A Quantitative Framework to Evaluate and Translate Proarrhythmic Risk November 15th 2016
Lars Johannesen, PhD
Disclaimer • The opinions expressed in this presentation are mine and do not necessarily reflect the official views of the U.S. Food and Drug Administration (FDA)
Outline • Background • QTc prolongation and torsade • Impact of regulatory response • Limitations of current approach
• The Comprehensive In vitro Proarrhythmia Assay (CiPA) • Overview • Translation of preclinical findings to the clinical ECG
• Summary
Background • In the late 1980s and early 1990s a growing awareness for the potential for non‐cardiac drugs to cause torsade and QT prolongation was observed
Stockbridge, Drug Saf, 2013
QTc prolongation and torsade hERG potassium block
Early after depolarization (EAD) (calcium and sodium current)
Action potential
ECG
Example of torsade de pointes QT
Wu, Cardiovasc Res, 2008; January, Circ Res, 1989; Chokr, Arg Bras Cardiol 2014
Regulatory response • The resulting regulatory response included two ICH guidances: • ICH S7B: Non‐clinical assessment of potential for QT prolongation • ICH E14: Clinical assessment of the potential for QT prolongation • The guidances were adopted in the US in 2005, and describe the assessment of QT prolongation potential of new drugs.
Impact of regulatory response • A consequence of placing the hERG assay early in development is that candidates might be down prioritized or discarded based on their affinity for hERG • As many as 60% of new molecular entities developed test positive for hERG and as a result are discarded early in development De Ponti 2008
• Using the hERG assay as a gate keeper would be appropriate if block of the hERG potassium channel was a good predictor of torsade.
Limitations of prediction of QT risk using hERG • Redfern et al. compared the relationship between hERG block and QT / torsade • Their work suggested a 30‐fold safety margin
• A study by Kramer et al. showed importance of considering other ion channels for risk prediction • This is due to drugs such as verapamil and ranolazine, which both block not only hERG but also inward currents (L‐type calcium/late sodium) Redfern, Cardiovasc Res, 2003; Kramer, Sci Rep 2013; Hoekstra, Front Physiol 2012
Role of multichannel block • Block of inward currents (L‐type calcium and late sodium) is important • These currents have been implicated in the genesis of EADs, a proposed trigger of torsade January, Circ Res, 1989; Wu, Cardiovasc Res, 2008
• Focus on just hERG and QT are therefore unlikely to be successful for prediction of torsade risk
Gintant, Nat Rev Drug Discov, 2016
Comprehensive In vitro Proarrhythmia Assay (CiPA) Goal: Develop a new in vitro paradigm for cardiac safety evaluation of new drugs that provides a more accurate and comprehensive mechanistic‐based assessment of proarrhythmic potential
Gintant, Nat Rev Drug Discov, 2016
10
CiPA – a global effort Nonprofits ‐ Public‐Private Partnerships • Health and Environmental Sciences Institute • Cardiac Safety Research Consortium • Safety Pharmacology Society Global Regulatory Agencies • U.S. Food and Drug Administration • European Medicines Agency • Japan Pharmaceuticals and Medical Devices Agency / NIHS • Health Canada Academia / Industry • Numerous Pharmaceutical and Device Companies • Numerous Academic Groups • Contract Research Organizations
Slide courtesy of David Strauss
Drug effects on multiple cardiac ion channels in vitro • Goal: Deliver robust, reliable and reproducible ion channel protocols to support in silico reconstruction of drug effects on the human ventricular cell • Challenges & Limitations: • Which ion channels should be selected? • What properties should be studied: % block vs. dynamic drug‐ion channel interaction? • What requirements are needed to deliver robust, reliable and reproducible ion channel data in a high throughput screening environment? Gintant, Nat Rev Drug Discov, 2016
Modified slide from David Strauss
In silico reconstruction of ventricular repolarization changes • Goal: Identify an appropriate human cardiomyocyte model and modify and calibrate model as necessary and select torsadogenic markers. • Challenges & Limitations: • Which human cardiomyocyte model should be used? • What model modifications if any are necessary? • Which markers of torsadogenic potential should be extracted from the model? Gintant, Nat Rev Drug Discov, 2016
Modified slide from David Strauss
In silico reconstruction and risk assessment • At a consensus meeting in July 2013 the O’Hara‐Rudy model was selected • Using CiPA compound training drugs, risk metrics based on the in silico reconstruction are being developed
O’Hara, PLoS Comput Biol, 20117
Confirmation using human stem cell cardiomyocytes • Goal: Compare to responses from in vitro ionic current studies that guide in silico reconstructions and identify unexpected differences would require further consideration • Challenges & Limitations: • Current commercial cell lines includes a mixed phenotype (ventricular, atrial) • Which parameters should be extracted? Action potential duration? Field potential duration? • How should impact on rate be accounted for?
Gintant, Nat Rev Drug Discov, 2016 Modified slide from David Strauss
• What requirements are needed to deliver robust, reliable and reproducible data?
Cardiomyocyte evaluation • Assessment of drug‐induced effects using MEA and VSO, include: • Action potential duration/field potential duration • Beat frequency • Upstroke velocity
• A pilot study is currently ongoing assessing drug‐induced effects at 12 different sites using CiPA compound training drugs. Gintant, Nat Rev Drug Discov, 2016 Modified slide from David Strauss
Phase 1 ECG assessment •
Goal: Use human Phase 1 ECG data to determine if there are unexpected ion channel effects compared to preclinical ion channel data
•
Challenges & Limitations: • Which biomarkers can be used to detect cardiac ion channel effects beyond hERG? • Are they affected by heart rate and if so how to account for rate changes • Can they be evaluated using exposure‐response using data from a SAD/MAD study
Gintant, Nat Rev Drug Discov, 2016
Modified slide from David Strauss
E‐R assessment using Phase 1 ECGs • Efforts to support analysis of ECG effects using E‐R and Phase I clinical ECGs was undertaken by IQ/CSRC, and a prospective study was completed in 2014. • The study met its primary analysis, detecting QTc effect in all positive drugs included. Darpo, Clin Pharmacol Ther, 2015
• ICH E14 Q&A was revised in December 2015 and a white paper is in process. • The results of the IQ‐CSRC study are encouraging, but the question of which ECG biomarkers that can be used to confirm the effects on other cardiac ion channels and if they can be evaluated using E‐R similarly to QT remain.
Assessment of ECG biomarkers
Vicente, J Am Heart Assoc 2015
Mechanistic ECG biomarkers
While QTc reflects overall repolarization, J‐Tpeakc reflects the part of the action potential influenced mostly by inward currents, which can mitigate drug‐induced torsade.
J‐Tpeakc represents a balance of inward and outward current block
Johannesen et al, Clin Pharmacol Ther 2014
J‐Tpeakc reflects a balance of outward (hERG) and inward current block (late sodium or calcium), of importance for torsade risk.
Prospective assessment of new biomarkers • The results of the retrospective analysis are encouraging and shows the potential of new ECG biomarkers • To confirm the retrospective observations two prospective clinical trials were conducted: • Study 1 – Comparison of ECG signatures between hERG blockers with high and low risk for torsade • Study 2 – Evaluation of ability of late sodium or calcium to shorten hERG‐induced QTc prolongation
Prospective study 1 • The goal of the first prospective study was to compare changes in the ECG between:
Vicente, J Am Heart Assoc 2015
Additional inward current block is reflected in the J‐Tpeakc interval
The findings of the first prospective study are concordant with the retrospective study confirming J‐Tpeakc as an ECG biomarker of inward current block. Johannesen et al, Clin Pharmacol Ther 2014
Prospective study 2 • The goal of the second prospective study was to evaluate if drug‐induced QTc prolongation could be mitigated by inward current block • hERG (dofetilide) either alone or in combination with a late sodium current blocker (mexiletine or lidocaine) • hERG (moxifloxacin) either alone or in combination with a calcium channel blocker (diltiazem)
Late sodium current block reduces QTc via J‐Tpeakc shortening
Late sodium block reduces drug‐induced QTc prolongation by shortening the J‐Tpeakc interval. Johannesen et al, Clin Pharmacol Ther 2016
No observed QTc shortening with calcium channel block
No QTc shortening with co‐administration of a L‐type calcium channel blocker was observed, however, this could be confounded by study design. Johannesen et al, Clin Pharmacol Ther 2016
Next steps for assessment of drug effects with novel biomarkers • The results from the retrospective analysis and two prospective studies are overall supportive of the use of novel ECG biomarkers to detect the presence of inward current block • However, the impact of L‐type calcium channel block combined with a hERG potassium channel block is not well understood.
• A third prospective clinical study is currently being designed and is targeted for completion by late 2017 focusing on the assessment of novel ECG biomarkers in a small phase 1 like study.
Summary • Assessment of risk for torsade based on QT assessment and hERG potassium channel block is associated with a high sensitivity but low specificity • CiPA is a comprehensive, mechanistic‐based approach, comprising of assessment of effects on multiple cardiac ion currents with in silico reconstruction and confirmation in stem cells • CiPA is being proposed as a replacement to the current paradigm (QT & hERG) • A key component of CiPA is the confirmation of the preclinical components in humans, which depend on assessment of ECG effects based on E‐R analysis of ECG biomarkers such as QT
Acknowledgments FDA • David Strauss • Christine Garnett • Norman Stockbridge • Jose Vicente • Jiang Liu • Jeffry Florian • Dhananjay Marathe • Meisam Hosseini • Zhihua Li • Sara Dutta • Ksenia Blinova
•
CiPA steering committee –
•
All CiPA Working groups – – – –
•
Ayako Takei, Bernard Fermini, Colette Strnadova, David Strauss, Derek Leishman, Gary Gintant, Jean-Pierre Valentin, Jennifer Pierson, Kaori Shinagawa, Krishna Prasad, Kyle Kolaja, Natalia Trayanova, Norman Stockbridge, Philip Sager, Tom Colatsky, Yuko Sekino, Zhihua Li Ion channel working group In silico working group Cardiomyocyte working group Phase 1 ECG working group
ALL contributors to CiPA – – – –
HESI, SPS, CSRC FDA, EMA, PMDA, NIHS, Health Canada Many pharmaceutical, CRO, and laboratory device companies Academic collaborators
References • • • • • • • • • • • • • • • •
Chokr MO, et al. Arg Bras Cardiol 2014;102(6):e60-4 Darpo B, et al. Clin Pharmacol Ther 2015;97(4):326-35 De Ponti 2008 Gintant G, et al. Nat Rev Drug Discov 2016;15(7):457-71 Hoekstra M, et al. Front Physiol 2012 January CT, et al. Circ Res 1989;64(5):977-90 Johannesen L, Clin Pharmacol Ther 2014;96(5):549-58 Johannesen L, Clin Pharmacol Ther 2014;95(5):501-8 Johannesen L, Clin Pharmacol Ther 2016;99(2):214-23 Kramer J, et al. Sci Rep 2013;3:2100 Monahan BP, et al. JAMA 1990;264(21):2788-90 O'Hara, et al., PLoS Comput Biol 2011;7(5):e1002061 Redfern WS, Cardiovasc Res, 2003;58(1):32-45 Stockbridge N, et al. Drug Saf 2013;36(3):167-182 Vicente J, et al., J Am Heart Assoc 2015;4(4) Wu L, et al. Cardiovasc Resc 2008;77(3):481-8