Accumulation of copy number alterations and clinical progression across advanced prostate cancer

Abstract

Background: Genomic copy number alterations commonly occur in prostate cancer and are one measure of genomic instability. The clinical implication of copy number change in advanced prostate cancer, which defines a wide spectrum of disease from high-risk localised to metastatic, is unknown.

Methods: We performed copy number profiling on 688 tumour regions from 300 patients, who presented with advanced prostate cancer prior to the start of long-term androgen deprivation therapy (ADT), in the control arm of the prospective randomised STAMPEDE trial. Patients were categorised into metastatic states as follows; high-risk non-metastatic with or without local lymph node involvement, or metastatic low/high volume. We followed up patients for a median of 7 years. Univariable and multivariable Cox survival models were fitted to estimate the association between the burden of copy number alteration as a continuous variable and the hazard of death or disease progression.

Results: The burden of copy number alterations positively associated with radiologically evident distant metastases at diagnosis (P=0.00006) and showed a non-linear relationship with clinical outcome on univariable and multivariable analysis, characterised by a sharp increase in the relative risk of progression (P=0.003) and death (P=0.045) for each unit increase, stabilising into more modest increases with higher copy number burdens. This association between copy number burden and outcome was similar in each metastatic state. Copy number loss occurred significantly more frequently than gain at the lowest copy number burden quartile (q=4.1 × 10^-6). Loss of segments in chromosome 5q21-22 and gains at 8q21-24, respectively including CHD1 and cMYC occurred more frequently in cases with higher copy number alteration (for either region: Kolmogorov-Smirnov distance, 0.5; adjusted P<0.0001). Copy number alterations showed variability across tumour regions in the same prostate. This variance associated with increased risk of distant metastases (Kruskal-Wallis test P=0.037).

Conclusions: Copy number alteration in advanced prostate cancer associates with increased risk of metastases at diagnosis. Accumulation of a limited number of copy number alterations associates with most of the increased risk of disease progression and death. The increased likelihood of involvement of specific segments in high copy number alteration burden cancers may suggest an order underlying the accumulation of copy number changes.

Trial registration: ClinicalTrials.gov NCT00268476 , registered on December 22, 2005. EudraCT 2004-000193-31 , registered on October 4, 2004.

Keywords: Advanced prostate cancer; Copy number alteration; Genomic biomarkers; STAMPEDE trial.

Fig. 1

CN-300 cohort and association of the burden of copy number alteration with metastatic states. A Map demonstrating all UK STAMPEDE trial sites recruiting patients included in the CN-300 cohort. B Distribution of burden of copy number (CN) alteration (%) in tumour-enriched region of index core split by metastatic states (N=284; 16 metastatic patients with unknown designation for low versus high volume were excluded). C–E Alteration frequency (%) of patients with at least one segment of loss mapped to denoted cytobands in C M0N1 versus M0N0; D M1 low versus M0N1; E M1 high versus M1 low. F–H Alteration frequency (%) of patients with at least one segment of gain mapped to denoted cytobands in F M0N1 versus M0N0; G M1 low versus M0N1; and H M1 high versus M1 low

Fig. 2

Association of copy number alteration with clinical outcome measures. A–D Cox survival model demonstrating adjusted estimate of impact of burden of copy number (CN) alteration as a continuous variable on hazard of A failure-free survival; B metastatic progression-free survival; C prostate cancer-specific survival; D overall survival. Variables included in adjusted analysis: (1) grading group; (2) metastatic state (M0N0, M0N1, M1 low and M1 high); (3) Pre-ADT serum PSA log transformed; (4) age at randomisation; (5) tumour cellularity (%). Black line represents the impact of the burden of copy number alteration on relative risk for 284/300 patients for which we could determine disease state (16 M1 patients with unknown designation for low versus high volume were excluded). Each coloured line represents sub-groups of the CN-300 cohort defined by metastatic state. E–H Kaplan-Meier estimates. CN-300 cohort split into quartile groups determined by the burden of copy number alteration. Time-to event; E failure-free survival; F metastatic progression-free survival; G prostate cancer-specific survival; H overall survival

Fig. 3

Proportion of copy number alteration attributable to loss as compared to gain. A–D Patients are ranked with ascending burden of copy number (CN) alteration (%) identified in the index core on the x-axis. Y-axis represents the proportion of copy number alteration (%) attributable to a gain (red dot) as compared to a loss (blue dot). The line represents the conditional mean (blue=loss, red=gain), i.e. an estimate of the mean proportions conditional on the number of patients (LOESS function). The grey band indicates the confidence interval (level of confidence 0.95) A M0N0 patients; B M0N1 patients; C M1 low patients; D M1 high patients. E Density plot representing the proportion of the genome altered by a copy number loss in the index core. Each line sub-groups the CN-300 cohort into burden of copy number alteration quartiles. F Boxplot demonstrating the difference in ratio of the burden of copy number alteration due to a gain as compared to a loss (log transformed) between non-metastatic (M0) and metastatic (M1) patients

Fig. 4

Frequency of copy number alteration. A Landscape of copy number alteration across the autosome. The CN-300 cohort is split into quartile groups defined by burden of copy number alteration (%) in index core (red to yellow=quartile1-4). Y-axis=Number of patients with an alteration (above midline=copy number gain; below midline=copy number loss). X-axis=genomic location. Regions of interest are annotated by chromosome followed by genomic location and mapped cytoband. B Stacked bar chart of selected copy number altered segments in the CN-300 cohort index cores. Regions ordered putatively ‘early’ (top) to ‘late’ alterations (bottom). Each bar divides the patients harbouring the specific genomic alteration (total number of patients annotated at the end of each bar) into burden of copy number alteration quartile groups. Regions of interest are represented by chromosome number, genomic location and cytoband (blue=copy number loss, pink=copy number gain). Regions containing known prostate cancer genes of interest are listed as follows: 8:23.4–24 (NXK3.1), 10:89.5–90 (PTEN), 13:48.5–49 (RB1), 17:7.5–8 (TP53), 8:128.5–129 (cMYC), 5:98–98.5 (CHD1). C Density plots demonstrating distribution of burden of copy number (CN) alteration (%) identified in the index core of patients with and without 8p segment deletions (low KS distance). All CN-300 patients harbouring 8:13–13.5 (8p22) and/or 8:11.5–12 (8p23) segment deletions are represented by a blue line versus no 8p22 and 8p23 segment deletion represented by a black line (8p22 deletion N=220, 8p23 deletion N=236). D Density plots demonstrating two regions with a high KS score that are associated with a higher burden of CN alteration (%); 8:128.5–129 (8q24 harbours cMYC) and 5:98–98.5 (5q21 harbours CHD1). All CN-300 patients harbouring 8:128.5–129 (8q24) gain are represented by a red line (N=145) versus no alteration at that segment (black line). All CN-300 patients harbouring 5:98–98.5 (5q21) loss are represented by a blue line (N=79) versus no alteration at that segment (black line)

Fig. 5

Variance in copy number alteration across multi-region diagnostic cores. A For a sub-set of 112 patients within the CN-300 cohort, we were able to copy number profile multiple diagnostic core biopsies from the same prostate (N=500, median 4 diagnostic core biopsies per patient). We calculated the variance in burden of copy number alteration (%) defined as the standard deviation across cores from the same prostate, represented as a dot. The colour and size of the dot represents the number of diagnostic cores copy number profiled per patient (grey=2, brown=>2) ranked in ascending order of burden of copy number alteration identified in the index core. Metastatic status of each patient is annotated (green=non-metastatic, blue=metastatic). Burden of copy number alteration (PGA=percentage genome altered) is represented as a bar with each patient split by proportion of gain (red) and loss (blue). Bottom bar chart represents number of diagnostic cores sequenced per patient. B Distribution of variance (%) of burden of copy number alteration per patient compared between non-metastatic (green violin plot) and metastatic (blue violin plot). Dot size represents number of cores sequenced per patient. C Boxplot demonstrating intra-patient heterogeneity of selected regions of interest. Regions annotated on x-axis labelled with chromosome number and genomic location mapped to cytoband. Blue=segment loss, red=segment gain. Within each bar, patients are only included if we sequenced more than one core and they harbour at least one core with the annotated alteration (numbers of patients annotated beneath x-axis). Y-axis represents the percentage of cores within patients harbouring the alteration and the boxplot line represents the median across patients