Search

Appendix

Figure S1. Sample flow-chart describing the number of biological samples and clinical data obtained through an international multi-institutional effort and the distribution of samples in each microarray

Figure S3. Ploidy values obtained from allele-specific copy number analysis are depicted. Scatter plot of CPCs (red), CPPs (yellow) and aCPPs (light blue) showing two distinct groups in CPCs: hypodiploid CPCs (dark blue), and hyperdiploid CPCs (green) exhibiting significant differences (Mann-Whitney test p=1.00x10-4).

Figure S2. Unsupervised clustering analysis of CPTs showed significant segregation of CPCs from CPPs with aCPPs (subgroups delineated by red rectangles p=0·05). Clustering was performed using PVCLUST and 1000 gene expression normalized intensities and methylation Beta values with the largest MAD. Colors: Red: CPC, Yellow: CPP, Light blue: aCPP

Table S1-S3. S1Choroid plexus tumor chromosomal instability reported in literature S2-Characterization of patient and sample cohort used in analysis S3 Frequency of TP53 mutations in CPTs and characterization of mutation types. Abbreviations: CPC: choroid plexus carcinoma, CPP: choroid plexus papilloma, aCPP: atypical choroid plexus papilloma, SH3: SRC Homology 3 Domain

supplemental figure and table legend

Purpose: To investigate molecular alterations in choroid plexus tumors (CPT) using a genome-wide high-throughput approach to identify diagnostic and prognostic signatures that will refine tumor stratification and guide therapeutic options.Experimental Design: One hundred CPTs were obtained from a multi-institutional tissue and clinical database. Copy-number (CN), DNA methylation, and gene expression signatures were assessed for 74, 36, and 40 samples, respectively. Molecular subgroups were correlated with clinical parameters and outcomes.Results: Unique molecular signatures distinguished choroid plexus carcinomas (CPC) from choroid plexus papillomas (CPP) and atypical choroid plexus papillomas (aCPP); however, no significantly distinct molecular alterations between CPPs and aCPPs were observed. Allele-specific CN analysis of CPCs revealed two novel subgroups according to DNA content: hypodiploid and hyperdiploid CPCs. Hyperdiploid CPCs exhibited recurrent acquired uniparental disomy events. Somatic mutations in TP53 were observed in 60% of CPCs. Investigating the number of mutated copies of p53 per sample revealed a high-risk group of patients with CPC carrying two copies of mutant p53, who exhibited poor 5-year event-free (EFS) and overall survival (OS) compared with patients with CPC carrying one copy of mutant p53 (OS: 14.3%, 95% confidence interval, 0.71%–46.5% vs. 66.7%, 28.2%–87.8%, respectively, P = 0.04; EFS: 0% vs. 44.4%, 13.6%–71.9%, respectively, P = 0.03). CPPs and aCPPs exhibited favorable survival.Discussion: Our data demonstrate that differences in CN, gene expression, and DNA methylation signatures distinguish CPCs from CPPs and aCPPs; however, molecular similarities among the papillomas suggest that these two histologic subgroups are indeed a single molecular entity. A greater number of copies of mutated TP53 were significantly associated to increased tumor aggressiveness and a worse survival outcome in CPCs. Collectively, these findings will facilitate stratified approaches to the clinical management of CPTs. Clin Cancer Res; 21(1); 184–92. ©2014 AACR.

Figure S5. Kaplan-Meier curves depicting (A) overall survival and (B) event-free survival estimates of CPCs by ploidy (hyperdiploid: green, hypodiploid: blue). Statistical values were obtained with the Log-rank (Mantel-Cox) test.

Figure S4. Gene set enrichment analysis (GSEA) comparing the expression of different biological pathways and processes between hypodiploid (red) and hyperdiploid (blue) CPCs. (5% FDR, p

Purpose: To investigate molecular alterations in choroid plexus tumors (CPT) using a genome-wide high-throughput approach to identify diagnostic and prognostic signatures that will refine tumor stratification and guide therapeutic options. Experimental Design: One hundred CPTs were obtained from a multi-institutional tissue and clinical database. Copy-number (CN), DNA methylation, and gene expression signatures were assessed for 74, 36, and 40 samples, respectively. Molecular subgroups were correlated with clinical parameters and outcomes. Results: Unique molecular signatures distinguished choroid plexus carcinomas (CPC) from choroid plexus papillomas (CPP) and atypical choroid plexus papillomas (aCPP); however, no significantly distinct molecular alterations between CPPs and aCPPs were observed. Allele-specific CN analysis of CPCs revealed two novel subgroups according to DNA content: hypodiploid and hyperdiploid CPCs. Hyperdiploid CPCs exhibited recurrent acquired uniparental disomy events. Somatic mutations in TP53 were observed in 60% of CPCs. Investigating the number of mutated copies of p53 per sample revealed a high-risk group of patients with CPC carrying two copies of mutant p53, who exhibited poor 5-year event-free (EFS) and overall survival (OS) compared with patients with CPC carrying one copy of mutant p53 (OS: 14.3%, 95% confidence interval, 0.71%–46.5% vs. 66.7%, 28.2%–87.8%, respectively, P = 0.04; EFS: 0% vs. 44.4%, 13.6%–71.9%, respectively, P = 0.03). CPPs and aCPPs exhibited favorable survival. Discussion: Our data demonstrate that differences in CN, gene expression, and DNA methylation signatures distinguish CPCs from CPPs and aCPPs; however, molecular similarities among the papillomas suggest that these two histologic subgroups are indeed a single molecular entity. A greater number of copies of mutated TP53 were significantly associated to increased tumor aggressiveness and a worse survival outcome in CPCs. Collectively, these findings will facilitate stratified approaches to the clinical management of CPTs. Clin Cancer Res; 21(1); 184–92. ©2014 AACR.