Childhood Acute Lymphoblastic Leukemia, part 2

The COG AALL03N1 (NCT00268528) trial studied the impact of nonadherence to mercaptopurine during maintenance therapy in 327 children and adolescents (169 Hispanic patients and 158 non-Hispanic White patients).[99]

Sources: AALL03N1 (NCT00268528) · 99

The COG AALL03N1 (NCT00268528) trial studied the impact of nonadherence to mercaptopurine during maintenance therapy in 327 children and adolescents (169 Hispanic patients and 158 non-Hispanic White patients).[99]

A progressive increase in relapse was observed with decreasing adherence to mercaptopurine, with HRs ranging between 4.0% to 5.7% for adherence rates ranging from 94.9% to 90%, 89.9% to 85%, and less than 85%. After adjusting for other prognostic factors (including NCI risk group and chromosomal abnormalities), a progressive increase in relapse was observed with decreasing adherence to mercaptopurine. MRD data were unavailable in this study population, so they were not included in the analysis of prognostic factors.

Adherence was significantly lower among Hispanic patients, patients older than 12 years, and patients from single-mother households. However, among adherent patients, Hispanic ethnicity remained an independent predictor of adverse outcome.

The AALL03N1 trial also included 71 Asian American patients and 68 African American patients, in addition to the non-Hispanic White and Hispanic patients described above.[100]

Sources: 100

The AALL03N1 trial also included 71 Asian American patients and 68 African American patients, in addition to the non-Hispanic White and Hispanic patients described above.[100]

Using an adherence rate of less than 90% to define nonadherence, 20.5% of the participants were not adherent.

An adherence rate of less than 90% was associated with increased relapse risk (HR, 3.9).

Adherence rates were significantly lower in Asian American and African American patients than in non-Hispanic White patients.

In a third publication from the AALL03N1 study, the following key observations were made:[130]

Sources: 130

In a third publication from the AALL03N1 study, the following key observations were made:[130]

Patients with mercaptopurine nonadherence (defined as mean adherence rate of <95%) were at a 2.7-fold increased risk of relapse compared with adherers.

Among adherers, high intra-individual variability in thioguanine levels (due to varying dose-intensity and drug treatment interruptions) was associated with increased risk of relapse.

The authors of the above studies also found that self-reporting was not a reliable measure of adherence, with 84% of patients overreporting compliance with taking mercaptopurine at least some of the time.[131] The data suggest that additional measures of adherence besides self-reporting are needed.

Sources: 131

Another publication from the AALL03N1 study described mercaptopurine ingestion habits, red cell thioguanine nucleotide (TGN) levels, adherence, and relapse risk.[103]

Sources: 103

Another publication from the AALL03N1 study described mercaptopurine ingestion habits, red cell thioguanine nucleotide (TGN) levels, adherence, and relapse risk.[103]

The findings showed that certain ingestion habits (e.g., taking with dairy and taking at varying times throughout the day) were associated with nonadherence. However, after adjusting for adherence and other prognostic factors, ingestion habits were not associated with relapse risk.

For adherent patients, there was no association between TGN levels and ingestion habits.

The authors concluded that commonly practiced restrictions surrounding mercaptopurine ingestion do not appear to impact outcome but may hinder adherence.

Intrathecal chemotherapy.

Sources: Intrathecal chemotherapy

CNS-directed systemic chemotherapy.

Sources: CNS-directed systemic chemotherapy

Cranial radiation therapy.

Sources: Cranial radiation therapy

Patients with 5 or more WBC/µL and blasts in the cerebrospinal fluid (CSF) (CNS3), obtained at diagnosis.

Patients with blasts in the CSF but fewer than 5 WBC/µL (CNS2) may be at increased risk of CNS relapse,[2] although this risk appears to be nearly fully abrogated if they receive more doses of intrathecal chemotherapy, especially during the induction phase.[3]

Sources: 2 · 3

Patients with T-ALL, especially those with high presenting peripheral blood leukocyte counts.

Patients who have a traumatic lumbar puncture showing blasts at the time of diagnosis may have an increased risk of CNS relapse. These patients receive more intensive CNS-directed therapy on some treatment protocols.[3,4]

Sources: 3 · 4

Methotrexate alone.

Methotrexate with cytarabine and hydrocortisone (triple intrathecal chemotherapy).

Dexamethasone.

L-asparaginase (does not penetrate into CSF itself, but leads to CSF asparagine depletion).[7]

Sources: 7

High-dose methotrexate with leucovorin rescue.

Escalating dose intravenous (IV) methotrexate without leucovorin rescue.

In a randomized Children's Cancer Group (CCG) study of standard-risk patients who all received the same dose and schedule of intrathecal methotrexate without cranial irradiation, oral dexamethasone was associated with a 50% decrease in the rate of CNS relapse compared with oral prednisone.[8]

Sources: 8

In another standard-risk ALL trial (COG-1991), escalating dose IV methotrexate without leucovorin rescue significantly reduced the CNS relapse rate compared with standard, low-dose, oral methotrexate given during each of two interim maintenance phases.[9]

Sources: COG-1991 · 9

In a randomized clinical trial conducted by the former Pediatric Oncology Group, T-ALL patients who received high-dose methotrexate experienced a significantly lower CNS relapse rate than did patients who did not receive high-dose methotrexate.[10]

Sources: 10

The CCG-1952 study for National Cancer Institute (NCI) standard-risk patients compared the relative efficacy and toxicity of triple intrathecal chemotherapy (methotrexate, cytarabine, and hydrocortisone) with methotrexate as the sole intrathecal agent in nonirradiated patients.[21]

Sources: CCG-1952 · 21

The CCG-1952 study for National Cancer Institute (NCI) standard-risk patients compared the relative efficacy and toxicity of triple intrathecal chemotherapy (methotrexate, cytarabine, and hydrocortisone) with methotrexate as the sole intrathecal agent in nonirradiated patients.[21]

There was no significant difference in either CNS or non-CNS toxicities.

Although triple intrathecal chemotherapy was associated with a lower rate of isolated CNS relapse (3.4% ± 1.0% compared with 5.9% ± 1.2% for intrathecal methotrexate; P = .004), there was no difference in event-free survival (EFS) rates.

The CCG-1952 study for National Cancer Institute (NCI) standard-risk patients compared the relative efficacy and toxicity of triple intrathecal chemotherapy (methotrexate, cytarabine, and hydrocortisone) with methotrexate as the sole intrathecal agent in nonirradiated patients.[21] › Although triple intrathecal chemotherapy was associated with a lower rate of isolated CNS relapse (3.4% ± 1.0% compared with 5.9% ± 1.2% for intrathecal methotrexate; P = .004), there was no difference in event-free survival (EFS) rates.

The reduction in CNS relapse rate was especially notable in patients with CNS2 status at diagnosis (lymphoblasts seen in CSF cytospin, but with <5 WBC/high-power field on CSF cell count). The isolated CNS relapse rate was 7.7% (± 5.3%) for CNS2 patients who received triple intrathecal chemotherapy, compared with 23.0% (± 9.5%) for those who received intrathecal methotrexate alone (P = .04).

There were more bone marrow relapses in the group that received the triple intrathecal chemotherapy, leading to a worse overall survival (OS) rate (90.3% ± 1.5%) compared with the intrathecal methotrexate group (94.4% ± 1.1%; P = .01).

When the analysis was restricted to patients with B-ALL and rapid early response (M1 marrow on day 14), there was no difference between triple and single intrathecal chemotherapy in terms of rates of CNS relapse, OS, or EFS rates.

The findings of this trial need to be interpreted within the context of other therapy administered to patients. Dexamethasone, which has been associated with lower CNS relapse rates and improved EFS rates in standard-risk patients in other trials,[8,22] was not used in CCG-1952 (prednisone was the only steroid administered to patients).[23] It is not clear whether the results of the CCG-1952 trial are generalizable to protocols that include the use of dexamethasone and/or other CNS-directed systemic therapies.

Sources: 8 · 22 · 23

The CCG-1952 study for National Cancer Institute (NCI) standard-risk patients compared the relative efficacy and toxicity of triple intrathecal chemotherapy (methotrexate, cytarabine, and hydrocortisone) with methotrexate as the sole intrathecal agent in nonirradiated patients.[21]

In a follow-up study of neurocognitive functioning in the two groups, there were no clinically significant differences.[24][Level of evidence A3]

Sources: 24 · Level of evidence A3

The COG AALL1131 (NCT02883049) study for NCI high-risk patients and NCI standard-risk B-ALL patients with slow early response (defined by day 8 peripheral blood MRD and/or day 29 marrow MRD) randomly assigned patients aged 1 to 30 years to receive either postinduction intrathecal methotrexate or triple intrathecal chemotherapy (methotrexate, cytarabine, and hydrocortisone). Patients with CNS3 disease were not eligible, and patients on this trial did not receive cranial radiation. Postinduction intrathecal therapy was administered for a total of 21 to 26 doses. Neurocognitive assessments were performed in a subset of patients aged 6 to 12 years at initial diagnosis.[25]

Sources: AALL1131 (NCT02883049) · 25

The COG AALL1131 (NCT02883049) study for NCI high-risk patients and NCI standard-risk B-ALL patients with slow early response (defined by day 8 peripheral blood MRD and/or day 29 marrow MRD) randomly assigned patients aged 1 to 30 years to receive either postinduction intrathecal methotrexate or triple intrathecal chemotherapy (methotrexate, cytarabine, and hydrocortisone). Patients with CNS3 disease were not eligible, and patients on this trial did not receive cranial radiation. Postinduction intrathecal therapy was administered for a total of 21 to 26 doses. Neurocognitive assessments were performed in a subset of patients aged 6 to 12 years at initial diagnosis.[25]

The 5-year postinduction disease-free survival (DFS) rates were 93.2% (± 2.1%) for patients randomly assigned to intrathecal methotrexate and 90.6% (± 2.3%) (P = .85) for patients assigned to triple intrathecal chemotherapy.

The OS rates were 96.3% (± 1.5%) for patients who received intrathecal methotrexate and 96.7% (± 1.4%) (P = .77) for patients who received triple intrathecal chemotherapy.

There were no differences in the cumulative incidence of isolated bone marrow relapse, isolated CNS relapse, or combined bone marrow and CNS relapse between the two arms.

There were no significant differences in neurological toxicities or in assessments of neurocognitive functioning during treatment for patients who received intrathecal methotrexate compared with patients who received triple intrathecal chemotherapy.

Patients with T-cell phenotype and high initial WBC count.

Patients with high-risk B-ALL and extremely high presenting leukocyte counts and/or adverse cytogenetic abnormalities.

In a trial conducted between 1990 and 1995, the Berlin-Frankfurt-Münster (BFM) group demonstrated that a reduced dose of prophylactic radiation (12 Gy instead of 18 Gy) provided effective CNS prophylaxis in high-risk patients.[26]

Sources: 26

In the follow-up trial conducted by the BFM group between 1995 and 2000 (BFM-95), cranial radiation therapy was administered to approximately 20% of patients (compared with 70% on the previous trial), including patients with T-cell phenotype, a slow early response (as measured by peripheral blood blast count after a 1-week steroid prophase), and/or adverse cytogenetic abnormalities.[18]

Sources: 18

In the follow-up trial conducted by the BFM group between 1995 and 2000 (BFM-95), cranial radiation therapy was administered to approximately 20% of patients (compared with 70% on the previous trial), including patients with T-cell phenotype, a slow early response (as measured by peripheral blood blast count after a 1-week steroid prophase), and/or adverse cytogenetic abnormalities.[18]

While the rate of isolated CNS relapses was higher in the nonirradiated higher-risk patients compared with historical (irradiated) cohorts, their overall EFS rates were not significantly different.

Several groups, including the St. Jude Children's Research Hospital (SJCRH), the Dutch Childhood Oncology Group (DCOG), and the European Organization for Research and Treatment of Cancer (EORTC), have published results of trials that omitted cranial radiation therapy for all patients, including high-risk subsets.[12,13,27] Most of these trials have included at least four doses of high-dose methotrexate during postinduction consolidation and an increased frequency of intrathecal chemotherapy. The SJCRH and DCOG studies also included frequent vincristine/dexamethasone pulses and intensified dosing of pegaspargase,[12,13] while the EORTC trials included additional high-dose methotrexate and multiple doses of high-dose cytarabine during postinduction treatment phases for patients with CNS3 status (CSF with ≥5 WBC/µL and cytospin positive for blasts).[27]

Sources: 12 · 13 · 27 · 12 · 13 · 27

Several groups, including the St. Jude Children's Research Hospital (SJCRH), the Dutch Childhood Oncology Group (DCOG), and the European Organization for Research and Treatment of Cancer (EORTC), have published results of trials that omitted cranial radiation therapy for all patients, including high-risk subsets.[12,13,27] Most of these trials have included at least four doses of high-dose methotrexate during postinduction consolidation and an increased frequency of intrathecal chemotherapy. The SJCRH and DCOG studies also included frequent vincristine/dexamethasone pulses and intensified dosing of pegaspargase,[12,13] while the EORTC trials included additional high-dose methotrexate and multiple doses of high-dose cytarabine during postinduction treatment phases for patients with CNS3 status (CSF with ≥5 WBC/µL and cytospin positive for blasts).[27]

The 5-year cumulative incidence of isolated CNS relapse on those trials was between 2% and 4%, although some patient subsets had a significantly higher rate of CNS relapse. On the SJCRH study, clinical features associated with a significantly higher risk of isolated CNS relapse included T-cell phenotype, the t(1;19) translocation, and the presence of blasts in the CSF at diagnosis.[12]

Sources: 12

The overall EFS rate was 85.6% for the SJCRH study and 81% for the DCOG study—both in line with outcomes achieved by contemporaneously conducted clinical trials on which some patients received prophylactic radiation therapy—but was lower on the EORTC trial (8-year EFS rate, 69.6%).[27]

Sources: 27

On the SJCRH study, 33 of 498 patients (6.6%) in first remission with high-risk features (including 26 patients with high minimal residual disease [MRD], 6 with BCR::ABL1-positive ALL, and 1 with near haploidy) received an allogeneic hematopoietic stem cell transplant, which included total-body irradiation.[12]

Sources: 12

In a meta-analysis of aggregated data from more than 16,000 patients treated between 1996 and 2007 by ten cooperative groups, the use of cranial radiation therapy did not appear to impact 5-year OS or cumulative incidence of any event.[14]

Sources: 14

In a meta-analysis of aggregated data from more than 16,000 patients treated between 1996 and 2007 by ten cooperative groups, the use of cranial radiation therapy did not appear to impact 5-year OS or cumulative incidence of any event.[14]

In subgroup analyses of high-risk subsets, only those with CNS3 status at diagnosis appeared to benefit from cranial radiation, with a significantly lower rate of CNS relapses (isolated/any) in irradiated patients. However, even within this subgroup, OS rates were similar with or without the use of radiation therapy.

This study suggests that cranial radiation therapy may not be an essential component of treatment, even for high-risk patients. However, interpretation is limited by the considerable variation in treatment administered to patients by the different cooperative groups.

The EORTC-58832 trial that was conducted between 1983 and 1989 included patients with medium-risk and high-risk ALL. Patients were randomly assigned to receive or not to receive cranial radiation after intensification and before maintenance therapy.[28][Level of evidence A1]

Sources: 28 · Level of evidence A1

The EORTC-58832 trial that was conducted between 1983 and 1989 included patients with medium-risk and high-risk ALL. Patients were randomly assigned to receive or not to receive cranial radiation after intensification and before maintenance therapy.[28][Level of evidence A1]

The 25-year EFS and OS rates in the two arms of the trial were similar: the EFS rate was 59.5% and the OS rate was 78.1% for patients who did not receive cranial radiation; the EFS rate was 60.5% and the OS rate was 78.5% for patients who received cranial radiation.

There was a dramatic decrease in the rate of late CNS adverse events in the patients who did not receive cranial radiation therapy.

The incidence of second neoplasms was also decreased in patients who did not receive radiation (7.3%) compared with patients who did receive radiation (13%) (HR, 0.43). Meningiomas accounted for the increased incidence of second neoplasms in the cranial radiation group.

The SJCRH, DCOG, and the EORTC have published results of trials that omitted cranial radiation therapy for all patients, including high-risk subsets.[12,27] These trials have included at least four doses of high-dose methotrexate during postinduction consolidation and an increased frequency of intrathecal chemotherapy. The SJCRH study also included higher cumulative doses of anthracycline than on Children’s Oncology Group (COG) trials, and frequent vincristine/dexamethasone pulses and intensified dosing of pegaspargase,[12] while the EORTC trials included additional high-dose methotrexate and multiple doses of high-dose cytarabine, during postinduction treatment phases for CNS3 (CSF with ≥5 WBC/µL and cytospin positive for blasts) patients.[27]

Sources: 12 · 27 · 12 · 27

The SJCRH, DCOG, and the EORTC have published results of trials that omitted cranial radiation therapy for all patients, including high-risk subsets.[12,27] These trials have included at least four doses of high-dose methotrexate during postinduction consolidation and an increased frequency of intrathecal chemotherapy. The SJCRH study also included higher cumulative doses of anthracycline than on Children’s Oncology Group (COG) trials, and frequent vincristine/dexamethasone pulses and intensified dosing of pegaspargase,[12] while the EORTC trials included additional high-dose methotrexate and multiple doses of high-dose cytarabine, during postinduction treatment phases for CNS3 (CSF with ≥5 WBC/µL and cytospin positive for blasts) patients.[27]

On the SJCRH Total XV (TOTXV) study, patients with CNS3 status (N = 9) were treated without cranial radiation therapy (observed 5-year EFS rate, 43% ± 23%; OS rate, 71% ± 22%).[12] On this study, CNS leukemia at diagnosis (defined as CNS3 status or traumatic lumbar puncture with blasts) was an independent predictor of inferior EFS.

Sources: Total XV (TOTXV) · 12

On the DCOG-9 trial, the 5-year EFS rate of CNS3 patients (n = 21) treated without cranial radiation therapy was 67% (± 10%).[13]

Sources: 13

On the EORTC trial, the 8-year EFS rate of CNS3 patients (n = 49) treated without cranial radiation therapy was 68%. The cumulative incidence of isolated CNS relapse for those patients was 9.4%.[27][Level of evidence B4]

Sources: 27 · Level of evidence B4

A meta-analysis of aggregated data from more than 16,000 patients treated between 1996 and 2007 by ten cooperative groups evaluated whether the use of cranial radiation therapy affected outcome in high-risk patient subsets.[14]

Sources: 14

A meta-analysis of aggregated data from more than 16,000 patients treated between 1996 and 2007 by ten cooperative groups evaluated whether the use of cranial radiation therapy affected outcome in high-risk patient subsets.[14]

In subgroup analyses of high-risk subsets, only those with CNS3 status at diagnosis appeared to benefit from cranial radiation therapy, with a significantly lower rate of CNS relapses (isolated/any) in irradiated patients. However, even within this subgroup, OS rates were similar with or without the use of radiation therapy.

A SJCRH study of 567 adult long-term survivors of childhood ALL underwent neurocognitive testing (mean time from diagnosis, 26 years).[40]

Sources: 40

A SJCRH study of 567 adult long-term survivors of childhood ALL underwent neurocognitive testing (mean time from diagnosis, 26 years).[40]

Patients treated with 24 Gy of cranial radiation therapy showed the highest rates of impairment. Up to one-third of these patients demonstrated impairments (defined as test scores 2 or more standard deviations below age-adjusted national norms) in attention, memory, processing speed, and executive function.

Significantly fewer patients who had received 18 Gy of cranial radiation therapy demonstrated severe impairments compared with those who had received 24 Gy. In general, there was no significant difference in rates of impairment between nonirradiated survivors and those who received 18 Gy of cranial radiation therapy. However, the 18-Gy group was at increased risk of academic problems.

In addition to being dose-related, the neurocognitive impact of cranial radiation therapy was also dependent on age at diagnosis, with higher frequency of impairments in patients diagnosed at a younger age.

A study compared memory impairment in patients who received 18 Gy of cranial radiation therapy (n = 127) versus 24 Gy of cranial radiation therapy (n = 138).[45]

Sources: 45

A study compared memory impairment in patients who received 18 Gy of cranial radiation therapy (n = 127) versus 24 Gy of cranial radiation therapy (n = 138).[45]

Long-term survivors who received 24 Gy of cranial radiation therapy demonstrated significant impairments in immediate and delayed memory, compared with survivors who received 18 Gy.

In a randomized trial comparing irradiated (at a dose of 18 Gy) and nonirradiated standard-risk ALL patients, the following was observed: [34][Level of evidence A3]

Sources: 34 · Level of evidence A3

In a randomized trial comparing irradiated (at a dose of 18 Gy) and nonirradiated standard-risk ALL patients, the following was observed: [34][Level of evidence A3]

Cognitive function for both groups (assessed at a median of 6 years postdiagnosis) was in the average range, with only subtle differences noted between the groups in cognitive skills.

In a randomized trial, hyperfractionated radiation therapy (at a dose of 18 Gy) did not decrease neurological late effects when compared with conventionally fractionated radiation therapy. Cognitive function for both groups was not significantly impaired.[46]

Sources: 46

In the SJCRH long-term follow-up study of 567 adult long-term survivors, some nonirradiated patients also demonstrated neurocognitive impairments.[40]

Sources: 40

In the SJCRH long-term follow-up study of 567 adult long-term survivors, some nonirradiated patients also demonstrated neurocognitive impairments.[40]

The age-adjusted mean test scores for nonirradiated patients were very similar to that of expected national norms. However, approximately 15% of the nonirradiated survivors participating in this study demonstrated impairments in some domains, including attention, memory, processing speed, and executive function.

Despite the impairments noted on neurocognitive testing, overall, the educational attainment and employment status of the tested ALL survivors were similar to age- and sex-adjusted expected proportions using census data for the U.S. population.

In a second study from SJCRH, patients enrolled on Total Study XV (which omitted cranial radiation therapy in all patients) underwent comprehensive neuropsychological assessments at induction, end of maintenance, and 2 years after completion of therapy.[47]

Sources: 47

In a second study from SJCRH, patients enrolled on Total Study XV (which omitted cranial radiation therapy in all patients) underwent comprehensive neuropsychological assessments at induction, end of maintenance, and 2 years after completion of therapy.[47]

Neurocognitive function was largely age appropriate 2 years after completing therapy, without evidence of excess impairment on measures of intellectual functioning, academic abilities, learning, and memory. Problems with sustained attention were observed at an increased frequency in this population compared with normative expectations.

High-risk patients who received more intensive CNS-directed chemotherapy (including high-dose methotrexate and more doses of intrathecal chemotherapy) were at greater risk of difficulties in attention, processing speed, and academics.

In a subsequent study from the SJCRH, 400 patients in the Total XVI study (enrolled between 2007 and 2017) underwent neurocognitive testing after the completion of therapy (approximately 3 years from diagnosis). None of the patients received cranial radiation, but the number of doses of triple intrathecal therapy (ITT) chemotherapy varied by risk group and CNS status. Patients with low-risk ALL were prescribed 13 to 21 doses, while patients with standard- and high-risk ALL were prescribed 16 to 27 doses.[48]

Sources: 48

In a subsequent study from the SJCRH, 400 patients in the Total XVI study (enrolled between 2007 and 2017) underwent neurocognitive testing after the completion of therapy (approximately 3 years from diagnosis). None of the patients received cranial radiation, but the number of doses of triple intrathecal therapy (ITT) chemotherapy varied by risk group and CNS status. Patients with low-risk ALL were prescribed 13 to 21 doses, while patients with standard- and high-risk ALL were prescribed 16 to 27 doses.[48]

Compared with age-normative expectations, the entire patient cohort exhibited neurocognitive impairments in multiple domains, including attention, working memory, executive function, fine motor skills, and adaptive skills.

For patients with low-risk ALL, there was no correlation between the number of ITT doses (<20 vs. >21) and neurocognitive measures.

For patients with standard- and high-risk ALL, those who received the highest number of ITT doses (>27) had an increased risk of impairment than those who received fewer doses, especially in working memory, attention, and fine motor skills.

Patients with public or no insurance had worse neurocognitive outcomes than those with private insurance.

Among patients with standard- and high-risk ALL, boys had a higher risk of neurocognitive impairments than girls, as did patients who were younger at diagnosis.

The clinical manifestations of the neurocognitive impairments (if any) found by end-of-treatment neurocognitive testing, and their trajectories over time, were not evaluated in this study.

In the COG AALL06N1 (NCT00437060) trial, patients with high-risk B-ALL who had been enrolled in the AALL0232 trial received a neurocognitive evaluation. These patients had been randomly assigned to receive either high-dose methotrexate with leucovorin rescue or escalating-dose IV methotrexate with pegasparaginase during the interim maintenance phase of therapy in the AALL0232 trial. A neurocognitive evaluation, which included an assessment of intellectual function (estimated intelligence quotient [IQ]), working memory, and processing speed, was conducted after the completion of therapy.[49]

Sources: AALL06N1 (NCT00437060) · 49

In the COG AALL06N1 (NCT00437060) trial, patients with high-risk B-ALL who had been enrolled in the AALL0232 trial received a neurocognitive evaluation. These patients had been randomly assigned to receive either high-dose methotrexate with leucovorin rescue or escalating-dose IV methotrexate with pegasparaginase during the interim maintenance phase of therapy in the AALL0232 trial. A neurocognitive evaluation, which included an assessment of intellectual function (estimated intelligence quotient [IQ]), working memory, and processing speed, was conducted after the completion of therapy.[49]

The method of methotrexate delivery was unrelated to differences in neurocognitive outcomes after controlling for ethnicity, race, age, gender, insurance status, and time off treatment.

Survivors younger than 10 years at diagnosis had a statistically significantly lower estimated IQ and processing speed scores than older patients.

Additionally, patients with public health insurance had significantly lower estimated IQ scores than participants with U.S. private or military insurance.

Chemotherapy with or without prophylactic cranial radiation therapy.

Protocols of the former Pediatric Oncology Group (POG) treated children with T-ALL differently from children with B-ALL. On the POG-9404 protocol, patients with T-ALL were treated using a regimen based on the DFCI 87-001 protocol, which included multiple doses of L-asparaginase and doxorubicin during postinduction therapy, as well as prophylactic cranial radiation therapy. Patients were randomly assigned to either the high-dose methotrexate arm or the control arm.[4]

Sources: POG-9404 · 4

Protocols of the former Pediatric Oncology Group (POG) treated children with T-ALL differently from children with B-ALL. On the POG-9404 protocol, patients with T-ALL were treated using a regimen based on the DFCI 87-001 protocol, which included multiple doses of L-asparaginase and doxorubicin during postinduction therapy, as well as prophylactic cranial radiation therapy. Patients were randomly assigned to either the high-dose methotrexate arm or the control arm.[4]

Patients randomly assigned to receive high-dose methotrexate achieved a superior outcome (10-year EFS rates were 78% with high-dose methotrexate vs. 68% for the control arm).[5,6]

Sources: 5 · 6

In the POG-9404 study, patients were also randomly assigned to receive doxorubicin with or without dexrazoxane to determine the efficacy of dexrazoxane in preventing late cardiac mortality.[7][Level of evidence B1]

Sources: 7 · Level of evidence B1

In the POG-9404 study, patients were also randomly assigned to receive doxorubicin with or without dexrazoxane to determine the efficacy of dexrazoxane in preventing late cardiac mortality.[7][Level of evidence B1]

There was no difference in EFS between patients with T-ALL who were treated with dexrazoxane and patients who were not treated with dexrazoxane (cumulative doxorubicin dose, 360 mg/m2).[7]

Sources: 7

The frequency of grade 3 and grade 4 toxicities that occurred during therapy was similar between the randomized groups, and there was no difference in cumulative incidence of second malignant neoplasms. Three years after initial diagnosis, left ventricular shortening fraction and left ventricular wall thickness were both significantly worse in patients who received doxorubicin alone than in patients who received dexrazoxane, indicating that dexrazoxane was cardioprotective.[7]

Sources: 7

With combined data from three COG trials that randomized dexrazoxane with doxorubicin therapy (P9404, P9425, and P9426) and had a median follow-up of 12.6 years, dexrazoxane did not appear to compromise long-term survival.[8][Level of evidence A1]

Sources: 8 · Level of evidence A1

On protocols of the former Children’s Cancer Group (CCG), children with T-ALL received the same treatment regimens as did children with B-ALL. Protocol and treatment assignment were based on the patients' clinical characteristics (e.g., age and white blood cell [WBC] count) and the disease response to initial therapy. Most children with T-ALL met National Cancer Institute (NCI) high-risk criteria.

On protocols of the former Children’s Cancer Group (CCG), children with T-ALL received the same treatment regimens as did children with B-ALL. Protocol and treatment assignment were based on the patients' clinical characteristics (e.g., age and white blood cell [WBC] count) and the disease response to initial therapy. Most children with T-ALL met National Cancer Institute (NCI) high-risk criteria.

Results from the CCG-1961 trial for high-risk ALL, which included patients with T-ALL, showed that an augmented Berlin-Frankfurt-Münster (BFM) regimen with a single delayed intensification course produced the best results for patients with morphological rapid response to initial induction therapy (estimated 5-year EFS rate, 83%).[9,10] With this approach, patients with a presenting WBC count greater than 200,000 had similar outcomes to those with a WBC count of less than 200,000.[11][Level of evidence B1]

Sources: CCG-1961 · 9 · 10 · 11 · Level of evidence B1

Overall results from POG-9404 and CCG-1961 were similar, although POG-9404 used a higher cumulative dose of anthracyclines and cranial radiation therapy for every patient, while CCG-1961 used cranial radiation therapy only for patients with slow morphological response.[10,5]

Sources: 10 · 5

Among children with NCI standard-risk T-ALL treated on the CCG-1952, COG-1991, and POG-9404 trials, the 7-year EFS rates were comparable with those of patients treated with the CCG regimens that used significantly less anthracycline in a less intensive chemotherapy backbone without the prophylactic cranial irradiation included in POG-9404.[12] However, the patients with NCI standard-risk T-ALL had an inferior EFS and OS compared with patients with NCI standard-risk B-ALL treated with the same regimens on the CCG-1952 and COG-1991 trials.[12]

Sources: CCG-1952 · COG-1991 · 12 · 12

In the COG, children with T-ALL are not treated on the same protocols as children with B-ALL.

In the COG, children with T-ALL are not treated on the same protocols as children with B-ALL.

Pilot studies from the COG have demonstrated the feasibility of incorporating nelarabine (a nucleoside analog with demonstrated activity in patients with relapsed and refractory T-cell lymphoblastic disease) in the context of a BFM regimen for patients with newly diagnosed T-ALL.[13-15]

Sources: 13 · 15

In the COG, children with T-ALL are not treated on the same protocols as children with B-ALL. › Pilot studies from the COG have demonstrated the feasibility of incorporating nelarabine (a nucleoside analog with demonstrated activity in patients with relapsed and refractory T-cell lymphoblastic disease) in the context of a BFM regimen for patients with newly diagnosed T-ALL.[13-15]

The pilot study showed a 5-year EFS rate of 73% for all patients who received nelarabine and 69% for those patients with a slow early response.[16]

Sources: 16

In the COG, children with T-ALL are not treated on the same protocols as children with B-ALL.

The COG AALL0434 (NCT00408005) trial enrolled 1,562 evaluable patients aged 1 to 31 years with T-ALL. Patients received an augmented BFM regimen and were randomly assigned to receive an interim maintenance phase with either high-dose methotrexate with leucovorin rescue or escalating-dose methotrexate without leucovorin but with pegaspargase.[3] Intermediate-risk and high-risk patients were also randomly assigned to receive either six courses of nelarabine during postinduction treatment or no nelarabine.[17] Nearly all patients received either prophylactic (12 Gy) or therapeutic (18 Gy) cranial irradiation. Only 10% of patients considered to be low risk were not irradiated. Patients assigned to the escalating-dose methotrexate arm received cranial radiation therapy earlier than did patients assigned to the high-dose methotrexate arm (week 8 vs. week 26). Patients on the escalating-dose methotrexate arm also received two additional doses of pegaspargase. Results were as follows:[3,17,18]

Sources: AALL0434 (NCT00408005) · 3 · 17 · 3 · 17 · 18

In the COG, children with T-ALL are not treated on the same protocols as children with B-ALL. › The COG AALL0434 (NCT00408005) trial enrolled 1,562 evaluable patients aged 1 to 31 years with T-ALL. Patients received an augmented BFM regimen and were randomly assigned to receive an interim maintenance phase with either high-dose methotrexate with leucovorin rescue or escalating-dose methotrexate without leucovorin but with pegaspargase.[3] Intermediate-risk and high-risk patients were also randomly assigned to receive either six courses of nelarabine during postinduction treatment or no nelarabine.[17] Nearly all patients received either prophylactic (12 Gy) or therapeutic (18 Gy) cranial irradiation. Only 10% of patients considered to be low risk were not irradiated. Patients assigned to the escalating-dose methotrexate arm received cranial radiation therapy earlier than did patients assigned to the high-dose methotrexate arm (week 8 vs. week 26). Patients on the escalating-dose methotrexate arm also received two additional doses of pegaspargase. Results were as follows:[3,17,18]

The overall 5-year EFS rate was 83.8%, and the OS rate was 89.5%. On multivariable analysis, high end-induction MRD (>0.1%) and elevated presenting WBC count (>200 × 103) were each independent predictors of inferior outcome.[18]

Sources: 18

Results indicated a better disease-free survival (DFS) for patients who were randomly assigned to the escalating-dose methotrexate arm (5-year DFS rate, 91.5%) than for patients randomly assigned to the high-dose methotrexate arm (5-year DFS rate, 85.3%; P = .005).

For intermediate-risk and high-risk patients, treatment with nelarabine was associated with a superior outcome (5-year DFS rate, 88.2% with nelarabine vs. 82.1% without nelarabine; P = .029). The 5-year cumulative incidence of central nervous system (CNS) relapse was significantly lower for patients treated with nelarabine (1.3% vs. 6.9% in the no-nelarabine arm).

The best outcome for intermediate-risk and high-risk patients was observed in those randomly assigned to both the escalating-dose methotrexate and nelarabine arms (5-year DFS rate, 91.4%). The worst outcome was observed in patients randomly assigned to receive high-dose methotrexate and no nelarabine (5-year DFS rate, 78.1%).

For patients with CNS3 disease, all of whom were assigned to receive high-dose methotrexate and 18 Gy of cranial radiation therapy, nelarabine was associated with significantly superior DFS.[19]

Sources: 19

Patients with initial induction failure (M3 marrow at day 29, n = 43) were nonrandomly assigned to receive high-dose methotrexate and nelarabine; 20 of these patients were removed from the protocol therapy to undergo allogeneic hematopoietic stem cell transplant (HSCT) in first complete remission (CR). The overall 5-year EFS rate for induction-failure patients was 53%, with no difference in outcomes between HSCT and chemotherapy.

In the COG, children with T-ALL are not treated on the same protocols as children with B-ALL.

In the successor COG AALL1231 (NCT02112916) trial for T-ALL, patients were randomly assigned to receive or not receive the proteasome inhibitor bortezomib during induction and delayed intensification phases. Patients with initial induction failure (M3 marrow at day 29) or high MRD (≥0.1%) at end of consolidation were nonrandomly assigned to receive three additional intensified chemotherapy blocks postconsolidation. Other nonrandomized changes to treatment included the use of dexamethasone during induction (instead of prednisone) and omission of cranial radiation therapy for all but the very high-risk patients. Because the results of AALL0434 trial were not known at the time that AALL1231 opened, nelarabine was not included in the chemotherapy backbone. The trial was designed to enroll 1,200 patients but closed to accrual early (after 824 evaluable patients had enrolled) once the results of the nelarabine randomization on AALL0434 were released.[20]

Sources: AALL1231 (NCT02112916) · 20

In the COG, children with T-ALL are not treated on the same protocols as children with B-ALL. › In the successor COG AALL1231 (NCT02112916) trial for T-ALL, patients were randomly assigned to receive or not receive the proteasome inhibitor bortezomib during induction and delayed intensification phases. Patients with initial induction failure (M3 marrow at day 29) or high MRD (≥0.1%) at end of consolidation were nonrandomly assigned to receive three additional intensified chemotherapy blocks postconsolidation. Other nonrandomized changes to treatment included the use of dexamethasone during induction (instead of prednisone) and omission of cranial radiation therapy for all but the very high-risk patients. Because the results of AALL0434 trial were not known at the time that AALL1231 opened, nelarabine was not included in the chemotherapy backbone. The trial was designed to enroll 1,200 patients but closed to accrual early (after 824 evaluable patients had enrolled) once the results of the nelarabine randomization on AALL0434 were released.[20]

Comparing patients with T-ALL assigned to receive bortezomib versus patients who did not receive bortezomib, there was no difference in the 4-year EFS rate (82.9% vs. 81.5%; P = .396) or the 5-year OS rate (87.9% vs. 88.3%; P = .469).

Subset analyses were performed comparing similar patients with T-ALL scheduled to receive 12 Gy cranial radiation therapy on the AALL0434 trial (90.8% of patients), but not on the AALL1231 trial (9.5% of patients). Excluding patients who received nelarabine (AALL0434) or bortezomib (AALL1231), the 4-year EFS (P = .412), OS (P = .600), cumulative incidence of CNS relapse (P = .456), and overall relapse (P = .836) rates were not significantly different between the studies, indicating that the omission of cranial radiation (in the context of the AALL1231 chemotherapy backbone) did not adversely impact outcome.

While the 4-year EFS rates were similar between the AALL1231 and AALL0434 trials (P = .131), the 4-year OS rate on the AALL1231 trial was inferior to that on the AALL0434 trial (87% vs. 90%, P = .006).

While the cumulative incidence of relapse was similar in both studies (P = .562), there was a higher rate of treatment-related mortality on the AALL1231 trial, compared with the AALL0434 trial (6.1% vs. 2.0%). Rates of treatment-related mortality were higher on the AALL1231 trial during both induction (1.5% vs. 0.4%, P = .002) and postinduction treatment phases (3.9% vs. 2.1%, P = .008).

Despite intensification of the chemotherapy backbone, outcome was poor for patients with T-ALL assigned to the very high-risk group (5.2% of patients). The 4-year EFS rate was 31.3% for patients on the no-bortezomib arm, compared with 7.8% for patients on the bortezomib arm (P = .033).

Younger age at diagnosis (≤90–180 days).

Extremely high presenting leukocyte count (≥200,000–300,000/μL).

Poor early response, as reflected by a poor response to a prednisone prophase or high levels of minimal residual disease (MRD) at the end of induction and consolidation phases of treatment.

The international Interfant-99 trial used a cytarabine-intensive chemotherapy regimen, with increased exposure to both low- and high-dose cytarabine during the first few months of therapy.[23]

Sources: 23

The international Interfant-99 trial used a cytarabine-intensive chemotherapy regimen, with increased exposure to both low- and high-dose cytarabine during the first few months of therapy.[23]

The 5-year EFS rate was 37% for infants with KMT2A rearrangements.

The COG tested intensification of therapy with a regimen that included multiple doses of high-dose methotrexate, cyclophosphamide, and etoposide.[22]

Sources: 22

The COG tested intensification of therapy with a regimen that included multiple doses of high-dose methotrexate, cyclophosphamide, and etoposide.[22]

The 5-year EFS rate was 34% for infants with KMT2A rearrangements.

On the COG P9407 (NCT00002756) trial, infants were treated with a shortened (46-week) intensive chemotherapy regimen.[24][Level of evidence B4]

Sources: P9407 (NCT00002756) · 24 · Level of evidence B4

On the COG P9407 (NCT00002756) trial, infants were treated with a shortened (46-week) intensive chemotherapy regimen.[24][Level of evidence B4]

The 5-year EFS rate was 36% for infants with KMT2A rearrangements.

The international Interfant-06 study tested whether acute myeloid leukemia (AML)-style consolidation chemotherapy was superior to ALL-style chemotherapy.[30][Level of evidence B4]

Sources: 30 · Level of evidence B4

The international Interfant-06 study tested whether acute myeloid leukemia (AML)-style consolidation chemotherapy was superior to ALL-style chemotherapy.[30][Level of evidence B4]

The 6-year EFS rate was 46.1%, and the OS rate was 58.2%. These rates were not statistically different from the rates observed in the predecessor Interfant-99 protocol.

For infants with KMT2A rearrangements, the 6-year EFS rate was 36.4%, with no significant difference between the AML and ALL approach.

In a subsequent analysis of the subset of KMT2A-rearranged patients in whom MRD was evaluated, both end-of-induction (EOI) and end-of-consolidation (EOC) MRD were strongly predictive of outcome. Patients who were EOI MRD positive, but became EOC MRD negative, had similar outcomes to those who were EOI MRD negative (6-year DFS rates, 65.7% and 72.0%, respectively). Patients with high EOC MRD had dismal outcomes (6-year DFS rate, 13.1%). Patients with negative EOI MRD had a higher 6-year DFS rate when treated with the ALL approach (78.2%) than did patients who were treated with the AML approach (45%). However, patients with high EOI MRD (≥5 × 10-4) who were treated with an AML approach had a superior 6-year DFS rate (45.9%) when compared with patients who were treated with the ALL approach (23.2%).[34][Level of evidence B1]

Sources: 34 · Level of evidence B1

In a follow-up pilot trial conducted by the Interfant study group, 30 infants with KMT2A-rearranged ALL were treated using the Interfant-06 chemotherapy backbone, with the addition of one 28-day course of blinatumomab after the induction phase.[35]

Sources: 35

In a follow-up pilot trial conducted by the Interfant study group, 30 infants with KMT2A-rearranged ALL were treated using the Interfant-06 chemotherapy backbone, with the addition of one 28-day course of blinatumomab after the induction phase.[35]

Blinatumomab was well tolerated, without any excessive or unexpected toxicity in these infant patients.

With short median follow-up (26.3 months), the 2-year DFS rate was 81.6% (95% CI, 60.8%–92.0%), and the 2-year OS rate was 93.3% (95% CI, 75.9%–98.3%).

These outcomes were superior to the 2-year DFS and OS rates of historical controls who were treated in the Interfant-06 protocol (DFS and OS rates, 49.4% and 65.8%, respectively).

On the MLL-10 trial conducted by the Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG), infants with KMT2A rearrangements were treated with an intensified chemotherapy backbone that included multiple phases using high-dose methotrexate, cyclophosphamide, etoposide, and high-dose cytarabine. Patients classified as high-risk on the basis of age and CNS status (75% of KMT2A-rearranged patients) were allocated to HSCT in first CR.[25]

Sources: 25

On the MLL-10 trial conducted by the Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG), infants with KMT2A rearrangements were treated with an intensified chemotherapy backbone that included multiple phases using high-dose methotrexate, cyclophosphamide, etoposide, and high-dose cytarabine. Patients classified as high-risk on the basis of age and CNS status (75% of KMT2A-rearranged patients) were allocated to HSCT in first CR.[25]

The 5-year EFS rate was 66%, and the 5-year OS rate was 82%.

In the COG AALL0631 (NCT00557193) trial, infants with KMT2A-rearranged ALL were assigned to receive an intensive chemotherapy regimen with or without lestaurtinib, an FLT3 inhibitor (administered during postinduction treatment phases).[28]Exploratory studies were conducted to evaluate the impact of sufficient lestaurtinib blood levels to achieve FLT3 inhibition and to evaluate the impact of ex vivo sensitivity of leukemia cells to lestaurtinib.

Sources: AALL0631 (NCT00557193) · 28

In the COG AALL0631 (NCT00557193) trial, infants with KMT2A-rearranged ALL were assigned to receive an intensive chemotherapy regimen with or without lestaurtinib, an FLT3 inhibitor (administered during postinduction treatment phases).[28]Exploratory studies were conducted to evaluate the impact of sufficient lestaurtinib blood levels to achieve FLT3 inhibition and to evaluate the impact of ex vivo sensitivity of leukemia cells to lestaurtinib.

The overall 5-year EFS rate for all participants was 34%, and the 5-year OS rate was 41%.

There was no difference in outcome between patients on the lestaurtinib arm versus those who received chemotherapy only.

Only 38% of lestaurtinib-treated patients demonstrated in vivo FLT3 inhibition. This subset of patients appeared to have a better outcome than other patients on the lestaurtinib arm, as did those whose blasts demonstrated ex vivo sensitivity to FLT3 inhibition.

On a trial conducted by the JPLSG between 1998 and 2002, all infants with KMT2A rearrangements were intended to proceed to allogeneic HSCT from the best available donor (related, unrelated, or umbilical cord) 3 to 5 months after diagnosis.[36]

Sources: 36

On a trial conducted by the JPLSG between 1998 and 2002, all infants with KMT2A rearrangements were intended to proceed to allogeneic HSCT from the best available donor (related, unrelated, or umbilical cord) 3 to 5 months after diagnosis.[36]

The 3-year EFS rate for all enrolled infants was 44%. This outcome resulted, in part, from the high frequency of early relapses, even with intensive chemotherapy. Of the 41 infants with KMT2A rearrangements on that study who achieved CR, 11 infants (27%) relapsed before proceeding to transplant.

In a follow-up trial conducted by the JPLSG between 2011 and 2015, allocation to HSCT in first CR was restricted to the patients with KMT2A rearrangements and classified as high risk (age <6 months and/or CNS3 status at diagnosis).[25]

Sources: 25

In a follow-up trial conducted by the JPLSG between 2011 and 2015, allocation to HSCT in first CR was restricted to the patients with KMT2A rearrangements and classified as high risk (age <6 months and/or CNS3 status at diagnosis).[25]

Of the 56 patients that were classified as high risk (75% of all KMT2A-rearranged patients enrolled on the trial), 49 achieved CR, and 38 underwent HSCT.

In an intent-to-treat analysis of all high-risk patients enrolled on the trial, the 5-year EFS rate was 56.6%.

For the subset of patients who also met Interfant-06 criteria for high risk (age <6 months and WBC >300,000/µL or a prednisolone poor responder), the 5-year EFS rate was 45.2%.

In a COG report that included 189 infants treated on CCG or POG infant ALL protocols between 1996 and 2000, there was no difference in EFS between patients who underwent HSCT in first CR and those who received chemotherapy alone.[37]

Sources: 37

The Interfant clinical trials group, after adjusting for waiting time to transplant, also did not observe any difference in DFS in high-risk infants (defined by prednisone response) with KMT2A rearrangements treated on the Interfant-99 trial with either allogeneic HSCT in first CR or chemotherapy alone.[23]

Sources: 23

The Interfant clinical trials group, after adjusting for waiting time to transplant, also did not observe any difference in DFS in high-risk infants (defined by prednisone response) with KMT2A rearrangements treated on the Interfant-99 trial with either allogeneic HSCT in first CR or chemotherapy alone.[23]

In a subset analysis from the same trial, allogeneic HSCT in first remission was associated with a significantly better DFS for infants with KMT2A rearrangements who were younger than 6 months at diagnosis and had either a poor prednisone response at day 8 or leukocyte counts of at least 300,000/µL.[38] In this subset, HSCT in first remission was associated with a 64% reduction in the risk of failure resulting from relapse or death compared with chemotherapy alone.

Sources: 38

On the Interfant-06 study, infants considered to be high risk (all of the following: KMT2A rearrangements, age <6 months, and WBC ≥300,000/μL) were considered eligible for allogeneic HSCT in first CR.[30][Level of evidence B4]

Sources: 30 · Level of evidence B4

On the Interfant-06 study, infants considered to be high risk (all of the following: KMT2A rearrangements, age <6 months, and WBC ≥300,000/μL) were considered eligible for allogeneic HSCT in first CR.[30][Level of evidence B4]

About one-half of the high-risk patients did not proceed to transplant in the first CR primarily because of early relapse.

The 6-year EFS rate of the entire high-risk group was 21%.

For the highly-selected population who underwent HSCT, the 4-year DFS rate was 44%.

On the Interfant-99 trial, patients without KMT2A rearrangements achieved a relatively favorable outcome with the cytarabine-intensive treatment regimen (4-year EFS rate, 74%).[23]

Sources: 23

The COG P9407 (NCT00002756) trial of intensified chemotherapy reported a 5-year EFS rate of 70% in infants without the KMT2A rearrangement.[24][Level of evidence B4]

Sources: P9407 (NCT00002756) · 24 · Level of evidence B4

A favorable outcome for this subset of patients was obtained in a Japanese study using therapy comparable to that used to treat older children with ALL.[27] However, that study was limited by small numbers (n = 22) and a highly unusual sex distribution (91% males).

Sources: 27

On the Interfant-06 study, the 6-year EFS rate for infants without KMT2A rearrangements was 73.9%, and the OS rate was 87.2%.[40]; [30][Level of evidence B4]

Sources: 40 · 30 · Level of evidence B4

On the COG AALL0631 (NCT00557193) study, infants without KMT2A rearrangements received the same intensified chemotherapy backbone as those with the KMT2A rearrangement.[41]

Sources: AALL0631 (NCT00557193) · 41

On the COG AALL0631 (NCT00557193) study, infants without KMT2A rearrangements received the same intensified chemotherapy backbone as those with the KMT2A rearrangement.[41]

The 5-year EFS rate was 87.3% (± 4.7%), and the 5-year OS rate was 93.6% (± 3.5%) for 64 infants without KMT2A rearrangements.

T-cell immunophenotype.

BCR::ABL1 and BCR::ABL1-like disease.

Lower incidence of favorable cytogenetic abnormalities.

The CALGB-10403 (NCT00558519) trial prospectively studied the feasibility and efficacy of using a pediatric treatment regimen (administered by medical oncologists) for adolescent and young adult patients with newly diagnosed ALL. Of the 318 patients enrolled, 295 were eligible and evaluable for response. The median age was 24 years (range, 17–39 years).[56,58]

Sources: CALGB-10403 (NCT00558519) · 56 · 58

The CALGB-10403 (NCT00558519) trial prospectively studied the feasibility and efficacy of using a pediatric treatment regimen (administered by medical oncologists) for adolescent and young adult patients with newly diagnosed ALL. Of the 318 patients enrolled, 295 were eligible and evaluable for response. The median age was 24 years (range, 17–39 years).[56,58]

Use of the pediatric regimen (from the COG AALL0232 study, which included interim maintenance phases with escalating doses of methotrexate without leucovorin followed by asparaginase) was deemed safe, and the overall treatment-related mortality was 3%.

Sources: AALL0232

The median EFS was 78.1 months, which is more than double the historical control of 30 months. The 3-year EFS rate was 59%, and the median OS was not reached. The estimated 3-year OS rate was 73%.

Pretreatment risk factors associated with a worse outcome were obesity and the presence of the BCR::ABL1-like expression signature. Of the evaluable patients, 31% had a BCR::ABL1-like signature. These patients had a significantly worse outcome, with a 3-year EFS rate of 42%, compared with an EFS rate of 69% for patients without BCR::ABL1-like ALL (hazard ratio, 2.06; log-rank P = .008).

End-induction MRD was a significant predictor of outcome. In the subset of patients evaluable for MRD, 44% had absence of detectable end-induction MRD using an Ig-TCR PCR-MRD assay (sensitivity of detection of 1 in 104 to 105). These patients had a 3-year DFS rate of 85%, compared with a 3-year DFS rate of 54% for those with detectable end-induction MRD (P = .001).

In a study of patients aged 16 to 39 years, the outcomes of those who received postremission, standard, pediatric-inspired chemotherapy per the CALGB-10403 protocol were compared with the outcomes of patients who received myeloablative allogeneic HSCT. Patients who received chemotherapy had superior OS, DFS, and nonrelapse mortality rates.[59][Level of evidence B4]

Sources: 59 · Level of evidence B4

Investigators reported on 197 patients aged 16 to 21 years treated on the CCG study (a pediatric ALL regimen) compared with 124 adolescents and young adults treated on the Cancer and Leukemia Group B (CALGB) study (an adult ALL regimen).[48]

Sources: 48

Investigators reported on 197 patients aged 16 to 21 years treated on the CCG study (a pediatric ALL regimen) compared with 124 adolescents and young adults treated on the Cancer and Leukemia Group B (CALGB) study (an adult ALL regimen).[48]

For the patients treated with a pediatric ALL regimen, the 7-year EFS rate was 63%.

For the patients treated with an adult ALL regimen, the 7-year EFS rate was 34%.

In a Canadian population-based cohort study, the effect of adapting pediatric protocols for adolescent and young adult patients with ALL was determined over a 20-year period.[60]

Sources: 60

In a Canadian population-based cohort study, the effect of adapting pediatric protocols for adolescent and young adult patients with ALL was determined over a 20-year period.[60]

The 5-year EFS rate was 72% for adolescent and young adult patients treated at pediatric centers, compared with 56% for adolescent and young adult patients treated at adult centers (P = .03).

In the most recent period (2006–2011), the outcome of adolescent and young adult patients treated at adult centers with pediatric protocols was superior to those treated with adult protocols (EFS rate, 72% vs. 60%), but inferior to adolescent and young adult patients treated at pediatric centers (EFS rate, 81%; P = .02).

The authors conclude that besides protocol therapy, there may be other differences between adult and pediatric centers that may explain the disparate outcomes.

Treatment setting (i.e., site experience in treating ALL).

Adherence to protocol therapy.[47]

Sources: 47

The components of protocol therapy.

In the CCG-1961 high-risk ALL study, alternate-week dosing of dexamethasone was compared with standard continuous dexamethasone during delayed intensification to determine whether the osteonecrosis risk could be reduced.[65]

Sources: CCG-1961 · 65

In the CCG-1961 high-risk ALL study, alternate-week dosing of dexamethasone was compared with standard continuous dexamethasone during delayed intensification to determine whether the osteonecrosis risk could be reduced.[65]

The median age at symptom onset was 16 years.

The cumulative incidence was higher in adolescents and young adults aged 16 to 21 years (20% at 5 years) than in those aged 10 to 15 years (9.9%) or in patients aged 1 to 9 years (1%).

Operative interventions were needed for management of symptoms and impaired mobility in more than 40% of cases.

The use of alternate-week dosing of dexamethasone as compared with standard continuous dexamethasone during delayed intensification in CCG-1961 reduced the risk of osteonecrosis. The greatest impact was seen in females aged 16 to 21 years, who showed the highest incidence of osteonecrosis with standard therapy containing continuous dexamethasone; osteonecrosis was reduced with alternate-week dexamethasone postinduction (57.6% to 5.6%).

In the COG AALL0232 (NCT00075725) high-risk ALL trial, patients were randomly assigned during induction to receive either 14 days of dexamethasone or 28 days of prednisone.[67]

Sources: AALL0232 (NCT00075725) · 67

In the COG AALL0232 (NCT00075725) high-risk ALL trial, patients were randomly assigned during induction to receive either 14 days of dexamethasone or 28 days of prednisone.[67]

The incidence of osteonecrosis in patients older than 10 years who received dexamethasone was 24.3%, compared with an incidence of 15.9% in those who received prednisone (P = .001).

For patients aged 10 years and older, there was no difference in outcome between arms by corticosteroid (5-year EFS rate, 73.1% on the dexamethasone arm and 73.9% on the prednisone arm; P = .78).

However, among all patients in the analysis cohort, patients with osteonecrosis had superior 5-year EFS and OS rates than patients without osteonecrosis (EFS rates, 89.8% vs. 77.8%, P < .0001; OS rates, 95.8% vs. 86.1%, P < .0001).[68] Similar differences were seen in patients older than 10 years. Improved survival was directly attributed to reduced relapse rates.

Sources: 68

A large retrospective study that included 653 patients with Down syndrome and ALL who were treated between 1995 to 2004 reported the following results:[74]

Sources: 74

A large retrospective study that included 653 patients with Down syndrome and ALL who were treated between 1995 to 2004 reported the following results:[74]

Patients with Down syndrome had lower CR rates (97% vs. 99%, P < .001), higher cumulative incidence of relapse rates (26% vs. 15%, P < .001), and higher treatment-related mortality rates (7.7% vs. 2.3%, P < .001), compared with patients without Down syndrome.

For patients with Down syndrome and ALL, treatment-related mortality occurred at a higher rate during all phases of therapy, including maintenance. The rates of treatment-related mortality during induction were 2.8% and 4.9% during the remaining treatment phases after CR had been achieved. The most common cause of treatment-related mortality was infection.

Among the patients with Down syndrome, age younger than 6 years, WBC count of less than 10,000/μL, and the presence of the ETV6::RUNX1 fusion (observed in 8% of patients) were independent predictors of favorable EFS.

In a report of 2,811 children with ALL enrolled on the COG P9900 classification study, 80 patients (3%) had Down syndrome. Age, sex, WBC count, and risk group were similar between patients with and without Down syndrome, but a lower percentage of patients with Down syndrome had ETV6::RUNX1 (2.5% vs. 24%, P < .001) or trisomies 4 and 10 (7.7% vs. 24%, P < .001).[71]

Sources: 71

In a report of 2,811 children with ALL enrolled on the COG P9900 classification study, 80 patients (3%) had Down syndrome. Age, sex, WBC count, and risk group were similar between patients with and without Down syndrome, but a lower percentage of patients with Down syndrome had ETV6::RUNX1 (2.5% vs. 24%, P < .001) or trisomies 4 and 10 (7.7% vs. 24%, P < .001).[71]

The 5-year EFS and OS rates were inferior in children with Down syndrome: 69.9% versus 78.1% (P = .078), and 85.8% versus 90.0% (P = .033), respectively.

However, when children with KMT2A rearrangements, BCR::ABL1, ETV6::RUNX1, and trisomies 4 and 10 were excluded from the analysis, the EFS and OS rates were similar for children with and without Down syndrome (EFS rates, 68.0% [± 9.3%] vs. 70.5% [± 1.9%], P = .817; OS rates, 86.7% [± 6.7%] vs. 85.4% [± 1.5%]; P = .852).

In the CCG-1991 (NCT00005945) study for children with standard-risk B-ALL (2000–2005), patients (including those with Down syndrome) were randomly assigned to receive an interim maintenance phase consisting of either escalating doses of intravenous methotrexate without leucovorin rescue or standard low-dose oral methotrexate given with vincristine, dexamethasone, and mercaptopurine.[85]

Sources: CCG-1991 (NCT00005945) · 85

In the CCG-1991 (NCT00005945) study for children with standard-risk B-ALL (2000–2005), patients (including those with Down syndrome) were randomly assigned to receive an interim maintenance phase consisting of either escalating doses of intravenous methotrexate without leucovorin rescue or standard low-dose oral methotrexate given with vincristine, dexamethasone, and mercaptopurine.[85]

The 10-year EFS rates were 94.4% (± 5.4%) for patients with Down syndrome randomly assigned to receive escalating-dose IV methotrexate (N = 31), versus 81.5% (± 6.6%) for patients who received oral low-dose methotrexate (N = 44).

The mean total tolerated dose of methotrexate on the escalating IV arm was lower in patients with Down syndrome than patients without Down syndrome.

The incidence of mucositis was increased in patients with Down syndrome who received escalating-dose methotrexate, compared with patients without Down syndrome, but there was no increase in hepatic toxicity, infections, or treatment-related deaths.

On DFCI ALL Consortium protocols, children with Down syndrome receive the same risk-stratified therapy as other patients, without any dose reductions or modifications. An analysis of two consecutive trials conducted between 2000 to 2011 found the following:[84]

Sources: 84

On DFCI ALL Consortium protocols, children with Down syndrome receive the same risk-stratified therapy as other patients, without any dose reductions or modifications. An analysis of two consecutive trials conducted between 2000 to 2011 found the following:[84]

The 5-year EFS and OS rates of patients with Down syndrome (N = 38) were similar to that of patients without Down syndrome (N = 1,248) (EFS rates, 91% vs. 84%; OS rates, 97% vs. 91%).

All patients with Down syndrome achieved complete remission, and none of the patients experienced treatment-related mortality.

Patients with Down syndrome had significantly higher rates of mucositis (52% vs. 12%, P < .001), non-CNS thrombosis (18% vs. 8%; P = .036), and seizure (16% vs. 5%, P = .010). Patients with Down syndrome also had a higher incidence of infections during all therapy phases.

The COG-AALL0031 study evaluated whether imatinib mesylate could be incorporated into an intensive chemotherapy regimen for children with BCR::ABL1 ALL. Patients received imatinib mesylate in conjunction with chemotherapy during postinduction therapy. Some children proceeded to allogeneic HSCT after two cycles of consolidation chemotherapy with imatinib mesylate, while other patients received imatinib mesylate in combination with chemotherapy throughout all treatment phases.[97,102]

Sources: COG-AALL0031 · 97 · 102

The COG-AALL0031 study evaluated whether imatinib mesylate could be incorporated into an intensive chemotherapy regimen for children with BCR::ABL1 ALL. Patients received imatinib mesylate in conjunction with chemotherapy during postinduction therapy. Some children proceeded to allogeneic HSCT after two cycles of consolidation chemotherapy with imatinib mesylate, while other patients received imatinib mesylate in combination with chemotherapy throughout all treatment phases.[97,102]

The 5-year DFS rate for the 25 patients who received intensive chemotherapy with continuous dosing of imatinib mesylate was 70% (± 12%). These patients fared better than historical controls who were treated with chemotherapy alone (without imatinib mesylate), and at least as well as the other patients on the trial who underwent allogeneic transplant. The 5-year DFS rate was 66% for patients who underwent sibling-donor transplant (n = 21) and 59% for those who underwent unrelated donor transplant (n = 13).

Patients with additional cytogenetic abnormalities had worse outcomes (P = .05).

The COG-AALL0622 (NCT00720109) study tested the use of dasatinib (instead of imatinib) combined with a chemotherapy backbone similar to that used in the COG-AALL0031 trial.[106][Level of evidence B4] On this trial, dasatinib was started on day 15 of induction, while on the AALL0031 trial, imatinib was started postinduction.

Sources: COG-AALL0622 (NCT00720109) · 106 · Level of evidence B4

The COG-AALL0622 (NCT00720109) study tested the use of dasatinib (instead of imatinib) combined with a chemotherapy backbone similar to that used in the COG-AALL0031 trial.[106][Level of evidence B4] On this trial, dasatinib was started on day 15 of induction, while on the AALL0031 trial, imatinib was started postinduction.

Introduction of a TKI during induction on the AALL0631 trial resulted in better early response compared with the AALL0031 trial (no TKI during induction). The end-induction (day 29) CR rate was 98% (n = 59) on the AALL0622 trial, compared with 89% (n = 91) on the AALL0031 trial (P = .01), and rates of low end-induction MRD (<0.01%) were superior on the AALL0622 trial compared with the AALL0031 trial (59% vs. 25%; P < .001).

Outcomes in the two trials were similar: the 5-year OS rates were 81% and 86%, and the 5-year DFS rates were 68% and 60% for AALL0031 and AALL0622, respectively.

Excessive toxicity with dasatinib was not observed.

In a subset analysis that included patients who had diagnostic banked samples available, the IKZF1 deletion was identified in 57% of patients and was associated with inferior EFS and OS.

The EsPhALL2004 trial tested whether imatinib (administered discontinuously) given in the context of intensive chemotherapy improved outcome for pediatric BCR::ABL1 ALL patients, most of whom (80%) received an allogeneic HSCT in first CR. Patients were classified as either good risk or poor risk on the basis of early response measures and remission status at the end of induction. Good-risk patients (n = 90) were randomly assigned to receive imatinib or no imatinib; poor-risk patients (n = 70) were directly assigned to treatment with imatinib. Interpretation of this study is limited because of the high noncompliance rate with randomized assignment in good-risk patients and early closure before reaching goal accrual because of the publication of the results of the COG AALL0031 trial on which imatinib had been given continuously with chemotherapy.[103]

Sources: 103

The EsPhALL2004 trial tested whether imatinib (administered discontinuously) given in the context of intensive chemotherapy improved outcome for pediatric BCR::ABL1 ALL patients, most of whom (80%) received an allogeneic HSCT in first CR. Patients were classified as either good risk or poor risk on the basis of early response measures and remission status at the end of induction. Good-risk patients (n = 90) were randomly assigned to receive imatinib or no imatinib; poor-risk patients (n = 70) were directly assigned to treatment with imatinib. Interpretation of this study is limited because of the high noncompliance rate with randomized assignment in good-risk patients and early closure before reaching goal accrual because of the publication of the results of the COG AALL0031 trial on which imatinib had been given continuously with chemotherapy.[103]

The overall DFS of patients treated on this trial appeared to be better than historical controls, and when analyzed as-treated (and not by intent-to-treat), good-risk patients who received imatinib had a superior DFS (4-year DFS rate was 75% for patients who received imatinib and 56% for patients who did not receive imatinib).[107]

Sources: 107

The subsequent EsPhALL2010 (NCT00287105) trial was a result of an amendment to the 2004 trial, which included earlier initiation of imatinib therapy at day 15 of induction and continuous dosing of imatinib until the end of therapy or 1 year after transplant. A subsequent amendment in the trial also changed the indication for HSCT in first CR to only the poor-risk patients.[108]

Sources: EsPhALL2010 (NCT00287105) · 108

The subsequent EsPhALL2010 (NCT00287105) trial was a result of an amendment to the 2004 trial, which included earlier initiation of imatinib therapy at day 15 of induction and continuous dosing of imatinib until the end of therapy or 1 year after transplant. A subsequent amendment in the trial also changed the indication for HSCT in first CR to only the poor-risk patients.[108]

Earlier introduction of imatinib (during induction) resulted in an increased rate of CR to 97% at the end of induction (from 78% in the previous trial) and fewer patients being allocated to HSCT (38% on amended trial vs. 81% on initial trial).

The EFS and OS rates were similar between the amended trial and the initial trial, even though significantly fewer patients received HSCT in first CR on the amended trial.

The EsPhALL chemotherapy backbone combined with continuous dosing of imatinib was associated with a high rate of toxicity (primarily infections) and treatment-related mortality.

The CA180-372/COG AALL1122 (NCT01460160) study was a joint COG and European intergroup single-arm, phase II trial that tested the combination of dasatinib and the EsPhALL chemotherapy backbone for patients with newly diagnosed BCR::ABL1–positive ALL. Dasatinib was given starting at day 15 of induction therapy, and only CNS3 patients or those allocated to HSCT received cranial radiation. Nineteen of 109 patients (18%) were considered high risk based on high time point 2 (end-induction IB) MRD or detectable MRD (any level) after three consolidation blocks of chemotherapy. All other patients were considered standard risk (82%). Fifteen patients (14%), all assigned to the high-risk group, underwent HSCT in first CR. All remaining patients were allocated to receive 2 years of chemotherapy with dasatinib.[90]

Sources: CA180-372/COG AALL1122 (NCT01460160) · 90

The CA180-372/COG AALL1122 (NCT01460160) study was a joint COG and European intergroup single-arm, phase II trial that tested the combination of dasatinib and the EsPhALL chemotherapy backbone for patients with newly diagnosed BCR::ABL1–positive ALL. Dasatinib was given starting at day 15 of induction therapy, and only CNS3 patients or those allocated to HSCT received cranial radiation. Nineteen of 109 patients (18%) were considered high risk based on high time point 2 (end-induction IB) MRD or detectable MRD (any level) after three consolidation blocks of chemotherapy. All other patients were considered standard risk (82%). Fifteen patients (14%), all assigned to the high-risk group, underwent HSCT in first CR. All remaining patients were allocated to receive 2 years of chemotherapy with dasatinib.[90]

The 5-year EFS rate was 54.6%, and the 5-year OS rate was 91.5%, indicating a high rate of salvageability after relapse.

For patients with standard-risk ALL, the 5-year EFS rate was 52.1%, and the OS rate was 82.3%.

For patients with high-risk ALL, the 5-year EFS rate was 56.8%, and the OS rate was 78.2%.

The 3-year EFS rate on the AALL1122 trial (65.5%) was statistically noninferior to that of the EsPhALL 2010 trial (59.1%), on which patients received imatinib combined with the same chemotherapy backbone. However, fewer patients on the AALL1122 trial underwent HSCT in first CR, compared with the EsPhALL 2010 trial (14% vs. 38%). Fewer patients on the AALL1122 trial received cranial radiation (15% vs. 100%).

Frequency of infections were high (grade 3 or higher sepsis, 34%; grade 3 or higher fungal infection, 14%), and the treatment-related mortality rate was 8% (8% in those receiving chemotherapy with dasatinib, 13% in those undergoing HSCT in first CR).

MRD on this trial was assessed by three methodologies: IG/TCR PCR, flow cytometry, and BCR::ABL1 PCR. Concordance between assays was highest for IG/TCR PCR and flow cytometry, while BCR::ABL1 PCR results were more frequently discordant. The BCR::ABL1 PCR assay more often demonstrated persistent positivity when the other assays yielded nondetectable results. The outcome of patients who continued to have detectable BCR::ABL1 PCR results, but were MRD negative by the other two assays, was similar to that of patients with nondetectable results by all three assays.

The Chinese Children’s Cancer Group ALL-2015 trial randomly assigned 189 patients to receive either dasatinib or imatinib combined with a multiagent regimen based on SJCRH ALL protocols. On this trial, dasatinib was administered at a higher dose (80 mg/m2 instead of 60 mg/m2) and imatinib was administered at a lower dose (300 mg/m2 instead of 340 mg/m2) than they were administered on previous pediatric BCR::ABL1 ALL trials conducted by COG and EsPhALL.[109][Level of evidence A1]

Sources: 109 · Level of evidence A1

The Chinese Children’s Cancer Group ALL-2015 trial randomly assigned 189 patients to receive either dasatinib or imatinib combined with a multiagent regimen based on SJCRH ALL protocols. On this trial, dasatinib was administered at a higher dose (80 mg/m2 instead of 60 mg/m2) and imatinib was administered at a lower dose (300 mg/m2 instead of 340 mg/m2) than they were administered on previous pediatric BCR::ABL1 ALL trials conducted by COG and EsPhALL.[109][Level of evidence A1]

With a median follow-up of 26.4 months, the 4-year EFS and OS rates were 71% and 88.4%, respectively, for patients randomly assigned to receive dasatinib, compared with 48.9% and 69.2%, respectively, for patients randomly assigned to receive imatinib.

Toxicity was similar between the two treatment arms.

Caution is needed in interpreting these results because of the short median follow-up time and because the outcomes of patients who received imatinib on this trial were inferior to the outcomes of imatinib-treated patients reported in previous trials.

Site of relapse.

Sources: Site of relapse

Time from diagnosis to relapse.

Sources: Time from diagnosis to relapse

Patient characteristics (e.g., age and peripheral blast count at time of relapse).

Sources: Patient characteristics

Risk group classification at initial diagnosis.

Sources: Risk group classification at initial diagnosis

Response to reinduction therapy.

Sources: Response to reinduction therapy

Cytogenetics/genomic alterations.

Sources: Cytogenetics/genomic alterations

Immunophenotype.

Sources: Immunophenotype

The BFM group showed that since 2000, improvements in supportive care have led to decreases in treatment-related mortality in children with Down syndrome, but the risk of relapse remains high.[22]

Sources: 22

An analysis of data from the Center for International Blood and Marrow Transplant Research (CIBMTR) on 27 patients with Down syndrome and ALL who underwent hematopoietic stem cell transplant (HSCT) between 2000 and 2009 substantiated this finding. They noted that with current transplant practices, hematopoietic recovery, graft-versus-host disease (GVHD), and transplant-related mortality were within the expected range compared with patients without Down syndrome. However, relapse was higher than expected (>50%) and was the primary cause of treatment failure, leading to poor survival (disease-free survival [DFS] rate at 3 years, 24%).[23][Level of evidence C1]

Sources: 23 · Level of evidence C1

In a study of 2,053 patients with ALL initially treated in COG studies who subsequently experienced a relapse, several recurrent cytogenetic abnormalities were significant predictors of OS. Patients with either ETV6::RUNX1 gene fusions or high hyperdiploidy with trisomies 4 and 10 had a longer median time to first relapse (43 months) and superior 5-year postrelapse OS rates (74.4% and 70.2%, respectively), compared with other patients. Conversely, those with hypodiploidy, KMT2A rearrangements, or TCF3::PBX1 gene fusions were more likely to have earlier relapses and had inferior 5-year postrelapse OS rates (14.2% ± 6.1%, 31.9% ± 7.7%, and 36.8% ± 6.6%, respectively). In multivariable analysis controlling for time to relapse, each of these cytogenetic abnormalities retained significance as an independent prognostic factor after relapse.[15]

Sources: 15

In the ALL-REZ BFM 2002 (NCT00114348) study, an EFS rate of 84% (± 7%) was observed for patients with ETV6::RUNX1 ALL with bone marrow relapse.[32] In this study, 94% of patients with ETV6::RUNX1 had a duration of first remission that extended at least 6 months beyond completion of their primary treatment, and on multivariate analysis, time to relapse (and not the presence of ETV6::RUNX1) was an independent predictor of outcome.

Sources: NCT00114348 · 32

Similarly, the 5-year OS rate was 81% for patients with ETV6::RUNX1 enrolled on the French Acute Lymphoblastic Leukaemia Study Group (FRALLE) 93 trial who relapsed at any site more than 36 months after diagnosis. The presence of ETV6::RUNX1 was associated with a favorable survival outcome compared with other late relapsing patients.[33] However, the 3-year OS rate was only 31% for patients with ETV6::RUNX1 who experienced an early relapse (<36 months).

Sources: 33

Reinduction chemotherapy.

Sources: Reinduction chemotherapy

Postreinduction therapy for patients achieving a second CR.

Sources: Postreinduction therapy for patients achieving a second CR

A COG study used three blocks of intensive reinduction therapy with an initial four-drug combination that included doxorubicin followed by two intensive consolidation blocks before either HSCT or chemotherapy continuation.[24]

Sources: 24

A COG study used three blocks of intensive reinduction therapy with an initial four-drug combination that included doxorubicin followed by two intensive consolidation blocks before either HSCT or chemotherapy continuation.[24]

Second remission was achieved after block 1 in 68% of patients with early relapse (<36 months from initial diagnosis) and in 96% of those with later relapse.

Blocks 2 and 3 reduced MRD in 40 of 56 patients who were MRD positive after block 1.

A United Kingdom–based randomized trial of ALL patients in first relapse compared reinduction with a four-drug combination using idarubicin versus mitoxantrone.[42][Level of evidence A1] The potential benefit of mitoxantrone in relapsed ALL regimens requires further investigation.

Sources: 42 · Level of evidence A1

A United Kingdom–based randomized trial of ALL patients in first relapse compared reinduction with a four-drug combination using idarubicin versus mitoxantrone.[42][Level of evidence A1] The potential benefit of mitoxantrone in relapsed ALL regimens requires further investigation.

There was no difference in second CR rates or end-reinduction MRD levels between the two study arms.

A significant improvement in the OS rate in the mitoxantrone arm was reported (69% vs. 45%; P = .007), which resulted from decreased relapse after transplant.

Investigators from the ALL-REZ BFM group used a six-drug reinduction approach, which included high-dose methotrexate.[43]

Sources: 43

Investigators from the ALL-REZ BFM group used a six-drug reinduction approach, which included high-dose methotrexate.[43]

A randomized comparison of 1 g/m2 of methotrexate delivered over 36 hours versus 5 g/m2 of methotrexate delivered over 24 hours with reinduction showed no difference in outcome, but the 36-hour infusion was associated with a higher incidence of mucositis.

The combination of clofarabine, cyclophosphamide, and etoposide was reported to induce remission in 42% to 56% of patients with refractory or multiply relapsed disease.[44,45]; [46][Level of evidence B4]

Sources: 44 · 45 · 46 · Level of evidence B4

The combination of bortezomib plus vincristine, dexamethasone, pegaspargase, and doxorubicin has been reported to induce complete response (with or without platelet recovery) in 70% to 80% of multiply relapsed patients with B-ALL.[47][Level of evidence C1]; [48][Level of evidence C3]

Sources: 47 · Level of evidence C1 · 48 · Level of evidence C3

The combination of bortezomib plus vincristine, dexamethasone, pegaspargase, and doxorubicin has been reported to induce complete response (with or without platelet recovery) in 70% to 80% of multiply relapsed patients with B-ALL.[47][Level of evidence C1]; [48][Level of evidence C3]

This combination (using prednisone instead of dexamethasone) was tested in B-ALL patients with first relapse occurring less than 36 months from initial diagnosis.[49]

Sources: 49

The combination of bortezomib plus vincristine, dexamethasone, pegaspargase, and doxorubicin has been reported to induce complete response (with or without platelet recovery) in 70% to 80% of multiply relapsed patients with B-ALL.[47][Level of evidence C1]; [48][Level of evidence C3] › This combination (using prednisone instead of dexamethasone) was tested in B-ALL patients with first relapse occurring less than 36 months from initial diagnosis.[49]

The second CR rate was 68%, which was not significantly different from that observed on a predecessor trial using the same reinduction platform without bortezomib.

In subset analyses, the addition of bortezomib to the four-drug reinduction platform did not result in significantly better second CR rates for patients with either very early relapses (<18 months from diagnosis) or early relapses (18–36 months from diagnosis) when compared with historical controls.

In a randomized phase III trial conducted between 2015 and 2019, patients with early-relapsing B-ALL were randomly assigned to receive either one 28-day course of blinatumomab or a cycle of intensive cytotoxic chemotherapy as a third consolidation course before proceeding to HSCT. The study was intended to enroll 202 patients but was terminated early (after 108 patients enrolled) when the superiority of blinatumomab was demonstrated on the basis of a prespecified stopping rule.[65]

Sources: 65

In a randomized phase III trial conducted between 2015 and 2019, patients with early-relapsing B-ALL were randomly assigned to receive either one 28-day course of blinatumomab or a cycle of intensive cytotoxic chemotherapy as a third consolidation course before proceeding to HSCT. The study was intended to enroll 202 patients but was terminated early (after 108 patients enrolled) when the superiority of blinatumomab was demonstrated on the basis of a prespecified stopping rule.[65]

The 24-month EFS rate was 66.2% for patients who received blinatumomab, compared with 27.1% for patients who received the intensive chemotherapy block (hazard ratio [HR], 0.33; P < .001).

The 24-month cumulative incidence of relapse was significantly lower in patients who received blinatumomab than in patients who received intensive chemotherapy (24.9% vs. 70.8%; HR, 0.24).

For patients with high MRD (≥10-4) at the time randomized treatment began, more patients achieved MRD negativity with blinatumomab treatment than with intensive chemotherapy treatment (93% vs. 24%).

A COG phase III clinical trial compared two courses of blinatumomab with two blocks of intensive chemotherapy in patients with B-ALL in first relapse. Treatment randomization occurred after a single block of intensive reinduction chemotherapy. Eligible patients included those with early relapse or those with late relapse but MRD levels greater than 0.1% after reinduction chemotherapy. Study enrollment ended early because of loss of clinical equipoise, in favor of blinatumomab. This was determined by the study’s data and safety monitoring committee. Key findings from the trial include the following:[66]

Sources: 66

A COG phase III clinical trial compared two courses of blinatumomab with two blocks of intensive chemotherapy in patients with B-ALL in first relapse. Treatment randomization occurred after a single block of intensive reinduction chemotherapy. Eligible patients included those with early relapse or those with late relapse but MRD levels greater than 0.1% after reinduction chemotherapy. Study enrollment ended early because of loss of clinical equipoise, in favor of blinatumomab. This was determined by the study’s data and safety monitoring committee. Key findings from the trial include the following:[66]

The 2-year DFS rate was 54.4% for patients who received blinatumomab versus 39.0% for patients who received chemotherapy (HR for disease progression or mortality, 0.70; 95% confidence interval [CI], 0.47–1.03).

The 2-year OS rate was 71.3% for patients who received blinatumomab versus 58.4% for patients who received chemotherapy (HR for mortality, 0.62; 95% CI, 0.39–0.98).

Negative MRD rates were higher for patients after one or two courses of blinatumomab (75% and 66%, respectively) than for patients after one or two courses of intensive chemotherapy (32% and 32%, respectively).

Adverse event rates for toxicities such as sepsis, mucositis, and febrile neutropenia were much lower for patients who received blinatumomab than for patients who received intensive chemotherapy.

In BFM studies, patients are considered to be intermediate risk if they have a late isolated marrow relapse or an early or late combined marrow/extramedullary relapse. In the ALL-REZ BFM P95/96 study from this group, end-reinduction MRD (assessed by a polymerase chain reaction–based assay) significantly predicted outcomes of children with intermediate-risk relapsed B-ALL treated with chemotherapy alone in second CR (no HSCT).[28]

Sources: 28

In BFM studies, patients are considered to be intermediate risk if they have a late isolated marrow relapse or an early or late combined marrow/extramedullary relapse. In the ALL-REZ BFM P95/96 study from this group, end-reinduction MRD (assessed by a polymerase chain reaction–based assay) significantly predicted outcomes of children with intermediate-risk relapsed B-ALL treated with chemotherapy alone in second CR (no HSCT).[28]

Patients with low MRD (<0.1%) had a 10-year EFS rate of 73%, while those with high MRD (≥0.1%) had a 10-year EFS rate of 10%. On multivariate analysis, end-reinduction MRD was the strongest independent prognostic factor.

In a subsequent BFM study (ALL-REZ BFM 2002 [NCT00114348]), patients with intermediate-risk relapse were allocated to allogeneic HSCT if they had high MRD at the end of the first month of treatment. Those who had low end-reinduction MRD were treated with chemotherapy only (no HSCT).[71]

Sources: ALL-REZ BFM 2002 [NCT00114348] · 71

In a subsequent BFM study (ALL-REZ BFM 2002 [NCT00114348]), patients with intermediate-risk relapse were allocated to allogeneic HSCT if they had high MRD at the end of the first month of treatment. Those who had low end-reinduction MRD were treated with chemotherapy only (no HSCT).[71]

The EFS rate was 64% for patients with high end-reinduction MRD treated with allogeneic HSCT in second CR, which was significantly better than what had been observed on the previous P95/96 trial, during which such patients received chemotherapy without HSCT. The improvement in EFS was primarily because of a significantly lower risk of relapse in the cohort who underwent HSCT in second CR (cumulative incidence of relapse, 27% on the 2002 trial compared with 59% on the P95/96 trial).

Patients with late marrow-involved relapses and low end-reinduction MRD, treated with chemotherapy only, had a 5-year EFS rate of 76%, confirming the results seen in the previous P95/96 trial. However, the chemotherapy-only strategy resulted in a significantly worse outcome for patients with early-combined relapses (marrow plus extramedullary site) and low end-reinduction MRD. The 5-year EFS rate was only 37% for these patients. Because of these data, patients with early-combined relapses are now considered high risk on BFM trials.

For patients with late marrow-involved relapses and high end-reinduction MRD (defined as ≥0.1%), level of MRD was prognostically significant. The 10-year EFS rate was significantly worse for those with MRD levels of ≥1% compared with those with MRD levels of ≥0.1% to <1% (56% vs. 74%, respectively; P = .02). Conversely, for patients with low end-reinduction MRD (<0.1%), there was no significant difference in EFS between those with very low MRD (<0.01%) versus those with MRD levels between 0.01% and <0.1%.[21]

Sources: 21

The United Kingdom ALLR3 trial assigned patients who relapsed more than 6 months after completion of front-line therapy to HSCT if their end-reinduction MRD was ≥0.01% or to chemotherapy for those with MRD of <0.01%.[18]

Sources: 18

The United Kingdom ALLR3 trial assigned patients who relapsed more than 6 months after completion of front-line therapy to HSCT if their end-reinduction MRD was ≥0.01% or to chemotherapy for those with MRD of <0.01%.[18]

Of the 228 patients treated, 220 patients achieved CR; 127 patients were allocated to HSCT (high MRD) and 93 were allocated to chemotherapy (low or not evaluable MRD). The 5-year EFS rates were 72% for patients with low MRD versus 56% for patients with high MRD. The 5-year OS rates were 87% for low MRD and 64% for high end-reinduction MRD.

On the basis of these data, investigators from the United Kingdom recommend HSCT for late-relapsing patients with B-ALL who have MRD of ≥0.01% after reinduction therapy and chemotherapy alone for those with end-reinduction MRD of <0.01%.

In the COG AALL1331 (NCT02883049) trial, patients with low-risk relapsed ALL were randomly assigned to receive either standard chemotherapy or standard chemotherapy with three blocks of blinatumomab replacing one block of intensive chemotherapy. Low risk was defined as end-reinduction MRD levels of <0.1% in patients with either bone marrow relapse occurring more than 36 months after diagnosis or extramedullary relapse occurring more than 18 months after diagnosis.[72]

Sources: AALL1331 (NCT02883049) · 72

In the COG AALL1331 (NCT02883049) trial, patients with low-risk relapsed ALL were randomly assigned to receive either standard chemotherapy or standard chemotherapy with three blocks of blinatumomab replacing one block of intensive chemotherapy. Low risk was defined as end-reinduction MRD levels of <0.1% in patients with either bone marrow relapse occurring more than 36 months after diagnosis or extramedullary relapse occurring more than 18 months after diagnosis.[72]

For all 255 patients in the protocol, the 4-year DFS rates were 61.2% for patients who received blinatumomab and 49.5% for patients who received standard chemotherapy (P = .0849). The 4-year OS rates were 90.4% for patients who received blinatumomab, compared with 79.6% for patients who received standard chemotherapy (P = .11).

For patients with bone marrow relapse with or without extramedullary involvement, the 4-year DFS rates were 72.7% for patients who received blinatumomab and 53.7% for those who received standard chemotherapy (P = .015). The 4-year OS rates were 97.1% for patients who received blinatumomab, compared with 84.8% for the chemotherapy-only group (P = .02).

The outcomes were poor for patients with isolated extramedullary relapse, regardless of treatment arm. Outcomes for these patients were inferior to those of similar patients in previous trials. The 4-year DFS rates were 36.6% for patients who received blinatumomab and 38.8% for those who received chemotherapy only (P = .62). The 4-year OS rates were 76.5% for the blinatumomab group, compared with 68.8% for the chemotherapy-only group (P = .53).

Patients who received blinatumomab experienced much lower rates of toxicities such as mucositis, febrile neutropenia, and sepsis than patients who received standard chemotherapy. For example, 11% of patients who received standard chemotherapy developed sepsis, compared with less than 2% of patients who received blinatumomab. Similarly, febrile neutropenia occurred in 3% of patients who received blinatumomab, compared with 48% of those who received standard chemotherapy.

The COG AALL0433 (NCT00381680) trial included patients with intermediate-risk relapse of B-ALL (defined as late marrow-involved relapse occurring >36 months from initial diagnosis, or very early isolated extramedullary relapse occurring <18 months from initial diagnosis). Patients received three induction blocks, followed either by an intensive chemotherapy backbone (with radiation therapy for patients with CNS or testicular involvement) or allogeneic HSCT (if matched-sibling donors were available). The trial also included a randomized comparison of standard versus intensified vincristine dosing. Of the 271 eligible patients, 257 (95%) achieved second CR. Of these patients, 74 (29% of those in second CR) received HSCT, 47 of whom had matched-sibling donors and 27 of whom had other donors.[31]

Sources: AALL0433 (NCT00381680) · 31

The COG AALL0433 (NCT00381680) trial included patients with intermediate-risk relapse of B-ALL (defined as late marrow-involved relapse occurring >36 months from initial diagnosis, or very early isolated extramedullary relapse occurring <18 months from initial diagnosis). Patients received three induction blocks, followed either by an intensive chemotherapy backbone (with radiation therapy for patients with CNS or testicular involvement) or allogeneic HSCT (if matched-sibling donors were available). The trial also included a randomized comparison of standard versus intensified vincristine dosing. Of the 271 eligible patients, 257 (95%) achieved second CR. Of these patients, 74 (29% of those in second CR) received HSCT, 47 of whom had matched-sibling donors and 27 of whom had other donors.[31]

For all 271 eligible patients, the 3-year EFS and OS rates were 63.6% and 72.3%, respectively.

For patients with late marrow-involved relapses (n = 242), the 3-year EFS and OS rates were 66.3% and 74.8%, respectively.

For patients with late marrow relapses, MRD at the end of the first month of treatment (end-reinduction MRD) was a significant prognostic factor. For patients with end-reinduction MRD less than 0.1%, the 3-year EFS rate was 84.9%, versus 53.7% for those with MRD higher than 0.1%.

Adjusting for median time to HSCT, patients who received HSCT had an improved 3-year DFS rate (P = .03), but not OS rate (P = .46), compared with patients who received chemotherapy alone.

The vincristine randomization was stopped early because of excessive toxicity, particularly in older patients.

Among patients treated with either an intensive or nonintensive induction regimen, 96% obtained a second CR.

Approximately one-half of the patients received consolidation therapy, and 78% of the patients proceeded to allogeneic HSCT.

At the time of relapse, the TKI was changed for 23 of the 26 (88%) patients, 15 of whom (58%) received dasatinib. The 4-year EFS rate was 60.9%, and the OS rate was 76.1%.

Total-body irradiation (TBI)-containing transplant preparative regimens.

Sources: Total-body irradiation (TBI)-containing transplant preparative regimens

MRD detection just before transplant.

Sources: MRD detection just before transplant

MRD detection posttransplant.

Sources: MRD detection posttransplant

Donor type and HLA match.

Sources: Donor type and HLA match

Role of GVHD/graft-versus-leukemia (GVL) in ALL and immune modulation after transplant to prevent relapse.

Sources: Role of GVHD/graft-versus-leukemia (GVL) in ALL and immune modulation after transplant to prevent relapse

Two registry studies and a small randomized trial showed that transplant conditioning regimens that included TBI resulted in higher cure rates than chemotherapy-only preparative regimens.[56,90,91]

Sources: 56 · 90 · 91

An international study (United States, Europe, and Australia) that combined data sets from prospective trials and single-center data showed that the use of non-TBI regimens was an independent risk factor for poor outcome.[74,81,92]

Sources: 74 · 81 · 92

An international phase III trial (the FORUM study) conducted between 2013 and 2018 enrolled 417 pediatric patients aged 4 to 21 years with ALL. Patients were randomly assigned to receive either TBI (12 Gy) with etoposide or chemotherapy only (fludarabine, thiotepa, and busulfan or treosulfan) as a preparative regimen. All patients were in morphological CR at the time of transplant and received a 9/10 or 10/10 matched sibling donor (MSD) or unrelated matched donor (URD) transplant. The intent was to enroll 1,000 patients, but the study was closed to enrollment early when a futility early stopping rule was met, indicating that the chemotherapy-only preparative regimen was associated with an inferior outcome.[93]

Sources: 93

An international phase III trial (the FORUM study) conducted between 2013 and 2018 enrolled 417 pediatric patients aged 4 to 21 years with ALL. Patients were randomly assigned to receive either TBI (12 Gy) with etoposide or chemotherapy only (fludarabine, thiotepa, and busulfan or treosulfan) as a preparative regimen. All patients were in morphological CR at the time of transplant and received a 9/10 or 10/10 matched sibling donor (MSD) or unrelated matched donor (URD) transplant. The intent was to enroll 1,000 patients, but the study was closed to enrollment early when a futility early stopping rule was met, indicating that the chemotherapy-only preparative regimen was associated with an inferior outcome.[93]

The 2-year OS rate was 91% for patients who received a TBI-containing regimen, compared with 75% for those who received a chemotherapy-only preparative regimen (P < .001).

The 2-year EFS rate was also superior in patients who received TBI (86% vs. 58%, P < .001), primarily because of a lower cumulative incidence of relapse in the TBI cohort (2-year cumulative incidence of relapse, 12% for TBI patients vs. 33% for chemotherapy-only patients).

In subgroup analysis, the TBI regimen was superior regardless of donor type (MSD vs. URD), immunophenotype (B-ALL vs. T-ALL), and whether transplant occurred in first CR or second CR.

There was no difference in the rates of acute or chronic GVHD or frequency of grades 3 or 4 nonhematologic adverse events between the two randomized treatment arms.

A substudy of the FORUM trial compared two chemotherapy preparative regimens (busulfan/thiotepa/fludarabine with treosulfan/thiotepa/fludarabine) in 191 children (aged <4 years) who were nonrandomly assigned to non-TBI regimens. The two regimens were also nonrandomly assigned by center/country preference, as some countries did not have access to treosulfan.[94]

Sources: 94

A substudy of the FORUM trial compared two chemotherapy preparative regimens (busulfan/thiotepa/fludarabine with treosulfan/thiotepa/fludarabine) in 191 children (aged <4 years) who were nonrandomly assigned to non-TBI regimens. The two regimens were also nonrandomly assigned by center/country preference, as some countries did not have access to treosulfan.[94]

The 3-year probabilities of EFS and OS were 0.52 (95% CI, 0.44–0.59) and 0.69 (95% CI, 0.61–0.76), respectively.

Multivariate analysis showed no difference in EFS and OS according to preparative regimen used, but decreased OS for patients with KMT2A rearrangements and lower EFS in children younger than 1 year at diagnosis.

A historical comparison was conducted between patients aged 2 to 4 years in this trial and patients aged 2 to 4 years who received TBI regimens in two previous international studies performed by the same group. In the ALL-SCT BFM 2003 trial, 3-year probabilities of EFS and OS were 0.83 (95% CI, 0.66–0.92) and 0.91 (95% CI, 0.76–0.97), respectively. In the ALL-SCT BFM International trial, the 3-year probabilities of EFS and OS were 0.87 (95% CI, 0.57–0.97) and 0.93 (95% CI, 0.61–0.99), respectively. In the FORUM trial, the 3-year probabilities of EFS and OS for children who received either chemotherapeutic conditioning regimen were 0.55 (95% CI, 0.44–0.64) and 0.68 (95% CI, 0.57–0.76), respectively (P = .007 and P = .008 for comparison with the TBI-based trials, respectively).

The authors concluded that there is strong evidence that TBI-based regimens are superior in children older than 2 years, and this result should be considered as parents and clinicians evaluate the risks and benefits of TBI in children aged 2 to 4 years.

In one report, 13 patients with ALL and high MRD at the time of planned transplant received an additional cycle of chemotherapy in an attempt to lower MRD before proceeding to HSCT. Ten of the 13 patients (77%) remained in CR post-HSCT, with no relapses observed in the eight patients who achieved low MRD after the additional chemotherapy cycle. In comparison, only 6 of 21 high-MRD patients (29%) who proceeded directly to HSCT without receiving additional pre-HSCT chemotherapy remained in CR.[102]

Sources: 102

In a COG trial of transplant for children with ALL, grades I to III acute GVHD were associated with lower relapse risk (HR, 0.4; P = .04) and better EFS (multivariate analysis, HR, 0.5; P = .02). Any effect of grade IV acute GVHD in decreasing relapse risk was obscured by a marked increase in transplant-related mortality (HR, 6.4; P = .003), while grades I to III acute GVHD had no statistically detectable effect on transplant-related mortality (HR, 0.6; P = .42).[136]

Sources: 136

In a multivariate model, both pretransplant MRD and acute GVHD were independent predictors of relapse, with the lowest risk of relapse observed in patients with both low pretransplant MRD and grades I to III acute GVHD.[112] For patients who did not develop acute GVHD by day 55 post-HSCT, nearly all relapses occurred between days 100 and 400 post-HSCT.

Sources: 112

One study showed that in 46 patients with increasing recipient chimerism, the 31 patients who underwent immune suppression withdrawal, donor lymphocyte infusion, or both therapies had a 3-year EFS rate of 37% versus 0% in the nonintervention group (P < .001).[142]

Sources: 142

Other studies have shown better-than-expected rates of survival of pre-HSCT, MRD-positive patients when tapering of immunosuppression medication has occurred for MRD detected after HSCT.[143]

Sources: 143

A large international study showed a marked decrease in relapse in patients who experienced MRD-positive disease after HSCT and developed acute GHVD (HR, 0.29; P < .001), resulting in improved EFS (HR, 2.9; P = .01). Acute GVHD significantly decreased relapse and improved EFS in both MRD-positive and MRD-negative patients. Chronic GVHD was also associated with less relapse in both MRD-positive and MRD-negative patients.[92]

Sources: 92

Blinatumomab. Blinatumomab is a bispecific monoclonal antibody with one site for CD3 (T cells) and the other site for CD19 (present on most B-ALL cells). Thus, blinatumomab promotes the binding of the patient’s own cytotoxic T cells to B lymphoblasts, resulting in the tumor being killed. In a phase I/II trial of children younger than 18 years with relapsed/refractory B-ALL, 27 of 70 patients (39%) treated at the recommended phase II dose achieved a CR with single-agent blinatumomab; 52% of those achieving CR were MRD negative.[162]In a pooled analysis of five trials that included 166 pediatric and 517 adult patients, those with less than 50% bone marrow blasts at the start of treatment had better responses to blinatumomab. Among pediatric patients, CR rates (including CR with partial hematologic recovery [CRh] and CR with incomplete hematologic recovery [CRi]) were 65.3% for patients with less than 50% baseline bone marrow blast percentage, compared with 38.3% for patients with 50% or more baseline bone marrow blast percentage. Similarly, MRD responses were more frequent in patients with less than 50% baseline bone marrow blast percentage than in those who had 50% or more bone marrow blast percentage (51.4% vs. 25.5%). There was no significant difference in these end points when comparing patients with 5% to <25% blasts and 25% to <50% blasts at the start of blinatumomab treatment.[163]

Sources: 162 · 163

Inotuzumab. Inotuzumab is an anti-CD22 monoclonal antibody that is conjugated to calicheamicin. In trials of adult patients with relapsed/refractory B-ALL, CR was achieved in approximately 80% of patients.[164,165] There have been two phase II trials of single-agent inotuzumab (both trials used 1.8 mg/m2 total dose in the first course and 1.5 mg/m2 in subsequent courses) used in the treatment of pediatric patients with relapsed (second or greater relapse) or refractory ALL.[87]Expert panel recommendations for the prevention of SOS associated with HSCT after inotuzumab include limiting inotuzumab to two courses, avoiding dual-alkylator HSCT regimens, avoiding hepatotoxic agents, and considering SOS prophylactic agents.[167]

Sources: 164 · 165 · 87 · 167

Inotuzumab. Inotuzumab is an anti-CD22 monoclonal antibody that is conjugated to calicheamicin. In trials of adult patients with relapsed/refractory B-ALL, CR was achieved in approximately 80% of patients.[164,165] There have been two phase II trials of single-agent inotuzumab (both trials used 1.8 mg/m2 total dose in the first course and 1.5 mg/m2 in subsequent courses) used in the treatment of pediatric patients with relapsed (second or greater relapse) or refractory ALL.[87]Expert panel recommendations for the prevention of SOS associated with HSCT after inotuzumab include limiting inotuzumab to two courses, avoiding dual-alkylator HSCT regimens, avoiding hepatotoxic agents, and considering SOS prophylactic agents.[167]

In the COG phase II trial (N = 48), the overall response rate (ORR) (CR + CRi) was 58.3% after the first course, and 62.5% after two courses. Of those achieving complete response or complete response with incomplete count recovery, 70% were MRD negative (<0.01%). Overall, inotuzumab was well tolerated, without any significant hepatoxicity or sinusoidal obstructive syndrome (SOS) during treatment. However, of 21 patients who subsequently went on to receive HSCT, 6 (28.6%) developed SOS.[87]

Sources: 87

On a phase II trial conducted by the Innovative Therapies for Children With Cancer group (N = 28), the ORR was 81.5% (all responses occurring after the first cycle). Of the patients achieving a response, 81.8% achieved MRD negativity (<0.01%), 59.1% after the first course, and the remaining patients after the second course. SOS was observed in seven (28%) patients, six after subsequent HSCT and one during inotuzumab therapy. There was no correlation between the extent of CD22 expression (mean fluorescence index or percentage of CD22 positive cells) and response. In a post-hoc analysis, combining patients from the phase I and phase II trials, the authors found that having a shorter interval between last dose of inotuzumab and undergoing HSCT was associated with a higher risk of developing SOS.[166]

Sources: 166

Rituximab or ofatumumab. These monoclonal antibodies target CD20, which is poorly expressed in younger children with ALL, but more common in teenagers and young adults. A number of studies have suggested benefit when adding one of these agents to reinduction chemotherapy in adults,[168,169] but very little data have been published in children.

Sources: 168 · 169