Tobacco Documents Online

tdeas in Pathology Pivotal Role of Increased Cell Proliferation in Human Carcinogenesis Samuel M. Cohen, David T. Purtilo, and Leon B. Ellvvein DepartmeN:s of Pathology and Microbiology and of Pediatrics, and the Eppley Institute for Cancer Research, University of Nebraska Medical Center, Omaha, Nebraska Cancer develops secondary to multiple genetic events. Each time a cell divides there is a rare chance that a genetic error related to the carc:ir ogenic process will occur. Thus, environmental agents or disease processes that produce sustained increased cell prolif- eration can enhance the likelihood of cancer development by pro- viding additional cell divisions, each with an opportunity for spontaneous genetic error. Studies of hereditary cancers and of various ]D:NA-damaging agents, such as radiation and certain vi- ruses and chemicals, have provided insight into identification of the essential genes, but many examples of carcinogenesis in hu- mans do not involve direct DNA damage. Also, most preneoplastic lesions in human carcinogenesis show increased proliferation compared with normal tissues, whether from increased mitotic rate, blocked differentiation, prolonged cell survi l A or other mechanisnis. Selected examples of proliferation-related carcino- genesis aria described, including certain infectious agents, defec- tive immune surveillance, hormonal imbalances, chronic inflam- matory-rei;enerative processes, and exposure to various chemi- cals. A common biologic mechanism for these diverse stimuli is increased cell proliferation as a prelude to cancer. This mechanism seems essential to the genesis of many cancers in humans. Key words: Cell proliferation, Carcinogenesis, Viral cancer, Chemical carcinogens, Genetics, Immune surveillance. Modern Pathology, Vol. 4, No. 3 C ancer, the second leading cause of death in the United States, may be inc,reasing, particularly in elderly individuals (1). Although progress has been made in the treatment of patients with cancer, prevention offers greater opportu- nities for reducing the death toll. Cigarette smoking, responsible for a majority of cancers of the respi- ratory tract and cancers of other organs, rem.a:ins the leading known 0893-3952/91/0403-0<71S03.00/0 MODERN PATHOLOGI' Copyright O 1991 hy',"he United States and Canadian Academy of PatholDgy, Inc. Vol. 4, No. 3, P. 371, 1991 Printed in the U.S.A. cause of cancer (2). Specific chem- icals known to be carcinogens in humans, such as 2-naphthylamine, benzidine, 4-aminobiphenyl, vinyl chloride, and diethylstilbesterol, account for only a small percentage of cancers (3). Infectious organisms have also been implicated as being etiologic agents of specific cancers (4), including enteric bacteria, par- asites, such as Schistosoma and Clonorchis, and viruses, such as Ep- stein-Barr virus (EBV), human T- lymphotropic viruses I and II, hep- atitis B virus (HBV), and human papilloma virus (HPV). Mounting evidence strongly sup- ports the contention developed in 1914 that cancer results from ge- netic alterations (5). Utilizing mo- lecular biologic techniques, numer- ous genetic alterations, including specific genes, have been identified in several cancers. However, many etiologic agents do not directly cause genetic damage. Similarly, some environmental agents asso- ciated with cancers do not directly damage DNA. Thus, although ge- netic damage is most likely an eventual common pathway to the development of cancer, other piv- otal mechanisms contribute to car- cinogenesis. That multiple events are essen- tial for the development of cancer has been demonstrated in experi- mental animal models, in in vitro systems, and in certain human can- cers. Nearly 50 years ago, Beren- blum and Shubik [6) conducted classical experiments in mouse cu- taneous carcinogenesis that re- sulted in the formulation of the two-stage carcinogenesis concept. Alfred Knudson (7) hypothesized that two genetic events occur for retinoblastomas to emerge in chil- dren. His hypothesis has been con- firmed through numerous genetic analyses and ultimately by the mo- lecular cloning of. a specific Rb ~ gene. Although cancer arises from de- ~ fective control of cell proliferation, the etiologic and pathogenetic role CA of cell proliferation has received Ll relatively little attention. Never- ~ theless, as early as 1953, Nordling (8) stated that, although genetic ~ alterations were necessary, the likelihood that certain cance*s 4~ would develop could be greatly a::g- mented by sustaining cell prolifer- CELL PROLIFERATION AND CANCER 371

atiorn of the target tissue. A decade ago, a specifically define role for cell proliferation was integrated into a carcinogenesis model devel- oped by Moolgavkar and coworkers (9, 10)„ which was derived from ep- idemiologic data, and into a biolog- ically similar model formulated by Greenfield et al. (11) and by Cohen and El.lwein (12), using data from animal experiments. Although de- rived from two different perspec- tives, the biologic framework of both models is strikingly similar. They offer a basis for interpreting a wide variety of carcinogenesis data in animal models and humans. Both models quantify genetic and proliferative events and thus offer insight into assessments dealing with the risk of developing cancer. The framework of these models is presented below and then selec- tively illustrated in human carci- nogenes; s. We have attempted to identify the common biologic thread of increased cell prolifera- tion as a common prelude to car- cinogenesis. This perspective is not intended to be definitive and, thus, the important work of many inves- tigators is not cited nor is the bur- geoning information being pub- lished regarding multiple molecular events being discovered for specific histologic types of cancer. . CELL PROLIFERATION AND CARCINOGENESIS The model of carcinogenesis dis- cussed and illustrated herein is shown Ln Fig. 1. For any theoretic model, assumptions are made in defining qualitative and quantita- tive aspects. The assumptions of this model are the following: (a) cancer arises from normal cells through two irreversible genetic events; (b) these genetic events oc- cur only during active cell prolif- eration or are irreversibly fixed only during cell division; (c) the carcinogenic events occur only in a susceptible subpopulation of cells within the target tissue (frequently referred to as stem cells); and (d) the two genetic events occur in a random fashion with non-zero spontaneous probabilities. Note that the word "transformation" is used to mean the development of malignant cells. I Normal Stem Celts I ~*tea c«m,rnea Ceft Vll rormol c«,., tn«d co+. Norm®I Mehre ceft Falated Matire C®is Figure 1. Diagrammatic representation of the biologic model of carcinogenesis originally described by Greenfield et al. (11). The bottom three circles represent the normal differ- entiation of a tissue. Ascending along the left side are the two stages of carcinogenesis, initiation and transformation. Down ward-pointing arrows represent cell death, whereas the other arrows represent the various combinations of possible results of cells following cell division. In the context of this model, an agent can alter the likelihood of developing a cancer in only two ways: it can increase the probabil- ity of irreversible genetic damage occurring during cell division; or it can increase cell proliferation, usu- ally accompanied by an increase in cell number, and consequently in- crease the number of opportunities for spc:~itaneous genetic damage. It also can do both. Although other models postulating more than two critical genetic events in the car- cinogenic process have been pro- posed, our analyses reveal that two critical events seem to be sufficient for cancer to occur. We recognize that the size of the susceptible pop- ulation of cells and their suscepti- bility can be altered by a variety of genetic and nongenetic events and stimuli. Also, further genetic alter- ations may occur in subclones of a malignancy, producing considera- ble heterogeneity with respect to several aspects of its biologic be- havior. Clearly, other events occur during progression of cancers that endow the malignancies with in- creased survival advantage. Under normal circumstances, the probability for either of the two critical genetic events to occur is exceedingly low (probably in the range of 10-10 to 10-6 per cell divi- sion); otherwise everyone would develop cancer at an early age. On the other hand, these probabilities are not zero, or no one would de- velop cancer. Thus, this model pre- dicts that, if people lived suffi- ciently long, all would develop can- cer. However, because these probabilities are so low, the odds are in favor of an individual not developing cancer, even with a life span of 100 yr. Approximately 25% of persons develop malignancy in the United States during their life- time. INHERITED CANCERS Studies of hereditary cancers of children have provided experi- ments of nature illuminating how carcinogenesis can occur. As origi- nally advanced by Knudson (7), ge- netic events occur in the two alleles of the Rb gene that give rise to retinoblastoma (Fig. 2). Normally, the likelihood of developing reti- noblastoma in an individual with- out an inherited retinoblastoma gene defect is rare, given that two rare events are required for the tu- mor. In contrast, individuals who inherit the defect in the Rb gene have nearly a 100% occurrence of the tumor. Although this pheno- typic expression initially suggested a dominant trait, Knudson postu- lated autosomal recessive Rb gene inheritance. With retinoblastic proliferation during development, a genetic error eventually occurs in the second Rb allele. Although rare during any one mitotic event, the 372 MODERN PATHOLOGY

probability that a mutation will oc- cur is sufi:iciently high that nearly all genetically susceptible individ- uals develop retinoblastoma. Inci- dences frequently are bilateral, and/or pe:rsons develop more than one tumor per eye at an early age. Retinoblasts only proliferate during development of the eye, and cell division is necessary for either of the two genetic events in the genesis of retinoblastoma to occur (unless one allele is defective be- cause of a germ line mutation). Thus, the chance of developing a retinoblastoma is eliminated once these cells stop proliferating. Sim- ilar arguments can be advanced for neuroblastoma, since neuroblasts also cease~ proliferating during childhood. Knudson's hypothesis prompted the search for other tumor suppres- sor genes (also referred to as an- tioncogenes) (13, 14). Increased susceptibility to the development of tumors in other tissues, such as osteogenic sarcomas, has been ob- served in patients with retinoblas- toma, although it remains unclear as to why tumors do not increase in all tissues. A second suppressor gene (wit:h protein product p53) might be involved with the genesis of these sarcomas. Other possible candidates for tumor suppressor genes include Wilms' tumor, renal cell carcinoma, and at least two forms of inherited colonic carci- noma. Polyposis coli (Pc) is an autoso- mal dominant, inherited suscepti- bility to adenomatous polyps and adenocarcinoma of the colon (15). Individuals with the Pc genetic de- fect (chromosome 5q) develop nu- merous co;lonic polyps that often evolve into carcinomas within a few decades. Sim.ilar genetic events oc- cur in some nonpolyposis coli pa- tients, who more commonly de- velop colon cancer at a later age. In addition, at least six other autoso- mal dominant hereditary traits predispose to colon cancer (16). Adenoma,tous polyps, which are preneoplastic lesions, exhibit in- creased proliferative capacity, pre- sumably due to enhanced prolifer- ation of the colonic crypts. Most (if not all) preneoplastic lesions in- volved in human carcinogenesis show increased proliferation com- pared with normal tissue, whether ~ 00 ~ WILD PREDISPOSED ~ TYPE PHENOTYPE HEREDITARY .. W_ TUMOR NON-HEREDITARY o.~. --40- Figure 2. Genetics of retinoblastoma. Tumors occur when defects occur in both alleles, whether caused by absence of the entire chromosome, deletion of a portion or all of the gene segment, or mutation of the gene. Individuals with hereditary retinoblastoma are born with one defective allele in all of their cells, requiring only a defect to develop in the second allele for malignancy to occur. Nonhereditary individuals must generate defects in both alleles beginning with cells having two normal alleles at conception. A - B--~ C - D-_E -= F---Po G Figure 3. Alternative explanations of multiple genetic events occurring during carcinogen- esis. lf each of the identified genetic events occurs sequentially, the upper diagram pertains. However, more likely is a mechanism similar to that presented in the lower diagram where there are multiple genetic events that will affect the proliferative rates, genetic stability, and/or the cell population sizes of A, B, or C, but are not essential for the carcinogenic process itself. However, these additional genetic effects will greatly accelerate the carcin- ogenic process overall. it comes from increased mitotic rate, blockage in differentiation, or other mechanisms. Fearon and Vogelstein (17) have recently postulated a multistep process for colonic carcinoma. However, their multiple stage model could also be consistent with only two critical events being re- quired for carcinogenesis (Fig. 3). The additional genetic alterations that they observe in other genes may enhance the proliferative ca- pacity or alter the differentiation of cells in the preneoplastic, ade- nomatous polyp. Although these CELL PROLIFERATION AND CANCER 373

secondary genetic alterations may give a proliferative advantage to preneoplastic cells, and may thus significantly decrease the time to the development of an actual ma- lignancy, they nevertheless are not required, rate-limiting events in the development of the tumor. HICIRMONES AND CANCER Hor -mones govern numerous cel- lular :functions, including prolifer- ation, growth, and maintenance of bodily functions. Clinical and epi- dem.iologic studies demonstrate that sustained hormonal stimula- tion (]:8) and consequent enhanced cell proliferation result in estrogen- dependent endometrial (19) and brea st carcinomas (20), thyroid- stimulating hormone (TSH)-de- pendent thyroid tumors (21), and andre€;en and estrogen interactions in the development of prostatic cancer (22). Endometrial carcinoma fre- quently results from chronic estro- gen stimulation of cellular prolif- eration.. For example, an increased incidence of endometrial adenocar- cinoma.s results from exogenous es- trogen therapy, such as seen with hormone replacement for meno- pausal women (19) and, possibly, from the use of older-type, estro- gen-containing contraceptives (23). In addition, obesity is a risk factor i`or endometrial carcinoma, possibly due to hyperestrogenism from ;increased production or stor- age of estrogen by adipose cells (24). Moreover, chronic estrogen stimulation is associated with en- dometrial hyperplasia and carci- noma in women with the polycystic ovary syndrome (25). Characteris- tically, estrogen stimulates the en- dometrium to proliferate (Fig. 4). Normally, this proliferative stimu- lus is tempered in midcycle by the increased production of progester- one, ultimately resulting in the shedding of cells during menstrua- tion. In the circumstances de- scribed above, estrogen stimulation is sustained rather than cyclic. Estrogen-related substances, such as diethylstilbesterol (DES), stimulatia estrogen-responsive cells. In experimental animals, DES in- duces a variety of estrogen-related tumors (26). In humans, the devel- r - A;7 '+ } ~ - ~ lf , J ~ ~' Y . .pl r..' s~'=f . ~ ~ . ~ 11 ~ . l + Y, . ~ r'S' L ~ E3~t t y 1 ~ :`h ''~ ~ , i i b. ~ y s ~ ' ~ , il „ ~ :U: ' , ',r' = ~ :: ...~ . ~, t!i ~04'S': • L:.•.. !~.. <a- ) - Y IJ,*Qt~ ~t 4sI ~~ .4q ,,:"' 4f ,ti!6 -4 y+ 1~=Zvv- s..+ 4k.Y'_'ko ~~- ~`'334'R 7~wv a ~ *aas.a 1, lir 7 \ Ti ~? ~-Z_•s ~t ` . .n :.C e ~ ~ i ~ ,At „~4 . ..- n. ! :{ i+.r , f ;. .,'a• ~ ~• 1ry % ''l1Y,t si ~ ~.~ r ~ S~~•®- •.. !'t! +. ~ f . . '3 -. . ~ r~ l~ ..4 ~ ' `ly_~j:filll.: ~ ?9',Y;" . . t-~`ro 1'r~.%a.~a~ ~.1~ q. 4Y:•.- oo.-• s 'To a 'C n r ij! j J `i ', ~~+1y DS. Figure 4. Estrogens have a proliferative effect on the endometrium. During the normal reproductive stage of a woman's life, this produces a proliferative endometrium (A), which is converted to secretory endometrium by progestational effects. However, if there is hyperestrogenism secondary to exogenous or endogenous sources, a hyperplastic endo- metrium results (8), since the proliferative effects of estrogen are not impeded by the normal or subnormal levels of progestins. If the adenomatous hyperplasia continues, a malignant tumor, adenocarcinoma, arises (C). This is particularly striking in postmenopausal women who have hyperestrogenism secondary to exogenous sources, but also can occur secondary to endogenous production. In contrast, postmenopausal women usually have lost the proliferative-stimulatory effects of estrogen, and their endometrium becomes atrophic (D). opment of vaginal adenocarcino- mas in the offspring of mothers who were exposed to DES during pregnancy is a notorious example ' (27). DES given to experimental animals, particularly in the ham- ster kidney model, undergoes met- abolic activation and DNA adduct formation (28). However, the prin- cipal role of DES in tumorigenesis appears not to involve DNA adduct formation but increased cell prolif- eration of estrogen-responsive cells (29). The situation in the human may reflect both of these compo- nents. The initial event in utero is likely due to interaction of DES with the DNA of specific vaginal cells. These initiated cells undergo rapid cell proliferation following 374 M0DERN PATHOLOGY

menarache, often leading to the carcino;enic second event. The role of estrogen, and possi- bly othe:r hormones, in causing breast cancer is similar to that in the enclometrium (20). However, other cr.it:ical factors may affect the responsiveness of cells to estrogen stimulation. For example, the age at which a woman has her first child significantly influences her susceptibility to breast cancer: the younger t','ae woman is at the time of her initial pregnancy, the less likely she is to develop breast can- cer. During pregnancy, terminal differentiation of the breast duc- tules occurs, removing a large num- ber of cells from the cancer-suscep- tible po;pulation. Thus, even if these cells are subsequently stim- ulated to increased proliferation, they are not susceptible to devel- oping cancer. Human chorionic go- nadotropin (HCG) also appears to be involved with the induction of this differentiation process. An ini- tial pregnancy at a later age pro- longs the susceptible period of these cells. Not only is the rate of cell proliferation important, but the size of the susceptible cell pop- ulation and the period of time over which it persists influence the chances for experiencing the criti- cal events necessary for developing cancer. Chronic increased cell prolifera- tion induced by estrogen also in- creases the appearance of benign and malignant hepatocellular tu- mors in experimental animals and in humans (30). Hepatocytes with estrogen receptors respond to in- creased estrogen levels by dividing more frequently. Hormonal effects on cell prolif- eration also greatly affect the like- lihood of developing prostatic ade- nocarcinoma (22). Animal models have been developed wherein the prostate is provoked into a burst of cell proliferation. This prolifera- tion can thEn be hormonally sus- tained, leading to the development of adenocarcinomas. In these in- stances, it remains unclear what the interactions between andro- gens and estrogens are, but both and possibly other hormones ap- pear to be involved. Thyroid carcinogenesis involves the interaction between the thyroid and the pituitary in a feedback loop (18, 21). As the thyroid produces more thyroid hormone (T3 or T4), it inhibits the pituitary, reducing TSH production. If thyroid hor- mone levels decrease, TSH levels produced by the pituitary increase, resulting in increased thyroid pro- liferation. In animal models, this process is frequently seen following the administration of chemicals (21) that decrease levels of thyroid hormone by a variety of mecha- nisms. This ablates negative feed- back on the pituitary and, conse- quently, overproduction of TSH re- sults and thyroid follicular cell proliferation arises. Ultimately, thyroid tumors result. Again, there is no evidence that TSH damages DNA by itself; these tumors arise as a consequence of chronic in- creased cell proliferation of the tar- get tissue. This mechanism is non- genotoxic, but a thyrotoxic chemi- cal can ultimately evoke tumors in the target organ. INFECTIOUS ORGANISMS AND CANCER Several microbial agents in- crease cell proliferation and in- crease the risk of developing can- cer. The intriguing possibility that infectious organisms might cause cancer has been investigated for more than a century. The first transmissible carcinogenic viral agents were identified in experi- ments by Rous and by Ellerman and Bang during the early part of this century (31). Cell-free extracts were found to transmit cancer from diseased to disease-free animals. Some fungi have been implicated in cancer development by produc- ing specific carcinogenic toxins, for example, aflatoxin (32). Bacteria have also been associated with the production of carcinogenic chemi- cals. For example, enteric bacteria occasionally are involved in the metabolic activation of certain car- cinogens, such as cycasin (33). Other organisms more directly cause specific cancers. Immunoproliferative small in- testinal disease (IPSID) is found in males in Third World countries. The initial benign appearing hy- perplastic lymphoid lesion is thought to arise from chronic an- tigenic stimulation by bacterial li- popolysaccharides or enterotoxins of Vibrio cholerae. Supporting this view is the regression of the lesions following a 6-mo trial of tetracy- cline (34). Without treatment, these lesions can convert to mono- clonal malignant lymphoma that secretes a heavy chains of immu- noglobulin. Certain parasitic diseases in- crease susceptibility to cancer most notably schistosomiasis (35) and clonorchiasis (36). Chronic Schis- tosoma hematobium infection is as- sociated with a markedly increased risk of developing bladder cancer. This agent causes chronic inflam- mation, fibrosis, squamous meta- plasia, and sustained, increased, squamous cell proliferation com- pared with the normal, mitotically quiescent transitional epithelium (Fig. 5). The majority of the tumors that develop within these infected bladders are squamous cell carci- nomas, rather than the usual tran- sitional cell carcinomas. Although specific carcinogens, such as nitro- samines, may be produced in schis- tosomiasis, sustained increased cell proliferation is pivotal to generat- ing these tumors. Schistosomal infections of the lower gastrointestinal tract (S. mansoni and S. japonicum), com- mon in the Far East and elsewhere, are associated with development of colonic carcinomas (37). This as- sociation is considerably less fre- quent than with schistosomiasis and bladder cancer. Again, sus- tained increased proliferation of the colonic epithelium may be a mechanism responsible for these cancers. Chronic biliary tract infections with the flukes, Clonorchis sinensis (36) or Opisthorchis viverrini (38), evoke destruction, epithelial regen- eration, and an increased preva- lence of cholangiocarcinoma (Fig. 6). A specific carcinogen is not im- plicated in this process, whereas increased cell proliferation is sus- tained in bile ducts and ductules. Although on a worldwide basis these infections and tumors are common, they seldom affect per- sons in economically developed countries. In contrast, specific RNA and DNA viruses infecting populations globally can be carcin- ogenic (31, 39). CELL PROLIFERATION AND CANCER 375

Figure 5. Squamous cell metaplasia (A) of the urinary bladder in a patient with chronic schistosomiasis. There is also ulceration of the epithelium. Note the numerous schistosome organisms in the wall of the bladder. The organisms can also be seen in the poorly differentiated squamous cell carcinoma that arose in another patient with schistosomiasis (B). Figure 6. Bile ductular proliferation (right) secondary to Clonorchis infection, which given rise to a cholangiocarcinoma (left). Numerous oncogenic RNA retro- viruses affect animals. Retrovi- ruses convert viral RNA to DNA by utilizing the reverse transcrip- tase er; zyme; the DNA is then in- corporated into the genome of the host. It has been hypothesized that specific viral oncogenes have been incorporated into the human ge- nome as protooncogenes or cellular oncogenes (39, 40). These viral and human cellular homologs act in a genetical;(y dominant fashion. The Rous sarcoma virus carries the viral sn° gene. Several other viral oncoger.tes of retroviruses that in- fect humans have been identified and can result in leukemia or lym- phoma. Other retroviruses can produce cancers without carrying a specific has i oncogene as part of their RNA (39, 40) by increasing the proliferation of the target tissue. Transmission of virus occurs from cell to cell, eventually resulting in the inter- position of virally generated DNA next to a cellular oncogene. Thus, sustained cell proliferation and, ul- timately, tumors can arise. A single oncogene, such as bcl-2 involved in follicular lymphoma, appears inca- pable of producing cancer without a second event. This oncogene, which is activated by the reciprocal translocation t(14;18) involving the breakpoint at bcl-2 on chro- mosome 18 and the heavy chain locus at 14q32, enhances follicular center cell proliferation. A second event is needed for the malignant counterpart to emerge. The human oncogenic retrovi- rus, human T-cell leukemia virus (HTLV) I, chronically infects nearly 1 million people in Japan. Given that only 400 patients de- velop adult T-cell leukemias yearly in Japan, it is evident that a mul- tistep process prevails. It has been postulated that immune deficiency and genetic events are involved in this leukemogenicity (41). Infection with human immuno- deficiency virus (HIV) results in an increased susceptibility to malig- nant lymphomas, squamous cell carcinomas, and Kaposi's sarcoma, but does not seem to be directly oncogenic (42) (see below under "Immune Surveillance of Cancer"). Several DNA viruses, including hepatitis B virus (HBV), human papilloma virus (HPV), and Ep- stein-Barr virus (EBV), are asso- ciated with certain types of cancers in humans. In each instance, the development of the malignancies results from a sustained prolifera- tion of the target cell. Herpes vi- ruses I and II have also been asso- ciated with an increased risk of cervical cancer, although this as- sociation has not been confirmed by recent research (43). HBV infection is mostly asymp- tomatic and transient. However, in susceptible individuals, the acute infection leads to sequelae of chronic active hepatitis (Fig. 7), which can progress to cirrhosis (44). Likely, the virus persists pre- dominantly in males owing to an inadequate immune response to the virus. The male:female ratio of pri- mary hepatomas is 4:1. Females have superior immunocompetence to HBV than do males, as has been shown in studies done in Taiwan (45). Increased androgens may also enhance hepatocarcinogenesis. The characteristic features of chronic active hepatitis and cirrho- sis are hepatocellular necrosis si- multaneously with regenerative re- pair. The normal liver is a mitoti- cally quiescent tissue as is the urinary bladder epithelium. With chronic active hepatitis or cirrho- sis, hepatocyte. proliferation is markedly increased and is sus- tained for the life of the patient. In a significant number of such pa- tients, hepatoma arises. World- wide, hepatomas are caused pri- marily by chronic HBV (46). 376 MODERN PATHOLOGY

HBV--related hepatocarcinoge- nesis is probably not related di- rectly to a specific oncogenic DNA alteration induced by the virus it- self. Transgenic mice that overpro- duce the large envelope polypeptide of HBV accumulate hepatitis B surface antigen and develop chronic active hepatitis, regenera- tive nodules, and ultimately hepa- tomas (47). This protein has none of the characteristics of oncogenes or tumor suppressor genes, but rather appears to be involved with the development of hepatocellular necrosis, chronic active hepatitis, and sustained, increased hepato- cyte proliferation. Any situation resulting in a chronic inflammatory or cirrhotic process is associated with an in- creased proliferative rate, regener- ative nodules, and an increased risk of hepatomas. Examples include chronic alcoholism and a variety of hereditary disorders, such as hem- ochromatosis. Not one of these conditions causes specific genetic damage, but they have in common increased sustained cell prolifera- tion. HPV infects squamous epithelia and is most commonly associated with cervical squamous cell carci- noma. Also, squamous cell carci- nomas of the penis, skin, anus, and oral cavity frequently contain the virus (48). HPV blocks differentia- tion of the infected epithelium, giv- ing features of dysplasia (Fig. 8). Figure 7. Chronic active hepatitis secondary to HBV infection. This is a chronic necroin- flammatory process with sustained regeneration, occasionally leading to the development of hepatoma. Increased cell proliferation and ex- pansion of the basal cell compart- ment result. Since HPV infections are usually persistent, events lead- ing to the continued presence of dysplasia can evolve, occasionally leading to carcinoma in situ and squamous cell carcinoma (49). Again, HPV causes a greatly in- creased risk of carcinoma owing to enhanced cell proliferation. Smok- ing cigarettes and defective im- mune responsiveness to the virus also appear to play a role (48, 50). EBV is a well-known B-cell mi- togen. The virus infects B-cells through the C3d (CR2) receptor and immortalizes them in vitro. During acute infectious mononucleosis (Fig. 9), approximately one per 104 B-cells is infected, whereas during latency approximately one per 106 B-cells is infected. Multiple im- mune responses, especially by T- cells, bring the B-cell proliferation under control (51). However, if the polyclonal B-cell proliferation is not brought under control, Burk- itt's or other non-Hodgkin's lym- phomas can arise. Immune-defi- cient individuals chronically im- munosuppressed by holoendemic malaria, children with inherited immunodeficiency, HIV-infected persons, or transplant recipients frequently develop EBV-carrying tumors. Klein and Klein (52) postulated a multistep scenario in the genesis of African Burkitt's lymphoma. Holoendemic malaria suppresses cytotoxic T-cells against EBV-in- fected B-cells while simultaneously causing polyclonal B-cell prolifer- Figure 8. The normal squamous epithelium of the cervix shows a "stem cell" basal layer with differentiation progressing to the surface (A). With HPV infection, there is blockage of this differentiation process leading to an expansion of the proliferative pool of cells extending higher in the epitheliurn, above the basal layer (B). In this figure, there is clear evidence of HPV infection as indicated by the koilocytes and chronic inflammation, with increasing degrees of dysplasia progressing from right to left. CELL PROLIFERATION AND CANCER 377

~~ ~ i.p ~ jA4 3 Y I1 1C6~A( rTr ~` KK `_i ^'~ ~t~2~~f~• ~ Rt~ ,A •' A Rt2 `t / *, i®~' •6 _, . ,. z ~ ~ ~. ~~~' '~ `f ~ ~ ~V4 ~ ~ w -a'.-WAktol I _Vk~~ 4=1* P, -31 & ., - ro lit . . :.. . .z:6,44 A: ~ btA~ ~~ 4 1 .~,,~ ~ ~ ~ ~" M .~, t;! • / ~t'it' i ~ ~ ~e~ ~- ~ ~ + ~ u^ ~ i • -.W~-~*_P--. ' ~ ' 1 `t '7 a'=vw.`, : ~~ • 1` 7r ~1 't''~~ ~ ' fi. ~ ~ ~ ~ * r 'A ~ '~ ~ ~ ~ . . . ; ~~ _ -. ,116'v .#Qt0~ ® `: ~, a W ~ ~~.,, ~'•, ~ 4A41;~,~w~a~~.~t^, fr~~~,~ 8p ~ .~ b~1 ~. ~" :t9~ .. s_- *W /~'.. ts _ Figure 9. I nfectious mononucleosis (A) is a markedly proliferative acute infection secondary to EBV, which is normally brought under control by variou!; immune factors. In immunosuppressed patients, this control of B-cell proliferation does not occur. The sustained proliferation can eventuate in the development of B-cell lymphomas, such as Burkitt's lymphoma (8). ation. This increased cell prolifer- ation increases the random chance that a specific chromosomal trans- location might occur involving the c-myc p rotooncogene at 8q24 with corresponding breakpoints involv- ing imraunoglobulin loci (14q32, 2p12, or 22q11). Juxtaposition of an Ig gene with c-myc promotes expression of the c-myc gene prod- uct. Concurrently, the major his- tocompa.tibility complex (MHC) and EBV viral targets for T-cell surveillance are down regulated. Identical chromosomal transloca- tions are seen in mouse and rat immurtocytomas and plasmacyto- mas. Moreover, the Ig-myc trans- gene results in transgenic mice that develop pre-B-cell malignant lym- phomas (53). In ce:ll culture systems, EBV readily produces a mitogenic re- sponse in B-cells but does not re- sult in the production of malignant transformation (54). This suggests that E:$•V does not have a specific malignant transforming gene and that t;he specific chromosomal translocation leading to Burkitt's lymphoma is an exceedingly rare event. The probability that any particular cell division will produce a malignant transformation is not increased, but, in the patient wherein uncontrolled polyclonal B- cell proliferation persists, the num- ber of,ce1l divisions is enormously increased. The odds of a chromo- somal translocation occurring within the susceptible pre-B-cell population are thus increased. IMMUNE SURVEILLANCE OF CANCER In 1957, Burnet proposed the no- tion that the immune system rou- tinely recognizes and eliminates newly generated cancer cells. This hypothesis was extended by Thomas in 1959 (see Ref. 55). Can- cer was proposed to arise chiefly because of a breakdown in the im- mune surveillance against cancer cells. This theory was based on sev- eral experimental observations. In mice, tumor-specific antigens were identified, suggesting that tumor cells had specific antigens that could be detected by the immune system and eliminated. Further- more, several viral and chemical carcinogens were demonstrated to have immunosuppressive proper- ties in animal models. Further, sup- porting this theory, clinical obser- vations indicated that patients with congenital immunodeficien- cies, such as Wiskott-Aldrich syn- drome, or acquired immunodefi- ciency secondary to immunosup- pressive therapy for renal transplantation had a markedly in- creased occurrence of malignan- cies. Although superficially plausible, additional observations and exper- imentation have revealed that the immune surveillance theory of car- cinogenesis is not correct. The tu- mor-specific antigens that were discovered in mice were determined to be related primarily to tumori- genic viruses or H-2 antigens. Tu- mor-specific antigens in human cancers have only been discovered in multiple myeloma (56). A wide variety of tumor-associated anti- gens have been identified, which are embryonic or differentiation antigens of normal cells. Although qualitative differences have not been identified, quantitative differ- ences between normal cells and cancer cells have been demon- strated. The immunosuppressive effects of a variety of carcinogens, partic- ularly the polycyclic aromatic hy- drocarbons, and some of the carcin- ogenic viruses are immunosuppres- sive only at doses far in excess of those known to cause cancer, or they are immunosuppressive using routes of administration unrelated to the carcinogenicity studies (57). For other chemicals or viruses shown to be carcinogenic in various animals, immunosuppressive prop- erties could not be demonstrated. Also, many experimental models were shown to be unaffected when immunosuppressants, such as aza- thioprine or cyclophosphamide, were administered concurrently or sequentially with the carcinogenic agent (58). Although a marked increased prevalence of malignancies occurs in immunocompromised patients (59), all types of malignancies are not increased; only B-cell lympho- mas, Kaposi's sarcoma, and cuta- neous, oral, anal, and uterine cer- vical squamous cell carcinomas (Fig. 10) are increased (59, 60). As 378 MODERN PATHOLOGY

Itrenune Surveillance of Infectious Organisms r Immunosuppression New Infections T or Reactivation T 1 T Epstein Bai^r Virus (EBV) Human Papilloma Virus (HPV) Hepatitis B Virus (HBV) Other B-Cell Proliferation r r Squamous Cell Dysplasia Chronic Active Hepatitis T 7 B-Ce11 f.ymphoma Squamous Cell Carcinoma Hepatoma Figure 10. Diagrammatic representation of the role of immune surveillance of various infectious organisms in the etiology of specific cancers. The immune system normally controls the acute infection. If the patient is immunosuppressed (hereditary, transplantation, AIDS), chronic proliferative effects ensue from the uncontrolled infections, occasionally resulting in malignancy. described above, these malignan- cies hava been associated with vi- ruses. El3'J is present in the B-cell malignancies and HPV in the squa- mous cell carcinomas. Kaposi's sar- coma occu rs in AIDS patients, pos- sibly arising from the Tat gene of HIV or growth factors liberated by stimulated T-cells, or other, yet to be identified, factors (61). The common biologic theme among these tumo:rs is a chronic increased proliferation of the target cells re- sulting from persistent, new, or reactivated viral infections. Im- munity controls the extent of pro- duction and persistence of these viruses. Thus, immune surveillance is germane to carcinogenesis with these sper,il"Ic, virus-induced tu- mors. Surveillance, however, is not against malignant cells but against the infectiou s organisms. As indi- cated above, excellent examples of this are the E BV-induced lymphoid malignancies in immunosup- pressed patients that result from a failure of T-cells to recognize Ep- stein-Barr viral antigens on the surface of infected B-cells and to eliminate them. The sustained cell proliferation leads to a substan- tially increased risk of cancer. Although there is little evidence to support the immune surveillance theory of carcinogenesis as origi- nally described in the 1950s and 1960s, immune surveillance is plau- sible for microbe-induced cancers that act primarily through produc- ing mitogenesis. The immune sys- tem under Darwinian evolutionary pressures evolved to protect the host against life-threatening infec- tions and not cancer. Malignancies occur largely during the postre- productive period and, thus, natu- ral selection would not occur. How- ever, immunologic regulation of the invasiveness and metastatic poten- tial of cancers, such as melanomas, bladder carcinomas, leukemia, lymphoma, and renal cell carcino- mas, may be important (62). CHRONIC INFLAMMATORY PROCESSES As summarized above, many chronic inflammatory processes in- crease the risk that cancer might develop. In addition, in the gas- trointestinal tract, notable exam- ples are the association of chronic atrophic gastritis with gastric car- cinoma (63) and chronic ulcerative colitis with colonic carcinoma (64). A chronic necroinflammatory proc- ess results in sustained regenera- tive proliferation of cells that gain a proliferative, and possibly a sur- 1 vival, advantage greater than the surrounding normal tissue. Gastric intestinal metaplasia or colonic ep- ithelial dysplasia ensues that can develop into proliferative foci, ad- enomatous polyps, and adenocar- cinomas. The presence of agents that en- hance the proliferative process, such as high salt intake or Helico- bacter infections associated with the stomach (63, 65), or bile acids, high fat, and low fiber diets with the colon (66), predisposes to de- velopment of cancer. Conversely, dietary calcium, high fiber, and low fat are associated with decreasing colonic cancer risk by decreasing proliferation (66, 67). Gallstones and gallbladder and biliary tract cancer (68), tropical phagedenic ulcer and squamous cell carcinoma of the skin (69), and chronic esophagitis secondary to gastric reflux leading to Barrett's esophagus and adenocarcinoma (70) are other situations character- ized by sustained cellular prolifer- ation and frequent carcinogenesis (Table 1). In a similar vein, high rates of growth of normal tissues are also associated with an increased risk of cancer. For example, osteogenic sarcoma incidence peaks during ad- olescence when growth is marked (71). Osteogenic sarcomas in older individuals are frequently associ- ated with Paget's disease, a disease associated with an increased prolif- erative process of the osteoblasts (72). CELL PROLIFERATION AND CHEMICAL CARCINOGENS Numerous chemicals and chem- ical mixtures increase the risk of developing cancer, including ciga- rette smoking, snuff use, betel quid chewing, aromatic amines, polycy- clic aromatic hydrocarbons, nitro- samines, and others as detailed in a recent IARC monograph (26). Most of these chemicals are both mutagenic in short-term screening assays and carcinogenic in a variety of species. They are also cytotoxic to the target tissue, resulting in regeneration and increased cell proliferation (Fig. 11). At toxic doses, a sharp increase in the rate of tumor formation is observed. CELL PROLIFERATION AND CANCER 379

TABLE 1. SOME OF THE CHRONIC CONDITIONS ASSOCIATED WITH INCREASED CELL PROLIFERATfON AND INCREASED RISK OF CANCER DEVELOPMENT Organ Chronic Condition Skin Phagedenic ulcer Esophagus Reflux esophagitis with Barrett's esophagus Stomach Chronic atrophic gastritis Colon Chronic ulcerative colitis Liver Cirrhosis Gallbladder Cholelithiasis Bone Paget's disease CHEMICAL CARCINOGEN GEWOTOXIC 1. Threshokf unlikely 2. Dose-response may be affected by cell pioliferation (usually toxicity related at high dDses) NON-GENOTOXIC PROLIFERATIVE 1. Threshold questionable 2. Usually effective at !ow doses PROLIFERATIVE 1. Threshold likely 2. Usually related to toxicity and regeneration Figure 1111. Diagrammatic representation of proposed classification of chemical carcinogens based on their ability to react directly with DNA or cell receptors. Cell proliferation affects the dose-response of all classes of chemical carcinogens [From Cohen and Ellwein (12)]. Similarly, UV radiation is asso- ciated with skin carcinomas (73); high energy radiation with cancers of the bone marrow (leukemia), thyroid, breast, and other tissues (73); and thorotrast with liver and kidney cancers (74). These forms of radiation are obviously geno- toxic, but the development of tu- mors secondary to radiation is fre- quently associated with chronic, destructive-regenerative processes. Although the carcinogenic syn- ergism associated with the prolif- erative effects of chemicals has been best documented in experi- mental wtimals, this likely holds for humar.ts also. For example, cig- arette smoke is toxic to the respi- ratory epithelium, leading to chronic b:ronchitis and squamous metaplaaia, associated with in- creased cell proliferation (75). In addition, cigarette smoking pro- duces hyperplasia and carcinoma of the urinar;~ bladder (76). Snuff and other oral]y used tobacco products contain rnany carcinogens, but they are also associated with chronic inflammatory, regenera- tive processes in the oral cavity and pharynx (77). Likely, the combi- nation of their genotoxicity and in- creased cell proliferation results in cancer in these tissues in humans. When tested in experimental bioassays for carcinogenic activity, a large proportion of chemicals that are negative in various short-term mutagenicity screens show carcin- ogenic activity in mice and/or rats (78). This raises concern regarding the interpretation of these data and the extrapolation of potential risk to humans for compounds in the environment or food supply. A ma- jor complication in assessing risk from these chemicals is that tu- mors usually occur only at high, often toxic, doses that are fre- quently associated with increased cell proliferation of the target tis- sue (12). For nongenotoxic chemicals that demonstrate proliferative effects only at very high doses, a no-effect threshold might exist (12). For ex- ample, when melamine is adminis- tered to rats or mice at high doses, urinary calculi form, and ulti- mately, bladder tumors develop (12, 79, 80). When lower doses of melamine are administered and calculi do not form, cell prolifera- tion is not increased and no tumors form. The data regarding melamine and related compounds that induce calculi in experimental animals only at high doses imply that there is no carcinogenic risk for humans exposed at low doses, where calculi do not form. The Environmental Protection Agency (EPA) has re- cently followed this pragmatic logic and an understanding of biologic mechanisms in interpretating data for melamine (81). Another example with a similar mechanistic action is sodium sac- charin. It induces bladder cancer only in rats, particularly in males, but not in mice, hamsters, or mon- keys (12, 82). The tumorigenic ef- fect of sodium saccharin is likely due to formation of silicates in the urine of male rats. Factors, includ- ing pH, protein, sodium, silicate, and saccharin, reach critical levels in the urine following feeding of high doses of sodium saccharin to rats. A threshold effect is likely. If lower doses of sodium saccharin are administered, silicate precipitates and crystals do not form, cell pro- liferation is not increased, and there is no increased tumor for- mation. Moreover, the critical set of urinary parameters in the rat following high doses of sodium sac- charin is not present in humans, mice, or monkeys. Hence, humans appear to be resistant and would not be expected to develop bladder cancer even at extremely high doses of sodium saccharin. Assessment of risk of chemical compounds is controversial. Regu- latory agencies are determining whether differences should be made in interpreting data between nongenotoxic and genotoxic com- pounds. Genotoxic compounds generally do not appear to have a threshold regarding their genotoxic effects, but the tumorigenic re- sponse is greatly augmented at doses producing increased cell pro- liferation. In contrast, most (if not all) nongenotoxic compounds likely require a threshold dose for increasing cell proliferation, and consequently, they are likely to 380 MOAIERN PATHOLOGY