Dear Mr Covell,

I write in response to your regulation 28 report, sent to NICE on 4 September 2023, regarding the very sad death of Ms Talia Phillips. I would like to offer my sincere condolences to Ms Phillips’ family. We have reflected on the circumstances surrounding Ms Phillips’ death and the concerns raised in your report regarding monitoring requirements for fluoxetine. We have made recommendations on the use of antidepressants in our guidelines on the treatment of anxiety and we have also published guidance on safe prescribing of antidepressants in our guideline on medicines associated with dependence or withdrawal symptoms. However, we consider that the Medicines and Healthcare products Regulatory Agency (MHRA), as the regulator or medicines, would be best placed to address concerns you have raised regarding monitoring requirements as these are covered by the summary of product characteristics (SmPC) for a drug, a document which is agreed by the MHRA. We would therefore suggest you send the regulation 28 report to the MHRA for their consideration. If the MHRA issues revised prescribing advice, we would consider whether any relevant NICE guidance needs to be amended. Please do let me know if you require any further information.

Dear Mr Covell,

Regulation 28 Report concerning Talia Evania Phillips, DOB 18/08/2000 Thank you for your Regulation 28 Report relating to the death of Talia Evania Phillips. I would like to offer my sincere condolences to Ms Phillip’s’ family on their tragic loss.

In the Matters of Concern section of the report relating to the tragic death of Talia Evania Phillips you request that guidance in relation to the prescribing of fluoxetine and management of patients on fluoxetine should be reviewed to consider in what circumstances a blood test to establish the level of fluoxetine in the patient's blood would be advisable.

We have reviewed the available evidence from the fluoxetine Summary of Product Characteristics (SmPC), data from the UK Yellow Card Scheme, literature1-8 as well as the advice of our Expert Advisory Group (EAG) of the Commission on Human Medicines on the monitoring of blood levels of antidepressants in patients on fluoxetine treatment which was sought in May 2020 in response to a fatal case report and a Regulation 28 request.

The EAG previously advised that the evidence from the analyses of Yellow Card data and published information on antidepressant drug level monitoring was not sufficiently robust to advise clinicians to routinely monitor blood levels of antidepressants for all patients on treatment. The Group recommended however that blood level monitoring of antidepressants may be helpful in certain circumstances, for example in the event of symptoms suggestive of toxicity or when concomitant medicines may interact to increase antidepressant drug levels.

Circumstances which can have an impact on fluoxetine levels are described in the SmPC and the medications which are known to alter plasma levels of fluoxetine are detailed in the

fluoxetine SmPC section 4.5 on Interactions with other medicinal products and other forms of interactions. The use of fluoxetine concomitantly with a number of medicines requires caution, and these are listed in the attached Annex.

When fluoxetine drug level testing was previously discussed, the EAG commented that fluoxetine has a good safety margin in overdose with even a 10-fold increase in dose causing only low toxicity. The EAG acknowledged that fluoxetine has a wide therapeutic range and that currently there is no robust data regarding therapeutic plasma levels to support guidance. Overall, the EAG concluded that routine therapeutic drug level monitoring for fluoxetine would not be recommended unless clinically indicated based on risk factors for toxicity and QT prolongation in the patient. In addition, the fluoxetine SmPC describes that steady state plasma concentrations are dependent on body weight. Fluoxetine is extensively metabolised by the liver and excreted by the kidneys. A lower dose, e.g., alternate day dosing, is recommended in patients with significant hepatic dysfunction. When given fluoxetine 20mg/day for 2 months, patients with severe renal failure (GFR <10ml/min) requiring dialysis showed no difference in plasma levels of fluoxetine or norfluoxetine compared to controls with normal renal function.

We note that the Deceased had a normal ECG during fluoxetine treatment in relation to an episode of palpitations. Palpitations at normal therapeutic doses are listed in the fluoxetine SmPC as commonly occurring in association with fluoxetine use and Electrocardiogram QT prolonged (QTcF≥450msec) is also listed as common.

The fluoxetine SmPC describes that signs of toxicity in overdose of fluoxetine alone usually have a mild course. Symptoms of overdose have included nausea, vomiting, seizures, cardiovascular dysfunction ranging from asymptomatic arrhythmias (including nodal rhythm and ventricular arrhythmias) or ECG changes indicative of QTc prolongation to cardiac arrest (including very rare cases of Torsade de Pointes), pulmonary dysfunction, and signs of altered CNS status ranging from excitation to coma. Fatality attributed to overdose of fluoxetine alone has been extremely rare.

The approved fluoxetine SmPC contains information reflecting the currently available data on known interactions and clinical circumstances which may predispose a person to fluoxetine toxicity and describes symptoms of toxicity in overdose. The fluoxetine SmPC does not make specific recommendations on when to perform blood tests to establish the level of fluoxetine as this is a clinical judgement depending on the unique individual patient circumstances and therefore would be a matter for clinical guidelines.

In order to complete our assessment of the adequacy of the current fluoxetine product information, we have sought details of any other medical indications in addition to anxiety, any concomitant medications the Deceased was prescribed, any information about the prescribed dose and duration of fluoxetine use in addition to the blood levels found and information on previous tests performed and a copy of the postmortem report via a response for further information to the registered Yellow Card.

If the follow up information indicates that additional guidance would be beneficial, a further review of the adequacy of the fluoxetine product information will be undertaken.

Finally, I take this opportunity to confirm that this case report has been added to the Yellow Card database (reference number ADR 28344736), which is the UK’s system for collecting and monitoring information on suspected adverse drug reactions (ADRs) and medical device adverse incidents.

Should you have any further questions, please do not hesitate to contact me.