peripheral neuropathy. Curr Neurol Neurosci Rep. (2017) 17:47. 10.1007/s11910-017-0757-7 [DOI] [PubMed] [Google Scholar]
- 3.Zajączkowska R, Kocot-Kępska M, Leppert W, Wrzosek A, Mika J, Wordliczek J. Mechanisms of chemotherapy-induced peripheral neuropathy. Int J Mol Sci. (2019) 20:1451. 10.3390/ijms20061451 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Banach M, Juranek JK, Zygulska AL. Chemotherapy-induced neuropathies-a growing problem for patients and health care providers. Brain Behav. (2017) 7:e00558. 10.1002/brb3.558 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Molinares D, Kurtevski S, Zhu Y. Chemotherapy-induced peripheral neuropathy: diagnosis, agents, general clinical presentation, and treatments. Curr Oncol Rep. (2023) 25:1227–35. 10.1007/s11912-023-01449-7 [DOI] [PubMed] [Google Scholar]
- 6.Loprinzi CL, Lacchetti C, Bleeker J, Cavaletti G, Chauhan C, Hertz DL, et al. Prevention and management of chemotherapy-induced peripheral neuropathy in survivors of adult cancers: aSCO guideline update. J Clin Oncol. (2020) 38:3325–48. 10.1200/JCO.20.01399 [DOI] [PubMed] [Google Scholar]
- 7.Mücke M, Phillips T, Radbruch L, Petzke F, Häuser W. Cannabis-based medicines for chronic neuropathic pain in adults. Cochrane Database Syst Rev. (2020) 2020:Cd012182. 10.1002/14651858.CD012182.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.McDonagh MS, Morasco BJ, Wagner J, Ahmed AY, Fu R, Kansagara D, et al. Cannabis-based products for chronic pain: a systematic review. Ann Intern Med. (2022) 175:1143–53. 10.7326/M21-4520 [DOI] [PubMed] [Google Scholar]
- 9.Ligresti A, De Petrocellis L, Di Marzo V. From phytocannabinoids to cannabinoid receptors and endocannabinoids: pleiotropic physiological and pathological roles through complex pharmacology. Physiol Rev. (2016) 96:1593–659. 10.1152/physrev.00002.2016 [DOI] [PubMed] [Google Scholar]
- 10.Sevigny EL. Cannabis and driving ability. Curr Opin Psychol. (2021) 38:75–9. 10.1016/j.copsyc.2021.03.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Cavaletti G, Marmiroli P, Renn CL, Dorsey SG, Serra MP, Quartu M, et al. Cannabinoids: an effective treatment for chemotherapy-induced peripheral neurotoxicity? Neurotherapeutics. (2021) 18:2324–36. 10.1007/s13311-021-01127-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Blanton HL, Brelsfoard J, DeTurk N, Pruitt K, Narasimhan M, Morgan DJ, et al. Cannabinoids: current and future options to treat chronic and chemotherapy-induced neuropathic pain. Drugs. (2019) 79:969–95. 10.1007/s40265-019-01132-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lynch ME, Cesar-Rittenberg P, Hohmann AG. A double-blind, placebo-controlled, crossover pilot trial with extension using an oral mucosal cannabinoid extract for treatment of chemotherapy-induced neuropathic pain. J Pain Symptom Manage. (2014) 47:166–73. 10.1016/j.jpainsymman.2013.02.018 [DOI] [PubMed] [Google Scholar]
- 14.Ward SJ, Lichtman AH, Piomelli D, Parker LA. Cannabinoids and cancer chemotherapy-associated adverse effects. J Natl Cancer Inst Monogr. (2021) 2021:78–85. 10.1093/jncimonographs/lgab007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hanuš LO, Meyer SM, Muñoz E, Taglialatela-Scafati O, Appendino G. Phytocannabinoids: a unified critical inventory. Nat Prod Rep. (2016) 33:1357–92. 10.1039/C6NP00074F [DOI] [PubMed] [Google Scholar]
- 16.Nachnani R, Sepulveda DE, Booth JL, Zhou S, Graziane NM, Raup-Konsavage WM, et al. Chronic cannabigerol as an effective therapeutic for cisplatin-induced neuropathic pain. Pharmaceuticals (Basel). (2023) 16:1442. 10.3390/ph16101442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Raup-Konsavage WM, Sepulveda DE, Wang J, Dokholyan NV, Vrana KE, Graziane NM. Antinociceptive effects of cannabichromene (CBC) in mice: insights from von frey, tail-flick, formalin, and acetone tests. Biomedicines. (2023) 12:83. 10.3390/biomedicines12010083 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sepulveda DE, Morris DP, Raup-Konsavage WM, Sun D, Vrana KE, Graziane NM. Cannabigerol (CBG) attenuates mechanical hypersensitivity elicited by chemotherapy-induced peripheral neuropathy. Eur J Pain. (2022) 26:1950–66. 10.1002/ejp.2016 [DOI] [PubMed] [Google Scholar]
- 19.Wade QW, Morris N, Sepulveda DE, Cortez-Resendiz A, Brandl C, Lausch JE, et al. Cannabigerol reverses mechanical allodynia through α(2A)-adrenergic modulation of thalamocortical signaling in chemotherapy-induced neuropathy. Br J Pharmacol. (2025). 10.1111/bph.70281 [DOI] [PubMed] [Google Scholar]
- 20.Ward SJ, McAllister SD, Kawamura R, Murase R, Neelakantan H, Walker EA. Cannabidiol inhibits paclitaxel-induced neuropathic pain through 5-HT1A receptors without diminishing nervous system function or chemotherapy efficacy. Br J Pharmacol. (2014) 171:636–45. 10.1111/bph.12439 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ward SJ, Ramirez MD, Neelakantan H, Walker EA. Cannabidiol prevents the development of cold and mechanical allodynia in paclitaxel-treated female C57Bl6 mice. Anesth Analg. (2011) 113:947–50. 10.1213/ANE.0b013e3182283486 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.King KM, Myers AM, Soroka-Monzo AJ, Tuma RF, Tallarida RJ, Walker EA, et al. Single and combined effects of Δ(9) -tetrahydrocannabinol and cannabidiol in a mouse model of chemotherapy-induced neuropathic pain. Br J Pharmacol. (2017) 174:2832–41. 10.1111/bph.13887 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Li H, Ward SJ. Paclitaxel-associated mechanical sensitivity and neuroinflammation are sex-, time-, and site-specific and prevented through cannabigerol administration in C57Bl/6 mice. Int J Mol Sci. (2024) 25:4277. 10.3390/ijms25084277 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hayduk SA, Hughes AC, Winter RL, Milton MD, Ward SJ. Single and combined effects of cannabigerol (CBG) and cannabidiol (CBD) in mouse models of oxaliplatin-associated mechanical sensitivity, opioid antinociception, and naloxone-precipitated opioid withdrawal. Biomedicines. (2024) 12:1145. 10.3390/biomedicines12061145 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gokhale M, Stürmer T, Buse JB. Real-world evidence: the devil is in the detail. Diabetologia. (2020) 63:1694–705. 10.1007/s00125-020-05217-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sendor R, Stürmer T. Core concepts in pharmacoepidemiology: confounding by indication and the role of active comparators. Pharmacoepidemiol Drug Saf. (2022) 31:261–9. 10.1002/pds.5407 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Yoshida K, Solomon DH, Kim SC. Active-comparator design and new-user design in observational studies. Nat Rev Rheumatol. (2015) 11:437–41. 10.1038/nrrheum.2015.30 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Sepulveda DE, Vrana KE, Graziane NM, Raup-Konsavage WM. Combinations of cannabidiol and Δ(9)-tetrahydrocannabinol in reducing chemotherapeutic induced neuropathic pain. Biomedicines. (2022) 10:2548. 10.3390/biomedicines10102548 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Harris HM, Sufka KJ, Gul W, ElSohly MA. Effects of delta-9-tetrahydrocannabinol and cannabidiol on cisplatin-induced neuropathy in mice. Planta Med. (2016) 82:1169–72. 10.1055/s-0042-106303 [DOI] [PubMed] [Google Scholar]
- 30.Mattar M, Umutoni F, Hassan MA, Wamburu MW, Turner R, Patton JS, et al. Chemotherapy-induced peripheral neuropathy: a recent update on pathophysiology and treatment. Life (Basel). (2024) 14:991. 10.3390/life14080991 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Masocha W. Targeting the endocannabinoid system for prevention or treatment of chemotherapy-induced neuropathic pain: studies in animal models. Pain Res Manag. (2018) 2018:5234943–9. 10.1155/2018/5234943 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Alkislar I, Miller AR, Hohmann AG, Sadaka AH, Cai X, Kulkarni P, et al. Inhaled cannabis suppresses chemotherapy-induced neuropathic nociception by decoupling the raphe nucleus: a functional imaging study in rats. Biol Psychiatry Cogn Neurosci Neuroimaging. (2021) 6:479–89. 10.1016/j.bpsc.2020.11.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Rosner J, De Andrade DC, Davis KD, Gustin SM, Kramer JLK, Seal RP, et al. Central neuropathic pain. Nat Rev Dis Primers. (2023) 9:73. 10.1038/s41572-023-00484-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Peirs C, Williams SG, Zhao X, Arokiaraj CM, Ferreira DW, Noh MC, et al. Mechanical allodynia circuitry in the dorsal horn is defined by the nature of the injury. Neuron. (2021) 109:73–90.e7. 10.1016/j.neuron.2020.10.027 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Peirs C, Williams SP, Zhao X, Walsh CE, Gedeon JY, Cagle NE, et al. Dorsal horn circuits for persistent mechanical pain. Neuron. (2015) 87:797–812. 10.1016/j.neuron.2015.07.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Li Y, Zhang H, Zhang H, Kosturakis AK, Jawad AB, Dougherty PM. Toll-like receptor 4 signaling contributes to paclitaxel-induced peripheral neuropathy. J Pain: Official journal of the American Pain Society. (2014) 15:712–25. 10.1016/j.jpain.2014.04.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Araldi D, Khomula EV, Bonet IJM, Bogen O, Green PG, Levine JD. Role of pattern recognition receptors in chemotherapy-induced neuropathic pain. Brain J Neurol. (2024) 147:1025–42. 10.1093/brain/awad339 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Tanga FY, Nutile-McMenemy N, DeLeo JA. The CNS role of toll-like receptor 4 in innate neuroimmunity and painful neuropathy. Proc Natl Acad Sci U S A. (2005) 102:5856–61. 10.1073/pnas.0501634102 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Yang Y, Zhao B, Huo J, Wei G, Liu X, Lan H, et al. Microglial macrophage-derived ds-HMGB1 in DRG orchestrates neuropathic pain through immune-neural signaling. Cell Rep. (2025) 44:116671. 10.1016/j.celrep.2025.116671 [DOI] [PubMed] [Google Scholar]
- 40.Sylantyev S, Jensen TP, Ross RA, Rusakov DA. Cannabinoid- and lysophosphatidylinositol-sensitive receptor GPR55 boosts neurotransmitter release at central synapses. Proc Natl Acad Sci U S A. (2013) 110:5193–8. 10.1073/pnas.1211204110 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Sánchez-Zavaleta R, Ávalos-Fuentes JA, González-Hernández AV, Recillas-Morales S, Paz-Bermúdez FJ, Leyva-Gómez G, et al. Presynaptic nigral GPR55 receptors stimulate [(3) H]-GABA release through [(3) H]-cAMP production and PKA activation and promote motor behavior. Synapse (New York, N.Y.). (2022) 76:e22246. 10.1002/syn.22246 [DOI] [PubMed] [Google Scholar]
- 42.Li X, Guo Q, Ye Z, Wang E, Zou W, Sun Z, et al. PPAR γ prevents neuropathic pain by down-regulating CX3CR1 and attenuating M1 activation of microglia in the spinal cord of rats using a sciatic chronic constriction injury model. Front Neurosci. (2021) 15:620525. 10.3389/fnins.2021.620525 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Morgenweck J, Griggs RB, Donahue RR, Zadina JE, Taylor BK. PPARγ activation blocks development and reduces established neuropathic pain in rats. Neuropharmacology. (2013) 70:236–46. 10.1016/j.neuropharm.2013.01.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Maldonado R, Baños JE, Cabañero D. The endocannabinoid system and neuropathic pain. Pain. (2016) 157(Suppl 1):S23–32. 10.1097/j.pain.0000000000000428 [DOI] [PubMed] [Google Scholar]
- 45.Jeong HJ, Mitchell VA, Vaughan CW. Role of 5-HT(1) receptor subtypes in the modulation of pain and synaptic transmission in rat spinal superficial dorsal horn. Br J Pharmacol. (2012) 165:1956–65. 10.1111/j.1476-5381.2011.01685.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Choi IS, Cho JH, Jang IS. 5-Hydroxytryptamine 1a receptors inhibit glutamate release in rat medullary dorsal horn neurons. Neuroreport. (2013) 24:399–403. 10.1097/WNR.0b013e3283614cbf [DOI] [PubMed] [Google Scholar]
- 47.Ueda M, Oyama T, Kuraishi Y, Akaike A, Satoh M. Alpha 2-adrenoceptor-mediated inhibition of capsaicin-evoked release of glutamate from rat spinal dorsal horn slices. Neurosci Lett. (1995) 188:137–9. 10.1016/0304-3940(95)11397-F [DOI] [PubMed] [Google Scholar]
- 48.Pan YZ, Li DP, Pan HL. Inhibition of glutamatergic synaptic input to spinal lamina II(o) neurons by presynaptic alpha(2)-adrenergic receptors. J Neurophysiol. (2002) 87:1938–47. 10.1152/jn.00575.2001 [DOI] [PubMed] [Google Scholar]
- 49.Kawasaki Y, Kumamoto E, Furue H, Yoshimura M. Alpha 2 adrenoceptor-mediated presynaptic inhibition of primary afferent glutamatergic transmission in rat substantia gelatinosa neurons. Anesthesiology. (2003) 98:682–9. 10.1097/00000542-200303000-00016 [DOI] [PubMed] [Google Scholar]
- 50.Zeilhofer HU, Werynska K, Gingras J, E G. Yévenes, glycine receptors in spinal nociceptive control—an update. Biomolecules. (2021) 11:846. 10.3390/biom11060846 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Zeilhofer HU. The glycinergic control of spinal pain processing. CMLS. (2005) 62:2027–35. 10.1007/s00018-005-5107-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Imlach WL. New approaches to target glycinergic neurotransmission for the treatment of chronic pain. Pharmacol Res. (2017) 116:93–9. 10.1016/j.phrs.2016.12.019 [DOI] [PubMed] [Google Scholar]
- 53.Dos Santos R, Veras F, Netto G, Elisei L, Sorgi C, Faccioli L, et al. Cannabidiol prevents chemotherapy-induced neuropathic pain by modulating spinal TLR4 via endocannabinoid system activation. J Pharm Pharmacol. (2023) 75:655–65. 10.1093/jpp/rgad023 [DOI] [PubMed] [Google Scholar]
- 54.Ippolito M, Hayduk SA, Kinney W, Brenneman DE, Ward SJ. KLS-13019, a novel structural analogue of cannabidiol and GPR55 receptor antagonist, prevents and reverses chemotherapy-induced peripheral neuropathy in rats. J Pharmacol Exp Ther. (2024) 391:231–40. 10.1124/jpet.124.002190 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Islas-Espinoza AM, Ramos R, Escoto-Rosales MJ, Pizaña-Encarnación JM, Morales-Galindo DK, Caram-Salas NL, et al. Cannabidiol reduces neuropathic pain and cognitive impairments through activation of spinal PPARγ. J Pain: Official journal of the American Pain Society. (2025) 30:105378. 10.1016/j.jpain.2025.105378 [DOI] [PubMed] [Google Scholar]
- 56.Silva NR, Gomes FIF, Lopes AHP, Cortez IL, Dos Santos JC, Silva CEA, et al. The cannabidiol analog PECS-101 prevents chemotherapy-induced neuropathic pain via PPARγ receptors. Neurotherapeutics. (2022) 19:434–49. 10.1007/s13311-021-01164-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Xiong W, Cui T, Cheng K, Yang F, Chen SR, Willenbring D, et al. Cannabinoids suppress inflammatory and neuropathic pain by targeting α3 glycine receptors. J Exp Med. (2012) 209:1121–34. 10.1084/jem.20120242 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Guenther KG, Wirt JL, Oliva I, Saberi SA, Crystal JD, Hohmann AG. The cannabinoid CB(2) agonist LY2828360 suppresses neuropathic pain behavior and attenuates morphine tolerance and conditioned place preference in rats. Neuropharmacology. (2025) 265:110257. 10.1016/j.neuropharm.2024.110257 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Xu Y, Yao L, Guo Y, Shi C, Zhou J, Hua M. The potential antinociceptive effect and mechanism of cannabis sativa L. Extract on paclitaxel-induced neuropathic pain in rats uncovered by multi-omics analysis. Molecules. (2024) 29:1958. 10.3390/molecules29091958 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Schwarz AM, Kobeci D, Mancuso JA, Moreno-Rodríguez V, Seekins C, Bui T, et al. Select minor cannabinoids from cannabis sativa are cannabimimetic and antinociceptive in a mouse model of chronic neuropathic pain. J Pharmacol Exp Ther. (2024) 391:214–21. 10.1124/jpet.124.002212 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Gu J, Lu H, Chen C, Gu Z, Hu M, Liu L, et al. Diabetes mellitus as a risk factor for chemotherapy-induced peripheral neuropathy: a meta-analysis. Support Care Cancer. (2021) 29:7461–9. 10.1007/s00520-021-06321-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Gewandter JS, Kleckner AS, Marshall JH, Brown JS, Curtis LH, Bautista J, et al. Chemotherapy-induced peripheral neuropathy (CIPN) and its treatment: an NIH collaboratory study of claims data. Support Care Cancer. (2020) 28:2553–62. 10.1007/s00520-019-05063-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Magge RS, DeAngelis LM. The double-edged sword: neurotoxicity of chemotherapy. Blood Rev. (2015) 29:93–100. 10.1016/j.blre.2014.09.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.