Abstract
Adolescents and young adult (AYA) patients with acute lymphoblastic leukemia (ALL) face worse outcomes than children. While pediatric-inspired protocols have improved outcomes, the ability of patients to complete these intensive regimens and the reasons for discontinuation are unknown. We analyzed a cohort of 332 AYA patients (aged 15-49 years) and 1159 children (aged 1-14 years) with Ph-negative ALL treated on DFCI consortium protocols. We found that AYA patients completed treatment at lower rates than children (60.8% vs. 89.7%, p < 0.001), primarily due to higher rates of early treatment failure (14.5% vs. 2.4%, p < 0.001). Withdrawal from treatment for toxicity, social/personal, or unknown reasons was uncommon, but higher among AYA patients (9.3% vs 4.7%, p= 0.001). Patients who remained on assigned therapy for one year had favorable overall survival (AYA 5-year OS 88.9%; children 5-year OS 96.4%; p < 0.001). Among patients who continued treatment for 1 year, AYA patients completed asparaginase (defined as receiving 26+ weeks) at lower rates than children (79.1% vs. 89.6%, p < 0.001). Patients who received more weeks of consolidation asparaginase had higher overall and event-free survival. Efforts should focus on identifying patients at risk for early treatment failure and optimizing asparaginase delivery.
Introduction
Acute lymphoblastic leukemia (ALL) is the leading cause of cancer death in the United States among adolescents and young adults (AYA) aged 15-29 years and the third leading cause among those aged 30-39 years.1 While ALL is cured in >90% of children,2,3,4 fewer adults survive the disease. Five-year overall survival (OS) for adults ≥ 18 years with newly diagnosed ALL treated on traditional adult protocols is approximately 50%.5,6 Both tumor biology and treatment approach are known to contribute to differential outcomes between children and adults with ALL.7,8 In 2008, Stock et al. observed that adolescents aged 16-20 years with ALL treated on Children’s Cancer Group (CCG) trials from 1998 to 2001 had superior outcomes compared to contemporaneously treated adolescent patients enrolled on adult Cancer and Leukemia Group B (CALBG) trials.9 Multiple prospective studies subsequently demonstrated that young adults up to the age of 40 or 50 years could be successfully treated with pediatric-type regimens with survival superior to similar-aged patients treated on adult protocols.10,11,12 Despite these improvements, survival differences between children and young adults with ALL persist.
The prospective single-arm, phase II CALGB 10403 study, which applied the Children’s Oncology Group (COG) AALL 0232 regimen to 295 young adults aged 17 to 39 years (median age 24 years), is the largest trial of pediatric chemotherapy in AYA patients ever conducted.13 The 3-year overall survival (OS) of patients treated on CALGB 10403 was 73%. Notably, only 39% of patients completed treatment, defined as finishing all phases of the protocol treatment. This is markedly lower than reported in the companion pediatric trial AALL 0232 where 57% of AYA patients (aged 18-30 years) and 74% of patients under 18 years completed protocol treatment.14 This finding raises concern that AYA patients may frequently not be able to complete complex pediatric regimens, either due to toxicity or personal/social reasons, and that treatment discontinuation may contribute to the worse oncologic outcomes of this population, with particular concern about truncation of asparaginase therapy.15, 16, 17
The Dana-Farber Cancer Institute (DFCI) Consortium has conducted trials of chemotherapy for children and adolescents with ALL since the 1970s. Patients treated on the DFCI regimens receive 30 consecutive weeks of asparaginase during post-induction therapy (see methods section for details on asparaginase regimens). Beginning in 2001, the DFCI Consortium began studying these pediatric regimens in older AYA patients and adults up to age 50 years, and found that the DFCI pediatric ALL regimen was tolerable in older AYA patients with disease-free survival (DFS) and OS superior to outcomes of similarly aged patients treated on adult regimens.18, 11 In this report we describe patterns of treatment completion in children, adolescents, and young adults treated on DFCI ALL pediatric regimens to understand the prevalence of and reasons for early treatment discontinuation. We specifically explore prevalence and impact of early asparaginase discontinuation in different age groups.
Methods
Patients:
Patients aged 1-50 years were treated on four sequential multi-center DFCI ALL Consortium protocols between 2000 and 2011. We identified all trial participants for the current analysis. We also identified patients 15 years and older not enrolled on these studies but treated per the same protocols at DFCI/Brigham and Women’s Hospital (BWH), Massachusetts General Hospital (MGH), and Boston Children’s Hospital (BCH) before 2022 using electronic medical record review. Earlier patients were enrolled on parallel pediatric protocol 00-001 (2000-2004) and adult protocol 01-175 (2002-2008) trials while later patients were enrolled on parallel pediatric protocol 05-001 (2005-2011) and adult protocol 06-254 (2007-2011) trials. Patients who were treated on trial signed informed consent. We obtained approval from the DFCI Institutional Review Board to access the charts and abstract the data of patients treated as per protocols 05-001 and 06-254.
Treatment:
The treatment protocols of 00-001, 05-001, and 01-175 have been previously published. 19, 19, 20, 21 In terms of risk stratification: in protocol 00-001, patients aged < 10, with presenting WBC < 50,000, B-precursor phenotype, no mediastinal mass, and CNS1 status or fewer than five leukocytes per high power field in the diagnostic lumbar puncture were considered standard risk (SR); all others were considered high risk (HR). In protocol 05-001, patients aged < 10 years with highest pre-treatment WBC < 50,000, no evidence of CNS leukemia, and absence of t(9;22), KMT2A rearrangement or hypodiploidy were considered standard risk (SR). Patients aged 10 or older, with highest pre-treatment WBC ≥ 50,000, evidence of CNS leukemia, or T-cell immunophenotype were considered HR. Patients with KMT2A rearrangement or hypodiploidy, or positive minimal residual disease (MRD) at the end of induction (defined as ≥ 0.001) were considered very high risk (VHR). The VHR arm of 05-001 incorporated two additional cycles of consolidation (IB and IC) compared to the HR arm. Protocol 01-175 was based on the HR arm of 00-001 and protocol 06-254 was based on the VHR arm of 05-001. The treatment pathways are summarized in Supplementary Table 1. All trials used prednisone during induction. Protocol 00-001 randomized patients between dexamethasone and prednisone during post-induction treatment phases, while the other three trials used dexamethasone. All patients were intended to receive 30 consecutive weeks of asparaginase during post-induction therapy (Figure 1). In the early era studies (protocols 00-001 and 01-175), patients received 30 weekly doses of E. coli asparaginase. In protocol 05-001, pediatric patients were randomized to 30 weeks of weekly E. coli asparaginase or PEG-asparaginase administered every 14 days for 15 doses. In the adult trial 06-254, PEG-asparaginase was administered every 3 weeks for a total of 10 doses.
Figure 1: Schematic representation of patient treatment disposition.
The figure shows the number and percentage of patients in each disposition subgroup. “Completed treatment” includes patients who were able to remain on protocol therapy even if they required early discontinuation of asparaginase in cases of severe pancreatitis, hypersensitivity with no available alternative, and other severe adverse events. “Withdrawal from tx” includes patients who were designated as ineligible, lost to follow up, withdrew for toxicity or other reasons, and those with unknown reasons for protocol exit.
Minimal residual disease:
Minimal residual disease (MRD) at the end of induction chemotherapy was assessed for patients who achieved complete remission. In 00-001, MRD was measured by PCR and not used for treatment decisions. In 05-001, MRD was measured by both PCR and flow cytometry and used for risk stratification. There is no MRD data available in 01-175. In 06-254, MRD was measured by PCR and did not change protocol-specified treatment. For patients treated as per 06-254 and as per 05-001, MRD was measured by PCR and/or flow cytometry and treatment decisions were at the discretion of the treating physician.
Treatment completion:
A patient was designated as having “completed treatment” if they received the full course of treatment (induction, consolidation, and maintenance). Reasons for not completing treatment were categorized as due to induction failure, transplantation at first complete remission (CR), relapse during treatment, death during treatment, and withdrawal from planned treatment. “Withdrawal from planned treatment” included all patients who were taken off study due to clinician decision, personal reasons, social reasons, or toxicity. Our study did not capture additional treatments that patients might have received after withdrawal from planned treatment.
Patients were able to remain on protocol therapy even if they required early discontinuation of asparaginase in cases of severe pancreatitis, hypersensitivity with no available alternative, and other severe adverse events; affected patients completed protocol treatment without remaining asparaginase doses. On the pediatric protocol 05-001, low-risk patients who were not able to receive at least 10 weeks of asparaginase received 3 cycles of high-risk consolidation chemotherapy (doxorubicin and dexrazoxane), and a higher dose of dexamethasone. On the other three protocols, patients who had to discontinue asparaginase early due to toxicity did not have additional treatment modifications. Completion of asparaginase was defined as receipt of more than 26 weeks of asparaginase during consolidation based on a previous study.17 This included patients who received alternative asparaginase preparations.
Statistical analysis:
Chi squared test was used to compare categorical variables. Median follow-up was estimated using the reverse Kaplan Meier method. Kaplan Meier survival analysis was used to plot OS and the log-rank test was used to compare groups. P-values were considered statistically significant if lower than 0.05. Univariate Cox proportional hazard regression was performed to identify associations between patient covariates and overall survival. Variables that were found to be statistically associated with survival were then used in a multivariable Cox proportional hazard regression using backwards selection, in which variables are sequentially eliminated if they are not statistically associated with the outcome. Cumulative incidence of relapse with death as a competing risk was plotted, and the groups were compared using the Gray test.
Results
Patients
Between 2000 and 2021, 1491 Philadelphia-chromosome negative ALL patients were treated on or as per pediatric DFCI Consortium trials. This included 332 AYA patients, aged 15-50 years at time of diagnosis, and 1159 children, aged 1-14 years at diagnosis (Table 1). Among all patients, 37.4% were treated on the early-era protocols 00-001 and 01-175 with native E. coli asparaginase, while 62.6% were treated on or as per the later-era protocols 05-001 and 06-254 using pegylated asparaginase. The median age of AYA patients was 23 years and the median age of children was 4.5 years. There were several differences between AYA patients and children: among AYA patients, 62.7% were male, compared to 53.9% of children (p = 0.005). AYA patients had a higher proportion of T-cell lineage leukemia (28.0%) compared to children (10.0%, p< 0.001), and more frequently presented with a mediastinal mass (21.6% for AYA patients compared to 5.4% for children, p < 0.001). Cytogenetic profiles also differed, with AYA patients having higher incidence of KMT2A-rearranged (7.2% vs. 3.1%, p = 0.001) leukemia, and lower incidence of hyperdiploid leukemia (10.2% vs. 26.1%, p < 0.001). AYA patients were more likely to be overweight or obese at diagnosis (43.7% combined) compared to children (24.3%, p < 0.001).
Table 1:
Patient characteristics
| AYA patients (15-49 years) | Children (1-14 years) | All patients | |
|---|---|---|---|
|
|
|||
| Total, eligible patients | 332 | 1159 | 1491 |
| Protocol/risk stratification | |||
| 00-001 | 32 (9.6) | 451 (38.9) | 483 (32.4) |
| Standard risk | 283 (62.8) | 283 (58.6) | |
| High risk | 32(100) | 168 (37.2) | 200 (41.4) |
| 05-001 (10 pts as per) | 74 (22.3) | 708 (61.1) | 782 (52.4) |
| Standard risk | 418 (59.0) | 418 (53.5) | |
| High risk | 68 (91.9) | 230 (32.5) | 298 (38.1) |
| Very high risk | 6 (8.1) | 60 (8.5) | 66 (8.4) |
| 01-175 | 74 (22.3) | 74 (5.0) | |
| 06-254 (63 pts as per) | 152 (45.8) | 152 (10.2) | |
| Age, median (range) | 23 (15-51) | 4.5 (1-14.9) | 5.8 (1-50) |
| WBC, median (range) | 10.9 (.7-1424) | 11.5 (0.6-905) | 11.4 (0.6-1424) |
| Blast %, median (range) (n = 1446) | 36 (0,98) | 28 (0,99) | 29 (0,99) |
| Sex | |||
| Male | 208 (62.7) | 625 (53.9) | 833 (55.9) |
| Female | 124 (37.3) | 534 (46.1) | 658 (44.1) |
| Immunophenotype | |||
| B-cell | 237 (71.4) | 987 (85.2) | 1224 (82.1) |
| T-Cell 1 | 93 (28.0) | 116 (10.0) | 209 (14.0) |
| Mixed/Other | 2 (0.6) | 56 (4.8) | 58 (3.9) |
| CNS status | |||
| CNS 1 | 257 (77.4) | 926 (79.9) | 1183 (79.3) |
| CNS 2 | 41 (12.3) | 158 (13.6) | 199 (13.4) |
| CNS 3 | 10 (3.0) | 23 (2.0) | 33 (2.2) |
| Traumatic/Unknown | 24 (7.3) | 52 (4.5) | 76 (5.1) |
| Mediastinal mass (n = 1481) | |||
| Yes | 71 (21.6) | 62 (5.4) | 133 (9.0) |
| No | 257 (78.4) | 1091 (94.6) | 1348 (91.0) |
| Cytogenetics | |||
| Normal | 98 (29.5) | 219 (18.9) | 317 (21.3) |
| MLL rearranged | 24 (7.2) | 36 (3.1) | 60 (4.0) |
| Hypodiploid | 0 | 10 (0.9) | 10 (0.7) |
| Hyperdiploid | 34 (10.2) | 302 (26.1) | 336 (22.5) |
| Complex karyotype | 7 (2.1) | 10 (0.9) | 17 (1.1) |
| Other abnormality | 121 (36.5) | 476 (41.1) | 597 (40.0) |
| Unknown | 48 (14.5) | 106 (9.2) | 154 (10.3) |
| BMI at diagnosis | |||
| Normal/Underweight | 187 (56.3) | 781 (67.4) | 968 (65.0) |
| Overweight | 79 (23.8) | 147 (12.7) | 226 (15.2) |
| Obese | 66 (19.9) | 134 (11.6) | 200 (13.4) |
| Not evaluable/unknown 2 | 97 (8.4) | 97 (6.5) | |
| MRD end-induction (n = 799 pts achieving CR w/ available MRD data) | |||
| Negative | 104 (83.9) | 619 (91.7) | 723 (90.5) |
| Positive | 20 (6.1) | 56 (8.3) | 76 (9.5) |
Patient characteristics.
Includes 4 early T-cell patients.
Includes 96 children younger than 2 and 1 child with unknown height at time of diagnosis. BMI: Body Mass Index. MRD: Minimal Residual Disease. CR: Complete Remission.
Patients treated on pediatric protocol 00-001 and on or as per 05-001 underwent risk stratification based on characteristics at presentation and response to induction chemotherapy. Of 1,255 patients treated on those protocols, 701 (55.9%) were designated SR, 488 (38.9%) as HR, and 66 (5.2%) as VHR. All AYAs were designated HR or VHR. Protocol 01-175 followed the HR arm of 00-001 and 06-254 followed the VHR arm of 05-001. Of the 1,491 patients included in our study, 1,414 (94.8%) achieved a complete remission. MRD data was available for 799 (56.5%) patients; among those, 723 (90.5%) were MRD-negative at the end of induction and 76 (9.5%) were MRD-positive.
Treatment completion
Of the 1491 patients, 1241 (83.2%) completed treatment, including 150 (12.1%) whose post-remission asparaginase was truncated. Treatment completion rates were lower for AYA patients aged 15-50 years (60.8%) compared to children less than 15 years (89.7%) (p < 0.001) (Table 2, Figure 1). Among 185 AYA patients aged 18-39 years at time of diagnosis, 57.3% completed treatment (data not shown). Lower treatment completion rates in AYA patients were primarily driven by a higher rate of early treatment failure, defined as induction failure or relapse within 1 year, which occurred in 48 (14.5%) AYA patients compared to 28 (2.4%) children (p < 0.001). Additionally, AYA patients were more likely to receive a BMT in first complete remission (n=28, 8.4%) compared to children (n=0, 0%, p < 0.001). Of note, the protocols did not recommend BMT in first remission for any patient subgroup included in this analysis (Philadelphia-chromosome negative ALL). Among 14 patients who underwent BMT in first CR and had available end-induction MRD information, 4 (29%) were MRD-positive at the end of induction. Withdrawal from planned treatment for reasons related to toxicity, social, or unknown reasons was rare but more common among AYA patients (n=31, 9.3%) compared to children (n=54, 4.7%) (p= 0.001). Of the 85 patients who withdrew from planned treatment, 49 (57%) did so within the first year of treatment (68% among AYA patients and 52% among children). Supplementary Table 2 shows further detail on the reasons for protocol exit for these 85 patients. 35 patients died during the first year of treatment, with 21 deaths occurring during induction (6 AYAs and 15 children). Relapse or death within the second and third year after initiation of therapy was also rare (2.4% of AYA patients, 1.6% of children, p = 0.29).
Table 2:
Patient treatment disposition by age group
| AYA | Children | All patients | |
|---|---|---|---|
|
|
|||
| Completed treatment 1 | 202 (60.8) | 1039 (89.7) | 1241 (83.2) |
| Transplant at CR1 | 28 (8.4) | 28 (1.9) | |
| Year 1 of protocol 2 | |||
| Induction failure | 26 (7.8) | 21 (1.8) | 47 (3.2) |
| Relapse during treatment | 22 (6.6) | 7 (0.6) | 29 (2.0) |
| Death during treatment 4 | 15 (4.5) | 20 (1.7) | 35 (2.4) |
| Withdrawal from treatment 3 | 21 (6.3) | 28 (2.4) | 49 (3.3) |
| Year 2 and 3 of protocol | |||
| Relapse during treatment | 7 (2.1) | 18 (1.6) | 25 (1.7) |
| Death during treatment | 1 (0.3) | 1 (0.1) | |
| Withdrawal from treatment 3 | 10 (3.0) | 26 (2.2) | 36 (2.4) |
|
| |||
| Total | 332 | 1159 | 1491 |
Patient treatment disposition by age group. The table shows the number and percentage of patients in each disposition subgroup.
Includes patients who were able to remain on treatment even if they required early discontinuation of asparaginase in cases of severe pancreatitis, hypersensitivity with no available alternative, and other severe adverse events
Within 1 year from trial registration or diagnosis (for as-per patients).
Withdrawal from treatment includes patients who were designated as ineligible, lost to follow up, withdrew for toxicity or other reasons, and those with unknown reasons for protocol exit.
Among 35 patients who died during the first year of treatment, 21 deaths occurred during induction, before remission was achieved (6 AYAs and 15 children).
Survival outcomes
Among patients who experienced early treatment failure (induction failure or relapse within 1 year of diagnosis), outcomes were poor: at 5 years, overall survival (OS) was 23.4% for AYA patients (95% CI 11.7-37.5) and 33.8% for children (95% CI 16.5-52.1) (log-rank p-value 0.4) (Supplementary Figure 1a). Conversely, patients who completed at least one year of protocol assigned treatment had favorable survival, which was statistically better among children: 5-year OS among AYA patients 88.9% (95% CI 83.4-92.7) versus children 96.4% (95% CI 95.2-97.4), p < 0.001 (Figure 2). In that group, 5-year event free survival (EFS) was 82.7% among AYA patients and 90.1% among children (p = 0.0025). Notably patients without early treatment failure who withdrew from planned treatment for toxicity, social, or unknown reasons also had favorable long-term outcomes: 5-year OS among AYA patients 87.8% (95% CI 59.5-96.8) and among children 94.2% (95% CI 77.9-98.6) (p = 0.2) (Supplementary Figure 1b). Our study did not record post-protocol treatment for these patients.
Figure 2:
Long term outcomes of patients who stayed on protocol for at least 1 year. This includes patients who discontinued asparaginase during consolidation. a) Overall survival. b) Event free survival.
Asparaginase completion
To determine the prevalence and impact of asparaginase truncation during ALL treatment, we focused on 1303 patients who stayed on the protocol assigned treatment for at least 1 year. This excluded patients who experienced induction failure, were transplanted in first CR, experienced early relapse, or death within 1 year, or were lost to follow-up/withdrew within the first year. We defined asparaginase completion as receipt of more than 26 weeks of asparaginase during consolidation. We found that AYA patients completed asparaginase at lower rates than children (79.1% vs. 89.6%, p < 0.001) (Table 3). Among the 46 AYA patients who did not complete 26 weeks of asparaginase, 26 stopped due to toxicity (including 8 cases of pancreatitis and 1 case of allergy), and 20 for unknown reasons.
Table 3.
| AYA | Children | All patients | |
|---|---|---|---|
| Completed asparaginase | 174 (79.1) | 970 (89.6) | 1144 (87.8) |
| Required switch to Erwinia | 18 (8.1) | 27 (2.4) | 45 (3.4) |
| Did not complete asparaginase | 46 (20.9) | 113 (10.4) | 159 (12.2) |
| Toxicity NOS | 17 (7.7) | 56 (5.2) | 73 (5.6) |
| Allergy | 1 (0.5) | 2 (0.2) | 3 (0.2) |
| Pancreatitis | 8 (3.6) | 1 (0.1) | 9 (0.7) |
| Unknown discontinuation reason | 20 (9.1) | 54 (5) | 74 (5.7) |
Asparaginase completion status (defined as receiving > 26 weeks of post-induction asparaginase) by age group. NOS: Not otherwise specified. N = 1303 patients who remained on protocol treatment for at least 1 year
Association of asparaginase completion and survival
In a Kaplan Meier analysis, the OS at 5 years among all 1303 patients included in this analysis was higher for those who completed asparaginase (95.8%, 95% CI 94.4-96.7) than those who did not (91.0%, 95% CI 84.9-94.7) (p = 0.027) (Figure 3). When patients were split into additional categories, those with 0-4 weeks of consolidation asparaginase had a 5-year OS of 80.9%, compared to 91.9% for those with 5-14 weeks, 94.6% for those with 15-25 weeks, and 95.8% for those with 26 weeks or more (log rank p-value = 0.0234) (Supplementary Figure 2a). The trend towards higher OS for those who completed asparaginase was present, but not statistically significant, when the analyses were done separately for AYA patients and children (Supplementary Figure 2b,c).
Figure 3:
Overall survival of patients based on asparaginase completion status. Asparaginase completion defined as receipt of 26 weeks of asparaginase in consolidation. This includes patients who had to switch asparaginase preparation to Erwinia. n = 1303 patients who remained on protocol therapy for at least 1 year.
To assess the impact of asparaginase after controlling for other co-variates, we performed univariate and multivariable Cox proportional hazard analyses with OS as the dependent variable. We included the 1303 patients who stayed on the protocol assigned treatment for at least 1 year. We built models with asparaginase completion as a binary variable (26 weeks or more vs. less) and as a continuous variable of the number of consolidation asparaginase weeks received. In the univariate analysis, we found that asparaginase (both as a binary and continuous variable), WBC at diagnosis (> 30 *109/L vs. not), age group (child vs. AYA), BMI (overweight/obese vs. normal), and hyperdiploid karyotype were statistically associated with OS (Table 4). In a multivariable analysis using backward selection of variables, asparaginase as a continuous variable, age group, and WBC remained significantly associated with OS. For every week of consolidation asparaginase received, patients had a 3% lower hazard of death in the multivariable analysis.
Table 4:
| Univariate HR [95% CI] |
p-value | Multivariable HR Model 1 [95% CI] |
p-value | Multivariable HR Model 2 [95% CI] |
p-value | |
|---|---|---|---|---|---|---|
| Asparaginase completion (yes vs. no) | 0.54 [0.30-0.94] | 0.029 | 0.67 [0.38-1.18] | 0.16 | - | - |
| Weeks of asparaginase | 0.96 [0.94-0.99] | 0.003 | - | - | 0.97 [0.95-0.997] | 0.030 |
| Switched to Erwinia (yes vs. no) | 0.83 [0.36-1.90] | 0.66 | ||||
| Age (AYA vs. child) | 2.98 [1.85-4.80] | <0.0001 | 2.77 [1.71-4.49] | <0.0001 | 2.69 [1.66-4.37] | <0.0001 |
| Sex (Male vs. Female) | 1.21 [0.77-1.89] | 0.41 | - | - | - | - |
| BMI (overweight/obese vs. normal) | 1.76 [1.10-2.83] | 0.018 | - | - | - | - |
| WBC (>30 vs. ≤30*109/L) | 2.38 [1.53-3.72] | <0.0001 | 2.29 [1.47-3.58] | <0.0001 | 2.27 [1.46-3.55] | <0.0001 |
| CNS disease (CNS-2 or CNS-3 vs. CNS-1) | 1.25 [0.71-2.19] | 0.45 | - | - | - | - |
| Immunophenotype | ||||||
| B-ALL vs. T-ALL | 1.29 [0.59-2.81] | 0.52 | - | - | - | - |
| Mixed/Other vs. T-ALL | 1.57 [0.46-5.36] | 0.47 | - | - | - | - |
| Cytogenetics | ||||||
| Complex (yes vs. no) | 1.13 [0.16-8.14] | 0.90 | - | - | - | - |
| Hypodiploid (yes vs. no) | 4.04 [0.99-16.45] | 0.051 | - | - | - | - |
| Hyperdiploid (yes vs. no) | 0.51 [0.28-0.94] | 0.032 | - | - | - | - |
| MLL rearranged (yes vs. no) | 2.47 [1.00-6.11] | 0.050 | - | - | - | - |
| Other abnormality (yes vs. no) | 0.98 [0.62-1.54] | 0.92 | - | - | - | - |
| Treatment protocols (newer vs. older) | 0.81 [0.52-1.27] | 0.37 | - | - | - | - |
Cox proportional hazard regression with overall survival as a dependent variable incorporating asparaginase completion. Multivariable model #1 contains asparaginase completion (yes vs. no), while multivariable model #2 contains number of weeks of asparaginase.
We also investigated the association between asparaginase completion and event-free survival (EFS) (Table 5). In univariate analyses, asparaginase completion (both as a binary and continuous variable), age group, WBC at diagnosis, B-cell (vs. T-cell) immunophenotype, and treatment protocols (newer vs. older) were associated with EFS. In a multivariable model, the same variables retained statistical significance. For every week of consolidation asparaginase received, patients had a 2% lower hazard of relapse, death, or second malignancy in the multivariable analysis.
Table 5:
| Univariate HR [95% CI] |
p-value | Multivariable HR Model 1 [95% CI] |
p-value | Multivariable HR Model 2 [95% CI] |
p-value | |
|---|---|---|---|---|---|---|
| Asparaginase completion (yes vs. no) | 0.69 [0.45-1.07] | 0.10 | 0.75 [0.48-1.17] | 0.21 | - | - |
| Weeks of asparaginase | 0.97 [0.95-0.99] | 0.011 | - | - | 0.98 [0.96-0.998] | 0.034 |
| Switched to Erwinia (yes vs. no) | 0.94 [0.53-1.66] | 0.83 | - | - | - | - |
| Age (AYA vs. child) | 1.50 [1.01-2.24] | 0.047 | 1.78 [1.17-2.69] | 0.007 | 1.73 [1.14-2.62] | 0.009 |
| Sex (Male vs. Female) | 1.30 [0.94-1.80] | 0.11 | - | - | - | - |
| BMI (overweight/obese vs. normal) | 1.25 [0.88-1.79] | 0.22 | - | - | - | - |
| WBC (>30 vs. ≤30*109/L) | 1.63 [1.17-2.27] | 0.004 | 1.86 [1.33-2.61] | <0.0001 | 1.86 [1.33-2.60] | <0.0001 |
| CNS disease (CNS-2 or CNS-3 vs. CNS-1) | 0.94 [0.60-1.46] | 0.77 | - | - | - | - |
| Immunophenotype B- vs. T-ALL | ||||||
| B-ALL vs. T-ALL | 2.19 [1.12-4.30] | 0.023 | 3.10 [1.55-6.19] | 0.001 | 3.08 [1.54-6.15] | 0.001 |
| Mixed/Other vs. T-ALL | 2.03 [0.72-5.69] | 0.18 | 2.71 [0.94-7.80] | 0.07 | 2.70 [0.94-7.79] | 0.07 |
| Cytogenetics | ||||||
| Complex (yes vs. no) | 1.82 [0.58-5.72] | 0.30 | - | - | - | - |
| Hypodiploid (yes vs. no) | 1.88 [0.46-7.65] | 0.38 | - | - | - | - |
| Hyperdiploid (yes vs. no) | 0.73 [0.49-1.08] | 0.11 | - | - | - | - |
| MLL rearranged (yes vs. no) | 1.71 [0.80-3.66] | 0.17 | - | - | - | - |
| Other abnormality (yes vs. no) | 1.24 [0.90-1.71] | 0.18 | - | - | - | - |
| Treatment protocols (newer vs. older) | 0.71 [0.52-0.97] | 0.031 | 0.70 [0.51-0.97] | 0.032 | 0.70 [0.50-0.96] | 0.028 |
Cox proportional hazard regression with event-free survival as a dependent variable incorporating asparaginase completion. Multivariable model #1 contains asparaginase completion (yes vs. no), while multivariable model #2 contains number of weeks of asparaginase.
Lastly, we investigated whether asparaginase completion is associated with leukemia relapse. We performed a cumulative incidence of relapse (CIR) analysis with death as a competing risk (Figure 4). We showed that those who completed asparaginase had a numerically lower CIR at 5 years (9.1%, 95% CI 7.5-10.9) compared to those who did not (11.2%, 95% CI 6.8-16.9), but this difference did not meet statistical significance (p = 0.08).
Figure 4:
Cumulative incidence of relapse with death as competing risk by asparaginase completion status (defined as receiving > 26 weeks of post-induction asparaginase). N=1303 patients who remained on protocol without relapse for at least 1 year.
Discussion:
In this retrospective analysis of 332 AYA patients and 1159 children with ALL treated on or as per DFCI Consortium pediatric protocols, we found that AYA patients completed their treatment at significantly lower rates than children and determined that lack of completion was driven primarily by higher rates of induction failure, early relapse, and transplant at first CR among AYA patients. We found that treatment withdrawal for regimen-related toxicity or social/personal reasons was less frequent than biological treatment failure but higher among AYA patients than children. Regarding asparaginase, we found that AYA patients without early treatment failure were frequently able to complete asparaginase consolidation, but at lower rates than children due to more frequent toxicity. The number of asparaginase weeks received was associated with superior OS and EFS in this cohort.
Overall, we found that AYA patients in the DFCI Consortium completed treatment at significantly lower rates than children but found that they completed treatment at substantially higher rates (60.8% overall, 72.9% in those without early biologic treatment failure or transplant in first CR) than those previously reported by the CALGB 10403 investigators (39%).13 This difference appears to be driven by more patients in the CALGB cohort withdrawing from study for toxicity, social/personal or unknown reasons compared to DFCI Consortium patients: 27% of patients in CALGB 10403 vs. 9% of AYA patients in our study fall in this category. Rates of induction failure, death while on treatment, and relapse while on treatment are similar between the two studies. The improved treatment completion may account for the higher DFS and OS in the AYA patients treated on DFCI consortium trials.
The etiology of the difference in rates of treatment completion between the CALGB and DFCI cohorts cannot be ascertained from our study. Potential hypotheses include differences in patient and treatment center characteristics. The DFCI Consortium is made up of 27 centers with experience treating pediatric ALL patients. This more limited network and center/clinician experience may have improved the ability of care teams to support young adults through pediatric-style ALL therapy. In contrast, the CALGB protocol was open at 71 centers, many with smaller case volume and potentially less experience using pediatric-style ALL regimens. Additionally, more patients enrolled in DFCI Protocols may have resided in locations, including Massachusetts, USA, with high insurance rates, higher median household income, and more robust social services. Overall, these findings suggest treatment at high volume, experienced centers may increase the chance a young adult is able to complete recommended pediatric-style ALL treatment which has implications for referral patterns and insurance coverage.
Our study found that AYA patients experienced early treatment failure (including induction failure and early relapse within one year) at a significantly higher rate than children and that AYA patients and children who experienced early treatment failure had similarly poor outcomes. This suggests that the major factor in worse outcomes in AYA ALL is driven by disease biology and determined early. Efforts should be focused on identifying these patients early, via improved characterization of genetic subgroups (Philadelphia chromosome-like, etc.), and refined minimal residual disease (MRD) monitoring, and directing these patients to alternate consolidation approaches such as novel agents and/or BMT.
Conversely, AYA patients who remained on protocol-assigned treatment for at least 1 year (with the majority completing all 2 years of therapy) without experiencing early treatment failure had very favorable oncologic outcomes, with 88.9% experiencing long-term overall survival and 82.7% event free survival at 5 years, supporting the application of a pediatric regimen to AYA patients, where cure can be achieved without transplantation.
The DFCI Consortium protocols uniquely consolidate patients with 30 weeks of continuous asparagine depletion. We showed that AYA patients completed asparaginase at lower rates than children (79.1% vs. 89.6%), primarily due to toxicity, but rate of completion is quite high overall highlighting the ability to apply asparaginase-intensive DFCI pediatric regimens to AYA patients. At DFCI, adult oncologists practice in proximity and in collaboration with the pediatric center allowing knowledge about management of asparaginase toxicities to be easily shared, possibly accounting for the high asparaginase completion rates among young adults.
The OS benefit of asparaginase completion among patients without early treatment failure is consistent with prior findings from the DFCI group in children,16 as well as studies from other pediatric treatment protocols, such as AALL0232, and NOPHO.15, 16 In CALGB 10403, a numerically lower OS for patients who discontinued asparaginase early was found but that finding did not reach statistical significance.22 In our dataset, we observe a statistically significant association between the number of consolidation asparaginase weeks and both OS and EFS.
Our study has several limitations. First, its retrospective nature makes causal inference difficult. Our large dataset and ability to compare similarly treated children and AYA patients helps to address this. Second, we do not have complete data on the reasons for withdrawal from treatment or for asparaginase discontinuation for all patients. Notably, the total number of patients who fell into those categories is relatively small. Third, information on the treatments that patients received after withdrawal from, or completion of the protocol is not available, except for receipt or non-receipt of BMT. Thus, oncologic outcome analyses might be confounded by post-protocol treatments. Fourth, AYAs were exposed to higher intensity chemotherapy than children, as all AYAs were assigned to at least the high-risk arms of their treatment protocols. Lastly, our study cannot assess the impact of race and ethnicity on treatment completion and outcomes, given a primarily white patient cohort.
This study of a large cohort of children and AYA patients treated on DFCI Consortium trials demonstrates that treatment of AYA patients on pediatric inspired protocols can achieve high treatment completion rates, excellent survival, and very low treatment withdrawal rates for toxicity or social/personal reasons. Given that most of the poor outcomes among AYA patients in our study were driven by early treatment failure, ongoing efforts to identify high risk patients early and allocate them to novel approaches will be crucial to improve the outcomes of this population. For patients without early treatment failure, the need for novel agents in consolidation, which has been demonstrated to be beneficial in older patients, is less certain given extremely low risk of relapse. Finally, ongoing research in optimizing delivery of asparaginase in AYA patients and reducing the incidence of toxicities resulting in asparaginase truncation will also help further improve outcomes in responding patients with the hope of closing the gap between children and AYA patients in the future. Lastly, further research is needed to determine the optimal treatment of patients who cannot tolerate asparaginase and whether novel approaches, such as the incorporation of blinatumomab consolidation, can be used to improve outcomes.23
Supplementary Material
Disclosure of Conflicts of Interest
YV received a consultancy fee from EastRx. LS was on the advisory board of Jazz Pharma, Takeda, Servier, and Syndax. AB received consultancy funding from Acceleron Pharma, Biogen, Celgene/BMS, Forty Seven, Jazz Pharma, Novartis, Takeda, and Xcenda, and research funding from Celgene/BMS, Novartis, Takeda, GSK, Janssen, and Astra Zeneka. DD received consultancy funding from Amgen, Autolos, Agios, Blueprint Medicines Corporation, Forty Seven, Incyte Corporation, Jazz Pharma, Novartis, Pfizer, Shire, and Takeda, and research funding from Bluperint Medicines Corporation, Novartis, Abbvie, and Glycomimetics. MRL received research funding from AbbVie, Novartis and was on advisory boards Pfizer, Novartis.
Data availability statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request as long as privacy considerations are maintained.
References
- 1.U.S. Cancer Statistics Working Group. U.S. Cancer Statistics Data Visualizations Tool, based on November 2018 submission data (1999–2016): U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute; www.cdc.gov/cancer/dataviz, June 2019. [Google Scholar]
- 2.Vrooman LM, Blonquist TM, Stevenson KE, Supko JG, Hunt SK, Cronholm SM, et al. Efficacy and Toxicity of Pegaspargase and Calaspargase Pegol in Childhood Acute Lymphoblastic Leukemia: Results of DFCI 11-001. J Clin Oncol. 2021. Nov 1;39(31):3496–3505. doi: 10.1200/JCO.20.03692. Epub 2021 Jul 6. [DOI] [PubMed] [Google Scholar]
- 3.Hunger SP, Lu X, Devidas M, Camitta BM, Gaynon PS, Winick NJ, et al. Improved survival for children and adolescents from 1990-2005: a report from the Children’s Oncology Group. J Clin Oncol 2012;30:1663–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pui CH, Campana D, Pei D, Bowman WP, Sandlund JT, Kaste SC, et al. Treating childhood acute lymphoblastic leukemia without cranial irradiation. N Engl J Med 2009;360:2730–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Patel B, Rai L, Buck G, Richards SM, Mortuza Y, Mitchell W, et al. Minimal residual disease is a significant predictor of treatment failure in non T-lineage adult acute lymphoblastic leukaemia: final results of the international trial UKALL XII/ECOG2993. Br J Haematol. 2010. Jan;148(1):80–9. doi: 10.1111/j.1365-2141.2009.07941.x. Epub 2009 Oct 26. [DOI] [PubMed] [Google Scholar]
- 6.Larson RA, Dodge RK, Burns CP, Lee EJ, Stone RM, Schulman P, et al. A five-drug remission induction regimen with intensive consolidation for adults with acute lymphoblastic leukemia: cancer and leukemia group B study 8811. Blood. 1995. Apr 15;85(8):2025–37. [PubMed] [Google Scholar]
- 7.Roberts KG. Genetics and prognosis of ALL in children vs adults. Hematology Am Soc Hematol Educ Program. 2018. Nov 30;2018(1):137–145. doi: 10.1182/asheducation-2018.1.137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Roberts KG, Li Y, Payne-Turner D, Harvey RC, Yang YL, Pei D, et al. Targetable kinase-activating lesions in Ph-like acute lymphoblastic leukemia. N Engl J Med. 2014. Sep 11;371(11):1005–15. doi: 10.1056/NEJMoa1403088 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Stock W, La M, Sanford B, Bloomfield CD, Vardiman JW, Gaynon, et al. ; Children's Cancer Group; Cancer and Leukemia Group B studies. What determines the outcomes for adolescents and young adults with acute lymphoblastic leukemia treated on cooperative group protocols? A comparison of Children's Cancer Group and Cancer and Leukemia Group B studies. Blood. 2008. Sep 1;112(5):1646–54. doi: 10.1182/blood-2008-01-130237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Stock W, Luger SM, Advani AS, Geyer S, Harvey R, Mullighan C, et al. Favorable outcomes for older adolescents and young adults with acute lymphoblastic leukemia: early results of U.S. Intergroup trial C10403. Blood. 2014;124(21):796. Abstract 796. [Google Scholar]
- 11.DeAngelo DJ, Stevenson KE, Dahlberg SE, Silverman LB, Couban S, Supko JG,, et al. Long-term outcome of a pediatric-inspired regimen used for adults aged 18-50 years with newly diagnosed acute lymphoblastic leukemia. Leukemia. 2015;29(3):526–534. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Huguet F, Leguay T, Raffoux E, Thomas X, Beldjord K, Delabesse E, et al. Pediatric-inspired therapy in adults with Philadelphia chromosome-negative acute lymphoblastic leukemia: the GRAALL-2003 study. J Clin Oncol. 2009;27(6):911–918. [DOI] [PubMed] [Google Scholar]
- 13.Stock W, Luger SM, Advani AS, Yin J, Harvey RC, Mullighan CG, et al. A pediatric regimen for older adolescents and young adults with acute lymphoblastic leukemia: results of CALGB 10403. Blood. 2019. Apr 4;133(14):1548–1559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Advani AS, Larsen E, Laumann K, Luger SM, Liedtke M, Devidas M, et al. Comparison of CALGB 10403 (Alliance) and COG AALL0232 toxicity results in young adults with acute lymphoblastic leukemia. Blood Adv. 2021. Jan 26;5(2):504–512. doi: 10.1182/bloodadvances.2020002439. . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gupta S, Wang C, Raetz EA, Schore R, Salzer WL, Larsen EC, et al. Impact of Asparaginase Discontinuation on Outcome in Childhood Acute Lymphoblastic Leukemia: A Report From the Children’s Oncology Group. J Clin Oncol. 2020. Jun 10;38(17):1897–1905. doi: 10.1200/JCO.19.03024. Epub 2020 Apr 10. . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gottschalk Højfeldt S, Grell K, Abrahamsson J, Lund B, Vettenranta K, Jønsson ÃG, et al. Relapse risk following truncation of pegylated asparaginase in childhood acute lymphoblastic leukemia. Blood. 2021. Apr 29;137(17):2373–2382. doi: 10.1182/blood.2020006583. . [DOI] [PubMed] [Google Scholar]
- 17.Silverman LB, Gelber RD, Dalton VK, Asselin BL, Barr RD, Clavell LA, et al. : Improved outcome for children with acute lymphoblastic leukemia: Results of Dana-Farber Consortium Protocol 91- 01. Blood 97:1211–1218, 2001 [DOI] [PubMed] [Google Scholar]
- 18.Barry E, DeAngelo DJ, Neuberg D, Stevenson K, Loh ML, Asselin BL, et al. Favorable outcome for adolescents with acute lymphoblastic leukemia treated on Dana-Farber Cancer Institute Acute Lymphoblastic Leukemia Consortium Protocols. J Clin Oncol. 2007. Mar 1;25(7):813–9. doi: 10.1200/JCO.2006.08.6397. . [DOI] [PubMed] [Google Scholar]
- 19.Vrooman LM, Blonquist TM, Harris MH, Stevenson KE, Place AE, Hunt SK, et al. Refining risk classification in childhood B acute lymphoblastic leukemia: results of DFCI ALL Consortium Protocol 05-001. Blood Adv. 2018. Jun 26;2(12):1449–1458. doi: 10.1182/bloodadvances.2018016584. . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Vrooman LM, Stevenson KE, Supko JG, O'Brien J, Dahlberg SE, Asselin BL, et al. Postinduction dexamethasone and individualized dosing of Escherichia Coli L-asparaginase each improve outcome of children and adolescents with newly diagnosed acute lymphoblastic leukemia: results from a randomized study--Dana-Farber Cancer Institute ALL Consortium Protocol 00-01. J Clin Oncol. 2013;31(9):1202–1210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.DeAngelo D, Dahlberg S, Silverman L, Couban S, Amrein P, Seftel M, et al. A Multicenter Phase II Study Using a Dose Intensified Pediatric Regimen in Adults with Untreated Acute Lymphoblastic Leukemia. Blood. 2015;110. 587–587. [Google Scholar]
- 22.Aldoss I, Yin J, Wall A, Mrózek K, Liedtke M, Claxton DF, et al. The impact of early PEG-asparaginase discontinuation in young adults with ALL: a post hoc analysis of the CALGB 10403 study. Blood Adv. 2023. Jan 24;7(2):196–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Litzow MR, Sun Z, Paletta E, Mattison R, Lazarus H, Rowe J, et al. Consolidation therapy with blinatumomab improves overall survival in newly diagnosed adult patients with B-lineage acute lymphoblastic leukemia in measurable residual disease negative remission: Results from the ECOG-ACRIN E1910 randomized phase III National Cooperative Clinical Trials Network trial. Blood. 2022. 140 (Supplement 2):LBA-1. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request as long as privacy considerations are maintained.