Precision medicine with prediction tools in high CV risk patients - - PowerPoint PPT Presentation

Precision medicine with prediction tools in high CV risk patients

Symposium “New concepts and models in CV risk management” ESC, Munich August 28, 2018 Frank L.J. Visseren

Faculty Disclosure

I I have received a research grant(s)/ in kind support

A From current sponsor(s) YES NO B From any institution YES NO

II I have been a speaker or participant in accredited CME/CPD

A From current sponsor(s) YES NO B From any institution YES NO

III I have been a consultant/strategic advisor etc

A For current sponsor(s) YES NO B For any institution YES NO

IV I am a holder of (a) patent/shares/stock ownerships

A Related to presentation YES NO B Not related to presentation YES NO

Declaration of financial interests For the last 3 years and the subsequent 12 months:

Faculty Disclosure

Declaration of non-financial interests:

• Professor of Medicine, epidemiologist
• University Medical Center Utrecht
• Member of Dutch guideline committees on cardiovascular prevention.

Investigator in phase II/III clinical trials

Great challenge for a clinician: translating the results

• f large clinical trials to individual patients

Lipid-lowering? Glucose-lowering? Antithrombotics? Blood pressure- lowering?

Much to consider, much to choose: what for who and when?

Riskfactor Drugs Dose / combi Treatment goal Lipids Statin Ezetimibe PCSK9-mab Dose? Combination? LDL-c <2.5 mmol/l LDL-c <1.8 mmol/l Even lower Blood pressure ACEi/ARB, Diuretics CCB, Betablocker, Spironolactone Dose? Combination? SBP <140 mmHg SBP <130 mmHg Elderly goal Antithrombotics Antiplatelet (COX, P2Y12, cAMP) DOAC Dose? Combination? Diabetes Metformin, SU,DPP-4 insulin, GLP-1, SGLT-2 HbA1c <53, <58, <64 mmol/mol Inflammation Triglycerides, Lp(a)

CV prevention starts with healthy lifestyle.

Why predicting risk?

Identifying high risk patients with modifiable risk factors To improve prognosis Shared and informed decision-making

But …..

Predicting risk is difficult Which score to use? How to interpret and communicate risk? Time consuming in clinical practice?

Accessible risk prediction tools

Tool Patient categories Geographical region Heart SCORE Healthy people Europe high and low risk regions QRISK3 Healthy people United Kingdom JBS-3 risk calculator Healthy people United Kingdom ASSIGN score Healthy people Scotland PROCAM score, Various websites Healthy people Germany CUORE Healthy people Italy ASCVD risk-estimator plus Healthy people United States Framingham risk score Healthy people United States Reynolds risk score Healthy people United States Globorisk Healthy people Worldwide UKPDS risk engine V2 Type 2 diabetes United Kingdom ADVANCE risk engine Type 2 diabetes Europe, Asia, Australasia and North America SMART risk score Vascular patients Europe and United States MAGGIC risk calculator Heart failure patients Worldwide Seattle Heart Failure model Heart failure patients Northern-America U-Prevent Healthy people Type 2 diabetes patients Vascular patients Elderly Europe and Northern-America

Distribution of 10-year risk for recurrent CV events in CVD patients

Dorresteijn ea, Heart. 2013 Jun;99(12):866-72 Kaasenbrood ea, Circulation. 2016;134:1419–1429 www.escardio.org Piepoli ea, EHJ 2016;37, 2315–2381

Total CVD risk = unmodifiable + modifiable Risk Factors

Adapted from: Ridker ea, Eur Heart J. 2016;37(22):1720-2

“Residual blood pressure risk” “Residual thrombotic risk” “Residual smoking risk” “Residual triglyceride risk” “Residual cholesterol risk”

Age Sex Family Hx eGFR

Challenges in CV risk prediction in apparently healthy people.

OK problem problem

Specific risk score for elderly Lifetime CVD risk score

Clin Res in Cardiol. 2017 Jan;106(1):58-68 De Vries et al, ESC congress 2018, abstract 114

10-year CV risk (adjusted for competing risks) Lipid-lowering

CVD risk prediction in elderly (>70 yrs)

Not all elderly at high risk

BP lowering

Competing risks Level of modifiable risk-factors Treatment horizon

Therapy benefit Costs and harms

Baseline risk Therapy effectiveness

Wouldn’t it be great to have lifetime predictions?

Lifetime predictions with Age as Time-Scale

Geskus, Biometrics 2011;67:39-49 Dorresteijn ea, BMJ. 2016 Mar 30;352:i1548

10-year risk Lifetime risk

Lifetime prediction of CV events in vascular patients: SMART-REACH model

C-statistic 0.67 (95% CI 0.66-0.68) C-statistic 0.68 (95% CI 0.67-0.70) Kaasenbrood ea, JAHA 2018 epub Dorresteijn, ESC congress 2018, abstract 3141

Lifetime risk prediction for CVD patients!

CV risk prediction in apparently healthy people <70 years: LIFE-CVD model

Lifetime CV risk score in primary prevention

N Jaspers, ESC congress 2018, Young Investigators Award Finalist, abstract 1149

• Apparently healthy people: LIFE-CVD model 1
• Derived and externally validated in cohorts: ARIC, MESA, EPIC, Heinz Nixdorf Recall (total n=69,523)

Externally validated scores for Lifetime CV risk and Lifetime therapy benefit

1 Jaspers, ESC congress 2018, abstract 1149 2 Kaasenbrood et al, JAHA 2018 3 Berkelmans et al, in revision

• Patients with CV disease: SMART-REACH model 2
• Derived and externally validated in cohorts: SMART, REACH (total n = 40,388)
• Patients with DM2: DIAL model 3
• Derived and externally validated in cohorts: Swedish NDR, Scottish diab reg, ADVANCE, ACCORD,

ASCOT, ALLHAT, SMART (total n = 587,151)

Lifetime CV risk score in primary prevention Lifetime risk prediction for CVD patients!

• Elderly patients: Elderly model 4,5
• Derived and externally validated in cohorts: PROSPER, SMART, ASCOT, HYVET (total n=11,090)

CV risk prediction in elderly

4 Clin Res in Cardiol. 2017 Jan;106(1):58-68 5 DeVries et al, ESC congress 2018, abstract 114

Lifetime risk Treatment effects CVD-free life

Individual lifetime treatment effects: gain in CVD-free life

• Patients free of CV disease or DM2: LIFE-CVD model 1
• Derived and externally validated in cohorts: ARIC, MESA, EPIC, Heinz Nixdorf Recall (total n=69,523)

Externally validated Lifetime CV risk scores and lifetime therapy benefit

• Patients with CV disease: SMART-REACH model 2
• Derived and externally validated in cohorts: SMART, REACH (total n = 40,388)
• Patients with DM2: DIAL model 3
• Derived and externally validated in cohorts: Swedish NDR, Scottish diab reg, ADVANCE, ACCORD,

ASCOT, ALLHAT, SMART (total n = 587,151)

Lifetime CV risk score in primary prevention Lifetime risk prediction for CVD patients!

• Elderly patients: Elderly model 4,5
• Derived and externally validated in cohorts: PROSPER, SMART, ASCOT, HYVET (total n=11,090)

CV risk prediction in elderly

Much to consider, much to choose: what for who and when?

Riskfactor Drugs Dose / combi Treatment goal Lipids Statin Ezetimibe PCSK9-mab Dose? Combination? LDL-c <2.5 mmol/l LDL-c <1.8 mmol/l Lower better? Blood pressure ACEi/ARB, Diuretics CCB, Betablocker, Spironolactone Dose? Combination? SBP <140 mmHg SBP <130 mmHg Elderly? Antithrombotics Antiplatelet (COX, P2Y12) DOAC Dose? Combination? Diabetes SGLT-2 GLP-1 HbA1c <53, <58, <64 mmol/mol New treatments Inflammation Triglycerides, Lp(a)

CV prevention starts with healthy lifestyle.

+ =

Interactive calculator

www.U-Prevent.com

Patient example A

www.U-Prevent.com

Systolic blood pressure 145 mmHg

Patient example A

www.U-Prevent.com

Patient example A

www.U-Prevent.com

Patient example A

www.U-Prevent.com

Patient example A

www.U-Prevent.com

Conclusion

• Prediction tools facilitate Precision Medicine
• Lifetime CVD risk prediction for:

– Apparently healthy people – Patients with vascular disease – Patients with Diabetes Mellitus

• Specific elderly 10-year CVD risk score
• Estimating gain in CVD-free life from (combination of) lipid lowering, blood

pressure lowering, antithrombotic treatment

Treating based on level of risk factors Treating based on risk Treating based on level of risk factors Treating based on individual treatment benefit Benefit-based Medicine!

www.U-Prevent.com