Synthetic Cannabinoids Found in Legal Cannabis Products

A product can be sold through a legal-looking website, arrive in polished packaging, and still contain a cannabinoid profile that bears little resemblance to its label. That is the central warning from a new Italian study¹ examining 26 low-THC cannabinoid products purchased online.

Researchers found that 24 of the 26 products, or 92.3%, were incorrectly or incompletely labelled. The problems extended beyond modest differences in advertised CBD or THC potency. Several products contained undeclared semi-synthetic cannabinoids, while four contained fully synthetic cannabinoids. In three cases, researchers detected EDMB-4en-PINACA, a relatively new synthetic compound that was not disclosed on any product label.

The sample was small and limited to products available through Italian websites, so the findings do not establish that 92.3% of all European cannabinoid products are mislabelled. They do, however, reveal a regulatory blind spot with implications well beyond Italy. A label built around familiar terms such as CBD, hemp, or low THC may no longer describe the actual chemical complexity of the product.

Study Finds Widespread Mislabeling In Legal Cannabis Products

The researchers purchased cannabis flowers, resins, and vape liquids between September 2024 and January 2026. These products were marketed as low-THC or otherwise legal cannabinoid products serving the Italian market.

Testing combined several analytical techniques. Gas chromatography with flame ionization detection was used to measure CBD and delta-9 THC. Gas chromatography-mass spectrometry was used to detect synthetic and semi-synthetic cannabinoids and estimate their abundance. The researchers also used ultra-high-performance liquid chromatography with high-resolution mass spectrometry to support or confirm some identifications.

This multi-method approach matters because no single routine test necessarily captures every emerging cannabinoid. Closely related compounds can produce similar signals, while low concentrations, matrix effects, thermal changes, and the absence of reference standards can complicate confirmation.

Why CBD And THC Numbers Tell Only Part Of The Story

All samples contained CBD, with a median concentration of 7.7%. Delta-9 THC had a median concentration of 0.4%, and two products exceeded the study’s referenced Italian threshold of 0.6%.

CBD content was incorrectly represented in 14 products. Seven did not disclose CBD even though it was detected, four contained less CBD than claimed, and three contained more. Delta-9 THC was misreported in 13 products. Nine contained more than declared, two contained less, and two did not declare THC despite its presence.

Potency discrepancies matter because dose expectations shape consumer decisions. They can also determine whether a product falls on one side or the other of a legal threshold. A buyer may believe a product is non-intoxicating or compliant with local rules when its measured composition suggests otherwise.

Yet the larger issue is that the traditional CBD-versus-THC framework is becoming outdated. Manufacturers can chemically modify plant-derived cannabinoids into new substances, while fully synthetic cannabinoids can be applied to flower or included in other formulations. A product marketed through the language of hemp may therefore contain molecules with limited human safety data and effects that cannot be inferred from the stated CBD concentration.

Semi-Synthetic Cannabinoids Complicate Product Safety

Semi-synthetic cannabinoids generally begin with a naturally occurring cannabinoid such as CBD or THC that is chemically altered. Fully synthetic cannabinoids are built through different chemical routes and can have structures and pharmacological properties that differ substantially from plant cannabinoids.

H4CBD was the most frequently detected semi-synthetic cannabinoid in the study, appearing in six products. Delta-9 THCP appeared in five, while delta-8 THCP appeared in four. The researchers also found compounds including HHC, HHCP, HHCPM, and oxygenated derivatives.

These findings do not mean every detected compound was deliberately added. Semi-synthetic manufacturing can produce reaction by-products, and degradation or storage conditions can alter chemical profiles. In some products, however, the relative concentration made accidental trace contamination a less convincing explanation.

The distinction is important for regulators and laboratories. A rule that bans one named molecule may be overtaken by a closely related substitute. A testing panel designed around yesterday’s cannabinoids may miss tomorrow’s products. The market can change faster than reference standards, toxicological research, and legislation.

What A More Useful Product Standard Would Require

A stronger system would evaluate the complete cannabinoid profile rather than checking only whether delta-9 THC falls below a single threshold. At minimum, that system would need:

• Batch-specific testing by an independent laboratory
• Disclosure of detected intoxicating and modified cannabinoids
• Methods updated as new substances reach the market
• Traceability from raw material through the finished product

A certificate of analysis is useful only if the tested batch matches the product being sold, the laboratory looks for relevant compounds, and the reporting threshold is low enough to identify meaningful contamination. A clean result from a narrow panel is not proof that an emerging cannabinoid is absent.

Undeclared Synthetic Cannabinoids Raise The Stakes

The most concerning result was the detection of fully synthetic cannabinoids in four products. EDMB-4en-PINACA appeared in three flower samples, with an estimated median amount of 1.93% and a maximum of 4.45%. MDMB-4en-PINACA appeared in one flower sample at an estimated 1.84%. Neither substance was declared.

The authors identified MDMB-4en-PINACA as a potent synthetic cannabinoid and cautioned that EDMB-4en-PINACA could carry adverse health risks. The study did not test these products in people, establish clinical outcomes, or directly measure the potency of every detected compound. Its safety signal comes from undisclosed exposure to pharmacologically uncertain substances, not from a clinical trial demonstrating a particular rate of harm.

This is precisely why label accuracy is more than a marketing issue. Consumers cannot make an informed choice if the active ingredients are missing from the package. Clinicians and poison-control specialists may also struggle to interpret symptoms when a patient believes they used CBD but was exposed to a synthetic cannabinoid.

There is a legal dimension as well. Undeclared THC above a jurisdictional limit or the presence of a controlled synthetic cannabinoid could expose a consumer to consequences despite an apparently lawful purchase. The boundary between legal and illegal cannot reliably protect buyers when product composition is unknown.

Why Regulators Need Surveillance Instead Of Static Lists

The study points toward a regulatory model closer to infectious-disease surveillance than conventional product registration. Authorities need continuous market sampling, shared spectral libraries, rapid alerts, and a way to regulate families of related compounds when evidence supports it.

That does not mean every minor cannabinoid should automatically be prohibited. It means regulators need enough analytical visibility to distinguish naturally occurring constituents, manufacturing by-products, deliberately added semi-synthetic compounds, and potent synthetic receptor agonists. Without that visibility, policymakers are regulating labels rather than products.

Manufacturers also have an incentive to improve. Accurate, batch-level disclosure would help legitimate producers separate themselves from sellers operating in regulatory grey zones. Standardized testing could become a competitive advantage, particularly as consumers become more aware that familiar hemp terminology does not guarantee familiar chemistry.

What The Italian Cannabis Study Does And Does Not Prove

The findings are compelling, but their boundaries matter. Only 26 products were analyzed, purchasing was limited to websites serving Italy, and the sample was not designed to represent every legal cannabinoid product. Four samples also lacked enough material for high-resolution mass spectrometry.

Concentrations for synthetic and semi-synthetic cannabinoids were semi-quantitative estimates. The researchers assumed broadly comparable mass-spectrometry responses between CBD and the other compounds, even though different structures may produce different responses. Some emerging substances also lacked analytical reference standards, reducing the certainty of those identifications.

Those limitations prevent broad claims about prevalence or exact exposure. They do not erase the verified central problem: products openly marketed as legal contained unexpectedly complex mixtures, and nearly all had at least one label failure under the study’s criteria.

The practical lesson is straightforward. Legal availability, low-THC branding, and professional packaging are not substitutes for verified composition. As cannabinoid chemistry evolves, product safety will depend increasingly on whether testing systems can identify what is present, not merely confirm what the manufacturer expected to find.

References:

^{1. Giorgetti, A., Caprari, C., Pelletti, G., Rossi, F., Pulver, B., Cannazza, G., Pascali, J. P., & Auwärter, V. (2026). Forensic profiling of synthetic and semi-synthetic cannabinoids in legal cannabis products in Italy. Forensic Science International, 113114. https://doi.org/10.1016/j.forsciint.2026.113114}