Antibiotikatherapie der Sepsis Tobias Welte Klinik fr Pneumologie

Antibiotikatherapie der Sepsis Tobias Welte Klinik für Pneumologie Medizinische Hochschule Hannover

Outline • Consequences of inappropriate MRSA NP therapy • Appropriate MRSA NP therapy – Vancomycin MIC creep – PK of vancomycin and linezolid – Nephrotoxicity of vancomycin • ZEPHy. R study and results • Conclusions MRSA, methicillin-resistant S. aureus; NP, nosocomial pneumonia Welte – Bremen 20. 02. 2014

Outline • Consequences of inappropriate MRSA NP therapy • Appropriate MRSA NP therapy – Vancomycin MIC creep – PK of vancomycin and linezolid – Nephrotoxicity of vancomycin • ZEPHy. R study and results • Conclusions MRSA, methicillin-resistant S. aureus; NP, nosocomial pneumonia Welte – Bremen 20. 02. 2014

Adequate vs inadequate antibiotic therapy Hospital mortality (%) p<0. 001 88/169 71/169 114/486 52. 1% 23. 5% 42. 0% 17. 7% 86/486 Mortality type Hospital mortality and infection-related mortality rates for infected patients from all causes (n=655) receiving either initially inadequate or adequate antimicrobial treatment Prospective US cohort study evaluating 2000 consecutive patients Kollef et al. Chest 1999; 115: 462 -74 Welte – Bremen 20. 02. 2014

Adequate versus inadequate initial antibiotic treatment and mortality p<0. 001 Hospital mortality (%) 70 60 91/147 50 40 30 20 98/345 10 0 Adequate Inadequate Initial antimicrobial treatment Prospective US study in ICUs (492 bloodstream infections) ICU, intensive care unit Welte – Bremen 20. 02. 2014 Ibrahim et al. Chest 2000; 118: 146 -55

Outline • Consequences of inappropriate MRSA NP therapy • Appropriate MRSA NP therapy – Vancomycin MIC creep – PK of vancomycin and linezolid – Nephrotoxicity of vancomycin • ZEPHy. R study and results • Conclusions MRSA, methicillin-resistant S. aureus; NP, nosocomial pneumonia Welte – Bremen 20. 02. 2014

Changes in vancomycin MIC have been observed in the clinical setting • • TEST programme = international surveillance with standardised BMD methodology Looking for progression of vancomycin ‘creep’ into vancomycin resistance Year Isolates Phenotype S. aureus MRSA MSSA 2004 -2009 All (n) Vancomycin MIC ≥ 2 µg/m. L, n (%) 20, 004 797 (4. 0) 8249 439 (5. 3) 11, 755 358 (3. 0) 2004 All (n) Vancomycin MIC ≥ 2 µg/m. L, n (%) 2525 101 (4. 0) 1158 65 (5. 6) 1367 36 (2. 6) 2005 All (n) Vancomycin MIC ≥ 2 µg/m. L, n (%) 2930 62 (2. 1) 1411 39 (2. 8) 1519 23 (1. 5) 2006 All (n) Vancomycin MIC ≥ 2 µg/m. L, n (%) 3612 94 (2. 6) 1531 50 (3. 3) 2081 44 (2. 1) 2007 All (n) Vancomycin MIC ≥ 2 µg/m. L, n (%) 4944 160 (3. 2) 2028 78 (3. 8) 2916 82 (2. 8) 2008 All (n) Vancomycin MIC ≥ 2 µg/m. L, n (%) 4348 253 (5. 8) 1481 136 (9. 2) 2867 117 (4. 1) 2009 All (n) Vancomycin MIC ≥ 2 µg/m. L, n (%) 1645 127 (7. 7) 640 71 (11. 1) 1005 56 (5. 6) p<0. 001 Welte – Bremen 20. 02. 2014 Hawser et al. Int J Antimicrob Agents 2011; 37: 219 -24

Vancomycin treatment failures increase with MIC 60 51 Failure rate (%) 50 40 30 22 20 31 27 10 0 0. 5 1. 0 1. 5 2. 0 MIC (μg/m. L) Relationship between MIC and vancomycin treatment MIC, minimum inhibitory concentration Stevens. Clin Infect Dis 2006; 42 Suppl 1: S 51 -7 Welte – Bremen 20. 02. 2014

Outline • Consequences of inappropriate MRSA NP therapy • Appropriate MRSA NP therapy – Vancomycin MIC creep – PK of vancomycin and linezolid – Nephrotoxicity of vancomycin • ZEPHy. R study and results • Conclusions MRSA, methicillin-resistant S. aureus; NP, nosocomial pneumonia Welte – Bremen 20. 02. 2014

Importance of PK-PD parameters for optimal treatment of MRSA infection • For vancomycin and linezolid Antibiotic (C) – 24 -hour AUC appears to be the most important PKPD parameter 1, 2 Peak AUC, area under the curve; MIC, minimum inhibitory concentration; MRSA, methicillin-resistant S. aureus; PD, pharmacodynamic; PK, pharmacokinetic Welte – Bremen 20. 02. 2014 24 -hour AUC MIC Time (h) 1. Andes et al. Antimicrob Agents Chemother 2002; 46: 3484 -9; 2. Moise PA et al. Am J Health-Syst Pharm. 2000; 57: S 4 -S 9.

Time to MRSA eradication with vancomycin Culture-positive patients (%) AUC/MIC<400 (n=16) AUC/MIC≥ 400 (n=18) 100 80 60 40 20 p=0. 04 0 0 10 20 Days of therapy until bacterial eradication AUIC, area under the inhibitory curve; MIC, minimum inhibitory concentration; MRSA, methicillin-resistant S. aureus Welte – Bremen 20. 02. 2014 30 Moise-Broder et al. Clin Pharmacokinet 2004; 43: 925 -42

Percent Probability of achieving AUIC/MIC ≥ 400 for vancomycin regimens of varying intensity when Cmin was between 15 and 20 mg/L MIC value (mg/L) AUIC, area under the inhibitory curve; Cmin, trough concentration; q 12 h, every 12 hours; MIC, minimum inhibitory concentration Welte – Bremen 20. 02. 2014 Patel et al. Clin Infect Dis 2011; 52: 969 -74

Vancomycin: penetration in the lung tissue Vancomycin 6 hours post-injection ELF: mean 2. 03 µg/ml (1 -2. 77 µg/ml)1 * Healthy volunteers 52%2 * Patients with pneumonia 9 -18%1, 3 No relationship between clinical outcomes and pharmacokinetics has been established * Data represent the drug level in ELF as a percentage of the simultaneous levels in plasma ELF, epithelial lining fluid Welte – Bremen 20. 02. 2014 1. Georges et al. Eur J Clin Microbiol Infect Dis 1997; 16: 385 -8; 2. Rybak. Clin Infect Dis 2006; 42 Suppl 1: S 35 -9; 3. Lamer et al. Antimicrob Agents Chemother 1993; 37: 281 -6

Linezolid: high penetration in the lung tissue Linezolid 8 hours post-injection ELF: mean 31. 4 µg/ml (8. 3 -89. 2 µg/ml)1 *Healthy volunteers ~400%1 *Patients with pneumonia 105%2 No relationship between clinical outcomes and pharmacokinetics has been established * Data represent the drug level in ELF as a percentage of the simultaneous levels in plasma ELF, epithelial lining fluid Welte – Bremen 20. 02. 2014 1. Conte Jr et al. Antimicrob Agents Chemother 2002; 46: 1475 -80; 2. Boselli et al. Crit Care Med 2005; 33: 1529 -33;

Linezolid: high bioavailability in ELF 16 critically ill VAP patients studied at steady state – All with late-onset VAP – 12 with organisms: 3 MRSA, 1 MSSA, 8 enteric Gram-negatives (4 enterobacteriaceae and 4 P. Aeruginosa) • • Serum and ELF concentrations after 2 days of therapy Blood at 10, 20, 30, 45 minutes and 1, 2, 4, 8, 12 hours after infusion BAL 1 and 12 hours after infusion Similar levels in serum and ELF – Range of peak penetration: 34 -188% – Range of trough penetration: 28 -220% Serum ELF 20 Linezolid concentration (mg/L) • 16 12 8 4 0 − 1 0 2 4 6 Time (hours) 8 10 12 No relationship between clinical outcomes and pharmacokinetics has been established Mean steady-state plasma (red circles) and ELF (blue circles) concentrations with NP (n=16). The dotted line represents the susceptibility breakpoint (4 mg/L) of staphylococci for linezolid (1). Error bars represent standard deviations BAL, bronchoalveolar lavage; ELF, epithelial lining fluid; MRSA, methicillin-resistant S. aureus; MSSA, methicillin-sensitive S. aureus; VAP, ventilator-associated pneumonia Boselli et al. Crit Care Med 2005; 33: 1529 -33 Welte – Bremen 20. 02. 2014

• • Prospective randomised study Designed to compare continuous vs intermittent vancomycin in 119 critically ill patients with MRSA infections Microbiological and clinical outcomes and safety were similar No statistically significant difference was found between the two treatment groups Patients (%) Continuous vs intermittent infusion vancomycin n=30 n=28 n=15 n=13 n=7 n=5 Adapted from Wysocki et al. Antimicrob Agents Chemother 2001; 45: 2460 -7 Welte – Bremen 20. 02. 2014

Outline • Consequences of inappropriate MRSA NP therapy • Appropriate MRSA NP therapy – Vancomycin MIC creep – PK of vancomycin and linezolid – Nephrotoxicity of vancomycin • ZEPHy. R study and results • Conclusions MRSA, methicillin-resistant S. aureus; NP, nosocomial pneumonia Welte – Bremen 20. 02. 2014

Renal effects related to vancomycin concentration Nephrotoxicity (%) p=0. 002 Renal toxicity was defined as increased creatinine by 0. 5 mg/d. L or doubling of baseline value Maximum vancomycin steady-state trough concentrations (mg/m. L)a Aggressive dosing and prolonged administration of vancomycin are associated with a greater risk of nephrotoxicity in patients with MRSA HCAP a. Maximum vancomycin serum trough concentrations ≥ 15 µg/m. L (n=49); maximum vancomycin serum trough concentrations <15 µg/m. L (n=45) Retrospective US observational single-centre study HCAP, healthcare-associated pneumonia; MRSA, methicillin-resistant S. aureus Adapted from Jeffres et al. Clin Ther 2007; 29: 1107 -15 Welte – Bremen 20. 02. 2014

Vancomycin relationships: toxicity and target attainment AUC/MIC ratio ≥ 400 MIC value Nephrotoxic event 0. 5 mg/L (%) 1. 0 mg/L (%) 2. 0 mg/L (%) Non-ICU (%) 500 mg IV q 12 h 57 15 0. 7 3 10 1000 mg IV q 12 h 90 57 15 6 16 1500 mg IV q 12 h 97 79 38 9 25 2000 mg IV q 12 h 98 90 57 14 34 AUC, area under the curve; ICU, intensive care unit; MIC, minimum inhibitory concentration Welte – Bremen 20. 02. 2014 Patel et al. Clin Infect Dis 2011; 52: 969 -74

Vancomycin concentration-time profile Lodise et al. Clin Infect Dis 2009; 49: 507 -14 • Retrospective US study correlating the vancomycin nephrotoxicity with its pharmacokinetics in 166 non-neutropenic patients Logistic regression-derived nephrotoxicity probability functions – Baseline creatinine <2. 0 mg/d. L – Vancomycin >48 h • 21 patients with nephrotoxicty (50% or 0. 5 mg/d. L increase in the serum creatinine level from baseline) • The results indicate that a vancomycin exposure–toxicity response relationship exists. The vancomycin trough value is the pharmacodynamic index that best describes this association ICU, intensive care unit Welte – Bremen 20. 02. 2014 ICU patients Non-ICU patients

Outline • Consequences of inappropriate MRSA NP therapy • Appropriate MRSA NP therapy – Vancomycin MIC creep – PK of vancomycin and linezolid – Nephrotoxicity of vancomycin • ZEPHy. R study and results • Conclusions MRSA, methicillin-resistant S. aureus; NP, nosocomial pneumonia Welte – Bremen 20. 02. 2014

ZEPHy. R study: design overview and objective Design overview • Phase IV, randomised, double-blind, multicentre, international comparator-controlled study in MRSA VAP, HAP or HCAP – Fixed-dose linezolid vs dose-optimised vancomycin – Non-inferiority study with a nested superiority hypothesis Study objective • To prospectively assess the efficacy, safety and tolerability of linezolid compared with vancomycin in MRSA nosocomial pneumonia HAP, hospital-acquired pneumonia; HCAP, healthcare-associated pneumonia; MRSA, methicillin-resistant S. aureus; VAP, ventilator-associated pneumonia Welte – Bremen 20. 02. 2014 Wunderink et al. Clin Infect Dis 2012; 54: 621 -9

ZEPHy. R study: randomisation and interventions Linezolid IV 600 mg q 12 h 1: 1 randomisation 7 -14 days EOT visit EOS visit Within 5 days 7 -30 days after EOT of EOT Follow-up call: survival status at day 60 Vancomycin IV 15 mg/kg q 12 h § Vancomycin dose adjusted by unblinded pharmacist per local protocols based on trough levels and renal impairment § Gram-negative coverage (not MRSA active) EOT, end of treatment; EOS, end of study; MRSA, methicillin-resistant S. aureus Welte – Bremen 20. 02. 2014 Wunderink et al. Clin Infect Dis 2012; 54: 621 -9

Patients with clinical response (%) Statistically superior clinical efficacy of linezolid vs vancomycin in MRSA NP in the ZEPHy. R study p=0. 042 95% CI 0. 5 -21. 6 95% CI 4. 9 -22. 0 95% CI 0. 1 -19. 8 150/180 161/201 130/186 95/165 95% CI 4. 0 -20. 7 145/214 102/186 81/174 Primary end point 92/205 Secondary end points Patients with EOS outcome of ‘indeterminate’ were excluded from the efficacy analysis CI, confidence interval; EOS, end of study; EOT, end of treatment; m. ITT, modified intent-to-treat; MRSA, methicillin-resistant S. aureus; NP, nosocomial pneumonia; PP, per protocol Wunderink et al. Clin Infect Dis 2012; 54: 621 -9 Welte – Bremen 20. 02. 2014

Statistically-significant higher rates of microbiological success with linezolid vs vancomycin in the ZEPHy. R study PP population Patients with respiratory secretions for culture Patients with microbiological response (%) 95% CI 12. 3 -30. 2 95% CI 0. 4 -21. 5 58. 1% (97/167) 47. 1% (82/174) 63. 9% (62/97) 68. 3% (56/82) 36. 1% (35/97) 31. 7% (26/82) 82. 6% (76/92) 81. 9% (149/182) 49. 0% (73/149) 60. 6% (114/188) 61. 4% (35/57) 50. 0% (26/52) 54. 1% (59/109) 48. 2% (55/114) 51. 0% (76/149) 51. 8% (59/114) Linezolid Vancomycin EOS EOT CI, confidence interval; EOS, end of study; EOT, end of treatment; PP, per protocol Welte – Bremen 20. 02. 2014 EOS EOT Wunderink et al. Clin Infect Dis 2012; 54: 621 -9

Clinical success rate 1 (%) Response differences between linezolid and vancomycin remained across most subgroups in the ZEPHy. R study APACHE, Acute Physiological Assessment and Chronic Health Evaluation; EOS, end of study; MIC, minimum inhibitory concentration; MRSA, methicillin-resistant S. aureus; MV, mechanical ventilation Welte – Bremen 20. 02. 2014 Wunderink et al. Clin Infect Dis 2012; 54: 621 -9

Median vancomycin trough concentration (µg/m. L) No relationship between vancomycin trough level and Clinical Microbiological outcomes in the ZEPHy. R study 73 Range: 82 3. 4 -50. 8 2. 8 -43. 2 47 52 5. 1 -45. 0 2. 7 -41. 4 End of study 23 15 2. 0 -42. 6 4. 1 -46. 9 76 Range: 80 2. 8 -50. 8 3. 3 -43. 2 50 49 5. 1 -45. 0 2. 7 -36. 4 22 16 2. 9 -42. 6 3. 0 -46. 9 End of study Adapted from Niederman et al. Am J Respir Crit Care Med 2011; 183: A 3932 Welte – Bremen 20. 02. 2014

Nephrotoxicity was nearly twice as common in vancomycin patients in the ZEPHy. R study (m. ITT population) Patients with nephrotoxicity (%) Laboratory evidence of nephrotoxicity (0. 5 mg/m. L increase in serum creatinine if normal at baseline, or 50% increase if abnormal at baseline) GFR, glomerular filtration rate Adapted from Wunderink et al. Clin Infect Dis 2012; 54: 621 -9 Welte – Bremen 20. 02. 2014

Linezolid has a comparable tolerability profile with vancomycin in the ZEPHy. R study. Linezolid Vancomycin AE, n (%) (n=597) (n=587) Anaemia 30 (5. 2) 42 (7. 2) Renal failure/impairment/azotaemiaa 22 (3. 7) 43 (7. 3) Cardiac arrest 11 (1. 8) 13 (2. 2) Thrombocytopenia 8 (1. 3) 13 (2. 2) Pancreatitis 5 (0. 8) 1 (0. 2) — 1 (0. 2) 4 (0. 6) 2 (0. 4) — 1 (0. 2) Polyneuropathy Pancytopenia/neutropenia Paresthesia a. Patient was reported to have ≥ 1 of the following: renal failure, renal impairment and/or azotaemia AE, adverse event Wunderink et al. Clin Infect Dis 2012; 54: 621 -9 Welte – Bremen 20. 02. 2014

Similar 60 -day mortality with linezolid and vancomycin: the ZEPHy. R study § Comparable all-cause 60 -day mortality rates 1 – Linezolid arm 15. 7%; vancomycin arm 17. 0% (ITT population) – Linezolid arm 28. 1%; vancomycin arm 26. 3% (m. ITT population) No pneumonia trial has ever demonstrated an overall mortality difference between antibiotics 2 1. Wunderink et al. Clin Infect Dis 2012; 54: 621 -9; 2. Wunderink et al. Clin Infect Dis 2012; 55: 163 -5 [letter] Patients surviving (%) 100 80 60 40 Linezolid censored 20 ITT, intent-to-treat; m. ITT, modified intent-to-treat Welte – Bremen 20. 02. 2014 Kaplan–Meier survival curves for the m. ITT population 0 Vancomycin censored 0 10 20 30 40 50 60 70 Time (days) 80 90 100 110 120

Conclusions • MRSA NP remains an important healthcare burden • Linezolid demonstrated statistically superior clinical efficacy versus vancomycin in the treatment of MRSA NP in the ZEPHy. R study 1 • Overall, linezolid demonstrated an acceptable safety and tolerability profile for the treatment of proven MRSA nosocomial pneumonia 1 • Linezolid was associated with lower rates of pneumonia-related rehospitalisation versus vancomycin in a US retrospective study 2 MRSA, methicillin-resistant S. aureus; NP, nosocomial pneumonia Welte – Bremen 20. 02. 2014 1. Wunderink et al. Clin Infect Dis 2012; 54: 621 -9 2. Mullins et al. Poster PIN 56 presented at ISPOR 2012

Welte – Bremen 20. 02. 2014