Table 1.

In vivo preclinical models for Immunotherapy

Model	Advantages	Limitations	References
Immunocompetent models
Carcinogen-induced models: Tumor bearing mice induced after administration of carcinogen	+ Sporadic cancer development  + High heterogeneity  + Natural tumor microenvironment  + Tumors develop from normal cells  + Easy to work with  + A wide range of methodologies can be incorporated to induce tumors	‒ Difficult to monitor tumor growth ‒ Variability in tumor progression and high heterogeneity ‒ Large cohorts needed for data interpretation ‒ Do not mimic tumor formation from chronic inflammation ‒ Severe DNA damage ‒ Limited human cancers purely derived from carcinogen exposure	[22–25, 31]
GEMMs Tumor bearing mice established through genetic manipulation of cancer causing genes	+ Natural tumor microenvironment  + Tumors development from normal cells  + Modeling cancer at a variety of stages	‒ Low immunogenicity ‒ Difficult to monitor tumor growth ‒ Lengthy and variable tumor progression ‒ Costly and challenging breeding and gene manipulation process ‒ Genomic homogeneity	[22, 25- 27, 31]
Syngeneic Models Tumor bearing mice established through injection of murine cancer cell lines	+ Reproducible  + Easy establishment of large cohorts  + Accurate tumor monitoring  + Non-immunogenic  + Low cost	‒ Lack of native tumor microenvironment ‒ Methodology linked alteration of immunophenotype ‒ Lack heterogeneity ‒ Limited host strains	[26, 29–31]
Humanized Models
Hu-CDX Immunocompromised mice bearing human tumor cell line xenograft and reconstituted with a HIS	+ High engraftment rates and reproducibility  + Inexhaustible tumor source  + Potential for metastasis when transplanted orthotopically  + CDX models of a variety of tumor types readily available commercially	‒ Highly selective in vitro expansion resulting in genetic and phenotypic aberrations ‒ Low predictive power and correlation to clinical results ‒ Limited by the simplicity of 2D cell cultures	[13, 25, 46, 52–54]
Hu-PDX Immunocompromised mice bearing whole tissue human tumor xenografts and reconstituted with a HIS	+ Retention of tumor cell heterogeneity and stromal tissue (at low passages)  + Reproduces the complexity of the original tumor and immune system  + Have been established for a wide variety of tumor types, including drug refractory tumors  + Allogeneic models readily available commercially	‒ Low tumor engraftment success rate (approximately 49%) ‒ Engraftment favors aggressive tumors ‒ Long establishment times (at least 3 months) ‒ Low rates and duration of immune reconstitution (dependant on humanization method) ‒ Onset of GvHD shortening experimental window ‒ Costly	[10, 14, 44]
Hu-CAR Immunocompetent mice bearing human Tumor xenograft and HIS and administered a CAR therapeutic	+ Recapitulate post treatment immune changes  + Measure CAR mediated killing (both direct and indirect via activation of resident immune cells)  + Facilitates the design of new CAR therapeutics	‒ Limited IL-6 expression in these models ‒ Rapid onset of GvHD ‒ Not able to model resistance over time	[17, 18, 63]