Study population

Only few clinical data exist regarding the development/improvement of neuropsychiatric comorbidities in patients with CD under treatment. Therefore, in order to determine the appropriate sample size, we cannot make use of any well-known or estimated variances or mean value differences of the target variables.

To determine the sample sizes that are needed to detect a difference between two proportions, an a priori power calculation has been conducted. Here, the expected proportion of mental disorders/pain syndromes in the study sample has been based on the aforementioned references. Specifying type I error as 0.05 and a power ≥80%, (δ=42, σ=28), a sample size of ±14 patients will be needed to detect the specified differences. Assuming a safe margin, we deem 20 patients to be more than sufficient for our study target.

Following previous rationale, we will recruit 20 (of the initial 80 patients enrolled in the cross-sectional part of the study) patients with previously diagnosed CD, who will show biochemical control of the disease 24 months after study enrolment, including patients after surgical, after irradiation therapy and/or under current medical treatment for CD. Patients will be recruited from the endocrine outpatient clinic of the Max Planck Institute of Psychiatry, Munich and the Department of Internal Medicine, Ludwig-Maximilians-University, Munich.

Inclusion criteria include:

1. Adult patients with confirmed CD, who will show biochemical control of the disease 24 months after initial study enrolment including patients after surgical, before/under/after irradiation therapy and/or under current medical treatment.

2. Written informed consent.

Exclusion criteria include:

1. Female pregnant patients.

Primary objectives

In order to document arithmetic and percentage change in mean urinary free cortisol (UFC) values between uncontrolled state/before treatment (baseline) versus biochemical control of CD (2-year follow-up), we will measure the mean of absolute changes in UFC values in both groups.

In order to document arithmetic and percentage change in specific pain patterns between uncontrolled state/before treatment (baseline) versus biochemical control of CD (2-year follow-up), we will measure the change in painDETECT scores.

Secondary/exploratory

In order to document the change in neuropsychiatric comorbidities between uncontrolled/before treatment (baseline) versus biochemical control of CD (2-year follow-up), we assess the following domains with the associated standardised questionnaires:

• Depression using Beck's Depression Inventory (BDI);

• Anxiety using State-Trait Anxiety Inventory (STAI);

• Personality using Cloninger Temperament and Personality Questionnaire (TPQ) and Eysenck Personality Questionnaire (EPQ-RK);

• Daytime sleepiness using Epworth Sleepiness Scale (ESS);

• Subjective sleep using Pittsburgh Sleep Quality Index (PSQI);

• Specific pain patterns using Migraine Disability Assessment (MIDAS) and German Society for the Study of Pain (DGSS);

• Body image using Fragebogen zur Beurteilung des eigenen Körpers (FBeK) and Fragebogen zum Körperbild (FKB-20);

• HRQoL using SF-36, EuroQoL and CushingQoL;

• Sleep EEG, cognition and functional connectivity (MRI) (only in patients with a clinical indication, for example, severe cognitive impairment or severe sleep disorder as set in the former cross-sectional part of the study).

In order to document the change in cardiometabolic comorbidities between uncontrolled/before treatment (baseline) versus biochemical control of CD (2-year follow-up), we will assess the following domains:

• Anthropometric parameters including height, weight, body mass index, waist-and-hip circumference and waist-to-height ratio;

• Cardiometabolic risk biomarkers including fasting plasma glucose, glycosylated haemoglobin (HbA1c), triglycerides, total cholesterol, high-density lipoprotein, low-density lipoprotein and lipoprotein(a);

• Insulin resistance using the homoeostatic model assessment index;

• Bone mineral density, fat-free mass, fat mass and fat mass percentage using dual-energy X-ray absorptiometry (DXA) In addition, fracture risk assessment within the study population will be carried out with the FRAX algorithm (http://www.shef.ac.uk/FRAX) as suggested by Trementino et al;17

• Safety .

To assess the safety of current medical treatment for CD, we will report all adverse and serious adverse events under treatment with severity graded according to the Common Toxicity Criteria for Adverse Events (CTCAE) V.3.0. If CTCAE grading does not exist, CTC grades 1–4 will be used (higher scores indicating greater severity).

Assessment of safety under current medical treatment will include:

• Assessment of liver enzymes (alanine transaminase (ALT), aspartate transaminase (AST), alkaline phosphatase (ALP) and total bilirubin);

• Assessment of ECG abnormalities (QTcF interval);

• Assessment of hyperglycaemia (fasting plasma, glucose, insulin, HbA1c).

Current medical treatment will be discontinued if:

• ALT or AST >3×upper limit of the normal range (ULN) and total bilirubin ≥2×ULN and ALP <2×ULN;

• ALT or AST >5×ULN and ≤8×ULN persistent for more than 2 weeks;

• ALT or AST 8×ULN;

• QTcF >480 ms;

• Uncontrolled diabetes mellitus.