Epilepsy & Behavior 55 (2016) 21–23
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Brief Communication
Use of computer simulations to test the concept of dose forgiveness in the era of extended-release (XR) drugs John M. Pellock a,1, Scott T. Brittain b,⁎ a b
Virginia Commonwealth University, Richmond, VA, USA Supernus Pharmaceuticals, Inc., Rockville, MD, USA
a r t i c l e
i n f o
Article history: Received 25 August 2015 Revised 24 November 2015 Accepted 25 November 2015 Available online xxxx Keywords: Forgiveness Extended-release Simulation Adherence
a b s t r a c t “Forgiveness” – the difference between a drug's postdose duration of action and its prescribed dosing interval – estimates the margin of therapeutic effect following a missed dose. Because this margin presumably decreases as dosing becomes less frequent, QD dosing of an antiepileptic drug (AED) is expected to be less forgiving than more frequent (e.g., BID) dosing of that same AED. However, if the AED is reformulated as an extended-release (XR) preparation, drug input may be prolonged relative to its immediate-release (IR) counterpart. It therefore stands to reason that forgiveness could be increased by an XR AED that extends the period during which therapeutic plasma concentrations are maintained if a dose is missed. Computer simulation was used to estimate forgiveness for an IR formulation of a hypothetical AED and its XR counterparts reformulated for less frequent dosing. Simulations determined forgiveness when the hypothetical IR AED was dosed TID, BID, and QD and when suitably designed XR formulations were dosed BID and QD. Simulations showed that forgiveness for an XR formulation can equal or exceed that for an IR formulation dosed more frequently. © 2015 Published by Elsevier Inc.
1. Introduction Patients and physicians agree that efficacy and safety/tolerability are the most important attributes when selecting an antiepileptic drug (AED). However, for patients, the next most important characteristic is dosing convenience [1], an important determinant of patient adherence [2]. Because extended-release (XR) formulations offer the potential for more constant plasma concentration–time profiles with less frequent dosing, they may positively impact tolerability, adherence, and therefore, overall effectiveness. It is therefore not surprising that most AEDs that were initially administered in multiple daily doses as immediaterelease (IR) formulations are now available as XR formulations for patients with epilepsy, including phenytoin, carbamazepine, valproate, lamotrigine, topiramate, oxcarbazepine, and levetiracetam. Although they use different technologies to slow/control drug release and thereby prolong drug absorption, XR formulations are designed to minimize plasma concentration fluctuations over the course of a day despite less frequent dosing than their IR counterparts [3]. However, even with less frequent dose administration, adherence can be imperfect as patients take doses late or miss doses altogether [4]. Recognition that
⁎ Corresponding author at: Supernus Pharmaceuticals, Inc., 1550 East Gude Drive, Rockville, MD 20850, USA. Tel.: +1 301 838 2581. E-mail addresses:
[email protected] (J.M. Pellock),
[email protected] (S.T. Brittain). 1 Division of Child Neurology, Virginia Commonwealth University School of Medicine, 1001 E. Marshall St., Richmond, VA 23298, USA.
http://dx.doi.org/10.1016/j.yebeh.2015.11.029 1525-5050/© 2015 Published by Elsevier Inc.
less frequent dosing with XR formulations can be associated with better outcomes is often tempered by expectations that QD dosing, for example, is not as “forgiving” as BID dosing, implying that a late or missed dose of an XR AED is more likely to compromise seizure control than a dose of its more frequently administered IR counterpart [5–7]. “Drug forgiveness” (F) – a measure of the margin in therapeutic effect if a dose is missed – is conceptualized as the difference between a drug's postdose duration of action (D) and the prescribed dosing interval (I), i.e., F = D − I [8]. The perception that an XR AED would invariably be less forgiving than its more frequently administered IR counterpart assumes that duration of action is constant, as if it is simply an intrinsic characteristic of the molecule. The increased availability of XR AEDs merits a closer look at postdose duration of action and its impact on forgiveness when dosing frequency changes. To that end, we used PK modeling and simulation to determine plasma concentration–time profiles for a hypothetical AED administered as an IR formulation and as two XR counterparts appropriately reformulated for less frequent daily dosing. Based on plasma concentration–time profiles and a threshold pharmacokinetic–pharmacodynamic (PK/PD) model, postdose duration of action (D) was determined, allowing forgiveness (F) to be calculated. 2. Methods For the hypothetical AED, modeling assumed that the PK characteristics of IR and XR formulations were identical in terms of elimination half-life (8.7 h), total daily dosage (180 mg), and systemic parameters
22
J.M. Pellock, S.T. Brittain / Epilepsy & Behavior 55 (2016) 21–23
(i.e., clearance and volume of distribution). Absorption rate was the only PK characteristic assumed to be different between IR and XR formulations. Absorption rate was assumed to be limited by the rate of drug release from the formulation and was modeled with a first-order absorption rate constant (ka). For the IR formulation, ka was kept constant, regardless of the simulated dose interval (8, 12, or 24 h). For XR formulations, ka reflected the formulation's customization to release drug at a rate appropriate for the desired dosing interval (XRBID: 12 h; XRQD: 24 h). Concentration vs. time curves were simulated using a script in R (www.R-Project.org) for a one-compartment model with first-order elimination and first-order absorption. Simulations were performed for a hypothetical IR formulation (ka = 5/h) dosed 60 mg TID (I = 8 h), 90 mg BID (I = 12 h), and 180 mg QD (I = 24 h) and for two hypothetical XR formulations dosed 90 mg BID for XRBID (ka = 0.5/h; I = 12 h) and 180 mg QD for XRQD (ka = 0.1/h; I = 24 h). Concentration– time curves were used to determine postdose duration of action when dosing was abruptly stopped, where D was defined as the interval between the time of the last administered dose and the time at which plasma AED concentration fell below a hypothetical minimum effective concentration, i.e., the therapeutic threshold. Forgiveness in each simulation was calculated as F = D − I. 3. Results Fig. 1 shows simulated concentration–time curves for both IR and XR formulations. Regardless of the formulation, duration of action increased when the total daily dose of 180 mg was administered in increasingly larger individual dosages (i.e., 60, 90, or 180 mg per dose) in conjunction with less frequent dosing, i.e., longer dose intervals (I = 8, 12, or 24 h). Fig. 1 (panels A through C) shows simulated concentration–time profiles for the IR formulation dosed TID, BID, and QD. Based on these simulations, D increased from 18 h with TID dosing to 24.5 h with QD dosing. However, the small increases in postdose duration of action did not offset the impact of longer intervals between doses. The forgiveness
TID Dosing (60 mg q8hr)
interval fell from 10 h when the IR formulation was dosed TID (I = 8 h) to 0.5 h when dosed QD (I = 24 h) (Table 1). Fig. 1 (panels D and E) shows simulated concentration–time profiles for the two XR formulations customized with slower absorption rates for BID and QD administration. The simulations predicted that for a given dosing regimen, the XR formulations had longer postdose duration of action than the IR formulation (Fig. 1B vs. Fig. 1D; Fig. 1C vs. Fig. 1E). Accordingly, the XR formulations had longer forgiveness intervals (Table 1). 4. Discussion With IR formulations, peak drug concentration occurs shortly after administration, resembling the plasma concentration–time profile of a parenterally administered drug. Thus, elimination half-life often determines dosing interval for an orally administered IR drug while the absorption phase is typically disregarded. However, in the case of oral XR formulations, plasma concentration–time profiles diverge significantly from those with parenteral dosing such that half-life alone no longer defines the dosing interval. Since the rate of drug release from an XR formulation is controlled, absorption continues long after dosing, with the peak concentration occurring much later. The optimal dosing interval with an XR formulation is therefore determined by the interplay between controlled drug release and drug elimination. A welldesigned XR product maintains effective drug levels longer than its IR counterpart, increasing postdose duration of action and effectively extending the XR formulation's half-life. The simulations with an IR AED formulation reported here demonstrate the dose dependency of postdose duration of action. With the IR formulation, the only parameters that changed in each dosing simulation were the magnitude of the dose and dose interval. Compared with TID dose frequency, which was consistent with the hypothetical AED's half-life of 8.7 h, the larger doses with BID and QD administration maintained therapeutic plasma concentrations for longer periods. However, increased duration of action with larger doses administered BID and QD relative to TID administration came at the expense of greater
BID Dosing (90 mg q12hr)
QD Dosing (180 mg q24hr)
Immediate-Release (IR) Formulation, Ka=5 hr -1
A.
B. D=20
Time, days
D=24.5
Threshold
Concentration
Threshold
Concentration
Concentration
D=18
C.
Time, days
Threshold
Time, days
Extended-Release (XR) Formulation
D.
XR Designed for BID Dosing, Ka=0.5 hr -1
E.
D=37
Threshold
Time, days
Concentration
Concentration
D=22
XR Designed for BID Dosing, Ka=0.5 hr -1
Threshold
Time, days
Fig. 1. Postdose duration of action (D) vs. dose interval determined from plasma concentration–time curves for a hypothetical AED (half-life, 8.7 h) administered as an IR formulation (panels A, B, C) or XR formulations customized according to dose frequency (panels D, E). Modeling assumed same systemic parameters and same total daily dosage (180 mg) administered TID, BID, or QD for IR AED and BID or QD for XR AED formulations.
J.M. Pellock, S.T. Brittain / Epilepsy & Behavior 55 (2016) 21–23 Table 1 Forgiveness calculation (F = D − I) for hypothetical AED as IR or customized XR formulations. Formulation
Interval, h (I)
Duration, h (D)
Forgiveness, h (F)
Immediate-release (IR), ka = 5 h−1
8 12 24 12
18 20 24.5 22
10 8 0.5 10
24
37
13
Extended-release (XR) designed for BID dosing (XRBID), ka = 0.5 h−1 Extended-release (XR) designed for QD dosing (XRQD), ka = 0.1 h−1
fluctuation [(Cmax − Cmin) / Cavg], producing not only higher, potentially more toxic peak concentrations but also lower trough concentrations that led to a smaller therapeutic margin and shorter forgiveness [9]. The simulations with the XR AED formulations illustrate the influence of a formulation's drug release profile, characterized by the absorption rate constant (ka), on postdose duration of action. In simulations of BID and QD dosing, ka was the only parameter that differed for XR formulations relative to the IR formulation. When the rate of drug release was customized to the desired dosing frequency, postdose duration of action and forgiveness were longer with the XR formulations. Compared with TID dosing of the IR formulation, the forgiveness interval was the same (XRBID) or even longer (XRQD) with the XR AEDs. These simulations were designed to explore forgiveness as the dosing interval of a hypothetical IR AED is lengthened without reformulation compared with XR formulations that are specifically designed to achieve a desired dosing frequency. The XR formulations combined a larger total dose with a slower release rate. This design extended postdose duration of action and offset the increased dosing interval to produce equivalent or longer forgiveness periods. For the IR formulation, the effect of the larger dose alone was not sufficient to justify less frequent dosing. Although TID administration of the IR AED was the only simulation in which the forgiveness period was longer than the dosing interval, this outcome arises because of the interplay of the parameters (e.g., ka, half-life, threshold) selected for modeling and simulation. A limitation of these simulations is the use of time to subtherapeutic plasma concentration as a measure of postdose duration of action, particularly since mechanism(s) of action and duration of an AED's therapeutic effect are generally not known. Moreover, the breakpoint between therapeutic and subtherapeutic concentrations is highly individualized and influenced by a multitude of factors such as genetics, age, and disease severity/refractoriness. This study illustrates the potential value of computer modeling and simulation in exploring drug dosing. For several AEDs, including some with XR formulations, modeling and simulation has been a tool for determining dosing strategies when doses are taken late or missed completely [10–13]. As our simulations illustrate, appropriately designed XR AEDs can be as forgiving, if not more so, as their more frequently dosed IR counterparts.
23
5. Conclusion Extended-release formulations of AEDs offer potential advantages over their IR counterparts in terms of simplified dosing, improved tolerability, and increased adherence. The sum of these advantages may be greater overall effectiveness and better outcomes for patients with epilepsy. The simulations illustrated here should allay concerns that a missed dose of an appropriately designed XR formulation poses, a priori, a greater risk to seizure control than a missed dose of its IR counterpart. Acknowledgments Editorial support was provided by Verna Ilacqua (ID&A), funded by Supernus Pharmaceuticals, Inc. Disclosures Dr. Pellock is an investigator in studies conducted by Acorda, Eisai, GW Pharmaceuticals, Marinus Pharmaceuticals, Lundbeck, Pfizer, Questcor/Mallinckrodt, UCB, and Upsher-Smith. He is a consultant to Acorda, Eisai, GW Pharmaceuticals, Marinus Pharmaceuticals, Neuropace, Lundbeck, Pfizer, Questcor/Mallinckrodt, Sepracor, Sunovion, UCB, and Upsher-Smith. He has served on advisory boards for Acorda, Eisai, Lundbeck, Pfizer, Questcor/Mallinckrodt, Sepracor, UCB, and UpsherSmith. Dr. Brittain is an employee of Supernus Pharmaceuticals, Inc. References [1] Groenewegen A, Tofighy A, Ryvlin P, Steinhoff BJ, Dedeken P. Measures for improving treatment outcomes for patients with epilepsy — results from a large multinational patient–physician survey. Epilepsy Behav 2014;34:58–67. [2] Claxton AJ, Cramer J, Pierce C. A systematic review of the associations between dose regimens and medication compliance. Clin Ther 2001;23:1296–310. [3] Pellock JM, Smith MC, Cloyd JC, Uthman B, Wilder BJ. Extended-release formulations: simplifying strategies in the management of antiepileptic drug therapy. Epilepsy Behav 2004;5:301–7. [4] Osterberg L, Blaschke T. Adherence to medication. N Engl J Med 2005;353:487–97. [5] Bialer M. Extended-release formulations for the treatment of epilepsy. CNS Drugs 2007;21:765–74. [6] Perucca P, Carter J, Vahle V, Gilliam FG. Adverse antiepileptic drug effects: toward a clinically and neurobiologically relevant taxonomy. Neurology 2009;72:1223–9. [7] Werz MA. Pharmacotherapeutics of epilepsy: use of lamotrigine and expectations for lamotrigine extended release. Ther Clin Risk Manag 2008;4:1035–46. [8] Urquart J. The electronic medication event monitor: lessons for pharmacotherapy. Clin Pharmacokinet 1997;32:345–56. [9] Levy G. A pharmacokinetic perspective on medicament noncompliance. Clin Pharmacol Ther 1993;54:242–4. [10] Chen C, Wright J, Gidal B, Messenheimer J. Assessing impact of real-world dosing irregularities with lamotrigine extended-release and immediate-release formulations by pharmacokinetic simulation. Ther Drug Monit 2013;35:188–93. [11] Gidal BE, Majid O, Ferry J, et al. The practical impact of altered dosing on perampanel plasma concentrations: pharmacokinetic modeling from clinical studies. Epilepsy Behav 2014;35:6–12. [12] Reed RC, Dutta S. Predicted serum valproic acid concentrations in patients missing and replacing a dose of extended-release divalproex sodium. Am J Health Syst Pharm 2004;61:2284–9. [13] Brittain S. Once-daily Trokendi XR® (SPN-538) vs. twice-daily Topamax®: impact of nonadherence on topiramate concentrations. Epilepsy Curr 2014;14(Suppl. 1): 2.122.
