Personalise Antidepressant Treatment for Unipolar Depression Combining Individual Choices, Risks and Big Data

Depression Clinical Trial

— PETRUSHKA  

Official title:

Personalise Antidepressant Treatment for Unipolar Depression Combining Individual Choices, Risks and Big Data

Clinical Trial Details — Status: Not yet recruiting

Administrative data

• NCT number: NCT05608330
• Other study ID #: 286484
• Status: Not yet recruiting
• Phase: N/A
• Start date: November 2022
• Est. completion date: November 2023

Study information

• Verified date: October 2022
• Source: University of Oxford
• Contact: n/a
• Is FDA regulated: No
• Study type: Interventional

Clinical Trial Summary

PETRUSHKA is aimed at developing and subsequently testing a personalised approach to the pharmacological treatment of major depressive disorder in adults, which can be used in everyday NHS clinical settings.

We have collected data from patients with major depressive disorder, obtained from diverse datasets, including randomised trials as well as real-world registries (registers that hold routinely collected NHS data from the UK). These data summarise the most reliable and most up-to-date scientific evidence about benefits and adverse effects of antidepressants for depression and have been used to inform the PETRUSHKA prediction model to produce individualised treatment recommendations. The prediction model underpins a web-based decision support tool (the PETRUSHKA tool) which incorporates the patient's and clinician's preferences in order to rank treatment options and tailor the treatment to each patient.

This trial will recruit participants from the NHS within primary care in England and investigate whether the use of the PETRUSHKA tool is better than 'usual care' treatment in terms of adherence to antidepressant treatment, clinical response and quality of life, and its cost-effectiveness over a 6-months follow up.

Description:

The PETRUSHKA tool, employs a bespoke algorithm to identify the best antidepressant for each individual patient. The algorithm: (a) is based on a prediction model which uses a combination of advanced analytics (statistics) and machine learning methods (artificial intelligence); (b) uses a dataset which is a combination of real-world data (QResearch: https://www.qresearch.org/) from over 1 million primary care patients with depression in England and Wales, and individual participant data from about 40,000 patients recruited in randomised controlled trials; (c) incorporates preferences from patients and clinicians (especially about adverse events); (d) generates a ranked list of personalised treatment recommendations that will inform the clinical discussion between clinicians and patients, and the final treatment decision. The clinical decision aid tool is implemented in the form of a web-based application, accessible from any computer or tablet.

Recruitment information / eligibility

• Status: Not yet recruiting
• Enrollment: 504
• Est. completion date: November 2023
• Est. primary completion date: July 2023
• Accepts healthy volunteers: No
• Gender: All
• Age group: 18 Years to 74 Years

Eligibility

Inclusion Criteria:

• Aged 18 - 74 years inclusive;

• Willing and able to give informed consent for participation in the trial;

• Clinical diagnosis of depression (either single episode or recurrent), for which an antidepressant is clinically indicated;

• Willing to start antidepressant treatment as monotherapy;

• Able to read/understand and/or complete self-administered questionnaires online in English;

• Willing to meet any clinical requirements related to taking a specific medication

Exclusion Criteria:

• Prescribed any antidepressant in the preceding 4 weeks;

• Current or historical diagnosis of ADHD, Alcohol/Substance Use Disorder, bipolar disorder, dementia, eating disorders, mania/hypomania, OCD, PTSD, psychosis/schizophrenia, Treatment Resistant Depression (having tried 2 or more antidepressants for the same depressive episode at adequate dose and time);

• Diagnosis of arrhythmias (including Q-T prolongation, heart block), recent MI, poorly controlled epilepsy, acute porphyrias;

• Require urgent mental care or admission (including suicidal intent/plans);

• Concurrently enrolled in another investigational medicinal product (IMP) trial or an interventional trial about depression;

• Participants who are currently pregnant, planning pregnancy or lactating;

• Has a medical, social or other condition which, in the investigator's opinion , may make the participant unable to comply with all the trial requirements (e.g., terminal illness - motor neuron disease).

Related Conditions & MeSH terms

• Depression
• Depressive Disorder

Intervention

Other:
• PETRUSHKA tool
In the experimental arm, the PETRUSHKA tool will automatically select the antidepressants that have the best profile in terms of efficacy and acceptability for each individual participant (based on their baseline demographic and clinical characteristics) and then ask the participant to provide their preferences about common (and non-serious) adverse events. Based on patient's preferences and their individual characteristics, the PETRUSHKA tool will then identify the three best antidepressants for the participant. The clinician and the participant will be presented with an overall recommendation (in the format of a pictogram) showing how strongly each antidepressant is recommended for that individual patient. Via a shared decision-making process, the participant and the clinician will then agree on which antidepressant to choose from the shortlist.
• Usual Care
Any antidepressant prescribed by clinician based upon their clinical judgement.

Locations

Country	Name	City	State
n/a

Sponsors (2)

Lead Sponsor	Collaborator
University of Oxford	National Institute for Health Research, United Kingdom

References & Publications (8)

Austin PC, Harrell FE Jr, Steyerberg EW. Predictive performance of machine and statistical learning methods: Impact of data-generating processes on external validity in the "large N, small p" setting. Stat Methods Med Res. 2021 Jun;30(6):1465-1483. doi: 10.1177/09622802211002867. Epub 2021 Apr 13. — View Citation

Califf RM, Robb MA, Bindman AB, Briggs JP, Collins FS, Conway PH, Coster TS, Cunningham FE, De Lew N, DeSalvo KB, Dymek C, Dzau VJ, Fleurence RL, Frank RG, Gaziano JM, Kaufmann P, Lauer M, Marks PW, McGinnis JM, Richards C, Selby JV, Shulkin DJ, Shuren J, Slavitt AM, Smith SR, Washington BV, White PJ, Woodcock J, Woodson J, Sherman RE. Transforming Evidence Generation to Support Health and Health Care Decisions. N Engl J Med. 2016 Dec 15;375(24):2395-2400. — View Citation

Chekroud AM, Krystal JH. Personalised pharmacotherapy: an interim solution for antidepressant treatment? BMJ. 2015 May 14;350:h2502. doi: 10.1136/bmj.h2502. — View Citation

Chekroud AM, Zotti RJ, Shehzad Z, Gueorguieva R, Johnson MK, Trivedi MH, Cannon TD, Krystal JH, Corlett PR. Cross-trial prediction of treatment outcome in depression: a machine learning approach. Lancet Psychiatry. 2016 Mar;3(3):243-50. doi: 10.1016/S2215-0366(15)00471-X. Epub 2016 Jan 21. — View Citation

Christodoulou E, Ma J, Collins GS, Steyerberg EW, Verbakel JY, Van Calster B. A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models. J Clin Epidemiol. 2019 Jun;110:12-22. doi: 10.1016/j.jclinepi.2019.02.004. Epub 2019 Feb 11. — View Citation

Lewis G, Pelosi AJ, Araya R, Dunn G. Measuring psychiatric disorder in the community: a standardized assessment for use by lay interviewers. Psychol Med. 1992 May;22(2):465-86. — View Citation

Riley RD, Ensor J, Snell KIE, Harrell FE Jr, Martin GP, Reitsma JB, Moons KGM, Collins G, van Smeden M. Calculating the sample size required for developing a clinical prediction model. BMJ. 2020 Mar 18;368:m441. doi: 10.1136/bmj.m441. — View Citation

Tervonen T, Naci H, van Valkenhoef G, Ades AE, Angelis A, Hillege HL, Postmus D. Applying Multiple Criteria Decision Analysis to Comparative Benefit-Risk Assessment: Choosing among Statins in Primary Prevention. Med Decis Making. 2015 Oct;35(7):859-71. doi: 10.1177/0272989X15587005. Epub 2015 May 18. — View Citation

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	To determine whether using the PETRUSHKA tool to "personalise" antidepressant treatment, results in an increased proportion of patients continuing the allocated treatment, compared to usual care.	The number of participants who are still taking the allocated antidepressants after 8 weeks.	8 Weeks
Secondary	Self-rated change in depressive symptoms from baseline	Self-rated change in depressive symptoms measured using the 9-item Patient Health questionnaire	Baseline, week 2, 4, 6, 8, 12, 16, 20, 24
Secondary	Observer-rated change in depressive symptoms from baseline	Change in depressive symptoms measured using the observer-rated 17-item Hamilton Depression Rating Scale	Baseline, week 2, 4, 6, 8, 12, 16, 20, 24
Secondary	The number of participants who discontinue from treatment at 8 weeks due to any cause	Discontinuation from treatment due to any cause	Week 8
Secondary	The number of participants who discontinue from treatment at 24 weeks due to any cause	Discontinuation from treatment due to any cause	Week 24
Secondary	The number of participants who discontinue from treatment at 8 weeks due to adverse events	Discontinuation from treatment due to adverse events only	Week 8
Secondary	The number of participants who discontinue from treatment at 24 weeks due to adverse events	Discontinuation from treatment due to adverse events only	Week 24
Secondary	Self-rated change in anxiety symptoms from baseline	Self-rated change in anxiety symptoms measured using the 7-item Generalised Anxiety Disorder Assessment	Baseline, week 2, 4, 6, 8, 12, 20, and 24,
Secondary	Observer-rated change in anxiety symptoms from baseline	Observer-rated change in anxiety symptoms using the Hamilton Anxiety Rating Scale	Baseline, week 2, 4, 6, 8, 12, 20, and 24,
Secondary	The impact of depression on quality of life and capability wellbeing	EQ-5D-5L questionnaire (self-rated)	Baseline, week 4,8,12,24
Secondary	A reduction in risk of suicidality from baseline	Columbia Suicide Severity Rating Scale (observer-rated), ranging 1-5, where 1 is least severe and 5 is most severe.	Baseline, week 8 and 24
Secondary	An improvement in the functional outcome from baseline, with 0 being not at all and with 40 being very severely impaired.	Work and Social Adjustment Scale (self-rated)	Baseline, week 4, 8, 12 and 24
Secondary	A change in the health/social care costs of depression (direct and indirect) from baseline	Health Economics Questionnaire (self-rated)	Baseline, week 4,8,12 and 24