With a 7% likelihood of regulatory approval, oncology drug registration failure rates lead all therapeutic areas. Further complicating the drug development process, preclinical oncology models poorly reflect tumor tissue biology and produce a high level of false positive data. Yet, monocellular two-dimensional (2D) tissue culture remains the preferred platform for most laboratory preclinical studies. The popularity of 2D tissue culture was driven predominantly by fast and dependable proliferation of human tumor cells rather than alignment to the pathophysiology of human cancer. As a consequence of this artificial selection bias, many registered oncology drugs today are highly toxic with a propensity to attack any dividing cell (i.e., non-targeted). Fortunately, recent advances in genomics, metabolomics and immunology have aided in development of new targeted agents, many of which have achieved regulatory approval.

Nevertheless, rational drug combinations and identification of clinically meaningful drug targets persist as major challenges in oncology, and each can be meaningfully addressed using more diverse and biologically relevant preclinical models. Ex vivo tumor tissue transplant models, such as patient-derived xenografts (PDX) offer greater model diversity and more faithfully reflect patient tumor genetics. However, cost and scalability barriers tend to limit widespread and practical model utility. In addition, there remains a strong selection bias for autocrine human tumors that can quickly adapt to a cross-species mouse host.

As a practical alternative to PDX models, we established, serially propagated and molecularly characterized 300 ex vivo 3D (3DX) models spanning 15 tumor indications. Given patient tumor seeding success rates of over 95%, the 3DX-TGA model enabled diverse pharmacologic evaluation for the vast majority of cancer patient models attempted. After achieving sufficient tumor biomass, typically within 7 to 8 weeks, thirty-one drugs were screened for proliferation and viability endpoints over a three-log dose range. Resistance and sensitivity profiles mirrored population-based response rates for indication-aligned FDA-registered drugs. Taken together, the 3DX-TGA model represents a powerful preclinical ex vivo model that: (1.) more faithfully recapitulates human tumor biology, (2.) can be scaled at a fraction of time and cost compared to PDX models, and (3.) provide a superior screening platform for novel drugs and drug-drug combinations.

Citation Format: Praveen Nair, Dileep Nair, Kaede Hinata, Cyrus Mirsaidi, Junjie Wu, Yong Hu, Brett M. Hall. Ex vivo three-dimensional tumor growth assay: 3DX-TGA [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5767. doi:10.1158/1538-7445.AM2017-5767