Pharmacogenetic-Guided Antidepressant Prescribing in Adolescents (PGx-GAP)

This is a parallel arm randomized (1:1) controlled trial. Adolescents aged 12-17 years (n=452) that did not respond or tolerate first-line fluoxetine therapy will be randomly allocated to receive 12-weeks of pharmacogenetic-guided antidepressant therapy (experimental intervention) or GLAD-PC guided prescribing (control intervention).

Goal: To test the efficacy of pharmacogenetic-guided antidepressant prescribing for adolescents with depression. Background: For an adolescent with moderate to severe depression, antidepressant medication is prescribed, often in combination with psychotherapy. The class of antidepressants recommended for use is selective serotonin reuptake inhibitors (SSRIs) with fluoxetine recommended as the first-line medication, and four other SSRIs recommended for consideration (sertraline, citalopram, escitalopram, fluvoxamine) if the adolescent does not respond or tolerate fluoxetine. For most adolescents, medication prescribing, and monitoring will be managed by a primary care physician or community pediatrician rather than by a mental health care provider, and guidelines exist to support this management (Guidelines for Adolescent Depression in Primary Care, GLAD-PC). However, GLAD-PC does not account for SSRI metabolism phenotypes that could change whether the SSRI selected is efficacious or tolerated. Our team of researchers, clinician scientists, patient partners, and primary care providers has designed a trial to test the impact of accounting for metabolism phenotypes, through pharmacogenetic-guided antidepressant prescribing, on adolescent outcomes, experiences, and health care utilization. Principal Question: Compared to GLAD-PC informed prescribing, does pharmacogenetic-guided prescribing for depressed adolescents who have not responded or tolerated first-line fluoxetine therapy, have superior efficacy following 12-weeks of therapy with an alternative SSRI? The Trial: This is a parallel arm randomized controlled trial. Adolescents aged 12-17 years (n=452) that did not respond or tolerate first-line fluoxetine therapy will be randomly allocated to receive pharmacogenetic-guided antidepressant therapy (experimental intervention) or GLAD-PC guided prescribing (control intervention). Participants and prescribing physicians will be blinded to which intervention was received. The primary outcome is depressive symptom remission at 12 weeks measured using the Quick Inventory of Depressive Symptomatology - Adolescent (17-item) (QIDS-A17). Secondary outcomes include side effects, role functioning, medication adherence, and health-related quality of life measured 4-, 8-, and 12-weeks after intervention initiation as well as cost-effectiveness.

Participants, their prescribing physician, and the investigator will all be blinded to study arm. The study coordinator will be the only one unblinded to study arm allocation.

Participants and their physician will receive a one-time prescribing report after completing baseline for second-line selective serotonin reuptake inhibitors with dosing information based on CYP2B6, CYP2C19, and CYP2D6 genotype data.

Participants and their physician will receive a one-time prescribing report after completing baseline for second-line selective serotonin reuptake inhibitors based on GLAD-PC dosing guidelines.

Quick Inventory of Depressive Symptomatology - Adolescent - 17-item (QIDS-A17) total score < 6. Scores range from 0-27, with higher scores indicative of more severe depression.

Frequency, Intensity, Burden of Side Effects Rating (FIBSER) scale. Total scores range from 0-6 (3 items); cut-points are used to indicate moderate (score of 3) or severe (score of 5) adverse drug reaction/side effect interference with activities.

WHO Disability Assessment Schedule. Scores range from 0 to 48, with higher scores indicative of worse role functioning.

Quick Inventory of Depressive Symptomatology - Adolescent - 17-item (QIDS-A17). Scores range from 0-27, with higher scores indicative of more severe depression.

Change in Clinical Global Impression Severity (CGI-S) scale. Scores range from 0-7, with higher scores indicative of more severe illness.

Resource use questionnaire that captures number of visits and out-of-pocket costs for various mental health services.

Administrative data will be obtained on medication information (agent, dose, duration) and health care utilization (doctor visits, hospitalizations, emergency room visits).

EuroQoL 5 Dimension - Youth (EQ-5D-Y). Five descriptive items code level of perceived problems in health states and a visual analog scale has a score from 0-100, with higher scores indicative of better health.

Medication Adherence Report Scale (MARS-5) scores. Scores range from 5-25 with higher scores indicative of better medication adherence.

Emergence of activation based on Treatment-Emergent Activation and Suicidality Assessment Profile. Total scores range from 0-114 (38 items) with higher scores indicating greater behavioral activation.

Participant-reported, Global Rating of Change Scale (GRCS) (11-point Likert scale ranging from +5 to -5) to indicate the degree to which symptoms and role functioning changed for the better, for the worse, or no change was experienced.

Physician-reported, 1-item survey about the perceived allocation of each of their participating patients; response options are 'PGx-guided prescribing', 'don't know' or 'GLAD-PC prescribing'

Inclusion Criteria: Age 12-17 Depression as the primary concern, confirmed by the treating physician QIDS-A17 score greater than or equal to 11 indicating moderate-to-severe symptoms Prior failure of fluoxetine therapy due to inefficacy or intolerance Intention to start a new SSRI English fluency Exclusion Criteria: Co-occurring psychosis, bipolar disorder, eating disorder, autism spectrum disorder, fetal alcohol spectrum disorder, or intellectual disability A score of 2 or 3 on suicide item 13 of the QIDS-A17 High-risk alcohol or substance use (excluding cannabis and tobacco) as indicated by a score of monthly or more on the S2BI History of non-response to 3 or more antidepressants (including fluoxetine, i.e. failure of fluoxetine and two other agents) as confirmed by the treating physician Psychotherapy or brain stimulation-based therapy initiated within 8 weeks of referral, or plans to initiate/change these therapies during study participation History of liver or hematopoietic cell transplant History of CYP2B6, CYP2C19, or CYP2D6 testing

Anonymized, individual participant PGx-GAP data will be shared using a controlled-access model. Under this model, the data will be released to a researcher if access criteria are met. All requests for data sharing should be made to the Co-Principal Investigators who will be responsible for reviewing and granting requests. Requestors should provide a research proposal for review. In the event that a data sharing request is declined, reasons will be provided to the requestor. If the data sharing request is granted, a data-sharing agreement will be initiated by the Co-Principal Investigators alongside the University of Calgary (lead institution). This agreement will include information on the individual data to be shared; if other documents will be available (e.g., statistical codes, data dictionary), when the data will be available and for how long, and how data access will be provided (e.g., file transfer).

The requestor is affiliated with an academic institution as an independent investigator or trainee of an independent investigator; The proposed research question(s) and hypothesis(es) are specific, measurable, and achievable; The proposed research project will be governed/overseen by a local legal/regulatory body; The proposed research project poses no risk of invasion of privacy or breaches of confidentiality for trial participants; A member of the proposed research team has sufficient statistical skills to carry out the proposed analytic plan; The requestor has sufficient financial and/or human resources to see the proposed research project to completion; The proposed research question(s) and hypothesis(es) do not conflict with active or planned studies of the Co-Principal Investigators.