In preclinical oncology, a tumor study can show little benefit even when a candidate has strong activity in cultured cells. Before abandoning the compound, researchers need to establish what the result means. Did enough drug reach the tumor? Was the proposed target actually responsible for the cellular response? Did the model contain the immune components required for treatment to work?
These questions call for different experiments. This article proposes a practical way to connect in vitro absorption, distribution, metabolism and excretion (ADME) studies with genetic models and humanized tumor models. The focus is preclinical oncology programs evaluating small molecules, including combinations with immunotherapies. Antibodies and cell therapies require different disposition assays; a conventional small molecule ADME panel should not be applied to them unchanged.
Begin with the decision the experiment must support
Before selecting an assay for a preclinical oncology study, write down the uncertainty that could change the next development decision. For an oral compound, that might be whether intestinal absorption is plausible. For a combination regimen, it could be whether the partner drug changes exposure or whether the proposed immune mechanism can operate in the chosen host.
Then specify the result that would trigger a change. Would poor permeability prompt medicinal chemistry work, a different formulation, or a different administration route? Would loss of activity in a genetically altered model challenge the mechanism? Defining these responses in advance makes an unfavorable result useful and helps avoid repeating studies that cannot distinguish competing explanations.
Use ADME results to frame the exposure question
For small molecules, solubility and permeability are central considerations in assessing oral absorption. Artificial membrane assays such as PAMPA primarily address passive permeation, whereas cellular systems can provide additional information about transport processes. A review by Koziolek and colleagues explains the strengths and limitations of available permeability methods and why assay selection must reflect the molecule and development question [1].
Consider a hypothetical compound with strong cellular potency but poor oral efficacy. A low permeability result would justify investigating absorption, but would not establish it as the sole explanation. Researchers should plan pharmacokinetic measurements to test whether the administered regimen produces exposure compatible with the intended pharmacological effect. In vitro results guide that investigation; they do not establish tumor exposure or clinical bioavailability by themselves.
Creative Biolabs’ in vitro ADME service overview describes assays spanning physicochemical characterization, permeability, binding and metabolism. When commissioning such work, teams should specify how each result will influence compound selection or the subsequent animal study. A smaller panel linked to a clear decision can be more informative than a broad panel without an interpretation plan.
Use genetic models to test a specific mechanism
Genetic perturbation can help investigate why a treatment works or fails. Chariou and colleagues used CRISPR/Cas9 to alter genes involved in antigen processing or interferon signaling in murine tumor cells, then evaluated the resulting tumors in syngeneic mice. Their study provides an example of using defined genetic changes to investigate resistance to PD-1/PD-L1 blockade [2]. Importantly, these were edited tumor cells implanted into mice, rather than newly engineered germline mouse strains.
That distinction matters when selecting a model. Ask whether the alteration needs to occur in the tumor, the host, or both. Editing a tumor cell line and generating a genetically engineered mouse answer different questions and require different validation plans. Creative Biolabs’ CRISPR/Cas9 genome editing mouse model page describes the latter service category.
Model acceptance criteria should extend beyond a positive genotyping result. Oliver and colleagues demonstrated mosaicism in founder mice generated through genome editing, including discrepancies between tail genotyping and germline transmission [3]. For a heritable model, researchers should therefore require confirmation of the intended allele and its transmission, followed by evidence that the relevant tissue shows the expected biological change. Study planning should also address genetic background and appropriate controls.
Define what humanization needs to achieve
Humanization describes several different approaches. A mouse expressing a human target protein does not automatically possess a reconstituted human immune system. Researchers evaluating an antibody against a human checkpoint may need target compatibility, while a study of human immune cell interactions requires consideration of the relevant immune populations.
PBMC-engrafted models generally provide relatively rapid establishment of human T cells, but xenogeneic graft-versus-host disease can restrict the experimental window. CD34-positive hematopoietic stem cell approaches take longer and can support broader immune reconstitution, although human immune function remains incomplete. Cogels and colleagues review these tradeoffs and the limitations of different humanization methods [4].
Creative Biolabs’ humanized mouse tumor modeling and pharmacodynamics overview distinguishes humanized PDX, humanized CDX and humanized tumor target models. Teams comparing these options should request a precise description of what is humanized and how it is verified.
Before dosing, agree on the immune measurements required for the proposed mechanism and the criteria for enrollment. For donor-dependent studies, specify how donor variation will be represented in the design and analysis. Plan monitoring that can help distinguish treatment-related effects from complications of the model. A tumor response is easier to interpret when the report also documents whether the required biological system was present during treatment.
Bring the evidence together before expanding the study
For a hypothetical small molecule and checkpoint inhibitor combination, a useful sequence is to investigate the compound’s disposition, test a defined mechanistic hypothesis, and then select an immune model capable of addressing the combination question. This is a planning framework, not a mandatory sequence for every program. Some questions require parallel experiments, while others can be resolved without a humanized mouse study.
The study report should connect dose and measured exposure with target engagement, relevant immune observations and tumor outcomes wherever feasible. If efficacy is absent, those measurements help determine which explanation deserves further testing. If efficacy is present, they help assess whether it is consistent with the proposed mechanism.
Before expanding into additional models, the project team should be able to state what the first study established, what remains unresolved, and which experiment would most directly resolve it. That discipline makes each model a source of interpretable evidence in preclinical oncology and gives researchers a clearer basis for advancing, revising or stopping a program.
Author Bio
Dr. Emily R. Coleman is a senior scientist at Creative Biolabs with a background in immunology, oncology research, and translational biotherapeutic development. Her work focuses on translating complex biological mechanisms into practical experimental strategies for next generation therapeutic discovery, with particular expertise in tumor immunology and antibody engineering.
At Creative Biolabs, she provides scientific insight across multiple R&D domains, including antibody discovery and development, gene and cell therapy research, and translational assay design. She is particularly focused on improving the connection between mechanistic biology and predictive preclinical models to support more effective therapeutic development across immune related disease areas.
About Creative Biolabs
Creative Biolabs is a biotechnology service provider supporting academic, pharmaceutical, and biotechnology researchers worldwide. Its capabilities span antibody discovery, vaccine development, mRNA and RNA technologies, cell and gene therapy, and related preclinical research.
Creative Biolabs’ products and services are intended for research use only and are not for diagnostic or therapeutic use.
