Evaluation of fluoroquinolone reduced dosage regimens in elderly patients by using pharmacokinetic modelling and Monte Carlo simulations

Transcription

1 J Antimicrob Chemother 2012; 67: doi: /jac/dks195 Advance Access publication 30 May 2012 Evaluation of fluoroquinolone reduced dosage regimens in elderly patients by using pharmacokinetic modelling and Monte Carlo simulations Bertrand Leroy 1, Mathieu Uhart 1, Pascal Maire 1,2 and Laurent Bourguignon 1,2 * 1 Hospices Civils de Lyon, Groupement Hospitalier de Gériatrie, Hôpital Antoine Charial, Service Pharmaceutique, 40 avenue de la Table de Pierre, Francheville, France; 2 Laboratoire de Biométrie et Biologie Evolutive, CNRS UMR 5558, Université de Lyon, Villeurbanne, France *Corresponding author. Tel: ; Fax: ; laurent.bourguignon@chu-lyon.fr Received 28 March 2012; returned 11 April 2012; revised 20 April 2012; accepted 23 April 2012 Objectives: Fluoroquinolones are widely used in geriatric patients, but elderly patients are known to be at increased risk of decline in renal function. As fluoroquinolones usually exhibit a dominant renal elimination pathway, reduced dosage regimens are often used in geriatric patients. Our objective was to assess the capability to reach a pharmacokinetic pharmacodynamic target of efficacy with such reduced dosage regimens of ofloxacin, levofloxacin and ciprofloxacin in elderly patients. Methods: Using Monte Carlo simulations, 1000 simulated elderly patients were created, based on published pharmacokinetic and pharmacodynamic data, and measured demographic data. Three usually proposed drug regimens taking renal function into account were evaluated using compartmental models. The probability of reaching an fauc/mic.100 was calculated for each regimen. Results: For MICs,1 mg/l, all simulated patients reach the efficacy target. However, with higher values of MIC, the proposed regimens were inefficient for patients with moderate or severe renal impairment: 3.4% and 30.2% of patients with moderate renal impairment reached the efficacy target for ciprofloxacin and ofloxacin, respectively, for an MIC of 2 mg/l. For ciprofloxacin, more than 80% of patients with severe renal impairment were unable to reach the target fauc/mic with an MIC as low as 1 mg/l, whereas for levofloxacin, all simulated patients reached the efficacy target until an MIC of 4 mg/l. Conclusions: This suggests that the proposed dosage reduction does not allow the same exposure to be achieved in elderly patients with renal impairment, eventually leading to treatment failure or development of resistant strains. Keywords: pharmacodynamics, geriatric, minimum inhibitory concentrations, renal impairment Introduction Fluoroquinolones are widely used in geriatric patients, both in community and healthcare facilities. Whereas norfloxacin, ofloxacin and ciprofloxacin are mainly used for their predominant Gram-negative activity, the latest agents, such as levofloxacin and moxifloxacin, have good Gram-negative activity and, in addition, further improved activity against Gram-positive and atypical and anaerobic bacteria. 1,2 Fluoroquinolones can also be used to treat infections such as urinary tract infections, community-acquired pneumonia, acute exacerbation of chronic bronchitis and sinusitis. 3 The emergence of resistant strains is a growing problem worldwide, 4 6 especially with multidrug-resistant Pseudomonas aeruginosa as a cause of nosocomial infections. 7 A careful use of currently available agents and the choice of an optimal dosage regimen are critical to maximize microbiological and clinical outcomes, and avoid further development of resistance. 8 Elderly patients are known to be at increased risk of drug adverse events, and drug interactions, mainly because of polymedication and age-associated physiological changes. Between age 30 years and age 80 years, a decline in the water:lipid ratio, cardiac output, renal function and hepatic function are observed. Drug absorption can also be impacted by a decline in acidity and alterations in intestinal motility and transit time that occur progressively with age. 9 As fluoroquinolones can exhibit both renal excretion and hepatic metabolism, with usually a renal-dominant elimination pathway, elderly patients were shown to have changes in # The Author Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please journals.permissions@oup.com 2207

2 Leroy et al. fluoroquinolone pharmacokinetic parameters, with a reduced rate of renal elimination, partially explained by the common decline in creatinine clearance. Therefore, reduced dosage regimens are often used to account for the decline in renal elimination. Our objective was to assess the capability to reach a pharmacokinetic pharmacodynamic target of efficacy with reduced dosage regimens of ofloxacin, levofloxacin and ciprofloxacin in elderly patients, using Monte Carlo simulations. Methods Pharmacokinetic parameters reported in the literature for levofloxacin, ciprofloxacin and ofloxacin were collected. A selection of studies undertaken in humans, reporting mean (or median) and standard deviation (or coefficient of variation) for each parameter, was done. Because of the methodology used, compartmental pharmacokinetic studies were favoured. When such information was available, the parameters obtained specifically in elderly patients were selected. Finally, the statistical relationships between the pharmacokinetic parameters and physiological variables (such as age, weight and creatinine clearance) identified in these studies were also incorporated in our models. The pharmacokinetic parameters and statistical relationships finally retained are presented in Table 1. From this information, three compartmental pharmacokinetic models were built, including an absorption compartment and a central compartment for ofloxacin, with a supplemental peripheral compartment for levofloxacin and ciprofloxacin. The systems of ordinary differential equations for the three models were determined. Anthropometric data were collected from a cohort of 300 elderly patients of a geriatric university hospital (S. Goutelle and R. Faure, University Hospitals of Lyon, France, personal communication). A mean age of years, a mean weight of kg and a mean creatinine clearance of ml/min (estimated with the Cockcroft Gault formula) were recorded and used in the simulation step of the study. From the mean values and standard deviations of pharmacokinetic parameters and anthropometric variables, 1000 simulated patients were created by Monte Carlo simulations with an assumption of a normal distribution. The range of possible values was restricted to the 95% CI to discard marginal variable values in the random process. Except when statistical relationships between parameters and/or variables were identified in the literature, independence between parameters was assumed. The simulations were performed using the normrnd instruction of MATLAB software (MathWorks, Natick, MA, USA). We assumed no intra-individual variability of pharmacokinetic parameters and variables during the time of the simulation. As a dosage regimen reduction is recommended in patients with altered renal function, a situation that is common in elderly patients, three different regimens were included in the simulation of 10 day therapies (Table 2), based on actual recommendations Among different recommendations available in Europe, EUCAST guidelines were used to set the standard dosage regimens. Since no specific regimen is proposed for patients with renal impairment in these guidelines, the French Summaries of Product Characteristics were used to set the reduced dosage regimens. The time course of fluoroquinolone serum concentrations was described using the corresponding system of differential equations. For ofloxacin, dx(1) = Ka X(1) (1) dx(2) = F.Ka X(1) Kel X(2) (2) Table 1. Pharmacokinetic parameters and statistical relationships incorporated in the models Ka (h 21 ) Volume of distribution (L) Kcp (h 21 ) Kpc (h 21 ) Total drug clearance (L/h) Reference Ciprofloxacin % a % a [( CL b CR ) weight] 60/ Levofloxacin V1/F¼ (weight256.0) /V (CL b CR ) 60/ ,12 Ofloxacin (+0.78) weight NA NA 1.69 CL CR b ,14 Ka, absorption rate constant; Kcp and Kpc, inter-compartmental transfer rate constants; CL CR, creatinine clearance; V1, volume of central compartment; F, bioavailability; NA, not applicable. a Coefficient of variation arbitrarily set to 30%. b Creatinine clearance estimated with the Cockcroft Gault formula. Table 2. Dosage regimens included in the simulation Regimen 1 Regimen 2 Regimen 3 Renal function a CL CR.60 ml/min CL CR ¼30 60 ml/min CL CR,30 ml/min ciprofloxacin 500 mg BID 250 mg BID 250 mg QD Renal function a CL CR.50 ml/min CL CR ¼20 50 ml/min CL CR,20 ml/min levofloxacin 500 mg QD 250 mg QD 125 mg QD ofloxacin 200 mg BID 200 mg QD 200 mg q48h CL CR, creatinine clearance; QD, once daily; BID, twice daily; q48h, once in 2 days. a Renal function is estimated with the Cockcroft Gault formula. 2208

3 Efficacy of fluoroquinolone reduced dosage regimens in the elderly JAC where X(1) is the amount of ofloxacin in the absorption compartment (mg), X(2) is the amount of ofloxacin in the central compartment (mg), F is the bioavailability, Ka is the absorption rate constant (per hour) and Kel is the elimination rate constant (per hour). For ciprofloxacin and levofloxacin, dx(1) = Ka X(1) (3) dx(2) = F.Ka X(1) X(2) (Kel + Kcp)+Kpc X(3) (4) dx(3) = X(2) Kcp Kpc X(3) (5) where X(1), X(2) and X(3) are the amount of ciprofloxacin or levofloxacin in the absorption, central and peripheral compartment (mg), respectively, F is the bioavailability, Ka is the absorption rate constant (per hour), Kel is the elimination rate constant (per hour) and Kcp and Kpc are intercompartmental transfer rate constants (per hour). The serum concentration was calculated using the following formula: C = X(2) V where C is the serum concentration, X(2) is the amount of fluoroquinolone in the central compartment and V is the volume of distribution of the central compartment for ciprofloxacin and levofloxacin and the apparent volume of distribution for ofloxacin. We used MATLAB software to solve the system of ordinary differential equations with the function ode45. The pharmacokinetics pharmacodynamics of fluoroquinolone have been well defined and several parameters have been proposed to predict clinical efficacy (C max /MIC, AUC/MIC, AUC/MPC and t.mic; where MPC stands for mutant prevention concentration) Because the AUC/MIC ratio has been reported to have the strongest correlation with clinical outcomes and development of resistance to fluoroquinolones in a non-clinical model of infection, it was chosen as a substitution criterion to evaluate treatment efficacy in our model. The attainment of a total fauc/mic ratio.100, which is the higher target recognized by EUCAST, was set as the pharmacodynamic target. The unbound fraction (f) was set to 0.7 for ciprofloxacin and levofloxacin and to 0.75 for ofloxacin. Typical MICs 29 were considered during the simulations, ranging from 0.5 to 4 mg/l. A breakpoint concentration 90 (BC 90 ), defined as the MIC value above which,90% of patients reach the efficacy target, was calculated. Results Mean AUCs estimated for patients treated with ciprofloxacin were mg.h/l, mg.h/l and mg. h/l, respectively, with regimen 1, 2 and 3. Mean AUCs estimated for patients treated with levofloxacin were mg.h/l, mg.h/l and mg.h/l, respectively, with regimen 1, 2 and 3. Mean AUCs estimated for patients treated with ofloxacin were mg.h/l, mg.h/l and mg.h/l, respectively, with regimen 1, 2 and 3. Whatever the antimicrobial agent was, there were significant differences in the mean AUC according to the regimen considered (t-test, P, 0.001). The percentage of patients reaching the efficacy target as a function of MIC is reported in Table 3 and the evolution of this percentage is represented in Figure 1. When considering an MIC of 2 mg/l only 3.4% and 30.2% of patients with moderate (6) Table 3. Percentage of patients reaching the efficacy target for four typical MICs renal impairment reached the efficacy target for ciprofloxacin and ofloxacin, respectively. This figure dropped to 0% and 0.8% in cases of severe renal impairment for ciprofloxacin and ofloxacin, respectively. The BC 90 s of each regimen and each antimicrobial agent are reported in Table 4. It should be noted that for ciprofloxacin,.80% of patients with severe renal impairment were unable to reach the target fauc/mic with an MIC as low as 1 mg/l, whereas for levofloxacin, all simulated patients reached the efficacy target until an MIC of 4 mg/l. Discussion Ciprofloxacin total population CL CR.60 ml/min CL CR ¼30 60 ml/min CL CR,30 ml/min Levofloxacin total population CL CR.50 ml/min CL CR ¼20 50 ml/min CL CR,20 ml/min Ofloxacin total population CL CR.50 ml/min CL CR ¼20 50 ml/min CL CR,20 ml/min Fluoroquinolones represent an important class of antibiotics, but the emergence of bacterial resistance requires optimizing fluoroquinolone use. The pharmacokinetics of these drugs show a significant renal elimination, which can be reduced in the case of renal impairment, frequently encountered in the geriatric population. Therefore, dosage reduction is recommended in patients with renal failure, whether with a reduction in dose unit or an increase in the dosing interval. It seems necessary to verify if such reduced doses were suitable for clinical practice. From published pharmacokinetic data, and anthropometric data measured in a cohort of hospitalized elderly patients, a simulation study was conducted to test the usefulness of dose reductions based on renal function in elderly patients. Three fluoroquinolones were modelled, and the dosages commonly used have been tested. Efficacy was evaluated by observing the value of the index fauc/mic, considering MICs ranging from 0.5 to 4 mg/l. Our study showed that dose reductions according to renal function are generally well adapted to geriatric patients when considering bacteria with low MICs. Indeed, for MICs,1 mg/l, 2209

4 Leroy et al. (a) (b) (c) Regimen 1 Regimen 2 Regimen 3 Figure 1. Percentage of patients treated with ciprofloxacin (a), levofloxacin (b) or ofloxacin (c) reaching the efficacy target as a function of MIC. all simulated patients reach the efficacy target (fauc/mic.100) for all fluoroquinolones tested. However, with higher values of MIC, the proposed regimens were inefficient for patients with moderate or severe renal impairment, mainly to ciprofloxacin and ofloxacin (0% and 0.8% in cases of severe renal impairment when considering an MIC of 2 mg/l). More generally, the AUC values observed in patients with renal impairment after administration of a reduced regimen are significantly lower than those observed in patients without renal 2210

5 Efficacy of fluoroquinolone reduced dosage regimens in the elderly JAC Table 4. Calculated BC 90 s Population BC 90 (mg/l) ciprofloxacin levofloxacin ofloxacin Regimen Regimen Regimen Total population impairment. This suggests that the proposed dose reduction does not allow the same exposure to be achieved in these patients as in patients without renal impairment. This relative under-exposure could be responsible for treatment failure or facilitate the development of resistant strains. 30 The methodology used here, although it has already been applied on several occasions to evaluate the usefulness of therapeutic strategies, 7,31 34 has some limitations that should be discussed. First, the blood concentrations of fluoroquinolones were estimated using pharmacokinetic mono-compartmental (ofloxacin) or bi-compartmental models (levofloxacin and ciprofloxacin). More complex and perhaps more efficient models could have been used. However, the structural models used in this study were reported to present satisfactory predictive capabilities for the blood levels of these drugs in their original publications. On the other hand, it was not possible to find publications specifically mentioning pharmacokinetic parameters obtained in elderly patients for all drugs studied. A slight difference in predictions could be considered for this reason, but taking into account the weight and renal function of elderly patients in these models should have reduced this potential source of bias. Moreover, in the absence of a single source for all relevant parameter values required for modelling (mean and standard deviation), it was sometimes necessary to extract information from several different publications. Finally, some assumptions were necessary to perform simulations: Gaussian distribution of pharmacokinetic parameters and physiological variables, assumption of independence of parameters and assumption of absence of intra-individual variability during the 10 days of simulated therapy. This study brings up the possibility that dose adjustments based on renal function are not always suitable for elderly patients. This could lead to a risk of inefficacy. In particular, doses of ciprofloxacin appear to be too strongly reduced in patients with renal failure to ensure the effectiveness of treatment for infection with bacteria presenting an MIC.1 mg/l. However, as our model does not take into account the antibiotic toxicity, which has been reported as dose-dependent, toxic risk when using higher doses has not been studied. The simulations were undertaken assuming the choice of the most conservative drug regimen proposed either in EUCAST recommendations or in the French Summary of Product Characteristics for each evaluated drug. While more aggressive strategies could be used if necessary, these conservative schemes were selected for simulation because their use was considered very likely in the case of elderly patients for whom a greater risk of adverse effects is expected. 38 More favourable results would have been found with more aggressive regimens. Similarly, the choice of a target of at least 100 for fauc/mic could be discussed: regardless of the susceptibility of the bacteria causing the infection, a higher or lower target may be more appropriate depending on the location and severity of infection. 39 Finally, even if the criterion used (fauc/mic) is considered to be correlated with clinical response, a clinical study would be needed to validate this hypothesis and to suggest a strategy more suited to this population. Conclusions Fluoroquinolones are widely used in geriatric patients, but because of a renal-dominant elimination pathway, elderly patients have been shown to have changes in fluoroquinolone pharmacokinetic parameters, with a reduced rate of renal elimination. Therefore reduced dosage regimens are often used to account for the decline in renal elimination. Using Monte Carlo simulations, our study showed that dosage reductions according to renal function are generally well adapted to geriatric patients when considering bacteria with low MICs, but with higher MICs, the proposed regimens for ciprofloxacin and ofloxacin could be inefficient for patients with moderate or severe renal impairment. This suggests that the proposed dosage reduction does not allow the same exposure to be achieved in these patients as in patients without renal impairment. Whereas a clinical study is needed to validate this hypothesis, this relative under-exposure could be responsible for treatment failure or facilitate the development of resistant strains. When resistant organisms are found, tailored therapy should be considered. Funding This study was carried out as part of our routine work. Transparency declarations None to declare. References 1 Blondeau JM. Expanded activity and utility of the new fluoroquinolones: a review. Clin Ther 1999; 21: 3 40; discussion Pickerill KE, Paladino JA, Schentag JJ. Comparison of the fluoroquinolones based on pharmacokinetic and pharmacodynamic parameters. Pharmacotherapy 2000; 20: Bartlett JG, Dowell SF, Mandell LA et al. Practice guidelines for the management of community-acquired pneumonia in adults. Infectious Diseases Society of America. Clin Infect Dis 2000; 31: Rossolini GM, Mantengoli E. Antimicrobial resistance in Europe and its potential impact on empirical therapy. Clin Microbiol Infect 2008; 14 Suppl 6: Rice LB. The clinical consequences of antimicrobial resistance. Curr Opin Microbiol 2009; 12: Hawkey PM, Jones AM. The changing epidemiology of resistance. J Antimicrob Chemother 2009; 64 Suppl 1: i Ludwig E, Konkoly-Thege M, Kuti JL et al. Optimising antibiotic dosing regimens based on pharmacodynamic target attainment against 2211

6 Leroy et al. Pseudomonas aeruginosa collected in Hungarian hospitals. Int J Antimicrob Agents 2006; 28: Gillespie EL, Kuti JL, Nicolau DP. Pharmacodynamics of antimicrobials: treatment optimisation. Expert Opin Drug Metab Toxicol 2005; 1: Schentag JJ, Goss TF. Quinolone pharmacokinetics in the elderly. Am J Med 1992; 92: 33S 7S. 10 Forrest A, Ballow CH, Nix DE et al. Development of a population pharmacokinetic model and optimal sampling strategies for intravenous ciprofloxacin. Antimicrob Agents Chemother 1993; 37: Deguchi T, Nakane K, Yasuda M et al. Microbiological outcome of complicated urinary tract infections treated with levofloxacin: a pharmacokinetic/pharmacodynamic analysis. Int J Antimicrob Agents 2010; 35: Zhang J, Xu J-F, Liu Y-B et al. Population pharmacokinetics of oral levofloxacin 500 mg once-daily dosage in community-acquired lower respiratory tract infections: results of a prospective multicenter study in China. J Infect Chemother 2009; 15: Rademaker CM, Jones RW, Notarianni LJ et al. Pharmacokinetics of a single dose of ofloxacin in healthy elderly subjects using noncompartmental and compartmental models. Pharm Weekbl Sci 1989; 11: Fillastre JP, Leroy A, Humbert G. Ofloxacin pharmacokinetics in renal failure. Antimicrob Agents Chemother 1987; 31: Gupta K, Hooton TM, Naber KG et al. International clinical practice guidelines for the treatment of acute uncomplicated cystitis and pyelonephritis in women: a 2010 update by the Infectious Diseases Society of America and the European Society for Microbiology and Infectious Diseases. Clin Infect Dis 2011; 52: e EUCAST. Ciprofloxacin: Rationale for the EUCAST Clinical Breakpoints, Version files/rationale_documents/ciprofloxacin_rationale_1.9.pdf (20 March 2012, date last accessed). 17 EUCAST. Levofloxacin: Rationale for the EUCAST Clinical Breakpoints, Version EUCAST_files/Rationale_documents/Levofloxacin_rationale_1.5.pdf (20 March 2012, date last accessed). 18 EUCAST. Ofloxacin: Rationale for the EUCAST Clinical Breakpoints, Version files/rationale_documents/ofloxacin_rationale_1.4.pdf (20 March 2012, date last accessed). 19 AFSSAPS. Antibiotic therapy in general and current practices in upper respiratory tract infections in adults and children. Recommendations. Med Mal Infect 2005; 35: Blondeau JM, Hansen G, Metzler K et al. The role of PK/PD parameters to avoid selection and increase of resistance: mutant prevention concentration. J Chemother 2004; 16 Suppl 3: Forrest A, Nix DE, Ballow CH et al. Pharmacodynamics of intravenous ciprofloxacin in seriously ill patients. Antimicrob Agents Chemother 1993; 37: Madaras-Kelly KJ, Ostergaard BE, Hovde LB et al. Twenty-four-hour area under the concentration-time curve/mic ratio as a generic predictor of fluoroquinolone antimicrobial effect by using three strains of Pseudomonas aeruginosa and an in vitro pharmacodynamic model. Antimicrob Agents Chemother 1996; 40: Lister PD, Sanders CC. Pharmacodynamics of trovafloxacin, ofloxacin, and ciprofloxacin against Streptococcus pneumoniae in an in vitro pharmacokinetic model. Antimicrob Agents Chemother 1999; 43: Drusano GL, Preston SL, Fowler C et al. Relationship between fluoroquinolone area under the curve:minimum inhibitory concentration ratio and the probability of eradication of the infecting pathogen, in patients with nosocomial pneumonia. J Infect Dis 2004; 189: Fantin B, Leggett J, Ebert S et al. Correlation between in vitro and in vivo activity of antimicrobial agents against gram-negative bacilli in a murine infection model. Antimicrob Agents Chemother 1991; 35: Drusano GL, Johnson DE, Rosen M et al. Pharmacodynamics of a fluoroquinolone antimicrobial agent in a neutropenic rat model of Pseudomonas sepsis. Antimicrob Agents Chemother 1993; 37: Mouton JW, van Ogtrop ML, Andes D et al. Use of pharmacodynamic indices to predict efficacy of combination therapy in vivo. Antimicrob Agents Chemother 1999; 43: Ambrose PG, Bhavnani SM, Owens RC Jr. Clinical pharmacodynamics of quinolones. Infect Dis Clin North Am 2003; 17: Fung-Tomc JC, Minassian B, Kolek B et al. Antibacterial spectrum of a novel des-fluoro(6) quinolone, BMS Antimicrob Agents Chemother 2000; 44: Shams WE, Evans ME. Guide to selection of fluoroquinolones in patients with lower respiratory tract infections. Drugs 2005; 65: Zelenitsky S, Ariano R, Harding G et al. Evaluating ciprofloxacin dosing for Pseudomonas aeruginosa infection by using clinical outcome-based Monte Carlo simulations. Antimicrob Agents Chemother 2005; 49: Burgess DS, Hall RG II. Simulated comparison of the pharmacodynamics of ciprofloxacin and levofloxacin against Pseudomonas aeruginosa using pharmacokinetic data from healthy volunteers and 2002 minimum inhibitory concentration data. Clin Ther 2007; 29: Koomanachai P, Bulik CC, Kuti JL et al. Pharmacodynamic modeling of intravenous antibiotics against gram-negative bacteria collected in the United States. Clin Ther 2010; 32: Montgomery MJ, Beringer PM, Aminimanizani A et al. Population pharmacokinetics and use of Monte Carlo simulation to evaluate currently recommended dosing regimens of ciprofloxacin in adult patients with cystic fibrosis. Antimicrob Agents Chemother 2001; 45: Kashida Y, Kato M. Characterization of fluoroquinolone-induced Achilles tendon toxicity in rats: comparison of toxicities of 10 fluoroquinolones and effects of anti-inflammatory compounds. Antimicrob Agents Chemother 1997; 41: Stahlmann R. Clinical toxicological aspects of fluoroquinolones. Toxicol Lett 2002; 127: Bertino J Jr, Fish D. The safety profile of the fluoroquinolones. Clin Ther 2000; 22: ; discussion Nicolle LE. Quinolones in the aged. Drugs 1999; 58 Suppl 2: Schentag JJ, Gilliland KK, Paladino JA. What have we learned from pharmacokinetic and pharmacodynamic theories? Clin Infect Dis 2001; 32 Suppl 1: S