Principles and Practice of·
`
`nco ogy
`
`Derek Raghavan, MBBS, PhD, FRACP, FACP
`Chief, Departments of Solid Tumor Oncology
`and Investigational Therapeutics
`Roswell Park Cancer Institute and
`Professor of Medicine and Urology
`State University of New York at Buffalo
`Buffalo, New York
`
`Howard I. Scher, MD
`Chief, Genitourinary Oncology Service
`Associate .Attending Physician
`Division of Solid Tumor Oncology
`Department of Medicine
`Memorial Sloan-Kettering Cancer Center
`New York, New York
`
`Steven A. Leibel, MD
`Vice Chairman and Clinical Director
`Attending Radiation Oncologist
`Department of Radiation Oncology
`Memorial Sloan-Kettering Cancer Center
`New York, New York
`
`Paul Lange, MD, FACS
`Professor and Chair .
`Department of Urology
`University of Washington
`Seattle, Washington
`
`With 226 Additional Contributors
`
`r Lippincott - Raven
`
`_ , P U B L
`
`I S H E R S
`
`Philadelphia • New York
`
`West-Ward Exhibit 1032
`Dinney-Raghavan 1997
`Page 001
`
`

`

`- 1
`
`Developmental Editor: Eileen Wolfberg Jackson
`Project Editor: EUen M. Campbell
`Production Manager. C11TCn Erlichman
`Production Coordinator: MaryClare Malady
`Design Coordinator: Doug Smock
`Indexer. Steve Sorenson
`Compositor. Maryland Composition Company, Inc.
`Primer: Courier Book Company/Westford
`
`Copyright Cl I 997 by Lippincott-Raven Publishm. All rights reserved. This book is
`protected by copyright. No pan of it may be reproduced, stored in a retrieval system,
`or transmitted, in any form or by any means-electronic, mechanical, photocopy,
`recording, or otherwise-without the prior written peimission of the publisher, except
`for brief quotations embodied in critical articles and reviews. Printed in the United
`States of America. For information write Lippincott-Raven Publishers, 227 East
`Washington Square, Philadelphia, PA 19106-3780.
`
`Materials appearing in this book prepareed by individuals as part of their official duties
`as U.S. Gov~ent employees are not covered by the above-mentioned copyright.
`
`Library of Congress Cataloging-in-Publication Data
`Principles and practice of genitourinary oncology I Derek Raghavan ...
`[et al.] ; with 226 additional contributors.
`p. cm.
`Includes bibliographical references and index.
`ISBN 0-397-51458-1 (alk. paper)
`I. Raghavan, Derek.
`I. Genitourinary organs- Cancer.
`I. Urogenital Neoplasms. WJ 160 P9573 1996)
`(DNLM:
`RC280.U74P746 1996
`616.99'26-dc20
`DNLM/DLC
`for Library of Congress
`
`96-8893
`CIP
`
`Care has been taken to confirm the accuracy of the information presented and 10
`describe generally accepted practices. However, the authors, editors, and publisher are
`not responsible for errors or omissions or for any consequences from application of the
`information in ibis book and make no warranty, express or implied, with respect to the
`contents of the publication.
`The authors, editors, and publisher have exerted every effort t.o ensure that drug
`selection and dosage set forth in this text are in accordance with current
`recommendations and prac1ice at the time of publication. However, in view of ongoing
`research, changes in government regulations, and the constant flow of information
`relating to drug therapy and drug reactions, the reader is urged to check the package
`insert for each drug for any change in indications and dosage and for added warnings
`and precautions. Th.is is panicularly imponant when the recommended ageot is a new
`or infrequently employed drug.
`Some drugs and medical devices presented in this publication have Food and Drug
`Administration (FDA) clearance for limited use in restricted research settings. It is the
`responsibility of the health care provider to ascertain the FDA status of each drug or
`device planned for use in their clinical practice.
`
`9 8 7 6 5 4 3 2 I
`
`West-Ward Exhibit 1032
`Dinney-Raghavan 1997
`Page 002
`
`

`

`...
`
`CHAPTER 2
`
`-
`
`Principles and Practice of Genitourinary Oncology, edited by
`Derck Roghavan, Howard L Scher, Steven A. Leibel, and Paul H.
`Lange. l..ippiocon-R~ven Publishe"l, Philadelphia, C 1997.
`
`Biology of Metastasis: Studies in Renal Cancer
`
`Colin P. N. Dinney and Isaiah]. Fidler
`
`For many patients with cancer, metastasis has occurred by the
`time of diagnosis. The major barrier to the effective treatment
`of their metastases is the biologic heterogeneity of neoplastic
`cells. This heterogeneity is exhibited in numerous genetic, bio(cid:173)
`chemical, immunologic, and biologic characteristics, such as
`cell-surface receptors, enzymes, karyotype, cell morphology,
`growth properties, sensitivities to various therapeutic agents,
`and ability to invade and produce metastasis.1- 7
`Understanding the mechanisms responsible for the develop(cid:173)
`ment of the biologic heterogeneity of cancer and the process
`by which tumor cells invade local tissues and spread to distant
`organs is a primary goal of cancer research. This · review de(cid:173)
`scribes the way in which the development of relevant preclinical
`models for human renal cell carcinoma has facilitated the study
`of the biology of metastasis.
`
`metastasis, partly because of the elimination of disseminating
`tumor cells that fail to complete any step in metastasis. 12 Using
`radiolabeled B 16 melanoma cells, researchers observed that, by
`24 hours after entry of the cells into the circulation, < 1 % of
`the cells were still viable, and <0.01 % of tumor cells placed
`into the circulation survived to produce metastases. t2 These
`observations brought into question whether the development of
`metastases represents the chance survival and growth of neo(cid:173)
`plastic cells or the selective growth of unique subpopulations
`of malignant cells endowed with special properties. In other
`words, can any cell growing in a primary neoplasm produce
`metastases, or do only specific and unique cells possess the
`appropriate properties that enable them to metastasize? Most
`recent data conclude that neoplasms are biologically heteregen(cid:173)
`ous and the process of metastasis is i_?deed selective.
`
`PATHOGENESIS OF A METASTASIS
`
`ROLE OF THE ORGAN ENVIRONMENT
`
`The process of cancer metastasis consists of a series of se(cid:173)
`quential steps, each of which can be rate limiting.6 After cellular
`transfonnation, either unicellular or multicellular, growth of
`neoplastic cells is progressive, but extensive vascularization
`must occur if a tumor mass is to exceed 2 mm in diameter. 8
`Local invasion of the host stroma by some tumor cells, the next
`step, can occur by several mechanisms that are not mutually
`exclusive.9 For example, thin-walled venules and lymphatic
`channels, which offer little resistance to penetration by tumor
`cells, provide the most common pathways for tumor cell entry
`irito the circulation. Detachment of small tumor cell aggregates
`and embolization occur next The tumor cells that survive the
`circulation must lodge in the capillary beds of organs. Extrava(cid:173)
`sation follows, probably by the same mechanisms that influence
`initial invasion. Proliferation within the organ parenchyma
`completes the metastatic process (Fig. 2-1). To produce detecta(cid:173)
`ble lesions, the metastases must develop a vascular network,
`evade the host immune.system,2 and respond to organ-specific
`factors that influence their growth. io. t t Once they do so, meta(cid:173)
`static cells can invade host stroma, penetrate blood vessels, and
`enter the circulation to produce further metastases, the so-called
`i
`" metastasis of metastases. " 4·s
`Only a few cells io a primary tumor can give rise to clinical
`
`Clinical observations of cancer patients and preclinical stud(cid:173)
`ies with experimental rodent tumors have demonstrated that
`certain tumors produce metastasis to specific organs -indepen(cid:173)
`dent of vascular anatomy, rate of blood flow, or the number of
`tumor cells delivered to each organ. For instance, the distribu(cid:173)
`tion and fate of hematogenously disseminated, radiolabeled
`melanoma cells in experimental rodent systems amply demon(cid:173)
`strate that tumor cells reach the microvasculature of many or(cid:173)
`gans.12- 15 Extravasation into the organ parenchyma and prolif(cid:173)
`eration of tumor cells occur only in certain organs, however.
`Therefore, the mere presence of viable tumor cells in a particular
`organ does not always predict that the cells will proliferate to
`produce metastases. 10•13- 16
`The search for the mechanisms that regulate the pattern of
`7
`metastasis began more than a century ago when, in 1889, t
`Paget questioned whether the distribution of metastases was
`due to chance. Paget therefore anaJyzed 735 autopsy records
`of women with breast cancer, and the nonrandom pattern of
`visceral metastases suggested to him that the process was not
`due to chance, but rather; certain tumor cells (the "seed") had
`a specific affinity for the milieu of certain organs (the ''soil").
`Metastases resulted only when the appropriate seed and soil
`were matched.17
`
`17
`
`~ ii
`
`West-Ward Exhibit 1032
`Dinney-Raghavan 1997
`Page 003
`
`

`

`18 I CHAPTER 2
`
`Primary Malignant
`Neoplasm
`
`Vascularlzatlon
`
`Invasion
`
`~~-----~~; ~
`
`Interaction with:
`Lymphatics
`Platelets, Lymphocytes and
`Venules
`Caplllarles Other Blood Components
`
`r Extravasatlon
`
`-~'J
`
`Adherence of
`Tumor Cells
`
`Arrest In Organs
`
`Transpor.t
`
`Establishment of
`Microenvlronment
`
`Lung
`
`FIG. 2-1. The pathogenesis of cancer metastasis. To produce clinically relevant metastases, tumor
`cells In a primary neoplasm must complete a series of sequential, selective, and rate·limiting steps.
`
`Metastases
`
`Experimental data supporting the ''seed and soil'' hypothesis
`came from studies on the preferential invasion and growth of
`B 16 melanoma metastases in specific organs. 18 When the B 16
`melanoma cells were injected into the circulation of syngeneic
`mice, tumors developed in the lungs and in fragments of lung or
`ovarian tissue implanted intramuscularly. In contrast, metastatic
`lesions did not develop in renal tissue implanted as a control
`organ or at the site of surgical trauma. 18 This study confirmed
`that the production of metastasis was determined not only by
`the characteristics of the neoplastic cells, but also by the micro(cid:173)
`environrneot of the host tissue. In vitro experiments demonstrat(cid:173)
`ing organ-selective adhesion, invasion, and growth also support
`Paget's hypothesis. 17 Using the B16 melanoma system, cells
`with increased capacity for organ adhesion, invasion, and
`growth have been isolated.10·19- 24 Moreover, experiments. with
`organ tissue-derived soluble growth factors indicate that soil
`factors can have profound effects on certain tumor cell subpopu(cid:173)
`lations.10
`Although clinical observations have suggested that carcino(cid:173)
`mas frequently metastasize through the lymphatic system, and
`malignant tumors of mesenchymal origin more often spread by
`the hematogenous route, the presence of numerous venolym(cid:173)
`phatic anastomoses invalidates this belief.25 The circulatory
`anatomy influences the dissemination of many malignant ceUs;
`however, it cannot, as Ewing proposed,26 fully explain the pat(cid:173)
`terns of distribution of numerous tumors. Ethical considerations
`rule out the experimental analysis of cancer metastasis, as stud(cid:173)
`ied in laboratory animals, in patients. The introduction of perito(cid:173)
`neovenous shunts for palliation of malignant ascites has, how(cid:173)
`ever, provided an opportunity to study some factors affecting
`
`metastatic spread in humans.27·28 Good palliation with minimal
`complications was reported for 29 patients with various neo·
`plasms. The autopsy findings in 15 patients substantiated the
`clinical observations that the shunts do not significantly increase
`the risk of visceral organ metastasis. In fact, despite continuous
`entry of hundreds of millions of tumor cells into the circulation,
`metastases in the lung (the first capillary bed encountered) were
`unusual.27·28 These results provide compelling verification of
`the seed and soil hypothesis.
`
`METASTATIC HETEROGENEITY
`
`Populations of cells that differ in metastatic potential have
`been isolated from the parent neoplasm, a finding that supports
`the hypothesis that not all the cells in a primary tumor can
`successfully disseminate. Two general approaches have been
`used. In the first, me.tastatic cells are selected in vivo: tumor
`cells are implanted into mice, and metastatic lesions are har(cid:173)
`vested. The cells recovered can be expanded in culture or used
`immediately to repeat the process. The cycle is repeated several
`times, and the behavior of the cycled cells is compared with
`that of the cells of the parent tumor. This procedure was origi(cid:173)
`nally used to isolate the Bl6·Fl0 line from the wild-type BI6
`melanoma, 20 and it has since also been successful in producing
`tumor cell lines with increased metastatic capacity from many
`other experimental tumors. 6 In the second approach, cells are
`selected for the enhanced expression of a phenotype believed
`to be important in one or another step of the metastatic sequence,
`and then they are tested in the appropriate host to detennine
`
`·l
`
`West-Ward Exhibit 1032
`Dinney-Raghavan 1997
`Page 004
`
`

`

`BIOLOGY OF METASTASIS: STUDIES IN RENAL CANCER
`
`/ 19
`
`whether concomitant metastatic potential has been increased or
`decreased.
`The first experimental proof of metastatic heterogeneity in
`neoplasms was provided by Fidler.and Kripke in 1977, in stud(cid:173)
`ies with mouse B 16 melanoma.27 Using the modified fluctua(cid:173)
`tion assay of Luria and Delbruck,28 Fidler and Kripke showed
`that distinct tumor cell clones, each derived from individual
`cells isolated from the parent tumor, varied dramatically in their
`potential to fonn pulmonary nodules after intravenous inocula(cid:173)
`tion into syngeneic mice. Control subcloning procedures dem(cid:173)
`onstrated that the observed diversity was not a consequence of
`the cloning procedure.27
`To exclude the possibility that the metastatic heterogeneity
`found in the B 16 melanoma cells may have been introduced as
`a result of lengthy cultivation, ~pke studied the biologic and
`metastatic heterogeneity of a mouse melanoma induced in C:iH
`mice by long-term exposure to ultraviolet B radiation and paint(cid:173)
`ing with croton oil.29 One mouse thus treated developed a mela(cid:173)
`noma designated K-1735. The original K-1735 melanoma was
`established in culture and immediately cloned.30•31 The clones
`differed from each other and from the parent tumor in their
`ability to produce lung metastases. Moreover, the metastases
`demonstrated significant variability in their size and pigmenta(cid:173)
`tion. Metastases to the lymph nodes, brain, heart, liver, and skin
`were found in addition to lung metastases; those growing in
`the brain were unifonnly melanotic, whereas those growing in
`31
`other organs generally were not.30
`·
`
`MODELS FOR HUMAN RENAL CANCER
`METASTASIS
`
`The concept that neoplasms are heterogeneous and contain
`subpopulations of cells with different patterns of biologic be(cid:173)
`havior, including metastatic ability, is no longer controversial.
`Studies in most rodent tumor systems have also demonstrated
`that metastasis is a selective process,20•32•33 metastases can have
`a clonal origin, 34-3s metastases can develop from the expansion
`of a single cell,38 and the host organ microenvironment can
`profoundly influence the growth of metastatic tumor cells:6 Al(cid:173)
`though most data on the metastatic heterogeneity of neoplasms
`and on host-tumor cell interactions during the metastatic process
`have been derived from studies in nonhuman systems, evidence
`is now accumulating about the biology of metastasis by cells
`isolated from surgical specimens of human cancers. In large
`measure, this accumulation of evidence has been due to recent
`improvements in the use of in vivo models for the isolation of
`metastatic subpopulations of human cancer cells and for testing
`of their metastatic potential. With the discovery of the athymic
`T cell-deficient nude mouse, its use as a recipient for human
`xenografts, and its adaptation for the study of human cancer
`growth and metastasis, 39 many aspects of the biology of human
`cancer can now be studied. Reports on the metastatic ability of
`human tumors subsequent to implantation into nude mice have
`
`recently increased,39•40 and several reports have concluded that
`the metastatic capacity of human tumor cells in nude mice is
`42 Fidler
`influenced by variations in experimental techniques.41
`•
`has shown that the capacity to produce metastasis from human
`tumor cell lines of long duration43•44 or of recent origin45
`-48
`depends both on the injection site an~ on the intrinsic properties
`of the cells. 39
`
`An appropriate model for studies of human cancer metastasis
`must meet two rigid demands: it must use metastatic cells
`(seed), and it must grow in the relevant organ environment
`(soil). To study the properties of metastatic subpopulations from
`surgical specimens of human cancers, methods for their isola(cid:173)
`tion, propagation, and testing must be developed.
`The study of the biology of human renal cell carcinoma
`(HR.CC) has been facilitated by the development of relevant in
`vivo models. Orthotopic ilnplantation of HR.CC cells allows for
`the isolation of tumor cells capable of producing metastasis to
`the lymph nodes, lung, and other organs. Furthermore, this sys(cid:173)
`tem allows for analysis of the properties that identify metastatic
`HR.CC under in vivo conditions.
`·
`
`ESTABLISHMENT OF RENAL CANCER CELL
`LINES
`
`We have used two HRCC cell lines, the SN12 and KG-2, to
`study the metastatic potential of HR.CC. Naito and associates
`established the SN12 renal cell carcinoma line from a radical
`nephrectomy specimen of a renal cell carcinoma in a 43-year(cid:173)
`old man.49 A cell suspension with high viability was divided
`into three aliquots. One was used to produce a tissue culture line
`designated SN12C. The second was injected subcutaneously in
`the lateral aspect of the anterior thoracic wall of several nude
`mice. Two months _later, a subcutaneous tumor was excised and
`dissociated enzymatically. The cells were established in culture,
`and the line was designated SN12Sl. The third aliquot was
`injected into the kidneys of nude mice. Several weeks later, a
`large tumor was harvested: In one mouse, a grossly visible
`tumor nodule was found in the liver, and the peritoneum con(cid:173)
`tained ascitic fluid. Tumor cells isolated from the kidney tumor
`and the liver nodule were established as individual cell lines in
`culture and designated lines SN12Kl an!! SN12Ll , respec(cid:173)
`tively. The cell line established from the ascitic cells was desig(cid:173)
`nated SN12Al.
`The cells of the HRCC lines grew on plastic as monolayer
`cultures with similar morphologic features. The hwnan origin
`of all five cell lines was ascertained by detailed karyotypic
`analysis and isoenzyme determinations. No contamination with
`mouse cells was fpund in any of the lines.
`Similarly, the KG-2 cell line was isolated from a surgical
`specimen subsequent to radical nephrectomy in a 58-year-old
`man. Cells were injected into the subcutis, kidney, cecal wall,
`and spleen of nude mice. Tumor grew in the subcutis and kid(cid:173)
`ney, and only kidney tumors produced distant metastases.
`
`BIOLOGIC BEHAVIOR OF THE DIFFERENT
`HRCC LINES
`
`Subsequent experiments showed that the five HR.CC SN12
`cell lines consisted of cells with different biologic properties,
`including growth in vivo at ectopic and orthotopic sites and
`production of experimental and spontaneous metastasis (Table
`2-1).49
`The in vitro tumor cell doubling time did not differ signifi(cid:173)
`cantly among the cell lines. All lines were tumorigenic at subcu(cid:173)
`taneous sites, but SN12Ll cells (developed from a metastasis)
`grew at the fastest rate, whereas SN12Al cells (ascites) were the
`
`West-Ward Exhibit 1032
`Dinney-Raghavan 1997
`Page 005
`
`

`

`20
`
`I CHAPTER 2
`
`HRCC
`
`In vitro·
`
`SN12-culture
`SN12-skin
`SN12-kidney
`SN12-kidney-liver
`SN 12-kidney-ascites
`KG-2
`
`Best
`++
`++
`++
`++
`++
`
`Table 2-1. Biologic properties of HRCC lines
`
`Growth
`
`Subcutaneous
`tissue
`++
`Best
`++
`Best
`++
`++
`
`Kidney
`+++
`+++
`+++
`+++
`++
`+++
`
`From
`in vitro
`
`+
`+
`
`Metastasis
`
`From
`subcutaneous
`+
`Highest
`+
`+
`
`From kidney
`++
`++
`+++
`Highest
`
`slowest. A large part of the subcutaneous tumor was necrotic,
`irrespective of the tumor cell line and its growth rate.
`Injection of the SN12 HRCC cells into ectopic organs, such
`as the subcutis, peritoneum, or the spleen, produced a high rate
`of turnorigenicity but not distant metastases. This finding was
`remarkable because intrasplenic injection has been shown to
`be optimal for the study of metastatic behavior of human colon
`carcinoma cells.4s-<1s The injection of HRCC cells into the kid(cid:173)
`ney environment produced the highest incidence of tumorige(cid:173)
`nicity regardless of the cell line, a finding suggesting that the
`kidney is the best environment for expression of tumorigenicity
`of HR.CC cells. The growth rate of the HRCC cells in the kidney
`differed among the: lines, with SN12Ll cells producing the fast(cid:173)
`est-growing tumors. The slowest-gr-owing tumors were pro·
`duced by cells of the SN! 2C (tissue culture) or SNl 2Al (ascites
`derived) lines. The production of lung metastasis from kidney
`tumors varied among the lines. A high incidence of Jung metas(cid:173)
`tasis was observed with cells of line SN12Ll (derived from a
`liver metastasis), whereas cells of the SN12Al line (derived
`from ascites) produced few lung metastases. Cells from the
`SN 12L1 line were also highly invasive and produced metastases
`in the omental and mesenteric lymph nodes, pancreas, dia(cid:173)
`phragm, and seminal vesicles.
`These data suggest a correlation between the method for iso·
`lation of the HRCC lines and their biologic behavior. The
`SN12C line was isolated by culture, and these cells exhibited
`the fastest in vitro growth rate. The SN12Sl cells were isolated
`from a subcutaneous tumor; they grew well in the subcutis to
`produce a high incidence of spontaneous lung metastasis. The
`SN12Kl and SN12Ll cells were isolated from kidney and liver
`tumors, respectively, and grew readily in the kidney; these cells
`produced a high incidence of spontaneous lung metastasis. The
`SN12Al cells isolated from ascitic fluid were the least malig(cid:173)
`nant of the five lines.
`
`ORTHOTOPIC VERSUS ECTOPIC
`IMPLANTATION FOR STUDIES OF
`SPONTANEOUS METASTASIS
`
`The implantation of HRCC cells into the subcutis of nude
`mice has resulted in tumors that failed to metastasize.50- 57
`Therefore, even ifHRCC cells growing subcutaneously in nude
`mice maintain their original morphologic and biochemical char(cid:173)
`acteristics, whether the subcutaneous environment of the nude
`mouse is appropriate for studies of metastasis is questionable.
`The ideal in vivo model for studying this disease should allow
`
`the interaction of the tumor cells with their relevant organ envi·
`ronment, the kidney. Naito and associates specifically examined
`this issue by injecting SN12Ll cells (isolated in culture from
`a liver metastasis produced in a nude mouse by cells from the
`surgical specimen growing in the kidney) into the subcutis and
`the kidney.58 Tumor growth was studied histologically, and the
`production of metastasis was recorded.
`Two weeks after injection of tumor cells, palpable nodules
`were found at the subcutaneous injection site. These tumors
`were surrounded by a fibrous capsule. Although the tumors
`were well vascularized, small areas of necrosis were also found.
`At this time, the HRCC cells implanted in the kidney were
`within the renal subcapsular space. Prominent areas of invasive
`growth· into the renal parencbyma were also noted, however.
`These tumors were free of necrosis and were not encapsulated.
`By 4 or 6 weeks after injection, the tumors growing subcuta(cid:173)
`neously were encapsulated by a fibrous connective tissue. Al(cid:173)
`though the periphery of the tumor mass was well vascularized,
`central areas of necrosis were prominent. The tumor growing
`in the kidney was highly iflvasive, with a moderate degree of
`vascularization. The tumors were neither necrotic nor encapsu(cid:173)
`lated.
`Six weeks after tumor cells were injected into the kidney,
`most of the renal parenchyma was destroyed by the growing
`tumor. Grossly apparent pulmonary metastases were found, and
`the diaphragm, mesenteric and omental lymph nodes, and pan(cid:173)
`creas contained growing tumors.
`Experiments using the KG-2 cell line also found differences
`between tumors arising from orthotopic implantation and those
`arising from ectopic implantation. For instance, the morpho(cid:173)
`logic pattern of HRCC growing in the subcutis was different
`from that of HRCC growing in the kidney.50 The KG-2 HRCC
`cell line was highly vascularized (in the patient) and consisted
`of a clear cell carcinoma with a 20% sarcomatoid component.
`This histologic pattern was maintained in KG-2 tumors growing
`in the kidney of nude mice. Metastases also developed in the
`lung and lymph nodes, so the KG-2 behavior in nude mice
`recapitulated the clinical behavior. Within the kidney, the tumor
`retained a predominant clear cell histologic pattern, with only
`10% of the tumor composed of a sarcomatoid component. The
`tumor was well vasculariz.ed and grew rapidly, with no evidence
`of central· necrosis or fibrous encapsulation. In contrast, the KG-
`2 tumor in the subcutis consisted primarily of granular cells
`(70%), with a minor component of clear (20%) and sarcomatoid
`(10%) cells. These tumors were not vascularized and had central
`necrosis and a surrounding fibrous pseudocapsule. Metastasis
`occurred only in nude mice after orthotopic implantation of
`cells.58
`
`West-Ward Exhibit 1032
`Dinney-Raghavan 1997
`Page 006
`
`

`

`~ . .--==-u~-=~~~-========~~=------========---~==~-------=--i
`
`I
`J
`
`BIOLOGY OF METASTASIS: STUDIES IN RENAL CANCER
`
`/
`
`21
`
`GROWTH IN DIFFERENT ORGAN
`ENVIRONMENTS SELECTING FOR DIFFERENT
`SUBPOPULATIONS OF HRCC CELLS
`
`The "ata discussed in the preceding paragraphs demonstrate
`that cell lines derived from HRCC by growth in the subcutis or
`kidney differ in their biologic behavior. Moreover, the biologic
`behavior of HRCC cells isolated directly in culture differed
`from that of HRCC cells first grown in nude mice and then
`established in culture.
`To determine whether the different methods of isolation
`yielded different subpopulations of HR.CC cells, the five SN12
`lines were subjected to detailed cytogenetic analysis. Celis har(cid:173)
`vested from semiconfluent cultures were prepared for G- and Q(cid:173)
`bandings by well-established techniques.59 Samples were also
`stained with Giemsa to detennine modal chromosome nwnber
`and the presence of double-minute chromosomes. Complete
`karyotypes were prepared from at least 50 metaphase spreads
`of each cell line, and clonality of a marker (altered) chromosome
`was further examined w.hen it was present in more than one
`cell. This detailed cytogenetic analysis indicated that the SN12
`lines each consist of unique subpopulations of cells derived
`from the original HR.CC tumor.
`Today, with the development of molecular techniques, we
`have much more sophisticated methods for establishing clon(cid:173)
`ality and clonal dominance within a tumor. Viral and plasmid
`vectors can be used to insert selectable genetic markers into
`tumor cells, and clones with unique insertion sites can be
`tracked as the tumor grows and metastasizes. This technique
`has been used to determine whether the tumor clones that domi(cid:173)
`nate an SN12C tumor and its metastasis are the same and
`whether clonal dominance is influenced by the microenviron(cid:173)
`ment.60 Transduction of the SN12C human renal ceU carcinoma
`line with the NEOR gene produced a genetically "tagged" cell
`population within which individual clones can be ideptified if
`they dominate the tumor during its growth in vivo. These experi(cid:173)
`ments showed that clonal dominance is influenced by the organ
`environment in which the primary tumor grows, that is, distinct
`clones dominated in the kidney, colon, and subcutaneous sites.
`In addition, tumors grown in the orthotopic site (kidney) were
`all populated by the same dominant clones, and each distant
`visceral metastasis retained the same c!onality. SN12C NEOR
`cells grown in an epithelial, ectopic site (colon) produced tu(cid:173)
`mors with unique dominant clones, and their visceral metastases
`retained the dominant pattern ex.pressed by the parent tumor
`from which they were derived. In contt:ast, SN12C tumors
`growing subcutaneously showed a random pattern of clonal
`dominance in both their primary and metastatic sites.
`Some progress has been made in understanding the molecular
`mechanisms of the heterogeneity of human neoplasms in vivo.
`These experiments have reinforced the concept that metastasis
`depends on the interaction of tumor cells with the appropriate
`microenvironment. For instance, the inability of the KG-2 cells
`to produce metastasis from a subcutaneous tumor may have
`been due to multiple additive factors, among which is vasculari(cid:173)
`zation. The growth of tumors and production of metastasis are
`known to depend on angiogenesis.61 In many rodent and human
`cancers, the probability of metastasis increases after vasculari(cid:173)
`zation.61 Of several diffusible cytokines that are mitogenic to
`vascular endothelium, one produc~d by many HR.CC, basic fi(cid:173)
`broblast growth factor (bFGF), has been shown to stimulate
`
`the proliferation and migration of capillary endothelial cells. 50
`Gohji and associates found that, under in vitro conditions, KG-
`2 cells stained positive for bFGF. The highly metastatic KG-
`2 LM6 cell lines were most positive. Angiogenesis, however,
`depended on the organ environment Regardless of which of
`the KG-2 lines were injected, subcutaneous tumors were not
`well vascularized, whereas kidney tumors were. Whether this
`characteristic was due to differences in levels of bFGF (or other
`factors) is uncertain.
`Singh and associates62 found that, after the orthotopic and ec(cid:173)
`topic implantation of both SN l 2C and the highly metastatic
`SN l 2PM6, renal tumors were highly vascularized, as revealed by
`immunohistochemistry using antibodies to factor VIII, whereas
`subcutaneous tumors were not. The expression ofbFGF in renal
`tumors was 10- to 20-fold that in subcutaneous tumors. Similar
`data were obtained at the protein level by using fluorescence-acti(cid:173)
`vated cell sorting (FACS), immunohistochemistry, and enzyme(cid:173)
`linked irnmunosorbent assay. bFGF was detected in the urine of
`mice with HR.CC tumors growing in the kidney but not the subcu(cid:173)
`tis. The serum level ofbFGF was two- to threefold greater in mice
`with HR.CC tumors in the kidneys. These data strengthen the ar(cid:173)
`gument that the organ microenvironment influences the expres(cid:173)
`sion level of bFGF in HRCC.
`Gohji and associates also showed that highly metastatic renal
`cell carcinoma cells expressed higher levels of both 72-kd colla(cid:173)
`genase type IV and urokinase, and these researchers observed
`a direct correlation between the production of these enzymes
`and metastatic potential.63 They examined the effect of the mi(cid:173)
`croenvironment on the expression of collagenase type IV and
`found that the 72-kd collagenase type IV level in the culture
`supematants of KG-2 cells was increased by their cultivation
`with mouse kidney or lung fibroblasts. In contrast, cocultivation
`of KG-2 cells with mouse skin fibroblasts significantly reduced
`collagenase type IV activity. Similar results were obtained by
`culturing KG-2 cells in the medium conditioned by different
`mouse fibroblasts. Gohji and colleagues i,nvestigated the effects
`on KG-2 cells of cytokines and growth factors known to be
`produced by fibroblasts of various origins. Cytokines and
`growth factors tested, bFGF, hepatocyte growth factor, and
`transforming growth factor-beta, (TGF-,81) often stimulated
`collagenase type IV expression by the cultured KG-2 cells. Par(cid:173)
`allel immunohistochemical analyses revealed that mouse kid(cid:173)
`ney and lung fibroblasts produced higher levels of TGF-(31 than
`did skin fibroblasts. These results indicated that collagenase
`type IV production by KG-2 renal cell carcinoma cells is influ(cid:173)
`enced by the organ microenvironrnent