Tumor slices as a robust pre-clinical platform

From the Gujral lab, Human Biology Division

Limiting the likelihood of failure for subsequent clinical trials is the main challenge when assessing the therapeutic potential of new compounds or new combinations in pre-clinical studies. Given the essential role of a tumor’s microenvironment (e.g matrix, fibroblasts, immune cells) in tumor progression and response to treatment, in vitro studies using cancer cell line only or even coculture assays do not represent a reliable platform to identify translatable therapeutic candidates. Immune-competent in vivo systems, albeit being physiologically relevant are time-consuming and expensive.  These limitations make rapid identification of promising strategies difficult. To address these challenges, the Gujral lab in the Human Biology Division established a screening platform using organotypic tumor slice culture. Bulk tumors are sliced into very thin layers of tissue and cultured in vitro. Dr. Gujral explained the significance of this strategy: “Among the new wave of comprehensive in vitro methodologies, including tumor organoids, and patient-derived xenograft (PDX) models, organotypic tumor slice cultures represent a physiologically-relevant culture system that is feasible and convenient for studying the tumor microenvironment. However, the utility of the slice cultures had been marred by their poor characterization in the past.” They published their results in Oncoimmunology.

Schematic showing how organotypic tissue slice culture is an excellent compromise between cell culture and in vivo preclinical models, in terms of physiological relevance, analytical complexity and sample cost.
Schematic showing how organotypic tissue slice culture is an excellent compromise between cell culture and in vivo preclinical models, in terms of physiological relevance, analytical complexity and sample cost. Illustration provided by Dr. Gujral.

The authors started a careful characterization of different syngeneic tumor models (pancreas, breast, colon, melanoma), as well as human tumor models, primarily focusing on the immune contexture. First, the authors  determined that all the immune populations assessed (CD4 and CD8 T cells, B cells, NK cells, macrophages, neutrophils, dendritic cells) were present in similar proportions when comparing bulk-tumor cores and tumor slices before culture. In the study, human liver tumor slices were characterized by a significant loss of CD4 and CD8 T cells compared to the bulk-tumor core, but neutrophils and macrophages remained unchanged. The authors think this difference in cellular proportions between tumor slices and bulk-tumor core is due to intra-tumoral heterogeneity.

The objective of this platform is to test the efficacy of therapeutic strategies; the researchers sought to determine immune status of the tumor slices over time in culture. They observed that T cells were lost rapidly after three to four days of culture, whereas macrophages were surprisingly persistent up to seven days of in vitro culture. Dr. Sivakumar and colleagues treated tumor slices with either IFNg or immune checkpoint inhibitors (e.g. anti-PDL-1 antibody) during the four-day window where T cells persisted. They demonstrated that T cell, macrophage and dendritic cell activation was achievable after IFNg stimulation and that PDL-1 blockade was successful at increasing the T cell pool in these tumor slices. Breast and colon PDX-derived tumor slices also responded to other targeted therapies. Dr. Gujral explains the perspective: “We demonstrated that we can assess response to both targeted and immune checkpoint therapies using patient’s primary tumor samples in less than 4 days, therefore enabling precision medicine. We anticipate that our immune profiling data and demonstration of the organotypic tumor slice culture as a platform for drug testing will enable researchers to appreciate its utility as a preclinical model system.”

Dr. Gujral highlights that the collaborative work has been instrumental in this study: “Our collaborators at UW provided us with the resected liver from hepatocellular carcinoma patients. The resected liver sections were critical for understanding the feasibility of preparing viable tumor slices from human samples and for studying their response to immunomodulatory drugs. Similarly, our collaborators from Cornell, Dana Farber and Harvard helped in ascertaining the similarity and differences in cytotoxic drug response between tumor organoids and tumor slices prepared from the same colon cancer PDX model.”

The Gujral lab is already looking ahead to new challenges to further accelerate the translational studies using their platform. Gujral says: “We are combining organotypic tissue slices with machine learning approaches to predict response to hundreds of drugs using limited amount of primary tumor samples.”

This work was supported by the National Institutes of Health.

Fred Hutch/UW Cancer Consortium members Drs Gujral and Yeung, contributed to this research.

Sivakumar R, Chan M, Shin JS, Nishida-Aoki N, Kenerson HL, Elemento O, Beltran H, Yeung R, Gujral TS. 2019. Organotypic tumor slice cultures provide a versatile platform for immuno-oncology and drug discovery. Oncoimmunology. 10;8(12): e1670019. doi: 10.1080/2162402X.2019.1670019.