PROJECT RESULTS

Developing Limits for THC in Products From Hemp Seeds

Applicant: 

Gero Leson
Leson & Associates
Berkeley, California  U.S.A.  94709

Table of Contents:

• Background and Objectives
• Procedure and Project Activities
• Results and Discussion
• Conclusions

ARDI Project:

#01-482

Total Approved:

$5,400

Date Approved:

February 7, 2003

Project Status:

Completed March, 2003

Background and Objectives:

In recent years, the presence of trace residual delta‑9‑tetrahydrocannabinol (THC) the main psychoactive ingredient of marijuana, in food products incorporating hemp seed and seed derivatives (whole and hulled seeds [also called hemp nuts], oil, flour, meal, and protein isolate) has raised concerns over THC's potentially adverse impacts on human health.

The study's main objectives were to: 

1. Prepare a hazard assessment for the intake of THC via hemp foods, including the establishment of Lowest Observed Adverse Effects Level (LOAEL) and No Observed Adverse Effect Levels (NOAEL) for oral ingestion of THC, derivation of safety factors, and estimation of the corresponding acceptable daily intake (ADI).

2. Develop an exposure assessment for the intake of THC via hemp foods, assuming their extensive daily consumption with trace residual THC levels now commonly achieved by Canadian suppliers in hemp seed and seed derivatives (less than: 2 µg/g for whole seeds, meal and flour; 1.5 µg/g for hulled seeds and protein powder; 5 µg/g for hemp oil).

3. Assess whether THC intake may, under highly conservative assumption of consumption patterns, present the potential for adverse health effects.

4. Comment on the acceptability of currently achieved THC levels in hemp seed and seed derivatives as health‑based THC limits.

Procedure and Project Activities:

The hazard assessment for oral THC uptake was based on a current, critical review of the original scientific literature on the subject, including several recent studies and previous reviews, such as the Health Risk Assessment conducted by Health Canada in 1998/99.

The exposure assessment used statistical data on food consumption by North Americans.  Daily food intakes by food group were estimated based on the most recent United States Department of Agriculture's (USDA) Continuing Survey of Food Intakes by Individuals (CSFII) 1996.  The potential for substitution of conventional foods by hemp foods and the typical THC content in hemp seed derivatives was established based on previous Canadian reports, industry sources and analytical data.  Several food intake scenarios were evaluated.  Scenario I assumed an average recommended daily caloric intake with complete replacement of protein by hemp protein, without further differentiation of hemp products.  Scenario 2 assumed a typical North American diet in which all food items, except meat, were completely replaced by technically feasible hemp foods.  Scenario 3, the "reasonable worst case", assumed a high caloric intake by a vegetarian with complete replacement of animal protein by hemp protein.

Results and Discussion:

The hazard assessment provided the following major findings and conclusions:

• The lowest single oral THC dose, at which acute adverse neurological effects, i.e. slightly reduced psychomotoric performance, have been observed, is 5 mg (for a body weight of 70 kg).  This dose represents the LOAEL for acute effects caused by THC.

• The same single dose of 5 mg also did not cause a difference to a placebo with respect to psychotropic effects and thus constitutes the NOAEL for this effect.

• Adverse chronic effects, such as cognitive changes, structural brain changes, mutagenicity, carcinogenicity, significant changes to hormone levels in males and females, congenital effects, and adverse impact on child development were either not found in humans or were found only at doses significantly higher than the equivalent of oral doses of 10 mg/day, in which cases observed effects were moderate.

• The relevance of animal studies, which found increased risk of stillbirth and other adverse effects on the fetus following peritoneal injection of THC, to humans, is highly questionable.  No such effects had been found with humans after oral or inhalative administration of much higher doses.  The same applies to the reported impact of low THC doses on hormone levels in pregnant rats.  These studies had been the basis for the conclusion by Health Canada's 1999 draft risk assessment that inadequate margins of safety exist to protect the population from the assumed neuroendocrine disruption caused by THC.

• Since chronic and subchronic adverse effects require doses higher than those for acute neurological effects, determination of an acceptable daily intake should be based on the LOAEL for reduced psychornotoric performance of 5 mg for a single dose, or 2 x 5 mg, taken orally over the course of a day.  Considering that the observed psychomotoric effects are not severe and according to scientific practice, selection of a safety factor of 20 provides a sufficient margin of safety from acute adverse neurological effects.

• Based on the above, an acceptable daily intake (ADI) for orally ingested THC of 500 µg/day was assumed to provide protection from both acute and chronic adverse effects to humans.

• Less efficient transfer to the fetus and suckling infant of THC orally ingested by the mother, compared to inhaled THC, provides additional protection to both by limiting THC uptake.  Children also appear to be less susceptible to THC compared to adults.  Thus, the proposed ADI appears to provide sufficient protection to both fetuses and children of mothers who routinely consume hemp foods.

• The accumulation of THC in body tissue represents a source of THC to the plasma even after cessation of THC uptake.  The establishment of a dynamic equilibrium between accumulation and remobilization and the slow rediffusion process indicate that corresponding THC levels in body tissue will be insufficient to supply THC to the plasma at rates that could result in or contribute to adverse effects.

• Other cannabinoids present in industrial hemp in relevant quantities appear to be effective either at much higher concentrations than THC, e.g., cannabinol (CBN) or may act as an antagonist to the neurological effects of THC, as with cannabidiol (CBD).  Regulating the THC level in hemp raw seed and seed derivatives (whole and hulled seeds, oil, flour, meal, and protein isolate) thus would provide comprehensive protection from potentially adverse health effects caused by ingestion of hemp food products.

The exposure assessment for THC intake via hemp foods generated the following major conclusions:

• Complete replacement of conventional food items in a "typical American diet", including meat products, by currently available hemp food items containing common levels of THC will, even under "reasonable worst‑case" Scenario 3 assumptions, not cause a daily THC uptake via hemp food in excess of 500 µg.  This reasonable worst‑case scenario makes the following assumptions:

• Complete substitution of all meat and non‑meat food items by hemp foods, wherever technically feasible.

• A high daily caloric intake at the 95th percentile of the United States population (3182 kcal/day).

• The use of the maximum technically conceivable hemp content in all food products, irrespective of the higher relative cost of hemp seed ingredients.

• The more realistic typical daily THC uptake by individuals who consume hemp food items regularly and extensively will rarely exceed the lower level of Scenario 2, i.e. 100 µg/day.  This implicitly assumes increased future commercial availability of these items and the maintenance of the current THC levels.

• The corresponding range of daily intake of cannabidiol (CBD) in Scenarios 2 and 3 is estimated at 1 to 5 mg, respectively.

• Consequently, the daily THC ingestion even by extensive users of hemp foods will remain consistently and, in general, significantly, below the proposed ADI for oral THC, and thus will not cause any acute or chronic adverse health impacts.  Specifically, the highest conceivable intake of THC via hemp foods is far below the psychoactive threshold for THC.

Generally achieved THC levels in hemp seed derivatives thus represent a conservative choice for achievable and enforceable THC limits in these materials.

The estimated 10‑20% contribution by the two non‑psychoactive THC acids A and B to total THC in hemp seed derivatives, predominantly measured by gas chromatography/mass spectrometry (GC/MS), provides an additional small margin of safety from potentially adverse effects of THC.

THC uptake from the use of hemp oil cosmetics is still lower than from hemp food ingestion.  A recent study estimated that exclusive and extensive use of hemp oil cosmetics containing high amounts of hemp oil, or pure hemp oil, on compromised skin will not contribute more than 10 µg/day to total THC uptake.  Typical THC uptake from the extensive application of commercially available hemp oil cosmetics to healthy skin is typically less than I µg/day.  Thus, compared to hemp foods, hemp cosmetics do not contribute significantly to total THC intake.

Extensive hemp food consumption also no longer appears to have the potential for causing confirmed positive urine tests for marijuana.  A recent study showed that daily THC ingestion with hemp oil, in single doses of up to 600 µg/day and over a 40‑day period, failed to cause confirmed positive urine test according to the protocol used by most public and private employers in the United States.  Positive screening tests at a lower cutoff level are conceivable but unlikely.

Little representative information on the content in hemp seed derivatives of cannabinoids other than THC, notably cannabidiol (CBD) and cannabinol (CBN) is currently available.  It is estimated that CBD intake is typically 10 times that for THC.  CBD is considerably less pharmacologically active than THC.  Studies suggest that typical CBD intake via food is far too low to cause measurable effects on humans.  Findings of low‑dose adverse effects of CBN on the hormone secretion of male rats are contradicted by human studies at higher doses.  Thus, uptake via hemp foods of other relevant cannabinoids does not appear to pose the risk of adverse health effects.  However, this subject requires further study.

Conclusions:

ARDI Home - Contact ARDI