Document QJ4n4zDk0y2R26j9wpxpDZXXk

American Journal of Industrial Medicine 27:293-300 (1995) COMMENTARY Mechanisms, Chemical Carcinogenesis, and Risk Assessment: Cell Proliferation and Cancer James Huff, PhD Mechanisms of carcinogenesis-and in particular chemically associated carcinogenic- ity-have attracted considerable scientific and public attention in the last decade. Much insight has been gained that will lead to more reasoned and better prevention, interven- tion, and treatment for the reduction of environmentally caused cancers. However, there seems to be an exaggerated tendency to embrace "mechanisms" not yet fully charac- terized, completely tested, unequivocally proven, and consensus accepted. More than 100 agents and exposure circumstances have been identified as causally or strongly associated with human cancers; for many the evidence was discovered first in experimental animals. More chemicals have been uncovered as carcinogenic in experimental animals, with as yet no or little available information in exposed human populations. Additional and expanded mechanistic and epidemiological studies should further elucidate the relevance of these agents to adverse human health effects, including cancers. Claims are being posed that certain chemical-specific "mechanisms" in experimental systems are irrelevant to humans, and thus chemicals thought to be aberrantly carcinogenic in animals would present no cancer hazard to exposed humans. Nonetheless before undeniable proof becomes available, we must continue to proceed with sensitive and responsible caution. This commentary offers a central and personal view of one such k- mechanism: cell proliferation and cancer. Q 1995 Wiley-Lis. Inc.* Ll,.i Key words: carcinogenesis; cell proliferation; risk assessment; experimental carcinogenesis; bioassays; animal carcinogens; human carcinogens INTRODUCTION Strategies for accomplishing risk assessments for protecting workers and the general public from exposures to harmful circumstances are being revamped to include "mechanistic information." A major difficulty confronting researchers and regulators centers on what is actually meant by the oft-used term "mechanism": to many this term connotes simply pharmacokinetics, whereas to others the word means molecular bases of action. Some have additional notions with respect to genotoxicity, Environmental Carcinogenesis Program, National Institute of Environmental Health Sciences, Research Triangle Park, NC. Address reprint requests to James Huff, PhD, Environmental Carcinogenesis Program,National Institute of Environmental Health Sciences, P.O. Box 12233, Research Triangle Park, NC 27709. Accepted for publication October 14, 1993. This updated commentary has been taken in part from Ramazzini Newsletter 3:47-50, 1992 [June 19931. 0 1995 Wiley-Lis, Inc. *This article is a US Government work and. as such, is in the public domain in the United States of America. 294 Huff metabolism, toxicity, and so forth. In chemical carcinogenesis, for example, discussion and research center on the role that exogenously enhanced cell proliferation may have on the carcinogenesis process. No consensus exists on this particular issue; for some, confusion persists because without cell proliferation cancer could not occur, and thus individuals might conclude that if a chemical induces cell proliferation this must then be mechanistically influential in the carcinogenic process. Not true. CELL PROLIFERATION Cell proliferation is typically estimated (assumed) by quantifying S-phase nuclei. Most reported cell proliferation data reflect replicative DNA synthesis, without demonstrating that the S-phase nuclei proceed to divide and accumulate. Thus, the term cell proliferation and other synonyms should be replaced by more specific terms such as what was specifically measured: replicative DNA synthesis, S-phase labeling index, BrDU labeling index, and so on. Clearly, reporting mitotic indexes allows more confidence in cell division taking place; yet even in this instance, without cell death rate measurements, there may be no net increase in cells. Several views expressed at a recent National Institute of Environmental Health Sciences (NIEHS) international symposium on cell proliferation and carcinogenesis placed this important issue in better perspective: while cellular replication must take place for cancer development, chemical-induced cell proliferation per se may or may not influence chemical carcinogenesis; enhanced cell division is not reliably predictive of carcinogenicity; and no generic postulate can be formulated.' Nonetheless, some scientists and administratorscontinue to insist that chemicals causing increases in cell replication would also lead to cancer, and studies of cell proliferation would divulge the mechanism of tumor formation. This notion has been used to "explain" why certain chemicals induced cancer in laboratory animals; but a testable or biologically based "mechanistic hypothesis" has yet to be systematically formulated and evaluated. At times we develop and use certain terminology for communicative purposes that tends to lack precision: e.g., "tumor promoters'' and "nongenotoxic carcinogens." Unfortunately, these terms often become "mechanistic-based,'' rather than only operational, and we begin to "believe" our own hypothesis without having the necessary supporting experimental information (re: satisfaction of Koch's postulates). This seems to be where we are at the moment regarding the impact of enhanced cell division on chemical carcinogenesis. Much work is being accomplished; yet considerably more needs to be done, however, to enable us to make better and more scientific-seated mechanistic decisions. EXPERIMENTAL STUDIES Few studies have been designed and conducted to answer the question: "What influence does chemically enhanced cell proliferation have on chemical carcinogenesis?" In the aggregate, these have been relatively short-term experiments of comparatively limited duration: 1-13 weeks. However, in most reports the connection between chemically augmented cell proliferation and cancer has not been proven; and in most, has been shown to play no obvious role in the carcinogenic process. Better `Proceedings comprising 46 papers was published in Environrnenral Health Perspectives 101 (Suppl. 51. and more ex tions of carc the propose` chemical an' Thus, portant and not sustain associates c, plasia. Incrc explain or c Cell p arguments f factors that towards car be no cancl be repaired carcinogen, would be n experimenI mal" cell CHEMICP Cher the multisl genic sub: more or a both of tv obviously and 2) prc mechanis: no inhere by increa carcinoge biologica Av: examinec cell replit eration i: vational 4 iting no without i instance: or in the one of tt decipher ISCUS- 1 may e; for iccur, n this )e nuithout s, the terms ieling llows it cell `s ex- EHS) nporancer Jence cino- iicals f cell been i; but ically y for ' and ianis1 hy: satment vlluch :r, to hat igen- :om- :tion ; and etter )I. 51. I I Chemical Carcinogenesis and Cell Proliferation 295 and more experiments are needed (e.g., using the same long-term exposure conditions of carcinogenesis bioassays) before we will be able to reliably conclude whether the proposed mechanistic hypothesis is or is not correct. And these will likely be chemical and organ specific. Thus, cell replication (proliferation, turnover, mitogenesis, division) is an important and essential factor in chemical carcinogenesis, yet available scientific data do not sustain the hypothesis that enhanced cellular proliferation in a particular organ associates consistently with or mechanistically causes an increased induction of neoplasia. Increased cellular proliferation is but one of a number of hypotheses that might explain or contribute to the resultant neoplastic transformation. Cell proliferation per se may influence the carcinogenic processes but there are arguments for and against this hypothesis. As mentioned, one of the major confusing factors that leads some to the notion that agent-enhanced cell proliferation persuades towards cancer comes from the knowledge that without cell proliferation there would be no cancer. That is, in the absence of cell division, any "damaged" DNA would be repaired or remain without consequence. Because cell proliferation is necessary for carcinogenesis, then the implication is frequently made that enhanced cell turnover would be mechanistically associated with the resultant cancer. However, the available experimental data do not support the view that simply increasing division of a "normal" cell population is or would be sufficient for carcinogenesis. CHEMICALS AND CELL TURNOVER Chemicals that only enhance cell proliferation may have no or little influence on the multistep and multimechanistic carcinogenesis processes. However, if a carcinogenic substance also induces sustainable cell proliferation, it might influence one or more or a combination of steps in the overall process by potentially involving one or both of two possible effects: 1) reduce the latency period for tumor development by obviously increasing populations of both transformed and nontransformed cells or/ and 2) promote an increase in incidence of detectable tumors induced by alternative mechanisms by simply stimulating more potential stem cells. Still, chemicals having no inherent carcinogenic potential would not be expected to "cause" cancer simply by increasing cell proliferation. Chemical combination experiments including "noncarcinogenic cell proliferators" could lead to better understanding of the role this biological effect has in carcinogenesis processes. Available studies of cellular proliferation of known carcinogens have been examined and the unambiguous conclusion is that short time-point measurements of cell replication do not predict long-term effects, and that mitogenesis or cell proliferation is merely a part of the carcinogenic process. In many instances, the observational data actually contradict the hypothesis: Le., tumors occur in organs exhibiting no "increases" in cell division, cell proliferation is significantly enhanced without any tumors being discovered, or combinations of these two findings. In some instances, DNA synthesis is substantially increased in the same organ of both sexes or in the same sex in two species-yet tumors develop in only one of the sexes or in one of the species. Experimental findings such as these need to be fittingly pursued, deciphered, and resolved. 296 Huff RESEARCH QUESTIONS AND CELL REPLICATION Answers to the following (and other) questions may allow a better foundation to formulate a mechanistic-based hypothesis about cell proliferation and cancer: 1. whether chemically induced cell proliferation in normal tissue can be taken to indicate that a greater incidence of ``spontaneous" mutations (initiating events??) would occur 2. whether measuring cell turnover in normal tissue (vs. preneoplastic or neoplastic tissue) has any relevance to the carcinogenesis process 3. whether an "adaptive" short-term (e.g., a few to several days) increase in cellular proliferation would influence the carcinogenic process (especially and typically observed for nongenotoxic chemical carcinogens thought to need "continuous" and long-term exposures for carcinogenesis) 4. whether enhanced cell replication must take place in stem cells vs. those terminally differentiated (and unable to clonally expand) 5 . whether "pure" mitogenic (and nontoxic) chemicals are carcinogenic by the simple mechanism of cell replication (and hence increase the probability of "mutations" or "genetic errors") 6. whether the impact of chemically induced cell proliferation always, never, or only sometimes leads to or accelerates cancer 7. whether cell replication influences each (or one) step of the multistage and multimechanistic processes of carcinogenesis 8. whether this mechanism would likewise increase the incidence of cancers in humans exposed to drugs, alcohol, tobacco, industrial chemicals, environmental. estrogens, "promoters," viruses, irritants, heat, and any number of other potential agents known or capable of inducing cellular damage or death and subsequent regenerative physiological adaptive processes, including cellular replacement. MECHANISMS AND RISK ASSESSMENT As with pharmacokinetics, metabolism, toxicity, genotoxicity, other mechanisms, and auxiliary experimental data (apoptosis; proto-oncogene activation; tumor suppressor gene inactivation) and carcinogenesis hypotheses, one would be ill-advised to seriously consider using cell proliferation or other mechanistic findings to significantly influence the risk assessment and risk management processes necessary to protect public health. Available data are incomplete, hypotheses being promoted are under test, and the scientific debates rightly continue. Meanwhile, we need to honestly communicate not only what we do (or think we) know, but also we must disclose our lack of conclusive knowledge to the public and to the public health administrators who must deal with these frequently contentious and subjective issues. Interestingly enough, as we learn more and more about the intricacies and multiplicities of "mechanisms of carcinogenesis" and carcinogenesis processes, we have not been able to conclusively and unequivocally describe-after only four or five decades of intense searching-the complete mechanistic process for a single chemical carcinogen. Undoubtedly significant gains have been and are being made, but we still await a solitary and inarguable mechanism of action of carcinogenicity for even one chemical. For a few we are close. Some believe, including myself, that a mechanism able inform cia1 (or cla: cancer indu effects; diet tion differer disproven a CY-2p-G LOI Mean. cautious be hypothetica sions. One 1 that induce comitant wi ulin protein U.S.Envirc these experi whereas 0th erly cautiou ample, and specific nep produce can tantly, gaso with the o[ epidemiolol ney in coho epidemiolol answers. MECHANI: Thus, using experi to permit ir. known to c should cont dence, mor1 cause these 4 to believe, thraquinone other chemi tently to be classical ant in experime these need r ation to &en to lents??) aplastic celluIar ipically nuous' ' ? termi- by the if "mu- or only ?d rnul- icers in nmental otential sequent nent. rnecha; tumor &adlings to :cessary omoted need to de must : health issues. ies and ses, we four or I single : made, city for .. that a Chemical Carcinogenesis and Cell Proliferation 297 mechanism or "unified theory" of carcinogenesis may not be congruent with available information; perhaps more logically, we must continue to approach each chemcia1 (or class of chemicals) as if it exhibits a distinct mechanism or mechanisms of cancer induction (e.g., individual benzene metabolites; dioxin and receptor-mediated effects; diethylstilbestrol (DES) and hormonal perturbations), and may indeed function differently among different species or genders. Of course this hypothesis may be disproven as well, as we gather further information and more knowledge. (u-2p-GLOBULIN Meanwhile regulators and administrators should continue to be considerably cautious before embracing a purported mechanism-even a reasonable proposed hypothetical mechanism of carcinogenesis-as being pivotal to public health decisions. One example still garnering considerable debate centers on whether chemicals that induce cancers of the tubular cell epithelium of the kidney in male rats, concomitant with an increase in cell proliferation and testosterone-mediated a-2p-globulin protein, are relevant or useful to overall hazard identification for humans. The U .S. Environmental Protection Agency (for example) has embraced the concept that these experimental carcinogenic responses are not important to the human situation, whereas others (e.g., the International Agency for Research on Cancer) appear properly cautious, or have proposed equally reasonable alternative hypotheses. For example, and yet to be clarified, there are several chemicals known to induce this specific nephropathy syndrome that do not cause kidney cancers, and may or may not produce cancers in other organs; this lack of consistency must be explained. Importantly, gasoline is one complex mixture that induces these "rodent-unique'' cancers with the opine of having no real or potential risk to humans; yet accumulating epidemiological evidence shows an association with increases in cancers of the kidney in cohorts exposed occupationally to aviation fuels. Continuing experimental and epidemiological efforts should lead to more definitive mechanistic and public health answers. MECHANISM, LOGIC AND PUBLIC HEALTH Thus, practicality and prudence commands us to be convincingly certain before using experimentaldata to promote or formulate mechanistic hypotheses in an attempt to permit increased (or to promulgate decreased) levels of exposures to chemicals known to cause cancers in laboratory animals. Conversely (and meanwhile), one should continue to use in vivo data to protect public health and to reduce the incidence, morbidity, and mortality of cancers from exposures to chemicals known to cause these diseases in animals or in humans. On the other hand, we should be willing to believe, for example, that another benzidine-based dye or nitrosoamine or anthraquinone or aniline or phenylenediamine would behave toxicologically like all the other chemicals belonging to those groups already shown convincingly and consistently to be carcinogenic. Likewise, one should be reasonbly confident that the next classical antihistamine or penicillirdtetracycline antibiotic would not be carcinogenic in experimental animals. Obviously, empirical relationships or predictions such as these need mechanistic or bioassay substantiation. 298 Huff Ergo, mechanisms of action are not considered to be understood well enough to become significant factors in the evaluations of chemically induced carcinogenesis, but certainly all relevant and available biological data and other pertinent information including available mechanistic hypotheses information should be considered when deciding on the overall evidence of the carcinogenic responses. And, thus, for extrapolating or interpolating among mammalian species in risk assessment and risk management approaches. The cascade of events collectively termed carcinogenesis may eventually be deciphered fully enough to allow better predictions of carcinogenic risks to humans. After all, enriched human health and prolongation of life remain the ultimate collective goals of biomedical research efforts. ACKNOWLEDGMENTS I thank J. Carl Barrett, Sharon Bryant, John Bucher, Robert Maronpot. Ronald Melnick, and Christopher Portier for reviewing this commentary and for making encouraging and worthwhile suggestions. Jill Brazier was most gracious to initially request this topic for the Collegium Ramazzini Newsletter, and for allowing a major part to be used for this commentary. And to this Journal for suggesting a revised commentary, including an expanded reading list based mainly on relevant papers from NIEHS. REFERENCES Ames BN. Gold Ls [ 19901: Too many carcinogens: Mitogenesis increases mutagenesis. Science 249: 970-97I . Bannaxh P, Hacker H. Klimek F, Mayer D, Stumpf H, Zerban H [1991]: Cytochemical, microbiochemical and molecular genetic analysis of chemical carcinogenesis. Prog Histochem Cytochem 23~45-60. Barrett JC (19921: Mechanism of action of known human carcinogens. IARC Sci Pub1 116:115-134. Barrett JC [ 19931: Mechanisms of multistepcarcinogenesis and risk assessment. Environ Health Perspect 100:9-20. Barrett JC, Wiseman RW [1992]: Molecular carcinogenesis in humans and rodents. Comp Mol Carcinog 376: 1-30. Baserga R [1990]: The cell cycle: Myths and realities. Cancer Res 50:6769-6771. Boyd JA, Barrett JC [19901: Genetic and cellular basis of multistep carcinogenesis. Pharmacol Ther 461469-486. Foley JF, Tuck PD, Thai-vu 'IT, Frost M, Kari F, Anderson MW, Maronpot RR [1993]: Inhalation exposure to a hepatocarcinogenic concentration of dichloromethane does not induce sustained replicative DNA synthesis in hepatocytes of female B6C3F1 mice. Carcinogenesis 14:811-8 17. Fung VA, Huff JE, Weisburger E, Hoe1 DG [1993]: Predictive strategies for selecting 379 NCVNTP chemicals evaluated for carcinogenic potential: Scientific and public health impact. Fund. Appl Toxicol 20:413-436. Fung VA, Barrett JC, Huff JE (1995): Carcinogenesis bioassays in perspective: application in identifying human cancer hazards. Environ Health Perspect [in press]. Hard GC, Rodgers IS,Baetcke KP, Richards WL, McGaughy RE, Valcovic LR [1993]: Hazard evaluation of chemicals that cause accumulation of a2p.-globulin, hyaline droplet nephropathy, and tubule neoplasia in the kidneys of male rats. Environ Health Perspect 99:313-349. Hoe1 DG,Haseman JK. Hogan MD. Huff JE, McConnell EE [1988]: The impact of toxicity in carci- nogenicity studies: Implications for risk assessment. Carcinogenesis 9:2045-2052. Huff JE [ 1992al: Design strategies, results, and evaluations of long-term chemical carcinogenesisstudies. Scand J Work Environ Health 18 [Suppl. 1]:31-37. Huff JE [l992b]: A historical perspective of the classification developed and used for chemical carcin- ogens b 1]:74-I Huff JE [ 1992 parative Huff JE [ 1992i the fore Med Tc Huff JE [1993. ing cam HuffJE[1993t Perspec Huff JE [ I994 perfect I Huff JE [ 19941 rodents Press. Huff JE [ 1994i Upton Tribute Huff JE [ 1993 genesis Huff JE, Barre carcinoi Huff JE. Hoe1 the inita 18 [Sup Huff JE. Rall cology ventive IARC [ 19931: Agency Kari FW,Folc regimen 1 4 319Lijinsky W [ 1' Lucier G [ 199: Lucier GW, H epidemi Melnick RL [ 1 FASEB Melnick RL [ I 16:109- Melnick RL [ 1 globulin Melnick RL [ I carcinop Melnick RL, t hepatoc! Melnick RL, 1 chemica Monticello TM Health I Short BG 1199 115-12( Strauss BS [ 19 Tennant RW, E associatc ... .... .. - Chemical Carcinogenesis and Cell Proliferation 299 >ugh to ogens by the National Toxicology Program: 1983-1992. Scand J Work Environ Health 18 [Suppl. .enesis, mation 1 when for ex- 1]:74-82. Huff JE [ 1992~1:Chemical toxicity and chemical carcinogenesis. Is there a causal connection? A com- parative morphological evaluation of 1500 experiments. IARC Sci Pub1 116:437-475. Huff JE [1992d]: Applicability to humans of rodent-specific sites of chemical carcinogenicity: Tumors of the forestomach and of the harderian. preputial, and zymbal glands induced by benzene. J Occup nd risk Med Toxicol 1:109-141. genesis iogenic kain the Huff JE [ I993al: lssues and controversies surrounding qualitative strategies for identifying and forecasting cancer causing agents in the human environment. Pharmacol Toxicol 72 [Suppl. 13312-27. Huff JE [1993bl:Chemical and cancer in humans: First evidence in experimental animals. Environ Health Perspect 100:201-2 10. Huff JE [19944: Chemicals causally associated with cancers in humans and in laboratory animals-a perfect concordance. In Waalkes MP, Ward JM [eds]: "Carcinogenesis." New York: Raven Press. Huff JE [1994~1I:dentification of environmental agents that cause cancer: Respiratory carcinogenesisin rodents and in humans. In Waalkes MP. Ward JM [eds]: "Carcinogenesis." New York: Raven Press. Ronald Huff JE [1994dl: Chemicals and the human cancer tragedy. Tribute to Irving Selikoff. In Mehlman MA, naking Upton A [eds]: "Identification and Control of Environmental and Occupational Diseases: A nitially I Tribute to Dr. Irving J. Selikoff." Princeton, NJ: Princeton Sci. Pub. Co. (in press). I major Huff JE [ 1993fJ: Absence of morphologic correlation between chemical toxicity and chemical carcinogenesis. Environ Health Perspect 101 (suppl 5):45-54. revised Huff JE, Barrett JC [ 19941: Concepts and value of using optimum exposure levels in long-term chemical papers carcinogenesis experiments. [submitted]. Huff JE, Hoe1 DG 119921: Perspective and overview of the concepts and value of hazard identification as the inital phase of the risk assessment for cancer and human health. Scand J Work Environ Health 18 [SUPPI.1]:83-89. Huff JE. Rall DP [1992]: Relevance to humans of carcinogenesis results from laboratory animal toxi- cology studies. In Last JM, Wallace RB [eds]: "Maxcy-Rosenau-Last's Public Health and Pre- nce 249: ventive Medicine." Thirteenth Edition. Norwalk, CT:Appleton & Lange, pp 433-457. IARC [1993]: IARC [1972-19941: IARC Monogr Eva1 Carcinog Risks Hum. Vols: 1-60. International iicrobioytochem Agency for Research on Cancer. Lyon. France. Kari FW,Foley JF, Seilkop SK, Maronpot RR, Anderson M W [1993]: Effect of varying exposure regimens on methylene chloride-induced lung and liver tumors in female B6C3F1. Carcinogenesis 5-134. Perspect 14:8 19-826. Lijinsky W [19931: Species differences in carcinogenesis. In Vivo 7:65-72. Lucier G (19921: Receptor-mediated carcinogenesis. IARC Sci Pub1 116:87-112. hcinog Lucier GW, Huff JE, Tritscher AM [ 19941: Carcinogenicity of TCDD: Experimental, mechanistic and epidemiologic evidence. Ann Rev Pharmacol Toxicol 3433-377. Melnick RL [1992al: Does chemically induced hepatocyte proliferation predict liver carcinogenesis? :ol Ther FASEB J 612698-2706. Melnick RL [1992bl: Mechanistic data in scientific public health decisions. Regul Toxicol Pharmacol ihalation 16:109-1 IO. ustained Melnick RL [1992c]: An alternative hypothesis on the role of chemically induced protein droplet [a2m- 11-817. globulin] nephropathy in renal carcinogenesis. Regul Toxicol Pharmacol 16:I 1 1-125. ICYNTP Melnick RL 119931: Critique does not validate assumptions in the model on a2p-globulin and renal id. Appl carcinogenesis. Regul Toxicol Phannacol 18:365-368. Melnick RL, Huff JE 119931: Liver carcinogenesis is not a predicted outcome of chemically induced :ntifying hepatocyte proliferation. Toxicol Ind Health 9:415-438. Melnick RL, Huff JE, Bamett JC, Maronpot RR. Lucier G. Portier CJ [1993]: Cell proliferation and d evalu- chemical carcinogenesis: A symposium overview. Mol Carcinog 7: 135-138. hy, and Monticello TM. Gross EA, Morgan KT [1993]: Cell proliferation and nasal carcinogenesis. Environ Health Perspect 101 (Suppl 5):121-124. n carci- Short BG [1993]: Cell proliferation and renal carcinogenesis. Environ Health Perspect 101 (Suppl 5): 115-120. studies. Strauss BS [19921: The origin of point mutations in human tumor cells. Cancer Res 52249-253. Tennant RW, Elwell MR, Spalding JW, Griesemer RA (19911: Evidence that toxic injury is not always carcin- associated with induction of chemical carcinogenesis. Mol Carcinog 4:420-440. 300 Huff Tomatis L [1993]: Cell proliferation and carcinogenesis: A brief history and current view based on an IARC workshop report. Environ Health Perspect 101 (Suppl 5):149-152. Tomatis L, Aitio A, Day NE, Heseltine E. Kaldor J, Miller AB, Parkin DM, Riboli E (eds.)[1990]: Cancer: causes, Occurrence and control. IARC Sci Publ 100:1-352. Vainio H, Heseltine E, McGregor D, Tomatis L. Wilbourn J [1992]: Working group on mechanisms of carcinogenesis and the evaluation of carcinogenic risks [Meeting Report]. Cancer Res 52:23572361. Vainio H,Magee P, McGregor D, McMichael AJ (eds) [1992]: Mechanisms of carcinogenesis in risk evaluation. IARC Sci Publ 116:l-615. Ward JM, Uno H, Kurata Y, Weghorst CM, Jang J-J [1993]: Cell proliferation not associated with carcinogenesis in rodents and humans. Environ Health Perspect 101 (Suppl 5):125-135. Weinstein IB [19881: The origins of human cancer: Molecular mechanisms of carcinogenesis and their implications for cancer prevention and treatment. Twenty-seventh G. H. A. Clowes Memorial Award Lecture. Cancer Res 48:4135-4143. Weinstein 1B [19911: Mitogenesis is only one factor in carcinogenesis. Science 251:387-388. Weinstein IB [19921: Toxicity, cell proliferation, and carcinogenesis. Mol Carcinog 52-3. COMMEN Comme the Stei Nationa Theodor D Key words: 11 INTRODUC When difference u indeed, then Sterling/Wei approaches : Our or analysis of tl by comparir elevated lun failed to fin relative risk cumulative z 1) the total : [19861 focu: their analysi experiences males, whic white males simple circu This p( nonwhite mz exposure as Faculty of App: Canada. Address reprint Science, Simon Accepted for pi 8 1995 Wiley.