Chapter 186 – GENETICS OF CANCER
Raju S.K. Chaganti
Cancer is a multistep process, the earliest step of which in some cases is an inherited mutation, followed by one or more somatic mutations in the target cell. Each cancer ultimately traces back to a single transformed normal cell and hence is clonal in nature. This chapter examines the role of genetics in the natural history and clinical implications of cancer.
The aggregation of cancer in a family can be due to genetic or nongenetic causes, the former through mendelian (single-gene mutation; Chapter 39) or nonmendelian (polygenic or multifactorial; Chapter 37) inheritance of genes that predispose to cancer and the latter related to common exposure to carcinogenic agents (Chapter 187), lifestyle, or simply coincidence. The modern understanding of familial aggregation of cancer has required increasingly sophisticated epidemiologic and statistical methods in combination with genetic concepts and methods.
Although mendelian inheritance accounts for a small minority of all cancers, mutations that predispose to cancer have provided some of the most penetrating insights into the understanding of the genetic basis of normal as well as abnormal development; these mutations manifest the classical recessive or dominant modes of inheritance (Table 186-1). Nonmendelian inheritance, which also plays a major role in the overall incidence of cancer, has been more difficult to characterize. In addition, the interaction of mutated genes with the environment adds another level of complexity in deciphering the role of genetics of cancer in individuals as well as in families.
Syndrome | Mode of Inheritance | Gene(s) |
HEREDITARY BREAST CANCER SYDROMES | ||
Hereditary breast and ovarian cancer syndrome | Dominant | BRCA1 |
BRCA2 | ||
Li-Fraumeni syndrome | Dominant | TP53 |
Cowden's syndrome | Dominant | PTEN |
Bannayan-Riley-Ruvalcaba syndrome | Dominant | PTEN |
HEREDITARY GASTROINTESTINAL MALIGNANCIES | ||
Hereditary nonpolyposis colon cancer | Dominant | MLH1 |
MLH2 | ||
MSH6 | ||
Familial polyposis | Dominant | APC |
Hereditary gastric cancer | Dominant | CDH1 |
Juvenile polyposis | SMAD4/DPC4 | |
BMPR1A | ||
Peutz-Jeghers syndrome | Dominant | STK11 |
Hereditary melanoma–pancreatic cancer syndrome | Dominant | CDKN2A |
Hereditary pancreatitis | Dominant | PRSS1 |
Turcot's syndrome | Dominant | APC |
MLH1 | ||
PMS2 | ||
Familial gastrointestinal stromal tumor | Dominant | KIT |
GENODERMATOSES WITH CANCER PREDISPOSITION | ||
Melanoma syndromes | Dominant | CDKN2A |
CDK4 | ||
CMM | ||
Basal cell cancer, Gorlin's syndrome | Dominant | PTCH |
Cowden's syndrome | Dominant | PTEN |
Neurofibromatosis 1 | Dominant | NF1 |
Neurofibromatosis 2 | Dominant | NF2 |
Tuberous sclerosis | Dominant | TSC1 |
TSC2 | ||
Carney's complex | Dominant | PRKAR1A |
Muir-Torre syndrome | Dominant | MLH1 |
MSH2 | ||
Xeroderma pigmentosum | Recessive | XPA,B,C,D,E,F,G |
POLH | ||
Rothmund-Thomson syndrome | Recessive | RECOL4 |
LEUKEMIA/LYMPHOMA PREDISPOSITION SYNDROMES | ||
Bloom's syndrome | Recessive | BLM |
Fanconi's anemia | Recessive | FANCA,B,C |
FANCA,D2 | ||
FANCE,F,G | ||
FANCL | ||
Ataxia-telangiectasia | Recessive | ATM |
Shwachman-Diamond syndrome | Recessive | SBDS |
Nijmegen breakage syndrome | Recessive | NBS1 |
Canale-Smith syndrome | Dominant | FAS |
FASL | ||
Wiskott-Aldrich syndrome | X-linked recessive | WAS |
Common variable immune deficiency | Recessive | |
Severe combined immune deficiency | X-linked recessive | IL2RG |
Recessive | ADA | |
JAK3 | ||
RAG1 | ||
RAG2 | ||
IL7R | ||
CD45 | ||
Artemis | ||
X-linked lymphoproliferative syndrome | X-linked recessive | SH2D1A |
GENITOURINARY CANCER PREDISPOSITION SYNDROMES | ||
Hereditary prostate cancer | Dominant | HPC1 |
HPCX | ||
HPC2/ELAC2 | ||
PCAP | ||
PCBC | ||
PRCA | ||
Simpson-Golabi-Behmel syndrome | X-linked recessive | GPC3 |
von Hippel–Lindau syndrome | Dominant | VHL |
Beckwith-Wiedemann syndrome | Dominant | CDKN1C |
NSD1 | ||
Wilms' tumor syndrome | Dominant | WT1 |
Wilms' tumor, aniridia, genitourinary abnormalities, mental retardation (WAGR) syndrome | Dominant | WT1 |
Birt-Hogg-Dub? syndrome | Dominant | FLCL |
Papillary renal cancer syndrome | Dominant | MET,PRCC |
Constitutional t(3;8) translocation | Dominant | TRCB |
Hereditary bladder cancer | Sporadic | |
Hereditary testicular cancer | Possibly X-linked | |
Rhabdoid predisposition syndrome | Dominant | SNF5INI1 |
CENTRAL NERVOUS SYSTEM/VASCULAR CANCER PREDISPOSITION SYNDROMES | ||
Hereditary paraganglioma | Dominant | SDHD |
SDHC | ||
SDHB | ||
Retinoblastoma | Dominant | RB1 |
Rhabdoid predisposition syndrome | Dominant | SNF5/INI1 |
SARCOMA/BONE CANCER PREDISPOSITION SYNDROMES | ||
Multiple exostoses | Dominant | EXT1 |
EXT2 | ||
Leiomyoma/renal cancer syndrome | Dominant | FH |
Carney's complex | Dominant | PRKAR1A |
Werner's syndrome | Recessive | WRN |
ENDOCRINE CANCER PREDISPOSITION SYNDROMES | ||
Multiple endocrine neoplasia 1 | Dominant | MEN1 |
Multiple endocrine neoplasia 2 | Dominant | RET |
Familial papillary thyroid cancer | Dominant | Multiple loci |
Modified from Garber JE, Offit K: Hereditary cancer predisposition syndromes. J Clin Oncol 2005;23:276–292.
Some of the best known dominantly inherited syndromes are retinoblastoma, the hereditary breast cancer syndromes (Chapter 208), hereditary gastrointestinal malignancies (Chapter 203), and the endocrine tumor syndromes (Chapter 250). In many of the dominant syndromes, a defect in a tumor suppressor gene is generally inherited as a germline mutation from one parent; a subsequent acquired (somatic) mutation in the second copy of the gene from the other parent leads to complete loss of function and the development of cancer. In others, a gain-of-function mutation leads to formation of an oncogene, which promotes cancer even when only one copy of the gene is affected. Likewise, the so-called chromosome breakage (Bloom's syndrome [Chapter 38], ataxia telangiectasia [Chapter 271], Fanconi's anemia [Chapter 171]) and DNA repair disorders (xeroderma pigmentosum) illustrate recessively inherited cancer syndromes.
Hereditary retinoblastoma, which is a paradigm of a rare cancer syndrome, accounts for only 1% of pediatric cancers, although it is the most common eye cancer in children. It may be unilateral or bilateral, with bilateral tumors arising at a median age of 8 months whereas unilateral tumors arise at a median age of 2 years. Bilateral tumors are generally hereditary, although only 10% of patients present with a positive family history. Epidemiologic features of this disease suggested the “two-hit” hypothesis, which was validated upon cloning the gene and showing that the hereditary form represents a germline (present in one copy of the gene in all somatic cells of the gene carrier) mutation in the retinoblastoma gene as well as a second mutation, deletion, or other genetic event in the other copy of the retinoblastoma gene occurring in the target retinal cells and leading to suppression of the gene's function. In contrast, nonhereditary tumors arise when both mutations occur in the target cells, thus also fulfilling the criteria for the two-hit model. Functional characterization of the retinoblastoma gene showed that it regulates a key checkpoint in mitotic cell cycle progression from the G1 to S phase. Mutational inactivation of the gene eliminates the checkpoint, thereby leading to abnormal cell cycle progression and uncontrolled mitotic proliferation. The recognition that loss of function of a key regulatory gene leads to the development of a tumor by abrogating the normal control of a key cell function led to the concept that such genes represent tumor suppressor genes, of which the retinoblastoma gene was the first to be discovered.
Another cell cycle checkpoint regulator, which functions as a potent tumor suppressor when functionally inactivated, is the TP53 (p53) gene. The p53 protein is a transcription factor that regulates a number of growth regulatory genes, mediates apoptosis, and modulates cellular response to DNA damaging agents. In recognition of its multiple cellular regulatory functions, it is sometimes called the cell's gatekeeper. Germline mutations of the TP53 gene underlie the Li-Fraumeni syndrome (Chapter 187), which is characterized by bone or soft tissue sarcomas, as well as multiple other cancer types. Several other dominantly inherited cancer syndromes, such as hereditary breast cancer syndromes and hereditary gastrointestinal malignancies, are caused by dysfunction of tumor suppressor genes.
CANCER AS A SOMATIC CELL DISEASE
Chromosomal Changes
Cytogenetic studies of various leukemias established four cardinal attributes of genetic change in cancer: (1) specific or nonrandom chromosomal changes may characterize individual cancer types; (2) tumor genomes are genetically unstable and subject to continuing change, a feature now recognized as genomic instability; (3) all cells in a given tumor trace back to a single progenitor cell and therefore are clonal; and (4) tumor progression is often associated with additional specific or nonrandom chromosomal changes, presumably “selected” from the genomic instability, in subpopulations of tumor cells that lead clonal diversity and evolution. Chromosomal changes are of many types, the most common being gain of an entire chromosome (aneuploidy) or a region of it (duplication), loss of an entire chromosome (monosomy) or a region of it (deletion), translocation or inversion (rearrangement), and amplification (Fig. 186-1).
FIGURE 186-1 Common cytogenetic changes in cancer. The chromosome (at metaphase) is traditionally distinguished by its short and long arms separated by a centromere. Stylized bands (dark and light stripes along the length of the chromosome) produced by special treatments are also shown. The abnormality (right) and the corresponding normal image of the chromosome are illustrated in each panel. A, Gain of a chromosome leading to aneuploidy. B, Deletion of a chromosomal segment from one of the two homologues. C, Translocation showing exchange of segments between nonhomologous chromosomes. D, Amplification, with an increase in a region of a chromosome by replicating many times in place.
Thousands of human tumors belonging to all five cancer lineages (hematopoietic, epithelial, mesenchymal, neuronal, germ cell) have been investigated for their chromosomal changes using molecular cytogenetic methods. These studies generally confirmed the principles of chromosomal change deduced from studies of leukemia. In addition, detailed cytogenetic characterization showed that all tumors display all the types of aberrations illustrated in Figure 186-1, and hematopoietic and mesenchymal tumors tend to exhibit specific translocations that are associated with specific histologies.
Oncogenes
Early in the 20th century, Peyton Rous showed that infection of chickens with a transforming retroviral sequence, termed the viral oncogene (v-onc gene), transduced and activated by mutation its normal cellular counterpart (c-onc gene) and caused the development of sarcomas. This particular c-onc gene was thus named c-src (sarcoma). Over the years, more than 30 isolated retroviruses have been shown to cause acute transformation in eukaryotic, although not human, cells. In humans, c-onc genes (homologous to ras), which are members of the so-called ras family of c-onc genes, cause bladder, lung, colon, and other cancers. Other transforming genes that are neither members of the ras family nor homologous with transforming sequences in retroviruses include multiple receptor tyrosine kinases, growth factors, and transcription factors. All these genes are now generically referred to as oncogenes. Oncogenes as a class tend to be highly conserved and, in their normal state, regulate development, growth, and other important cell functions.
A reciprocal translocation between two nonhomologous chromosomes leads to exchange of chromosomal segments following breakage and healing of the broken ends (see Fig. 186-1). In cancer-causing translocations, genes situated at the breakpoints are subject to either juxtaposition to other key genes with subsequent deregulation or the formation of fusion genes that generate abnormal proteins. For example, the Ph chromosome translocation, which involves the ABL oncogene on chromosome 9 (at 9q34) and the BCR gene (at 22q11), leads to the generation of one of two fusion genes that code for an 210-kD protein (p210) in chronic myelogenous leukemia (Chapter 195) or a 190-kD (p190) protein in acute lymphoblastic leukemia (Chapter 194) (Fig. 186-2). Another example is Burkitt's lymphoma, in which breakpoints in a translocation affecting chromosomes 8 and 14, with breaks at bands 8q24 and 14q32, cause MYC (a homologue of the avian myelomatosis retroviral oncogene) situated at 8q24 to rearrange with IGH (immunoglobulin heavy chain gene) at 14q32, thereby leading to the generation of a hybrid IGH: MYC transcriptional unit. The protein-coding region MYC remains intact but is brought under the transcriptional control of the IGH gene, thereby leading to deregulated expression of MYC, which then sets the stage for neoplastic transformation. This general pattern of translocations is common in lymphomas, where a variety of oncogenes are deregulated by juxtaposition with IG or TCR (T-cell receptor) antigen loci, the latter constitutively active in immune cells. Specific translocations also characterize sarcomas (Chapters 212 and 213), all of which generate fusion proteins involving oncogenes.
FIGURE 186-2 Analysis of a chromosomal translocation. A translocation between chromosomes 9 and 22 with breaks through the ABL (chromosome 9) and BCR (chromosome 22) genes results in abnormal expression of the ABL gene and development of chronic myelogenous leukemia (CML). A, The molecular structure of ABL and BCR genes, the breakpoints in the genes, and the P190 and P210 proteins encoded by the fusion genes. B, Cytogenetic depiction of the translocation. Top, The normal and translocated chromosome illustrated in a cartoon form. Bottom, Fluorescence in situ images of the translocation. Left, A normal metaphase and interphase showing two red (ABL) and two green (BCR) signals. Right, A tumor metaphase and interphase showing one red (ABL), one green (BCR), and two fusion red-green signals.
Finally, amplification of an oncogene to form multiple copies of itself can lead to overexpression of the gene product. This frequent, tumor-specific mechanism is associated with aggressive behavior or poor outcome. Thus, the MYC family of oncogenes (MYC, MYCL, MYCN) is amplified in small cell lung cancer, MYCN in neuroblastoma, ERBB1 (estrogen growth factor receptor) in glioblastoma, and ERBB2 (HER2), MYC, and CCND1 (cyclin D1) in breast cancer (Chapter 208).
Epigenetics
The traditional view of gene function is based on the concept that both parental (paternal and maternal) alleles of a gene are equivalent. Genomic imprinting is an epigenetic modification that occurs in only one parental allele of the gene in germline or somatic cells and that leads to differential expression of the two parental alleles in an individual cell or all of its progeny. Loss of imprinting may activate a normally inactive growth-enhancing gene or inactivate a tumor suppressor gene. Genomic imprinting is normally maintained by methylation of cytosine within DNA sequences that are rich in CpG dinucleotides (CpG islands). Loss of imprinting, which is detectable as abnormal methylation of CpG islands, is found in many types of tumors and affects many genes (e.g., APC, CDKN2A, MLH1, RASFA1, MGMT).
Genetic Counseling
Risk assessment based on testing for the mutation status of the relevant gene in a family is an integral component of management of affected and at-risk individuals in a family (Chapter 38). In the communication of genetic risk information to a patient or family, the provider must be equipped to deal with the medical, psychological, and social consequences in all their complexity. Furthermore, as ability to identify individuals at risk increases, especially at early ages, the responsibility to institute effective surveillance and prevention strategies also increases.
Genetic Testing for Diagnosis and Prognosis
Cytogenetic chromosomal analysis or fluorescence in situ hybridization is now routine for all newly diagnosed leukemias (Chapters 194 and 195) because both good and bad prognostic cytogenetic markers have been identified. Cytogenetic analysis is equally important in the differential diagnosis of lymphomas, sarcomas, and other tumors and in the prognostic evaluation of neuroblastoma and metastatic breast cancer. In addition, cytogenetic analysis after chemotherapy or stem cell transplantation can assess residual tumor burden and engraftment. Prognostic markers include the EGFR gene in lung cancer (Chapter 201) and the IGHV mutation in chronic lymphocytic leukemia (Chapter 195). Molecular methods such as the polymerase chain reaction also provide increasing precision to detect chromosomal translocations and other abnormalities in treated patients to detect minimal residual disease.
GENETICS IN CANCER THERAPY
The greatest hope for therapy is that when a causal mutation or other genetic aberration for any cancer is discovered, it should be possible either to replace a defective gene with a normal one by genetic or cellular engineering or to develop drugs that block the action of an aberrant gene. In the latter area, the growth factor gene ERBB2 (HER2) is amplified and the HER2 protein encoded by it is overexpressed in a proportion of metastatic breast cancers with poor prognosis (Chapter 208). The drug trastuzumab, an antibody against HER2, specifically blocks the overexpressed protein in HER2-positive breast cancers and is a highly effective therapy. The normal ABL gene encodes a protein tyrosine kinase, the Abl tyrosine kinase, whereas the BCR-ABL fusion gene in chronic myelogenous leukemia (Chapter 195) encodes the aberrant Bcr-Abl tyrosine kinase. Inhibitors of Bcr-Abl tyrosine kinase, such as imatinib mesylate, are highly effective in this otherwise refractory leukemia as well as in gastrointestinal stromal tumors (Chapter 202). Another kinase inhibitor, sunitinib, was introduced as a potential drug for kidney cancer, with promising results. Given the current pace of research, more genetically based targeted therapeutic agents can be expected in the near future.
No hay comentarios:
Publicar un comentario