Tracking the Origin of Metastatic Prostate Cancer

EURURO-5845; No. of Pages 4 EUROPEAN UROLOGY XXX (2014) XXX–XXX

available at www.sciencedirect.com journal homepage: www.europeanurology.com

Platinum Priority – Brief Correspondence Editorial by XXX on pp. x–y of this issue

Tracking the Origin of Metastatic Prostate Cancer Johan Lindberg a,y, Anna Kristiansen b,y, Peter Wiklund c, Henrik Gro¨nberg d, Lars Egevad b,* a

Department of Medical Epidemiology and Biostatistics, Science for Life Laboratory, Karolinska Institutet, Stockholm, Sweden; b Department of Oncology-

Pathology, Karolinska Institutet, Stockholm, Sweden; d

c

Department of Urology, Division of Surgery, Karolinska Hospital, Stockholm, Sweden;

Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden

Article info

Abstract

Article history: Accepted September 6, 2014

Metastatic prostate cancer is a monoclonal disease. We previously failed to identify a common somatic denominator between primary tumor tissue and two lymph-node metastases by exome sequencing [Lindberg J, et al. Eur Urol 2013;63:702–8]. To track the seeding clone we performed copy-number alteration analysis on 34 morphologically distinct tissue areas in one prostatectomy specimen. Using break-point regions to infer phylogenetic relationships, the clone most closely related to the metastases was found in intraductal carcinoma of the prostate. Although the majority of tumor areas harbored events also found in the metastases, three carried none. This emphasizes the importance of intraprostatic tumor heterogeneity for prediction of prognosis. These findings also support recent evidence that intraductal carcinoma is a marker of aggressive disease. Patient summary: We identified the area in the prostate that gave rise to metastases by searching for metastatic-specific DNA alterations in multiple regions of the prostate. The metastasizing component grew within prostatic ducts, suggesting that intraductal cancer should be reported when found in needle biopsies. It is also important to be aware of tumor heterogeneity when assessing somatic changes linked to tumor aggressiveness. # 2014 European Association of Urology. Published by Elsevier B.V. All rights reserved.

Keywords: Prostate cancer Heterogeneity Metastasis Intraductal carcinoma

y These authors contributed equally. * Corresponding author. Department of Pathology, Radiumhemmet P1:02, Karolinska University Hospital, 171 76 Stockholm, Sweden. Tel. +46 707 205979. E-mail address: [email protected] (L. Egevad).

Prostate-specific antigen screening saves lives by lowering the probability of dying from prostate cancer [1]. Nevertheless, many indolent age-related cancers are treated without any apparent benefit for the patient. Therefore, the currently direst aim of clinical prostate cancer research is to establish an improved treatment rationale at primary diagnosis. Histopathological appearance and Gleason grading are currently important tools in risk stratification and management of localized prostate cancer. However, personalized genomic profiling for selection of the most efficient treatments is the future of clinical practice. Although

conceptually demonstrated by RNA profiling for assessment of proliferation rates [2], clinical implementation is limited by pronounced intraprostatic tumor heterogeneity. The majority of prostate cancers are multifocal [3], and we previously used whole-exome sequencing to demonstrate that spatially separated tumor foci are somatically independent [4]. This is in contrast to metastatic prostate cancer, which is of monoclonal origin [5], and has implications for any approach using tumor biopsies for prognostication. As part of a genomics study investigating correlations between different tumor characteristics that was published

http://dx.doi.org/10.1016/j.eururo.2014.09.006 0302-2838/# 2014 European Association of Urology. Published by Elsevier B.V. All rights reserved.

Please cite this article in press as: Lindberg J, et al. Tracking the Origin of Metastatic Prostate Cancer. Eur Urol (2014), http:// dx.doi.org/10.1016/j.eururo.2014.09.006

EURURO-5845; No. of Pages 4 2

EUROPEAN UROLOGY XXX (2014) XXX–XXX

in this journal [6], we set out to verify the monoclonal relationship between primary tumor tissue and metastases by exome sequencing in two patients. The first patient carried an aggressive neuroendocrine tumor that exhibited a clear monoclonal relationship to bone metastasis. Surprisingly, in the second patient (SWE-54), we could not find any common somatic denominator between fresh-frozen primary tumor tissue obtained after radical prostatectomy (SWE-54A) and fresh-frozen right/left pelvic lymph-node metastases (SWE-54-B/C). Therefore, in an attempt to identify the clone seeding the metastases, we conducted laser-capture microdissection of 45 morphologically distinct areas using formalin-fixed paraffin embedded tissue blocks (Supplementary Text, Supplementary Fig. 1, Supplementary Table 1). For tracking of the metastases we adopted a similar strategy as Navin et al. [7], performing low-pass wholegenome sequencing to detect break-point regions (BPRs), marking the start of an amplification or a deletion event (Supplementary Text). The presence of BPRs was used to infer

E-5

ID

C

4 SWE-54 B C 21_PZ_T1

2 12 6_PZ _T1 _P Z_ T1

C

1_

Proportion portion LN-Met k-point regions break-point 0.8

17_PZ

ID

_T

SW

C

1_

PZ

B _T1

_T

9_

PE 18_PZ_T1_E PZ_T1 16_

PZ

T1_ID

8_

7_PZ_

A

phylogenetic somatic relationships. Eleven areas were lost during tissue harvesting, library preparations, or quality control of the data. Additionally, nine tissue areas contained less than five BPRs and were excluded, leaving 25 tumor areas for further analysis. In total, 385 BPRs were identified as being present in two or more tumor areas. These BPRs were used to construct a phylogenetic tree (Fig. 1A) and the somatic relationships were visualized by coloring the proportion of BPRs shared with metastases (Fig. 1B, Supplementary Text). Remarkably, tissues with the highest level of similarity to the lymph-node metastases were the three areas where an intraductal carcinoma (IDC) component was microdissected for sequencing (Supplementary Figs. 2 and 3). To verify the diagnosis of IDC, immunohistochemistry for p63 and alpha-methylacyl-CoA-racemase (AMACR, p504s) was performed. All glandular elements of the IDC areas had either a complete or fragmented basal cell layer, confirming the morphological diagnosis of IDC (Fig. 2A,B). Interestingly, one area of

0.6 0.4

T1

_

PZ

_ 14

27

_P

Z_

_ T1

SV

0.2

I

0.0 LN-Mets

1

Z_T

25_P

2_PZ_T1 30_P

Z_T2

15_PZ_T1 1 1 Z_T Z_T 4_P _P 3 1

SW 20

1

T Z_

_T

_P

Z _P

10

T1

22_PZ_T1

1 19_ PZ_

_T

PZ

4A

SWE-54B/C

VI

3_

E

EP

1_

S T1_ PZ_ 29_ Z_T1 1_P

_P

28

I

SV

1_

T Z_

E-5

8_PZ_T1_IDC 7_PZ_T1_IDC 9_PZ_T1_IDC 21_PZ_T1 17_PZ_T1 18_PZ_T1_EPE 26_PZ_T1 12_PZ_T1 27_PZ_T1_SVI 22_PZ_T1 16_PZ_T1 14_PZ_T1 28_PZ_T1_SVI 3_PZ_T1 10_PZ_T1_EPE 19_PZ_T1 1_PZ_T1 29_PZ_T1_SVI 25_PZ_T1 13_PZ_T1 2_PZ_T1 15_PZ_T1 4_PZ_T1 SWE−54A 30_PZ_T2 20_PZ_T1

Fig. 1 – The somatic relationships of the tumor areas. (A) A phylogenetic tree was constructed using a neighbor-joining algorithm to track the clone from the primary tumor tissue that seeded the metastases. (B) To visualize the somatic relationship to metastases, samples were colored according to the proportion of break-point regions shared with lymph-node metastases. SWE-54A (primary fresh-frozen tissue), SWE-54B (right fresh-frozen lymph-node metastasis), and SWE-54C (left fresh-frozen lymph-node metastasis) represent tissues previously profiled by low-pass whole-genome sequencing and whole-exome sequencing [1]. Tumor areas are labeled according to area, zone, focus, and additional annotation; for example, 8_PZ_T1_IDC corresponds to area 8, peripheral zone, tumor focus 1, intraductal carcinoma. TZ = transition zone; PZ = peripheral zone; IDC = intraductal carcinoma; EPE = extraprostatic extension; SVI = seminal vesicle invasion.

Please cite this article in press as: Lindberg J, et al. Tracking the Origin of Metastatic Prostate Cancer. Eur Urol (2014), http:// dx.doi.org/10.1016/j.eururo.2014.09.006

EURURO-5845; No. of Pages 4 EUROPEAN UROLOGY XXX (2014) XXX–XXX

3

Fig. 2 – Histological sections of areas subjected to laser capture microscopy and copy-number alteration analysis. (A) Intraductal carcinoma (area 8, peripheral zone, tumor focus 1); hematoxylin and eosin (HE) staining, 20T magnification. (B) The same area immunohistochemically stained for p63 (brown) and alpha-methylacyl-CoA-racemase (AMACR; red); 20T magnification. (C) Invasive carcinoma with perineural invasion (area 21_PZ_T1). The arrow indicates a nerve; HE staining, 20T magnification. (D) The same area stained for p63 and AMACR; 20T magnification.

invasive carcinoma with Gleason score 4 + 4 = 8 (21_PZ_T1, area 21, peripheral zone, tumor focus 1; Fig. 2C,D) was highly related to the IDC component, although located at some distance (Supplementary Figs. 2 and 3). This area displayed perineural invasion, giving a rationale for the somatic similarity observed. These data suggest that tumor cells of the main focus, in area 17, acquired the ability to spread throughout the ducts and along the nerves to colonize other parts of the prostate, as well as nearby lymph nodes (Supplementary Fig. 2). As demonstrated previously [6], the fresh-frozen primary tumor sample did not share any somatic denominator with the metastases. A single area of invasive carcinoma (20_PZ_T1) displayed a high degree of similarity to SWE-54A, carrying 83% of its BPRs (Supplementary Fig. 3). SWE-54A was originally

macroscopically sampled from the same region in the posterior part of the peripheral zone. The relatedness to 20_PZ_T1 strengthens the validity of this approach for identifying phylogenetic relationships, as the two were prepared for sequencing using different chemistry and quality of the DNA. Furthermore, this prostate carried three tumor areas somatically isolated from the lymph-node metastases (including 31_TZ_T3, which contained <5 BPRs and was not included in the phylogenetic analysis). The level of intraprostatic tumor heterogeneity, demonstrated by our group and others [8], indicates that prognostication based on tumor tissue will not be easy. Pathological features associated with poor outcome may potentially provide means to prioritize areas for somatic analysis. There is accumulating evidence that the outcome

Please cite this article in press as: Lindberg J, et al. Tracking the Origin of Metastatic Prostate Cancer. Eur Urol (2014), http:// dx.doi.org/10.1016/j.eururo.2014.09.006

EURURO-5845; No. of Pages 4 4

EUROPEAN UROLOGY XXX (2014) XXX–XXX

of invasive prostate cancer is worse when there is concomitant IDC [9]. To the best of our knowledge, this is the first genomics-based evidence connecting a histopathological characteristic, correlated to poor outcome, with metastatic seeding potential, giving a rationale for prioritizing analysis of intraductal components when found.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.eururo.2014.09.006. References

Author contributions: Lars Egevad had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

[1] Hugosson J, Carlsson S, Aus G, et al. Mortality results from the Go¨teborg randomised population-based prostate-cancer screening trial. Lancet Oncol 2010;11:725–32.

Study concept and design: Lindberg, Egevad, Gro¨nberg. Acquisition of data: Kristiansen, Wiklund.

[2] Cuzick J, Swanson GP, Fisher G, et al. Prognostic value of an RNA expression signature derived from cell cycle proliferation genes in

Analysis and interpretation of data: Lindberg, Egevad.

patients with prostate cancer: a retrospective study. Lancet Oncol

Drafting of the manuscript: Lindberg, Egevad, Kristiansen.

2011;12:245–55.

Critical revision of the manuscript for important intellectual content: Lindberg, Egevad, Kristiansen. Statistical analysis: Lindberg. Obtaining funding: Gro¨nberg, Egevad. Administrative, technical, or material support: Lindberg. Supervision: Lindberg, Egevad. Other (specify): None. Financial disclosures: Lars Egevad certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: None. Acknowledgments: This study was supported by grants from the Linne´

[3] Arora R, Koch MO, Eble JN, Ulbright TM, Li L, Cheng L. Heterogeneity of Gleason grade in multifocal adenocarcinoma of the prostate. Cancer 2004;100:2362–6. [4] Lindberg J, Klevebring D, Liu W, et al. Exome sequencing of prostate cancer supports the hypothesis of independent tumour origins. Eur Urol 2013;63:347–53. [5] Liu W, Laitinen S, Khan S, et al. Copy number analysis indicates monoclonal origin of lethal metastatic prostate cancer. Nat Med 2009;15:559–65. [6] Lindberg J, Mills IG, Klevebring D, et al. The mitochondrial and autosomal mutation landscapes of prostate cancer. Eur Urol 2013;63: 702–8. [7] Navin N, Kendall J, Troge J, et al. Tumour evolution inferred by singlecell sequencing. Nature 2011;1–6. [8] Haffner MC, Mosbruger T, Esopi DM, et al. Tracking the clonal origin of lethal prostate cancer. J Clin Invest 2013;123:4918–22. [9] Trudel D, Downes MR, Sykes J, Kron KJ, Trachtenberg J, van der Kwast

Centre for Breast and Prostate Cancer (CRISP, no. 70867901), Karolinska

TH. Prognostic impact of intraductal carcinoma and large cribriform

Institutet, the Swedish Research Council (no. K2010-70X-20430-04-3),

carcinoma architecture after prostatectomy in a contemporary co-

and the Swedish Cancer Society (no. 11-0287).

hort. Eur J Cancer 2014;50:1610–6.

Please cite this article in press as: Lindberg J, et al. Tracking the Origin of Metastatic Prostate Cancer. Eur Urol (2014), http:// dx.doi.org/10.1016/j.eururo.2014.09.006