14/10/2015 14th ICTDMCT - Rotterdam, The Netherlands 1

Looking into the future: routine TDM for beta-lactams in ICU ?

Paul M. Tulkens, MD, PhD

Cellular and Molecular Pharmacology& Center for Clinical Pharmacy

Louvain Drug Research InstituteHealth Science Sector

Université catholique de LouvainBrussels, Belgium

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands

Do we have a problem ?

1. Infections are (most often) treated with an antibiotic dosing regimen related to the severity of the disease rather than the susceptibility of the micro-organism ...

What is a "severe disease" ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 3

Problem ... #2 (of many)

Clinicians tend to ask (and clinical microbiologists to provide only) "S – I – R" answers based on accepted breakpoints …

But, what is a breakpoint ?

GoodEvil

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 4

In the good old time…

MIC (mg/L)

1 2 4 8 16 6432 2561280.50.250.125

mean serum concentration

Good !!

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 5

No so old but still good time ….

MIC (mg/L)

Good !!

Bad !!

1 2 4 8 16 6432 2561280.50.250.125 1 2 4 8 16 6432 2561280.50.250.125

effective serum concentration

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 6

Still good old time ….

MIC (mg/L)

Good !!

Bad !!

1 2 4 8 16 6432 2561280.50.250.125 1 2 4 8 16 6432 2561280.50.250.125

effective serum concentration

This is why microbiologists used the 2-fold dilution progression !

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 7

But now, what do you do with this ?

MIC (mg/L) May be?

effective serum concentration ?

1 2 4 8 16 6432 2561280.50.250.125

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 8

But now, what do you do with this ?

MIC (mg/L)

effective serumconcentration ?

1 2 4 8 16 6432 2561280.50.250.125or here ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 9

But now, what do you do with this ?

MIC (mg/L)

1 2 4 8 16 6432 2561280.50.250.125

This where we have

problem #2

effective serumconcentration ?

or there ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 10

Breakpoints do not make things easy

what do you do

with that ?

Problem #3: which is the correct parameter to take into account ?

• Why are β-lactams time-dependent ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 11

oxacillin

-2 -1 0 1 2

-5

-4

-3

-2

-1

0

1

2

3

MIC

log concentration (mg/L)

∆lo

g C

FU (2

4h -

0h)

gentamicin

-2 -1 0 1 2

-5

-4

-3

-2

-1

0

1

2

3

MIC

log concentration (mg/L)

∆lo

g C

FU (2

4h -

0h)

phamacodynamic model of antibiotic response24 incubation at fixed concentrations

full susceptible MSSA

Problem #3: which is the correct parameter to take into account ?

• Why are β-lactams time-dependent ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 12

phamacodynamic model of antibiotic response24 incubation at fixed concentrations

oxacillin

-2 -1 0 1 2

-5

-4

-3

-2

-1

0

1

2

3

Cmin - CmaxCmin - Cmax

MIC

log concentration (mg/L)

∆lo

g C

FU (2

4h -

0h)

gentamicin

-2 -1 0 1 2

-5

-4

-3

-2

-1

0

1

2

3

Cmin - Cmax

MIC

log concentration (mg/L)

∆lo

g C

FU (2

4h -

0h)

maximal activity in the Cmin-Cmax range

activity is concentration-dependent in the Cmin-Cmax

range

full susceptible MSSA

Problem #3: which is the correct parameter to take into account ?

• Why are β-lactams time-dependent ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 13

phamacodynamic model of antibiotic response24 incubation at fixed concentrations

meropenem

-2 -1 0 1 2

-5-4-3-2-101234

Cmin - Cmax

MIC

log concentration (mg/L)

∆lo

g C

FU (2

4h -

0h)

What do you do

with this ?

Clinical P. aeruginosa

the Cmin/MIC ratio

becomes critical !

Problem #3: which is the correct parameter to take into account ?

• Are β-lactams really time-dependent ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 14

phamacodynamic model of antibiotic response24 incubation at fixed concentrations

Clinical P. aeruginosa of increasing MIC

meropenem vs P. aeruginosa

-2 -1 0 1 2

-5-4-3-2-101234

Cmin - Cmax

MIC = 1

log concentration (mg/L)

∆lo

g C

FU (2

4h -

0h)

meropenem vs P. aeruginosa

-2 -1 0 1 2

-5-4-3-2-101234

Cmin - Cmax

MIC = 4MIC = 1

log concentration (mg/L)

∆lo

g C

FU (2

4h -

0h)

meropenem vs P. aeruginosa

-2 -1 0 1 2

-5-4-3-2-101234

Cmin - Cmax

MIC = 8

MIC = 1

log concentration (mg/L)

∆lo

g C

FU (2

4h -

0h)

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 15

Solution for β-lactams: T > MIC…

• The same for all beta-lactams ?(Free fractions of the drug [Fu] ) ?

• The same for all micro-organisms ?• The same for all infections ?• Can you apply to all patients ?• How much / How frequent ?

(Static dose vs maximum effect ?)

You know it is "time above MIC", but…

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 16

Concentration-time profile of a beta-lactam in volunteersVd = 20 L, ka = 1.2 h-1, ke = 0.3 h-1

There are variations of PK in individuals...

Unlike the Belgian 400 m run team, we are not all

(almost) equal

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 17

6 18

What is, indeed, a standard patient ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 18

Mouton, Int J Antimicrob Agents april 2002

0.20

1.20

2.20

3.20

4.20

5.20

6.20

7.20

8.20

9.20

10.20

11.20

12.20

13.20

14.20

15.20

16.20

17.20

18.20

19.20

0.00

1.25

2.50

3.75

5.00

6.25

7.50

8.75

10.0

0

11.2

5

12.5

0

13.7

5

15.0

0

16.2

5

17.5

0

18.7

5

20.0

0

21.2

5

22.5

0

23.7

5

0.95-1.000.90-0.950.85-0.900.80-0.850.75-0.800.70-0.750.65-0.700.60-0.650.55-0.600.50-0.550.45-0.500.40-0.450.35-0.400.30-0.350.25-0.300.20-0.250.15-0.200.10-0.150.05-0.100.00-0.05

Conc

Time

Prob

4h

10h

Concentration-time profile of a beta-lactam in patients with a simulation with a coefficient var. of 20 %

Variation of PK in individuals...

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 19

6 18

What is, indeed, a standard patient ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 20

Continuous Infusion of Ceftazidime (4 g/day) vs Conventional Schedule and Dosis (3 X 2 g/day) for Treatment of Ventilator-associated Pneumonia in Intensive Care

Units.P.F. Laterre, N. Baririan, H. Spapen, T. Dugernier, M. Simon, D. Pierard, H. Servais, C. Seral and P.M. Tulkens

Cliniques universitaires St-Luc & Université catholique de Louvain, Brussels; Akademische Ziekenhuis, Vrije Universiteit Brussel, Brussels; Clinique St-Pierre, Ottignies; Clinique St Joseph, Arlon; Belgium.

• ICAAC 2002 Poster no. A1 1402

• target level: 24 mg/L (max. MIC: 6 mg/L [EUCAST bkpt = 8 mg/L])

• loading dose: 10.8 mg/kg (assumed Vd: 0.4 L/kg)

• infusion: 4 g/day • assumed clearance: 102 ml/min (6.12 L/h)• drug diluted in 48 ml of water• infusion through motor-operated syringe at a

rate of 2 ml/h; • temperature 25°C or lower

target

observed(mean)

patients with continous administration of ceftazidime

But even then, serum levels remain difficult to predict with accuracy…

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 21

Solution for β-lactams: T > MIC…

• The same for all beta-lactams ?(Free fractions of the drug [Fu] ) ?

• The same for all micro-organisms ?• The same for all infections ?• Can you apply to all patients ?• How much / How frequent ?

(Static dose vs maximum effect ?)

You know it is "time above MIC", but…

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 22

How much time above MIC ?

• cefotaxime

• neutropenic mice

• K. pneumoniae

• pulmonary infection

100 % - Maximal effect ?

40 %Static dose ?

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 23

It all depends on your patient !

40 %

Moderately severe infectionin a non-immunosuppressed patient

100 % ?

Severe infectionin an immunosuppressedpatient

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 24

It all depends on your patient !

40 %

Moderately severe infectionin a non-immunosuppressed patient

100 % ?

Severe infectionin an immunosuppressedpatient

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 25

But back to MIC …!

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 26

But back to MIC …!

an ICU patient may require Cminat 4 x MIC !

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 27

And do not forget about changes in MIC (low-level resistance) during treatment !

meropenem (n=28)

D0 DL0.125

0.25

0.5

1

2

4

8

16

32

64

128

256

*

piperacillin-tazobactam (n=31)

D0 DL

2

4

8

16

32

64

128

256

512

1024

*

MIC

(mg/

L)

Change in MIC of antibiotics used in empiric antipseudomonal therapy (nosocomial pneumonia; intensive care units) towards the isolate identified before onset of therapy (D0) vs. the last isolate (DL) collected from the same patient and with clonal similarity with the first isolate. Differences were analyzed using both raw and log2 transformed data and found significant by both non-parametric (Wilcoxon matched pair test) and parametric (two-tailed paired t-test) analysis.

Riou et al. Int J Antimicrob Agents. 2010 Dec;36(6):513-22.

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 28

And do not forget about changes in MIC (low-level resistance) during treatment !

meropenem (n=28)

D0 DL0.125

0.25

0.5

1

2

4

8

16

32

64

128

256

*

piperacillin-tazobactam (n=31)

D0 DL

2

4

8

16

32

64

128

256

512

1024

*

MIC

(mg/

L)

Change in MIC of antibiotics used in empiric antipseudomonal therapy (nosocomial pneumonia; intensive care units) towards the isolate identified before onset of therapy (D0) vs. the last isolate (DL) collected from the same patient and with clonal similarity with the first isolate. Differences were analyzed using both raw and log2 transformed data and found significant by both non-parametric (Wilcoxon matched pair test) and parametric (two-tailed paired t-test) analysis.

Riou et al. Int J Antimicrob Agents. 2010 Dec;36(6):513-22.

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 29

As a result, monitoring the serum levels of β-lactams has been proposed …

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 30

But available methods are slow and complex,and do not measure the free concentration ...

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 31

A clinical algorithm or a path to success...

Knowledge or ou “educated” suspicion of the causative agent

Pathology andepidemiology Local MIC data

Obtain an MICand free serum

levels

no

Use common dosage but with attention to PK/PD

yes

Adjust the dosage on a full PK/PD basis and continue

monitoring free blood levels

S / I / Ris

insufficient !!

Is the organism probably highly

susceptible ?

But what do we need ?

• a fast and reliable assay of the serum free fraction… results available within the period of the medical shift !

• a clear definition of the desired target for efficacy … and prevention of emergence of resistance… Cmin (or Css) at 4 x the MIC ?

• a clear definition of the maximal doses without unacceptable toxicity (convulsions…) … Cmax not exceed the value of an approved mode of administration ?

• an algorithm that calculates the next dose based on population PK but also on real data from the previous administration… adaptive PK/PD modeling

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 32

Adjust the dosage on a full PK/PD basis and continue

monitoring free blood levels

A clinical algorithm or a path to success...

in ICU, the patient's situation changes rapidly !

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 33

We can always dream …

difficult machinery

acrobatic algorithms

dead ends…

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 34

But at the end …

light at the end of the tunnel…

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 35

But at the end …

light at the end of the tunnel…

… and far above sea levelhttp://www.spiegel.de/international/zeitgeist/peak-of-insanity-dutch-dream-of-building-artificial-mountain-a-784085.htmlLast accessed: 13-10-2015

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 36

Back-up

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 37

But even then, serum levels remain are difficult to predict with accuracy…

Mouton, unpublished

5 10 15 2010

100

time h

conc

entra

tion

mg/

Lpatients with continous administration of ceftazidime

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 38

0 1 3 42

Free

ser

um c

once

ntra

tion

(mg/

L)

Time after administration (h)

Which pharmacokinetic parameter drives the activity of β-lactams ?

MICfT > CMI

Time during whichthe free concentrationremains > MIC

14/10/2015 14th ICTDM&CT - Rotterdam, The Netherlands 39

Solution for β-lactams: fT > MIC…

• The same for all beta-lactams ?(Free fractions of the drug [Fu] ) ?

• The same for all micro-organisms ?• The same for all infections ?• Can you apply to all patients ? • How much / How frequent ?

(Static dose vs maximum effect ?)

You know it is "free time above MIC", but…