Anorexia Nervosa (AN) has the highest mortality of any psychiatric disorder and a paucity of effective treatments. AN becomes intractable in around 20%, resulting in huge individual and healthcare costs. Exploration of underlying processes and novel treatment strategies is thus crucial. This is a pilot study of a novel translational approach to the investigation and treatment of severe AN. The aims are (1) to explore the safety, acceptability and feasibility of Deep Brain Stimulation (DBS) for AN (2) to map neural mechanisms underpinning aberrant reward and optimise DBS targets. The study will involve 10 consenting adults with full mental capacity, and the nain protocol will last 15 months. . There is then optional annual follow ups for up to 5 years .It incorporates an ethical substudy including assessment of capacity and informed consent. It combines complementary forms of cutting edge neuroimaging including fMRI and MEG (magnetoencephalography). These will be complimentary in helping identify the best strategy for treating severe AN using DBS. Individuals with severe intractable AN will be eligible to take part. The study will be conducted in Oxford, United Kingdom at the John Radcliffe and Warneford Hospitals. The individuals will have preoperative ethical, neuroimaging and psychological assessments, A DBS operation in month 2, DBS switch on in month 3 month postoperatively . DBS will be targeted to the Nucleus accumbens, followed by a 12 month post switch on follow-up period with monthly joint neurosurgical psychiatric and psychological assessments and postoperative MEG scans . The study will further our understanding of food reward processes in general and AN in particular indeed promises to provide important information which may revolutionize future treatments. The proposal builds upon a body of research investigating the role of aberrant reward processes in AN and exploits our complementary research experience in AN using experimental behavioural strategies, fMRI , MEG and DBS to establish a powerful translational research strategy.

Our hypothesis, based on prior research, is that DBS to the nucleus accumbens will have beneficial neural and symptomatic effects in severe Anorexia Nervosa.

As a test of this hypothesis, and to evaluate acceptability and ethical issues we aim to initially study up to 10 patients.

This study may provide important clinical and scientific information to contribute to larger studies.

DESIGN SUMMARY The design will consist of a case series of 10 patients with repeated measures. The individuals will be aged 20-65 with at least 7 years history of severe treatment resistant AN, who are fully consenting and motivated to recover.

The patients will be selected on the basis of specific suitability criteria. A small case series study design is needed as this is a novel and invasive strategy, and further information about acceptability, feasibility and neural processes involved is required to inform more extensive studies. This can optimally be obtained by initial detailed study of a small number of patients. As some heterogeneity of response is likely, as advised by our external reviewer Professor Walter Kaye, we are aiming to study 10 patients ..

The intervention will be bilateral deep brain stimulation targeted at the nucleus accumbens with stimulation at the ventral anterior limb of the internal capsule. Following DBS each patient will be followed up every month weeks for 13 months.

First, after recruitment there will be a 2 month preoperative period of thorough preoperative examination of ethical issues including qualitative and semi structured interviews (MacCATCR (MacArthur Competence Assessment Tool for Clinical Research))to assess competence and informed consent will be completed by an independent researcher experienced in ethical issues in AN (Jacinta Tan).

The study participants will then be have the initial research assessment and will be reviewed every month in joint eating disorders psychiatry -neurosurgery clinics . In month 10 there will be two randomly assigned 2 week periods of active or sham stimulation,.

monthly assessments will include : • BMI • Semi structured interviews : Eating Disorder Eating Disorder Examination (EDE)

• Yale Brown CornellEating Disorder Scale (YBCEDS)
• Observer rated comorbidity (SCID / Hamilton Depression rating scale)
• Computerized behavioural tasks: Food reward (Leeds-Oxford food preference task) habit formation tasks
• Self report schedules: EDE Questionnaire, functional impairment (CIA) , quality of life scale (WHOQUALBREF); starvation dependence scale, Beck Depression Inventory (BDI, Beck et al. 1961), the State Trait Anxiety Inventory (Spielberger, 1998), the Snaith-Hamilton pleasure scale (SHAPS, Snaith et al. 1995) and the Ruminative Response Scale for Eating Disorders (RRSED). Pre-operative imaging will involve fMRI and MEG. Post-operative imaging will involve 3 MEG scans, two in month 10 and at study end, 15 months.

DETAILS OF METHODOLOGY:

Subjects selection and recruitment:

A list of eligible patients, fulfilling inclusion and exclusion criteria will be made in discussion with Specialist Eating Disorder services clinicians. Patients will be ranked in order of suitability according to specific suitability criteria:

i. Severity and intractiblity of Anorexia Nervosa ii. Restrictive anorexia nervosa - without binging/purging iii. Absence of currently severe comorbid depression or self-harm iv. Ideally on no, or minimal Psychotropic medication v. No recent history of compulsory treatment ( i.e. in the last year) vi. Has tried existing treatment options without success vii. Has never had periods of full remission viii. Intensity of desire to recover ix. Good intellectual function and education - well able to comprehend the facts of the intervention (e.g. healthcare background, university education) x. Poor social and occupational function, high levels of distress - not functioning well, having poor quality of life.

The highest ranked patients will be asked if they consent to receiving information about the project. If they are interested they will be approached by Dr Park with information about the project, given an information sheet and the opportunity to discuss with family members. If they remain interested they be will be offered the opportunity to meet with Dr Park and Professor Aziz, for further information and to address questions , and involve their family if they wish. Following this they will be reviewed by Dr Park a week later and thus given a up to week to decide if they would like to participate. Consent to participate in the project will be taken by Dr Park and can be freely withdrawn at any point with no effect on their usual clinical care. Throughout the study the patients will receive supportive treatment as usual from the clinical team who will be independent from the research team, and it will be made clear they can opt out of the study at any point without having to give explanations.

If the patient consents to participation they will have preoperative assessments in the University Research Department at the Warneford hospital:

I: PREOPERATIVE PHASE: 2 months

1. Initial assessment with Research team. BMI: height and weight Semi structured interviews of Eating Disorder and comorbidity: EDE, SCID, Yale-Brown-Obsessive Compulsive scale ( YBOCS); Self reported questionnaires as above indexing mood and quality of life Computerised Behavioural tasks: Leeds -Oxford food preference task , Habit formation task

2. Weekly self report measures of symptoms done at home, weekly (5 minutes) using the' True Colours' online self monitoring system pioneered at the University of Oxford Department of Psychiatry, adapted for Eating disorders by Dr Park including the OXBREAD brief rating scale. http://oxtext.psych.ox.ac.uk/true-colours. This self report system will continue from entry to the study end. These self report schedules will be augmented by brief weekly on - line face to face reviews by skype or face time with the aim of offering participants maximal support and monitoring.

3. Neuropsychological assessment battery, duration approximately 1 hour: (Neuropsychological assessments are routine for all DBS operations done in Oxford).

   • WAIS IV subtests : Vocabulary , Matrix Reasoning , Digit Span , Coding (Wechsler 2008): Measures of general intellectual skills that is needed to make sense of performance across other cognitive domains.

   • BIRT Memory and Information Processing Battery : List Learning Task. Baseline measure of Verbal Memory and Learning( Coughlan, Oddly, & Crawford, , 2007).

   • Verbal Fluency. Baseline measure of Phonemic and Semantic Fluency(Spreen ,& Strauss 1998).

   • CANTAB: IED and SWM subtests: Assess executive faculties such as set shifting and spatial working memory (Robbins, James, Owen, Sahakian, McInnes& Rabbitt, 1994)

4. Preoperative Capacity assessment Once each suitable patient is identified, and consented , Dr Jacinta Tan will interview the patient independently to do a full MacCAT-CR assessment plus in depth interview to explore the patient's experience of treatment, rationale for participation and motivations for taking part in research from the participants' perspective, to ensure a high level of capacity , voluntariness and informed consent.

   The assessment is expected to take 2.5 -3 hours and can be done in sections if the patient tires. It will be tape recorded and transcribed. Jacinta will score the MacCAT-CR and an independent observer will also score the MacCAT-CR. Both will also provide global assessments of capacity to consent to research. Dr Tan will provide a detailed report of the interview and her assessment of ethical acceptability of participation for that individual. If she decides the patient is not suitable, the process will be repeated with the next individual on the list who agrees.

5. Neuroimaging:

Pre-operatively the patient will have:

a whole brain MRI scan including T1, T2, Inversion Recovery sequences and fMRI in resting state and task based using our food reward task: These scans will occur at the John Radcliffe Hospital , Oxford; total duration 1 hour.

a whole brain MEG scan- resting state and task based food reward task , which will be done at the Warneford Hospital, Oxford, duration 30 minutes.

Medical optimisation: patients will have normalisation of electrolytes and ECG prior to being considered fit for the operation. They will need to be a BMI of> 13.

II OPERATIVE PHASE: Month 2.

The DBS operation-will require up to 4 days of inpatient stay at the John Radcliffe Hospital On the day of surgery, the patient will be anaesthetised and the base ring fixed to the skull and a CT scan performed with a localiser fixed to the ring. The patient will then be transferred to theatre and the scalp cleansed.

By fusing the structural MRI scan to the stereotactic CT scan the trajectory to implant electrodes into the nucleus accumbens bilaterally will be calculated. Then through bilateral scalp incisions and a twist drill skull perforation, deep brain electrodes will be passed to target bilaterally and fixed to the skull with titanium mini-plates. The scalp will then be closed and a repeat stereotactic CT scan will be obtained to confirm electrode placement.

Having done so, back in theatre the electrodes will be connected to extension cables that will then be passed sub-cutaneously down one side of the head, behind the ear to a sub-cutaneous pouch below the collar bone and connected to a rechargeable pacemaker and all the wounds closed. Video recordings are used as part of the assessment process during routine implantation of DBS wires whether taking part in a study or not and is therefore considered part of standard operating procedure. They will be assessed as part of the study to measure the response to DBS.

The day after surgery, the pacemaker will be turned on to ascertain any immediate effects of stimulation on symptoms and then switched off again. Having done so the patient will be discharged two days after implant to be assessed six weeks after implant when all wounds are healed and any acute effects of the DBS electrode insertion operation such as headaches, scalp soreness or wound tenderness have worn off.

III: POST OPERATIVE FOLL0W-UP PHASE: Month 3 ( off): Month 4-15: DBS On

In the first post -operative month the stimulator will remain off. At the end of Month 3: post op MEG, followed by DBS switch on Month 4-6: 3 month dose optimisation period: dose adjustment may require more frequent visits to Oxford, up to weekly, for neurosurgical review Months 7-9: Dose stabilization period Month 10: Two fortnight periods of double blind either DBS On or off. MEG scans in each condition.

Months 11-15: DBS at stable optimised dose. At month 15: Final MEG scan, neuropsychology

Post operative assessments, aiming to track the neural and symptomatic change as a result of DBS will involve:

1. Continuation of weekly self report rating scale of eating disorder, anxiety and depression involving the True Colours text messaging system started in phase I: http://oxtext.psych.ox.ac.uk/true-colours (5 minutes)

1. Monthly reviews (up to 2 hours) in the Warneford Research department incorporating:

• Interview with Dr Park , research assistant and/or DPhil student: BMI: Semi structured interviews including Eating Disorder Eating Disorders Examination (EDE), comorbid psychopathology (SCID) Yale-Brown-Obsessive Compulsive Scale (YBOCS), . Behavioural paradigms: food reward (Oxford-Leeds food preference task, habit task; Self report schedules as in phase 1 assessing eating disorder symptoms, starvation dependence, functional impairment and quality of life .

• Joint Neurosurgery-psychiatric reviews with Professor Aziz , and/ or Senior nurse practitioner , and Dr Park: During this review neurosurgical parameters such as the degree of stimulation, and psychiatric parameters such as experience of eating disorder and comorbid psychiatric symptoms will be jointly assessed. Any necessary adjustments to stimulator intensity will then be programmed.

Post operative neuroimaging :

MEG resting state and food reward task based: (LO food preference task (OHBA, Warneford Hospital) - 30 minutes; at month 3 ( pre DBS switch on , 2 scans in month , 10 in DBS ON and OFF conditions , and end of study, month 15 Post operative Neuropsychological assessment 12 month after DBS on ie at 15 months. approximately 1 hour.

Optional post operative Ethics sub-study:

• A separate, post operative ethics sub study will be optional and participants will be informed about this option in the main study patient information sheet, In the main study consent form they will be asked if they agree to being contacted about the ethics sub-study post operatively or not.
• At the end of the study it will thus be possible to produce a detailed report of the ethics of the research, including the participant's perspective of the research experience

Optional Annual follow up phase for up to 5 years postop: repeating interview and questionnaires measures taken at entry and protocol end .

Inclusion Criteria:

1. Primary diagnosis: Anorexia Nervosa according to the DSMIV (Diagnostic and Statistical Manual of Mental Disorders Fourth Edition)criteria based on a psychiatric interview

2. Illness duration of severe AN > 7 years

3. Disabling severity with substantial functional impairment

4. Female Gender and right handed

5. Treatment refractoriness, defined as lack of response to two or more typical modes of treatment, such as inpatient weight restoration, psychotherapy and psychopharmacology

6. Severely underweight: BMI >13 <16

7. Age: 20-65 years old

8. Written informed consent

9. Able to fully understand the consequences of the procedure

10. English speaking and able to answer the study questions fluently

11. Having the mental capacity to make provide informed consent to research participation

    -

Exclusion Criteria:

1. Unstable physical condition (severe electrolyte disturbances, cardiac failure, other physical conditions due to underweight in which surgery/anaesthesia is contraindicated)

2. Treatable underlying cause of anorexia/underweight

3. Parkinson's disease, dementia, epilepsy

4. History of schizophrenia/ psychosis, bipolar disorder

5. Alcohol or substance abuse (including benzodiazepines) during the last 6 months

6. Current severe major depressive or Tic disorder

7. Antisocial or Severe Borderline Personality Disorder

8. Standard MRI scan exclusion criteria (pregnancy, pacemaker and metals contraindicated for MRI except for the DBS implantation and stimulator itself)

9. Current Involuntary patient

   -