Studies To Establish Quality Control Ranges for SB-265805 (LB20304) When Using National Committee for Clinical Laboratory Standards Antimicrobial Susceptibility Test Methods

SB-265805, formerly LB20304, is a novel naphthridone compound in the fluoroquinolone class of antimicrobial agents. This antimicrobial possesses excellent activity against gram-positive cocci, with a potency 32- to 64-fold greater than that of ciprofloxacin, especially when tested against oxacillin-resistant staphylococci and penicillin-resistant Streptococcus pneumoniae (6). Results have been described for the Enterobacteriaceae whereby the MIC at which 50% of the isolates were inhibited (MIC₅₀) was 0.03 μg/ml, activity twofold less than that of ciprofloxacin (1). SB-265805 was also observed to be very active against Pseudomonas aeruginosa (MIC₅₀s, 0.25 to 1 μg/ml), Stenotrophomonas maltophilia (MIC₅₀s, 0.5 to 1 μg/ml), anaerobes (MIC₅₀s, 0.12 to 8 μg/ml), vancomycin-resistant enterococci (MIC₅₀s, 0.12 to 2 μg/ml), Haemophilus influenzae (MIC₅₀s, ≤0.004 μg/ml), Moraxella catarrhalis (MIC₅₀s, ≤0.004 to 0.015 μg/ml), and Neisseria gonorrhoeae (MIC₅₀s, 0.008 to 0.015 μg/ml (1, 6).

SB-265805 appears to be a promising agent that should be rapidly explored in clinical trials. To facilitate the accurate testing of SB-265805, quality control (QC) parameters will be necessary for reference broth microdilution and standardized disk diffusion methods (3, 4). Trials to establish QC ranges have rigid designs recommended by the National Committee for Clinical Laboratory Standards (NCCLS) in the previously approved and the more recent tentative M23 documents (2, 5). These study guidelines (2, 5) have been applied to SB-265805 QC protocols initiated in late 1997 and again in early 1998. Our experience with SB-265805 MICs derived from two MIC tray lots produced by commercial manufacturers (AccuMed International, Inc., Westlake, Ohio, and PML Microbiologics, Wilsonville, Oreg.) was also addressed.

The QC strains tested (seven) are listed in Table 1, and each nonfastidious strain was tested in two MIC QC trials using the differing NCCLS study designs (2, 5). The principal differences between the trials were in the number of participating centers and the number of medium lots tested; e.g., the 1997 trial used six laboratories (≥5 participants required) and six medium lots (one common lot tested on all) (2), and the 1998 trial used nine laboratories (≥7 participants required and ≥3 medium or base lots were tested by all laboratories) (5). Disk diffusion tests and MIC studies with H. influenzae ATCC 49247 and S. pneumoniae ATCC 49619 were conducted by utilizing the most recent protocol design (5).

TABLE 1.

Proposed ranges for MIC and disk diffusion tests with SB-265805 (formerly LB20304)

QC strain	MIC QC results		Disk diffusion QC results
	Range (μg/ml)	% Results in range	Zone diam range (mm)	% Results in range
E. coli ATCC 25922	0.004–0.016	99.6	30–36	96.5
	(0.004–0.016)a	100.0a
E. faecalis ATCC 29212	0.016–0.12	99.6	NTb	NTb
	(0.016–0.12)a	100.0a
P. aeruginosa ATCC 27853	0.25–1	100.0	19–25	97.0
	(0.25–1)a	100.0a
S. aureus ATCC 29213/25923c	0.008–0.03	100.0	27–33	99.4
	(0.008–0.03)a	99.2a
H. influenzae ATCC 49247	0.002–0.008	100.0	31–37	95.4
S. pneumoniae ATCC 49619	0.008–0.03	100.0	28–34	97.3

^aRange proposed from the ≥5 laboratory study design (NCCLS approved guideline (M23-A, 1994 [4]). 

^bNT, not tested. 

^cATCC 29213 was used for MIC testing, and ATCC 25923 was used for disk testing. 

Table 1 lists the proposed QC ranges from the study with six laboratories and the study with nine laboratories. On two occasions among 15 organism-drug study sets, a single participant’s results were declared nonvalid because of significant variation (results >2 log₂ dilutions from the all-participant mode) from peer data. The MIC ranges for QC established were the modal MIC ± 1 log₂ dilution in all cases except for Enterococcus faecalis (range, 0.016 to 0.12 μg/ml). For E. faecalis the mode was broad (equal occurrences at 2 log₂ dilutions), therefore necessitating a 4-dilution range (2). More than 99% of all qualifying laboratory results were within the uniform 7-mm ranges suggested for each control strian. Lastly, SB-265805 MIC QC ranges generated by the two protocol NCCLS designs (2, 5) were identical.

Two commercially prepared broth microdilution tray lots were prepared, and 16 to 105 replicate QC tests were determined for each lot by using four QC strains (Escherichia coli ATCC 25922, Staphylococcus aureus ATCC 29213, P. aeruginosa ATCC 27853, and E. faecalis ATCC 29212). SB-265805 MICs with the two gram-negative strains were approximately twofold lower on commercial lots than on the NCCLS trial lots (2, 5), but the number of determinations falling outside the proposed control ranges was only 1 in 413 total tests (0.2%; data not shown). The results with the other QC strains were identical for the commercial system and reference lots. These data appear to indicate the validity of the proposed QC ranges for SB-265805 antimicrobial susceptibility testing, and their favorable application to clinical trials, possibly using commercially prepared reagents. They suggest that the two NCCLS QC study designs (2, 5) for establishing MIC ranges functioned equally well in the case of this new fluoroquinolone.