Current methods of preclinical testing of potential therapeutics have been, for the most part, underwhelming in terms of their ability to yield a clinical impact. Moreover, preclinical investigation of cancer therapy resistance mechanisms faces similar challenges due, at least in part, to both limitations of the preclinical model systems and lack of reliable biomarkers. A prime example is glioblastoma (GBM) with few therapeutic options yielding incredibly poor outcomes (5-year survival <4%) despite nearly eight decades of research. Indeed, current standard of care (SOC) therapy includes maximal safe surgical resection followed by fractionated irradiation and temozolomide (TMZ) chemotherapy. These patients tend to fall into two categories: those who have inherent resistance to radiation and TMZ and those who acquire resistance to those therapies (typically within 6 months). While many groups have tried to develop more effective therapeutics or overcome GBM resistance mechanisms, virtually all attempts have relied on highly artificial models under major growth promoting conditions that select for highly proliferative tumors that no longer resemble the patient?s tumor. To address these issues, investigators are increasingly utilizing patient-derived models of cancer (PDMC) coupled with comprehensive molecular profiling to build more reliable models for examining therapeutic resistance and for developing novel therapies. Building on a series of collaborations and funded projects, our investigative team has developed a large panel of GBM xenografts (PDX) that can be cultured without serum as derivative PDMC models including spheroids (neurospheres) and matrix-embedded microtumors. Our parent U01 (U01-CA223976) seeks to investigate tumor microenvironmental (TME) stressors on these 3 PDMC models in terms of molecular biology (e.g., transcriptome and kinome similarity) and phenotype (e.g., radiation and TMZ response) fidelity. In this U01 Revision Application, we seek to extend our work by developing new in vitro and in vivo models to investigate inherent and acquired resistance to SOC GBM therapy. We have characterized baseline in vivo radiation and TMZ sensitivity in 20 of our GBM PDX and have developed 8 radiation resistant and 5 TMZ resistant isogenic lines from initially sensitive tumors. Transcriptomic and kinomic testing of these pairs have identified several genes and kinases associated with resistance. We will leverage this very unique resource to examine SOC therapeutic resistance.
Aim 1 utilizes a high-throughput geospatially controlled 3D bioprinting system to replicate in vivo conditions by using co-culture of PDX cells and vascular endothelial cells in a variety of matrices with and without other TME stressors (e.g. hypoxia and nutrient deprivation seen in patients). These constructs will be tested in up to 384-well format to facilitate inhibition of high-priority targets of SOC resistance as identified from our transcriptomic and kinomic profiling.
Aim 2 will explore the role of the gut microbiome, an understudied host TME factor, in SOC therapeutic resistance as our preliminary data indicate that the gut microbiome alters TMZ sensitivity. Selected PDX will be grown in the brains of Rag1-/- mice with human fecal transplants for SOC testing.
Cancer therapies that appear successful in preclinical models often fail in clinical trials suggesting that a knowledge gap exists regarding both modeling and understanding cancer therapy resistance mechanisms. We will investigate the role that the tumor microenvironment plays in resistance to the clinical standard of care therapies of temozolomide and radiation using advanced patient-derived models of glioma.
