Bad Bugs, New Drugs – But No Tests! Contemporary Challenges Associated with Detection & Testing of MDR Gram Negative Bacteria Romney Humphries, PhD D(ABMM) UCLA Clinical Microbiology [email protected]

Disclosures • Dr. Humphries is a full time employee of the University of California, Los Angeles. • Dr. Humphries is a paid consultant to Merck, Allergan, Cepheid, Roche, SlipChip, MicrobeDx, DiaSorin, and receives grant support from bioMerieux, Beckman Coulter, BD, GenMark, Accelerate Diagnostics, Luminex, Merck, Allergan, Curetis, and Applied BioCode. • Dr. Humphries is on the Merck speaker’s bureau. • Dr. Humphries is a voting member of the CLSI AST Subcommittee

Objectives • Define key Gram negative resistance threats encountered by the Clinical Microbiology Laboratory • Describe antimicrobial agents available to treat MDR Gram negative bacteria, and testing options for these • List key challenges associated with susceptibility testing of MDR Gram negative bacteria

Case 62 year old woman with advanced pancreatic cancer

Vomiting & fever after surgery

CT scan: fluid collection in liver, inflammatory ascites

Blood cultures: Gram negative rods = K. pneumoniae

Amikacin Ampicillin Cefazolin Cefepime Ceftazidime Ceftazidime-avi Ceftriaxone Ceftolozane-tazo Ciprofloxacin Colistin Ertapenem Gentamicin Imipenem Meropenem Pip-tazo Tigecycline Tobramycin Trim-sulfa

2 >32 >32 >32 >32 32 >32 >32 >2 >4 >2 ≤0.5 >8 >16 >128 1 1 >4

S R R R R R R R R * R S R R R S S R

Carbapenem resistance in GNR UCLA, 2015, 1 isolate per patient* 40

% Resistance

35 30

Enterobacteriaceae (n=7614)

25

K. pneumoniae (n=957)

20 P. aeruginosa (n=744)

15 10

A. baumannii (n=55)

5 0 Ertapenem

Meropenem

Imipenem *note % R increases by 10% if each isolate counted for Psa

Carbapenem activity in GNRs

Review: mechanisms of beta-lactam resistance in GNR

“mutational resistance”

“acquired resistance”

So why all the fuss about CRE? • Before 2000: Mostly β-lactamase + porin loss • Extended spectrum β-lactamases (ESBLs) & AmpC enzymes • Did not increase over time

• After 2000, CRE incidence increased • 2001: 1% of Klebsiella, Enterobacter, and E. coli reported to NHSN were NS to carbapenems • 2011: 4% of Enterobacter, E. coli NS to carbapenems 10% of Klebsiella sp. NS to carbapenems

• Much of this increase appears to be due to the spread of carbapenemaseproducing CRE • Rates vary region to region in the US 8

GNR Beta-lactamases Class

A

B

C

D

Examples ESBLs [TEM, SHV, CTX-M] KPC SME

Produced by: K. pneumoniae and other Enterobacteriaceae S. marcescens

Notes Most inhibited by clavulanic acid Usually plasmid-mediated (not SME) Endemic in USA

Metallo-β-lactamases (MBL) (e.g. NDM, VIM, IMP, GIM, SPM carbapenemases)

P. aeruginosa Enterobacteriaceae Acinetobacter S. maltophilia

AmpC

Enterobacteriaceae Some non-fermenters

Inducible in some genera Not inhibited by clavulanic acid

OXA carbapenemases

Acinetobacter baumannii Enterobacteriaceae

Hydrolyze carbapenems to some degree Endemic in Asia, Europe

Adapted from Queenan & Bush. 2007. Clin Microbiol Rev. 20:440. Bush & Jacoby. 2010. AAC. 54:969; Bush, K. 2013. Ann NY Acad Sci 1277:84.

Inhibited by EDTA Do not hydrolyze aztreonam Endemic in India / Asia

9

KPC (Klebsiella pneumoniae Carbapenemase) • First report 1997 North Carolina • High level of enzyme typically produced • Mostly K. pneumoniae, but also other GNR • Plasmid with KPC gene generally has other R genes including ESBLs

Metallo β-Lactamase (MBL) Type of Carbapenemase • Zinc required for activity • NDM (New Delhi MBL) most common worldwide; • India and Pakistan • First report 2008 in Swedish patient hospitalized in India

• VIM (Verona integron-encoded MBL) • Europe, SE Asia

• IMP (Imipenemase) • Europe

NDM

72 NDM in IL ERCP outbreak1

IMP

VIM

6 IMP in CA Stanford outbreak2

8 VIM in KY NICU outbreak3 1. 2. 3.

MMWR 2014 62(51);1051-1051 JCM 2011 49: 4239 3. MMWR 2016. 65(7);190

OXA Carbapenemases • First described in Acinetobacter baumannii in 1985 • OXA-48 • • • •

Commonly found in Europe and Africa; rare in USA First report 2008 in Turkey Mostly K. pneumoniae, E. coli Variants include OXA-181 and OXA-232 “OXA-48-like”

• Weakly hydrolyze carbapenems & cephalosporins (need porin defect and presence of ESBL for full CRE phenotype)

OXA-48-like Carbapenemases

Resistance is complex… • BEST way to detect resistance is to use current CLSI/ FDA breakpoints NO FDA enforcement mechanism to ensure commercial AST systems update breakpoints. As of January 2017:

Phoenix (BD) has clearance on 4/4 updated breakpoints MicroScan (Beckman Coulter) has clearance for 2 / 4 updated breakpoints Vitek 2 (BioMerieux) has clearance on 0 / 4 updated breakpoints

Impact of delay…

Computer simulation data for Orange County, CA 5 year delay in implementing breakpoints = 8, 500 more CRE carriers in one county

Further confusion…. CPE vs. CRE? CDC Surveillance Definition

Enterobacteriaceae resistant to: • ertapenem, doripenem, meropenem, or imipenem* OR… • Positive for carbapenemase (by MHT, CarbaNP, PCR)

UCLA Clinical Definition

Enterobacteriaceae that are: • Not susceptible (“I or R”) to one of the following: doripenem, meropenem, or imipenem* Remember! Not all CRE have a “carbapenemase” *Proteus/Providencia/Morganella exceptions for imipenem

CDC Definition Goal: best possible sensitivity for carbapenemase Study: 307 Enterobacteriaceae with elevated (I or R) carbapenem MICs collected from EIN sites Not – S to Imipenem or Meropenem

Not-S to imipenem or meropenem & R to 3rd gen cephs (Old CDC Defn)

R to Ertapenem or Meropenem or Imipenem (New CDC Defn)

New with MHT

False positives

38.1%

26.7%

55%

12%

False negatives

3.9%

4.9%

0.7%

0.7%

EID 2015 21 (9)

67 KPC - 13 imipenem & meropenem “S” but erta-R

Use of a carbapenemase test CLSI Methods

FDA Cleared

Modified Hodge Test

CarbaNP

mCIM

Xpert CarbaR

Easy

Difficult

Easy

Easy

Good for KPC K. pneumoniae

Good for KPC, MBL

Good for almost all

Good for all tested

Poor for NDM

Poor for OXA-48-like

?

No sensitivity issues

UCLA Data: Use of Carbapenemase Test to confirm CPE CDC Definition CDC Definition + mCIM CDC Definition + Xpert

Sensitivity 98.9% 95.6% 98.9%

Specificity 11.7% 100% 100%

UCLA protocol: - Identify CRE by UCLA definition (some KPC are “S” to erta) - Perform Xpert Carba-R - Knowledge of carbapenemase is very useful for epidemiology!

Now you’ve found it… what else? CLSI, M100S 27th Edition Instructions for Use: “…each laboratory should develop a protocol to address isolates that are confirmed as resistant to all agents on their routine test panels. This protocol should include options for testing additional agents in-house or sending the isolate to a reference laboratory.” CAP- MIC.21944: “There are written policies for testing supplemental agents when needed on isolates resistant to routinely tested antimicrobial agents.”

Supplemental Drugs to Consider for AST of MDR GNR Antimicrobial Agent Colistin (or polymyxin B) Fosfomycin Minocycline

Tigecycline2 Ceftazidime-avibactam Ceftolozane –tazobactam

Enterobacteriaceae Yes1 (ex. Pro/Prov/Morg/Ser) Urine (E. coli BPs) Yes Yes (ex. Pro/Prov/Morg) Yes Yes

IR, intrinsic resistance; BP, breakpoints 1 No CLSI breakpoints; ECV only 2 Not on urine isolates

P. aeruginosa

A. baumannii

Yes

Yes

No BPs IR

IR Yes

IR

No BPs

Yes Yes

No BPs No BPs

Supplemental Drugs to Consider for AST of MDR GNR

Colistin (polymyxin B) Fosfomycin Minocycline Tigecycline Ceftazidime-avibactam Ceftolozane-tazobactam

Automated AST?

Disk?

Gradient Strip?

Sensititre

No

No!

RUO, poor

RUO

No Some Yes No No

Yes Yes Yes Yes Yes

RUO Yes Yes RUO Liofilchem

No Yes Yes Yes Yes

Colistin/Polymyxin B: CLSI M100 S27 MIC (µg/ml)

Organism Acinetobacter spp.

Zone (mm)

Susc ≤2

Int -

Res 4

≤2

-

4

Pseudomonas aeruginosa Enterobacteriaceae

Organism Enterobacteriaceae

Susc

Int none

none Insufficient clinical and PK/PD data to set “breakpoint”

ECV (µg/ml) WT ≤2

NOT a NWT clinical ≥4 breakpoint!!

♦ Deleted breakpoints for “non-Enterobacteriaceae” CLSI M100 27th ed.

Res

Commercially Available Tests for Colistin Test Method

Manufacturer

Regulatory Status

Notes

Disk

BD

RUO

Not recommended by CLSI/EUCAST

Gradient Diffusion

bioMerieux, Liofilchem

RUO

Not recommended by CLSI/EUCAST

Sensititre Broth Microdilution

ThermoFisher

RUO

Broth microdilution Custom panel

RUO for AST

Some prelim data to suggest may work

MicroScan colistin well on Beckman Coulter dried Gram negative ID panel

Options (for now) that work: Sensititre • RUO GNX2F & GNX3F panels • 0.25 – 4.0 μg/ml colistin & polymyxin B wells • One study2: 3/10 VME (30%) and 2/40 ME (5%) vs. BMD • All 3 VME were MIC of 2 μg/mL by Sensititre, 4 μg/ml by BMD • ME were MIC of 4μg/mL by Sensititre, 4 μg/ml by BMD

• Prudent practice: use caution if MIC is 2 – 4 μg/ml

1 Sader 2015 DMID 83: 379 2 Hindler 2013 JCM 15:1678

mcr-1 MIC region!

Options that work: MicroScan Colistin Well • NOT FDA-cleared for this indication • 4 μg/mL colistin well present on dried Gram Negative panel Modification of an FDA-cleared test (RUO?) Landman et al 2007 IDSA Annual Meeting • MUST do a verification study! Comparator: BMD (Tween?) • Disclaimers to results? Panel: MicroScan NBPC 30 • How aboutA.QC? Isolates: baumannii (n=10) N=14 colistin P. aeruginosa (n=10) MIC >4 μg/mL K. pneumoniae (n=14) E. cloacae (n=6) Results: 40/40 (100%) correlated with BMD

Do we need to routinely test Colistin? • Resistance is not common among P. aeruginosa, but can occur • Resistance more common among A. baumannii and CRE UCLA Data, 2016 Organism

N

%S

Enterobacteriaceae1

7117

96.7

Meropenem-NS Enterobacteriaceae1

48

66

A. baumannii

150

98.6

P. aeruginosa

59

94.9

1

excludes species intrinsically R to colistin

Colistin When/how to test? • Scenario when testing might be requested: • Carbapenem-resistant Enterobacteriaceae • MDR A. baumannii, P. aeruginosa

• No FDA cleared AST: ˗ ˗ ˗ ˗

Etest  RUO, poor performance Disk  RUO, poor performance Sensititre  OK Agar screen  Under evaluation @ CLSI

29

Why no FDA cleared tests for Colistin? • No FDA breakpoint (colistin) • No sponsor for colistin or polymyxin B (generic) But – good news! - 21st Century Cures Act: passed in 2016! - Includes language to remove breakpoints from drug label - FDA to recognize breakpoints set by other agencies (e.g., CLSI) - Device manufacturers can get clearance for devices with these BPs - … hope for near future!

Current timelines for new drug AST development/clearance Time to clearance Vitek 2 Phoenix MicroScan Etest Trek

Note: time to market may be longer

BD DISK Hardy Disk 0

5

10

15

20

Data derived from FDA website clearance dates for drug X

25 Months

30

35

40

45

50

Contrast with abx development

# approvals

Antibiotics Approved by FDA for Human Use 18 16 14 12 10 8 6 4 2 0

No new antibiotic classes since 1987

Generating Antibiotics Now (GAIN) - For Qualified Infectious Diseases Products - Extra years exclusivity - Fast track through FDA - Priority review by FDA

May need similar incentives for AST manufacturers Bottom line: developing new ASTs is not a good business decision 83-'87 88-'92 93-'97 98-'02 03-'07 08-'12 13-'16

In discussion @ CLSI, FDA, CDC, IDSA

Fosfomycin • Oral “sachet” formulation • Clinical trials underway for IV formulation • Excellent activity vs. MDR Gram negative infections

Acinetobacter

• Only FDA cleared for E. coli and E. faecalis • CLSI breakpoints only for these two organisms Organism E. coli

MIC (µg/ml) Zone (mm) Susc Int Res Susc Int Res ≤64 128 256 16 13-15 ≤12

- Testing requires glucose-6-phosphate (Enterobacteriaceae) - CLSI endorses disk diffusion or agar dilution ONLY for fosfomycin

Fosfomycin: AST considerations Fosfomycin MIC Distributions 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 0

E. coli ECV

“S” P. aeruginosa ECV

CLSI E. coli

≤ 64

EUCAST Enterobacteriaceae (IV and oral)

≤ 32

P. aeruginosa

none*

* EUCAST: use in combo with other abx if MIC ≤128

P. aeruginosa Data from EUCAST

Breakpoints?

E. coli

K. pneumoniae

Fosfomycin: AST considerations (EUCAST) • EUCAST recommends MIC method for fosfomycin, except E. coli

Ignore colonies in zone for E. coli (except panel D)

Why? Mutation frequency for fosfomycin is low (1:104)  more cells in 0.5 McFarland used for disk diffusion vs. agar dilution

What about P. aeruginosa ?

• inoculum issues with P. aeruginosa by diffusion methods (Etest and disk) • Use of 0.5 McFarland resulted in 22% ME •  suggest use of 1:100 dilution of inoculum • BUT, still saw ~30% VME

Fosfomycin When/how to test? • Scenario when testing might be requested: • Carbapenem-resistant Enterobacteriaceae in urine • MDR P. aeruginosa in urin

• Only FDA cleared AST: ˗ Disks  only for E. coli! ˗ Etest  RUO currently. ˗ Poor performance for P. aeruginosa ˗ Unknown performance for other species ˗ * at this point, testing other species may not generate useful data. DISCLAIM results! 37

Minocycline • PO, IV formulation • FDA approved for Acinetobacter infections • Considerably more active than doxycycline and tetracycline against A. baumannii • 67% of MDR isolates are “S”1

• Recent studies demonstrating clinical efficacy • Bishburg et al. 2014. Infect Dis Clin Pract. 22:26. • Jankowski et al. 2012. Infect Dis Clin Pract. 20:184.

• Option for multidrug resistant strains 1Hawser

et al. 2013. ICAAC. Abstract #C2-1625.

Pseudomoanas a aeruginosa

Acinetobacter baumannii Minocycline vs. Tetracycline (n=5478) 1

Minocycline

Tetracycline Susceptible (≤4 µg/ml) Resistant (>8 µg/ml)

Intermediate (8 µg/ml)

Resistant (>8 µg/ml)

0 (0%)

Intermediate (8 µg/ml) Susceptible (≤4 µg/ml)

1 (