**4. Discussion**

The presented work investigates the ability of inhaled THC-predominant (chemotype-1) cannabis flos to improve health-related quality of life (HRQoL) and mitigate symptomatology in a treatment-resistant population of patients diagnosed with chronic painful conditions and anxiety-related disorders. Our results indicate that sustained inhalation of cannabis flos KHIRON 20/1 was associated with a robust and long-lasting improvement in HRQoL, mood and quality of sleep. The pharmacokinetics of orally ingested cannabinoids typically display erratic intestinal absorption, high inter- and intra-individual variability, extensive hepatic metabolism, and a delayed onset of effects between 90 and 120 min. In contrast, vaporized cannabinoids are rapidly and reliably absorbed into the bloodstream, achieving peak concentrations in blood generally in under 10 min [37]. These differences in pharmacokinetic properties afford patients a greater degree of control over dosage and speed of onset. Accordingly, both quantitative and qualitative research report on the ability of inhaled cannabis flos to quickly relieve symptoms of depression, stress, and anxiety. Patients describe the bodily sensation of cannabis inhalation as a "sigh of relief", which leads

to a state of relaxation promoting a reduction in pain sensation and, subsequently, improved sleep, motility, mood and acceptance [5,38]. For this reason, inhalation of cannabis flos is typically recommended as a rescue medication for acute or "breakthrough" symptoms [39]. However, our results showed sustained reduction in pain severity and interference of chronic pain with daily activities after 3 months of daily administration of KHIRON 20/1, which was maintained at the 6-month follow-up.

This finding is coherent with recent prospective observational studies investigating medical outcomes in chronic pain patients combining different formulations of medicinal cannabis, which also reported significantly lower levels of pain severity and pain interference, improved mood, sleep duration and sleep quality, and overall quality of life at 3 months compared to baseline [40,41]. Besides the 3-month follow up, Wang and collaborators incorporated ecological momentary assessment (EMA) to measure real-time health outcomes once daily for one week before (baseline) and for up to three weeks immediately after starting the treatment. Authors reported a significant reduction in real-time pain intensity (16.5-point reduction in a 0–100 VAS) and anxiety, longer sleep duration and better sleep quality in the first 3 weeks of treatment [40]. In similar studies, chronic pain patients treated exclusively with oral CBPMs also showed maximal improvement 3 months after the treatment initiation, which was sustained for over 6 months [22,42], suggesting that tolerance to the beneficial effects of cannabinoid therapy does not commonly occur. This observation is further supported by results from RCTs leading towards the clinical approval and commercialization of Sativex, in which MS patients showed sustained improvements in pain for more than 12 months without developing tolerance [43].

It is notable that greater than 95% of patients included in our data review were using cannabis illegally to treat their conditions at baseline and yet, we found a marked improvement in all PROMS analyzed. We interpret this finding to indicate that the administration of cannabis flowers in a clinical environment, under the supervision of a trained healthcare provider, further improves the clinical outcomes associated with legally prescribed CBPMs when compared to chronic patients self-medicating with illicit cannabis. This interpretation is further supported by similar findings from different jurisdictions where legally protected access to medical cannabis had recently become available [44,45]. The effect of such regulatory changes may have a greater impact in those experiencing anxiety-related disorders as it eliminates several major concerns for these patients, such as product availability, product reproducibility and the fear of potential legal consequences [5]. Our results indicate indeed that the largest clinical improvements associated with the inhalation of THC-predominant cannabis flos were reported by patients with a primary indication of generalized or social anxiety. First, we found a robust reduction in the GAD-7 scoring, from a baseline value of 12.7 to a value of 6.28 at the 3-month follow up. This remarkable result contrast with those reported by a Canadian group who applied the GAD-7 scale to a large cohort of adults authorized to use cannabis between 2014 and 2019. Although a statistically significant decrease in GAD-7 scoring was noted (from 9.11 to 9.04), it did not meet the threshold to be considered clinically significant [46]. In contrast, participants in our cohort diagnosed with generalized anxiety displayed higher baseline levels of moderate-to-severe anxieties, which could be potentially exacerbated by their illicit use of cannabis. Second, our result show that the cohort of patients diagnosed with anxiety-related disorders had a significantly larger contribution to the improvement in mood captured by the PHQ-9 scale, a measure of clinical depression. It could be postulated that the overall anxiolytic effect of whole flower CBPM could result from the combination of a rapid pharmacological activation of central type-1 cannabinoid (CB1) receptors together with the reassurance of pharmaceutical quality CBPM, legally prescribed by a clinician.

Functional imaging studies in humans have shown a correlation between THCmediated analgesia and a reduction in neural connectivity between the anterior cingulate cortex (ACC) and cortical areas involved in pain processing, the dorsolateral prefrontal cortex in particular, which are two key brain regions for the modulation of cognitive and emotional inputs [47,48]. Accordingly, results from human lab experiments sugges<sup>t</sup> that

THC prevents the onset of pain sensation by slightly increasing pain threshold but does not effectively reduce the perceived intensity of experimental pain [49]. Instead, THC seems to influence affective processing, thus making pain sensation less unpleasant and more tolerable, which resonates with qualitative assessments made by patients treated with CBPMs [38]. Available evidence also suggests that inhaled THC can potentiate the extinction of fearful and aversive memories in humans and reduce anxiety responses without eliciting psychotic effects [50], although it remains unclear if this effect is mediated by the activation CB1 receptors in the same brain regions. However, significantly increased circuit coupling between the ACC and the amygdala has been described during the processing of fearful stimuli in anxious (but not in healthy) individuals, which also correlated positively with self-reported symptoms of anxiety [51]. The key regulatory role of CB1 receptors in the amygdala, activated by endogenously produced anandamide on fear processing and aversive memory extinction has also been characterized both in preclinical and clinical studies [52,53]. Taken together, this evidence highlights the role of the ACC as a critical mediator in the analgesic and anxiolytic actions of THC, which could also explain why frequent and transient activation of central CB1 receptors could lead to sustained improvement in the emotional processing and a reduction in negative affect and physical symptoms associated with chronic illnesses [19].

The occurrence of adverse side effects experienced by participants was relatively rare, likely because most participants (95.6%) had previous experience with cannabis inhalation. In fact, the one patient reporting transient, mild, memory loss was naïve to cannabis. Adverse CNS-related side effects following cannabis inhalation are typically related to the dose of THC [54]. To counter this we have detailed an administration protocol to guide naïve patients and prescribing doctors following the mantra of "start low and go slow" [39], and based on the number and frequency of inhalations as opposed to the total amount of herbal cannabis loaded in the vaporizer. Pharmacokinetic studies on medically vaporized herbal cannabis have previously been performed with a tabletop model, S&B Volcano, which has a greater capacity to evaporate cannabinoids due to the instrument design and the range of working temperatures [30]. Human pharmacokinetic information for handheld devices is not readily available and it can largely depend on cannabinoid extraction efficiency, which may vary between devices [15]. Therefore, we aimed at providing simple instructions for first-time users to quickly gain control over cannabinoid dosing and speed of onset while minimizing the risk of involuntary overdosing. However, it is worth noticing the relatively safe profile of the inhaled route compared to the sublingual or oral administration. Firstly, due to their lipophilic nature, sublingual absorption of cannabinoids in oily carriers is limited and almost identical to oral ingestion [55]. Secondly, several studies have reported that intoxication, acute psychiatric symptoms, and adverse cardiovascular events are more common in patients following oral ingestion of CBPMs, while hyperemesis syndrome (cycling vomiting) was more likely attributable to inhalation of herbal cannabis [56]. Finally, in response to the clinical requirement of prescribing THC-predominant cannabis flos for extended periods of time in patients experiencing benefit, clinicians should be aware of the relevant contraindications to this substance including psychotic vulnerability and cardiovascular instability, as well as the risks of patients developing cannabis use disorders (CUD).

This work presents several limitations, some of which are inherent to the way realworld data is collected and interpreted [57]. We used a convenience cohort which, while representative of the more than 3,000 patients enrolled by T21 over the last two years, still poses a high risk of selection bias [58]. Additionally, patients were grouped for analysis of PROMS by primary indication, but their diagnosis and etiology could differ. Although a dosing protocol was suggested, it is plausible to assume that each patient established their own individualize dosing regime and that some may not have used an herbal vaporizer to administer their CBPMs, which is also representative of real-world clinical practice [39]. Finally, our research design did not control for placebo effect, which is typically robust in studies using cannabis [58], although this could be partially mitigated by the flexible

dosing regimen [10]. Owing to the inherent psychoactivity associated with the central activation of CB1 receptors, complete blinding in studies using THC is virtually impossible. In fact, relief of spontaneous pain typically correlates with high drug-like scores in human lab studies [18]. Therefore, aiming at completely separating the therapeutic properties from the psychoactive properties of THC may be erroneous, as some level of mind alteration may be required for the analgesic effect of cannabis to occur.
