DNA content based flow sorting combined with genomic high-resolution profiling in the context of the development of castration resistance in prostate cancer

Prostate cancer (PCa) is a common malignant disease in men. Due to increased screening efforts, PCa is often diagnosed in early stages and patients are therefore eligible for curative surgical approaches. Yet many patients, mostly due to age or extended disease, are treated primarily using androgen deprivation therapy (ADT). While this approach usually leads to disease stability or even reduction, unfortunately, this effect is usually not sustainable and tumors acquire resistance toward ADT commonly referred to as castration resistance. The resulting tumor progression usually leads to long term disease progression with increased morbidity and eventually the death of the patient.
Among other factors, changes in the tumor genome may confer resistance towards ADT, similar to what has previously been demonstrated for therapy resistance in other malignancies. This thesis focuses on different translational approaches aimed at tracking down the phenotypic and genomic changes occurring in PCa upon ADT.
In the first paper included in the thesis (“ERG rearrangement and protein expression in the progression to castration-resistant prostate cancer”), we explored one of the prototypic genomic alterations, namely the TMPRSS2-ERG gene fusion, and its role in the development of castration-resistant PCa. We used a tissue microarray (TMA) that comprised clinical samples of PCa prior to and after ADT using classical molecular analysis methods such as immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH). Interestingly we were able to identify a PCa subgroup that appeared to be associated with a partial shutdown of the androgen receptor (AR)-driven signaling axis.
Further, the notion of tumor evolution and intra-tumor heterogeneity has spawned an increased interest in recent years. The advent of high resolution, high throughput genomic analysis methods have provided the necessary tools to decipher the clonal complexity of tumors. This led to an improved understanding of the clonal evolution of malignant diseases. While most computational approaches focus on the bulk of the tumor, we developed and refined a combined approach relying on prior flow sorting of cytogenetically distinct tumor subpopulations and subsequent molecular and computational analyses to obtain a different perspective on intra-tumor heterogeneity.
The resulting findings are summarized in the second paper entitled “Delineation of human prostate cancer evolution identifies chromothripsis as a polyclonal event selecting for FKBP4 driven castration resistance”. To the best of our knowledge, this was the first paper describing tumor populations exhibiting features of chromothripsis, the catastrophic shattering, and the reconnection of chromosomes, that are related but distinct in the different subpopulations. This gains further clinical significance as in the process the tumor acquired a new amplification of a chaperone protein, namely FKBP4, which is involved in AR signaling and highly correlates with patients’ survival. Taken together, these findings, therefore, might open new doors in the development of prognostic and therapeutic tools.