1
Introduction
Edward J. Estlin1, Richard J. Gilbertson2 and Robert F. Wynn3
Scope and aims
The aim of this textbook is to provide the reader with a focused but comprehensive overview of the clinical and scientific principles that guide current treatments for childhood cancer. For this purpose, the book is divided into four sections, namely central nervous system tumors, hematological malignancies, non-central nervous system (CNS) solid tumors of childhood, and a final section which covers psycho-social support, palliative care and survivorship issues.
For each of the disease specific chapters, our aim is to present the reader with information that is visually distinctive and should allow easy access to key factual information that relates to the epidemiology, presentation, diagnosis, treatment, and prognosis for individual categories of childhood malignancy, along with a brief overview given of the history of therapeutic developments for any given disease type. To facilitate this, the paragraphs will contain regular bullet points to highlight the presentation of key factual information, tables and fact boxes which we hope will enable any reader to quickly pick out important information in relation to the disease type they are interested in.
Progress for the treatment of childrenâs cancers has traditionally involved advances at the scientific interface such as the recognition of the importance of chromosomal abnormalities, oncogene amplification and aberrations of tumor suppressor gene functions. In more recent times, advances in our understanding of the cellular biological characteristics of cancers is leading towards new insights that can enable a more rational treatment stratification and is also leading towards the development of specific and targeted therapies. Therefore, each of the disease-specific chapters will focus on an integration of the scientific and clinical principles that guide the treatment for these individual cancer types, and introduce the reader to advances in the field that are at the level of the clinical interface. Therefore, in order to help orientate the reader with the scientific information that is incorporated into individual chapters in the rest of the text book, the aim of this introduction is to help define those key scientific principles that pertain to the contemporary management of children with cancer, and which are currently informing novel therapies that are now close to or actually at the clinical interface.
For the descriptions that will be presented below, our aim is not to provide an exhaustive text and reference resource for the reader, but really to highlight the key terms and scientific principles that will serve as a glossary for the main body of the text book to follow, and which will involve referencing against contemporary textbooks and review articles that act as a starting point for further reading.
The epidemiology of childhood cancer
When compared with the adult population, cancer in children is rare and comprises less than 1% of the national cancer burdens of industrialized countries [1]. Moreover, whereas most adult cancers are carcinomas, the cancers that occur in childhood are histologically very diverse and comprise [2]:
- Leukemia, myeloproliferative diseases and myelodysplastic diseases.
- Lymphomas and reticuloendothelial neoplasms.
- CNS tumors.
- Neuroblastoma and other peripheral nervous cell tumors.
- Retinoblastoma.
- Renal tumors.
- Hepatic tumors.
- Malignant bone tumors.
- Soft tissue and other extra osseous sarcomas.
- Germ cell tumors, trophoblastic tumors, and neoplasms of the gonads.
- Other malignant epithelial neoplasms and malignant melanoma.
- Other unspecified malignant neoplasms.
Thus, for children, carcinomas are rare, and the majority of cancers present as acute leukemia, lymphoma (non-Hodgkinâs lymphoma, Hodgkinâs disease), sarcoma (osteogenic sarcoma, rhabdomyosarcoma), germ cell tumor and embryonal malignancies (neuroblastoma, nephroblastoma, medulloblastoma, hepatoblastoma). Embryonic tumors, which are thought to arise during intra-uterine or early post-natal development from an organ rudiment or immature tissue, and form structures that are characteristic of the affected part of the body, are rare in adults.
The age-standardized incidence of all cancers in children under the age of 15 years is between 70 and 160 per million children per year, corresponding to a risk of 1 in 100 to 1 in 400, and an annual worldwide incidence of approximately 160 000 new cases per year [2]. The epidemiology of childhood cancers is characterized by [1, 2]:
- The incidence is highest in the first 5 years of life, reaches a nadir at 9â10 years of age and then rises thereafter.
- Boys are affected more than girls.
- While total incidence rates vary only modestly between world regions, there is more marked variation for diagnostic subgroups. For example, among children of North America and Europe, acute leukemia forms the largest diagnostic sub-group, accounting for one-third of the total number of cases, with a lower incidence seen for sub-Saharan Africa, where lymphomas predominate as the most frequent childhood cancer.
- Within geographical regions, there are indications that racial influences may form a part in the susceptibility of children to cancer. For example, whereas in the USA, the incidence rates for acute lymphoblastic leukemia are highest for children of Hispanic origin, and lower for those of Afro-American ethnicity, the incidence of Wilmsâ tumor is lowest for this latter ethnic group.
- Brain and spinal cord tumors are second only to leukemia in industrialized countries, where they account for 20â25% of childhood cancer. The lower recorded incidence in developed countries may represent under diagnosis.
- Neuroblastoma and nephroblastoma have a fairly constant incidence rate worldwide.
- Environmental influences play a part in the variations of the incidence rate for individual cancers types found worldwide.
- Embryonal tumors and common acute lymphoblastic leukemia tend to affect younger children, osteogenic sarcoma and Hodgkinâs disease are more a diagnosis of adolescence, and rhabdomyosarcoma has bi-modal peaks with both younger children and adolescents being affected.
However, for the great majority (>95%) of cases of childhood cancer the causation is unknown, but those factors that are known to increase the risk of childhood are indicated in Box 1.1, and can be generally categorized as:
- Genetic causes â The largest contributions here come from heritable retinoblastoma, neurofibromatosis type 1 (NF1), tuberous sclerosis and the Li-Fraumeni syndrome. For example, children with NF1 have a relative risk for glioma and soft tissue sarcoma of 40, and germline mutations of the tumor suppressor gene TP53 (Li-Fraumeni genetic abnormality) are present in most children with adrenocortical carcinoma and about 10% of children with rhabdomyosarcoma.
- Infection â associated with Epstein-Barr virus (Hodgkinâs lymphoma & nasopharyngeal carcinoma), hepatitis B (hepatocellular carcinoma) and Human Herpes Virus 8 (Kaposi sarcoma), a phenomenon that at least in part explains some of the worldwide differences for the incidence rates of these diseases.
- Although the subject of extensive investigation, the influence of other environmental factors such as ultraviolet light, electromagnetic fields is uncertain, although the Chernobyl nuclear disaster has seen an associated increase in thyroid cancer in children of the affected geographical region.
Box 1.1 The epidemiology of childhood cancer.
- Genetic causes â retinoblastoma, neurofibromatosis type 1, tuberous sclerosis, Li-Fraumeni cancer family syndrome
- Constitutional chromosomal disorders â Downâs syndrome, Turners syndrome
- Inherited immunodeficiency and bone marrow failure syndromes
- Irradiation
- Infection
Genetics in relation to childrenâs cancers
The study and investigation of the genetics of childrenâs cancers has led to important advances in the understanding of the epidemiology and causation of certain malignancies such as retinoblastoma; the recognition of karyotypic abnormalities provides a vital part of the diagnosis and risk stratification for therapy of many childrenâs cancers. Some examples are described as follows:
- Philadelphia chromosome, which represents translocation between chromosomes 9 and 22 [t(9;22)], confers an adverse prognosis when this is associated with the diagnosis of acute lymphoblastic leukemia in children [3].
- Other translocations such as the translocation involving chromosomes 11 and 22 [t(11;22)] can help define disease entities such as Ewingâs tumor and malignant peripheral neuroectodermal tumor from their differential diagnoses [4].
- Chromosome losses from 1p and 11q and gain of chromosomal material for 17q for neuroblastoma are associated with an adverse outcome for this cancer type [5].
The chromosomal translocations described for acute lymphoblastic leukemia (ALL) and Ewingâs tumor above promote the malignant phenotype of these cancers. For example, the Ewingâs tumor gene translocation results in an oncogenic transcription factor, EWS-Fli-1, and the Philadelphia translocation results in production of a constitutively active receptor kinase, bcr-abl, which has influences on proliferation, cell cycle control, and cell death. The recognition of the links between the genetics of cancer and subsequent cellular biological functions are leading to advances in therapy with mechanism-based compounds, such as imatinib in the case of bcr-abl positive leukemia [3].
Cellular and molecular biology in relation to childrenâs cancers (Box 1.2)
Receptor kinases and intracellular signalling
Although underlying genetic abnormalities such as loss of tumor suppressor function and gain of oncogenic gene activity may underlie the pathogenesis of individual cancers in children, science is now bringing us insights into the processes that are important as biological determinants of the malignant phenotype in cancer cells. For example, extracellular growth factor/cytokines or mitogens can bind to receptors on the cell surface that are linked to receptor kinases, or these cell surface receptors can be constitutively active [6]. Such ligand/receptor interactions can lead to:
- The activation of intercellular signalling pathways such as Akt/PI3, mTOR and MAP kinases, which in turn leads to:
- The dysregulation of various cellular activities such as gene expression, mitosis, differentiation, cell survival/apoptosis and motility, and invasiveness.
Moreover, genetic mutations can also lead to dysregulation of intracellular signalling, as in the example of loss of the inhibitory effect of PTEN function on Akt/PI3 signalling by gene deletion [7], or the loss of tumor suppressor gene function by gene promoter methylation as in the example of Ras-association domain family 1. Proteins in the ras family are very important molecular switches for a wide variety of signal pathways that control such processes as cellular skeletal integrity, proliferation, cell adhesion, apoctosis and migration. Ras-related proteins are often deregulated in cancers, leading to increased invasion and metastases, and decreased apoctosis. Ras activates a number of pathways, but especially an important one seems to be the mitogenactivated protein kinases, which themselves transmit signals downstream to other protein kinases and gene-regulatory proteins. Inappropriate activation of the gene can occur when tumor suppressor genes are lost, such as the tumor suppressor gene NF1, and ras oncogenes can be activated by point mutations to be constitutively activated [8].
Box 1.2 Cancer biology and immunology.
- Loss of tumor suppressor function or amplification of oncogenic function promotes malignant phenotype
- Dysregulation of cellular signalling
- Deregulation of cell cycle control
- Disordered proliferation, metastases and survival
- Knowledge informing treatment stratifications and novel therapies
- Immunological properties exploited in diagnosis and therapy.
The biological characteristics of cancers are also influenced by their environment. For example, tissue hypoxia is known to contribute to the pathogenesis and maintenance of the malignant phenotype, and interaction between the physicochemical properties of cancers and biological systems that control cellular proliferation, migration, and survival is now increasingly well understood. As a particular example of this, the vascular endothelial growth factor family of ligands and receptors are modulated by hypoxia, and has now been extensively studied for both childrenâs and adult malignancies. This in turn is bringing forward the rational introduction of novel mechanism-based therapies that aim to disrupt specific processes important for the pathogenesis of cancer [9], rather than attempting to cause cancer cell death by non-specific DNA damage as in the case of most conventional cytotoxic agents as will be discussed below.
Cell cycle control
Cell cycle progression is monitored by surveillance mechanisms, or cell cycles checkpoints, that ensure that initiation of a later event is coupled with the completion of an early cell cycles event. Dysregulation of the progression of cells through the cell cycles is also a feature of malignant cells, and defects in certain molecules such as p53, the retinoblastoma protein (pRb) and cyclin kinase inhibitors (e.g. p15, p16, p21) that control the cell cycle have been implicated in cancer formation and progression [1O]. Virtually all human tumors degregulate either the retinoblastoma (pRb/p16(INK4a)/cyclin D1) and/or p53 (p14 (ARF)/mdm2/p53) control pathways [10].
One of the most studied control systems in cancer involves p53, otherwise known as protein 53, a transcription factor that regulates the cell cycle, and hence functions as a tumor suppressor [11].
- p53 has many anti-cancer mechanisms in that it can activate DNA repair proteins when DNA has sustained damage, it can hold the cell cycle at the G1/S regulation point upon DNA damage recognition, and it can initiate apoptosis if the DNA damage proves to be irreparable [11].
- In normal cells, p53 is usually inactive, bound to the protein MDM2, which prevents its action and promotes its degradation by acting as a ubiquitin ligase. Upon DNA damage or other stress, various pathways will lead to the association of P53 and the MDM2 complex [12].
Once activated, P53 will either induce cell cycle arrest to allow repair and survival of the cell or apoptosis to discard the damaged cell. How p53 makes this choice is currently unknown.
If the p53 gene is damaged, then tumor suppression is severely impaired. People who inherit only one functional copy of the p53 gene, TP53, are at risk of developing tumors in early adulthood, a disease known as the Li-Fraumeni cancer family syndrome. Indeed, more than 50% of human tumors contain a mutation or deletion of the TP53 gene [2]. The occurrence of retinoblastoma serves as a paradigm for the effects of loss of the tumor suppressor function of pRb, and this will be discussed in Chapter 18 of the textbook.
Cancer biology and risk stratification
Advances in our understanding of the biology of childhood cancer are providing new insights into prognosis and will serve to underpin rational approaches to the stratification of treatments. For example, whereas an adverse prognosis for childhood medulloblastoma is found to relate to the growth factor receptor erb-b2 expression [13], nuclear accumulation of beta-catenin is associated with the activation of the Wnt/Wg signalling pathway and a more favorable prognosis [14]. This information may allow a staged reduction in the radiotherapy burden for the treatment of medulloblastoma, and also promote the development of specific mechanism-base therapies [15].
Cancer immunology
Knowledge of the immunology of childhood cancer is important for the diagnosis of childhood leukemias and can also be exploited for the therapy of childhood cancers. For example, the cluster of differentiation (CD nomenclature) has been developed to characterize the monoclonal antibodies that have been generated against epitopes on the surface molecules of leucocytes.
- The CD system is commonly used as cell markers; this allows cells to be defined based on what molecules are present on their surface.
- CD molecules are utilized in cell sorting using various methods, including flow cytometry, and can be used for the recognition of stem cells (CD34+, CD31â): all leukocyte groups, (CD45+): T lymphocytes, (CD3+) and B leukocytes, (CD19+).
The immune system can also be utilized to direct therapy against the cancers themselves. For example, dendritic cell-based immunotherapy utilizes dendritic cells, which are antigen-presenting cells that are harvested from patients to activate a cytotoxic response towards an antigen. Briefly, the dendritic cells are harvested from patients, these cells are then either incubated with an antigen or infected with a viral vector, and the activated dendritic cells are then transfused back into the patient. These cells then present the tumor-associated antigens to the effector lymphocytes, namely CD4+ T-cells and CD8+ T-cells, and some classes of B lymphocyte also. A similar procedure exists for T cell-based adoptive immunotherapy, which sees T-cells that have a natural or genetically-engineered reactivity to a patientâs cancer expanded in vitro and than adoptively transferred into the cancer patient [16].
In the situation of bone marrow transplantation, in particular in the case of allogeneic bone marrow transplantation, a balance is sought between rejection of the transplanted graft, eventual immune tolerance of the graft and maintenance of a controlled degree of graft versus host disease in order to maximize anti-leukemic effect [17].
The general principles of pharmacology in relation to childhood cancer
Radiotherapy continues to form an important part of the treatment of many childhood cancers, and this will be discussed further in the subsequent disease-specific chapters of this textbook. The cornerstone for the treatment of many cancers of childhood remains conventional chemotherapy agents, and a summary of these is presented below. Although a detailed description of the properties of anticancer agents in terms of their pharmacokinetic profiles, and a description of the cellular and molecular pharmacological processes that can limit or potentiate their effectiveness in cancer cells, is beyond the scope of this chapter, the reader is referred to the excellent reference text by Chabner and Longo for further information [18]. Basically, conventional chemotherapy agents usually act to promote DNA damage in all cells of the body, and the cure for cancer thus relies on there being a therapeutic index allowing eradication of the tumor at acceptable toxicity to the patient. The conventional cytotoxic age...