Anti-infectives: the evidence base or lack thereof!

Anti-infectives: the evidence

base or lack thereof!

Andrew Lovering

Antimicrobial Reference Laboratory

North Bristol NHS Trust

Where is TDM useful?

Where exposure predicts toxicity – Aminoglycosides, vancomycin, colistin, ethambutol,

cycloserine, flucytosine, voriconazole

Where exposure predicts clinical cure or

resistance emergence – Teicoplanin, vancomycin, posaconazole, itraconazole

Where dose poorly predicts exposure – Physiological abnormalities, extra-corporeal support, oral

agents

Data Sources to Support TDM

Linking exposure to toxicity – Animal studies

– Case reports

– Clinical trials

Linking exposure to outcome – Case reports

– In vitro modelling

– In vivo modelling (animals)

– Clinical trials (confirmatory)

– In silico

Pharmacokinetics/Pharmacodynamics

Nicolau. 2008 Crit Care 12(Suppl 4):S2

Pharmacodynamic Optimisation

Nicolau. 2008 Crit Care 12(Suppl 4):S2

The TDM Paradigm

0

2

4

6

8

10

12

14

16

0 20 40 60

Time (h)

Co

ncen

trati

on

(m

g/L

)

Gentamicin

0

1

2

3

4

5

6

7

8

9

0 10 20 30

Time (h)C

on

cen

trati

on

(m

g/L

)

Ciprofloxacin

Toxic Concentration Sub-therapeutic Concentration (E.coli)

TDM in Reality

Pascual A et al. Clin Infect Dis. 2012;55:381-390

TDM for Vancomycin

First isolated 1953 and approved 1958 – Pre 5-10 mg/L and Post 20-40 mg/L

Rybak et al. Jan 2009. (AJHSP 66:82-98)* – Pre dose after 4th dose

– Target exposure of AUC:MIC >400 (S.aureus)

– <10 mg/L promotes resistance

– 15-20 mg/L for isolate with MIC of 1mg/L

– Loading dose of 25-30 mg/kg in serious sepsis (ABW)

*Infectious Diseases Society of America

American Society of Health-System Pharmacists

Society of Infectious Diseases Pharmacists

Use of the 2009 Consensus

Guidelines

163 US hospital responses to survey

Most now only use trough levels

Most target levels >10 mg/L and 15-20

mg/L in complicated cases

Few hospitals use loading doses

Very few hospitals dose on the basis of

ABW

Davis . September 2013. Pharmacotherapy ePub.

Vancomycin Since 2009

Kullar (CID 2011); Brown (AAC 2012); Holmes (AAC

2013)

– AUC:MIC targets of 373-453

Casapao (AAC 2013); Jacob (IJAA 2013)

– High MIC (>1.5) associated with increased mortality and

treatment failure

VanHal (AAC 2013)

– Nephrotoxicity risk increases with trough concentration and

duration

Cianferoni (Infection 2013); Norton (JAC 2013)

– Continuous infusion levels >30 mg/L increase risk of AKI

Vancomycin TDM Post 2013

Consensus Guidelines II in draft (2014/5) – Reaffirm AUC:MIC of >400

– Confirm trough of 15-20mg/L and extend to children

– Loading algorithm for >2g and recommendations in

obesity

– Propose early monitoring (<24h)

– Propose dosing in dialysis

– Propose CI limit of 30 mg/L and closer monitoring

– Clarify link between exposure and nephrotoxicity

TBM for First-Line TB Agents?

Pasipanodya et al, August 2013 (JID).

– 142 patients, define targets associated with good outcomes

(PZA >58.3 mg/l, RIF>6.6 mg/L, INH >8.8 mg/L)

Prahl et al. September 2013 (ICAAC 1641)

– 32 patients, death/relapse associated with low RIF and INH

(P=0.013)

Pasipanodya & Gumbo. July 2013 (CID 57: 21-31)

– DOT is not associated with better outcomes

Pasipanodya et al. July 2012 (CID 55: 169-72)

– Pharmacokinetic variability best predicts failure

Babalik et al. Nov. 2013 (Int J Tub Lung Dis 17:1442-7)

– 21 patients with high pharmacokinetic variability (>40% sub

therapeutic)

Drug concentration levels predictive of optimal: 2-months and long-term outcomes, 142 patients

Drug Odds ratio (95% CI) for poor long-term outcome

Any one drug above threshold 7.57 (2.57 – 22.34)

Any two above threshold 2.65 (0.99 – 7.18)

All three above threshold *** Reference

Drug Odds ratio (95% CI) for poor long-term outcome

Any one drug above threshold 7.57 (2.57 – 22.34)

Any two above threshold 2.65 (0.99 – 7.18)

All three above threshold *** Reference

Drug Peak concentration threshold (mg/L): 2-months

Threshold AUC (mg*h/L) overall long-term outcomes

Pyrazinamide 58.3 363

Rifampin 6.6 13

Isoniazid 8.8 52

Drug Peak concentration threshold (mg/L): 2-months

Threshold AUC (mg*h/L) overall long-term outcomes

Pyrazinamide 58.3 363

Rifampin 6.6 13

Isoniazid 8.8 52

Pasipanodya et al 2013 JID

Drug Odds ratio (95% CI) for poor long-term outcome

Any one drug above threshold 7.57 (2.57 – 22.34)

Any two drugs above threshold 2.65 (0.99 – 7.18)

All three drugs above threshold Reference

Drug Odds ratio (95% CI) for poor long-term outcome

Any one drug above threshold 7.57 (2.57 – 22.34)

Any two drugs above threshold 2.65 (0.99 – 7.18)

All three drugs above threshold Reference

1

TDM for First-Line TB Agents

Pasipanodya et al. 2013. Journal Infectious Diseases

Evidence for TDM in TB

Identification of targets associated with good

outcomes

Confirmation that failure to reach the targets

results in treatment failure

Identification that a significant proportion of

patients have sub-therapeutic levels

Confirmation that TDM can correct sub-

therapeutic levels

No definitive RCT to show that TDM is

effective

Modelling of improved dosing in TB

Dose Optimization

n=25,000 (%)

Standard Therapy (ST)

n=25,000 (%)

Clinical measures Best-case scenario Best-case scenario

Cured 24,393 (98) 18,055 (73)

Cured-prolonged

treatments

15 (0.1) 4,486 (18)

Deaths 3 (0) 1,056 (4)

Antibiotics and Outcome in

Septic Shock

Kumar 2009. Chest 136: 1237-48

Therapeutic Drug Monitoring

– Beta Lactams 30 bed ICU over 11m

Adjusted levels 2x per week for patients

on intermittent or continuous infusion

Target 100% free drug 4-10 times mic

74.5% of patients outside range

– 50.4% below target

– 23.7% above target

Roberts. 2010. IJAA 36:332-339

TDM of Beta Lactams

Indication N Dose ↑ Dose↓

Bacteraemia 18 72% 17%

HAP 89 60% 25%

SSTI 16 50% 19%

Meningitis 17 47% 12%

CAP 47 32% 23%

Urosepsis 7 29% 57%

Roberts. 2010. IJAA 36:332-339

Outcomes of Extended or CI vs. Standard

IV Therapy – Systematic Review

1229 patients in 14 studies

Mainly non-randomised studies

Carbapenems or piperacillin/tazobactam

Moderate to severe sepsis

Mortality significantly (P<0.001) decreased

with extended or continuous infusion

– RR 0.59, 95% CI 0.41-0.83

Falagas. 2013. CID 56:272-282

BLING II: Continuous Infusion

vs. intermittent IV dosing Multicentre, double blind, double dummy

RCT

28 ICUs Australia/New Zealand

Severe sepsis treated with carbapenem or

piperacillin/tazobactam or

ticarcillin/clavulanate

420 recruitment powered to detect

difference in ICU free days

Due to report early 2014

Dulhunty. 2013. Crit Care Resus 15:179-85

Summary

Fundamental basis for dose optimisation of

many drugs established

Large literature base that is rapidly expanding

Systematic reviews and meta analyses

published, but few RCTs

Poor uptake and dissemination of knowledge

Lack of a mechanism to transfer knowledge to

the bedside