Abstract
Background
Metastatic prostate cancer is an incurable disease. During the development of this disease, prostate cancer cells enter the bloodstream as single cells or clusters of cells. Prostate fibroblasts, a cancer-promoting cell type in the prostate cancer microenvironment, could in theory incorporate into these migrating cell clusters or follow cancer cells into the bloodstream through holes in the tumor vasculature. Based on this idea, we hypothesized that fibroblast-like cells, defined here as cytokeratin 8/18/19−/DAPI+/CD45−/vimentin+ cells, are present in the blood of men with metastatic prostate cancer.
Methods
Veridex’s CellSearch® System was used to immunomagnetically capture EpCAM+ cells and clusters of cells heterogeneous for EpCAM expression from the blood of men with metastatic prostate cancer, localized cancer, and no known cancer, and immunostain them for the presence of cytokeratins 8/18/19, a nucleus, CD45, and vimentin. Fibroblast-like cells were then quantified.
Results
Fibroblast-like cells were present in 58.3% of men with metastatic prostate cancer but not in any men with localized prostate cancer or no known cancer. The presence of these cells correlated with certain known indicators of poor prognosis: ≥5 circulating tumor cells, defined here as cytokeratin 8/18/19+/DAPI+/CD45− cells, per 7.5 mL of blood, and a relatively high serum prostate-specific antigen level of ≥20 ng/mL.
Conclusions
The presence of fibroblast-like cells in the blood may provide prognostic information as well as information about the biology of metastatic prostate cancer.
Keywords: metastasis, blood, vimentin
Introduction
Metastatic prostate cancer poses a significant risk to men, with 67% of men dying within five years of diagnosis [1]. This disease is often treated initially with hormone therapies but almost always eventually becomes resistant [2]. In order to develop novel strategies for preventing and/or treating metastatic prostate cancer, we need to better understand the biology of metastasis, including the biology of the blood during metastasis.
Before cancer cells capable of metastasis enter secondary sites, they must first invade through the basement membrane at the primary tumor site and enter the bloodstream as single cells or in cell clusters [3]. Migrating clusters of cancer cells could, in theory, incorporate non-cancer cell types such as prostate fibroblasts, a cell type in the prostate cancer microenvironment that induces prostate cancer growth [4–7] and invasion [8]. Lending credence to this theory is the fact that metastatic prostate cancer cells express high levels of N-cadherin [9], a protein that mediates their adhesion to prostate fibroblasts [10].
It is also possible that prostate fibroblasts enter the bloodstream as single cells. Tumor vasculature is leaky [11], suggesting that cells from a tumor microenvironment could enter the bloodstream through holes in vessel walls. Furthermore, prostate fibroblasts invade through collagen I in response to prostate cancer-derived angiogenin [12,13], suggesting that they may follow migrating prostate cancer cells into the bloodstream.
Based on these concepts, we hypothesized that fibroblast-like cells are present in the blood of men with metastatic prostate cancer but not in the blood of men with localized prostate cancer or no known cancer. To test this hypothesis, we used Veridex’s CellSearch® System to capture circulating CK−/DAPI+/CD45−/vim+, or fibroblast-like, cells, which were then quantified. The presence of fibroblast-like cells in the blood may provide both prognostic information and information about the biology of metastatic prostate cancer.
Materials and Methods
Study participants
Participants with localized prostate cancer and those with metastatic, castration-resistant prostate cancer were identified in a medical oncology clinic, while cancer-free participants were recruited through advertising for normal controls. All participants consented to participate in this study per IRB-approved protocol. Their characteristics are delineated in Table 1.
Table 1.
Medical information of study participants. Participants 1–9 had no known cancer, and their ages ranged from 22 to 60.
| Cancer status | Participant | Age | PSA level (ng/mL) | Gleason score | Treatment | |||
|---|---|---|---|---|---|---|---|---|
| Localized PCa | 10 | 64 | 7.9 | 3+3=6 | none | |||
| 11 | 70 | 7.1 | 3+4=7 | none | ||||
| 12 | 64 | 10.3 | 4+3=7 | none | ||||
| 13 | 63 | 6.4 | 3+3=6 | none | ||||
| 14 | 70 | 8.4 | 4+3=7 | none | ||||
| 15 | 67 | 5.9 | 3+4=7 | none | ||||
| 16 | 53 | 14.0 | 3+4=7 | none | ||||
| 17 | 68 | 5.6 | 3+4=7 | none | ||||
| 18 | 54 | 4.1 | 3+4=7 | none | ||||
| 19 | 60 | 2.2 | 3+3=6 | none | ||||
| Metastatic PCa | 20 | 46 | 4.7 | 4+4=8 | H | |||
| 21 | 86 | 160.3 | 7 | RRP | C | H | ||
| 22 | 72 | 12.6 | 4+4=8 | RRP | RT | H | ||
| 23 | 69 | 75.3 | 4+3=7 | RT | C | H | ||
| 24 | 57 | 16.4 | 9 | RRP | H | |||
| 25 | 82 | 176.2 | 3+3=6 | RT | C | H | ||
| 26 | 65 | 124.2 | 3+4=7 | C | H | |||
| 27 | 54 | 4.9 | 4+5=9 | H | ||||
| 28 | 66 | 95.3 | 4+3=7 | RRP | RT | C | H | |
| 29 | 80 | 68.5 | 5+4=9 | RRP | C | H | ||
| 30 | 62 | 7.7 | 7 | RRP | H | |||
| 31 | 76 | 582.0 | 7 | RT | C | H | ||
RRP, radical retropubic prostatectomy; RT, radiation therapy at the primary tumor site; C, chemotherapy; H, hormone therapy.
Blood sample collection
Blood samples (10 mL/sample) from 14 men with metastatic prostate cancer (2 for an initial experiment, presented in Figures 1B and 1C), 10 with localized prostate cancer, and 9 without cancer were collected in CellSave Preservative tubes. During the participant visits when these blood samples were collected, blood samples were also collected for serum PSA measurements. All blood samples were collected under IRB-approved protocols at the Johns Hopkins University School of Medicine (JHSOM) and the University of Michigan Medical School. The CellSave tubes were gently inverted five times and then stored at room temperature for up to 72 h prior to blood sample processing as described in Veridex’s CXC kit protocol.
Figure 1.
CellSearch® System protocol and resultant images and cell counts. A, Diagram depicting the CellSearch® System protocol. Blood was collected from men with no known cancer, localized prostate cancer, and metastatic prostate cancer, and loaded onto a CellTracks® AutoPrep® System. This system captured EpCAM+ cells, and clusters of cells heterogeneous for EpCAM expression, and stained them for the epithelial markers cytokeratins 8/18/19 (CK); the presence of a nucleus (DAPI); the lymphocyte marker CD45; and the mesenchymal marker vimentin (vim). The captured, stained cells were transferred to a CellTracks® Analyzer II and scanned into CellSearch® software, which presented the user with images of cells that could then be quantified. All of the images in a single row depict the same cell or cell cluster. B and C, Subsets of images of captured, stained cells from one study participant (B) and a second study participant (C) who had metastatic prostate cancer. D, The number of CK+/DAPI+/CD45− cells CD45−/vim+ cells (i.e., circulating fibroblast-like cells) per 7.5 mL of blood in 9 (i.e., circulating tumor cells, or CTCs) and CK−/DAPI+/men with no known cancer, 10 with localized prostate cancer (PCa), and 12 with metastatic PCa. E, The percentage of study participants in each category (e.g., no known cancer, localized PCa, and metastatic PCa) who had CTCs or circulating fibroblast-like cells. ***, P < 0.001; NS, P > 0.05, compared to CK+/DAPI+/CD45−cells in participants with metastatic PCa. **, P < 0.01, compared to CK−/DAPI+/CD45−/vim+ cells in participants with metastatic PCa.
Immunomagnetic isolation of cells
Immediately after processing, the blood samples (7.5 mL/sample) were loaded onto a CellTracks® AutoPrep® System (Figure 1A), where they were immunomagnetically enriched for EpCAM+ cells and clusters of cells heterogeneous for EpCAM expression. The system stained the enriched cells with the CXC kit reagents DAPI, FITC-labeled anti-CK-8/18/19 antibody, and APC-labeled anti-CD45 antibody, as well as the user-defined reagent PE-labeled anti-vimentin antibody (Abcam, diluted to 4 or 2 μg/mL in 0.1% bovine serum albumin/PBS for Figures 1B and 1C, respectively, and 3 μg/mL in 0.1% bovine serum albumin/PBS for Figures 1D, 1E, and 2). A cartridge containing the enriched, stained cells was removed from the system and placed in the dark for 20 min to 24 h prior to analysis.
Figure 2.
Correlation of the presence of circulating fibroblast-like cells with known indicators of unfavorable prostate cancer prognosis: ≥5 CTCs per 7.5 mL of blood (A), a PSA level of ≥20 ng/mL (B), and both ≥5 CTCs per 7.5 mL of blood and a PSA level of ≥20 ng/mL (C). **, P < 0.01; *, P < 0.05, compared to no circulating fibroblast-like cells.
Analysis of captured images
The cartridge containing the enriched, stained cells was placed in the CellTracks® Analyzer II, where the cells were scanned (0.5 s for the PE channel). Images of single cells and small cell clusters were then loaded into the CellSearch® software and reviewed by multiple individuals, who enumerated CK−/DAPI+/CD45−/vimentin+/cells (i.e., circulating fibroblast-like cells) and CK+/DAPI+/CD45− cells (i.e., circulating tumor cells).
Statistics
For the data presented in Figures 1E and 2A–C, Fisher’s exact test was performed using a Microsoft Research online calculator, which generated two-sided P values.
Results
The overall goal of this study was to determine whether circulating CK−/DAPI+/CD45−/vimentin+, or fibroblast-like, cells are present in men with metastatic prostate cancer. In a preliminary study (outlined in Figure 1A), we found that these cells were indeed present in both men that we tested; images of fibroblast-like cells from these men are shown in Figures 1B and 1C. These results led us to question whether the presence of circulating fibroblast-like cells is a specific feature of metastatic prostate cancer or is generalizable to all prostate cancers or even all men. Thus, we performed a larger study in which we evaluated their presence in 12 men with metastatic prostate cancer, 10 with localized prostate cancer, and 9 with no known cancer (Table 1 contains relevant medical information of these participants). Circulating fibroblast-like cells were detected in 7 of 12 men with metastatic prostate cancer and in none of the men with localized prostate cancer or no known cancer (Figures 1D and 1E).
In men who had detectable levels of circulating fibroblast-like cells, the cell number ranged from 2 to 12 per 7.5 mL of blood (Figure 1D). The presence of these cells correlated with certain known indicators of poor prognosis: ≥5 circulating tumor cells (CTCs, or CK+/DAPI+/CD45− cells) per 7.5 mL of blood, and a relatively high serum prostate-specific antigen (PSA) level of ≥20 ng/mL (Figure 2).
Discussion
We found that circulating CK−/DAPI+/CD45−/vimentin+, or fibroblast-like, cells were indeed present in study participants with metastatic prostate cancer but not in those with localized prostate cancer or no known cancer. While others have found circulating cells expressing mesenchymal markers in men with prostate cancer [14], those cells also expressed epithelial markers such as cytokeratins. In contrast, we detected circulating cells that express the mesenchymal marker vimentin but not cytokeratins. To our knowledge, we are the first to detect such cells in prostate cancer patients. This finding may have both clinical and biological importance.
From a clinical standpoint, the presence of circulating fibroblast-like cells in men with prostate cancer may indicate a poor prognosis. These cells were detected only in men with metastatic prostate cancer, an incurable disease. Furthermore, the presence of these cells correlated with certain known indicators of poor prognosis: ≥5 circulating tumor cells (CTCs, or CK+/DAPI+/CD45− cells) per 7.5 mL of blood, and a relatively high serum prostate-specific antigen (PSA) level of ≥20 ng/mL (Figure 2). These cut-off values for CTCs and PSA are controversial: depending on the source, ≥4 [15] or ≥5 CTCs [16] per 7.5 mL of blood, or a post-prostatectomy PSA level of ≥0.2 ng/mL [17] or ≥0.4 ng/mL [18], to name just two, indicates poor prognosis. Circulating fibroblast-like cells differ from CTCs and PSA in that it is their presence rather than any arbitrary cut-off value that seems to correlate with poor prognosis. Thus, these cells may be more meaningful than CTCs and PSA as a prognostic indicator.
From a biological standpoint, our work raises several important questions about the pathology of metastatic prostate cancer. First, did the fibroblast-like cells that we detected originate from the prostate? If so, were they prostate fibroblasts, prostate myofibroblasts (i.e., cells having characteristics of both smooth muscle cells and fibroblasts), or prostate cancer cells that underwent epithelial-to-mesenchymal transition? We are working on characterizing the fibroblast-like cells to address these questions.
Another important question raised by our work is where the circulating fibroblast-like cells go, and more specifically, whether they go to the bone, a predominant site for metastatic prostate cancer. The current dogma is that the stromal cells, i.e., non-epithelial cells, in the bone of men with metastatic prostate cancer are bone-derived [19]. Our findings give rise to the possibility that some of the stromal cells in their bones are prostate-derived: if the circulating fibroblast-like cells that we found are indeed prostate fibroblasts, then perhaps prostate fibroblasts migrate to secondary sites such as the bone and help cancer grow at those sites. In support of this hypothesis, Duda et al. found in a mouse model that carcinoma-associated fibroblasts migrated with metastasizing cancer cells to the lungs, where they contributed to the formation of metastatic lesions [20].
In our study, men with detectable levels of circulating fibroblast-like cells had cell numbers ranging from 2 to 12 per 7.5 mL of blood (Figure 1D). These cell numbers are low, possibly due to the limited sensitivity of the CellSearch® method [21]. Future work should focus on evaluating the presence of circulating fibroblast-like cells using a technique that is more sensitive, such as the recently developed Herringbone-Chip method [22], and more direct (i.e., directly capture circulating fibroblast-like cells rather than EpCAM+ cells and cells that have adhered to EpCAM+ cells). Future studies should also involve larger numbers of study participants.
Conclusions
Overall, the results of our study suggest that the presence of fibroblast-like cells in the blood may distinguish men who have metastatic prostate cancer from men who do not. Thus, these cells may have potential as a prognostic marker and/or therapeutic target in men with prostate cancer. Further work needs to be done to evaluate the clinical utility of circulating fibroblast-like cells, including their ability to predict a poor outcome, and to better understand their biology, i.e., where they came from, where they end up, and how they contribute to metastatic progression.
Acknowledgments
This work was supported by the Patana Fund of the Brady Urological Institute and two National Cancer Institute Prostate Cancer SPORE grants (P50 CA58236, P50 CA69568).
We thank Dr. Lori Sokoll (JHSOM, Departments of Pathology, Oncology, and Urology), Deb Elliott (JHSOM, Department of Pathology), and Marty Brown (University of Michigan, Department of Internal Medicine, Division of Hematology/Oncology) for their assistance with the CellSearch® System. We thank Robin Gurganus, RN (JHSOM, Department of Urology) for drawing blood from two of the study participants.
Footnotes
This work was performed in the Departments of Urology and Pathology at the Johns Hopkins University School of Medicine and in the Department of Urology and the Comprehensive Cancer Center at the University of Michigan Medical School.
Disclosure Statement
The authors have no conflicts of interest to disclose.
References
- 1.Schoenstadt A. [Accessed May 12, 2010];Prostate cancer survival rates. [eMedTV website]. Available at: http://prostate-cancer.emedtv.com/prostate-cancer/prostate-cancer-survival-rates-p2.html. Updated September 1, 2006.
- 2.Walsh PC, Worthington JF. Guide to surviving prostate cancer. 2. New York, New York, USA: Wellness Central; 2007. [Google Scholar]
- 3.Friedl P, Gilmour D. Collective cell migration in morphogenesis, regeneration and cancer. Nat Rev Mol Cell Biol. 2009;10:445–457. doi: 10.1038/nrm2720. [DOI] [PubMed] [Google Scholar]
- 4.Gleave M, Hsieh JT, Gao CA, von Eschenbach AC, Chung LW. Acceleration of human prostate cancer growth in vivo by factors produced by prostate and bone fibroblasts. Cancer Res. 1991;51:3753–3761. [PubMed] [Google Scholar]
- 5.Kirschenbaum A, Wang J-P, Ren M, et al. Inhibition of vascular endothelial cell growth factor suppresses the in vivo growth of human prostate tumors. Urologic Oncology: Seminars and Original Investigation. 1997;3:3–10. doi: 10.1016/s1078-1439(97)00001-x. [DOI] [PubMed] [Google Scholar]
- 6.Dow JK, deVere White RW. Fibroblast growth factor 2: its structure and property, paracrine function, tumor angiogenesis, and prostate-related mitogenic and oncogenic functions. Urology. 2000;55:800–806. doi: 10.1016/s0090-4295(00)00457-x. [DOI] [PubMed] [Google Scholar]
- 7.Nakashiro K, Okamoto M, Hayashi Y, Oyasu R. Hepatocyte growth factor secreted by prostate-derived stromal cells stimulates growth of androgen-independent human prostatic carcinoma cells. Am J Pathol. 2000;157:795–803. doi: 10.1016/s0002-9440(10)64593-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Nishimura K, Kitamura M, Miura H, et al. Prostate stromal cell-derived hepatocyte growth factor induces invasion of prostate cancer cell line DU145 through tumor-stromal interaction. Prostate. 1999;41:145–153. doi: 10.1002/(sici)1097-0045(19991101)41:3<145::aid-pros1>3.0.co;2-r. [DOI] [PubMed] [Google Scholar]
- 9.Tanaka H, Kono E, Tran CP, et al. Monoclonal antibody targeting of N-cadherin inhibits prostate cancer growth, metastasis and castration resistance. Nat Med. 2010;16(12):1414–1420. doi: 10.1038/nm.2236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Tran NL, Nagle RB, Cress AE, Heimark RL. N-Cadherin expression in human prostate carcinoma cell lines. An epithelial-mesenchymal transformation mediating adhesion withStromal cells. Am J Pathol. 1999;155:787–798. doi: 10.1016/S0002-9440(10)65177-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hobbs SK, Monsky WL, Yuan F, Roberts WG, Griffith L, Torchilin VP, Jain RK. Regulation of transport pathways in tumor vessels: Role of tumor type and microenvironment. Proc Natl Acad Sci USA. 1998;95:4607–4612. doi: 10.1073/pnas.95.8.4607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Jones ML, Ewing CM, Isaacs WB, Getzenberg RH. Prostate cancer-derived angiogenin stimulates the invasion of prostate fibroblasts. J Cell Mol Med. 16(1):193–201. doi: 10.1111/j.1582-4934.2011.01283.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jones ML. The Migration and Invasion of Prostate Fibroblasts: a Potential Contributor to Prostate Cancer Metastasis [dissertation] Baltimore, MD: Johns Hopkins University School of Medicine; Dec, 2010. [Google Scholar]
- 14.Armstrong AJ, Marengo MS, Oltean S, et al. Circulating tumor cells from patients with advanced prostate and breast cancer display both epithelial and mesenchymal markers. Mol Cancer Res. 2011;9:997–1007. doi: 10.1158/1541-7786.MCR-10-0490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Goodman OB, Jr, Fink LM, Symanowski JT, et al. Circulating tumor cells in patients with castration-resistant prostate cancer baseline values and correlation with prognostic factors. Cancer Epidemiol Biomarkers Prev. 2009;18:1904–1913. doi: 10.1158/1055-9965.EPI-08-1173. [DOI] [PubMed] [Google Scholar]
- 16.de Bono JS, Scher HI, Montgomery RB, et al. Circulating tumor cells predict survival benefit from treatment in metastatic castration-resistant prostate cancer. Clin Cancer Res. 2008;14:6302–6309. doi: 10.1158/1078-0432.CCR-08-0872. [DOI] [PubMed] [Google Scholar]
- 17.Freedland SJ, Sutter ME, Dorey F, Aronson WJ. Defining the ideal cutpoint for determining PSA recurrence after radical prostatectomy. Prostate-specific antigen Urology. 2003;61:365–369. doi: 10.1016/s0090-4295(02)02268-9. [DOI] [PubMed] [Google Scholar]
- 18.Amling CL, Bergstralh EJ, Blute ML, Slezak JM, Zincke H. Defining prostate specific antigen progression after radical prostatectomy: what is the most appropriate cut point? J Urol. 2001;165:1146–1151. [PubMed] [Google Scholar]
- 19.Nannuru KC, Singh RK. Tumor-stromal interactions in bone metastasis. Curr Osteoporos Rep. 2010;8:105–113. doi: 10.1007/s11914-010-0011-6. [DOI] [PubMed] [Google Scholar]
- 20.Duda DG, Duyverman AMMJ, Kohno M, Snuderl M, Steller EJA, Fukumura D, Jain RK. Malignant cells facilitate lung metastasis by bringing their own soil. Proc Natl Acad Sci USA. 2010;107:21677–21682. doi: 10.1073/pnas.1016234107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Nagrath S, Sequist LV, Maheswaran S, et al. Isolation of rare circulating tumour cells in cancer patients by microchip technology. Nature. 2007;450:1235–1239. doi: 10.1038/nature06385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Stott SL, Hsu CH, Tsukrov DI, et al. Isolation of circulating tumor cells using a microvortex-generating herringbone-chip. Proc Natl Acad Sci USA. 2010;107:18392–18397. doi: 10.1073/pnas.1012539107. [DOI] [PMC free article] [PubMed] [Google Scholar]