An interview with Cynthia Ludwig
Analytical testing of medical cannabis is highly variable at present. Because cannabis is classified by the US Drug Enforcement Administration as a Schedule I drug, meaning that it has a high potential for abuse and no accepted medical use, there are no federal laws governing the testing of marijuana (other than forbidding it). In states where medical or recreational marijuana use is permitted, rules and regulations are only gradually being developed. Further, the federal prohibition on the transport of marijuana across state lines makes it difficult for many laboratories to participate in interstate proficiency testing programs. In this confusing landscape, some analytical laboratories are developing and performing validated methods, whereas others are unreliable. One way to improve the situation is to develop official methods for cannabis testing. One organization involved in this work is the American Oil Chemists’ Society (AOCS). Cynthia Ludwig, the director of technical services at AOCS, spoke to us about these efforts.
How would you characterize the current state of analytical testing for cannabis?
It is like firing a shotgun full of rock salt from 50 yards. There are some laboratories that will hit the bull’s-eye and many that will not. From what I have seen in the laboratory proficiency testing results from three programs, the majority of the laboratories participating in a proficiency program are performing well and delivering quality results. The crux of the matter is that those are the labs that are putting in the effort to be reliable-but what about the labs that are not participating? Well, it is unknown. It would behoove analytical laboratories to self-regulate and participate in lab proficiency programs as well as seek ISO 17025 accreditation as is typical of analytical labs serving other industries.
Can you please briefly describe the basic steps in developing an official AOCS method for a cannabis test?
The need for a new Official Method, Recommended Practice, or Standard Procedure is determined by an industry’s requirements or requests. AOCS is contacted by industry professionals about a need for a new method or with suggestions to replace or revise methods already in place. After a proposal is submitted, AOCS coordinates with the industry experts to assemble any additional information needed.
A method validation and any additional scientific references, publications, or related materials are then submitted to an Expert Panel or a subcommittee of the Uniform Methods Committee (UMC) for evaluation and determination that the proposal should move forward. The panel or subcommittee then reviews the validation data and gives a recommendation to move forward, though it may request more validation data.
After the preliminary information is adopted by the subcommittee, a collaborative study is administered that follows the following AOCS guidelines:
Documents describing these steps are available on the AOCS website (http://www.aocs.org/attain-lab-services/methods/methods-development-x2021).
Have any laboratories developed methods for cannabis testing that have been approved as official methods by AOCS? If not, have any begun the process, and how far have they progressed?
No laboratories have yet completed the process to submit a cannabis-related method for AOCS Official Method or Standard Procedure status, but a few are getting close. Five laboratories are working on validating methods that are key in the industry, namely two liquid chromatography methods for cannabinoids, a headspace gas chromatography method for solvent residue, an inductively coupled plasma method for heavy metals, and a polymerase chain reaction (PCR) kit for microbial contamination and sample preparation. The first methods will likely be approved in summer of 2017.
How closely does AOCS work with cannabis testing laboratories interested in developing official AOCS methods? What assistance is provided?
We work closely with the laboratories, assisting with anything ranging from validation protocol suggestions and technical consulting to sourcing supplies, editing documents, and statistical analysis for the validation step.
When a method is approved to move to the collaborative study phase, AOCS coordinates recruiting the participating laboratories, facilitates shipping samples, and handles all of the data and results. We also manage the communications between the collaborating groups so there is a onestop shop for all questions. It really helps in the organization of a large study to have key people with experience (about 100 years for us) in running these types of studies.
What is the most critical step in developing an official AOCS method for cannabis?
All the steps are critical, and you cannot move from one to another without completing them all properly. One particular question keeps me up at night: How do we get a homogeneous “in-matrix” sample to all the laboratories for testing? With the ban on shipping cannabis samples and edibles across state lines, the industry is prevented from developing “in-matrix” methods on a national scale. States are doing a good job trying to work within their regulations, but having the ability to harmonize across state lines would put efforts in the fast lane.
A key to analytical study is numbers. The more laboratories you can have running a new method, the more eyes and hands that use it, and the more minds that contribute to the fine details, the better the method you will have at the end of the day. The only way to perform an effective study is to have the ability to test the same sample and work with “inmatrix” standard samples.
I did my graduate work in coal chemistry. Every sample I tested from Illinois #6 to Wyodak-Anderson came from Argonne National Labs coal sample program (1). A similar program for cannabis is needed to facilitate method development.
The collaborative study step in the process of developing an AOCS method is organized and directed by a technical committee. How are the members of the committee chosen? Is that process made any more difficult for a cannabis method study by the relative newness of the cannabis industry?
The expert technical committee is nominated and chosen by AOCS senior staff members in conjunction with top industry professionals. There are clearly industry leaders that have shown their expertise in publications and research papers.
Even though the industry is thought to be relatively young, some individuals have been working with cannabis for many, many years and we are working with those experts to further the analytical methods for cannabis testing.
What part of the collaborative study process would you consider to be the most difficult to complete for a cannabis testing method?
Getting a homogeneous sample to all the laboratories we need to participate in a collaborative study is the most difficult part. If the shipping restrictions could be lifted for these important research samples, a better set of data could be collected and used for evaluation of the methods.
An AOCS Standard Procedure is “a method that may be of interest or value, but does not have official method status.” Have any methods for cannabis testing been recognized as Standard Procedures by the AOCS?
Standard Procedures are also those methods that may require a certain brand of equipment. AOCS Official Methods are written to be neutral in equipment and reagents for the most part. As new technologies emerge and companies develop one-of-a-kind solutions for industry needs, the Standard Procedure gives us an opportunity to validate and study the method and offer our recommendation while waiting for the technology to be offered more broadly by other companies. Standard Procedures go through the same rigorous evaluation as is performed for Official Methods.
AOCS is currently partnering with a cannabis testing laboratory to provide ISO 13528-compliant reports for an inter-laboratory comparison and proficiency test for cannabis testing labs. What is the goal of this partnership, and what are the implications for the cannabis industry in terms of ensuring laboratory assurance and reliability?
The goal of the partnership is to share AOCS’s 100+ years of experience running lab proficiency programs with the budding new industry (pun intended). At an AOCS board meeting last year, it was decided that the compounds in cannabis fall within the scope of our nonprofit’s mission: “AOCS advances the science and technology of oils, fats, surfactants, and related materials, enriching the lives of people everywhere.” I was then given the task to build relationships and partner with the industry leaders to support them in the same fashion that we have supported other industries such as oilseeds and biofuels.
Method development and lab proficiency programs are two areas in which we thought AOCS could contribute and assist the industry to advance. As opportunities arise, we will be happy to participate and fully share our knowledge and expertise in other facets.
By changing the evaluation of laboratory proficiency results from a +/- 33.3% pass/fail result to the ISO 13528-compliant statistics report, laboratories receive a plethora of useful information, yet maintain each participant’s anonymity using a four-digit analyst number. A consensus mean from all the data, excluding the Grubbs outliers, is calculated and Z-scores are shown in both number and graphical form. The analysts also receive kernel density plots that show the distribution of all results about the consensus mean. These tools allow a lab to see where they fall in comparison to other laboratories in the program. Here is one example of the value of this more complete reporting system. I got a call from the head of a lab that had recently completed the lab proficiency program series, and he they wanted to talk to me about the laboratory’s results. Although the lab produced results that were within two standard deviations of the consensus mean (usually considered “passing” from an auditor’s standpoint), all the results were lower than the consensus mean. We discussed the implications and some possible reasons for this bias. The analysts in the lab went back and investigated their method, reagents, and other elements of the testing procedure. They discovered that the standards they were using for their calibrations curves had degraded. They changed their procedure to put a limitation on how long a standard is used and changed the storage conditions. They are excited to participate in the next round to see if their results are centered around the consensus mean, as they should be.
Cynthia Ludwig is the director of technical services of AOCS, an association that provides technical and scientific resources to the oils, fats, lipids, proteins, surfactants and related industries. In that role, she is responsible for all technical services activities including methods development, laboratory proficiency programs, and AOCS accreditations. Ludwig joined AOCS in 2014. Cynthia has a diverse background and long career as a research scientist. She started her journey as a large-scale production and analytical chemist at Sigma-Aldrich in St. Louis, Missouri. After five years, she joined Monsanto, where she spent the next 20 years contributing to over 20 patents, receiving numerous awards, and developing new products and technologies. Her degrees include a BS in Organic Chemistry from Southern Illinois University (US) and an MS in Physical Organic Chemistry from Eastern Illinois University (US).
This article was originally published in "Advancing The Science of Medical Cannabis," e-book for LCGC in September 2016. The full e-book is available for download here: www.chromatographyonline.com/lcgc-ebooks-09-01-2016