Chapter 7: Cancer Genetics

Introduction

Cancer is second only to cardiovascular disease as the most frequent cause of death in Western, industrialized nations. The “cost” of cancer for the individual, the family, and the nation is substantial and spans many elements, such as direct health care costs, loss of productivity, and costs associated with treatment, research, and development. Most cases of cancer are sporadic; fewer than 10% of all tumors result from a familial disposition. Yet each case, whether sporadic or familial, has a genetic cause. Cancer is a genetic disease.

Cancer Is a Genetic Disease

The term cancer is not a distinct disease but a general term for all malignant neoplasms. A proliferating cell mass is malignant when it grows independently of control mechanisms, being capable of transcending tissue boundaries, growing invasively, and metastasizing. Some types of malignant tumors (e.g., basal cell carcinoma) have a very low metastatic potential and may grow and infiltrate surrounding tissue without metastasizing for many years.

There are three major groups of cancers: carcinomas, which develop from epithelial tissue (e.g., skin, intestinal epithelium, bronchial epithelium, and the epithelium of the glandular ducts such as the mammary glands or pancreas); sarcomas, which originate from mesenchymal tissue (e.g., connective tissue, bones, muscles); and the malignant diseases of the hematological and lymphatic systems (leukemias and lymphomas, respectively).

Most types of cancer develop through the progressive accumulation of various mutations within a cell. These genetic changes are typically acquired somatically, although some can be transmitted through the germ line and are present at birth in every body cell. The genes responsible for cancer development can be classified into two major groups: protooncogenes and tumor suppressor genes.


Protooncogenes:
Genes that, through (dominant) activating mutations, can be turned into oncogenes. Oncogenes facilitate malignant transformation by synthesis of structurally altered or defective proteins.

Tumor suppressor genes:
Genes that are relevant for the regulation of growth, repair, and cell survival, with malignant transformation supported through (recessive) loss-of-function mutations on both copies of the gene.

They typically include DNA repair genes that are responsible for detecting and repairing genetic damage within a cell.

Malignant transformation:
The change from controlled to uncontrolled growth of a cell that is caused by mutations in oncogenes or tumor suppressor genes.


The coordinated interplay between growth-promoting and growth-controlling genes is required for maintaining the functional health of individual cells and tissue groups. Cancer development reflects the breakdown of this carefully balanced system. It is never a single mutation that causes cancer; rather, cancer is the result of the accumulation of several genetic and chromosomal changes. Only after a critical number of genetic changes have occurred does the initially controlled growth become an uncontrolled, malignant growth (malignant transformation).

The most extensive studies on this topic have been done for colorectal cancer. Bert Vogelstein and his colleagues described the tumor progression model (adenoma carcinoma sequence). Using this model, they were able to explain the impact of a succession of different gene defects on tumor development. It is remarkable how the accumulation of genetic defects in oncogenes and/or tumor suppressor genes correlates with morphological parameters. Tumor progression, from normal tissue to adenoma and carcinoma of the colon, takes approximately 10 years (Fig. 7.1).

Figure 7.1.Tumor Progression Model.

Tumor Progression Model.

(Adapted from Vogelstein et al. 1996.)

A Question of Balance

Uncontrolled growth is a common characteristic of all tumors, whether they are benign or malignant. Malignant tumors, however, also have the ability to grow invasively and metastasize. Uncontrolled growth of a tumor results from disruptions in intracellular, as well as intercellular, processing of information. Therefore, cancer does not necessarily result from increased cell proliferation.

Rather, it is a question of balance between cell division and growth, on one side, and apoptosis (programmed cell death), on the other. It is crucial for embryonic development, as well as adult organisms, that the different signaling pathways are accurately coordinated and regulated. There is an inverse relationship between the degree of differentiation and the extent of a cell’s proliferation. A malignant tumor tends to be less differentiated than its tissue of origin.

Oncogenes

From Protooncogenes to Oncogenes

Oncogenes develop from protooncogenes through so-called “hypermorphic” mutations that result in a gain of function. Such mutations are mostly missense mutations that cause permanent activation or altered function of the gene product (qualitative changes). Protooncogenes can also be duplicated or multiplied (amplification), resulting in increased gene copy numbers and thus more gene products in the cell (quantitative changes).

Also, translocations can turn a protooncogene into an oncogene by generating a fusion gene with novel function and/or placing it under the control of a new, constitutively active promoter, which might trigger abnormal expression with regard to organ system or developmental stage.

Intracellular Dominance

Typical protooncogenes are involved in pathways that regulate cellular growth, cell proliferation, and the cell cycle. This includes receptor tyrosine kinases, growth factors and their receptors, components of intracellular signaling cascades, and proteins that regulate the cell cycle. Oncogenes are dominant at the cellular level, which means that activation or overexpression of one single allele is sufficient to result in a change of the cell’s phenotype.

Two examples of oncogenes are the receptor tyrosine kinases RET and MET. RET is the receptor for growth factors of the glial cell line–derived growth factor (GDNF) family. Activating mutations in the RET gene are responsible for the development of MEN2, one of the multiple endocrine neoplasias (Section 32.5). It is interesting to note that inactivating mutations of the same gene can be a cause for Hirschsprung disease (Section 23.1). MET is the receptor for hepatic growth factor (HGF). Mutations in the MET gene can result in familial papillary renal carcinoma.

Among the most frequent oncogenic changes are mutations of genes of the RAS family (e.g., H-RAS, K-RAS, and N-RAS). RAS genes code for guanosine triphosphate (GTP) binding proteins that have a crucial regulating function for several important signaling cascades in the cell. K-RAS mutations occur in approximately 90% of all pancreatic carcinomas and in 50% of all colon cancers; mutations of N-RAS occur in roughly 30% of all cases of AML.

The classical example of a chromosomal translocation resulting in activation of a protooncogene is the translocation between chromosomes 9 and 22 that results in the so-called Philadelphia chromosome, t(9;22)(q34;q11). On a molecular level, the translocation causes a fusion of the BCR and ABL genes, leading to a fusion protein, BCR-ABL, and constitutional activation of the ABL tyrosine kinase (see Section 3.4 and Fig. 3.2).

Translocations of protooncogenes into chromosomal regions that are under the control of promoters for genes of the immunoglobulin chains (chromosomes 14, 22, and 2) result in uncontrolled, constitutive expression, as seen for the MYC protooncogene in Burkitt lymphoma (Table 7.1).

Table 7.1.Chromosomal Translocations as Causes of Malignant Diseases

Neoplasia Translocation Frequency Fusion Gene or Oncogene Involved
CML (chronic myeloic leukemia) t(9;22)(q34;q11) 90%–95% BCR-ABL
ALL (acute lymphatic leukemia) t(9;22)(q34;q11) 10%–15% BCR-ABL
Burkitt lymphoma t(8;14)(q24;q32)
t(8;22)(q24;q11)
80%
15%
MYC
CLL (chronic lymphatic leukemia) t(11;14)(q13;q32) 10%–30% BCL-1
Follicular lymphoma t(14;18)(q32;q21) >90% BCL-2

Tumor Suppressor Genes

Intracellular Recessiveness

Most familial cancer predisposition syndromes result from mutations in tumor suppressor genes in which loss of function favors development of a tumor. The products of these genes inhibit cellular growth, proliferation, or cell cycle progression (gatekeeper genes). Also, they can ensure genetic stability, for example, through DNA repair (caretaker genes). In contrast to oncogenes, which are activated by mutations on a single allele, a tumor-promoting phenotype associated with tumor suppressor genes is triggered if there are inactivating mutations in both alleles. These mutations are, therefore, recessive on a cellular level.

Especially frequent are null mutations that cause complete absence of a functional product, such as small frameshift deletions or nonsense mutations that cause aborted protein synthesis. Missense mutations are also commonly encountered; however, if those have not been characterized before, it may be difficult to determine their true functional impact and, consequently, they are labeled “variants of unknown significance” (or “unclassified variants”).


Variant of unknown significance (VUS):
A genetic variant identified in a patient with a particular disease or a suspected disease predisposition that may or may not be of functional importance. This type of variant is frequently encountered, for example, in mutation testing for cancer predisposition syndromes; as the clinical impact of the variant is uncertain, it cannot be used for predictive testing.

Two-Hit Hypothesis

In 1971, Alfred Knudson observed that 50% of the offspring of patients with bilateral retinoblastoma develop retinoblastomas themselves, while only 10% to 15% of unilateral retinoblastomas appear to be heritable. He postulated that all cases of bilateral retinoblastoma should be considered heritable. Children with familial retinoblastoma more frequently develop the disease in both eyes or simultaneously at multiple sites while also developing it earlier than children with single unilateral tumors. These observations led Knudson to establish the two-hit hypothesis of cancer development.

He postulated that children with bilateral tumors have inherited a constitutional mutation as a predisposition for development of the retinoblastoma. This single mutation is not enough to trigger the disease; a second hit, meaning a second somatic mutation, is required. Patients with sporadic retinoblastoma did not inherit a mutation and require two independent somatic mutations affecting the same cell. In both cases two hits are required for tumor development; in congenital retinoblastoma one of the mutations is already present in all cells.

Because, statistically, a somatic mutation is likely to occur in at least one relevant cell, the disease occurs almost inevitably and much earlier in constitutional mutation carriers than in noncarriers. Frequently, secondary tumors can develop independently (e.g., osteosarcoma and leukemia).

Two-hit hypothesis of cancer development:
Cancer development involves two successive mutations that affect the two alleles of a tumor suppressor gene. In familial cancer disposition syndromes, a mutation on one allele is inherited, and only one additional hit is required for cancer development.


The two-hit hypothesis applies to all known tumor suppressor genes. Mutations are required on both alleles to convert a normal cell into a tumor cell. In constitutional tumor predisposition syndromes, one mutated allele derives from the parental germ line so that offspring will have only one wild-type allele in all of their body cells (first hit). Each additional inactivating mutation of the second (wild-type) allele causes loss of function of the respective gene product within the affected cell (second hit) (Fig. 7.2).

Figure 7.2.Two-Hit Hypothesis Applied to Retinoblastoma.

Two-Hit Hypothesis Applied to Retinoblastoma.

Recessive Equals Dominant

Although inactivation of tumor suppressor genes reflects a recessive mechanism at the cellular level, the associated cancer predispositions are inherited as (autosomal) dominant disorders. This paradox is easily understood if one calculates that the human body contains approximately 1014 cells, and the average mutation rate per gene and generation is 1 in 106. Even though the specific target organ has much fewer cells, it is just a matter of time for at least one mutation event (one hit) to occur in at least one cell.

Normally this event does not have a functional effect, but an effect will occur if the cell already contains an inherited mutation on the other allele. Cancer development is, thus, a question of probability. Patients with an autosomal dominant tumor predisposition syndrome who never experienced a second hit in the relevant organ in their lifetime, therefore, would not develop any tumor. This is reflected by incomplete penetrance of the respective disease.

The seeming paradox of a dominant disorder with a recessive pathomechanism at the cellular level applies to all conditions where abnormal functioning of a single cell is sufficient for the development of clinical symptoms.


Epigenetic Processes

A modification of the two-hit hypothesis became necessary when it was realized that the second hit in the second allele did not necessarily have to involve a DNA change. Epigenetic processes, such as DNA hypermethylation, can also account for the inactivation of an allele of a tumor suppressor gene. Since the methylation status of a gene remains the same throughout all mitotic cell divisions, the effect resembles that of a true alteration of the DNA sequence.


DNA Repair Genes

The inactivation of DNA repair genes does not immediately trigger abnormal cellular growth or differentiation. However, it causes failure to identify and repair mutations in the entire genome, which may result in a 100-fold to 1,000-fold increase in the mutation rate. This has an impact on protooncogenes as well as other tumor suppressor genes or gatekeeper genes.

Since the accumulation of mutations is a decisive factor in tumor progression, inactivation of DNA repair genes significantly accelerates malignant transformation. One example of a hereditary tumor predisposition caused by mutations of DNA repair genes is Lynch syndrome, due to mutations in DNA mismatch repair genes such as MSH2 or MLH1 (Section 32.4). Other genetic disorders of DNA repair are discussed in Section 32.7.

Loss of Heterozygosity

In many cases, the functional loss of the second allele of tumor suppressor genes in cancer development is not caused by a point mutation (or epigenetic changes) but by larger deletions or chromosomal alterations that result in the total loss of that gene copy. This can be recognized through molecular genetic analysis of loss of heterozygosity (LOH). For example, when individuals with familial retinoblastoma and a disease-causing point mutation in the RB1 gene are investigated, all normal cells show heterozygosity for that mutation.

In contrast, tumor tissue from the same individual will only reveal the mutated allele. In the case of a deletion, the normal allele is absent and, hence, there is loss of heterozygosity. The mutation is really hemizygous, as the wild-type allele is not amplified by polymerase chain reaction (PCR) due to a large deletion or possibly loss of the entire chromosome. LOH is the most frequent mechanism for a second hit in heritable tumor predisposition syndromes and is often seen as part of the malignant progression of sporadic tumors.