Analytical Quality by Design: development and control strategy for a RP-HPLC method implemented in routine to evaluate the cannabinoids content in cannabis olive oil extracts

Cannabis sativa L. is characterized by having a remarkable number of chemical compounds: cannabidiol (CBD) and Δ-9-tetrahydrocannabinol (Δ-9-THC) are the cannabinoids considered as the most interesting for the clinical practice. Since 2015, the Italian law allows pharmacists to prepare and dispense cannabis extracts to patients under medical prescription. Cannabis olive oil extracts are prepared as magistral preparation from dried cannabis inflorescences, but a standard protocol is still missing. In fact, each pharmacist is allowed to prepare magistral preparations according to his/her own technical experience, leading to products with a wide variability in cannabinoids concentrations. The evaluation of Δ-9-THC and CBD content in cannabis extracts before sale is a regulatory requirement in Italy. Consequently, a quick and simple RP-HPLC/UV method has been developed to quantify them in the cannabis olive oil extracts. The analytical quality by design principles (AQbD) have been applied to the method development. The analytical target profile (ATP) was defined by the baseline separation of CBD and Δ-9-THC. The critical method attributes (CMAs) were set as the critical resolution between a peak pair (Rs) and the analysis time (t). Column temperature, pH of buffer part of the mobile phase and flow rate were selected as critical method parameters (CMPs); the other parameters were fixed according to the laboratory expertise and preliminary experiments. A response surface methodology (RSM) was used to optimize the method: a Doehlert design and a polynomial quadratic model were employed to approximate the relationship between the CMPs and the CMAs. Then, the method operable design region (MODR) was obtained by Monte-Carlo simulations and risk of failure maps. The probability of meeting the specifications (Rs ≥ 0.85 and t ≤ 6 min) was set at 95%. The method was validated according to the ICH Q2 guidelines and then implemented in routine analysis. A control strategy, based on system suitability tests (SST) and control charts, was planned. The evaluation was conducted over a period of six months of routine use. Both the CMAs, Rs and t showed values within the specifications in every analysis performed with this method. On the basis of such results, a new and more complete risk evaluation was achieved. It confirmed that the method performances have been maintaining the quality required by analysts during the method design. In the last part of this work, a study focusing on the concentrations monitoring of CBD and Δ-9-THC in the olive oil extracts was carried out. The analysis was conducted on 228 extracts: 39.91 % from Bedrocan®, 35.09 % from FM2®, 14.91 % from Bediol® and 10.09 % from Bedrolite®. The statistical analysis highlighted and confirmed a wide variability in the concentrations among different samples from the same starting typology of cannabis. This study underlines the compelling need of a standardized procedure to harmonize the preparation of the extracts.