ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Dec. 1988, p. 1875-1878 Vol. 32, No. 12 0066-4804/88/121875-04$02.00/0 Copyright © 1988, American Society for Microbiology Pharmacokinetics and Human Tissue Penetration of Flurithromycin G. BENONI,1* L. CUZZOLIN,l R. LEONE,' U. CONSOLO,2 G. FERRONATO,2 C. BERTRAND,3 V. PUCHETTI,3 AND M. E. FRACASSO1 Institute ofPharmacology,1 Surgical Clinic-Thoracic Surgery Unit,3 and Institute ofDental Medicine,2 University of Verona, 37134 Verona, Italy Received 13 May 1988/Accepted 21 September 1988 The relationship between concentrations in serum and levels in tissue of flurithromycin, a new fluorinated macrolide, was determined in patients undergoing maxillofacial surgery and thoracotomy. All patients received 500 mg of flurithromycin orally every 8 h. Drug levels in serum, bone, soft tissue, lung, and pericardial fluid were determined microbiologically. The total amount of antibiotic per gram of tissue was calculated on the basis of the concentration in the supernatant of the homogenate. From the parallel course between free concentrations in serum and calculated contents in interstitial fluid tissue, it was concluded that the tissues examined were easily accessible by flurithromycin; penetration values measured by the ratio of areas under the curve were 8.3 for lung, 3.6 for bone, and 0.8 for soft tissue. The results of the pharmacokinetic study suggest that accumulation of the drug during repetitive multiple doses is predictable. Mean residence times were 10.2 and 8.3 h in groups 1 and 2, respectively. For bacteriostatic drugs such as macrolides, not only very high but also prolonged concentrations in tissue lead to a favorable therapeutic result.
Flurithromycin (Pierrel S.p.A.), an (8S)-8-fluoroerythro- omy were studied. Informed consent was obtained from mycin, shows the same spectrum of antimicrobial activity as each patient. The patients had normal renal and hepatic does erythromycin and the same activity in vitro (4, 9). function, and none had received any antibiotics in the month Preliminary information on the kinetics of this drug in before the surgery. healthy volunteers suggests that flurithromycin has good These patients received a total of seven doses of 500 mg of bioavailability and attains suitable levels in serum (1). We flurithromycin at 8-h intervals. Blood samples were taken investigated the pharmacokinetics of flurithromycin after after one administration of a 500-mg dose and on the morning single and multiple doses and studied its penetration into of day 3 after the last dose. Blood samples were taken at 0, bone, soft tissue, lung, and pericardial fluid in surgery 0.5, 1, 2, 4, 6, and 7 h after drug administration. patients. Since the antibacterial activity of antibiotics in vivo and 7 cannot be predicted only from activity in vitro and free Lung tissue samples were collected at 1.5, 2.6, 4, 5, concentrations in serum, we studied the relationship be- h after drug administration and homogenized in an equal tween free flurithromycin concentrations in serum and levels volume of 0.1 M phosphate buffer (pH 8). The homogenates in interstitial fluid, calculated from total drug content in were centrifugated at 1,200 x g in a refrigerated centrifuge, tissue, and consequently the difference in drug penetration and the supernatants were immediately assayed. Pericardial into different tissues. fluid samples were collected at 1.5, 3.5, 3.8, 5.7, and 7 h during surgery and assayed in toto. Drug assay and kinetic model. All samples were assayed MATERIALS AND METHODS within 2 h of collection by a microbiological method, using Bone and soft tissue. Twelve male patients (20 to 31 years antibiotic medium 1 and Micrococcus luteus ATCC 9341 as old and weighing 58 to 72 kg) undergoing maxillofacial the indicator organism. Serum, tissue, and pericardial fluid surgery were included in the study. Each patient had a samples were assayed against standards prepared in serum, normal clinical examination, normal hematological findings, tissue, and fluid samples of subjects who had received no and normal renal and liver function tests. None of the antibiotic therapy. The limit of detection of the assay was subjects took other drugs or antibiotics at least 2 weeks 0.03 ,g/ml. The assay had a coefficient of variation of 6.5% before the surgery. Informed consent was obtained from and a mean relative error of 8%. Assays of hemoglobin in each patient before flurithromycin administration. Each pa- tissues and correction of flurithromycin concentrations for tient received a total of 10 doses of 500 mg of antibiotic at 8-h blood contamination were performed according to Kroening intervals. Blood samples were taken after one administration et al. (8). of 500 mg of flurithromycin and on the morning of day 4 after The sampling period was long enough to allow drug levels the last (10th) dose, during surgery. Blood samples were in serum after one dose to approach zero or at least low taken at 0, 0.25, 0.5, 1, 1.5, 3, and 8 h after drug adminis- values. Free concentrations in serum were calculated from tration. the total levels in serum by correction for protein binding, Bone and soft tissue samples, obtained from each patient which amounted to 70% (G. Bonardi, unpublished data) for at 1, 1.5, 2, and 3 h after drug administration, were pulver- flurithromycin. Binding of flurithromycin to serum protein is ized with a Spex freezer mill and diluted 1:3 in 0.1 M independent of drug concentration. phosphate buffer (pH 8) before assay. A model was assumed in which free antibiotic can diffuse Lung tissue and pericardial fluid. Nine male patients (56 to passively from plasma to interstitial fluid. Both the volume 68 years old and weighing 78 to 47 kg) undergoing thoracot- of interstitial fluid (as a fraction of total tissue volume) and the binding to other tissue constituents are unknown. In the * Corresponding author. model used by us, the following equations hold: 1875 1876 BENONI ET AL. ANTIMICROB. AGENTS CHEMOTHER.
I *Stde 10th dese (during surgery) =eg/=I
, -. I. Om sa« i a bm 0 4 TTii 8 mrs FIG. 1. Concentration-time curves after single and multiple doses (500 mg every 8 h) of flurithromycin in patients undergoing maxillofacial surgery. Symbols: 0, total concentration in serum; 0, free concentration in serum; *, total content in bone tissue; A, total content in soft tissue. dCildt = DaCp - DaCi (1) pericardial fluid, elimination of the drug, and the relationship in which C1 is the concentration in interstitial fluid, Cp is the between drug levels in blood and tissues. After correction concentration in plasma (nonprotein bound), D is the diffu- for protein binding, the free concentrations in serum were sion coefficient, and a is the area-of-diffusion boundary. calculated. Fig. 1 and 2 show the curves of the total free Integration leads to concentrations in serum after single and multiple doses of flurithromycin and drug content in lung, bone, soft tissue, and pericardial fluid. These curves are based on the means of Ci = Da(J Cpdt - Cidt) (2) the results of two values at each time point; standard deviations were not calculated. in which Cdt is the area under the concentration-time curve The AUC from 0 to 7 h for lung samples was 23.4 ,ug h/g, (AUC) when Ci returns to zero after the first dose. Equation and the AUCs from 0 to 3 h for bone and soft tissue samples 1 leads to were 2.4 and 1.93 ,ug- h/g, respectively. The ratios of flurithromycin AUC from 0 to T in tissues to that in serum C1dt= J Cpdt (3) were 2.15 for lung, 0.58 for bone, and 0.46 for soft tissue. Moreover, the AUC,,s in tissue (in micrograms hour per gram) were 54.8 for lung, 31 for bone, and 7.0 for soft tissue. The relationship between tissue content and concentrations The ratios (R) of AUCo. in tissues to that in serum were 8.3 in interstitial fluid and other tissue components is expressed for lung, 3.6 for bone, and 0.8 for soft tissue. by Concentrations in the interstitial fluid at individual time Ct = fiCi + (1 D)Cr (4) points were calculated by dividing through R total content in in which Ct is total tissue content, f; is the volume fraction of tissues. Figure 3 shows the curves of the free antibiotic interstitial fluid, and Cr is the mean concentration in other concentrations in serum and the calculated concentrations in tissue components. Combination of equations 3 and 4 leads interstitial fluid. The calculated contents in the interstitial to AUC,/AUCp = (R -f1)1(1 - f]), in which R = AUCt/ fluid of bone and lung tissues were similar to the free AUCp. In these equations, the subscripts r, p, i, and t refer concentrations in serum, whereas concentrations in soft to measurements in other tissue components, in plasma, in tissue were very different. interstitial fluid, and in tissue, respectively. The pharmacokinetic parameters (mean and standard de- The predicted accumulation factor (Rp) was calculated by viation) for each group after the first and last doses of dividing 1 by the quantity (1 - e-T), in which ,B was flurithromycin are presented in Table 1. The mean residence determined from the first dose and T was the dose interval. times were 10.2 h for group 1 and 8.3 h for group 2. The ratio The observed accumulation factor (Ro) was determined by of the AUC during a dose interval to the AUC for the first the ratio of the AUC from 0 h to infinity (AUC_OO) for the last dose averaged 0.25 for group 1 and 0.67 for group 2. The dose to the AUC0OC for the first dose. predicted and observed accumulation factors were 2.2 and Pharmacokinetic parameters were calculated by conven- 1.8, respectively, for group 1 and 2.0 and 1.4 for group 2. tional means from the plasma concentration-time data col- lected in a single-dose portion of the study and after admin- DISCUSSION istration of the last dose of the multiple-dose regimen The pharmacokinetics of flurithromycin and the relation- according to the formulas of Gibaldi and Perrier (5). ship between free flurithromycin concentration in serum and levels in tissues were studied in patients undergoing thora- RESULTS cotomy or maxillofacial surgery. In four patients undergoing The serum and tissue curves provide an indication of the maxillofacial surgery and in one patient undergoing thora- penetration offlurithromycin into bone, soft tissue, lung, and cotomy, treatment was suspended because of gastrointesti- VOL. 32, 1988 PHARMACOKINETICS OF FLURITHROMYCIN 1877
I St do" 7 th dome (during thorasteomy) meg/ml meg/ml (g)
4 hours houae FIG. 2. Concentration-time curves after single and multiple doses (500 mg every 8 h) of flurithromycin in thoracotomized patients. Symbols: 0, total concentration in serum; 0, free concentration in serum; *, concentration in pericardial fluid; A, total content in lung tissue. nal side effects. The correlation between incidence or sever- flurithromycin obtained from the last dose was similar to that ity of gastrointestinal symptoms, flurithromycin obtained from the first dose, consistent with linear and concentration in plasma, and therapeutic effect has not been predictable accumulation. studied. In pericardial fluid, the flurithromycin curve follows the The pharmacokinetic data derived from this study agree total-serum curve. The protein content of this fluid reflects with preliminary results discussed by Bonardi et al. (1). The the similar behavior of the drug observed in serum'. The ratio serum half-lives of flurithromycin were similar in the two of content in total tissue and free levels in serum ranged groups, ranging from 7.7 h after 1 dose to 9.9 h after 10 greatly in the tissues examined, being very high in bone and doses. This study shows that higher concentrations of fluri- lung tissues. In these cases, flurithromycin is present not thromycin in serum and tissues were achieved than would be only in interstitial fluid but also in cells, and it appears that expected from a comparable oral dose of erythromycin. differences in tissue binding may play a role. The large Comparative crossover studies should be performed to con- volume of distribution is in accordance with these values. firm this point. Flurithromycin is rapidly absorbed, giving a The amount of antibiotic per gram of tissue is not equal to peak level in serum by 1 to 2 h. In general, the observed the concentration of free, potentially active antibiotic. Con- accumulation factor was close to the predicted accumulation tent in total tissue is determined by the free aqueous antibi- factor. For both groups, the mean elimination half-life Qf otic concentration, the tissue-bound fraction of the drug, and
wumg/mI mg/mI
-A&
-de P *--OOO I-m lo - I 410-; /.I
, , ... , I v 0s0OOasf - I - 1. i2 a a OM as I La S 3 S ours hours FIG. 3. Concentration-time curves after multiple doses of flurithromycin showing free concentration in serum (0) and calculated content in interstitial fluid in bone (0), soft tissue (A), and lung tissue (U). 1878 BENONI ET AL. ANTIMICROB. AGENTS CHEMOTHER.
TABLE 1. Model independent pharmacokinetic parameters after single and multiple doses of flurithromycina Serum Treatmentb CL/F (liters/h) AUC0. V/F (liters/kg) (AgUCOA,uM UmlAin)freeRo. R (pg. h/mi) (pLg. h/mI) (pg h2/ml) (p.hm)t1in free () MT() R serum Group 1 Dose 1 30.8 ± 7.2 16.2 ± 0.8 5.7 ± 1.1 166.0 ± 10.2 8.6 ± 0.4 8.6 10.2 ± 0.8 1.8 ± 0.2 2.2 ± 0.3 Dose 10 22.3 ± 1.1 4.2 ± 0.3 9.9 Group 2 Dose 1 30.8 ± 7.2 16.2 ± 0.3 5.4 ± 2.0 134.5 ± 8.5 6.6 ± 0.5 7.7 8.3 ± 0.5 1.4 ± 0.3 2.0 ± 0.3 Dose 7 28.8 ± 1.5 10.9 ± 0.9 8.5 a CL, Clearance; AUCO, area under the concentration-time curve from 0 h to infinity; V, volume of distribution; AUCS,, area under the concentration-time curve from 0 h to , in which T equals 7 h for group 2 and 3 h for group 1; AUMC, area under the first moment of the serum concentration-time curve; '1/2, mean elimination half-life; MRT, mean residence time (AUMC/AUC); R0, observed accumulation factor; Rp, predicted accumulation factor. Values other than those for tj12 are means ± standard deviations. b Dosage regimen was 500 mg every 8 h. the tissue fraction that is impermeable to the antibiotic. LITERATURE CITED Since the amount ofbone and lung tissue that is impermeable 1. Bonardi, G., P. Del Soldato, and A. M. Lepore. 1986. Bioavail- to the antibiotic is unknown, tissue binding cannot be ability of flurithromycin after P.O. and I.V. dosing in healthy calculated. However, because the ratio (R) between tissue volunteers. Acta Pharmacol. Toxicol. 59(Suppl. 5):291. content and free levels in serum is different for different 2. Brun, Y., F. Forey, J. P. Gamondes, A. Tebib, J. Brune, and J. tissues, it may be concluded that if tissue binding occurs, it Fleurette. 1981. Levels of erythromycin in pulmonary tissue and must vary depending on the tissue. High levels of macrolides bronchial mucus compared to those of amoxicillin. J. Antimi- accumulate in the lungs within several hours (2, 3), an crob. Chemother. 8:459-466. observation which we confirmed for flurithromycin. More- 3. Fraschini, F., P. C. Braga, V. Copponi, G. Gattei, E. Guerrasio, over, the high flurithromycin concentration in bone observed F. Scaglione, F. Villa, and G. Scarpazza. 1980. Tropism of in our study was probably due to the ability of the drug to erythromycin for the respiratory system, p. 659-662. In J. D. bind to components of bone cells because of the presence of Nelson and C. Grassi (ed.), Current chemotherapy and infec- fluorine. tious disease. American Society for Microbiology, Washington, In experiments with animals, Grady and Stern (6) found D.C. detectable erythromycin concentrations in bone material 4. Gialdroni Grassi, G., R. Alesina, C. Bersani, A. Ferrara, A. from rats. Sandberg Sorensen et al. (10), studying humans, Fietta, and V. Peona. 1986. In vitro activity of flurithromycin, a detected erythromycin concentrations in all narrow and novel macrolide antibiotic. Chemioterapia 5:177-184. cancellous bone specimens which exceeded the MIC of 0.25 5. Gibaldi, M., and D. Perrier. 1985. Pharmacokinetics. Marcel to 0.5 ,ug/ml for susceptible strains of Staphylococcus au- Dekker, Inc., New York. reus. 6. Grady, J. E., and K. F. Stern. 1966. Penetration of lincomycin In soft tissue, the drug content in interstitial fluid is higher into bone, p. 201-205. Antimicrob. Agents Chemother. 1965. than the free levels in serum and similar to total content in 7. Hand, W. L., R. W. Corwin, T. H. Steinberg, and G. D. tissue. It is probable that the presence of inflammation or Grossman. 1984. Uptake of antibiotics by human alveolar mac- exudate and consequently of phagocytes in the interstitial rophages. Annu. Rev. Respir. Dis. 129:933-936. fluid of soft tissue and the ability of macrolides to penetrate 8. Kroening, U., S. Liebig, and M. Wundschock. 1978. Tobramy- into phagocytes account for this high content. In alveolar cin-Spiegel im menschlichen Lungengewebe. Infection 6:231- macrophages, the concentrations of macrolides in intracel- 235. 9. Periti, P., P. Nicoletti, and A. Novelli. 1985. Comparative in vitro lular fluid are 9 to 23 times higher than the concentrations in evaluation of the antimicrobial activity of flurithromycin by extracellular fluid (7). Even if high concentrations are means of MS-2 Abbott Research System. Chemioterapia 4:156- achieved intracellularly, the antimicrobial activity of the 160. antibiotic against phagocytosed bacteria is difficult to prove. 10. Sandberg Sorensen, T., H. Colding, E. Schroeder, and V. Tham- It is probable that for bacteriostatic drugs such as macro- drup Rosdahl. 1978. The penetration of cefarolin, erythromycin lides, both very high and prolonged concentrations in tissue and methicillin into human bone tissue. Acta Orthop. Scand. 49: are necessary for a favorable therapeutic result. 549-553.