Cannabinoid Vaping Products: Regulation, Composition, Toxicological Effects, and Emerging Research

Abstract

The 2018 U.S. Farm Bill inadvertently paved the way for a market of unregulated, hemp-derived cannabinoid vaping products, including cannabidiol (CBD) and Δ8-tetrahydrocannabinol (Δ8-THC). These products contain extremely high cannabinoid concentrations, contaminants, and potentially harmful byproducts from heating, raising concerns about respiratory toxicity. This review examines the regulatory landscape, manufacturing practices, composition, and toxicological mechanisms associated with hemp-derived cannabinoid vaping products. While vaping-related lung injuries, such as E-cigarette or Vaping, Product use-Associated Lung Injury (EVALI), have been linked to vitamin E acetate (VEA), a definitive mechanism of injury has not been established, and cases continue to be reported. Studies reveal multiple mechanisms of lung toxicity associated with cannabinoid vaping, including inflammatory responses, oxidative stress, and damage from contaminants like heavy metals and flavoring agents. Emerging evidence also highlights the formation of reactive cannabinoid quinones (e.g., CBDQ) during vaping, which form covalent adducts with protein cysteine residues, potentially altering their function, and also have the potential to drive oxidative damage through redox cycling. These electrophilic quinones may act as pleiotropic modifiers of cellular function and represent an important, yet understudied, contributor to cannabinoid vaping toxicity. This review identifies key research gaps, including the need for studies on chronic exposure models, mechanisms of lung injury, and the interplay between VEA, cannabinoid quinones, and other harmful byproducts. Additionally, given the potential for both therapeutic benefits and toxic effects, research should investigate optimal temperatures and formulations that balance efficacy and safety over potential toxicity caused by thermal oxidation. Overall, a comprehensive understanding of the toxicological mechanisms of cannabinoid vaping products is essential to guide public health decisions, inform regulatory frameworks, and support the development of safer products.

Graphical Abstract

INTRODUCTION

In 2018, the Agriculture Improvement Act, often known as the Farm Bill,¹ removed hemp and other cannabis derivatives containing less than 0.3% Δ9-tetrahydrocannabinol (Δ9-THC) from the definition of marijuana under the Controlled Substances Act (CSA) in the United States (U.S.). This bill represented a radical change in cannabis policy in the U.S. and unintentionally facilitated the emergence of a host of unstudied and unregulated alternative cannabinoid products. While the Farm Bill was originally intended to legalize certain products containing hemp and its main constituent, the noneuphoric cannabinoid cannabidiol (CBD), it also created a legal gray area and regulatory ambiguities that manufacturers exploited. By 2019, companies began using acid catalysis to convert hemp-derived CBD into cannabinoids such as Δ8-tetrahydrocannabinol (Δ8-THC), a compound with euphoric effects similar to Δ9-THC. This loophole allowed products containing Δ8-THC and other alternative cannabinoids to be sold nationwide, bypassing traditional cannabis regulations.

While hemp-derived cannabinoid products also include topicals, edibles, and tinctures, vaping products pose a particular public health concern due to the potential of lung injury resulting from inhalation of extremely high cannabinoid concentrations (often exceeding 95%), impurities such as flavoring chemicals, heavy metals, pesticides, and harmful byproducts generated from heated aerosolization.² These risks are reflected in real-world outcomes, as cannabinoid vaping products are highly associated with the outbreak of E-cigarette, or Vaping, Product use-Associated Lung Injury (EVALI).^3,4 In addition to EVALI, other adverse health effects including anxiety, respiratory impairment, hallucinations, and pediatric toxicity have been reported for both CBD and Δ8-THC vaping products.^5–14

Surveys from 2018–2022 indicate that CBD vaping is common among both adults and youth, with 18.9–38.5% of adult CBD users and over 21% of adolescent vapers reporting past CBD vaping product use. In addition, CBD was detected in 26% of analyzed vape devices confiscated from California high schools.^15–18 While vaping was the most common method of use in one study of adult Δ8-THC users, larger surveys show that 11.4% of U.S. 12th graders and 12.4% of young adults used Δ8-THC. However, though population-level data on Δ8-THC vaping specifically remain limited.^19–21 Within users, CBD and Δ8-THC vaping products are primarily used for anxiety, stress, sleep problems, and pain, with Δ8-THC additionally used for its euphoric effects, despite limited clinical evidence and regulatory oversight.^{15,16,19,22–27} Despite their growing popularity,²⁸ especially among adolescents, and their potential to induce lung injury, research on the respiratory health effects of cannabinoid vaping products remains limited. Moreover, the regulation, composition, and biological effects of these products have not been comprehensively reviewed in an integrated context, highlighting a critical knowledge gap.

LEGAL FRAMEWORK AND REGULATION OF HEMP-DERIVED CANNABINOIDS

The legal and scientific definitions of cannabis, marijuana, and hemp are complex, and are often misunderstood, and debated by the recreational cannabis and botanist communities.^29,30 However, they are important for understanding the current landscape of cannabinoid legality and regulation in the U.S. Cannabis is a genus that includes both hemp (legally classified as having <0.3% Δ9-THC on a dry weight basis) and marijuana (>0.3% Δ9-THC), but these terms do not represent different species. Although hemp is often classified as the cannabis species Cannabis sativa, genetic studies show it is more closely related to modern “Indica” (Crematogaster afghanica, wide-leaflet biotype)^30,31 (Figure 1). Because extensive cross-breeding has blurred distinctions between sativa and indica, chemical profiles—not species labels—provide a more accurate basis for differentiating hemp from other cannabis varieties.³⁰

Figure 1.

Explanation of modern Cannabis terminology.

When the Farm Bill removed hemp from the classification of marijuana, it did not explicitly legalize CBD but rather allowed for the sale of hemp-derived cannabinoids when they are produced following the guidelines of the Farm Bill, which includes certain stipulations such as licensing of hemp growers.^32–34 Manufacturers further argue that Δ8-THC and other alternative cannabinoid products are allowed because they are produced from hemp-derived CBD and are, therefore, “hemp products”.³⁴ This has led to the production and distribution of Δ8-THC and other hemp-derived cannabinoids, even in states where Δ9-THC-containing cannabis is not legal recreationally or medically. Hemp-derived cannabinoid products present a unique regulatory concern as they are not subject to cannabis control systems, which govern minimum purchasing age, quality (including testing for potency, consistency, and contaminants), and labeling standards in states with legalized cannabis.³⁴

Since the emergence of hemp-derived cannabinoid products on the U.S. market, there have been efforts to restrict or regulate them. Hemp-derived CBD products are still largely legal in all states, although several states go beyond federal limits and require no traceable amount of Δ9-THC.³⁵ While the legal landscape of Δ8-THC is constantly changing, it is currently available for sale in 23 states in the U.S. including North Carolina,^36,37 which is down from 28 states at the end of 2021.³⁴ States that have banned Δ8-THC have often done so by adding a broader definition of THC that includes isomers other than Δ9-THC.^34,36 The U.S. Food and Drug Administration (FDA) has further clarified that the Farm Bill explicitly preserves their authority over hemp products including foods, dietary supplements, human and veterinary drugs, and cosmetics.¹ CBD and Δ8-THC food products and dietary supplements in interstate commerce violate the Federal Food, Drug, and Cosmetic Act (FD&C Act), which prohibits substances that are active ingredients in drug products and nonapproved food additives.^38,39 This rule has been violated by many companies, resulting in the issuing of warning letters by the FDA.³⁸ The U.S. Drug Enforcement Administration (DEA) has also clarified that substances, including Δ8-tetrahydrocannabinol acetate (Δ8-THC-O-Acetate) and Δ9-tetrahydrocannabinol acetate (Δ9-THC-O-Acetate), are prohibited as they are not naturally occurring in the hemp plant and therefore do not fall under the definition of hemp⁴⁰ (Figure 2). However, there are currently no public plans to regulate CBD or Δ8-THC vaping products at the federal level by the FDA, DEA, or any other agency.

Figure 2.

Structures of CBD and Δ8-THC versus Δ9-THC and other “hemp-derived” cannabinoids, including those that are not naturally occurring and are prohibited by the DEA, and those that are still sold.

FORMULATIONS, DEVICE TYPES, AND CONTAMINANTS

Manufacturing Hemp-Derived Cannabinoids.

CBD-dominant hemp is milled and extracted with CO₂ or solvents (commonly ethanol) to produce crude oil, which undergoes winterization or dewaxing to remove high melting point components.⁴¹ Waxes are separated by filtration or centrifugation, and cannabinoids are further purified by chromatography or crystallization.⁴¹ CBD isolate can then be converted to Δ8-THC, naturally present at low levels in hemp, through acid-catalyzed cyclization using an acid, solvent, heat, and time, with different conditions yielding distinct reaction profiles.^42,43 Following the emergence of Δ8-THC, other alternative cannabinoids, most found naturally in low levels in hemp, began to emerge in the U.S. marketed as hemp-derived products and, therefore Farm Bill compliant. These currently include Δ10-THC, cannabinol (CBN), hexahydrocannabinol (HHC), tetrahydrocannabivarin (THCV), tetrahydrocannabiphorol (THCP), Δ9-tetrahydrocannabutol (THCB), and tetrahydrocannabinolic acid (THCA) (Figure 2).^44–46 The manufacturing process used to make these products is not well documented. However, it has been speculated that some of these cannabinoids may not actually be hemp-derived. For example, due to yields, THCA products are most likely extracted from marijuana rather than hemp.⁴⁶ Similarly, THCP products are likely to be produced completely synthetically.^45,46

Product and Device Types.

Previous research comparing cannabinoid vaping devices, has shown differential cellular responses and aerosol composition based on the device characteristics.⁴⁷ There are various types of cannabinoid vaping products, but previous research has not systematically categorized them or assessed their popularity. Cartridges and pods are popular product types⁴⁸ that are typically prefilled with cannabinoid distillate, which can range from 700 to 2000 mg/mL of cannabinoids and often contains naturally occurring terpenes, with some formulations including added terpenes for enhanced effects or flavor (Figure 3). Terpenes are a diverse class of organic hydrocarbons found in the essential oils of plants, including cannabis plants, where they contribute to characteristic aromas and flavors. Emerging evidence also indicates that these compounds may exert physiological and neuropsychological effects, influencing mood, perception, and aspects of health,⁴⁹ as well as having potential anti-inflammatory effects.⁵⁰ In formulations such as THC vape oil, terpenes play a critical role in defining strain-specific organoleptic properties, including distinct flavor profiles and volatile scent signatures.

Figure 3.

Cannabinoid vaping product and device types.

Disposable cartridges and pods, which contain a built-in coil (usually ceramic or quartz) and wick, are connected by the user to a battery that supplies power. Cartridges typically attach to 510-thread batteries, providing cross-compatibility between different cartridges and battery models. Pods, however, are typically proprietary to specific brands and are designed to work only with their corresponding batteries. Another product category, that based on online availability is less popular overall, is juices. Juices are usually sold in larger (often 30 mL) bottles and can be used to fill any refillable vaping device, although they are sometimes sold alongside recommended devices (Figure 3). Juices tend to be the lowest concentration product type and can range from 15–50 mg/mL. Unlike cartridges and pods, which are usually formulated with distillates, juices are propylene glycol (PG) and vegetable glycerin (VG) based, similar to nicotine e-cigarette formulations, and often contain added flavoring chemicals. Disposable cannabinoid vaping devices are another very popular product category (Figure 3). Unlike cartridges and pods where the battery can be reused, disposable products are designed to be completely replaced after each use. Disposable devices can be filled with distillate or be PG/VG based. Not all cannabinoid vaping devices have temperature settings; however, pod batteries sometimes have low (315–350 °F) and high (400–600 °F)⁵¹ settings. Additionally, cartridge and pod-based devices as well as disposables may be puff or button-activated.

Cannabinoid Content, Additives, and Contaminants.

Content and concentrations of components and additives in CBD and THC vaping products are often incorrect. Multiple studies have shown that CBD vaping products often contain CBD levels that vary by ± 20% from their labeled concentration.^52,53 Others have found actual concentrations ranged from −34 to −100% of that labeled.⁵⁴ In a recent study, the majority of products had CBD concentrations that deviated at least 10% from their label and 26% of the products did not align with the product type specified on the packaging.⁵⁵ In the same study, heavy metals, most commonly lead, were detected in 21.8% of the products tested, and residual solvents were detected in 89.6%, with hexane, m/p-xylene, methanol, and o-xylene being the most common.⁵⁵ A total of 26 pesticides were also found across 30 products (14.9% of all products tested).⁵⁵ Additionally, CBD vaping products have been found to contain unlabeled Δ9-THC, synthetic cannabinoids, silicones, chemical solvents, and various terpenes with unknown inhalation safety.^52,54,56 Conversely, one recent study found only low levels of THC and other cannabinoids in CBD vaped condensates.⁵⁷

Similar to CBD vaping products, Δ8-THC vaping products have been shown to deviate from their label and contain a number of additives and contaminants. In contrast to CBD vaping products, Δ8-THC products may also have unique contaminants due to technical and quality control issues associated with acid-catalyzed ring closure of cannabidiol.⁵⁸ A comprehensive listing of acid-catalyzed ring closure of cannabidiol byproducts based on available literature included 23 compounds, most of which have not been evaluated for safety or toxicity.⁵⁸ Out of 27 tested Δ8-THC products in one study, Δ8-THC concentration was not accurate for any, deviating as much as 40% from the labeled value, 11 samples contained unlabeled cutting agents, and all exhibited reaction byproducts including metals, olivetol, and other cannabinoids including novel cannabinoids.⁵⁹ Furthermore, while many retailers include testing results on their Web sites, it has been found that Δ8-THC products contain impurities in concentrations far higher than what was on their certificates of analysis.⁶⁰ This may at least in part be because high-performance liquid chromatography (HPLC) methods used by most laboratories to evaluate cannabinoid products cannot accurately separate, identify, and quantify byproducts of the acid-catalyzed ring closure of cannabidiol.⁵⁸ Anecdotal evidence also supports instances of certain retailers falsifying lab reports.⁶¹

Overall, a significant public health concern is posed by the deviation of cannabinoid products from their labels and the frequency of additives and contaminants. While the safety of many contaminants has not been evaluated, those that have been evaluated often show potential adverse effects. For example, some terpenes and diluents commonly found in cannabinoid vaping products have also been directly shown to have cytotoxic effects on human lung cells.^52,62 Risk assessment data further indicates noncarcinogenic risks associated with inhalation of heavy metals in vaping products including manganese, copper, and nickel.⁶³

BIOLOGICAL AND TOXICOLOGICAL EFFECTS OF CANNABINOID VAPING PRODUCTS

Cannabinoid Receptors, Transport, and Metabolism.

Primary effects of cannabinoids, including psychoactive effects, are thought to be mainly exerted through CB1 and CB2 receptors, which are extensively present throughout the central nervous system and peripheral tissues, including the lungs. THC and CBD have similar chemical structures differentiated by a cyclic ring on THC versus a hydroxyl group on CBD. While THC exists in a planar conformation, the rings in CBD are at right angles to one another.⁶⁴ The planar conformation of THC allows it to bind to CB1 receptors, explaining the euphoric effects of THC compared to CBD.⁶⁴ In addition to CB1 and CB2 receptors, CBD also acts on transient receptor potential (TRP) channels including TRPV1 and TRPV2, which are broadly involved in pain and sensory signaling. THC targets TRP channels (mainly TRPV2), the nuclear receptor PPARγ that regulates adipocyte differentiation and metabolism, the G-protein–coupled receptors GPR18, a GPCR involved in immune cell signaling, GPR55, a GPCR that regulates calcium signaling and neuronal activity, and glycine receptors, which mediate inhibitory neurotransmission in the central nervous system.^65,66 While little is known about Δ8-THC specifically, it is a mixed agonist of CB1 and CB2 but has less affinity for CB1 than Δ9-THC, potentially explaining why its euphoric effects are milder.⁶⁶ Once they enter the body, CBD and THC can be transported through the blood by lipoproteins and albumin.⁶⁷ Computational and ligand displacement studies have also indicated that both CBD and THC bind to fatty acid-binding proteins which facilitate their intracellular transport.⁶⁷

Metabolites of inhaled cannabinoids derived from metabolism in the lung may contribute to their effects in the lung. For example, the overall bioavailability of CBD is low (11–45%), and the major and active metabolite of CBD is 7-OH–CBD.^68,69 In addition to 7-OH–CBD, a number of other metabolites are generated via hydroxylation and carboxylation.^69,69–71 Some of these hydroxylated analogs have nanomolar affinities for the CB2 receptor.⁶⁴ The metabolism of Δ8-THC is very similar to that of Δ9-THC.^66,70 The primary metabolites of Δ8-THC are the euphoric 11-OH-Δ8-THC, produced by cytochrome P450 (CYP) enzyme hydroxylation, and inactive metabolite Δ8-THC–COOH(73, 73–75).⁶⁶ In phase-II metabolism of Δ8-THC–COOH, glucuronic acid is added to the 11-COOH or phenolic OH group, which increases water solubility and facilitates excretion.⁶⁶ Although the liver is a primary location of metabolism, side-chain hydroxylation of THC is also thought to be prominent in the lung,⁷⁰ but no specific data exist for Δ8-THC. While less is known about CBD and Δ8-THC metabolism in the lung specifically, major metabolic enzymes of CBD, including CYP2C19 (mainly expressed in alveolar type 2 progenitor cells) and CYP3A4 (lower expression across many cell types) and of Δ8-THC including CYP2C9 (alveolar type 2 progenitor, goblet, basal, and suprabasal cells) and CYP3A4 (across many cell types) are found in the lung.⁷²

Experimental Studies on the Effects of Vaped Hemp-Derived Cannabinoids.

Several studies have been published examining the effects of aerosolized CBD on lung epithelial cells. A 2019 study examined inflammatory responses in lung epithelial cells (BEAS-2B and NHBE), macrophages, and lung fibroblasts to CBD aerosol exposure (19.6 mg/mL, 2 L/min flow, 2 puffs/min, 30 min) using a chamber connected to a Scireq inExpose system.⁷³ They found that CBD aerosols produced acellular, cellular, and mitochondrial reactive oxygen species (ROS) and that they caused both an anti-inflammatory response, through a reduction in MCP-1 cytokine levels and inhibition of LPS induced NF-κB, and a pro-inflammatory response, including increased IL-8, CXCL1, and CXCL225.⁷³ Also in 2019, the Scripps Research Institute developed an e-Vape inhalation chamber for rats, which has been used to examine CBD.⁷⁴ It was found that CBD aerosols (100 and 400 mg/mL) induced dose-dependent hypothermia but not nociception.⁷⁴ Another study shortly thereafter used an LX-1 smoking machine to study the effects of commercial CBD-containing vaping products and refill solutions along with flavorings on a bronchial epithelial cell line.⁷⁵ They showed that CBD aerosols (1.7 mg/mL flavored, 33.3 mg/mL unflavored, 2s puff, 2 puffs/min, 30 min) produced cytotoxic effects and increased release of pro-inflammatory cytokines IL-1β, CXCL1, and CXCL10, and that combination with certain flavorings amplified these effects.⁷⁵ CBD aerosols with and without vitamin E acetate (VEA) (a potential causative agent of E-cigarette, or Vaping, Product use-Associated Lung Injury (EVALI)^3,4) have been further shown to decrease cell viability after 3 days of exposure, leave microscopic depositions on the epithelial surface prior to cell death, and alter protein secretions involved in detoxification, redox stress, carbon metabolism, and glycan degradation.⁷⁶ In the most recent study on CBD vaping, mice exposed to aerosolized CBD had more inflammation, severe lung damage, and oxidative stress than nicotine exposure.⁷⁷ Together, these studies indicate that vaporized CBD induces inflammatory mediators, cytotoxicity, and oxidative stress, and highlight safety concerns associated with the use of these products. However, we do not yet understand the mechanism of action of the effects, including whether they are due to CBD itself or other components of the mixture, such as additives and contaminants in commercial products or compounds formed through heated aerosolization. In addition, similar studies conducted using Δ8-THC vaping products are lacking.

There are limited studies on vaporized THC, most examining Δ9-THC. In 2016, the same group at The Scripps Research Institute used their e-Vape inhalation chamber to show that exposure to Δ9-THC aerosols (12.5 and 25 mg/mL) reduced body temperature and increased spontaneous locomotion, similar to injected Δ9-THC, in rats.⁷⁸ However, their model has not been used to explore the toxicologic effects of Δ9-THC in the lung. Rats exposed to vaped Δ9-THC were also shown to have sex-dependent differences in responses, where females had higher 11-OH-THC levels and behavioral effects.⁷⁹ Additional work in rodents has demonstrated dose-dependent serum THC levels and locomotor effects, and a recent murine inhalation model using commercial cannabis distillate vapes confirmed physiologically relevant systemic exposure and hypo-locomotion responses in mice.^80,81 One human pharmacokinetic modeling study further showed how puff patterns and aerosol characteristics influence THC deposition and plasma concentrations.⁸² Finally, two in vitro studies using human airway and immune cells found that exposure to commercial high-THC cannabis vape products induced oxidative stress responses, altered gene expression profiles, and promoted secretion of pro-inflammatory mediators.^83,84 A recent chamber study also reported that THC-containing vaping aerosols produce high concentrations of ultrafine particles that efficiently deposit in the lungs of nonusers, underscoring potential risks of secondhand exposure.⁸⁵

While one more recent study examined the behavioral effects of vaporized Δ8-THC in rats and found a hyper-locomotion effect,⁸⁶ to date, only one modern experimental study has investigated the biological effects of Δ8-THC exposure on the lung beyond behavioral outcomes. In this study, various cell lines were treated with media containing vaped condensates of either Δ8-THC alone or Δ8-THC combined with VEA.⁸⁷ Findings revealed that vaped Δ8-THC led to cytotoxic effects, reduced mitochondrial membrane potential, and triggered the release of multiple cytokines and chemokines, including FGF-2 and IL-18.⁸⁷ Therefore, while preliminary evidence indicates potential lung injury due to Δ8-THC vaped condensates, no research has specifically examined the respiratory impact of inhaled aerosolized Δ8-THC.

TOXICOLOGICAL MECHANISMS OF HEMP-DERIVED CANNABINOIDS

Most mechanisms of cannabinoid vaping product-mediated toxicity have been examined in the context of EVALI. The US Centers for Disease Control and Prevention (CDC) began an investigation into EVALI in August 2019. An outbreak of EVALI cases continued to unfold throughout that year with hospitalized cases reported to the CDC peaking at 237 in a day in September 2019.⁸⁸ As of February 2020, when the CDC stopped collecting case data, there had been 2,807 confirmed hospitalized cases including 68 deaths.⁸⁸ Despite the removal of VEA from many products, EVALI cases continue to occur, and some evidence suggests that COVID-19 cases—which share very similar clinical symptoms and imaging findings—may have obscured ongoing cases of EVALI.^89,90 Most products used by EVALI patients contained cannabinoids and within these, 16% were CBD vaping products.⁹¹ While case reports on the EVALI outbreak did not explicitly track Δ8-THC use, 82% of affected individuals reported using THC-containing vaping products.^91,92 Notably, many of the most frequently used brands also market Δ8-THC vape products and analysis of vaping products associated with EVALI cases identified Δ8-THC in 16–27% of tested samples.^91–94 In addition to CBD and THC, VEA was identified as a potential causative agent in EVALI. The FDA found VEA in 50% of THC products used by patients, often at high concentrations, and CDC analysis detected VEA in bronchoalveolar-lavage fluid from 94% of EVALI cases but none of the healthy controls.^3,4

Vitamin E Acetate Associated Mechanisms of Lung Injury.

Following the identification of the association between EVALI and VEA, there have been multiple studies on the pathogenesis of VEA in the lung. VEA is common in consumer products, especially dietary supplements and skin creams and is believed to be safe when ingested or applied dermally.⁴ However, it is possible that VEA may be damaging to the lung specifically or may be altered through heating in a vaping device. One theory is that as a tocopherol, VEA may cause phosphatidylcholines in the lung plasma membrane to transition from a gel to a liquid crystalline phase, disrupting the surface tension needed for respiration.^4,95,96 This could theoretically result in the acute to subacute lung injury seen in EVALI.

A secondary hypothesis is that heating VEA in vaping devices produces other toxic substances. Models predict that VEA decomposes into several highly reactive intermediates when heated to temperatures reached during “dry hit” vaping including 4-acetoxy-2,3,5-trimethyl-6-methylene-2,4-cyclohexadienone (ATMMC), duroquinone (DQ), duroquinone methide (DQM), and ketene, which is known to be acutely toxic.^97–99 In one study, the activation of genes associated with oxidative damage, including NQO1 and HMOX1, due to vaped VEA was larger than from duroquinone alone,¹⁰⁰ indicating that duroquinone may not be the only contributor to the adverse effects of VEA. Additionally, ketene has been observed in vaped condensates of cannabinoid acetates and one commercial Δ8-THC product.¹⁰¹

Despite the identification of VEA in products used by EVALI patients and potential mechanisms of toxicity due to VEA thermal degradation products, a definitive mechanism of EVALI has not been established. In addition to those that have already been considered, other cannabinoid vaping product constituents may be responsible for or contribute to EVALI pathology, especially as the CDC only evaluated known priority toxicants in their study. These substances could include terpenes, heavy metals, and flavoring additives. Cannabinoids themselves, including CBD and Δ8-THC, or vaping-induced cannabinoid oxidation products may also play a role in EVALI.

Cannabinoid Quinone Associated Mechanisms of Lung Injury.

Despite the identification of VEA in products used by EVALI patients and potential mechanisms of toxicity due to VEA degradation products, a definitive mechanism of EVALI has not been established. VEA was not detected in all products and patient samples analyzed and the CDC and FDA only evaluated known priority toxicants, likely missing other potential candidate causative agents. A wide spectrum of hazardous emissions, including metals¹⁰² and volatile carbonyls,¹⁰³ are also recognized as potential toxicants in cannabinoid vaping products. One emerging hypothesis is thermal oxidation of cannabinoids to reactive cannabinoid quinones which could exert toxicity through covalent modification of proteins or oxidative damage (Figure 4).^104–106 The respiratory toxicity of quinones, like CBDQ, is mediated through oxidative stress and by Michael adduction to cysteine residues, which disrupts protein structure, cellular redox state, and inflammatory signaling.^106–109 These pathways mirror the pathological processes observed in EVALI, supporting the plausibility that electrophilic and redox-active agents, via cysteine adduction, can contribute to acute lung injury seen in vaping-related syndromes.

Figure 4.

Proposed oxidation of CBD and Δ8-THC to quinones CBDQ and Δ8-THCQ and formation of quinone protein adducts through Michael addition.

CBD and Δ8-THC have low oxidation potentials of approximately 1.2 and 1.0 V, respectively, due to their electron-rich aromatic rings. The electron-rich rings of cannabinoids are likely to readily oxidize by a free radical chain reaction, which is a long chain sequence only requiring one initiating event, to electrophilic quinones CBDQ and Δ8-THCQ (Figure 4). Initiating pro-oxidant species can remove a phenolic hydrogen from the A ring¹¹⁰ of cannabinoids generating a phenoxyl radical. Reaction with an oxidizing metal such as Fe(III) or Cu(II)¹¹¹ and subsequent loss of a proton also generates a phenoxyl radical on the A ring. Oxidation to a quinone may then occur through the addition of oxygen to the phenoxyl radical. Heated aerosolization during vaping has the potential to contribute to or accelerate this oxidative transformation. Furthermore, initiating species may be present in cannabinoid vaping products as additives or contaminants.

CBDQ has been found after long-term storage of commercial and lab-made CBD e-liquids¹¹² and in high concentration distillate-based CBD vaping products (CBDQ concentration: average = 14 mM, maximum = 45 mM).^47,113 Generation of CBDQ after vaping CBD distillates has been found to be variable across products;^47,113 however, vaping-induced formation of CBDQ has since been independently replicated by another group.¹¹⁴ The variability in vaping-induced CBDQ observed across products, as well as the lack of CBDQ found in lab-made confections containing only CBD and PG/VG,^47,113 suggested the possibility that pro-oxidant initiators in the commercial products were responsible for the generation of vaping-induced CBDQ. While specific pro-oxidant initiators have not yet been identified, candidates with low oxidation potentials include metals, either derived from device coils or leftover from the manufacturing process, and Furaneol, 4-hydroxy-2,5-dimethyl-furan-3-on, known as strawberry or pineapple furanone, which has a low oxidation potential and has been implicated in the initiation of some cellular metal-dependent radical processes.¹¹⁵

While an initiator is required for CBD free radical chain oxidation, the amount of vaping-induced quinone formation in vaping products will depend on the concentration of CBD, the presence and concentration of other volatile compounds in the product and the configuration of a particular vaping apparatus. The rate of oxidation during vaping d[CBD]/dt, is directly proportional to CBD concentration so distillates, which are essentially neat CBD oils, are particularly vulnerable to oxidation in storage and throughout the vaping process. By the same token, any CBD diluent present in a commercial product, on the heating coil or in the condensate will reduce the rate of CBDQ formation during the vaping process. Thus, disposables or juices diluted with PG/VG having CBD levels as low as 15 mg/mL will undergo oxidation much less readily than distillates, given the same initiation and termination processes.

The possibility of stabilizing CBD distillates, juices and disposable products during storage and use has not been systematically examined. Metal chelating additives or antioxidants might reduce the rate of CBDQ formation. But CBD is a phenolic compound with a rate constant for H atom transfer to a peroxyl radical on the same order as anti-oxidants such as BHT or BHA, limiting their use as effective antioxidants in this instance. Vitamin E is a better antioxidant than BHT by 2 orders of magnitude, but its use in the vaping setting brings obvious risks, given the history of VEA.^3,4 Any approach to divert free radical chains from CBD to a more reactive additive would limit CBDQ formation. But the additive will be oxidized to give potentially reactive products in the process. Vitamin E itself gives reactive quinones while terpenes and other oxidizable organics would give potentially toxic peroxidic and electrophilic products of their own. A long-standing problem in quality control of oxidizable commercial products is known as adventitious initiation, and CBD distillates are textbook examples of this problem. The substrate is in high concentration, it is a good electron donor, and it has a high rate constant for maintaining a radical chain, competitive with many antioxidants.

High levels of cannabinoid quinones in commercial vaping products and additional accumulation following vaping is significant as quinones can exert toxicity through multiple mechanisms, primarily modification of proteins or other biological molecules and redox cycling.^104–106 Thiols represent the most commonly modified residues, although amine groups, which have much lower quinone reaction rate constants, in the form of lysine and arginine residues can also be modified.^104,105 Cystine modification by quinones can result in protein dysfunction, including modifications to structure, function, and irreversible inhibition.^105,116 Candidate proteins adducted by quinones include GAPDH, CK, papain, BSA, KEAP1, heat shock proteins, the 20/26S proteasome, AHR, SOD2, COX-2, COMT, NQO1, GST, and NF-κB. Cancer research on CBDQ (HU-331) also identified it as a likely noncompetitive inhibitor of TOP2A, stabilizing its N-terminal clamp and inhibiting ATPase activity, suggesting covalent adduction of cysteine residues.^117–121 Using click chemistry, CBDQ was recently shown to form adducts with KEAP1 and TOP2A in differentiated human bronchial epithelial cells exposed to CBD vaping products.^47,113 Transcriptomic and functional responses, including KEAP1-Nrf2 activation, cell-cycle downregulation, and reduced proliferation, further supported adduction of these targets.^47,113

Quinones exert toxicity not only through protein modification but also by generating ROS via redox cycling. They are reduced by P450/NADPH oxidoreductase to semiquinone radicals, which are reoxidized by oxygen to form superoxide.^105,122 This produces H₂O₂ and subsequently hydroxyl radicals via metal catalysis.^105,122 Quinone structure influences redox-cycling potential, with extended rings and electron-rich groups enhancing radical stability. The resulting ROS cause both indiscriminate and selective molecular damage.^105,122 While there are natural detoxification mechanisms, such as reduction of quinones back to hydroquinones by NADPH or NQO1, a strict balance of oxidation and reduction is normally maintained by cells and disruption to this balance can contribute to many pathologies and diseases.¹²² Therefore, redox cycling and formation of hydroxyl radicals or other ROS presents an additional mechanism of cannabinoid quinone toxicity and should be investigated in future studies, especially in THC products for which there are limited data on reactive electrophilic quinone formation.

RECOMMENDED RESEARCH DIRECTIONS

This review of the existing literature on hemp-derived cannabinoid vaping products reveals several important research gaps. Addressing these gaps is essential to better understand the potential respiratory health risks posed by these products, clarify the mechanisms of lung injury, and inform regulatory approaches to ensure public safety. The recommended research areas listed below highlight opportunities to advance the science on product composition, toxicological mecanisms, and therapeutic potential.

• Evidence of quinone formation in CBD vaping products, which are absent in clean lab-made formulations, suggests that pro-oxidant species drive oxidative stress and toxicity. Identifying and removing these species could improve product safety. As new cannabinoids (e.g., Δ8-THC, Δ10-THC, CBN, HHC, THCV, THCP, THCB, THCA) enter the market, their potential for quinone formation and related toxicity must be evaluated. Since quinones may also contribute to EVALI pathology, possibly alongside agents such as VEA, by inducing or supporting oxidative stress and protein modification.

• CBDQ has been shown to form protein adducts with key targets (e.g., TOP2A and KEAP1^47,113), but further work is needed to map adduct formation across other proteins, cysteine residues, and cell types beyond airway epithelium. It is also possible that quinone-protein modifications could occur in extracellular fluid.¹²³ Overall, future work should prioritize identification of specific cysteine residues that are modified¹²⁴ to uncover where adduction could significantly alter protein function. Beyond protein modification, quinones can also bind DNA bases,^122,125 and while less likely than cannabinoid quinones adducts, this possibility warrants investigation.

• CBDQ and other quinones react with glutathione (GSH),^106,120,126 but other thiol-containing metabolites may also be modified. Chronic exposure could disrupt redox homeostasis, amino acid metabolism, and cellular signaling, contributing to oxidative stress and metabolic dysregulation. Quinones redox cycle and form adducts, driving oxidative stress—a key toxicity pathway. CBD aerosols and CBDQ exposure both trigger oxidative stress,^73,76,77 suggesting CBDQ may contribute to vaping toxicity. Further study is needed on oxidative stress and mitochondrial dysfunction.

• Preclinical animal and in vitro studies are largely acute, yet chronic exposure better reflects real-world use¹⁶ and may reveal mechanisms of fibrosis, tissue remodeling, and oxidative stress. Importantly, no clinical studies exist on cannabinoid vaping health outcomes. Research is needed to establish exposure biomarkers, determine tissue dosing and dosimetry (to facilitate extrapolation of preclinical studies), assess lung function, and identify early injury indicators in human users.

• Current regulatory loopholes from the 2018 Farm Bill’s definition of hemp have enabled the sale of psychoactive hemp-derived cannabinoids like delta-8 and delta-10 THC with minimal federal oversight, allowing products to be sold without consistent testing, restrictions on additives, or labeling,¹²⁷ thus exacerbating potential toxicities and health risks. While federal and state efforts to close these gaps—such as redefining hemp to include all THC isomers—have faced political challenges and industry pushback, the regulation of hemp-derived cannabinoids remains inconsistent and vulnerable to exploitation. Knowledge on whether and how current regulatory loopholes can exacerbate quinone-related risks should factor into these conversations.

• Finally, the therapeutic potential of cannabinoid vaping products remains unclear. Studies should determine how heating influences toxicant formation, identify conditions that optimize safety and efficacy, and clarify whether any inhaled cannabinoids offer clinical benefit.

CONCLUSION

Cannabinoid vaping products may pose a significant public health risk due to their rising popularity, extremely high cannabinoid concentrations, contaminants, harmful byproducts generated from heated aerosolization, and their association with EVALI. Hence, a comprehensive understanding of the diverse toxicological mechanisms that may contribute to respiratory injury from cannabinoid vaping products is critical for public health and regulatory oversight. Studies on cannabinoid vaping have revealed multiple mechanisms of lung toxicity, including inflammatory responses, oxidative stress, and damage from byproducts such as heavy metals and flavoring chemicals. Additionally, VEA has been implicated in EVALI cases, and several mechanisms of VEA-mediated lung injury have been suggested. Emerging evidence suggests that reactive cannabinoid quinones may also contribute to lung toxicity through multiple mechanisms, including protein modification and oxidative stress. Given their potential to disrupt diverse cellular pathways, quinones represent an important area for future research to better understand their role in the health risks associated with cannabinoid vaping products. Future research should prioritize clarifying both the therapeutic potential and the toxicological effects of cannabinoid vaping products, focusing on chronic exposure models, mechanisms of lung injury, and the role of cannabinoid quinones and other harmful byproducts in order to inform regulatory decisions and guide the development of safer products.

METHODS

A systematic search was performed in PubMed between January and April 2025 to identify literature published on vaping or aerosolized hemp-derived cannabinoids. Search strategies combined cannabinoid-related and vaping-related terms, and additional studies were identified through manual screening of reference lists. Eligibility criteria included peerreviewed English-language articles and authoritative U.S. agency reports addressing product composition, device characteristics, contaminants, aerosol/exposure methods, respiratory outcomes, or inhalation-relevant mechanisms. Records were excluded if they addressed only oral or topical administration, lacked a vaping or aerosol component, or involved nicotine-only devices.

Example Queries.

• (“cannabidiol” OR CBD) AND (vape* OR “e-cig*” OR aerosol* OR “electronic cigarette*”)

• (“delta-8 THC” OR “Δ8-THC” OR “delta 8 tetrahydrocannabinol”) AND (vape* OR “e-cig*” OR aerosol*)

• (quinone OR “cannabinoid quinone” OR CBDQ) AND (vape* OR aerosol* OR condensate) AND (adduct* OR “oxidative stress” OR “redox cycling”)

• (“vitamin E acetate” OR VEA) AND (vape* OR “e-cig*” OR aerosol*)

• (terpene* OR “flavoring chemical*” OR “heavy metal*” OR pesticide*) AND (vape* OR aerosol*)

ABBREVIATIONS

CBD

cannabidiol

CBDQ

cannabidiol quinone

CBN

cannabinol

CB1

cannabinoid receptor type 1

CB2

cannabinoid receptor type 2

CSA

controlled substances act

EVALI

E-cigarette or vaping, product use-associated lung injury

GSH

glutathione

HHC

hexahydrocannabinol

KEAP1

Kelch-like ECH-associated protein 1

Nrf2

nuclear factor erythroid 2–related factor 2

HBEC

primary human bronchial epithelial cell

PG

propylene glycol

ROS

reactive oxygen species

THC

tetrahydrocannabinol

THCA

tetrahydrocannabinolic acid

THCB

tetrahydrocannabutol

THCP

tetrahydrocannabiphorol

THCV

tetrahydrocannabivarin

TOP2A

topoisomerase II α

CDC

U.S. Centers for Disease Control and Prevention

DEA

U.S. Drug Enforcement Administration

FDA

U.S. Food and Drug Administration

VG

vegetable glycerin

VEA

vitamin E acetate

Δ8-THC

Δ-8-tetrahydrocannabinol

Δ8-THCQ

Δ-8-tetrahydrocannabinol quinone

Δ9-THC

Δ-9-tetrahydrocannabinol

Δ9-THC-O-Acetate

Δ-9-tetrahydrocannabinol acetate

Δ10-THC

Δ-10-tetrahydrocannabinol

Biographies

Charlotte Love recently completed her doctorate in Toxicology and Environmental Medicine at the University of North Carolina at Chapel Hill. Her research examines the pulmonary health effects of aerosolized cannabinoids and their quinone oxidation products, using primary human airway models, transcriptomic profiling, and chemical biology approaches to investigate mechanisms of lung injury. She previously conducted toxicology research at the National Institute of Environmental Health Sciences on PFAS-associated developmental outcomes, at Rutgers University on nanoparticle inhalation during pregnancy, and at the Gulf Coast Research Laboratory on oil spill chemical exposures.

Ilona Jaspers has over 25 years’ experience in analyzing the toxicity of inhaled chemicals in the context of ambient air pollution and use of consumer products, such as vaping products. Using organotypic in vitro models and human clinical studies, she has investigated the effects of vaping products and specific components, such as flavorings, on respiratory host defense dysfunction. She has published over 50 papers on the mechanisms, toxicity, health effects, and public health consequences of vaping products and contributed to workshop reports and perspectives on these topics.

Hye-Young Kim has expertise characterizing electrohphile products generated from free radical damage during oxidative stress. These oxidation products are reactive toward proteins and DNA, subsequently altering their cellular and molecular functions. Her studies make it possible to determine the protein targets of these electrophiles in an unbiased manner to examine the biological consequences of adduction. She has extensive knowledge of LC-MS/MS and its use to analyze species from small molecules to macromolecules such as proteins. These methods are used to identify target proteins and their modification sites to provide the guidelines for reactivity and selectivity of electrophile adduction.

Ned Porter has authored papers on free radical reactions over five decades. His studies on lipid peroxidation have established fundamental pathways for these important free radical chain reactions. He has established mechanisms for the formation of an array of products formed in peroxidative processes, and he has developed methods for the isolation and identification of protein adducts formed with electrophilic species generated in free radical reactions of natural compounds. Among the electrophilic protein adducts studied have been those formed from 4-hydroxynonenal, a product of polyunsaturated fatty acid oxidation, secosterols A and B derived from cholesterol free radical oxidation and ozonolysis, and quinones derived from electron-rich cannabinoids.