Beyond the Well-Stirred Assumption: Prospective Model Qualification for Phage PK/PD, antimicrobial therapy and mechanistic dosing

Beyond the Well-Stirred Assumption reveals one of the main challenges of prospective qualification of phage PK/PD models: local tissue phage amplification, immune sequestration, biofilm, and hidden tissue compartments are a few factors that determine therapeutic success in the well-stirred assumptions. The more successful development strategy is based upon a measurable mechanism and not just an impressive assay result. That change equates chemistry, biology, exposure, resistance, safety, and production as part of one argument of translation.

The scientific basis is based on a process that involves a planned decision, prediction of data prior to observation, and assessment of if there is an acceptable level of error for the planned decision. The mechanistic claim must be expressed in terms of each individual component, otherwise no isolated measurement could carry the entire mechanistic claim. The connected evidence lessens the risk of confusing potency, binding, performance in the formulation or model fit with understanding of the therapeutic activity.

The not unreasonable assumption is most vulnerable when bacteria are part of a biofilm, live inside necrotic cavities, mucus in airways, poorly perfused tissues, or within cells. There may be large regional differences in the concentration of phagytes in blood plasma and susceptible bacteria may be present in regions with resistant or dormant bacteria. Prospective qualification should thus require site-specific sampling or dynamic surrogates that will determine if any contact between phages and bacteria is predicted as opposed to systemic titer as a full exposure measure.

The design variables need to be treated together, as it is possible to modify performance of the variables of access to the vasculature, tissue architecture, spatial distribution of bacteria, diffusion of phages, adsorption, immune capture, dose timing, and combination of antibiotics simultaneously. When potency is optimized without regard to other factors, a hidden penalty may be paid in another dimension, particularly when potency is being optimized against stability, selectivity, delivery or manufacturability. Multiparameter decision rules are thus a more realistic criterion for ranking candidates than a single number.

It is also important to measure failure routes early to increase mechanistic confidence. The poor penetration, receptor heterogeneity, dormant bacteria, bio-film barriers, rapid resistance, immune clearance and inadequate local replication are important escape mechanisms. Deliberate stress testing transforms these routes into design information which can be used for the next chemistry cycle, formulation, dosing or more model cycles.

Exposure is the link between the nominal dose and the biological effect. Significant parameters are the phage burden at the site, the number of active bacteria in the site, the adsorption opportunity, the replication gain, clearance, and interval between doses. Strong intrinsic activity and weak exposure at the causal site may present a candidate with a great performance in simplified assays, but be unsuitable for translation.

Interpretation of direct engagement or functional perturbation must be done at realistic concentrations. High doses can lead to secondary stress responses, off-target effects or assay artifacts that simulate a desired mechanism. Concentration-response analysis, coupled with exposure and occupancy, can be used to pinpoint a range where the proposed causal process is thought to have an action.

In real tissues with heterogeneous structures, the context of biological factors also influences interpretation of the results of the use of phage therapy. Target dependence and effective exposure can vary between individuals due to differences in genotype, cell state, tissue architecture, immune tone, metabolic environment and previous treatment. Repeated testing in a single system is not as compelling as representative model panels.

The way to measure resistance and adaptation is to prospective, not retrospective. Which mechanisms survive continued pressure can be determined in a serial passage, long-term selection, genetic perturbation, repeated dosing or environmental stress. A candidate who is not quite as strong but easier to lose is not quite as valuable as a wimpy screening leader.

The combination of safety assessment with mechanism provides a more informative assessment. The following concerns are relevant: systemic inflammatory response, endotoxin burden, off-targets bacterial lysis, immunogenicity and treatment related cytokine changes. A comparison of desired from unwanted biological responses informs on the distinction between mechanism-related pharmacology and broad-based cellular injury and formulation-related toxicity.

Manufacturing cannot continue as a downstream logistics problem due to titer accuracy, infective fraction, aggregation, storage stability, host-range consistency, genomic integrity and batch purity affecting the product delivered to the biological system. Even if the name of the product is not changed, small process changes can affect the composition, physical state, impurity burden, the stability, and/or the dose reproducibility. Critical process attributes are directly connected to the mechanistic performance, which leads to development gains.

Spatial PK/PD models, Agent-based components, Compartmental systems, Bayesian calibration, Virtual populations and scenario qualification are additional tools that give leverage to the computational approach. There is most usefulness in prediction when there is ambiguity and applicability limits still present, particularly where there are new chemical structures, unusual biological states or where there is little data. Models that reveal uncertainty allow for improved experiment selection than models that report confident rankings without revealing areas of uncertainty.

To validate prospective, predictions need to be made in advance, datasets need to be blinded and must be analyzed at other laboratories, and sensitivity stress tests must be performed and decision-oriented error limits must be set. Orthogonal tests are important because if repeated, tests from the same assay family can give the same hidden bias which doesn't strengthen the causality. When two or more independent methods agree there is greater confidence in the link from design to biological effect; when they disagree, then the assumptions must be changed.

It is important that data integration maintains traceability from raw experiment down to a development decision. Chemical structures, assay conditions, assay metadata, exposure data, omics data and resistance data and safety signals must be identified and versioned. This discipline will help in subsequent interpretations from straying from the exact circumstances that led to a result.

Decision metrics also should be not only laboratory show, but also translation. A helpful scorecard is a set of the following attributes: Reproducibility, Activity per exposure, Mechanism of action, Resistance frequency, Safety margin, Stability, Manufacturing robustness. This will encourage balanced candidates and allow for better trade-offs to be seen before major development investments are made.

In this context, negative result is a special value. The edge of a hypothesis is determined by: Failed synthesis, missing target engagement, unstable exposure, absence of rescue, emergence of resistance, poor batch reproducibility, loss of activity in realistic models. Documenting failures helps to enhance the design of future experiments, and avoids reinvestments in appealing, but unsupported explanations.

The reproducibility of results between laboratories and batches is another necessary filter. When it appears the results are significant, a mechanism relying on a single operator, single preparation, single cell line, or single modeling option is inherently fragile. When conditions are repeated independently, and with exposure matched, it is then a matter of durable biology or local experimental success.

For translation, it is also important to have a realistic perspective on the clinical or deployment situation. An advanced product may be a great idea in theory, but may not be viable in real-world circumstances due to route, administrative procedure, diagnostics, storage and monitoring, cost, or treatment timing. Mechanistic excellence has impact when the product design is appropriate in the context in which biological benefit needs to happen.

Temporal design introduces another dimension since biological systems react differently to short pulses, exposure over a longer period of time, washout and repeated dosing. Time resolved measurements can help to distinguish primary engagement from downstream adaptation, and can show if there is any early activity that lasts after the initiating signal has subsided. The selection of sampling schedules should reflect the anticipated time course of potential qualification of a phage PK/PD model, instead of a single endpoint.

An integrated development record helps to ensure better communication between different areas of medicinal chemistry and formulation science, pharmacology, computation, manufacturing and clinical planning. Rather than a single definitive story, the record can highlight clear connections, possible explanations, areas of uncertainty, and new experiments with the greatest information content. With that transparency, iteration becomes quicker and there's less pressure to stick to initial assumptions in the face of conflicting data.

Novelty is held in check by reference to credible reference conditions. Whether a new approach is an improvement on prediction, mechanism, durability or manufacturability is determined by established compounds (known inactive controls), standard formulations, baseline models, or accepted analytical methods. Head-to-head comparison also allows a comparison of gains that are lost after normalizing the data for exposure, assay sensitivity, batch quality, or experimental complexity.

Mechanistic dosing is now suddenly a testable prediction discipline, thanks to prospective qualification. The criteria are not whether a candidate wins one assay, but whether proposed mechanism is coherent across all exposure, biological variation, resistance, safety and production. That standard puts in place a more "disciplined" way to go from discovery signal to a therapeutic concept that will be able to stand the complexity of the real world.