The Future of Cannabis Testing

Cannabis products are becoming accepted as a medical treatment and a mainstream recreational product, with dozens of countries allowing legal or decriminalized use. Reliable analytical methods are critical to ensuring consistent dosing, efficacy, and safety.

The current state of play 

In the United States, cannabis remains classified under Schedule I of the Controlled Substances Act, precluding U.S. Food and Drug Administration (FDA) from research and regulatory oversight. Medical, recreational, and “low THC, high CBD” products are analyzed under nonuniform state-specific regulations and at present only four states have no cannabis access laws. In June 2018 the FDA approved Epidiolex® (cannabidiol), a drug that contains purified CBD in a strawberry-flavored formulation agreeable to young children, for treating specific pediatric epilepsy syndromes. On September 27, 2018, the FDA reclassified Epidiolex from Schedule I to Schedule V (the least restrictive class.) All other non-FDA-approved CBD products remain on Schedule I, but various CBD-products of lower concentration and dosing are allowed by prescription in states permitting medical cannabis. The Veterans Medical Marijuana Safe Harbor Act was introduced in the U.S. Senate in September 2018, and is intended to allow veterans to legally use, possess, and transport medical cannabis and to discuss its use with physicians in the Department of Veterans Affairs, for the treatment of chronic pain and post-traumatic stress disorder, notwithstanding state laws. The U.S. Senate is advancing the 2018 Farm Bill with language that allows hemp and hemp products. Industrial hemp, where only the fiber and/or seed fatty acids are sought, and with flowers that contain mostly CBD and little THC, can warrant testing that verifies the absence of THC.

Various sites aggregate cannabis regulations, including details of what substances must be tested for, and it is difficult to stay up-to-date for all jurisdictions, but if you bookmark one site, make it the National Conference of State Legislatures (NCSL). Anyone seeking information on analytical testing and other regulations in a specific state should examine the official laws of that state. The patchwork of state regulations leads to great variability. For example, the sampling protocol of raw flowers allows Washington growers to simply select and send their representative samples to a testing laboratory, while Oregon law requires the testing laboratory to go onsite and select the samples. 

As few certified methods exist, currently, each testing laboratory develops its own methods and incorporates the limit of detection (LOD) for the procedure. If the action level for a specific chemical is, say, 5 μg/g and a laboratory’s LOD is 100 μg/g, that laboratory cannot reasonably analyze for that chemical. Many agree that it is reasonable to require each testing laboratory to maintain state-certified competence to ensure that their methods are fit for purpose and performed competently. Particularly, as these labs face many novel regulatory challenges. Unless the United States adopts a national cannabis policy, it is reasonable to assume state-specific laws will be around for a while and communication between industry and state government officials can only be good. The critical period in crafting state regulations is soon after recreational or medical cannabis becomes legal, but it is very important to keep abreast of proposed changes and provide comment when appropriate. After a few years, laws and bureaucracy are established and difficult to change – therefore many argue it is best is for the industry to be proactive and at the table during rule-making. In the U.S. at least, maintaining cannabis on Schedule I remains the biggest hurdle confronting the development of standardized testing methods.

On October 17, 2018, recreational cannabis became legal in Canada, but Health Canada has yet to provide uniform testing requirements for cannabis products. The biggest difference between Canada and the United States is that Canadian law provides nationwide regulation and licensing of most aspects of cannabis. The Cannabis Act can be found here. Local jurisdictions will control the sale of cannabis products, with laws similar to tobacco and alcohol for restrictions on age, location, and advertising. Canada allows for new classes of licenses, e.g., micro-class for nurseries, producers, and processors, along with the existing classes that also include research, analysis, and industrial hemp. Previously, most large producers focused on similar products and the new Canadian micro-class regulations will provide space for innovative growers and processors to create specific strains and end products.

Challenges in developing reliable analytical methodologies

Hundreds of phytochemicals can be found in cannabis and the amounts of each can vary widely between strains. Quantitative analysis of pure substances is straightforward. In cannabis, the kernel is to collect representative samples, extract and separate the substances of interest, measure the amounts present, and relate the results back to the original form. The sample preparation is therefore quite different when extracting THC from a flower or a chocolate bar, for example. For method development, the question is “Was the extraction step 100% effective?”. Raw flowers contain high concentrations of THC in a cellulosic-like matrix, whereas a chocolate bar has a much lower concentration and in a matrix of fat and sugar. After extraction, quantitative analysis of the THC should be the same, except that analytical techniques are not always linear over the entire concentration range found in raw cannabis and products. Any analytical method should employ standards that are similar to the test specimen, which dictates that a THC standard used to analyze raw flower will require quantitative dilution, for example, to be used to test an edible product that is 1% THC. If smoking were the only delivery system, this would be less of a problem, but the huge variety of delivery vehicles requires accurate dosage testing from all formulations.

Dozens of other compounds with proven and potential physiological effects are also found in cannabis. Including terpenes, volatile hydrocarbons that have a strong odor that can be categorized as fruity, nutty, etc., and terpenoids, oxidized forms of terpenes. And, flavonoids which are in a class with vitamins and antioxidants including the substances that give berries their strong color. Pesticides, other agrochemicals, and heavy metals are also often present in low concentrations but are highly toxic and verifying their absence is often required, let alone prudent. Because terpenes are volatile, analysis by gas chromatography (GC) is traditionally used, but recently headspace solid-phase microextraction (HSPM) and GC–MS has gained popularity, with the advantages of being non-destructive and eliminating interferences from co-extracted matrices. Time-of-flight mass spectrometry (TOF-MS) can provide semi-quantitative information on an entire suite of compounds, but to determine reliable concentrations of all substances requires the methodology to specifically detect each substance. TOF-MS instrumentation prices begin at about $500k, a hefty investment for a typical cannabis analysis laboratory.

Prior to July 2018, California authorities relied on an honor system to regulate and test for pesticide contamination. The state now requires testing for 65 substances, with 21 at the non-detect (ND) level. This required California ND on pesticides (and six processing chemicals) can cause issues. Problematic on the low end is the LOD, i.e., what is the least amount that can be measured by a certain technique? If the results are “zero” is it truly because of the absence of the substance or sloppy lab work? This can give rise to a producer or extraction facility shopping for testing laboratories that will produce acceptable results. The end result is little, if any, confidence that a product offered to consumers is free of contaminants. On the high end, for the desired compounds, e.g., THC and CBD, if the measured results are incorrectly too low, the product has more than the labeled dose and if the measured results are too high, the product is substandard and can give rise to consumer complaints.

Current California cannabis regulations can be found in a 120-page document and regulations that speak to testing laboratories start on page 78, with definitions, licensing, accreditation, sampling, and standard operating procedures (sampling, etc.). Page 92, §5714, lists required testing: cannabinoids, foreign material, heavy metals, microbial impurities, mycotoxins, moisture content and water activity, residual pesticides, residual solvents, and processing chemicals, terpenoids (if applicable) and homogeneity (if applicable) and details are presented in subsequent pages. For solvents and processing chemicals, a sample is considered to “pass” if the six Category I chemicals are not detected and if the 14 Category II chemicals do not exceed the listed action levels. Similarly, for pesticides, 21 are in Category I (non-detect) and 45 (not to exceed the action level.) This begs the question “If a grower source can certify that it does not use one or more agrochemicals or solvents, why must they all be tested?”. 

Inductively coupled plasma (ICP) and atomic absorption (AA) spectrophotometry are reliable instruments for quantitative testing for heavy metals that can accumulate in cannabis: arsenic (As), cadmium (Cd), lead (Pb), and mercury (Hg) and it appears that tests for other metals need not be performed, but if your state requires testing for additional metals, ICP and AA are adaptable. Typically, metals are extracted from the matrix with acid solutions and microwave digestion, but until methods are certified, accepted, and required by state law, each laboratory has its own procedure. 

The development of analytical methods for all these testing requirements across all cannabis matrices should not be difficult, expensive, or time-consuming. According to Neal Langerman, a California-based chemical safety consultant: “Development of analytical procedures is not hard or time-consuming. Getting industry and regulatory acceptance is. That will take a few years. A concerted effort on the part of the industry could get this done in five years.”. The industry should embrace regulation. State law can require destruction of product that does not meet specification in THC or CBD, but that passes testing for all other chemicals, which begs the question, why not allow it for sale at a discount? The only way to resolve this is for the industry, in cooperation with bodies like the American National Standards Institute (ANSI), American Society for Testing and Materials International (ASTM) and government agencies to develop standard analytical procedures, including certification of standards. “Once reliable methods are established, the efficacy of the testing laboratories should be challenged by round-robin analysis cycles.” Says Langerman. Round-robin analysis is an inter-laboratory test performed independently by multiple facilities performing the test with the same method or a variety of methods and equipment, and subsequent comparison and agreement of results. With standards in place, this kind of lab proficiency testing can ensure competency and quality of testing is maintained whilst weeding out those using inappropriate or unreliable methodologies. 

Solutions to problems with existing regulations

Removal of cannabis from the FDA Schedule I would make it easier for facilities with analytical expertise, like universities, pharmaceutical research centers, and food safety testing labs, to develop reliable analytical methods. Testing methods for pesticides or Salmonella in spinach are not the same as in cannabis but it makes no sense to re-invent the same method entirely. If done properly, certified independently, and accepted by state regulators, all testing labs can be confident that a method is reliable whether it’s novel or borrowed and adapted from another industry.

More research on the specific substances in cannabis products and the consequences in human physiology are needed. For example, are there specific terpenes that affect the CB1 and CB2 receptors and should the specific ones be quantified, or is “total terpene” adequate? If specific claims are made, in particular for medical cannabis, it seems reasonable to quantify the proffered active component.

As regulatory structures mature many questions will emerge. If the absence of agrochemicals or processing solvents and chemicals can be demonstrated from inventory records, should testing for those substances be required? If testing soil samples with well-developed methods can show the absence of heavy metals, should testing for those metals be required in the product? If a grower repeatedly shows that its product is free of the 65 agrochemicals specified by California, should complete testing still be required for every batch? If a clean (free of agrochemicals and heavy metals) flower is extracted with carbon dioxide and combined with food-grade ingredients, should the final product only require testing for THC, CBD, and terpenes? 

State regulators should periodically meet and discuss what works and what doesn’t. Even if each state has their own laws, the laws could be more uniform. Examine your state regulations and if you find something that appears to be too restrictive, contact your officials and let them know why it is overly burdensome and ask if they can issue a waiver or modify the regulations to be more reasonable. Now is the time to act to ensure regulations are fair and effective.