Savino Sguera of Digamma consulting continues his analysis on the reasons why cannabinoid and contaminant reporting can vary heavily in the cannabis industry. Click here to see part one!

Instrument calibration is the process of generating what is called a calibration curve. A calibration curve is a slope that relates the instruments response to established values.

When an instrument is first installed, its response to chemicals needs calibration. The instrument is calibrated by running a series of standards, chemicals with a precisely known concentration. These series of standards produce a slope that sets the instruments’ response to known amounts of the chemical being analyzed. At this point, an unknown concentration can be run through the instrument and can compared against the slope to pinpoint the precise concentration of chemicals in the unknown sample. This is how all client samples are run by an analytical lab, as all client samples are treated as a sample of unknown concentration until the laboratory releases their report on them.

If calibration is performed properly, a laboratory’s reporting of an unknown sample will be very accurate. But if the calibration is slightly off, the potential to report incorrect values for unknown solutions becomes very high.

The main source of erroneous calibrations are the standards. If a lab has an accurate standard, it will generate an accurate curve. If this standard is higher in concentration than the label value of the standard, it will generate a curve that skews unknowns to lower values. If the standard is lower in concentration than the label value of the standard, it will generate a curve that skews unknowns to higher values.

Figure 2: A graph illustrates how a 25% degradation of the calibration standard produces a 33% increase in final results reported. The black line represents a calibration curve generated with a proper standard of 1000 ug/ml, the red line represents a calibration curve generated with a degraded standard of 750 ug/ml. The properly calibrated machine measures the unknown sample correctly as 800 ug/ml, and the improperly calibrated machine measures 1067 ug/ml.

An example is illustrated in Figure 2. The black line represents a calibration curve created with no errors. The black curve is generated off a standard labeled at 1000 ug/ml and with a true value of 1000 ug/ml. When an “unknown” standard with a true value of 800 ug/ml is tested against the black curve, it generates a value of 800 ug/ml.

Next look at the red calibration curve. This curve was generated using a standard which was labeled at 1000 ug/ml but with a true value of 750 ug/ml. This results in an erroneous calibration curve, and when an “unknown” sample with a true value of 800 ug/ml is run against it, it reports a value of 1067 ug/ml. That’s a 33% increase when compared to the black calibration curve. The red curve’s standard was 25% lower than the label value of 1000 ug, and caused a 33% increase over the value reported by the black calibration curve.

Standards can be lower than their label value for a variety of reasons, but the source of problem can be determined. A retailer of reference standards typically has certificates of precision for their products, and they can trace the sample back to the manufacturer with ease by lot and batch number. Most discrepancies between the true value and label value of a standard occur after the standard has been shipped from the manufacturer to the client laboratory. It is up to the laboratory to ensure that the standard is correct and consistent with previous standards in addition to being consistent with the same standard from a second manufacturer.

The most common reason that a standard would be at a lower concentration than the label value is due to degradation. THCA degrades above room temperature to THC easily. Labs can prevent this degradation by ensuring that standards are delivered to laboratories on on dry ice to prevent heat based degradation, and stored in a -20 C freezer.

Editor’s Note: Standards may eventually degrade even when used under appropriate conditions! Make sure that the standard is replaced and tested against on a regular basis.

Other sources of standard manipulation may be intentional. By adding excess solvent, even a drop, to a certified reference standard, a dishonest person can lower the concentration of chemicals in the standard through dilution. This increases the reported results and is very difficult to detect. Some labs may attempt to manipulate their standards and then attempt to use standard manufacturers’ certificates of precision as a legal defense. Though this manipulation can be very difficult to detect, a skilled administrator can discern when manipulation is present by cross-checking the manufacturer’s certificates against the laboratories result and as well by cross-checking the results of one lab with those of others for the same sample.

Resources:

Want to get in touch with Savino? He can be reached via the following methods:

1. Website: https://www.digammaconsulting.com/
2. Email: savino@digammaconsulting.com

Recently, the Bureau of Cannabis Control released its long-awaited cannabis regulations. At Medicinal Genomics, we are particularly interested in the microbial testing requirements, as it affects the laboratories in California that are using our PathogINDICAtor® qPCR Microbial Detection Platform.

All-in-all, our laboratory customers will be happy with the regulations. They are favorable to qPCR methods like PathogINDICAtor because the state requires testing for specific microbial targets and does not set detection thresholds using colony forming units (CFU).

California is the first state to require testing for only pathogenic microbial impurities:

1. Shiga toxin-producing E. coli (STEC)
2. Salmonella
3. PathogenicAspergillus species (A. fumigatus, A. flavus, A. niger, and A. terreus)

The state requires cannabis testing labs test inhaled cannabis products for all six targets, while non-inhaled products need only be tested for STEC and Salmonella.

Cannabis testing labs that are currently using culture-based tests for microbes will have a very difficult time validating their method on Aspergillus. Aspergillus is notoriously difficult to grow in a culture medium because it tends to grow very slowly and colonies clump together to form a heterogenous macrocolony (see below). As a result, there are no commercially available culture-based tests that can detect Aspergillus.

However, qPCR methods can detect Aspergillus DNA. Our PathogINDICAtor Aspergillus detection assays have been validated on cannabis flower, extract, and edibles. We have also multiplexed all four Aspergillus species into a single assay, so testing labs only need to use one well on a 96-well plate.

All six targets that California requires are presence/absence tests. If any of the target organisms is detected within a 1g sample, the entire batch fails and can’t be released for retail sale. A failed batch can be remediated if it fails microbial testing, but it must be resubmitted and pass all laboratory testing before it can be released for retail sale. A batch can be remediated twice.

The language used in the regulations is significant because it does not use the term CFU, which has been used in other state regulations to set detection thresholds. CFU is a unit of measure used to quantify the results of culture-based tests. However, since qPCR methods measure microbial contamination using DNA, results are given in C_q (quantification cycle), not CFU.

We have developed equations to convert C_q to CFU for states that use that particular unit of measurement, but it has always been like fitting a square peg into a round hole because the two values are measuring different things. qPCR measures the amount of DNA present in a sample, while CFU measures the number of colonies that grew in a culture medium after a period of time. We have demonstrated in peer-reviewed papers that the microbes present on a cannabis sample and the microbes that grow in a culture medium can be radically different (see image below). More importantly, the latter is not an accurate representation of the potential harm to cannabis users and patients.

At Medicinal Genomics, we have developed qPCR testing assays for all six of the required targets. We have multiplexed STEC and Salmonella into one assay and all four Aspergillus into another single assay. That means labs can test for all six microbial targets in two wells (not including positive and negative controls).

We have also documented our validation on cannabis flower, extract, and edibles, which is publically available for download on our Validation Documents page. No other microbial testing platform has gone through such a thorough validation.

If you are a cannabis testing lab in California, considering contacting us about microbial testing. Or if you prefer, you can review our validation data and protocols.

Resources:

1. Want to get in touch with Medicinal Genomics? They can be reached via the following methods:
   1. Website: https://www.medicinalgenomics.com/
   2. Phone: 866-574-3582
   3. Email: info@medicinalgenomics.com

Cannabinoid potency tests are performed using High Performance Liquid Chromatography (HPLC) machines. Yay, article’s over!

Just kidding. What does an HPLC machine actually do?

Let’s start with the “C” in HPLC, which stands for chromatography. Chromatography is an analytical technique used to separate compounds based on their physical and chemical properties. Imagine you have a bunch of ping-pong balls with different weights. These ping-pong balls are sitting in a bucket that has an air pump attached to the bottom. As you turn on the air pump, the lightest ping-pong balls will start to fly out of the bucket. You can collect and count all the ping pong balls that fly out of the bucket, turn up the air pressure, and count the next group of ping-pong balls that fly out. If you continue turning up the air pressure, even the heaviest ping pong balls will fly out of the bucket.

Chromatography essentially works the same way:

1. The ping pong balls represent different cannabinoids and their respective weights. Different cannabinoids have different chemical and physical properties.
2. The bucket represents an analytical column. Solvents travel up the analytical column and pull the cannabinoids with them.
3. The air pressure represents the strength of the organic solvent. Stronger solvents will pull up cannabinoids at greater speeds.
4. Counting the number of ping-pong balls represents the signal strength measured by a detector.

Diagram of an HPLC machine. On the right side, you see the separatory column and detector. On the left are the solvents. Image courtesy of Wikipedia.

Samples received here at AgriScience Labs are carefully weighed and extracted by our trained analysts. The extracted solutions are then injected into an HPLC instrument with a stream of organic solvent. The sample is pushed through the analytical column, separating the different cannabinoids based on their physical properties. The strength of the organic solvent is gradually increased and the separated cannabinoids are removed from the column at different times. The cannabinoid compounds are then sent to a detector which produces a signal strength that correlates with the amount of each cannabinoid. This data is then processed by our scientists to determine how much of each cannabinoid is present in your sample.

Of course, this is a simplification of the actual process for determining the amount of each cannabinoid, but hopefully you now have a general idea of how your samples are analyzed for potency.

Resources:

1. Want to get in touch with AgriScience Labs? They can be reached via the following methods:
   1. Website: http://agrisciencelabs.com/
   2. Phone: 303-292-3800
   3. Email: info@agrisciencelabs.com