Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
[CANCER RESEARCH 56. 3042-3046. July 1. 1996] CWR22: The First Human Prostate Cancer Xenograft with Strongly Androgen-dependent and Relapsed Strains Both in Vivo and in Soft Agar1 Moolky Nagabhushan, Casey M. Miller, Theresa P. Pretlow, Joseph M. Giaconia, Nancy L. Edgehouse, Stuart Schwartz, Hsing-Jien Kung, Ralph W. de Vere White, Paul H. Gumerlock, Martin I. Resnick, Saeid B. Amini, and Thomas G. Pretlow2 Departments of Pathology ¡M.N., C. M. M., T. P. P., J. M. G., N. L E., T. G. P.I Genetics ¡S.S.], Molecular Biology and Microbiology [H-J. K.I, Epidemiology and Biostatistics ¡S.B. A.I. and Urology [M. I. K.. T. G. PJ. Case Western Reserve University Medical Center. Cleveland, Ohio 44106, and Departments of Urology [R. W. d. V. W.¡ and Hemati/hgy/Oncology ¡P.H. G.]. University of California Davis Medical Center, Sacramento. California 95817 reported prostate cancer xenografts either (a) regress and disappear never to recur or (b) do not regress significantly after castration. After castration of animals that bear CWR22, serum PSA has fallen 153080-fold; in all but one animal, >50-fold. Physically measurable recurrence is preceded by an elevation of serum PSA. The general availability of this model will offer a unique opportunity to compare a human prostate cancer that regresses after castration with tumor that relapses several months to more than 1 year later in the castrated animals. ABSTRACT Most patients' prostate cancers respond to androgen deprivation but relapse after periods of several months to years. Only two prostate cancer xenografts, LNCaP and PC-346, have been reported to be responsive to androgen deprivation and to relapse subsequently. Both of these tumors shrink slightly, if at all, and relapse less than 5 weeks after androgen withdrawal. After androgen withdrawal, the human primary prostate cancer xenograft CWR22 regresses markedly, and prostate-specific anti gen (PSA) falls up to 3000-fold in the blood of mice. PSA usually returns to normal. In some animals, the tumor relapses and is then designated CWR22R. In these animals, PSA starts to rise approximately 2-7 months, and tumor begins to grow 3-10 months after castration. Animals with CWR22 need to be euthanized because of large tumors 6-12 weeks after the transplantation of CWR22. Androgen proximately 3-4-fold. withdrawal MATERIALS prolongs life ap INTRODUCTION In men, most prostate cancers respond to androgen withdrawal (1) but relapse after the initial response. Few, if any, prostate cancers in humans have been cured by hormonal manipulation (2). Models for the development of experimental therapy are limited. To our knowl edge, only two xenografts, LNCaP (3, 4) and PC-346 (5), have been reported previously to show initial responses to androgen withdrawal followed by resumption of growth. Both of these tumors exhibit deceleration of their rates of growth after androgen withdrawal by castration; occasionally, these tumors regress slightly after castration, usually with less than a 10% reduction in volume. Unlike most hormonally responsive prostate cancers in men, both of these tumors resume a more rapid rate of growth less than 5 weeks after castration (3, 5). PSA3 in the sera of animals bearing LNCaP tumors may decrease up to 8-fold after castration (3). In 1994, we (6) described CWR22, an androgen-dependent, serially transplantable xenograft derived from a primary human prostate can cer that has been used subsequently in many laboratories. After castration, CWR22 shows marked regression, not just deceleration or stabilization of growth, and may regress completely. We now describe serially transplanted CWR22R tumors, xenografts that have relapsed several months to more than 1 year after the castration of the hosts. To our knowledge, this is the first report of a human prostate cancer xenograft that both regresses markedly for long periods of time after androgen withdrawal, and, in a proportion of animals, evolves to a tumor that resumes growth in the castrated animal. All previously Received 4/1/96: accepted 5/1/96. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by NIH Grants DK45770. DK51347, CA57179. CA54031, AND METHODS The origin (7) and detailed characterization (6) of the serially transplantable human primary prostate cancer xenograft CWR22 have been described. CWR22 causes elevations of PSA in mouse peripheral blood that correlate with growth (size and weight) of the tumors in nude mice. PSA in 50 fil tail vein blood was assayed at intervals S5 weeks, as shown in the data from individual experiments, with a slight modification of the method used by us previously (8). Suspensions of single cells for transplantation and for culture in soft agar were obtained by digestion of xenografts with 0.1% Pronase (VWR Scientific. Bridgeport, NJ) as before (6). Nude mice were housed and cared for as described previously (6, 7). Mice with CWR22 were given 12.5-mg sus tained release testosterone pellets (Innovative Research of America, Sarasota, FL) s.c. before receiving tumors and at intervals of 3 months until death. Cells from freshly dissociated, serially transplanted CWR22 and CWR22R that were serially transplanted from CWR22 tumors that relapsed after andro gen deprivation were used in soft agar assays as described by Hamburger and Salmon (9) with the following modifications: MEM+ (culture medium with supplements as described in Ref. 10) with 20% normal (Life Technologies, Inc., Gaithersburg, MD) or charcoal-stripped (Calico Biologicals, Inc., Reamstown, PA) calf serum was used as the solvent for the noble agar layers (Difco, Detroit, MI). Cell suspensions were filtered through a single layer of Nitex (Tetko, Inc., Briarcliff Manor, NY) with a48-f¿m porosity immediately before being used for culture. After preliminary experiments to determine the con centrations of cells that would allow for the most satisfactory growth for counting, cells were plated at final concentrations of 0.05 and 0.025 million cells/ml. Cell suspensions were placed in a volume of 0.4 ml at the tip of a horizontal 5-ml tube. Dilutions of testosterone in 0.04 ml were placed near the opening of this horizontal tube, and 3.6 ml 0.33% soft agar at 44°Cwere added rapidly to the tube. The tube was mixed three times by inversion, and 1 ml soft agar (final concentration 0.3%) was plated on top of a previously plated 1-ml layer 0.5% agar. The cells in the several dilutions of testosterone (Sigma Chemical Co., St. Louis, MO) shown below and the control plates were cultured and counted in triplicate. The plates were incubated in a humidified chamber with 5% CO2 in air at 37°C.After 14 days in culture, the number of clusters per plate was counted and plotted as the mean ±SD. Groups of cells greater than eight cells per group were scored (11) and termed "clusters." In preliminary experiments, variables that might affect the soft agar assay were explored. After filtration through 48-(¿mmesh Nitex as described above, CA43703, CA57183. and CA55792. 2 To whom requests for reprints should be addressed, at the Institute of Pathology. Case cells were cultured in soft agar over a wide range of concentrations. Concen trations of 0.05 and 0.025 million cells/ml gave approximately 50-100 clus Western Reserve University. 2085 Adelbert Road, Cleveland, OH 44106. 3 The abbreviation used is: PSA, prostate specific antigen. ters/plate in the absence of testosterone. Cultures observed at intervals of 1 week for 4 weeks gave similar numbers of clusters (more than eight cells per 3042 Downloaded from cancerres.aacrjournals.org on August 2, 2017. © 1996 American Association for Cancer Research. ANDROGEN-DEPENDENT AND RELAPSED PROSTATE CANCER group) of cells at 2 and 3 weeks with decreased numbers of clusters at 4 weeks; the sizes of the clusters did not increase after 2 weeks. Assays were performed in each experiment at both concentrations, and the concentration that gave the number of clusters closest to 100 clusters/plate was selected for evaluation. In >80% of experiments, the 0.05 million cells/ml satisfied these conditions. At this concentration, fewer than two clusters per plate were observed in the day 0 control plates in most experiments. In the early phases of this series of experiments, we occasionally saw more clusters per plate. It became apparent that experiments with more than two clusters per control plate were experi ments in which longer intervals of time had elapsed between filtering the cells through 48-fim Nitex and plating the cells. When more than eight clusters per plate were observed in any control plate, the entire experiment was discarded. Viability assays showed that >99% of the clusters were viable at 2 weeks; viability was assessed with an iodonitrotetrazolium stain (Sigma Chemical Co.) as used by Rosenthal et al. (12). 2.5 5 7.5 DAYS POST RESECTION RESULTS PSA in the sera of mice declined much more rapidly after the resection of tumors than after the castration of mice bearing tumors (Fig. 1). The half-life of PSA after resection (Fig. 1A) appears to be less than 1 day, a value that is consistent with the values reported for a study of the LNCaP xenograft (3). Based on only five animals for which PSA values were obtained at intervals after castration and removal of sustained release testosterone pellets without resection of tumors, the half-life of the serum PSA appears to be variable over a range of approximately 3-8 days. The wide variation observed among different animals in this experiment may be related in part to the fact that PSA was elevated 1 day after castration as compared with the value obtained just before castration in some animals. In the most extreme example of this, seen in animal 2276 (Fig. Iß),the PSA was 522 ng/ml immediately before castration, 1504 ng/ml at 1 day, 558 ng/ml at 2 days, 417 ng/ml at 5 days, and 289 ng/ml at 7 days after castration. We speculate that the elevation of PSA in some animals on the day following androgen deprivation may have been related to injury of the neoplastic cells. The regression of tumors proceeded much more slowly than the decrease in PSA, i.e., the PSA usually declined by 50% in a maximum of 1 week after castration while the tumor volume usually took more than 1 month to decline by 50%; the tumor volume often continued to decline for more than 3 months before becoming stable. The experiment in which mice have been followed for the longest period after castration was a part of a separate ongoing study. His torically, of the 151 animals that have received injections of >1000 CWR22 cells in Matrigel and observed for at least 3 months over the past 3 years, all but 5 have developed tumors. In our longest (>600 days) experiment, CWR22 cells were transplanted by the injection of 8.3 million cells in 0.05 ml Matrigel into single subscapular sites bilaterally in nine animals. By design, when the largest tumor reached 7 mm in its smallest dimension, all nine animals were castrated, and testosterone pellets were removed. The larger of the two tumors was resected from each of five of the animals 1 day after castration; from each of the remaining four animals, 4 days after castration. At the time of castration, PSA values in the blood of these mice ranged between 41 and 282 ng/ml; PSA in the nine animals fell 190-2230-fold after resection of the larger of their two tumors and castration. After castration, five of the nine animals failed to develop recurrent tumors that could be measured with calipers; the remaining four animals developed relapsed tumors with the first evidence of elevation of blood PSA seen 92, 151 (two animals), and 221 days after castra tion (Fig. 2). As detailed below, some of the five mice in which relapsed tumors were not detected grossly showed evidence of the persistence of small amounts of residual tumor. To date, these nine animals have been followed for >600 days after castration or until g 10 1500- 0 5 10 20 DAYS POST CASTRATION Fig. 1. A, decline of PSA in serum of tail vein blood after resection of xenografts and removal of sustained release testosterone pellets. Immediately before resection. PSA ranged from 398 ng/ml in the mouse with the largest tumor (mouse 2264) to 145 ng/ml in the mouse with the smallest tumor (mouse 2282). PSA declines with a half-life in the mouse serum of < 1 day. D, mouse 2262; O. mouse 2264; O. mouse 2266; A, mouse 2280; •¿ mouse 2282. B. decline of PSA after castration of the mice and removal of sustained release testosterone pellets. In some mice, e.g., mouse 2276. PSA was sharply elevated 1 day after castration and removal of the exogenous testosterone. One might speculate that this kind of response to castration may reflect the rapid release of PSA by injured neoplastic cells. In other mice, e.g., mouse 2272, no such elevation was observed; however, there may have been release of PSA by the tumor, since PSA consistently fell more sharply during the second day than during the first day after castration. D, mouse 2270; O, mouse 2272; O, mouse 2274; A, mouse 2276; •¿ mouse 2284. death. Two of the nine animals, (mice 2144 and 2148) are currently alive and without recurrence of growing tumors or abnormal serum PSA. Mice 2146 and 2162 had the lowest PSA values at the time of resection and had tumor burdens estimated from external measure ments of 0.95 and 0.72 ml. After resection of the larger of the two tumors in these two animals, the estimated tumor burdens were 0.29 and 0.27 ml, respectively. Both of these mice died without tumors detectable by caliper measurements or abnormal circulating PSA. At autopsy, mouse 2146 showed no tumor at death 362 days after castration; mouse 2162 died 443 days after castration and was found to have minute amounts of histologically detectable tumor (estimated to be <1 mm3) in the organized Matrigel at the site of injection, despite the normal serum PSA. The fifth animal (mouse 2160) that died without recurrent growth of tumor that could be detected with calipers had a slight elevation (0.8 ng/ml) of PSA that developed 359 days after castration. It died 373 days after castration; despite the abnormal PSA, no tumor could be detected at autopsy including a thorough histopathological analysis of the site of injection. All other 3043 Downloaded from cancerres.aacrjournals.org on August 2, 2017. © 1996 American Association for Cancer Research. ANDROGEN-DEPENDENT AND RELAPSED 600 -i 0 200 400 PROSTATE CANCER measured by calipers for >6 months after castration. Although CWR22R tumors derived from relapsed tumors (a) cause less eleva tion of PSA/g tumor in blood and (b) are more heterogeneous than CWR22 tumors with respect to the amount of PSA/g tumor in blood, the amount of PSA released into the blood by CWR22R has always been sufficient to predict the relapse of tumors at least 1 month before the relapsed growth is detectable with caliper measurements. In other experiments, small numbers of animals with large tumor burdens and serum PSA between 369 and 1142 ng/ml have been castrated, and their sustained release testosterone pellets have been resected. Animals whose tumors have recurred have never shown progressive elevations of the PSA in their sera <2 months after castration. Enlargement of tumors that can be measured with calipers occurs > 1 month and usually many months after the elevation of PSA is first detected. 600 DAYS POST CASTRATION r 1000 Fig. 2. In an experiment designed for a parallel project, animals were injected in each of two sites, located in the subscapular areas bilaterally, with 8.3 million cells in Matrigel. i.e.. 8300-fold more cells than are required to produce tumor in >95% of the animals. When the largest tumor in the 18 injection sites reached 7 mm in its smallest dimension, animals were castrated, and the sustained release testosterone pellets were resected. The larger of the two tumors was resected from each mouse 1 or 4 days after castration, jr. deaths of animals. Four mice developed relapsed tumors after castration and regression of tumors with the elevation of PSA in the tail vein blood being observed first in mouse 2152 (PSA. 4.7 ng/ml at 92 daysl. next in mice 2142 and 2150 (PSA. 2.0 and 0.6 ng/ml at 151 days), and last in mouse 2154 (PSA. 1.2 ng/ml at 221 days). All four of these mice developed progressive elevation of their PSAs and growth of their tumors that were documented histopathologically at the time when they were euthanized 144. 193, 227, and 344 days after castration. D. mouse 2142; < . mouse 2144; O. mouse 2146; A, mouse 2148; •¿ mouse 2150; ».mouse 2152; ».mouse 2154; A. mouse 2160; T. mouse 2162. animals that have developed blood PSA levels of 0.8 ng/ml after months of normal blood PSA levels have developed recurrent tumors. In seven of the nine animals. PSA values dropped to 0.1 or 0.2 ng/ml. The remaining two animals (mice 2152 and 2154) never developed normal PSA values after castration. In mouse 2152, PSA dropped from 114 ng/ml just prior to castration to 0.6 ng/ml at 64 days, 4.7 ng/ml at 92 days, 8.8 ng/ml at 123 days, and 73.1 ng/ml at 144 days after castration; weekly measurements with calipers first suggested relapse 119 days after castration. In mouse 2154. PSA dropped from 201 ng/ml just prior to castration to 0.4 ng/ml at 92 days and again at 123 days after castration. In the blood of this mouse, PSA values were still 0.5 ng/ml at 186 days, 1.2 ng/ml at 221 days, and 13.7 at 254 days before rising progressively to 515.3 ng/ml on the day when the mouse was euthanized, 344 days after castration; measure ment with calipers showed relapse 279 days after castration. Four of the nine animals developed recurrent tumors that could be measured with calipers and resulted in their being euthanized 144, 193. 227, and 344 days after castration. These results are particularly interesting in light of the fact that all injection sites received 8.3 million cells, i.e., 8300 times as many cells as are tumorigenic in >95% of the animals maintained with s.c. sustained release testos terone. Most animals given 1000 cells and sustained release testos terone s.c. are euthanized with large tumors <3 months after the 1000 cells are injected. Fig. 3 shows representative examples of the relationships between PSA in the sera of mice and the growth of tumors as measured with calipers. Mouse 2150 (Fig. 3/4) was castrated on day 0; one of its two tumors was resected on day 4. The tumor that was not resected began growth as a relapsed tumor, as measured by calipers, approximately 180 days after castration; PSA indicated recurrent tumor 151 days after castration. Mouse 2274 (Fig. 3ß)was allowed to develop large tumors bilaterally before castration in an attempt to develop relapsed tumor after castration. The change in PSA as a function of time after castration reversed its course 79 days after castration, indicating relapsed tumor; however, the two tumors did not appear to relapse as L 100 0) 10£ SS o. 0.1 O 100 200 DAYS POST CASTRATION -1000 - 100 1 -100. 0 100 200 DAYS POST CASTRATION Fig. 3. Mouse 2150 and mouse 2274 were in experiments with different protocols that involved castration and resection of their sustained release testosterone pellets followed by the development of relapsed tumors. Although both mice showed elevation of PSA in their blood sera followed by growth of relapsed tumors as measured with calipers, the temporal relationships between these two phenomena were quite different in the two mice. A, animal 2150 had one tumor resected 4 days after castration and resection of the sustained release testosterone pellet, and the remaining tumor was never large. PSA remained normal (<0.25 ng/ml) at 64. 92. and 123 days after androgen withdrawal and became elevated only 1 month prior to the time when the growth of relapsed tumor could be detected with calipers. D, volume (cm1) of tumor in mouse 2150; •¿. level of PSA {ng/ml) in mouse 2150. B, in contrast, mouse 2274, an animal with bilateral large tumors at the lime of castration, developed its lowest PSA level {6.6 ng/ml) in tail vein blood 51 days after androgen withdrawal. It showed the first elevation of PSA with the development of a PSA of 10.4 ng/ml at 79 days after castration, 11.5 ng/ml at 114 days, and progressively higher levels until death at 225 days after castration. As we have observed in other animals with large tumors, the tumors continued to shrink long after PSA had begun to become progressively elevated; and the growth of tumors was not evident from measurement calipers until approximately 6 months after castration. D. volume (cm3) of tumors in mouse 2274; •¿ level of PSA {ng/ml) in mouse 2274. 3044 Downloaded from cancerres.aacrjournals.org on August 2, 2017. © 1996 American Association for Cancer Research. ANDROGEN-DEPENDENT AND RELAPSED PROSTATE CANCER In experiments designed to test the hormonal dependence of CWR22, three female mice and three male mice were given injections of the same suspension of minced tumor in 0.5 ml Matrigel. All three male mice were euthanized 6 weeks later because of large tumors. All three female mice were euthanized 4 months after injection without any evidence of tumor. In another experiment, two female mice and two male mice were given injections of 1.1 million CWR22 cells in Matrigel. The male mice were euthanized 6 and 9 weeks after trans plantation with large tumors. Both female mice were euthanized 6 months after they received these cells but failed to develop tumors. In contrast to CWR22, CWR22R grows in female mice. We have passaged relapsed tumors serially from five different mice. The relapsed tumors, CWR22R, have varied widely in their rates of growth; however, in general, CWR22R tumors have grown more slowly than CWR22. CWR22 tumors generally reach sizes that require that the animal be euthanized 6-12 weeks after transplanta tion. Some of the five strains of relapsed tumors reach similar sizes as early as 9 weeks; most require 3-7 months after transplantation. The basis for their slower growth and the marked differences among different relapsed tumors and their progeny is not yet known. CWR22 and CWR22R differed in their responses to stimulation with testosterone in soft agar (Fig. 4). Cluster formation by CWR22 was increased in the presence of testosterone in a doserelated fashion up to an optimal concentration in the range of 25-35 nvi testosterone. The dose-response curves were parallel in normal calf serum and in charcoal-stripped calf serum with the formation of more clusters in the normal serum than in charcoalstripped serum. In both kinds of sera, cluster formation was in creased 2-3-fold in the presence of testosterone. The response of CWR22R to testosterone was less consistent than that of CWR22. The experiments with CWR22R (Fig. 4) were carried out with different CWR22R tumors that arose as separate events in different animals over different intervals of time. In some experiments, small increases in cluster formation were observed in the presence of doses of testosterone that were optimal for CWR22. In other instances, cluster formation of CWR22R appeared to decrease slightly in the presence of testosterone. The only generalization about cells from all CWR22R tumors that seems very important in the light of the available data is that CWR22R cells responded much less vigorously than did CWR22 cells to testosterone in soft agar. A400i LU j E oc ni 300H O. 2000) CE LU 100O i 60 i 20 40 TESTOSTERONE (nM) B 250i LU ^ 200-1 OC LU o. 150W oc LU 100U —¿r-» 1 50 0 25 125 TESTOSTERONE K r625 (nM) Fig. 4. A, response of CWR22 lo testosterone in soft agar. G. O, O, and A, CWR22 growth with charcoal-stripped serum in four different trials. •¿ *. •¿, and A, CWR22 growth with normal serum in four different trials done in the same experiments as the above trials. B. response of CWR22R to teslosterone in soft agar. O, O, and A, CWR22R growth with charcoal-stripped serum in three different trials. *, •¿. and A. CWR22R growth with normal serum in three different trials done in parallel with the above trials. CWR22 and the tumors that have relapsed after the withdrawal of androgen. CWR22R. respond differently to testosterone when cultured in soft agar. The experiments shown with CWR22R were carried out with tumors that relapsed in different animals at different times. Although the responses of CWR22R were slightly more heterogeneous than those of CWR22. the magnitudes of the responses of CWR22R to testosterone were much smaller than those observed for CWR22 with normal serum and with charcoal-stripped serum. DISCUSSION CWR22R enhances the value of CWR22 as a relatively unique model of human prostate cancer. The availability of CWR22 and CWR22R as serially transplanted xenografts provides prostate cancer researchers, for the first time, a xenograft derived from a primary prostate cancer that regresses markedly after androgen deprivation and relapses as new tumor growth in approximately one quarter to one half of the mice usually 3-10 months after castration, and 1-5 months after serial measurements of PSA in the tail vein blood has heralded the impending relapse. Even very large CWR22 tumors shrink by >50% after tumor-bearing mice are castrated, and tumors of < l g at the time of castration usually regress to <3 mm in diameter even when they are fated to recur. Both LNCaP (3, 4) and PC-346 (5) are reported to be androgen responsive; however, they regress only slightly, if at all, and only for 3 to 5 weeks after tumor-bearing mice are castrated (3, 5). The decrease in PSA in the blood of castrated mice bearing CWR22 is much greater than has been reported for other prostate cancer xenografts that relapse following regression. PSA provides a very sensitive indication of the growth of CWR22 (6) that is independent of other indicators of tumor growth such as size as measured with calipers. CWR22 and the relapsed CWR22R provide investigators with an approach to the investigation of prostate cancer that is relatively unique. Differences between CWR22 and the two other prostate cancer xenografts for which relapsed and hormonally responsive variants are available include the marked physical regression of CWR22 that is measurable with calipers for months after castration and the marked drop in PSA in the peripheral blood of animals bearing CWR22 after castration. Note Added in Proof The two animals still alive and without evidence of tumor >600 days after castration (Fig. 2) were given sustained release testosterone s.c. 603 and 609 days after castration. One died without tumor 3 weeks later; the other was euthanized with tumor 7 weeks later. 3045 Downloaded from cancerres.aacrjournals.org on August 2, 2017. © 1996 American Association for Cancer Research. ANDROGEN-DEPENDENT AND RELAPSED REFERENCES 1. Resnick. M. I., and Grayhack. J. T. Treatment of stage IV carcinoma of the prostate. Ural. Clin. North Am., 2: 141-161, 1975. 2. Schroder, F. H. Endocrine therapy: where do we stand and where are we going? In: L. M. Franks (ed.). Prostate Cancer Cell and Molecular Mechanisms in Diagnosis and Treatment. Cancer Surveys. Vol. 11, pp. 177-194. Plainview, NY: Cold Spring Harbor Laboratory Press. 1991. 3. Cleave. M. E., Hsieh. J-T., Wu, H-C, von Eschenbach, A. C. and Chung, L. W. K. Serum prostate specific antigen levels in mice bearing human prostate LNCaP tumors are determined by tumor volume and endocrine and growth factors. Cancer Res., 52: 1598-1605, 1992. 4. Zhao, X.. van Steenbrugge. G. J.. and Schröder, F. H. Differential sensitivity of hormone-responsive and unresponsive human prostate cancer cells (LNCaP) to tumor necrosis factor. Urol. Res., 20: 193-197, 1992. 5. van Steenbrugge. G. J.. van Weerden. W. M., de Ridder. C. M. A., van der Kwast, T. H., and Schröder.F. H. Development and application of prostatic xenograft models for the study of human prostate cancer. In: M. Motta and M. Serio (eds.). Sex Hormones and Antihormones in Endocrine Dependent Pathology: Basic and Clinical Aspects, pp. 11-22. Amsterdam: Elsevier. 1994. 6. Wainstein, M. A.. He, F.. Robinson. D.. Kung. H-J.. Schwanz. S.. Giaconia, J. M., Edgehouse. N. L., Pretlow. T. P., Bodner. D. R.. Kursh. E. D.. Resnick. M. I., Seftel. 7. 8. 9. 10. 11. 12. PROSTATE CANCER A., and Pretlow. T. G. CWR22: androgen-dependenl xenograft model derived from a primary human proslatic carcinoma. Cancer Res.. 54: 6049-6052. 1994. Pretlow. T. G.. Wolman, S. R., Micale, M. A., Pelley, R. J.. Kursh. E. D.. Resnick, M. I., Bodner. D. R., Jacobberger. J. W.. Delmoro. C. M.. Giaconia, J. M., and Pretlow. T. P. Xenografts of primary human prostatic carcinoma. J. Nati. Cancer Inst., 85: 394-398. 1993. Pretlow. T. G., Pretlow. T. P.. Yang, B., Kaetzel, C. S.. Delmoro. C. M.. Kamis. S. M.. Bodner. D. R.. Kursh, E., Resnick, M. I., and Bradley. E. L., Jr. Tissue concentrations of prostate-specific antigen in prostatic carcinoma and benign prostatic hyperplasia. Int. J. Cancer. 49: 645-649, 1991. Hamburger, A. W.. and Salmon, S. E. Primary bioassay of human tumor stem cells. Science (Washington DC), 197: 461-463, 1977. Pretlow, T. G., Ogrinc. G. S., Amini. S. B.. Delmoro. C. M., Molkentin, K. F.. Willson, J. K. V., and Pretlow, T. P. A better defined medium for human prostate cancer cells. In Vitro Cell. Dev. Biol.. 29A: 528-530. 1993. Buick, R. N., Stanisic, T. H., Fry, S. E.. Salmon. S. E.. Trent, J. M.. and Kransovich. P. Development of an agar-methylcellulose clonogenic assay for cells in transitional cell carcinoma of the human bladder. Cancer Res.. 39: 5051-5056, 1979. Rosenthal. A.. Lindquist, P. B.. Bringman. T. S.. Goeddel, D. V., and Derynck. R. Expression in rat fibroblasts of a human transforming growth factor-a cDNA results in transformation. Cell, 46: 301-309, 1986. 3046 Downloaded from cancerres.aacrjournals.org on August 2, 2017. © 1996 American Association for Cancer Research. CWR22: The First Human Prostate Cancer Xenograft with Strongly Androgen-dependent and Relapsed Strains Both in Vivo and in Soft Agar Moolky Nagabhushan, Casey M. Miller, Theresa P. Pretlow, et al. Cancer Res 1996;56:3042-3046. Updated version E-mail alerts Reprints and Subscriptions Permissions Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/56/13/3042 Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected]. Downloaded from cancerres.aacrjournals.org on August 2, 2017. © 1996 American Association for Cancer Research.