Disentangling Pharmacological and Expectation Effects in Antidepressant Discontinuation (PHEA)

Disentangling Pharmacological and Expectation Effects in Antidepressant Discontinuation - a Randomized, Balanced Open-hidden Trial

Antidepressant medication is established as an evidence-based, guideline-recommended treatment for Major Depressive Disorders. In the past decades, prescriptions of antidepressant medication have markedly increased, with a specific surge in maintenance prescriptions and therefore, long-term intake, despite guideline recommendations to discontinue antidepressant medication after maintenance therapy has been completed. Over half of fully remitted patients who attempt to discontinue their antidepressant medication report adverse discontinuation symptoms. For many patients, discontinuation symptoms are so severe, that they do not manage to complete their discontinuation attempt. While discontinuation symptoms, deterioration of depressive symptoms, and recurrence can result from pharmacological effects of antidepressant discontinuation, patients' expectations towards discontinuation are likely to play an essential role in occurrence, too. The aim of the present study is to explore the interplay of expectations and pharmacological effects in antidepressant discontinuation. Participants who fulfill German national S3 guideline recommendations will receive a 1:1 chance to either discontinue their antidepressant medication or remain on their antidepressant medication. In addition, participants' expectations are intended to be manipulated by varying verbal instructions using the open-hidden paradigm. Within the open trial arms, participants will receive full information about their treatment (i.e., high expectation). Within the hidden trial arms, participants will be informed about a 50% chance of discontinuing versus remaining on their antidepressant medication (i.e., moderate expectation). Participants will have a 1:1:1:1 chance of being allocated to 1 of the 4 experimental groups: open discontinuation (OD), hidden discontinuation (HD), open continuation (OC), or hidden continuation (HC) of their antidepressant medication. This preregistration is part of the collaborative research center (CRC) SFB/TRR289 which aims to characterize the psychological and neurobiological effects of treatment expectations on health outcome (https://treatment-expectation.de) and is funded by the Deutsche Forschungsgemeinschaft (DFG).

Antidepressant medication is established as an evidence-based, guideline-recommended treatment for Major Depressive Disorders. Following initial response to antidepressant medication and full remission of depressive symptoms, treatment guidelines generally recommend maintenance therapy for several months in order to prevent relapse and subsequent discontinuation of antidepressant medication. German national S3 guidelines for treating Major Depressive Disorders recommend that patients with a single episode remain on maintenance therapy for at least 4 months, while those with recurring episodes and significant functional impairment persist for at least 24 months. In the past decades, prescriptions of antidepressant medication have markedly increased, with a specific upsurge in maintenance prescriptions and consequential long-term intake, despite guideline recommendations to discontinue antidepressant medication. Over half of fully remitted patients who attempt to discontinue their antidepressant medication report adverse discontinuation symptoms. For many patients, discontinuation symptoms are so severe, that they do not manage to complete their discontinuation attempt. Additionally, discontinuation of antidepressant medication is associated with elevated risks of deterioration of depressive symptoms and recurrence. While S3 guidelines recommend dose-tapering over at least four weeks when stopping antidepressant medication, precise recommendations on how to minimize the risk of recurrence and the potential burden associated with depressive and discontinuation symptoms are lacking. Discontinuation symptoms, deterioration of depressive symptoms, and recurrence can result from pharmacological effects of antidepressant discontinuation and patients' expectations towards discontinuation are likely to play an essential role in occurrence, too. The present study aims to explore the interplay of expectations and pharmacological effects in antidepressant discontinuation. Participants who fulfill German national S3 guideline recommendations will receive a 1:1 chance to either discontinue their antidepressant medication or remain on their antidepressant medication. In addition, participants' expectations are intended to be manipulated by varying verbal instructions using the open-hidden paradigm. Within the open trial arms, participants will receive full information about their treatment (i.e., high expectation). Within the hidden trial arms, participants will be informed about a 50% chance of discontinuing versus remaining on their antidepressant medication (i.e., moderate expectation). Participants will have a 1:1:1:1 chance of being allocated to 1 of the 4 experimental groups: open discontinuation (OD), hidden discontinuation (HD), open continuation (OC), or hidden continuation (HC) of their antidepressant medication. The trial will consist of a 13-week experimental phase (1 week run-in, 4 weeks of either discontinuation following a pre-specified tapered dose-reduction scheme or continuation of initially prescribed antidepressant medication, 8 weeks monitoring either off antidepressant medication or on initially prescribed antidepressant medication) and a 39-week clinical observation phase. During run-in, all participants will remain on their prescribed antidepressant medication and initial dose, though newly encapsulated to control for tablet appearance effects. All pills for all participants will look identical throughout the whole trial. During the subsequent 4 weeks, participants within the hidden arms (HD and HC) will be blinded as to whether they are receiving tapered dose-reduction or their initial antidepressant medication. During the following 8-week monitoring phase, participants within the hidden arms will receive double-blind placebo pills (HD) or double-blind antidepressant medication (HC). Participants within the open trial arms will either be aware of discontinuing their antidepressant medication during the first 4 weeks following run-in, followed by 8 weeks of receiving open-label placebo pills (OD), or will be aware of remaining on their antidepressant medication (OC) during the entire experimental phase, respectively. At 13-weeks post-baseline, the experimental phase will conclude with the primary outcome measure and patients will be debriefed. Detailed hypotheses are: Interaction effect of treatment and treatment expectation: Treatment (continuation vs. discontinuation of antidepressant medication) and treatment expectation (high vs. moderate) interact in modulating discontinuation symptom load among remitted MDD patients over the course of the experimental phase. Post-hoc comparison on the nocebo-determined effect of expectation: Remitted MDD patients who remain on their antidepressant medication will show a higher discontinuation symptom load with moderate than with high expectation. Post-hoc comparison on the pharmacological effect of treatment: Remitted MDD patients with moderate treatment expectation will show a higher discontinuation symptom load if antidepressant medication is discontinued versus if antidepressant medication is continued. Post-hoc comparison on the effect of treatment expectation: Remitted MDD patients who discontinue their antidepressant medication will show a higher discontinuation symptom load with high than with moderate treatment expectation. Posner task (spatial-cueing-paradigm at 13 weeks post baseline): Attentional and emotional processing differs between the experimental groups among remitted MDD patients. Modulating effects of further psychological, physiological, and medical factors on the relationship between treatment expectation and discontinuation symptom load: The relationship between treatment expectation and discontinuation symptom load will vary according to subjective stress ratings, objective stress measurements of salivary cortisol levels, prior side effects of antidepressant medication, prior discontinuation experience, personality traits, psychopathology, current mental well-being, reported levels of anxiety, and substance use.

Antidepressant medication, Discontinuation symptoms, Pharmacological effects, Expectation effects, Nocebo, Open-hidden trial

The PHEA-study is a randomized, balanced open-hidden discontinuation trial with a 2x2-factorial design (treatment: continuation versus discontinuation; expectation: high versus moderate). A stratified block-randomization will be applied, with participants receiving either open discontinuation (OD), hidden discontinuation (HD), open continuation (OC), or hidden continuation (HC) of their antidepressant medication. Participants will have a 1:1:1:1 chance of being allocated to one of the four experimental groups. With an anticipated dropout rate of 20 % our recruitment target is N=192 (n=48 per group). Randomization will be stratified by i) study center (Hamburg vs. Marburg, 1:1), ii) intake duration of antidepressant medication with 24 months as a marker for long-term intake (<24 months intake vs. ≥ 24 months intake, 3:7); and iii) risk of developing discontinuation symptoms associated with the antidepressant medication (moderate vs. higher or unknown risk, 1:1).

In case of assignment to one of the hidden arms, both participants and investigators (i.e. physicians, psychologists, and research assistants) will be blinded to group assignment. Accordingly, in case of assignment to one of the open trial arms, participants and investigators will be aware of group assignment. The primary outcome is a self-report measure assessed via an online-survey by the participants; the outcome assessor (= participant) will therefore only be blinded to group assignment in the hidden arms. The person responsible for randomization and preparation of the medication is not blinded, but has no contact with the study participants.

Participants will discontinue their antidepressant medication and will be fully informed about treatment (i.e., high expectation).

Participants will discontinue their antidepressant medication, but will be informed about a 50% chance of discontinuing versus remaining on their antidepressant medication (i.e., moderate expectation).

Participants will remain on their initial antidepressant medication and will be fully informed about treatment (i.e., high expectation).

Participants will remain on their initial antidepressant medication, but will be informed about a 50% chance of discontinuing versus remaining on their antidepressant medication (i.e., moderate expectation).

Psychological intervention: Participants' expectations will be manipulated by varying verbal instructions using the open-hidden paradigm. Within the open trial arms, participants will receive full information about their treatment (i.e., high expectation). expectation).

Psychological intervention: Participants' expectations will be manipulated by varying verbal instructions using the open-hidden paradigm. Within the hidden trial arms, participants will be informed about a 50% chance of discontinuing versus remaining on their antidepressant medication (i.e., moderate expectation).

Discontinuation symptom load over the course of the experimental phase from baseline T0 to primary endpoint T9 - 'Discontinuation Related Signs and Symptoms Scale' (DESS)

The Discontinuation Related Signs and Symptoms Scale is a self-report questionnaire to assess discontinuation symptoms, incorporating 43 discontinuation symptoms of antidepressants with intensity ratings ranging from 0 (not present) - 3 (severe) each. Total sum score ranges from 0-129 with higher scores indicating more pronounced discontinuation symptoms.

Measured as area under the curve (AUC) over the following time points: Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]

The Posner Task manipulates attentional resources, provokes emotional stimuli and robustly activates limbic, prefrontal, and visuo-spatial brain circuits. In short, participants respond as fast as possible to a dot target by button pressing while neutral, happy, sad or fearful face distractors are presented. Targets are preceded by either spatially-directing cues leading to covert shifts in the attentional focus (i.e., low attentional resources to process distractors) or non-spatial cues leaving the attentional focus on the distractors. This approach allows objective validation of the effects of the experimental groups (high/medium expectation during discontinuation and continuation, respectively) on discontinuation symptoms with respect to (i) emotional processing, (ii) top-down control of emotional input under high and low attentional demands, and (iii) mood state.

Discontinuation symptom load over the course of the trial from baseline T0 to study completion FU3 - 'Discontinuation Related Signs and Symptoms Scale' (DESS)

The Discontinuation Related Signs and Symptoms Scale is a self-report questionnaire to assess discontinuation symptoms, incorporating 43 discontinuation symptoms of antidepressants with intensity ratings ranging from 0 (not present) - 3 (severe) each. Total sum score ranges from 0-129 with higher scores indicating more pronounced discontinuation symptoms.

Measured as area under the curve (AUC) over the following time points: Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

The Generic rating scale for treatment effects is one of three subscales of the 'Generic rating scale for previous treatment experiences, treatment expectations, and treatment effects' (GEEE). The subscale 'Generic rating scale for treatment effects' consists of three items that assess current treatment effects, i.e., effects of antidepressant discontinuation or continuation, on a numeric rating scale with eleven response options (0-10) each; total scores range from 0-30 with higher scores indicating more pronounced effects of antidepressant discontinuation.

Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

Self-report measure to assess depressive symptoms, including 21 items with 4 response options (0-3), total scores range from 0-63 with higher scores indicating more pronounced depressive symptoms).

Screening; weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

Expert-rated interview to assess the severity of depression by 10 items with up to 7 rating categories (0-6) for each item, total scores range between 0-60 with higher scores indicating more pronounced depressive symptoms.

Screening; weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

Appearance of a new, depressive episode after full remission of depressive symptoms and a period of recovery. Potential recurrence will be monitored weekly during run-in and (dis-) continuation phase, biweekly during monitoring phase, and 6, 9 and 12 months post-baseline based on BDI-II and MADRS scores. If recurrence is suspected (as indicated by BDI-II scores >19 OR MADRS scores >21 over two weeks/study visits), corresponding SCID-5-CV sections will be conducted to (dis-)confirm recurrence.

Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

Treatment adherence assessed by a single item in the clinical interview about adherence to intake of study medication during the experimental phase and remaining on or off antidepressant medication during the clinical observation phase, respectively. Further details will be published elsewhere, a link to the registration of the overarching project Z03 will be added as soon as available.

Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

Functional and structural imaging data will be acquired using a 3T MR system (Trio, Siemens Healthcare, Erlangen, Germany at the Institute of Systems Neuroscience, director C. Büchel) with a 64-channel head/neck coil. All structural and functional images will be acquired using standardized MR protocols as described in, provided by and analyzed in the CRC-overarching central project Z03 to later allow for pooled analysis of harmonized multi-center data ('Mega-analysis'). These include (i) resting state functional MRI (rsfMRI), (ii) 3D-MPRage T1-weighted sequence, and (iii) Diffusion Tensor Imaging (DTI). Further details will be published in the registration of the CRC-overarching central project Z03, a link will be added as soon as available.

Self-reported measure to assess patients' treatment expectations, consisting of 15 items with 11 response options (0-10), total scores range from 0-150 with higher scores indicating more positive treatment expectations.

Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]

The Generic rating scale for treatment expectations is one of three subscales of the 'Generic rating scale for previous treatment experiences, treatment expectations, and treatment effects' (GEEE). The subscale Generic rating scale for treatment expectations consists of three items that assess expectations towards treatment on a numeric rating scale with eleven response options (0-10); total scores of all three subscales combined range from 0-30 with higher scores indicating higher treatment expectations.

Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

Self-report measure to assess side effects of antidepressant medication, including 36 items (symptom descriptions) with 4 response options (0-3). Items are additionally evaluated on their relation to antidepressant medication (Yes/ No questions). Total scores range from 0-108 with higher scores indicating stronger experiences of side effects.

Pre-experiences with antidepressant discontinuation - 'Generic rating scale for previous treatment experiences' (GEEEpre)

The Generic rating scale for previous treatment experiences is one of three subscales of the 'Generic rating scale for previous treatment experiences, treatment expectations, and treatment effects' (GEEE). It is a self-reported assessment of previous experiences with treatment, i.e., antidepressant discontinuation. If experience with antidepressant discontinuation is indicated, the subsequent 3 items rate the experiences on a numeric analogue scale each (ranging from 0-10), total score range from 0-30; higher scores indicate more pre-experiences with antidepressant discontinuation.

Self-reported measure of subjective stress, including 10 items with 5 rating categories (0-4), total scores range between 0-40 with higher scores indicating more pronounced stress.

10-item scale measuring the personality traits Extraversion, Agreeableness, Conscientiousness, Emotional Stability, and Openness to assess the influence of personality traits on treatment outcome at baseline. The items are rated on a five-step scale from 1 "disagree strongly" to 5 "agree strongly". The scale consists of 2 BFI items for each Big Five Dimension.

Behavioral Inhibition/Behavioral Approach (motivational systems) - 'Behavioral Inhibition/Behavioral Approach System' Scale (BIS/BAS)

Self-report questionnaire to assess sensitivity to approach or avoidance goals; 24 items; each item is rated on a 4-point scale from 1 (not at all true for me) to 4 (very true for me).

Self-report questionnaire to assess amount of somatosensory amplification tendencies; 10 items; items are rated on a 5-point scale from 1 (not at all true) - 5 (extremely true); total scores range from 10-50 with higher scores indicating more somatosensory amplification.

Self-report questionnaire to assess mental well-being by 7 statements about thoughts and feelings using 5 response options, total scores range from 7-35 with higher scores reflecting a higher level of mental well-being.

Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

Self-report measure comprising 4 items that assess core symptoms/signs of depression and anxiety. Two items assess depression; two items assess anxiety, all by means of a four level Likert scale (0-3), so that total scores for each subscale range from 0-6. A score of 3 or greater on the depression subscale indicates a higher probability of symptoms of depression; a score of 3 or greater on the anxiety subscale indicates a higher probability of symptoms of anxiety.

Self-report questionnaire as indicator of state and trait anxiety and depression, divided in 2 sections (state vs. trait) consisting of 20 statements with 4 response options (1-4), respectively. Total scores per scale range between 20 and 80 with higher sum scores indicating higher state/trait anxiety or depression.

At baseline (T0) and 13 weeks post baseline at primary endpoint (T9); Trait scale measured at baseline only

Self-report questionnaire to assess illness burden; 7 items; each item rated on a 0 (no disability)-10(maximum disability) standardized numerical analogue scale; total scores range from 0-70 with higher scores indicating more pronounced disability.

At baseline (T0), one week post baseline (T1), five weeks post baseline (T5), and 13 weeks post baseline at primary endpoint (T9)

Assesses the consumption (volume and frequency) of nicotine, caffeine, alcohol, pain medication, tranquilizers, or recreational drugs.

One week post baseline (T1), five weeks post baseline (T5), nine weeks post baseline (T7), and 13 weeks post baseline at primary endpoint (T9)

Perceived warmth and competence from both the assessor's perspective (11 items, 0 (not at all) to 5 (very much)) and participants' perspectives (12 items, 0 (not at all) to 5 (very much)) will be assessed. Higher scores indicate higher ratings of warmth and competence.

Interview-based measure assessing adverse events, followed by an expert-rating of the intensity and description of the adverse events (1-5) and causal relation to treatment (1-6). Higher scores indicate more and/or a higher intensity of adverse events.

Weekly during run-in and (dis-)continuation phase [5 weeks]; biweekly during monitoring phase [8 weeks]; follow-up [9 months]

Single item that asks participants to indicate whether they believe they discontinued or remained on their initial medication. Optional item of the 'Generic rating scale for previous treatment experiences, treatment expectations, and treatment effects' (GEEE).

Single item that asks participants to rate whether they perceive their depressive disorder as a biologically or a psychologically caused illness on a single item ranging from 0 (biologically caused) to 10 (psychologically caused).

Inclusion Criteria: Adult patients (18 to 75 years) with fully remitted Major Depressive Disorder, single or recurrent, as main diagnosis confirmed by prescribing physician and SCID-5 (American Psychiatric Association, 2013) Intake of SSRI/SNRI (citalopram: 20-40mg, escitalopram: 10-20mg, sertraline: 75-150mg, venlafaxine: 75-150mg, duloxetine: 60-100mg, paroxetine: 20-40mg) or NaSSA (mirtazapine: 30-45mg) Discontinuation wish by patient supported by prescribing physician Fulfils criteria of the German S3 national guideline recommendations for treatment of Major Depressive Disorders to discontinue antidepressant medication: a) response to antidepressant medication, b) symptom remission for at least four months (for a single episode) or two years (for two or more episodes with significant functional impairment) and c) concurrent intake of antidepressant medication (at least 4 weeks on a steady dose) Exclusion Criteria: Acute or chronic somatic illness and/or intake of medication which might interfere with depressive disorder, antidepressant medication or proposed study Acute suicidality, psychotic symptoms, substance abuse or addiction, current mania, or hypomania confirmed by SCID-5 (American Psychiatric Association, 2013) or other psychopathology which might interfere with depressive disorder, antidepressant medication or proposed study Any history of bipolar disorder or psychosis confirmed by SCID-5 (American Psychiatric Association, 2013) Severe stressful life events (e.g., death of a family member) within six months prior to study participation Insufficient German language proficiency No informed consent Upon optional participation: MRI-specific exclusion criteria (phobic anxiety, claustrophobia, ferromagnetic implants, etc.)

The PHEA-study is a project within a special research area funded by the DFG, namely the TRR-SFB 289 Treatment Expectation. Individual participant data will be shared with the study team of the TRR-SFB 289 after deidentification and will be available in this form for other researchers upon reasonable request. Only anonymized data in agglomerated form is used for publications. No personal data will be shared.