(G-243) CannibiSenS: An on-demand rapid screen for Tetrahydrocannabinol (THC) in human saliva

Introduction::
Management of substances that possess a high potential for abuse require a comprehensive understanding of the temporal effects of a corresponding volume of intake. Cannabis is deemed as one of the widely used drugs in the United States and studies related to the primary psychoactive compound present- Δ-9 tetrahydrocannabinol (THC) has been known to cause adverse health effects. In this study, we present a field-deployable electrochemical sensing system that can detect THC at the 5ng/mL cut-off level with a dynamic range of 0.1- 100ng/mL in human saliva. Considering the complexity of human saliva matrix, the specificity study demonstrated selectivity to THC with minimum interactions to ethanol and cannabidiol (CBD). Surface Plasmon Resonance (SPR) has been implemented to visualize and validate the capture probe as means for THC detection. A robust-compatible binary classifier model has been shown in this work to effectively group samples into THC+ (high) and THC- (low) groups from human saliva with an accuracy greater than 90% considering a limited dataset. Hence, we demonstrate the potential of an innovative end-to-end system to effectively regulate cannabis use and prevent substance abuse in our surroundings.






Materials and Methods:: Non-Faradaic Electrochemical Impedance Spectroscopy(nF-EIS), an AC-based electrochemical technique was employed in the evaluation of a THC dose response in human saliva. A Gamry Reference 600 Potentiostat and our designed portable CannibiSens device was used to measure the response by taking measurements after each immobilization step. The non-faradaic changes in the dielectric properties of the SAM formed at the electrode- electrolyte surface was analysed as a result of the binding of the THC- BSA hapten to the anti-THC functionalized sensor surface for nF-EIS. A 10mV sinusoidal input voltage was swept across the working electrode over the frequency range of 1Hz to 1MHz which takes approximately 1 minute. Based on the current response measured, the impedance is calculated to build an impedance profile based on the dose of THC in human saliva.

Results, Conclusions, and Discussions:: As a proof-of- feasibility, dynamic monitoring of THC on a THC sensitive platform for self- monitoring of marijuana consumption and for rapid, non-invasive on-site DUI testing with an additional capability of be able to quantify the presence and absence THC through a machine learning classification technique has been demonstrated. An AC technique (non-Faradaic Electrochemical Impedance Spectroscopy) was employed to evaluate the dose response of THC in human saliva. The dose dependent response primarily occurs due to the capacitance changes occurring within the double layer which are captured in the form of impedance-by-impedance spectroscopy. A highly specific response with minimal cross reactivity was obtained from the interaction of EtOH with THC immunoassay. An LOD of 0.1ng/ml was achieved with a dynamic range of 0.1- 100ng/ml encompassing the set threshold level of 5ng/ml THC in saliva. Analysis on the developed CannibiSenS platform showed a comparable response with the benchtop results indicating the feasibility of translation into real- life scenario use. Further, a classifier based on bagged trees algorithm was developed to classify between THC presence (THC +) and THC absence (THC -) in human saliva with accuracy- 95% and 94% respectively in the benchtop system and CannibiSenS device. We present this work with an outlook that such a sensing platform with a capability of classifying the presence and absence of THC in matter of minutes would allow not only allow law enforcement to rapidly screen for salivary THC on-site but also allow marijuana users to monitor and limit their THC consumption within the threshold levels to avoid fatal road incidents.


Acknowledgements (Optional): :

References (Optional): :