Search

The efficacy of asparaginase in acute lymphoblastic leukemia (ALL) is dependent on depletion of asparagine, an essential amino acid for ALL cells. The target level of plasma asparaginase activity to achieve asparagine depletion has been between 0.05 and 0.4 IU/mL. COG AALL07P4 examined the asparaginase activity and plasma and CSF asparagine concentration of pegaspargase when given intravenously in the treatment of NCI high risk ALL. Matched plasma asparaginase/asparagine levels of the clearance of 54 doses of pegaspargase given in induction or consolidation demonstrated that all patients who had a plasma asparaginase level >0.02 IU/mL had undetectable plasma asparagine. No difference was observed in CSF asparagine levels associated with matched plasma asparaginase levels of 0.02-0.049 versus 0.05-0.22 IU/mL ( p  = .25). Our data suggest that a plasma asparaginase activity level of 0.02 IU/mL can effectively deplete plasma asparagine. The data also indicate that the 95% CI for plasma asparagine depletion after a pegaspargase dose is 22-29 days. Clinical trial registration: clinicaltrials.gov identifier NCT00671034.

The cure rates for childhood acute leukemia have dramatically improved to approximately 70% overal, with treatments that include intensive cytotoxic chemotherapy and, in some cases, hematopoietic stem cell transplantation. However, many children still die of their disease or of treatment-related toxicities. Even in patients that are cured, there can be significant and, not uncommonly debilitating, acute and late complications of treatment. Improved understanding of the molecular and cellular biology of leukemia and the increasing availability of high-throughput genomic techniques have facilitated the development of molecularly targeted therapies that have the potential to be more effective and less toxic than the standard approaches. In this article, we review the progress to date with agents that are showing promise in the treatment of childhood acute leukemia, including monoclonal antibodies, inhibitors of kinases and other signaling molecules (e.g., BCR-ABL, FLT3, farnesyltransferase, mTOR and γ-secretase), agents that target epigenetic regulation of gene expression (DNA methyltransferase inhibitors and histone deacetylase inhibitors) and proteasome inhibitors. For the specific agents in each of these classes, we summarize the published preclinical data and the clinical trials that have been completed, are in progress or are being planned for children with acute leukemia. Finally, we discuss potential challenges to the success of molecularly targeted therapy, including proper target identification, adequate targeting of leukemia stem cells, developing synergistic and tolerable combinations of agents and designing adequately powered clinical trials to test efficacy in molecularly defined subsets of patients.

The Children's Cancer Group initiated risk-based allocation for childhood acute lymphoblastic leukemia 3 decades ago. Long-term survival data (minimum follow-up >10 years) is now available. About 3711 eligible children were enrolled in risk-adjusted treatment protocols (1983-1989). Ten-year event-free survival (EFS) and overall survival were 62% (standard deviation [SD] = 1%) and 73% (SD = 1%). These data showed a significant improvement (P < 0.0001) compared with the predecessor studies. Since 11% of patients with initial relapses survived without second events, these data predicted a cure rate of 73%. Ten-year EFS and survival were improved significantly for patients with intermediate risk (P < 0.0001), high risk (P < 0.0001) and lymphomatous features (P < 0.0001). Key components of therapies included delayed intensification and substitution of intrathecal chemotherapy for prophylactic/preventive cranial radiation in low- and intermediate-risk patients. This is the largest series of children on concurrent studies who were observed more than 10 years.

Multidrug resistance (MDR) is a phenomenon by which cells become resistant to an array of structurally unrelated chemotherapeutic agents. The prognostic value that P-glycoprotein (Pgp), multidrug resistance-related protein 1 (MRP1), and lung resistance protein (LRP) have in the setting of pediatric acute lymphoblastic leukemia (ALL) is controversial. In a retrospective study, we analyzed samples obtained from 295 similarly treated pediatric ALL patients to assess whether the overexpression and/or function of these proteins at diagnosis affects outcome. Most patients (70%, 207/295) did not overexpress an MDR protein. A small number of patients expressed functional Pgp (1%, 3/295) and some overexpressed functional MRP1 (10%, 19/295), with a statistically significant number of the latter being of T-lineage as opposed to pre-B (P < 0.001). A small number of patients (2%, 6/295) also overexpressed both Pgp and MRP1. Additional patients expressed increased levels of LRP. Elevated levels of these proteins at diagnosis did not correlate with risk factors and did not predict an adverse prognosis. Life-table estimates and Kaplan-Meier plots did not show any significant differences between patients who overexpressed an MDR protein compared with those who did not, nor was any difference noted when the different MDR + groups were compared with one another. These data strongly support the conclusion that the overexpression of these functional drug efflux pumps at diagnosis does not contribute to treatment failure in pediatric ALL.

To determine the maximum tolerated dose (MTD) and assess the toxicity profile and pharmacokinetics of weekly gemcitabine infusions in pediatric patients with refractory hematologic malignancies. Fourteen patients under 21 years old were given infusions of gemcitabine for escalating durations at 10 mg/m2/min weekly for three consecutive weeks. Two males and two females were studied at each dose level. Pharmacokinetics of the drug's metabolism were measured by high pressure-liquid chromatography (HPLC) for 24 h after the first dose. Intracellular difluorodeoxycytidine triphosphate formation in leukemic blasts was measured in selected patients. The MTD of gemcitabine in these patients was 3600 mg/m2/week for three consecutive weeks (10 mg/m2/min for 360 min). Hepatotoxicity was the dose limiting toxicity. Thirty to fifty percent of patients exhibited fever, rash, or myalgia. Rare instances of hypotension and pulmonary toxicity were observed. Two of six patients [one acute lymphoblastic leukemia (ALL) and one acute myelogenous leukemia (AML)] treated at the MTD had at least M2 marrows, although peripheral blood counts did not recover sufficiently for the patients to be considered in complete response. Pharmacokinetics of gemcitabine fit a two-compartment open model with terminal half-life and plasma clearance value of 62 min and 2.2 l/min/m2, respectively. No gender differences were observed. In conclusion, the MTD of gemcitabine was 10 mg/m2/min for 360 min every week for 3 weeks. This is the recommended phase II dose schedule for children with leukemia. The activity of the drug at this schedule in heavily pretreated, refractory patients warrants a phase II trial in hematologic malignancies.

The filamentous fungus Aspergillus nidulans nudC (nuclear distribution C) gene is required for movement of nuclei following mitosis and for normal colony growth. It is highly conserved, structurally and functionally, throughout most of evolution. The human homolog, called HnudC, has been cloned and has an important role in cell proliferation. In hematopoiesis, HNUDC is highly expressed in early hematopoietic precursors and declines during normal differentiation. Stimulation of proliferation of the erythroleukemia cell line TF-1 with GM-CSF enhances HnudC protein and mRNA expression and treatment with antisense (but not sense) oligonucleotides to HnudC mRNA significantly reduces cell division. A significant increase in HNUDC is present in bone marrow aspirates from patients with acute lymphoblastic leukemia (ALL) and acute myelogenous leukemia (AML) compared to the level in normal cellular counterparts, demonstrating dysregulated expression in leukemia. These data support the conclusion that HnudC plays a functional role in promoting hematopoietic cell growth and that it is involved in leukemogenesis.

We used a SCID mouse model of human B-lineage acute lymphoblastic leukemia to examine the antileukemic activity of temozolomide in comparison to as well as in combination with B43-PAP anti-CD19 immunotoxin. One hundred percent of the 20 PBS-treated control mice died of disseminated human B-lineage ALL at 32 to 64 days after the inoculation of 1x10(6) NALM-6 cells, with a median event free survival time of 43 +/- 1 days. Temozolomide, when administered i.p. for 5 consecutive days at a dose level of 411 mg/m2 or as a single 750 mg/m2 bolus dose, elicited significant antileukemic activity and improved survival in this SCID mouse model of human B-lineage ALL. The median survival times were 43 +/- 1 days for PBS-treated mice, 56 +/- 16 days for mice injected with the 5-day temozolomide program, and 64 +/- 15 days for mice treated with a single bolus dose of temozolomide. However, temozolomide was not as effective as B43-PAP. Whereas only 40 +/- 21% of mice treated with temozolomide survived beyond 120 days, B43-PAP treatment resulted in 74 +/- 7% survival in the same model system. The combination of temozolomide with B43-PAP was well tolerated by mice but it was not significantly more effective than B43-PAP alone. Temozolomide may have very limited potential as an antileukemic agent for treatment of B-lineage ALL.

Children with Philadelphia (Ph) chromosome positive (+) acute lymphoblastic leukemia (ALL) represent a subgroup at very high risk for treatment failure. This study included 1322 children enrolled between 1988 and 1994 on CCG risk-adjusted studies for ALL who had centrally reviewed cytogenetic data. Thirty patients had a t(9;22) and are referred to as Ph+; 1292 were Ph-. 23 of these 30 patients were treated on the CCG-1882 high risk ALL protocol. The event-free survival (EFS) outcome in CCG-1882 was significantly worse for Ph+ compared with Ph- patients, with 4-year estimates of 11.3% (SD = 9.8%) and 73.4% (SD = 2.3%), respectively (p < 0.0001).

Contemporary intensive therapies are effective for the majority of pediatric T-lineage acute lymphoblastic leukemia (ALL) patients, thus current challenge is to identify patients who may benefit from alternative treatment modalities. Previously, we demonstrated that human leukemic cell growth in the severe combined immunodeficiency (SCID) mouse was a significant prognostic factor for very high risk B-lineage ALL patients. In the current report we show that primary leukemic cells from 24 of 88 (27%) T-lineage ALL patients (SCID+) caused histopathologically detectable leukemia in SCID mice. These SCID+ patients were similar to SCID- (n = 64) patients with respect to virtually all presenting features, including age, sex, race, and leukocyte count. Growth of primary leukemic cells in SCID mice was not a significant predictor of outcome for the aggregate population of T-lineage ALL patients. Two-year event-free survival (EFS) outcomes for SCID+ patient and SCID- patients were 76.2% (SD = 5.6%) and a 64.0% (SD = 10.4%; p = 0.20). Overall survival also was similar between the two groups (p = 0.36). Among the subset of patients with M1 or M2 marrow status by day 7 of induction chemotherapy (rapid early responders), those who were SCID+ had poorer outcomes than those who were SCID-, with a 2-year EFS of 68.4% (SD = 11.9%) vs. 85.7% (SD = 6.0%) and relative hazard rate of 3.06 (p = 0.06). These data suggest that leukemic cell growth in SCID mice may identify a subset of T-lineage ALL patients who are at higher risk for relapse despite achieving a rapid early response to induction chemotherapy.

The SCID mouse represents a valuable tool for assessing growth characteristics and drug sensitivity of human leukemic cells. We have examined differences in the engraftment patterns in SCID mice of primary human leukemic cells isolated from children (< 21 years old) with either t(1;19)+/E2A-PBX1+ or t(9;22)+/BCR-ABL+ acute lymphoblastic leukemia. Leukemic cells from 13/24 t(1;19)+/E2A-PBX1+ patients caused overt leukemia in SCID mice. Macroscopic lesions were evident in 6/13 cases, with multiple sites involved in some mice: hepatomegaly,(3) splenomegaly(4), thymic enlargement; liver tumors(1), kidney tumors(1), abdominal tumors(1). Microscopic lesions in SCID mouse organs were present in all 13 cases and involved the bone marrow, brain, heart, gut, liver, kidney, lung, ovary, pancreas, skeletal muscle, spleen, and thymus. Leukemic cells from 5/20 t(9;22)+/BCR-ABL+ patients caused overt leukemia in SCID mice. Notably, macroscopic lesions (splenomegaly; leukemic bones; hepatic tumors) were observed in only 1 case. In all 5 cases, microscopic lesions were found in the mouse bone marrow. Additional microscopic lesions were restricted to skeletal muscle, spleen, and mesentery (1 case) or thymus (1 case). These findings differ markedly from those of t(1;19)+/E2A-PBX1+ leukemic cells due to the lack of involvement of major organs such as liver, pancreas, kidney, skin, or brain. These data illustrate the biological heterogeneity of childhood ALL and suggest that the differential risks associated with t(1;19)+/E2A-PBX1+ and t(9;22)+/BCR-ABL ALL might arise from unique engraftment and proliferation capabilities of the respective leukemic cell populations.

Primary leukemic cells isolated from children (N = 681 ) with newly diagnosed B-lineage ALL enrolled on risk-adjusted treatment protocols of the Children's Cancer Group (CCG) were injected via the tail vein into 7-10 week old SCID mice. Leukemic cells from 104 of 681 patients (15.3%) were able to engraft and proliferate in one or more SCID mouse organs. These SCID+ patients were somewhat more likely than SCID patients to be older than 10 years of age (p = 0.03) and have WBC counts >20,000/microL (p = 0.04), but the groups were similar with respect to all other presenting features. Event-free survival (EFS) outcome at 3 years of follow-up was similar for SCID+ patients compared with SCID- patients (79.2%, SD = 5. 1% vs. 84.8%, SD = 2.8%; p = 0.20). Overall survival also was similar between the two groups (p = 0.93). This result was maintained within the subgroups of lower risk (N = 448) and higher risk (N = 233) patients. However, there were trends for poorer outcome among patients whose cells caused overt leukemia in SCID mice and infiltrated either 6 or more organs (p = 0.03), skeletal muscle (p = 0.0003), kidney (p = 0.05), or spleen (p = 0.06). Thus, engraftment of primary leukemic cells in SCID mice was not a significant predictor of outcome for the aggregate population of B-lineage ALL patients, the majority of whom were low risk, treated according to contemporary intensive chemotherapy programs of the CCG. However, development of disseminated overt leukemia and infiltration of SCID mouse skeletal muscle by primary leukemic cells from adjacent bone marrow may reflect a biologically more aggressive disease and identify patients at higher risk for treatment failure.

The establishment of an in vivo animal model system for infant acute lymphoblastic leukemia (ALL) would allow the testing of new agents against primary leukemic cells from infant ALL patients. We have demonstrated previously that growth of B-lineage leukemic cells in mice with severe combined immunodeficiency (SCID) was a significant prognostic factor for children with high risk ALL. We now have examined the significance of this prognostic variable for 13 infants with newly diagnosed ALL treated at participating institutions of the Children's Cancer Group (CCG). Chromosomal translocations were detected in 10/12 evaluated cases, including five with t(4;11), one each with t(7;9) and t(7;11), t(1;19), and t(9;22), and two with t(11;19). Twelve of the thirteen infants with ALL achieved remissions following induction chemotherapy. Primary leukemic cells from 8 of the 13 infants caused overt leukemia in SCID mice. Among these 8 SCID+ infants, 7 were CD10- and seven had cytogenetic or molecular evidence of an 11q23 rearrangement. Six of the 8 SCID+ infants have relapsed; only 2 remain in remission following chemotherapy or bone marrow transplant. However, among the 5 SCID- infants there were also two relapses. These data are suggestive of a poorer outcome for SCID+ infants, but larger numbers of patients must be analyzed to assess their statistical significance. In summary, we have established a SCID mouse model for human infant ALL that will be useful for 1) predicting short-term and long-term outcome of patients, 2) testing pharmacokinetics, efficacy, and toxicity of new agents, and 3) elucidating the in vivo mechanisms of chemotherapeutic drug resistance in infant ALL.

The quinazoline antifolate N-(5-[N-(3,4-dihydro-2-methyl-4-oxoquinazolin-6-ylmethyl)-N- methylamino]-2-thenoyl)-L-glutamic acid (ZD1694; Tomudex) is a potent inhibitor of thymidylate synthase and causes cell death through disruption of DNA synthesis and repair by blocking the obligatory thymidine nucleotide synthesis. B43(anti-CD19)-PAP immunotoxin is a potent inhibitor of protein synthesis in CD19+ B-lineage acute lymphoblastic leukemia (ALL) cells and causes apoptosis. In this model, 100% of SCID mice challenged with 1 x 10(6) human NALM-6 B-lineage ALL cells develop overt and invariably fatal leukemia. All of the 22 control SCID mice treated with phosphate-buffered saline died of disseminated human leukemia between 31 and 61 days with a median survival of 41.2 days. Treatment with ZD 1694 resulted in improved leukemia-free survival with a median survival of 69.2 days (P < 0.001, log-rank test). B43-PAP treatment was more effective than ZD1694 (P=0.026) and resulted in 51.0% long-term leukemia-free survival with a median survival of 187.5 days (P < 0.0001. log-rank test). The combination of ZD1694 and B43-PAP was more effective than either agent alone and resulted in 100% long-term leukemia-free survival. To our knowledge, this preclinical study is the first to demonstrate the feasibility and therapeutic advantage of combining an anti-leukemia immunotoxin with a thymidylate synthase inhibitor.

We previously have reported on an experimental athymic mouse model in which regression of human Burkitt's lymphoma is induced by either coinjection with or intratumor inoculation of Epstein-Barr virus (EBV)-immortalized human B cells. In the current study, we were interested in determining whether the powerful antitumor effects of EBV-immortalized B cells could be effective against a variety of human tumors grown in athymic mice, including acute lymphocytic leukemia, malignant melanoma, acute promyelocytic leukemia, neuroblastoma, lung carcinoma, colon adenocarcinoma, Wilms tumor, Hodgkin's lymphoma, rhabdomyosarcoma and breast adenocarcinoma. We report here the results of experiments in nude mice that demonstrated the potent antitumor effect of EBV-immortalized B cells against human tumors derived from a variety of different tissues.

Less than fifteen years have passed since the practical applications of antibody-directed cell targeting were conceived in the laboratory. In vitro and in vivo applications for immunotoxins are being developed in a variety of protocols to treat human cancer. Immunotoxins may be particularly well-suited for the treatment of hematological cancers where malignant cells are more accessible than in solid tumors. Experimental results and clinical responses have provided encouragement, especially in light of the complexities of cancer biology, while the problems do not appear insurmountable. Continued basic research in cell biology combined with advances in manufacturing techniques will undoubtedly propose improvements in immunotoxins as well as new applications.