We are updating our Privacy Policy and Terms and Conditions to help you clearly understand how your personal information is collected, stored and managed. Learn More
Menu Back to Poster-Presentations-Details

M 07: Synergetic Prevention of Sudden Death by ACEI, Statin and Gliflozin in Type 2 Diabetes: A Simulation Study





Poster Presenter

      Hai-Ha Le

      • PhD Candidate, Pharmacology
      • Claude Bernard Lyon 1 University
        France

Objectives

Sudden cardiac death (SCD) is the first cause of cardiovascular deaths in type 2 diabetes (T2D). ACEI, statin and recently empagliflozin have shown efficiency for SCD prevention in T2D. We estimated the public health impact of this tri-therapy on SCD in a diabetic virtual realistic population (VRP).

Method

We developed a SCD risk score for T2D from 7 RCTs; estimated risk reductions on SCD by each of the tri-therapy through related meta-analyses and trials; finally, we integrated these results on a VRP generated from real French cohorts with risk-based and different therapeutic strategies scenarios.

Results

A predictive SCD score was built from 30560 individual patient data of 7 randomized controlled trials (INDANA database and DIABHYCAR trial), using Cox regression proportional hazards model, censored at 5-year period. The significant risk factors were age (HR 1.07, CI 1.06-1.09, p <1e-16 for 1year), male sex (HR 2.32, CI 1.65-3.26, p <1e-6), total cholesterol level (HR 1.13, CI 1.02-1.26, p= 0.02), systolic blood pressure (HR 1.011, CI 1.003-1.019, p <0.005), smoking status (HR 1.84, CI 1.40-2.41, p <1e-5), history of myocardial infarction (HR 2.15, CI 1.47-3.13, p <7e-5), and diabetes mellitus status (HR 2.93, CI 1.80-4.76, p =1.5e-5). Important interaction between diabetes status and sex was found, indicating that in diabetic patients, both sexes might have comparable risks of SCD. Area under the receiver-operating characteristics (ROC) curve of the model was of 70%. A clinical scoring system was also established for simple assessment of SCD risk. Our model was applicable for patients with T2D and/or hypertension and was the first score for SCD prediction in cardiovascular primary prevention. Latest related meta-analyses and trial indicated significant relative risks (RRs) of 90%, 85% and 69% on SCD risk in T2D by statin (Rahimi et al.2012), ACEI (Benoit et al. 2014) and gliflozin (EMPAREG-OUTCOME trial, 2015), respectively. Supposing no interaction existing between these drugs, the relative risk of the tri-therapy was RR = 0.90 * 0.85 * 0.69 = 0.53. A French diabetic RVP of 176 187, aged from 40 to 75 was generated from a 8 995-patient sample, giving an median of SCD risk of 1.7% at a 5-year time horizon. A simulation of public health impact on this platform estimated the numbers needed to treat (NNTs) at 221 people for the whole population and at 105 among individuals of the highest 10% predicted SCD risk, if treated simultaneously this tri-therapy for one year. The corresponding untreated risks of SD were of 1.9% and 4.0% respectively.

Conclusion

For the first time, we developed a SCD risk score for primary prevention (also applicable for those with diabetes and/or hypertension). Our score confirmed common risk factors as suggested by classical scorer (cardiovascular Framingham score). As well, we proposed an innovative approach to estimate impact of various drug strategies on a VRP and approved its feasibility by a simulation in T2D by the statin-ACEI-gliflozin tri-therapy. Such co-prescription appears to prevent 1 SCD in 221 treated diabetic patients in 1 year. However, interaction between concerned these drugs should be further verified. A study to elucidate this point by our team suggested no important interactions could have significant impact on their combination. In general, we suggested the OPTI-VRP (OPTImize therapeutic strategies on a Virtual Realistic Population) approach to simulate public health impact (PHI) step by step: 1. O (Outcome): Choose outcome(s) of interest 2. P (Population): Define the population on which optimizing PHI 3. T (Treatment): Choose treatments of interest 4. I (Integration): Generate the targeted VRP and integrate available information obtained from OPT (the three items above) to simulate PHI. This OPTI-VRP is a multi-component approach which allows accurate fitting to the characteristics of the particular population of interest. In perspectives, we could: i. Re-use VRP for other objectives, such as validation of risk scores; ii. Enrich the approach by external data/information source; and iii. Integrate various constraints of optimization, eg. cost, utility, side effects, etc… Our OPTI-VRP approach that gathers effect models (via meta-analyses and risk scores) and VRP simulation provides a clinical powerful tool. This could help valuing each evidence-based component, better transposing clinical trial results into practice, facilitating clinical decision in both public health and at individual levels, on both medical and economic aspects.