Divergent clonal evolution of castration-resistant neuroendocrine prostate cancer

Abstract

An increasingly recognized resistance mechanism to androgen receptor (AR)-directed therapy in prostate cancer involves epithelial plasticity, in which tumor cells demonstrate low to absent AR expression and often have neuroendocrine features. The etiology and molecular basis for this 'alternative' treatment-resistant cell state remain incompletely understood. Here, by analyzing whole-exome sequencing data of metastatic biopsies from patients, we observed substantial genomic overlap between castration-resistant tumors that were histologically characterized as prostate adenocarcinomas (CRPC-Adeno) and neuroendocrine prostate cancer (CRPC-NE); analysis of biopsy samples from the same individuals over time points to a model most consistent with divergent clonal evolution. Genome-wide DNA methylation analysis revealed marked epigenetic differences between CRPC-NE tumors and CRPC-Adeno, and also designated samples of CRPC-Adeno with clinical features of AR independence as CRPC-NE, suggesting that epigenetic modifiers may play a role in the induction and/or maintenance of this treatment-resistant state. This study supports the emergence of an alternative, 'AR-indifferent' cell state through divergent clonal evolution as a mechanism of treatment resistance in advanced prostate cancer.

Figure 1. Clinical and mutational profile of the cohort

(a) Schematic illustrating sites of biopsy for CRPC-NE (dark pink) and CRPC-Adeno (light pink) subgroups. Numbers in circles indicate numerosity of samples from each site. (b) AR signaling (right) based on abundance of mRNA transcripts included in the AR signaling signature described in ref 19. Violin plots show the density of AR signaling. Each dot represents a sample; diamonds and solid lines represent the mean and 95% confidence interval, respectively. Representative immunohistochemistry (left) shows AR protein expression. Scale bars, 50 μm. (c) Significantly mutated genes. Each row represents a gene and each column an individual subject. Top light green bars correspond to the total number of non-silent SNVs in an individual. Left light green bars indicate the number of subjects harboring non silent corresponding mutations in the genes indicated on the right. Bottom panel reports the copy number status of selected genes. (d) Genomic location of AR mutations in samples from SU2C-2015 and this study. (e) Copy number status of AR locus. Color intensity and location are indicative of level and focality of amplification. (f) Frequency of copy number aberrations; concordant fractions (gray), CRPC-NE specific (dark pink) and CRPC-Adeno specific (light pink). Data adjusted for tumor ploidy and purity. Highlighted genes are significantly preferentially aberrant in one class and demonstrate concordant differential mRNA levels (for DNA and mRNA: FDR <= 10% for deletions and p-value <= 1% for amplifications).

Figure 2. Tracing CRPC-NE emergence through allele specific analysis

(a) Potential models of evolution that occur during prostate cancer progression towards the neuroendocrine phenotype: linear progression from primary untreated adenocarcinoma to CRPC-Adeno to CRPC-NE; independent progression of two distinct clonal populations within the primary or metastatic CRPC-Adeno towards either CRPC-Adeno or CRPC-NE; divergent clonal evolution of CRPC-NE from either primary adenocarcinoma or CRPC-Adeno. *indicates favored model. (b) Allele specific analysis of primary prostate adenocarcinoma and local lymph node metastasis removed at time of radical prostatectomy (RP) and two metastatic CRPC-NE (treated) tumors (3 years after RP) from subject WCMC7520. H&E pathology images and intervening therapies are shown in the timeline. Scale bars, 100 μm. (c) Allele specific analysis of tumors at three time points from patient subject WCMC161 during castration resistance: lymph node (CRPC-Adeno), bone biopsy (CRPC-Adeno), and liver biopsy (small cell CRPC-NE). H&E pathology images and intervening therapies are shown in the timeline. ADT= androgen deprivation therapy; EP= etoposide and cisplatin chemotherapy; Abi= abiraterone acetate with prednisone. Circos plots summarize genome-wide allele specific DNA quantity in tumor cells. Individual's tumor phylogeny sketched upon allele-specific analysis including genome-wide amplification and ploidy assessment. Scale bars, 100 μm.

Figure 3. Methylation analysis of CRPC-NE and CRPC-Adeno

(a) Hierarchical clustering of 28 eRRBS samples data using (1 - Pearson's correlation) as distance measure on unselected sites. Clinical features of outlier cases are described. (b) Left, pie chart showing the number of differentially methylated genes, identified by annotating hyper- and hypo- methylated loci (number is reported between parentheses) on GENCODE version 19. Right, table shows a selection of terms enriched by differentially methylated genes. (c) Top, genome track of SPDEF. Hyper-methylated loci are reported in the annotation track. Bottom, box plot of expression levels of SPDEF samples for This Study (left) and SU2C/PCF 2015 (right) cohorts. (d) Bar plots highlight the effect of EZH2 transcription activity across 487 samples with different pathology classification. The bars are relative to the mRNA level fold (with respect to benign prostate tissue samples) of homeobox genes under-expressed in CRPC-NE versus CRPC-Adeno (FDR < 0.1); a selection of EZH2 target genes (DKK1, NKD1, AMD1, HOXA13, HOXA11, NKX3-1); DNA methyltransferase genes - indicated as DNMTs (DNMT1, DNMT3B, DNMT3A, DNMT3L); EZH2. Significance of differences between CRPC-NE and CRPC-Adeno subgroups are shown (max P = 3*10⁻⁵ for DNMTs). When significant, p-values in SU2C/PCF cohort are shown. The number of samples for each pathology classification is reported inside the square symbols of the legend. (e) Cell viability in prostate adenocarcinoma cell lines (DU145, LNCaP) the neuroendocrine prostate cell line NCI-H660 assessed at 48 hours after treatment with escalating does of the EZH2 inhibitor GSK343 (5, 7.5, 10uM).

Figure 4. Integrative DNA, RNA and Methylation analysis

(a) Weighted Venn diagram with the number of protein-coding genes significantly differentially observed in the three data layers. The superimposed pie chart reports the estimation of the impact of each layer upon the following priority rule: methylation overall and DNA over RNA. (b) Integrated NEPC score analysis across 604 samples from four different RNA-Seq prostate cancer datasets (This Study, SU2C/PCF 2015, WCMC 2011/2014 and TCGA). Samples are ordered by decreasing values of Integrated NEPC score (only a fraction of data is shown, entire data are reported in Supplementary Fig. 10). Top, annotation tracks report original dataset and pathology classification. Middle, plot reports Integrated NEPC score (black line) and AR signaling (grey line) across samples. Bottom, heat map of normalized FPKMs for a selection of the 70 genes (in rows) across samples (in columns). (c) Prediction accuracy of CRPC-NE samples by precision and recall statistics for Integrated NEPC Score (circles), AR signaling (squares), mRNA level of SPDEF (diamonds), AR (triangles) in RNA-seq datasets: This Study (green), SU2C/PCF 2015 (orange), WCMC 2011/14 (violet) and all datasets (black). Grey curves represent F-measure levels, defined as the harmonic mean of precision and recall. Due to the absence of CRPC-NE samples (positive events), TCGA data were not reported here. (d) AR signaling versus Integrated NEPC score across 730 samples from five independent prostate datasets using transcriptome data as proxy. The old-rose shaded area refers to significant values of Integrated NEPC Score. Predicted CRPC-NE percentages calculated by excluding benign samples.