Medical cannabis is already being used successfully for treatment of many diseases. Quality control is therefore of great importance. Among other things, total cannabinoid content and the ratio of the individual cannabinoids has to be determined, as different medical applications depend on the content of the individual cannabinoids. The optimization of regulated methods is difficult as changes, including co-injection with water, are not permitted.
In the German Pharmacopoeia 2018 (DAB 2018), the German Federal Institute for Drugs and Medical Devices (BfArM) published a revised monograph “Cannabis flowers” (cannabis flos) .
As the European Pharmacopoeia (Ph. Eur.) does not include a corresponding monograph, the method described in DAB 2018 for determination of the content of active ingredients in cannabis flowers by HPLC currently (06/2019) represents the binding EU-wide method . The monograph represents the recognized pharmaceutical rule according to the German Medicines Act and is used in decisions of the competent authorities in the EU.
Medical cannabis is classified into three different product groups depending on the ratio between cannabidiol (CBD) and Δ9-Tetrahydrocannabinol (THC) (tab. 1). These are used for different applications depending on the ratio.
Preparation of the standard solutions and sample follow the procedure described in the monograph: For sample preparation, 0.500 g powdered drug are shaken with 20 mL ethanol 96% for 15 min and then centrifuged. The clear supernatant is transferred into a 50 mL volumetric flask. The residual solution is treated twice with 12.5 mL ethanol 96% each in the same way. The organic solutions are combined and supplemented with 96% ethanol to make up 50.0 mL. This solution is filtered through a membrane filter made from regenerated cellulose with 0.45 µm nominal pore size. 1.0 mL filtrate is supplemented with ethanol 96% to 10.0 mL .
From the standard substances, 5 reference solutions of the individual active ingredients are prepared in methanol. For the calibration, at least 6 calibration solutions are prepared from the reference solutions, with a defined concentration being mandatory for each solution.
Suitability Test and Calibration
For the suitability test, a further reference solution is produced from Δ9tetrahydrocannabinol and Δ8tetrahydrocannabinol.
The resolution between the two peaks must be at least 1.2. The reproducibility of the area values is proven by a 6-fold injection of the THCA reference solution at RSD < 1%.
For the calibration, at least 6 calibration points of the respective substances are measured. Peak areas are plotted against concentration and fitted linearly.
Methods and Conditions
Separation of the components to be determined (cannabidiolic acid (CBDA), cannabidiol (CBD), cannabinol (CBN), Δ9-tetrahydrocannabinol (THC) and Δ9tetrahydrocannabinolic acid (THCA) is carried out on a C18 reversed phase. The analytical conditions applied in the practical example are summarized in table 2.
Preparation of standard solutions and sample follow the procedure described in the monograph. The reference solutions and standards are prepared in methanol, while the samples are prepared in ethanol. This leads to different peak shapes for standards and samples. For the ethanolic sample extracts, this results in a reduced separation efficiency of the column (fig. 1). The peak of the sample extract exhibits a widening due to a shoulder at the beginning of the peak. The standard, on the other hand, has an almost Gaussian peak.
Optimization by Co-Injection
Due to the binding nature of the procedure, no deviation from the sample preparation and preparation of the standard solutions as described in the monograph is permitted in pharmaceutical quality control. This is remedied by the co-injection method in which a few additional microliters of water are drawn up and injected together with the standard or sample (fig. 2).
The co-injection shifts the distribution equilibrium at the time of sample introduction briefly towards the stationary phase, resulting in the ideal peak shape being restored while slimming it simultaneously (fig. 3). In the co-injections shown here, the same amount of water was drawn up in addition to the sample volume of 10 µL.
Measurements of Real Samples
To compare the DAB method and co-injection, a real sample (cannabis of product group 2) was measured with and without co-injection. Results are shown in figure 4. Co-injection resulted in improved resolution. For direct comparison of the individual cannabinoids, widths of the individual peaks were summarized in table 3. With a peak width reduction of up to 56.1%, co-injection with water represents a significant improvement in chromatography.
In addition to the real sample of product group 2, real samples of product groups 1 and 3 were also measured. Cannabinoids content is summarized in table 4.
Classification of the cannabis flowers is based on the ratio of the relative content of THC to CDB (tab. 1), with the respective acid form being converted and included. In addition, the contents relate to dry matter.
Chromatographic differences occur using the DAB method, resulting in a poorer resolution of the real samples due to the different solvents of the standard solutions and of the extracts of the cannabis flowers. By coinjection with water, resolution of the real samples can be increased significantly and peaks with shoulders can be prevented. Automation of co-injection means no additional work or costs are to be expected in the routine.
With this method, three real samples with different cannabinoid contents could be measured and classified into the three product classes.
The analyses on which the data shown here are based were carried out in the laboratories of the German Federal Institute for Drugs and Medical Devices (BfArM) in Bonn.
Dr. Stefan Vosskötter
Dr. Stefan Vosskötter
Product Specialist HPLC
Shimadzu Deutschland GmbH
More on cannabis analytics!
 German Pharmacopoeia (Deutsches Arzneibuch 2018, DAB 2018) - first published in German Federal Gazette notice BAnz AT 24.04.2018 B5
 EU Guideline 2001/83 Consolidated Version, Annex I EC
 Shimadzu Application Note: Determination of the content of effective cannabis ingredients by HPLC according to DAB monography “Cannabis Flowers”