Xenograft Model

A xenograft model is a preclinical research method in which living cells, tissues, or organs from one species are transplanted into a host organism of a different species. In biomedical research, the most common form involves implanting human tumor cells into immunodeficient mice to study cancer biology, test drug efficacy, and evaluate therapeutic strategies in a living system before proceeding to human clinical trials.

Xenograft models occupy a critical position in the translational research pipeline, bridging the gap between laboratory cell culture experiments and human clinical trials. They allow researchers to study how human tumors grow, metastasize, and respond to treatment within a living organism that approximates physiological complexity. Under the ICH S9 guideline on nonclinical evaluation of anticancer drugs, used by regulators such as the FDA and the EMA, a drug's antitumor activity should have been characterized before phase I studies, and xenograft studies are a common way to do so; they support, but do not replace, the toxicology studies needed before a first in human trial. For systematic reviewers and evidence synthesis teams, understanding xenograft methodology is essential when evaluating preclinical evidence, assessing translational validity, and interpreting animal to human concordance in drug development pipelines.

Researchers obtain human tumor cells from established cell lines (cell line derived xenografts) or tumor tissue taken directly from patients' surgical specimens or biopsies (patient derived xenografts, or PDX models). The cells or tissue fragments are implanted into immunodeficient mice, typically subcutaneously, orthotopically (into the organ of tumor origin), or intravenously for metastasis studies.

Once tumors establish and reach a measurable size, the animals are randomized into treatment and control groups. Tumor volume is measured at regular intervals using calipers or imaging techniques, and treatment effects are assessed by comparing growth curves, survival times, and biomarker responses. At study completion, tumors are harvested for histological, molecular, and genomic analyses to characterize treatment response mechanisms.

Each approach has limits. Cell line xenografts are inexpensive and reproducible, but lines grown for years in culture can differ from the tumors they came from; PDX models keep more of the donor tumor's histology and genetic features but cost more and take months to establish. Because the host mice lack a functional immune system, standard xenografts cannot show how a treatment acts through the immune system, and the supporting tissue around the human tumor (the stroma) is of mouse origin.

Use xenograft models when evaluating anticancer drug candidates before clinical trials, when studying tumor biology in a physiologically relevant context, when comparing drug combinations or dosing schedules, and when developing biomarkers for treatment response. They are standard in oncology drug development, and patient derived xenografts have been explored in personalized medicine to test treatments for an individual patient, although the months needed to establish a model limit this use. On their own they are a poor fit for immunotherapies, which need a working immune system; immunocompetent (syngeneic) or humanized mouse models are used instead.

Assuming that xenograft results will directly predict human clinical outcomes ignores the immunological and microenvironmental differences between species. In an analysis of US National Cancer Institute data, Johnson and colleagues (2001) found that a drug's activity in xenografts of a given tumor type did not closely match its activity in phase II trials of the same human cancer. Using only subcutaneous models when the research question concerns metastasis or organ specific drug delivery limits translational relevance. Failing to report the immunodeficient mouse strain, tumor cell passage number, and implantation site makes it difficult to reproduce the study or compare results across laboratories. The ARRIVE 2.0 guidelines (2020) list what animal studies should report, including randomization, blinding and sample size, and the 3Rs (replacement, reduction and refinement) ask researchers to use no more animals than needed and to limit suffering, for example with humane endpoints for tumor size.

A pharmaceutical team evaluating a novel kinase inhibitor for colorectal cancer establishes PDX models from five patient tumors with different molecular profiles. Each PDX line is expanded and randomized into treatment and vehicle control groups. The inhibitor shrinks tumors significantly in the two PDX lines carrying the target mutation but shows no effect in the three lines without it. These preclinical results support enrolling or stratifying patients by the target mutation in subsequent early phase clinical trials, although five tumor lines are a small sample and the mice could not show any immune contribution to the drug's effect.