Daniel von Allmen

Historical Overview

Early strides in improving oncologic outcomes through the use of single chemotherapeutic agents were first reported by Farber in 1948¹ and Li in 1956.² As additional chemotherapeutic agents were developed, they were combined in multidrug regimens that demonstrated both significantly improved response rates and response duration.^3,4 By the late 1970s, multimodal therapy was shown to improve cure rates in children with Wilms tumor⁵ and was being adopted for the treatment of rhabdomyosarcoma, Ewing sarcoma, lymphoma, and other solid tumors.⁶ The close collaboration of multidisciplinary cooperative groups and the development of improved supportive care measures added impetus to progress. During the 1990s, dose-intensive chemotherapy programs were shown to be successful in improving outcome for patients with Burkitt lymphoma, neuroblastoma, and other advanced-stage solid tumors.^7–9 In addition, improvements in outcome were achieved either by alternating effective groups of chemotherapeutic agents to overcome or prevent resistance,¹⁰ or by administering agents by continuous infusion rather than bolus.¹¹

By 2001, noncytotoxic biologic therapies (e.g., signal transduction inhibitors, various tissue growth factor receptor inhibitors, antiangiogenesis agents, tumor-targeted antibodies, and adoptive immunotherapy techniques) were developed to target specific biologic pathways.¹² In addition, improvements in radiation therapy have led to the development of intraoperative radiation therapy and radiosurgery techniques. Collectively, these advancements have resulted in the continually increasing survival of children with solid tumors and a profound improvement in their quality of life.

The aim of this chapter is to present the key aspects of contemporary multimodal therapy, touching on treatment rationales, the impact of tumor biology on specific treatment approaches, and future trends.

Incidence and Survival Rates

Although childhood cancers account for only 2% of all reported cancer cases in the general population, they account for 10% of all deaths in children.¹³ On average, 1 to 2 of every 10,000 children in the USA develop cancer each year.¹⁴ The distribution of the types of cancer in childhood is different than in adults. Whereas most cancers in adults have an epithelial cell origin, <10% of childhood cancers fall into this category.

The incidence of specific cancers varies by age, gender, and race. Overall, however, it rose from 11.5 cases per 100,000 children in 1975 to 14.8 per 100,000 in 2004.¹⁴ The peak incidence (>200 cases per million) is in children younger than 2 years of age. The incidence then decreases to a low of 82.5 cases per million by age 9 years, at which point it begins to rise again through adolescence. In children younger than age 2, central nervous system malignancies, neuroblastoma, acute myeloid leukemia (AML), Wilms tumor, and retinoblastoma account for the majority of diagnoses. In children 2 to 4 years of age, acute lymphoblastic leukemia (ALL) is the most common childhood cancer. After age 9 years, the incidence of Hodgkin disease, osteosarcoma, and Ewing sarcoma begins to increase sharply.¹³

Over the past several decades, the mortality from childhood cancers has declined dramatically (40%),¹⁵ while the 5-year survival rate has risen from 58% in 1975–1977 to close to 80% in 1996–2003.¹⁴

Tumor Pathology and Chemotherapeutic Regimens

Although the spectrum of malignancies in children is more limited than that in adults, the exact diagnosis is often more difficult to establish due to the common histologic appearance of small round blue cells. As these primitive malignancies often lack morphologically distinguishing characteristics, Ewing sarcoma, neuroblastoma, lymphoma, small-cell osteosarcoma, and primitive neuroectodermal tumors may appear quite similar by light microscopy. Given that chemotherapy regimens must be carefully tailored to each specific tumor type, pediatric subspecialists have continued to better define prognostic subgroups for many tumors. These subgroups help to determine the best therapy and the dose intensity required for optimal outcomes. The initial step in the accurate diagnosis of a tumor is the availability of adequate tissue with which to make the diagnosis. The quality and quantity of this tissue should be discussed with the surgeon, the pathologist, and the pediatric oncologist before the procedure to ensure the proper handling of the specimen (e.g., the need for fresh tissue, frozen samples, and fixed specimens for histologic and biologic diagnostic use). Whereas light microscopy remains the primary tool of pathologists, immunohistology, electron microscopy, DNA content of tumor, cytogenetic abnormalities, and specific tumor gene expression are now used to determine specific tumor subgroups.

Tumor Biology

Genetic alterations within a single cell, such as the activation of an oncogene or the loss of a tumor suppressor gene, can lead to the accumulation of cells lacking the ability to respond to growth-regulating signals and the subsequent development of cancer.

Understanding normal cell growth and regulation is a prerequisite to understanding both the genetic basis for the development of childhood cancer and the mechanisms of action of chemotherapeutic agents designed to kill rapidly proliferating cancer cells (Fig. 64-1). Normal cell growth occurs by the regulated progression of the cell through the cell cycle of DNA replication and mitosis. This cycle is separated by two intervening growth phases, referred to as G₁ and G₂. Cells can temporarily leave the cell cycle and enter a resting state referred to as G₀. They are programmed to proceed through the cell cycle by a series of external and internal stimuli. Binding of proteins (growth factors) to cell surface receptors stimulates a cascade of cytoplasmic signaling proteins (membrane kinases and signal transducers) that carries the stimuli to the nucleus. Other proteins (transcription factors) then bind to the DNA, resulting in the expression of growth-regulating genes. When functioning normally, these genes promote or prevent cell division, direct the cell to differentiate, or initiate apoptosis, the process of programmed cell death.

Alterations in one or several of these signaling proteins can lead to the unregulated cell growth characteristics of cancer cells. Oncogenes result from mutation or overexpression of the normal growth-promoting proto-oncogenes. Tumor suppressor genes are normally present in cells and function as negative regulators to slow the process of proliferation and allow time for cellular repair. When oncogenes become activated or tumor suppressor gene function is lost, cells lose their ability to respond to the usual regulatory protein stimuli and proliferate rapidly. Rapid cell proliferation leads to accumulation of more genetic defects, activation of additional oncogenes, and loss of more negative regulators as the cells become increasingly more malignant. Through the study of chromosomal aberrations, more than 100 oncogenes and 25 tumor suppressor genes have now been identified.

Cancer Cytogenetics

With the discovery of chromosomal banding techniques in the 1970s, cancer cytogeneticists were able to identify sub-chromosomal deletions, inversions, and translocations occurring in cancer cells. Investigation of these aberrant regions led to the identification of oncogenes and tumor suppressor genes, a process that continues today.

The presence of consistent cytogenetic abnormalities associated with a specific childhood leukemia or solid tumor is useful in both cancer diagnosis and prognosis. Specific cytogenetic aberrations have been identified in rhabdomyosarcoma, Ewing sarcoma, synovial sarcoma, germ cell tumors, medulloblastoma, neuroblastoma, retinoblastoma, and Wilms tumor.¹⁶ Chromosomal aberrations can also be helpful in predicting prognosis. For example, the finding of a chromosome 1q deletion, the presence of double-minute chromatin bodies, or the presence of homogeneous staining regions in neuroblastoma confer a poor prognosis.¹⁷

The treatment of Wilms tumor based on risk stratification determined by multiple tumor characteristics that impact prognosis (including loss of heterozygosity at 1p and 16q) is a topic of ongoing study. In the future, specific tumors may be identified by a specific ‘fingerprint’ determined by microarray analysis that can simultaneously analyze expression of thousands of genes on a single chip.¹⁸ The ability to tailor therapy to individual patients based on the genetic characteristics of their particular tumor is quickly becoming a reality.

Clincial Trials

The goal of chemotherapy in children is to maximize tumor kill while maintaining acceptable side effects. Clinical trials have led to the development of combination chemotherapy regimens for most childhood cancers. Adjuvant therapy (initiated after local control measures) has remained the mainstay of treatment, though neoadjuvant chemotherapy (initiated before definitive local control measures) has proved to be effective in patients with metastatic disease as well as in those with large primary tumors. In the latter clinical setting, chemotherapy is used to decrease the size of the tumor, making it more amenable to complete resection.

The first step in the clinical development of an anticancer agent is to define a tolerable dose. Phase I clinical trials are designed as dose-escalation studies to determine the maximally tolerated dose of a new drug given as a single agent. In a phase II trial, a consistent dose of the agent is tested for efficacy in a variety of tumor types to establish the spectrum of activity of the agent. Once an agent has demonstrated activity toward a specific cancer, this agent is tested for efficacy when combined with other known active agents for that tumor system.

Standard therapy for a specific tumor type is established through phase III clinical trials. These trials use a prospective randomized design to compare two chemotherapy combinations. At the conclusion of the trial, the chemotherapy regimen with the greatest efficacy and the least toxicity is selected as the standard regimen for that tumor. Subsequent phase III trials compare new regimens with this newly established standard. It is through the development of phase I through III clinical trials within national cooperative groups that treatment advances are made.

Combination chemotherapy remains the mainstay of treatment. The likelihood of cure is maximized when all available active agents are administered simultaneously after local control measures have been undertaken and the tumor burden is as low as possible.¹⁹ This approach has led to increased survival rates in children with neuroblastoma, Ewing sarcoma, anaplastic Wilms tumor, and osteosarcoma.

Adjuvant and Neoadjuvant Chemotherapy

The use of adjuvant chemotherapy is supported by the finding that less than 20% of sarcoma and lymphoma patients with initially nonmetastatic solid tumors can be cured by operative or radiation therapy alone or combined.²⁰ Tumor recurrence is generally at a distant site, lending support to the hypothesis that micrometastatic disease exists at the time of presentation for most patients with clinically nonmetastatic disease. As many as 40% of patients with Wilms tumor can be cured with resection or radiation therapy alone. However, survival increases to 90% with the addition of combination chemotherapy.²¹

As the goal of adjuvant chemotherapy is to prevent the growth of metastatic disease, it is vital that chemotherapy begins as soon as possible after local control measures are employed. For this reason, most current chemotherapy protocols for childhood solid tumors advise that chemotherapy is initiated within 2 weeks of operation.²² In children with Wilms tumor, rapid assessment of tumor biology and risk stratification is important for determining the appropriate chemotherapy regimen.

Neoadjuvant chemotherapy has become standard in the treatment of Ewing sarcoma and osteosarcoma, and has the theoretical advantage of minimizing resistance to chemotherapy.10,23,24 Delayed surgical intervention may allow a more complete or less morbid resection, as well as histologic assessment of tumor responsiveness to the chemotherapy agents. Neoadjuvant chemotherapy is beneficial only in tumors for which a known highly effective combination chemotherapy program limits the risk of tumor progression at the primary site. For example, diagnostic biopsy followed by neoadjuvant chemotherapy and delayed resection of the primary tumor for complex neuroblastoma reduces the operative complication rate without compromising survival.²⁵

Dose Intensity and Duration of Chemotherapy

Advances in supportive care have enabled increased dose intensity of active chemotherapy agents in pediatric clinical trials. These advances decrease or minimize the toxic effects of higher-dose chemotherapy on normal tissues. The use of cytokines (granulocyte colony-stimulating factor [G-CSF] and interleukin-11 [IL-11]) to speed recovery of white blood cells and platelets,26,27 and the use of cardioprotectant agents to allow use of a higher cumulative dose of doxorubicin have helped in the development of new dose-intensive therapy for solid tumors.²⁸ Similarly, progress in bone marrow and stem cell transplantation have allowed dose intensities to be pushed to the upper limits.^29–32

The positive impact of increasing the dose intensity on improving response rate and survival has been demonstrated for Burkitt lymphoma, osteosarcoma, Ewing sarcoma, testicular cancer, and advanced ovarian cancer.7–9,33–36 The duration of chemotherapy programs for most pediatric solid tumors has been 1 year. However, as the dose intensity of chemotherapy programs has increased, the duration of therapy has concomitantly decreased. This downward trend is likely to continue.

Chemotherapeutic Agents

Chemotherapy drugs are divided into classes by their mechanism of action. These classes include alkylating agents (cisplatin and its analogs), antimetabolites, topoisomerase inhibitors, antimicrotubule agents, differentiation agents, miscellaneous nonclassified agents, and biologic agents. Understanding the mechanism of action for each of these agents helps in establishing combination therapies with synergistic antitumor effects. The most common agents from each class, their mechanism of action, common side effects, and tumors in which they are active are listed in Table 64-1.

Acute Chemotherapy Toxicity and Supportive Care

Most acute toxicities in childhood solid tumor therapy are reversible. Because toxicity is greatest in normal cells that have the highest rate of turnover, normal bone marrow cells, mucosal lining cells, liver cells, and hair cells are frequently affected. The most common side effects from combination chemotherapy include nausea and vomiting, myelosuppression, hair loss, mucositis, diarrhea, liver function abnormalities, and allergic reactions.

Myelosuppression is an expected side effect of almost all treatment protocols for solid tumors. Transfusions of packed cells and platelets are frequently needed. Of greatest concern is the risk of severe life-threatening bacterial or fungal infections that occur during episodes of neutropenia. In dose-intensive regimens, more than 75% of chemotherapy courses result in hospitalization for fever, with the incidence of bacteremia ranging from 10– 20% per course.³⁷

Several chemotherapeutic agents have specific toxicities. For example, vincristine and doxorubicin are vesicants and can cause severe skin and tissue necrosis if the drug extravasates into the subcutaneous tissue. Because doxorubicin and related anthracyclines have cumulative cardiotoxic effects, the total lifetime anthracycline dose must be limited for each patient to minimize the risk of developing congestive heart failure. Cisplatin has toxic renal effects and is often combined with another nephrotoxic agent, ifosfamide, in the treatment of osteosarcoma and neuroblastoma. Cisplatin also can cause hearing loss, especially in patients who receive high doses. Vincristine and vinblastine can cause cumulative peripheral neuropathies, and drug doses frequently must be altered to prevent significant morbidity. All of these toxicities must be considered when designing therapeutic programs.

Success in improving treatment outcomes is partially attributed to advances in supportive care. The routine use of hematopoietic growth factors, specifically G-CSF, results in more rapid granulocyte recovery and shorter hospitalizations for fever and neutropenia.²⁶ In addition, IL-11 enhances platelet recovery, decreases the depth of the platelet nadir, and decreases platelet transfusion requirements.^27,38,39 It is well tolerated and is beneficial in combination regimens that induce severe thrombocytopenia.^40,41

The gastrointestinal tract is injured by cytarabine, anthracyclines, and high-dose methotrexate; however, the folate derivative Leucovorin can be given to rescue normal mucosal and bone marrow cells from the effects of the latter. No rescue is known for the mucositis and diarrhea which occur with other agents. In addition to enhancing platelet production, IL-11 may help speed recovery from gastrointestinal injury after chemotherapy.⁴²

Renal toxicity can occur from the use of cisplatin, ifosfamide, and high-dose methotrexate. Cisplatin causes renal tubular damage, leading to elevation of levels of blood urea nitrogen and creatinine; this effect is generally reversible. Both ifosfamide and cisplatin cause renal electrolyte wasting in which hypokalemia, hypocalcemia, hypophosphatemia, and hypomagnesemia can occur. Renal injury from these agents can be improved by hyperhydration and forced diuresis. Mesna can prevent hemorrhagic cystitis resulting from cyclophosphamide and ifosfamide by binding to the bladder-toxic acrolein metabolites.⁴⁰

Additionally, a recent study demonstrates that amifostine administered prior to and during cisplatin infusion significantly reduces the risk of severe ototoxicity in children with average risk medulloblastoma who are undergoing treatment with dose-intense chemotherapy.⁴³

Targeting Biologic Pathways

Over the past several decades, many of the key genetic events that control carcinogenesis have been identified. More recently, progress has been made in the development of agents designed to target specific biologic pathways. These agents include signal transduction inhibitors (tissue growth factor receptor inhibitors, antiangiogenesis agents, and biologic response modifiers), individual cytokines, tumor-targeted antibody therapies, and adoptive immunotherapy techniques. The development of these agents differs from the development of agents used in standard cytotoxic therapy. Whereas standard phase I studies for cytotoxic agents are designed to define the maximal tolerated dose, in biologic targeted therapy the optimal therapeutic dose is well below the maximal tolerated dose. The challenge in evaluating these newer agents is how to select the optimal dose and schedule, combine tumor-targeted agents with classic cytotoxic therapy, and validate the intended effect on the selected target for these biologic compounds designed to treat minimal residual disease.