Cancer and Metastasis Reviews 17: 331–336, 1999. © 1999 Kluwer Academic Publishers. Printed in the Netherlands.
Mechanisms, hypotheses and questions regarding prostate cancer micrometastases to bone Paul H. Lange1 and Robert L. Vessella1,2 Department of Urology, University of Washington, Seattle, WA, USA; 2 Puget Sound VA Medical Center, Seattle, WA, USA
1
Key words: prostate cancer, prostate specific antigen, metastasis, bone, cachexia, bone remodeling Abstract The morbidity and mortality associated with prostate cancer can almost universally be attributed to the consequences of metastases to the bone. While clinically there have been descriptive reports of these lesions and their detection by bone scan, there is an embrrassing paucity of reports as to the mechanisms of prostate cancer cell trafficking to the bone, adaptation to the bone environment, pertubation of the normal bone reformation process and the events leading to cachexia and death. In recent years, there have been numerous in vitro studies suggesting that PSA and hK2 may play a significant biological role in these events. Also, recent data generated form reverse transcription polymerase chain reaction assays reveal that metastasis to the bone may be an early event which further underscores the need to better understand this complex and critically important process. This commentary highlights several general concepts and a few specific issues related to CaP bone metastasis with the intent of revealing numerous opportunities for further investigation and inquiry.
The mechanisms involved in the metastasis of prostate cancer are poorly studied. While they probably involve the same basic postulated processes of all solid tumors, namely (a) tumor bed escape, (b) intravasation of the vessels, (c) adhesion to the endothelial wall, (d) extravasation, and (e) invasion/seeding at a distant site as other solid tumor metastases, there are some intriguing features of prostate cancer that are distinctive. Before discussing some of the more recent molecular studies that have influenced our current thinking about prostate cancer metastases, it is worth revisiting some of the early medical literature on this subject. Clearly one of the fascinating aspects of prostate cancer is its tendency to exhibit metastases first and foremost to the bone. In a comprehensive autopsy study of patients who died from cancer between 1927 and 1941, Walther [1] found that bone was the most common site of metastases for prostate cancer and the second most common site for breast cancer. These two cancers had the highest metastasis rate to bone among solid tumors. In this series, the frequency of bone metastasis in prostate
cancer was about 60% although several other autopsy series put the frequency closer to 80–85% [2–5]. It was then presumed that primary cancers seed distant sites by hematogenous dissemination and according to the hemodynamic theory proposed in 1928 by Ewing [6], the frequency of metastasis to a given organ was often a reflection of blood flow and direction through that organ. Thus they assumed that the very high rate of metastases of all cancers to the liver and lung confirmed this hypothesis. In the Walther [1] study, this frequency is nearly double that of metastasis to bone in cancer overall. Also, it is for this reason that early investigators suggested that prostate cancer cells went to the bone because of venous drainage through the so called Batson’s Plexus – a serpiginous collection of veins that connect the pelvis with the vertebral column. However, this theory did not explain the high frequency of prostate cancer metastases outside the spinal column and bony pelvis; nor did it account for the lack of prostate cancer metastases to the lungs. A better explanation for prostate and breast metastases to the bone has always been that that cells have a ‘proclivity’ to seed,
332 adapt and grow in the bone but not the liver or lung. However, it took the pioneering work of Talmadge and Fidler [7] to show in animals that cancer cells ‘hone’ to special sites best adapted for them. In prostate cancer, such adaptation would involve processes of engraftment and growth in bone while it is apparently more difficult for prostate cancer cells to seed, adapt and grow in the liver and lung. Yet, we will comment later on findings from our laboratory that suggest only a small fraction of prostate cancer micrometastases that seed the red marrow, actually proliferate to the point of overt metastases. Though much more research activity is needed in this area, several comprehensive reviews of the field have recently appeared [8–11]. Also, specifics about the processes by which micrometastases adapt to their new environment has been difficult to unravel until the recent availability of better molecular tools. Our interest in this field stems from several years of work in the detection of prostate cancer micro-metastatic dissemination using the reverse transcription polymerase chain reaction (RT-PCR) methodology. Using PSA mRNA as a target of RT-PCR, we, and many others, initially had hoped to demonstrate that this molecular approach would help in staging of the patient at the time of diagnosis. The players in this field are too numerous to cite here, but a detailed review of the topic was recently written by our group [12]. While most of the investigations focused on detection of circulating PSA-positive cells in the peripheral blood, a few of us expanded the studies to include bone marrow aspirates. Except for the group of investigators at Columbia [13], nearly all others were unable to find a correlation between the RT-PCR results and staging or clinical outcome, although follow-up was premature in most studies. However, what was learned from the bone marrow studies was informative to the biology of prostate cancer dissemination. In all three investigations [14–16], there was a high positivity rate of detection early in the disease process. When compared to the detection rate in the peripheral blood, that in the bone marrow was at least double. This led our group to suggest that the majority, and perhaps most, men have prostate cancer cells in their bone marrow at the time of diagnosis. Yet, the discovery of a high rate of bone marrow positivity before radical prostatectomy does not correlate with clinical realities. In men who elect surgery, about 15–30% develop clinical recurrence and not all of these progress to osseous metastases. This is approximately a third to a fourth of the detection rate of prostate cancer
cells in the bone marrow prior to surgery as determined by RT-PCR which implies that a substantial proportion of the detected cells did not engraft/adapt sufficiently for subsequent growth or that host factors, such as the immune response, interfered with subsequent growth. We have continued to follow patients by RT-PCR after radical prostatectomy and have found that a few remain RT-PCR positive many months after surgery but the majority revert to RT-PCR negative status within 6–9 months. Of course RT-PCR negative status does not necessarily mean that there are no PSA-positive cells present, only that the number is below the detection capability of the assay. These studies on the whole suggest that (1) dissemination occurs very early in the disease process; (2) the disseminated cells most likely represent a heterogeneous population; and (3) processes such as engraffment, adaptation and growth are critical aspects of bone metastasis and there are likely selective forces upon the prostate cancer micrometastases within this new environment which influence which cells die off, remain dormant or develop into overt metastases. Cancer cells spread to the bone via the blood and enter an environment unlike that of the visceral organs. The bone matrix is hard and relatively acellular with low metabolic activity. Yet, within this matrix are a plethora of growth factors for the osteoclasts and osteoblasts, the key players in bone reformation [17,18]. As the bone is constantly undergoing resorption and reformation, this matrix is being destroyed and in the process releases these growth factors into the microenvironment. Not only are these factors favorable for growth, but they exert chemotactic responses [19–21]. In addition to the bone matrix, the red marrow is packed with hematopoietic cells at various stages of differentiation. This extremely active compartment also provides a rich bed of growth factors. What is intriguing but largely unexplored is why prostate and breast cancer find the environment of the bone marrow so attractive for growth while other cancer types prefer the visceral, high blood flow organs and some tumor types almost never show growth in the bone. As discussed more later, we are beginning to explore the possibility that prostate specific antigen (PSA) and hK2 may have a critical role in adapting (e.g. utilization of the available growth factors) to this environment. Although prostate cancer is one of several tumor types exhibiting metastases to bone, its affect on the bone is unique and perplexing. We previously mentioned that the bone is constantly undergoing the
333 processes of resorption and reformation by the osteoclasts and osteoblasts, respectively. In fact, ‘pitting’ of the hard bone matrix by the osteoclasts is considered by some to be a necessary first step in the attachment of cancer cells to the underlying integrins [22–24]. Once the cancer cells attach and begin to grow they can affect the bone in one of two ways: nearly all cancer metastases to the bone with the exception of prostate cause an osteolytic response characterized by breakdown of the matrix; in contrast, prostate cancer bone metastases are uniquely osteosclerotic, characterized by excessive bone formation. What factors are produced or perturbed to result in this osteolytic or osteoblastic imbalance? Gutman, some sixty years ago, recognized that prostate cancer caused an imbalance favoring an osteosclerotic response and proposed that prostate cancer cells produced a factor responsible for these clinical findings [25]. Today, there is considerable speculation based primarily on in vitro studies that PSA and hK2 may participate in mediating at least some of this imbalance. PSA and hK2 are prostate ‘specific’ serine proteases of the kallikrein family. They are secreted in an inactive form, activated within the microenvironment by as yet unknown factors and then inactivated by complexation with serine protease inhibitors soon thereafter [26]. However, during their active state, they may interact with critical growth and regulatory factors involved in the bone reformation process. For example, PSA has been shown to interact with transforming growth factor beta, insulin-like growth factor binding proteins and with parathyroid hormone related protein [27–29]. hK2 appears to have the ability to interact with the bone morphogenetic proteins (BMPs; manuscript in preparation) and these BMPs are known to activate osteoblasts. Recent data have shown that prostate cancer cells not only produce BMPs but they also have receptors for BMPs thus suggesting a possible complementary autocrine role [30–32]. Urokinase, another serine protease produced by prostate cancers may have a role in the aggressiveness of the bone metastasis [33–38]. As stated earlier, the bone marrow microenvironment is very rich in growth factors and one can easily envision numerous patterns of autocrine/paracrine interactions resulting in enhanced prostate cancer growth which in turn results in factor interactions favoring an osteosclerotic response by bone cells. An excellent review on the role of growth factors on prostate cancer has recently appeared for those readers seeking more information on this particular subject [39].
One of the common consequences of prostate cancer metastasis is cachexia, which effects a high proportion of patients with advanced disease. This syndrome of body wasting, anorexia and pain [40] has not been studied to any significant extent in prostate cancer and yet it is responsible for the very poor quality of life and morbidity in these advanced disease patients. One recent commentary to which we subscribe suggests that cachexia in prostate cancer may be a major cause of death, not just morbidity [41]. As part of our research program, we perform ‘rapid’ autopsies on patients within 2–3 h of death in an effort to harvest for research purposes viable bone and soft tissue metastases. We’ve been astonished to find a substantial number of patients who have few, if any, soft tissue metastases. Many of these patients exhibited the classic cachexia syndrome yet tumor burden was limited to bone; the extent of bone marrow involvement is ill-defined from the scientific literature and is currently under study as part of our investigations. However, it would appear that the interaction of prostate cancer cells within the bone marrow is the initiating event of cachexia in these patients and that the cachexia is overwhelming despite an unremarkable tumor burden in a large number of patients. These observations may suggest that for at least some of our patients with advanced prostate cancer, control of the cytokines or other factors involved in cachexia may be as important if not more so, then control of tumor burden in the management of the disease. This brief overview of prostate cancer metastasis has touched on several possible cell to cell interactions and molecular mechanisms involved in the cascade of processes ultimately required for micro-metastases to develop into overt clinical metastases. Studies of this progression in prostate cancer are far behind those of many other tumors because there is a general lack of metastatic tissues for study due to the manner in which the disease is managed (i.e. no surgical intervention if known metastases exist) and because animal models (spontaneous or xenograft) of prostate cancer metastases are few. Under such limitations, how do we envision making progress in this field? Largely due to the previously cited RT-PCR studies, it is our hypothesis that dissemination of prostate cancer cells into the blood and bone marrow is an early event in the disease process. This observation has spurred a number of investigations into the identity and characterization of those disseminated cells. For example, several colleagues at Johns Hopkins University have joined forces to develop a method of prostate
334 cancer cell isolation from the peripheral blood and have followed with studies of ploidy and other descriptors of possible clinical relevance [42]. We and others [43] have taken slightly different methodological approaches but have similar goals; that is, isolate and characterize those cells which disseminate early in the process and compare them to the primary tumor and to cells isolated during progressive disease. Some of the studies are straight-forward, such as determining the percentage of cells with mutated androgen receptor or those expressing her2/neu. Other studies underway in our group and elsewhere are of considerably greater challenge in attempting to molecularly profile individual cells using micro-array technologies and to propagate from single isolates, cell lines and xenografts. Mastering the isolation and propagation of these cells will open many opportunities for studying their interaction with osteoclasts, osteoblasts and other cell types within the bone environment. Comparing molecular profiles of cells captured early in the dissemination process with those in advanced, end-stage disease will establish molecular pathways of progression with the potential elucidation of opportunities for novel strategic interventions. There are very few animal models of prostate cancer metastasis that clearly mimic what is seen in man; preferentially to the bone and accompanied by high serum PSA levels. However, significant advances have been made in the last couple of years in developing improved models of bone metastasis using human prostate cancer xenografts. Three groups, including our own, have documented metastatic or quasi-metastatic xenograft systems of bone metastases that result in an osteosclerotic response [44–46]. These models are under intense study to elucidate cell-to-cell interactions and factors responsible for these events. An interesting question is why it is so difficult to achieve spontaneous bone metastasis in immune compromised mice bearing orthotopic human prostate cancer xenografts when in man, the proclivity for bone is so remarkable, reaching almost 100%. Sometimes, understanding why a biological system doesn’t work as expected opens doors as to the critical mechanisms involved. Along a similar line of intrique, these mice bearing large orthotopic xenografts rarely become cachexic. In summary, human prostate cancer cells seem to have acquired the multitude of processes and mechanisms associated with micrometastases to the bone, engraftment and robust growth while other cancer cells find the bone difficult to colonize. Are these cells
‘attracted’ to the bone or do they simply have the capability to take advantage of this environment once they find themselves lodged within the bone marrow? It is certainly exciting to speculate that PSA, hK2 and other products of the prostate cancer cell are active participants in these processes. However, we know by hard experience, that what is observed in vitro, may not be a primary factor in vivo. Nevertheless, the osteoslerotic response and the onset of severe cachexia in the absence of all but bone involvement remain perplexing, clinically relevant events worthy of more intense study. We believe such studies may be especially productive now because of such recent activities as rapid autopsies, isolation of prostate cancer cells from peripheral blood and bone marrow, development of metastatic and quasi-metastatic xenograft bone models, and the availability of molecular profiling databases along with micro-array technologies. We are hopeful that these studies will identify the critical pathways in progression and ultimately derive new approaches for improving the quality of life and perhaps, in some patients, long term stability if not cure.
References 1. Walther HE, Krebsmetastase A. Basel: Beno Schwabe & Co. 1948 2. Cook GB, Watson FR: Events in the natural history of prostate cancer: using salvage curves, mean age distributions and contingency. J Urol 96: 87–96, 1968 3. Cumming J, Hacking N, Fairhurst J, Ackery D, Jenkins JD: Distribution of bony metastases in prostatic carcinoma. Br J Urol 66: 411–414, 1990 4. Rana A, Chisholm GD, Khan M, Sekharjit SS, Merrick MV, Elton RA: Patterns of bone metastasis and their prognostic significance in patients with carcinoma of the prostate. Br J Urol 72: 933–936, 1993 5. Thalmann G, Anezinis P, Devoll R, Farach-Carson C, Chung LWK: Experimental approaches to skeletal metastasis of human prostate cancer. In: Raghavan D, Scher HI, Leibel SA and Lange PH (eds) Principles and practices of genitourinary oncology, Lippincott-Raven, Philadelphia, 1997, pp 409–416 6. Ewing JA: Treatise on Tumors. WB Saunders, Philadelphia, 1928 7. Talmadge JE, Fidler IJ: Cancer metastasis is selective or random depending on the parent tumour population. Nature 297: 593–596, 1982 8. Yoneda T: Cellular and molecular mechanisms of breast and prostate cancer metastasis to bone. Eur J Cancer 34: 240–245, 1998
335 9. Scher HI, Chung LWK: Bone metastases: improving the therapeutic index. Semin Oncol 21: 630–656, 1994 10. Shevrin DH, Kukreja SC: Pathophysiology of human prostate cancer metastasis to bone, studied in vitro and in animal models. In: Bone Metastasis – Mechanisms and Pathophysiology. R.G. Landes Company, Austin, pp 181–194 11. Stearns ME, Wang M: Bone metastases and prostate cancer. In: Bone Metastasis – Mechanisms and Pathophysiology, R.G. Landes Company, Austin, 1996, pp 195–220 12. Corey E, Corey MJ: Detection of disseminated prostate cells by reverse transcription-polymerase chain reaction (RT-PCR): technical and clinical aspects. Int J Cancer 77: 655–673, 1998 13. De la Taille A, Olsson CA, Katz AE: RT-PCR for PSA – Molecular staging of prostate cancer: Dream or reality? Oncology, 1999 (in press) 14. Wood DP, Banks ER, Humphreys S, McRoberts JW, Rangnekar VM: Identification of bone marrow micrometastases in patients with prostate cancer. Cancer 74: 2533–2540, 1994 15. Melchior SW, Corey E, Ellis WJ, Ross AA, Layton TJ, Oswin MM, Lange PH, Vessella RL: Early tumor cell dissemination in patients with clinically localized carcinoma of the prostate. Clin Cancer Res 3: 249–256, 1997 16. Gao C, Dean RC, Pinto A, Mooneyhan R, Connelly RR, McLeod DG, Srivastava S, Moul JW: Detection of circulating PSA-expressing prostatic cells in the bone marrow of radical prostatectomy patients by sensitive Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR). J Urol 1999 (in press) 17. Hauschka PV, Manrakos AE, Iafrati MD, Doleman SE, Klagsbrun M: Growth factors in bone matrix. Isolation of multiple types of affinity chromatography on heparin sepharose. J Biol Chem 261: 12665–12674, 1986 18. Pfeilschifter J, Mundy GR: Modulation of transforming growth factor beta activity in bone cultures by osteotropic hormones. Proc Natl Acad Sci USA 84: 2024–2028, 1987 19. Orr W, Varani J, Gondek MD, Ward PA, Mundy GR: Chemotactic responses of tumor cells to products of resorbing bone. Science 203: 176–179, 1979 20. Orr FW, Varani J, Gondek MD, Ward PA, Mundy GR: Partial characterization of a bone derived chemotactic factor for tumor cells. Am J Pathol 90: 43–52, 1980 21. Pfeilschifter J, Wolf O, Naumann A, Minne HW, Mundy GR, Ziegler R: Chemotactic response of osteoblast-like cells to transforming growth factor beta. J Bone Miner Res 5: 825–830, 1990 22. Manishen JW, Sivananthan K, Orr FW: Resorbing bone stimulates tumor cell growth: a role for the host microenvironment in bone metastasis. Am J Pathol 123: 39–45, 1986 23. Boyde A, Maconnachie E, Reid SA, Delling G, Mundy GR: Scanning electron microscopy in bone pathology: review of methods. Potential and applications. Scanning Electron Microsc 4: 1537–1554, 1986 24. Hiraga T, Nakajima T, Ozawa H: Bone resorption induced by a metastatic human melanoma cell line. Bone 16: 349– 356, 1995
25. Gutman EB, Sproul EE, Gutman AB: Significance of increased phosphatase activity of bone at the site of osteoplastic metastases secondary to carcinoma of the prostate gland. Am J Cancer 28: 485–495, 1998 26. Rittenhouse HG, Finlay JA, Mikolajczyk SD, Partin AW: Human Kallikrein 2 (hK2) and prostate specific antigen (PSA): Two closely related, but distinct, kallikreins in the prostate. Crit Rev Clin Lab Sci 35: 275–368, 1998 27. Cramer SD, Chen Z, Peehl DM: Prostate specific antigen cleaves parathyroid hormone-related protein in the PTH-like domain: inactivation of PTHrP-stimulated cAMP accumulation in mouse osteoblasts. J Urology 156: 526–531, 1996 28. Cohen P, Graves H, Peehl DM, Kamarei M, Giudice LC, Rosenfeld RG: Prostate-specific antigen (PSA) is an insulinlike growth factor binding protein-3 prostease found in seminal plasma. J Clin Endocrinol Metab 73: 1046–1052, 1992 29. Plymate SR, Rosen CJ, Paulsen CA, Ware JL, Chen J, Vessella RL, Birnbaum RS: Proteolysis of insulin-like growth factor-binding protein-3 in the male reproductive tract. J Clin Endocrinol Metab 81: 618–624, 1996 30. Ide H, Katoh M, Sasaki H, Yoshida T, Aoki K, Nawa Y, Osada Y, Sugimura T, Terada M: Cloning of human bone morphogenetic protein type IB rceptor (BMPR-IB) and its expression in prostate cancer in comparison with other BMPRs [published erratum appears in Oncogene 1997 Aug 28; 15(9): 1121]. Oncogene 14: 1377–1382, 1997 31. Harris SE, Harris MA, Mahy P, Wozney J, Feng JQ, Mundy GR: Expression of bone morphogenetic protein messenger RNAs by normal rat and human prostate and prostate cancer cells. Prostate 24: 204–211, 1994 32. Bentley H, Hamdy FC, Hart KA, Seid JM, Williams JL, Johnstone D, Russell RGG: Expression of bone morphogenetic proteins in human prostatic adenocarcinoma and benign prostatic hyperplasia. Br J Cancer 66: 1159–1163, 1992 33. Hienert G, Kirchheimer JC, Pfluger H, Binder BR: Urokinase-type plasminogen activator as a marker for the formation of distant metastases in prostatic carcinomas. J Urol 140: 1466–1469, 1988 34. Soff G, Sanderowitz J, Gately S, Verrusio E, Weiss I, Brem S, Kwaa H: Expression of plasminogen activator inhibitor type 1 by human prostate carcinoma cells inhibits primary tumor growth, tumor-associated angiogenesis, and metastasis to lung and liver in an athymic mouse model. J Clin Invest 96: 2593–2600, 1995 35. Achbarou A, Kaiser S, Tremblay G, Ste Marie LG, Brodt P, Goltzman D, Rabbani SA: Urokinase overproduction results in increased skeletal metastasis by prostate cancer cells in vivo. Cancer Res 54: 2372–2377, 1994 36. Crowley CW, Cohen RL, Lucas BK, Liu G, Shuman MA, Levinson AD: Prevention of metastasis by inhibition of the urokinase receptor. Proc Natl Acad Sci USA 90: 5021–5025, 1993 37. Liu DF, Rabbani SA: Induction of urinary plasminogen activator by retinoic acid results in increased invasiveness of human prostate cancer cells PC-3. Prostate 27: 269–276, 1995 38. Rabbani SA: Urokinase (uPA) as a mediator of osteoblastic skeletal metastasis. In: Bone Metastasis – Mechanisms
336
39.
40. 41.
42.
43.
44.
and Pathophysiology, R.G. Landers Company, Austin, 1996, pp 137–162. Russell PJ, Bennett S and Stricker P: Growth factor involvement in progression of prostate cancer. Clin Chem 44: 705– 723, 1998 Tisdale MJ: Biology of cachexia. J Natl Cancer Inst 89: 1763–1773, 1997 Chou E and Simons JW: The molecular biology of prostate cancer morbidity and mortality: accelerated death from ejaculate poisoning? Urol Oncol 3: 79–84, 1997 Polascik TJ, Wang Z, Di S, Gurganus RT, Hortopan SC, Eisenberger MA, Carducci MA, OP T’s P, Partin AW: Harvest of intact prostate cancer cells from blood: further characterization of a new assay. J Urol 159: 7, 1998 (Abstract) Racila E, Euhus D, Weiss AJ, Rao C, McConnell J, Terstappen LW, Uhr JW: Detection and characterization of carcinoma cells in the blood. Proc Natl Acad Sci USA 95: 4589–4594, 1998 Wu TT, Sikes RA, Cui Q, Thalmann GN, Kao C, Murphy CF, Yang H, Zhau HE, Balian G, Chung LW: Establishing
human prostate cancer cell xenografts in bone: induction of osteoblastic reaction by prostate-specific antigen-producing tumors in athymic and SCID/bg mice using LNCaP and lineage-derived metastatic sublines. Int J Cancer 77: 887–894, 1998 45. Klein KA, Reiter RE, Redula J, Moradi H, Zhu XL, Brothman AR, Lamb DJ, Marcelli M, Belldegrun A, Witte ON, Sawyers CL: Progression of metastatic human prostate cancer to androgen independence in immunodeficient SCID mice. Nat Med 3: 402–408, 1997 46. Quinn JE, Buhler KR, Brown JM, Bladou E, Lange PH, Vessella RL: A model for prostate cancer bone metastasis: Direct intra-bone injection of prostate cancer xenografts. Bone 23: S579, 1998 (Abstract)
Address for offprints: Paul H. Lange, Professor and Chairman, Department of Urology, Box 356510, Seattle, WA 98195; Tel: 206 543 3918; Fax: 206 543 3272; e-mail:
[email protected]