Veterinary Committee on Antimicrobial Susceptibility Testing Vet CAST
Veterinary Committee on Antimicrobial Susceptibility Testing (Vet. CAST) What is a MIC? Relationship between a MIC and conventional pharmacodynamic parameters P. L. Toutain VMAS symposium/workshop Uppsala 13 December 2017
MIC: definition (ISO 20776) • It is the lowest concentration (in mg/L) of an antimicrobial agent (AMD) that, under defined in vitro conditions, prevents the appearance of visible growth of a microorganism within a defined period of time
Minimum Inhibitory Concentration (MIC) is used for many purposes • For the susceptibility testing (AST) to guide clinician • MIC distributions to define wild type or non-wild type bacterial populations (ECOFF) • Resistance surveillance, • For PK/PD indices (f. AUC/MIC, f. T>MIC) • Comparative “potency” of new AMD during drug development • ……
MIC should be obtained in strictly defined in vitro conditions • Any calculation or expression of the MIC should include a description of the method by which the MIC was determined or a reference to a published method (e. g. CLSI or EUCAST) should be given. • Many factors of variability may impact the MIC value
MICs estimated with different inoculum densities, relative to that MIC at 2 x 105 Ciprofloxacin Gentamicin Linezolid Oxacillin Daptomycin Vancomycin PL Toutain Ecole
influence of p. H on AMD activity 40 Zone diameters (mm) x 30 x x x 20 Streptomycine 6 10 Lorian, p 35 7 x Chlortétracycline 8 p. H
The matrix effect • MHB is the standard matrix • What about MIC in a more relevant biological matrix as serum? • The case of macrolides
The case of tulathromycin CLSI CBP=16µg/m. L MIC Tulathromycine PL Toutain; Ecole
MICs for M. haemolytica (calf isolates) in biological fluid (calf serum) versus broth (n=6) MICserum / MICbroth ratios Serum MIC 29 times higher than broth MIC Serum MIC 50 times lower than broth Tulathromycin
“Macrolides show antimicrobial activity against P. aeruginosa in eukaryotic media through increased uptake and reduced efflux. These data may help explain the clinical efficacy of macrolides against pseudomonal Infections”. Clinical Infectious Diseases 2012 55 , 534 14
MIC of azithromycin (mg/L): CA-MBH vs. RPMI 1640 medium Macrolides are considered intrinsically inactive against P. aeruginosa, with high (� 256 mg·L− 1) minimal inhibitory concentrations (MICs) measured in the recommended conventional broth but not in RPMI 1640 Buyck et al 2012 CID 55, 534 15
Macrolides increase outer Membrane Permeability in RPMI, not CA-MHB by down regulating opr. M efflux pump • Macrolides have low MICs against P. aeruginosa when tested in the presence of serum, bronchoalveolar lavage fluid or culture media used for eukaryotic cell cultures. • Enhanced macrolide activity in these media results from an increased permeability of the bacterial outer membrane by downregulation of opr. M by macrolides in RPMI 1640 medium Buyck et al 2012 CID 55, 534 16
Impairment of opr. M Expression by Azithromycin in RPMI 1640 Medium The addition of a subinhibitory concentration of azithromycin (1 mg/L) caused a marked decrease of opr. M expression in bacteria grown in RPMI 1640 medium but no change in those grown in CA-MHB. 17
Clinical consequences of (1) ignoring the true potency of macrolides against bacteria and (2) seeking to explain macrolide efficacy by an antiinflammatory activity The risk of selecting macrolide-resistant P. aeruginosa 18
What is exactly a MIC? An hybrid variable obtained in standardized conditions
MIC is an activity • MIC provides a quantitative measure of an phenotypic activity over a certain period of time (most often 18 -24 h) • It is the end result of growth and kill/death of bacteria by the AMD • This activity reflects more fundamental pharmacodynamic (PD) parameters An antimicrobials can be fully describe by its fundamental PD parameters (as for any others drugs) and a MIC can be also viewed as an hybrid variable
Main PD parameters that can be estimated from killing curves • MIC: Minimum inhibitory concentration – is not a “parameter” but an hybrid variable – Dimension: concentration • Efficacy (Emax or Kmax) – Measured in vitro as a Maximum killing rate constant (Kmax)) – Dimension : per hour (1/h) • Potency – Measured as an EC 50 – Dimension : concentration • Sensitivity – Measured as a slope of the concentration -effect relationship – Dimension: scalar (without unit)
The 3 PD parameters of a concentration effect relationship Emax (Kmax) EC 50 Emax 2 2 Emax/2 2 all ow 1 1 sh Emax 1 Slope Time-dep Conc-dep stiff AMD Concentration Efficacy (pharmacological) EC 501 EC 502 Potency • Sensitivity • Allow to classify AMD as time-vs conc-dependent
Potency vs. efficacy • Not to be confused when doing pharmacology – MIC reflect both thus MIC is hybrid – Efficacy in a clinical context encompass all PD parameters • Advantages and disadvantages of a high potency – Advantage : Low dose to administrate when very potent – Limits: potency often positively correlated to lipophilicity
Potency of Fluoroquinolones Hydrophobicity vs MIC for S aureus 128 MIC (µg/m. L) 64 32 R 2 = 0, 6764 16 8 4 MIC SA 2 Expon. (MIC SA) 1 0, 5 0, 25 0, 125 0, 0625 -0, 5 0 Takenouchi et al AAC 1996 0, 5 1 1, 5 2 Hydrophobicity (Clog-P) 2, 5
How to model efficacy, potency and sensitivity from killing curves: the Emax model % of Maximal Effect Maximal effect Efficacy EC 50 [Concentration] Potency
Effect (%) Sigmoid Emax PD Model: the “n” of Hill n=5 n=2 n=1 n = 0. 5 n = 0. 1 EC 50 [Drug] EC 50
The Emax/Hill model parameters vs. variables Dependent variable Killing rate Efficacy (parameter) Potency (parameter) Sensitivity (parameter) Independent variable (µg/m. L or µMolar or MIC multiple)
Interpretation of the slope (Gamma or Hill coefficient)
MIC can be related to its more fundamental PD parameters by the modeling of killing curves
A semimechanistic model to analyse KC (see presentation by Lena and Anders)
The drug Killing rate is a function of basic PD parameters
The three PD parameters of florfenicol for M. haemolytica Parameters What is measured Symbols Units Values (median, bootstrap) 95% CI Efficacy Maximal Killing rate Emax Per hour 2. 88 2. 3 -3. 9 Potency Concentration for Emax/2 EC 50 mg/L 0. 74 0. 60 -0. 94 Sensitivity Steepness of the slope Hill coefficient Scalar (Gamma) 2. 4 1. 82 -3. 0
What are the relationship between the MIC and the three PD parameters?
Relationship between the MIC and the 3 PD parameters MIC is define as for ISO/EUCAST/CLSI With time=18 h, N(0)=5 x 105 and N(18 h)=108
What is the most influencing factors on the value of the MIC? How to document this issue? Monte Carlo simulation and sensitivity analysis
What is a MIC? Hand-on exercise using Monte Carlo simulations
Q 1: compute the MIC of florfenicol for M. haemolytica from its PD parameters and test tube conditions Parameters Units Test tube conditions Values (mean, bootstrap) Kgrowth (1/h) 0. 854 Emax Per hour 2. 830 Duration for the test 24 h EC 50 mg/L 0. 726 Initial inoculum (N 0) 5*105 2. 424 Final inoculum size i. e. N(24 h) 108 Hill coefficient Scalar (Gamma) Click here to access the Excel sheet
Q 1: Results It is generally accepted that broth MIC tests are reproducible to within one doubling dilution of the real end point (i. e. ± one well or tube in a doubling dilution series).
Q 2: what is the computed MIC for florfenicol if the initial inoculum is 2 x 107, not 5 x 105
Q 3: What is the Minimum bactericidal concentration (MBC)? • It is the concentration (mg/L) with no (or few) surviving bacteria in the MIC test • Concentration resulting in at least 99. 9% killing compared to initial inoculum • Operational definition final INO=5 x 102 CFU/m. L – (see Mouton et al Clin Pharmacokinet 2005 44 767 -768):
Q 3: What is the Minimum Bactericidal concentration (MBC)?
An infinity of what if questions • To systematically explore influence of the different PD parameters and test tube conditions, the only viable option is Monte Carlo Simulations • For a general presentation of MCS, see presentation by Ludovic Pelligand
• An add-in design to help Excel spreadsheet modelers perform Monte Carlo simulations • Others features – Search optimal solution (e. g. dose) by finding the best combination of decision variables for the best possible results 12/09/2021
MIC: Monte Carlo Simulations Replace point estimates by distribution to solve the equation
Crystal Ball
Open your Excel sheet with Crystal Ball and define assumptions
Assumptions for Emax: log-normal distribution; GSD=10% Also minimum
Assumption for Duration of the test : Time
Assumption for Final INO
Define forecast and start simulations (n=5000)
Forecast distribution for MIC
Q 4: What is the certainty to have a MIC between 0. 40 and 0. 60 mg/L Right Certainty Grabber Left Certainty Grabber
Sensitivity Charts • Sensitivity charts show the influence of each assumption on MIC. • The overall sensitivity of a forecast to an assumption is a combination of two factors: – The model sensitivity of the forecast to the assumption (structural features of the model) – The assumption’s uncertainty (here 10% for all factors)
Assumptions and their effects on MIC Major influence of EC 50 and Kgrowth (medium effect) No or minimal influence on initial and final INO, Time
Choose assumptions: here only test tube conditions
Choose your assumptions: Only test tube conditions
Choose your assumptions: Only test tube conditions
Choose assumptions: Here only PD parameters
Q 5: same questions but with Gamma=0. 5, 1 or 5 Gamma=0. 5 Gamma=1. 0 Gamma=5
Time vs. Concentration dependent AMD: value of the Hill coefficient
Time kill curves and kill rates as a function of AMD concentrations: value of the Hill coefficient Tobramycin Meropenem Time kill-curves Gamma=3. 32 005 Gamma=0. 7 Mouton & Vinks 2 Kill rate vs. conc
Conclusion of MCS • For our conditions of simulations we can conclude: – For a time-dependent AMD (high gamma) , EC 50 is the most influencing factor for computation of a MIC – For a concentration- dependent AMD (low gamma), Kgrowth is the most influencing factor for computation of a MIC
- Slides: 59