Development of 3-dimensional human pituitary corticotroph tumor cultures as a preclinical model for drug discovery

Abstract Cushing Disease (CD) is a life-threatening “orphan disease” caused by an adrenocorticotropic hormone (ACTH)-secreting pituitary adenoma driving excess adrenal cortisol production. There is a large unmet medical need for CD treatment. However, translational research has been greatly hampered due to unavailability of any

human pituitary corticotroph tumor cell models. Using single cell RNA-sequencing (scRNAseq) and microarray transcriptome analysis of surgically resected human corticotroph tumors, we observed that loss of pituitary corticotroph tumor ACTH secretion coincided with reduced angiogenesis, survival signals and immune

responses in parallel with increased collagen catabolism, cell adhesion and extracellular matrix organization. Guided by these findings, we developed a unique 3-dimensional (3D) pituitary tumor culture system and for the first time, we have been able to generate 3D human corticotroph tumor cultures that secrete ACTH >4 months.

We have assembled an experienced multidisciplinary team to complete 3 focused specific aims using this first of its kind resource. Firstly, we will use whole exome sequencing to characterize the genomic landscape of our corticotroph 3D culture biobank and compare genomic and genetic fidelity between the original corticotroph

tumor, normal blood and matched 3D corticotroph tumoroid cultures from the same individual patient. ScRNAseq analysis of serial passages of individual patient-derived corticotroph tumor cultures will monitor for transcriptome changes in a temporal fashion over the course of culture. The histopathological structure of our

3D corticotroph cultures at the single cell level will quantify tissue architecture so we can map corticotroph tumoroid cellular composition and distribution. A second aim will employ a miniaturized automated system to conduct a high throughput drug screen in our 3D corticotroph tumor cultures. Compounds will be subjected to

rigorous evaluation to define primary “hits” and validated by re-screening in triplicate using 20 concentrations from 100µM to 20pM (2-fold dilution) to reliably calculate an EC50 for each compound. Finally, three complementary approaches, computational cheminformatic profiling, scRNAseq to delineate transcriptomic

changes at the single cell level following drug treatment and functional genomics will be employed to explore the MOA of validated hit compounds. This integrated interrogation of our drug screen results and the genetic features of our patient-derived 3D tumor cultures as well as that of the original tumor tissue, will allow us to

disentangle an individual drug's mode(s) of action, and directly document drug sensitivity of individualized parental corticotroph tumors. In summary, we will use our unique biobank of comprehensively molecularly characterized pituitary corticotroph tumor tissues and paired derived 3D corticotroph tumor cultures to test

libraries of clincially relevant compounds. This pituitary 3D tumor culture system is transformative in the field due to the lack of any human pituitary corticotroph tumor cell models and will pave the path for much needed improved therapy for patients with this dreadfully disabling and often fatal disorder, Cushing disease.