INTRODUCTION

Cancer prevention and control has traditionally focused on modifiable risk factors for cancer such as cigarette smoking and diet and on screening for disease precursors such as dysplastic nevi and cervical dysplasia (1).  However, the identification and evaluation of inherited risk is likely to play an increasingly important role in cancer prevention strategies (2-5)

In recent years, the focus has shifted from rare cancer syndromes--for example, Li-Fraumeni syndrome and Von Hippel Lindau syndrome--to common, adult-onset disorders that are important causes of morbidity and mortality (6).  Examples discussed in this chapter include the identification of mutations in the hMSH2 and hMLH1 genes as causes of susceptibility to colorectal cancer and mutations in the BRCA1 and BRCA2 genes identified as causes of susceptibility to breast and ovarian cancer. 

The identification and modification of environmental risk factors among persons with an inherited susceptibility to cancer is a new paradigm for cancer prevention and control (2, 5).  For example, genetic testing for colorectal cancer may improve the predictive value of environmental factors such as high-fat diet, inadequate intake of vegetables and fruit, and physical inactivity (4).  The identification of gene-environment interactions in the etiology of breast cancer could result in preventive and therapeutic interventions for women at risk for the disease (3).  For example, interactions might be identified between ATM gene mutations and non-genetic factors such as ionizing radiation and cigarette smoking (7).  Genetic tests for other conditions may allow for the identification of subgroups of individuals who are more or less likely to benefit from preventive strategies that are of uncertain value in the general population, such as prostate specific antigen (PSA) testing for prostate cancer.

GENETIC POLYMORPHISMS AND CANCER SUSCEPTIBILITY 

Polymorphisms versus mutations

Highly penetrant mutations, such as those in the BRCA1 or MSH2 gene, are the most prominent examples of genetic traits causing a cancer predisposition.  However, polymorphisms, or common genetic variants, may also cause an increased risk for disease (8,9).  Only a few examples of this phenomenon have been documented, but with progress in mapping the human genome (10), more are likely to be found.  Mutations that produce small increases in risk--i.e., low penetrance mutations--are also likely to be found.  The distinction between a polymorphism and a low penetrance mutation is somewhat arbitrary, and is based on prevalence: genetic alterations are generally considered polymorphisms if their prevalence is greater than 1%. 

Three multiple gene "superfamilies" (cytochrome P450s, N-acetyl transferase, and glutathione S-transferase genes) are briefly described below because of their importance to cancer genetics and public health.  These polymorphically expressed genes may be positively (or inversely) associated with susceptibility to cancer and several other diseases because of their important role in the detoxification (or activation) of xenobiotics and environmental chemicals (9, 10).  Studies that have examined associations with genetic polymorphisms and specific cancer sites (for example, breast, colorectal, and lung) are discussed later in this chapter.

Cytochrome P450 enzymes are a multiple gene "superfamily" that plays an important role in the detoxification of xenobiotics such as polycyclic aromatic hydrocarbons, benzo(a)pyrene, arylamines, and heterocyclic amines (12).  Although P450 cytrochromes provide a line of defense against exposure to environmental chemicals, carcinogens may be activated by P450 metabolism.  Cytochrome P450 enzymes are primarily expressed in the liver and are expressed in other tissues (8,12). Of these, the CYP1A1 gene  (Aryl Hydrocarbon Hydroxylase Genotype), located on chromosome 15q, codes for Aryl hydrocarbon hydroxylase (AHH)(8).   AHH catalyzes the monooxygenation of polycyclic aromatic hydrocarbons to phenolic products and epoxides that may be carcinogenic.  The CYP2D6 gene (Debrisoquine Hydroxylase Genotype) is located on chromosome 22q and codes for debrisoquine hydroxylase (8,12).  Debrisoquine hydroxylase metabolizes a variety of drugs and other xenobiotics.  Like other polymorphically expressed P450 enzymes, it may activate procarcinogens or, conversely, detoxify carcinogens (8) 

N-acetyl transferase 1 and 2 

The N-acetyl transferase-1 (NAT1) and N-acetyl transferase-2 (NAT2) genes are located on chromosome 8q (8).  Both are polymorphically expressed in a variety of tissues. NAT2 detoxifies or, conversely, activates aromatic amines found in tobacco smoke such as 4-aminobiphenyl (13).  Both phenotypic assays and genotypic assays for NAT2 can be used to classify individuals as rapid or slow acetylators. 

Glutathione S-transferase  

The glutathione S-transferase-M1(GSTM1) gene is located on chromosome 1 and the gene for glutathione S-transferase-T1 (GSTT1) is located on chromosome 11q (8).  Glutathione S-transferases detoxify a variety of carcinogens and cytotoxic drugs (for example, benzo[a]pyrene, monohalomethanes, and solvents) by catalyzing the conjugation of a glutathione moiety to the substrate (14).  The incorporation of glutathione increases the molecule's water solubility and excretability (8).  Individuals who are homozygous carriers of deletions in the GSTM1 or GSTT1 genes may have a higher risk of cancer because of their impaired ability to metabolize and eliminate carcinogens (14). 

BREAST AND OVARIAN CANCER SUSCEPTIBILITY GENES  

Breast cancer is the most common cancer among U.S. women (15).  In 1998, about 178,700 new cases will be diagnosed and 43,500 women will die from the disease (15).  Although less common than breast cancer, ovarian cancer is also an important cause of premature mortality and morbidity in women. Two of the strongest risks factors for breast cancer are age and family history (16,17).  Women who have a first-degree relative with breast cancer have a two- to three-fold increased risk of developing breast cancer.  However, the degree of risk greatest when the affected relative is diagnosed at a young age and at or near average risk when the affected relative is elderly at the time of diagnosis.  Approximately 20% of breast cancer patients have a family history of the disease (18).  Of these cases, only a small minority have features characteristic of high risk families, such as early age at onset, bilaterality, and occurrence in multiple generations. (19).  

Breast cancer is likely to be etiologically heterogenous, as are other chronic diseases.(23).  However, the overall contribution of genetic factors to these cancers is likely to be considerable.  Mutations in the BRCA1 and BRCA2 genes are well established but rare causes of increased cancer susceptibility (3).  Other genetic traits, such as ATM gene mutations and genetic polymorphisms, are likely to account for a genetic predisposition to breast cancer in a larger number of people, though their effect on individual risk may be smaller (3,16). 

BRCA1 and BRCA2 gene mutations 

The BRCA1 gene, on chromosome 17, and the BRCA2 gene, on chromosome 13, were initially identified through linkage studies in families with early-onset familial breast and ovarian cancer (20,21).  Multiple different BRCA1 and BRCA2 mutations have been identified in such families.  These advances led to the rapid development and commercialization of genetic tests for breast cancer susceptibility.  

Laboratory Testing for BRCA1 and BRCA2 Gene Mutations 

Because of the heterogeneity of BRCA1 and BRCA2 mutations, laboratory testing for all possible mutations is challenging.  False positive results may occur as a result of benign polymorphisms that are not associated with an increased risk for breast cancer (30).  Most of these are missense mutations that do not result in a change in protein structure.  Frame-shift or nonsense mutations, which account for about 90% of known BRCA1 mutations, are assumed to predispose to disease because of their effect on protein structure; most produce truncation of the protein product (22).  False negative results of genetic testing may also occur, because testing for hundreds of mutations is not practical with current technology.  Some mutations in families with a cancer predisposition linked to a specific BRCA locus remain to be identified, possibly because they occur in non-coding regions of the gene. 

Several methods can be used to detect mutations.  Direct DNA sequencing can detect sequence variation.  Manual sequencing is labor intensive for a large gene such as BRCA1 or BRCA2 but rarely misses mutations in the coding sequence (23).  Single-strand conformation polymorphism (SSCP) assay does not detect all sequence changes but can detect most DNA sequence variation; fragments that have shifted mobility on SSCP gels can then be sequenced to determine the exact nature of the DNA sequence variation (23).  Once a mutation is known, allele-specific oligonucleotide hybridization can be used to rapidly screen many samples for that same mutation. 

Clinical Significance of BRCA1 and BRCA2 Gene Mutations 

Nearly all published studies of mutations in the BRCA1 and BRCA2 genes have involved members of high risk families. Research entry criteria have typically included multiple family members with early age at onset of cancer or multiple tumors, and pedigrees meeting stringent criteria for autosomal dominant inheritance of cancer predisposition (24).  Women from such cancer-prone families who test positive for a BRCA1 or BRCA2 gene mutation may have very high lifetime risks of cancer. Estimates of the penetrance in these high risk families indicate that the lifetime risk of breast cancer in mutation carriers is similar for both genes (84% cumulative risk by age 70 for BRCA2 compared to 85% for BRCA1)(24,25), but that the risk of early breast cancer is somewhat lower for BRCA2 (28% cumulative risk by age 50 compared to 51% for BRCA1.  The estimated risk of ovarian cancer is also lower for BRCA2 (27% cumulative risk by age 70 compared to 66% for BRCA1).  Men who are carriers of mutant BRCA1 and BRCA2 alleles are at increased risk for prostate cancer and also can transmit the cancer susceptibility gene mutation to their children (25).  Males with breast cancer have been seen in association with mutations from both genes, but are more commonly found with BRCA2 mutations.  The cancer risk for BRCA2 carriers may also include pancreatic cancer (25). 

Carrier Frequency of BRCA1 and BRCA2 Gene Mutations 

Using inferential procedures, Struewing et al. estimated that the overall carrier frequency of BRCA1 gene mutations is 1 in 500 in the general U.S. population (95% confidence interval: 1 in 300 to 1 in 800)(26).  A similar estimate (1 in 833, or between 1 in 500 and 1 in 2000) was obtained by Ford et al. (27). 

Similarly, among women with early breast cancer identified through a registry in the UK,  mutations of BRCA1 or BRCA2  were found in only 5.9% of those diagnosed before age 36 and in only 4.1% of  women between the ages of 36 and 45 (30).  Because some mutations may be missed by current molecular techniques, the investigators estimated that the true percentages of mutation carriers were 9.4% and 6.6% respectively for the two age groups; even with this correction, BRCA1 and BRCA2 mutations accounted for only a small proportion of breast cancer cases (30).

Some BRCA1 and BRCA2 mutations occur with increased frequency among certain ethnic groups.  The 185delAG mutation in the BRCA1 gene and the 6174delT mutation in the BRCA2 gene are each found in about 1% of persons of Ashkenazi Jewish descent (26,31,32).  The 999del5 mutation in the BRCA2 gene occurs in about 0.6% of the general population (33).    All these mutations have been found in cancer-prone families; they have also bee found in cancer patients with less dramatic family histories of cancer.   Estimates of penetrance for the mutations found in Ashkenazi Jewish populations, derived from a volunteer survey and from clinical case series, indicate a lifetime risk of breast cancer of 37-56% (34-36).  The 6174delT mutation has been observed less frequently than 185delAG among Jewish women with early onset breast cancer, so it may be less penetrant at early ages (26,31,32).  A population-based study of the 999del5 mutation in Iceland indicates a lifetime risk of breast cancer of 36% (33). 

  Interactions of BRCA1 and BRCA2 Gene Mutations With Modifiable Risk Factors For Breast Cancer

The variable age at onset of hereditary breast cancer and the incomplete penetrance of BRCA1 or BRCA2 gene mutations suggest that other host or environmental factors may modify the expression of these traits (37).  Data are scant on the role of environmental risk factors for breast cancer among women with BRCA1 or BRCA2  gene mutations.   For example, the risks and benefits of estrogen replacement therapy or oral contraceptives in women who carry these cancer susceptibility gene mutations are largely unknown (38), although one study found that women carrying BRCA1 or BRCA2 mutations experienced a reduction on ovarian cancer risk with oral contraceptive use, similar to the risk reduction seem among women in the general population (39).  Data are also lacking about the value of lifestyle modifications, such as a low-fat diet, adequate intake of vegetables and fruit, and regular exercise for these women.

There is also uncertainty about treatment recommendations for women with BRCA1 or BRCA2 gene mutations.  The only available means of primary prevention identified so far are prophylactic mastectomy and oophorectomy. The value of chemopreventive therapy, such as tamoxifen or roloxifene, in women who carry these breast and ovarian cancer gene mutations is unknown.  A task force convened by the NIH Cancer Genetics Studies Consortium (CGSC) suggested annual mammography and annual or semiannual clinical breast examination, beginning at age 25 to 35 years, for women who have BRCA1 and BRCA2 mutations, but noted that these recommendations are based upon expert opinion and are not of proven benefit (38).  The CGSC task force also recommended ovarian cancer screening with pelvic ultrasound and serum CA-125 measurements (38), again based on expert opinion.   Research aimed at improved measures to reduce risk in genetically susceptible persons was identified as a high priority by the task force (38).

ATM gene mutations

Ataxia-telangiectasia (A-T) is an autosomal recessive neurologic syndrome associated with unusual sensitivity to ionizing radiation (40,41).  Homozygotes have about a 100-fold greater risk of cancer (mostly leukemia, lymphoma, and other cancers arising in childhood and adolescence) as compared with the general population (41,42).  Women who are heterozygous carriers of mutations in gene for A-T (referred to as the ATM gene) have been reported to have an increased risk of breast cancer (40,43).  Easton (43) obtained a pooled relative risk estimate from four studies of 3.9 (95% CI 2.1 to 7.2).  More recent studies that have employed molecular techniques have provided inconsistent evidence of a role for ATM gene mutations in patients with breast cancer. Estimates of the heterozygote frequency of ATM gene mutations in the general population, obtained using estimates of the total number of individuals with A-T in the population, have ranged from 0.5% to 2.8% (43,44).  The ATM gene was cloned and sequenced in 1995 after the disease was mapped to a region of chromosome 11 in linkage studies (45).  More than 100 mutations of the ATM gene have been reported to date. 

Public Health Impact of ATM Gene Mutations

Data on the association between ATM mutations and breast cancer are conflicting.   Swift et al. (46) examined cancer incidence and mortality among 1599 adult blood relatives of patients with A-T and 821 of their spouses, in a sample of 161 families with A-T.  The estimated risk of cancer of all types among heterozygous carriers of ATM gene mutations, as compared with noncarriers, was 3.8 in men and 3.5 in women (p < .005).  The relative risk of breast cancer among women was 5.1 (p < .01)(46).

In a series of unrelated breast cancer cases with a family history of breast cancer and/or a family history of tumors previously associated with A-T homozygosity or heterozygosity,  3 of 88 (3.4%) of the breast cancer patients had a germ-line mutation in the ATM gene (47).  Athma et al. (48) determined the ATM gene carrier status of 775 blood relatives in 99 A-T families using DNA linkage studies.  They found 33 women with breast cancer who could be genotyped; of these, 25 were ATM heterozygotes compared with an expected number of 14.9 (odds ratio = 3.8, 95% CI 1.7 to 8.4) (48).  For the 21 breast cancers with onset before age 60, the odds ratio was 2.9 (95% CI 1.1 to 7.6) and for the 12 cases with onset at age 60 or older, the odds ratio was 6.4 (95% CI 1.4 to 28.8).

Conversely, a study by Fitzgerald et al. (49) of 401 women with early-onset breast cancer (onset before the age of 40), which employed a protein-truncation assay, found no evidence for a role for the ATM gene in the cases.  Only two mutations (representing 0.5% of the cases) were found in the ATM gene, as compared with only 1% (2 of 202) of the convenience (blood donor) controls (49).

One possible explanation for these conflicting findings across studies is that ATM gene mutations confer susceptibility to breast cancer only in later life, or only in combination with an exposure such as ionizing radiation (50).  False negative test results or chance variation could also account for this inconsistency.

Polymorphisms associated with breast cancer risk

N-acetyl transferase 2 genetic polymorphisms have been examined in relation to cigarette smoking and breast cancer susceptibility in women (13,50-53).  No consistent pattern has emerged.  In a study by Ambrosone et al. (13), risk for breast cancer was increased among cigarette smokers having the slow acetylator genotype.

Genes that code for cytochrome P450 enzymes involved in the metabolism and transport of estrogen, may influence breast cancer risk in women (11).  For example, the CYP17 A2/A2 and A2/A1 genotypes have been associated with a young age at first menstruation and an increased risk for breast cancer (54).  In a nested case-control study of Asian, African-American, and Latino women in Los Angeles, California and Hawaii, the odds ratio associated with the A2 allele was 2.5 (95% CI 1.07-5.94) for regional or metastatic breast cancer (54).  Because the allele is common in the population (roughly 40%), the risk for breast cancer attributable to this genetic polymorphism may be substantial (perhaps as high as 29% if preliminary findings are confirmed).

No consistent pattern has emerged from molecular epidemiology studies of breast cancer that have examined associations with CYP1A1, CYP2D6, or GSTM1 (8), perhaps because of ethnic differences in the frequencies of these alleles or population differences in other risk factors for breast cancer (55).

COLORECTAL CANCER SUSCEPTIBILITY GENES

Colorectal cancer is one of the most common cancers among U.S. men and women (15).  In 1998, an estimated 131,600 new cases will be diagnosed, and 27,900 men and 28,600 women will die from the disease. About 95% of colorectal cancer cases are sporadic and occur outside of the well-described syndromes (56).  About 10% of U.S. adults have a first degree relative with a history of colonic cancer.  Such persons have a twofold to threefold increased risk of developing colon cancer (57). Most people with a family history of colorectal cancer lack the pedigree characteristics of autosomal dominant inheritance of cancer predisposition (56).  Further, colorectal cancer, similar to other chronic diseases, is likely to have multiple causes and no single gene is likely to account for a significant attributable fraction of colorectal cancer cases.

The inherited syndromes of colorectal cancer can be roughly divided into those that exhibit colonic polyposis (e.g, familial adenomatous polyposis [FAP] and related syndromes) and those that do not (56-58).   The nonpolyposis syndromes include hereditary nonpolyposis colorectal cancer (HNPCC).  Persons with the nonpolyposis syndromes may have some adenomatous polyps, but these are fewer (56).  While the adenomatous polyposis syndromes account for only about 0.5% of colonic cancer cases, the nonpolyposis syndromes may account for 5% or more (56).  Familial cases outside of the syndromes may account for 10%-30% of colonic cancer cases.

Familial adenomatous polyposis (FAP)

FAP is an autosomal dominantly inherited disease that occurs in only about 1 in 10,000 persons (56).  Persons with this condition develop hundreds to thousands of adenomatous polyps of the colon at a young age, and colonic cancer occurs at an average age of 39 years (57).  Total colectomy and mucosal proctectomy is recommended because of the extreme risk for colonic cancer (57).   FAP is caused by highly penetrant mutations of the adenomatous polyposis coli (APC) tumor suppressor gene on the long arm of chromosome 5.  Persons in most families with FAP have distinct mutations of the APC gene, and these mutations are almost always associated with truncation of the APC protein (i.e., frameshift or nonsense mutations).  Some APC mutations cause additional distinctive clinical features in addition to polyposis (56).

Hereditary nonpolyposis colon cancer (HNPCC)

HNPCC, which is much more common than FAP, also occurs in an autosomal dominant fashion (58).  HNPCC affects about 1 in every 200 to 400 persons in the United States, which makes it one of the most common autosomal dominantly inherited diseases (59).  Colonic cancer occurs at an average age of 45 years in these persons (56).  The clinical criteria for diagnosis of HNPCC are three family members who have colonic cancer (two of whom are first degree relatives of the third person), colonic cancer cases spanning at least two generations, and at least one case diagnosed before age 50 (58).  In addition to an increased risk of colorectal cancer, women with HNPCC have an increased risk of endometrial cancer, and other cancers, including ovarian cancer, and urinary tract cancers have been observed in some families (60).

The discovery of DNA abnormalities associated with genomic instability (referred to as microsatellite instability) in persons with HNPCC led to findings of a defect in the DNA mismatch repair system (61).   The hMSH2 mismatch repair gene was identified and cloned in 1993, shortly after a region on chromosome 2p was associated with HNPCC (62,63).  A second mismatch repair gene associated with colorectal cancer, hMLH1, was subsequently identified on chromosome 3p (64).

HNPCC is now known to be caused by germline mutations of any of four (or more) DNA mismatch repair genes (hMSH2, hMLH1, PMS1, and PMS2; others may exist).  The first two of these genes account for about 80% to 90% of HNPCC cases, and the others account for only a small fraction of cases (65,66).  Linkage and mutational analyses have suggested that hMSH2 is the gene most commonly associated with HNPCC, accounting for 50% to 60% of HNPCC cases; hMLH1 accounts for an additional 30% of HNPCC cases (67).  These genes code for proteins that normally repair mutations introduced by the DNA polymerase during cell replication.  Mismatch repair dysfunction accelerates the accumulation of mutations in tumor suppressor genes and oncogenes (e.g., ras, APC, and p53) thereby speeding the development of malignancy (67,68).

Estimates of the cancer risks associated with hMSH2 and hMLH1 mutations are derived from families characterized by an early age of onset and/or multiple tumors and that meet stringent criteria for HNPCC that were established for research purposes (60,69).  Persons from families with HNPCC (including persons who test positive for a hMSH2 or hMLH1 gene mutation) may have a cumulative risk for colorectal cancer as high as 35% by age 50 and 80% by age 70 (60,67).  However, such estimates may be biased upward, because the people selected for study had a family history of cancer.  Estimates of the carrier frequency of hMSH2 and hMLH1 gene mutations among persons who do not have a family history of colorectal cancer are currently lacking.

Because of the heterogeneity of hMSH2 and hMLH1 mutations, laboratory testing for all possible mutations is difficult for persons not from families at high risk for colorectal cancer.  False-positive results may occur as a result of missense mutations that are not associated with an increased risk for colorectal cancer.  Frameshift or nonsense mutations are likely to predispose carriers to disease (22).  False-negative results will also occur in population screening since tests for all possible hMSH2 and hMLH1 mutations are not feasible on a large scale basis.

Polymorphisms associated with colorectal cancer risk

Genetic polymorphisms may account for why some people are more sensitive than others to environmental carcinogens or cancer promoters associated with colorectal cancer, such as a diet high in red meat and low in fiber (4).  In the Physicians' Health Study, for example, men with the homozygous mutation of the gene associated lower levels and with reduced activity 5-methyltetrahydrofolate, the primary circulating form of folate, had half the risk for colorectal cancer of men with the homozygous normal or heterozygous genotypes (70).  Among men who drank little of no alcohol, those with the homozygous mutation genotype had an eight fold lower risk than those with the homozygous normal genotype (OR 0.12, 95% CI 0.03 to 0.57), and those who were moderate drinkers had a twofold lower risk (OR 0.42, 95% CI 0.15 to 1.2)(70).  These results suggest that the gene mutation reduces colorectal cancer risk and that high alcohol consumption or low folate intake may negate some of the protective effect.

Polymorphisms in N-acetyltransferase have also been examined in relation to cancer of the colon and rectum (71,72).  This enzyme catalyzes the formation of mutagenic substances from foods such as cooked meat and fish.  People in whom acetylation is fast are at increased risk for colorectal cancer in some but not all studies (71-73).  Molecular epidemiology studies have also suggested that glutathione S-transferase-M1 genotypes may be associated with susceptibility to colorectal cancer (14).

A polymorphism of the APC gene (I1307K), recently reported to be present in 1 in 17 Ashkenazi Jewish persons, may be associated with an increased risk of colorectal cancer in that population (74).  This mutation does not directly alter the function of the APC gene but appears to make it unstable and prone to acquire mutations during cell division.  In the initial report, a risk relationship was suggested by differences in the frequency of the mutation among Ashkenazi Jewish controls and persons with colorectal cancer identified in referral centers (6% vs. 10%) and also between controls and a small subset of case subjects who had a positive family history of colorectal cancer (6% vs. 28%).  Additional studies are consistent with a weak association between this polymorphism and colorectal cancer as well as a potential association with breast or other cancers (75-77), but this association cannot be considered proven yet.  

LUNG CANCER GENETIC SUSCEPTIBILITY

Lung cancer is a major public health problem and an important part of the world-wide pandemic of smoking-related death and disability.  In 1998, an estimated 93,100 deaths from lung cancer will occur among men in the U.S. and an additional 67,000 deaths from lung cancer will occur among U.S. women (15).  The overall contribution of genetic factors to lung cancer is difficult to discern because of the strong associations with cigarette smoking and other environmental exposures (78).

The familial aggregation of lung cancer, which has been well-documented by results obtained from several studies, is not fully explained by the familial aggregation of smoking (79,80).  Although most segregation analyses of lung cancer have not provided strong evidence for cancer-predisposing mutations (80), a recent segregation analysis by Gauderman et al. (81) in 337 extended pedigrees provided evidence that a Mendelian gene is segregating in these families.  The estimated frequency of the high-risk allele was 2%; carriers were estimated to have a relative risk of lung cancer of 17.3 compared with noncarriers (81).  In a population study by Sellers et al. (82,83), the pattern of lung cancer occurrence in families with lung cancer was consistent with Mendelian codominant inheritance for early age-at-onset of a rare autosomal gene.

A variety of studies have examined polymorphically expressed genes in relation to lung cancer risk including those that code for cytochrome P450 enzymes, N-acetyl transferases, and glutathione S-transferases (8,78,84).  These have included transitional studies of convenience samples of lung cancer cases and controls, industry-based studies, and population-based studies.  Some transitional studies have examined both polymorphically expressed genes and genetic markers of exposure such as DNA adducts in relation to lung cancer risk (14,84).

Cytochrome P450 enzymes activate benzo(a)pyrene and may activate or metabolize other carcinogens found in cigarette smoke (8).  Lung cancer cases have been reported to have higher levels of AHH activity as compared with controls, although negative results have been obtained in some studies (84).  The results of more recent studies that have looked for associations with CYP1A1 genotypes have also been inconsistent, perhaps reflecting ethnic differences in the frequency of high-risk alleles.  Kawajiri et al. reported that Japanese patients with lung cancer were more likely to carry an isoleucine to valine mutation in the CYP1A1 locus (85).  The Msp1 allele has also been associated with lung cancer in Japanese individuals.  Studies in Caucasian populations have had negative results, however.

Poor metabolizers of debrisoquine, who are unable to metabolize environmental agents via the CYP2D6 pathway, have been reported to have a higher risk of lung cancer, but the strength of the association with this phenotype has varied across studies and some studies have failed to find an association (78).  Some studies have suggested an interactive effect with cigarette smoking (84).  A consistent pattern has not emerged from more recent lung cancer studies that have looked for associations with CYP2D6 genotypes.

McWilliams et al. (86) carried out a meta-analysis of ten case-control studies of GSTM1 deficiency and lung cancer risk and obtained a pooled odds ratio of 1.4 (95% 1.2-1.7).  In their meta-analysis, an association was found for each of the three major histologic sub-types of lung cancer (squamous cell carcinoma, adenocarcinoma, and small cell lung cancer)(86).  Studies to date have provided little evidence of an interactive effect between GSTM1 genotype and CYP1A1 in relation to lung cancer risk (14).  Exposure to cigarette smoke and dietary factors such as low vitamin C intake may influence the importance of GSTM1 genotypes to lung cancer risk.  Although the strength of the association between lung cancer and GSTM1 deficiency is likely to be modest at best, this gene could still account for a substantial number of lung cancer cases in the population because of the high frequency of GSTM1 deficiency in the general population.

Although N-acetyl transferase-2 detoxifies or activates aromatic amines found in tobacco smoke, studies of lung cancer have often failed to find an association with acetylator phenotype.  Oyama et al. (87) examined NAT2 genotype in 124 Japanese patients with lung cancer and in 376 controls; the slow acetylator genotype was present in 17 (14%) of the lung cancer patients as compared with 40 (11%) of the controls (odds ratio = 2.0, p = .05).

Somatic mutations such as p53 tumor suppressor genes and ras oncogenes are common in lung tumors and activation of myc oncogenes has been reported in small cell carcinomas of the lung (78).  Tumorigenesis in the lung is likely to involve a complex series of molecular genetic changes which lead to the transformation of normal cells to malignant cells.  These genetic changes in the lung that occur during tumorigenesis are somatic events that may be partly a consequence of lung cancer rather than a cause (78).

PROSTATE CANCER GENETIC SUSCEPTIBILITY

The public health burden of prostate cancer is substantial.  A total of 184,500 new cases of prostate cancer and 39,200 deaths from prostate cancer are anticipated in 1998, making it the most frequent cancer in males (15).  The incidence of prostate cancer is nearly two-fold higher in African-American men than in white men, and mortality is similarly elevated.  Environmental factors implicated as possible contributors to prostate cancer risk include high fat diet and vasectomy (88).  A genetic contribution to prostate cancer risk has also been documented, but knowledge of the molecular genetics of prostate cancer is still limited.

As with breast and colon cancer, familial clustering of prostate cancer has been observed (89-91).  A family history of a brother or father with prostate cancer increases the risk of prostate cancer two to three-fold, with the risk inversely related to the age of the affected relative (91-93).  A family history of prostate cancer also increases risk of breast cancer in female relates (94).  The association between prostate and breast cancer risks is explained in part by the increased risk of prostate cancer observed among male carriers of BRCA1 and BRCA2 mutations (25,95).

Families demonstrating autosomal dominant transmission of isolated susceptibility to prostate cancer have been observed, and have been estimated to account for 9% of prostate cancer cases (90).  Studies of such families have led to the identification of a possible genetic susceptibility locus on chromosome 1 (96,97), termed HPC1 (hereditary prostate cancer 1).  No significant increase in breast cancer or other cancers was found in these families (98), suggesting that the genetic susceptibility conferred by mutations at this locus differs from that conferred by mutations in the BRCA1 and BRCA2 genes. 

Linkage studies suggest the existence of other genetic loci potentially related to prostate cancer susceptibility on chromosomes 6 and 13 (99,100).  In addition, polymorphisms in the steroid 5 alpha-reductase type II and androgen receptor genes have been postulated to contribute to prostate cancer risk (11).  The genetic contribution to prostate cancer susceptibility thus appears to be complex.  No role has yet been defined for genetic screening related to prostate cancer susceptibility. 

MELANOMA GENETIC SUSCEPTIBILITY

The incidence of melanoma has increased markedly in the U.S. and in several other countries in recent decades (101).  In 1998, an estimated 41,600 new cases will be diagnosed in the U.S., and 4,600 men and 2,700 women will die from the disease (15).

Ultraviolet radiation and intense intermittent sun exposure resulting in sunburns in childhood are major environmental risk factors for melanoma that may interact with genetic influences.  Risk factors for melanoma that have a genetic basis include family history, number of atypical nevi, and pigmentary traits such as blue eyes, fair or red hair, and pale complexion (101).  Individuals who have one affected relative have about a two-fold increased risk of melanoma; those with two affected relatives have about a five-fold increased risk (101).  About 10% of cases occur in persons with a familial predisposition (102).

Genetic studies have suggested that the etiology of cutaneous malignant melanoma is heterogenous and complex. Dysplastic nevi syndrome (atypical mole syndrome phenotype) is a highly penetrant autosomal dominant disorder with underlying predisposition to melanoma (103).  Some melanoma-prone families have been reported to have an excess of pancreatic cancer (104). In 1989, Bale et al. reported evidence of linkage between the combined trait of dysplastic nevi and melanoma and genetic markers on the short arm of chromosome 1 (105).  More recent studies in other pedigrees have failed to confirm these findings, either for melanoma and dysplastic nevi combined or melanoma alone (101).

A linkage analysis in 11 cutaneous malignant melanoma pedigrees by Cannon-Albright et al. (106) provided strong evidence for a dominant, partially penetrant melanoma susceptibility locus on the short arm of chromosome 9.  The gene penetrance was estimated to be 53% by age 80 years.  Gene carriers had higher nevus counts and nevus densities (computed from mole size and number) than did those who were not carriers.  Among gene carriers, those with melanoma had more sunlight exposure than did those without melanoma suggesting a gene-environment interaction (107).

The CDKN2A tumor suppressor gene was localized on 9p21 in 1994 (108,109); that chromosomal region had previously been associated with malignant melanoma in linkage analyses and in loss of heterozygosity and cytogenetic studies.  The CDKN2A gene codes for a low-molecular-weight cell cycle control protein, p16^INK4A, that inhibits excessive cell proliferation by inhibiting the activity of the cyclin D1-cyclin-dependent kinase 4 or 6 complex (102).

Germline mutations of the CDKN2A gene have been found in roughly one fourth of melanoma-prone kindreds.  In a recent population-based study in Stockholm, Sweden, Platz et al. (110) found mutations of the CDKN2A gene in 7.8% of 64 families with two or more first-degree relatives with cutaneous malignant melanoma.  No mutations were detected in 36 families with melanoma in more distant relatives.  In 1996, a second melanoma susceptibility gene, CDK4, was identified on the long arm of chromosome 12 suggesting that there are diverse mechanisms of melanoma development in high risk families.  CDK4 mutations are rare and have only been identified in a few families to date (107).

SUMMARY AND CONCLUSIONS

Studies of genetic susceptibility to common cancers highlight the need for additional population-based molecular epidemiologic research, to better define the contribution of genetic factors to cancer, and to examine interactions with environmental factors that are amenable to preventive interventions.  The etiology of most cancers is likely to be due to an interaction between a variety of genetic factors and non-genetic risk factors such as smoking, diet, and exogenous and endogenous hormones. As Rebbeck (14) put it:

"The etiology of most commonly occurring cancers cannot be explained by allelic variability at a single locus.  Instead, the major burden of cancer in the general population probably results from the complex interactions of multiple genetic and environmental factors over time.  An understanding of the interplay of xenobiotic exposures, endogenous physiology, and genetic variability at multiple loci will facilitate knowledge about cancer etiology and the identification of individuals who are at increased risk for developing cancer."

The literature summarized in this chapter highlights gaps in our present understanding of the genetics of common malignancies.  Relatively few studies have looked for potential gene-environment interactions, explored associations between two or more genetic polymorphisms, or evaluated interactions between genetic polymorphisms and endogenous risk factors.  Studies designed to answer such questions are now underway in Los Angeles, Hawaii, North Carolina, and other localities (11,111).  They are likely to be particularly informative because they target geographically defined populations that are racially and ethnically diverse, evaluate incident cancers, and have adequate power sizes to explore ethnic and racial differences in cancer susceptibility.  Such research strategies represent an important new direction in cancer genetics.

References 

1.      Greenwald P.  Introduction: history of cancer prevention and control.  In: Greenwald P, Kramer BS, Weed DL, eds.  Cancer Prevention and Control.  New York: Marcel Dekker, Inc., 1995; 1-7.

2.      Khoury MJ.  From genes to public health: the applications of genetic technology in disease prevention.  Am J Public Health 1996;86:1717-22.

3.      Coughlin SS, Khoury MJ, Steinberg KK.  BRCA1 and BRCA2 gene mutations and risk of breast cancer: public health perspectives.  Am J Prev Med 1998;16:91-8.

4.      Coughlin SS, Miller DS.  Public health perspectives on testing for colorectal cancer susceptibility genes.  Am J Prev Med 1998;16:99-104.

5.      Coughlin SS.  The intersection of genetics, public health, and preventive medicine (editorial).  Am J Prev Med 1998;16:89-90.

6.      National Society of Genetic Counselors.  Predisposition genetic testing for late-onset disorders in adults.  A position paper of the National Society of Genetic Counselors.  JAMA 1997;278:1217-20.

7.      Swift M.  Ionizing radiation, breast cancer, and ataxia-telangiectasia (editorial).  JNCI 1994;86:1571-2.

8.      Smith G, Stanley LA, Sim E, et al.  Metabolic polymorphisms and cancer susceptibility.  Cancer Surveys.  Vol. 25: Genetics and Cancer: A Second Look.  Imperial Cancer Research Fund. 1995; 27-65.  [full citation needed]

9.      Windmill KF, McKinnon RA, Zhu X, et al.  The role of xenobiotic metabolizing enzymes in arylamine toxicity and carcinogenesis: functional and localization studies.  Mut Res 1997;376:153-60.

10.  Green ED, Waterston RH.  The Human Genome Project: prospects and implications for clinical medicine.  JAMA 1991;266:1966-75

11.  Feigelsen HS, Ross RK, Yu MC, et al.  Genetic susceptibility to cancer from exogenous and endogenous exposures.  J Cell Biochem 1996;25(Suppl):15-22.

12.  Strong LC, Amos CI.  Inherited susceptibility.  In: Schottenfeld D, Fraumeni JF, Jr, eds.  Cancer Epidemiology and Prevention, 2nd edn.  New York: Oxford University Press, 1996.

13.  Ambrosone CB, Freudenheim JL, Graham S, et al.  Cigarette smoking, N-acetytransferase 2 genetic polymorphisms and breast cancer risk.  JAMA 1996;276:1494-1501.

14.  Rebbeck TR.  Molecular epidemiology of the human glutathione S-transferase genotypes GSTM1 and GSTT1 in cancer susceptibility.  Cancer Epidemiol Biomarkers Prev 1997;6:733-43.

15.  Cancer facts and figures 1998, American Cancer Society, Atlanta, GA.

16.  Newman B, Millikan RC, King M-C.  Genetic epidemiology of breast and ovarian cancers.  Epidemiol Rev 1997;19:69-79.

17.  Mettlin C, Croghan I, Natarajan N, et al.  The association of age and familial risk in a case-control study of breast cancer.  Am J Epidemiol 1990;131:973-83.

18.  Andersen TI.  Genetic heterogeneity in breast cancer susceptibility.  Acta Oncol 1996;35:407-10.

19.  Xu CF, Solomon E.  Mutations of the BRCA1 gene in human cancer.  Sem in Cancer Biol 1996;7:33-40.

20.  Miki Y, Swensen J, Shattuck-Eidens D, et al.  A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1.  Science 1994;266:66-71.

21.  Wooster R, Neuhausen SL, Mangion J, et al.  Localization of a breast cancer susceptibility gene, BRCA2, to chromosome 13q12-13.  Science 1994;265:2088-90.

22.  Collins FS.  BRCA1--lots of mutations, lots of dilemmas.  N Engl J Med 1996;334:186-8.

23.  Shattuck-Eidens D, McClure M, Simard J, et al.  A collaborative survey of 80 mutations in the BRCA1 breast and ovarian cancer susceptibility gene.  JAMA 1995;273:535-41.

24.  Ford D, Easton DF, Stratton M, et al.  Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families.  Am J Hum Genet 1998;62:676-89.

25.  Easton DF, Steele L, Fields P, et al.  Cancer risks in two large breast cancer families linked to BRCA2 on chromosome 13q12-13.  Am J Hum Genet 1997;61:120-8.

26.  Struewing JP, Abeliovich D, Peretz T, et al.  The carrier frequency of the BRCA1 185delAG mutation is approximately 1 percent in Ashkenazi Jewish individuals (letter).  Nature Genetics 1995;11:198-200.

27.  Ford D, Easton DF, Peto J.  Estimates of the gene frequency of BRCA1 and its contribution to breast and ovarian cancer incidence.  Am J Hum Genet 1995;57:1457-62.

28.  Newman B, Mu H, Butler LM, et al.  Frequency of breast cancer attributable to BRCA1 in a population-based series of American women.  JAMA 1998;279:915-21.

29.  Malone KE, Daling JR, Thompson JD, et al.  BRCA1 mutations and breast cancer in the general population.  Analyses in women before age 35 years and in women before age 45 years with first-degree family history.  JAMA 1998;279:922-9.

30.  PetoJ, Collins N, Barfoot R,Seal S, Warren W, Rahman N, Easton DF et al.  Prevalence of BRCA1 and BRCA2 gene mutations in patients with early-onset breast cancer.  J Natl Cancer Inst 1999; 91:943-9.

31.  Roa BB, Boyd AA, Volcik K, et al.  Ashkenazi Jewish population frequencies for common mutations in BRCA1 and BRCA2.  Nature Genetics 1996;14:185-7.

32.  Oddoux C, Struewing JP, Clayton CM, et al.  The carrier frequency of BRCA2 6174delT mutation among Ashkenazi Jewish individuals is approximately 1%.  Nature Genetics 1996;14:188-90.

33.  Thorlacius Sstruewing JP, Hartge P et al/ Population-based study of risk of breast cancer in carriers of BRCA2 mutation.  Lancet 1998; 352: 1337-9.

34.  Struewing JP, Hartge P, Wacholder S, et al.  The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews.  N Engl J Med 1997;336:1401-8.

35.  Fodor FH, Weston A, Bleiweiss IJ, McCurdy LD, Walsh M, Tartter PI et al.  Frequency and carrier risk associated with common BRCA1 and BRCA2 mutations in Ashkenazi Jewish breast cancer patients.  Am J Hum Genet 1998; 63: 45-51.

36.  Levy-Lahad E, Catane R, Eisenberg S, et al.  Founder BRCA1 and BRCA2 mutations in Ashkenazi Jews in Israel: frequency and differential penetrance in ovarian cancer and in breast-ovarian cancer families.  Am J Hum Genet 1997;60:1059-67.

37.  Weber BL, Giusti RM, Liu ET.  Developing strategies for intervention and prevention in hereditary breast cancer.  Monogr Natl Cancer Inst 1995;17:99-102

38.  Burke W, Daly M, Garber J, et al.  Recommendations for follow-up care of individuals with an inherited predisposition to cancer.  II. BRCA1 and BRCA2.  JAMA 1997;277:997-1003.

39.  Narod S, Risch H, Mosleshi R et al. For the Hereditary Ovarian Cancer Study Group. Oral contraceptives and the risk of hereditary ovarian cancer.  New Engl J Med 1998;339:424-8.

40.  Swift M, Reitnauer PJ, Morrell D, et al.  Breast and other cancers in families with ataxia-telangiectasia.  N Engl J Med 1987;316:1289-94.

41.  Lavin MF, Shiloh Y. The genetic defect in ataxia-telangiectasia (review).  Ann Rev Immunol 1997;15:177-202.

42.  Morrell D, Cromartie E, Swift M.  Mortality and cancer incidence in 263 patients with ataxia-telangiectasia.  J Natl Cancer Inst 1986;77:89-92.

43.  Easton DF.  Cancer risks in A-T heterozygotes.  Int J Radiat Biol 1994;66(6 suppl):S177-82.

44.  Swift M.  The incidence and gene frequency of ataxia-telangiectasia in the United States.  Am J Hum Genet 1986;39:573-83.

45.  Rasio D, Negrini M, Croce CM.  Genomic organization of the ATM locus involved in ataxia-telangiectasia.  Cancer Res 1995;55:6053-7.

46.  Swift M, Morrell D, Massey RB, et al.  Incidence of cancer in 161 families affected by ataxia-telangiectasia.  N Engl J Med 1991;325:1831-6.

47.  Vorechovsky I, Luo L, Lindblom A, et al.  ATM mutations in cancer families.  Cancer Res 1996;56:4130-3.

48.  Athma P, Rappaport R, Swift M, et al.  Molecular genotyping shows that ataxia-telangiectasia heterozygotes are predisposed to breast cancer.  Cancer Genet & Cytogenet 1996;92:130-4.

49.  Fitzgerald MG et al.  Heterozygous ATM mutations do not contribute to early onset of breast cancer.  Nat Genet 1997;15:307-10.

50.  Swift M.  Ataxia telangiectasia and risk of breast cancer (letter).  Lancet 1997;350:739-40.

51.  Ambrosone CB, Freudenheim JL, Marshall JR, et al.  The association of polymorphic N-acetyltransferase (NAT2) with breast cancer risk.  Ann NY Acad Sci 1995;768:250-2.

52.  Ladero JM, Fernandez MJ, Palmeiro R, et al.  Hepatic acetylator polymorphism in breast cancer patients. Oncology 1987;44:341-4.

53.  Agundez JA, Ladero JM, Olivera M, et al.  Genetic analysis of the arylamine N-acetyltransferase polymorphism in breast cancer patients.  Oncology 1995;52:7-11.

54.  Feigelson HS, Coetzee GA, Kolonel LN, et al.  A polymorphism in the CYP17 gene increases the risk of breast cancer.  Cancer Res 1997;57:1063-5.

55.  Kelsey KT, Wiencke JK.  Growing pains for the environmental genetics of breast cancer: observations on a study of the glutathione S-transferases.  J Natl Cancer Inst 1998;90:484-5.

56.  Burt RW, Petersen GM.  Familial colorectal cancer: diagnosis and management.  In:  Young GP, Rozen P, Levin B, eds.  Prevention and Early Detection of Colorectal Cancer.  London: Saunders, 1996; 171-94.

57.  Petersen GM.  Genetic epidemiology of colorectal cancer.  Eur J Cancer 1995;31A:1047-50.

58.  Lynch HT, Smyrk T, Lynch JF.  Overview of natural history, pathology, molecular genetics and management of HNPCC (Lynch Syndrome).  Int J Cancer 1996;69:38-43.

59.  Han HJ, Yuan Y, Ku JL, et al.  Germline mutations of hMLH1 and hMSH2 genes in Korean hereditary nonpolyposis colorectal cancer (letter).  J Natl Cancer Inst 1996;88:1317-9.

60.  Vasen HF, Wijnen JT, Menko FH, et al.  Cancer risk in families with hereditary nonpolyposis colorectal cancer diagnosed by mutation analysis.  Gastroenterology 1996;110:1020-7.

61.  Marra G, Boland CR.  Hereditary nonpolyposis colorectal cancer: the syndrome, the genes, and historical perspectives.  J Natl Cancer Inst 1995;87:1114-25.

62.  Fishel R, Lescoe MK, Rao MR, et al.  The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer.  Cell 1993;75:1027-38.

63.  Leach FS, Nicolaides N, Papadopoulos N, et al.  Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer.  Cell 1993;75:1215-25.

64.  Bronner CE, Baker SM, Morrison PT, et al.  Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary nonpolyposis colon cancer.  Nature 1994;368:258-61.

65.  Aaltonen LA, Peltomaki P.  Genes involved in hereditary nonpolyposis colorectal cancer (review).  Anticancer Res 1994;14(4B):1657-60.

66.  Nystrom-Lahti M, Moisio AL, Hofstra RM, et al.  DNA mismatch repair gene mutations in 55 kindreds with verified or putative hereditary non-polyposis colorectal cancer.  Hum Mol Genet 1996;5:763-9.

67.  Kolodner RD, Hall NR, Lipford J, et al.  Structure of the human MLH1 locus and analysis of a large hereditary nonpolyposis colorectal carcinoma kindred for MLH1 mutations.  Cancer Res 1995;55:242-8.

68.  Cho KR, Vogelstein B.  Genetic alterations in the adenoma-carcinoma sequence.  Cancer 1992;70(6 Suppl):1727-31.

69.  Voskuil DW, Vasen HFA, Kampman E, et al.  Colorectal cancer risk in HNPCC families: development during lifetime and in successive generations.  Int J Cancer 1997;72:205-9.

70.  Ma J, Stampfer MJ, Giovannucci E, et al.  Methylenetetrahydrofolate reductase polymorphism, dietary interactions, and risk of colorectal cancer.  Cancer Res 1997;57:1098-102.

71.  Bell DA, Stephens EA, Castranio T, et al.  Polyadenylation polymorphism in the acetyltransferase 1 gene (NAT1) increases risk of colorectal cancer.  Cancer Res 1995;55:3537-42.

72.  Shibuta K, Nakashima T, Abe M, et al.  Molecular genotyping for N-acetylation polymorphism in Japanese patients with colorectal cancer.  Cancer 1994;74:3108-12.

73.  Roberts-Thomson IC, Ryan P, Khoo KK, et al.  Diet, acetylator phenotype, and risk of colorectal neoplasia.  Lancet 1996;347:1372-4.

74.  Laken SJ, Petersen G, Bruber SB, et al.  Familial colorectal cancer in Ashkenazim due to a hypermutable tract in APC.  Nature Genetics 1997;17:79-83.

75.  Frayling IM, Beck NE, Ilyas M et al.  The APC variants I1307K and E1317Q are associated with colorectal tumors, but not always with a family history.  Proc Natl Acad Sci 1998;95:10722-7.

76.  Woodage T, King SM, Wacholder S et al.  The APC I1307K allele and cancer risk in a community-based study of Ashkenazi Jews.  Nat Genet 1998;20:62-5.

77.  Abrahamson J, Moslei R, Vesprini D et al.  No association of the I1307K APC allele with ovarian cancer risk in Ashkenazi Jews.  Cancer Res 1998;58:2919-22.

78.  Blot WJ, Fraumeni JF, Jr.  Cancer of the lung and pleura.  In: Schottenfeld D, Fraumeni JF, Jr, eds.  Cancer Epidemiology and Prevention, 2nd edn.  New York: Oxford University Press, 1996; 652-3.

79.  Osann KE.  Lung cancer in women: the importance of smoking, family history of cancer, and medical history of respiratory disease.  Cancer Res 1991;51:4893-7.

80.  Yang P, Schwartz AG, McAllister AE, et al.  Genetic analysis of families with nonsmoking lung cancer probands.  Genetic Epidemiol 1997;14:181-97.

81.  Gauderman WJ, Morrison JL, Carpenter CL, et al.  Analysis of gene-smoking interaction in lung cancer.  Genetic Epidemiol 1997;14:199-214.

82.  Sellers TA, Bailey-Wilson JE, Elston RC, et al.  Evidence for Mendelian inheritance in the pathogenesis of lung cancer.  J Natl Cancer Inst 1990;82:1272-9.

83.  Sellers TA, Bailey-Wilson JE, Potter JD, et al.  Effect of cohort differences in smoking prevalence on models of lung cancer susceptibility.  Genet Epidemiol 1992;9:261-72.

84.  Economou P, Samet JM, Lechner JF.  Familial and genetic factors in the pathogenesis of lung cancer.  In: Samet JM, ed.  Epidemiology of Lung Cancer.  New York: Marcel Dekker, Inc., 1994; 353-96.

85.  Kawajiri K, Nakachi K, Imai K, et al.  Identification of genetically high risk individuals to lung cancer by DNA polymorphisms of the cytochrome P450IA1 gene.  FEBS 1990;263:131-3.

86.  McWilliams JE, Sanderson BJS, Harris EL, et al.  Glutathione S-transferase M1 (GSTM1) deficiency and lung cancer risk.  Cancer Epidemiol Biomarkers Prev 1995;4:589-94.

87.  Oyama T, Kawamoto T, Mizoue T, et al.  N-acetylation polymorphism in patients with lung cancer and its association with p53 gene mutation.  Anticancer Res 1997;17:577-82.

88.  Pienta K, Esper PS.  Risk factors for prostate cancer.  Ann Intern Med 1993;118:793-803.

89.  Steinberg GD, Carter BS, Beaty TH, et al.  Family history and the risk of prostate cancer.  Prostate 1990;17:337-47.

90.  Carter BS, Beaty TH, Steinberg GD, et al.  Mendelian inheritance of familial prostate cancer.  Proc Natl Acad Sci 1992;89:3367-71.

91.  Ghadrian P, Howe GR, Hislop TG, et al.  Family history of prostate cancer: a multi-center case-control study in Canada.  Int J Cancer 1997;70:679-81.

92.  Gronberg H, Damber L, Damber JE.  Familial prostate cancer in Sweden: a nationwide register cohort study.  Cancer 1996;77:138-43.

93.  Cannon L, Bishop DT, Skolnick M, et al.  Genetic epidemiology of prostate cancer in the Utah Mormon geneology.  Cancer Surv 1982;1:47-69.

94.  Sellers TA, Potter JD, Rich SS, et al.  Familial clustering of breast and prostate cancers and risk of postmenopausal breast cancer.  J Natl Cancer Inst 1994;86:1860-5.

95.  Ford D, Easton DF, Bishop DT, et al.  Risk of cancer in BRCA1-mutation carriers.  Lancet 1994;343:692-5.

96.  Smith JR, Freije D, Carpten JD, et al.  Major susceptibility locus for prostate cancer on chromosome 1 suggested by a genome-wide search.  Science 1996;274:1371-4.

97.  Cooney KA, McCarthy JD, Lange E, et al.  Prostate cancer susceptibility locus on chromosome 1q: a confirmatory study.  J Natl Cancer Inst 1997;89:955-9.

98.  Gronberg H, Isaacs SD, Smith JR, et al.  Characteristics of prostate cancer in families potentially linked to the hereditary prostate cancer 1 (HPC1) locus.  JAMA 1997;278:1251-55.

99.  Cooney KA, Wetzel JC, Merajver SD, et al.  Distinct regions of alleleic loss on 13q in prostate cancer.  Cancer Res 1996;56:1142-5.

100.         Cooney KA, Wetzel JC, Consolino CM, et al.  Identification and characterization of proximal 6q deletions in prostate cancer.  Cancer Res 1996;56:4150-3.

101.          Armstrong BK, English DR.  Cutaneous malignant melanoma.  In: Schottenfeld D, Fraumeni JF, Jr, eds.  Cancer Epidemiology and Prevention, 2nd edn.  New York: Oxford University Press, 1996; 1282-1312.

102.          Goldstein AM, Tucker MA.  Screening for CDKN2A mutations in hereditary melanoma.  J Natl Cancer Inst 1997;89:676-8.

103.          Albino AP.  Genes involved in melanoma susceptibility and progression.  Curr Opinion Oncol 1995;7:162-9.

104.          Fraser MC, Goldstein AM, Tucker MA.  The genetics of melanoma.  Sem Oncol Nurs 1997;13:108-14.

105.          Bale SJ, Dracopoli NC, Tucker MA, et al.  Mapping the gene for hereditary cutaneous malignant melanoma-dysplastic nevus to chromosome 1p.  N Engl J Med 1989;320:1367-72.

106.          Cannon-Albright LA, Godgar DE, Meyer LJ, et al.  Assignment of a locus for familial melanoma, MLM, to chromosome 9p13-p22.  Science 1992;258:1148-52.

107.          Greene MH.  Genetics of cutaneous melanoma and nevi.  Mayo Clin Proc 1997;72:467-74.

108.          Kamb A, Shattuck-Eidens D, Eeles R, et al.  Analysis of CDKN2 (MTS1) as a candidate for the chromosome 9p melanoma susceptibility locus (MLM).  Nature Genet 1994;8:22-6.

109.          Nobari T, Miura K, Wu DJ, et al.  Deletions of the cyclin-dependent kinase-4 inhibitor gene in multiple human cancers.  Nature 1994;368:753-6.

110.          Platz A, Hansson J, Mansson‑Brahme E, et al.  Screening of germline mutations in the CDKN2A and CDKN2B genes in Swedish families with hereditary cutaneous melanoma. J Natl Cancer Inst 1997;89:676-8.

111.      Newman B, Moorman PG, Millikan R, et al.  The Carolina Breast Cancer Study: integrating population-based epidemiology and molecular biology.  Br Cancer Res Treatment 1995;35:51-60.