J01: Antibacterials for systemic use
Time-kill in vitro
in vivo models
- Mouse thigh/lung
- Neutropenic or immuno-competent
- Dose fractionation
- 1 CFU per animal (homogenization)
- Mouse thigh/lung
Nielsen-model (susceptible–resistant strains)
ABX C_T should be high enough.
Antibacterials are drugs used to treat bacterial infections. We often use the words antimicrobials, antibacterials, and antibiotics for these compounds. They are similar terms, but each has a specific meaning:
- Antimicrobials target all kinds of microorganisms, such as bacteria, fungi, viruses, and protozoa.
- Antibacterials only act on bacteria. They include both antibiotics and some common disinfectants.
- Antibiotics are naturally produced by microorganisms and kill or inhibit other microorganisms, mainly bacteria.
The term “antibiotic” comes from the Greek words anti (“against”) and biotikos (“concerning life”). Strictly speaking, antibiotics refers only to naturally produced substances, and not synthetic or semi-synthetic types (e.g., quinolones). However, it is commonly used interchangeably with antibacterials.
How do antibacterials work?
Antibacterials interfere with vital processes in bacterial cells.
- Bactericidal antibacterials kill bacteria.
- Bacteriostatic antibacterials prevent bacteria from multiplying.
The host’s immune system then clears the remaining infecting pathogens.
Antibiotic targets in bacteria
Different classes of antibacterials act on different parts of bacterial cells. They mainly target:
- The cell wall or membranes surrounding the cell.
- The DNA and RNA production machinery.
- The protein production machinery (ribosomes and related proteins).
These structures do not exist in human or animal cells, so antibacterials typically do not harm our cells. However, side effects can still occur.
Narrow-spectrum and broad-spectrum antibacterials
Some antibacterials have a narrow spectrum, meaning they act on only certain groups of bacteria. Others are broad spectrum and affect many bacterial species. Narrow-spectrum antibacterials are often preferred since they leave non-disease causing bacteria alone. However, broad-spectrum antibacterials are sometimes used when doctors are unsure of the exact bacteria causing the infection.
Resistance
Antibiotic resistance is a bacterias ability to survive the effects of an antibiotic. In the presence of the antibiotic, resistant bacteria live and multiply. Clinically, resistance means the bacterium can grow at drug exposures reached during standard treatment, leading to treatment failure.
Bacteria gain resistance through:
- Mutations in their own DNA.
- Horizontal gene transfer, where they acquire DNA from other bacteria.
If the new trait offers an advantage, the bacterium may pass it on to future generations or spread it through horizontal transfer. Resistant bacteria travel through food, water, global trade, or human travel.
AMR terminology:
- MDR: Resistant to ≥1 agent in ≥3 classes
- XDR: Resistant to ≥1 agent in all but 2 classes
- PDR: Resistant to all agents in all classes
Resistance measurement: MIC
Translating antibiotic PKPD
Preclinical antibiotic studies provide essential PKPD information to help select effective clinical dosing regimens. There are two main approaches:
Index-based approaches
The classical method links unbound antibiotic PK to the MIC. Depending on the antimicrobial class, three relevant indices are traditionally described:
- fAUC/MIC: Exposure-dependent effect.
- fCmax/MIC: Concentration-dependent effect.
- fT>MIC: Time above MIC. Time-dependent effect.
Decided by plotting log-kill against AUC, and fitting a nonlinear (sigmoid) regression. Highest R^2 decides the index.
R^2 is used by convention, even though it’s not suitable for nonlinear regressions.
These indices help set efficacy targets based on outcomes like bacterial stasis or log reductions in bacterial count. Dose fractionation studies identify which index best predicts efficacy, and simulations then predict outcomes in diverse patient populations. The Probability of Target Attainment (PTA) guides clinical dose selection and determines susceptibility breakpoints.
However, limitations of PK/PD indices include:
- Simplification using a single metric and MIC
- Ignoring bacterial dynamics (rate of killing, regrowth, resistance)
- Limited applicability to antibiotic combinations
Model-based approaches
Model-informed Drug Development (MIDD) incorporates longitudinal data, capturing bacterial dynamics better than indices alone. PKPD models developed from in vitro experiments predict in vivo outcomes and clinical efficacy. Steps include:
- Building initial models from in vitro data (static time-kill studies)
- Refining models with dynamic in vitro and in vivo (mouse) studies
- Integrating human PK data for simulations to predict clinical outcomes
This approach considers variability in bacterial dynamics across strains, doses, and patient populations.
Impact of the immune system
The immune response significantly influences antibiotic efficacy. Most antibiotic studies occur in neutropenic models, isolating antibiotic effects but overlooking immune interactions. Immune variability among patients impacts real-world efficacy. Incorporating immune response data into PKPD models may enhance the translation of preclinical findings to clinical practice, especially in patients with varying degrees of immunocompetence.
Modeling notes
The index is class-specific. For antibacterials, exposure is summarized as unbound concentration referenced to the MIC: fT>MIC, fAUC/MIC or fCmax/MIC [1]. Which index predicts effect is settled by dose fractionation and differs between antibacterial classes, so no single index covers J01 [2,3].
Typical structure. Semi-mechanistic models give the bacteria a growth rate, a capacity limit and a drug-driven kill rate, fitted to whole time-kill curves rather than one summary index [4].
Resistance is a compartment. Regrowth is usually modeled as a pre-existing less susceptible subpopulation that treatment selects for, though published models differ in how resistance arises [3,5].