INFORMATION TO USERS
This manuscript has been reproduced from the microfihn master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type o f computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Infomiation Company 300 North Zed) Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600 i j DEVELOPMENT OF DRUGS FOR HUMAN PROSTATE CANCER
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the
Graduate School of The Ohio State University
By
Chiung-Tong Chen, M.S.
*****
The Ohio State University
1997
Dissertation Committee: Approved by
Dr. M. Guillaume Wientjes, Adviser
Dr. Jessie L.-S. Au
Dr. William L. Hayton Adviser
Dr. Kenneth K. Chan College of Pharmacy UMI Number: 9813235
UMI Microform 9813235 Copyright 1998, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, Ml 48103 ABSTRACT
Prostate cancer represents a growing public health problem in the United States and
Europe. In 1997, an estimate of 43% of total newly diagnosed cancers (334.500 new cases) in men were adenocarcinomas of the prostate, the most frequent malignancy in men currently.
Fifteen to twenty-five percent of newly diagnosed cases had been reported as advanced stages o f diseases. The benefit o f survival from currently available treatment is minimal. It is projected that 41,800 men will die of prostate cancer in the United States in 1997. Progress in the understanding of prostate tumor biology and the development of effective therapy has been hampered by lack of clinically relevant tumor models. The work presented in this dissertation focused on the development of clinically relevant prostate tumor models for drug activity evaluation, and on the evaluation of pharmacologic effects of two anticancer agents in human prostate tumors. The major conclusions and contributions of the research in this dissertation are as follows: (a) The establishment of pharmacodynamics of paclitaxel (Chapter
2) and doxorubicin (Chapter 4) in histocultures of patient and three human xenograft tumors
(Chapter 7). These data indicate that paclitaxel and doxorubicin are active against early stage and androgen dependent prostate cancer, with the antiproliferative effect being more prominent than the cytotoxic (i.e., cell kill) effect. It is noted that these results are the first to delineate the two effects separately, (b) The apoptotic effect of paclitaxel in patient tumors in their native 3-dimensional state is cell cycle specific (Chapter 3). These results confirm the previously reported cell cycle specificity of paclitaxel-induced apoptosis in monolayer cultures o f human cancer cells, and indicate that the microenvironment of solid tumors do not alter this property of paclitaxel effect, (c) the concentrations of doxorubicin delivered to prostate tissue in dogs, after an intravenous injection, are insufiBcient to deliver the concentrations needed to produce >50% antiproliferation and cytotoxicity, respectively, in patient tumors (Chapters
5 and 6). These results demonstrate that insufficient drug delivery to tumors is one cause of the lack of clinical response, and have led to the development of regional therapy where the drug is delivered in high concentrations directly to the prostate, (d) The CWR22, CWR22R and CWR91 human prostate xenograft tumors show pharmacodynamics of paclitaxel and doxorubicin that are qualitatively and quantitatively similar to those in patient tumors
(Chapter 7). It is noted that these models are the first to show responses to drugs similar to patient tumors; other models such as monolayer cultures of human cancer cells overpredict chemosensitivity by >10 fold. These results establish the three xenograft tumors as clinically relevant models for drug activity evaluation. As discussed in Chapter 8, Conclusions and
Perspectives, the results described in this dissertation has led to several avenues of research in our laboratory that focus on the development of new effective treatments for prostate cancer, a deadly disease with few treatment options.
Ill Dedicated to my parents and families
IV ACKNOWLEDGMENTS
I would like to express my most sincere gratitude to my adviser Dr. Wientjes for his considerate and supportiveness throughout this study. The completion of this study encompasses 5 years of my efforts at The Ohio State University. With as much sincerity, I am grateful to Dr. Au for the her advises and her intimate collaboration with Dr. Wientjes, from which intellectual interactions have been built upon and critical scientific demanding drove me toward being mature mind of sciences. I wish to thank my advisory committee members,
Drs. William L. Hayton and Kenneth K. Chan, for helpful advisory discussions.
I want to thank Dr. Gan, Y. and Lu, J. for technical support and sharing study materials and results; Drs. Yen, J., Kuh, H-J., Kuan, H.Y., and Johnson, A. and Millenbaugh,
N., and Jang, S.H. for stimulating discussions; Dr. Koo, P., Johnston, J., Kellough, D A.,
Jones, W., Jurcisek, J., Lembach, M., Koolemans-Beynen, A., and Cai, J. for technical support. Friendships from others in the laboratory and the college have been appreciated.
Special thanks to my parents, parents-in-law and families for their long-term supports in every way. Supports from my wife and the duty as a father have been the motives and joys that help me through. VITA
December 4, 1962 ...... Bom - Taiwan, Republic of China
1986 ...... B.S. in Pharmacy, China Medical College
1988 ...... M.S. in Pharmacology, National Yang-Ming Medical College
1990- 1991...... Vice Researcher, Shu-Lin Research Center, China Chemical & Pharmaceutical Group
1990 - 1991...... Registered Pharmacist
1992 ...... Assistant Research Fellow, Institute of Biological Chemistry, Academia Sinica
1992 - present ...... Graduate Research Associate, The Ohio State University
PUBLICATIONS
Papers
1. Chen, C.T., Au, J.L.-S., Gan, Y.B., and Wientjes, M.G. Differential time dependency of antiproliferative and apoptotic effects of taxol in human prostate tumors. Urol. Oncol., 3: 11-17, 1997.
2. Ho, C.L., Lin, Y.L., Chen, W.C., Yu, H.M., Wang, K.T., Hwang, L.L., and Chen, C.T. Immunogenicity of mastoparan B, a cationic tetradecapeptide isolated from the hornet (Vespa basalis) venom, and its structural requirements. Toxicon,
VI 33:1443-1451, 1995.
3. Ho, C.L., Hwang, L.L., Lin, Y.L., Chen, C.T., Yu, H.M., and Wang, K.T. Cardiovascular effects of mastoparan B and its structural requirements. Eur. J. Pharmacol., 259:259-264, 1994.
4. Ho, C.L., Hwang, L.L., and Chen, C.T. Edema-inducing activity of a lethal protein with phospholipase A, activity isolated from the black-bellied hornet (Vespa basalis) Venom. Toxicon, 31:605-613, 1993.
5. Chen, C.T., Chan, J.Y.H., Barnes, C.D., and Chan, S.H.H. Tonic suppression of baroreceptor reflex by endogenous neurotensin in the rat. Regul. Peptides, 28:23-37, 1990.
Abstracts
1. Gan, Y., Au, J.L.-S., Chen, C.T., and Wientjes, M.G. Cytotoxicity of suramin, geldanamycin, cytochalasin E, and tfiiacetazone in human prostate cancer xenografts. Proc. Am. Assoc. Cancer Res., 38:219, 1997.
2. Au, J.L.-S., Li, D., Gao, X., Gan, Y , Chen, C.T., Ge, S., Johnson, A.L., and Wientjes, M.G. Time-dependent taxol cytotoxicity. Proc. Am. Assoc. Cancer Res., 38:4, 1997.
3. Au, J.L.-S., Li, D., Gao, X., Gan, Y , Chen, C.T., Yen, W.C., Johnson, A.L., and Wientjes, M.G. Time-dependent taxol effect in human and rat solid tumors. Archives of Otolaryngology-Head and Neck Surgery, 4th International Conference on Head and Neck Cancer, 493:193, 1996.
4. Chen, C.T., Gan, Y.B., Au, J.L-S., and Wientjes, M.G. Response of human prostate tumors to taxol: Relationship with expression of multidrug resistance p- glycoprotein and bcl-2. Proc. Am. Assoc. Cancer Res., 37:368, 1996.
5. Weaver, J R., Kalns, J.E., Schmittgen, T.D., Chen, C.T., Lim, C., and Au, J.L.-S. Molecular pharmacodynamic endpoints for antiproliferative agents. Proc. Am. Assoc. Cancer Res., 36:360, 1995.
vn 6. Gan, Y.B., Balturshot, G., Millenbaugh, N.E., Chen, C.T., Kalns, J.E., Lim, C , and Wientjes, M.G., Au, J.L.-S. P-glycoprotein expression in human tumors and its correlation with chemosensitivity to taxol and cell mobility. Proc. Am. Assoc. Cancer Res., 36:334, 1995.
7. Chen, C.T., Au, J.L-S., Burgers, J.K., and Wientjes, M.G. Comparative activity of taxol and doxorubicin in human prostate tumors. Proc. Am. Assoc. Cancer Res. 36:320, 1995.
8. Chen, C.T., Au, J.L-S., Burgers, J.K., and Wientjes, M.G. Pharmacodynamics of taxol in human prostate tumors. Pharm. Res., 11:5387, 1994.
9. Au, J.L.-S., Millenbaugh, N.E., Kalns, J.E., Chen, C.T., and Wientjes, M.G. Pharmacodynamics of taxol in human solid tumors. Proc. Am. Assoc. Cancer Res., 35:A2540, 1994.
10. Chen C.T., Chan, J.Y.H., and Chan, S.H.H. Tonic suppression of baroreceptor reflex by endogenous neurotensin in the rat. 4th Annual Meeting of Joint Biomedical Sciences, Taipei, Taiwan, Republic of China. 1988.
FIELDS OF STUDY
Major Field: Pharmacy
vni TABLE OF CONTENT
CHAPTER
INTRODUCTION...... 1 I. I The prostate gland ...... 2 1.2 Overview of clinical status of prostate c a n c e r ...... 4 1.3 Clinical therapies for prostate cancer ...... 7 1.3.1 Therapies for localized prostate cancer ...... 7 1.3.2 Therapies for metastatic prostate c a n c e r ...... 8 1.3.2.1 Hormonal therapy ...... 8 1.3.2.2 Radiation therapy ...... 9 1.3.2.3 Chem otherapy ...... 10 1.4 Models for preclinical studies in prostate cancer ...... 12 1.5 Paclitaxel (Taxol*) ...... 14 1.6 Doxorubicin (Adriamycin®) ...... 16 1.7 Overview of the dissertation ...... 18
CHAPTER 2
DIFFERENTIAL TIME DEPENDENCY OF ANTIPROLIFERATIVE AND APOPTOTIC EFFECTS OF PACLITAXEL IN HUMAN PROSTATE TUMORS 26 2.1 Introduction ...... 26 2.2 Materials and methods ...... 29 2.2.1 Chemicals and reagents ...... 29 2.2.2 Tumor procurement ...... 29 2.2.3 Histoculture ...... 30 2.2.4 Pharmacologic effects of paclitaxel ...... 30 2.2.5 Pharmacodynamic data analysis ...... 32 2.2.6 Statistical analysis ...... 32 2.3 R esu lts...... 34 2.3.1 Histocultures ...... 34 2.3.2 Paclitaxel-induced inhibition of DNA synthesis ...... 34 2.3.3 Paclitaxel-induced apoptosis ...... 35 2.3.4 Tumor pathology and paclitaxel effects ...... 35 2.4 Discussion ...... 37
IX 2.4.1 Antiproliferative and apoptotic effects ...... 37 2.4.2 Pharmacodynamics - Effect of exposure time ...... 38 2.4.3 Pharmacodynamics - Effect of drug concentration ...... 39 2.5 Conclusions ...... 40 2.6 Acknowledgments ...... 41
CHAPTER 3
PACLITAXEL-INDUCED APOPTOSIS: A CELL CYCLE SPECIFIC EVENT ...... 47 3.1 Introduction ...... 47 3.2 Materials and methods ...... 49 3.2.1 Chemicals and supplies ...... 49 3.2.2 Monolayer cell cultures ...... 49 3.2.3 Procurement of tumor specimens and histocultures ...... 50 3.2.4 Drug treatment ...... 51 3.2.5 BrdUrd labeling index and inununohistochemistry of BrdUrd . 52 3.2.6 Apoptotic index and TUNEL assay ...... 52 3.2.7 Effects of hydroxyurea on cell cycle progression and paclitaxel- induced apoptosis ...... 53 3.2.8 Data analysis ...... 54 3.2.9 Statistical analysis ...... 54 3.3 R esu lts...... 55 3.3.1 Delayed cytotoxicity in PC3 and DU145 cells ...... 55 3.3.2 Histoculture: Inhibition of DNA synthesis by paclitaxel and hydroxyurea ...... 55 3.3.3 Histoculture: Induction of apoptosis by paclitaxel and hydroxyurea ...... 56 3.3.4 Histoculture: Correlation between cell proliferation and paclitaxel- induced apoptosis ...... 57 3.3.5 Histoculture: Kinetics of apoptotic induction by paclitaxel 58 3.4 Discussion ...... 59 3.5 Conclusions ...... 61 3.6 Acknowledgments ...... 62
CHAPTER 4
PHARMACODYNAMICS OF DOXORUBICIN IN HUMAN PROSTATE TUMORS ...... 72 4.1 Introduction ...... 72 4.2 Materials and methods ...... 75 4.2.1 Chemicals and supplies ...... 75 4.2.2 Tumor procurement ...... 75 4.2.3 Histoculture ...... 76 4.2.4 Detection of PSA in media and in histocuitured tissues ...... 76 4.2.5 Pharmacologic effects of doxorubicin ...... 77 4.2.6 Autoradiography ...... 78 4.2.7 TUNEL assay ...... 78 4.2.8 Pharmacodynamic data analysis ...... 79 4.2.9 Statistical analysis ...... 80 4.3 Results...... 81 4.3.1 Histocultures ...... 81 4.3.2 Antiproliferative effect ...... 81 4.3.3 Cytotoxic effect ...... 82 4.3.4 Comparison of pharmacodynamics of antiproliferative and cytotoxic effects...... 82 4.4 Discussion ...... 84 4.5 Acknowledgments ...... 86
CHAPTER 5
TISSUE PENETRATION OF DOXORUBICIN INTO PROSTATE XENOGRAFT TUMOR HISTOCULTURES...... 90 5.1 Introduction ...... 90 5.2 Materials and methods ...... 92 5.2.1 Chemicals and instruments ...... 92 5.2.2 CWR22 prostate tumor xenograft growth in vivo ...... 93 5.2.3 Histocultures and daig treatments ...... 93 5.2.4 Concentration analysis ...... 94 5.2.4.1 Tissue samples ...... 94 5.2.4.2 Medium samples ...... 94 5.2.4.3 Collagen gel and plate ...... 95 5.2.5 High performance liquid chromatography assay ...... 95 5.2.6 Data analysis ...... 95 5.3 R esu lts...... 97 5 .3 .1 Profiles of doxorubicin concentration in the culture medium . . 97 5.3.2 Tissue concentrations of doxorubicin in the CWR22 xenograft tumors ...... 97 5.3.3 Distribution ratio of doxorubicin between tumor specimens and m edium ...... 98 5.3.4 Destruction of tumor tissue by doxorubicin ...... 98 5.3.5 Distribution of doxorubicin to the culturing gel and plate .... 98 5.4 Discussion ...... 99 5.5 Conclusions ...... 101 5.6 Acknowledgments ...... 102
XI CHAPTER 6
DOXORUBICIN CONCENTRATIONS IN DOG PROSTATE AFTER SYSTEMIC ADMINISTRATIONS...... 109 6.1 Introduction ...... 109 6.2 Materials and methods ...... I l l 6.2.1 Chemicals and Instruments ...... I l l 6.2.2 Animal protocols ...... I l l 6.2.3 Sample extraction and analysis ...... 112 6.2.4 High performance liquid chromatography analysis ...... 113 6.2.5 Data analysis ...... 113 6.4 Results...... 115 6.4.1 Plasma concentrations ...... 115 6.4.2 Prostate concentrations ...... 115 6.5 Discussion ...... 117 6.6 Acknowledgment ...... 121
CHAPTER 7
ANDROGEN-DEPENDENT AND -INDEPENDENT HUMAN PROSTATE TUMOR XENOGRAFTS AS MODELS FOR DRUG ACTIVITY EVALUATION ...... 126 7.1 Introduction ...... 126 7.2 Materials and methods ...... 130 7.2.1 Chemicals and supplies ...... 130 7.2.2 PaticiiL tuniors ...... 130 7.2.3 Transplantation of xenograft tum ors ...... 131 7.2.4 Histoculture ...... 131 7.2.5 Double labeling of ^H-thymidine and BrdUrd ...... 132 7.2.6 Immunohistochemistry ...... 133 7.2.7 Dectection of Pgp, p53, Bcl-2 and PSA by Western blot analysis ...... 133 7.2.8 Pharmacologic effects of paclitaxel and doxorubicin ...... 134 7.2.9 Terminal deoxynucleotidy transferase (TdT)-mediated dUTP nick end labeling (TUNEL) a s s a y ...... 135 7.2.10 Quantitation of PSA secretion ...... 136 7.2.11 Data analysis ...... 137 7.2.12 Statistical analysis ...... 138 7.3 R esu lts ...... 139 7.3.1 Double labeling and the correlation between ^H-thymidine and BrdUrd L I ...... 139 7.3.2 Histocultures of xenograft tum ors ...... 139 7.3.3 Expression of Pgp, PSA, p53 and Bcl-2 proteins ...... 140
Xll 7.3.4 Daig effects in xenografts tum ors ...... 141 7.3.5 Comparison of drug effects in xenograft and patient tumors .. 143 7.4 Discussion ...... 144 7.5 Acknowledgments ...... 147
CHAPTER 8
CONCLUSIONS AND PERSPECTIVES ...... 155
APPENDICES ...... 157
APPENDIX A DATA RELEVANT TO CHAPTER 2 ...... 158 APPENDIX B DATA RELEVANT TO CHAPTER 3 ...... 163 APPENDIX C DATA RELEVANT TO CHAPTER 4 ...... 171 APPENDIX D DATA RELEVANT TO CHAPTER 5 ...... 174 APPENDIX E DATA RELEVANT TO CHAPTER 6 ...... 179 APPENDIX F DATA RELEVANT TO CHAPTER 7 ...... 183
BIBLIOGRAPHY...... 194
xin LIST OF TABLES
Table I. I. Clinical staging and current standard therapies ...... 22
Table 1.2. Preclinical models of human prostate cancers ...... 23
Table 2.1. Patient and tumor characteristics, and tumor sensitivity to 24 and 96 hr paclitaxel treatments ...... 42
Table 2.2. Relationships between tumor sensitivity to 24 hr and 96 hr paclitaxel treatments ...... 43
Table 3.1. Immediate and delayed effects of paclitaxel in monolayer PC3 and DU 145 cell c u ltu re s...... 63
Table 4.1. Patients and tumor characteristics and tumor sensitivity of tumors to 96 hr doxorubicin treatments ...... 87
Table 5.1. Kinetic parameters of doxorubicin uptake in CWR22 xenograft tumor histocultures ...... 108
Table 6.1. Summary of pharmacokinetic parameters of doxorubicin in the beagle dog after a 2 mg/kg intravenous bolus administration ...... 125
Table 7.1. Expression of PSA, Pgp, Bcl-2 and p53 proteins in patient and xenograft tu m o rs ...... 148
Table 7.2. Sensitivity of patient and xenograft tumors to doxorubicin and paclitaxel . 149
XIV LIST OF FIGURES
Figure 1.1. Prostatic acini with prostatic specific antigen immunostaining ...... 24
Figure 1.2. Chemical structures of paclitaxel and doxorubicin ...... 25
Figure 2.1. Prostate tumor growth in histoculture ...... 44
Figure 2.2. Inhibition of thymidine LI as a function of extracellular paclitaxel concentration ...... 45
Figure 2.3. Apoptotic index as a function of extracellular paclitaxel concentration . 46
Figure 3.1. Schematic illustration of protocols for paclitaxel treatments in monolayer cell cultures and histocultures ...... 64
Figure 3.2. Activity of paclitaxel on human prostate cancer cells by SRB assay .... 65
Figure 3.3. Effects of hydroxyurea pretreatment on paclitaxel activity in patient prostate tumors ...... 66
Figure 3.4. Effects of hydroxyurea pretreatment on paclitaxel activity in CWR22 tu m o rs ...... 67
Figure 3.5. Effects of sequential treatments with paclitaxel and hydroxyurea in CWR22 tumors ...... 68
Figure 3.6. Correlation between paclitaxel-induced apoptosis and BrdUrd L I ...... 69
Figure 3.7. Effects of hydroxyurea on paclitaxel-induced apoptosis in prostate tu m o r s ...... 70
Figure 3.8. Kinetics of apoptotic induction by paclitaxel in patient prostate tumors . 71
Figure 4.1. Prostate tumor histoculture ...... 88
Figure 4.2. Concentration-dependent antiproliferative and cytotoxic effects ...... 89
XV Figure 5.1. Concentration-time profiles of doxorubicin in culture medium ...... 103
Figure 5.2. Tissue concentrations of doxorubicin in CWR22 tumor histocultures . 104
Figure 5.3. Increase of AUG» ratios with initial drug concentration in culture m edium ...... 105
Figure 5.4. Structural changes in tumor histocultures after doxorubicin treatment . 106
Figure 5.5. Recovery of doxorubicin form the collagen gel matrix ...... 107
Figure 6.1. Doxorubicin concentration in plasma ...... 122
Figure 6.2. Doxorubicin concentration in prostate ...... 123
Figure 6.3. Attainment and maintenance of quasi-equilibrium between plasma and prostate concentrations of doxorubicin ...... 124
Figure 7.1. Correlation between tumor cells labeling by BrdUrd and ^H-thymidine 150
Figure 7.2. BrdUrd labeling; TUNEL assay; and PSA, p53 and Pgp immunohistochemistry ...... 151
Figure 7.3. Concentration-dependent antiproliferation induced by paclitaxel and doxorubicin ...... 152
Figure 7.4. Concentration-dependent cytotoxicity induced by paclitaxel and doxorubicin ...... 153
Figure 7.5. Concentration-dependent inhibition of PSA secretion ...... 154
XVI CHAPTER 1
INTRODUCTION
Prostate cancer represents a growing public health problem in the United States and
Europe. In 1997, an estimate of 43% of total newly diagnosed cancers (334,500 new cases) in men were adenocarcinomas of the prostate, the most frequent malignancy in men currently.
Fifteen to twenty-five percent of newly diagnosed cases had been reported as advanced stages of diseases. The benefit of survival from currently available treatment is minimum. It is projected that 41,800 men will die of prostate cancer in the United States this year. Progress in the understanding of prostate tumor biology and the development o f effective therapy has been hampered by lack of clinically relevant tumor models. The work presented in this dissertation focused on the development of clinically relevant prostate tumor models for drug activity evaluation, and on the evaluation of pharmacologic effects o f two anticancer agents in human prostate tumors. 1.1 The prostate gland
The prostate is the largest male accessory gland. It resembles a walnut in shape, surrounds the urethra and locates immediately inferior to the neck of the urinary bladder and anterior to the rectum. It weighs about 25 g in the adult. The prostate is divided into four histologically distinct zones, including (a) the anterior fibromuscular stroma zone that occupies the anterior surface of the prostate, (b) the peripheral zone, which accounts for 70% of prostatic glandular volume and is the most frequent (70%) site of origin of prostate carcinoma, (c) the central zone which accounts for 15-20% o f prostatic volume, is the most frequent site of benign prostatic hyperplasia, and is 5-10% o f the site for prostate cancer, and
(d) the transitional zone which accounts for 5% of the gland and 20% of prostate cancer incidence (Oesterling et al., 1997). The prostate is comprised of 30 to 50 tubuloaveolar glands that open into the prostatic urethra. Acinar cells composed of secretory and basal epithelial cells which line up around the lumen of the tubular structure as shown in Figure 1.1.
Stem cells and neuroendocrine cells are also found in the prostate. Prostatic secretion constitutes 20-30% of the normal human ejaculate and contains various enzymes, lipids, metal ions and amines (Newman 1996). Prostatic acid phosphatase and prostatic specific antigen
(PSA) are two prostate-specific enzymes secreted into the lumina of the prostate ducts as visualized by immunostaining in Figure 1.1. The serum levels of these two enzymes have been used as clinical markers for detection of prostate carcinomas. PSA is considered the most sensitive and specific tumor marker, compared to other biomarkers of all other tumors.
PSA is a serine protease o f the kallikrein family, originally identified in seminal fluid in 1971
(Hara et al., 1971), purified in 1973 (Li and Beling 1973) and isolated from human prostate tissue in 1979 (Wang et al., 1979).
Although the functions of the prostate gland are not fully known, several functions for the prostate gland have been identified. The prostate acts as an exocrine gland by secreting enzymes like fibrinolysin, coaguiase and other coagulum lysing enzymes to the seminal fluid to facilitate the access of sperm to ova. Prostatic fluid preserves sperm viability by reducing the acidity of the urethra and its constituents facilitate and enhance the sperm motility and penetration. The secreted prostatic acid phosphatase hydrolyses phosphorylcholine to choline that serves as a nutrient for spermatozoa. The high level of zinc in prostatic secretion provides antibacterial activity in the seminal fluid (Fair and Wehner, 1976). The prostate also acts as an endocrine gland by helping to convert testosterone to the more potent dihydroxytestosterone and thus affects both hypothalamic and hypophyseal functions. The prostate gland through its anatomical location, physically involved in the control of urine output and in the transmission of seminal fluid during ejaculation. 1.2 Overview of clinical status of prostate cancer
Prostate cancer has become a major public health problem, increasing in prevalence
from being the 10th most frequently diagnosed cancer in the world fourteen years ago to the
most frequent malignancy in men currently. In 1997, an estimate o f 43% o f total newly
diagnosed cancers in men were adenocarcinomas of the prostate (Parker et al., 1997). Death
due to prostate cancer accounts for about 14% of all cancer death and is only second to the
lung cancer in American men. It is projected that 334,500 new cases will be diagnosed and
41,800 men will die o f prostate cancer in the United States in 1997 (Parker et al., 1997).
In spite of its common occurrence, the etiology of the prostate cancer is still unclear.
Prostate cancer is a slowly progressing disease, and is usually diagnosed in older men at a
median age of 70.5 years with fewer than 2% of patients less than 50 years old. In addition
to age, the risk factors include race, environment, testosterone levels, family history, dietary
fat, and 5-alpha reductase activity. For example, prostate cancer is more common in
American blacks than in whites. In contrast, black Africans have a low incidence o f prostate
carcinoma and life expectancy is about 5 year in most African countries. However, the frve- year survival rate for American whites (58%) is higher than that (47%) for American blacks
(Murphy et al., 1982). The incidence of prostate cancer usually is low in Asia, but is increasing rapidly in Asians who have immigrated to the United States (Hanks et al., 1993).
Prostatic intraepithélial neoplasia (PEN) has been considered to be latent preneoplastic lesions
(Bostwick and Drawer, 1987). High grade PIN is considered to be a precursor to invasive carcinoma. Although the incidence of latent preneoplastic lesions is similar in Japan and in the United States, the incidence o f clinical prostate carcinoma in Japan is much lower than that in the United States. This suggests an important role for environmental factors in prostatic carcinogenesis (Carter and Cofifey, 1990).
An emphasis on early detection and the availability of new screening techniques, in particular the detection of PSA has increased the rate of diagnosis and lowered the stage and age at the time of diagnosis (Mettlin et al., 1993). Adenocarcinoma o f acinar origin accounts for 98% of all cancer of the prostate. Métastasés to bone is common with primary carcinomas of the prostate. In addition, prostate tumor frequently spreads to distant lymph nodes. Lung métastasés are uncommon. Liver métastasés are usually seen late in the course of the disease.
Statistics showed that 56, 15, and 15% of new cases present with localized, regional, and distant diseases, respectively, while the rest was not staged (Parker et al., 1997).
Accordingly, in some studies, 25% or higher percent of the patients were found to have advanced disease at the time of diagnosis (Alivizatos and Oosterhof, 1993; Daneshgari and
Crawford, 1993; Gudziak and Smith, 1994). Interestingly, although the incidence of metastatic prostate cancer is high, the prostate itself is rarely associated with métastasés from other tumors (Oesterling et al., 1997).
Compared to many other cancers, prostate cancer has been considered a disease of old age, perceived to have a limited effect on life expectancy. More recently, prostate cancer is receiving more research attention. The continuous increase in life expectancy, the emphasis on early detection, and the availability of new screening techniques are the causes for prostate cancer to become clinically more prevalent. Prostate cancer causes an average life span shortening o f 9 years (Horm and Sondik, 1989). The survival rate is significantly and inversely influenced by the tumor stage at diagnosis. The 5-year survival rates in patients of stage A, B, C and D are 78, 68, 55 and 23%, respectively (Murphy et al., 1982). The currently
available treatments for advanced disease have not significantly improved patent survival, and
are mainly effective as palliative therapy for symptomatic patients.
i 1.3 Clinical therapies for prostate cancer
Radical prostatectomy, radiation therapy, cryotherapy, and deferred therapy (watchful waiting), are used for localized prostate cancer whereas hormonal therapy, radiation therapy and chemotherapy are used for advanced prostate cancer. Neoadjuvant hormonal therapy for localized and neoadjuvant chemotherapy for advanced prostate cancer are under investigation.
Table 1.1 lists the regimens for prostate cancer of different stages (Gamick, 1994).
1.3.1 Therapies for localized prostate cancer
Current treatment options for locally confined disease include surgical radical prostatectomy, radiation therapy, cryotherapy, and deferred therapy (watchful waiting), with surgery being the most common treatment modality. It is controversial if these treatments provide improvement of quality life span (Walsh, 1992). Neoadjuvant hormone therapy prior to radical prostatectomy is under investigation to improve the control of localized prostate cancer. Some studies showed benefits of the neoadjuvant hormone therapy (Vailancourt et al., 1996; Abbas et al., 1996), although another study showed no improvement in the time to disease progression or overall survival (Abbas and Scardino, 1996).
Aggressive therapies like radical surgery or radiation therapy are controversial, because they compromise the quality of life with no apparent survival advantage in early stage diseases. A cancer-specific survival of 90% at 13.5 years has been reported for radical prostatectomy (Paulson, 1994), suggesting the ability o f radical prostatectomy to be effective at long-term cancer control in young patients who will live long enough to benefit from the surgical treatment. However, impotence and incontinence have been the side effects, and are of major concern for radical prostatectomy. Thirty percent of patients who undergo radical prostatectomy do not regain erectile function even with the use of nerve-sparing procedures
(Gamick, 1994). Aggressive therapy does not offer survival advantage compared to deferring
therapy until patients are symptomatic (Johansson et al., 1992). It has been suggested that
asymptomatic elderly (>65 year-old) patients with well-diflferentiated adenocarcinomas should
not receive therapeutic interventions until patients become symptomatic (Johansson, 1994;
Albertsen et al., 1995; Oesterling et al., 1997). However, in a clinical study that evaluated
deferred therapy, 75% of asymptomatic patients had clinically localized cancer at the time of
diagnosis. Most of these patients became symptomatic soon after diagnosis and only 11% of
patients survived 5 years without any further treatment (Handley et al., 1988). The choice for
the treatment of early stage disease is now suggested by the physician and the final decision
is made by the patient himself.
1.3.2 Therapies for metastatic prostate cancer
1.3.2.1 Hormonal therapy
Hormonal therapy has been the first choice of treatment for advanced prostate cancer.
Since 1941, androgen deprivation, either by surgical castration (bilateral orchiectomy) or by
medical castration (orally administered estrogens and, recently, luteinizing hormone-releasing
hormone (LH-RH) agonists), has been the main approach of hormonal therapy for advanced
prostate cancer. The majority (70-80%) of patients achieve symptomatic relief from initial hormone therapy. However, androgen insensitivity develops in nearly all patients with advanced prostate cancer. The median survival of patients that receive hormonal therapy ranges from 1.8 to 3 years (Alivizatos and Oosterhof, 1993; Gudziak and Smith, 1994).
Adrenal androgens accounts for 5 to 10% of circulating androgens and can not be suppressed
8 by castration. The concept of total androgen ablation was suggested in 1984. The total
androgen ablation using a combination of an LH-RH analog, leuprolide, and an anti-
androgenic agent, flutamide, has been tested in patients with stage D2 cancer in 1986. The
results showed a 5.8-month survival benefit for the combination-therapy group (Daneshgari
and Crawford, 1993). However, this improvement could not be demonstrated in other trials
in Europe and Canada (Denis et al., 1991; Beland et al., 1991).
As discussed above, early prostate cancer is mainly hormone sensitive, but after
androgen ablation, prostate cancer converts to androgen insensitive. Once relapse occurs in
prostate cancer patients after initial hormonal therapy, half of patients die within 6 months of
relapse. The response of patients with relapsed prostate cancer to further hormonal
combination therapies is minimal (Gudziak and Smith, 1994; Korman, 1989). Studies showed
that expression of bcl-2, a protooncogene that counteracts apoptosis (programmed cell
death), is associated with the development of androgen-independent prostate cancer and drug
resistance (McDonnell et al., 1992; Desoize, 1994; Berchem et al., 1995). Further efforts to
explore changes of biological characteristics and changes of sensitivity to anticancer agents
of the prostate cancer during and after the transition from androgen-dependent to androgen-
independent may be of importance.
1.3.2.2 Radiation therapy
Regional radiation therapy is primarily used for localized prostate cancers, staged A,
B and C. For patients with metastatic prostate cancer, radiation therapy is only used as palliative treatment for symptomatic osseous métastasés and for pain relief. Systemic radiation therapy is not used in advanced prostate cancer patients because of serious side efifects because it appears to be more life-threatening than hormonal therapy (Kuban et al.,
1991).
1.3.2.3 Chemotherapy
As discussed above, nearly all patients with advanced prostate cancer develops
androgen insensitivity and no longer responds to hormonal therapy. The final approach is to
use cytotoxic agents. Currently, few anticancer agents produce between 10% to 33% of
objective responses in hormone-refi-actory patients, with an overall median survival o f 30-40
weeks (Horm and Sondik, 1989; Oesterling et al., 1997), which is close to that (1.8 to 3 year)
of hormonal therapy. An important consideration for using cytotoxic drugs is the balance
between the risk of drug-induced side effects and the possible therapeutic benefits. For
example, the dose intensity of doxorubicin is limited due to the quick onset of
myelosuppression and severe cardiac toxicity. Among the agents used to treat advanced
prostate cancer, doxorubicin produces one of the highest combined partial and complete
response rate of 33% (Perez et al., 1989; Oesterling et al., 1997). However, doxorubicin, as
a single agent or in combination with other chemotherapeutics, has not improved the survival
rate (Oesterling et al., 1997, Rangel et al., 1992).
The combination of several anticancer drugs with different mechanisms may provide additive or synergism and is therefore a logical therapeutic approach. However, clinical trials of combination anticancer drug therapy did not show superior results relative to single-agent therapy in terms of survival benefit and sometimes showed increased host toxicity (Korman,
1989). The previously tested combinations primarily consist of several agents that interact with DNA replication and/or functions, and thereby, target rapidly growing cells. The failure
10 of these therapies may be due in part to the fact that prostate cancer is slow-growing; the estimated average proliferative fraction of prostate cancer using different markers of proliferative index ranges from 0.68 to 16.3% (Pretlow et al., 1994) and the in vivo doubling time of tumor volume is estimated at more than 2 years (Schmid et al., 1993). Therefore, treatment directed at DNA replication and rapidly growing cancer may be inappropriate for the treatment of the slow growing prostate cancer. The need for more effective agents with novel mechanisms not targeted at interrupting DNA replication may improve the therapeutic management of advanced prostate cancer.
11 1.4 Models for preclinical studies in prostate cancer
The limited availability of clinically relevant experimental prostate cancer models has contributed to the slow progress in the management of prostate cancer. An ideal clinically relevant animal model is not available due to the fact that prostate cancer rarely arises spontaneously in animals. Although clinically relevant, prostate tumor specimens obtained from individual patients are not renewable sources. Past efforts on developing optimal treatments have focused on evaluating potential agents in clinical trials and on developing relevant experimental models. Table 1.2 summarizes the existing cell lines and transplantable xenograft tumors derived from both primary and metastatic human prostate carcinomas, and their tumorigenicity in mice, androgen sensitivity, and PSA production.
Our laboratory uses histocultures of tumors directly from patients (Wientjes et al.,
1995; Gan et al., 1996a; Gan et al., 1996b), including prostate tumors (Chen et al., 1997a;
Chen et al., 1997b), to determine chemosensitivity. Histocultures of patient tumors in their native state, were first introduced in the 1950's and re-introduced by Hoffman and colleagues
(Vesico et al., 1987; Geller et al., 1992). Histoculture has several advantages over monolayer culture o f cell lines. Histoculture system allows growth and study o f actual patient tumor specimens in three dimension. Native tissue heterogeneity, tissue architecture and cell-cell
(epithelial-stromal) interactions are preserved. Proliferating and non-proliferating populations are maintained. Furthermore, unlike primary cell culture, histoculture does not require chemical or enzymatic dissociation and has a high success rate o f culture. The clinical relevance of the human tumor histoculture system was recently demonstrated by Hoffman and colleagues. These investigators showed in retrospective and semi-prospective preclinical and
12 clinical studies that drug responses in human tumor histocultures, using inhibition of DNA
precursor incorporation or inhibition of metabolic reduction of tétrazolium dye as endpoint,
correlates with the sensitivity, resistance and survival of head and neck, colorectal and gastric
cancer patients to treatment by mitomycin C, doxorubicin, 5-fluorouracii, or cisplatin
(Robbins et al., 1994; Furukawaet al., 1995; Kubota et al., 1995). The limitations o f patient tumor histocultures system are that patient tumors are not renewable source and cannot be readily propagated in animals. An alternative source is transplantable human prostate xenograft tumors that can be cultured and maintained in animals and, therefore, enable drug activity evaluation under in vitro and in vivo conditions. Dr. Thomas G. Pretlow at the Case
Westem Reserve Universit}' has established several human prostate xenograft tumors, i.e.. CWR22.
CWR22R and CWR91. CWR22 is derived from a patient with a Gleason grade 9 primarv- prostate carcinoma stage D and osseous metastasis (Pretlow et al.. 1993). CWR22R is the androgen- independent subline of CWR22 established as recurrence after androgen ablation (Wainstein et al..
1994; Nagabhushan et al.. 1996). CWR91 is an androgen-independent tumor derived from a primaix' prostate tumor of a second patient with Gleason grade 7 and stage C disease (Pretlow et al.. 1993). CWR22 is the first androgen-dependent prostate cancer model that has the relapsed androgen-independent counterpart, CWR22R These two xenograft tumors together with CWR91 represent clinically relevant models for studying changes of biological properties and chemosensitivity in androgen dependent and independent tumors. The present study used these xenograft tumors to evaluate tumor responses to two anticancer drugs, paclitaxel and doxorubicin. In addition, tumor specimens obtained directly from patients were also used.
13 1.5 Paclitaxel (Taxol*)
Paclitaxel is a diterpenoid anticancer drug isolated from the stem bark of the westem yew, Taxiis brevifolia (Wani et al., 1971). Its chemical structure is a taxane ring with a side chain at position C-13 (Figure 1.2). The side chain is essential for the biological activity
(Horwitz et al., 1992). Paclitaxel has a unique mechanism of action. It binds tightly to the
P'tubulin subunit of microtubules (Rao et al., 1992; Schiff et al., 1979). Paclitaxel lowers the critical concentration of tubulin dimers necessary for polymerization, therefore promoting microtubule polymerization and assembly. It also stabilizes the polymers after they are formed. Microtubules are normally in a state of dynamic equilibrium with the a,p tubulin dimers. Proper dynamic balance of microtubule assembly and disassembly is important for cell division, cell motility, cell structure determination and maintenance, organization and position of membranous organelles, and intracellular vesicle transport (Geifand and
Bershadsky, 1991). The alteration of the dynamic balance due to the enhanced microtubule assembly by paclitaxel has been suggested to be responsible for its antitumor activity
(Horwitz, 1994). The major dose-limiting systemic adverse effects of paclitaxel are myelosuppression and peripheral neuropathy. Paclitaxel-induced myelosuppression is countered by co-administration of granulocyte colony stimulating factor. Mucositis is another non-hematological dose-limiting toxicity. Mild toxicities include reversible complete alopecia, symptoms of hypersensitivity, bradycardia, and ventricular tachycardia (Francis, et al., 1995).
In clinical trials, paclitaxel has demonstrated significant activity against human tumors, including metastatic melanoma (Legas et al., 1990), refractory acute leukemia (Rowinsky et al., 1989), advanced ovarian cancer (McGuire et al., 1989; Einzig et al., 1992), metastatic
14 breast cancer (Holmes et ai., 1991), and non-small cell lung cancer (Murphy et al., 1993).
However, paclitaxel at doses of 135-170 mg/m'for 24-hr infusion has only minor activity against hormone-refractory prostate cancer in a phase II trial (Roth et al., 1993). Other clinical trials using the maximally tolerated doses (250-300 mg/m') of paclitaxel and comparing 24-hr infusion to 96-hr infusion are under investigation (Arbuck et al., 1993).
15 1.6 Doxorubicin (Adriamycin*)
Doxorubicin is an anthracycline with antibiotic and anticancer activities. It was
originally isolated from cultures of Streptomyces peiicetms var. caesiiis in 1969 (Arcamone
et ai-, 1969). Doxorubicin is a glycoside, consisting of an aglycone and sugar (Figure 1.2).
Doxorubicin has several action mechanisms, including topoisomerase II inhibition, DNA
intercalation, DNA-protein link breakage, cell membrane interaction and free radical
formation (Dollery, 1991). Doxorubicin is a potent agent against a wide spectrum of
malignancies, including leukemias, sarcomas, and solid tumors such as breast, prostate,
ovarian, and small cell lung cancers (Speth et al., 1988; Oesterling et al., 1997). When given
as systemic infusion, doxorubicin, like most other anticancer agents, induces acute
myelosuppression, stomatitis, esophagitis, reversible complete alopecia, hyperpigmentation
o f nail beds and phalangeal, dermal creases, nausea and vomiting (Dollery, 1991). The
administered cumulative dosage of doxorubicin is primarily limited by its most important chronic and cumulative cardiotoxicity, leading to the congestive heart failure (Crom, 1986).
Prolonged 96-hr infusion is associated with a lower incidence of toxicity compared to a shorter infusion periods (Legha et al., 1982a). Drug resistance is another limitation for the clinical use of doxorubicin. The several mechanisms of resistance to doxorubicin are: decreased topoisomerase H activity, early onset of DNA repair, overexpression of the multidrug resistance P-glycoprotein, hypoxia, decreased NADPHiCytochrome P-450 reductase activity, and increased activity of superoxide dismutase, DT-diaphorase, glutathione transferase, glutathione peroxidase, and catalase (Germann et al., 1993; Deffie et al., 1988;
Teicher, 1988). A great deal of research effort has been directed at overcoming tumor
16 resistance to doxorubicin, because it has the broadest clinical utilities in available anticancer agents and is among the most effective agents for treatment of hormone-refractory prostate cancer (Oesterling et al., 1997). Therefore, the present study selected doxorubicin for investigation and comparison with paclitaxel.
17 1.7 Overview of the dissertation
The remaining chapters in this dissertation are organized according to the study
subjects. Each chapter contains an Introduction which states the research problem, followed
by Materials and Methods, Results, Discussion and Conclusion. Related results of the
different studies are discussed in appropriate chapters. The major conclusions and
contributions of the research in this dissertation are as follows: (a) The establishment of
pharmacodynamics of paclitaxel (Chapter 2) and doxorubicin (Chapter 4) in histocultures of
patient and three human xenograft tumors. These data indicate that paclitaxel and
doxorubicin are active against early stage and androgen-dependent prostate cancer, with the
antiproliferative effect being more prominent than the cytotoxic (i.e., cell kill) effect. It is
noted that these results are the first to delineate the two effects separately, (b) The apoptotic
effect of paclitaxel in patient tumors in their native 3-dimensional state is cell cycle specific
(Chapter 3). These results confirm the previously reported cell cycle specificity of paclitaxel-
induced apoptosis in monolayer cultures of human cancer cells, and indicate that the
microenvironment of solid tumors does not alter this effect of paclitaxel. (c) The
concentrations of doxorubicin delivered to prostate tumor in dogs, after an intravenous
injection, are insufficient to deliver the concentrations needed to produce >50%
antiproliferation and cytotoxicity, respectively, in patient tumors (Chapters 5 and 6). These
results demonstrate that insufficient drug delivery to tumors is one cause of the lack of clinical
response, and have led to the development of regional therapy where the drug is delivered in
high concentrations directly to the prostate, (d) The CWR22, CWR22R and CW R91 human xenograft tumors show pharmacodynamics o f paclitaxel and doxorubicin that are qualitatively
18 and quantitatively similar to those in patient tumors (Chapter 7). It is noted that these models are the first to show responses to drugs as patient tumors; other models such as monolayer cultures of human cancer cells overpredict chemosensitivity by >10 fold. These results establish the three xenograft tumors as clinically relevant models for drug activity evaluation.
As discussed in Chapter 8 , Conclusions and Perspectives, the results described in this dissertation have led to several avenues of research in our laboratory that focus on the development of nine effective treatments for prostate cancer, a dead disease with few treatment options.
Chapter 2 describes the pharmacodynamics of paclitaxel in patient tumor histocultures, and the effect of short (24 hr) and long (96 hr) treatment periods. The results demonstrate (a) incomplete concentration-dependent antiproliferative and apoptotic effects of paclitaxel in human prostate tumors, (b) that neither effect was significantly enhanced by increasing the drug concentration 10 to 100 fold indicating the presence of resistant cells, and
(c) that the antiproliferative eflfect was affected more significantly by drug exposure time than was the apoptotic effect. These results have been published (Chen et al., 1997a).
Chapter 3 describes the cell cycle specificity of paclitaxel-induced apoptosis, in patient tumors and in androgen-dependent CWR xenograft tumors. Cotreatment with an S-phase blocking agent, hydroxyurea, prior to or after paclitaxel treatment prevented paclitaxel- induced apoptosis in these tumors, indicating the requirement of S phase entry and cell cycle progression for this paclitaxel effect.
Chapter 4 describes the pharmacodynamics of doxorubicin in histocultured patient prostate tumors. The results show that doxorubicin inhibits DNA synthesis and induces cell
19 death in a concentration-dependent manner, and that the concentration needed to induce significant cell death is 12 fold higher than the clinically achievable plasma concentration after systemic administration. The study suggested that neoadjuvant or adjuvant systemic doxorubicin therapy has limited value in treating early stage prostate cancer. An alternative treatment option is to use regional drug delivery such as direct intraprostatic injection to deliver high drug concentration in order to achieve appreciable antitumor effect. The manuscript describing these results have been accepted for publication (Chen et al.. 1997b).
Chapter 5 describes the kinetics of doxorubicin uptake into prostate histocultures of
CWR22 prostate tumors. The results indicate (a) drug binding to tumors resulting in an average tissue-to-culture medium cumulative of - 100, (b) drug uptake into tumor is concentration-dependent, and (c) a slow release of drug from tumor after treatment. These results are used with the results in Chapters 4 and 6 to depict that the drug concentrations in prostate tissue, after an intravenous infusion, are insufficient to produce significant antiproliferation and cytotoxicity.
Chapter 6 describes the doxorubicin concentration in prostate tissue of dogs after systemic intravenous bolus and infusion administrations of the maximally tolerated dose (2 mg/kg). The results show that (a) the tissue concentration rose rapidly, reaching a limit that was similar to that of the drug concentration in plasma, and (b) the bolus injection and the 96 hr infusion produced similar tissue concentrations, indicating that both regimens are equally efficient in delivering the drug to the prostate tissues. These data support the use of the 96 hr infusion, which has less cardiac toxicity than bolus injection. Furthermore, these results, together with the results of Chapters 4 and 5 suggest that a direct administration of
2 0 doxorubicin into the targeting sites of prostate tumor holds promise as an effective treatment of prostate cancer.
Chapter 7 describes the pharmacodynamics of paclitaxel and doxorubicin in patient and xenograft tumors. The results indicate that the three xenograft tumors, which show chemosensitivity comparable to the results of 2 50% patient tumors and wtiich encompass the majority of the heterogenous patient prostate tumors in the expression of p-glycoprotein. prostate specific antigen, p53 and Bcl-2 proteins, represent useful models for drug activity evaluation.
CWR xenografts represent renewable clinical relevant models for the development of effective treatments of prostate cancer.
21 STAGES OF DISEASE STANDARD THERAPY
A Microscopic A1 Cancer is confined to one site and is well- Observation, radiation or radical cancer within differentiated prostatectomy prostate gland A2 Cancer occurs in many sites or is moderately to Radiation or radical prostatectomy poorly differentiated
B Palpable lump B1 Cancer forms a small discrete nodule in one lobe Radiation or prostatectomy within prostate of gland gland B2 Cancer forms a large nodule or multiple nodules, or involves both lobes or is moderately to poorly differentiated N) to C Large mass C l Cancer occurs as a continuous mass that may Radiation; some physicians administer involving all or have extended somewhat beyond the gland hormonal therapy with radiation most of prostate gland C2 Large cancer occurs as a continuous mass that has invaded structures surrounding the gland
D Metastatic tumor D1 Cancer appears in the lymph nodes of the pelvis Hormonal therapy once symptoms arise (or possibly as soon as metastatic deposits are found) and palliative D2 Cancer involves tissue beyond the lymph nodes; therapy for pain and other discomforts usually including the bones Table 1.1. Clinical staging and current standard therapies. (Adopted from Gamick, 1994) Tissue of Ongm Tumongcruc in Mice Androgen PSA Year Reported Reference Sensiti\*e Produced
Cell Lines
EB-55 Prostate Yes No No 1974 Okada
DU-U5 Brain Yes No No 1977 Micke\-
PC-3 Vertebrae Yes No No 1978 Kaighn
LN'CAP Lymph Sode Yes Yes Yes 1980 Horoszewicz
PC-93 Prostate No No No 1983 Class
TSL'-PRI L>-mph N'ode Yes No No 1987 ItZ U lT U
JCA-1 Prostate Yes No No 1990 Muroki
DLTRO-I Lymph Node Yes No No I99I Gingnch
ND-l Prostate Yes No Trace 1992 Narayan
XcnoynAs
PC'S: Prostate Yes Yes Yes 1977 Hochn
Honda Tcstis Yes Yes No 1977 [to
9479 Bone Yes No No 1981 Graham
PC.EW Lymph Node Yes Yes Yes 1984 Hochn
PC-133 Bone Yes No No 1985 van.Siccnbrugge
PC-135 prostate Yes No No 1985 \-an Steenbrugge
PC-EG Prostate Yes Yes Yes 1988 Csapo
DC^683 L>mph Xcxie Yes Yes Yes 1982 Gingnch
CWR22 Prostate Yes Yes Yes 1993 (Yellow
CXVKSl Prostate Yes No Trace 1993 Pretlow
C\VK2\ Prostate Yes N.ANA 1995 Pretlow
rW H Zl prostate Yes N A NA 1993 Pretlow
c:\\TC IR Prostate Yes No Yes 1996 Nagabhushan
LuCiP 23 Liver. Yes Yes Yes 1996 EIlis
PC-295 Lymph Node Yes Yes Yes 1996 van VVecrden
PC-310 Prostate Yes Yes Yes 1996 van Weerdcn
PC-524 Prostate Yes No No 1996 %'an Wcerden
PC-329 Prostate Yes Yes Yes 1996 \-an Wcerden
PC-339 Prostate Yes No No 1996 van Wcerden
PC-346 Prostate Yes Yes Yes 1996 van Wcerden
PC-3 74 Skin Yes No Yes 1996 van Wcerden
Table 1.2. Preclinical models of human prostate cancers. N.A.. not available
23 %
L
Figure 1.1. Prostatic acini with prostatic specific antigen immunostaining. Prostatic acini with secretory (s) and basal (b) epithelial cells. Immunoreactivity of prostatic specific antigen is shown around acinus lumen (L).
24 -< Y
A. Paclitaxel (Taxol*)
OH CHgOH
OH
OH
O
CH
NH OH
B. Doxorubicin (Adriamycin*)
Figure 1.2. Chemical structures of paclitaxel and doxorubicin.
25 CHAPTER 2
DEFFERENTIAL TIME DEPENDENCY OF ANTIPROLIFERATIVE AND
APOPTOTIC EFFECTS OF PACLITAXEL IN HUMAN PROSTATE TUMORS
2.1 Introduction
The significant clinical activity of paclitaxel in several human solid tumors, including
metastatic melanoma, refractory acute leukemia, advanced ovarian, metastatic breast, and
nonsmall cell lung cancers (Arbuck and Blaylock, 1995), prompted the phase I/II trial of
paclitaxel as a single agent in advanced prostate cancer patients. The initial trial o f paclitaxel,
at doses of 135-170 mg/m' infused over 24 hr, showed only minor activity with a <5% rate
o f major responses (Roth et al., 1993). Another clinical trial using the maximally tolerated
dose o f paclitaxel at 250 mg/m' is ongoing (Arbuck et al., 1993). There is a controversy
regarding the optimal paclitaxel treatment schedule. The 3-hr treatment schedule is advocated in part because of its application in ambulatory setting (The Cancer Letter, 1993) whereas preclinical data suggest that the therapeutic effect is improved by prolonging treatment to 96 hr (Kelland and Abel, 1992).
Paclitaxel has multiple pharmacologic effects, of which the most pronounced are the enhanced polymerization and stabilization of microtubules and the blockade at the Gj/M
26 interphase (Horwitz, 1994). These actions result in antiproliferation (i.e. inhibition of DNA synthesis) and cell kill (i.e. apoptosis) effects in tumor cell lines, murine solid tumors, and leukemic cells in patients (Horwitz, 1994; Crossin and Carney, 1981; Liu et al., 1994; Bhalla et al., 1993; Milas et al., 1995; Li et al. 1994). Paclitaxel has been recently shown to produce antiproliferative and apoptotic effects in human head and neck tumors (Gan et al., 1996a).
The pharmacodynamics, i.e. drug concentration-exposure time-effect relationship, for these effects in prostate cancer are not known.
The goal of the present study was to determine the pharmacodynamics of the antiproliferative and apoptotic effects of paclitaxel in human prostate tumors. These studies required the evaluation of drug sensitivity in individual patient tumors, and were performed using histocultures of surgical specimens of prostate tumors. The major advantages of the histoculture system are the maintenance of a 3-dimensional tissue structure and organization, co-existence of tumor and stromal cells, cell-cell interaction, and inter- and intra-tumor heterogeneity (Vesico et al., 1987). The maintenance of tissue architecture is critical because the interaction between the tumor and normal cells may be important for prostatic epithelial growth and response to androgen stimulation (Schroeder and Mackensen, 1974; Mickey,
1988). Previous observations from our laboratory reported that human prostate tumor histocultures maintained their characteristics for at least 8 weeks as indicated by unchanged thymidine labeling index (LI) and secretion of prostate specific antigen (PSA) and prostatic acid phosphatase (Wientjes et al., 1995). The clinical relevance o f the human tumor histoculture system was recently demonstrated by Hoffman and colleagues. These investigators show in retrospective and semi-prospective preclinical and clinical studies that
27 drug responses in human tumor histocultures, using inhibition of DNA precursor incorporation or inhibition of metabolic reduction of tétrazolium dye as endpoint, correlates with the sensitivity, resistance and survival of head and neck, colorectal and gastric cancer patients to treatment by mitomycin C, doxorubicin, 5-fluorouracil, or cisplatin (Robbins et al.,
1994; Furukawa et al., 1995; Kubota et al., 1995).
28 2.2 Materials and methods
2.2.1 Chemicals and reagents
Paclitaxel was a gift from Bristol-Myers Squibb Co. (Wallingford, CT). Sterile pigskin collagen (Spongostan standard) was purchased from Health Designs Industries (Rochester,
NY), culture supplies (i.e., L-glutamine, sodium pyruvate, fetal bovine serum, gentamicin,
Dulbecco's Modified Eagle's Medium (DMEM), Minimum Essential Medium (MEM), MEM non-essential amino acids solution and MEM vitamin solution from GIBCO Laboratories
(Grand Island, NY), ^H-thymidine (specific activity, 65 Ci/mmole) from Moravek
Biochemicals Inc. (Brea, CA), NTB-2 nuclear track emulsion from Eastman Kodak Chemicals
(Rochester, NY), mouse normal IgG from Sigma (St. Louis, MO) and PSA detection kit from
Hybritech (San Diego, CA). All chemicals and reagents were used as received.
2.2.2 Tumor procurement
Surgical specimens of human primary prostate tumors were obtained from the peripheral zone of the prostate gland in patients who had organ-confined prostatic adenocarcinoma and who underwent radical prostatectomy. Specimens were obtained via the
Tumor Procurement Service at The Ohio State University Comprehensive Cancer Center, and from the neighboring Doctor’s Hospital. Tumor specimens were placed in MEM within 10 to
30 min after surgical excision, stored on ice and prepared for culturing within one hr from excision.
The histopathology of tumor specimens was established using frozen sections of tumor fragments grown as histocultures and/or adjacent paraffin-embedded sections. Tumors were graded according to the Gleason grading system with a summed score of 2 for well-
29 differentiated tumors and a summed score of 10 for poorly-differentiated tumors (Gleason,
1966).
2.2.3 Histoculture
The non-necrotic portions of tumor specimens were cultured as described previously
(Wientjes et al., 1995). In brief tumors were cut into < I mm^ pieces under sterile conditions.
Four to six tumor pieces were placed on a 1 cm^ presoaked collagen gel, and incubated at
37°C in a humidified atmosphere of 95% air and 5% CO,. The culture medium consisted of a 1:1 mixture of MEM and DMEM, 10% fetal bovine serum, 2 mM L-glutamine, I mM sodium pyruvate, 40 pg/ml gentamicin, 0.1 mM MEM non-essential amino acids, and concentrated MEM vitamin solution (100 fold concentrated, 10 ml per liter). The histocultures were fed every other day and used for paclitaxel pharmacodynamic studies on day 4 to 7. The secretion of PSA by human prostate tumor histocultures was determined by a "sandwich” immunoassay using two antibodies against two different epitope sites on PSA, performed by the James Cancer Hospital Immunology Laboratory on campus, The Ohio State
University, Columbus Ohio.
2.2.4 Pharmacologic effects of paclitaxel
Tumor histocultures were incubated with paclitaxel. Paclitaxel stock solution was prepared in ethanol. SuflBcient volume of stock solution was added to the culture medium so that the final ethanol concentration was < 0 .1%.
The antiproliferative effect of paclitaxel was measured by the inhibition of DNA precursor (i.e. thymidine) incorporation in tumor cells. Tumor cells were distinguished from non-epithelial cells by their morphology and organization. Tumor histocultures were exposed to various concentrations of paclitaxel ranging from 0.00012 to 12 for 24 and 96 hr.
After drug treatment, the medium was exchanged and the tumors were washed 3 times with
5 ml of drug-free medium. Tumors were incubated with 0.03 //M ^H-thymidine for 96 hr,
washed 3 times with PBS, then fixed in 10% neutralized formalin and embedded in paraffin.
The embedded tissues were cut into 5 fu.m sections using a microtome, deparaffinized and
processed for autoradiography. Controls were processed similarly, with the exception o f drug
treatment. Tissue sections were examined microscopically to score the ^H-thymidine-labeled
tumor cells for calculation of the LI. A typical experiment used a total of 10 to 20 tumor
pieces for each drug concentration. A minimum of 50 cells per tumor piece, or >500 cells
were counted per concentration.
Apoptosis was measured using light microscopy by monitoring morphological changes, i.e., chromatin condensation and margination, disappearance of nucleoli, formation o f membrane blebs, apoptotic bodies and/or cell shrinkage (Kerr et al., 1994). Our laboratory and others have shown that this method gives the same results as the terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) method (Gan et al.,
1996a; Gold et al., 1994).
Two treatment times were used, i.e. 24 hr because it is the clinically used infusion duration for paclitaxel (Dorr and Von Hoff 1994), and 96 hr because this treatment duration yields higher cytotoxicity in human ovarian carcinoma cells (Kelland and Abel, 1992). The 96 hr regimen was well tolerated in patients with doxorubicin/mitoxantrone-refractory breast cancer (Wilson et al., 1991). Although the 3 hr treatment is approved by the Food and Drug
Administration, it was not evaluated because minimal or undetectable cytotoxicity in human
31 prostate PC-3 and DU 145 tumor cells was observed, when the effect was measured immediately after a 3 hr treatment (unpublished results). Three of 34 specimens provided sufficient material to allow concurrent treatments for 24 and 96 hr. The remaining 31 specimens were treated for either treatment. In total, 26 tumors were treated with paclitaxel for 24 hr and 11 tumors for 96 hr.
2.2.5 Pharmacodynamic data analysis
The relationship of paclitaxel-induced infiibition of DNA synthesis and drug concentration was analyzed by computer-fitting the following equation to the experimental data.
E = (E^-Re) • ( 1- — -----) f Re (Eq 2.1) Ar"+C" where E is the LI of drug-treated tissues, C is the drug concentration, E„ is the LI of untreated controls, K is the drug concentration at one-half of (E„-Re), n is a curve shape parameter, and Re is the residual fraction. Values for IC 30 (the drug concentrations needed to produce 30% inhibition) were determined. IC 30 was used instead oflC^ because 50% inhibition was not achieved in -30% of tumors. Equation 2.1 is a modification of the more commonly used equation that describes a sigmoidal concentration-effect relationship that encompasses a spectrum of effect from 0% to 100%. Inclusion of the Re term is necessary to described the less-than-complete effect, i.e., the average maximum inhibition of DNA synthesis by paclitaxel was about 50% to 70% (see below).
2.2.6 Statistical analysis
Differences in mean values between groups were analyzed using Student's t test for
32 unpaired data, or the Mann-Whitney U-test. Frequencies were compared by the chi square test or Fisher’s exact test. Software for statistical analysis (TTEST, NPARIWAY, and
FREQUENCY procedures) was by SAS (Cary, NC).
33 2.3 Results
2.3.1 Histocultures
A total of 39 tumors from 39 patients were cultured. Thirty-four or 87% of tumors
were successfully cultured, defined as having >40 tumor cells per microscopic field and
having a thymidine LI of >5%. Figures 2.1 illustrate the growth of prostate tumor cells in
histoculture, and the clear distinction between prostate cancer cells with round nuclei which
grew in acinus-like structures, and non-epithelial cells with elongated nuclei which grew in
stromal tissue. After drug exposure, the fraction of thymidine-labeled cells decrease and the
fraction of apoptotic cells increase. Table 2.1 shows the patient and histopathological
characteristics of the 34 tumors. All tumors were from chemotherapy-naive patients. The
average [^H]-thymidine LI was about 40%. The histocultures secreted PSA. The
concentration of PSA in the histoculture medium of some untreated controls varied from 0.5
to 48.8 ng/ml, reflecting the inter-tumor (inter-patient) variation. The average PSA
concentration was 13±I5 ng/ml (mean±SD, n=12) on day 9 and 11±16 ng/ml (n= 6 ) on day
12, indicating no significant changes in the PSA secretion by histocultures over the duration
of the experiment.
2.3.2 Paclitaxel-induced inhibition of DNA synthesis
Paclitaxel produced a sigmoidal, concentration-dependent inhibition of the thymidine
LI (Figure 2.2). In all except one tumors, the inhibition was incomplete. The data in individual
tumors are summarized in Table 2.1. The 24 and 96 hr treatments showed a >300,000 and
a 14,000 fold range in ICjo, respectively, indicating significant inter-tumor (inter-patient)
variation.
34
i Table 2.2 compares the pharmacodynamies of the 24 and 96 hr treatments. The most significant diflference between the 24 and 96 hr treatments was the maximal inhibition in individual tumors. The average increased from 47% for the 24 hr treatment to 70% for the 96 hr treatment (p<0.001), and the incidence of tumors showing an E„u^ exceeding
50% increased from 70% (18/26) for the 24 hr treatment to 100% (11/11) of the tumors for the 96 hr treatment (p=0.04). The same trend of a higher antiproliferative effect for the longer treatment time was observed in the three tumors that were evaluated for both treatments (i.e., tumors 4, 15, and 19). These data indicate that the antiproliferative effect of paclitaxel can be enhanced by prolonging the treatment duration. For both 24 and 96 hr treatments, E^^^ was reached at between 120 and 1,200 nM, and further increases in concentrations to 12,000 nM did not significantly increase the E^.^ (p> 0 .2 ).
2.3.3 Paclitaxel-induced apoptosis
Figure 2.1 shows the paclitaxel-induced apoptotic cells identified by morphological changes. The same cells were labeled by the TUNEL method, confirming the morphologic identification. Paclitaxel induced apoptosis in all tumors (Table 2.1); the plateau maximal apoptotic index in individual tumors occurred at between 120 to 1,200 nM (Figure 2.3). In comparison, the untreated controls showed negligible apoptosis (<1%). Increasing the drug concentration to 12,000 nM did not significantly increase the apoptotic indices for either 24 or 96 hr treatments (p>0. 1). The effect of exposure time on apoptosis was different from that on the antiproliferative effect. Prolonging the exposure time from 24 to 96 hr did not affect the paclitaxel-induced apoptosis (p=0.48).
2.3.4 Tumor pathology and paclitaxel effects
35 Multivariate analysis showed that there was no significant correlation between paclitaxel pharmacological effects (i.e. IQ q , maximal apoptotic index) and tumor histopathology (i.e. Gleason grade, thymidine LI in untreated controls) or patient age (data not shown).
36 2.4 Discussion
The present study used tumor specimens derived from radical prostatectomies, a
treatment for the early disease, whereas paclitaxel chemotherapy has been used for advanced,
hormone refractory tumors. While the in vitro chemosensitivity in the early stage tumors may
not directly reflect that in advanced disease, these specimens served a primary purpose which
was to evaluate the time-dependency of paclitaxel effects. The histoculture system has several
advantages, including the use of patient-derived tumor tissues, and the maintenance of 3-
dimensional structure and cell-cell interaction. In addition, drug sensitivity at a range of drug
concentrations can be evaluated for a single tumor, and the variability in sensitivity between
patient tumors can be determined. The limitation of the histoculture system are the lack of
blood perfusion, and the usual difference between in vitro and in vivo growth conditions.
2.4.1 Antiproliferative and apoptotic effects
This study shows that paclitaxel inhibited DNA synthesis and induced apoptosis in all
34 human prostate tumors. In general, neither effect was complete. The incomplete inhibition
of DNA synthesis by paclitaxel is different from the effect of the complete inhibition observed
for doxorubicin, another agent used to treat hormone refractory prostate cancer (Chen et al.,
1995) and described in chapter 3. The finding of an incomplete inhibition o f DNA synthesis
in human prostate tumors is consistent with the finding in human head and neck tumors (Gan
et al., 1996) and bladder tumors (Au et al., 1996), and with literature data showing that
paclitaxel-treated cells can proceed with DNA synthesis (Lopes et al., 1993; Liebmann et al.,
1994). These DNA synthesizing cells did not complete cytokinesis, yielding hyperdiploid cells with up to 16 N of DNA. Further studies to define the DNA ploidy and the fate of the
37 paclitaxel-treated cells in human prostate tumors are needed.
In general, the inhibition of DNA synthesis showed a greater variability than apoptosis
(Table 2.1). For example, for the 24 hr treatment, the of DNA inhibition ranged from
14% to 100%, with IC 30 ranging from 0.04 nM to >12,000 nM, or a >300,000-fold range.
The maximum apoptotic index showed a 4-fold range of 5% to 20%, which occurred within a 10-fold concentration range of 120 to 1,200 nM. The high variations among individual tumors is presumably due to patient-related tumor heterogeneity.
2.4.2 Pharmacodynamics - Effect of exposure time
Increasing the exposure time from 24 to 96 hr substantially increased the antiproliferative effect, but only marginally enhanced the apoptotic effect. It is speculated that the different pharmacodynamics of the two effects may have contributed to the controversy regarding the optimal treatment schedule. For example, studies that measure mainly the antiproliferative action will likely find a time-dependent effect, whereas studies that measure mainly the apoptotic effect will not find such a relationship.
The pharmacodynamics of paclitaxel in human tumors at clinically achievable concentrations at 24 and 96 hr were compared to make a qualitative comparison of the relative effectiveness of the two treatments in prostate cancer patients. The steady state paclitaxel concentration at the maximally tolerated total dose in patients by the 96 hr schedule is 70 nM for 160 mg/m’ (Wilson et al., 1994) which is more than 10-fold lower than the 900 nM for the 250 mg/m’ by the 24 hr schedule (Wiemik et al., 1987). Under these conditions, the inhibition of tumor cell LI was approximately equal at about 57% (Figure 2.2, p=0.47).
On the other hand, the apoptotic index at 900 nM and 24 hr exposure is 11%, which is
38 slightly but significantly higher than the index of 7% at 70 nM and 96 hr exposure (Figure 2.3, p<0.03). These data imply that after adjusting for the difference in the doses and the infusion rates, the increase in the treatment duration from 24 to 96 hr will not enhance the antiproliferative effect, but will decrease the apoptotic effect. It should be emphasized that the above comparisons are done using an assumption that the in vitro concentrations are equal to in vivo concentrations. There are several known differences for paclitaxel pharmacokinetics under in vitro and in vivo conditions. For example, it has been reported that (a) there are differences in the extent of protein binding in culture medium and in plasma (Song et al.,
1996), (b) depletion of paclitaxel from culture medium is dependent on cell density and (c ) uptake of paclitaxel is saturable (Kang et al., 1996). A pharmacokinetic model is under development in Dr. Au’s laboratory to incorporate these differences among in vitro and in vivo conditions, in order to provide more accurate quantitative analysis and for extrapolation of in vitro pharmacodynamic data to in vivo situations.
2.4.3 Pharmacodynamics - Effect of drug concentration
In all tumors, and maximal apoptotic index were reached at between 120 and
1,200 nM paclitaxel concentrations, indicating that increasing the dose to elevate plasma concentrations beyond these levels may not enhance the therapeutic effect.
39 2.5 Conclusions
The data of present study indicate (a) inhibition of DNA synthesis and induction of apoptosis by paclitaxel in human prostate tumors, with substantial variation among individual patient tumors, and (b) different pharmacodynamics of the antiproliferative and apoptotic effects of taxol, with the antiproliferation being affected more significantly by treatment time compared to the apoptotic effect. These data also provide the first pharmacological evidence that prolonging paclitaxel treatment beyond 24 hr or increasing the plasma concentration above 1 pM may not offer additional benefits in the treatment o f prostate cancer.
40 2.6 Acknowledgments
This study was supported in part by research grant R0ICA63363 from the National
Cancer Institute, NIH, DHHS, an award from the CaP CURE Foundation, and a gift from
Bristol-Myers Squibb Company. The Ohio State University Comprehensive Cancer Center
Tissue Procurement Service was supported in part by P30CA16058 from the National Cancer
Institute, NIH, DHHS. We thank Drs. Robert Badalament, John Burgers, and Jack Perez for providing the tumor specimens.
41 Tumor .A.ge Race AUA Gleason Control ^nutxF IC,o Maximal stage Grade thvmidine LI (%) (nM) .Apoptotic ' (%) Index, (%) 24 hr treatment 1 63 Caucasian C2 8 34 100 0.04 6.7 2 63 Caucasian B2 7 34 71 2.0 6.9 3 59 Caucasian B2 7 45 65 14.0 14.0 4 70 Black B2 7 48 62 6.2 9.9 5 55 Caucasian B2 7 65 61 2.8 9.3 6 71 Caucasian C2 7 42 60 1.1 17.4 7 73 Caucasian B2 8 47 59 0.1 14.4 8 72 Caucasian Cl 5 29 57 0.4 7.3 9 66 Caucasian B2 6 43 56 10.0 15.4 10 66 Caucasian B1 5 31 55 264 18.1 11 66 Caucasian Cl 6 54 53 26.1 15.8 12 70 Caucasian B2 5 25 53 1.2 7.9 13 70 Black B2 7 58 52 9.2 11.5 14 49 Caucasian B2 5 60 51 6.4 8.3 15 58 Black B2 6 36 49 17.0 12.0 16 64 Caucasian D1 7 27 45 115 12.2 17 59 Caucasian C2 7 52 44 170 19.9 18 67 Caucasian B2 7 54 43 14.4 14.4 19 55 Caucasian B2 7 53 30 >12,000 12.1 20 65 Caucasian B2 6 24 30 >12,000 8.6 21 65 Caucasian B2 7 28 30 >12,000 17.0 in 59 Caucasian C2 8 45 27 >12.000 117 23 63 Caucasian B2 7 34 22 >12,000 19.8 24 60 Caucasian Cl 8 34 20 >12,000 10.1 25 58 Caucasian D1 9 59 19 >12,000 11.3 26 73 Caucasian B2 6 25 14 >12.000 7.2 96 hr treatment 27 55 Caucasian B2 5 40 87 0.8 10.7 28 59 Caucasian Cl 6 43 85 17.4 18.2 29 62 Caucasian B2 5 38 80 7.9 14.6 30 67 Caucasian B2 5 48 79 1.1 13.6 4 70 Black B2 7 50 73 7.6 5.5 31 71 Caucasian B2 6 24 70 0.02 9.9 19 55 Caucasian B2 7 39 65 11,000 13.7 32 71 Caucasian B2 6 54 63 7,7 11.9 33 55 Caucasian B2 5 34 59 1.1 10.6 15 58 Black B2 6 40 58 8.2 10.5 34 63 Caucasian Cl 6 45 53 103 15.4 Table 2.1. Patient and tumor characteristics, and tumor sensitivity to 24 and 96 hr paclitaxel treatments. All tumors were obtained from radical prostatectomies. 5^^% is the maximal reduction of labeling index (LI), as percent of control. IC 30 is the concentration of paclitaxel needed to produce a 30% inhibition of LI. American Urological Association (AUA) stage is same as the Whitmore and Jewett staging system.
42 Control thymidine LI, % IC,o, nM Maximal apoptotic median (range) index, %•
24 hr 42±12 15.7 (0.04-> 12,000) 47±19 12±4 (n=26)
96 hr 41±8 7.7(0.02-11,000) 70±I2 I2±3 (n=ll)
P 0.92 0.098 0.001 0.98
Table 2.2. Relationships between tumor sensitivity to 24 hr and 96 hr paclitaxel treatments.
Data are represented as mean±SD, except for ICjq. Median and range are used because 8 tumors had IC 30 values exceeding 12,000 nM. Differences between the 24 and 96 hr treatments were analyzed by Student’s unpaired t-test, except for IC^g, which was analyzed by the non-parametric Mann-Whitney U-test.
' Maximum apoptotic index occurred at 120 to 1,200 nM concentrations in individual tumors.
43 Figure 2.1. Prostate tumor growth in histoculture. A human prostate specimen (Tumor # 4 in Table 2.1), grown in histoculture for 9 days and labeled with ^H-thymidine is shown at 16x (A) and 40x (B) magnification. The thymidine labeling shows as black dots (arrows). The typical prostate tumor architecture with acinus-like structures surrounded by stromal material was maintained. The apoptotic cells, resulted from treatment with 1,200 nM paclitaxel for 96 hr, are indicated by chromatin condensation and margination and disappearance of nucleoli (arrow, Panel C). The same cells were labeled brown by the TUNEL method. 100 2 -4—» 80 - C o
O 60 - y — O
40 -
- J 2 0 -
0 0.001 0.1 10 Paclitaxel, |iM
Figure 2.2. Inhibition of thymidine LI as a function of extracellular paclitaxel concentration. Human prostate tumor histocultures were treated with paclitaxel for 24 or 96 hr. Inhibition of cumulative ^H-thymidine LI was expressed as percent of untreated controls. Data represent mean ± SEM of 26 tumors treated for 24 hr, and 1 1 tumors treated for 96 hr. Sigmoidal lines are computer-fitted according to Equation
2 . 1.
45 12 n 24 hr 96 hr
ü
O 4 —* Q. O <Û.
0 0.001 0.1 10 Paclitaxel,
Figure 2.3. Apoptotic index as a function of extracellular paclitaxel concentration. Human prostate tumor histocultures were treated with paclitaxel. The drug-induced apoptosis was measured by morphological changes. Data represent mean ± SEM of 26 tumors treated for 24 hr, and 11 tumors treated for 96 hr. Note that the average apoptotic indices at various paclitaxel concentrations differed slightly from the average maximal apoptotic indices outlined in Table 2.1 because maximal indices in individual tumors occurred at different concentrations.
46 CHAPTER 3
PACLITAXEL-INDUCED APOPTOSIS: A CELL CYCLE SPECIFIC EVENT
3.1 Introduction
Paclitaxel is the first agent from the novel class of anticancer drugs, taxoids. It is first isolated in 1967 (Wall and Wani, 1967) from the bark of the yew and structurally characterized in 1971 (Wani et ai., 1971). Paclitaxel was later found to be a product of an endophytic fungus Taxomyces cmdreanae in the yew bark (Stierle et al., 1993). Significant activities of paclitaxel have been reported in patients with advanced ovarian (McGuire et al.,
1989), breast (Gianni et al., 1994), non-small cell lung (Chang et al., 1993; Murphy et al.,
1993) and head and neck cancers (Forastiere and Urba, 1995). In addition, preclinical studies have shown the activity of paclitaxel in several other neoplastic diseases.
The G2/M blockade in the cell cycle induced by paclitaxel is believed to be the putative mechanism of its anticancer effects (SchifF et al., 1979). Paclitaxel also promotes abnormal microtubule assembly, stabilizes microtubule, elevates intracellular calcium concentration, inducing phosphorylation of proteins and activates the lipopolysaccharide-like signal transducing pathway (Horwitz, 1992; Johnson and Byerly, 1993; Ding et al., 1993;
Carboni et al., 1993; Ding et al., 1990; Manthey et al., 1993). One or more of these effects.
47 as well as other yet unknown mechanisms, can result in its antiproliferative and cytotoxic (i.e., cell kill) effects. It is not known whether antiproliferation or cytotoxicity contributes to the clinical activity of paclitaxel in patients. There are data to suggest that the paclitaxel-treated tumor cells can escape from the G2/M block (Kuriyama et al., 1986; Lopes et al., 1993), that the cytostasis resulting from mitotic block by paclitaxel may not be a sufficient signal for paclitaxel-induced apoptosis (Donaldson et al., 1994), and that apoptosis, not mitotic block, is correlated with its antitumor effects (Milross et al., 1996).
Paclitaxel induces apoptosis in a number of human cancer cells and tumors, including leukemia cells (Bhalla et al., 1993), retinoblastoma cells (Inomata et al., 1995), ovarian cancer cells and xenograft tumors (Liu et al., 1994; Havrilesky et al., 1995; Frankel et al., 1997), gastric cancer cells (Chang et al., 1996), breast cancer cells (Sumantran et al., 1995; Saunders et al., 1997), head and neck cancer cells and tumors (Pulkkinen et al., 1996; Gan et al.,
1996a), bladder tumors (Au et al., 1997), and prostate cancer cells and tumors (Danesi et al.,
1995; Chen et al., 1997a). Although paclitaxel exerts cytotoxicity in many different cancer cells and tumors, the sensitivity o f paclitaxel cytotoxicity may be cell type dependent (Roberts et al., 1990; Gangemi et al., 1995).
The present study evaluated the relationship between paclitaxel-induced apoptosis and cell cycle progression. The effect of co-treatment with hydroxyurea which blocks cells in the
S phase, on paclitaxel-induced apoptosis was investigated. The relationship between paclitaxel treatment periods and drug effects was also investigated.
48 3.2 Materials and methods
3.2.1 Chemicals and supplies
Paclitaxel was a gift from Bristol Myers Squibb Co. (Wallingford, CT). Human PC3
and DU 145 prostatic cancer cells were from American Type Culture Collection (Rockville,
MD). Male athymic Nu/Nu mice, BALB/C retired breeders, were purchased from National
Cancer Institute (Frederick, MD); testosterone (12.5 mg/tablet) from Innovative Research
of America (Toledo, OH); Matrigel* from Becton Dickinson Labware (Bedford, MA);
sulforhodamine B (SRB), Tris-Base, hydroxyurea (HU), bromodeoxyuridine (BrdUrd) and proteinase K from Sigma Chemical Co (St. Louis, MO); trichloroacetic acid (TCA) and glacial acetic acid from the chemical store at The Ohio State University (Columbus, OH); automated microtiter plate reader, model EL340, from BIO-TEK instruments Inc (Winooski,
VT); sterile pigskin collagen (Spongostan standard) from Health Designs Industries
(Rochester, NY); culture supplies (i.e., L-glutamine, sodium pyruvate, fetal bovine serum, gentamicin, Dulbecco's Modified Eagle's Medium (DMEM), Minimum Essential Medium
(MEM), MEM non-essential amino acids solution and MEM vitamin solution were from
GIBCO Laboratories (Grand Island, NY); terminal deoxynucleotidyl transferase (TdT), digoxigenin-dUTP and anti-digoxigenin-peroxidase in ApopTag*" from Oncor Inc.
(Gaithersburg, MD); monoclonal mouse IIB5 anti-BrdUrd antibody and liquid 3,3’- diaminobenzidine substrate kit from BioGenex (San Ramon, CA); and LSAB universal detection kit from Dako Inc. (Carpiteria, CA). All chemicals and reagents were used as received.
3.2.2 Monolayer cell cultures
49 PC3 and DU145 cells were used to determine the effect of treatment durations on the
antitumor effect of paclitaxel. Drug-induced reduction in cell number was measured by the
SRB assay, as previously described (Skehan et al., 1990). Cells were seeded in 96-well plate
at 3000 cells per 200 pi per well, and were allowed to attach to culture flasks for 24 hr. The
treatment protocol is shown in Figure 3.1 (A). Afterwards, cells were treated with paclitaxel
for 24 hr and 96 hr followed by fixation with 50 pi of cold (4°C) 50% TCA gently layered on
top of the growth medium in each well (the final concentration of TCA was 10%). This set
of cells measured the immediate effects of the 24 and 96 hr treatments. Another set of cells were treated with paclitaxel for 24 hr, followed by replacing drug-containing medium with fresh medium and incubation for an additional 72 hr. This second set of cells measured the delayed effect of the 24 hr treatment. The cultures were then incubated in 4°C for 1 hr and then washed five times with tap water to remove TCA and growth medium. The fixed cultures were stained with 50 pi 0.4% SRB in 1% glacial acetic acid for 30 min at room temperature. After staining, SRB was removed and the culture plates were rinsed five times with 1% glacial acetic acid to remove unbound SRB. The residual solution was removed by shaking the culture plates and the plates were air-dried. The bound SRB was solubilized with
200 pi of 10 mM Tris buffer (pH 10) for 15 minutes at room temperature. Optical density
(CD) was measured with a microtiter plate reader with a wavelength at 490 nm. The CD values of the drug-treated cells were expressed as % of the control.
3.2.3 Procurement of tumor specimens and histocultures
Surgical specimens of human primary prostate tumors were obtained from the peripheral zone of the prostate gland in patients who underwent radical prostatectomy.
50 Human prostate tumor specimens were obtained via the Tumor Procurement Service at The
Ohio State University Comprehensive Cancer Center. The androgen-dependent CWR22 xenograft was originally derived from a primary prostate carcinoma of a patient with a
Gleason grade 9 and stage D tumor and osseous metastasis (Pretlow et al., 1993). CWR22 was provided by Dr. Thomas G. Pretlow at the Case Western Reserve University Medical
Center (Cleveland, OH). CWR22 xenograft tumors were maintained in male nude mice according to the previously published procedures (Pretlow et al., 1993; Nagabhushan et al.,
1996). Briefly, subcutaneous implantation of a testosterone pallet was performed 3 days before the implantation of CWR22 cells. CWR22 cells in suspension ( 1 0* cells) or minced xenograft tissues were mixed with equal volume o f Matrigel. The mixture was subcutaneously implanted into the two flanks of a mouse, at 0.3 ml per site. When a tumor reached the size of about 1 gram in about 1.5 to 2.5 month, it was harvested.
Human and xenograft tumor specimens were placed in MEM within 10 to 30 min after surgical excision and prepared for histoculturing within one hr after excision. Tumor specimens were cultured as described previously (Chapter 2; Chen et al., 1997a). In brief, tumors were cut into < 1 mm^ pieces under a sterile condition. Five to six tumor pieces were placed on a I cm^ presoaked collagen gel, and incubated at 37°C in a humidified atmosphere of 95% air and 5% CO,. The culture medium consisted of a 1:1 mixture of MEM and
DMEM, 10% fetal bovine serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 40 pg/ml gentamicin, 0.1 mM MEM non-essential amino acids, and concentrated MEM vitamin solution ( 100-fold concentrated, 10 ml per liter).
3.2.4 Drug treatment
51 Tumors were treated with paclitaxel at different concentrations, i.e., 0.012, 0.12, 1.2,
and 12 pM, for different exposure times, 1, 3, 8 , 16, 24, 48, 72, 96 and 120 hr in patient
tumor histoculture.
3.2.5 BrdUrd labeling index and immunohistochemistry of BrdUrd
Procedures to detect BrdUrd labeling were performed as described in the LSAB
universal detection kit with minor modifications, as previously reported (Gan et al., 1996b).
After being deparaflBnized and rehydrated sequentially in xylene, ethanol, and water, tissue
sections on slides were boiled in a 10 mM citrate buffer (pH 6.0) in a microwave oven and
then cooled and washed in PBS for 15 min. After wiping off the excess PBS, the tissue sections were incubated with Dako blocking solution for 10 min followed by an incubation of 45 min with the BrdUrd antibody (1:250 dilution in 0.5 % BSA in PBS). Negative controls used mouse IgG instead. After two washes with PBS, the tissue sections were covered with the linker solution and then with peroxidase-conjugated streptavidin solution for 20 min each.
After two washes with PBS, the tissue sections were incubated with 3 -3 '-aminobenzidine for
5-7 min and then counterstained with hematoxylin followed by dehydration. All incubation processes were carried out in a humidified chamber so that no or only minimal liquid evaporation occurred. The tissue section was scored microscopically for BrdUrd-labeled tumor cells and the labeling index (LI), defined as the number of BrdUrd-labeled tumor cells per total tumor cells, was determined.
3.2.6 Apoptotic index and TUNEL assay
Apoptotic cells were identified by their morphologies, i.e, chromosome condensation and apoptotic bodies and by the TUNEL assay (Kerr et al., 1994; Chapter 2). The apoptotic
52 index, defined as the number o f apoptotic tumor cells per total tumor cells, was determined.
TUNEL assay were carried out as previously described (Chapter 2; Chen et al., 1997a). The tissue sections on slides were deparafiBnized, rehydrated and incubated with proteinase K (20
pg/ml) in PBS for 15 minutes at room temperature followed by 4 washes with distilled water for 2 minutes each. Protease treatment is necessary to make DNA fi-agments accessible for the reaction with enzyme and substrate. Endogenous peroxidase activity in tissues was quenched by incubating tissue sections with 2% hydrogen peroxide in PBS for 5 minutes at room temperature. Tissue slides were then washed twice with PBS for 5 minutes each.
Equilibration buffer from the kit was applied directly onto tissue sections and incubated for
10 minutes at room temperature. TdT-containing liquid from the kit was then applied and incubated at 37°C for 1 hour. After the enzyme reaction, slides with tissues were incubated with 37°C-prewarmed Stop solution from the kit for 10 minutes at room temperature followed by 3 washes with PBS for 5 minutes each. Anti-digoxigenin-peroxidase from the kit was applied and incubated for 30 minutes at room temperature followed by 3 washes with
PBS for 5 minutes each. The volume of applied liquid was enough to cover the whole section.
Between processes, excess liquid was tapped off gently and the tissue was blotted with dry towels. The slides were kept in a humidified chamber so that no concentration changes occurred due to the loss of water during the incubation periods. Brown deposits were developed by incubating the tissue sections with 3-3'-diaminobenzidine and the substrate, hydrogen peroxide, for 3 to 6 minutes. The tissue sections were counterstained with hematoxylin, dehydrated and coverslipped for microscopic examination.
3.2.7 Effects of hydroxyurea on cell cycle progression and paclitaxel-induced apoptosis
53 Hydroxyurea was used as a tool to cause cell cycle arrest at early S phase (Hill et al..
1977). The effect of hydroxyurea on paclitaxel-induced apoptosis was examined. Patient and
the xenograft tumors were pretreated with hydroxyurea (0, 2, and 10 mM) and paclitaxel (1.2
pM) for different periods of time according to procedures shown in Figures 3 .1 (B) and 3 .1
(D). Simultaneous treatments of paclitaxel and hydroxyurea for 48 and 96 hr in the presence
of BrdUrd were also performed as shown in Figure 3.1 (C).
3.2.8 Data analysis
Relationships between the cytotoxicity, measured by the SRB assay and paclitaxel
concentrations were analyzed by computer-fitting the experimental data with Equation 3.1.
E = (100%-/?e) • (1- —^ —-) + Re (Eq. 3.1) K." where E is the fraction of survival cells compared with control, C is the drug concentration,
K is the drug concentration at 50% effect, n is a curve shape parameter, and Re is the residual survival fraction. Values for IC% (the drug concentrations needed to produce 50% inhibition) were determined. Equation 3.1 is a modification of the more commonly used equation that describes a sigmoidal concentration-effect relationship that encompasses a spectrum of effect from 0% to 100%. Inclusion of the Re term is necessary to described the less-than-complete effect, i.e., the average maximum cytotoxicity of 24-hr paclitaxel treatment was about 70% to 85%.
3.2.9 Statistical analysis
Differences among groups were detected by analysis of variance and by Student's t test. Software for statistical analysis (ANOVA, TTEST procedures) was by SAS (Cary, NC).
54 3.3 Results
3.3.1 Delayed cytotoxicity in PC3 and DUI45 cells
Paclitaxel showed a concentration-dependent cytotoxicity in PC3 and DU 145 cells
(Figure 3.2). As shown in Table 3.1, the 96-hr treatment produced a maximum effect
of ~ 100%, while the 24-hr treatment showed a 6 ^^% of less than 50% in both cell lines.
Interestingly, for the 24 hr treatment, the addition of a 72-hr waiting period before drug effect
measurement significantly enhanced the apparent drug effect with the E^^ increased to about
100% and the IC 50 decreased by >400 fold. The concentration-eflfect profiles of the immediate
effect (i.e. no delay in effect measurement) o f the 96 hr treatment and the delayed effect (i.e.
72 hr delay in effea measurement) of the 24 hr treatment were nearly superimposable in PC3 and DU 145 cells (Figure 3.2), and there are no differences in the IC 50 and E^ix of the two treatments (Table 3.1). The SRB assay, because it indirectly measures the cell number, can not distinguish between cytotoxic and antiproliferative effects. A suppression in the relative optical absorbance of SRB may be simply due to the effect of cell growth arrest. However, because no viable cells were found in culture flask nor culture medium at 96 hr, we conclude that the reduction in the SRB reading was due to cell kill by paclitaxel.
3.3.2 Histoculture: Inhibition of DNA synthesis by paclitaxel and hydroxyurea
Untreated controls of patient tumor histocultures showed BrdUrd LI of 34%.
Treatment by hydroxyurea (2 and 10 mM) or paclitaxel ( 1.2 pM) for 48 to 96 hr reduced the
LI in patient tumors to <5% and 15%, respectively (Figure 3.3 (A)). The same treatments reduced the LI in xenograft tumor from 40% to < 16% and 21%, respectively (Figure 3.4
(A)). The incomplete inhibition of BrdUrd LI in histocultures by paclitaxel is qualitatively
55 different from the response in monolayer cultures of cancer cells that is shown above. This
difference has been observed in other patient tumors and other xenograft tumors (unpublished
results), and may be related to the differences of 2-dimensional versus 3-dimensional
structures. Muiitlayered structure-related chemoresistance is well recognized (Durand, 1990;
Kerr et al., 1987; Pizao et al., 1994).
Addition of hydroxyurea to paclitaxel affected the paclitaxel-induced inhibition of
BrdUrd LI in patient and xenograft tumors, in a concentration- and sequence-dependent
manner. In patient tumors, treatment by hydroxyurea before paclitaxel reduced the LI to the
level that was similar to the effect of hydroxyurea alone, in both tumors. The effect of
sequencing of the two drugs was further studied in the xenograft tumor, because it was
readily available whereas the patient tumors were not. Figure 3.5 (A) shows the results.
Paclitaxel treatment prior to hydroxyurea treatment showed a lower reduction of LI, compared to the reverse sequence. Interestingly, simultaneous treatment by the two drugs produced the lowest inhibition that was even lower than the inhibition induced by paclitaxel alone. In summary, the rank order of BrdUrd LI inhibition was hydroxyurea = hydroxyurea- before-paclitaxel > hydroxyurea-after-paclitzixel > paclitaxel alone > simultaneous hydroxyurea and paclitaxel.
3.3.3 Histoculture: Induction of apoptosis by paclitaxel and hydroxyurea
Untreated controls of patient and xenograft tumor histocultures contained <0.2% apoptotic tumor cells. In patient tumors, both paclitaxel and hydroxyurea significantly increased the apoptotic index, whereas only paclitaxel induced apoptosis in the xenograft tumor (Figures 3.3 (B) and 3.4 (B)). In both tumors, hydroxyurea pretreatment reduced
56 paclitaxel-induced apoptosis, with a greater reduction at higher hydroxyurea concentration.
The effect of sequencing the two drugs was further studied in the xenograft tumor.
The results, shown in Figure 3.5 (B), indicates that the hydroxyurea-afler-paclitaxel combination also reduced the paclitaxel-induced apoptosis, although there is no apparent relationship with the hydroxyurea concentration, a finding that is different for the hydroxyurea-before-paclitaxel combination. Interestingly, the simultaneous treatment with hydroxyurea and paclitaxel, although it also reduced the paclitaxel-induced apoptosis, the reduction was less pronounced than the other two combinations. In summary, the rank order of apoptosis was paclitaxel > simultaneous treatment by the two drugs > hydroxyurea before or after paclitaxel combinations.
Paclitaxel did not appear to alter the apoptosis induced by hydroxyurea in patient tumors, as there are no statistical differences in the apoptotic indices after treatment by hydroxyurea alone or after hydroxyurea-plus-paclitaxel combinations (Figure 3.3 (B)).
3.3.4 Histoculture: Correlation between cell proliferation and paclitaxel-induced apoptosis
It has been shown that in monolayer cell cultures, entry into cell cycle is required for apoptosis induction by paclitaxel (Liebmann et al., 1993 and 1994). Our laboratory has shown a correlation between the LI of tumor histoculture with the extent of paclitaxel-induced apoptosis in patient solid tumors (Gan et al., 1996a), thus confirming the link between cell c>cle and apoptotic effect of paclitaxel in 3-dimensional solid tumors. The present study evaluated if hydroxyurea alters this relationship. The results showed that over 80% of the paclitaxel-induced apoptotic cells were labeled by BrdUrd in both patient and xenograft tumors. Addition of
57 hydroxyurea did not alter this relationship (Figure 3.6). These findings confirmed our earlier finding o f a significant positive correlation between the two parameters and that entry into cell cycle appears to be correlated with paclitaxel-induced apoptosis.
Hydroxyurea significantly reduced the fi-action o f paclitaxel-induced apoptotic cells that were labeled by BrdUrd in patient and CWR22 xenograft tumors (Figure 3.7).
3.3.5 Histoculture: Kinetics of apoptotic induction by paclitaxel
Figure 3.8 shows the kinetics of apoptosis induced by paclitaxel. The maximum apoptotic index increased with drug concentration up to 0.12 pM. No further increases were observed at higher drug concentration at 1.2 and 12 pM. In general, the maximum apoptotic index was reached after 24 hr treatment. The less-than-complete apoptotic fraction indicates that only a certain fraction of tumor cells are susceptible to paclitaxel-induced apoptosis.
58 3.4 Discussion
The study using cell lines demonstrated that paclitaxel has two effects, i.e. immediate and delayed. The study on the kinetics of paclitaxel apoptosis in patient tumors showed that apoptosis reached its maximum level at 24 hr, which persisted for 96 hr. Taken together, these data indicate that the delayed eflfect of paclitaxel may be a reflection of the delayed manifestation of apoptosis. This hypothesis was later confirmed in a separate study in our laboratory. This later study further showed that the delayed drug effect is also partly due to the extensive intracellular accumulation of paclitaxel (10% of the dose at 10 nM initial drug concentration) and the subsequent slow release of paclitaxel from its intracellular binding site
(Kuh et al., 1996).
Entry in cell cycle is considered a prerequisite of apoptosis induction by paclitaxel. For example, blocking cells in G1 phase by Cremophor EL and in S phase by L-buthionine sulfoxime antagonizes the cytotoxic effect of taxol (Liebmann et al., 1994). Stationary phase
(i.e. confluent) cells (lung A549 and breast MCF-7) are less sensitive than cells in the exponentially growing phase to paclitaxel (Liebmann et al., 1993). It has been shown that cells in M phase are sensitive to paclitaxel-induced apoptosis. The requirement of an exposure period to paclitaxel may be related to the necessity that the cells cross G2/M to undertake a death decision (Gangemi et al., 1996). These previous studies were performed using monolayer cultures of cancer cells. The link between cell cycle and apoptosis in solid tumors was confirmed by our present and previous (Gan et al., 1996a) findings of the positive correlation between LI and paclitaxel-induced apoptosis.
The present study also evaluated if agents that block cells in the cell cycle affect
59 apoptosis induction by paclitaxel. Hydroxyurea, which has been used to cause cell cycle arrest at early S phase, inhibited the LI of patient and xenograft tumor histocultures. Hydrox>airea also affected the inhibition by paclitaxel, in a concentration- and sequence-dependent manner.
Addition of hydroxyurea before and after paclitaxel enhanced the inhibition. Interestingly, addition of hydroxyurea simultaneous with paclitaxel reduced the inhibition by single agents.
Likewise, the apoptosis induction by paclitaxel was reduced by hydroxyurea treatment given prior to or after paclitaxel treatments. Simultaneous addition of hydroxyurea reduced the apoptosis by paclitaxel, although to a lesser extent compared to the other two sequences. The finding that hydroxyurea added before paclitaxel inhibited paclitaxel-induced apoptosis indicates that the apoptotic signal is exerted after S phase. The finding that hydroxyurea added after paclitaxel also inhibited paclitaxel-induced apoptosis indicates that successful completion of the S phase is required for paclitaxel-induced apoptosis. Additional cell cycle distribution studies to confirm these hypotheses are warranted.
60 3.5 Conclusions
In conclusion^ the results in this Chapter demonstrate that paclitaxel has immediate and delayed effects. These data have led to an additional study which demonstrated that the delayed effect is in part due to the delayed manifestation of apoptosis and in part due to the release of drug from its binding site. The finding o f a positive correlation between BrdUrd LI and maximum induction of apoptosis by paclitaxel confirms that apoptosis induction by paclitaxel is linked to cycling cells. We further found that agents that block cells in the cell cycle, i.e. hydroxyurea which has been shown to block cells in early S phase, can inhibit paclitaxel-induced apoptosis.
61 3.6 Acknowledgments
This study was supported in part by research grant R01CA63363 from the National
Cancer Institute, NIH, DHHS, an award from the CaP CURE Foundation, and a gift from
Bristol-Myers Squibb Company. The Ohio State University Comprehensive Cancer Center
Tissue Procurement Service was supported in part by P30CA16058 from the National Cancer
Institute, NIH, DHHS. We thank Dr. Robert Badalament for providing the tumor specimens and Dr. Thomas G. Pretlow for providing the CWR22 tumors.
62 Exposure Time (hr) Cells Effects Regimens 24 96
ICwXnM) Immediate NM”'' 5.84=3.5' Delayed 4.2±1.7 NA' PC3 (%) Immediate 31.2±13.9' 98.1±1.9' Delayed 98.5±2.1 NA
IC50 (nM) Immediate NM”-' 9.5±6.9‘* Delayed 29.4±27.1 NA DU145 Em .(% ) Immediate I7.8±6.r 92.34=1.0' D elved 89.7±2.6 NA
Table 3.1. Immediate and delayed effects of paclitaxel in monolayer PC3 and DÜI45 cell cultures.
Cells were treated with paclitaxel for 24 and 96 hr. Drug effect was measured immediately after drug treatment and, in the case of the 24 hr treatment, was also measured with a 72 hr delay (i.e. effect measured at 96 hr after treatment was initiated). Hence, we measured the immediate effect of the 96 hr treatment and the immediate and delayed effects of the 24 hr treatment.
is the maximal cytotoxic effect. IQ is the concentration of paclitaxel needed to produce a 50% of cytotoxicity. Values are represented as mean ± sd, n^4. NM. can not be measured because the effect was less than 50%. p<0.001 compared to 72-hr delayed and 96 hr regimens. No significant difference between the 72-hr delayed and 96 hr regimens. NA, ; not applicable.
63 A. add O paclitaxel fix
24 hr e fix
replaced w/ ® fresh medium fix
24 hr 96 hr B- HU w/ & w/o continuous HU paclitaxel w/ BrdUrd fix
24 hr 72 In c. simultaneous paclitaxel & HU w/ BrdUrd fix fix
48 hr 96 hr D. paclitaxel paclitaxel paclitaxel w/ BrdUrd HU HU fix
24 hr 48 hr 96 hr
Figure 3.1. Schematic illustration of protocols for paclitaxel treatments In monolayer cell cultures and histocultures. A; Samples in O and @ were fixed and analyzed by the SRB assay immediately after 24- and 96-hr paclitaxel treatments, respectively. Samples in ® were treated with paclitaxel for 24 hr and fixed 72 hr later. B; The histocultures were pretreated with hydroxyurea (HU) at 0, 2 and 10 mM for 24 hr and then continuously exposed after adding 1.2 pM paclitaxel for 48 hr in the presence of BrdUrd. The control group was without paclitaxel treatment. C: Samples were treated with paclitaxel and HU simultaneously. Two groups of histocultures were fixed at 48 hr and 96 hr, respectively. D; Samples were pre treated with paclitaxel for 24 hr, followed by HU treatment for another 24 hr, and then followed by a continuous 48 hr exposure to both drugs and to BrdUrd.
64 PC3 DU 145
120 - 120 -
— 100 - y ) 24, Immediate O 24, Immediate C 80 - o o 60 - O 60 40 24, delayed 24, delayed 20 - 96, Immediate LAOn 96, Immediate 0 - // T I I I r 0 0.001 0.01 0.1 1 0.001 0.01 0.1 1 Paclitaxel, |j,M Paclitaxel, laM
Figure 3.2. Activity of paclitaxel on human prostate cancer cells by SRB assay. A 40 -1
without paclitaxel ] with paclitaxel
20 - “E Z) 10 - “E # T ÛÛ # # B 8
X (D 6 T3 C o 4 a4—» CL 2 O Q. < 0 %%
Figure 3.3. Effects of hydroxyurea pretreatment on paclitaxel activity in patient prostate tumors. Inhibition of DNA synthesis (A) and induction of apoptosis (B) by paclitaxel 1.2 pM in the absence or presence of hydroxyurea (HU) 24-hr pretreatments at 0, 2 and 10 mM. Data represent mean ± sem, n=4. p<0.05 for *: vs. without paclitaxel; #: vs. HU 0 mM.
6 6 A o 40 - n CZZ] without paclitaxel with paclitaxel
"2 20 II
3 # "S 10 - T Gû pj R IS % %% % B %'
12 - X m "O t a 8 - _ c (sfi o
2a. 4 - o a CL < wM
%. % % f ' b % %
Figure 3.4. Effects of hydroxyurea pretreatment on paclitaxel activity in CWR22 tumors. Inhibition o f DNA synthesis (A) and induction of apoptosis (B) by paclitaxel 1.2 pM in the absence or presence of hydroxyurea (HU) 24-hr pretreatments at 0, 2 and 10 mM. Data represent mean ± sd, n=4. p<0.05 for *: vs. without paclitaxel; #: vs. HU 0 mM.
6 7 A 40 -
■E tI ZD 20 - " 2 00
0 B 16 -1 X 12 - o 8 - 2 Q. 4 - O Q. < t 2,10 48,96 2,10 2,10 2,102,10 48,96 . y :r .d » -* ' y y Figure 3.5. Effects of sequential treatments with paclitaxel and hydroxyurea in CWR22 tumors. Inhibition of DNA synthesis (A) and induction of apoptosis (B) by sequential treatments of paclitaxel 1.2 pM and of hydroxyurea (HU) at 0 , 2 and 10 nuVI according to regimens in Figure 1 are shown. HU-*paclitaxel; regimen B; paclitaxel+HU (simultaneous); regimen C; paclitaxel-tHU regimen D. Data represent mean ± sd, n>3. Exposure periods (48 or 96 hr) or doses of HU (2 or 20 mM) are as indicated. *: p<0.05 vs. paclitaxel alone. 68 r^ = 0.86 10 - p<0.01 X (D 7 3 C o o oÛ. Q. < 0 10 20 30 40 BrdUrd LI, % Figure 3.6. Correlation between paclitaxel-induced apoptosis and BrdUrd LI. Control LI and paclitaxel-induced apoptosis in the patient (A ) and CWR22 (■) tumors were also included. 6 9 A without paclitaxel c 6 - X g with paclitaxel il 4 - Il 2 - < — ' j % % % % % B 1 2 -1 CNJ m S 8 - 3 g Ü 4 - s-l Q - 03 # < _J %% % f %' Figure 3.7. EfTects of hydroxyurea on paclitaxel-induced apoptosis in prostate tumors. Paclitaxel (1.2 pM)-induced apoptotic cells mainly (>82%) are BrdUrd-labeled, whereas hydroxyurea (HU) also induces, to a less extent, apoptosis of which less BrdUrd- labeled cells. HU selectively decreases BrdUrd-labeled apoptotic cell fraction induced by paclitaxel. Data are meanisd, n=4. p<0.05 for *: vs. without paclitaxel; #: vs. HU 0 mM. 7 0 15 - X 0 TJ 1.2(pM) O O 0.12 Q. O Q. < ▼ 0.012 20 40 60 80 100 120 Time, hr Figure 3.8. Kinetics of apoptotic induction by paclitaxel in patient prostate tumors. Six tumors were treated with paclitaxel in various concentrations, untreated control (♦), 0.012 pM (T), 0.12 pM (•), 1.2 pM (■), and 12 pM (A). Data are mean±sd, n>3. 7 1 CHAPTER 4 PHARMACODYNAMICS OF DOXORUBICIN IN HUMAN PROSTATE TUMORS 4.1 Introduction Prostate cancer is the most common malignancy in man. An estimate of 43% of total newly diagnosed cancers in men are adenocarcinomas of the prostate and about 60% of the 334,500 new cases of prostatic adenocarcinomas will present with disease confined to the organ (Parker et al., 1997). Prostate cancer is a slowly progressing disease, and is usually diagnosed in older men at a median age of 70.5 years with fewer than 1% of patients less than 50 years old. The survival rate is significantly and inversely influenced by the tumor stage at diagnosis. An emphasis on early detection and the availability of new screening techniques, in particular the detection of prostate specific antigen (PSA) have increased the rate of diagnosis and lowered the stage and age at the time of diagnosis (Mettlin et al., 1993). The current treatment options for locally confined disease include surgical radical prostatectomy, radiotherapy and cryotherapy, with surgery being the most common treatment modality. Neoadjuvant hormone therapy prior to radical prostatectomy is under investigation to improve the control of localized prostate cancer. Some studies showed benefits of the 7 2 neoadjuvant hormone therapy (Vailancourt et al., 1996; Abbas et al., 1996), although time to progression or overall survival did not improve in another study (Abbas and Scardino, 1996). Androgen ablation therapy and systemic chemotherapy are usually reserved for metastatic disease and are not used for local disease. Among the agents used to treat advanced prostate cancer, doxorubicin produces one of the highest combined partial and complete response rate of 33% (Perez et al., 1997). However, doxorubicin, as a single agent or in combination with other chemotherapeutics, has not improved the survival rate (Oesterling et al., 1997; Rangel et al., 1992). Because chemotherapy is not commonly used in early prostate cancer, the activity of doxorubicin against early disease is not known. The present study was designed to determine the pharmacologic effects o f doxorubicin in surgical specimens of early stage human prostate tumors obtained by radical prostatectomy. The drug-induced antiproliferative and cytotoxic (cell kill) effects were quantified. The effective drug concentrations were compared with the literature data on clinically achievable drug concentrations in patients to infer potential therapeutic effectiveness of neoadjuvant or adjuvant systemic doxorubicin therapy. The study was performed using histocultures of surgical specimens of human prostate tumors obtained by radical prostatectomy. The histoculture system provides advantages over monolayer cell culture system. Histocultures maintain the 3-dimensional tissue structure and organization and thus co-existence of tumor and stromal cells, cell-cell interaction, and inter- and intra-tumor heterogeneity are preserved (Vesico et al., 1987). The maintenance of tissue architecture is critical because the interaction between the tumor and normal cells may be important for prostatic epithelial growth and 73 response to androgen stimulation. A previous study from our laboratory reported that human prostate tumor histocultures maintained their characteristics for at least 8 weeks as indicated by unchanged ^H-thymidine labeling index (LI) and secretion of prostate specific antigen (PSA) and prostatic acid phosphatase (Wientjes et al., 1995). The clinical relevance o f the human tumor histoculture system was recently demonstrated by Hoffinan and colleagues. These investigators show in retrospective and serai-prospective preclinical and clinical studies that drug responses in human tumor histocultures, using inhibition of DNA precursor incorporation or inhibition of metabolic reduction of tétrazolium dye as endpoint, correlates with the sensitivity, resistance and survival of head and neck, colorectal and gastric cancer patients to treatment by mitomycin C, doxorubicin, 5-fluorouracil, or cisplatin (Robbins et al., 1994; Furukawa et al., 1995; Kubota et al., 1995). 74 4.2 Materials and methods 4.2.1 Chemicals and supplies Doxorubicin was a gift from Pharmacia (Albuquerque, NM). Sterile pigskin collagen (Spongostan standard) was purchased from Health Designs Industries (Rochester, NY); tissue culture supplies (i.e., L-glutamine, sodium pyruvate, fetal bovine serum, gentamicin, Dulbecco's Modified Eagle's Medium (DMEM), Minimum Essential Medium (MEM), MEM non-essential amino acids solution and MEM vitamin solution) from GIBCO Laboratories (Grand Island, NY); ^H-thymidine (specific activity, 65 Ci/mmole) from Moravek Biochemicals Inc. (Brea, CA); NTB-2 nuclear track emulsion from Eastman Chemicals (Rochester, NY); terminal deoxynucleotidyl transferase (TdT), digoxigenin-dUTP and anti- digoxigenin-peroxidase in ApopTag* for in situ detection of dead and dying cells from Oncor Inc. (Gaithersburg, MD); proteinase K and mouse normal IgG from Sigma Co. (St. Louis, MO); the liquid 3,3'-diaminobenzidine substrate kit from BioGenex (San Ramon, CA); the mouse monoclonal antibody ER-PR 8 against human PSA and LSAB universal detection kit from Dako Inc (Carpiteria, CA); and the PSA detection kit from Hybritech (San Diego, CA). All chemicals and reagents were used as received. 4.2.2 Tumor procurement Surgical specimens of human primary prostate tumors were obtained from the peripheral zone of the prostate gland in patients who had organ-confined prostatic adenocarcinoma and who imderwent radical prostatectomy. Specimens were obtained via the Tumor Procurement Service at The Ohio State University Comprehensive Cancer Center, and from the neighboring Doctor's Hospital. Tumor specimens were placed in MEM within 10 to 75 30 min after surgical excision, stored on ice and prepared for culturing within one hr after excision. The histopathology of tumor specimens was established using frozen sections o f tumor fragments adjacent the histocultured specimens. Tumors were graded according to the Gleason grading system with a score of 2 for well-difierentiated tumors and a score of 10 for poorly-differentiated tumors (Gleason, 1966) by pathologists at the James Cancer Hospital (Columbus, OH). 4.2.3 Histoculture Tumor specimens were cultured as described previously (Wientjes et al., 1995). In brief, tumors were cut into < I mm^ pieces under sterile condition. Five to six tumor pieces were placed on a 1 cm^ presoaked collagen gel, and incubated at 37°C in a humidified atmosphere of 95% air and 5% COi The culture medium consisted of a 1:1 mixture of MEM and DMEM, 10% fetal bovine serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 40 pg/ml gentamicin, 0.1 mM MEM non-essential amino acids, and concentrated MEM vitamin solution ( 100 fold concentrated, 10 ml per liter). The histocultures were fed every other day and used for doxorubicin pharmacodynamic studies on day 4. 4.2.4 Detection of PSA in media and in histocultured tissues The secretion of PSA by human prostate tumor histocultures to the culture medium collected at the end of doxorubicin exposure was measured by a sandwich-immunoassay using two antibodies against two different epitope sites on PSA, performed by Immunology Laboratory of the James Cancer Hospital (Columbus, OH). PSA expression in histocultured tumors was measured by immunohistochemistry using a universal detection kit from Dako 76 Inc. .4Aer de-waxing and rehydration sequentially in xylene, ethanol and water, tissue sections were boiled in a 10 mM citrate buffer, pH 6.0, in a microwave oven for 15 min, cooled at room temperature for 15 minutes and washed once in PBS. Tissue sections were then incubated with the blocking solution for 10 min and subsequently with the mouse anti-human PSA (1:20 dilution in PBS containing 5 mg/ml BSA) in a humidified chamber at room temperature for 2 hr. Mouse normal IgG was used as the antibody for negative controls. After washing with PBS, the tissue sections were covered with the biotin-anti-IgO linker solution, and then with peroxidase-conjugated streptavidin solution. After washing twice with PBS, tissue sections were incubated for 3 to 7 min with the chromogen 3-3'-diaminobenzidine and with the substrate hydrogen peroxide, and then counterstained with hematoxylin followed by dehydration and coverslipping for microscopic examination. 4.2.5 Pharmacologic effects of doxorubicin The antiproliferative effect was measured by the inhibition of DNA precursor ([^H]- ihymidine) incorporation. The cytotoxic effect was measured by monitoring dead and dying cells using the TUNEL assay. Tumor histocultures were exposed to various concentrations of doxorubicin ranging from 0.00017 to 17 nM. for 96 hr. After drug treatments, the doxorubicin-containing medium was removed and tumor histocultures were washed three times with drug-free medium. Tumors were incubated with 0.03 ^M pH]-thymidine for 96 hr, washed 3 times with PBS, then fixed in 10% neutralized formalin, dehydrated and embedded in parafihn. The embedded tumor tissues were cut onto microscope slides at thickness using a microtome. Two sets of the slides with consecutively cut specimens were collected for autoradiography and for TUNEL assay separately. Untreated controls were 77 processed similarly. 4.2.6 Autoradiography Tumor sections on microscope slides were deparaflSnized, rehydrated and stained with hematoxylin. Slides were then coated with a thin layer o f NTB-2 nuclear track emulsion and exposed for 7 to 10 days in the dark in a cold room. Slides were developed, dehydrated and then coverslipped. The ^H-thymidine labeled cells were scored and the labeling index was calculated as (labeled cells) divided by (total cells). 4.2.7 TUNEL assay Procedures were carried out as described in the ApopTag*" detection kit. Tissue sections on slides were deparaflSnized, rehydrated and incubated with proteinase K (20 pg/ml) in PBS for 15 minutes at room temperature followed by 4 washes with distilled water for 2 min each. This protease treatment is necessary to make DNA fragments accessible for reaction with enzyme and substrate. Endogenous peroxidase was quenched by incubating tissue sections with 2% hydrogen peroxide in PBS for 5 min at room temperature followed by 2 washes with PBS for 5 min each. Equilibration buffer from the kit was applied directly onto tissue sections and incubated for 10 min at room temperature. TdT-containing solution from the kit was then applied and incubated in a humid chamber at 37°C for 1 hr. After the enzyme reaction, slides with tissues were incubated with 37°C-prewarmed Stop solution for 10 minutes at room temperature followed by 3 washes with PBS for 5 min each. Anti- digoxigenin-peroxidase from the kit was applied and incubated for 30 min at room temperature followed by 3 washes with PBS for 5 min each. Brown deposits were developed by incubating the tissue sections with chromogen 3,3'-diaminobenzidine and the substrate, 78 hydrogen peroxide, for 3 to 7 min. The volume of applied liquid was enough to cover the whole section. Between processes, excess liquid was tapped off gently and the tissue was blotted around carefully. The slides were kept in a humid chamber so that no concentration changes occurred due to the loss of water during the incubation periods. The tissue sections were counterstained with hematoxylin, dehydrated and coverslipped. The TUNEL-positive cells were scored. 4.2.8 Pharmacodynamic data analysis The drug concentration-efifect relationships were analyzed using Equations 4.1 ^annprohferauve (1 ) (Eq. 4.1) /C j + c (antiproliferative effect) and 4.2 (cytotoxic effect). cytotoxic „ control (Eq. 4.2) where Eggjpn^;^, is the LI of doxorubicin-treated specimens expressed as a percent of the LI from the untreated control, C is the drug concentration, K, and are the drug concentrations at 50% effect, n is a curve shape parameter, and E„„^i is the fraction of dead cells in untreated control. Values for IC 50 or IC% (drug concentration needed to produce 50 or 90% inhibition of DNA synthesis) and LCjo or LQo (drug concentration needed to induce 50 or 90% cell death) were determined. 7 9 4.2.9 Statistical analysis Differences in mean values between groups were analyzed using Student's t test. Software for statistical analysis (TTEST procedures) was by SAS (Cary, NC). 80 4.3 Results 4.3.1 Histocultures Twenty tumors, obtained from 20 patients, were studied. O f these, 17 tumors or 85% were successfully cultured. A successful culture is defined as having ^40 tumor cells per microscopic field and having a LI of ^5%. Table 4 .1 shows the patient and histopathological characteristics of the cultured tumors. All tumors were from chemotherapy-naive patients. The average control LI was about 39%. This value is similar to the average value of 41% in 34 human prostate tumors shown in the previous chapter (Chen et al., 1997a). Figure 4 .1 shows the presence of ^H-thymidine-labeled tumor cells in an untreated control tumor specimen. The cultured tumors secreted PSA as shown by the positive immunohistochemical staining (Figure 4 .1) and the presence of PSA in culture medium. The PSA concentration in cultured medium varied by - 130-fold, reflecting the inter-tumor variation (Table 4 . 1). 4.3.2 Antiproliferative effect The 96-hr doxorubicin treatment produced a sigmoidal, concentration-dependent inhibition of LI, reaching complete inhibition in all 17 tumors (Figure 4.2). Figure 4.1 also shows the micrographs of untreated control and doxorubicin-treated tumors. The drug-treated tumor showed a nearly complete inhibition of DNA synthesis but no morphological difference compared to the control. Among the 17 tumors, the doxorubicin concentrations that caused 50% and 90% inhibition of DNA synthesis (IC 50 and IC%) showed a 68 - and 18-fold range, respectively, indicating substantial inter-tumor variation. The antiproliferative effect of doxorubicin was inversely correlated with the age of patient (p<0.02) and with the LI of untreated controls and Gleason grade (p=0.07 and 0.06, 81 respectively), but had no correlation with tumor stage, PSA secretion, and race of patients (p>0.12). While age has been related to the incidence and progression o f prostate cancer, the reason for its inverse correlation with the antiproliferative effect of doxorubicin is not apparent. 4.3.3 Cytotoxic effect The 96-hr doxorubicin treatment produced a sigmoidal, concentration-dependent cytotoxic effect (Figure 4.2). Doxorubicin induced 100% cell death in all 17 tumors. In comparison, untreated controls showed insignificant cell death at 3.1±2.5%. Figure 1 shows the cells labeled by the TUNEL assay, which identifies cells with fi'agmented DNA. Interestingly, the dead and/or dying cells did not show the typical apoptotic features such as chromosome condensation and apoptotic bodies (Kerr et al., 1994). Table 4.1 summarizes the drug concentrations that produced a cytotoxic effect. The LC values were significantly correlated with the Gleason grade (p<0.05 for both LC^ and LCgo) and were weakly correlated with tumor stage (p=0.06 for LC % and 0.08 for LC% , respectively), but were not correlated with PSA secretion, baseline cell death, nor age and race of patients (p>0.18 for all cases). 4.3.4 Comparison of pharmacodynamics of antiproliferative and cytotoxic effects There was a correlation between the IC and LC values for individual tumors (i^=0.23, p=0.05). The doxorubicin concentrations that caused 50% and 90% of maximal death, i.e. LCjo and LCgo, were 35- and 6-fold higher than the IC% and ICg<, for the antiproliferative effect, respectively. For example, LC#was significantly higher than IC % (p<0.0001). The inter-tumor variation in the LC values was 2-fold lower than the IC values. These 82 comparisons indicate that human prostate tumors are more sensitive to the antiproliferative effect than the cytotoxic effect of doxorubicin. In addition, the opposite correlations between IC% and LC% with tumor Gleason grade suggest different determinants o f the antiproliferative and cytotoxic effects. 8 3 4.4 Discussion The purpose of the present study was to investigate if the clinically achievable doxorubicin concentrations are sufiBcient to produce significant antitumor effects in early stage human prostate tumors, in order to determine the potential value of neoadjuvant therapy. The results demonstrate that doxorubicin was able to completely inhibit DNA synthesis and induce cell death in histocultures of prostate tumors obtained via radical prostatectomy. In spite of the 13-fold higher inter-tumor variations for the antiproliferative effect than for cytotoxic effect, there was a correlation between the IC and LC values. This indicates a parallel tumor sensitivity to the two effects. The lower IC values compared to the LC values indicates a higher tumor sensitivity to the antiproliferative effect than the cytotoxic effect of doxorubicin. The drug concentrations needed to produce 50 and 90% antiproliferative and cytotoxic effects are between 60 and 4200 nM. These concentrations are 17-fold higher than the effective concentrations in monolayer cultures of human prostate cancer cells (Li and Au, 1997), probably due to biologic differences between continuous cell lines and human tumors and/or differences in culture conditions (monolayer versus 3- dimensional). The clinically achievable peak plasma concentration of doxorubicin during a 96-hr infusion varies from 22 to 178 nM (Legha et al., 1982a; Speth et al., 1987; Bugat et al., 1989), which is about the same as the average IC^ but is much lower than the average values of ICgo, LCjo and LC,,, (660, 2100 and 4,200 nM, respectively). The present study was performed using radical prostatectomy specimens which typically represent early stage disease. While the pharmacodynamics of doxorubicin in advanced prostate tumors have not 8 4 been studied, the general chemoresistance of advanced tumors suggests that the advanced tumors are likely to be more chemoresistant than the early stage tumors. Extrapolation of the results of the present study suggests inadequate drug delivery together with tumor chemoresistance as the reasons for the ineffectiveness of doxorubicin therapy in the treatment of advanced prostate cancer. In conclusion, results of the present study suggest that neoadjuvant or adjuvant systemic doxorubicin therapy has limited value in treating early stage prostate cancer. .An alternative treatment option is via regional drug delivery such as direct intraprostatic injection to provide high drug concentration in the prostate tissue to achieve appreciable antitumor effect. 85 4.5 Acknowledgments This study was supported in part by a research grant ROIC A74179 from the National Cancer Institute, DHHS. The Ohio State University Comprehensive Cancer Center Tissue Procurement Service was supported in part by P30CAI6058 from the National Cancer Institute, NIH, DHHS. We thank Drs. Robert Badalament, John Burgers, and Jack Perez for providing the tumor specimens. 86 Tumor AU A Gleason PSA in Control IC,q Control LC,„" LC,. No. Age Race Stage Grade medium thymidine (nM) (nM) TUNEL (nM) (nM) (ng/ml) LI (%) index (%) 1 58 Black 82 6 21.8 37 138 746 1.1 2598 4160 2 59 Caucasian B2 8 44.0 31 14 1103 0.7 910 1435 3 59 Caucasian C2 7 1.8 49 9 959 0 1496 6373 4 60 Caucasian C2 8 4.0 38 15 1107 3.0 853 2039 5 62 Caucasian 82 7 8.3 37 270 494 0 1448 6397 6 63 Caucasian 82 7 12.7 20 135 1440 2.5 2873 5396 7 63 Caucasian 82 7 32.9 56 18 240 7.2 1685 3582 8 65 Caucasian 82 6 8.9 46 15 1039 6.0 5674 7374 9 66 Caucasian 82 6 0.03 32 140 419 0 1771 5572 10 67 Caucasian 82 5 12.5 53 56 531 3.2 2987 5047 11 67 Caucasian 82 7 13.0 38 85 794 5.8 2552 4547 12 70 Black 82 7 5.8 43 11 165 2.6 3636 5222 13 70 Black 82 7 15.6 50 75 270 2.5 2527 4952 14 70 Caucasian D1 9 1.0 26 5 1078 5.1 881 1649 15 71 Caucasian 82 6 23.8 55 26 212 2.5 2261 4943 16 71 Caucasian C2 7 1.7 27 4 534 8.3 921 1715 17 73 Caucasian 82 6 0.5 31 20 79 2.8 1033 1592 Mean NA NA NA NA 12.3 39 61 659 3.1 2124 4235 SDNANANANA 12.4 11 73* 408" 2.5 1256* 1905" Median NANANANA 8.9 38 20 534 2.6 1771 4943 Table 4.1. Patients and tumor characteristics and tumor sensitivity of tumors to 96 hr doxorubicin treatments. iC;„ and IQu are the concentrations that caused 50% and 90% inhibition of thymidine labeling. LC 50 and LQo &re the concentrations that caused 50% and 90% of cell death, as determined by the TUNEL assay. American Urological Association (AUA) stage is same as the Whitmore and Jewett staging system. ■* p<0,000l between IC 50 and LCjq.*’ p<0.0001 between IC 90 and LC%. 87 00 00 Figure 4.1. Proitate tumor hbtocuHure. A human prostate specimen (Tumor #15 in Table 4.1), grown in histoculture for 12 day and labeled with ’H-thymidine (A, 400*), The ’H-thymidine labeling shows as black dots (enowheads). The expression of PSA was detected using iromunohistochemistiy (B, 400%). The same tumor treated with 1.7 pM doxorubicin for % hr had very few remaining ’H-thymidine-labeled cells (arrowheads in C), Cells in the same tumor as in C with fragmented DNA labeled by the TUNEL assay (D). Note the lack o f morphological changes in cells in the treated tumor in spite of the nearly complete inhibition of DNA synthesis. 100 100 o 80 - 80 V-. c «4—»c z o m o 60 M— o 3 40 40 Q. 0 X 0 0.001 0.1 Doxorubicin, Figure 4.2. Concentration-dependent antiproliferative and cytotoxic effects. Human prostate tumor histocultures were treated with doxorubicin for 96 hr. Inhibition of 96- hr cumulative ^H-thymidine LI ( • ) was expressed as percent o f untreated controls. Induction of dead or dying cell deaths (O) was assessed by TUNEL as percent o f total tumor cells. The sigmoidal lines are computer-fitted lines according to Equations 4 .1 and 4.2. Data represent mean ± SEM of 17 tumors. 89 CHAPTER 5 TISSUE PENETRATION OF DOXORUBICIN INTO PROSTATE XENOGRAFT TUMOR HISTOCULTURES 5.1 Introduction In previous and ongoing studies we have used the histoculture system to study the pharmacodynamics of doxorubicin in prostate cancer (Wientjes et al., 1995; Chen et al., 1997a&b; Chapters 2, 3, 4, and 7). The histoculture system enables us to evaluate drug activity in patient tumors or xenograft tumors. This possibility to use patient specimens gives the method the potential to be clinically relevant, as was supported by recent studies by Hoffman and colleagues. These investigators showed in prospective studies a correlation between chemosensitivity in the histoculture system with clinical outcome in head and neck, colorectal and gastric cancer patients treated with mitomycin C, doxorubicin, 5-fluorouracil, or cisplatin (Robbins et al., 1994; Furukawa et al., 1995; Kubota et al., 1995). The drug concentrations effective for doxorubicin in histocultures were several orders of magnitude higher than were found in cell line studies (Li and Au, 1997). One of the differences between monolayer and three-dimensional systems such as the histoculture systems is the kinetics of drug uptake and efflux. The aim of the current study is to evaluate the rate of equilibration 9 0 of doxorubicin between medium and the histocultures, and the kinetics of doxorubicin degradation during the in vitro studies. The human prostate tumor xenograft CWR22, which is a renewable source of tumor specimens, was used. The selected drug concentrations include the drug concentrations that have shown significant antipoliferative and cytotoxic effects in histocultures of patient and xenograft tumors. This tumor, established by Pretlow and associates, is an androgen-dependent xenograft originally derived from a prostate tumor of Gleason grade 9 from a patient with Stage D disease and osseous metastasis (Pretlow et al., 1993). 91 5.2 Materials and methods 5.2.1 Chemicals and instruments Male athymic Nu/Nu mice, BALB/C retired breeders, were purchased from National Cancer Institute (Frederick, MD). Testosterone tablets, 12.5 mg/tablet, were purchased from Innovative Research of America (Toledo, OH). Matrigel* from Becton Dickinson Labware (Bedford, MA). Sterile pigskin collagen (Spongostan standard) was purchased from Health Designs Industries (Rochester, NY). Tissue culture supplies (i.e., L-glutamine, sodium pyruvate, fetal bovine serum, gentamicin, Dulbecco's Modified Eagle's Medium (DMEM), Minimum Essential Medium (MEM), MEM nonessential amino acids solution and MEM vitamin solution) were from GEBCO Laboratories (Grand Island, NY). Doxorubicin and epirubicin were gifts from Pharmacia (Albuquerque, NM). Solid phase extraction tubes, Supelclean LC-18, were from Supelco (Bellefonte, PA). A rotor-stator type of tissue homogenizer, Lfltra-Turrax^Tissumizer, was from Tekmar (Cincinnati, OH). A centrifuge, GS-6R, was from Beckman (Fullerton, CA). A solvent delivery pump, Spectroflow 400, was from ABI Analytical, BCratos Division (Ramsey, NJ). A 3 pm particle size, 83 mm x 4.6 mm Pecosphere ™ reversed-phase C,g column for separation was from Perkin-Elmer (Norwalk, CT). An automated injector, WISP 712, and a scanning fluorescence detector. Model 470, for detection and quantitation of doxorubicin and epirubicin were from Waters (Milford, MA). The excitation and emission wavelengths were 480 run. and 550 nm, respectively. Data acquisition and integration were performed with a computer using the HP ChemStation software from Hewlett-Packard (Wilmington, DE). All chemicals and reagents were used as received. 9 2 5.2.2 CWR22 prostate tumor xenograft growth in vivo CWR22 is an androgen-dependent human prostate xenograft tumor. Growth of the CWR22 xenograft was maintained in male nude mice according to the previously published procedures (Pretlow et al., 1993; Nagabhushan et al., 1996). Subcutaneous implantation of a testosterone pellet was performed 3 days before the implantation of tumor fragments. Minced tumor tissues mixed with the same volume of Matrigel were implanted into both flanks of the nude mice at 0.3 ml per site. Tumors were harvested when they reached a size of 1 g at about 1.5 to 2 month after implantation. 5.2.3 Histocultures and drug treatments Tumor specimens were cultured as described previously (Chapter 2; Chen et al., 1997a). In brief, tumors were cut into < 1 mm^ pieces under a sterile condition. Five tumor pieces were placed on a 1 cm^ medium-presoaked collagen gel, and incubated at 37°C in a humidified atmosphere of 95% air and 5% CO,. The culture medium consisted of a 1:1 mixture of MEM and DMEM, 10% fetal bovine serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 40 pg/ml gentamicin, 0.1 mM MEM nonessential amino acids, and concentrated MEM vitamin solution (100 fold concentrated, 10 ml per liter). The histocultures were fed every other day for 3 to 4 days and then treated with doxorubicin (0.02, 0.06, 0.12, 0.5, 2 and 10 pM) for 96 hr. After exposure, the cultures were washed and replenished with fresh medium for another two days. The media and cultured tumor specimens were collected at 3, 6, 12, 24, 48, and 96 hr after exposure and at 24 and 48 hr after removal of the drug-containing medium and kept at -20 °C until doxorubicin assay. Histocultures of tumor specimens at 2 days after doxorubicin treatment were kept in formalin-fixed parafiin embedded blocks and sectioned onto slides for 9 3 microscopic examination. 5.2.4 Concentration analysis Samples of culture medium or tumor tissues were collected from the histoculture experiments until assay. Sample preparation and extraction were according to previously published methods with slight modifications (Chai et al., 1994; Cox et al., 1991). Epirubicin was used as the internal standard. 5.2.4.1 Tissue samples The weighted tumor samples (-5 mg), collected in centrifuge tubes with 2 ml acidified methanol, 50 mM phosphate buffered 95 % pH 3.0 methanol (5:95) pH 3.0 and homogenized for 1 minutes. The homogenizer was washed with 3 ml methanol. All methanol fractions wee combined and reduced to a volume of less than 2 ml by evaporation, followed by a solid phase extraction. The solid phase extraction tubes were conditioned with 3 ml 100% methanol followed by 3 ml of a 1:3 mixture of methanol:20 mM potassium phosphate, pH 8.0. The tumor extracts were diluted with 3 ml 10 mM potassium phosphate, pH 8 .0, and 2 ml methanol, and applied to the conditioned C,g solid phase extraction tubes. After washes with 1 ml of water followed by 2 ml of 50% methanol in water, the analytes were eluted with 6 ml of a 95:5 mixture of methanol:50 mM potassium phosphate, pH 3.0. The elution fraction was then evaporated to dryness. The residue was reconstituted with 200 or 500 pi of the mobile phase for samples of low or high concentration, respectively. The prepared samples were then analyzed by high performance liquid chromatography (HPLC). 5.2.4.2 Medium samples Medium samples were processed by direct protein precipitation using 10% trichloroacetic 9 4 acid followed by centrifugation at 7,000 g for 10 min. The supernatant was analyzed by HPLC. S.2.4.3 Collagen gel and plate The collagen gels used to histoculture growth were analyzed by the same method as in tissue samples. Culture plates were washed with methanol three times. The washes were collected and evaporated to dryness. The residue was reconstituted with mobile phase and then ready for high performance liquid chromatography assay. 5.2.5 High performance liquid chromatography assay HPLC analysis of medium was performed isocratically using a Pecosphere™ reversed- phase Cix column, and a mobile phase of 30% acetonitrile in 20 mM potassium phosphate, pH 3.0. The flow rate was 0.8 ml/min. Doxorubicin and epirubicin in the eluent were detected with a scanning fluorescence detector, using excitation and emission wavelengths of 480 nm and 550 nm, respectively. The detection limit for doxorubicin was 0.3 nM. The retention time was ~1 min for doxorubicin and '9 min for epirubicin. Standard curves were linear within the range of 1 to 100 ng/ml (i.e. 0.002 to 0.17 /xM) for culture medium. 5.2.6 Data analysis Equation 5.1 was used to calculate the half-life, t^.^^for the doxorubicin concentration in tumor to reach one-half of its maximal value. Co,medium's the initial drug concentration in culture medium, a is the tissue-to-medium concentration ratio at equilibrium, and p is the rate constant describing the attainment of equilibrium. 9 5 “ ' Cwmm* ( 1 ' ^ ' 0 (Eq. 5. I) The computer software, WINNONLIN by Scientific Consulting Inc. (Apex, NC), was used for data fitting analysis. Differences in mean values between groups were analyzed using Student's t test (TTEST procedure) by SAS (Cary, NC). 9 6 i 5.3 Results 5.3.1 Profiles of doxorubicin concentration in the culture medium The concentration of doxorubicin in culture medium with time (Figure 5. 1). The maximal decline of doxorubicin concentration in the medium after 96 hr incubation was 44±6% (mean±SD) for the six initial concentrations. The half-life of doxorubicin concentration decline in the medium was about 62 hr. Figure 5.1 also shows the drug concentration in culture medium after replacing drug- containing medium with drug-ftee medium. The results show appreciable drug concentrations at 24 and 48 hr (i.e. 120 and 144 hr after incubation was initiated). These results indicate release of doxorubicin from tumor histocultures and from the collagen gel matrix supporting the histocultures. 5.3.2 Tissue concentrations of doxorubicin in the CWR22 xenograft tumors The doxorubicin concentration in xenograft tumor histocultures increased with time (Figure 5.2). The kinetic parameters describing drug uptake are summarized in Table 5.1. The time to reach 50% of the maximal drug concentration in histocultures, Tn,^.\.i % increased with decreasing initial drug concentration in culture medium, ranging from -10 hr at > 2 pM initial concentration to -40 hr at < 0.12 pM. The results indicated that the accumulation of doxorubicin in tumors was dependent on the extracellular drug concentration. Figure 5.2 also shows the decline of drug concentration in tumors after replacing the drug-containing medium with drug-free medium at 96 hr. The tumor concentrations at 24 and 48 hr after removing drug-containing solution were 61 and 42% of the tumor concentrations prior to drug removal, indicating a slow release and/or degradation of doxorubicin in tumors. 97 5.3.3 Distribution ratio of doxorubicin between tumor specimens and medium Table 5 .1 summarizes the ratios of drug concentration in tumor and in culture medium at the end of 96 hr incubation. The results indicate similar ratios for all six initial medium concentrations ranging from 0.02 to 10 pM, and a 85-120 fold higher tumor concentration at 96 hr. On the other hand, the ratio of areas under the doxorubicin concentration-time curves (AUC) in tumor and medium, from time 0 to 96 hr increased with the initial medium concentrations (Table 5.1, Figure 5.3). 5.3.4 Destruction of tumor tissue by doxorubicin Figure 5.4 shows the tumor structure after doxorubicin treatment. The results show- decreasing cell density with increasing initial medium concentration. At the highest concentration of 10 pM, the histoculture contained empty space and few live cells. 5.3.5 Distribution of doxorubicin to the culturing gel and plate The amount of doxorubicin in the tumor tissue accounted for < 7% of the applied dose. The amount in the collagen gel, expressed as a fraction of the applied dose, decreased with increasing initial medium concentration and ranged from 27% at 0.02 pM initial concentration to <5% at 10 pM concentration (Figure 5.5). The results suggest saturable binding of doxorubicin to the collagen gel. In comparison, <1% of the applied dose was recovered from the culture flask by exhaustive methanol washes. 98 5.4 Discussion The results of the present study indicate that the penetration of doxorubicin into tumor histocultures was relatively slow. The rate of penetration was positively correlated with the initial medium concentration. Furthermore, we found drug concentration-dependent histological changes in tumor architecture and tumor cell density. It has been shown that the penetration of doxorubicin in multicellular system such as the spheroids with penetration is slow and limited to the several peripheral cell layers (Durand, 1990), which suggest a penetration barrier for doxorubicin. Hence, we propose that the more rapid rate of penetration at higher drug concentrations is due to destruction of tumor architecture and consequently the penetration barrier. This hypothesis is supported by other studies in our laboratory on paclitaxel using image analysis; these studies showed a correlation between the extent o f drug uptake in tumor histoculture with tumor architecture and with tumor cell density (Kuh et al., 1996). The 85-120 fold higher concentration in tumors compared to culture media at 96 hr indicates significant drug accumulation in tumor cells. This is consistent with the reported tissue accumulation in vivo (Timuor et al., 1987; Terasaki et al., 1984). This study further provides the tumor concentrations that are achieved at extracellular effective drug concentrations as defined in Chapter 4. The tumor concentrations at drug concentrations that produce 50% and 90% antiproliferation are 3 and 20 pM, respectively. The tumor concentrations at drug concentrations that produce 50% and 90% cytotoxicity are > 100 pM. This information was used together with the results in Chapter 6 , i.e. drug concentration in prostate tissue after an intravenous dose, to conclude that the drug 99 concentration that is delivered to the prostate by an intravenous bolus injection or slow infusion is insufficient to elicit a significant drug effect. These findings have led us to the major conclusion that an improvement of doxorubicin efficacy in the treatment of prostate cancer would require improvement of drug delivery to the prostate, and therefore suggest the approach of regional intraprostatic drug administration. 100 5.5 Conclusions This study has defined the kinetics of doxorubicin uptake in solid tumors and has provided data on the drug accumulation at the extracellular concentrations that are known to produce antiproliferation and cytotoxicity. These data were used, together with data o f Chapters 4 and 6, to design strategies tor improving the efficacy of doxorubicin treatment in prostate cancer patients. 101 5.6 Acknowledgments We thank JefF Johnston and Dr. Peter Koo for providing some of the xenograft tumors used in the study. 102 drug removal 10 ■ O E 1 3 10 (nM) y (D E c 0.1 c o !o 3 u. 0.5 O 0.01 X o 0.12 a 0.06 0.02 .001 0 48 96 144 Exposure Time, hr Figure 5.1. Concentration-time profiles of doxorubicin in culture medium. Histocultures of CWR22 tumor were Incubated with doxorubicin for 96 hr at the following concentrations: 0.02 (•), 0.06 (■), 0.12 (A), 0.5 (▼), 2 (♦), and 10 pM (O). The drug- containing medium was replaced with drug-free medium at 96 hr. Data are expressed as mean ± sd of ns3. 103 drug removal 1000 1 10 (mM) 100 0 ) 3 05 05 C 0.5 y \a 0.12 L_3 O X o 0.06 Q 0.02 0 48 96 144 Exposure Time, hr Figure 5.2. Tissue concentrations of doxorubicin in CWR22 tumor histocultures. Histocultures of CWR22 tumor were incubated with doxorubicin at the following concentrations: 0.02 (•), 0.06 (■), 0.12 (A), 0.5 (T), 2 (♦), and 10 ^iM (O). The drug- containing medium was replaced with drug-free medium at 96 hr. Data are expressed as mean ±sd of na3. 104 I -B 60 - (D E 55 - (D 3 50 - CO CO 45 - 40 - 0.01 0.1 1 10 Doxorubicin, Medium, pM Figure 5.3. Increase of AUCm„„/AUC„,j,um ratios with initial drug concentration in culture medium. A U C j^and ALfC„„*„ were calculated from the concentration-time profiles from time 0 to 96 hr, using trapezoidal rule. 105 Figure 5.4. Structural changes in tumor histocultures after doxorubicin treatment. CWR22 tumor histocultures were treated with doxorubicin at 0.12 (A), 0.5 (B), and 10 fiM (C) concentrations for 96 hr. Micrographs of tissue were obtained after incubating the histocultures for additional 48 hr with drug-free medium. 106 20 - ■ a 0 L_ 0 > O O 10 - 0 a : 0.01 0.1 10 Doxorubicin, Medium, pM Figure 5.5. Recovery of doxorubicin form the collagen gel matrix. Doxorubicin recovered from the collagen gel matrix supporting the histocultures are expressed as a % of the applied dose. 107 c T '-O .m cdium ma.x. 1/2 Ciissue^^medium (pM) (hr) (pM) (0-96 h) 0.02 37 0.79 94 35.4 0.06 40 2.87 112 39.8 0.12 42 5.85 105 44.8 0.5 23 21.55 119 47.5 2 9 100.90 113 55.9 10 10 458.20 84 60.4 Table 5.1. Kinetic parameters of doxorubicin uptake in CWR22 xenograft tumor histocultures. mcdiun,; The doxorubicin concentration in culture medium at time zero. 2 • Half-time to reach the maximal tumor tissue concentration. Cmax The maximal doxorubicin concentration in tumor tissue. Ctosu 108 CHAPTER 6 DOXORUBICIN CONCENTRATIONS IN DOG PROSTATE AFTER SYSTEMIC ADMINISTRATIONS 6.1 Introduction Doxorubicin, although it is the most active single agent for treatment of prostate cancer, produces only 33% response with no survival advantage (Perez et al., 1989). The low response can be due to insufficient drug delivery to the prostate and tumor resistance to the drug. We have shown in a previous chapter (Chapter 4; Chen et al., 1997b) that doxorubicin is active against histocultures of prostate tumors obtained from patients and can produce 100% inhibition of proliferation and cell kill, although the extracellular concentrations that are required to produce these effects are at least 4 fold higher than the clinically achievable plasma concentrations in patients. The present study examined the role of the tissue kinetics of doxorubicin in the prostate after an intravenous dose as a determinant of the low response. This study was performed in dogs because the dog prostate is comparable to human prostate in that the dog is the only animal showing spontaneous prostate cancer as human and has a glandular morphology of the prostate gland similar to humans (Waters and Bostwick, 1997; Bertsch and Rohr, 1980). To determine if the 96 hr infusion schedule, which is used because of its lesser cardiac 109 toxicity compared to shorter infusions (Legha et al., 1982a), compromises the drug accumulation, we compared the drug accumulation in the prostate after a bolus injection and a 96-hr infusion of the same total dose of 2 rag/kg. This is the maximal tolerated dose in dogs with comparable plasma concentrations to that found in human patients (Legha et al., 1982a; Speth et al., 1987; Bugat et al., 1989). The drug accumulation in prostate under in vivo conditions was further compared to the results in Chapter 5 on the in vitro drug accumulation in histocultures of the human prostate CWR22 xenograft tumor. 110 6.2 Materials and methods 6.2 .1 Chemicals and Instruments Doxorubicin and epirubicin were gifts from Pharmacia Inc. (Albuquerque, NM). The equipment used included a portable infusion pump (CADD-PLUS Pump, SIMS Deltec, St. Paul, MN), a rotor-stator type of tissue homogenizer (Ultra-Turrax* Tissumizer, Tekmar Co., Cincinnati, OH), and a GS- 6R centrifuge from Beckman (Fullerton, CA). The high performance liquid chromatography (HPLC) consisted of a solvent delivery pump (Spectroflow 400, ABI Analytical, Kratos Division, Ramsey, NJ), an automated injector (WISP 712, Waters .Associates, Milford, MA), and a scanning fluorescence detector (Model 470, Waters Associates). HPLC data acquisition and integration were performed using HP ChemStation software from Hewlett-Packard (Wilmington, DE). HPLC solid phase extraction tubes (Supelclean LC-18) were purchased from Supelco Inc. (Bellefonte, PA), HPLC reversed-phase C,x column (3 pm particle size, 83 mm x 4.6 mm Pecosphere™ ) from Perkin- Elmer (Norwalk, CT). All other chemicals were purchased from Sigma Co. (St. Louis, MO) and were used as received. 6.2.2 Animal protocols Fourteen male beagle dogs, 1 to 1.5 yr old and weighing 10.5±0.9 kg, were used. Animals were housed in a facility with controlled 12-hr light cycle for a week before experiments. Animals were allowed free access to food and water. Animals were studied in two groups. One group received intravenous bolus injection of doxorubicin. For these animals, an angiocatheter (16 gauge, 2 inches in length) was inserted into a cephalic vein for bolus administration. The second group of animals received intravenous doxorubicin over 96 111 hr infusion. These animals were sedated with 1 mg/kg of acepromazine, administered subcutaneously, and then catheterized in the right jugular vein using a 15 cm long flexible polyurethane catheter. The drug solution was infused through the jugular vein catheter using the CADD-PLUS Pump placed in the animal jacket. Both groups of animals received 2 mg/kg doxorubicin. In all animals, a cephalic vein was catheterized for blood sampling in the first 24 hr. Later less frequent blood sampling was performed by needle puncture of alternate peripheral veins. At predetermined times, blood samples (2-3 ml each) were obtained. After anesthetizing animals with an intravenous dose of 26 mg/kg pentobarbital, prostate tissues were harvested and animals were euthanized by pentobarbital overdose. 6.2.3 Sample extraction and analysis Plasma and tissue samples were stored fi-ozen at -20°C. Samples were extracted using previously published methods (Chai et al., 1994; Shinkai et al., 1996) with minor modifications. The tissue extraction method produced a 70 to 85% recovery (Shinkai et al.. 1996). Epirubicin was used as the internal standard. Thawed prostate tissues were sectioned and weighed. About 100 mg of tissues was placed in 50 ml centrifuge tubes, and homogenized with epirubicin (2 pg/g) and 20 ml acetone for 1 minute. The tissue fi-agments remaining in the homogenizer were recovered by washing with 20 ml of acetone. The two acetone fractions were combined and centrifuged for 10 minutes at 2000 g. The supernatant was transferred, evaporated to dryness, and the residue was reconstituted with 500 pi of HPLC mobile phase. Plasma samples were processed by solid phase extraction. The C,* extraction tubes were preconditioned with 3 ml 100% methanol, followed by 3 ml of a 1:3 mixture of 112 methanol;20 mM potassium phosphate, pH 8.0. Plasma samples (0.5 or I ml) were mixed with epirubicin (20 ng/ml), diluted with 1 ml 10 mM potassium phosphate, pH 8.0, and 1 ml 100% methanol, and then applied to the extraction tubes. After washes with 1 ml of water followed by 2 ml of 50% methanol in water, doxorubicin and epirubicin were eluted with 3 ml of a 95:5 mixture of methanol:50 mM potassium phosphate, pH 3.0. The eluents were then evaporated to dryness and the residue was reconstituted with 200 pi of HPLC mobile phase. 6.2.4 High performance liquid chromatography analysis HPLC analysis of plasma and tissue extracts was performed isocratically using a Pecosphere™ reversed-phase C,* column, and a mobile phase of 30% acetonitrile in 20 mM potassium phosphate, pH 3.0. The flow rate was 0.8 ml/min. Doxorubicin and epirubicin in the eluent were detected with a scanning fluorescence detector, using excitation and emission wavelengths o f480 nm and 550 nm, respectively. The detection limit for doxorubicin was 0.3 nJVI in plasma. The retention time was -7 min for doxorubicin and -9 min for epirubicin. Standard curves were constructed separately for prostate tissue and plasma. Both standard curves were linear within the range of 0.05 to 6 pg/ml (i.e. 0.09 to 10 pM) for prostate tissue and 1 to 80 ng/ml (i.e. 0.0017 to 0.0136 pM) for plasma. To confirm the tissue extraction yield o f 70 to 85% as reported previously (Shinkai et al., 1996), the residual tissue after extraction was twice re-extracted using the same procedures. The results showed a 5 to 10% recovery from the residual tissue, thus indicating an extraction recovery of >90%. 6.2.5 Data analysis The plasma pharmacokinetics of doxorubicin were analyzed using a three compartment open-body model, as shown by others (Chan et ai., 1978; Iguchi et al., 1986). 113 Equation 6 .1 was used to estimate the pharmacokinetic parameters, including total body clearance (Cl), area under the plasma concentration curve (AUC), area under the first moment plasma concentration curve (AUMC), mean residence time (MRT), half-life of each disposition phase (t,,^„, t,op and tj^,^) and steady state volume of distribution (Vd^). = /( e i g (Eq. 6.1) where Cp is the plasma doxorubicin concentration, a, P, and y are the rate constants of the three disposition phases, respectively. The area under the concentration-time profile in prostate (AUCprosmJ for the bolus dose was calculated using the prostate concentration-time data and the trapezoidal rule. The AUCp„j„,j for the 96 hr infusion was estimated as the product of the steady state tissue concentration and the duration of infusion. The attainment of a quasi-equilibrium between prostate and plasma concentrations was analyzed by Equation 6.2. J 2 - L - • rime = Ratio [I - e (Eq. 6.2) C Plasma where Ratio is the concentration ratio when the equilibrium is achieved, and t,o is the half- life of approach to the equilibrium. The computer software for data fitting and analysis was by SAS (Cary, NC) and WINNONLIN was by Scientific Consulting Inc. (Apex, NC). Differences in mean values between groups were analyzed using the Student's t test (TTEST procedure, SAS). 114 6.4 Results 6.4.1 Plasma concentrations Figure 6.1 shows the plasma concentration-time profiles of doxorubicin after intravenous administration by bolus injection and by infusion over 96 hr. Following the bolus injection, the drug concentration declined rapidly from 2150 nM at 5 min to 19 nM at 3 hr, followed by a slow terminal phase with a half-life of ~110 hr. After the 96 hr infusion, the plasma concentration at 5 hr was about one-half of the maximum value of 17 nM. The pharmacokinetic data obtained after the bolus dose were used to calculate the pharmacokinetic parameters, summarized in Table 6. 1. These results indicate a rapid a phase and slow elimination of doxorubicin. The large volume of distribution indicates extensive tissue binding. The long mean residence time could not be due to slow drug elimination, because of the total high body clearance (i.e. equal to 58% of the liver blood flow in dogs, ref. Greenway and Stark, 1971). The long mean residence time, together with the large volume of distribution, indicate that the long y phase is due to the slow release of drug from tight tissue binding sites. 6.4.2 Prostate concentrations Figure 6.2 shows the prostate tissue concentration-time profiles of doxorubicin. After bolus injection, drug concentration in prostate was at its maximum value of 6 pM at the first sampling time of 5 min, declining to 1.1 pM at 96 hr. After the 96 hr infusion, the maximum tissue concentration was 2.87 pM. The estimated values of AUCp^,,,. for the bolus dose and the 96 hr infusion were 195 and 276 pM hr. This indicates that doxorubicin is more efficiently delivered to the prostate by slow infusion than by bolus administration. 115 To confirm the slow release of doxorubicin fi'om tissues, a prostate was obtained from one animal at 48 hr after the termination of the 96 hr infusion. The results showed a 60% decline in prostate concentration from 2.87 to 1.05 pM during the 48 hr period and the plasma concentration decreased by a 70% from 19.8 to 5.6 nM. This confirms the significant drug retention in tissues suggested by the plasma pharmacokinetic data. Figure 6.3 shows the changes of ratios with time, after the intravenous bolus injection and infusion. Analysis of these data by Equation 6.2 showed that the prostate and plasma concentrations reached a quasi equilibrium at 18 hr, and that the t, ^ was 1.6 hr. The equilibrating ratio was 180, indicating extensive binding o f doxorubicin to prostate tissue. The constant Cp^,„,.:Cpi,o,„ ratio at 48 hr after the termination o f the 96 hr infusion indicates the maintenance of the concentration equilibrium after drug infusion. 116 6.5 Discussion The present study examined the kinetics of doxorubicin penetration into prostate tissue. The results indicate (a) relatively rapid drug penetration into prostate, (b) that after an intravenous bolus injection, the ti^, for the prostate concentration to reach a quasi equilibrium with the plasma concentration was 1.6 hr, (c) extensive drug accumulation in prostate, resulting in the up to 180 fold higher drug concentration in prostate than in plasma, and (d) slow drug release from prostate. The kinetic properties of doxorubicin accumulation in prostate accounted for the comparable AUCpro,^ic attained by the bolus injection and the 96 hr infusion. The latter finding indicates that the slow infusion is at least equally effective or better in delivering the drug to the prostate, which support the use of the 96 hr infusion without compromising its anticancer activity, especially in view of its lower cardiac toxicity compared to the bolus injection (Legha et al., 1982a). The time needed for the prostate concentration to reach equilibrium with the plasma concentration can be due to the relatively slow perfusion rate of the prostate, slow drug penetration from interstitial fluid to prostate tissue, and/or saturable tissue binding. Saturable tissue binding would result in a low tissue concentration in the presence of initially high plasma/interstitial concentration and hence a lower ratio initially, followed by increasing ratios as the plasma/interstitial concentration reduced to below the saturable range. A separate study in our laboratory found that the doxorubicin concentrations in other rapidly perfused tissues such as heart, liver, and muscle reached a quasi equilibrium with the plasma concentrations within 1 hr after an intravenous bolus dose. Furthermore, the concentrations in liver at equilibrium were 7 fold higher than the prostate concentration at equilibrium. The 117 rapid attainment of the higher equilibrium concentrations in the liver at plasma concentrations that are identical to those in the present study further indicates that saturable tissue binding is not likely the cause of the slow equilibrium in prostate. In fact, the higher equilibrium concentrations and the more rapid attainment of equilibrium in the liver indicate that slow perfusion is a likely possible cause of the relatively slow equilibrium in prostate tissue. It has been shown that doxorubicin slowly penetrates multicellular systems such as spheroids, with drug localization in the outer cell layers of the spheroids. Hence, slow penetration of doxorubicin into prostate tissue might have also caused the slow attainment of equilibrium. The Cp„„ 3,^;Cp|„nu ratio of 180 is within the range of the tissue-to-plasma ratios of doxorubicin reported in other animal tissues which range from 20 ; 1 in rabbit adipose tissue to 1,800:1 in guinea pig spleen (Harris and Gross, 1975; Terasaki et al., 1984). The slow release of doxorubicin from prostate is consistent with the detection of doxorubicin in several tissues (liver, lung, lymph node, kidney, spleen and heart) after the plasma concentration declines below the assay detection limit of 1 ng/ml (1.7 nM) at 21 days after bolus injection (Timour et al., 1987). The differential concentrations of a series of antibiotics used for prostatitis in blood and prostatic acinar fluid has led to the hypothesis that there is a blood-prostate barrier of drug transfer, similar to the blood-brain barrier (Barza, 1993). It has since been shown that the pH gradient between the acinar fluid and interstitial fluid results in a higher concentration of basic drugs (e.g. trimethoprim, erythromycin and clindamycin) and a lower concentration of acidic drugs (e.g. rifampicin, penicillin G and ampicillin) in the acidic prostatic acinar fluid compared to the plasma (Stamey et al., 1973; Meares, 1982). Another study showed 118 insignificant amount of doxorubicin found in the central nervous system after systemic administration (Neuwelt et a!., 1983). This finding, together with our current results, indicates the different transport mechanisms of doxorubicin to the central nervous system and to the prostate. The current finding of the 180-fold higher prostate tissue concentration of doxorubicin compared to its plasma concentration supports the absence of a blood-prostate barrier. In addition, a comparison of the doxorubicin in prostate to the results of a separate study in our laboratory which measured the doxorubicin concentrations in other tissues indicate the prostate concentrations rank between the concentrations in heart and muscle, which further suggests that there is no blood-prostate barrier for doxorubicin transfer. The maximum ratio of 180 for doxorubicin observed in this in vivo study is higher than the in vitro accumulation ratio of ~ 100 in CWR22 tumor histocultures exposed to several drug concentrations which represent the IC,q, IC%, LQo, and LQ„ values (see Results in Chapters 4 and 5). This difierence may be due to the difference in tissue structures under in vivo and in vitro conditions. As shown in Chapter 5, the CWR22 tumor increasingly lost its structural integrity and cell density as a function of doxorubicin concentration. The diminished cell density, because of the extensive binding of doxorubicin to intracellular macromolecules such as DNA, might have resulted in the lower accumulation ratio in CWR22 tumor under in vitro conditions. Further studies to confirm the drug accumulation under in vitro and in vivo conditions are warranted. The plasma concentrations of doxorubicin in dogs after the intravenous bolus injection or the 96 hr infusion are at the low end o f the range of concentrations in humans that receive a similar dose (i.e. 19 nM in dogs vs 22 to 178 nM in humans, ref. Legha et al., 1982a; Speth 119 et al., 1987; Bugat et al., 1989). This corresponds to a lower drug clearance in dogs compared to humans. The present study is one in a series to determine if the low response o f prostate cancer to doxorubicin therapy may be due to insuflBcient drug delivery to the prostate. We have shown that in prostate tumor histocultures derived from patients, the doxorubicin concentrations that produce 50% and 90% inhibition of DNA synthesis (IC 50 and ICg„) are 60 and 660 nM and the concentrations that produce 50% and 90% cell kill (LC5,, and LC,,,) are 2100 and 4200 nM. These extracellular IC and LC exceed the clinical achievable plasma concentrations in patients, indicating that insufficient drug delivery to the prostate as a possible mechanism of the low response to doxorubicin therapy. To further confirm this hypothesis, we compared the prostate tissue concentrations after an intravenous dose, obtained in the present study, with the drug accumulation in CWR22 xenograft tumor histocultures treated with IC 5,,, IC%, LC%, and LC%, as shown in Chapter 5. The comparison shows that the prostate tissue concentrations are comparable to the concentration in CWR22 tumor attained after exposure to IC 50 but are 25 fold lower than the concentration attained after exposure to LCj,,. This indicates that the drug concentration delivered to the prostate by an intravenous dose of 2 mg/kg is sufficient to produce 50% antiproliferation but not 50% cytotoxicity. It has been proposed that for slow growing tumors such as prostate cancer, cytotoxicity is the more important effect that leads to tumor shrinkage (Isaacs et al., 1981). We conclude that improvement of delivery of doxorubicin to the prostate, such as direct intraprostatic drug injection, may improve the treatment outcome. 120 6.6 Acknowledgment This study was supported in part by a research grant from the CaP CURE Foundation. We thank the Battelle Memorial Institute at Columbus OH, for providing the infusion pump. 121 10 -3 1 c ü 3 u . O X o Û 0.01 0.001 0 24 72 96 Time, hr Figure 6 . 1. Doxorubicin concentration in plasma. Doxorubicin, 2 mg/kg, was administered to dogs intravenously by bolus injection ( • ) or by infusion over 96 hr (■). Data represent mean ± sd (n^S). Some standard deviations are smaller than the size of the symbols. 122 10 B 1 B (/) p O .1 4 144 Time, hr Figure 6.2. Doxorubicin concentration in prostate. Doxorubicin, 2 mg/kg, was administered to dogs intravenously by bolus injection (O) or by infusion over 96 hr (□). A prostate was removed at 48 hr after termination of the 96 hr infusion (0). Data represent mean ± sd or average ± range. Numbers of observations (one animal per observation) are indicated for each time point. 123 2 0 0 • 3 ▲ 1 (0 I 150 iS ■ 4 Q. O 0} 2 100 - Q. o d CO a: 50 - 0 48 96 144 Exposure Time, hr Figure 6.3. Attainment and maintenance of quasi-equilibrium between plasma and prostate concentrations of doxorubicin. Doxorubicin. 2 mg/kg, was administered to dogs intravenously by bolus injection ( • ) or by infusion over 96 hr (■). A prostate was removed at 48 hr after termination of the 96 hr infusion (A). Data represent mean ± sd or average ± range. Numbers of observations (one animal per observation) are indicated for each time point. 124 Parameters Values t,o,„, min 3.2 min 66.9 h n .r hr 109.6 Vd^, L/kg 181.5 Cl, ml kg 'm in ' 25.4±4.2 AUC, pg-ml''min 80.0±12.4 AUMC, pg-ml‘'-min" 580734±225624 MRT, hr 119.5^=31.1 A, ng/ml 3893.2=tI208.0 B, ng/ml 37.5±15.6 C, ng/ml 6 .0±0.6 a, hr ' 13.0±2.3 P, hr-' 0 .6=k0.2 Y, hr-' 0.0063±0.0011 Table 6.1. Summary of pharmacokinetic parameters of doxorubicin in the beagle dog after a 2 mg/kg intravenous bolus administration. t^^p, and the half-lives of a, P, and y phases, respectively. Cl; clearance; AUC: area under the plasma concentration curve, AUMC: area under the first moment curve, MRT: mean residence time, Vd^: volume of distribution at steady state. Data were expressed as mean±sd. 125 CHAPTER? ANDROGEN-DEPENDENT AND -INDEPENDENT HUMAN PROSTATE TUMOR XENOGRAFTS AS MODELS FOR DRUG ACTIVITY EVALUATION 7.1 Introduction Prostate cancer is the most common malignancy in man. Most (>70%) of patients symptomatically respond to initial hormone therapy for two to three years. In spite of hormonal therapy, androgen insensitivity develops in nearly all patients with advanced prostate cancer. The median survival of patients that receive a hormonal therapy is between 1.8 to 3 years (Alivizatos and Oosterhof, 1993; Gudziak and Smith, 1994). Hormone retfactory prostate cancer has a poor prognosis with an overall median survival of 9-18 months, which is not prolonged by currently available treatments (Gudziak and Smith, 1994; Oesterling et al., 1997). Secondary hormonal manipulation may provide symptomatic relief in about 38% of patients and demonstrate a subjective response varied from 40 to 90% in 30% patients, with a limited response duration ranging from 3 to 16 months (Daniel et al., 1990; Matzkin and Soloway, 1992). Systemic chemotherapy produces a response rate that is similar to hormonal therapy (Perez et al., 1989). The low response rate and short-lived response in patients indicates the need of developing more effective treatments. The development of new treatments for hormone refractory prostate cancer has been 126 slow, in part because of the lack of appropriate experimental models. Several series of rat prostate tumor models are available (Dunning, 1963; Isaacs et al., 1986), but their clinical relevance is questionable because of the known difference between rat and human prostate (Newhall et al., 1990). Our laboratories have used histocultures of surgical specimens of prostate tumors from patients to evaluate drug activity (Wientjes et al., 1995; Chen et al., 1997a&b). The major advantages of the histoculture system over monolayer cell culture are the maintenance of 3-dimensional tissue architecture with intact stroma, cell-cell interaction, and inter- and intra-tumoral heterogeneity (Vesico et al., 1987). These characteristics are important for prostate tumor growth because the interaction between tumor and stromal cells is thought to play a role in epithelial growth and response to androgen stimulation (Schroeder and Mackensen, 1974; Mickey, 1988). The clinical relevance of the histoculture model has been shown by Hofl&nan and colleagues in retrospective and semi-prospective preclinical and clinical studies; drug responses in human tumor histocultures correlate with the tumor sensitivity, resistance, and patient survival of head and neck, colorectal and gastric cancer of patients treated with mitomycin C, doxorubicin, 5-fluorouracil, or cisplatin (Robbins et al., 1994; Furukawa et al., 1995; Kubota et al., 1995). The three limitations of human prostate tumor specimens as models for drug activity evaluation are: (a) not renewable sources of tumors, (b) cannot be readily grown in animals and therefore cannot be extended to in vivo studies, and (c ) because they are mainly obtained through radical prostatectomy of locally confined, early stage tumors and are therefore not representative of the advanced disease. Additional experimental models that retain the advantages and are devoid of the limitations of patient tumor histocultures may improve therapy development. 127 There are nine human prostate cancer cell lines and 21 transplantable human prostate xenograft tumors, derived from tumors in primary and metastatic sites (Ellis et al, 1996; Van Weerden et al., 1996; Pretlow et al., 1993). Three o f the xenografts, i.e. CWR22, CWR22R and CWR91, were used in the current study (Pretlow et al., 1993). CWR22 was derived from a Gleason grade 9, Stage D, prostate carcinoma with osseous metastasis. It is androgen- dependent, does not grow in female mice and regresses in male mice after orchiectomy. CWR22R, a sub-line of CWR22 that recurred after regression following androgen withdrawal, is not dependent on androgen and is able to grow in female and castrated animals. CWR22 together with CWR22R is the first pair of human xenografts representing both androgen-dependent prostate cancer and its relapsed androgen-independent counterpart. CWR91 is derived from an Gleason grade 7, Stage C, androgen-independent prostate carcinoma (Pretlow et al., 1993; Wainstein et al., 1994; Nagabhushan et al., 1996). The availability of these three xenograft tumors provides the opportunity to evaluate drug activity in multilayered systems under in vitro and in vivo conditions, and to evaluate the relationship between chemosensitivity and tumor progression to androgen-independent state. The first purpose of the present study was to evaluate whether CWR22, CWR22R and CWR91 tumors are clinically relevant models for drug activity evaluation. Doxorubicin and paclitaxel were used as the test drugs. Doxorubicin produces one of the highest combined partial and complete response rates for single agents of about 33% (Perez et al., 1989). While paclitaxel as a single agent has limited clinical activity in prostate cancer (Roth et al., 1993), more recent data indicate encouraging clinical response for the combination of paclitaxel with estramustine (Hudes et al., 1995). The reason for selecting these two drugs is that the 128 available data showed these drugs produce qualitatively different pharmacodynamics in tumors obtained from patients with locally confined and early stage disease (Chen et al.. I997a&b). For example, doxorubicin produced complete antiproliferation and cytotoxicity (i.e. cell kill) whereas paclitaxel produces only partial cytotoxicity. Similar qualitative differences in the response of xenograft and patient tumors to these two drugs will suggest that the chemosensitivity of xenograft tumors is representative of the chemosensitivity of patient tumors. In addition, comparison of chemosensitivity in CWR22 with the chemosensitivity in CWR22R tumor may suggest the effect of tumor progression to the androgen-independent state on tumor sensitivity to doxorubicin and paclitaxel. The second purpose of the present study is to evaluate the effects of tumor progression from androgen-dependent state to androgen-independent state on chemosensitivity. The expression of the multidrug resistance P-glycoprotein (Pgp), p53, and Bcl-2 in the three CWR xenograft tumors and in patient tumors was studied. The relationship between the expression of these protein markers with chemosensitivity was also compared. These proteins were selected because doxorubicin and paclitaxel are substrates for Pgp (Jachez et al., 1993; Germann et al., 1993; Chen et al., 1994), and because p53 and Bcl-2 play a role in drug-induced apoptosis (Hale et al., 1996). 129 7.2 Materials and methods 7.2.1 Chemicals and supplies Paclitaxel was a gift from Bristol-Myers Squibb Co. (Wallingford, CT), and doxorubicin from Pharmacia (Dublin, OH). Testosterone tablets (12.5 mg each) were purchased from Innovative Research of America (Toledo, OH); Matrigel* from Becton Dickinson Labware (Bedford, MA); sterile pigskin collagen (Spongostan standard) from Health Designs Industries (Rochester, NY); tissue culture supplies (i.e., L-glutamine, sodium pyruvate, fetal bovine serum, gentamicin, Dulbecco's Modified Eagle's Medium (DMEM), Minimum Essential Medium (MEM), MEM nonessential amino acids solution and MEM vitamin solution) from GEBCO Laboratories (Grand Island, NY); ^H-thymidine (specific activity, 65 Ci/mmole) from Moravek Biochemicals Inc. (Brea, CA); NTB-2 nuclear track emulsion from Eastman Kodak Chemicals (Rochester, NY); ApopTag*” In Situ Apoptosis Detection Kit from Oncor Inc. (Gaithersburg, MD); bromodeoxyuridine (BrdUrd), proteinase K, mouse normal IgG and Bicinochoninic Acid Protein Determination Kit from Sigma Co. (St. Louis, MO); mouse monoclonal antibody HB5 against BrdUrd; JSB-1 against Pgp and liquid 3,3'-diaminobenzidine substrate kit from BioGenex (San Ramon, CA); mouse monoclonal antibody ER-PR 8 against human PSA D 07 against p53, Bcl-2 antibody and LSAB universal detection kit from Dako Inc (Carpiteria, CA); and PSA detection kit from Hybritech (San Diego, CA). Al chemicals and reagents were used as received. 7.2.2 Patient tumors Archived specimens of human primary prostate tumors were used for immunohistochemical and Western blot analysis of the expression of PSA Pgp, Bcl -2 and p53 130 (Chen et al., 1997c). The pharmacodynamie data of doxorubicin and paclitaxel in these tumors have been reported previously (Chapters 2 and 4; Chen et al., I997a&b). These tumors were obtained from the peripheral zone of the prostate gland in patients who had organ-confined prostatic adenocarcinoma, during radical prostatectomy. The processing of these tumors was as described elsewhere (Chapters 2 and 4; Chen et al., 1997a&b). 7.2.3 Transplantation of xenograft tumors Male athymic Nu/Nu Balb/C retired breeder mice were purchased from the National Cancer Institute (Bethesda, MD). Three human prostate xenograft tumors (CWR22, CWR22R and CWR91) were kindly provided by Dr. Thomas G. Pretlow at the Case Western Reserve University. The tumors were transplanted in nude mice according to the method previously described (Pretlow et al., 1993; Nagabhushan et al., 1996). Briefly, after micing the tumor using two scalpel blades, tumor fragments were mixed with an equal volume of Matrigel. About 0.3 ml of the mixture was injected subcutaneously into each flank of a mouse, through a 18-gauge neecHe. The host animals for the androgen-dependent CWR22 requires testosterone supplementation. Three days prior to transplantation of CWR22 tumor, animals were subcutaneously implanted with sustained-release testosterone pellets to maintain testosterone levels in the range found in humans (i.e. 6 ng/ml, ref. Doering et al., 1975). The host animals for the androgen-independent CWR22R and CWR91 tumors did not require testosterone supplementation. Three days before the CWR22R implantation, animals were castrated to ensure the androgen-independency of CWR22R. When tumor reached about 1 g in size, in about 1.5 to 2.5 months, they were harvested and used for histoculture. 7.2.4 Histoculture 131 Patient and xenograft tumor specimens were cultured as described previously (Chapters 2 and 4; Chen et al., 1997a&b). In brie( tumors were cut into < 1 mm^ pieces under a sterile condition. Five to six tumor pieces were placed on a 1 cm^ medium-presoaked collagen gel, and incubated at 37°C in a humidified atmosphere o f 95% air and 5% CO,. The culture medium consisted of a 1:1 mixture of MEM and DMEM, 10% fetal bovine serum, 2 mM L-glutamine, 1 mM sodium pyruvate, 40 pg/ml gentamicin, 0.1 mM MEM nonessential amino acids, and concentrated MEM vitamin solution (100 fold concentrated, 10 ml per liter). The histocultures were fed every other day and used for the double labeling and pharmacodynamic studies on day 4. We have shown that testosterone supplement had no effect on the LI of patient tumor histocultures, presumably because these were short-term cultures (Wientjes et al., 1995). Hence the culture medium for CWR22 tumor was not supplemented with testosterone. 7.2.5 Double labeling of ^H-thymidine and BrdUrd CWR xenograft histocultures were simultaneously incubated with 0.03 yuM ^H- thymidine and with 40 pM BrdUrd for 48 hr, washed 3 times with PBS, then fixed in 10% neutralized formalin, embedded in paraffin, and processed for immunohistochemistry followed by autoradiography, respectively to visualize the incorporated BrdUrd and ^H-thymidine as previously described (Chen et al., 1997a; Gan et al., 1996). Briefly, for BrdUrd immunohistochemistry, tissue slides were de-waxed, rehydrated, washed and incubated with the mouse BrdUrd monoclonal antibody (1:250 dilution in PBS containing 5 mg/ml bovine serum albumin (BSA) using the LSAB universal staining kit as described in the next section. For autoradiography, tumor sections on microscope slides after BrdUrd immunostaining were 132 stained with hematoxylin. Slides were then coated with a thin layer of NTB-2 nuclear track emulsion and exposed for 7 to 10 days in the dark in a cold room. Slides were developed, dehydrated and then coverslipped for microscopic examination. ^H-thymidine- and BrdUrd- labeled cells were scored under microscope and the fraction of labeled cells (LI) was determined as previously described (Chapters 2 and 4). The correlation between the two Li's was evaluated. 7.2.6 Immunohistochemistry The expression o f PSA, Pgp, p53 and Bcl-2 proteins were detected by immunohistochemical methods as previously described (Gan et al., 1996a). To minimize data variability due to intratumoral heterogeneity, adjacent sections of tumors were used to simultaneously detect these markers. Briefly, de-waxed and rehydrated tissue sections were boiled in a 0.1 M citrate buffer (pH 6.0) for 15 min, washed in phosphate buffered saline, and incubated with DAKO blocking solution for 10 min and subsequently with one of the following antibody solutions for 2 hr, i.e. PSA antibody (1:100 dilution), p53 antibody (D07, 1:100 dilution), Bcl-2 antibody (1:50 dilution), and Pgp antibody (1:100 dilution) in PBS containing 5 mg/ml bovine serum albumin. After applying biotin-anti-IgO linker solution and peroxidase-conjugated streptavidin solution, tissue sections were incubated for 3 to 7 min with 3-3'-diaminobenzidine and hydrogen peroxide, and then counterstained with hematoxylin. The negative controls used mouse IgG as primary antibody. 7.2.7 Detection of Pgp, p53, Bcl-2, and PSA by Western blot analysis The levels of Pgp, p53, Bcl-2 and PSA in CWR xenografts were analyzed by Western blotting, as previously described (Gan et al., 1996a). Breitly, tumors were homogenized and 133 DNA was sheared by passing through a 21 gauge needle. Cell debris were removed by centrifugation, and the protein concentration in the supernatant was measured by the Bicinochoninic Acid Protein Determination Kit (Sigma Chemical Co.). Twenty-five ng of protein was loaded on a 7.5% (for Pgp detection), 10% (for p53), or 12% (for Bcl-2 and PSA) polyacrylamide slab gel and then transferred onto a nitrocellulose filter by elecro-blotting. After blocking the nonspecific binding sites with 5% skim milk in Tris-buffered saline containing 0.1% Tween 20 (TEST), the filter was incubated with antibody in TEST containing 2% BSA. The antibodies for Pgp (JSB-1, 4E3, and ab- 1), p53 (DO-7) and PSA were diluted 750 fold, and the Bcl-2 antibody was diluted 250 fold in PBS containing 5 mg/ml BSA. The linker solution and the peroxidase-conjugated streptavidin (1:5 dilution in TEST containing 1% BSA) were applied sequentially. The signal was developed using the Amersham ECL chemiluminescence method. 7.2.8 Pharmacologic effects of paciitaxel and doxorubicin Drug effects in the three xenograft tumors were studied. Tumor histocultures were exposed to paciitaxel (0.00012 to 12 pM) for 24- and 96-hr and to doxorubicin (0.00017 to 17 pM) for 96 hr. Paciitaxel was first dissolved in ethanol and then transferred to the culture medium. The final ethanol concentration in the medium was 0 . 1%. Same concentrations of ethanol was added to the controls. Doxorubicin was dissolved in distilled water as a 10 mg/ml stock solution. After drug treatment, tumors were washed three times with 5 ml of drug-free medium, each for 10 min and incubated in medium containing 40 pM BrdUrd for 48 hr. Tumor tissues were then fixed in 10% neutralized formalin, embedded in paraffin and cut into 5 pm sections using a microtome. Controls were processed similarly, with the exception of 134 drug treatment. Drug-induced antiproliferation was measured by inhibition of BrdUrd labeling. BrdUrd-labeled ceils were detected by immunohistochemical method using the LSAB kit. The BrdUrd-labeled tumor cells were scored under microscope, and the LI was determined as above described. A typical experiment used a total of 8 to 15 tumor pieces for each drug concentration. A minimum o f 200 cells per piece, or >1,600 cells, was counted per concentration. For tumor specimens treated with the highest doxorubicin concentration, all tumor cells survived were counted. Drug-induced cytotoxicity was measured by counting terminally damaged cells, which were identified by morphological changes and by the TUNEL assay. The morphologies of apoptotic cells included chromatin condensation and margination, membrane blebbing, apoptotic bodies and cell shrinkage (Kerr et al., 1994). Necrotic morphologies included cytoplasmic vacuolation, cell swollen, loss of membrane integrity, and gross cytolysis. Apoptotic cells are usually scattered throughout the tumor fragment, whereas necrotic cells usually appear as groups o f adjoining cells (Weedon et al., 1979). The TUNEL assay labels fragmented DNA and therefore labels both apoptotic and necrotic cells. TUNEL was performed according to manufacture’s instructions for ApopTag and as described as below. 7.2.9 Terminal deoxynucleotidy transferase (TdT)-mediated dUTP nick end labeling (TUNEL) assay Procedures were carried out as described in the ApopTag* detection kit. Tissue sections on slides were deparaffinized, rehydrated and incubated with proteinase K (20 pg/ml) in PBS for 15 minutes at room temperature followed by 4 washes with distilled water for 2 135 min each. The protease treatment is necessary to make DNA fragments accessible for reaction with enzyme and substrate. End7ogenous peroxidase was quenched by incubating tissue sections with 2% hydrogen peroxide in PBS for 5 min at room temperature followed by 2 washes with PBS for 5 min each. Equilibration buffer from the kit was applied directly onto tissue sections and incubated for 10 min at room temperature. TdT-containing solution from the kit was then applied and incubated in a humid chamber at 37°C for 1 hr. After the enzyme reaction, slides with tissues were incubated with 37°C-prewarmed Stop solution for 10 minutes at room temperature followed by 3 washes with PBS for 5 min each. Anti- digoxigenin-peroxidase from the kit was applied and incubated for 30 min at room temperature followed by 3 washes with PBS for 5 min each. Brown deposits were developed by incubating the tissue sections with chromogen 3,3 '-diaminobenzidine and the substrate, hydrogen peroxide, for 3 to 7 min. The volume of applied liquid was enough to cover the whole section. After each step, excess liquid was tapped off gently and the tissue was blotted around carefully. The slides were kept in a humid chamber so that no concentration changes occurred due to the loss of water during the incubation periods. The tissue sections were counterstained with hematoxylin, dehydrated and coverslipped. The TUNEL-positive cells were scored. 7.2.10 Quantitation of PSA secretion Reduction of PSA secretion is a clinically relevant and widely used objective criterion for monitoring patient response (Oesterling et al., 1997). Drug-induced reduction of PSA secretion from CWR tumors was measured. After incubating tumors with drugs and BrdUrd, aliquots of culture media of control and treated samples were collected. The PSA 136 concentrations in the medium were measured by a sandwich-immunoassay using two antibodies against two different epitope sites on PSA, performed by Immunology Laboratory of the James Cancer Hospital (Columbus, OH). 7.2.11 Data analysis The drug concentration-effect relationships were analyzed using Equations 7 . 1 and 7.2 for antiproliferation and cytotoxicity, respectively. C ” ^antiproliferatton ^max ^^ ^ (100 ^m w c^ (EQ- 7.1) Kf + C" ^cyro,ox,c.n- = (Eq. 7.2) where 'S the LI of doxorubicin-treated specimens expressed as a percent of the LI from the untreated control, C is the drug concentration, 6 ^^ is the maximum drug effect, K, and K, are the drug concentrations at 50% effect, n is a curve shape parameter, is the fraction of TUNEL-labeled cells, and basal is the fraction of dead cells in untreated control. Values for IC 137 76% (Table 7.2) and the basal term to correct for the fraction of dead cells in untreated controls. 7.2.12 Statistical analysis Differences among groups were detected using analysis of variances. Differences between groups were analyzed using the Mann-Whitney U-test or t-test. The correlation between the ^H-thymidine LI and BrdUrd LI was also performed. Softwares for statistical analysis (ANOVA, TTEST, NPARIWAY, and REG procedures) were developed by SAS (Cary, NC). 138 7.3 Results 7.3.1 Double labeling and the correlation between ^H-thymidine and BrdUrd LI Figure 7 .1 (A) shows the labeling of tumor cells simultaneously by BrdUrd and thymidine. Figure 7 .1 (B) shows the correlation between the BrdUrd LI and ^H-thymidine LI in three CWR91 tumors. The linear regression line was BrdUrd LI = 0.96 * (^H-thymidine LI) + 0.004 (r^ 0.996). This indicates that BrdUrd LI is identical 4o H-thymidine LI. Because the ^H-thymidine labeling technique requires the use of radioactive ^ H-thymidine, which is expensive and environmentally undesirable, subsequent studies use the BrdUrd labeling methods to monitor DNA synthesis. 7.3.2 Histocultures of xenograft tumors Figure 7.2 shows the micrographs of histocultures of CWR22, CWR22R, and CW R91 tumors. The histocultures maintained the 3-dimensional structure, showed presence of epithelial tumor cells and normal stromal (fibroblasts and muscle cells), and retained scattered acinus-like structures in CWR22 tumors. The BrdUrd LI of the xenograft tumor histocultures was in the following rank order: CWR91 > CWR22R > CWR22 (Table7. 1), i.e. the clinically occurred androgen-independent tumors (i.e. CWR91) was the highest followed by the relapsed androgen-independent CWR22 counterpart (i.e. CWR22R), and the LI in the androgen-dependent CWR22 tumor was the lowest (p<0.001 between xenografts). The LI in histocultures derived from xenografts obtained from 7 successive generations were not significantly different (p>0.1, ANOVA). The coefficient of variation of LI was smaller (<20%) in xenografts than that (-30%) in heterogenous patient tumors, indicating small inter generation differences during successive transplantation. Conversely, the large differences in 139 the LI among androgen-dependent and -independent xenografts tumors suggest biological difierences resulting in higher proliferative fractions in the androgen-independent tumors. The LI in the histoculture system is higher than the proliferation index of between 1.6 to 16% found in other studies that used pulse labeling with DNA precursors for 5 min to 1 hr or a snapshot measurement of the expression markers such as proliferation cell nuclear antigen or Ki-67 (Pretlow et al., 1994). The higher BrdUrd LI in histocultures is likely due to the 48 to 500-fold longer labeling period compared to the short-term labeling studies (Nemoto et al., 1989; Scrivner et al., 1991). 7.3,3 Expression of Pgp, PSA, p53 and Bcl-2 proteins Patient tumois were analyzed for expression o f p-glycoprotein, PSA p53, and Bcl -2 proteins by immunohistochemistry, whereas the three CWR xenografts were analyzed by both immunohistochemistry and Western blotting. The immunohistochemical results in membranous and cytoplasmic immunostaining for Pgp, PSA and Bcl- 2, and nucleus immunostaining for p53 protein (Figure 7.2). The discrete nucleus p53 staining enabled the determination of the fraction of cells showing different staining intensity. .A1 three CWR tumors and 85% of patient tumors showed immunostaining for Pgp, with a homogeneous staining in the less differentiated tumors and a localized staining at the luminal side of apical cells in highly differentiated tumors. Sixty % of patient tumors did not shown immunostaining for Bcl-2 protein, whereas 40% were stained positive for Bcl-2. In comparison, Bcl-2 was not detected immunohistochemically and was detected only by Western blot analysis in all three xenograft tumors with the highest expression in CWR22R tumor. This indicates that a lower Bcl-2 140 expression in the xenograft tumors as compared to 40% of patient tumors. CWR22 and CWR22R tumors contained <10% scattered p53-positive cells. In contrast, CWR91 tumor did not show p53-positive cells by immunohistochemical staining and showed only a faint band by Western blot analysis. The majority of patient tumors (-90%) contained < 10% p53-positive cells, which is comparable to the results in the three xenograft tumors. Immunohistochemical detection of PSA in CWR22 tumor was achieved with a lower antibody concentration (dilution 1:100) compared to that in CWR22R and CWR91 tumors (dilution 1:20), indicating a higher PSA secretion in CWR22 tumor. This is consistent with the rank order o f the signals in the Western blot (Chen et al., 1997c), the PSA concentrations in the culture media of untreated controls (i.e. 33±14 ng/ml, n= 8 , in CWR22 tumor, 4.4±1.3 ng/ml, n= 6, in CWR22R tumor, and below the detection limit o f 0.03 ng/ml in CWR9I tumor), and the rank order of the reported plasma PSA concentrations in mice bearing these xenograft tumors (Wainstein et al., 1994). The PSA concentrations in the culture medium o f the three xenograft tumors are comparable to our reported PSA concentrations in the culture media of patient tumors (i.e. range of 0.03 to 48.8 ng/ml and average of 12.3 ng/ml, ref. Chen et al., 1997a&b). 7.3.4 Drug efTects in xenografts tumors Figure 7 .1 (c) shows control BrdUrd LI in CWR22 and CWR91 tumors and the drug- induced TUNEL-labeled CWR22R tumor cells. In all three CWR xenograft tumors, paciitaxel and doxorubicin produced antiproliferation, reduction in PSA secretion and cytotoxicity; the three effects are in descending order with respect to both drug potency and (Table 7.2). 141 For example, 0.1 doxorubicin produced -50% antiproliferation and s5% cytotoxicity. Likewise, 0.1 pM paciitaxel produced 30 to 70% antiproliferation and<30% cytotoxicity. Figure 7.3 shows the sigmoidal, concentration-dependent antiproliferation induced by doxorubicin and paciitaxel. The most notable difference among the two drugs is the different extent of In all three xenograft tumors, doxorubicin produced 100% antiproliferation, whereas paciitaxel treatment for 24 or 96 h produced a maximal inhibition of -50% with no further increase even when drug concentrations were increased to 12 pM. A comparison of the IC 5,, values shows a higher molar potency for doxorubicin in CWR22 and CWR22R tumors compared to paciitaxel, whereas CWR91 tumors were more sensitive to paciitaxel. Figure 7.4 shows the cytotoxicity induced by doxorubicin and paciitaxel in the three xenograft tumors. Doxorubicin produced concentration-dependent and nearly complete cytotoxicity, whereas paciitaxel produced incomplete effect. Untreated controls showed an average of 1% TUNEL-labeled cells. A comparison of the IC 50 values shows a higher molar potency for doxorubicin in all three xenograft tumors compared to paciitaxel. Drug-induced reduction of PSA concentrations was studied in CWR22 and CWR22R tumors but not in CWR91 tumors because the PSA concentration of the latter tumor could not be quantified by immunoassay. Doxorubicin produced concentration-dependent and nearly complete elimination of PSA concentrations in CWR22 and CWR22R tumors (Figure 7.5). In contrast, treatment with paciitaxel for 96 hr produced incomplete inhibition of PSA secretion in CWR22 tumor (Figure 7.5), and highly variable PSA concentrations in CWR22R tumor (data not shown). In general, the extent of drug-induced reduction in PSA levels are consistent with the extent of drug-induced antiproliferation and cytotoxicity. 142 7.3.5 Comparison of drug effects in xenograft and patient tumors Figures 7.3 and 7.4 compare the chemosensitivity of the three xenograft tumors with our previous data on the antiproliferative and cytotoxic effects of doxorubicin and paciitaxel in histocultures of human patient tumors obtained via radical prostatectomy (Chen et al., I997a&b). For doxorubicin, the antiproliferative effect in the three xenograft tumors and the cytotoxic effect in the two androgen-independent tumors are nearly identical with its effects in patient tumors, whereas the CWR22 tumor was 2-3 fold less sensitive than patient tumors to drug-induced antiproliferation and cytotoxicity. For paciitaxel, the antiproliferative effect in CWR22 and CWR22R tumors and the cytotoxic effect in CWR22 and CWR22R tumors are nearly identical to its effects in patient tumors (Figures 7.3 and 7.4), whereas CWR9I tumor is several fold more sensitive than patient tumors to the antiproliferative effect, and the maximum drug-induced cytotoxicity in CWR9I tumors is 2.5 times that in patient tumors. 143 7.4 Discussion The heterogeneity in patient tumors limits the likelihood of finding experimental models that precisely predict the chemosensitivity o f all patient tumors. Our initial goal is to develop models that represent a majority of patient tumors. The present study was to determine whether the three human prostate xenograft tumors, CWR22, CWR 2 2 R and CWR91, represent clinically relevant models for drug activity evaluation. Our results show that the three xenograft tumors have proliferation status and expression of PSA, Pgp, Bcl-2 and p53 proteins that are consistent with a majority of patient tumors. More importantly, the sensitivity of the three xenograft tumors to doxorubicin and paciitaxel are qualitatively similar to the median response of patient tumors. For example, in all three xenograft tumors and patient tumors, doxorubicin produced complete antiproliferation and cytotoxicity whereas paciitaxel produced incomplete effects. Furthermore, a comparison of the concentration-effect relationships in xenograft and patient tumors indicates that the chemosensitivity observed in patient tumors is represented by the chemosensitivity of one or more of the three xenograft tumors, as follows; (a) the three xenograft tumors and patient tumors responded equally to doxorubicin-induced antiproliferation, (b) CWR22R, CWR91 and patient tumors responded equally to doxorubicin-induced cytotoxicity, whereas CWR22 was >3 fold less sensitive, (c) CWR22, CWR22R and patient tumors responded equally to paclitaxel-induced antiproliferation, whereas CWR91 was several fold more sensitive, and (d) CWR22, CWR22R and patient tumors responded equally to paclitaxel-induced cytotoxicity, whereas CWR91 tumors was 2.5 times the maximum cytotoxicity in that in patient tumors. The similar chemosensitivity between the xenograft tumors and patient tumors is noteworthy because 144 other more commonly used experimental models such as monolayer cultures of human prostate cancer PC3 and DU145 cell lines typically show 50% effective drug concentration that are several orders of magnitude lower than the IC 50 and LC 50 values in patient tumors (i.e. 4 or 30 nM in monolayers, ref. Li and Au, 1997). The second goal of the present study was to compare the effects of tumor progression to the androgen-independent state on chemosensitivity. Such comparison in patient tumors has not been possible because advanced prostate tumor samples are not readily available. Our results show that progression of the androgen-dependent CWR22 tumor to androgen- independent CWR22R tumor resulted in no change in Pgp expression, over-expression of p53 and Bcl-2, and no change in tumor sensitivity to the antiproliferation effect but a higher sensitivity to the cytotoxic effect of doxorubicin and paciitaxel. In addition, the CWR91 tumor was also more sensitive than CWR22 tumor to the antiproliferative and/or cytotoxic effects of these two drugs. The higher drug-induced cytotoxicity in CWR22R and CWR91 tumors compared to CWR22 tumor was further confirmed in a separate study using two new agents, i.e. cytochalasin E and geldanamycin (unpublished results). The enhanced drug- induced antiproliferation and cytotoxicity in the two androgen-independent tumors compared to the androgen-dependent tumor dispels the general impression that androgen-independent tumors are less sensitive than androgen-dependent tumors to chemotherapy. This finding supports the use of chemotherapy to eliminate the small subset of androgen-independent tumor cells, in the early course of the disease rather than after patients have failed androgen ablation therapy. There are no apparent cause-and-effect relationships between p53 and Bcl-2 145 overexpression and enhanced drug-induced cytotoxicity. In fact, a comparison of the p53 and Bcl-2 levels among the three xenograft tumors indicate no consistent relationship between the expression of these proteins, androgen-dependency, and tumor sensitivity. For example, CWR91 had the lowest levels of p53 and Bcl-2 among the three tumors, yet it is equally or more sensitive than CWR22 tumor to the antiproliferative and cytotoxic effects of doxorubicin and paciitaxel. A separate study in our laboratory, which examined the relationship between the expression of p53 and Bcl-2 and the paciitaxel sensitivity in about 100 bladder, breast, head and neck, ovarian and prostate tumors from patients, indicates no relationship between these parameters (Gan et al., 1996a and unpublished results). An additional noteworthy finding is the differential sensitivity among the three xenograft tumors to doxorubicin and paciitaxel. This suggests that these two drugs exert dififerent effects on different subsets of prostate tumors, and raises the interesting possibility that patients failing one drug may respond to the other and that combination o f the two drugs may increase the fraction of responding tumor cells and thereby enhance the response rate. In summary, the results of the present study support using the three xenograft tumors as renewable sources of human prostate tumors for in vitro and in vivo drug activity evaluation. In addition, because the immunohistochemical results suggest that the three CWR xenograft tumors express a lower level of Bcl-2 compared to 40% of patient tumors, development of additional xenograft tumors that express a higher level of Bcl-2 may be warranted. 146 7.5 Acknowledgments This study was supported in part by research grants RO1CA63363 and ROICA74179 from the National Cancer Institute, NIH, DHHS, an award from the CaP CURE Foundation, and a grant from the Bristol Myers Squibb, Co. Thanks to Dr. Thomas G. Pretlow for providing the three prostate xenograft tumors and to Jane Cai, Jeff Johnston and Dr. Peter fCoo for maintaining the xenografts and to special thanks to Dr. Yuebo Gan for his generousness of sharing the related results. 147 l\SA licl-2 p53 Tumors l.l, % iiiununoslaining Wcsleni immunostuiniiig Western Immunostuining W estern immnnosiiimmg W estern (lllCUDlSl)) Inlciisily F. % Intensity intensity F. % Intensity intensity F. % intensity Intensity & % F.% littenstty stained eclis PuticiK 42±12* +/++ too NA 15 NA 60 NA - & 0% I t s NA (n= 26) + 85 +/-H- 40 t & <10% 77.0 +++ & >40% tt.5 Xenogralt tumors CW R22 35±5* ■H-r too -H-++ + too ++ - 0 +/-H- + & 3 4% too +/++ (n= 36) CW K22K 5 0 ± I0 * +/++ too ++/+++ + too ++ - 0 +++ + &7,t% too + + /+ 1 ’> (n=31) 00 CWIWI 63±8* + too + + too + - 0 + - & 0% too + (n=32) Table 7.1. Expression of PSA, Pgp, Bcl-2 and p53 proteins in patient and xenograft tumors. Expression of proteins in archived patient tumor slides was detected using immunohistochemical methods. Protein expression in xenograft tumors was also analyzed by Western blotting of tissue extracts. The membranous and cytoplasmic staining of PSA, Pgp, Bcl-2 proteins prohibited the scoring of individual stained cells, hence the tumors were scored as either positively or negatively stained and the data was expressed as fraction (F) of tumors stained positive for these proteins. The discrete nucleus immunostaining of p53 enabled the scoring of the cell fractions with different staining intensity. Results were obtained from our previous study (Chen et al., 1997c). * Differences in the LI between patient tumors and the three xenograft tumors are significant (p<0.05). Tutnurs and dtug Antiprulifcration Cvtotoxicitv PSA icduction treatments IC,„, nM PP l.C,u. nM PP 1C,„. ,iM 1 ». (X)xotubicin (96iir) Patient (11^17) 6 1 ,7 .) lOOiO 2I24H 256 100,0 NA NA CW R22(n-5) I5 U 4 0 0.01 lOOiO 622411023 -0 0 1 80i4 -0(101 282412464 87,4 CWR22R (n-5) 62,15 0.37 101)10 1050,430 007 00i2 0007 7211333 0611 CWR9I (n-sl 7llt24 0 2 0 lOOlO 8221405 0 0 3 0012 0007 NA NA Paciitaxel (24 hr) Patient (n-26) ■III pM (0.2 >12.00(1) 4 7 H 9 12 pM 12,4 NA NA CWR22 (n-5) K‘J‘t (202 -12,000) 0.57 40110 0.68 •12 pM 8 ,2 0 0 3 >12 pM 15,0 CWK22K(ii-5) -12 (iM 0.05 4316 0.64 12 (,M 10,5 -0 01 NA NA \0 CWR9I (n-4) )76 (82 ->12.000) 0 0 7 5018 0.20 -12 pM 20,2 -0.01 NANA I'aelitaxel (% hr) I'atient ( i f 11 ) 7.) (1.5 >12.000) 70112 •12 pM 12,3 NA NA C W R22(n-5) 185(100 >12.001)) 0.04 5 8 ,8 0.05 -12 pM 0 ,3 1)08 >12 pM 36i|3 CWR22R(n-5) 5)8(2) 12.000) 0(18 56,12 0.04 12 pM 15,4 0 1 NA NA L'WKVI (n- 4) 26,12 024 76,4 0 3 6 l2pM 30,3 •0 01 NA NA Table 7.2. Sensitivity of iiatient and xenograft tumors to doxorubicin and {laclituxel. Drug acliviiy in xciuigratt (uiiiors was cslablislicd in the present study. Drug activity iitpaticiu tumors were obtained from our previous studies (Chen ct al., IW7a&b). Data represent mean±SD with tlie exception tliat for like paclitaxel-induced aiuiprolitcralioti iti patient tutnurs, the Jala represent median with ranges in parentlicses. tttaxittiutn drug ctfeci at the liigliesi drug concentration (4 to 12 pM ). 80 intercept = 0.004 slope = 0.96 r ^ = 0.996 60 D E Cû 40 40 60 80 3 H-Thymidine Ll, % Figure 7.1. Corrélation between tumor cells labeling by BrdUrd and ^H-thymidine. A. Double-labeled CWR91 cells with BrdUrd in brown and ^H-thymidine in black dots as indicated by arrowheads (40Qx). B. Linear regression line: BrdUrd LI = 0.96*(^-thymidine LI) + 0.004 (r^O.996, p <0.001), observed values and 95% confidence interval. 150 en Figure 7.2. BrdUrd labeling; TUNEL assay; and PSA, p53 and Pgp immunohistochemistry. Untreated CWR22 (a, 160x) and CWR91 (b) with BrdUrd-labeled cells, CWR22R with TUNEL-staining cells (c), CWR22 with PSA (d), CWR22R with p53 (e) and CWR91 with Pgp (f) immunostaining were illustrated (400x). Acinus-like structures in CWR22 (a) as arrows indicated. Doxorubicin oe hr) CWR22R C^R22 y CWR91 Human C 0 0.001 0.1 10 o Ü Paciitaxel 24 hr 96 hr 120 I- CWR22R CWR22R -CWR22 I —CWR22 Human Human CWR91 CWR91 I I L 0.001 0.1 10 0.001 0.1 10 Drug Concentration (jxM) Figure 7.3. Concentration-dependent antiproliferation induced by paciitaxel and doxorubicin. CWR22 (•), CWR22R (O) and CWR91 (□) tumor histocultures were treated with paciitaxel for 24 or 96 hr and doxorubicin for 96 hr. Inhibition of cumulative BrdUrd LI was expressed as percent of untreated controls. Results in xenograft tumors were obtained in the present study. Results in patient tumors ( 0 ) were obtained from results in Chapters 2 and 4. The sigmoidal lines are computer-fitted lines according to Equation 7.1. Data represent mean ± SD. 152 Doxorubicin (96 hr) 1 0 0 Human CWR91 CWR22R CWR22 40 0 0.001 0.1 10 CO Q) Paciitaxel Q 24 hr 96 hr (D 40 -1 40 CWR91 O CWR91 CWR22R 20 - CWR22R 20 Human CWR22 \ Human 0 CWR22 0 0.001 0.1 10 0 0.001 0.1 10 Drug Concentration(| jM) Figure 7.4. Concentration-dependent cytotoxicity induced by paciitaxel and doxorubicin. CWR22 (•), CWR22R (O) and CWR91 (□) tumor histocultures were treated with paciitaxel for 24 or 96 hr and doxorubicin for 96 hr. Drug-induced death fraction of tumor cells was expressed as the cytotoxic effect. Results in xenograft tumors were obtained in the present study. Results in patient tumors ( 0 ) were obtained from results in Chapters 2 and 4. The sigmoidal lines are computer-fitted lines according to Equation 7.2. Data represent mean ± SD. 53 120 -I O 4—' C o 80 - ü ^ Paciitaxel "o 40 - CWR 22 CWR 22R CL • Doxorubicin - y / —\— 0 0.001 0.1 10 Drug (|jM) Figure 7.5. Concentration-dependent inhibition of PSA secretion. CWR22 tumor histocultures were treated with paciitaxel for 96 hr (O) and with doxorubicin for 96 hr (•); CWR22R was treated with doxorubicin for 96 hr ( 0 ). PSA levels in culture medium collected at 48 hr after drug treatment were determined. The sigmoidal lines are computer-fitted lines according to Equation 7.1. Data represent mean ± SEM. 154 CHAPTER 8 CONCLUSIONS AND PERSPECTIVES The research in this dissertation has made several contributions in the development of effective treatment of prostate cancer, as follows, (a) The establishment of pharmacodynamics of paciitaxel (Chapter 2) and doxorubicin (Chapter 4) in histocultures of patient and three human xenograft tumors. These data indicate that paciitaxel and doxorubicin is active against early state and androgen dependent prostate cancer, with the antiproliferative effect being more prominent than the cytotoxic (i.e., cell kill) effect. It is noted that these results are the first to delineate the two effects separately, (b) The apoptotic effect of paciitaxel in patient tumors in their native 3-dimensional sketch is cell cycle specific (Chapter 3). These results confirm the previously reported cell cycle specificity of paclitaxel- induced apoptosis in monolayer cultures of human cancer cells, and indicate that the microenviornment of solid tumors do not alter this property of paciitaxel effect, (c ) the concentrations of doxorubicin delivered to prostate tumor in dogs, after an intravenous injection, are insufficient to produce >50% antiproliferation and cytotoxicity, respectively, in patient tumors (Chapters 5 and 6). These results demonstrate that insufficient drug delivery to tumors is one cause of the lack of clinical response, and have led to the development of 155 regional therapy where the drug is delivered in high concentrations directly to the prostate, (d) The CWR22, CWR22R and CWR91 human xenograft tumors show pharmacodynamics of paciitaxel and doxorubicin that are qualitatively and quantitatively similar to those in patient tumor (Chapter 7). It is noted that these models are the first to show responses to drugs as patient tumors; other models such as monolayer cultures of human cancer cells overpredict chemosensitivity by >10 fold. These results establish the three xenograft tumors as clinically relevant models for drug activity evaluation. Furthermore, the data on paciitaxel and doxorubicin pharmacodynamics show that drug-induced antiproliferation occurred at a lower concentration compared to drug-induced cytotoxicity, indicating antiproliferation as their major effect. Because prostate cancer is slow growing, development of drugs that have cytotoxicity as major effect is warranted. Other studies in our laboratory have identified geldanamycin as an agent that shows cytotoxicity as its major effect in the androgen- independent (CWR22R and CWR91) prostate xenograft tumors. The research in this dissertation further demonstrate the value of pharmacokinetics and pharmacodynamics studies in the development of effective drugs. For example, research of these studies demonstrate that intravenous injection does not deliver effective drug concentrations to the prostate, thus suggesting the novel approach of regional therapy which is currently being researched in our laboratory. In addition, the separate measurement of the two drug effects, i.e. antiproliferation and cytotoxicity, enables the distinction and the specificity of the two effects, and consequently the identification of drugs that specifically produce the desired effect (e.g. cytotoxicity for slow growing tumors and antiproliferation for rapidly growing tumors). 156 APPENDICES DATA FOR FIGURES AND TABLES IN CHAPTERS 157 APPENDIX A DATA RELEVANT TO CHAPTER 2 158 i Tumor 10 (|ig/ml) 1 0.1 0.01 0.001 0.0001 24 hr treatment 1 33.7±I4.3 14.146.5 30.543.8 73.5414.8 66.0415.4 85 1415 6 2 50.44:20.7 20.646.8 17.347.6 36.848.5 82.7436.8 121.2417.1 3 32.3x13.6 38.0411.7 39.5411.2 88.2417.6 91.0418.7 NM 4 36.1x5.6 46.6411.3 35.446.8 68.447.6 81.8413.8 93.448.3 5 41.04:4.3 37.345.8 44.146.5 58.949.1 78.245.0 NM 6 43.5±11.9 37.7412.8 52.348.6 51.5410.7 71.648.7 84.8412.7 7 33.04:8.3 44.3413.4 30.847.1 51.747.5 37.8410.8 72.0414.1 8 50.34:7.4 36.748.2 34.445.1 54.846.4 55.445.9 NM 9 60.14:17.8 33.948.1 40.5410.3 52.1416.1 136.3416.0 83.7418.0 10 43.34:6.8 75.6412.2 65.9414.1 NM 104.7419.0 142.6422.1 11 44.94:7.0 47.749.3 53.147.5 85.3412.5 119.6416.4 103.4413.6 12 54.54:16.4 44.647.2 53.540.9 54.3410.7 74.447.8 NM 13 43.2±6.4 45.145.0 59.246.9 75.846.2 79.749.0 80.348.9 14 52.4±10.8 37.646.7 56.7411.8 64.046.7 89.749.3 NM 15 55.1±10.8 50.649.7 48.249.0 94.8420.3 56.0424.4 140.7423.4 16 64.64:8.4 52.0411.3 68.448.0 110.9418.1 92.5412.3 NM 17 57,44:8.4 68.2410.9 64.248.6 96.4413.8 82.3413.5 115.4411.9 18 60.54:10.5 50.645.5 79.9410.1 54.9412.7 85.049.9 75.4411.9 19 77.84=14.4 64.9411.0 90.3411.9 89.148.7 NM 102.049.5 20 67.94:11.4 52.349.8 68.0412.6 75.8417.5 67.7411.3 NM 21 69.04:8.7 84.2411.5 77.649.4 65.3413.7 83.4418.0 NM 22 73.84:8.1 NM 75.1410.1 70.748.4 85.046.2 94.745.2 23 54.24:9.2 95.649.2 71.949.7 81.2412.5 115.1418.0 107.0429.8 24 81.24Ü8.9 80.2434.4 93.3416.9 50.6410.9 88.4 NM 25 70.44:6.8 85.946.6 84.64.4 79.446.2 113.148.4 NM 26 125.94:12.1 83.2424.6 89.9421.5 48.6410.3 81.3421.5 NM 96 hr treatment 27 24.94:3.8 55.7410.0 32.547.2 61.346.4 79.249.3 73.146.5 28 21.64:5.2 16.042.5 44.948.0 91.7414.8 68.2414.5 101.8414.7 29 22.24:6.5 19.743.0 32.345.4 NM 90.9414.1 99.5410.9 30 20.7410.9 45.2419.6 37.645.4 56.6415.1 64.3417.7 89.4418.5 4 35.744.1 33.046.5 38.046.7 65.145.8 97.647.8 89.4411.2 31 24.4412.6 37.3418.8 65.3418.4 68.7423.4 36.847.7 NM 19 65.5411.8 99.349.6 92.8412.1 95.249.8 NM 101.9415.9 32 36.243.7 48.346.5 48.046.4 69.245.8 73.846.0 108.347.8 33 39.946.9 53.1415.0 33.845.8 35.847.6 64.549.7 105.5418.9 15 55.7416.0 34.448.4 58.8412.8 79.4411.2 78.249.0 133.7410.9 34 7.1 47.4410.8 64.149.0 96.8413.7 104.2421.9 NM Table A I Raw data of LI (mean^SEM) for individual tumors in Table 2.1. NM; not measured. 159 Tumor 10 (ng/ml) 1 0.1 0.01 0.001 0.0001 24 hr treatment 1 6.4±4.0 6.74:2.7 4.24:4.1 0.740.4 0.040.0 NM 2 4 .3 il.7 3.04:1.3 6.94:2.8 5.142.7 0.040.0 0.240.2 3 9.0±2.9 14.04:4.8 13.142.8 MM 0.340.3 0.040.0 4 8.2±2.1 9.44:1.6 3.241.4 0.540.4 1.540.7 0.040.0 5 9.3±1.6 5.5 5.741.5 4.542.4 0.840.8 NM 6 12.0±5.5 15.64:3.0 17.442.6 0 .0± 0.0 0.040.0 0.040.0 7 11.1±3.3 14.44:3.1 7.343.3 1.7±1.3 2.541.5 0.040.0 8 7.3±1.6 3.7±1.5 3.241.2 1.140.8 0.040.0 NM 9 10.74:3.4 13.34:2.0 15.443.7 10.342.3 0.040.0 0.040.0 10 11.54:4.2 18.14:4.4 5.941.8 NM 3.942.3 0.040.0 11 12.54:1.1 15.84:3.7 14.343.7 6.242.2 0.740.7 0.040.0 12 7.94:4.1 7.34:2.2 3.143.2 3.242.0 3.342.6 NM 13 11.54:1.3 9.84:1.4 6.140.9 2.840.8 0.540.3 0.340.2 14 8.34:1.8 7.44:2.5 7.942.5 3.841.2 0.540.5 NM 15 9.04:1.8 12.04:3.0 6.940.6 0.140.1 0.440.4 0.040.0 16 6.14:2.9 10.34:1.4 12.242.5 7.842.4 1.540.7 NM 17 12.04:1.5 19.946.2 14.042.4 1.140.7 0.040.0 0.040.0 18 13.94:2.5 14.4±1.6 7.742.3 0.040.0 1.941.1 0.040.0 19 10.74:3.7 12.14:1.6 4.741.4 2.440.7 2.440.9 NM 20 8.64:1.0 8.244)9 7.042.0 0.540.5 0.040,0 NM 21 17.04:3.0 15.54:4.4 0.040.0 0.540.4 0.040.0 NM 22 11.74:2.6 5.34:2.9 2.041.3 0.040.0 0.040.0 NM 23 17.74:3.8 19.84:2.7 0.440.4 1.040.6 0.040.0 0.040.0 24 10.74:3.6 6.8 8.4 6.440.2 0.040.0 NM 25 10.54:1.9 11.34:2.0 3.541.2 1.941.0 0.040.0 NM 26 7.2±1.5 3.84:1.4 0.840.8 0.040.0 0.040.0 NM 96 hr treatment 27 9.244).9 10.74:2.1 10.541.5 0.140.1 0.040,0 0.040.0 28 18.24:2.5 9.1±2.8 15.143.1 2.442.4 0.740.5 0.040.0 29 13.84:2.1 14.14:3.4 14.641.7 NM 0.040.0 0.240.2 30 5.84:2.9 13.644)9 12.042.9 3.942.5 3.641.8 0.040.0 4 5.24:1.8 5.54:1.6 4.742.4 0.240.2 0.040.0 0.040.0 31 6.1±0.7 9.944)9 7.846.6 2.941.9 0.040.0 NM 19 13.74:1.8 9.94:1.0 ^ 9.241.2 0.840.4 NM 0.940.6 32 11.2±1.6 11.94:1.5 3.441.0 0.840.5 0.040.0 0.340.3 33 9.04:2.2 10.64:2.8 3.942.4 4.742.6 0.040.0 0.040.0 15 5.34:1.9 10.54:1.8 9.843.7 0.040.0 0.040.0 NM 34 15.44:2.2 0.044)0 0.040.0 NM 0.040.0 0.040.0 Table A.2 Raw data of apoptotic index (mean±SEM) for individual tumors in Table 2.1. NM: not measured. 160 Paclitaxel (|iM) LI, % of Control 24 hr 96 hr 1.2 x 10-* 97.04=6.7 10I.3±5.4 1.2 x 10'^ 83.7 ±4.4 75.7±6.1 1.2 x 10'- 70.7±3.7 72.1±5.9 1.2 x 10'* 59.9±4.0 50.3±5.7 1.2 x 10** 53.9±4.2 44.8±6.6 1.2 x 10' 55.6±4.2 3l.9±5.1 Table A 3 Data (niean±SEM) for Figure 2.2. 161 Paclitaxel (|iM) Apoptotic Index, % 96 hr 24 hr 0 0.3±0.1 0.3±0.1 1.2 x 10-* 0 .2±0.1 O.liO.l 1 2 x 10^ 0 .8 ±0.2 0.4±0.4 1.2 x 10-- 2 .6±0.6 1.8 ±0.6 1.2 x 10-* 8.1±0.9 8.3±1.5 1.2 x 10" 10.7±1.0 10.2±0.8 1.2 x 10* 10. 1+0.6 10.3±1.4 Table A.4 Data (mean±SEM) for Figure 2.3. 162 APPENDIX B DATA RELEVANT TO CHAPTER 3 163 PC3 Paclitaxel 24 hr Paclitaxel 96hr Immediate Paclitaxel 24 hr (pM) Immediate (pM) (% of control) (liM) delayed (% of control) (% of control) 2.3x10-’ 97.5±4.5 NM NM NM NM 4.5x10-' 108.3*8.8 4.6x10-* 106.9*7.2 4.6x10-* 93.7*4.6 9|xlO ^ 99.9*14.9 9.1x10-* 96.2*10.4 9.1x10-* 97.2*1.4 1.8x10-3 103.2*7.5 1.2x10-3 93.0*7.6 1.8x10-3 94.4*20.5 3.7x10-3 98.1*10.1 2.3x10-3 88.6*9.1 2.3x10-3 78.4*15.5 7.3 X10-3 102.5*12.1 9.2x10-3 13.5*0.9 9.2x10-3 21.8*15.9 1.5x10-' 90.7*6.1 1.8 x 10-: 29*2.4 1.5x10-: 5.6* 1.9 2.9* 10-: 83.2*12.1 3.7x10-: 2.2*2.4 2.9x10-: 0.5*16 5.9x10-- 72.5*8.5 7.3x10-: 1.8*2.5 5.9x10-: -0.2* 2.0 1.2x 10-' 68.3*9.3 1.2 x 10-' 2.1*2.5 1.2 x 10' -0.3*2.7 1.2 x 10" 67.4*12.1 1.2 x 10° 04*1.3 1.2 x 10° 05*1.0 DU145 2.3x10-' 102.7*8.3 1.2 x 10-* 97.0*0.3 1.2 x 10-* 108.9*8.0 4.5x10-* 98.6*7.7 NM NM NM NM 9.1x10-* 100.7*15.2 NM NM NM NM 1.8x10-3 106.7*8.4 1.2x10-3 98.4*7.4 1.2x10-3 101.6* 0.0 3.7x10-3 99.6*10.5 NM NM NM NM 7.3x10-3 91.8*13.2 NM NM NM NM 1.5x10-: 89.0*12.7 1.2 x 10-: 35.8*16.5 1.2x 10-: 47.4*21.5 2.9x10-: 84.9*3.1 NM NM NM NM 5.9x10-: 85.3*4.1 NM NM NM NM 1.2 x 10' 79.9*11.7 1.2 x 10' 10.5*4.2 1.2x 10-' 19.8*12.3 1.2 x 10" 78.8*10.2 1.2 x 10° 9.7*2.4 1.2 x 10° 14.9*6.3 Table B.l Data (mean±SEM) for Figure 3.2. NM; not measured. 164 BrdUrd LI, % Apoptotic Index, % HU(raM ) Without With Without With Paxlitaxel Paxlitaxel Paxlitaxel Paxlitaxel 0 34.2±3.9 15.6±3.6 0 .2±0.2 6.0±0.5 2 5.2±1.9 2 .0d=l.l 2.7±I.l 2 .0± 0.6 10 I.4±0.9 0.4±0.3 2.1±I.I 3.5±1.2 Table B.2 Data (mean±SEM) for Figure 3.3. 165 BrdUrd LI, % Apoptotic Index, % HU (itiM) Without With Without With Paxlitaxel Paxlitaxel Paxlitaxel Paxlitaxel 0 39.5±6.5 21.1±5.4 0.0± 0.0 8.8±3.3 2 15.9±6.0 12.6±3.8 O.liO.l 2.3±2.0 10 6.6±4.9 6.9±2.8 0 .2±0.2 0.7±0.7 Table B.3 Data (mean±SD) for Figure 3.4. 166 1 BrdUrd .A.poptotic Regimens Labeling Index, % Index, % A Treatments Treatments None, 24 hr None, 48 hr 39.5±6.5 0 HU 2mM, 24 hr HU 2 mM, 48 hr 15.9±6.0 O.liO.l HU 10 mM, 24 hr HU 10 mM, 48 hr 6.6±4.9 0.2±0.4 None. 24 hr TXL, 1.2 pM 21.1±5.4 8.8±3.3 48 hr HU 2 mM, 24 hr TXL, 1.2 pM 12.6±3.8 2.3±2.0 HU 2 mM, 48 hr HU 10 mM, 24 hr TXL. 1.2 pM 6.9±2.8 0.7±0.7 HU 10 mM, 48 hr C Simultaneous HU plus TXL, 1.2 pM, treatments HU 2 mM, 48 hr 27.8±10.2 4.8±2.5 HU 10 mM. 48 hr 32.8±7.4 5.5±3.7 HU 2 mM, 96 hr 32.4±11.9 6.0±3.9 HU 10 mM, 96 hr 29.9±6.7 6.1±2.7 D Treatments Treatments Treatments TXL. 1.2 pM TXL, 1.2 pM TXL, 1.2 pM 24.3±4.4 11.7±3.4 24 hr 24 hr 48 hr TXL, 1.2 pM TXL, 1.2 pM TXL, 1.2 pM 14.8±5.5 1.0±0.6 24 hr HU 2 mM, 24 hr HU 2 mM, 48 hr TXL. 1.2 pM TXL, 1.2 pM TXL, 1.2 pM 12.3±7.4 1.7±0.3 24 hr HU 10 mM, 24 hr HU 10 mM. 48 hr Table B.4 Data (mean±sd) for Figure 3.5. TXL; paclitaxel; HU: hydroxyurea. Regimens are shown as in Figure 3. IB, C and D. 167 BrdUrd LI, % Apoptotic Index, % 6.9 0.7 12.3 1.7 12.6 2.3 14.8 l.O 27.8 4.8 29.9 6.1 32.4 6.0 32.8 5.5 Control LI, CWR 48 hr, 39.5 8.8 Control LI, CWR 96 hr, 39.5 11.7 Control LI, patient 48 hr, 34.2 6.0 Table B.5 Data for Figure 3.6. 168 Apoptotic Index, % Apoptotic Index, % HU (mM) Labeled (Human) Labeled (CWR22) Without With Paxlitaxel Without With Paxlitaxel Paxlitaxel Paxlitaxel 0 0.02±0.04 4.63±0.1.51 O.OOiO.OO 8.18±3.00 2 0.66±0.91 0.27±0.42 O.lOiO.lO 2.14±1.86 10 0.41±0.55 0.13±0.10 0.20±0.37 0.60±0.60 Table B.6 Data (meanisd) for Figure 3.7. 169 Apoptotic Index, % Time (hr) 0 pM 0.012 pM 0.12 pM 1.2 pM 12 pM 1 NM NM NM 1.4±1.6 NM 3 NMNMNM 2.4±2.3 NM 8 NMNMNM 1.8±2.5 NM 16 0.4±0.8 NM NM 6.8±3.7 NM 24 0 .8 ±0.6 NM NM 15.9±3.0 11.2±3.6 48 0.4±0.2 NM 10.5±3.7 12.2±4.8 7.6±1.5 72 0.2±0.3 2.6±0.5 11.4±4.2 13.54:3.1 9.8±3.3 96 0.4±0.5 3.2±2.l 13.2±3.1 12.9±2.7 10.5±1.0 120 O.OiO.O 3.3±3.0 7.0±0.1 10.4±0.3 NM Table B.7 Data (meaniSD) for Figure 3.8. NM; not measured. 170 APPENDIX C DATA RELEVANT TO CHAPTER 4 171 Tumor 10 (ug/ml) 1 0.1 0.01 0.001 0.0001 0.00001 LABELING INDEX 1 O.OiO.O 7.1*4.0 43.2*11.2 96.3*18.2 106.2*12.3 98.6*14.1 NM 6 O.OiO.O 11.2*4.1 42.4*20.3 87.9*18.5 91.3*24.8 103.9*14.0 NM 7 O.OiO.O 0.7*07 8.6*58 55.1*15.2 79.5*17.1 93.8*14.8 NM 8 O.OiO.O 00*0.0 6.2*6.2 45.9*12.0 NM 78.8*31.8 NM 9 O.OiO.O NA 39.0*14.6 113.7*11.9 118.0*15.4 102.1*12.1 NM 10 O.OiO.O 3.9*39 27.1*10.4 74.5*9.9 90.1*6.1 98.8*6.0 NM 11 O.OiO.O 3.4*1.3 34.3*8.6 81.5*18.3 NA NA NM 13 O.OiO.O 2.1±1.5 19.7*3.2 95.2*8.4 99.4*6.1 109.6*7.2 NM 15 O.OiO.O 0.0*0.0 9.8*1.5 62.2*9.3 82.7*12.5 87.9*9.0 NM Tumor 10.5 (ug/ml) 3.5 0.35 0.035 0.0035 0.00035 0.000035 2 O.OiO.O 00*0.0 9.9*45 46.3*10.8 44.0*12.1 105.4*25.3 88.3*14.6 3 O.OiO.O 00*0.0 11.5*3.6 34.0*7.6 44.6*12.8 85.3*8.4 86.9*10.4 4 O.OiO.O 00*0.0 6.2*62 45.9*12.0 NM 78.8*31.8 NM 5 O.OiO.O O.OiO.O 5.0*33 NM 118.1*30.2 NM 99.3*19.4 12 6.6±6.6 0.0*0.0 4.6*1.2 24.2*5.4 57.4*7.5 105.4*11.5 84.2*10.8 14 O.OiO.O 00*0.0 3.3*16 30.7*4.9 58.8*18.6 65.6*13.3 73.6*24.9 16 O.OiO.O 00*0.0 1.3*13 17.8*5.4 57.8*10.0 65.3*10.3 NM 17 O.OiO.O 00*0.0 0.0*00 13.9*3.9 92.9*21.4 92.3*16.5 131.8*20.3 TUNEL-POSITIVE FRACTION Tumor 10 (ug/ml) 1 0.1 0.01 0.001 0.0001 0.00001 1 lOOdtO 13.8*3.4 24*1.1 1.3*0.8 2.0*2.0 1.0*0.6 NM 6 lOOdbO 17.6*3.3 14.6*3.7 5.8*1.4 1.0*0.9 2.4* 1.7 NM 7 lOOiO 51.5*13.4 45.9*13.2 4.8*1.2 4.8*1.5 2.0* 1.2 NM 8 lOOiO 62.3*15.8 8.6* 1.6 6.0*09 4.8*1.2 4.7*0.7 4.5*0.7 9 100±0 NA r 27.1*10.9 3.1*13 3.3*08 2.9* 1.4 NM 10 lOOiO 15.6*3.8 12.0*3.6 7.7*2.0 3.8*15 5.5*2.4 NM 11 100±0 21.2*4.9 8.0*09 4.6*0.8 NA NA NM 13 100*0 23.3*4.7 5.6* 1.2 2.8*10 2.3*08 1.5*0.7 NM 15 100*0 32.3*11.6 4.2* 1.3 2.6*15 1.8*0.5 1.4*0.5 NM Tumor 10.5 (ng/ml) 3.5 0.35 0.035 0.0035 0.00035 0.000035 2 100*0 100*0 13.4*3.7 2.6*1.4 5.3*1.3 0.3*0.3 04*0.3 3 100*0 88.5*11.5 21.7*9.0 4.6*0.7 4.7*0.9 3.2*1.2 2.1*1.3 4 100*0 100*0 31.9*6.6 16.7*2.5 NM 4.3* 1.5 NA 5 100*0 NA 22.7*5.1 16.8*2.5 0.0*0.0 NM 0.0*0.0 12 100*0 95.7*4.3 6.6* 1.0 13.9*6.5 30*1.3 11.8*4.1 2.5*09 14 100*0 100*0 24.4*10.5 11.0*2.0 6.3*1.2 10.1*4.3 7.6*1.2 16 100*0 100*0 24.1*1.4 16.8*3.7 7.1*1.3 13.8*3.1 NM 17 100*0 100*0 8.2*28 6.5*2.0 2.3*1.3 1.4*1.4 0.6*06 Table C l Raw data of LI and TUNEL index (mean±SEM) for individual tumors in Table 4.2. NM: not measured; NA: not applicable. 172 Doxorubicin (piM) LI, % of Control TUNEL Index, % 0 NA 3.1±0.6 l.TxlO-* 96.0±2.0 4.9±0.8 1.7x10-^ 87.0±4.0 S .liO .8 1.7x10'- 67.0±6.0 5.4±0.7 1.7x10'* 26.0±5.0 10.7±2.6 1.7x10" 4.6±0.9 50.3±8.6 6 .0 x 10" O.OiO.O 94.4±2.2 1.7x10' 0.4±0.4 99.7±0.2 Table C.2 Data (mean±SEM) for Figure 4.2. NA; not applicable. 173 APPENDIX D DATA RELEVANT TO CHAPTER 5 174 Exposure Doxorubicin, Medium (pM) Time (hr) 0.02 nM 0.06 pM 0.12 pM 0.5 pM 3 0.019db0.004 0.064±0.01I 0.131=1=0.010 0.462=1=0.133 6 0.022±0.006 0.06I±0.022 0.142±0.039 0.538±0.136 12 0 .022±0.001 0.060=fc0.013 0.084±0.055 0.506=1=0.141 24 0.014±0.005 0.053±0.022 0.09l±0.025 0.367±0.077 48 0.013±0.007 0.045±0.0I0 0.068=fc0.0l4 0.271±0.025 96 0.008^=0.003 0.025±0.009 0.055±0.009 0.181=1=0.084 120 0.003±0.002 0.005±0.001 0.008=1=0.005 0.030=1=0.013 144 0 .002=^0 .0 0 1 0.003=1=0.001 0.004±0.001 0.016±0.010 Exposure Doxorubicin, Medium (pM) Time (hr) 2 pM 10 pM 3 2.355±0.748 1I.174±4.032 6 1.876±0.805 13.005=1=6.671 12 1.977±0.995 9.819±5.252 24 1.589±0.453 6.69861.887 48 1.406±0.163 6.14060.628 96 0.888±0.150 5.43660.534 120 0.088±0.026 0.39260.011 144 0.096±0.057 0.54760.061 Table D.l Data (mean±SD) for Figure 5.1. 175 Exposure Doxorubicin, Tissue (|iM) Time (hr) 0.02 pM 0.06 pM 0.12 pM 0.5 pM 3 0.136±0.134 0.304±0.069 0.7154:0.728 2.8344=0.808 6 0.I5I±0.076 0.424±0.I2I 1.254±0.205 4.9934=0.298 12 0.374±0.I58 0.505^=0.236 1.4164:0.531 7.498± 1.695 24 0.405±0.I97 1.3334=0.468 1.956±0.659 10.95±1.67 48 0.414±0.135 1.8854=0.355 3.5764=0.679 15.844=0.77 96 0.794±0.150 2.867±0.508 5.8474=0.795 21.55±4.52 120 0.495±0.031 1.615±0.357 3.2974=1.762 12.36±2.10 144 0.239±0.098 0.8984=0.240 2.686± 1.179 7.39±1.10 Exposure Doxorubicin, Tissue (pM) Time (hr) 2 pM 10 pM 3 18.434=2.34 104.9614.2 6 39 14615 07 186.4616.2 12 60.43617.76 216.1612.4 24 82.74616.26 376.7626.6 48 94.33621.33 433.9626.6 96 100.90612.31 458.2691.3 120 69.48614.04 304.1647.0 144 55.28610.13 249.3633.2 Table D.2 Data (mean±SD) for Figure 5.2. 176 Doxorubicin, Medium (|aM) Ratio 0.02 35.4 0.06 39.8 0.12 44.8 0.5 47.5 2 55.9 10 60.4 Table D.3 Data (mean±SD) for Figure 5.3. 177 Doxonibicin, Medium (|aM) Recovered, % 0.02 27.2±1.1 0.06 5.0±0.8 0.12 2 . 1±0.8 0.5 2.8±3.3 2 3.2±2.8 10 2.8±2.4 Table D.4 Data (mean±SD) for Figure 5.4. 178 APPENDIX E DATA RELEVANT TO CHAPTER 6 179 Time (hr) Doxorubicin, Plasma, na ml Bolus 0.03 2988.2=354.1 0.08 1248.1=462.2 0.17 468.5=225.3 0.25 2032=40.1 0.33 99.4 0.42 49.7 0.5 42.8=0.3 0.75 33.8=7.0 1.00 21.8=4.4 1.75 17.2 2.00 15.0=4.1 2.62 12.0 4.00 10.4=3.8 8.00 7.7= 1.0 12.00 5.0=0.2 15.00 5.6 24.00 5.1=0.9 48.00 46=1.2 72.00 3.7=0.1 96.00 3.5=03 Infusion 2.00 3.19 3.00 4.71=2.82 24.00 7.10=28.2 48.00 9.54=3.20 72.00 7.29=3.28 96.00 11.49=4.30 Table E.l Data (mean±SD) for Figure 6.1. 180 Time Prostate Concentration (min) (ng/g) bolus 5 3467±612 60 2072 120 1912 240 I656±790 5760 6186168 infusion over 5760 min 16666675 2 day after 5760 min infusion 610 Table E.2. Data (niean±sd) for Figure 6.2. 181 Time Cprostate/Cplasma (min) Ratio bolus 5 3.3390 60 100.6385 120 79.5242 240 144.1697 5760 185.3284 infusion, 96 hr 5760 144.8994 48 hr after a 96-hr infusion 5760 + 2880 186.2971 Table E.3. Data for Figure 6.3. 182 APPENDIX F DATA RELEVANT TO CHAPTER 7 183 CWRZi 1 Tumor A iTumor A Tumor A Control LI 49.3 Control LI 34.2 Control LI 40.9 paclitaxel 96 hr ±SD paclitaxel 24 hr mean ±SD doxoiubtcin 96 hr mean ±SD 40 46.9 13.4 40 39.5 110 10 0.0 0.0 10 51.3 11.3 10 45.0 10.8 1 0.6 11 I 53J 13.8 1 37.1 15.0 0.1 39.5 27.5 0.1 49.9 20.5 0.1 67.9 18.2 0.01 98.5 24.4 0.01 75.0 18.9 0.01 63.5 13.9 0.001 108.6 17.5 0.001 76.5 24.1 0.001 78.7 141 0.0001 87.2 16.5 0.0001 911 24.5 0.0001 84.6 39.5 Tumor B Tumor B Tumor B Control Li 37.2 Control LI 29.5 Control Li 36.6 paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 37.5 11.8 10 44.2 21.5 10 0.0 0.0 1 41.g 219 1 60.5 241 1 7.5 19.5 0.1 47.6 13.9 0.1 58.2 241 0.1 218 14.1 0.01 94.7 25.4 0.01 83.9 23.5 0.01 106.4 31.7 0.001 94.4 20.3 0.001 1017 27.0 0.001 98.4 11.7 0.0001 109.6 271 0.0001 97.6 272 0.00001 103.0 16.6 Tumor C Tumor C Tumor C Control LI 39.3 Control Li 33.8 Control Li 40.4 paclitaxel 96 hr ±SD paclitaxel 24 hr mean ±SD doxonthicin 96 hr mean ±SD 10 39.1 17.5 10 51.1 20.0 10 0.0 0.0 1 34.5 119 1 50.3 16.3 1 3.6 9.9 0.1 48.1 22.2 0.1 79.4 271 0.1 46.2 31.8 0.01 80.8 18.4 0.01 73.5 211 0.01 85.0 19.2 0.001 76.1 21.2 0001 89.4 29.7 0.001 96.3 26.4 0.0001 88.3 17.6 0.0001 94.3 38.4 0.00001 107.7 230 Tumor D Tumor D Tumor D Control LI 31.8 Control LI 31.6 Control LI 31.8 paclitaxel 96 hr ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 43.1 11.3 10 46.7 18.7 10 0.0 0.0 1 50.1 16.8 1 49.6 26.4 1 7.7 9.1 0.1 54.6 27.9 01 53.2 112 0.1 44.8 33.8 0.01 97.6 44.0 0.01 95.2 16.7 0.01 104.4 39.8 0001 88.8 10.1 0001 970 21.8 0.001 91.4 27.8 0.0001 91.2 8.3 0.00001 93.1 15 Tumor E Tumor E Tumor E Control LI 39.9 Control LI 27.7 Control LI 43.2 paclitaxel 96 hr ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 50.8 14.6 10 67.2 24.4 10 0.0 0.0 1 513 11.3 1 65.5 318 1 0.1 01 0.1 58.9 20.1 0.1 68.5 13.4 0.1 28.1 212 0.01 80.9 17.1 0.01 81.3 21.0 0.01 88.5 34.2 0.001 94.1 20.6 0.001 94.8 24.9 0.001 90.7 34.6 0.0001 911 215 Tumor F Control LI 45.2 doxorubicin 96 hr mean ±SD 10 0.0 0.0 1 0.6 10 0.1 28.9 16.9 0.01 84.8 21.7 0.001 93.4 29.1 0 0001 83.2 24.7 Table F.l Raw data of LI in CWR22 xenograft tumors (meanjrSD) in Table 7.2. 184 CWR22R 1 Tumor A Tumor A Tumor A Control LI 39.3 Control LI 34.3 Control LI 34.3 paclitaxel 24 h mean ±SD paclitaxel 96 hr mean ±SD doxorubicin 96 hr mean ±SD 10 50.7 15.3 10 44.0 20.2 10 0.0 0.0 1 52.7 27.1 1 52.1 26.1 1 3.7 11.0 0.1 70.4 22.8 0.1 43.7 28.0 0.1 24.3 11.9 0.01 101.7 57.0 0.01 88.7 21.1 0.01 92.2 57.8 0.0001 119.4 36.8 0.0001 118.6 34.0 0.0001 94.1 60.7 Tumor B Tumor B Tumor B 38.9 Control LI 37.6 Control LI 40.2 paclitaxel 24 h mean ±SD paclitaxel 96 hr mean ±SD doxorubicin 96 hr mean ±SD 10 6 Z 5 25.1 10 25.0 7.1 10 0.0 0.0 1 54.2 37.8 1 44.4 26.2 1 0.0 0.0 0.1 68.1 33.4 0.1 65.0 15.9 0.1 26.7 14.7 0.01 98.8 35.2 0.01 92.2 35.0 0.01 70.0 45.7 0.0001 107.4 36.7 0.0001 100.7 16.5 0.0001 74.3 29.1 Tumor C Tumor C Tumor C Control LI 49.3 Control LI 36.2 Control LI 36.2 paclitaxel 24 h mean ±SD paclitaxel 96 hi mean ±SD doxorubicin 96 hr mean ±SD 10 65.4 21.5 10 42.9 25.2 10 0.0 0.0 1 57.0 5.5 1 38.5 25.2 1 0.0 0.0 0.1 68.0 28.0 0.1 58.1 18.9 0.1 17.7 16.6 0.01 103.0 23.7 0.01 78.5 26.3 0.01 79.4 30.6 0.0001 110.7 26.0 0.0001 106.2 47.7 0.0001 78.4 48.3 Tumor D Tumor D Tumor D Control LI 51.8 Control LI 46.9 Control LI 24.4 paclitaxel 24 h mean ±SD paclitaxel 96 hr mean ±SD doxorubicin 96 hr mean ±SD 10 50.0 28.9 10 59.0 17.9 10 0.0 0.0 1 52.4 17.7 1 51.0 12.3 1 0.0 0.0 0.1 68.6 24.0 0.1 71.8 25.9 0.1 37.2 18.8 0.01 93.5 25.9 0.01 84.1 29.5 0.01 70.0 13.3 0.001 99.2 12.9 0.001 98.0 14.5 0.001 94.5 27.3 Tumor E Tumor E Tumor E Control LI 52.0 Control LI 41.0 Control U 39.3 paclitaxel 24 h mean ±SD paclitaxel 96 hr mean ±SD doxorubicin 96 hr mean ±SD 10 66.2 19.3 10 46.3 20.0 10 0.0 0.0 1 67.3 20.5 1 61.1 19.6 1 1.7 4.5 0.1 72.4 15.3 0.1 63.2 12.9 0.1 30.0 16.6 0.01 86.6 27.2 0.01 75.5 22.1 0.01 91.7 44.6 0.0001 87.8 12.9 0.001 102.8 30.4 0.001 94.8 32.3 Table F.2 Raw data of LI in CWR22R xenograft tumors (mean±SD) in Table 7.2. 185 CWR91 Tumor A Tumor A Tumor A Control LI 71.5 Control LI 83.3 Control I I 70.2 paclitaxel 96 br mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 18.2 6.1 10 41.0 10.6 10 0.0 0.0 1 26.4 9.9 1 30.9 10.2 1 6.4 7.9 0.1 25.1 7.4 0.1 65.5 16.4 0.1 16.8 243 0.01 90.9 12.2 0.01 94.4 4.6 0.01 85.6 37.0 0.0001 99.1 15.7 0.0001 96.0 3.4 0.0001 101.7 21.4 Tumor B Tumor B Tumor B C aitro lII 73.7 Control LI 74.0 Control I I 74.5 paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 27.2 11.0 10 34.4 9.8 10 0.0 0.0 1 21.1 14.3 1 42.6 10.6 1 3.5 6.3 0.1 23.4 14.7 0.1 74.3 16.7 0.1 23.7 27.7 0.01 86.0 17.1 0.01 92.8 13.9 0.01 95.0 13.5 0.0001 94.4 17.7 0.0001 94.3 16.0 0.001 94.6 28.0 0.0001 109.9 9.2 Tumor C Tumor C Tumor C COTitrol LI 71.0 Control I I 79.2 Control II 68.7 paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 33.5 12.7 10 35.6 7.4 10 0.0 0.0 1 26.3 7.2 1 39.7 10.1 1 6.2 6.7 0.1 30.0 11.1 0.1 44.9 19.4 0.1 27.5 10.2 0.01 97.8 8.7 0.01 93.5 9.1 0.01 91.8 19.3 0.0001 99.5 11.6 0.0001 105.2 6.5 0.001 76.5 18.0 0.0001 91.5 15.9 Tumor D Tumor D Tumor D Control LI 63.2 Control I I 57.4 Control II 64.1 paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 16.6 11.3 10 61.0 7.4 10 0.0 0.0 1 23.2 3.9 1 42.1 38.0 1 9.7 14.7 0.1 27.2 7.4 0.1 53.7 18.3 0.1 14.9 6.8 0.01 55.2 22.1 0.001 105.2 13.0 0.001 95.6 14.8 0.0001 104.2 29.7 0.0001 106.8 14.7 Tumor E Control I I 61.2 doxorubicin 96 hr mean ±SD 10 0.0 0.0 1 5.1 9.9 0.1 5.2 13.8 0.01 117.9 20.8 0.001 118.2 21.1 Table F.3 Raw data of LI in CWR91 xenograft tumors (mean±SD) in Table 7.2. 186 CWR22 Tumor A Tumor A Tumor A pacUtnel 96 hr meaa ±SD paclitaxel 24 hr mean USD doxoiubicin 96 hr mean ±SD 40 S.013 5.4 40 7.1 12 10 83.26 24.7 10 14.95 3.9 10 8.21 18 1 1.009 1 1 1137 43 1 7397 3 0.1 1327 28 01 9.174 3.4 0.1 9.661 4.2 0.01 1.058 25 0.01 4.699 3 0.01 3345 13 0.001 0.129 0.4 0.001 3.664 3.3 0.001 1137 1.5 0.0001 0.21 04 0.000! 0 0 0.0001 0.884 0.7 control 032 0.5 control 0.1 0.3 Control 0 0 Tumor B Tumor B Tumor B paclitaxel 96 hr ±SD paclitaxel 24 hr mean ±SD doxoiubicin 96 hr mean ±SD 10 7.959 35 10 6218 4.4 10 79.29 31.6 1 10.58 6.2 1 6.166 3.2 1 3.799 15.4 0.1 8.475 4.4 0.1 5.863 3.1 0.1 0.217 0.5 0.01 0.282 0.4 001 0.348 0.5 0.01 0.797 1.98 0.001 0 0 0001 0.422 0.7 0.001 0.291 0.76 conttol 0 0 control 0.1 0.2 0.0001 0.162 0.3 control 0.06 0.1 Tumor C Tumor C Tumor C paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD doxoiubicin 96 hr mean ±SD 10 1317 5.4 10 11.25 5.9 10 81 65 31.2 1 13.11 3.1 1 5.495 3.8 1 4667 15 0.1 10.98 5.5 0.1 5.211 13 0.1 0 488 1.3 001 0096 0.2 001 0.566 06 0.01 0.519 1.4 0.001 0 0 0.001 0 0 0.001 0.085 0.3 control 0.03 0.1 control 0 0 0.0001 0.119 0.2 0.00001 1.061 1.7 control 0.028 0.1 Tumor D Tumor D Tumor D paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 4.474 5.3 10 7.843 5.4 10 74.18 17.5 1 7.578 3.4 1 6.307 16 1 1.467 1.3 0.1 4.481 18 0.1 7.489 5.7 0.1 12 22 0.01 1.988 17 0.01 1.067 0.5 0.01 0.692 0.9 0.001 0.915 0.6 0001 0 0 0.001 0.289 0.3 0.0001 2.637 1.8 control L 0 0 0.0001 0.257 0.5 control 0.08 0.9 0.00001 0 0 control 0.512 1.1 Tumor E Tumor E paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD 10 4.054 1.86 10 10.19 5.7 1 4984 17 1 8.144 46 0.1 1.903 1.5 0.1 6.676 4 0.01 0.332 0.37 0.01 0 0 0.001 0.806 0.68 0.001 0.351 0.3 0.0001 0.557 0.8 control 0.1 0.2 control 0.8 0.8 Table F.4 Raw data of cell death in CWR22 xenograft tumors (mean±SD) in Table 7.2. 187 CWR22R Tumor A Tumor A Tumor A | paclitaxel 24 h mean ±SD paclitaxel 96 hi mean ±SD doxorubicin 96 hr mean ±SD 10 15.53 4.9 10 11.62 5.5 10 98.77 4.4 1 14.76 8.3 1 12.83 8.3 1 30.3 31 0.1 8.066 9.2 0.1 10.04 8.5 0.1 3.906 4.7 0.01 1.165 1.03 O.OI 5.352 3.3 0.01 1.289 2.7 0.0001 0 0 0.0001 0 0 0.0001 0 0 control 0.4 0.6 control 0.06 0.3 control 2.204 2.4 Tumor B Tumor B Tumor B paclitaxel 24 b mean ±SD paclitaxel 96 hr mean ±SD doxorubicin 96 hr mean ±SD 10 14.55 11.3 10 5.725 3.1 10 100 0 1 14.31 9.7 1 11.93 8.1 1 69.18 42 0.1 9.413 13.4 0.1 17.65 4.7 0.1 7.974 26 0.01 0 0 0.01 4.419 4.3 0.01 0 19 0.0001 0 0 0.0001 1.111 1.2 0.0001 0 24 control 0.189 0.5 control 0.17 0.6 control 0.515 0.9 Tumor C Tumor C Tumor C paclitaxel 24 h mean ±SD paclitaxel 96 hr mean ±SD doxorubicin 96 hr mean ±SD 10 19.48 9.6 10 12.82 8.3 10 100 0 1 19.08 5.5 1 10.12 7.9 1 43.18 11 0.1 19.92 9.3 0.1 15.72 6.6 0.1 26.8 34 0.01 3.958 4.6 0.01 4.699 8.9 0.01 1.095 2.1 0.0001 0.114 0.2 0.0001 0.325 0.7 0.0001 0 0 control 0.06 0.2 control 0 0 control 0 0 Tumor D Tumor D Tumor D paclitaxel 24 h mean ±SD paclitaxel 96 hr mean ±SD doxorubicin 96 hr mean ±SD 10 19.88 13.5 10 20.1 8.5 10 96.15 14 1 18.79 5.7 1 19.5 4.6 1 52.5 25 0.1 21.73 9.9 0.1 22.77 11.5 0.1 38.47 43 0.01 4.767 2.7 0.01 5.828 2.1 0.01 0 0 0.001 2.46 1.1 0.001 1.691 3.2 control 0 0 control 1.36 1.8 control 0.42 1.4 Tumor E Tumor E Tumor E paclitaxel 24 h mean ±SD paclitaxel 96 hr mean ±SD doxorubicin 96 hr mean ±SD 10 21.81 11.1 10 13.22 6.3 10 100 0 1 27.84 9.8 1 19.32 7.4 1 52.66 30 0.1 25.26 9.9 0.1 16.53 7.8 0.1 22.5 33 0.01 7.896 4.7 0.01 6.302 3.2 0.01 1.38 2.3 0.0001 1.1 1.2 0.001 0.852 0.9 0.001 0 0 control 1.68 1.1 control 0.15 0.6 control 0.458 1 Table F.S Raw data of cell death in CWR22R xenograft tumors (mean±SD) in Table 7.2. 188 CWR91 Tumor A Tumor A Tumor A paclitaxel 96 h mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 20.8 8 10 23.47 3.1 10 96.88 5.9 I 26.7 9.88 1 34.09 10.7 1 48.21 2 0 J 0.1 31.9 I I J 0.1 10.55 8.2 0.1 28.21 12.4 0.01 5.56 3.87 0.01 4.279 4.2 0.01 11.15 9.3 0.0001 1.27 0.59 0.0001 2.476 1.9 0.001 3.753 3.2 control 1.68 1.67 control 2.091 2.8 0.0001 3.515 3 3 control 2.49 2.7 Tumor B Tumor B Tumor B paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 25.62 12.9 10 30.12 14.5 10 100 0 1 38.26 18 1 22.65 I I J 1 57.64 2 1 2 0.1 24.78 9.8 0.1 14.99 8.5 0.1 17.52 8.5 0.01 3.281 1.7 0.01 1.701 1.6 0.01 3.878 4.8 0.0001 4.811 3.6 0.0001 1.935 1.86 0.0001 5.425 6.03 control 2.022 1.1 control 1.688 1.7 control 3.7 3.6 Tumor C Tumor C Tumor C paclitaxel 96 hr mean ±SD paclitaxel 24 hr mean ±SD doxorubicin 96 hr mean ±SD 10 35.24 13.9 10 22.88 8.1 10 100 0 1 29.13 29.1 1 30.22 15.7 1 43.18 10.7 0.1 27.86 27.9 0.1 11.45 11.3 0.1 23.15 5.8 0.01 9.598 5.6 0.01 6.889 4.4 0.001 8.706 8.7 0.0001 2.556 1.7 0.0001 2.878 1.03 0.0001 3 J1 4 2.9 control 3.574 2.4 control 2.792 2.88 control 5.068 3.98 Tumor D doxorubicin 96 hr mean ±SD 10 100 0 1 34.26 9 2 0.1 20.39 14.9 0.01 2.069 2.06 0.0001 1.054 13 control 0.973 1.3 Table F.6 Raw data of cell death in CWR91 xenograft tumors (mean±SD) in Table 7.2. 189 ^H-Thymidine LI BrdUrd LI 0.7255 0.7059 0.7895 0.7530 0.6928 0.6627 0.6563 0.6406 0.5433 0.5197 0.7419 0.6935 0.6054 0.5830 0.4494 0.4494 0.6477 0.6250 0.7108 0.6867 0.6000 0.5800 0.7595 0.7511 0.6230 0.6066 0.6081 0.5946 0.6646 0.6584 Table F.7. Data for Figure 7.2. 190 Paclitaxel, 24 hr LI (% of Control) (pM ) CWR22 CWR22R CWR91 Human 1.2x10'^ 92.5±9.1 99.2±0.0 105.2^=0.0 83.7+4.4 1.2x10-- 79.5^=11.8 96.8±6.8 93.6±0.8 70.7±3.7 1.2x10-' 65.4±10.2 69.4±1.9 59.6±12.9 59.9=1=4.0 1.2x10“ 52.5±11.2 56.7±6.2 38.8±5.4 53.9±4.2 1.2x10' 50.8±9.5 58.9±7.9 43.0±12.3 55.6±4.l Paclitaxel, 96 hr 1.2x10-' 96.7±1.4 108.5±9.1 99.3=1=4.0 101.3±5.4 1.2x10'^ 86.0±9.1 100.4±3.4 NM 75.7=1=6.1 1.2x10-- 85.8±9.8 83.8±6.9 82.5=1:18.8 72.1=1=5.9 1.2x10' 51.8±4.8 60.4±10.5 26.4±2.9 50.3=1=5.7 1.2x10“ 46.4±8.0 49.4±8.5 24.3±2.6 44.8±6.6 1.2x10' 44.4±6.5 43 4± 12.2 23.9=1=7.9 31.9±5.1 Doxorubicin, 96 hr l.TxlO-' 90.8±5.0 91.7±26.7 102.5±8.1 96.0±2.0 1.7x10-' 96.5±6.0 94.7±0.2 87.8±9.0 87.0±4.0 1.7x10'- 94.6±9.7 80.5±13.3 97.6±14.1 67.0=1=6.0 1.7x10-' 35.1±9.8 29.0±7.4 17.6=1=8.6 26.0=1=5.0 1.7x10“ 3.4±3.5 1.1±1.6 62=1=2.3 4.6=1=09 6.0x10“ NMNM NM 0.0=t0.0 1.7x10' O.OiO.O O.OiO.O 0 0 =1=0.0 04=1=0.4 Table F.8. Data (mean±sd) for Figure 7.3. NM; not measured. 191 Doxorubicin Cell Death, % (HM) CWR22 CWR22R CWR91 Human 0 0.2±0.2 3.1±1.7 0.6±0.9 3.14:2.6 1.7x10“* 0.2±0.1 3.3±1.8 2.4±4.1 4.9±0.8 1.7x10-^ 0.2±0.1 6.2±3.5 NM 5.1±0.8 1.7x10-2 0 .8±0.2 ll.3±11.9 0.7±0.7 5.44:0.7 1.7x10'* 0.8±0.5 22.3±4.6 19.9±14.1 10.74:2.6 1.7x10° 2.7±1.8 45.8±9.8 49.64:14.3 50.34:8.6 6.0x10° NM NM NM 94.44:2.2 1.7x10* 79.6±4.0 99.2±1.6 99.04:1.7 99.74:0.2 Paclitaxel Cell Death, % (ixM) 24 hr 96 hr CWR22 CWR22R CWR91 Human CWR22 CWR22R CWR91 Human 0 0±0 0.7±0.7 22±0.6 0.4±0.7 0±0 03±02 2.4±1.0 0.3±0.4 1.2» lO-" 0.9±0 0.3±0.5 2.4±0.5 0J±0.6 1.1±1.4 0.5±0.6 2.9±1.8 Q2±03 1.2» 10-^ 0.6±0.9 2.5±0.0 NM 0.4±1.1 1.1±1.5 1.3±0.6 NM 0.8±1.1 1.2» 10-: 1.1±1.4 3.6±3.1 4.3±2.6 1.8±1.8 1.5±1.9 5.3±0.8 6.1±3.2 2.6±2.8 1.2» 10-' 7.1±1.7 16.9±7.7 123±2.3 8.3±4.8 7.0±3.7 16.5±4.6 28.2±3.6 8.1±4.6 1.2» 10“ 6.7±1.0 19.0±5.4 29.0±5.8 10J±2.6 9.5±32 14.7±4.4 31.4±6.1 10.7±5.0 1.2»10‘ 8.7±2.0 18.3±3.1 25.S±4.0 10Jx4.5 8.7±4.8 12.7±5.1 27.2±7.3 10.1±3.2 Table F.9. Data (meanisd) for Figure 7.4. NM; not measured. 192 PSA (% of Control) Drug Concentration CWR22 CWR22R (gg/ml) Paclitaxel Doxorubicin Doxorubicin 96 hr 96 hr 96 hr 10 67.6Ü1.3 I3.3±4.4 3.8±0.7 1 66.8±23.9 74.3±0.9 NA 0.1 78.2±6.4 76.1 ±23.3 77.8±12.2 0.01 92.3±I4.3 NA NA 0.001 87.4±9.1 106.0±23.5 84.8±0.5 Table F. 10. Data (mean±sd) for Figure 7,5. NA ; not available. 193 BIBLIOGRAPHY Abbas, F. and Scardino, P.T. Why neoadjuvant androgen deprivation prior to radical prostatectomy is unnecessary. Urol. Clin. N. Am., 23:587-604, 1996. Abbas, P., Kaplan, M., and Soloway, M.S. Induction androgen deprivation therapy before radical prostatectomy for prostate cancer. Br. J. Urol., 77:423-428, 1996. .'Mbertsen, P.C., Fryback, D.G., Storer, B.E., Kolon, T.F., and Fine, J. Long-term survival among men with conservatively treated localized prostate cancer. JAMA, 274:626-631, 1995. .Alivizatos, G. and Oosterhof, GO. Update of hormonal treatment in cancer of the prostate. Anti-Cancer Drugs, 4:301-309, 1993. Arbuck, S.G., Christian, M.C., Fisherman, J.S., Cazenave, L.A., Sarosy, G., Suffhess, M., Adams, J., Canetta, R., Cole, K.E., and Friedman, M.A. Clinical development of taxol. Monogr. Natl. Cancer Inst., 15:11-24, 1993. Arbuck, S.G., and Blaylock, B.A Taxol: Clinical results and current issues in development. In SuflBiess, M., (ed.), Taxol-Science and applications. pp379-415, Boca Raton, CRC Press 1995. Arcamone, F., Cassinelli, G., Fantini, G., Grein, A., Orezzi, P., Pol, C., and Spalla, C. Adriamycin, 14-hydroxydaunoraycin, a new antitumor antibiotic from S. peucetius var. caesius. Biotech. Bioeng., 11:1101-1110, 1969. Au, J.L.-S., Kalns, J., Gan, Y., and Wientjes, M.G. Pharmacologic effects of taxol in human bladder tumors. Cancer Chemother. Pharmacol., In Press, 1997. Bartsch, G. and Rohr, H P. Comparative light and electron microscopic study o f the human, dog and rat prostate. An approach to an experimental model for human benign prostatic hyperplasia. Urol. Intlis., 35:91-104, 1980. Barza, M. Anatomical barriers for antimicrobial agents. Eur. J. Clin. Microbiol. Inf. Dis., 12:s31-35, 1993. 194 Beland, G., Elhilali, M., Fradet, Y., Laroche, B., Ramsey, E.W., Trachtenberg, J., Venner, P.M., and Tewari, H.D. Total androgen ablation: Canadian experience, Urol. Clin. N. Am. 18: 75-82, 1991. Berchem, G.J., Bosseler, M., Sugars, L.Y., Voeller, H.J., Zeitlin, S., and Gelmann, E.P. Androgens induce resistance to bcl-2-mediated apoptosis in LNCaP prostate cancer cells. Cancer Res., 55:735-738, 1995. Bhalla, K., Ilerado, A_M., Tourkina, E., Tang, C., Mahoney, M E., and Huang, Y. Taxol induces intemucleosomal DNA fragmentation associated with programmed cell death in human myeloid leukemia cells. Leukemia, 7:563-568, 1993. BischofF, K.B., Dedrick, R.L., Zaharko, and Longstreth, J.A. Methotrexate pharmacokinetics. J. Pharm. Sci., 60:1128-1133, 1971. Bostwick, DG. and Brawer, MK.: Prostatic intra-epithelial neoplasia and early invasion in prostate cancer. Cancer, 59: 788-794, 1987. Bugat, R., Robert, J., Herrera, A , Pinel, M.C., Huet, S., Chevreau, C , Boussin, G , Roquain, J , and Carton, M. Clinical and pharmacokinetic study of 96-h infusions of doxorubicin in advanced cancer patients. Eur. J. Cancer Clin. Oncol., 25:505-511, 1989. Carboni, J.M., Singh, C., and Tepper, M.A. Taxol and lipopolysaccharide activation of a murine macrophage cell line and induction of similar tyrosine phosphoproteins. Monogr. Natl. Cancer Inst., 15:95-101, 1993. Carter, HB. and Coflfey, DS.: The prostate: an increasing medical problem. The prostate, 16: 39-48, 1990. Chai, M., Wientjes, M.G., Badalament, R.A., Burgers, J.K. and Au, J.L.-S. Pharmacokinetics o f intravesical doxorubicin in superficial bladder cancer patients. J. Urol., 152:374-378, 1994. Chan, K.K., Cohen, J.L., Gross, J.F., Himmelstein, K.J., Bateman, J R., Tsu-Lee, Y., and Marils, A.S. Prediction of adriamycin disposition in cancer patients using a physiologic, pharmacokinetic model. Cancer Treat. Rept., 62:1161-1171, 1978. Chang, Y.F., Li, L.L., Wu, C.W., Liu, T.Y., Lui, W.Y., P'eng, F.K., and Chi, C.W. Paclitaxel-induced apoptosis in human gastric carcinoma cell lines. Cancer, 77:14-18, 1996. Chang, AY., Kim, K., Glide, J., Anderson, T., Karp, D., and Johnson, D. Phase II study of taxol, merbarone, and piroxantrone in stage IV non-small-cell lung cancer: The Eastern Cooperative Oncology Group Results. J. Natl. Cancer Inst., 85:388-394, 1993. 195 Chen, C.T., Au, J.L-S., Burgers, J.K., and Wientjes, M.G. Comparative activity of taxol and doxorubicin in human prostate tumors. Proc. Am. Assoc. Cancer Res., 36:320, 1995. Chen, C.T., Au, J.L.-S., Gan, Y , and Wientjes, M.G. Differential time dependency of antiproliferative and apoptotic effects of taxol in human prostate tumors. Urol. Oncol., 3:11- 17, 1997a. Chen, C.T., Au, J.L.-S., and Wientjes, M.G. Pharmacodynamics of doxorubicin in human prostate tumors. Clin. Cancer Res., In press, 1997b. Chen, C.T., Au, J.L.-S., Gan, Y., and Wientjes, M.G. Androgen-dependent and -independent human prostate tumor xenografts as models for drug activity evaluation. In preparation, 1997c. Chen, G , Jafffezou, J,P., Fleming, W,H., Duran, G,E., and Sikic, B,I. Prevalence of multidrug resistance related to activation of the mdri gene in human sarcoma mutants derived by single-step doxorubicin selection. Cancer Res., 54:4980-4987, 1994. Class, P., and Van Steenbrugge, G.J. Expression of HLA-like structures on a permanent human turmor line PC-93. Tissue Antigens, 21:227-232, 1983. Cox, S.K., Wilke, A.V., and Frazier, D. Determination of adriamycin in plasma and tissue biopsies. J. Chromatogr. 564:322-329, 1991. Crom, W.R. Doxorubicin, In Taylor, W.J. and Caviness, M.H.D. (eds), A textbook for the clinical application of therapeutic drug monitoring, pp421-429, Irving, Texas: Abbott Laboratories, 1986. Crossin, K.L., and Carney, D.H. Microtubule stabilization by taxol inhibits initiation o f DNA synthesis by thrombin and epidermal growth factor. Cell, 27:341-350, 1981. Csapo, Z , Brand, K., Walther, R., and Fokas, K., Comparative experimental study of the serum prostate specific antigen and prostatic acid phosphatase in serially transplantable human prostatic carcinoma lines in nude mice. J. Urol. 140:1032-1038, 1988. Daneshgari, F. and Crawford, E D. Endocrine therapy of advanced carcinoma of the prostate. Cancer, 71:1089-1097, 1993. Danesi, R., Figg, W.D., Reed, E., and Myers, C.E. Paclitaxel (taxol) inhibits protein isoprenylation and induces apoptosis in PC-3 human prostate cancer cells. Mol. Pharmacol., 47:1106-1111, 1995. 19 6 Deffie, A.M., Alam, T., Seneviratne, C., Beenken, S.W., Batra, J.K., Shea, T.C., Henner, W.D., and Goldenberg, G.J. Multifactorial resistance to adriamycin; relationship of DNA repair, glutathione transferase activity, drug efflux, and P-glycoprotein in cloned cell lines of adriamycin-sensitive and -resistant P388 leukemia. Cancer Res., 48:3595-3602, 1988. Denis, L., Smith, P., Cameiro de Moura, J.L., Newling, D , Bono, A., Keuppens, F., Mahler, C., Robinson, M., Sylvester, R., De Pauw, M. et. al. Total androgen ablation: European experience. The EORTC GU Group., Urol. Clin. N. Am. 18:65-73, 1991. Denmeade, S R., Lin, X.S., and Isaacs, J.T. Role of programmed (apoptotic) cell death during the progression and therapy for prostate cancer. Prostate, 28:251-265, 1996. Desoize, B. Anticancer drug resistance and inhibition of apoptosis. Anticancer Res., 14:2291-2294, 1994. Ding, A., Porteu, F., Sanchez, E., and Nathan, C.F. Shared actions of endotoxin and taxol on TNF receptors and TNF release. Science, 248:370-372, 1990. Ding, A., Sanchez, E., and Nathan, C.F. Taxol shares the ability of baterial lipopolysacharide to induce tyrosine phosphorylation of microtubule-associated protein kinase. J. Immunol., 151:5596-5602, 1993. Dollery, S.C. Doxorubicin hydrochloride. Therapeutics Drugs, 1:D221-D225, Churchill Livingstone, 1991. Donaldson, K.L., Goolsby, G.L., Kiener, P A., and Wahl, A.F. Activation of p34cdc2 coincident with taxol-induced apoptosis. Cell Growth Diff., 5:1041-1050, 1994. Dorr, R.T., and Von HoK D.D. Paclitaxel. In: Cancer chemotherapy handbook. 2nd edition pp761-767, Norfolk, CT, Appleton & Lange, 1994. Dunning, W.F. Prostate cancer in the rat. Monogr. Natl. Cancer Inst., 12:351-369, 1963. Durand, R.E. Slow penetration of anthracyclineis into spheroids and tumors: a therapeutic advantage? Cancer Chemother. Pharmacol. 26:198-204, 1990. Einzig, A.I., Wiemik, P H., Sasloff, J., Runowicz, C D., and Goldberg, G.L. Phase II study and long-term follow-up of patients treated with taxol for advanced ovarian adenocarcinoma. J. Clin. Oncol., 10:1748-1753, 1992. Ellis, W.J., Vessella, R.L., Buhler, K. R., Bladou, F., True, L.D., Bigler, S. A., Curtis, D., and Lange, P H. Charactization of a novel androgen-sensitive, prostate-specific antigen- producing prostatic carcinoma xenograft: LuCaP 23. Clin. Cancer Res., 2:1039-1048, 1996. 197 Fair, W.R. and Wehner, N. The prostatic antibacterial factor: identity and significance. Proa. Clin. Biol. Res., 6:383-403, 1976. Forastiere, A.A. and Urba, S.G. Single-agent paclitaxel and paclitaxel plus ifosfamide in the treatment of head and neck cancer. Sem. Oncol., 22 (3 Suppl 6):24-27, 1995. Francis, P.A., Kris, M.G., Rigas, J R., Grant, S.C , and Miller, V.A. Paclitaxel (Taxol) and docetaxel (Taxotere): active chemotherapeutic agents in lung cancer. Lung Cancer 12 (Suppl 1): S I63-172, 1995. Frankel, A., Buckman, R., and Kerbel, R.S. Abrogation of taxol-induced G2-M arrest and apoptosis in human ovarian cancer cells grown as multicellular tumor spheroids. Cancer Res., 57:2388-2393, 1997. Furukawa, T., Kubota, T., and Hofifinan, R.M. Clinical applications o f the histoculture drug response assay. Clin. Cancer Res., 1:305-311, 1995. Gan, Y., Wientjes, M.G., Schuller, D.E., and Au, J.L. Pharmacodynamics of taxol in human head and neck tumors. Cancer Res., 56:2086-2093, 1996a. Gan, V., Wientjes, M.G, Badalament, R.A., Au, J.L.-S. Pharmacodynamics o f doxorubicin in human bladder tumors. Clin Cancer Res., 2:1275-1283, 1996b. Gangemi, R.M., Costa, G., Fulco, R.A., d'Aquino, S., and Aronadio, O. Apoptosis susceptibility of human carcinoma and leukemia cell lines to taxol. Relationship with cell cycle and drug concentration. Ann. N Y. Acad. Sci., 784:550-554, 1996. Gangemi, R.M., Tiso, M., Marchetti, C., Severi, A.B., and Fabbi, M. Taxol cytotoxicity on human leukemia cell lines is a function o f their susceptibility to programmed cell death. Cancer Chemother. Pharmacol., 36:385-392, 1995. Gamick, M B. The dilemmas of prostate cancer. Sci. Am., 270:72-81, 1994. Gelfand, V.I. and Bershadsky, A.D. Microtubule dynamics: mechanism, regulation, and function. Annu. Rev. Cell Biol., 7:93-116, 1991. Geller, J., Sionit, L.R., Connors, K., and Hoffman, R.M. Measurement of sensitivity in the human prostate in in vitro three-dimensional histoculture. Prostate 21:269-278, 1992. Germann, U.A., Pastan, I., and Gottesman, M.M. P-glycoproteins: mediators of multidrug resistance. Sem. Cell Biol., 4:63-76, 1993. 198 Gianni, L„ Capri, G., Munzone, E., and Straneo, M. Paclitaxel (Taxol) efficacy in patients with advanced breast cancer resistant to anthracyclines. Sem. Oncol., 21(5 Suppl 8):29-33, 1994. Gingrich, J., Tucker, J., Walther, P., Day, J., Poulton, S., and Webb, K. Establishment and characterization of a new human prostatic carcinoma cell line (DuPro-I). J. Urol., 146:915- 919, 1991. Gleason, D.F. Classification of prostatic carcinomas. Cancer Chemother. Rep., 50:125-128, 1966. Gold, R., Schmied, M., Giegerich, G., Breitschopf, H., Hartung, H P., Toyka, K.V., and Lassmann, H. Differentiation between cellular apoptosis and necrosis by the combined use o f in situ tailing and nick translation techniques. Lab. Invest., 71:219-225, 1994. Graham, S.D., Poulton, S.H., Linder, J., Woodard, B.H., Lyles, K.W., and Paulson, D.F. Establishment of a long-term adenocarcinoma of the prostate cell line in the nude mouse. Prostate, 7:369-376, 1985. Greenway, C.V. and Stark, R.D. Hepatic vascular bed. Physiol. Rev, 51:23-65, 1971. Gudziak, M R. and Smith, A.Y. Hormonal therapy for stage D cancer o f the prostate. West. J. Med., 160:351-359, 1994. Hale, A.J., Smith, C.A., Sutherland, L.C., Stoneman, V.E., Longthome, V.L., Culhane, A C., and Williams, G.T. Apoptosis: molecular regulation of cell death. Eur. J. Biochem., 236:1-26, 1996. Handley, R, Carr, T.W., Travis, D , Powell, P H., and Hall, R.R. Deferred treatment of prostate cancer. Br. J. Urol., 62:249-253, 1988. Hanks, GE., Myers, CE. and Scardino, PT.: Cancer of the Prostate. In DeVita, V.T. Jr., Heilman, S., and Rosenberg, S. A. (eds.), Cancer:Principles and Practice of Oncology, 4th edition, pp 1073-1113. Philadelphia: Lippincott-Raven Publishers, 1993. Hara, M., Koyanagi, Y., Inoue, T., and Fukuyama, T. Some physico-chemical characteristics of "-seminoprotein", an antigenic component specific for human seminal plasma. Forensic immunological study of body fluids and secretion. Jpn. J. Legal Med., 25:322-324, 1971. Harris, P.A and Gross, J.F. Preliminary pharmacokinetic model for adriamycin (NSC-123127). Cancer Chemother. Rep., 59:819-825, 1975. 199 Havrilesky, L.J., Elbendary, A., Hurteau, J.A., Whitaker, R.S., Rodriguez, G.C., and Berchuck, A. Chemotherapy-induced apoptosis in epithelial ovarian cancers. Obst. Gynecol., 85:1007-1010, 1995. Hill, B ,T, Whelan, R.D., Hemmings, V.J., and Franks, L.M. A method of synchronization o f normal and malignant human cells in culture. Cell Biol. Intl. Rep., 1:379-384, 1977. Hoehn, W., Wagner, M„ Riemann, J.F., Hermanek, P., Williams, E., Walther, R., and Schrueflfer, R. Prostatic adenocarcinoma PC EW, a new human tumor line transplantable in nude mice. Prostate, 5:445-452, 1984. Hoehn, W., Schroeder, F.H., Reimann, J.F., Joebsis, A C., and Hermanek, P. Human prostatic adenocarcinoma: some characteristics of a serially transplantable line in nude mice (PC 82). Prostate, 1:95-104, 1980. Holmes, F.A., Walters, R.S., Theriault, R.L., Forman, A.D., Newton, L.K., Raber, M.N., Buzdar, A.U., Frye, D.K., and Hortobagyi, G.N. Phase II trial of taxol, an active drug in the treatment of metastatic breast cancer. J. Natl. Cancer Inst., 83:1797-1805, 1991. Horm, J.W. and Sondik, E.J. Person-years of life lost due to cancer in the United States, 1970 and 1984. Am. J. Public Health, 79: 1490-1493, 1989. Horoszewicz, J.S., Leong, S.S., Chu, T.M., Wajsman, Z.L., Freedman, M., Papsidero, L., Kim, U., Chai, U.S., Kakati, S., Arya, S.K., and Sandberg, A.A. The LnCap cell line-a new model for studies on human prostatic carcinoma. Prog. Clin. Biol. Res., 37:115-132, 1980. Horwitz, SB. Taxol (paclitaxel): Mechanism of action. Annals Oncol., 5:53-56, 1994. Horwitz, SB. Mechanism of action of taxol. Trends Pharmacol. Sci., 13:134-136, 1992. Hudes, G.R., Nathan, F.E., Khater, C , Greenberg, R., Gomella, L., Stem, C , and McAleer, C. Paclitaxel plus estramustine in metastatic hormone-refractory prostate cancer. Sem. Oncol., 22(5 Suppl l2):41-45, 1995. Iguchi, H., Tone, H., Kiyosaki, T., and Ishikura, T. Pharmacokinetics and disposition of (2"R)-4'-0-tetrahydropyranyladriamycin in dogs. Jpn. J. Antibiotics, 39:638-652, 1986. lizumi. T., Yazaki, T., Kanoh, S., Kondo, I., and Koiso, K. Establishment o f a new prostatic carcinoma cell line (TSU-PRl). J. Urol., 137: 1304-1306, 1987. I no mata, M., Saijo, N., Kawashima, K., Kaneko, A., Fujiwara, Y., Kunikane, H., and Tanaka, Y. Induction of apoptosis in cultured retinoblastoma cells by the protein phosphatase inhibitor, okadaic acid. J. of Cancer Res. Clin. Oncol., 121:729-38, 1995. 2 0 0 Isaacs, J.T., Isaacs, W.B., Feitz, W.F.J., and Scheres, J. Establishment and characterization of seven Dunning rat prostate cancer cell lines andtheir use in developing methods for predicting metastatic abilities of prostatic cancers. Prostate, 9:261-281, 1986. Isaacs, J.T. and CofiFey, D.S. Adaptation versus selection as the mechanism responsible for the relapse of prostatic cancer to androgen ablation therapy as studied in the Dunning R-3327-H adenocarcinoma. Cancer Res., 41:5070507-5, 1981. Ito, Y.Z., Mashimo, S., Nakazato, Y., and Takikawa, H. Hormone dependency of a serially transplantable human prostatic cancer (HONDA) in nude mice. Cancer Res., 45:5058-5063, 1985. Jachez, B., Nordmann, R., and Loor, F. Restoration of taxol sensitivity o f multidrug-resistant cells by the cyclosporine SDZ PSC 833 and the cyclopeptolide SDZ 280-446. J. Natl. Cancer Inst., 85:478-483, 1993. Janssen, M.J.H., Crommelin, D.J.A., Storm, G., and Hulshoff, A. Doxorubicin decomposition on storage. Effect of pH, type of buffer and liposome encapsulation. Intl. J. Pharm., 23:1-11, 1985. Johansson, J.E. Watchful waiting for early stage prostate cancer. Urol., 43:138-142, 1994. Johansson, J.E., Adami, H.O., Andersson, S.O., Bergstrom, R., Holmberg, L., and Krusemo, U.B. High 10-year survival rate in patients with early, untreated prostatic cancer. JAMA, 267:2191-2196, 1992. Johnson, B.D., and Byerly, L. A cytoskeletal mechanism for Ca’* channel metabolic dependence and inactivation by intracellular Ca’". Neuron, 10:797-804, 1993. Kaighn, M E., Narayan, K.S., Ohnuki, Y., Lechner, J.F., and Jones, L.W. Establishment and characterization of a human prostatic carcinoma cell line (PC-3). Invest. Urol., 17:16-23, 1979. Kang, H.-J., Tran, A , Schuller, D.E., and Au, J.L.-S. Taxol uptake and retention in human cancer cells and solid tumors. Proc. Am. Assoc. Cancer Res., 37:367, 1996. ■ Kelland, L.R., and Abel, G. Comparative in vitro cytotoxicity of taxol and taxotere against cisplatin-sensitive and -resistant human ovarian carcinoma cell lines. Cancer Chemother. Pharmacol. 30:444-450, 1992. Kerr, D.J., Wheldon, T.E., Kerr, AM., and Kaye, S B. In vitro chemosensitivity testing using the multicellular tumor spheroid model. Cancer Drug Delevery, 4:63-74, 1987. 2 0 1 Kerr, J.F.R., Clay, M,, and Harmon, B.V. Apoptosis: Its significance in cancer and cancer therapy. Cancer, 73:2013-2026, 1994. King, F.G. and Dedrick, R.L. Physiologic model for the pharmacokinetics of 2'Deoxycoformycin in normal and leukemic mice. J. Pharmacok. Biopharm., 9:519-534, 1981. Korman, L.B. Treatment o f prostate cancer. Clin. Pharm., 8:412-424, 1989. Kuban, D.A., el-Mahdi, A.M., and Dchellhammer, P.F. Hemibody irradiation in advanced prostatic carcinoma. Urol. Clin. N. Am., 18:131-137, 1991. Kubota, T., Sasano, N., Abe, O., Nakao, 1., Kawamura, E., Saito, T., Endo, M., Kimura, K., Hiroshi, D., Sasano, H., Nagura, H., Ogawa, N., Hoffman, R M ., and the chemosensitivity study group for the histoculture drug-response assay. Potential of the histoculture drug- response assay to contribute to cancer patient survival. Clin. Cancer Res., 1:1537-1543, 1995. Kuh, H.-J., Tran, A., Schuller, D.E., and Au, J.L.-S. Taxol uptake and retention in human cancer cells and solid tumors. Proc. Am. Assoc. Cancer Res., 37:A2508, 1996. Kuriyama, R, Dasgupta, S., and Borisy, G.G. Independence of centriole formation and initiation of DNA synthesis in Chinese hamster ovary cells. Cell Motil. Cytoskeleton, 6:355-362, 1986. Kyprianou, N., English, H.F., and Isaacs, J.T. Programmed cell death during regression of PC-82 human prostate cancer following androgen ablation. Cancer Res., 50:3748-3753, 1990. Le Bot, M A., Riche, C., Guedes, Y., Kemaleguen, D., Simon, S., Begue, J.M., and Berthou, F. Study of doxombicin photodegradation in plasma, urine and cell culture medium by HPLC. Biomed. Chromatogr., 2:242-244, 1988. Lefrak, E.A., Pitha, J., Rosenheim, S., and Gottlieb, J.A. A clinicopathologic analysis of adriamycin cardiotoxicity. Cancer, 32:302-3014, 1973. Legas, S.S., Ring, S., Papadopulos, N., Raber, M., and Benzamin, R.S. A phase II trial of taxol in metastatic melanoma. Cancer 65:2478-2481, 1990. Legha, S.S., Benjamin, RS., Mackay, B., Yap, H.Y., Wallace, S., Ewer, M., Blumenschein, G.R, and Freireich, E.J. Adriamycin therapy by continuous intravenous infusion in patients with metastatic breast cancer. Cancer, 49:1762-1766, 1982b. 2 0 2 Legha, S.S., Benjamin, R.S., Mackay, B., Ewer, M., Wallace, S., Valdivieso, M., Rasmussen, S.L., Blumenschein, G.R., and Freireich, E.J. Reduction of doxorubicin cardiotoxicity by prolonged continuous intravenous infusion. Ann. Intern. Med., 96:133- 139, 1982a. Lenaz, L. and Page, J.A. Cardiotoxicity of adriamycin and related anthracyclines. Cancer Treat. Rev., 3:111-120, 1976. Li, W.W., Cordon-Cardo, C , Chen, Q., Jhanwar, S.C., and Bertino, J R. Establishment, characterization and drug sensitivity o f four new human soft tissue sarcoma cell lines. Intl. J. Cancer, 68:514-519, 1996. Li, T.S., and Beling, C.G. Isolation and characterization of two specific antigens of human seminal plasma. Fertility & Sterility, 24:134-144, 1973. Li, X., Gong, J., Felman, E., Seiter, K., Traganos, F., and Darzynkiewicz, Z. Apoptotic cell death during treatment of leukemias. Leuk. Lymphoma, 13:65-70, 1994. Li, D , and Au, J.L.-S. Characterization of taxol-resistant prostate cancer cell lines. Proc. Am. Assoc. Cancer Res., 38:229, 1997. Liebmann, J.E., Cook, J.A., Lipschultz, C., Teague, D., Fisher, J., and Mitchell, J B Cytotoxic studies of paclitaxel (Taxol) in human tumour cell lines. Br. J. Cancer, 68:1104-1109, 1993. Liebmann, J.E., Cook, J.A., Lipschultz, C., Teague, D , Fisher, J., and Mitchell, J.B. The influence of Cremophor EL on the cell cycle effects of paclitaxel (Taxol) in human tumor cell lines. Cancer Chemother. Pharmacol., 33:331-339, 1994. Liu, Y., Bhalla, K., Hill, C., and Priest, D.G. Evidence for involvement of tyrosine phosphorylation in taxol-induced apoptosis in a human ovarian tumor cell line. Biochem. Pharmacol , 48:1265-1272, 1994. Lopes, N.M., Adams, E.G., Pitts, T.W., Bhuyan, B.K. Cell kill kinetics and cell cycle effects of taxol on human and hamster ovarian cell lines. Cancer Chemother. Pharmacol., 32:235- 242, 1993. Manthey, C.L., Qureshi, N., Stutz, P.L., and Vogel, S.N. Lipopolysaccharide antagonists block taxol-induced signaling in murine macrophages. J. Exp. Med., 178:695-702, 1993. Matzkin, H., and Soloway, M.S., Response to second-line hormonal manipulation monitored by serum PSA in stage D2 prostate carcinoma. Urol., 40:78-80, 1992. 203 McDonnell, T.J., Troncoso, P., Brisbay, S.M., Logothetis, C., Chund, L.W.K., Hsieh, J.-T., Tu, S.-M., and Campbell, M L. Expression of the protooncogene bcl-2 in the prostate and its association with emergence of androgen-independent prostate cancer. Cancer Res., 52:6940-6944, 1992. McGuire, W.P., Rowinsky, E.K., Rosenshein, N.B., Grumbine, F.C., Ettinger, D.S., Armstrong, O.K., and Donehower, R.C. Taxol: A unique antineoplastic agent with significant activity in advanced ovarian epithelial neoplasms. Ann. Intern. Med., 111:273- 279, 1989. Meares, E.M. Jr. Prostatitis: review of pharmacokinetics and therapy. Rev. Inf. Dis., 4 :475- 483, 1982. Mettlin, C , Murphy, G.P., Lee, P., Littrup, P.J., Chesley, A., Babaian, R., Badalament, R., Kane, R. A, and Mostofi, F.K. Characteristics of prostate cancers detected in a multimodality early detection program. Cancer, 72:1701-1708, 1993. Mettlin, C. et al., and investigators of the American Cancer Society-National Prostate Cancer Detection Project. Cancer, 71:891-898, 1993. Mickey, D.D., Stone, K.R., Wunderli, H., Mickey, G.H., Vollmer, R.T., and Paulson, D.F. Heterotransplantation of a human prostatic adenocarcinoma cell line in nude mice. Cancer Res., 37:4049-4058, 1977. Mickey, D.D. Growth and diflferentiation factors in prostatic tissue. In Motta, M. and Serio, M. (eds). Hormonal therapy of prostatic diseases: basic and clinical aspects, pp 13-47, Bussum, The Netherlands, Medicom, 1988. Milas, L., Hunter, N.R., and Kurdoglu, B. Kinetics of mitotic arrest and apoptosis in murine mammary and ovarian tumors treated with taxol. Cancer Chemother. Pharmacol., 35:297- 303, 1995. Milross, C.G., Mason, K.A., Hunter, N.R., Chung, W.K., Peters, L.J., and Milas, L. Relationship of mitotic arrest and apoptosis to antitumor effect o f paclitaxel. J. Natl. Cancer Inst., 88:1308-1314, 1996. Muraki, J., Addonizio, J., Choudhury, M., Fischer, J., Eshghi, M., Davidian, M., Sharppiro, L., Wilmot, P., Nagamatsu, G., and Chiao, J. Establishment o f new human prostatic cancer cell (JCA-1). Invest. Urol., 36:79-84, 1990. Murphy, W.K., Fossella, F.V., Winn, R.J., Shin, D M., Hynes, H E., Gross, H.M., Davilla, E , Leimert, J., Dhingra, H., Raber, M.N., et al. Phase II study o f taxol in patients with untreated advanced non-small-cell lung cancer. J. Natl. Cancer Inst., 85:384-388, 1993. 204 Murphy, G.P., Natarajan, N., Pontes, J.E., Schmitz, R.L., Smart, C.R., Schmidt, J.D., and Mettlin, C. The national survey of prostate cancer in the United States by the .American College of Surgeons. Journal of Urology, 127:928-934, 1982. Nagabhushan, M., Miller, C M., Pretlow, T.P., Giaconia, J.M., Edgehouse, N.L., Schwartz, S., Kung, H.-J., de Vere White, R.W., Gumerlock, P H., Resnick, M l., Amini, S.B., and Pretlow, T.G. CWR22: the first human prostate cancer xenograft with strongly androgen- dependent and relapsed strains both in vivo and in soft agar. Cancer Res., 56:3042-3046, 1996. Narayan, P. and Dahiya, R Establishment and characterization o f a human primary prostatic adenocarcinoma cell line (ND-1). J. Urol. 148:1600-1604, 1992. Nemoto, R., Uchida, K., Shimazui, T., Hattori, K., Koiso, K., and Harada, M. Immunocytochemical demonstration of S phase cells by anti-bromodeoxyuridine monoclonal antibody in human prostate adenocarcinoma. J. Urol., 141:337-340, 1989. Neuwelt, E.A., Glasberg, M., Frenkel, E., and Bamett, P. Neurotoxicity o f chemotherapeutic agents after blood-brain barrier modification: neuropathological studies. Ann. Neurol., 14:316-324, 1983. Newhall, K.R., Isaacs, J.T., Wright, G.L., Jr. Dunning rat prostate tumors and cultured cells lines fail to express human prostate carcinoma-associated antigens. Prostate, 17:317-325, 1990. Newman, J. Epidermiology, diagnosis and treatment of prostate cancer. Radiol. Techno 1., 68:39-64, 1996. ODAC prefers three-hour taxol infusion. The Cancer Letter, 19:38, 1993. Oesterling, J., Fuks, Z , Lee, C.T., and Scher, H I. Cancer of the Prostate. In\ DeVita, V.T. Jr., Heilman, S., and Rosenberg, S.A. (eds.). Cancer, principles and practice of oncology. 5th edition, ppI322-1386. Philadelphia: Lippincott-Raven Publishers, 1997. Parker, S.L., Tong, T., Bolden, S., and Wingo, P.A. Cancer Statistics, CA Cancer J. Clinic., 47:5-27, 1997. Paulson, D.F. Impact of radical prostatectomy in the management o f clinically localized disease. Journal of Urology, 152:1826-1830, 1994. Peehl, D M., Erickson, E., Malspeis, L., Orr, J., Mayo, J., Camalier, R., Monks, A., Cronise, P., Pauli, K., and Grever, M.R. Prostate cancer: NCI drug discovery efforts. Proc. Am. Assoc. Cancer Res., 34:369. 1993. 205 Perez, C.A., Fair, W.R., and Ihde, D C. Carcinoma of prostate. In: DeVita, V.T. Jr.. Heilman, S. and Rosenberg, S. A. (eds). Cancer, principles and practice of oncology. 3rd edition. Lippincott, Philadelphia, 1989. Pizao, P.E., Smitskamp-Wilms, E., Van Ark-Otte, J., Beijnen, J.H., Peters, G.J., Pinedo, H.M., Giaccone, G. Antiproliferative activity of the topoisomerase I inhibitors topotecan and camptothecin, on sub- and postconfluent tumor cell cultures. Biochem. Pharmacol., 48:1145- 1154, 1994. Pretlow, T.G., Pelley, R.J., and Pretlow, T.P. Biochemistry of prostatic carcinoma. In Pretlow, T.G. and Pretlow T.P. (eds.) Biochemical and molecular aspects of selected cancers. 2: 169-237, San Diego: Academic Press, 1994. Pretlow, T.G., Wolman, S R., Micale, M.A., Pelley, R.J., Kursh, E D., Resnick, M l., Bodner, D R., Jacobberger, J.W., Delmoro, C M., Giaconia, J.M., and Pretlow, T.P. Xenografts of primaiy human prostatic carcinoma. J. Natl. Cancer Inst., 85:394-398, 1993, Pulkkinen, J O , Elomaa, L., Joensuu, H., Martikainen, P., Servomaa, K., and Grenman, R. Paclitaxel-induced apoptotic changes followed by time-lapse video microscopy in cell lines established from head and neck cancer. J. Cancer Res. Clin. Oncol., 122:214-218, 1996. Rangel, C., Matzkin, H., and Solo way, M.S. Experience with weekly doxorubicin (adriamycin) in hormone-refractory stage D2 prostate cancer. Urol., 39:577-582, 1992. Rao, S., Horwitz, S B ., and Ringel, I. Direct photoaffinity labeling of tubulin with taxol. J. Natl. Cancer Inst. 84:785-788, 1992. Robbins, K.T., Connors, K.M., Stomiolo, AM., Hanchett, C., and Hoffman, R.M. Sponge-gel-supported histoculture drug-response assay for head and neck cancer. Correlations with clinical response to cisplatin. Arch. Otolaryngol. Head Neck Surg., 120:288-292, 1994. Roberts, J R., Allison, D C , Donehower, R.C., and Rowinsky, E.K. Development of polyploidization in taxol-resistant human leukemia cells in vitro. Cancer Res., 50:710-716, 1990. Roth, B.J., Yeap, B Y., Wilding, G., Kasimis, B., McLeod, D. and Loehrer, P.J. Taxol in advanced, hormone-refractory carcinoma of the prostate, A phase II trial of the Eastern Cooperative Oncology Group. Cancer, 72:2457-2460, 1993. Rowinsky, E.K., Burke, P.J., Karp, J.E., Tucker, R.W., Ettinger, D.S., and Donehower, R.C. Phase I and pharmacodynamic study of taxol in refractory acute leukemias. Cancer Res., 49:4640-4647, 1989. 206 Saito, Y-, Milross, C.G., Hittelman, W.N., Li, D., Jibu, T., Peters, L.J., and Milas, L. Effect of radiation and paclitaxel on p53 expression in murine tumors sensitive or resistant to apoptosis induction. Inti. J. Rad. Oncol. Biol. Phy., 38:623-631, 1997. Saunders, D.E., Lawrence, W.D., Christensen, C., Wappler, N.L., Ruan, H., and Deppe, G. Paclitaxel-induced apoptosis in MCF-7 breast-cancer cells. Intl. J. Cancer, 70:214-220, 1997. Schifl^ P.B., Pant, J., and Horwitz, S B. Promotion o f microtubule assembly in vitro by taxol. Nature 277:665-667, 1979. Schmid, H.-P., McNeal, J.E., and Stamey, T.A. Observation on the doubling time o f prostate cancer. Cancer, 71:2031-2040, 1993. Schroeder, F.H., and Mackensen, S.J. Human prostatic adenoma and carcinoma in cell culture-effects of androgen-ffee culture medium. Invest. Urol., 12:176-181, 1974. Scrivner, D.L., Meyer, J.S., Rujanavech, N., Fathman, A., and Scully, T. Cell kinetics by bromodeoxyuridine labeling and deoxyribonucleic acid ploidy in prostatic carcinoma needle biopsies. J. Urol., 146:1034-1039, 1991. Shinkai, H., Takahashi, H., Miyamoto, K., Uchida, T., and Tokiwa, T. Comparative pharmacokinetics of BCRN8602, a new morpholino anthracycline, and adriamycin in tumor- earing mice. Cancer Chemother. Pharmacol., 38:417-424, 1996. Skehan, P., Storeng, R., Scudiero, D., Monks, A , McMahon, J., Vistica, D , Warren, J.T., Bokesch, H., Kenney, S., and Boyd, M.R. New colorimetric cytotoxicity assay for anticancer-drug screening. J. Natl. Cancer Inst., 82:1107-1112, 1990. Song, D., Hsu, L.-F., and Au, J.L.-S. Binding of taxol to plastic and glass containers and protein under in vitro conditions. J. Pharm. Sci., 85:29-31, 1996. Speth, P.A.J., Linssen, PCM, Holdrinet, R.S.G., and Haanen, C. Plasma and cellular adriamycin concentrations in patients with myeloma treated with ninety-six-hour continuous infusion. Clin. Pharmacol. Then, 41:661-665, 1987. Speth, P.A.J., van Hoesel, Q.G.C.M., and Haanen, C. Clinical pharmacokinetics of doxorubicin. Clin. Pharmacok., 15:15-31, 1988. Stamey, T.A, Bushby, S.R., and Bragonje, J. The concentration of trimethoprim in prostatic fluid: nonionic diffusion or active transport?. J. Inf. Dis., 128:s686-692, 1973. 207 Stierle, A., Strobel, G., and Stierle, D. Taxol and taxane production by Taxomyces andreanae, an endophytic fungus of Pacific yew. Science, 260:214-216, 1993. Strobel, T,. Swanson, L., Korsraeyer, S., and Cannistra, S.A. BAX enhances paclitaxel-induced apoptosis through a p53-independent pathway. Proc. Natl. Acad. Sci. USA, 93:14094-14099, 1996. Sumantran, V.N., Ealovega, M.W., Nunez, G., Clarke, M.F., and Wicha, M.S. Overexpression of Bcl-XS sensitizes MCF-7 cells to chemotherapy-induced apoptosis. Cancer Res., 55:2507-2510, 1995. Teicher, BA. Drug resistance in oncology. New York, Marcel Dekker, 1993. Terasaki, T., Iga, T., Sugiyama, Y., and Hanano, M. Pharmacokinetic study on the mechanism of tissue distribution of doxorubicin: interorgan and interspecies variation of tissue-to-plasma partion coefficeints in rats, rabbits, and guinea pigs. 73:1359-1363, 1984. Tewey, K.M., Rowe, T.C., Yang, L., Halligan, B.D., and Liu, L.F. Adriamycin-induced DNA damage mediated by mammalian DNA topoisomerase II. Science, 226:466-468, 1984. Timour, Q , Nony, P., Lang, J., Lakhal, M., Trillet, V., and Faucon, G. Doxorubicin concentration time course in the myocardium after single administration to the dog. Cancer Chemothr. Pharmacol., 20:267-269, 1987. Trielli, M.O., Andreassen, PR., Lacroix, F.B., and Margolis, R.L. Differential taxol-dependent arrest of transformed and nontransformed cells in the G1 phase of the cell cycle, and specific-related mortality of transformed cells. J. Cell Biol., 135:689-700, 1996. Triton, T.R. and Yee, G. The anticancer agent adriamycin can be actively cytotoxic without entering cells. Science, 217:248-250, 1982. Vailancourt, L., Ttu, B., Fradet, Y., Dupont, A., Gomez, J., Cusan, L., Suburu, E.R,. Diamond, P., Candas, B., and Labrie, F. Effect of neoadjuvant endocrine therapy (combined androgen blockade) on normal prostate and prostatic carcinoma. Am. J. Surg. Pathol., 20:86-93, 1996. van Steenbrugge, G.J., Groen, M., Bolt-de Vries, J., Romijn, J.C., and Schroeder, F.H. Human prostate cancer (PC-82) in nude mice: a model to study androgen regulated tumor growth. Prog. Clin. Biol. Res., 185A:23-50, 1985. 208 van Weerden, W.M., de Ridder, C.M.A., Verdaasdonk, C.L., Romijn, J C , van der Kwast, T.H., Schroder, F.H., and van Steenbmgge, G J Animal model: Development of seven new human prostate tumor xenograft models and their histopathological characterization. Am. J. Pathol., 149:1055-1062. Vesico, R.A., Redfem, C.H., Nelson, T.J., Ugoretz, S., Stem, P.H., and Hoffman, R.M. In vivo-like drug responses of human tumors growing in three-dimensional gel-supported primary culture. Proc. Natl. Acad. Sci. USA., 84:5029-5033, 1987. Wahl, A.F., Donaldson, K.L., Fairchild, C., Lee, F.Y., Foster, S.A., Demers, G.W., and Galloway D A. Loss of normal p53 function confers sensitization to Taxol by increasing G2/M arrest and apoptosis. Nature Med., 2:72-79, 1996. Wainstein, M.A.„ He, F., Robinson, D., Kung, H.-J., Schwartz, S., Giaconia, J.M., Edgehouse, N.L., Pretlow, T.P., Bodner, D R., Kursh, E D., Resnick, M.I., Seftel, A., and Pretlow, T.G. CWR22: androgen-dependent xenograft model derived from a primary human prostatic carcinoma. Cancer Res., 54:6049-6052, 1994. Wall, M E. and Wani, M.C. A preliminary report dealing only with the isolation was reported at the 153 rd National Meeting of the American Chemical Society, Miami Beach, M-006, 1967. Walsh, P C Why make an early diagnosis of prostate cancer. J. Urol., 147:853-854, 1992. Wang, M.C , Valenzuela, L.A., Murphy, G.P., and Chu, T.M. Purification of a human prostate specific antigen. Invest. Urol., 17:159-163, 1979. Wani, M.C., Taylor, H.L., Wall, M.E., Coggon, P., and McPhail, A T Plant antitumor agents. VI. The isolation and structure of taxol, a novel antileukemic and antitumor agent from Taxus brevifolia. J. Am. Chem. Soc., 93:2325-2327, 1971. Waters, D.J. and Bostwick, D.G. Prostatic intraepithélial neoplasia occurs spontaneously in the canine prostate. J. Urol., 157:713-716, 1997. Weedon, D., Searle, J., and Kerr, J.F. Apoptosis. Its nature and implications for dermatopathology. Am. J. Dermatopathol., 1:133-144, 1979. Wientjes, M.G., Pretlow, T.G., Badalament, R.A., Burgers, J.K., and Au, J.L.-S. Histocultures of human prostate tissues for pharmacologic evaluation. J. Urol., 153:1299- 1302, 1995. 209 Wiemik, P.H., Schwartz, E.L., Einzig, A., Strauman, J.J., Lipton, R.B., and Dutcher, J.P Phase I trial of taxol given as a 24-hour infusion every 21 days; Responses observed in metastatic melanoma. J. Clin. Oncol., 5:1232-1239, 1987. Wilson, W.H., Berg, S.L., Bryant, G., Wittes, R.E., Bates, S., Fojo, A , Steinberg, S.M., GoldspieL, B.R., Herdt, J., O ’Shaughnessy, J., Bails, P.M., and Chabner, B.A. Paclitaxel in doxorubicin-refractory or mitoxantrone-refractory breast cancer: a phase I/II trial of 96-hour infusion. J. Clin. Oncol. 12:1621-1629, 1994. Woods, C M., Zhu, J., McQueney, P.A., Bollag, D., and Lazarides, E. Taxol-induced mitotic block triggers rapid onset of a p53-independent apoptotic pathway. Mol. Med., 1:506-526, 1995. 2 1 0