Next generation patient-derived ex vivo models drive functional precision medicine

Next generation patient-derived ex vivo models drive functional precision medicine

by Imagen Therapeutics|August 24, 2022 at 8:19 PM

Typical ex vivo chemosensitivity assays initially designed to expose patient-derived cancer cells to certain drugs seemed an obvious and compelling strategy to improve the translatability of in vitro pharmacology findings to patients. Despite this promise, data generated using ex vivo models were soon deemed ineffective due to the lack of benefits in patient care. What went wrong and have we made any improvements?

The rise and fall of ex vivo models

When first designed 20 years ago, ex vivo pharmacology assays on patient-derived cancer cells were used to predict patient response to chemotherapeutics. Few drugs were available for testing which were designed to hit fast proliferating cells, regardless of their genetic make-up. Additionally, ex vivo culture conditions were relatively primitive and normal cells typically took over the culture over time. Technology was also not as advanced, such that assays mostly measured metabolic read outs which loosely correlated with cell death. The results were therefore inaccurate, and prospective clinical trials demonstrating patient benefits were missing.

Twenty years later while the first draft of the human genome was about to be completed, a kinase inhibitor drug was found to specifically target a universal genetic alteration in patients with chronic myeloid leukemia (CML). The ability to perform genomic analysis on any tumour thanks to advanced sequencing techniques gave rise to the precision medicine era. Identifying sequence abnormalities on patient tumour samples and targeting them with the right drug seemed a promising new approach. Aside from CML, precision medicine started to be applied to non-small cell lung cancer with EGFR mutations, BRAF was targeted in melanoma, and TRK mutations became drug targets for several tumours. This strategy is not rid of challenges. Finding the target mutation alone was difficult, when a target is found, designing the right drug or drug combination is also complex. It soon became evident that relying solely on genomic-based precision oncology was too simplistic as it didn’t take in consideration the complex biology of a cell in its in vivo environment.

The birth of functional precision medicine

While genome-driven drug development failed to deliver the anticipated clinical success, ex vivo culture systems were improved in both two-dimensional and three-dimensional formats. Additionally, pharmacology assays were optimized to measure perturbations induced by drugs. Nowadays we can study single cells in detail, and we have advanced bioinformatics tools to investigate the link between drug response and clinical or molecular cancer features. Finally, there are simply more cancer drugs to test, greatly increasing the chances of finding at least one active drug.

This led to an evolution of precision medicine to the so-called Functional Precision Medicine (FPM) that combines a cancer cell genetic background with its physiological features to better predict drug response. FPM has been applied to hematologic malignancies such as leukaemia (AML) and B- and T-cell non-Hodgkin lymphoma. These studies yielded significantly positive results, making FPM a promising strategy to supplement the widely used fundamental genomic-based assays.

Importantly, adequate preclinical models are needed to enable FPM approaches and accelerate the research. Ex vivo patient-derived cell models (PDCs) preserve in culture patient tumour characteristics, histological and genomic features, and can accurately predict clinical responses, as shown by a growing body of retrospective studies. These next generation models are well-characterized and can be used to select the candidate drug with the most promising clinical profile to move forward to clinical trials.

FPM has increasingly been adopted to reduce oncology drug attrition rates. To know more about Imagen's precision medicine preclinical platform, head to predictTx.