Skip to main content
main-content

01.12.2018 | Research article | Ausgabe 1/2018 Open Access

BMC Cancer 1/2018

Cribriform and intraductal prostate cancer are associated with increased genomic instability and distinct genomic alterations

Zeitschrift:
BMC Cancer > Ausgabe 1/2018
Autoren:
René Böttcher, Charlotte F. Kweldam, Julie Livingstone, Emilie Lalonde, Takafumi N. Yamaguchi, Vincent Huang, Fouad Yousif, Michael Fraser, Robert G. Bristow, Theodorus van der Kwast, Paul C. Boutros, Guido Jenster, Geert J. L. H. van Leenders
Wichtige Hinweise

Electronic supplementary material

The online version of this article (https://​doi.​org/​10.​1186/​s12885-017-3976-z) contains supplementary material, which is available to authorized users.

Abstract

Background

Invasive cribriform and intraductal carcinoma (CR/IDC) is associated with adverse outcome of prostate cancer patients. The aim of this study was to determine the molecular aberrations associated with CR/IDC in primary prostate cancer, focusing on genomic instability and somatic copy number alterations (CNA).

Methods

Whole-slide images of The Cancer Genome Atlas Project (TCGA, N = 260) and the Canadian Prostate Cancer Genome Network (CPC-GENE, N = 199) radical prostatectomy datasets were reviewed for Gleason score (GS) and presence of CR/IDC. Genomic instability was assessed by calculating the percentage of genome altered (PGA). Somatic copy number alterations (CNA) were determined using Fisher-Boschloo tests and logistic regression. Primary analysis were performed on TCGA (N = 260) as discovery and CPC-GENE (N = 199) as validation set.

Results

CR/IDC growth was present in 80/260 (31%) TCGA and 76/199 (38%) CPC-GENE cases. Patients with CR/IDC and ≥ GS 7 had significantly higher PGA than men without this pattern in both TCGA (2.2 fold; p = 0.0003) and CPC-GENE (1.7 fold; p = 0.004) cohorts. CR/IDC growth was associated with deletions of 8p, 16q, 10q23, 13q22, 17p13, 21q22, and amplification of 8q24. CNAs comprised a total of 1299 gene deletions and 369 amplifications in the TCGA dataset, of which 474 and 328 events were independently validated, respectively. Several of the affected genes were known to be associated with aggressive prostate cancer such as loss of PTEN, CDH1, BCAR1 and gain of MYC. Point mutations in TP53, SPOP and FOXA1were also associated with CR/IDC, but occurred less frequently than CNAs.

Conclusions

CR/IDC growth is associated with increased genomic instability clustering to genetic regions involved in aggressive prostate cancer. Therefore, CR/IDC is a pathologic substrate for progressive molecular tumour derangement.
Zusatzmaterial
Additional file 1: Figure S1. Comparison of tumour cell percentage in whole-slide reference images for both TCGA and CPC-GENE cohorts, stratified by CR/IDC status. Figure S2. PGA for deletion events in the TCGA cohort per chromosome arm for GS ≥ 3 + 4 = 7 with and without CR/IDC. Figure S3. PGA for amplification events in the TCGA cohort per chromosome arm for GS ≥ 3 + 4 = 7 with and without CR/IDC. Figure S4. PGA for deletion events in the CPC-GENE cohort per chromosome arm for GS ≥ 3 + 4 = 7 with and without CR/IDC. Figure S5. PGA for amplification events in the CPC-GENE cohort per chromosome arm for GS ≥ 3 + 4 = 7 with and without CR/IDC. Figure S6. Overview of ERG expression in TCGA [log10(TPM)] stratified by CR/IDC status (A) and deletion of the genomic region between TMPRSS2 and ERG (B). (PDF 3140 kb)
12885_2017_3976_MOESM1_ESM.pdf
Additional file 2: Table S1. Overview of genomic instability of individual chromosome arms in both TCGA and CPC-GENE datasets. Genomic instability was calculated based on a modified PGA formula (see methods). P-values are based on a Wilcon–Mann–Whitney test while log2FC represents the log2 ratio of the average PGA scores for CR/IDC positive samples and CR/IDC negative samples. PGA scores for deletions and amplifications were calculated and tested separately. (XLS 139 kb)
12885_2017_3976_MOESM2_ESM.xlsx
Additional file 3: Table S2. Gene-wise copy number alterations associated with CR/IDC growth using any CR/IDC presence for patient stratification. Columns contain: Symbol – official gene symbol, Chromosome / Start / End – genomic coordinates of gene locus, FDR – Boschloo’s exact test p-value after correcting for multiple tests using the Benjamini–Hochberg procedure. amplifications_case – number of CR/IDC positive samples with an amplification spanning gene locus, amplifications_control – number of control samples with an amplification spanning gene locus, cases – total number of CR/IDC positive samples, controls – total number of control samples. All entries are sorted by genomic location. Deletions are presented in the same format and listed separately. (XLS 226 kb)
12885_2017_3976_MOESM3_ESM.xls
Additional file 4: Table S3. Gene-wise copy number alterations associated with CR/IDC growth using a ≥ 30% CR/IDC threshold to stratify samples. Columns contain: Symbol – official gene symbol, Chromosome / Start / End – genomic coordinates of gene locus, FDR – Boschloo’s exact test p-value after correcting for multiple tests using the Benjamini–Hochberg procedure. amplifications_case – number of CR/IDC positive samples with an amplification spanning gene locus, amplifications_control – number of control samples with an amplification spanning gene locus, cases – total number of CR/IDC positive samples, controls – total number of control samples. All entries are sorted by genomic location. Deletions are presented in the same format and listed separately. (XLS 161 kb)
12885_2017_3976_MOESM4_ESM.xls
Additional file 5: Table S4. Gene-wise copy number alterations detected in the TCGA cohort and validated in the CPC-GENE cohort using a ≥ 30% CR/IDC threshold to stratify samples. Columns contain: Symbol – official gene symbol, Chromosome / Start / End – genomic coordinates of gene locus, FDR – Boschloo’s exact test p-value after correcting for multiple tests using the Benjamini–Hochberg procedure for specified dataset. amplifications_case – number of CR/IDC positive samples in specified dataset with an amplification spanning gene locus, amplifications_control – number of control samples in specified dataset with an amplification spanning gene locus, cases – total number of CR/IDC positive samples in specified dataset, controls – total number of control samples in specified dataset. All entries are sorted by genomic location. Deletions are presented in the same format and listed separately. (PDF 12328 kb)
12885_2017_3976_MOESM5_ESM.xls
Additional file 6: Figure S7. Overview heatmap of copy number alterations in CPC-GENE cohort. Clinical variables are displayed on the left, while percent genome altered (PGA) is displayed on the right. Samples are ordered by CR/IDC percentage, with two thresholds chosen to discriminate between negative (0%) and intermediate (< 30%) CR/IDC status. (XLSX 14 kb)
12885_2017_3976_MOESM6_ESM.pdf
Additional file 7: Table S5. Significant CNAs identified by logistic regression analysis accounting for genomic instability as confounding factor in the TCGA dataset. Columns contain: Symbol – official gene symbol, Chromosome / Start / End – genomic coordinates of gene locus, p-alue / FDR – p-value of logistic regression before and after correction for multiple tests via FDR, odds ratio / 95% CI – odds ratio and 95% confidence interval based on logistic regression. Deletions and amplifications are presented in the same format and listed separately. All entries are sorted by genomic location. (XLS 266 kb)
12885_2017_3976_MOESM7_ESM.xls
Literatur
Über diesen Artikel

Weitere Artikel der Ausgabe 1/2018

BMC Cancer 1/2018 Zur Ausgabe

Neu im Fachgebiet Onkologie

Mail Icon II Newsletter

Bestellen Sie unseren kostenlosen Newsletter Update Onkologie und bleiben Sie gut informiert – ganz bequem per eMail.

Bildnachweise