The notable success of therapy for acute lymphoblastic leukemia (ALL) in children, for whom cure rates now exceed 80% to 90%, is even more remarkable when one considers that much of this success has occurred through alteration in drug dosing and schedule rather than the introduction of new agents. In adults, morphologic complete remission rates are now more than 90%. Relapse is common and cure rates, up until recently, have only been in the 30% to 40% range. Despite these poor results, treatment outcomes in all age groups with ALL have improved when one compares results from the early 1980s to the early 2000s. The poorer outcomes in adults are multifactorial, but largely relate to a higher incidence of poor-risk cytogenetic features and lessened ability to tolerate intensive chemotherapy regimens that are often used in pediatric patients.

Patients presenting to adult oncologists traditionally were treated with similarly intensive regimens regardless of whether they were adolescents, young adults, or middle-aged to older individuals. In the past 10 years, multiple retrospective comparisons of the outcomes of adolescents and young adults comparing those of similar age treated on pediatric-intensive regimens versus adult-intensive regimens have been made. Most of these studies have reported that survival outcomes are significantly better in adolescents and young adults (AYA) who receive therapy with pediatric regimens as compared with adult regimens. One of the first reports came from a review of data from the Children’s Cancer Group (CCG) in comparison with data from the Cancer and Leukemia Group B (CALGB). A comparison of 197 CCG patients with 124 CALGB patients showed that although the complete response (CR) rates were identical at 90% between the two groups, in the CCG study there was a 67% overall survival (OS) at 7 years compared with 46% for the CALGB cohort ( P <.001). For the CCG patients, central nervous system (CNS) prophylaxis was more intensive and given earlier in the course; moreover, doses of nonmyelosuppressive drugs such as asparaginase, vincristine, and corticosteroids were much higher in the CCG regimens than in the CALGB regimens. At least 6 other retrospective comparisons have largely demonstrated similar results, although the reader is cautioned that treatments in some of these studies were often quite different and incorporated blood and marrow transplantation (BMT) in different ways. There has been much speculation as to why patients treated on pediatric regimens have improved outcomes, and whether this relates to the higher doses of nonmyelosuppressive drugs, greater adherence to treatment schedules, and/or the greater experience of pediatric hematologists and oncologists in treating this disease.

These reports have led to the development of pediatric-intensive regimens for the treatment of adults. The French GRAALL (Group for Research and Adult Acute Lymphoblastic Leukemia) 2003 Trial was a phase 2 study in which a pediatric-inspired regimen was used to treat patients between the ages of 15 and 60 years. The regimen contained higher doses of prednisone, asparaginase, and vincristine in comparison with the group’s previous adult protocols. With a median follow-up of 42 months, the event-free survival (EFS) and OS were 55% and 60%, respectively. However, patients older than 45 years had a treatment-related death rate of 23% in contrast to a 5% rate in patients younger than 45. EFS in the younger patients was 58% versus 46% in the older patients ( P = .03). Other phase 2 studies have reported similar outcomes.

Randomized trials comparing pediatric-intensive with adult-intensive regimens in adults with ALL have not been reported, and may be logistically difficult to do. An alternative approach to addressing this question is being conducted in the United States through the Intergroup Trial led by CALGB, protocol 10403. This trial is taking patients with Philadelphia chromosome–negative (Ph-neg) up to the age of 40 years and treating them with one arm of a high-risk CCG protocol. The CALGB 10403 protocol will compare the outcomes of patients treated with this regimen with pediatric patients receiving therapy on a designated arm of the CCG protocol. End points for this study include use of traditional end points such as CR rates and survival, but will also assess whether patients and physicians are able to adhere to the scheduling and dosing requirements of the protocol.

Several new approaches in the therapy for ALL, including established agents used in other diseases and new agents in development, represent hope for therapeutic advances in the treatment of ALL and are outlined in detail in this review ( Box 1 ).

Monoclonal antibody therapy

Immunophenotypic analysis of ALL plays a fundamental role in the diagnosis and classification of the disease. It is assuming an increasingly prominent role in the assessment of minimal residual disease in both pediatric and adult patients. A prominent antigen, CD20, is expressed on B-lineage lymphocytes in the mid-stages of development and is found on 40% to 50% of B-lineage ALL, with expression rates rising to 80% to 90% in mature B-cell or Burkitt-type leukemia and lymphoma. In pediatric ALL, expression of CD20 was found to be of conflicting prognostic significance with some studies suggesting inferior EFS, which was independent of other prognostic factors such as age and karyotype. Another study from St Jude Children’s Hospital suggested that expression of CD20 led to better outcomes. In adults, a retrospective analysis of 253 adult patients, of whom 47% expressed CD20 at a level of 20% or more, demonstrated that although CR rates were similar, disease recurrence was higher in the CD20-positive group. These patients had less lymphadenopathy but more thrombocytopenia and a poorer performance score than the CD20-negative patients. Remission duration and OS were poorer in the CD20-positive patients, and CD20 expression remained an independent predictor of outcome in multivariate analysis. The GRAALL Group has also reported that in Ph-negative B-lineage ALL in patients up to the age of 55 years, CD20 expression was associated with an increased risk of relapse (39% vs 20%, P = .02). Overall, it did not significantly affect EFS or OS (55% vs 59% at 42 months and CD20-positive and CD20-negative groups, respectively, P = .65). However, in patients with a white blood cell count of greater than 30 × 10 ⁹ /L, EFS was much worse in the CD20-positive cohort (70% vs 24% at 42 months, P = .006).

Rituximab, which has dramatically changed the therapy for non-Hodgkin lymphoma, has more recently been applied to therapy in CD20-positive B-lineage ALL. In a retrospective study of 282 adolescents and adults with de novo Ph-neg B-lineage ALL received a combination of hyper-CVAD plus rituximab versus a cohort who received hyper-CVAD alone. The CR duration and OS with hyper-CVAD combined with rituximab as compared with hyper-CVAD alone were significantly better. The CR duration was 70% with hyper-CVAD plus rituximab versus 38% with hyper-CVAD alone ( P <.001), and for OS 75% with hyper-CVAD plus rituximab versus 47% for hyper-CVAD alone ( P = .003) ( Fig. 1 ). These results were reported in patients younger than 60 years. Patients older than 60 years did not have any significant differences in outcome, whether or not they received rituximab. In a series of patients between the ages of 15 and 55 years from the German ALL Study Group, 41% of patients had CD20 expression of greater than 20%. In these patients, rituximab was added to the standard combination chemotherapy regimen. The CR rates were similar between the 181 patients who received rituximab plus chemotherapy versus the 82 patients who received chemotherapy alone. Early death rates were also similar. In standard-risk patients, achievement of minimum residual disease (MRD) negativity at day 24 was 57% with rituximab versus 24% without rituximab. Continuous CR at 5 years had a probability of 80% with rituximab versus 47% without rituximab, and overall survival was 71% with rituximab versus 57% without. For conventionally defined high-risk patients, the OS at 3 years was 55% with rituximab and 36% without rituximab. This improvement in outcome was thought to be attributable to fewer relapses in those who received rituximab. Of interest, CD20 expression may actually increase following exposure to corticosteroids, as demonstrated in a group of pediatric ALL patients in whom CD20 expression was upregulated from 45% in initial samples to 81% at the end of induction therapy. This result suggests that additional patients may benefit from rituximab. Rituximab is currently being studied in the therapy for newly diagnosed Ph-neg B-lineage ALL patients in phase 3 randomized trials.

Outcomes by therapy for younger patients (age younger than 60 years) with Philadelphia chromosome–negative precursor B-lineage acute lymphoblastic leukemia. In the CD20-positive subset, ( A ) complete remission (CR) duration and ( B ) survival by inclusion or exclusion of rituximab therapy, and ( C ) survival by hyper-CVAD (fractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone) regimen (standard, modified hyper-CVAD 1 with rituximab inclusive of anthracycline intensification, or modified hyper-CVAD 2 with rituximab eliminating anthracycline intensification) are depicted. ( D ) Survival by regimen (without rituximab) for the CD20-negative group is also depicted.

( From Thomas DA, O’Brien S, Faderl S, et al. Chemoimmunotherapy with a modified hyper-CVAD and rituximab regimen improves outcome in de novo Philadelphia chromosome-negative precursor B-lineage acute lymphoblastic leukemia. J Clin Oncol 2010;28:3880; with permission.)

Another antigen of interest in the therapy for ALL is CD22. This 135-kDa transmembrane sialoglycoprotein, a member of the immunoglobulin superfamily, is present in the cytoplasm of developing B cells. Cell surface expression occurs at a later stage of B-cell development. CD22 expression on B-cell malignancies is high, in the range of 60% to 80%. A CD22 IgG ₁ humanized monoclonal antibody known as epratuzumab has been developed. It rapidly internalizes into the cell on binding to CD22, and appears to stimulate more of an immunomodulatory effect than an antiproliferative effect with rituximab. Epratuzumab has single-agent activity in aggressive B-cell non-Hodgkin lymphoma. It has been combined with rituximab and chemotherapy, demonstrating a high CR rate of 62% and a partial response (PR) rate of 33%, with EFS that appears superior to rituximab and chemotherapy alone.

The Children’s Oncology Group (COG) has evaluated epratuzumab in pediatric ALL patients in a pilot trial. Patients with leukemic blast expression of CD22 of greater than 25% received 4 doses of intravenous epratuzumab, 360 mg/m ² per dose twice weekly for 4 doses and then the same dose weekly for 4 weeks in combination with chemotherapy. All patients were in marrow relapse and, of the 12 evaluable patients who received single-agent therapy, 2 patients had a minimal cytolytic response, 8 had stable disease, and 2 progressed. Nine of the 12 patients achieved a CR after combined epratuzumab and chemotherapy, for which 7 were negative by MRD. The COG is conducting a phase 2 trial of 4 doses of epratuzumab in combination with chemotherapy, and other trials combining epratuzumab with chemotherapy are ongoing.

Other CD22 antibodies of interest include a combination of an antibody that is fused to a 38-kDa fragment of Pseudomonas aeruginosa exotoxin A. This agent is known as CAT-3888. A phase 1 trial of 23 pediatric ALL patients with heavily relapsed disease gave doses ranging from 10 to 40 μg/kg every other day for 3 to 6 doses and repeated every 21 to 28 days. Adverse events were noted to be mild and reversible; and the trial did not demonstrate a maximum tolerated dose (MTD). Objective responses were not noted, but transient clinical activity was found. This immunotoxin has been modified to create higher CD22 binding affinity and is known as CAT-8015 or moxetumomab pasudotox. It has been tested in vitro and found to be highly cytotoxic to B-lineage ALL cells in preclinical studies. A humanized anti-CD22 monoclonal antibody has also been conjugated to calicheamicin, and is known as inotuzumab ozogamicin (CMC544). Studies in patients with refractory non-Hodgkin lymphoma have demonstrated an objective response rate of 87% in patients with follicular lymphoma and 80% in large B-cell lymphoma. In a phase 1 trial in patients with refractory and relapsed ALL of 36 evaluable patients, CD22 was expressed in more than 50% of the blasts in all patients and in more than 90% in 20 patients. Nine of 36 patients achieved a conventional CR, and an additional 11 had a marrow CR for an overall response rate of 56%. The agent was administered as an intravenous infusion over 1 hour every 3 weeks, and the median of courses administered was 2 (range 1–5). Fever and mild hypotension occurred in most patients on the first and second days of therapy. Nine patients or 25% had liver function abnormalities that were thought to be drug related, and these were severe in 4 patients. Two patients had liver biopsies showing periportal fibrosis. Twelve patients were able to proceed to allogeneic BMT, and 3 of these developed sinusoidal obstruction syndrome post transplant. Further trials using the drug in alternative schedules and in combination with other agents are planned.

An interesting approach to treating B-lineage ALL is to combine antibodies in a bispecific fashion. DT22199ARL is made up of two scFv ligands recognizing CD19 and CD22, which are linked to a hybrid gene encoding the first 390 amino acids of the catalytic diphtheria toxin (DT390). This molecule has been enhanced genetically so as to improve the in vivo antileukemic activity via reverse orientation of the VH-VL domains with the addition of linkers that reduce and stabilize aggregation. In an imaging model of a bioluminescent xenograft in which progression of a human Raji-Burkitt lymphoma cell line could be tracked, it was shown that this unique molecule could prevent hindlimb paralysis by blocking metastases to the spinal cord. This agent is given intravenously for 4 doses every other day for 4 hours on an inpatient basis, and is in the midst of phase 1 testing.

The anti-CD52 antibody, alemtuzumab (Campath-1H), is an unconjugated humanized monoclonal antibody. CD52 is expressed on virtually all normal and malignant T and B lymphocytes as well as monocytes and macrophages.

Experience with alemtuzumab in the treatment of ALL has been relatively limited. In a phase 2 study, 6 patients with ALL and 9 patients with relapsed-refractory acute myelogenous leukemia (AML) were treated with 30 mg intravenously 3 times a week for a total of 4 to 12 weeks. All patients demonstrated myelosuppression, but 10 patients progressed while on study, and there were frequent infections.

In a trial assessing the efficacy of alemtuzumab in eradicating MRD in ALL, the CALGB tested sequential cohorts of CD52-positive patients with a dose escalation of alemtuzumab to a target dose of 30 mg administered 3 times a week subcutaneously for 12 doses (4 weeks) during postremission therapy. Twenty-four patients ranging in age from 18 to 77 years were treated. The drug was generally well tolerated with mild hematologic toxicities, but grade 3 to 4 myelosuppression was evident. Two patients had transient viremia with cytomegalovirus. In 11 of 24 patients, serial assessment of MRD demonstrated a median 1-log decrease in disease burden. During subsequent postremission therapy, 8 patients developed cytomegalovirus viremia, 3 had herpes zoster reactivation, and 2 had herpes simplex infections. Further studies with a 30-mg dose are planned.

Another novel approach to using monoclonal antibody therapy in the treatment of ALL is to engage the cytotoxic benefit of T cells by constructing a bispecific T-cell engaging (BiTE) antibody construct. The agent, blinatumomab, combines the variable regions of an anti-CD19 antibody directed against B cells and an anti-CD3 antibody directed against T cells. This agent physically links T cells and malignant B cells, thus triggering the T-cell receptor complex signaling cascade. Recruitment of T cells with subsequent activation occurs when the anti-CD19 antibody binds its target antigen.

A phase 2 trial of blinatumomab at 60 μg/m ² /d given by continuous intravenous infusion for 4 to 8 weeks demonstrated 7 PR and 4 CR for a total of 11 responses in 12 patients.

In the trials with lymphoma, it was noted that in patients treated with 15 μg/m ² /d, dosing had bone marrow clearance of disease. This finding provided the rationale for using this dose by continuous infusion for 4 weeks in patients with B-lineage ALL. In a recently reported trial, patients in complete hematologic remission who had evidence of disease by MRD testing that was either persistent or recurrent were treated. Of 20 evaluable patients, 16 became MRD-negative after one 4-week cycle of therapy. Twelve of these 16 had been molecularly refractory to previous chemotherapy. The most common grade 3 or 4 toxicity was lymphopenia, with an incidence of 33%. Other adverse events included fever, chills, decrease of blood immunoglobulins, and hypokalemia. Most of these were transient. One patient had to permanently discontinue treatment because of a grade 3 seizure; this was reversible. Another patient had syncope with a convulsion. Blinatumomab appears to be a promising therapy in B-lineage ALL, and further trials in patients with MRD positivity, those with morphologic relapse, and those on front-line therapy with chemotherapy are planned.