Determination of Disk Diffusion and MIC Quality Control Guidelines for GSK2140944, a Novel Bacterial Type II Topoisomerase Inhibitor Antimicrobial Agent

Abstract

GSK2140944 is a novel bacterial type II topoisomerase inhibitor in development for the treatment of conventional and biothreat pathogens, including Gram-positive pathogens and methicillin-resistant Staphylococcus aureus. This quality control study was performed to establish ranges for selected control strains: S. aureus ATCC 29213 and ATCC 25923, Escherichia coli ATCC 25922, Haemophilus influenzae ATCC 49247, and Streptococcus pneumoniae ATCC 49619. The control ranges will be crucial for the accurate evaluation of GSK2140944 potency as it progresses through clinical trial development.

TEXT

GSK2140944 (Fig. 1) is a bacterial type II topoisomerase inhibitor that has demonstrated activity against contemporary Gram-positive pathogens (including methicillin-resistant Staphylococcus aureus [MRSA]), especially isolates resistant to the existing classes of antimicrobials (quinolones and glycopeptides) (1, 2). GSK2140944 selectively inhibits bacterial DNA replication by interacting with the GyrA subunit of bacterial DNA gyrase and the ParC subunit of bacterial topoisomerase IV. This highly specific interaction with the bacterial topoisomerases is evidenced by weak inhibition of human topoisomerase II, supporting the selective activity of GSK2140944 against the bacterial targets. A Clinical and Laboratory Standards Institute (CLSI) M23-style quality control (QC) study was performed to establish disk diffusion and broth microdilution QC ranges for five bacterial strains to assist clinical laboratories in monitoring the activity of this compound during ongoing clinical trials (3–6).

FIG 1.

Chemical structure of GSK2140944.

Eight laboratories (seven are required in the CLSI M23-A3 guidelines) were used in each study, with seven sites being used in both studies, to establish a QC range (3). These laboratories were experienced microbiology facilities, and each followed the CLSI procedures for disk diffusion (4) and broth microdilution (5) methods. The sites that participated (and study directors) were Wheaton Franciscan Laboratory, Wauwatosa, WI (E. Munson), JMI Laboratories, North Liberty, IA (R. N. Jones), Thermo Fisher Scientific, Cleveland, OH (C. Knapp), University of Alberta, Edmonton, AL, Canada (R. Rennie), University of Washington, Seattle, WA (S. Swanzy), Robert Wood Johnson Medical School, New Brunswick, NJ (M. Weinstein), the Cleveland Clinic Foundation, Cleveland, OH (G. Hall/G. Procop), University of Texas Medical Center, Houston, TX (A. Wanger, disk diffusion study only), and Duke University Medical Center, Durham, NC (S. Mirrett, MIC study only).

The reference frozen-form broth microdilution panels were prepared by Thermo Fisher Scientific (Cleveland, OH, USA), according to good manufacturing practice (GMP) guidelines, and shipped frozen to all participating sites. The panels contained four lots of cation-adjusted Mueller-Hinton broth (Oxoid, Hampshire, United Kingdom; BBL, Sparks, MD, USA; and Difco [2 lots], Detroit, MI, USA). Also, panels containing four lots of Haemophilus test medium (HTM) and four lots of Mueller-Hinton broth supplemented with 2.5 to 5% lysed horse blood were provided by the same three vendors listed above. Azithromycin, levofloxacin, and linezolid were utilized as the control agents (6). For the broth microdilution testing, each laboratory tested 10 replicates of S. aureus ATCC 29213, Escherichia coli ATCC 25922, Haemophilus influenzae ATCC 49247, and Streptococcus pneumoniae ATCC 49619. The colony counts of the inoculum were performed on drug-free agar medium and resulted in the following average counts by QC organism: 3.2 × 10⁵ CFU/ml for S. aureus ATCC 29213, 3.6 × 10⁵ CFU/ml for E. coli ATCC 25922, 5.0 × 10⁵ CFU/ml for H. influenzae ATCC 49247, and 2.7 × 10⁵ CFU/ml for S. pneumoniae ATCC 49619.

For the disk diffusion tests, two different lots of 10-μg GSK2140944 disks were manufactured by two companies, lot 309972 (MAST Group, Merseyside, United Kingdom) and lot 2340189 (BD, Franklin Lakes, NJ, USA). Single lots of comparator disks from BD were used: azithromycin 15 μg (lot 2304320), levofloxacin 5 μg (lot 2272187), and linezolid 30 μg (lot 2291112). Three manufacturers (Remel, Lenexa, KS, Hardy Diagnostics, Santa Maria, CA, and BBL) were used to produce lots of Mueller-Hinton agar (lots 302846, 13042, and 2356035, respectively) and Mueller-Hinton agar with 5% sheep blood (lots 304123, H12361, and 3024248, respectively). For the disk diffusion tests, each laboratory tested 10 replicates of S. aureus ATCC 25923, E. coli ATCC 25922, and S. pneumoniae ATCC 49619.

The GSK2140944 broth microdilution MIC QC results are summarized as CLSI-approved ranges in Table 1. All S. aureus ATCC 29213 MIC results were included in the approved 0.12- to 1-μg/ml range (Fig. 2A). According to the CLSI M23-A3 document (3), a “shoulder” off the modal value, which is ≥60% of the data points at the mode, indicates that the population is considered to have spread across 2 dilutions, and the QC range is extended to 4 dilutions. The population of MICs at 0.25 μg/ml for S. aureus ATCC 29213 represents 61.7% of the modal MICs at 0.5 μg/ml. Therefore, a 4-dilution range is warranted in this case. All E. coli ATCC 25922 results were within the approved range of 1 to 4 μg/ml with a clear mode at 2 μg/ml (255 of 320, 79.7% of results) (Fig. 2B). For H. influenzae ATCC 49247 and S. pneumoniae ATCC 49619, all GSK2140944 MIC results were within the approved 3 doubling dilution ranges (Table 1; Fig. 2C and D).

TABLE 1.

CLSI approved quality control ranges of GSK2140944 broth microdilution and disk diffusion susceptibility tests

QC organism	Broth microdilution results		Disk diffusion zone diameter results:
	Approved MIC range (μg/ml)	% of results in approved range	Approved zone range (mm)	% of results in approved range
S. aureus ATCC 29213	0.12–1	100.0	NAa	NA
S. aureus ATCC 25923	NA	NA	23–29 (24–29)b	96.4 (96.4)
E. coli ATCC 25922	1–4	100.0	18–26 (17–27)	98.7 (99.6)
H. influenzae ATCC 49247	0.25–1	100.0	NA	NA
S. pneumoniae ATCC 49619	0.06–0.25	100.0	22–28	99.4

^aNA, not applicable.

^bRange Finder results, if different, are in parentheses (7).

FIG 2.

CLSI-approved GSK2140944 MIC quality control ranges for S. aureus ATCC 29213 (A), E. coli ATCC 25922 (B), H. influenzae ATCC 49247 (C), and S. pneumoniae ATCC 49619 (D).

All broth medium lots and interlaboratory modal values were within 1 MIC dilution step with no skewing observed, regardless of the QC strain tested. Only one value among the 1,120 control results was outside the CLSI published range (6). That single levofloxacin result for S. pneumoniae ATCC 49619 was out of range; however, the remaining 99.9% of the control results provided valid internal quality assurance for this portion of the study. All other comparator control values were within the published ranges (6).

Using the CLSI document M23-A3 guideline (3) for establishing QC ranges, we also performed a disk diffusion study to determine zone diameter ranges for GSK2140944, and those results are summarized in Table 1. The zone diameters reported by eight laboratories for E. coli ATCC 25922 are shown in Fig. 2A. Using the M23-A3 criteria (3), a 7-mm range of 19 to 25 mm included only 91.3% of the reported zone diameters. Following the CLSI guidelines to include >95% of all reported values, we added 1 mm to the range (19 to 26 mm), resulting in inclusion of 95.6% of the data, which was considered acceptable. The Range Finder statistical program, an additional tool to analyze the data using the mean, median, and mode to determine outliers (7), suggested a range of 17 to 27 mm (99.6% of the results included). The CLSI antimicrobial susceptibility committee (January 2014) approved a QC range of 18 to 26 mm (98.7% of the results included). The zone diameters reported for S. aureus ATCC 25923 by the eight laboratories produced a QC range of 23 to 29 mm (7-mm range), which included 96.4% of these results (Fig. 3B). The Range Finder program recommended a range of 24 to 29 mm which included the same percentage of results but identified laboratory G as an outlier laboratory with a median zone diameter statistically separated from those of all other laboratories. Deleting laboratory G and recalculating the range without the outlier laboratory data resulted in the same range, but included 99.3% of all results for the S. aureus QC strain.

FIG 3.

CLSI-approved GSK2140944 disk diffusion quality control zone diameter ranges for E. coli ATCC 25922 (A), S. aureus ATCC 25923 (B), and S. pneumoniae ATCC 49619 (excluding laboratory C) (C).

H. influenzae ATCC 49247 was tested by disk diffusion, but variations in the medium prevented a QC range from being established. One medium manufacturer contributed to modestly larger zones than the other manufacturers. This may be an outlier lot from this medium producer, but that remains to be determined, pending further investigations to compare lots from that manufacturer. A range of 22 to 28 mm was approved for S. pneumoniae ATCC 49619 and included 99.4% of all zone diameters. However, the Range Finder program identified laboratory C as an outlier site that could be deleted from the analysis (Fig. 3C). The range without laboratory C was 22 to 28 mm, which included all reported zone diameters.

The control disks (azithromycin, levofloxacin, and linezolid) provided valid internal assurance for this study with 99.5% of the reported participant zones found within the CLSI published QC ranges (6). There was minimal variance in the results between the two lots of GSK2140944 disks (≤1-mm difference in median values) for the three QC organisms tested.

The results (Table 1) from these two multilaboratory investigations provide initial GSK2140944 QC ranges for routine susceptibility testing using disk diffusion and broth microdilution methods (4, 5), as this new bacterial type II topoisomerase inhibitor (GSK2140944) is being developed as an oral and intravenous treatment for community-acquired bacterial pneumonia (CABP) and acute bacterial skin and skin structure infections (ABSSSI) caused by conventional and biothreat pathogens such as Yersinia pestis, Bacillus anthracis and Francisella tularensis. The CLSI Subcommittee on Antimicrobial Susceptibility Testing approved these MIC QC ranges in January 2012 and disk diffusion ranges in January 2014 for publication after the selection of the compound's chemical name.