AFFINITY Study Protocol: Methods and Eligibility
The AFFINITY Trial was an Australasian, investigator-driven, multi-centre, randomised, double-blind, placebo-controlled clinical trial investigating whether routine treatment with fluoxetine could improve functional recovery following recent stroke. The trial was funded by the National Health and Medical Research Council (NHMRC) and conducted across hospital stroke units and rehabilitation centres in Australia, New Zealand and Vietnam.
Trial Overview
The formal title of the study was “An Australasian, investigator-driven, NHMRC funded, multi-centre, prospective, randomised, parallel group, double-blind, placebo-controlled trial to establish the effect(s) of routine administration of fluoxetine (20 mg once daily) in patients with recent stroke.”
The short title was Assessment of Fluoxetine in Stroke Recovery (AFFINITY) Trial, with the acronym AFFINITY.
The study was classified as a Phase IIIb clinical trial. Although fluoxetine was an established treatment for depression, its routine use for stroke recovery represented a potential new indication.
Primary Research Question
The primary research question was whether treatment with fluoxetine 20 mg once daily, commenced between 2 and 15 days after stroke onset and continued for 180 days, could improve functional outcomes at the 180-day assessment.
The trial was designed to assess the effects of treatment in adults who had experienced a recent ischaemic or haemorrhagic stroke and had persistent measurable neurological impairment.
Trial Design and Setting
AFFINITY was a prospective, parallel-group, randomised, placebo-controlled and double-blind clinical trial.
Participants were randomly assigned to receive either fluoxetine or a matching placebo for 180 days. The study was conducted through hospital stroke units and rehabilitation centres across Australia, New Zealand and Vietnam.
The trial was led by Associate Professor Maree Hackett of The George Institute for Global Health and The University of Sydney, and Professor Graeme Hankey of The University of Western Australia and Sir Charles Gairdner Hospital.
Participant Eligibility
The trial included men and women aged 18 years or older who had experienced a stroke between 2 and 15 days before randomisation.
Eligible participants had brain imaging consistent with either ischaemic or haemorrhagic stroke and persistent measurable focal neurological deficits. These could include motor, sensory, visual, language or cognitive impairments. The impairment had to be sufficient to produce a modified Rankin Scale score of at least 1 and to warrant treatment from the perspective of the participant or their carer.
Participants were excluded when factors such as previous epileptic seizures, bipolar disorder, previous drug overdose or attempted suicide, current SSRI treatment, significant hepatic or renal impairment, severe hyponatraemia, or clinically important medication interactions made fluoxetine treatment unsuitable.
Other exclusion criteria included limited availability for follow-up, life-threatening illness likely to affect survival, pregnancy or breastfeeding without appropriate contraception, and participation in another interventional clinical trial involving an investigational medicine or device.
Randomisation and Treatment
Randomisation was performed using a password-protected central computerised randomisation service available 24 hours a day. A minimisation algorithm was used to maintain balance between the treatment groups according to key prognostic factors.
These factors included the time from stroke onset, presence of a motor deficit, presence of aphasia, and predicted probability of survival free of dependency at six months.
Participants were assigned to one of two treatment groups:
- Fluoxetine 20 mg once daily for 180 days
- Matching placebo once daily for 180 days
For participants unable to swallow capsules, the contents of an opened capsule could be administered through an enteral feeding tube.
Outcome Measures
The primary outcome was functional status at 180 days after randomisation, measured using the simplified modified Rankin Scale questionnaire (smRSq).
Secondary outcomes were assessed at 180 and 365 days and included survival, mood, cognitive function, communication, motor function, overall health status, health-related quality of life and functional recovery.
The study also assessed new diagnoses of depression requiring antidepressant treatment and fatigue using established clinical assessment tools.
Serious adverse events were monitored throughout follow-up. These included recurrent stroke, acute coronary syndrome, major bleeding, falls, fractures, epileptic seizures, symptomatic abnormalities in blood glucose or sodium levels, attempted suicide or self-harm, and death.
The trial also assessed healthcare costs during the first year and the cost-effectiveness of the intervention.
Follow-Up
Participants were followed at 28, 90, 180 and 365 days after randomisation.
Site investigators conducted assessments at 28 and 90 days in hospital, outpatient clinics or by telephone. The AFFINITY Trial Coordinating Centre conducted the 180- and 365-day assessments by telephone or postal questionnaire.
This follow-up schedule allowed the study team to assess both the early and longer-term effects of the intervention after stroke.
Sample Size and Statistical Analysis
The planned sample size was 1,600 participants, with 800 participants allocated to each treatment group.
The study was designed with 90% statistical power to detect an absolute increase in the proportion of participants achieving a modified Rankin Scale score of 0–2 at six months from 50% to 57.5%.
The primary statistical analysis used ordinal logistic regression of the modified Rankin Scale outcome, adjusted for baseline variables incorporated into the randomisation minimisation algorithm.
Trial Duration
The original protocol described a planned study period; actual trial recruitment ran from January 2013 to June 2019.
The protocol summary used for the study was Version 4, dated 18 November 2015. The trial was designed to provide evidence on whether routine administration of fluoxetine after recent stroke could influence functional recovery and other important patient outcomes.
Trial Coordination and Collaboration
The AFFINITY Trial was coordinated by the AFFINITY Trial Coordinating Centre, Stroke Research Group, Harry Perkins Institute of Medical Research, Sir Charles Gairdner Hospital, Western Australia.
Researchers and institutions interested in collaboration or further information were invited to contact the AFFINITY Trial Coordinating Centre.
Protocol Sources and Related Pages
This page summarises the historical protocol, including its planned sample size of 1,600. The completed trial recruited 1,280 participants in Australia, New Zealand and Vietnam. Consult the ANZCTR registration ACTRN12611000774921, the published FOCUS, AFFINITY and EFFECTS protocol and the AFFINITY primary results to distinguish planned methods from the final study report.
For a shorter introduction, read the study design overview. The exclusion criteria describe participation restrictions, and the research team page explains coordination and funding.