Breakout #4 Phase 1 ECG:

Transcription

1 Breakout #4 Phase 1 ECG: Potential Role of ECG under CiPA HESI CSRC CiPA Meeting Washington, DC - May 22, 2018 Jose Vicente, PhD Christine Garnett, PharmD Division of Cardiovascular and Renal Products Center for Drug Evaluation and Research U.S. Food and Drug Administration

2 Disclaimer This presentation reflects the views of the authors and should not be construed to represent FDA s views or policies. 2

3 Role of Phase I ECG Data Under CiPA To determine if there are unexpected ion channel effects compared to preclinical ion channel data Vicente et al. Clinical Pharmacology & Therapeutics 2018 (doi: /cpt.896) 3

4 Potential impact of CiPA Can CiPA nonclinical proarrhythmia assessment combined with clinical ECG assessment better inform ECG monitoring (e.g., in specific populations and in the range of 10-20ms mean ΔΔQTc prolongation)? 4

5 Discussion Items Summary of results supporting J-Tpeakc use to inform multi-ion channel effects Role of ECG in a potential implementation of CiPA Integrated risk assessment, context of use and potential thresholds How to handle discrepancies Limitations/Challenges Challenges in concentration-ecg-response models Identifying nonlinear relationships in small sized trials Impact on interpretation of results Heart rate dependency of J-Tpeak interval J-Tpeak/RR adaptation and hysteresis Population vs. subject-specific corrections Are we ready? 5

6 ECG Biomarkers to Differentiate herg vs. Balanced Ion Channel Block Johannesen et al. Clinical Pharmacology & Therapeutics 2014;96:

7 herg vs. balanced ion channel block Dofetilide (predominant herg block) Ranolazine (herg + late sodium) herg block prolonged QTc by prolonging J-Tpeakc and Tpeak-Tend Balanced ion channel block prolonged QTc by prolonging Tpeak-Tend with no effect on J-Tpeakc Johannesen et al. Clinical Pharmacology & Therapeutics 2014;96:

8 Challenges Nonlinear concentration-response relationships Diltiazem (calcium block) did not shorten QTc prolongation by dofetilide (herg block) Better J-Tpeak/RR characterization may help to reduce variability of J-Tpeakc measures 8

9 Potential Role of ECG under CiPA 9

10 Challenges in concentration-ecg-response Can the SAD study provide sufficient data to characterize the ECG profile for drugs with nonlinear exposure-response relationships? Phase 1 vs. larger studies? ECG signatures Proarrhythmic risk vs. ion channel effects balanced ion channel block vs. proarrhythmic risk Use of other intervals (e.g., PR, QRS) as indicators of ion channel effects (e.g., calcium, peak sodium)? When to default back to TQT or C-QT? 10

11 Different herg + Calcium ECG signatures Diltiazem (calcium block) did not shorten QTc prolongation by dofetilide (herg block) However, verapamil (herg + calcium) prolonged QTc without prolonging J-Tpeakc Thus, ECG signature of herg + calcium is unclear Although verapamil QTc and J-Tpeakc signature is similar to ranolazine (herg + late sodium) Calcium block has been shown to prevent EADs in nonclinical assays No TdP with verapamil 11

12 Heart rate correction (J-Tpeak) Like QT, J-Tpeak is heart rate dependent Population (n=431 subjects) correction available 1 J-Tpeakc = J-Tpeak/RR 0.58 Adaptation to HR is not instantaneous J-Tpeak/RR hysteresis not characterized yet Characterization of HR dependency and adaptation in a larger population is ongoing Feasibility of individual/study-specific correction in Phase 1 studies? [1] Johannesen et al. Clin Pharm Ther

13 BACKPUP 13

14 CHALLENGES IN CONCENTRATION-ECG- RESPONSE 14

15 Nonlinear concentration-response (I) Ranolazine - Study 1 Ranolazine - Study 3 QTc Day 1 Day 3 Nonlinearity vs. limited number of subjects informing high concentrations? 15

16 Nonlinear concentration-response: Delayed effects ( Hockey stick )? QTc Day 1 Day 3 Delayed effects vs. different ion channel effects at different (higher) concentrations? 16

17 Nonlinear concentration-response: large intercepts with flat slope QTc J-Tpeakc Day 1 Day 3 Flat, linear or Emax concentration-j-tpeakc relationship? Not enough data at low concentrations? Not large enough sample size? 17

18 HERG + LATE SODIUM VS. HERG + CALCIUM 18

19 herg+calcium vs. herg+late sodium with different Drugs Verapamil (herg + Calcium) Ranolazine (herg + late sodium) QTc QTc J-Tpeakc J-Tpeakc 19

20 herg+calcium vs. herg+late sodium with Drug Combinations herg + calcium herg QTc J-Tpeakc Tpeak-Tend +12 ms herg -10 ms herg + calcium herg + calcium herg herg -20 ms herg herg herg + late sodium -20 ms herg + late sodium herg + late sodium 20

21 HEART RATE DEPENDENCY OF J-TPEAK 21

22 Heart rate correction (QT) Heart rate correction of QT approaches described in ICH/E14 Adaptation to new HR is not instantaneous (QT/RR hysteresis effect) 22

23 Heart rate correction (J-Tpeak) J-Tpeak is also heart rate dependent Population (n=431 subjects) correction available 1 J-Tpeakc = J-Tpeak/RR 0.58 Like for QT, adaptation to HR is not instantaneous J-Tpeak/RR hysteresis not characterized yet Characterization of HR dependency and adaptation in a larger population is ongoing Feasibility of individual/study-specific correction in Phase 1 studies? [1] Johannesen et al. Clin Pharm Ther

24 QT and J-Tpeak Heart Rate Dependency Biomarker-RR relationship for two different subjects (red vs. blue) at baseline 450 QT 400 J-Tpeak JJ TTTTTTTTTTTT = JJ TTTTTTTTTT RRRR 0.58 QT (ms) 400 J-Tpeak (ms) For 22 Subjects QT J-Tpeak Mean Slope SD of Slope SD/Mean (%) 24% 23% RR interval (ms) RR interval (ms) Population based (n=431 subjects) heart rate correction for J-T peak available (Johannesen, et al. Clin Pharmacol Ther 2014) Ongoing collaboration with Dr. Malik to characterize J-Tpeakc/RR hystereis 24

25 Sources of J-Tpeakc variability (I) HR: 78 bpm J-Tpeak(c): 216 (251) ms Ranolazine: 2.9 µg/ml HR: 67 bpm J-Tpeak(c): 203 (216) ms Ranolazine: 3.4 µg/ml Ranolazine: heart rate increase in 2 time-points in 2 subjects (i.e., 6 ECGs) drive wider CIs ~Cmax 25