EUCAST 2010 Why European breakpoints?

EUCAST 2010 Why European breakpoints? Gunnar Kahlmeter Chairman of EUCAST Clinical microbiology Växjö, Sweden gunnar.kahlmeter@ltkronoberg.se

Antimicrobial susceptibility testing to predict success and failure in antimicrobial therapy (individually or as a basis for empiric therapy) To provide early warning (for infection control) for epidemiology (to describe, compare, etc resistance development and to determine the effect of measures to counteract resistance development)

Wild type Susceptible Clinical resistance Clinical resistance Clinical resistance Clinical resistance Phenotypically detectable resistance Genetically detectable resistance Resistant

Clinical breakpoints to categorise a microorganism S, I or R to the drug in question S implies a high likelihood of clinical success using recommended standard doses. R implies a high likelihood of clinical failure even at the maximum dosage. Clinical breakpoints may change over time.

Breakpoint committees 2001 Committee Country Disk test? BSAC United Kingdom Yes CA-SFM France Yes CLSI USA Yes CRG The Netherlands No DIN Germany No NWGA Norway No SRGA Sweden Yes

S/I-breakpoints I/R-breakpoints

Europe Increased cooperation between countries Patient exchange (travelling, migrating workforces) Professional exchange Streamlining of diagnostic and treatment guidelines Pan-European surveillance of antimicrobial resistance ESCMID, ECDC, EMEA

EUCAST 1996/2002-2010 ESCMID (1996 - ) ESCMID & EU (2003 06, 07) ESCMID & ECDC (2008 10) ESCMID & ECDC (2011 - )

National Breakpoint Committees D, F, N, NL, S, UK, EUCAST General Committee All European Countries + ISC/FESCI EUCAST Steering Committee BSAC, CA-SFM, CRG, DIN, NWGA, SRGA And 2 reps from the General Committee Subcommittees Antifungals Anaerobes Expert Rules Expert groups

Tasks 1. Determine clinical breakpoints for existing and new antibacterials, antifungals, antimycobacterials. 2. Define wild type MIC distributions and epidemiological cutoff values for bacteria and fungi. 3. Develop susceptibility testing methods and systems for internal QC. 4. Liaise with EMEA, ECDC, EFSA, EARSS and others involved in antimicrobial resistance. 5. Liaise with national committees involved in antimicrobial resistance and susceptibility testing, to facilitate implementation of European breakpoints

Decision process EUCAST Steering Committee National breakpoint committees EUCAST General Committees and Expert groups, Industry (pharmaceutical + AST) EUCAST Steering Committee in agreement with National breakpoint Committees New consultation (?) Decision

EUCAST Subcommittees Antifungals Antifungal drugs in need of breakpoints Breakpoints and rationale documents Methods Anaerobe bacteria Antibiotics in need of breakpoints Breakpoints Methods Expert Rules Tables of intrinsic resistance Expert rules (IF/THEN)

Consultation with expert groups Neisseria spp (finalised) Anaerobes (ESGARAB, ongoing) Helicobacter pylori (EHSG, ongoing) Clostridium difficile (ESGCD, ongoing) Campylobacter (VetCast, ongoing) Listeria monocytogenes

CLSI and EUCAST agree but only on the rarest of occasions

EUCAST and CLSI breakpoints are different Target microorganism No of breakpoints Same breakpoint(s) for S and R S R Enterobacteriaceae 36 2* 3 3 Pseudomonas 18 1 (imipenem) 5 2 Acinetobacter 11 1 (colistin) 4 2 Staphylococci 31 4 5 2 Enterococci 14 0 2 3 Streptococci 25 2 5 2 S.pneumoniae 29 3 1 5 H.influenzae 27 0 3 0 *In 2010 when CLSI implemented new breakpoints for cephalosporins, cefotaxime and ceftriaxone agree.

EUCAST and CLSI are different

Tools for determining clinical breakpoints do not differ widely 1. Dose or doses 2. Target organisms 3. MIC-distributions for target organisms - breakpoints not to divide MIC-distributions of WT target organisms - MIC distribution and ECOFFs determined for each species 4. Resistance mechanisms in target organisms 5. Clinical indications 6. Pharmacokinetics (Cmax, AUC, T½, Protein binding, Vd..) 7. Pharmacodynamic properties (peak conc/mic, AUC/MIC, TA, MCs) 8. Clinical outcome (clinical outcome/mic) 9. Epidemiological cutoffs, Pk/Pd-breakpoints and clinical data together determine the CLINICAL BREAKPOINT

EUCAST and CLSI are different EUCAST CLSI Clinical microbiologists and ID Funded by ESCMID, ECDC and national breakpoint committees. Industry consultative role. Decision by consensus. Five meetings per year. EUCAST=EMEA brpt committee. Clinical breakpoints and ECOFFs Rationale for decisions published Documents in public domain and free of charge Industry, the profession Funded by industry and sales of output. Industry part of decision process Decision by vote. Two meetings per year. CLSI technical standing with FDA. Clinical breakpoints Rationale for decisions not published. Documents for sale EUCAST has a systematic process of review!

Why European breakpoints in Europe? based on European standard and maximum dosages based on EMEA approved indications and outcome evaluation, Pk/Pd, multiple MIC distributions, and modern principles for determining breakpoints accepted by the European regulatory agencies Developed as part of the EMEA regulatory process and the only breakpoints in European SPCs European Centre for Disease Prevention and Control (ECDC) case definitions for antimicrobial resistance surveillance rationale behind decisions transparent and published (RDs) reviewed at intervals: with every new member of class and on the initiative of EMEA, the Company, EUCAST driven by the profession and experts on microbiology, infectious diseases, Pk/Pd and susceptibility testing methods independent of commercial interests in the public domain and free of charge

EUCAST Website www.eucast.org free of charge no login

www.eucast.org

ECOFF ECOFF

The EUCAST website now has >20 000 MIC distributions

2007 2009

ECOFF

EUCAST and existing antimicrobials Aminoglycosides Carbapenems & aztreonam Cephalosporins iv Cephalosporins oral Fluoroquinolones Glycopetides Macrolides and lincosamides Penicillins Tetracyclines Miscellaneous antimicrobials Antifungal drugs (flu- and voriconzole)

Topicals and less commonly used drugs 1. Mupirocin (Topical) 2. Polymyxin B (Topical) 3. Bacitracin (Topical) 4. Streptomycin (hlr for enterococci) 5. Neomycin (Topical) 6. Sulfamethoxazole (UTI) 7. Cephalothin (expert rules?) 8. Sulfadiazine 9. Spiramycin 10.Nalidixic acid (screening) 11.Cefoperazone 12.Pefloxacin 13.Cefradine 14.Cefamandole 15.Sulfisoxazole 16.Pipemidic acid 17.Kanamycin 18.Ceftizoxime 19.Cefprozil + 45 others

Microorganisms without breakpoints Define relevant drugs, breakpoints, methodology and MIC-distributions. Helicobacter spp Campylobacter spp Clostridium difficile Legionella spp Pasteurella multocida Listeria monocytogenes

Phoenix EUCAST breakpoints in automated AST

Phoenix, BD - EUCAST breakpoints in 2009 - Evaluations (3): 2009 2010 - EUCAST compliant panels ie. Panels containg drugs which have breakpoints from EUCAST Vitek2, BM - EUCAST breakpoints in 2010 (?) - Evaluation ongoing - Problems with concept EUCAST breakpoints Microscan, Siemens - EUCAST breakpoints 2010 - No known evaluation - Breakpoints panels and concept declared - Launch April 2010

Organism group CA on1st test (n) CA on1st test (%) Final CA (n)* Final CA (%)* Total No of tests MRSA 414 98,6 418 99,5 420 MSSA 888 99,1 893 99,7 896 CNS 353 98,1 353 98,1 360 S. lugdunensis 108 100,0 108 100,0 108 S. pneumoniae 256 99,2 258 100,0 258 GAS 244 99,6 244 99,6 245 E. faecalis 141 100,0 141 100,0 141 E. faecium 109 97,3 109 97,3 112 total Gram-positive (a) 2513 98,9 2524 99,4 2540 P.aeruginosa 305 94,1 314 96,9 324 ESBL(b) 443 92,3 463 96,5 480 E coli 937 97,6 952 99,2 960 K. pneumoniae 652 98,8 660 100,0 660 K. oxytoca 176 97,8 180 100,0 180 P. mirabilis 321 97,3 328 99,4 330 M. morganii 176 97,8 179 99,4 180 E. cloacae 172 95,6 177 98,3 180 E. aerogenes 147 98,0 149 99,3 150 Total Gram-negative 3329 96,7 3402 98,8 3444 Overall 5842 97,6 5926 99,0 5984 * Final CA is calculated after discrepancies were resolved either after re-test or by etests.

EUCAST disk diffusion method 2009 Based on MH medium 0.5 MF inoculum 16-20h incubation Most disk contents same as CLSI Most control strains same as CLSI (and QC ranges) Extensive database of MIC v Zone diameters available Significant differences to CLSI Calibrated to EUCAST MIC breakpoints Some disk contents lower MH-F for fastidious organisms

MH-F Mueller-Hinton agar + 5% defibrinated horse blood and 20 mg/l β-nad S. pneumoniae ATCC 49619 H. influenzae NCTC 8468

EUCAST breakpoint tables MIC (mg/l) brpts* S 2 R>2 Zone (mm) brpts* S 22 R<22 Insufficient evidence (Literature: not enough evidence for a breakpoint or no indication ) Inappropriate drug (Literature: poor drug don t use! Numbered footnotes Lettered footnotes IE Can not be substituted. Can be supplemented with an MIC without interpretation. Can be substituted with an automatic R. MIC-breakpoints Zone diameter breakpoints *when numbers are the same = no intermediate category

Web-links to MIC-distributions Web-links to Zone diameter distributions Web-links to EUCAST Rationale Documents

EUCAST March 2010 Harmonised breakpoints for all major antibacterial and antifungal drugs. Orphan drugs and microorganisms identified and prioritized Breakpoints for new drugs as part of the approval process with EMEA (daptomycin, tigecycline, doripenem). Epidemiological cut off values determined for all drugs. SOPs to describe formal relationship with EMEA. EUCAST breakpoints mandatory in European SPCs. ISO-standardized MIC-determination. Software and database for MIC- and zone distributions. Breakpoints implemented in national (F, D, N, NL, S, UK) systems 2007 2010. EUCAST disk diffusion test launched 2009. Breakpoint tables, QC-tables, methodology documents available on website.

Breakpoint reviews EUCAST - All breakpoints 2002 10 - Glycopeptides - VAN 2/2 mg/l - Carbapenems - No change - Report as tested - Cephalosporins - Ceftazidime, cefepime and aztreonam - Report as tested CLSI - Penicillins - raised - Carbapenems - Lowered - Glycopeptides - Lowered - Cephalosporins - Lowered - Report as tested

Implementation of EUCAST breakpoints on a national level Take a decision (health authorities, societies, laboratories) and form a national strategy Inform all stakeholders Form a National Antibiotic Committee (NAC)

National strategies and joint decisions on AST are needed! NAC National Antimicrobial Committee Chairperson, secretary, educational officer Representatives (5 6)

National Antimicrobial Committees tasks Subcommittee on Antimicrobial susceptibility testing Strategy at national level Implementation of breakpoints and methods Education Liaison and consultation with EUCAST Liaison with groups involved in AMR-surveillance (ECDC, EARSS,.). QA Antimicrobial Policies Antimicrobial Resistance Surveillance Antimicrobial Consumption and Policies

EUCAST breakpoints EUCAST: France Germany Norway Sweden The Netherlands The UK Decisions for 2010/11: Denmark Belgium Austria Estonia Ireland Finland Scotland Wales Switzerland Discussion: Spain Greece Hungary Italy Turkey Israel

ENAC European Network of Antimicrobial Committees The EUCAST General Committee.

EUCAST April 2011 EUCAST disk diffusion method implemented in 5-6 countries NACs in 10 15 countries. National Educational Workshops on European AST in several countries. EUCAST breakpoints in all major systems for AST (BSAC, CA-SFM; Commercial systems Phoenix, Vitek2, Microscan, BioMic). All Rational Documents available on website. SOPs to describe formal relationship with ECDC. ECDC decided on European breakpoints as mandatory in surveillance of antimicrobial resistance and HCAI. Breakpoints and methods for Campylobacter, Helicobacter, C.difficile, and others. Breakpoints and methods several topical antimicrobials and several less commonly used drugs. Formal decision on the future relationship between EUCAST, ECDC, EMEA and ESCMID.

CLSI breakpoints 2010 for cephalosporins and Enterobacteriacea will not require ESBL-testing* Cefepime </= 8 - >/= 16 *Laboratories may wish to continue ESBL testing for epidemiologic or infection control purposes.

Breakpoints can fail in several ways! Fail to predict failure (undercall resistance) CLSI piperacillintazobactam breakpoints in Pseudomonas Fail to predict success (overcall resistance) Penicillin breakpoints in S.pneumoniae in pneumonia Generally fail to be useful (lack of correlation with either success or failure) Erythromycin breakpoints in H.influenzae (dividing a WT population in three SIR-categories)

Breakpoints that failed to predict failures! Classical cases. Carbapenem breakpoints in MRSA (70ies) Chloramphenicol breakpoints in H.influenzae (70ies) Cephalosporin breakpoints in Enterobacteriaceae Erythromycin breakpoints in S.pneumoniae Piperacillintazobactam breakpoints in Pseudomonas Vancomycin breakpoints in S.aureus Fluoroquinolone breakpoints in critical (Salmonella) infections.. With few exceptions, revisions of breakpoints have brought breakpoints down!

...and that does not fit MRSA or KPC-producers Initial imipenem breakpoints were set to include Pseudomonas (S 4 mg/l).

and that did not go well with CAT-producing H.influenzae Initial breakpoints were set to include E.coli (S 8mg/L)

CLSI EUCAST Cephalosporin MIC distributions With current EUCAST and new (2010) CLSI breakpoints EUCAST CLSI EUCAST CLSI

CLSI breakpoints 2010 for cephalosporins and Enterobacteriacea will not require ESBL-testing* Cefepime </= 8 - >/= 16 *Laboratories may wish to continue ESBL testing for epidemiologic or infection control purposes.

Alternative A: Reduce the R breakpoint for ceftazidime and cefepime from >8 to >2 mg/l. Implications for expert rules: Report as found regardless of the presence of a resistance mechanism (ESBL or other) EUCAST is currently reviewing cephalosporin breakpoints For consultation: Cefotaxime and ceftriaxone : S 1, R>2 mg/l Ceftazidime, cefepime and aztreonam: S 1, R>4 mg/l Implications for expert rules: Report as found regardless of the presence of a resistance mechanism (ESBL or other)

Gram-negative bacteremia and cefepime Cefepime 1-2 g x 2 28 d mortality, Cart-analysis EUCAST S 1, R>8 mg/l CLSI S 8, R>16 mg/l N=204 patients Bhat et al, AAC 51:4390 (2007)

Pseudomonas bacteremia and pip/taz Tam et al, CID 46:862 (2008) Retrospective study 2002-06 Patients on piperacillin/tazobactam or another antimicrobial (controls) High MIC (32 64 mg/l) vs. low MIC (<16 mg/l) CLSI breakpoint for Pseudomonas is S <64 mg/l. EUCAST breakpoint for Pseudomonas is S <16 mg/l.

CLSI Carbapenem breakpoints EUCAST S R> S R> ECOFF * Imipenem 4 8 2 8 0.5 Meropenem 4 8 2 8 0.125 Ertapenem 2 4 0.5 1 0.06 Doripenem ND ND 1 4 0.125 CLSI 2010 (?) EUCAST (review 2009) S R> ECOFF * Imipenem 1 2 0.5 Meropenem 1 2 0.125 Ertapenem 0.25 0.5 0.06 Doripenem 1 2 0.125

Recently revised vancomycin breakpoints for Staphylococci EUCAST -breakpoints: S 4, R>8 mg/l to 2, R>2 mg/l* CLSI-breakpoints from to 8, R 32 mg/l to 2, R 16 mg/l** *EUCAST: impaired response may be seen already at MIC 2 mg/l **CLSI: impaired response at 4 and 8 and during prolonged treatment.

S.aureus bacteremia S.aureus bacteremia S,aureus bacteremia S.aureus bacteremia S.aureus bacteremia Nosocomial MRSA infections The significance of vancomycin breakpoints MIC mg/l N= Reference 2 42 12 % mortality Fridkin et al, 2003, CID; 4 21 63 % mortality 36: 429 0.5 87 44 % failure Sakoulas et al, 2004, 1-2 90.5 % failure JCM; 42: 2398 AUC/PAP 0.9 vancomycin MIC 5 MIC 2-4mg/L had longer Charles et al, 2004, CID; 0.5-2mg/L 48 bacteraemia, more fever 38: 448 AUC/PAP >0.9 days but equivalent vancomycin MIC mortality at end of therapy 2-4mg/L < 2 79 15% failure Hidayat et al, 2006, Arch > 2 38% failure Intern Med; 166: 2138 2 414 MIC 2 predicted Soriano et al, 2008, CID; 0.5 1 1.5 2 40 39 mortality (OR 6.4) 15 (10 % mortality) 38 (24 % mortality) 46: 193 Hidayat et al (2006) Vancomycin breakpoints for Staphylococci were recently revised by boh CLSI and EUCAST

Thank you!

Why European breakpoints in Europe? breakpoints for European minimum and maximum dosages based on EMEA approved indications and outcome evaluation, Pk/Pd, multiple MIC distributions, and modern principles of determining breakpoints accepted by European regulatory authorities (EMEA, ECDC) and the only breakpoints in European SPCs European (ECDC) case definitions for antimicrobial resistance surveillance rationale behind decisions transparent and published independent of commercial interests reviewed at intervals: with every new member of class and on the initiative of EMEA, the Company, EUCAST in the public domain and free of charge

EUCAST and existing antimicrobials Aminoglycosides Carbapenems & aztreonam Cephalosporins iv Cephalosporins oral Fluoroquinolones Glycopetides Macrolides and lincosamines Miscellaneous antimicrobials Penicillins Tetracyclines Antifungal drugs (flu- and voriconzole)

EUCAST breakpoint committee for new drugs through EMEA* Daptomycin Tigecycline Garenoxacin ( ) Doripenem Cefalosporine (1 ongoing) Glycopeptides (ongoing) Fluoroquinolone (1 ongoing) Diaminopyrimidine (1 ongoing) Extensions of indications *EMEA = European Medicines Agency

Recent changes in breakpoints Vancomycin Cephalosporins Carbapenems

Enterobacteriaceae vs. 3rd generation Cephalosporins in CLSI An increasing array of resistance mechanisms (ESBLs, AmpC (plasmid/chromosmal), KPC, MBL, ) in an increasing array of micro-organisms make screening for enzymes as the basis for SIR-categorisation an increasingly difficult (expert) task. Screening and confirmation often delay the susceptibility report by 24 h or more. The current breakpoints require the exclusion of an ESBL prior to use. There is clinical data to suggest that ESBL-producers with low MICs (MICs 1 mg/l for cefotaxime & ceftriaxone, and 4 mg/l for ceftazidime and 8 mg/l for cefepime) can be treated with standard doses despite the presence of an ESBL.

Current ESBL detection strategy delays reporting there are many arguments for a robust breakpoint The current breakpoints require exclusion of an ESBL prior to use. The definition of an ESBL is shaky new enzymes are regularly described The number of species that will produce an ESBL is increasing. The testing needed to exclude an ESBL is increasingly difficult and time consuming There is frequently a delay of 24 h or more in the SIRreporting of cephalosporins The evidence for success in infections with non-esbl producers with MIC in the high of the susceptible range is weak.

Carbapenem breakpoints CLSI 2009 EUCAST 2009 S R> S R> ECOFF * Imipenem 4 8 2 8 0.5 Meropenem 4 8 2 8 0.125 Ertapenem 2 4 0.5 1 0.06 Doripenem ND ND 1 4 0.125

Carbapenem breakpoints CLSI 2010 EUCAST 2010 S R> S R> ECOFF * Imipenem 1 2 2 8 0.5 Meropenem 1 2 2 8 0.125 Ertapenem 0.25 0.5 0.5 1 0.06 Doripenem 1 2 1 4 0.125

EUCAST and CLSI breakpoints are different Target microorganism No of breakpoints Same breakpoint(s) for S and R S R Enterobacteriaceae 36 2 3 3 Pseudomonas 18 1 (imipenem) 5 2 Acinetobacter 11 1 (colistin) 4 2 Staphylococci 31 4 5 2 Enterococci 14 0 2 3 Streptococci 25 2 5 2 S.pneumoniae 29 3 1 5 H.influenzae 27 0 3 0 *In 2010 when CLSI implements their 2005-decision, two cephalosporin breakpoints will agree.

Streptococcus pneumoniae EUCAST vs. CLSI Antimicrobial EUCAST S /R> (mg/l) CLSI S /R> (mg/l) Benzylpenicillin Meningitis 0.064 / 0.064 0.064 / 0.064 Benzylpenicillin Pneumonia 0.064/ 2* 2 / 4** Penicillin V PcG 0.064 / 0.064 0.064 / 1 Ampi/Amoxicillin 0.5 / 2 2 / 4 Cefotaxime 0.5 / 2 1 / 2 Moxifloxacin 0.5 / 0.5 1 / 2 Erythromycin 0.25 / 0.5 0.25 / 0.5 Azitromycin 0.25 / 0.5 0.5 / 1 *MIC-related (0.125 2 mg/l) variable dosing for pneumonia **High dose for pneumonia