Lineage Relationship of Gleason Patterns in Gleason Score 7 Prostate Cancer

Published OnlineFirst May 21, 2013; DOI: 10.1158/0008-5472.CAN-12-2803

Cancer Molecular and Cellular Pathobiology Research

Irina V. Kovtun1, John C. Cheville2, Stephen J. Murphy3, Sarah H. Johnson3, Shabnam Zarei3, Farhad Kosari3, William R. Sukov2, R. Jeffrey Karnes4, and George Vasmatzis3

Abstract Gleason score 7 (GS7) prostate cancer [tumors with both Gleason patterns 3 (GP3) and 4 (GP4)] portends a significantly more aggressive tumor than Gleason score 6 (GS6). It is, therefore, critical to understand the molecular relationship of adjacent GP3 and GP4 tumor cell populations and relate molecular abnormalities to disease progression. To decipher molecular relatedness, we used laser capture microdissection (LCM) and whole- genome amplification (WGA) to separately collect and amplify DNA from adjacent GP3 and GP4 cell populations from 14 cases of GS7 prostate cancer. We then carried out massively parallel mate-pair next generation sequencing (NGS) to examine the landscape of large chromosomal alterations. We identified four to 115 DNA breakpoints in GP3 and 17 to 480 in GP4. Our findings indicate that while GP3 and GP4 from the same tumor each possess unique breakpoints, they also share identical ones, indicating a common origin. Approximately 300 chromosomal breakpoints were localized to the regions affected in at least two tumors, whereas more than 3,000 were unique within the set of 14 tumors. TMPRSS2–ERG was the most recurrent rearrangement present in eight cases, in both GP3 and GP4. PTEN rearrangements were found in five of eight TMPRSS2–ERG fusion–positive cases in both GP3 and GP4. Hierarchical clustering analysis revealed that GP3 has greater breakpoint similarity to its partner GP4 compared with GP3 from different patients. We show evidence that LCM, WGA, and NGS of adjacent tumor regions provide an important tool in deciphering lineage relationships and discovering chromosomal alterations associated with tumor progression. Cancer Res; 73(11); 1–10. 2013 AACR.

Introduction molecular changes, such as deregulation of PTEN have also – The elucidation of causal DNA rearrangements is critical to been implicated in prostate cancer (7 12). Despite the discov- the understanding of cancer development and progression, ery of some recurrent DNA rearrangements, association of and clinically relevant in regards to prognosis and treatment. these changes with clinical and pathologic features, particu- Although, the precursor lesions for prostate cancer are not larly with tumor grade, has been limited. completely defined (1, 2), significant progress has been made in Morphologic heterogeneity in prostate cancer, assessed with identifying the molecular events underlying initiation and Gleason pattern (GP) and the summation score (GS; refs. 13, progression using genomic and functional studies. The most 14), is strongly related to prognosis. Gleason score 6 (GS6) frequent chromosomal rearrangement in prostate cancer, the prostate cancer is either cured by treatment or needs no fi TMPRSS2–ERG gene fusion, is considered to be an early event treatment at all (15, 16). In up to 30% of cases, the nding of fi in prostate tumorigenesis. The significance of TMPRSS2–ERG GS6 on needle biopsy specimen is associated with insigni cant fusion protein as a prognostic or predictive marker has been cancer. In contrast, men with Gleason score 7 (GS7) and higher fi controversial (3–5), but recent evidence suggests that ERG have signi cant prostate cancer, and are at increased risk of status defines unique subtypes of prostate cancer (6). Other cancer progression. Despite the close association of GP to tumor behavior, the relationship of morphologic heterogeneity and molecular heterogeneity has not been elucidated. While there is a possibility that GP3 and GP4 regions in GS7 prostate Authors' Affiliations: Departments of 1Molecular Pharmacology and Experimental Therapeutics, 2Laboratory Medicine and Pathology, 3Molec- cancer are of different origin, it is generally accepted that GP4 ular Medicine, and 4Urology, Mayo Clinic, Rochester, Minnesota adjacent to GP3 are related and that GP3 progresses to GP4, Note: Supplementary data for this article are available at Cancer Research which shows more aggressive behavior. However, the transi- Online (http://cancerres.aacrjournals.org/). tion supposition has only been inferred by determining the I.V. Kovtun and J.C. Cheville contributed equally to this work. ERG status using FISH or immunostaining. In addition, there is a poor understanding of genomic differences between GP3 that Corresponding Authors: George Vasmatzis, Department of Molecular Medicine, Mayo Clinic, 200 First Street SE, Rochester, MN 55905. Phone: progresses and GP3 that does not. 507-284-2511; Fax: 507-266-5193; E-mail: [email protected]; Several studies have addressed the issue of tumor hetero- and John C. Cheville, E-mail: [email protected] geneity in prostate cancer by comparing changes in multiple doi: 10.1158/0008-5472.CAN-12-2803 separate tumor nodules within a prostate gland (17–21). 2013 American Association for Cancer Research. Assessment of the occurrence of the TMPRSS2–ERG fusion by

www.aacrjournals.org OF1

Kovtun et al.

FISH in multifocal prostate cancer show that 40% to 50% of foci variability in chromosomal rearrangements between different are discordant in regards to the presence or absence of the GPs within the same tumor and to determine whether these fusion, but within any one focus the absence or presence of the tumors exhibited a common origin. To investigate lineage fusion is consistent (18–21). These studies support the notion relationships between adjacent GPs, we carried out laser that multifocal primary prostate cancers are composed of capture microdissection (LCM), whole-genome amplification multiple clones. A study of metastatic prostate cancer using (WGA), and massively parallel mate-pair next generation high-resolution genome-wide single polymorphism and copy sequencing (NGS), and compared molecular alterations in GP3 number survey approach has shown the lineage relationship of and GP4 within the same tumor and between tumors from metastases to the primary tumor (22). different patients. Although, it is accepted that multiple separate tumors within the prostate gland arise from different clones, there is Materials and Methods no data regarding a common origin of different GPs within a Isolation of prostate GP3 and GP4 DNA, mate-pair unifocal dominant tumor. Indeed, no study has conclusively library construction and sequencing shown whether cells within a single tumor harbor identical Fourteen cases of fresh-frozen GS7 prostate cancer com- molecular alterations. For instance, the TMPRSS2–ERG fusion posed of GP3 adjacent to GP4 were selected for study. The can occur at a number of breakpoints that indicate a dif- tumors were the dominant tumor nodules within each gland. ferent clonal origin, so that identification of the fusion by In each case, 10 mm unstained sections were cut from fresh- FISH or by immunoperoxidase staining in different GPs within frozen blocks of tissue of patient with prostate cancer collected the same tumor does not conclusively prove a common origin from a radical prostatectomy specimen. LCM (Arcturus) was as these studies are not specific to any breakpoint. Moreover, used to separately isolate cells specific to the GP3 and GP4 molecular changes in the context of morphologic heterogene- populations minimizing contamination of adjacent cells ity within a tumor, specifically related to the differences within (Fig. 1A). Histologically normal glands were additionally col- GPs, have not been analyzed. It is unclear whether molecular lected from same slides at distances minimizing the chances of events characteristic of aggressive GP4 are present in the contaminating with the tumor tissues. The cells were lysed associated GP3 in GS7 cancer. It is also unknown whether the directly on the cap, and whole-genome DNA sequencing was chromosomal changes in the associated GP3 in GS7 cancer are carried out as previously described (23). A mate-pair library more similar to GP3 from other patients or more similar to its was prepared using the Mate-pair Library Prep Kit (Illumina), associated GP4. In this study, we identified patients with a following the manufacturer's recommendations. Briefly, 10 mg unifocal GP7 prostate cancer to determine the extent of of WGA DNA was fragmented using Covaris sonication to 4 and

A B

GP4 GP3 GP3

GP4

GP3 GP4

Figure 1. An example of GS7 prostate cancer specimen used for LCM. A, hematoxylin and eosin staining. B, microdissected areas corresponding to pattern GP3 (C) and GP4 (D), light images.

OF2 Cancer Res; 73(11) June 1, 2013 Cancer Research

Relationship of Gleason Patterns in Gleason Score 7 Prostate Cancer

5 kb, fragments and mate-pair libraries were prepared as FISH previously described and sequenced at one or 2 libraries per FISH studies were conducted on formalin-fixed, paraffin- lane on the Illumina GAII or HiSeq platforms, respectively (23). embedded tissue using a probe designed to assess PTEN status. Bacterial artificial chromosomes RP11-165M8, RP11-380G5, Bioinformatics analysis RP11-793O2, and RP11-348E17 selected using publicly avail- A set of algorithms developed to detect unsuspected large able database (28) were obtained from Invitrogen. DNA was chromosomal aberrations was used (24, 25). The read-to- isolated following a standard protocol using a Qiagen Plasmid reference–genome–mapping algorithm (24) was modified to Purification Kit (Qiagen). The extracted DNA was fluorescently map both mate-pair reads across the whole genome as recently labeled in Spectrum Orange-dUTP via nick translation (Abbott described (25). We used the protocol that allowed to sequence Molecular) and was tested on normal male blood metaphases. the ends of large fragments of genomic DNA (2.5–5KB), thus PCR was conducted to confirm the clones covered the regions "effectively covering" breakpoints about 15 on the average. of interest. Centromere 10 status was assessed using a com- However, the coverage at the nucleotide level was usually mercially available probe (Abbott Molecular). Tissue sections less than 5, precluding us from making specific comments 5-mm thick were deparafinized in a 90C oven, immersed twice regarding mutation calls (Supplementary Fig. S1). Breakpoints in xylene and twice in 100% ethanol, dried, the region to be covered by at least 3 mate-pairs in each sample were collected probed marked, and then microwaved in 10 mmol/L citric acid for further analysis. for 10 minutes. The slides were subsequently pretreated with Agglomerative hierarchical clustering was used to quantify 2x saline-sodium citrate buffer for 5 minutes, digested for 40 phylogenic relationships between samples. A distance matrix minutes with DigestALL (Invitrogen), and washed. Following was built to capture genetic distance between all possible pairs. dehydration and drying, the slides were denatured at 80C for A distance between 2 samples was defined as 1.1 raised to the 5 minutes and hybridized at 37C overnight in a humidified power of the lesser number of the unique abnormalities incubator. DAPI1 (10%; Abbott Molecular) was used for count- multiplied by 0.1, minus the number of shared abnormalities er staining. multiplied by 2. Clustering used the "hclust" function in R package using the McQuitty Similarity Analysis method. Results To analyze whether the breakpoints correlate with binding To identify genetic similarities and differences that occur sites for ERG, androgen receptor (AR), or cluster within open or in GP3 associated with GP4, we used LCM to collect cells closed chromatin, we used published ChIP-Seq data (26) for the within GP3 and adjacent GP4 in 14 cases of fresh-frozen AR, ERG, RNA polymerase II binding sites and trimethylated prostate cancer. Figure 1 shows a representative example of histone H3K4, trimethylated histone H3K36, trimethylated the hematoxylin and eosin-stained region of GP3 and GP4 histone H3K9, and trimethylated histone H3K27 chromatin tumor in the frozen tissue samples (Fig. 1A), together with the marks. Data were downloaded from Gene Expression Omnibus section image after LCM (Fig. 1B) and the captured GP3 and (27) Intervals of 50 kb (as in ref. 12) surrounding correspond- GP4 cells engrafted on the LCM caps (Fig. 1C and D, respec- ing sets of ChIP-Seq peaks were used to estimate depletion or tively). The DNA was amplified directly from the engrafted cells enrichment of the breakpoints. Significance of enrichment or (23) and massively parallel mate-pair library sequencing was depletion of observed breakpoints compared with background carried out. Binary algorithms developed in our group (24, 25) was calculated according to the binomial distribution. were used to analyze the data and detect large chromosomal abnormalities including translocations, deletions, and inser- Validation by PCR and Sanger sequencing tions. The discovered breakpoints were localized to the genes PCR was conducted on WGA DNA using combinations of as well as intragenic regions. Interchromosomal translocations breakpoint-specific primers (sequences are listed in Supple- and intrachromosomal rearrangements including amplifica- mentary Table S1) using the HotStar Taq DNA Polymerase Kit tions, deletions, or inversions were equally represented in GP3 (Qiagen), the Long template PCR Kit (Roche), or EasyA High and GP4 within the same tumors. Clusters of breakpoints Fidelity polymerase (Stratagene) on a GeneAmp PCR System encompassing areas of a few megabases were also observed 9700 (Applied Biosystems). Amplification products were visu- in several samples (16/28 samples), suggesting consequences alized on a 0.8% to 1% agarose gel, single bands were excised, of chromothripsis, a recently described phenomenon (29–31). DNA was extracted, and Sanger sequenced using the appro- In 7 cases, both GP3 and GP4 showed similar patterns for priate primers on a 3730 1 DNA Analyzer (Applied Biosys- chromothripsis, whereas 2 cases showed chromothripsis only tems). A mixed pool Genomic DNA sample was used as a in GP4. normal DNA control (Promega). The number of genomic breakpoints (i.e., junctions of the rearranged chromosomal regions) ranged from 10 to 180 per Immunostaining tumor between the 28 GP3 and GP4 samples, and significant Sections of 5 mm tissue sections were incubated in EDTA- variation was observed between adjacent GP3 and GP4 within containing antigen retrieval buffer for 30 minutes, followed by the same tumor (Fig. 2A). Evaluation of genomic breakpoints application of primary antibody (ERG clone 9FY, Biocare) at showed that recurrent events were uncommon between 1:25 dilution for 32 minutes at 37C. Immunohistochemical patient tumors (Fig. 2B), but were common between adjacent staining was completed on the Ventana Benchmark XT auto- GP3 and GP4 within the same tumor. Specifically, within the 14 mated platform using standard reagents. patient tissues, 176 chromosomal breakpoints were shared by

www.aacrjournals.org Cancer Res; 73(11) June 1, 2013 OF3

Kovtun et al.

A 200 GP4 GP3 Shared

150 Figure 2. Commonality and differences in chromosomal alterations in paired GP3 and GP4

100 in 14 patients with prostate cancer. A, a graphical representation of the extent of similarity of

500 subpopulations of cells from the same patient in terms of large chromosomal abnormalities. The number of alterations is shown PR22 PR7 PR13 PR21 PR25 PR10 PR11 PR24 PR26 PR6 PR23 PR27 PR12 PR28 on the y-axis, and the cases are depicted on the x-axis. B, B molecular intertumoral and

PR6 PR10 PR12 PR22 PR25 PR27 PR7 PR11 PR13 PR21 PR23 PR24 PR26 PR28 intratumoral heterogeneity in ERG 14 patients with prostate cancer. TMPRSS2 Each row represents a gene locus, PTEN and each column is GS7 prostate SP3 FAM126B cancer case. Blue squares indicate NDUFB3 FOXP1 that alteration is present in both EIF4E3 GP3 and GP4, green squares PDE4D indicate that alteration is present TRDN only in GP3, and red squares PACRG indicate that alteration is present LAPTM4B VPS13B only in GP4. PR indicates a case RIMS2 number. ATAD1 CFLP1 RNLS LIPJ GPC5 AVEN RORA CHST9 TTC28

adjacent GP3 and GP4 pairs, whereas more than 1,000 events observed. Among these were vacuolar sorting protein 13BB were unique to individual GS7 tumors (Fig. 2). (VPS13B), transcription factor SP3, integral membrane protein As expected, ERG was the most frequently rearranged gene, triadin (TRDN), and cAMP-speciﬁc phosphodiesterase 4D present in 8 of the 14 cases (Fig. 2B). Deletion of PTEN and (PDE4D; Fig. 2B). Interestingly, breakpoints at the SP3 loci in FOXP1, 2 genes often altered in prostate cancer (12, 32–35), 2 cases were present in both GP3 and GP4 in affected tumors, were next most common, present in 4 and 3 cases, respectively whereas mate-pair reads involving TRDN gene were found (Fig. 2B). In addition, recurrent alterations involving genes that either only in GP4 or only in GP3, and breakpoints at have not been previously implicated in prostate cancer were PDE4D were only present in GP4. This suggests that some 0.8 Figure 3. Phylogenic relationships between the tumors and between GPs within the tumor. Height PR12_GP3 PR12_GP4 PR28_GP3 PR28_GP4 0.4 Agglomerative hierarchical clustering was used to quantify relationships between samples. A

PR27_GP3 PR27_GP4 distance matrix was built to PR23_GP3 PR23_GP4 capture genetic distance between PR6_GP3 PR6_GP4 0.0 all possible pairs. PR26_GP3 PR26_GP4 PR24_GP3 PR24_GP4 PR11_GP3 PR11_GP4 PR10_GP3 PR10_GP4 PR25_GP3 PR25_GP4 PR7_GP3 PR7_GP4 PR21_GP3 PR21_GP4 PR13_GP3 PR13_GP4 PR22_GP3 PR22_GP4

OF4 Cancer Res; 73(11) June 1, 2013 Cancer Research

Relationship of Gleason Patterns in Gleason Score 7 Prostate Cancer

A Schematic of ERG-TMPRSS2 rearrangements ERG TMPRSS2 39,817.5 39,870.3 - 39,870.4 39,947.6 42,861.5 42,870.0 - 42,870.1 42,879.9

Exon 4 Exon 3 Exon 3 Exon 2 Exon 1 PR7 39,827 42,874 PR11 39,878 42,871 PR13 39,835 42,869 PR21 39,861 42,880 PR23 39,873 42,876 PR24 39,827 42,871 PR26 39,874 42,883 PR28 39,847 42,871

B ERG fusion gene junctions C ERG staining Ladder ANGN GP3 GP4 PR 26 Case PR28 Gleason pattern 4 TMPRSS2 ERG

PR28

42870192 39848620 chromosome 21 TMPRSS2 ERG

Normal gland PR24 Gleason pattern 3

42867740 39826975 chromosome 21

Figure 4. Characterization of ERG gene fusion status in adjacent GPs in the same tumor. A, a schematic of putative TMPRSS2–ERG rearrangements in all ERG gene fusion-positive cases as determined by mate-pair analysis. B, representative chromatograms of sequence analysis of the breakpoints (identical in GP3 and GP4 of depicted case) involving ERG and TMPRSS2 in 2 selected cases. The position on the chromosome is indicated. C, shows ERG immunostaining in depicted case. Arrows point to GP3 and GP4 of the tumor as well as to AN. GP3 and GP4 are adjacent GP3 and GP4, respectively. AN indicates adjacent histologically normal tissue and GN is normal genomic DNA. rearrangements play a role in early steps of tumorigenesis, In each of the GP3 and GP4 pairs from the 14 cases studied, a whereas others play a greater role in disease progression. We set of shared recombination events was predicted by our also compared genes that harbored breakpoints, affected in 2 algorithm (Fig. 2A). This observation is consistent with the or more prostate tumors in our dataset, to the list of genes notion that the adjacent GP share a common origin. The available in cosmic Census database (36) that are accepted as number of shared rearrangements, however, varied extensively important "drivers" in tumorigenesis (Supplementary Table between the GP within each tumor and did not correlate with S2). A total of 9 genes recurrently affected in our 14 patients the total numbers of abnormalities in each tumor. Likewise, the with GS7 prostate cancer have been previously annotated as number of breakpoints between ERG-positive and ERG-nega- "drivers" in cosmic database. These included ERG–TMPRSS2 tive cases did not significantly differ (Fig. 2A). Interestingly, 3 fusion, PTEN deletion, immunoglobulin kappa locus rearran- cases PR12, PR27, and PR28, presented little commonality gements (IGK@), FOXP1 translocation, ALK translocation, between GP3 and GP4, raising the possibility that the adjacent transcription factor 12 translocation (TCF12), p53 deletion, patterns in these cases may have developed independently (Fig. and musashi homologue 2 (MSI2) translocation (Supplemen- 2A). The majority of cases showed a significant number of tary Table S2A). Four more genes (Supplementary Table S2B) common breakpoints between GP3 and GP4 within a tumor, are likely to have similar functional importance as their close indicating a common origin. In 3 cases, PR11, PR12, and PR28, homologs listed as "drivers" in cosmic database. We consider no breakpoints unique to GP3 were identified (Fig. 2A). Cases the rest of the genes shown in Fig. 2 as potential novel PR6, PR24, and PR26 showed just a small number of unique contributors (drivers) to tumorigenesis in prostate cancer, GP3 events. In contrast, significant numbers of unique chro- whereas their specific role is to be identified in future studies. mosomal rearrangements were observed in the GP4 in each

www.aacrjournals.org Cancer Res; 73(11) June 1, 2013 OF5

Kovtun et al.

Table 1. Occurrence of validated breakpoints across different Gleason patterns

Case PR28 Case PR12 Case PR27

AN G3 G4 AN G3 G4 AN G3 G4 Event Genes involved # Chr 21 deletion ERG– Chr13 deletion STARD13 Chr7 deletion no gene mate-pairs Present TMPRSS2 n/a 8 20 n/a 0 20 n/a 0 16 þþþþþþ Event genes involved # Ch 8-Chr 11 KCNK9-CLMP Chr4-Chr13 PDS5B Chr2-Chr9 no gene mate-pairs present n/a 1 14 n/a 0 8 n/a 1 22 n/a þþþþþþ Event genes involved # Chr12-Chr22 CACNA1C– Chr4 deletion SLIT2 Chr6-Chr8a TUBB mate-pairs present ZNRF3 n/a 0 27 n/a 0 11 n/a 10 9 þþþþþþþ Event genes involved # Chr8-Chr6a TUBB mate-pairs present n/a 10 9 þþþ Event genes involved # Chr6 deletion TRDN mate-pairs present n/a 8 0 þ

Abbreviation: n/a, not tested aTwo breakpoints representing balanced translocation.

case, with PR22 and PR27 showing a high number of abnor- histologically normal tissue (AN), we postulate that TMPRSS2– malities specific to GP4. Only case PR10 presented more ERG gene fusion is a strong marker of lineage relationship for unique breakpoints in the GP3 as compared with GP4 (Fig. 2A). adjacent GPs. A high degree of commonality between adjacent GP3 and Previously, it has been reported that AR and ERG protein GP4 was also supported by independent agglomerative hier- cooperate to facilitate the formation of DNA double-strand archical clustering analysis (Fig. 3). This analysis showed that breaks (37, 38). In addition, breakpoints were shown enriched the GP3 is more closely associated with its associated GP4 than near open chromatin, AR, and ERG DNA-binding sites in the with GP3 from other patient's tumors. As expected, the GP3 setting of the TMPRSS2–ERG fusion, but inversely correlated and GP4 in PR12, PR28, and PR27 were the least related, as with these regions in tumors lacking ETS fusions (12). There- reflected by their height scores in the dendrogram (Fig. 3). fore, we investigated whether the breakpoints in our set of Conventional PCR followed by Sanger sequencing was used cases correlated with binding sites for ERG, AR, or cluster to identify the exact structure of the chromosomal fusions within open or closed chromatin. By using published ChIP-Seq predicted by our bioinformatics algorithm. We first validated data (26) for the AR, ERG, RNA polymerase II binding sites and ERG fusion events for which mate-pair reads were always histone modifications, marks for open and closed chromatin, observed in adjacent GP3 and GP4 (Figs. 2 and 4A), and in we examined their association with our breakpoint data. A these cases, GP3 and GP4 shared the exact breakpoints for significant enrichment for AR and ERG-binding sites in the TMPRSS2–ERG, indicating a common origin (Fig. 4 and Sup- vicinity of the breakpoints in half of ERG-positive cases was plementary Fig. S2). Importantly, DNA from the adjacent observed (Supplementary Fig. 3SA and S3B). In case, PR21 G4 histologically normal tissue (AN) did not contain the ERG depletion for both AR and ERG CHIP-Seq peaks was found. fusion events. Both GPs in each tumor showed strong ERG ERG-negative cases did not show a consistent trend. With the protein expression (Fig. 4C and Supplementary Fig. S2). The exception of GP4 in PR27, in which a depletion for each of the 3 specific position of the breakpoint within the ERG gene varied binding sites was found, no significant correlation was between cases, but localized within in the commonly reported observed for any CHIP-Seq peaks in ERG-negative cases. intron 2 or 3 regions (Fig. 4A). Similarly, each position of the Trimethylated histone H3K4 and trimethylated histone TMPRSS2 breakpoint predicted the association of the promot- H3K36 (Supplementary Fig. 3SD), marks of open chromatin, er and resultant androgen responsive expression of the resul- were enriched in the majority of ERG-positive cases with a tant ERG fusion gene product. On the basis of comparison of significant P value. A few ERG-negative cases have also shown TMPRSS2–ERG gene fusion status in GS7 cases and adjacent an enrichment for H3K4me3, in contrast with the previously

OF6 Cancer Res; 73(11) June 1, 2013 Cancer Research

Relationship of Gleason Patterns in Gleason Score 7 Prostate Cancer

published data (12). Only distribution of breakpoints in ERG- clones of the common predecessor. We also compared the negative case PR27 GP4 positively correlated with closed relationship between GP3 and GP4 in 2 cases, PR10 and PR11 chromatin marks H3K9me3 and H3K27me3 (Supplementary that had a large number of shared breakpoints by mate-pair Fig. 3S3E) and negatively correlated with marks of the open analysis (Fig. 2A). A validation of selected alterations revealed chromatin (Supplementary Fig. S3D). their presence in both GP3 and GP4 and their absence in To better understand the relationship of GP3 and GP4 in the histologically normal tissue (AN) correlating with mate-pair cases that showed little or no commonality by mate-pair analysis (Fig. 5D). The shown deletion in chromosome 1 analysis (Fig. 3B), we validated a set of selected breakpoints resulted in a fusion between the first exon of LPAR3 gene, (Table 1). In PR28, the identical TMPRSS2–ERG fusion and an coding for a G protein-coupled receptor family member highly event involving chromosomes 8 and 11 was validated in both expressed in normal prostate, and zinc finger factor ZZZ3 gene the GP3 and GP4 but was not present in AN (Fig. 4B and Fig. 5A, (Fig. 5D). Unlike cases in which GP3 and GP4 did not cluster respectively). The STARD13 gene on chromosome 13 was well together, GPs showing more commonality by the mate- validated in GP3, GP4, and AN of case PR12 (Fig. 5B). Similar pair analysis had lineage relationship. findings were shown for the chromosome 4 and 13 rearrange- To show heterogeneity/homogeneity of the cell population ment involving PDS5B in this case (Fig. 5B). For both break- for a given alteration within a single GP area, we conducted points, more product was observed in GP4 than in the asso- FISH analysis. A specific probe for PTEN was used to compare ciated GP3 and AN, suggesting that a smaller fraction of cells in the signal in adjacent histologically normal cells, GP3 and GP4 the latter 2 populations harbor the alteration. Because the cells (Fig. 6). Consistent with our validation results using PCR and from each population were isolated by LCM and DNA was Sanger sequencing (Fig. 6A), cells in both GP3 and GP4 showed processed independently, the chance of cross-contamination heterozygous deletion of PTEN, whereas adjacent histologi- was very unlikely. In addition, for a subset of rearrangements, cally normal cells harbored intact signals for both PTEN alleles we have conducted a semiquantitative PCR with varying (Fig. 6D). Because cells from histologically normal tissue, GP3 number of cycles (Supplementary Fig. S4). There was a corre- and GP4 for genomic analysis of alterations were collected lation between the amount of the amplified product and the individually using LCM and processed separately, we conclud- number of cycles, indicating that a different proportion of cells ed that the presence of certain rearrangements in adjacent to within samples harbor the alteration. tumor histologically normal cells is a true observation and not Individual rearrangements in case PR27 showed various a result of DNA contamination. distributions (Fig. 5C). Interestingly, a few breakpoints in PR27 were present only in GP3, a few only in GP4, and several were Discussion detected in GP3, GP4, as well as adjacent histologically normal. This is the first study to examine the chromosomal altera- As no events were identified shared by GP3 and GP4 exclu- tions in adjacent GPs within the same tumor using a unique sively, this raised a possibility that they represent different approach enabling interrogation of small tumor samples with

A Case PR28 C Case PR27 D Case PR11 Chr 12-Chr 22 event Chr 2-Chr 9 event Chr 1 event Ladder GP4GP3ANGN LadderGN GP3 GP4 AN Ladder GN GP3AN GP4 Figure 5. Validation analysis of selected breakpoints identified by mate-pair in the selected GS7 cases. Chr8-Chr 11 event – Chr 7 event A C, representative gel images of Ladder GN GP3 GP4 AN validated breakpoints between ANGNGP3GP4 Ladder adjacent GP3 and GP4 patterns. GN LPAR3 ZZZ3 is normal genomic DNA, AN is adjacent histologically normal tissue, collected independently, and B Case PR12 Chr 6 event GP3 and GP4 are adjacent GPs. D, ANGN GP3 GP4 Ladder validation of a deletion on Chr 13, STARD13 event chromosome 1 identified in PR11, Ladder GP3ANGN GP4 85351520 78124122 which is classified as a case with Chromosome 1 Chromosome 1 greater commonality between GP3 Chr 8-Chr 8 event and GP4. Designations are as Chr 4-Chr 13 event ANGN GP3 GP4 Ladder – described in A C. A chromatogram Ladder GP3ANGN GP4 illustrating sequence analysis and a schematic of the putative fusion ZZZ3 LPAR3 gene are also shown. Chr 6-Chr 8 event Chr 4 event GN GP3 GP4AN Ladder GN AN GP4GP3 Ladder Exon 15 Exon 1 Exon 2 Exon 1 Deletion

www.aacrjournals.org Cancer Res; 73(11) June 1, 2013 OF7

Kovtun et al.

A B

Breakpoint Breakpoint

Chr 10-Chr 10 event GNGP4GP3 Case PR23

Figure 6. FISH analysis of PTEN deletion in GS7 prostate cancer case. A, a schematic showing a map of the region on chromosome 10 containing PTEN gene (top). Speciﬁc breakpoints validated by PCR (bottom) are shown by red arrows. G3, G4, and GN are GP3, GP4, and normal genomic DNA, respectively. B, normal prostatic duct with nuclei showing a normal FISH pattern with 2 intact PTEN (red) and 2 intact centromere 10 (green) signals (arrows). C and D, GP3 and 4 areas, respectively. Arrows point to the cells with PTEN loss (only one red signal is present).

great specificity. Our findings indicate that while GP3 and GP4 SNVs. SNVs are known to be present not only in tumors, but from the same tumor each possess unique breakpoints, they also in normal tissue and normal peripheral blood. This also share identical ones, indicating a common origin. In requires an additional profiling of blood and normal tissue addition, defining these lineage relationships allows a better adjacent and distant to the tumor to decipher somatic tumor- understanding of tumor progression. igenic changes. In contrast, chromosomal rearrangements, Characterization of lineage (i.e., descent from a common related to tumorigenicity, are usually found in the tumor but precursor) relationships between concurrent GP3 and GP4 in not in blood. Therefore, in our efforts to show tumoral lineage, prostate cancer in the past has been hampered by the difficulty we took advantage of sequencing mate-pair libraries to identify in separately obtaining pathologically well-characterized cell unique abnormalities between different cell populations in the populations in sufficient quantity. Although large-scale same tumor. sequencing efforts have already been applied to prostate We analyzed 14 cases of GS7 prostate cancer and found a cancer, none of these studies has focused on progression from significant heterogeneity in GPs between different patient's GP3 to GP4. Whole-genome sequencing of several cancers tumors in agreement with previous reports (11,12). Recurrent revealed that solid tumors harbor many chromosomal rear- chromosomal alterations were not common and much less rangements and thousands of single nucleotide variations frequent than chromosomal changes unique to a specific (SNV). We have used both strategies to investigate lineage tumor. Similar observations were made in a recent study relationships between phenotypically different parts of a by Berger and colleagues (12) where complete sequencing of tumor from a same individual by enumerating common altera- 7 cases of GS7 cancer showed numerous chromosomal tions that are not found in normal distant tissue. We found that changes as well as point mutations. However, point mutations using analysis of chromosomal rearrangements helps to elu- within GP3 and GP4 were not examined in the Berger study. cidate lineage relationships in more specific manner than Our findings indicate that GS7 tumors show significant

OF8 Cancer Res; 73(11) June 1, 2013 Cancer Research

Relationship of Gleason Patterns in Gleason Score 7 Prostate Cancer

alized medicine (39), the application NGS in conjunction with Scenarios of prostate cancer initiation and progression careful selection subset of relevant population of cells for A B C interrogation seems to be a useful tool for identification of GP4 GP4 GP4 lineage relationship and recurrent molecular events within heterogeneous tumors (40). Such knowledge can be of great value in stratifying "indolent" from "significant" prostate cancers and can help to improve patient management, particularly GP3 GP3 GP3 when applied to needle biopsy specimens that sample only a very small portion of tumor. In conclusion, we used LCM, massively parallel NGS and a N N GP3 N unique bioinformatics algorithm to better define the genomic landscape within and between GS7 prostate cancer. In most cases, the GPs within each tumor had a common origin and were more similar to each other than GPs in other patient's tumors. Future work will help to determine the use of discov- Figure 7. A schematic showing different scenarios of prostate cancer formation and progression based on the presence of molecular change. ered molecular markers in the diagnosis of "indolent" and N is nonneoplastic tissue. "significant" prostate cancers.

Disclosure of Potential Conflicts of Interest heterogeneity within GP3 and GP4, yet share identical genomic No potential conflicts of interest were disclosed. alterations, indicating a common origin. Moreover, we show that in the majority of cases, GP3 shares more similarities with Authors' Contributions its partner GP4 than GP3 from other patients' tumors (Fig. 2 Conception and design: I.V. Kovtun, J.C. Cheville, S.J. Murphy, R.J. Karnes, G. Vasmatzis and 3). Some of the molecular changes, such as loss of PTEN, Development of methodology: I.V. Kovtun, J.C. Cheville, S.J. Murphy, R.J. were evident in GP3 as well as GP4, indicating that genomic Karnes, G. Vasmatzis Acquisition of data (provided animals, acquired and managed patients, changes associated with tumor aggressiveness occur before the provided facilities, etc.): I.V. Kovtun, J.C. Cheville, S.J. Murphy, S. Zarei, W.R. morphologic transition from GP3 to GP4. Sukov On the basis of our molecular analyses of associated GPs, we Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): I.V. Kovtun, R.J. Karnes, G. Vasmatzis propose a model to explain the role of chromosomal rearran- Writing, review, and/or revision of the manuscript: I.V. Kovtun, J.C. gements in cancer initiation and progression (Fig. 7). In the Cheville, S.J. Murphy, S.H. Johnson, F. Kosari, R.J. Karnes, G. Vasmatzis first scenario, a subpopulation of cells with identical molecular Administrative, technical, or material support (i.e., reporting or orga- nizing data, constructing databases): S.H. Johnson, G. Vasmatzis alterations within heterogeneous GP3 gives rise to GP4 in Study supervision: R.J. Karnes, G. Vasmatzis which cells acquire more changes as the tumor progresses (Fig. 7A). PR28 and PR11 (Figs. 4B and 5D, respectively) Grant Support represent examples of this scenario. Alternatively, GP3 and This work was supported by a Waterman Biomarker Discovery grant. Funding for this work was also provided by the Center of Individualized Medicine, the GP4 might arise from different clones by using different subsets office of Intellectual Property, and the Department of Laboratory Medicine and of molecular changes (Fig. 7B). Finally, a specific subset of Pathology at Mayo Clinic. Tissues were provided by the Specialized Program of molecular rearrangements in adjacent nonneoplastic area Research Excellence in Prostate Cancer grant from the National Cancer Institute, NIH. leads to transformation into GP3, which does not progress The costs of publication of this article were defrayed in part by the and becomes a "dead end" (Fig. 7C). A different subset of payment of page charges. This article must therefore be hereby marked advertisement rearrangements gives rise to GP3, which progresses into GP4, in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. such as in PR27 (Fig. 5C). Although tumor heterogeneity within and between cancers Received July 18, 2012; revised February 14, 2013; accepted March 24, 2013; poses a number of problems related to treatment and person- published OnlineFirst May 21, 2013.

References 1. Greene DR, Wheeler TM, Egawa S, Weaver RP, Scardino PT. Rela- 5. Tomlins SA, Bjartell A, Chinnaiyan AM, Jenster G, Nam RK, Rubin MA, tionship between clinical stage and histological zone of origin in early et. al. ETS gene fusions in prostate cancer: from discovery to daily prostate cancer: morphometric analysis. Br J Urol 1991;68:499–509. clinical practice. Eur Urol 2009;56:275–86. 2. Cheng L, Song SY, Pretlow TG, Abdul-Karim FW, Kung HJ, Dawson 6. Karnes RJ, Cheville JC, Ida CM, Sebo TJ, Nair AA, Tang H, et al. The DV, et al. Evidence of independent origin of multiple tumors from ability of biomarkers to predict systemic progression in men with high- patients with prostate cancer. J Natl Cancer Inst 1998;90:233–37. risk prostate cancer treated surgically is dependent on ERG status. 3. FitzGerald LM, Agalliu I, Johnson K, Miller MA, Kwon EM, Hurtado-Coll Cancer Res 2010;70:8994–9002. A, et al. Association of TMPRSS2–ERG gene fusion with clinical 7. Yoshimoto M, Cunha IW, Coudry RA, Fonseca FP, Torres CH, Soares characteristics and outcomes: results from a population-based study FA, et al. FISH analysis of 107 prostate cancers shows that PTEN of prostate cancer. BMC Cancer 2008;8:230. genomic deletion is associated with poor clinical outcome. Br J Cancer 4. Demichelis F, Fall K, Perner S, Andren O, Schmidt F, Setlur SR, et al. 2007;97:678–85. TMPRSS2:ERG gene fusion associated with lethal prostate cancer in a 8. Reid AH, Attard G, Ambroisine L, Fisher G, Kovacs G, Brewer D, et al. watchful waiting cohort. Oncogene 2007;26:4596–99. Molecular characterisation of ERG, ETV1 and PTEN gene loci identiﬁes

www.aacrjournals.org Cancer Res; 73(11) June 1, 2013 OF9

Kovtun et al.

patients at low and high risk of death from prostate cancer. Br J Cancer ALK-negative anaplastic large cell lymphomas by massively parallel 2010;102:678–84. genomic sequencing. Blood 2011;117:915–19. 9. Verhagen PC, van Duijn PW, Hermans KG, Looijenga LH, van Gurp RJ, 25. Vasmatzis G, Johnson SH, Knudson RA, Ketterling RP, Braggio E, Stoop H, et al. The PTEN gene in locally progressive prostate cancer is Fonseca R, et al. Genome-wide analysis reveals recurrent structural preferentially inactivated by bi-allelic gene deletion. J Pathol 2006;208: abnormalities of TP63 and other p53-related genes in peripheral T-cell 699–707. lymphomas. Blood 2012;120:2280–89. 10. Reid AH, Attard G, Brewer D, Miranda S, Riisnaes R, Clark J, et al. 26. Yu J, Yu J, Mani RS, Cao Q, Brenner CJ, Cao X, et al. An integrated Novel, gross chromosomal alterations involving PTEN cooperate with network of androgen receptor, polycomb, and TMPRSS2-ERG gene allelic loss in prostate cancer. Mod Pathol 2012;25:902–10. fusions in prostate cancer progression. Cancer Cell 2010;17:443–54. 11. Taylor BS, Schultz N, Hieronymus H, Gopalan A, Xiao Y, Carver BS, 27. http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc¼GSE14097 et al. Integrative genomic profiling of human prostate cancer. Cancer 28. http://genome.ucsc.edu, 2006, genome build hg18 Cell 2010;18:11–22. 29. Forment JV, Kaidi A, Jackson SP. Chromothripsis and cancer: causes 12. Berger MF, Lawrence MS, Demichelis F, Drier Y, Cibulskis K, Siva- and consequences of chromosome shattering. Nat Rev Cancer chenko AY, et al. The genomic complexity of primary human prostate 2012;12:663–70. cancer. Nature 2011;470:214–20. 30. Kloosterman WP, Tavakoli-Yaraki M, van Roosmalen MJ, van Binsber- 13. Bailar JC III, Mellinger GT, Gleason DF. Survival rates of patients with gen E, Renkens I, Duran K, et al. Constitutional chromothripsis rear- prostatic cancer, tumor stage, and differentiation—preliminary report. rangements involve clustered double-stranded DNA breaks and non- Cancer Chemother Rep 1966;3:129–36. homologous repair mechanisms. Cell Rep 2012;1:648–55. 14. Gleason DF. Classification of prostatic carcinomas. Cancer Che- 31. Stephens PJ, Greenman CD, Fu B, Yang F, Bignell GR, Mudie LJ, et al. mother Rep 1966;3:125–8. Massive genomic rearrangement acquired in a single catastrophic 15. Freedland SJ. Screening, risk assessment, and the approach to event during cancer development. Cell 2011;144:27–40. therapy in patients with prostate cancer. Cancer 2011;117:1123–35. 32. Feilotter HE, Nagai MA, Boag AH, Eng C, Mulligan LM. Analysis of 16. Albertsen PC. Treatment of localized prostate cancer: when is active PTEN and the 10q23 region in primary prostate carcinomas. Oncogene surveillance appropriate? Nat Rev Clin Oncol 2010;7:394–400. 1998;16:1743–48. 17. Bastacky SI, Wojno KJ, Walsh PC, Carmichael MJ, Epstein JI. Path- 33. Suzuki H, Freije D, Nusskern DR, Okami K, Cairns P, Sidransky D, et al. ological features of hereditary prostate cancer. J Urol 1995;153: Interfocal heterogeneity of PTEN/MMAC1 gene alterations in multiple 987–92. metastatic prostate cancer tissues. Cancer Res 1998;58:204–9. 18. Arora R, Koch MO, Eble JN, Ulbright TM, Li L, Cheng L. Heterogeneity 34. Wang SI, Parsons R, Ittmann M. Homozygous deletion of the PTEN of Gleason grade in multifocal adenocarcinoma of the prostate. Cancer tumor suppressor gene in a subset of prostate adenocarcinomas. Clin 2004;100:2362–66. Cancer Res 1998;4:811–15. 19. Andreoiu M, Cheng L. Multifocal prostate cancer: biologic, prognostic, 35. McMenamin ME, Soung P, Perera S, Kaplan I, Loda M, Sellers WR. and therapeutic implications. Hum Pathol 2010;41:781–93. Loss of PTEN expression in paraffin-embedded primary prostate 20. Barry M, Perner S, Demichelis F, Rubin MA. TMPRSS2–ERG fusion cancer correlates with high Gleason score and advanced stage. heterogeneity in multifocal prostate cancer: clinical and biologic impli- Cancer Res 1999;59:4291–96. cations. Urology 2007;70:630–33. 36. http://www.sanger.ac.uk/genetics/CGP/Census/ 21. Guo CC, Zuo G, Cao D, Troncoso P, Czerniak BA. Prostate cancer of 37. Lin C, Yang L, Tanasa B, Hutt K, Ju BG, Ohgi K, et al. Nuclear receptor- transition zone origin lacks TMPRSS2–ERG gene fusion. Mod Pathol induced chromosomal proximity and DNA breaks underlie specific 2009;22:866–71. translocations in cancer. Cell 2009;139:1069–83. 22. Liu W, Laitinen S, Khan S, Vihinen M, Kowalski J, Yu G, et al. Copy 38. Mani RS, Tomlins SA, Callahan K, Ghosh A, Nyati MK, Varambally S, number analysis indicates monoclonal origin of lethal metastatic et al. Induced chromosomal proximity and gene fusions in prostate prostate cancer. Nat Med 2009;15:559–65. cancer. Science 2009;326:1230. 23. Murphy S, Cheville JC, Zarei S, Johnson SH, Sikkink RA, Kosari F, et al. 39. Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos Mate pair sequencing of whole genome amplified DNA following laser E, et al. Intratumor heterogeneity and branched evolution revealed by capture microdissection of prostate cancer. DNA Res 2012;19: multiregion sequencing. N Engl J Med 2012;366:883–92. 395–406. 40. Barbieri CE, Demichelis F, Rubin MA. Molecular genetics of prostate 24. FeldmanAL,DoganA,SmithDI,LawME,AnsellSM,JohnsonSH, cancer: emerging appreciation of genetic complexity. Histopathology et al. Discovery of recurrent t(6;7)(p25.3;q32.3) translocations in 2012;60:187–98.

OF10 Cancer Res; 73(11) June 1, 2013 Cancer Research

Lineage Relationship of Gleason Patterns in Gleason Score 7 Prostate Cancer

Irina V. Kovtun, John C. Cheville, Stephen J. Murphy, et al.

Cancer Res Published OnlineFirst May 21, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-12-2803

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2021/03/20/0008-5472.CAN-12-2803.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2013/05/21/0008-5472.CAN-12-2803. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.