Elizabeth Fox
Peter C. Adamson
Introduction
Despite substantial improvements in survival, the poor treatment outcome for many high-risk patients, coupled with the acute and late effects of current therapy, emphasize the need to develop more targeted and less toxic therapies for children with cancer.
Development of cytotoxic anticancer agents
In order to appreciate the paradigm shift that has occurred in anticancer drug discovery, it is important to understand the history of drug development that formed the foundation of today's successful therapy, which predominantly relies upon the use of cytotoxic agents.
The successful identification of active anti-leukaemia drugs began in the late 1940s, in part through a process of rational drug discovery of antimetabolites.1,2 As a result of that effort, methotrexate and mercaptopurine remain the cornerstone of acute lymphoblastic leukaemia (ALL) maintenance therapy to this day.3,4
Many of the cytotoxic agents in use today were identified through a large-scale screening process carried out at the National Cancer Institute (NCI). The initial screening programme began in 1955 and relied on L1210 and P388 murine leukaemia in vivo models. An in vitro screening programme for pure compounds and natural products was implemented in 1985, and by 1990 was fully operational as the NCI-60 cell panel. The 60 human tumour cell lines used in this in vitro cytotoxicity screening are exclusively adult histologies.5 Since the NCI-60 panel screen does not include cell lines from paediatric malignancies, its utility for prioritizing new agents for paediatric drug development is limited.6
Only a small number of preclinical animal models have been successfully utilized for paediatric drug development. Therefore other criteria have been relied upon in selecting agents for development in paediatric malignancies,7 including selecting agents with a novel mechanism of action, a unique resistance profile, or an improved toxicity profile. Agents have also been studied because of favourable pharmacologic properties such as good penetration across the blood–brain barrier, improved oral bioavailability, or a new formulation that altered the distribution and toxicity profile. Most commonly, agents that have shown promising activity in early adult trials have been prioritized for paediatric development.
Unfortunately, activity in adult cancers does not necessarily predict for activity in childhood cancers. This has recently been observed with paclitaxel, a taxane that is highly effective in ovarian, breast, non-small-cell lung, and head and neck cancers8 but has not demonstrated significant activity in a spectrum of paediatric solid tumours.9,10
Identification of novel cytotoxic agents for the treatment of childhood tumours continues to be an active area of clinical research. The current status of novel microtubular toxins in paediatric oncology serves as an example of how continued investigation of cytotoxic agents may improve the care of children with cancer.
Microtubular toxins
Microtubules are critical to cell division, intracellular structure, transport, and cell signalling. Before the advent of molecularly targeted therapy, agents that interfered with tubulin function by disrupting the dynamic equilibrium (polymerization and depolymerization) of tubulin were identified as potent anticancer agents11 (Fig. 5.1) In 1967, the vinca alkaloids, vincristine and vinblastine, were isolated from the periwinkle plant Catharanthus roseus and demonstrated cytotoxic activity via inhibition of tubulin polymerization.12 Vincristine has a broad spectrum of activity in paediatric cancer including ALL, Hodgkin and non-Hodgkin lymphoma, Wilms tumour, rhabdomyosarcoma, Ewing sarcoma, brain tumours, and neuroblastoma. Vinblastine is active in testicular cancer and Hodgkin's disease. Newer microtubular toxins now expand upon this foundation.
Vinorelbine
Vinorelbine, a semisynthetic derivative of vinblastine, has a broad spectrum of cytotoxic activity, may not be cross-resistant to the other vinca alkaloids, and is orally bioavailable. Vinorelbine appears to selectively inhibit mitotic microtubule formation rather than neural axonal microtubule formation and therefore may be less neurotoxic. In 1994, vinorelbine was approved by the US Food and Drug Administration (FDA) for the treatment of non-small-cell lung cancer in adults. In xenograft models of central nervous system (CNS) malignancies, vinorelbine demonstrated antitumour activity against several adult and paediatric gliomas.13
|
|
|
Fig. 5.1 Microtubular toxins. Tubulin, which is composed of α and β subunits, exists in a dynamic equilibrium of polymerization and depolymerization. Microtubulin toxins interfere with microtubule function by disrupting the dynamic equilibrium. |
Paediatric phase I and II studies have now been completed. In a paediatric phase I trial of vinorelbine the dose-limiting toxicity was neutropenia and the recommended dose was 33.5mg/m2.14 A Children's Oncology Group (COG) paediatric phase II trial of vinorelbine completed accrual in June 2002. Children with relapsed rhabdomyosarcoma, extraosseous Ewing sarcoma, neuroblastoma, and selected CNS malignancies were treated with i.v. vinorelbine weekly for 6 weeks followed by a 2-week break. Response and toxicity data from this trial are pending. Evidence of activity of vinorelbine in paediatric patients with recurrent sarcoma has been reported in a paediatric trial in Italy in which i.v. vinorelbine 30 mg/m2 was administered on days 1 and day 8 of a 21-day schedule. Partial responses were observed in six of twelve patients with rhabdomyosarcoma, one of five patients with osteosarcoma, and one of seven patients with Ewing sarcoma.15
Taxanes
The taxanes, paclitaxel and docetaxel, inhibit tubulin depolymerization, disrupt the polymerization and depolymerization equilibrium, and thus interfere with microtubule function. As stated above, paclitaxel is active in a variety of adult malignancies but activity has not been demonstrated in paediatric solid tumours. Docetaxel, a semisynthetic taxane, may have activity against selected paediatric solid tumours. Two phase I trials of docetaxel have been completed in children with refractory solid tumours. In the initial trial, docetaxel was administered every 21days. The MTD was 65 mg/m2 in heavily pretreated patients and 125 mg/m2 in less heavily pretreated patients. Dose-limiting toxicity was neutropenia and fatigue.17 A subsequent study, using the same schedule of docetaxel with granulocyte colony-stimulating factor support, reached an MTD of 185 mg/m2 with dose-limiting toxicity of desquamating rash and myalgia18 (Table 5.1) Recently, a phase II study of docetaxel (125 mg/m2 i.v. every 21 days) in children with refractory solid tumours was completed. Preliminary observations include activity in patients with Ewing sarcoma and osteosarcoma.
Investigational microtubular toxins
Other classes of agents that interfere with microtubule function are being investigated in children with cancer. BMS247550 is an epothilone analogue, a non-taxane microtubule-stabilizing compound, extracted from the fermentation broth of Sorangium spp. Epothilones block mitosis, resulting in cell death.19 BMS247550 has potent cytotoxic activity in paediatric cell lines20 and is currently being evaluated in a paediatric phase I trial. ABT-751 is a novel orally bioavailable sulfonamide antimitotic agent that binds to the colchicine binding site on β-tubulin and inhibits polymerization of microtubules. It has demonstrated a broad spectrum of activity in vitro and in xenograft models of human tumours in vivo including those that are resistant to paclitaxel, vincristine, and doxorubicin because of the multidrug-resistant phenotype. It was most active in a preclinical murine sarcoma model.21 The unique binding site for an antimitotic agent, broad spectrum of activity in preclinical studies, and oral bioavailability make ABT-751 a potentially important new agent for evaluation in the paediatric population. A paediatric phase I trial of ABT-751 is being performed.
|
Table 5.1. Dose-limiting toxicities and recommended doses of docetaxel |
||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||
Molecularly targeted anticancer agents in paediatric oncology
Rapidly accumulating knowledge of the molecular pathogenesis of cancer provides new targets for drug discovery and development resulting in a shift from an empiric random screening of cytotoxic anticancer agents to a more mechanistic target-based approach.
Identification of active molecularly targeted agents often involves high-throughput screening in which large numbers of compounds can be tested for activity as inhibitors or activators of a biologic target. In vitro biochemical assays, such as ligand–receptor and protein–protein interactions, or cell-based assays using cell lines or yeast reconstituted with specific targets have been utilized for screening. Specificity, potency, and novel chemical structures are critical factors in screening for lead compounds. Once identified, a lead compound can be optimized and ultimately formulated into a drug.22
The clinical development of target-based anticancer drugs may require changes to the traditional clinical trial design and endpoints that have been used for cytotoxic drugs. In the phase I and II setting, the traditional endpoints of toxicity and response may not be appropriate for target-based agents which gain selectivity, in part, in a dose-dependent (concentration-dependent) manner. Traditional endpoints may need to be replaced by biologic or pharmacokinetic endpoints to define the optimal dose and the therapeutic effect of the drug on its target.23
Early phase clinical trials of molecularly targeted drugs performed in adults and children must attempt to answer several critical questions. These include the following.24
1. Can a sufficient concentration of a drug be achieved safely in blood and in target tumour tissue?
2. Can target inhibition by the agent be demonstrated in the tumour or surrogate tissue?
3. Does target inhibition by the agent result in downstream effects that modulate activity or toxicity?
4. Is the desired biologic effect, such as inhibition of angiogenesis, induction of apoptosis, or inhibition of proliferation or metastasis, observed?
Our increased understanding of the malignant process has identified a spectrum of potential drug targets. Currently, a robust area of drug development is the signal transduction inhibitors, several of which may have a role in the treatment of childhood cancer. Other targets, including the proteosome and histone deacetylase, are actively being pursued as drug targets.
Signal transduction inhibition
Aberrant signal transduction pathways are a hallmark of malignant transformation, tumour initiation, and progression. Numerous agents have been synthesized to target molecules in signal transduction pathways in adult malignancies. Determining the potential role of currently available signal transduction inhibitors for childhood malignancies is an area of ongoing research, with a number of signal transduction inhibitors in early phases of paediatric clinical trials.
Imatinib (Gleevec™)
Imatinib mesylate (STI571, Gleevec™) targets the Bcr–Abl fusion protein, a constitutively activated tyrosine kinase in chronic myelogenous leukaemia (CML). The Bcr–Abl fusion protein is present in 95 per cent of patients with CML, and its tyrosine kinase activity is essential to the malignant transformation in CML. The quest for inhibitors of this tyrosine kinase began in the 1980s using high-throughput screens of chemical libraries. Largely through the research of Dr Brian Druker of The Oregon Health and Science University Cancer Institute, imatinib mesylate was found to inhibit the Bcr–Abl tyrosine kinase as well as the platelet-derived growth factor receptor (PGDF-R) and mutated c-kit in gastrointestinal stromal tumours (GISTs).25,26 In vitro experiments demonstrate that imatinib inhibition of specific tyrosine kinases appears to be concentration dependent26 (Table 5.2). In 1998, clinical trials with imatinib commenced and the agent received accelerated FDA approval based on response for use in adults with CML in 2001 and GIST in 2002.
A number of paediatric solid tumours are known to over express PDGF-R or c-kit. PDGF-R is expressed on some osteosarcomas, desmoplastic small round blue cell tumours, and synovial cell sarcoma. Some Ewing sarcoma family tumours and neuroblastomas overexpress c-kit or its ligand stem cell factor. Although it is well recognized that expression of protein does not necessarily predict clinical activity of a targeted inhibitor, initial laboratory and paediatric clinical trials of imatinib mesylate are beginning to assess the role of signal transduction inhibitors for children with cancer. A COG phase I trial of imatinib mesylate in paediatric patients with Ph+ leukaemia has been completed; no dose-limiting toxicities were observed and preliminary response rates were similar to those obtained with imatinib in adults with CML. Based in part on this data, imatinib mesylate received FDA approval for the treatment of children with Ph+ CML in May 2003. A COG phase II trial of imatinib in children with refractory or relapsed solid tumours opened in June 2002. In addition to estimating response rates in selected paediatric solid tumours, the trial seeks to determine the time to progression and correlate response to expression of PGDF-R and c-kit in these tumours.
Epidermal growth factor receptor inhibitors
Epidermal growth factor receptor (EGFR) tyrosine kinase is a critical component in the signal transduction cascade related to invasion and metastasis of many tumours.27 In May 2003, the FDA approved the EGFR inhibitor gefitinib (ZD1839, Iressa™) for adults with refractory non-small-cell lung cancer. A paediatric phase I trial of gefitinib is in progress in the COG. In addition to determining the maximum tolerated dose, toxicity spectrum, and pharmacokinetics in paediatric patients with refractory solid tumours, the expression and activity of EGFR and downstream signalling pathway mediators will be documented and the biologic effects of gefitinib on normal epithelial cells will be studied as pharmacodynamic endpoints. By coupling pharmacokinetic and pharmacodynamic endpoints, this trial aims to determine the optimal biologic dose of gefitinib rather than the maximum tolerated dose of the agent in the paediatric population. A phase I trial of another EGFR inhibitor, OSI-774, is also planned. Since EGRF inhibitors may enhance the cytotoxic effects of standard chemotherapy, OSI-774 will be administered in combination with an alkylating agent (temozolomide) in children with refractory solid tumours.
|
Table 5.2. Concentration-dependent inhibition of tyrosine kinases by imatinib mesylate |
||||||||||||||
|
||||||||||||||
Farnesyl transferase inhibitors
A number of agents that target ras are also being developed for paediatric malignancies. Ras genes encode a family of guanosine triphosphate (GTP) binding proteins that play a critical role in the regulation of cell growth and differentiation. Ras proteins are activated by receptor tyrosine kinases on the cell surface and initiate phosphorylation cascades that sequentially activate effectors including Raf-1 and the MAPK pathway, the Rac–Rho pathway, MEK-1 and the JNK pathway, and PI3 kinase. These signal transduction pathways are critical to growth regulation in normal and malignant cells. In order to participate in signal transduction, Ras proteins must be associated with the inner surface of the cell membrane. Membrane association is facilitated by the post-translational addition of a lipid moiety, farnesyl, catalysed by the enzyme farnesyl transferase (FTase) (Fig. 5.2). In normal cells, growth factors bind cell surface receptors causing membrane-bound ras to switch from an inactive guanine diphosphate (GDP) bound form to an active GTP-bound form, initiating the signal transduction cascade. In many tumour cells, ras is mutated and remains in the active GTP-bound form in the absence of external growth signals. One strategy to inhibit ras-mediated signal transduction is to prevent its association with the cell membrane by inhibiting FTase. Two FTase inhibitors are being evaluated in paediatric clinical trials. In studies coordinated by the Paediatric Oncology Branch of the NCI, tipifarnib (R115777, Zarnestra2) is being evaluated in solid tumours, plexiform neurofibroma, and acute leukaemia.28 In addition to pharmacokinetic studies, pharmacodynamic studies including measurement of FTase activity and inhibition of farnesylation of the protein HDJ-2 in leukaemic blasts or peripheral blood mononuclear cells (PBMCs) from patients are incorporated as surrogate markers of the activity of tipifarnib. A phase I clinical trial of the FTase inhibitor SCH66336 is being conducted for paediatric CNS malignancies by the Paediatric Brain Tumour Consortium.
Vascular endothelial growth factor inhibitors
The growth of solid tumours is dependent, in part, on the tumour's ability to induce the formation of new blood vessels through the process of angiogenesis.29,30 Tumours secrete proteins, including basic fibroblast growth factor (bFGF) and vascular endothelial growth factor (VEGF), which activate microvascular endothelial cells to proliferate, migrate, and organize into capillary structures.31 Activated endothelial cells produce cytokines that inhibit cell death (apoptosis) and thus enhance malignant progression.32 Interaction of growth factors and cytokines from these cells is critical in establishing blood supply to a tumour.
Paediatric solid tumours may be potential targets for anti-angiogenic therapy. Childhood solid tumours can be highly vascular and, compared with adult carcinoma, paediatric sarcomas have higher blood flow.33 Additionally, p53 mutations have been noted in some neuroblastomas, Ewing sarcoma, and osteosarcoma, and since wild-type p53 may induce angiogenesis inhibitors, the loss of p53 gene activity could promote vascularization of paediatric solid tumours.34
|
|
|
Fig. 5.2 Farnesyl transferase inhibitors. Ras proteins are activated by receptor tyrosine kinases on the cell surface and initiate signal transduction cascades that are critical to cellular growth regulation. To function, Ras proteins must be associated with the inner surface of the cell membrane. Membrane association is facilitated by the addition of a lipid moiety (farnesyl) to Ras by farnesyl transferase. In normal cells, growth factors bind cell surface receptors causing membrane-bound Ras to switch from an inactive GDP-bound form to an active GTP-bound form initiating the signal transduction cascade. In many tumour cells, Ras is mutated and the Ras protein remains in the active GTP-bound form in the absence of external growth signals. |
Many agents, both old and new, can inhibit angiogenesis in laboratory models. Numerous classes of compounds, including small molecules, cytokines, and antibodies, are currently in clinical trials in adults. Selection of agents for clinical trials in paediatric patients requires careful consideration. Small-molecule inhibitors of angiogenesis, including SU5416 and SU6668, showed promising preclinical activity, but did not produce sufficient activity in early adult clinical trials. Development of these agents was discontinued before clinical trials in the paediatric population could be completed. Other agents, including a thrombospondin-1 mimetic and a thalidomide analogue, are currently in phase I clinical trials in adults, and studies in the paediatric population are being considered.
A paediatric phase I trial of the chimeric monoclonal anti-VEGF antibody bevacizumab (Avastin™) will open soon in the COG Phase I Consortium. Bevacizumab has been shown to selectively inhibit VEGF-mediated tumour angiogenesis in laboratory models. In an adult phase I study, bevacizumab was well tolerated and has subsequently shown activity in phase II studies in adults with colorectal cancer.35,36
In addition to the promising results from early clinical trials in adults, the decision to move bevacizumab into paediatric clinical trials was supported by studies in paediatric xenograft models including anaplastic Wilms tumour,37 neuroblastoma,38 hepatoblastoma, and rhabdomyosarcoma.39 The paediatric trial of bevacizumab will combine traditional phase I objectives (maximum tolerated dose, toxicities, pharmacokinetics) with measures of potential surrogate antiangiogenic markers including total and free serum VEGF, FGF, ICAM-1, V-CAM-1, and thrombospondin-1.
Other novel agents in development for childhood malignancies
Proteosome inhibitors
Phase I trial designs for molecularly targeted drugs can incorporate both traditional toxicity endpoints and novel methods of measuring target inhibition, as illustrated by the development of the proteosome inhibitor bortezomib (PS-341, Velcade2). Proteosomes are cellular organelles that degrade intracellular proteins and regulate the activity of proteins involved in signal transduction, cell cycle regulation, and metastasis. Bortezomib binds the active site on proteosomes, leading to reversible inhibition of this degradative pathway. No assay is available to measure bortezomib concentrations in plasma. However, a sensitive, specific, and reproducible assay measuring proteosome proteolytic activity in whole blood or PBMCs has been developed to measure the percentage proteosome inhibition by bortezomib in clinical trials. This assay has been used to demonstrate a dose-dependent reversible inhibition of proteosome activity in patients' blood or PBMCs.40,41 In May 2003, the FDA approved bortezomib for the treatment of multiple myeloma in adults. A phase I trial of bortezomib has been completed in children. As in the clinical trials in adults, percentage proteosome inhibition was used to determine the recommended dose. Clinical trials to combine bortezomib with cytotoxic agents in children with solid tumours and a phase I study in childhood leukaemia are being planned.
Heat shock protein
The ansamycin antibiotics geldanamycin and herbimycin A target the intracellular chaperone heat shock protein 90 (Hsp90). Hsp90 assists proteins, including kinases (Erb-B2, EGFR, Src family kinases, c-Raf-1 and Cdk-4), steroid hormone receptors, and cell cycle and apoptosis mediators, in maintaining stability and conformation. Geldanamycin prevents these proteins from binding the chaperone by competing for the APT binding site on Hsp90. Therefore these proteins are prone to degradation in the presence of geldanamycin.42 Because of its hepatic toxicity, geldanamycin has limited clinical potential. However, 17-allylamino-17-demethoxygeldanamycin (17AAG), an analogue with reduced liver toxicity, has shown promising clinical activity in adult malignancies and a phase I trial in childhood solid tumours is planned.
Development of drug resistance modifiers in paediatric oncology
Intrinsic or acquired drug resistance poses substantial challenges to treatment in patients with newly diagnosed or recurrent tumours. Our improved understanding of the mechanisms of drug resistance provides possible targets to modulate resistance to chemotherapy.
P-glycoprotein inhibitors
A common mechanism of drug resistance is the multidrug-resistant phenotype (MDR) which is conferred by the presence of P-glycoprotein, a 170-kDa membrane glycoprotein which functions as an energy-dependent drug efflux pump. Chemotherapeutic agents that are substrates for P-glycoprotein include vinca alkaloids, taxanes, anthracyclines, and epipodophyllotoxins. Strategies to block the P-glycoprotein efflux pump in patients have met with limited success, in part due to the use of non-specific competitive inhibitors including cyclosporin or cyclosporin analogues such as valspodar.43 Recently, agents that specifically block P-glycoprotein have entered clinical trials in adults and paediatric patients with refractory cancers. Tariquidar is a potent inhibitor of basal ATPase activity associated with P-glycoprotein, suggesting that its modulatory effect is derived from inhibition of substrate binding and ATP hydrolyis. In vitro tariquidar reverses resistance to doxorubicin, vincristine, and paclitaxel with a potency 10-fold greater than that of the competitive P-glycoprotein inhibitor valspodar. Additionally, tariquidar enhances the cellular accumulation of vinblastine and paxlitaxel in a cell line expressing P-glycoprotein.44
A pharmacodynamic assay to determine the function of the P-glycoprotein drug efflux pump has been developed and incorporated into clinical trials of tariquidar in adults and children.45,46 Lymphocytes (CD56+) which express P-glycoprotein are used as the surrogate tissue. Rhodamine 123 (Rh123), a fluorescent dye and substrate for P-glycoprotein, is added to whole blood collected before and 24 h after the patient receives the P-glycoprotein inhibitor tariquidar. The intracellular accumulation of fluorescent Rh123 in CD56+ lymphocytes is measured using flow cytometry. The percentage inhibition of Rh123 efflux, determined by comparison of the intracellular fluorescence intensity of Rh123 before and after tariquidar administration, increases in a dose-dependent (concentration-dependent) manner.
A phase I trial of tariquidar combined with doxorubicin, vinorelbine, or docetaxel is currently being conducted in paediatric patients with refractory solid tumours. In addition to determining the recommended paediatric dose of tariquidar, the pharmacokinetics of tariquidar alone and in combination with the anticancer agent are being studied, and studies of P-glycoprotein function in surrogate CD56+ lymphocytes and in tumour are being conducted before and after tariquidar administration using [99mTc]sestamibi scanning.
Bcl-2 antisense oligonucleotide
Another strategy being studied in children with cancer is modulation of the expression of the anti-apoptotic protein Bcl-2. Many anticancer agents induce tumour cell death by apoptosis or programmed cell death. The process of apoptosis is regulated by pro-apoptotic and antiapoptotic proteins. Many tumours, including neuroblastoma, Ewing sarcoma, Wilms tumour, and synovial cell sarcoma, overexpress the anti-apoptotic protein Bcl-2.47,48 Bcl-2 antisense (G3139, Genasense™) is an oligodeoxynucleotide designed to bind the first six codons of the human Bcl-2 mRNA. Decreasing the expression of Bcl-2 in tumour cells may increase susceptibility to chemotherapy-induced apoptosis.49 A phase I trial of Bcl-2 antisense in children with relapsed solid tumours, in which Bcl-2 antisense is administered in combination with doxorubicin and cyclophosphamide, is being conducted. The trial aims to determine the dose-limiting toxicities and recommended dose of Bcl-2 antisense, characterize the pharmacokinetics, and assess the biologic activity of Bcl-2 antisense in PBMCs and tumour tissue by determining Bcl-2 and related protein expression.
Conclusion
The principles for rational drug discovery were first demonstrated more than 50 years ago with the development of a number of antimetabolites including methotrexate, 6-MP, and 6-TG. In the past, limitations on our understanding of the malignant process precluded our ability to develop molecularly targeted therapies, and hence we relied predominantly on empiric screening of compounds for non-specific cytotoxic activity. Recently, we have entered an era of anticancer drug development which focuses on the identification of a spectrum of specific targets integral to cellular proliferation and malignant transformation. Certain targets, such as the PDGF-R pathway, may have similar roles in selected adult and paediatric cancers. Other targets, whose functional significance is yet to be identified, may be specific to paediatric malignancies and will present formidable challenges for paediatric drug development.
Target-based chemotherapy holds the promise of selectivity and hope of decreased acute and chronic toxicity. In early clinical trials the aim is to assess the ability of the drug to inhibit the target, determine the relevance of the target to paediatric malignancies, and carefully monitor for toxicities, particularly in agents that may require chronic dosing. For all agents that prove to be active in paediatric malignancies, an additional challenge will be to combine them with currently active regimens to improve overall survival and decrease the toxic side effects of therapy for children with cancer.
References
1. Heinle RW, Welch AD (1948). Experiments with pteroylglutamic acid and pteroylglutamic acid deficiency in human leukemia. J Clin Invest 27, 539 (abstr).
2. Farber S, Diamond LK, Mercer RD, et al. (1948). Temporary remissions in acute leukemia in children producted by folic acid antagonist, 4-aminopteroyl-glutamic acid (aminoptrin). N Engl J Med238, 787–93.
3. Goldin A (1968). Preclinical methodology for the selection of anticancer agents. In: Busch H (ed) Methods in Cancer Research, Vol. IV. New York: Academic Press, 193–254.
4. Elion GB, Hitchings GH (1965). Metabolic basis for the actions of analogs of purines and pyramidines. Adv Chemother 2, 91–177.
5. Boyd M (1996). The NCI in vitro anticancer drug discovery screen: concept, implementation, and operation, 1985–1995. In: Teicher B (ed) AntiCancer Drug Development Guide: Preclinical Screening, Clinical Trials, and Approval. Totowa, NJ: Humana Press, 23–42.
6. Weitman S, Carlson L, Pratt CB (1996). New drug development for pediatric oncology. Invest New Drugs 14, 1–10.
7. Houghton P, Adamson PC, Blaney S, et al. (2002). Testing of new agents in childhood cancer preclinical models: meeting summary. Clin Cancer Res 8, 3646–57.
8. Rowinsky E (1994). Update on the antitumor activity of paclitaxel in clinical trials. Ann Pharmacother 28 (Suppl 5), S18–22.
9. Hurwitz C, Strauss LC, Kepner J, et al. (2001). Paclitaxel for the treatment of progressive or recurrent childhood brain tumors: a Pediatric Oncology Phase II study. J Pediatr Hematol Oncol 23, 277–81.
10. Harris M, Hurwitz C, Sullivan JG, Larsen EC, Pratt CB (1999). Taxol in pediatric solid tumors: a Pediatric Oncology Group (POG) phase II study (POG 9262). Proc Am Soc Clin Oncol 18, abstr 2170.
11. Dumontet C, Sikic B (1999). Mechanisms of action and resistance to antitubulin agents: microtubule dynamics, drug transport and cell death. J Clin Oncol 17, 1061–70.
12. Johnson IS, Armstrong JG, Gorman M, et al. (1967). The vinca alkaloids: a new class of oncolytic agents. Cancer Res 23, 1390.
13. Friedman H, Colvin OM, Ludeman SM, et al. (1986). Experimental chemotherapy for human medulloblastoma. Cancer Res 46, 2827–33.
14. Madden T, Bleyer WA, Hohneker J, Johansen MJ, Wargin W, Reaman G (1995). The pharmacokinetics of vinorelbine (Navelbine) in pediatric cancer patients. Proc Am Soc Clin Oncol 30, abstr. 357
15. Casanova M, Ferrari A, Spreafico F, et al. (2002). Vinorelbine in previously treated advanced childhood sarcomas: evidence of activity in rhabdomyosarcoma. Cancer 94, 3263–8.
16. Cortes JE, Pazdur R (1995). Docetaxel. J Clin Oncol 13, 2643–55.
17. Blaney S, Seibel NL, O'Brien M, et al. (1997). Phase I trial of docetaxel administered as a 1-hour infusion in children with refractory solid tumors: a collaborative Pediatric Branch, National Cancer Institute and Children's Cancer Group trial. J Clin Oncol 15, 1538–43.
18. Seibel N, Blaney SM, O'Brien M, et al. (1999). Phase I trial of docetaxel with filgrastim support in pediatric patients with refractory solid tumors: a collaborative Pediatric Oncology Branch, National Cancer Institute and Children's Cancer Group trial. Clin Cancer Res 5, 733–7.
19. Bollag D, McQueney P, Zhu J, et al. (1995). Epothilones, a new class of microtubule stabilizing agents with a taxol-like mechanism of action. Cancer Res 55, 2325–33.
20. Fox E, Stover E, Widemann B, Fojo T, Balis FM (2002). Cytotoxicity of the novel epothilone B analog, BMS247550 in pediatric solid tumor cell lines and comparison with other tubulin binding agents. Proc Am Assoc Cancer Res, 43, abstr 3916.
21. Koyanagi N, Nagasu T, Fujita M, et al. (1994). In vivo tumor growth inhibition produced by a novel sulfonamide, E7010, against rodent and human tumors. Cancer Res 54, 1702–6.
22. Broach JR, Thorner J (1996). High-throughput screening for drug discovery. Nat Genet 384, 14–16.
23. Fox E, Curt GA, Balis FM (2002). Clinical trial design for target-based therapy. Oncologist 7, 401–9.
24. Workman P (2002). Challenges of PK/PD measurements in modern drug development. Eur J Cancer 38, 2189–93.
25. Druker BJ (2002). STI571 (Gleevec™) as a paradigm for cancer therapy. Trends Mol Med 8 (Suppl), S14–17.
26. Buchdunger E, Cioffi C, Law L, et al. (2000). Abl protein–tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and platelet derived growth factor receptors. J Pharmacol Exp Ther 295, 139–45.
27. Salomon DS, Brandt R, Ciardello F, Normanno N (1995). Epidermal growth factor peptides and their receptors in human malignancies. Crit Rev Oncol Hematol 19, 183–232.
28. Widemann BC, Fox E, Goodspeed W, et al. (2003). Phase I trial of the farnesyl transferase inhibitor (FTI) R115777 in children with refractory leukemias. Proc Am Soc Clin Oncol 22, abstr 3250.
29. Folkman J (1985). Angiogenesis and its inhibitors. Important Adv Oncol 1985, 42–62.
30. Kumar S (1980). Angiogenesis and antiangiogenesis. J Natl Cancer Inst 64, 683–7.
31. Zetter B (1998). Angiogenesis and tumor metastasis. Annu Rev Med 49, 407–24.
32. Rak J, Filmus J, Kerbel RS (1996). Reciprocal paracrine interactions between tumor cells and endothelial cells: the ‘angiogenic progression’ hypothesis. Eur J Cancer 32A, 2438–50.
33. Jain R (1991). Hemodynamic and transport barriers to the treatment of solid tumors. Int J Radiat Biol 60, 85–100.
34. Schweigerer L (1995). Antiangiogenesis as a novel therapeutic concept in pediatric oncology. J Mol Med 73, 497–508.
35. Gordon MS, Margolin K, Talpaz M, et al. (2001). Phase I safety and pharmacokinetic study of recombinant human anti-vascular endothelial growth factor in patients with advanced cancer. J Clin Oncol 19, 843–50.
36. Kabbinavar F, Hurwitz H, Fehrenbacher L, et al. (2003). Phase II, randomized trial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer. J Clin Oncol 21, 60–5.
37. Benson AB, Catalano PJ, Meropol NJ, et al. (2003). Bevacizumab (anti-VEGF) plus FOLFOX4 in previously treated advanced colorectal cancer: an interim toxicity analysis of the Eastern Cooperative Oncology Group Study E3200. Proc Am Soc Clin Oncol 22, abstr 975.
38. Rowe DH, Huang J, Li J, et al. (2000). Supression of primary tumor growth of a mouse model of human neuroblastoma. J Pediatr Surg 35, 977–81.
39. Gerber HP, Kowalski J, Sherman D, Eberhard DA, Ferrara N (2000). Complete inhibition of rhabdomyosarcoma xenograft growth and neovascularization requires blockade of both tumor and host vascular endothelial growth factor. Cancer Res 6, 6253–8.
40. Orlowski RZ, Sinchcombe TE, Mitchell BS, et al. (2002). Phase I trial of the proteosome inhibitor PS341 in patients with refractory hemologic malignancies. J Clin Oncol 20, 4420–7.
41. Aghajanian C, Soignet S, Dizon DS, Pien CS, Adams J, Elliott PJ (2002). A phase I trial of the novel proteosome inhibitor PS341 in advanced solid malignancies. Clin Cancer Res 8, 2505–11.
42. Neckers L, Schulte TW, Mimnaugh E (1999). Geldanamycin as a potential anticancer agent: its molecular target and biochemical activity. Invest New Drugs 17, 361–73.
43. Bradshaw DM, Arceci RJ (1998). Clinical relevance of transmembrane drug efflux as a mechanism of multidrug resistance. J Clin Oncol 16, 3674–90.
44. Martin C, Berridge G, Mistry P, Higgins C, Carlton P, Callaghan R (1999). The molecular interaction of the high affinity reversal agent XR9576 with P-glycoprotein. Br J Pharmacol 128, 403–11.
45. Witherspoon S, Emerson DL, Kerr BM, Lloyd TL, Dalton WS, Wissel PS (1996). Flow cytometric assay of modulation of P-glycoprotein function in whole blood by the multidrug resistance inhibitor GG918. Clin Cancer Res 2, 7–12.
46. Stewart A, Steiner J, Mellows G, Laguda B, Norris D, Bevan P (2000). Phase I trial of XR9576 in healthy volunteers after oral and intravenous administration. Clin Cancer Res 6, 4186–91.
47. Dole M, Nunez G, Merchant AK, et al. (1994). Bcl-2 inhibits chemotherapy induced apoptosis in neuroblastoma. Cancer Res 54, 3253–9.
48. Re GG, Hazen-Martin DJ, El Bahtimi R, et al. (1999). Prognostic significance of Bcl-2 in Wilms' tumor and oncogenic potential of Bcl-XL in rare tumor cases. Int J Cancer 84, 193–200.
49. Reed JC (1995). Bcl-2: prevention of apoptosis as a mechanism of drug resistance. Hematol Oncol Clin North Am 9, 451–73.
