ECOLE NATIONALE VETERINAIRE TOULOUSE Plasma vs tissue concentration

ECOLE NATIONALE VETERINAIRE TOULOUSE Plasma vs. tissue concentration to predict antibiotic efficacy PL Toutain UMR 181 Physiopathologie et Toxicologie Experimentales INRA/ENVT Fourth International conference on AAVM Prague, Czech republic 24 -28, 2008 1

The inadequate tissue penetration hypothesis In veterinary medicine, there are many publications on tissular concentrations to promote the idea that some antibiotics having a high tissular concentration accumulate in biophase (quinolones, macrolides) and are more efficacious as suggested by their low or undetectable plasma concentrations 2

The inadequate tissue penetration hypothesis: Schentag 1990 • Two false assumptions 1. tissue is homogenous 2. bacteria are evenly distributed through tissue Þ spurious interpretation of all important tissue/serum ratios in predicting the antibacterial effect of AB Schentag, 1990 3

Statements such as ‘concentrations in tissue x h after dosing are much higher than the MICs for common pathogens that cause disease’ are meaningless Mouton & al JAC 2007 4

Objectives of the presentation: To address some basic questions 1. Where are located the bugs ? • 2. Extracellular vs. intracellular Where is the biophase? • Interstitial space fluid vs. intracellular cytosol vs. intracellular organelles 3. What is a tissue and what is a tissue concentration 4. How to assess the biophase antibiotic concentration • 5. Total tissular concentration vs. ISF concentration. The issue of lung penetration 1. 2. 6. Epithelial lining fluid (ELF): ? ? The hypothesis of targeted delivery of the active drug at the infection site by phagocytes Plasma as the best surrogate of biophase concentration for PK/PD interpretation 5

Q 1: Where are located the pathogens 6

Where are located the pathogens Cell ISF Most bacteria of clinical interest (most often in phagocytic cell) • S. pneumoniae • E. coli • Klebsiella • Mannhemia ; pasteurella • Actinobacillius pleuropneumoniae • Mycoplasma hyopneumoniae • Bordetala bronchiseptia • • mycoplasma (some) Chlamydiae Brucella Cryptosporidiosis Listeria monocytogene Salmonella Mycobacteria Rhodococcus equi 7

Q 2: Where is the biophase 8

The interstitial space fluid is the biophase 1. Most bacterial infections are located in the extracellular compartment. 2. Except few cases, In acute infections in nonspecialized tissues, where there is no abscess formation, interstitial space fluid (ISF) must be considered as the actual target space for antiinfective agents 3. ISF concentrations are of primary interest Muller et al. AAC , 2004, 48: 1441 -1453 9

Q 3: what is a tissue & what is a tissular concentration 10

Historical definition of a tissue drug concentration • In the past, it was used to characterize the total concentration in a homogenized biopsy sample – For vet medicine: a by-product of regulatory residue studies • It was assumed that: – tissue is homogenous – that antibiotics is evenly repartited in tissue – That bacteria are evenly repartited in tissue – Each of these assumptions is false and can be very misleading 11

Why a total drug tissue concentrations may be misleading? 1. Drug distributed mainly extracellularly • • β-lactams and aminoglycosides, grinding up the tissue means dilution of the drug by mixing intracellular and extracellular fluids, resulting in underestimation of its concentrations at the site of infection. 2. Drug accumulated by cells • • • fluoroquinolones or macrolides assay of total tissue levels will lead to gross overestimation of the extracellular concentration. The opposite is true for intracellular infections. 12

Methods for studies of target site drug distribution in antimicrobial chemotherapy 13

“Tissue concentrations” • Total tissue – homogenates – biopsies • Extracellular fluids – implanted cages – implanted threads – wound fluid – blister fluid – ISF (Microdialysis, Ultrafiltration) 14

The tissue cage model for in vivo and ex vivo investigations 15

The tissue cage model • Perforated hollow devices • Subcutaneous implantation • development of a highly vascularized tissue • fill up with a fluid with half protein content of serum (delay 8 weeks) • C. R. Clarke. J. Vet. Pharmacol. Ther. 1989, 12: 349 -368 16

The tissue cage model : PK limitations • A foreign body – Not a physiological space – Clinical counterparts? • Ascitic fluid, effusions ( pericardial, pleural…) • Interpretation may be difficult because PK determined by: – diffusion capacity across the TC – TC size and geometry • surface area/volume ratio is the major determinant of peak and trough drug level 17

The tissue cage model Drug administration T 1/2 varies with the surface area / volume ratio of the tissue cage Penicillin 5 to 20 h Danofloxacin 3 to 30 h Greko, 2003, Ph. D Thesis Slow equilibration inoculation (C) Time 18

Microdialysis & ultrafiltration Techniques 19

What is microdialysis (MD)? • Microdialysis, a tool to monitors free antibiotic concentrations in the fluid which directly surrounds the infective agent 20

Microdialysis: The Principle • The MD Probe mimics a "blood capillary". • There is an exchange of substances via extracellular fluid • Diffusion of drugs is across a semipermeable membrane at the tip of an MD probe implanted into the ISF of the tissue of interest. 21

Microdialysis: The Principle • the implanted MD probe is perfused with the perfusate, ie, a physiologic liquid at a very slow rate. • Substances present in the interstitial space fluid of the investigated site can diffuse into the perfusate through a semipermeable membrane at the tip of the MD probe and appear in the dialysate. • Afterward the concentration in the dialysate is chemically analyzed and the true concentration in the interstitial space fluid can be calculated. Antibiotic 22

Microdialysis materials CMA 60 Microdialysis 1. Introducer with CMA 60 Microdialysis Catheter 2. Outlet tube 3. Vial holder 4. Microvial 5. Inlet tube 6. Luer lock connection 7. Puncture needle. 23

Microdialysis : Limits • MD need to be calibrated • Retrodialysis method – Assumption: the diffusion process is quantitatively equal in both directions through the semipermeable membrane. – The study drugs are added to the perfusion medium and the rate of disappearance through the membrane equals in vivo recovery. – The in vivo percent recovery is calculated (CV of about 10 -20%) 24

MD need to be calibrated: A small experimental error in recovery estimate results in a relatively larger error in drug concentration estimates which is probably responsible for the greater interanimal variability observed in lung tissue than in the other media Marchand & al AAC June 2005 25

Ultrafiltration • Excessive (in vivo) calibration procedures are required for accurate monitoring • Unlike MD, UFsample concentrations are independent on probe diffusion characteristics 26

Microdialysis vs. Ultrafiltration Vacuum The driving force is a pressure differential (a vacuum) applied across the semipermeable membrane Microdialysis : a fluid is pumped through a membrane; The analyte cross the membrane by diffusion The driving force is a concentration gradient 27

Marbofloxacin : plasma vs. ISF In vivo filtration Microdialysis • Not suitable for long term in vivo studies Ultrafiltration • Suitable for long term sampling (in larger animals, the UF permits complete freedom of movement by using vacutainer collection method) Bidgood & Papich JVPT 2005 28 329 28

What we learnt with animal and human microdialysis studies 29

Concentration (ng/m. L) Plasma (total, free) concentration vs interstitial concentration (muscle, adipose tissue) (Moxifloxacin) Total (plasma, muscle) free (plasma) interstitial muscle interstitial adipose tissue 1000 100 2 Muller AAC, 1999 6 10 12 20 30 40 Time (h) 30

Plasma (total, free) concentration vs muscle (free) concentration cefpodoxine Total (plasma) free (muscle) free (plasma) cefixime Liu J. A. C. 2002 31

What we learnt with animal and human MD studies • MD studies showed that: 1. the concentrations in ISF of selected antibiotics correspond to unbound concentrations in plasma 2. They are generally much lower than total concentrations reported from whole-tissue biopsy specimens especially for macrolides and quinolones 32

What we learnt with MD studies: Inflammation 33

Tissue concentrations of levofloxacin in inflamed and healthy subcutaneous adipose tissue Hypothesis: Accumulation of fibrin and other proteins, oedema, changed p. H and altered capillary permeability may result in local penetration barriers for drugs Inflammation No inflammation Methods: Free Concentrations measured in six patients by microdialysis after administration of a single intravenous dose of 500 mg. Results: The penetration of levofloxacin into tissue appears to be unaffected by local inflammation. Same results obtained with others quinolones Bellmann & al Br J Clin Pharmacol 2004 57 34

What we learnt with MD studies: Inflammation • Acute inflammatory events seem to have little influence on tissue penetration. • “These observations are in clear contrast to reports on the increase in the target site availability of antibiotics by macrophage drug uptake and the preferential release of antibiotics at the target site a concept which is also used as a marketing strategy by the drug industry” Muller & al AAC May 2004 35

In acute infections in nonspecialized tissues, where there is no abscess formation, free serum levels of antibiotics are good predictors of free levels in tissue fluid 36

The issue of lung penetration 37

Animal and human studies MD: The issue of lung penetration • Lung MD require maintenance under anesthesia, thoracotomy (patient undergoing lung surgery). . • Does the unbound concentrations in muscle that are relatively accessible constitute reasonable predictors of the unbound concentrations in lung tissue (and other tissues)? 38

Cefpodoxime at steady state: plasma vs. ISF (muscle & Lung) Plasma Free plasma Muscle Lung Free muscle concentrations of cepodoxime were similar to free lung concentration and therefore provided a surrogate measure of cefpodoxime concentraion at the pulmonary target site Liu et al. , JAC, 2002 50 Suppl: 19 -22. 39

The issue of lung penetration: Imipem imipenem distribution in muscle and lung interstitial fluids Marchand & al AAC June 2005 • The major finding of this study was the observation of virtually superimposed free IPM concentration-versus-time profiles in the three media investigated, • This result not only is in agreement with theory but also is consistent with most of the data in the literature. 40

The issue of lung penetration 41

Lung infections • Uncertainty of the relevant actual location of proliferating bacteria – Alveoli, pulmonary interstitium, bronchioles, blood? ? • What is the biophase? ? – Epithelium lining fluid (ELF) – Lung IF, alveolar macrophages, tisue biopsies, blood, bronchial secretion, sputum? ? • ELF seems the most relevant specimen but potential sources of error: dilution, release of AB from alveolar macrophage in the sample 42

43

• Fenestrated pulmonary capillary bed • expected to permit passive diffusion of antibiotics with a molecular weight 1, 000 The blood-alveolar barrier Alveolar macrophage Alveolar space ISF Epithelial lining fluid AB Capillary wall AB Alveolar Epithelium Thigh junctions ELF (protein: <10%) The alveolar epithelial cells would not be expected to permit passive diffusion of antibiotics between cells, the cells being linked by tight junctions 44

Drug passage in the lung ELF Drug passage through the alveolar epithelial cells will depend on the lipophilicity and diffusibility of the antibiotics, similar to the drug entry into the central nervous system. Kiem & Schentag AAC 2008 Jan 24 -36 45

ELF concentration: possible biais ELF • Measurement problems may confound the interpretation of the ELF concentrations of antibiotics. • Cells, especially AM cells (that constitute 3. 8 to 10. 0% of ELF volume) are included in the composition of ELF • The cells may be lysed during the measurement of antibiotic concentration in BAL-derived fluids. Kiem & Schentag AAC 2008 Jan 24 -36 46

BETA-LACTAMS ELF Measured ELF concentrations of the beta-lactams are well below serum concentrations, and their respective concentrations in AM cells were negligible The low measured ELF concentrations of betalactams in comparison to their corresponding serum levels could be the result of low capacity of their unbound free fractions for penetration through the alveolar epithelial cell barriers. Kiem & Schentag AAC 2008 Jan 24 -36 47

MACROLIDES AND KETOLIDES ELF Measured ELF concentrations of macrolides and ketolides and their derived AUCs were consistently higher than serum levels by as much as 10 -fold the high ratios of ELF concentration to serum concentration for macrolides and ketolides could not be explained solely on the basis of good penetration across the alveolar epithelium. The high concentrations of macrolides and ketolides in ELF might be explained by the possible contamination of intracellular antibiotics occurring during the process of BAL. Kiem & Schentag AAC 2008 Jan 24 -36 48

FLUOROQUINOLONES Fluoroquinolones achieved higher ELF levels than their free serum concentrations Kiem & Schentag AAC 2008 Jan 24 -36 49

Kiem & Schentag’ Conclusions (1) • The high ELF concentrations of some antibiotics, which were measured by the BAL technique, might be explained by possible contamination from high achieved intracellular concentrations and subsequent lysis of these cells during the measurement of ELF content. • This effect is similar to the problem of measuring tissue content using homogenization 50

Kiem & Schentag’ Conclusions (2) • Fundamentally, ELF may not represent the lung site where antibiotics act against infection. • In view of the technical and interpretive problems with conventional ELF and especially BAL, the lung microdialysis experiments may offer an overall better correlation with microbiological outcomes. • We continue to express PK/PD parameters using serum concentration of total drug because these values do correlate with microbiological outcomes in patients. 51

The site of infection: Intracellular pathogens 52

Intracellular location of bacteria Fusion 3 Phagosome 1 B Lysosome B p. H=7. 4 4 B 2 B B Chlamydiae No fusion with lysosome Phagolysosome B S. aureaus B Brucella B Salmonella Coxiella burneti p. H=5. 0 Listeria Cytosol 53

Intracellular location of antibiotics Phagolysosome Cytosol p. H=7. 2 Fluoroquinolones(x 2 -8) beta-lactams (x 0. 2 -0. 6) Rifampicin (x 2) Aminoglycosides (slow) volume 1 to 5% of cell volume p. H=5. 0 Macrolides (x 10 -50) Aminoglycosides (x 2 -4) Ion trapping for weak base with high p. Ka value 54

What are the antibiotic intracellular expressions of activity Phagolysosome Cytosol p. H=7. 2 Fluoroquinolones beta-lactams Rifampicin Aminoglycosides Good Macrolides Aminoglycosides Low or nul 55

The hypothesis of targeted delivery of active drug at the active site by the phagocytes 56

Drug delivery by the phagocytes • Transport by "non professional" phagocytes (e. g. azithromycin by fibroblast) • Fibroblast acts as a reservoir for drug and macrophages 57

Neutrophils as antibiotic delivery system The ability of neutrophils to migrate preferentially to sites of infection makes them attractive as a delivery mechanism for antibiotics; theoretically , with the proper cellular PK, an antibiotic could be taken up by neutrophils, which would then transport the drug and later release it. This could provide a mechanism for achieving higher levels of antibiotic in tissues i. e. directly at the nidus of infection. Scorneaux & Shryock tilmicosine in pigs; JVPT 1998, 21: 257 -258 & tilmicosine in cattle in: J. Dairy Sci. 1999, 82: 1202 -1212) 58

Macrolides: how to explain efficacy of low plasma concentrations? PMN 1. PK hypothesis M M Sink or reservoir? ? ? High local M concentration in the vicinity of the bug M Nucleus 2 -PD hypothesis 59

PK model for exposure to macrolides: The limits No In vivo data to support the hypothesis 2. Slow or very slow efflux M M+ M M M Sink rather than reservoir M M (e. g. 20% within 3 h for azithromycin) more rapid (90% within 1 h) for erythromycin) Transit time in a (normal) lung capillary: 26 sec Mass balance considerations (Tylosine in piglet 10 Kg BW, dose: 10 mg/kg) • Total pool of PMN=2 m. L/L of blood • Peak Plasma concentration: 1 µg/m. L • Accumulation ratio: 50 • Total amount of drug located in PMN : 100µg in toto =1/1000 of the dose 60

Macrolides: how to explain efficacy of low plasma concentrations? PMN 1. PK hypothesis M M Sink or reservoir? ? ? Cytokine modulation, ERK signaling M High local M concentration in the vicinity of the bug MIC in vivo < MIC in vitro M Nucleus 2 -PD hypothesis 61

Conclusions: 1. In acute infections in non-specialized tissues, where there is no abscess formation, free plasma levels of antibiotics are good predictors of free levels in interstitial fluid 2. PK/PD indices predictive of antibiotic efficacy should be based on free plasma concentration 3. People who truly understand tissue concentration work in corporate marketing departments (Apley, 1999) 62