Design and setting

The study is a randomised, placebo-controlled, double-blind parallel-group adaptive trial to examine the clinical efficacy and safety of add-on CBD treatment in patients with clozapine-resistant schizophrenia. It will recruit 88 individuals with clozapine-resistant schizophrenia, who will be randomised at the point of providing written informed consent. The study design was discussed with our local lived experience reference group.

Participants will be given 1000 mg daily of CBD or placebo, in addition to routine care. A battery of validated clinical and physical health measures will be conducted throughout the trial period. Adverse events will be recorded fortnightly (i.e. weeks 0, 2, 4, 6, 8, 10 and 12), and trial medication will be distributed to participants fortnightly (i.e. weeks 0, 2, 4, 6, 8 and 10).

The authors assert that this study complies with the ethical standards of the relevant national and institutional committees on human experimentation and with the World Medical Association Declaration of Helsinki Ethical Principles for Medical Research Involving Human Subjects.^{Reference World Medical Association21} All procedures involving human participants were approved by the Metro South Human Research Ethics Committee (reference number: HREC/2022/QMS/83530). The study sponsor is the University of Queensland, which provides overarching governance. We have separate site governance approvals for each recruiting site. The trial is registered on the Australia New Zealand Clinical Trials Registry (registration number: ACTRN12622001112752). The study will follow the CONSORT guidelines for parallel-group randomised trials.^{Reference Moher, Hopewell, Schulz, Montori, Gøtzsche and Devereaux22}

Participants

Eighty-eight eligible participants with TRS will be recruited through mental health services at the Gold Coast, Metro North, Metro South, and West Moreton Hospital and Health Services and in south-east Queensland, Australia. Participant eligibility is detailed in Box 1. Participants must meet all the inclusion criteria and none of the exclusion criteria. Capacity to consent will be determined by the treating clinician in collaboration with clinical trial staff. Only participants with this capacity will be recruited following their provision of consent.

Box 1 Participant eligibility

Allocation concealment and randomisation

Participants will be randomised using a chronological process and only after written consent has been provided and baseline assessments have been completed. Participants will be randomised to either the intervention or control group in a 1:1 ratio using a computer-generated randomisation table with block randomisation. They will be allocated a unique identification number, which will be linked to a specific number based on the site they have been recruited from. If a participant withdraws from the study, their identification number will not be re-used, nor will they be allowed to re-enter the study.

Participants, clinical trial staff, investigators and treating clinicians will be blinded to the intervention. An independent biostatistician will generate the randomisation list, which will be provided to the manufacturer and a study site pharmacist. The pharmacist will hold the closed randomisation list and will be the only one who is able to unblind participants. In the case of an emergency, where it is critical that medical staff know whether a participant is on CBD or placebo, participants will be provided with contact information for unblinding (i.e. a 24-h number).

Pharmacological treatment

Participants will receive five gel capsules per day containing either 200 mg CBD (99% purity; <0.1 mg THC per capsule) suspended in medium-chain triglyceride oil or matched placebo (medium-chain triglyceride oil only). Capsules were manufactured by Linnea SA, Lavertezzo, Switzerland, a Therapeutic Goods Administration licensed facility, in accordance with current Good Manufacturing Practice and distributed by BOD Science. Dispensation, by a clinical trial staff member, will occur on a fortnightly basis, with six dispensations total per participant.

Outcomes

Measurements

Table 1 illustrates the trial schedule of visits and assessments. A range of validated clinical and physical health measures will be administered at different points throughout the trial period. Clinical measures will be administered by clinical trial staff, whom the clinical trial research manager will supervise. These staff are experienced mental health clinicians trained to administer these measures. Depending on participants' preferences, clinical measures will be administered in the clinic or at the participant's residence.

Table 1 Schedule of trial visits and assessments

SAFTEE-SI, Systematic Assessment for Treatment Emergent Events – Systematic Inquiry; HDL, high-density lipoprotein; LDL, low-density lipoprotein; HbA1c, glycated haemoglobin; ELFT, electrolyte and liver function tests; NAFLD, non-alcoholic fatty liver disease; FIB-4, fibrosis-4; PANSS, Positive and Negative Syndrome Scale; GAF, Global Assessment of Function; SIMPAQ, Simple Physical Activity Questionnaire; AQoL, Assessment of Quality of Life; CGI, Clinical Global Impression; TOPF, Test of Premorbid Function; BACS, Brief Assessment of Cognition in Schizophrenia; HRSA, Hamilton Rating Scale for Anxiety; CDSS, Calgary Depression Scale for Schizophrenia; TMT, trail making test; FCI, Food Craving Inventory; AUDIT, Alcohol Use Disorders Identification Test; PSQI, Pittsburgh Sleep Quality Index; FBC, full blood count; WCC, white cell count; CBD, cannabidiol; THC, tetrahydrocannabinol.

a. Assessment schedule may vary by plus or minus 5 days for operational convenience.

b. Providing a saliva sample is optional for participants and will be used to explore the presence of known correlates of clozapine and obesity, including variants in genes such as LEP and HTR2C.

Efficacy measures

The PANSS positive score will be the primary outcome.^{Reference Kay, Fiszbein and Opler26} Secondary outcome measures will include the following:

• • PANSS total and other subscale scores (negative scale and general psychopathology scale);^{Reference Kay, Fiszbein and Opler26}

• • Global Assessment of Function: a numeric scale from 1 to 100 used by clinicians to subjectively rate participants' social, occupational and psychological functioning;^{Reference Hall27}

• • Simple Physical Activity Questionnaire: a five-item instrument designed to measure physical activity and sedentary behaviour in various populations, including clinical samples with high levels of sedentary behaviour;^{Reference Rosenbaum, Morell, Abdel-Baki, Ahmadpanah, Anilkumar and Baie28}

• • Assessment of Quality of Life 4 Dimensions: a 12-item instrument that captures four dimensions of quality of life, comprising independent living, mental health, relationships and senses;^{Reference Richardson, Lezzi, Khan, Sinha, Mihalopoulos and Herrman29}

• • Clinical Global Impression: a measure of symptom severity, treatment response and treatment efficacy in treatment studies of people with schizophrenia;^{Reference Busner and Targum30}

• • Test of Premorbid Functioning: a measure of pre-injury IQ and memory ability;^{Reference Holdnack and Drozdick31}

• • Brief Assessment of Cognition in Schizophrenia (BACS) Verbal Memory and Learning: a measure of episodic verbal learning memory;^{Reference Keefe, Harvey, Goldberg, Gold, Walker and Kennel32}

• • BACS digit sequencing task: a measure of working memory;^{Reference Keefe, Harvey, Goldberg, Gold, Walker and Kennel32}

• • BACS symbol coding task: a measure of non-verbal functions such as attention, flexibility, speed of processing and abstract reasoning;^{Reference Keefe, Harvey, Goldberg, Gold, Walker and Kennel32}

• • BACS semantic and letter fluency: a verbal fluency test that measures spontaneous production of words that belong to the same category or begin with the same letter;^{Reference Keefe, Harvey, Goldberg, Gold, Walker and Kennel32}

• • Hamilton Rating Scale for Anxiety: a measure of anxiety symptom severity;^{Reference Hamilton33}

• • Calgary Depression Scale for Schizophrenia: a nine-item measure of depression in schizophrenia, independent of psychosis symptoms;^{Reference Addington, Addington and Schissel34}

• • Trail Making Test: a neuropsychological test of visual attention and task switching that can provide information about visual search speed, scanning and speed of processing, mental flexibility, and executive functioning;^{Reference Gaudino, Geisler and Squires35}

• • Food Craving Inventory: a 28-item self-reported measure of general and specific food cravings;^{Reference White, Whisenhunt, Williamson, Greenway and Netemeyer36}

• • Alcohol Use Disorders Identification Test: a ten-item screening tool to assess alcohol consumption, behaviours, and problems associated with alcohol use;^{Reference Saunders, Aasland, Babor, De La Fuente and Grant37}

• • Pittsburgh Sleep Quality Index: a 19-item self-reported measure of overall sleep quality assessing seven domains comprising subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication and daytime dysfunction;^{Reference Buysse, Reynolds, Monk, Berman and Kupfer38}

• • physical health measures including, BMI, waist circumference, waist-to-hip ratio, heart rate, blood pressure, HDL, LDL, triglycerides, total cholesterol, HbA1c, ELFTs, NAFLD and FIB-4.

Other measures

In addition to the abovementioned measures, we will assess adherence to the trial investigational product. This will be based on self-report and capsule count at each visit. We will perform additional blood tests including a full blood count and measure clozapine/ norclozapine levels, as well as levels of CBD, THC and their metabolites.

Participants will also have the option of providing a saliva sample at baseline. This will be used to explore the presence of known genetic correlates of clozapine and obesity, including variants in genes such as LEP and HTR2C.^{Reference Vasudev, Choi, Norman, Kim and Schwarz25} The DNA samples collected in this study will be used to validate associations between DNA single-nucleotide polymorphisms and treatment-resistant clozapine patient populations.

Our research team has used this suite of measures in other clinical trials^{Reference Siskind, Russell, Suetani, Flaws, Kisely and Moudgil39–Reference Siskind, Russell, Gamble, Winckel, Mayfield and Hollingworth41} and found them to be acceptable to study participants with schizophrenia, with low rates of dropout due to burden of clinical measures.

Statistical analysis

Interim analysis

Three interim analyses will be performed at 25%, 50% and 75% recruitment, corresponding to n = 22, 44 and 66 participants having completed their 12-week assessments. Each interim analysis will provide an ospportunity to evaluate whether the trial can be concluded at an earlier stage, rather than progressing until the target sample has been fully recruited. The interim analysis will evaluate the level of efficacy achieved, conditional on a linear trajectory of change in PANSS positive scores and an end-of-trial Cohen's d effect size of 0.48. The upper (early efficacy) and lower (early futility) bounds and their corresponding mean difference at interim analysis are summarised in Table 2. Conditional power at each interim analysis will be evaluated using bootstrap resampling (n = 10 000). Dependent on conditional power at interim analyses, we will consider randomisation reallocation to address the level of imbalance to achieve the desired effect size. The interim analysis will also provide an opportunity to reallocate the proportion of participants allocated to placebo and CBD if the study is found to be underpowered. Results of each interim analysis will be non-binding, with results of interim analyses provided to the Data Safety Monitoring Board (DSMB) for further evaluation.

Table 2 Summary interim analysis upper and lower bounds

β_c = probability of type II error at differing time points; δ, mean difference achieved at interim analysis.

Transparency declaration

D.S. (lead author and manuscript guarantor) affirms that the manuscript is an honest, accurate and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned and registered have been explained.