3
Radiation-Induced Cancer: Mechanisms, Quantitative Experimental Studies, and the Role of Genetic Factors

INTRODUCTION

The process of cancer development (tumorigenesis) is recognized to involve multiple changes in genes involved in cell signaling and growth regulation, cell cycle checkpoint control, apoptosis, differentiation, angiogenesis, and DNA damage response or repair. Changes in these genes can involve (1) gene mutations or DNA rearrangements, which result in a gain of function as in the case of the conversion of proto-oncogenes to oncogenes; (2) mutations or DNA deletions or rearrangements, which result in loss of gene function as in the case of tumor-suppressor genes (Kinzler and Vogelstein 1998).

The long latent period between radiation exposure and cancer development together with the multistage nature of tumorigenesis make it difficult to distinguish radiation-induced changes from those alterations that occur once the process has been initiated. Radiation-induced cancers do not appear to be unique or specifically identifiable (UNSCEAR 2000b). The mutations in tumors and their growth characteristics are not readily distinguishable from those in spontaneously occurring tumors of the same site or from tumors at the same site induced by other carcinogenic agents. Attempts to identify radiation-specific changes in human tumors have not been particularly successful despite fairly extensive investigation (UNSCEAR 1993, 2000b). There are, however, clues to possible underlying mechanisms of radiation-induced cancer that emerge from epidemiologic and experimental investigations.

Based mainly on experimental studies, it is generally believed that complex forms of DNA double-strand breaks are the most biologically important type of lesions induced by ionizing radiation, and these complex forms are likely responsible for subsequent molecular and cellular effects (see Chapters 1 and 2). Attempts to repair complex DNA double-strand lesions are judged to be error prone, and there is evidence that this error-prone repair process can lead to gross chromosomal effects and mutagenesis. Molecular analyses of radiation-induced mutations have found a full range of mutations including base-pair substitutions, frameshift mutations, and deletions. Importantly, the most common radiation-induced mutations are deletions rather than base-pair changes in genes (point mutations; Chapters 1 and 2). Therefore, theories of radiation-induced cancer have generally centered on postirradiation tumor-suppressor gene inactivation that would be expected to occur through DNA deletion rather through the induction of point mutations. Oncogene activation through specific forms of induced chromosome translocation is also a candidate radiation-associated event, particularly for leukemia and lymphoma (UNSCEAR 2000b). Thus, mechanisms involving gene and/or chromosome rearrangements and loss of heterozygosity (signaling specific regions of DNA loss) are considered the most likely radiation-induced events that contribute to cancer development (UNSCEAR 2000b).

More recently, experimental studies have questioned whether the initiating events produced by radiation are indeed direct effects on specific genes (e.g., Little 2000). Rather, it has been proposed that the gene or chromosomal mutations involved in radiation tumorigenesis arise indirectly as a consequence of persistent genomic instability (Chapter 2) induced by the radiation exposure.

This chapter focuses first on studies relevant to mechanisms of radiation-induced tumorigenesis, with particular emphasis on the potential implications for low-dose risks. Subsequently, experimental studies addressing the quantitative relationship between radiation dose and cancer development are reviewed with particular regard to their consistency with proposed underlying mechanisms and the overall implications for cancer risk at low doses.

Advances in human and animal genetics have also highlighted the contribution made to cancer risk by heritable factors (Ponder 2001). Much of the available information concerns germline genes that influence the risk of spontaneous cancer and the mechanisms through which they act. However, evidence is also emerging on the impact of such genes



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 65
3 Radiation-Induced Cancer: Mechanisms, Quantitative Experimental Studies, and the Role of Genetic Factors INTRODUCTION of radiation-induced mutations have found a full range of mutations including base-pair substitutions, frameshift mu- The process of cancer development (tumorigenesis) is tations, and deletions. Importantly, the most common radia- recognized to involve multiple changes in genes involved in tion-induced mutations are deletions rather than base-pair cell signaling and growth regulation, cell cycle checkpoint changes in genes (point mutations; Chapters 1 and 2). There- control, apoptosis, differentiation, angiogenesis, and DNA fore, theories of radiation-induced cancer have generally cen- damage response or repair. Changes in these genes can in- tered on postirradiation tumor-suppressor gene inactivation volve (1) gene mutations or DNA rearrangements, which that would be expected to occur through DNA deletion rather result in a gain of function as in the case of the conversion of through the induction of point mutations. Oncogene activa- proto-oncogenes to oncogenes; (2) mutations or DNA dele- tion through specific forms of induced chromosome translo- tions or rearrangements, which result in loss of gene func- cation is also a candidate radiation-associated event, particu- tion as in the case of tumor-suppressor genes (Kinzler and larly for leukemia and lymphoma (UNSCEAR 2000b). Thus, Vogelstein 1998). mechanisms involving gene and/or chromosome rearrange- The long latent period between radiation exposure and ments and loss of heterozygosity (signaling specific regions cancer development together with the multistage nature of of DNA loss) are considered the most likely radiation-in- tumorigenesis make it difficult to distinguish radiation-in- duced events that contribute to cancer development duced changes from those alterations that occur once the pro- (UNSCEAR 2000b). cess has been initiated. Radiation-induced cancers do not More recently, experimental studies have questioned appear to be unique or specifically identifiable (UNSCEAR whether the initiating events produced by radiation are in- 2000b). The mutations in tumors and their growth character- deed direct effects on specific genes (e.g., Little 2000). istics are not readily distinguishable from those in spontane- Rather, it has been proposed that the gene or chromosomal ously occurring tumors of the same site or from tumors at the mutations involved in radiation tumorigenesis arise indi- same site induced by other carcinogenic agents. Attempts to rectly as a consequence of persistent genomic instability identify radiation-specific changes in human tumors have (Chapter 2) induced by the radiation exposure. not been particularly successful despite fairly extensive in- This chapter focuses first on studies relevant to mecha- vestigation (UNSCEAR 1993, 2000b). There are, however, nisms of radiation-induced tumorigenesis, with particular clues to possible underlying mechanisms of radiation-in- emphasis on the potential implications for low-dose risks. duced cancer that emerge from epidemiologic and experi- Subsequently, experimental studies addressing the quantita- mental investigations. tive relationship between radiation dose and cancer develop- Based mainly on experimental studies, it is generally be- ment are reviewed with particular regard to their consistency lieved that complex forms of DNA double-strand breaks are with proposed underlying mechanisms and the overall im- the most biologically important type of lesions induced by plications for cancer risk at low doses. ionizing radiation, and these complex forms are likely re- Advances in human and animal genetics have also high- sponsible for subsequent molecular and cellular effects (see lighted the contribution made to cancer risk by heritable fac- Chapters 1 and 2). Attempts to repair complex DNA double- tors (Ponder 2001). Much of the available information con- strand lesions are judged to be error prone, and there is evi- cerns germline genes that influence the risk of spontaneous dence that this error-prone repair process can lead to gross cancer and the mechanisms through which they act. How- chromosomal effects and mutagenesis. Molecular analyses ever, evidence is also emerging on the impact of such genes 65

OCR for page 65
66 BEIR VII on radiation cancer risk (ICRP 1998). Relevant data on ge- tumorigenic development, with phase transitions being de- netic susceptibility to cancer are reviewed in the final sec- pendent on the selection and overgrowth of clonal neoplastic tion of this chapter, and some interim judgments are devel- variants best fitted for the prevailing in vivo conditions. Al- oped about their implications for radiation cancer risk in the though there are exceptions, the consensus view is that tu- population. mor initiation or promotion is a monoclonal process having its origin in the appearance of a single aberrant cell (Levy MECHANISMS OF TUMORIGENESIS and others 1994; Rabes and others 2000). The tumor initiation phase is most difficult to study di- Gene and Chromosomal Mutations in Spontaneously rectly, but in recent years it has become evident that a rela- Arising Human Tumors tively tissue-specific set of so-called gatekeeper genes Studies on the cellular and molecular mechanisms of tu- (Kinzler and Vogelstein 1997; Lengauer and others 1998) morigenesis have in recent years cast much light on the com- may be critical mutational targets for cellular entry into neo- plex multistep processes of tumorigenesis and its variation plastic pathways. Table 3-1 provides examples of such genes among tumor types. There is a vast literature on tumor biol- and their principal associated neoplasms. These gatekeepers ogy and genetics (Bishop 1991; Loeb 1991, 1994; Hartwell are frequently involved in intracellular biochemical signal- 1992; Levine 1993; Vogelstein and Kinzler 1993; Hinds and ing pathways, often via transcriptional control, and are sub- Weinberg 1994; Weinberg 1994; Boland and others 1995; ject primarily to productive loss-of-function mutations. They Karp and Broder 1995; Levine and Broach 1995; Skuse and fall into the tumor-suppressor gene category consistent with Ludlow 1995; Kinzler and Vogelstein 1998; Rabes and the germline role of many of these genes in autosomal domi- others 2000; Khanna and Jackson 2001; Balmain and others nant familial cancer (see “Genetic Susceptibility to Radia- 2003), and it is sufficient to highlight the principal points of tion-Induced Cancer,” later in this chapter). The somatic loss current fundamental knowledge that may serve to guide of function associated with gatekeeper gene inactivation can judgments on the impact of ionizing radiation on cancer risk. arise by point mutation (often of the chain-terminating type), Tumor development is generally viewed as a multistep intragenic deletion, or gross chromosomal loss events clonal process of cellular evolution that may be conveniently (Sidransky 1996; Kinzler and Vogelstein 1997, 1998). For but imprecisely divided into a number of overlapping phases: some genes, epigenetic silencing events may also be impor- (1) tumor initiation, which represents the entry via mutation tant (Jones and others 1992; Feinberg 1993, 2004; Ranier of a given normal somatic cell into a potentially neoplastic and others 1993; Merlo and others 1995; Issa and Baylin pathway of aberrant development; cellular targets for this 1996; Roth 1996). process are generally held to have stem cell-like properties; It is evident from Table 3-1 that the gatekeeper gene hy- (2) tumor promotion, which may now be viewed as the early pothesis applies principally to the genesis of solid tumors. clonal development of an initiated cell; cell-cell communi- For lymphomas and leukemia a somewhat different mecha- cation, mitogenic stimulation, cellular differentiating factors, nism appears to apply. In these neoplasms, the early produc- and mutational and nonmutational (epigenetic) processes tive events often involve chromosomally mediated gain-of- may all play a role in this early pre-neoplastic growth phase; function mutations in tissue-specific proto-oncogenes (i.e., (3) malignant conversion, which represents the tumorigenic gene activation or intragenic fusion involving juxtaposition phase where the evolving clonal population of cells becomes of DNA sequences by specific chromosomal exchange; increasingly committed to malignant development; mutation Rabbitts 1994; Greaves and Wiemels 2003). In many in- of genes that control genomic stability is believed to be par- stances, these leukemia- or lymphoma-associated chromo- ticularly important; and (4) malignant progression, which is somal events involve the DNA sequences (TCR [T cell re- itself multifaceted, is a relatively late tumorigenic phase dur- ceptor] and IG [immunoglobin]) involved in immunological ing which neoplastic cells become increasingly autonomous and gain a capacity for invasion of surrounding normal tis- sue and spread to distant sites (metastasis); the development of tumor vasculature is important for the development of TABLE 3-1 Examples of Human Tumor-Suppressor solid cancers (Folkman 1995). In addition, there is evidence Genes of the Gatekeeper Type that inflammatory processes and the microenvironment in which tumors develop are important cofactors for malignant Gene Principal Cancer Type Mode of Action progression (Coussens and Werb 2002). Overall, it is clear that only a small fraction of cells that enter tumorigenic path- APC Colon carcinoma Transcriptonal regulator ways complete the above sequence that results in overt ma- NF1 Neurofibromas GTPase-activator VHL Kidney carcinoma Transcriptional regulator lignancy (Rabes and others 2000), and that the whole pro- WT-1 Nephroblastoma Transcription factor cess can take many years. PTCH Skin (basal cell) Signaling protein The balance of evidence suggests that sequential gene and chromosomal mutations act as the principal driving force for NOTE: GTPase = guanosine 5′-triphosphatase.

OCR for page 65
RADIATION-INDUCED CANCER 67 response (Rabbitts 1994). Tumorigenic chromosomal ex- the capacity to evade or tolerate antitumorigenic defenses change events are less well characterized in solid tumors but (Tomlinson and Bodmer 1999). These defenses would in- do occur in certain sarcomas and in thyroid tumors (Rabbitts clude cell-cell communication, apoptosis, terminal differen- 1994; Mitelman and others 1997). However, in accord with tiation, cell senescence, and immune recognition (Rabes and data from solid tumors, gene deletion and other loss-of- others 2000). Gene and chromosomal mutations conferring function mutations are not uncommon in lymphohemopoietic enhanced tumor cell survival or growth characteristics have tumors (Rabbitts 1994; Mitelman and others 1997). been identified in a range of malignancies (Greenblatt and In relation to tumorigenesis in general, a second broad others 1994; Branch and others 1995; Kinzler and Vogelstein category of so-called caretaker genes has also been identi- 1998; Greider 1996; Orkin 1996). fied, although it is important to stress that the distinction In summary, gene and chromosomal mutations of the gen- between gatekeeper and caretaker genes is somewhat artifi- eral types induced by ionizing radiation are known to play a cial—there are examples of genes that fulfill both criteria. role throughout the multistep development of tumors. Loss Caretaker genes are those that play roles in the maintenance of function of gatekeeper genes may be of particular impor- of genomic integrity (Kinzler and Vogelstein 1997, 1998). tance in the initiation of common solid tumors, while gain- Table 3-2 provides examples of such tumor genes and their of-function chromosomal exchanges and gene loss events associated neoplasms. In such cases, mutational loss of func- can arise early in lymphoma and leukemia. The relatively tion can lead to deficiency in DNA damage response and early spontaneous development of genomic instability via repair, repair or recombination, chromosomal segregation, specific mutation of caretaker genes is believed to be impor- cell cycle control, and/or apoptotic response (Loeb 1991; tant for tumorigenesis in many tissues, but epigenetic gene Hartwell and others 1994; Fishel and Kolodner 1995; Kinzler silencing or activation events have also been characterized. and Vogelstein 1996, 1998). Almost irrespective of the spe- The emphasis placed here on early events in tumorigenesis cific nature of the tumor gene in question, the net result of derives from the prevailing view from epidemiologic and caretaker gene mutation is to elevate the frequency of gene animal studies that ionizing radiation acts pri.cipally as a or chromosomal mutations in the evolving neoplastic clone, tumor-initiating agent. and there is evidence that in some tumors this phenotype can arise at a relatively early point in neoplastic growth Mechanisms of Radiation Tumorigenesis (Schmutte and Fishel 1999). This increased mutation fre- quency can be seen to provide the high level of dynamic Data from quantitative animal tumorigenesis (UNSCEAR clonal heterogeneity characteristic of tumorigenesis, thereby 1988; Rabes and others 2000) and human epidemiologic facilitating the selection of cellular variants that have gained studies (UNSCEAR 1994) imply that low-LET (linear en- ergy transfer) ionizing radiation acts principally as a tumor- initiating agent. Specifically, in humans and animals, single acute doses of low-LET radiation produce a dose-dependent TABLE 3-2 Examples of Human Tumor Genes of the increase in cancer risk with evidence that chronic and frac- Caretaker Type tionated exposures usually decrease that risk. Also, experi- mental animal data show that radiation only weakly pro- Gene Principal Cancer Type Mode of Action motes the development of chemically initiated tumors, and TP53 Multiple types Transcription factor the generally greater tumorigenic sensitivity of humans to (DNA damage response) acute irradiation at young ages is more consistent with ef- fects on tumor initiation than with promotional effects that ATM Lymphocytic leukemia PI-3 kinase accelerate the development of preexisting neoplasms. (DNA damage response) In this section, molecular and cytogenetic data on radia- MSH2, Colon or endometrial DNA mismatch repair tion-associated human and animal tumors are summarized in MLH1, carcinoma the context of the mutagenic and tumorigenic mechanisms PMS discussed previously. Particular attention is given to the proposition, based on somatic mutagenesis data, that early BRCA1/2 Breast or ovarian Transcription factor arising, radiation-associated events in tumors will tend to carcinoma (DNA damage response) take the form of specific gene or chromosomal deletions or XPA-G Squamous, basal cell Nucleotide excision repair rearrangements. carcinoma, melanoma MYH Familial adenomatous Removes adenines Gene and Chromosomal Mutations in Radiation- polyposis in families that misincorporated opposite Associated Human Tumors lack the inherited the mutagenic lesion mutation in the APC gene 8-oxoguanine The acquisition of data on TP53 tumor-suppressor gene mutational spectra in human tumors associated with ultra-

OCR for page 65
68 BEIR VII violet radiation (UVR) and chemical exposures was followed of post-Chernobyl PTC (Nikiforov and others 1996; Smida by searches for potential TP53 mutational signatures in ex- and others 1997). cess lung tumors arising in Japanese A-bomb survivors and Some informative molecular data are also available for radon-exposed uranium miners (Vahakangas and others basal cell skin carcinomas (BCCs) arising in X-irradiated 1992; UNSCEAR 1993; Taylor and others 1994b; Venitt and tinea capitis patients (Burns and others 2002). In five out of Biggs 1994; Bartsch and others 1995; Lo and others 1995; five tumors analyzed there was evidence of DNA loss events Rabes and others 2000). Subsequently, attention was also which encompassed the Ptch gene (the gatekeeper for BCC given to TP53 mutations in liver tumors arising in excess in development) plus the closely linked XPA gene. patients receiving the alpha-emitting radiographic contrast Overall, the studies summarized above, together with re- agent Thorotrast (Iwamoto and others 1999). Interpretation ports on the cytogenetic characterization of acute myeloid of these data are problematical, and although one study of leukemias in A-bomb survivors (Nakanishi and others 1999) lung tumors from uranium miners was suggestive of a pos- and radiotherapy-associated solid tumors (Chauveinc and sible codon-specific mutational signature of radiation (Tay- others 1997) do not provide clear evidence on the causal lor and others 1994b), this finding was not confirmed by gene-specific mechanisms of radiation tumorigenesis. In others (Venitt and Biggs 1994; Bartsch and others 1995; Lo general however, they do support a monoclonal basis for and others 1995). The studies on liver tumors from postirradiation tumor development and suggest that the Thorotrast patients provide some comment on secondary characteristics of induced tumors are similar to those of spon- TP53 mutation and possible instability effects but, overall, taneously arising neoplasms of the same type. A possible the studies cited above do not give consistent evidence that exception to this is that an excess of complex chromosomal TP53 is a primary target for ionizing radiation. events and microsatellite sequence instability was observed A cytogenetic-molecular data set is available on papillary in late-expressing myeloid leukemias arising in A-bomb sur- thyroid cancer (PTC) (Bongarzone and others 1997) arising vivors exposed to high radiation doses (Nakanishi and oth- in excess in 131I-exposed children in areas contaminated by ers 1999); these data are discussed later in this chapter. the Chernobyl accident (UNSCEAR 2000a). These mecha- nistic studies were guided by the knowledge that chromo- Gene and Chromosomal Mutations in Animal Tumors somally mediated rearrangement and activation of the ret proto-oncogene is a frequently early arising feature of PTC Although radiation-induced tumors from experimental (Richter and others 1999). Three different forms of ret gene animals have been available for study for many years, it is rearrangement have been characterized at the cytogenetic only through advances in cytogenetics, molecular biology, and molecular levels (i.e., ret/PTC1, ret/PTC2, and ret/ and mouse genetics that it has become possible to investi- PTC3), and the prevalence of these events has been investi- gate early events in the tumorigenic process. The most infor- gated in post-Chernobyl childhood PTC (Klugbauer and oth- mative data on such early events derives from studies of tu- ers 1995; Bongarzone and others 1997; Williams 1997; mors induced in F1 hybrid mice in which specific DNA loss Smida and others 1999a, 1999b). As expected, ret activation events may be analyzed by loss of heterozygosity for events were found to be recurrent in Chernobyl-associated genomically mapped polymorphic microsatellites. childhood PTC, and a similarly high frequency has been re- ported in adult thyroid cancer of patients with a history of Mouse Lymphoma and Leukemia radiation (Bounacer and others 1997). These studies suggest that the spectra of ret mutations differ between tumors of Early studies with radiation-induced thymic lymphoma adults and children. Some investigations suggest that ret/ provided evidence of recurrent RAS gene activation and PTC3 events in post-Chernobyl childhood cases are more some indication that the RAS gene mutational spectra dif- frequent than expected. However this view is questioned by fers between X-ray and neutron-induced lymphoma (Sloan the study of 191 cases by Rabes and colleagues (2000), and others 1990). Other molecular studies include the find- which provides evidence that the spectrum of ret rearrange- ing of recurrent chromosome (chr) 4 deletions in thymic and ments may be dependent on postirradiation latency, degree nonthymic lymphomas (Melendez and others 1999; of tumor aggression, and possibly, dose to the thyroid. Kominami and others 2002) and T-cell receptor (Tcr) gene At present, causal relationships between ret gene rear- rearrangements and chromosomal events in thymic lym- rangement, childhood PTC, and radiation remain somewhat phoma. However, the above and other somatic mutations in uncertain. However, a possible clue to radiation causation is mouse lymphoma have yet to be specifically associated with the finding that breakpoints in the majority of ret rearrange- initial radiation damage. ments carry microhomologies and short direct or inverted The situation in mouse acute myeloid leukemia (AML; repeats characteristic of the involvement of nonhomologous Silver and others 1999) is clearer. AML-associated, region- endjoining (NHEJ) mediated misrepair (Klugbauer and oth- specific deletion of chr2 has been shown by cytogenetic ers 2001). Other investigations have reported that TP53 gene analysis of in vivo irradiated bone marrow cell populations mutation does not play a significant role in the development to be a direct consequence of radiation damage; clonal pre-

OCR for page 65
RADIATION-INDUCED CANCER 69 neoplastic growth of carrier cells has also been reported mice (Mao and others 2004). This study has raised the hy- (Hayata and others 1983; Breckon and others 1991; Bouffler pothesis that after radiation, the wt Trp53 gene in +/– mice and others 1997). These deletions, which are characteristic activates the Fbxw7 gene, leading to genome instability, of ~90% of AML induced by various radiation qualities, aneuploidy, and thereby increased Trp53 loss. have been analyzed in detail, and a putative myeloid sup- Radiation-induced intestinal tumorigenesis has been stud- pressor gene target was identified within a chr2 interval of ied in F1 hybrid mice of the Apc+/– genotype (Luongo and ~1 centimorgan (cM; 1 centimorgan equals about 1 million Dove 1996; van der Houven van Oordt and others 1999; base pairs; Clark and others 1996; Silver and others 1999). Haines and others 2000). In this mouse model, DNA may be Site-specific breakage of chr2 is characteristic of early ra- sampled from very small, early arising adenomas, thus fo- diation-induced events in AML, and there are cytogenetic cusing attention on early clonal events in tumor develop- and molecular data that support the involvement of telomere- ment (Levy and others 1994). Loss of wt Apc with the whole like repeat (TLR) sequence arrays in chr2 breakage and of the encoding chr18 is a relatively common early event in rearrangement at fragile sites (Finnon and others 2002). Ini- spontaneous intestinal tumorigenesis in Apc+/– mice. How- tial hypotheses on this form of postirradiation chromosomal ever, in tumors arising in low-LET-irradiated mice, the spec- fragility centered on increased recombinational activity of trum of wt Apc loss events was dominated by interstitial such TLR sequence arrays (Bouffler and others 1997). How- chromosome deletions. One study (Haines and others 2000) ever, the data of Finnon and colleagues (2002) are more con- implicated a second chr18 locus in these early radiation-as- sistent with a mechanism of domain-specific chromosomal sociated losses and also identified loss of the Dpc4 gene as a rearrangement involving chromatin remodeling that is me- common secondary event in spontaneous and induced tu- diated by TLR-associated matrix attachment sequences. mors. In some genetic backgrounds, mammary, ovarian, and With regard to radiation-induced osteosarcoma, Nathrath skin tumors also arise in excess in Apc+/– mice (van der and colleagues (2002) have provided evidence for the involve- Houven van Oordt and others 1999). ment of two tumor-suppressor gene loci, but whether these The same molecular genetic approach to experimental loci are direct targets for radiation remains to be determined. radiation tumorigenesis has been used in tumor-prone ro- Mouse genetic models of tumorigenesis have also proved dents that are heterozygous for the Ptch and Tsc-2 tumor- to be instructive about the nature of radiation-associated suppressor genes. early events in tumor induction. In these models, the Mice deficient in the patched gene (Ptch+/–) are suscep- germline of the host mouse carries an autosomal deficiency tible to both spontaneous and radiation-induced BCC and in a given tumor-suppressor or gatekeeper gene, thus expos- medulloblastoma (Hahn and others 1998; Aszterbaum and ing the remaining functional (wild-type) copy to spontane- others 1999; Pazzaglia and others 2002). Of particular note ous or induced mutation and thereby tumor initiation (see are the recent data of Pazzaglia and colleagues (2002) show- “Genetic Susceptibility to Radiation-Induced Cancer”). The ing that neonatal mice are highly susceptible to X-ray- nature of these tumor gene-inactivating events has been induced medulloblastoma and that the predominant muta- studied in models of different tumor types. tional event in these tumors is loss of Ptch+. In mice deficient in the Trp53 tumor suppressor gene Loss of Tsc-2+ was similarly observed in many X-ray- (Trp53+/– and Trp53–/–), quantitative tumorigenesis studies induced renal carcinomas of Tsc-2+/– rats (Hino and others implied that loss of the wild-type (wt) gene of Trp53+/– het- 2002), although intragenic deletions and point mutations erozygotes was a critical early event for the radiation induc- were also observed. Importantly, the data available in this tion of lymphoma and sarcoma (Kemp and others 1994). rodent genetic model (Hino and others 2002) reveal differ- Molecular analysis confirmed the loss of wt Trp53 from tu- ent spectra of tumor-associated Tsc-2+ mutations in sponta- mors but also showed a high frequency of concomitant du- neous, X-ray, and ethylnitrosourea (ENU) induced renal car- plication of mutant (m) Trp53—such duplication was much cinomas, which strongly suggests that the wt gene in target less frequent in spontaneous tumors (Kemp and others kidney cells is a direct target for carcinogens. As predicted 1994). Subsequent cytogenetic studies showed that Trp53+/– from in vitro studies on somatic mutagenesis (Thacker mice were highly prone to radiation-induced whole chro- 1986), tumors induced by the powerful point mutagen ENU mosome loss and gain (aneuploidy), and that the molecular were not characterized by Tsc-2+ gene loss events. data on tumorigenesis could be explained by radiation-in- Studies with gene knockout mice are providing further duced loss of the whole chromosome (chr11) bearing wt evidence on the role of DNA damage response genes in de- Trp53, with duplication of the copy bearing mTrp53 being termining the in vivo radiosensitivity of cells and tissue, to- necessary to regain cellular genetic balance (Bouffler and gether with the impact on growth or development and spon- others 1995). Thus, in this genetic context, Trp53 loss and taneous tumorigenesis (Deng and Brodie 2001; Kang and tumorigenesis were relatively high-frequency events depen- others 2002; Spring and others 2002; Worgul and others dent upon the cellular tolerance of aneuploidy. However a 2002). It is expected that such animal genetic models will, recent study poses questions about whether Trp53 is indeed in due course, yield more detailed information on the in vivo a direct target for radiation tumorigenesis in these knockout mechanisms of radiation tumorigenesis.

OCR for page 65
70 BEIR VII In summary, although studies with radiation-associated on current mechanistic knowledge. In this respect, the fol- human tumors have yet to yield unambiguous data on the lowing section summarizes data concerning novel aspects nature of causal gene and chromosomal mutations, animal of radiation response that may have relevance to unconven- studies are providing valuable guidance on the issue. Three tional mechanisms of radiation tumorigenesis. principal points may be made. First, mechanistic studies on murine1 AML, lymphoma or sarcoma in Tp53+/– mice, in- RADIATION-INDUCED GENOMIC INSTABILITY IN testinal adenoma in Apc+/– mice, medulloblastoma in Ptch+/– RADIATION TUMORIGENESIS mice, and renal carcinoma in Tsc-2+/– rats all argue that the induction of critical cellular events by radiation occurs early As noted earlier in this chapter, the spontaneous develop- in the tumorigenic process—a conclusion that is consistent ment of tumors is frequently accompanied by the acquisition with previous judgments on the issue. Second, the cytogenetic of genomic instability phenotypes that serve to promote the and molecular data cited for AML and intestinal tumors pro- mutational evolution of more aggressive neoplastic clones. vide evidence for early monoclonal development of charac- This form of genomic instability is increasingly well under- teristic radiation-induced pre-neoplastic changes implying stood, and many of the responsible tumor gene mutations an initial, single-cell target. Third, for induction of AML have been identified. Also noted in Chapter 2 is the large and intestinal, medulloblastoma, and renal tumors, the body of data showing that initial radiation-induced lesions radiation-associated events are predominantly DNA losses are processed rapidly and expressed as chromosome aberra- targeting specific genomic regions harboring critical genes. tions at first postirradiation mitoses. However, during the This in vivo DNA deletion mechanism is consistent with last decade, evidence has accumulated that under certain that understood in greater detail from in vitro somatic muta- experimental conditions, the progeny of cells surviving ra- tion systems. Also, many of the radiation-associated DNA diation appear to express an excess of new chromosomal and loss events recorded in tumors are of cytogenetic dimen- gene mutations over many postirradiation cell generations. sions. It is therefore possible to draw parallels with in vitro This feature of cellular response (reviewed in Chapter 2) is data on chromosome aberration induction where the pre- generically termed radiation-induced persistent genomic in- dominant importance of DNA DSB induction and post- stability. There are a variety of different manifestations of irradiation error-prone NHEJ repair has been used in this this phenomenon, and the developing field has been the sub- report to argue against the proposition of a low-dose thresh- ject of a number of recent reviews (Morgan and others 1996; old in the dose-response. Mothersill and Seymour 1998b; Wright 2000). The avail- Evidence on the single-cell origin of radiogenic animal able data do not allow for generalizations on the onset and tumors, the in vivo gene or chromosomal loss mechanism duration of such phenomena. On the basis of these data and for tumor initiation that appears to apply, and the close par- previous reports of high-frequency neoplastic cell transfor- allels that may be drawn with mechanisms and dose-re- mation (Clifton 1996), it has been suggested that epigenetic sponse for in vitro induction of gene or chromosomal muta- changes affecting a substantial fraction of irradiated cells tions argue in favor of a no-threshold relationship between can serve to destabilize their genomes and that the elevated radiation dose and in vivo tumor risk. In the examples cited, postirradiation mutation rates in cell progeny, rather than there is generally concordance between gene loss or muta- gene-specific initial mutations, act to drive radiation tumori- tional events recorded in spontaneous and radiation-associ- genesis (Little 2000; Wright 2000). This section of the chap- ated tumors of a given type; although the data are more lim- ter focuses attention on in vivo studies of induced genomic ited, such concordance tends to apply to other tumorigenic instability that address the relevance of the phenomenon to agents. A degree of gene specificity for different tumor radiation tumorigenesis. types is also evident. An obvious caveat to this conclusion is the degree to Chromatid Instability in Hematopoietic Cells which these limited mechanistic data provide support for broad judgments about radiation risk at low doses. For ex- Radiation-induced genomic instability in hematopoietic ample, the data cited on the tolerance of aneuploidy in the cells was first revealed by studies showing a persistent ex- bone marrow of irradiated Trp53-deficient mice can explain cess of chromatid-type aberrations in the progeny of mouse the high-frequency development of lymphoma but may not bone marrow cells irradiated in vitro with α-particles and be wholly relevant to other tissues and/or other genetic set- subsequently grown in culture (Kadhim and others 1992). tings. Data discussed in the following section on the poten- Alpha particles were considered to be substantially more ef- tially powerful effects of genetic background on tumori- fective than low-LET radiation in inducing this form of ge- genic risk in irradiated mice also caution against a dogmatic nomic instability (Wright 2000), which has also been re- approach to judgments about low-dose risk that are based ported in the progeny of cells that had not been traversed by an α-particle track (i.e., a bystander effect for instability; Lorimore and others 1998). Posttransplantation growth 1Genus mus. A rat or mouse. in vivo of in vitro irradiated bone marrow cells was also re-

OCR for page 65
RADIATION-INDUCED CANCER 71 ported to result in excess chromatid instability (Watson and ated BALB/c mice persistently expressed substantially more others 1996). However, on the basis of the data summarized chromatid aberrations during passage than those derived below, the consequences of postirradiation chromatid insta- from irradiated C57BL/6 animals (Ponnaiya and others 1997; bility of bone marrow cells for hematopoietic neoplasia re- Ullrich and Ponnaiya 1998). In follow-up investigations, the mains somewhat doubtful. chromatid instability phenotype of BALB/c was shown to be Cytogenetic characterization of myeloid leukemia in- associated with a partial deficiency in the NHEJ repair pro- duced in the same mouse strain by α-particles, neutrons, and tein DNA-dependent protein kinase catalytic subunit (DNA X-rays did not reveal evidence of the LET-dependent cyto- PKcs) together with compromised postirradiation DNA DSB genetic footprint of induced chromatid instability that might repair (Okayasu and others 2000). This study, which in- be expected from in vitro cellular studies with bone marrow cluded an intercomparison of inbred mouse strains, showed cells (Bouffler and others 1996). In addition, the very high deficiency of DNA-PKcs and DNA DSB repair to be re- α-particle relative biological effectiveness (RBE) for stricted to BALB/c suggesting genetic associations with per- induced genomic instability in bone marrow cells in culture sistent genomic instability and mammary tumor susceptibil- (Kadhim and others 1992) is somewhat inconsistent with the ity. In accord with this, molecular genetic analyses showed low α-particle RBE suggested to apply to leukemogenic risk BALB/c to carry a rare variant form of the gene (Prkdc) in vivo (Breckon and Cox 1990; UNSCEAR 2000b). encoding DNA-PKcs, and subsequent analysis of recombi- Early studies of this form of induced instability depended nant mice provided strong evidence that variant Prkdc di- on in vitro irradiation. Studies with humans exposed in vivo rectly determined DNA-PKcs deficiency and postirradiation to low- and high-LET radiation (Tawn and others 2000b; chromatid instability in mammary epithelial cells (Yu and Whitehouse and Tawn 2001) have found no evidence of in- others 2001). On the basis of these data it was proposed that duced chromatid instability in hemopoietic cells. The same induced genomic instability and mammary tumor suscepti- negative result was obtained experimentally in the CBA/H bility were genetically codetermined. Importantly, these mouse strain (Bouffler and others 2001). However Watson investigations provide genetic evidence that deficiencies in and colleagues (2001) provided data that suggested variable the repair of DNA DSB, rather than as-yet-undefined epige- expression of in vivo induced chromatid instability in the netic phenomena, are likely to determine persistent chroma- CBH/H mouse strain. Since CBH/H is a highly inbred strain, tid instability in this mouse. The question as to whether such such variable expression of chromatid instability cannot be instability is a primary causal element in mammary tumori- ascribed to genetic variation. Experimental factors may genesis or a secondary in vitro consequence of DNA repair therefore be of considerable importance, and relevant to this deficiency and clonal growth selection remains to be are the data of Bouffler and colleagues (2001), which indi- resolved. cate the existence of confounding stress factors that may ac- Recent studies have also suggested a linkage between count for in vitro and in vivo differences in the apparent ex- DNA-PKcs and maintenance of functional telomeres (Bailey pression of such instability. and others 2004a, 2004b). As noted elsewhere in this report, These in vivo observations cast considerable doubt on the the products of telomere dysfunction are dicentric chromo- relevance of radiation-induced chromatid instability for risk somes created by end-to-end fusion and sister-chromatid fu- of lymphohematopoietic tumors. This view is strengthened sions, both of which can be associated with breakage-fusion- by studies showing that the genetic determinants of induced bridge cycles. More recently, a second product of telomere chromatid instability in mouse bone marrow cells differ from dysfunction, fusions between telomeres and the ends of bro- those of susceptibility to induced lymphohematopoietic neo- ken DNA strands (i.e., DNA DSBs), have been described. plasia (Boulton and others 2001). A similar degree of doubt Since telomere-DSB fusions have properties that differ from has been expressed following reanalysis of genomic insta- both chromosomal end fusions and ordinary chromosome bility data (Nakanishi and others 1999, 2001) relating to aberrations, such fusions offer a potentially important new myeloid leukemia arising in A-bomb survivors (Cox and mechanism for induction of instability. These fusions appear Edwards 2002; Little 2002). to occur only under conditions of telomere dysfunction resulting from defects in the NHEJ pathway (Bailey and others 1999; Mills and others 2004). This suggests that Chromatid Instability in Mouse Mammary Epithelial Cells genomic instability as a mechanism in radiation-induced Differences in radiosensitivity and susceptibility to radia- cancer may be limited to specific circumstances in which tion induction of specific tumors among inbred mouse strains individuals harbor specific DNA-repair deficiencies. are well recognized, and there is good evidence that the BALB/c mouse is unusually sensitive to the induction of Telomere-Associated Persistent Chromosomal Instability tissue injury and mammary tumors (Roderick 1963; Storer and others 1988); on these criteria the C57BL/6 mouse falls Telomeric repeat sequences (Bertoni and others 1994) cap into the radioresistant category. Initial cytogenetic studies the ends of mammalian chromosomes and serve to protect showed that mammary epithelial cells cultured from irradi- against replicative erosion and chromosomal fusion; in nor-

OCR for page 65
72 BEIR VII mal human cells in culture, telomere shortening and instabil- gue against the direct involvement of telomeric sequence ity is a natural feature of replicative cell senescence (Harley instability in these events (Bouffler and others 1996; Finnon and Villeponteau 1995; Bacchetti 1996). In often degenerate and others 2002). forms, telomeric repeats are also found in subtelomeric and In conclusion, although the position regarding radiation- interstitial chromosomal locations, and there is some evi- induced persistent genomic instability and its causal asso- dence that these loci may act as sites at which radiation- ciation with tumorigenesis is not well understood, a few spe- induced and other forms of genomic damage are preferen- cific points can be made: tially resolved (Bouffler 1998). Early studies of the postirradiation development of chro- 1. In the case of radiation-associated persistent telomeric mosomal instability in in vitro passaged human diploid fi- rearrangement and unstable chromosome translocation junc- broblasts were suggestive of instability effects in a high pro- tions, a coherent case can be made that a certain fraction of portion of irradiated cells (Sabatier and others 1992). misrepaired genomic damage after radiation may be prone However, subsequent detailed cytogenetic analyses sug- to ongoing secondary change in clonal progeny. There is gested that passage-dependent instability in cultured human evidence that such secondary genomic rearrangement can be fibroblasts primarily takes the form of telomeric events ex- a normal component of tumor development, in which case it pressed in cell clones naturally selected by growth rate dur- is reasonable to assume that excess instability of this type ing passage (Ducray and others 1999). Overall, the data ob- could be a feature of some radiation-associated tumors, par- tained may be interpreted as initial radiation exposure ticularly those arising after high-dose irradiation where mul- bringing forward in time the natural process of clonal tiple or complex rearrangements may be expected. telomeric sequence instability associated with cell senes- 2. The genetic evidence from mouse studies that post- cence and telomere shortening. irradiation chromatid instability can be associated with mam- A different form of postirradiation telomere-associated mary tumor development is also persuasive, although it instability is expressed in a hamster-human hybrid cell sys- leaves unanswered questions on the causal role of the excess tem (Marder and Morgan 1993) where, in some clones, chro- chromatid damage observed in vitro. Thus, in certain genetic mosomal instability is persistently expressed at transloca- settings of DNA repair deficiency, a role for postirradiation tions that have telomeric sequences at their junction (Day chromatid instability in tumorigenesis appears reasonable, and others 1998). Similarly, unstable structures have been and the potential linkage with telomere dysfunction could observed in unirradiated hamster cells undergoing gene am- also be important. plification (Bertoni and others 1994), and again it may be 3. Based on the negative or inconsistent data on in vivo that radiation is inducing genomic structures that enhance induced genomic instability in bone marrow cells, the non- the natural expression of instability. sharing of genetic determinants, and the contention on data There is good evidence that telomeric sequence instabil- regarding A-bomb leukemias, induced genomic instability ity is a recurrent feature of tumorigenic development is judged unlikely to impact appreciably on the risk of (Bacchetti 1996; Chang and others 2001; Murnane and lymphohematopoietic tumors after low-dose radiation. Sabatier 2004). Of particular relevance to the question of unstable translocation junctions are the so-called segmental There are very few data on radiation-associated human jumping translocations that have been well characterized in solid tumors from which to assess the potential contribution spontaneously arising human leukemias (Shippey and others of induced genomic instability. The central problem is the 1990). In respect of radiation tumorigenesis, detailed cyto- inherent difficulty in distinguishing this specific radiation- genetic analyses suggest an excess of complex aberrations induced phenotype from spontaneously developing genomic and segmental jumping translocations in myeloid leukemias instability as a natural consequence of clonal selection dur- arising at old ages in high-dose-exposed atomic bomb survi- ing tumor development. Stated simply, does tumor instability vors (Nakanishi and others 1999). These and other data on correlate with initial radiation damage or with neoplastic excess microsatellite instability in A-bomb myeloid leuke- phenotype? mias (Nakanishi and others 2001) have been reanalyzed in This problem is well evidenced by molecular studies on respect of dose and probability of tumor causation (Cox and post-Chernobyl (Belarus) childhood thyroid cancer. Initial Edwards 2002; Little 2002). These reanalyses largely un- studies showed evidence of excess microsatellite alterations couple the expression of leukemia-associated jumping trans- in these radiation-associated tumors when compared with a locations and microsatellite instability from radiation causa- reference group of adult thyroid cancers (Richter and others tion and argue that the potential contribution of induced 1999). However, more detailed follow-up studies showed instability to leukemogenic risk is likely to be small. that the principal correlation was between microsatellite Telomeric sequence instability at radiation-associated alterations and the aggression of early arising tumors. When deletion or translocation breakpoints in mouse myeloid leu- this factor was taken into account, microsatellite loss or kemia has also been recorded; this is not a general character- mutation in the early Belarus tumors was shown to be similar istic of these tumor-associated events, and recent studies ar- to that of the adult reference cases (Lohrer and others 2001).

OCR for page 65
RADIATION-INDUCED CANCER 73 Based on consideration of the available in vivo data it is Any critical analysis of quantitative data on radiation-in- concluded that, at present, only a weak scientific case can be duced cancer requires informed selection of data sets. First, made for a discernible impact of induced genomic instability the adequacy of a study with respect to statistical power and on radiation cancer risk. This conclusion is strengthened use of appropriate analytical methodology must be consid- when account is also taken of the uncertainties noted in ered. Second, biological factors involved in the pathogen- Chapter 2 regarding the biological basis and generality of esis of specific neoplasms must be considered with respect the expression of induced genomic instability in cultured to the applicability of the experimental model to carcinogen- mammalian cells. esis in general and to cancer risk in humans in particular. Given these caveats, there are relatively few studies on ani- mal carcinogenesis where the data are sufficient to address QUANTITATIVE STUDIES IN EXPERIMENTAL the issue of dose-response relationships or the issue of dose- TUMORIGENESIS rate effects and/or fractionation effects. Those studies in which such analyses are possible are limited mainly to ro- General Aspects of Dose-Response dent studies, principally mice. Biological factors in neoplas- The preceding discussion of potential mechanisms for tic development must also be noted. radiation-induced cancer has indicated an important role for As discussed later in this chapter genetic background has radiation-induced DNA DSBs, damage response pathways, a major role in determining neoplastic development at the and gene or chromosomal mutations in the initial events lead- level of sensitivity to both initiating events and events in- ing to cancer development. On this basis it would be pre- volved in expression. Therefore even in mouse studies in dicted that the form of the dose-response for radiation-in- which there is sufficient statistical power to address ques- duced cancer and the effects of fractionation or reduced dose tions of low-dose effects and time-dose relationships, the rate on this dose-response would be compatible with such data are limited to mouse strains that are highly susceptible underlying mechanisms unless factors involved in the ex- to specific forms of neoplasias. While variations in suscepti- pression of initiated cells are limiting in neoplastic devel- bility must be considered potential confounding factors in opment. Such a mechanistic model provides specific pre- applying animal data to human risks, careful analyses of dictions with respect to dose-response and time-dose human and animal data suggest that animal data do in fact relationships for initial events and provides a framework for have predictive value—for example, they can guide judg- prediction of dose-response and time-dose effects for radia- ments on the choice of cancer risk models (Carnes and others tion-induced cancer (Ullrich and others 1987). Animal stud- 1998; Storer and others 1988). On the other hand, there are ies can be used to test these predictions. This framework is specific murine neoplasms whose pathogenesis appears to based on the αD+βD2 dose-response model for chromosome be unique to the mouse. In these specific instances it is aberration induction described in Chapter 2. For single acute unlikely that data derived using these systems would be ap- exposures the dose-response would be predicted to follow plicable to human risks. These neoplasms are identified in this model such that at low doses the relationship between sections below. cancer incidence and dose would be linear, while at higher doses this relationship would follow a function more closely Specific Murine Neoplasms related to the square of the dose. It is unlikely from a statis- tical standpoint alone that such a function could be proven to Leukemia and Lymphoma hold to the exclusion of all other dose-response models for any set of experimental data. The induction of leukemia and lymphoma has been ex- Because of this, time-dose studies using both fractionated amined in a number of murine systems, but the most exten- and low-dose-rate exposure regimens are important compo- sive quantitative data on both dose-effects and time-dose nents in testing mechanistic predictions. On the basis of this relationships are for myeloid leukemia and thymic lym- model, it would be predicted that the dose-response follow- phoma. The most comprehensive data for myeloid leukemia ing low-dose-rate exposures would be linear, with the same with respect to dose-response relationships, and fractionation slope as the linear portion of the acute dose response model. and dose-rate effects are in CBA male mice and RFM male In other words, at low doses the risk of radiation-induced mice (Upton and others 1970; Mole and Major 1983; Mole cancer is independent of the time over which exposure oc- and others 1983). Interestingly, susceptibility in female mice curs and is a cumulative function of dose. Fractionated ex- of the same strains is markedly lower. The CBA mouse has posures can further test these time-dose relationships and also been used as an important model to dissect underlying also provide information on the kinetics of processes in- radiation-induced molecular events described earlier volved. Such kinetic information, while limited, can provide (Bouffler and others 1991; Clark and others 1996; Silver and insight into the nature of cellular versus tissue effects as others 1999). For both strains, studies have been conducted major components in cancer risks in the specific experimen- over the dose range 250–3000 mGy (Upton and others 1970; tal model under study. Mole and Major 1983; Mole and others 1983). Analyses of

OCR for page 65
74 BEIR VII data sets from both strains have yielded similar conclusions. RFM mice, increased incidences of pituitary and Harderian Briefly, a number of dose-response models were seen to de- gland tumors were reported. In spite of the large numbers of scribe the data sets adequately. Data on incidence as a func- animals used, analyses of the data with respect to dose-re- tion of dose for both strains could be described by quadratic, sponse models could not distinguish between linear and lin- linear-quadratic, and simple linear dose-responses with in- ear-quadratic models (Ullrich and Storer 1979b). sufficient statistical power to exclude any of these three However, when the data for low-dose-rate exposures were models on the basis of acute exposure data alone. Fraction- considered as well, they were most compatible with a linear- ation of the dose or low-dose-rate exposures resulted in a quadratic model (Ullrich and Storer 1979c). Importantly, linear dose-response consistent with expectations of radio- with respect to low-dose effects, these data support a linear biological theory in which the dose-response is linear qua- response at low doses that is independent of exposure time. dratic for acute exposures and linear for low-dose-rate expo- Such a response is consistent with predictions of the mecha- sures, with the linear slope of the linear quadratic predicting nistic model outlined earlier in this chapter. Although the the low-dose-rate and fractionation responses. These results number of animals used was smaller, a study examining ra- are compatible with the apparent role of alterations in chro- diation-induced lung and mammary adenocarcinomas in fe- mosome 2 in initial events for murine myeloid leukemogen- male Balb/c mice reached similar conclusions with respect esis and consistent with mechanistic predictions of dose and to dose-response functions and low-dose risks (Ullrich and time-dose relationships described previously. Storer 1979c; Ullrich 1983). This model was tested further This is not the case for studies on thymic lymphoma. In in a series of experiments comparing the effectiveness of contrast to myelogenous leukemia, for which male mice are single acute exposures, acute fractionated exposures, and the most sensitive, female RFM mice are significantly more low-dose-rate exposures on the induction of lung and mam- sensitive to the induction of thymic lymphoma following mary tumors in the Balb/c mouse (Ullrich and others 1987). radiation exposures (Ullrich and Storer 1979a). For radia- Importantly, in this study the hypothesis of time indepen- tion-induced thymic lymphoma in female RFM mice, the dence of effects at low doses was critically tested and found data suggest a more complex relationship between radiation to hold. Specifically, similar effects were observed whether exposure and neoplastic development. Following single the same total dose was delivered as acute low-dose frac- acute exposures over the 100–3000 mGy dose range, no tions or as low-dose-rate exposures. simple dose-response model was found to describe the data While the data for solid tumors described above are com- (Ullrich and Storer 1979a). Low-dose-rate exposures, al- patible with mechanistic models detailed earlier in this chap- though significantly less effective with respect to induction ter, there are data sets that do not support a linear-quadratic of thymic lymphoma than single acute exposures, still re- dose-response model. Extensive data for mammary cancer sulted in a complex dose-response with a clear suggestion of induction in the Sprague-Dawley rat appear more consistent a large threshold (Ullrich and Storer 1979c). These results with a linear model over a wide range of doses and with should not be unexpected since the development of thymic linear, time-independent effects at low doses, low-dose frac- lymphoma in mice following irradiation is an extremely tions, and low dose rates (Shellabarger and others 1980). complex process. The target cells for induction of thymic Although questions have been raised about the applicability lymphoma are thought to be in the bone marrow rather than of this model system to radiation-induced breast cancer in the thymus, and the pathogenesis of the disease appears to be humans, much of the data from this rat model, from the largely mediated through indirect mechanisms with cell kill- mouse model in Balb/c mice, and from epidemiologic stud- ing playing a major role (Kaplan 1964, 1967; Haran-ghera ies in exposed human populations appear to be consistent 1976). For example, the expression of thymic lymphoma can with respect to low-dose risk functions (Preston and others be substantially reduced or eliminated by protection of bone 2002b). marrow stem cells from radiation-induced cell killing. The In contrast to the data for leukemia and for pituitary, complex nature of the pathogenesis of this disease and the Harderian gland, lung, and mammary cancer described lack of a comparable counterpart in humans argues against above, data from studies examining radiation-induced ova- thymic lymphoma as an appropriate model for understand- rian cancer in mice and bone and skin cancer in various ani- ing dose-response and time-dose relationships in humans. mal species are more compatible with threshold dose-re- sponse models. In each instance it appears that an important role for cell killing in the process of neoplastic development Solid Tumors and progression may explain these observations. Data from experimental studies examining dose-response Analysis of the dose-response for radiation-induced ova- and time-dose relationships are also available for a limited rian tumors following single acute or low-dose-rate expo- number of solid cancers in female RFM and BALB/c mice, sures in RFM female mice indicated a marked sensitivity to including pituitary, Harderian gland, lung, and breast can- induction at relatively low radiation doses, but equally im- cers (Ullrich and Storer 1979b, 1979c; Ullrich 1983). In a portantly the analysis of the data strongly supported a thresh- large study examining dose and dose-rate effects in female old dose-response model (Ullrich and Storer 1979b, 1979c).

OCR for page 65
RADIATION-INDUCED CANCER 75 In fact, this is one of the few instances for which a linear Several lines of evidence support this view. Hoshino and relationship could be rejected statistically. Studies in other Tanooka have demonstrated that small doses of beta irradia- mouse strains, while having less statistical power, also sug- tion are capable of inducing initiating alterations in mouse gest a high sensitivity to induction of ovarian tumors at rela- skin that required subsequent promotion with 4- tively low doses but with an apparent threshold (Lorenz and nitroquinoline N-oxide (4NQO) for tumors to develop. Jaffe others 1947; Ullrich and Storer 1979c). This relatively un- and Bowden (1986) have demonstrated the initiating poten- usual dose-response combining a threshold with high sensi- tial of single doses of electrons when followed by multiple tivity to induction is unique to the mouse. Ovarian cancer in exposures to the tumor-promoting agent TPA (12-O- the mouse appears to involve an indirect mechanism for in- tetradecanoylphorbol-13-acetate). Fry and his coworkers duction involving oocyte cell killing and subsequent alter- (1986) have shown that X-ray-initiated cells can be promoted ations in the pituitary ovarian hormonal interactions (Kaplan to develop skin tumors by exposure to ultraviolet light. This 1950; Foulds 1975; Bonser and Jull 1977). The hormonal group has demonstrated further that the apparent threshold alterations are the proximate cause of tumor formation, with dose-response for skin tumorigenesis can be converted to a the role of radiation being relatively indirect as a result of its linear UVR dose-response when promotion is used to maxi- cell-killing effects. Because mouse oocytes are uniquely sen- mize the expression of latent initiated cells. sitive to the killing effects of radiation (the LD50 [lethal Based on such observations it is logical to speculate that dose—50%] is ~50 mGy), ovarian tumors occur at very high the multiple high-dose fractions of radiation that are gener- frequencies following relatively low doses of ionizing radia- ally required to induce skin tumors in mouse skin are acting tion (Ullrich and Storer 1979c). A threshold appears to exist not only to initiate cells but also to induce tissue damage via because a certain level of oocyte killing is required to cause cell killing, which in turn acts as a promoting stimulus to the hormonal alterations that result in ovarian tumor forma- facilitate the progression of these initiated cells into skin tu- tion. The principal effect of lowering the dose rate is to in- mors. Likewise in the rat, the high doses required to produce crease the threshold. In the RFM mouse, estimates of thresh- tumors are likely to produce both transformation of cells and olds were reported as 110 mGy for acute exposures and sufficient cell killing to promote the transformed cells. This 700 mGy for low-dose-rate exposures (Ullrich and Storer phenomenon does not appear to be unique to these animal 1979b, 1979c). In contrast to the mouse, oocytes in humans systems. Most evidence suggests that relatively high doses are relatively resistant, with an LD50 of several grays. This of radiation are necessary to induce skin tumors in humans difference in sensitivity is apparently because mouse and and that these effects can be enhanced by exposure to UV human oocytes are at different stages of differentiation in the light from the Sun (Shore 2001). It is also important to note ovary (Brewen and others 1976). The unique sensitivity of studies by Jaffe and Bowden demonstrating that multiple low the mouse ovary to radiation makes it unlikely that results doses of radiation to the skin that did not produce tissue dam- using this model system would have general applicability to age were not effective in promoting skin tumors initiated by risks in humans. chemical agents (Jaffe and Bowden 1986). These data sup- Radiation-induced skin cancer has been studied in both port the view that the predominant role for low-dose radia- mice and rats, although the majority of such studies have tion is tumorigenic initiation. focused on the rat model because the rat is significantly more Studies of bone cancer also suggest a threshold response sensitive to skin tumor induction than the mouse (Burns and and a requirement for prolonged exposure for tumor devel- others 1973, 1975, 1989a, 1989b). In both rats and mice, opment from exposure to low-LET radiation (NCRP 1990). relatively high total doses are required to induce skin cancer, Unfortunately most of the available data have focused on and there is a clear threshold below which no tumors are observations of effects rather than dissecting potential un- seen. Multiple repeated radiation exposures are generally derlying mechanisms. Attempts have been made to model required for tumors to develop in mouse skin, while a single bone tumorigenesis however, and these models have again high dose (>10 Gy) is capable of inducing tumors in rat skin. focused on an important role for a mechanism involved in It was for skin tumorigenesis that many of the concepts of the expression of initiated cells in controlling tumor devel- multistage carcinogenesis were developed, including con- opment (Marshall and Groer 1977). Although speculative, it cepts related to initiation, promotion, and progression, and it is likely that mechanisms similar to those proposed for skin is within this framework that the data for radiation-induced tumorigenesis involving the cell-killing effects of radiation skin tumors are best considered (Jaffe and Bowden 1986; are likely involved in producing a threshold response for Burns and others 1989b). It appears from a variety of studies bone tumors. that single doses of ionizing radiation are capable of initiat- ing cells with neoplastic potential, but that these cells re- Fractionation Kinetics quire subsequent promotion in order to develop into tumors (Hoshino and Tanooka 1975; Yokoro and others 1977; Jaffe Studies using fractionation regimens have been useful in and Bowden 1986). Without this promotion these latent ini- addressing issues of time-dose relationships in radiation car- tiated cells will not express their neoplastic potential. cinogenesis. In a few instances, investigators have also used

OCR for page 65
80 BEIR VII TABLE 3-3 Examples of Autosomal Recessive Disorders of DNA Damage Response Genes Approximate Prevalence Disorder or Locus Defect Proposed Major Clinical Features Cancer (per live births) Xeroderma XP-A to XP-G Excision or Photosensitivity and cancer Squamous cell skin 1 in 250,000 pigmentosum and XPV postreplication repair of UVR-exposed skin carcinoma, basal cell carcinoma, and melanoma Cockaynes CS-A, CS-B Transcribed strand Photosensitivity, dwarfism No excess a syndrome repair Trichothiodystrophy XP-D Excision repair Photosensitivity, abnormal Variable excess (skin) a sulfur-deficient hair Ataxia- ATM Kinase activity Radiosensitivity, neuro- Lymphoma 1 in 100,000 telangiectasia and immunodeficiency Nijmegen breakage NBS NHEJ factor Radiosensitivity, Lymphoma a syndrome (Mrell/RAD50/nbs) microencephaly, immunodeficiency Fanconi’s anemia FA-A to FA-C DNA cross-link repair Bone marrow deficiency, Leukemia 1 in 300,000 skeletal abnormalities aLess than 1 in 100,000. TABLE 3-4 Examples of Autosomal Dominant Disorders of Tumor Suppressor Genes, Proto-oncogenes, and DNA Damage Response or Repair Genes Genes Approximate Prevalence Disorder or Locus Defect Proposed Cancer (per live births) Tumor-Suppressor Disorders Familial adenomatous polyposis APC Transcriptional regulation Colorectal cancer 1 in 8000 (multiple polyps) Von Hippel-Lindau disease VHL Transcriptional regulation Renal cancer 1 in 30,000 Denys Drash syndrome WT1 Transcriptional regulation Nephroblastoma (+ others) ? Neurofibromatosis type 1 NF-1 GTPase regulation Neurofibroma Schwannoma 1 in 3000 Neurofibromatosis type 2 NF-2 Cytoskeletal linkage Meningioma Neurofibroma 1 in 30,000 Nevoid basal cell carcinoma syndrome PTC Cellular signaling Basal cell skin cancer 1 in 50,000 Medulloblastoma Tuberous sclerosis TSC1 Cellular signaling Benign lesions of skin, nervous 1 in 20,000 TSC2 Cellular signaling tissue, heart, and kidneys Retinoblastoma RB1 Transcriptional regulation Retinal tumors, bone or soft- 1 in 25,000 tissue sarcoma, brain cancer, and melanoma Proto-oncogene Disorders Multiple endocrine neoplasia (2A and 2B) RET Cellular signaling Thyroid or parathyroid ? and familial medullary thyroid cancer neoplasms DNA Damage Response or Repair Disorders Hereditary nonpolyposis colon cancer MLH1, MSH2, DNA mismatch repair, Colon cancer, endometrial 1 in 2000 PMS1, PMS2 apoptosis cancer Li-Fraumeni syndrome TP53 (others?) DNA damage recognition Various 1 in 50,000 Heritable breast or ovarian cancer BRCA-1 Transcriptional regulation, Breast or ovarian cancer 1 in 1000 BRCA-2 DNA repair Breast cancer (also male)

OCR for page 65
RADIATION-INDUCED CANCER 81 degree of spontaneous tumor risk that is imposed must be NBS in Table 3-3 might be judged to exhibit increased can- sufficient to distinguish that family from others that are non- cer risk after ionizing radiation, whereas XP would not. carriers. Given that, on average, spontaneous cancer in- Stated simply, germline deficiency in the recognition and/or cidence in the general population is around 30%, the infor- repair of induced DNA damage of specific forms is expected mation currently available is restricted largely to mutations to increase the abundance of genome-wide damage in the where the cancer in question is expressed at a high relative somatic cells of body tissues. This increased mutational load frequency in gene carriers (i.e., so-called high-penetrance will tend to increase cancer risk, albeit with differing de- mutations). grees of expression among tissues. It is important to recog- Other features of importance are (1) the organ specificity nize, however, that a number of autosomal dominant condi- of many cancer-predisposing mutations, (2) the age of onset tions, particularly Li-Fraumeni syndrome (TP53+/–), are of given neoplasms in gene carriers that usually occurs at determined by genes that play more general roles in the younger ages than in noncarriers, (3) the frequent occurrence control of stress responses, apoptosis, and/or coordination of of multiple tumors in gene carriers, and (4) the substantial the cell reproductive cycle (Chapter 2). Abnormal cellular variation for cancer risk between carriers of a given gene response or cancer risk in such disorders might be expected mutation, suggestive of major influences from the genetic for a range of DNA-damaging agents including ionizing background and/or life-style of the host. These issues of heri- radiation. table cancer risk have been summarized by the International Commission on Radiological Proterction (ICRP 1998) and Tumor-Suppressor Genes more recently by Ponder (2001). The crucial point, to be For tumor-suppressor genes such as VHL and NF1 in developed later, is that current knowledge of heritable can- Table 3-4 there is no specific association with DNA damage cer susceptibility in humans is restricted largely to relatively response or repair. Accordingly there is no expectation of rare mutations of high penetrance. Cancer may be regarded increased genome-wide sensitivity to the mutagenic effects as a multifactorial disorder (see Chapter 4), and genetic of radiation. In these instances increased radiation cancer views developed from the study of other multifactorial con- risk may be anticipated on the basis of the now well-sup- ditions, such as coronary heart disease, suggest strongly that ported hypothesis of Knudson (1986). In brief, there is good there will be many more variant cancer genes having lower evidence that many tumor-suppressor type genes act as tis- penetrance than those listed in Tables 3-1 and 3-2. The cur- sue-specific gatekeepers to neoplastic pathways (Kinzler and rent lack of knowledge about the nature, frequency, and im- Vogelstein 1997). Since loss or mutation of both autosomal pact of such genes imposes fundamental limitations in re- copies of such genes from single cells is believed to be rate spect of the objectives stated earlier. limiting for the initiation of neoplastic development, tumor initiation in normal individuals is expected to be a rare cellu- Mechanistic Aspects of Genetically Determined Radiation lar event. Response A carrier of a germline mutation in a given tumor-sup- pressor gene will however show loss of function of one such In making judgments on the radiation response of can- gene copy, thus “unshielding” the second copy in all target cer-prone individuals it is valuable to consider first the somatic cells. The lifetime risk of spontaneous loss or muta- theoretical expectations that follow from current knowledge tion of that second copy from any given population of target of the cellular mechanisms that are likely to be involved in cells will be relatively high—hence the often dramatic in- cancer susceptibility. Germline mutations in DNA damage crease in organ-specific cancer risk. response or repair genes, tumor-suppressor genes, and There is also a clear expectation that exposure of the car- proto-oncogenes are considered in turn. rier individual to ionizing radiation or indeed other genotoxic carcinogens would, via the same genetic-somatic mecha- DNA Damage Response-Repair Genes nism, result in a greater-than-normal risk of organ-specific cancer. Stated simply, the enhanced radiation cancer risk in As outlined in Chapters 1 and 2, different forms of DNA the carrier individual would be driven by a reduction in the damage are recognized and processed in mammalian cells target gene number from two to one; in a given disorder the by different biochemical pathways, which share few genetic organs at increased risk would tend to be the same as those determinants. Accordingly, there is no expectation of a glo- involved in spontaneous neoplasia. bal association between DNA damage response or repair deficiency and sensitivity to the tumorigenic effects of ra- Proto-oncogenes diation. Rather, the expectation is that a deficiency of genes associated with recognition or repair of the form of damage There are few well-characterized germline, gain-of-func- that is critical for cellular response to radiation (i.e., DNA tion mutations in proto-oncogenes that have unambiguous DSB) will be of greatest significance for radiation cancer associations with cancer risk; a series of characterized ret risk. On this basis the autosomal recessive disorders AT and gene mutations are however known to increase the risk of

OCR for page 65
82 BEIR VII thyroid neoplasia (Table 3-4). As in the case of tumor-sup- In summary the evidence available on human chromo- pressor gene loss, germline ret mutation may be viewed as somal radiosensitivity suggests that AT and NBS may be up removing one early rate-limiting step in multistage thyroid to tenfold more sensitive than normal; some uncertainty sur- tumorigenesis such that the carrier individual is at increased rounds the chromosomal radiosensitivity of other cancer- risk of neoplastic development via the accumulation of fur- prone disorders, but any such increase in sensitivity appears ther mutations in other genes. Again, greater-than-normal to be modest—not more than two- to threefold. Although radiation risk to the target organ should be anticipated. critical data are lacking, it is a reasonable assumption that, in In the following sections, the above propositions are ex- general, a heritable increase in chromosomal radiosensitiv- amined on the basis of available cellular, animal, and epide- ity would be associated with increased radiation cancer risk, miologic data. albeit with possible differences in the response of different tissues. Data from G2 chromosomal radiosensitivity assays are generally supportive of this association, but some data Cellular Data on Heritable Human Radiosensitivity remain controversial. Cellular data on heritable radiosensitivity in respect of cell inactivation have been reviewed recently (ICRP 1998). Animal Data on Radiosensitivity and Tumorigenesis In brief, although there are isolated instances of cancer and/ or radiotherapy patients showing clear evidence of radiosen- The experimental data available about the impact of heri- sitivity, it is only for AT and NBS that there is unambiguous table factors on radiosensitivity and tumorigenesis derive evidence of profoundly increased radiosensitivity to cell principally from studies on the genetic homologues of some killing associated with known human disorders of DNA of the human disorders listed in Tables 3-3 and 3-4. These damage response or repair and cancer. Claims for increased studies are summarized in Table 3-5 with references. radiosensitivity in other cancer-prone disorders remain con- Although there are some differences in the patterns of troversial and do not provide clear guidance on radiation phenotypic expression, in the main the rodent genetic homo- cancer risk. logues of AT, Li-Fraumeni syndrome (LFS), familial Although sensitivity to cell killing after radiation may at adenomatous polyposes, neroid basal cell carcinoma syn- present not be a particularly useful surrogate for cancer risk, drome (NBCCS), and tuberous sclerosis recapitulate many there are closer parallels between the induction of chromo- of the features of their human counterparts. In respect of some damage and cancer. Although not without some uncer- early responses, Atm–/– mice show extreme radiosensitivity; tainty, the data accumulating on the patterns of chromosomal there is also evidence of moderate in vivo radiosensitivity in radiosensitivity in human cancer-prone disorders are worthy Atm+/– mice. Studies with Atm+/– knockout mice (Barlow and of some attention. These data, considered by Scott and col- others 1999) provided evidence of increased in vivo radio- leagues (1998) and reviewed by the National Radiological sensitivity but failed to demonstrate differences in radiation Protection Board (NRPB 1999) show that, compared with induced tumorigenesis between +/– and +/+ genotypes. healthy controls, cells cultured from AT and NBS patients However, more recent data on spontaneous tumorigenesis typically exhibit two- to threefold greater chromosomal ra- (Spring and others 2002) imply that such studies are best diosensitivity, but in some cytogenetic assays, the increased conducted with Atm knock-in mice, which recapitulate sensitivity can be up to tenfold (Taalman and others 1983; known human mutations. Taylor 1983). The NRPB has summarized a large body of Data on BRCA1- and BRCA2-deficient mice have yet to cytogenetic data on which claims of associations between provide clear evidence on the role of these genes in radiation chromosomal radiosensitivity and human cancer suscepti- tumorigenesis. The principal benefit of the referenced stud- bility have been based. As in the case of cell killing, some of ies noted in Table 3-5 is the provision of a growing associa- these claims remain controversial. More recent studies on tion between the Brca genes, Rad51, cell cycle perturbation, the possible radiosensitivity of cells from breast cancer-sus- and DNA damage response. ceptible BRCA1 and BRCA2 patients have also provided The most valuable animal genetic data on radiation tum- conflicting evidence (Buchholz and others 2002; Trenz and origenesis have been developed from studies on mice others 2002; Powell and Kachnic 2003). Of additional inter- heterozygously deficient in the tumor-suppressor genes est are the data on G2 cell cycle radiosensitivity, which Tp53, Apc, and Ptch and in a rat strain (Eker) heterozygously among other findings suggest that AT heterozygotes are deficient in Tsc2 (see Table 3-5 for references). In all in- indeed radiosensitive and that up to 40% of unselected breast stances, the germline mutational loss of one copy of the re- cancer cases also exhibit modestly elevated radiation-in- spective tumor-suppressor gene leads not only to an increase duced chromosome damage (Scott and others 1994; Parshad in the rate of spontaneous tumorigenesis but also to increased and others 1996). There is also some evidence of elevated sensitivity to the induction of the same tumor types by whole- chromosomal radiosensitivity in cells from patients with body low-LET radiation with doses up to around 5 Gy. malignant gliomas (Bondy and others 1996) and colorectal These data provide strong support for the contention, dis- cancer (Baria and others 2001). cussed earlier, that the unshielding of tumor-suppressor

OCR for page 65
RADIATION-INDUCED CANCER 83 TABLE 3-5 Radiation Response and Tumorigenesis in Rodent Homologues of Cancer-Prone Human Genetic Disorders Radiation Response Genotype Human Homologue Early response Tumorigenesis Comment Key References Mouse Atm–/– Ataxia- Radiosensitive May be dependent upon Defects in meiosis, Barlow and others (1996) telangiectasia (AT) in vivo or in vitro Atm genotype (see text) immunity, and behavior Elson and others (1996) Xu and others (1996) Mouse Brcal–/– Heritable breast Cellular and No published study Embryolethal; Gowen and others (1998) BRCA+/– cancer in embryonic identified association with Sharan and others (1997) heterozygotes radiosensitivity Rad51–/– phenotype Mizuta and others(1997) Connor and others (1997) Mouse Tp53+/– Li-Fraumeni Excess aneuploidy Highly sensitive to induction Tumorigenesis Kemp and others (1994) syndrome (LFS) and G2/M of lymphoma or sarcoma associated with loss Bouffler and others (1995) checkpoint defect of Tp53+ in bone marrow cells Mouse Apc+/– Familial None reported Highly sensitive to induction Tumorigenesis Luongo and Dove (1996) adenomatous of intestinal adenoma (breast associated with loss Ellender and others (1997) polyposis and other cancers in some of Apc+ and other loci van der Houven van Oordt genetic backgrounds) and others (1999) Haines and others (2000) Mouse Ptch+/– Nevoid basal cell Some evidence Sensitive to induction of Tumorigenesis Hahn and others (1998) carcinoma of cellular medulloblastoma associated with loss Pazzaglia and others (2002) syndrome radiosensitivity of Ptch+ Rat Tsc2+/– Tuberous sclerosis None reported Sensitive to induction of Tumorigenesis Hino and others (1993, 2002) renal neoplasia associated with loss of Tsc2+ genes by germline mutation will lead to a significant increase During the last few years the impact of such modifier in individual susceptibility to radiation tumorigenesis. Criti- genes on the expression of tumorigenesis in mice has been cal mechanistic support for this hypothesis has been pro- demonstrated more clearly (Balmain and Nagase 1998). The vided by molecular analysis of tumors arising in irradiated principal message from this experimental work is that be- Tp53+/–, Apc+/–, and Ptch+/– mice and Tsc-2+/– rats; as pre- cause of the strongly modifying effects of genetic back- dicted, such analyses strongly suggest that radiation acts by ground, rodent homologues are unlikely to provide a quanti- inactivating the wild-type tumor-suppressor gene copy in tatively reliable representation of radiation tumorigenesis in target somatic cells. These wild-type genes appear to be cancer-prone human genetic disorders. Such genetic modifi- mutated by radiation through mechanisms principally in- cation is to be expected in humans, but the specific nature volving substantial DNA loss events, although there are ex- and impact of the modifier genes are likely to differ among amples of whole chromosome losses as well as intragenic species. The issue of genetic modification of radiation re- deletions and point mutations. sponse is considered further in the section of this chapter Although the above studies provide proof-of-principle that deals with cancer-predisposing mutations of low pen- experimental evidence of strong genetic effects on radiation etrance. tumorigenesis in mammalian species, quantification of the genetically imposed radiation risk is most problematical. An Human Data on Radiosensitivity and Tumorigenesis ICRP (1998) Task Group, in reviewing much of the data of Table 3-5, suggested that radiation tumor risk in such sup- As noted earlier in this chapter unambiguous evidence of pressor-suppressor gene-deficient mice might be elevated by human genetic disorders showing hypersensitivity to tissue up to a hundredfold or more but cautioned against firm judg- injury after radiation is confined to AT and NBS, where con- ments because of (1) problems associated with experimental ventional radiotherapy procedures have proved disastrous to design and (2) preliminary evidence that natural variation in patients. Adverse, but less profound, reactions to radio- the genetic background of host animals can have major modi- therapy are however reported to occur in around 5% of can- fying effects on tumor yield. cer patients (Burnet and others 1998). Studies on in vitro

OCR for page 65
84 BEIR VII cellular radiosensitivity in such radiotherapy patients have, the large size of the U.S.-based epidemiologic studies of Eng so far, failed to reveal evidence of strong correlations be- and colleagues (1993) and Wong and coworkers (1997a) al- tween in vivo and in vitro responses although subsets of these lows some judgments to be developed on the degree to which patients do show statistically significant increases in cellular this suppressor gene disorder predisposes to (second) radio- radiosensitivity under some assay conditions (Burnet and genic soft-tissue sarcoma and bone cancer. Although there is others 1998). Similarly limited molecular studies show no a clear dose-response for radiation tumorigenesis, these data correlation between adverse reactions to radiotherapy and imply that excess relative risk (ERR) in heritable RB pa- heterozygous ATM gene mutation (Appleby and others 1997; tients may be lower than in the nonheritable controls. Burnet and others 1998). The question as to whether adverse The background rate of tumorigenesis in RB is, as ex- tissue reaction to radiotherapy signals potentially increased pected, rather high, and for the purposes of this report, ex- risk of therapy-related second tumors has yet to be addressed cess absolute risk (EAR) may be a more useful measure of in epidemiologic studies. tumorigenic radiosensitivity than ERR. In considering this Postradiotherapy observations on specific sets of cancer issue, the ICRP (1998) and NRPB (2000) suggest that the patients have, however, revealed valuable information on EAR in heritable RB is around fivefold higher than in the genetic associations with risk of second tumors (Meadows nonheritable group. It is notable that low values of ERR for 2001). These data are summarized and referenced in Table 3- radiogenic cancer in such cancer-prone conditions are con- 6. In brief, there is evidence of an excess of radiotherapy sistent with other epidemiologic data on radiation tumori- (RT)-related tumors in the human cancer-prone conditions genesis where high background cancer rates also tend to be heritable retinoblastoma, NBCCS, and LFS plus related con- accompanied by lower ERRs. Abramson and colleagues ditions, as well as in children from families with a history of (2001) have also reported on third tumors in RB patients early onset cancer. In addition there are reports suggesting after radiotherapy. As might be expected, the sites of these that neurofibromatosis is a positive factor for RT-related tu- additional tumors generally accorded with the irradiated vol- morigenesis (Robison and Mertens 1993). By contrast, a ume of normal tissue. variety of studies discussed by Mark and colleagues (1993) In summary, although clinical and epidemiologic data on provide no clear evidence that genetic factors are important RT patients are limited, they are sufficient to confirm the for RT-related breast cancer. Recent studies provide no evi- view developed from mechanistic knowledge and experi- dence that the status of BRCA genes influences post- mental studies that human genetic susceptibility to sponta- radiotherapy outcomes at 5 years (Pierce and others 2000). neous tumorigenesis is often accompanied by an increase in In Table 3-6 the data suggesting that NBCCS and LFS absolute cancer risk after ionizing radiation. Quantifying that patients have substantial increases in tumorigenic radiosen- risk is problematical, but the single study on RB patients that sitivity are in accord with data obtained experimentally with has this capacity is suggestive of relatively modest (about their rodent genetic homologues. For retinoblastoma (RB), fivefold) increases over that of normal individuals. In the TABLE 3-6 Postradiotherapy Observations on Risk of Second Tumors in Humans Genetic Disorder or Study Group First Tumor Observations Key References Retinoblastoma Retinoblastoma Excess bone tumors and soft-tissue sarcomas, large cohorts; Tucker and others (1987a) some dose, dose-response, and risk estimates possible Eng and others (1993) Wong and others (1997a) Abramson and others (2001) NBCCS Medulloblastoma Excess basal cell skin neoplasms and ovarian fibromas, Strong (1977) short latency; case reports only Southwick and Schwartz (1979) LFS and related Various Follow-up of children developing posttherapy soft-tissue sarcoma, Strong and Williams (1987) conditions bone tumors, and acute leukemia—linkage with family histories of cancer Heyn and others (1993) Robison and Mertens (1993) Malkin (1993) Case-control study Various Excess posttherapy tumors in children from non-LFS families with a Kony and others (1997) of therapy-related history of early onset cancer second tumors

OCR for page 65
RADIATION-INDUCED CANCER 85 future, the growing capacity of molecular screening tech- appreciably distort current estimates of radiation cancer risk niques to detect cancer-susceptible genotypes in the general in the population. Stated simply, only a very small fraction population will, in principle, allow the radiation risk of such of excess cancers in an irradiated human population are ex- genotypes to be assessed in a number of suitable human co- pected to arise in individuals carrying familial cancer genes. horts. A summary of such molecular epidemiologic ap- The ICRP (1998) and NRPB (1999) stressed, however, proaches to spontaneous cancer risk is given later in this that this conclusion took no account of the presence of po- chapter. tentially more common cancer genes of low penetrance that do not express familial cancer. The ICRP and NRPB reports also commented on the problems inherent in identifying and Population Modeling of Radiation Cancer Risk: Impact of making judgments about radiation cancer risk in genetic sub- Strongly Expressing Genetic Disorders groups carrying such weakly expressing genes and consid- In conjunction with the work of an ICRP (1998) Task ered the issue of genetically imposed risk to individuals. Group, Chakraborty and colleagues (1997, 1998a) have con- These matters are discussed in subsequent sections. structed and illustrated the use of a population-based com- putational model that serves to describe the impact of can- Genes of Low Penetrance cer-susceptible genotypes on radiation cancer risk in the population. For reasons of data sufficiency, breast cancer As noted earlier in this chapter, knowledge of heritable risk in typical Western populations was considered and il- factors in tumorigenesis stems largely from studies on lustrated. This approach, which is based on established Men- strongly predisposing autosomal dominant familial traits and delian principles, employed best estimates of the prevalence autosomal recessive disorders having unambiguous pheno- of known, high-penetrance breast cancer-predisposing genes types. The problem of estimating the heritable impact on (BRCA1 and BRCA2), the relative risk of spontaneous cancer risk from weakly expressing genes of low penetrance breast cancer in such genotypes, and a range of factors that and other genetic modifiers of the cancer process has been describe in a hypothetical fashion the increase in radiation with us for some time. However, not unexpectedly, an un- risk imposed by the given gene mutations; the risk of radio- derstanding of this issue is proving difficult to obtain. To a genic breast cancer in normal individuals was based on data large measure this is due to the likelihood that, individually, from Japanese atomic bomb survivors. polymorphic variant genes probably contribute small addi- Other issues that were considered included increased gene tional cancer risks to each carrier in a largely tissue-specific frequency in certain genetically isolated populations (Ash- manner. These will tend to escape detection by conventional kenazi Jews) and the influence of reduced penetrance on medical genetic and epidemiologic studies. A combination population risk. The following points summarize the out- of such genes and their interaction with environmental risk come of these modeling exercises. factors may, however, provide a substantial genetic compo- nent to both spontaneous and radiation-associated risk. The • Using best estimates of breast cancer gene frequencies, magnitude of this risk in a given human population would the genetic impact on excess breast cancer in an irradiated then be determined by gene frequencies together with the Western population would be small even if these mutations pattern or strength of gene-gene and gene-environment in- were to impose a radiation risk that was as much as a hun- teractions. dredfold greater than that of normal genotypes. These issues of population cancer risk have been dis- • Using estimates of the higher gene frequencies in cussed widely in the context of epidemiologic and molecular Ashkenazi Jewish populations, the genetic impact on radia- genetic findings (Hoover 2000; Houlston and Tomlinson tion-associated breast cancer can become significant but only 2000; Lichtenstein and others 2000; Peto and Mack 2000; if the genetically imposed radiation risk is very high. Shields and Harris 2000; Dong and Hemminki 2001; • The genetic impact of such mutations will be diluted in Nathanson and Weber 2001; Ponder 2001). Here it is suffi- proportion to decreasing penetrance. cient to illustrate some of the progress being made in respect of the weakly expressing genetic component of human and This model and its predictions have been used by the animal tumorigenesis. Where possible, emphasis is placed ICRP (1998) and NRPB (1999) to provide interim judgments on data having some connection with cancer risk after ioniz- on the implications of genetic susceptibility to cancer for ing radiation. radiological protection. Since the overall prevalence of highly penetrant cancer- Human Breast Cancer predisposing mutations in typical human populations is judged to be 1% or less (ICRP 1998) and since available data BRCA1 and BRCA2 genes have been identified as the tend to argue against extreme increases in genetically im- principal genetic determinants of the 2–5% of breast cancer posed radiation cancer risk, there is reason to believe that the that expresses in multiple-case families; other, more weakly presence of these rare, highly penetrant mutations will not expressing genes involved in familial breast cancer remain

OCR for page 65
86 BEIR VII to be uncovered (Nathanson and Weber 2001; Ponder and colleagues (2000) specifically considered a total of 89 2001). However, epidemiologic evidence is highly sugges- second breast cancer cases. None of the cases studied car- tive of a more extensive genetic component to breast cancer ried ATM mutations. risk (Peto and Mack 2000), and much effort is being ex- The second line of evidence concerns the inheritance of pended to identify the functional gene polymorphisms that chromosomal radiosensitivity and its association with breast might be involved. Although some of the evidence remains cancer risk (Roberts and others 1999). In brief, in studies on controversial, Dunning and colleagues (1999) and Na- cultured blood lymphocytes, up to around 40% of un- thanson and Weber (2001) note the potential involvement selected breast cancer cases were shown to exhibit an ab- of polymorphic genes that encode steroid hormone recep- normal excess of chromatid aberrations following X-irradia- tors and paracrine growth factors (e.g., AR, CYP19) together tion in the G2 phase of the cell cycle. By contrast, this with genes involved in the metabolism of chemical species chromosomal trait was seen in only around 5% of age- (e.g., GSTP1) and in DNA damage response (e.g., ATM, matched controls. Follow-up family studies provided evi- RAD51, TP53). The most persuasive evidence on breast dence on the heritability of the trait, which, although not of cancer genes other than BRCA1 and BRCA2 concerns the a simple Mendelian form, could be genetically modeled. As cell cycle checkpoint kinase gene CHEK2. A truncating yet there is no evidence on the specific genes involved. germline deletion of this gene is present in around 1% of In summary, advances in breast cancer genetics do allow healthy individuals and is estimated to result in about a two- the construction of a general scheme to describe the interac- fold increase of breast cancer risk in women and about a tive genetic component of familial risk, including some al- tenfold increase in men (Meijers-Heijboer and others 2002). lowance for common genes of low penetrance (Ponder Two data sets have some association with cancer risk after 2001). Polygenic computational models describing the radiation. overall genetic component of spontaneous breast cancer risk First is the question of breast cancer risk in individuals in the population are also under development (Antoniou and who are heterozygous carriers of the ATM mutation of the others 2002). Although gene candidates and cellular pheno- highly radiosensitive disorder AT. ATM carriers (ATM+/–) types may prove to be instructive, there is at present little to might represent 0.25–1% of the general population, and guide specific conclusions on the question of the common there is evidence of modestly increased cellular radio- genetic component of radiation-associated cancer risk. The sensitivity in ATM+/– genotypes. It is therefore reasonable evidence available would tend to argue against a major to consider an increased risk of radiogenic breast cancer in overall impact on radiation breast cancer risk from the ATM these carriers. Considerable effort has been expended on gene in its heterozygous form, although specific ATM geno- molecular epidemiologic analysis of spontaneous breast types may, in principle, carry substantially increased risk. cancer risk in ATM+/– women (Bishop and Hopper 1997; ICRP 1998; Broeks and others 2000; Laake and others Human Colonic and Other Neoplasms 2000; Geoffroy-Perez and others 2001; Olsen and others 2001; Teraoka and others 2001). Although the position re- There is evidence that the genetic component of colonic mains somewhat uncertain, it seems reasonable to conclude cancer also includes a significant contribution from genes of that while increased breast cancer risk may be associated low penetrance. In a recent review of 50 studies on the po- with ATM+/– in some cohorts, the relative risk is likely to be tential impact of common polymorphisms, Houlston and modest (<3), and the overall impact on spontaneous breast Tomlinson (2001) identified significant associations with cancer risk in the population is rather small. Some data sug- risk for APC-I1307K, HRAS1-VNTR, and MTHFR-Val/Val. gest, however, that it is only certain dominant negative mis- For TP53, NAT1, NAT2, GSTM1, GSTT1, and GSTP1 poly- sense mutations of ATM that predispose to cancer (Khanna morphisms, the evidence was weaker. Specific data relating 2000; Chenevix-Trench and others 2002), and for these, the to gene polymorphisms and radiation risk are lacking al- relative risk may be substantially higher. The critical ques- though, as for breast cancer, there is some evidence of an tion is whether the ATM+/– genotype may more specifically association between colon cancer risk and lymphocyte chro- and significantly increase breast cancer risk after radiation. mosomal radiosensitivity (Baria and others 2001). For good scientific reasons, some early claims on substan- Finally, in illustration of ongoing work, it is relevant to tial risks at low doses are not regarded as being well mention polymorphic associations between GSTP1 and che- founded (see ICRP 1998). While a modestly increased con- motherapy-related leukemia (Allan and others 2001), tribution of the ATM+/– genotype to radiogenic cancer risk MCUL1 and uterine fibroma (Alam and others 2001), should not be discounted, three recent studies on patients GFRalpha1 and medullary thyroid carcinoma (Gimm and developing second cancers after RT argue against a major others 2001), PPARG and endometrial carcinoma (Smith and impact from the ATM gene (Nichols and others 1999; others 2001), and TP53 and adrenal cortical carcinoma Broeks and others 2000; Shafman and others 2000). In total, (Ribeiro and others 2001). In their review of gene-environ- these studies considered 141 patients with second cancers; ment interactions, Shields and Harris (2000) focus on lung the studies of Shafman and colleagues (2000) and Broeks cancer risk, and in this area, Bennett and colleagues (1999)

OCR for page 65
RADIATION-INDUCED CANCER 87 have provided evidence on the potential impact of GSTM1 functional gene polymorphisms identified in mice may not allelic status on tobacco-related lung cancer risk. predict those of humans precisely, the overall pattern of can- The broad but incomplete picture that emerges from these cer risk modification should provide broad guidance on the studies is of some associations between gene polymorphisms potential for such effects in humans. and risk for a range of human tumor types, as well as the Much of the research on the role of germline polymor- clear need for larger and more definitive studies. phic loci in mouse tumorigenesis has centered on spontane- ous and chemically induced neoplasms. These studies in- clude tumors of the skin (e.g., Nagase and others 2001; Human DNA Repair Gene Polymorphisms Peissel and others 2001), lung (e.g., Lee and others 2001; It has already been noted that DNA repair genes play a Tripodis and others 2001), and intestinal tract (e.g., van crucial role in cellular responses to radiation and that major Wezel and others 1996; Angel and others 2000). The most germline deficiencies in these genes can lead to heritable important messages to emerge from these studies are that predisposition to cancer. Accordingly, considerable effort is multiple common loci can exert complex patterns of control being expended in the search for common functional poly- over tumor susceptibility and resistance (synergistic and an- morphisms that might act as low-penetrance cancer suscep- tagonistic interaction), that the loci tend to be relatively tis- tibility genes. sue specific in their activity, and that genetic determinants of A series of studies have identified common and less com- spontaneous and induced tumorigenesis are often shared. A mon polymorphisms in around ten DNA repair genes, some particularly revealing conclusion from the study of Tripodis of which appear to have cellular consequences (Price and and colleagues (2001) is that as many as 60 loci may interact others 1997; Shen and others 1998; Mohrenweiser and Jones to determine the risk of a single tumor type; specific pairwise 1998; Duell and others 2000). The associations between interaction of a proportion of these loci was also demon- these polymorphisms and radiosensitivity and/or tumor risk strated. remain unclear, although there are some positive indications A second approach used in mouse genetic studies is to (Duell and others 2001; Hu and others 2001). Much of this seek evidence of natural polymorphic loci that modify the work has centered on genes involved in base- or nucleotide- tumorigenic expression of a major cancer-predisposing excision repair (Miller and others 2001). Studies on genes germline mutation. In this way, evidence has been obtained controlling DNA DSB repair are less well developed. How- for substantial genetic modification of tumorigenesis in ever, there are indications that a relatively common (in ~6% Trp53- (Backlund and others 2001) and Apc-deficient mice of the population) functional polymorphism in the XRCC2 (van der Houven van Oordt and others 1999; Moser and oth- gene of the homologous recombinational repair pathway for ers 2001). In the case of Apc, one of these modifier genes DNA DSBs associates with a modestly increased risk of (Pla2g2a) has been identified provisionally (Cormier and breast cancer (Kuschel and others 2002; Rafii and others others 2000). In general, these effects of genetic modifiers 2002). A significant association between breast cancer risk are again consistent with the potential interaction of multiple and certain polymorphisms of NHEJ DNA repair has also tissue-specific loci, and some of the data relate to tumors been reported (Fu and others 2003). A recent review of DNA induced by ionizing radiation. repair gene polymorphisms and cancer risk recommends Some studies in this area have the specific objective of large, well-designed studies that include consideration of mapping and characterizing the polymorphic loci that influ- relevant exposures (Goode and others 2002). ence tumorigenic radiosensitivity and tumor characteristics. Multiple loci have been shown to influence susceptibility to radiation-induced lymphoma and leukemia (Balmain and Genetic Studies with Animals Nagase 1998; Szymanska and others 1999; Saito and others The recognized difficulties of resolving the modifying 2001; Santos and others 2001). One study of Boulton and effects of low-penetrance genes on human cancer risk have colleagues (2001) provided evidence that the AML loci de- prompted experimental genetic studies with rodent models termining leukemia or lymphoma susceptibility were distinct in which genetic-environmental interactions can be more from those that influenced genomic instability in bone mar- closely controlled. row cells. However, no candidate genes were identified. This approach has been applied principally in mice for Genetic loci influencing the susceptibility of mice to the study of naturally arising polymorphic variation that in- α-particle (227Th)-induced osteosarcoma have also been fluences spontaneous cancer risk and the risk after exposure mapped (Rosemann and others 2002), but again, no candi- to chemical carcinogens and, in a few instances, ionizing date genes were specifically identified. radiation (Balmain and Nagase 1998). These studies have By contrast, another set of investigations has associated a the capacity to provide proof-of-principle evidence of the strain-specific functional polymorphism of the gene Prkdc impact of such common loci, together with their possible encoding DNA PKcs with induced genomic instability, DNA interactions and tissue specificity, as well as the classes of DSB repair deficiency, and susceptibility to radiation-in- genes and mechanisms involved. Thus, although specific duced breast cancer (Okayasu and others 2000; Yu and oth-

OCR for page 65
88 BEIR VII ers 2001). This same Prkdc polymorphism has also been founded, but rather the extent to which genetic distortion of implicated in radiation-induced lymphomagenesis, as a the distribution of this risk might lead to underprotection of modifier of induced intestinal neoplasia in Apcmin mice (Degg an appreciable fraction of the population. In this respect, and others 2003), and as a candidate gene for the Rapop1 some initial guidance for thought is already available from apoptosis-controlling locus (Mori and others 2001). Other the data discussed in this chapter. tissue-specific loci that control apoptosis have also been These data suggest large numbers of loci of low pen- genomically mapped (e.g., Weil and others 2001). etrance with relatively small individual effects and a signifi- With respect to breast cancer susceptibility in mice, it is cant degree of locus-specific interaction and tissue specific- already clear that loci other than Prkdc can be involved ity that may apply to their activity. Projecting this scenario (Moser and others 2001). From recent studies, it seems likely to a range of radiogenic tumors in a genetically heteroge- that one such gene is ATM, which in the heterozygous form neous human population would tend to lead to a situation in can enhance the frequency of both genomic instability and which the balance between a certain set of tumor susceptibil- ductal dysplasia of the breast of irradiated mice (Weil and ity (S) and resistance (R) loci in a given subgroup might others 2001). serve to emphasize risk in a given set of organs. Equally, however, the balance of additional S and R locus combina- tions might provide a degree of resistance to the induction Conclusions and development of cancer in other organs. Thus, with this Although much remains to be learned about genetic sus- first genetic scenario, major distortions of the distribution of ceptibility to the tumorigenic effects of radiation, it is pos- overall cancer risk after radiation might not apply simply sible to frame some interim conclusions of the role it may because different genetic susceptibilities would tend to “av- play in determining radiation cancer risk at the individual erage out” across organs. By contrast, a second hypothetical and population levels. scenario involves a small subset of common polymorphic The principal point to emphasize is that cancer is a multi- loci that exert organ-wide effects on tumor susceptibility or factorial set of diseases, and as such, there is expected to be resistance, which might be particularly strong in the specific a complex interplay between multiple germline genes and a instance of radiation exposure (e.g., functional polymor- plethora of other host- and environment-related factors. The phisms for genes involved in initial tissue-wide cellular re- data available, although far from complete, tend to support sponse to radiation damage). In this instance, genetically this basic expectation. The key issues and arguments are determined distortion of the distribution of overall cancer given here in brief summary. risk might be expected. At present, the data available are For rare major gene deficiencies in humans and mice, insufficient to distinguish the likely contributions from these there can be strong effects on radiation cancer risk, and for two genetic scenarios. individual carriers, it seems likely that the greatest implica- Finally, the large study of cancer concordance in 90,000 tions may be for the risk of second cancers after RT (see Nordic twin pairs should be noted. Lichtenstein and col- ICRP 1998). Although the data are sparse, such high-dose leagues (2000) and Hoover (2000) make some important radiation exposure in childhood may carry the greatest risk. points about the difficulties that exist in separating the ge- However, due to differences in genetic background, a uni- netic and environmental components of cancer. In essence, formity of tumorigenic response in RT patients with major Hoover notes that this Nordic study, like others, is consistent gene deficiencies should not be expected. with the presence of low-penetrance cancer-predisposing The fact that strongly expressing cancer-prone disorders genes in the general population. However, the confidence are so rare argues against a significant impact and distorting intervals for the heritable component of cancers at common effect on estimates of cancer risk in irradiated populations; sites were wide—all ranged from around 5 to 50%. It was population genetic modeling fully supports this view (see also pointed out that for cancer at common sites, the rate of ICRP 1998). By contrast, at the level of whole populations it concordance in monozygotic twins was generally less than is feasible that certain inherited combinations of common 15%. Thus, the absolute risk of concordance of site-specific low-penetrance genes can result in the presence of subpopu- cancer in identical genotypes sharing some common envi- lations having significantly different susceptibilities to spon- ronmental factors is rather low. In addition to this, a study taneous and radiation-associated cancer. In due course, the based on the Swedish Family Cancer Database (Czene and accumulation of sufficient molecular epidemiologic data others 2002) has provided further information on the genetic may allow for some meaningful theoretical modeling of the component of organ-specific cancer. With the exception of distribution of radiation cancer risk and the possible impli- the thyroid, the environment appears to have the principal cations for radiological protection. Irrespective of such mod- causal role for cancer at all sites. eling, risk estimates based on epidemiologic evaluation of One important message that emerges from current data whole populations will encompass this projected genetic on cancer genes of low penetrance and the overall genetic heterogeneity of response. Therefore, the key issue is not component of cancer is that predictive genotyping of indi- whether the estimate of overall cancer risk is genetically con- viduals for the purposes of radiological protection may not

OCR for page 65
RADIATION-INDUCED CANCER 89 be feasible in the medium term. The likely involvement of for some in vitro manifestations of induced genomic in- multiple and relatively organ-specific sets of polymorphisms stability. The data considered in this chapter did not reveal and gene-gene or gene-environment interactions makes the consistent evidence for the involvement of induced genomic provision of meaningful judgments on risk most uncertain. instability in radiation tumorigenesis, although telomere-as- For these reasons it may be more realistic at this stage of sociated processes may account for some tumorigenic phe- knowledge to focus attention on general patterns of gene- notypes. A further conclusion was that there is little evidence radiation interactions and their implications for population of specific tumorigenic signatures of radiation causation, but risk, rather than risk for specific individuals. rather that radiation-induced tumors develop in a tumor-spe- cific multistage manner that parallels that of tumors arising spontaneously. However, further cytogenetic and molecular SUMMARY genetic studies are needed to reduce current uncertainties In this chapter, the committee has reviewed cellular-mo- about the specific role of radiation in multistage radiation lecular and animal studies relevant to the complex multi- tumorigenesis; such investigations would include studies stage process of radiation tumorigenesis. Attention has also with radiation-associated tumors of humans and experimen- been given to evidence from various studies on the inherited tal animals. factors that influence radiation cancer risks. The principal objective of this work was to provide judgments on radiation Quantitative Studies of Experimental Tumorigenesis cancer risk of prime importance to radiological protection, particularly where these judgments serve to couple informa- Quantitative animal data on dose-response relationships tion about the action of radiation on cells (Chapters 1 and 2) provide a complex picture for low-LET radiation, with some with the epidemiologic measures of risk considered in sub- tumor types showing linear or linear-quadratic relationships sequent chapters. while other studies are suggestive of a low-dose threshold, particularly for thymic lymphoma and ovarian cancer. How- ever, since the induction or development of these two cancer Mechanisms of Radiation Tumorigenesis types is believed to proceed via atypical mechanisms involv- A critical conclusion on mechanisms of radiation tumori- ing cell killing, it was judged that the threshold-like re- genesis is that the data reviewed greatly strengthen the view sponses observed should not be generalized. that there are intimate links between the dose-dependent in- Radiation-induced life shortening in mice is largely a re- duction of DNA damage in cells, the appearance of gene or flection of cancer mortality, and the data reviewed generally chromosomal mutations through DNA damage misrepair, support the concept of a linear dose-response at low doses and the development of cancer. Although less well estab- and low dose rates. Other dose-response data for animal tu- lished, the data available point toward a single-cell (mono- morigenesis, together with cellular data, contributed to the clonal) origin for induced tumors and indicate that low-dose judgments developed in Chapters 10 and 12 on the choice of radiation acts predominantly as a tumor-initiating agent. a DDREF for use in the interpretation of epidemiologic in- These data also provide some evidence on candidate, radia- formation on cancer risk. tion-associated mutations in tumors. These mutations are Adaptive responses for radiation tumorigenesis have been predominantly loss-of-function DNA deletions, some of investigated in quantitative animal studies, and recent infor- which are represented as segmental loss of chromosomal mation is suggestive of adaptive processes that increase tu- material (i.e., multigene deletions). This form of tumorigenic mor latency but not lifetime risk. However, these data are mechanism is broadly consistent with the more firmly estab- difficult to interpret, and the implications for radiological lished in vitro processes of DNA damage response and mu- protection remain most uncertain. tagenesis considered in Chapters 1 and 2. Thus, if as judged in Chapters 1 and 2, error-prone repair of chemically com- Genetic Susceptibility to Radiation-Induced Cancer plex DNA double-strand damage is the predominant mech- anism for radiation-induced gene or chromosomal injury The review of cellular, animal, and epidemiologic or clini- involved in the carcinogenic process, there can be no expec- cal studies on the role of genetic factors in radiation tumori- tation of a low-dose threshold for the mutagenic component genesis shows that there have been major advances in under- of radiation cancer risk. standing, albeit with some important knowledge gaps. An One mechanistic caveat explored was that novel forms of important conclusion is that many of the known, strongly cellular damage response, collectively termed induced ge- expressing, cancer-prone human genetic disorders are likely nomic instability, might contribute significantly to radiation to show an elevated risk of radiation-induced cancer, prob- cancer risk. The cellular data reviewed in Chapter 2 identi- ably with a high degree of organ specificity. Cellular and fied uncertainties and some inconsistencies in the expres- animal studies suggest that the molecular mechanisms un- sion of this multifaceted phenomenon. However, telomere- derlying these genetically determined radiation effects associated mechanisms did provide a coherent explanation largely mirror those that apply to spontaneous tumorigenesis

OCR for page 65
90 BEIR VII and are consistent with knowledge of the somatic mecha- tumorigenesis. Attention has also been given to recent mo- nisms of tumorigenesis reviewed earlier in this chapter. In lecular epidemiology data on associations between func- particular, evidence has been obtained that major deficien- tional polymorphisms and cancer risk, particularly with re- cies in DNA damage response and tumor-suppressor-type spect to DNA damage response genes. Some issues of study genes can serve to elevate radiation cancer risk. Limited epi- design have been discussed, and although much work has demiologic data from follow-up of second cancers in gene been reported on cancer risk in heterozygous carriers of the carriers receiving radiotherapy were supportive of the above ATM gene, clear judgments about radiation risks remain conclusions, but quantitative judgments about the degree of elusive. increased cancer risk remain uncertain. However, since ma- Given that functional gene polymorphisms associated jor germline deficiencies in the genes of interest are known with cancer risk may be relatively common, the potential for to be rare, it is possible to conclude from published analyses significant distortion of population-based risk was explored, that they are most unlikely to create a significant distortion with emphasis on the organ specificity of the genes of inter- of population-based estimates of cancer risk. The major prac- est. A preliminary conclusion is that common polymor- tical issue associated with these strongly expressing cancer phisms of DNA damage response genes associated with or- genes is judged to be the risk of radiotherapy-related cancer. gan-wide radiation cancer risk would be the most likely A major theme developing in the whole field of cancer source of major interindividual differences in radiation genetics is the interaction and potential impact of more response. weakly expressing variant cancer genes that may be rela- Although good progress is being made, there are impor- tively common in human populations. Knowledge of such tant gaps in understanding the extent of genetic influences gene-gene and gene-environment interactions, although at on radiation cancer risk. Accordingly, further work is needed an early stage, is developing rapidly. The animal genetic data in humans and mice on gene mutations and functional poly- reviewed in this chapter provide proof-of-principle evidence morphisms that influence radiation response and cancer risk. of how such variant genes with functional polymorphisms Human molecular genetic studies should, where possible, be can influence cancer risk, including limited data on radiation coupled with epidemiologic investigations.