Clinical and Laboratory Standards Institute guideline M23—Development of In Vitro Susceptibility Test Methods, Breakpoints, and Quality Control Parameters offers guidance for developing breakpoints and QC ranges for antimicrobial susceptibility tests against aerobic and anaerobic bacteria, as well as selected fungi, according to CLSI antimicrobial susceptibility testing documents. It describes the data used by the CLSI Subcommittees on Antimicrobial Susceptibility Testing and Antifungal Susceptibility Tests to establish these breakpoints and QC ranges for antimicrobial agents, including microbiological data, pharmacokinetic and pharmacodynamic characteristics, and clinical data. As additional experience is accrued with the individual antimicrobial agents, new data may be used to reassess breakpoints or QC ranges. Users of this guideline should understand that susceptibility test results cannot predict clinical outcomes with absolute certainty. They should be used along with best clinical judgment and laboratory support to best serve the patient.

This guideline replaces the previous edition of the approved guideline, M23-Ed5, published in 2018. Several changes were made in this edition, including: 

• Updating pharmacokinetic/pharmacodynamic (PK/PD)–related definitions (see Subchapter 1.3.1) 

• Replacing Subchapter 2.3 (Developing Disks for Disk Diffusion Tests) with new Subchapter 2.2 (Assessing Variability of the Reference Method) to provide rationale for, factors to be assessed in, and suggested design of studies to assess reproducibility of reference broth microdilution and disk diffusion assays – Subchapter 2.2.2 cites CLSI document M23S12 to provide guidance on evaluating and selecting appropriate content (potency) of antimicrobial agents for the development of disks for disk diffusion assays. 

• Clarifying the importance of the four cutoffs used to establish clinically relevant breakpoints (see Chapter 5 and Table 6) 

• Clarifying the methods and guidance on targets, number of isolates to test, etc. (see Subchapter 5.2) – Clarifying language for static experiments (see Subchapter 5.2.1.1) – Adding significant detail for dynamic experiments (eg, chemostat and hollow-fiber model studies) (see Subchapter 5.2.1.2) – Adding significant detail on in vivo systems (see Subchapter 5.2.2), including: o Study design o Need for high-quality pharmacokinetic (PK) data o Information on PK at infection sites (eg, epithelial lining fluid [ELF]) o Mathematical modeling o Information on the numbers and types of isolates needed for studies o Information on the selection of relevant end points to determine PK/PD target o Information on assessment of pharmacodynamic index and PK/PD for efficacy of ß-lactam combination agents – Including information on free-drug concentrations in ELF (see Subchapters 5.2.2 and 5.2.3) – For Monte Carlo simulations, adding a statement about inclusion of an assessment of PK/PD target variability and suggestions regarding the potential use of weighted mean or random assignment following a probability distribution (see Subchapter 5.2.4.1) – Providing a new table (see Table 7) and figure (see Figure 3) on how to present target attainment data (see Subchapter 5.2.4.2) 

• Clarifying clinical exposure–response cutoff (see Subchapter 5.3) 

• Adding text regarding presentation of outcomes, including discordance between clinical and microbiological outcomes, and evaluating minimal inhibitory concentration increases between baseline and postbaseline isolates (see Subchapter 5.4.2) 

• Adding a sponsor checklist for CLSI breakpoint proposals (see Appendix A) 

• Adding new Appendixes C (Example of a Reference Broth Microdilution Reproducibility Study) and D (Example of a Disk Diffusion Reproducibility Study), which describe the suggested design of studies to assess reproducibility of reference broth microdilution and disk diffusion assays, respectively

This guideline provides recommendations for determining breakpoints and QC parameters for antimicrobial agents that directly act on microorganisms. The intended audience includes sponsors (eg, antimicrobial agent manufacturers), CLSI volunteers, and various third-party groups planning to submit data to establish or revise QC ranges and antimicrobial susceptibility testing (AST) breakpoints and interpretive categories for inclusion in CLSI AST documents only. Appendix A provides a sponsor checklist for CLSI breakpoint proposals. The methods described do not apply to: 

• Antimicrobial agents formulated for direct administration to skin or mucous membranes, topical or intraocular injection, intrathecal injection, or inhalation 

• Antimicrobial agents that are intended to exert activity in the gut lumen In some instances, recommendations may be based on limited data (eg, some species of nontuberculous mycobacteria). Also, CLSI recognizes that nontraditional agents (eg, those that can have a beneficial effect in disease but do not directly inhibit bacterial growth and survival, such as virulence inhibitors and/or non–small molecule agents) are being explored and developed for use in infectious diseases. However, at this time, there is no clear path for establishing breakpoints that are not based on minimal inhibitory concentrations (MICs).