PREcision meDICine in Treatment in CardioVascular Disease (PREDICT-CVD)

Maastricht University Medical Centre (MUMC)

Status: Not yet enrolling

Conditions: Chronic Heart Failure; Acute Decompensated Heart Failure

Treatments: Other: Standard of Care (Investigator Choice); Other: ASSIST-HF AI- algorithm

Study type: Interventional

Funder types: Other

Identifiers

NCT07369141

No 10088702 (Other Grant/Funding Number)

No 101112022 (Other Grant/Funding Number)

PREDICT-CVD

NL 009912 (Other Identifier)

Details and patient eligibility

About

The main goal of PREDICT-CVS is to demonstrate that a personalized, AI-derived biomarker-guided medication treatment algorithm that provides advice to the treating professional is superior to prevent clinical events and improve quality of life (QoL) when compared to patients receiving standard treatment of care.

Participants will be randomized into two groups

• Group 1 - Standard treatment Participant will receive the usual treatment for heart failure, based on current medical guidelines.
• Group 2 - AI-supported treatment The professional will receive a personalized treatment plan generated by the AI program. This plan is based on medical information and biomarkers. The professional can choose whether or not to follow the AI's advice, reasons for not following the AI-based advice will be collected.

Participants will visit the outpatient clinic at 1, 2, 3 and 9 months after being randomized.

Full description

The PREDICT-CVD is a prospective, randomized controlled, open label, multinational trial, in which patients with HF will be randomized to either the control arm or the intervention arm. The control arm is the standard of care and the intervention arm is where an AI-derived biomarker-guided treatment algorithm will be applied, to assist the treating professional in treatment decision. Assessment of the primary endpoint will be at the 9 month onsite follow-up visit, further long-term outcomes will be collected via telephone visits every 6 months until the last subject has completed the 9 month onsite follow-up.

The main objective is to demonstrate that a personalized, AI-derived biomarker-guided medication treatment algorithm that provides advice to the treating HF professional is superior to prevent clinical events and improve quality of life (QoL) when compared to patients receiving standard treatment of care.

The primary outcome of the trial is a clinical benefit, defined as a hierarchical composite of death from any cause, number of heart failure events and or a greater difference in change from baseline in the Kansas City Cardiomyopathy Questionnaire Total Symptom Score at 9 months, as assessed using a win ratio.

The expected duration for patients will be between 9 and max 24 months from the moment of enrollment. From 9 months onwards patients will be contacted by phone every 6 months and information regarding (HF) events and concomitant medication will be collected.

Enrollment: 430 estimated patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

1. Provide written and dated informed consent for participation prior to trial admission,

2. Age ≥18 years, female or male

3. Diagnosis of heart failure* either before discharge of a heart failure hospital admission or in the outpatient setting,

4. Symptomatic, defined as NYHA class II - IV,

5. LVEF < 50% measured in the last 12 months (using any modality), with no occurrence of HF event(s) since that measurement) and

6. Those not previously treated with evidence-based therapies or sub-optimally treated with GDMT defined as; • Not treated with one of the foundational four** HF medication unless absolute contraindication or documented intolerance

   • Definition of heart failure according to the most recent Heart Failure Guidelines of the European Society of Cardiology

     • Foundational four heart failure medication: BBs, ARNi, MRAs and SGLT2i. NB. This does not include ACE and ARBs these are permitted.

Exclusion criteria

1. Inability to understand and sign informed consent,
2. Scheduled or on renal replacement therapy,
3. Clearly documented intolerance to BBs, ARNI, SGLT2i or MRAs
4. Diagnosis of acute myocarditis or (non)obstructive hypertrophic cardiomyopathy,
5. The presence of a mechanical assist device,
6. Scheduled for mechanical assist device or heart transplant,
7. Acute coronary syndrome (ACS) in the past 3 months or current angina pectoris ≥class III,
8. Requiring valvular surgery or revascularization in the upcoming 3 months or revascularization within the past 3 months,
9. Anticipated need for surgery or any cardiovascular intervention, except implantable cardioverter defibrillator and/or cardiac resynchronization therapy, within 4 weeks,
10. Other non-cardiac conditions with limited life expectancy (≤ duration of the trial/ 1 year),
11. Participation in another (intervention) clinical trial (registry studies not included) until the onsite EoS visit (V5) at 9 months has been completed.
12. Women who are pregnant, breastfeeding or may be considering pregnancy during the study duration.

Trial design

Primary purpose: Other

Allocation: Randomized

Interventional model: Parallel Assignment

Masking: Single Blind

---

430 participants in 2 patient groups

Standard of Care

Active Comparator group

Description:

Standard treatment for HF

Treatment:

Other: Standard of Care (Investigator Choice)

Intervention

Experimental group

Description:

AI algorithm to assist the treating HF professional in treatment decision

Treatment:

Other: ASSIST-HF AI- algorithm