Evaluating the Renal Safety of Investigational New Drugs: Where Should We Be Going?

We the authors work in the US Food and Drug Administration (FDA) review division responsible for the therapeutic agents for primary renal disease. We also field consultative inquiries regarding off-target adverse renal effects of drugs intended to treat other diseases. We do neither basic science research on renal diseases nor clinical studies of new drugs, but we are professional spectators of both. We offer here our thoughts on the challenge of identifying renal safety signals in the preclinical space and in the earliest phases of clinical development.

Reliable and developmentally appropriate study end points are needed to achieve drug development for treatment of pediatric pulmonary arterial hypertension.

OBJECTIVE: To identify suitable end points and surrogates for pediatric pulmonary arterial hypertension (PAH) as the lack of developmentally appropriate end point and clinical trials contribute to the unmet medical need. STUDY DESIGN: Reviewed the efficacy end points and surrogates for all trials (1995 to 2013) that were submitted to the Food and Drug Administration (FDA) to support the approval of PAH therapy and conducted literature search. RESULTS: An increase in the 6 min walking distance (6MWD) was used as a primary end point in 8/9 adult PAH trials. This end point is not suitable for infants and young children because of performance limitations and lack of control data. One adult PAH trial used time to the first morbidity or mortality event as a primary end point, which could potentially be used in pediatric PAH trials. In the sildenafil pediatric PAH trial, the change in pulmonary vascular resistance index or mean pulmonary artery pressure was used as a surrogate for the 6MWD to assess exercise capacity. However, two deaths and three severe adverse events during the catheterizations made this an unacceptably high-risk surrogate. The INOmax persistent pulmonary hypertension of the newborn trial used a reduction in initiation of extracorporeal membrane oxygenation treatment as a primary end point, which is not feasible for other pediatric PAH trials. A Literature review revealed none of the existing noninvasive markers are fully validated as surrogates to assess PAH efficacy and long-term safety. CONCLUSIONS: For pediatric PAH trials, clinical end points are acceptable, and novel validated surrogates would be helpful. FDA seeks collaboration with academia, industry and parents to develop other suitable and possibly more efficient efficacy end points to facilitate pediatric PAH drug development.

Late sodium current block for drug-induced long QT syndrome: Results from a prospective clinical trial.

Drug-induced long QT syndrome has resulted in many drugs being withdrawn from the market. At the same time, the current regulatory paradigm for screening new drugs causing long QT syndrome is preventing drugs from reaching the market, sometimes inappropriately. In this study, we report the results of a first-of-a-kind clinical trial studying late sodium (mexiletine and lidocaine) and calcium (diltiazem) current blocking drugs to counteract the effects of hERG potassium channel blocking drugs (dofetilide and moxifloxacin). We demonstrate that both mexiletine and lidocaine substantially reduce heart-rate corrected QT (QTc) prolongation from dofetilide by 20 ms. Furthermore, all QTc shortening occurs in the heart-rate corrected J-Tpeak (J-Tpeak c) interval, the biomarker we identified as a sign of late sodium current block. This clinical trial demonstrates that late sodium blocking drugs can substantially reduce QTc prolongation from hERG potassium channel block and assessment of J-Tpeak c may add value beyond only assessing QTc.

Results from the IQ-CSRC prospective study support replacement of the thorough QT study by QT assessment in the early clinical phase.

The QT effects of five "QT-positive" and one negative drug were tested to evaluate whether exposure-response analysis can detect QT effects in a small study with healthy subjects. Each drug was given to nine subjects (six for placebo) in two dose levels; positive drugs were chosen to cause 10 to 12 ms and 15 to 20 ms QTcF prolongation. The slope of the concentration/ΔQTc effect was significantly positive for ondansetron, quinine, dolasetron, moxifloxacin, and dofetilide. For the lower dose, an effect above 10 ms could not be excluded, i.e., the upper bound of the confidence interval for the predicted mean ΔΔQTcF effect was above 10 ms. For the negative drug, levocetirizine, a ΔΔQTcF effect above 10 ms was excluded at 6-fold the therapeutic dose. The study provides evidence that robust QT assessment in early-phase clinical studies can replace the thorough QT study.

Differentiating drug-induced multichannel block on the electrocardiogram: randomized study of dofetilide, quinidine, ranolazine, and verapamil.

Block of the hERG potassium channel and prolongation of the QT interval are predictors of drug-induced torsade de pointes. However, drugs that block the hERG potassium channel may also block other channels that mitigate torsade risk. We hypothesized that the electrocardiogram can differentiate the effects of multichannel drug block by separate analysis of early repolarization (global J-Tpeak) and late repolarization (global Tpeak-Tend). In this prospective randomized controlled clinical trial, 22 subjects received a pure hERG potassium channel blocker (dofetilide) and three drugs that block hERG and either calcium or late sodium currents (quinidine, ranolazine, and verapamil). The results show that hERG potassium channel block equally prolongs early and late repolarization, whereas additional inward current block (calcium or late sodium) preferentially shortens early repolarization. Characterization of multichannel drug effects on human cardiac repolarization is possible and may improve the utility of the electrocardiogram in the assessment of drug-related cardiac electrophysiology.

Improving the assessment of heart toxicity for all new drugs through translational regulatory science.

Fourteen drugs have been removed from the market worldwide because they cause torsade de pointes. Most drugs that cause torsade can be identified by assessing whether they block the human ether à gogo related gene (hERG) potassium channel and prolong the QT interval on the electrocardiogram. In response, regulatory agencies require new drugs to undergo "thorough QT" studies. However, some drugs block hERG potassium channels and prolong QT with minimal torsade risk because they also block calcium and/or sodium channels. Through analysis of clinical and preclinical data from 34 studies submitted to the US Food and Drug Administration and by computer simulations, we demonstrate that by dividing the QT interval into its components of depolarization (QRS), early repolarization (J-Tpeak), and late repolarization (Tpeak-Tend), along with atrioventricular conduction delay (PR), it may be possible to determine which hERG potassium channel blockers also have calcium and/or sodium channel blocking activity. This translational regulatory science approach may enable innovative drugs that otherwise would have been labeled unsafe to come to market.

A proposal for integrated efficacy-to-effectiveness (E2E) clinical trials.

We propose an "efficacy-to-effectiveness" (E2E) clinical trial design, in which an effectiveness trial would commence seamlessly upon completion of the efficacy trial. Efficacy trials use inclusion/exclusion criteria to produce relatively homogeneous samples of participants with the target condition, conducted in settings that foster adherence to rigorous clinical protocols. Effectiveness trials use inclusion/exclusion criteria that generate heterogeneous samples that are more similar to the general patient spectrum, conducted in more varied settings, with protocols that approximate typical clinical care. In E2E trials, results from the efficacy trial component would be used to design the effectiveness trial component, to confirm and/or discern associations between clinical characteristics and treatment effects in typical care, and potentially to test new hypotheses. An E2E approach may improve the evidentiary basis for selecting treatments, expand understanding of the effectiveness of treatments in subgroups with particular clinical features, and foster incorporation of effectiveness information into regulatory processes.

Defining the regulatory role of ambulatory blood pressure monitoring. ABPM Steering Committee.

Our division is in a privileged position to study the usefulness of ABPM in the clinical development of antihypertensive agents. Most trials involving ABPM are commercially sponsored, and in due course, much of that data is presented to the Division. As such studies become more prominent in drug development, the Agency must make decisions on how such data are analyzed and interpreted. These should be seen as this project's short-term goals. In the longer term, the accumulation of ABPM data from clinical trials will be a valuable resource that researchers can use to address questions about what aspects of hypertension are most associated with prognosis and which treatments affect those aspects.

Potassium currents of rat basilar artery smooth muscle cells.

Primary isolates of smooth muscle cells from the basilar artery of the rat were studied using whole-cell and single-channel patch-clamp techniques. Two distinct potassium currents were characterized. With low intracellular calcium, depolarization above 0 mV elicited an outward current of a few hundred pA (at +120 mV) with sigmoidal onset and little inactivation during 1.25 s steps. This current was reduced by bath application of 1 mM procaine or 1 mM strychnine, but not by 500 nM charybdotoxin. These are characteristics of the delayed rectifier potassium current in other preparations. With higher intracellular calcium, depolarization above 0 mV elicited a non-inactivating potassium current of several nA (at +120 mV). This current persisted in the presence of 1 mM procaine or strychnine but was reduced by bath application of 100 nM charybdotoxin. In whole-cell recordings in which intracellular calcium was unbuffered with EGTA, spontaneous transient outward currents were manifest and displayed voltage dependence and tail currents similar to the calcium-dependent current. The spontaneous transient current and the calcium-dependent current had similar sensitivity to charybdotoxin. Cell-free membrane patches contained one or more channels of 220 pS (in solutions symmetrical with respect to potassium) with similar voltage and calcium dependence. These are characteristics of the large conductance calcium-activated potassium current in other preparations.

Vasodilatation of canine cerebral arteries by nicorandil, pinacidil and lemakalim.

1. Nicorandil, pinacidil and lemakalim relaxed precontracted rings of canine cerebral artery. 2. The order of potency was lemakalim greater than nicorandil approximately equal to pinacidil, but all these agents were less effective than nimodipine. 3. The effects of nicorandil were inhibited by methylene blue but not by glibenclamide, while the effects of pinacidil and lemakalim were inhibited by glibenclamide but not by methylene blue. 4. Thus nicorandil probably causes relaxation mostly by effects on guanylate cyclase while lemakalim and pinacidil produce the same effect by action at ATP-dependent potassium channels.

Effects of K+ channel agonists cromakalim and pinacidil on rat basilar artery smooth muscle cells are mediated by Ca(++)-activated K+ channels.

Whole-cell and cell-free inside-out patch-clamp recording techniques were used to examine the actions of potassium channel openers pinacidil and cromakalim in enzymatically isolated smooth muscle cells of rat basilar artery. Delayed rectifier and calcium-dependent potassium currents were identified from the whole-cell recordings. Only the calcium-dependent potassium current was increased by cromakalim and pinacidil. Recordings from inside-out membrane patches revealed a large conductance voltage- and calcium-dependent potassium channel, which was blocked by charybdotoxin but unaffected by ATP less than 10 mM. Cromakalim and pinacidil increased the open probability of this channel. On the basis of these results, we suggest that such drugs, acting on cerebral arterial smooth muscle cell potassium channels, may be of some benefit in the treatment of cerebral vasospasm following subarachnoid hemorrhage.

Glibenclamide relaxes vascular smooth muscle constriction produced by prostaglandin F2 alpha.

The present study has demonstrated: (1) glibenclamide can reduce resting tension in canine cerebral arteries but has no effect on resting tension in the rat aorta; (2) glibenclamide can relax prostaglandin F2 alpha-induced contractions in the rat aorta, and in canine femoral, mesenteric, renal, coronary, basilar and middle cerebral arteries; (3) the relaxation produced by glibenclamide in rat aorta is comparable to that of glyceryl trinitrate and stronger than that of papaverine; (4) canine femoral arteries are less sensitive to glibenclamide than the other arteries; (5) in cerebral arteries glibenclamide was as effective as papaverine, but less effective than glyceryl trinitrate; (6) the actions of glibenclamide on cerebral arteries are not mediated by cGMP as they were not blocked by methylene blue, an inhibitor of guanylate cyclase; (7) the effects of glibenclamide are not endothelium-dependent. The mechanism by which glibenclamide produces relaxation is not clear; while the drug is known to block ATP-dependent potassium channels, in vascular smooth muscle this would cause contraction, not dilation. The action of glibenclamide may be at the level of the receptor or the signal transduction process.

Free radicals mediate actions of oxyhemoglobin on cerebrovascular smooth muscle cells.

Single smooth muscle cells were isolated from the basilar artery of the rat by enzymatic dispersion. The membrane properties of the cells were assessed using the patch-electrode voltage-clamp technique, and cell viability was monitored using fluorescein diacetate uptake. Exposure of the cells to oxyhemoglobin (5 microM) resulted in 1) contraction, 2) the appearance of membrane blebs, 3) an increase in the outward potassium currents, 4) a decrease in the membrane resistance, and 5) cell death. In contrast, no effect of oxyhemoglobin on cultured murine neuroblastoma cells was observed. Methemoglobin (100 microM) had no effects on the smooth muscle cells. Catalase (300 units/ml) or dimethyl sulfoxide (0.5%) protected against the effects of oxyhemoglobin; superoxide dismutase (100-1,000 units/ml) provided only partial protection. Exposure of the cells to superoxide anions generated by xanthine (1 mM) plus xanthine oxidase (10 units/l) or to hydrogen peroxide (500 microM) caused an increase in the outward potassium currents without affecting membrane resistance. Generation of hydroxyl radicals by metal ions plus hydrogen peroxide caused the same effects as oxyhemoglobin, that is, an increase in the potassium currents, followed by a decrease in the membrane resistance and cell death. In conclusion, it appears that oxyhemoglobin exerts its effects on vascular smooth muscle cells by the generation of free radicals, chiefly hydroxyl radicals.

Excitability changes in the crustacean motor axons following activity.

It has been shown experimentally that the crustacean motor axon is supernormally excitable following a train of action potentials (Zucker 1974). Such a phenomenon can lead to recruitment of terminals which are unexcited at low rates of stimulation. Although currents underlying the crustacean motor axon have been characterized (Connor et al. 1977), it is not known whether this membrane model accounts for a supernormal period, what might cause superexcitability in this model, or how excitability might change during repetitive stimulation. In present study, it is demonstrated that the crustacean motor axon model does predict a supernormal period, that the supernormal period results from slow recovery from inactivation of the transient potassium, or A, current, and that supernormal excitability is enhanced by repetitive stimulation.

Theoretical response of a bifurcating axon with a locally altered axial resistivity.

A short region of high axial resistivity at one daughter branch of an axonal bifurcation can produce frequency dependent differential conduction of action potentials. The increase in resistivity need be only a few times that in the rest of the axon and length of the affected region need be only a fraction of a resting length constant (based on the local value of axial resistivity). The typical pattern observed will be that the unaffected daughter branch will conduct action potentials from the parent axon normally at all frequencies of stimulation, but the branch with the high resistance region will only follow action potentials within a restricted frequency range. In that band-pass region, the branch may conduct nearly all or only a small percentage of the action potentials from the parent axon.

Solution of the Hodgkin-Huxley and cable equations on an array processor.

This paper presents a method for the solution of equations for membrane ionic currents and the cable equation which describes the interaction of various segments in a compartmentalized model of the neuron. The method embodies some innovations which would speed calculation of propagating action potentials in any computing environment, but were specifically designed to permit the use of a relatively inexpensive array processor. The array processor produces an improvement in speed of calculation which permits exploration of a wider range of neuronal phenomena than has been previously feasible.

Etiology of the supernormal period.

In many excitable cells, there is a time after the action potential when the threshold for eliciting a second action potential is lower than it is in the steady state. The Hodgkin and Huxley (1952) equations predict such a supernormal period. Using their model, it is shown that the supernormal period results from the slow kinetics of the potassium current and does not depend on sodium current activation or inactivation or on the after-depolarization.

Theoretical response to trains of action potentials of a bifurcating axon with one short daughter branch.

It has been shown both experimentally (Stockbridge, N., and L. L. Stockbridge. 1988. J. Neurophysiol. 59:1277-1285) and theoretically (Stockbridge, N. 1988. J. Neurophysiol. 59:1286-1295) that the second of two closely spaced action potentials may be differentially conducted into a short daughter branch. Using numerical methods, the response to trains was examined in axons with a single bifurcation and uniform membrane properties. Short daughter branches conduct at higher rates of stimulation than do long branches. Under some conditions the longer daughter branch is always silent. Under other conditions, one or both branches will begin to conduct action potentials only when the stimulus frequency is high enough.