Two important factors are combining to propel the explosive growth of cannabis products and science: First, there is huge demand-pull fueled by reports of efficacy for the management of several ailments, including chronic pain, epilepsy, PTSD, and some cancers. Medicinal cannabis was one focus of the 2017 Cannabis Science Conference, held in August in Portland, OR. The other track, analytical cannabis, covered a wide range of topics—from seed to smoke. With two parallel tracks, I decided to focus on the analytical.
Starting with seed, a panel discussed the design and management of legal industrial-scale grows. Peter Maguire of Lighthouse Worldwide Solutions (Fremont, CA) described built-for-purpose facilities, including grow buildings. Each needs to be designed with a specific purpose and optimized for the plant strain. Details are important and often crucial. For example, one speaker pointed out that different strains have different growth patterns. Thus, if factors like high THC content or particular flavor are the goal, specific strains should be selected before the facility is designed. In contrast, medicinal cannabis products are selected to favor production of cannabis diol, alcohol, acid, or any of more than 300 cannabis compounds.
Generally, one starts with the purpose and works backward to the strain, number of plants required, optimized growth conditions, and then design of the grow room. Buildings are designed for positive pressure to reduce contamination. Nearly all enrich the CO2 partial pressure to promote photosynthesis. Even then, unexpected results are too common. For example, a grow operation in Israel needed four buildings. Three were situated facing east/west. The fourth had a north/south exposure perpendicular to the other three. Production in the latter was consistently inferior to the other three. Perhaps the solar flux was more uniformly distributed for the east/west facing buildings.
After the plants mature, they must be dried, and quickly. There is less than three days to remove 50% of the water content. This reduces mold or fungus contamination. If the product is intended for medicinal or extraction products, hotter temperatures can be used, but heat drives off volatile terpenes. Terpene content is a critical quality attribute for the flavor of smoke products.
Scientists experienced in cGMP will recognize that material flow is an important consideration. Product material should flow along unique paths to the processing and packaging operation to preserve batch traceability and reduce risk of interbatch contamination. Street clothing should not be worn into the growing, drying, and downstream processing facilities. These are not new concepts, but their implementation in growing cannabis plants is new. Plus, there is a wide range of uses of cannabis products. Most will have unique quality attributes.
Laboratory operations should follow cGLP standards. The workflow should be engineered and managed to produce reliable data to support decisions. Since cannabis products are potentially pharmaceutically active, staff must treat each contact or exposure with appropriate protocols for safety.
Michael Hogan, Ph.D., of PathogenDx (Tucson, AZ) described technology to identify and control fungal and bacterial content of plant material by using PCR and DNA microarray technology. One example: Lavatories should not be situated in or near the grow rooms. Hogan was also critical of the FDA for not providing more guidance in adopting analytical protocols for plant materials. Leaving this to the individual states leads to regulatory chaos, with increased safety risk.
Mycotoxin analysis
A new mycotoxin analyzer from aokin AG (Berlin, Germany) is designed for rapid assay of mycotoxins—THC or CBD (cannabidiol). The sample matrix ranges from beverages, baked goods, candies, and other edibles. aokin’s mycontol THC is an automated fluorescence polarimeter supported by application-specific reagent kits.
Sample workflow is simple. The weighed sample is extracted in a blender and then filtered. The filtrate is centrifuged with aokin’s quickclean column. Sample and reagent vials are placed in the instrument, which automatically performs the fluorescence polarization experiment and calculates the analyte amount. Cannabis-relevant assays include THC, CBD, and aflatoxins. Percent RSDs range from 4% to 8%.
Structure/activity relationships in cannabis components
Professor David Meiri (Technion–Israel Institute of Technology, Haifa) updated his keynote of last year on treating cancers with phytocannabinoids. He showed the cannabis matrix is complex; for instance:
- Cannabis plants vary greatly in expression of individual cannabanoids. This is also true for terpenes.
- Cannabis compounds number over 100 and cover a concentration range of nondetect to ~21%, which is the THC content of flowers of some commercial strains. By class, cannabinoids can be divided into 11 groups totaling 144 entries. Terpenes number more than 100.
- Different extraction protocols of the same cannabis batch produce different cannabanoid profiles.
- Potential pesticide residues number several hundred.
- The total list of analytes of potential interest exceeds 500.
- Therapeutic doses of cannabanoids for cancer therapy can be in the low-nanogram range.
- For cancer therapy, improved efficacy with combinations of cannabanoids rather than a single cannabinoid was demonstrated.
- Professor Meiri’s work usually involved multiple patients. In contrast, many of the other reports seemed to be n = 1. I did not see any report that would qualify as a proper clinical trial, a reflection of the fragmented and underfunded nature of cannabis research today.
Analysis of cannabis products
The remainder of the analytical track consisted of a collage of reports that seemed to lack a central theme. There are at least two contributing factors. The material enters the supply chain from different sources, including individuals tending clandestine “grows” with little training in horticulture, pest control, or safety. The other source is from licensed farmers using the best available farming technology.
One potent comment by Christian Sweeney of Cannabistry Labs (Denver, CO) was “One cannot manage if you do not measure.” But there is little consensus on what, why, and how to measure. In contrast to blood alcohol, the standards for impairment by THC are not in place.
Problematic extractables: pesticide residues and heavy metals
Cannabinoids are generally nonpolar polyoxygenated compounds that can be extracted and then concentrated with nonpolar solvents including ethanol, carbon dioxide, propane, butane, and hexane. However, extraction is a complicated and potentially dangerous procedure.
As chair of the session on extraction, Dr. A.C. Braddock of Eden Labs (Seattle, WA) showed the evolution of medicinal cannabis, with an emphasis on the last 200 years in the U.S.A. In the 1850s, Lilly, Upjohn and Park Davis marketed tinctures (ethanol extracts) of marijuana for a variety of ailments, including pain relief. This continued until the Marijuana Tax Act of 1937, which criminalized possession for recreational use. Recently cannabis has been regulated by the Controlled Substances Act, which is intertwined with international treaties. These prohibitions have not worked, but the policies have given rise to a highly profitable drug trade that leads to corruption and violence and huge risks to public safety. Dr. Braddock concluded her lecture with this observation: “It is criminal that cannabis is criminal.”
Applications of Eden extractors range from bench-scale (1 liter) on up. Dr. Braddock compared butane extraction to ethanol. The latter is favored for the extraction of CBD from hemp. Eden’s Coldfinger extractor facilitates the low-temperature vacuum distillation of cannabinoids and terpenes.
Pesticides and residues
Traditionally, cannabis materials were grown secretly, often on public lands such as national forests. The growers had little formal training or concern about product safety. Psychoactive potency was the only figure of merit. So for pest control, the MO was often, “if a little is good, then a bit more is bound to be better.” Slides demonstrated visible deposits of a fungicide on plant leaves at harvest.
A lecture by Jason Strull of 374 Labs, LLC (Sparks, NV) showed that extraction of dried plants for cannabis could selectively concentrate pesticide residues in the extract. Depending on the sample and process, extraction can achieve products with residue concentration extending into the 100-ppm range.
Heavy metals
Soil for grows is a key ingredient and should be analyzed for heavy metals. It might come from a Superfund site. Extraction hardware, such as condensers, might be fabricated with lead-based solder. This is also a problem in distilling ethanol for moonshine. Heavy metals can bioaccumulate from trace levels. Andrew Fornadel of Shimadzu Scientific Instruments (Columbia, MD) described the quantitative analysis of cannabis samples for heavy metals including As, Cd, and Hg, and by ICP/MS following microwave digestion. The limit of detection was single-digit ppt.
A paper by Ewa Pruszkowski of PerkinElmer (Shelton, CT) described the use of hops as a surrogate for cannabis while developing an ICP/MS method for 31 analytes ranging from 9Be to 238U, thus avoiding potential legal problems with possession of cannabis in any form (Pruszowski, E. and Bosnak, C. Analysis of plant materials for toxic and nutritional elements with the NexION 350 ICP-MS. 2017 Cannabis Science Conference Final Program, pp 62–63, Aug 28-30, Portland OR).
Professor Jack Henion of Cornell University (Ithaca, NY) and president of Advion (Ithaca) presented a keynote lecture on determination of cannabanoid composition by mass spectrometry. His point echoed other lecturers, i.e., base your analytical method on the purpose, not the fad. There are needs where triple-quads are justified, but if one is dealing with known analytes, the combination of retention time and single-ion monitoring (SIM) using a single quadrupole would be more than sufficient.
QA/QC for cannabis
A similar message was reported by Graham Shelver, Ph.D. of 908 Devices, Inc. (Cambridge, MA), who described the design and application of a small, portable GC/MS with a novel ion trap mass analyzer. The GC uses narrow metal columns from Restek (Bellefonte, PA). These can be heated rapidly (up to 5 °C/sec) to drive off analytes to the ion trap detector. The detector has multistage ion traps that provide a significant ion cloud despite being only the size of a walnut. Additionally, the ion trap operates with a modest vacuum of ~1 × 10-2 atm. Pump-down during startup takes around 25 sec. A hydrogen generator provides hydrogen carrier gas. Power consumption is less than 400 W at 24 V from a lithium battery. Normal operation is dilute-and-shoot with a run time of 120 sec. The company has developed a software package called JetStream for control and data processing. Applications include assay of terpenes and residual solvents.
If small instruments are not suitable, the alternative is triple-quadrupole MS with a GC or LC inlet. Quantitation can be improved with reference standards of analytes labeled with stable isotopes of 13C, 2H, and 18O, according to a lecture by Professors Brett Tripple and James Ehleringer of the University of Utah (Salt Lake City).
Conclusion
It is clear that the cannabis industry is struggling with new public acceptance leading to strong demand-pull. The 5000+ year history of therapeutic efficacy is impossible to ignore. Neglect of the Department of Agriculture, FDA, NIST, and NIH is complicating the process of developing and distributing safe and effective therapies. For critical ailments like chronic pain, PTSD, and epilepsy, cannabanoid therapy seems superior to, and safer than, other alternatives, including opioids.
Safe and effective cannabis products will require new science-based technology from seed to smoke. For this, the Cannabis Science Conference is a suitable forum. The next conference will be held August 27–29, 2018, at the Oregon Convention Center in Portland. Visit https://www.cannabisscienceconference.com/ for more information.
Robert L. Stevenson, Ph.D., is Editor Emeritus, American Laboratory/Labcompare; e-mail: [email protected]