Universal premedication and therapeutic drug monitoring for asparaginase-based therapy prevents infusion-associated acute adverse events and drug substitutions

Abstract

Background:

Asparaginase is a critical component of lymphoblastic leukemia therapy, with intravenous pegaspargase (PEG) as the current standard product. Acute adverse events (aAEs) during PEG infusion are difficult to interpret, representing a mix of drug-inactivating hypersensitivity and noninactivating reactions. Asparaginase Erwinia chrysanthemi (ERW) is approved for PEG hypersensitivity, but is less convenient, more expensive, and yields lower serum asparaginase activity (SAA). We began a policy of universal premedication and SAA testing for PEG, hypothesizing this would reduce aAEs and unnecessary drug substitutions.

Procedure:

Retrospective chart review of patients receiving asparaginase before and after universal premedication before PEG was conducted, with SAA performed 1 week later. We excluded patients who had nonallergic asparaginase AEs. Primary end point was substitution to ERW. Secondary end points included aAEs, SAA testing, and cost.

Results:

We substituted to ERW in 21 of 122 (17.2%) patients pre-policy, and 5 of 68 (7.4%) post-policy (RR, 0.427; 95% CI, 0.27–0.69, P = 0.028). All completed doses of PEG yielded excellent SAA (mean, 0.90 units/mL), compared with ERW (mean, 0.15 units/mL). PEG inactivation post-policy was seen in 2 of 68 (2.9%), one silent and one with breakthrough aAE. The rate of aAEs pre/post-policy was 17.2% versus 5.9% (RR, 0.342; 95% CI, 0.20–0.58, P = 0.017). Grade 4 aAE rate pre/post-policy was 15% versus 0%. Cost analysis predicts $125 779 drug savings alone per substitution prevented ($12 402/premedicated patient).

Conclusions:

Universal premedication reduced substitutions to ERW and aAE rate. SAA testing demonstrated low rates of silent inactivation, and higher SAA for PEG. A substantial savings was achieved. We propose universal premedication for PEG be standard of care.

1 |. INTRODUCTION

Asparaginase has been a critical component of therapy for children and young adults with acute lymphoblastic leukemia (ALL) for more than 30 years, when the addition of asparaginase was shown to significantly improve event-free survival.^1,2 Subsequent studies demonstrated that duration of asparagine depletion was an independent prognostic factor, with significantly improved survival in those patients tolerating at least 26 weeks of asparaginase therapy compared with those with 25 weeks or fewer.^3–5

The forms and routes of asparaginase therapy have varied, with intramuscular (IM)/intravenous (IV) native E. coli L-asparaginase initially used, with IM pegylated E. coli L-asparaginase (PEG) initially introduced as a second-line agent for E. coli asparaginase hypersensitivity. PEG later became first line, given its longer half-life and association with greater asparaginase activity levels (an accepted measure of potency).^6,7 IV PEG, with a similar half-life and potency to the IM form of PEG, has now become the standard product, given its equal efficacy to the IM form, but with improved patient comfort. IM or IV L-asparaginase derived from Erwinia chrysanthemi(ERW) is also available, and indicated for patients with hypersensitivity to E. coli asparaginase products.^8,9 It is appropriate that ERW is indicated as a second-line agent for a number of reasons. First, ERW requires at least six administrations to replace a single dose of PEG, and so is less convenient. Second, the magnitude and duration of serum asparaginase activity (SAA) achieved by a single dose of PEG relative to six doses of ERW is higher, and so total drug exposure (as measured by AUC, which represents a universally accepted measure of relative potency) is greater with PEG, although impact of relative potency upon outcomes remains a point of debate.^10,11 Finally, the cost of ERW is significantly greater than PEG. For all of these reasons, it is preferred to avoid substitution of PEG with ERW in the absence of contraindications.

As foreign proteins, all asparaginase products are associated with systemic clinical hypersensitivity reactions, manifested on the spectrum of urticaria, bronchospasm, angioedema, or anaphylaxis.^12,13 The incidence of hypersensitivity varies widely based on form, routes, frequency, etc., but is usually estimated at 10% to 30%.^13–18 These clinical reactions have been shown to be strongly associated with the production of neutralizing antibodies and lack of asparaginase activity, although the severity of the reaction does not correlate with the risk of neutralization. In fact, there are patients who develop neutralizing antibodies without any clinical manifestations, which is known as “silent inactivation.”¹⁹ Historically, the incidence of silent inactivation with IV PEG was relatively high at 8% to 15%, especially compared with the rates seen with the IM formulations (3%), although this was in the setting of IV PEG being used following native E. coli L-asparaginase.^13,17,20 Therefore, given this potentially high incidence as well as the inability to perform therapeutic drug monitoring (TDM), premedication with histamine antagonists or steroids has generally been avoided, for fear of “masking” this clinical reaction as the sole indicator of potential neutralization.

With the almost universal adoption of IV PEG as the standard formulation in ALL from the beginning of treatment, two important facts have become clear. First, when it is used beginning in induction, the rate of silent inactivation with IV PEG is much lower than with other formulations, with 0% to 3% incidence.^17,21,22 Second, with the use of IV PEG, a distinct type of acute clinical reaction (a nonallergic infusion reaction) is becoming increasingly recognized in addition to antibody-producing hypersensitivity reactions. These nonallergic infusion reactions often occur shortly into the infusion (within minutes or even seconds) and have a great deal of clinical overlap with the hypersensitivity reactions, manifesting with flushing, hypotension, tachycardia, dyspnea, tachypnea, and anxiety. Although true urticaria and bronchospasm are rare, it is, nevertheless, nearly impossible to clinically distinguish this reaction from allergic hypersensitivity. The occurrence of these reactions with the first dose of IV PEG is further support of a nonantibody-mediated mechanism. It is possible that PEG-induced acute hyperammonemia may mediate at least some of the symptoms and signs associated with these nonallergic infusion reactions.^23,24

Concurrent with the increased use of IV PEG, TDM for asparaginase therapy has become widely available. The SAA assay has been demonstrated to be a valid surrogate for asparagine depletion. Additionally, it is now available as a CLIA-certified test with a turnaround time of less than one week, allowing real-time decision-making and therapeutic adjustments. Generally accepted SAA assay targets include a minimum trough of ≥ 0.1 IU/mL.²⁵ However, data indicate that sustained asparagine depletion is critical for improved outcomes.^5,26 Furthermore, pharmacokinetic studies indicate that when SAA levels fall below 0.4 IU/mL, asparagine is no longer completely depleted, and begins to rebound, indicating an optimal trough ≥ 0.4 IU/mL.^11,27 These considerations have been discussed previously.²⁸

Given the aforementioned potential benefits of IV PEG relative to ERW (in terms of convenience, potency, and cost), the availability of TDM and the difficulty distinguishing “true” hypersensitivity reactions from infusion reactions, we began to rechallenge patients with prior adverse acute reactions to PEG asparaginase after administering premedication with histamine antagonists and corticosteroids, and then assessing efficacy using SAA assays at trough time points. We found that this intervention allowed these patients to tolerate the infusion without clinical sequelae, and with SAA levels well above the optimal range. We therefore began a policy of universal premedication and TDM for all patients at our institution who received IV PEG, hypothesizing that this would reduce unnecessary drug substitutions and adverse events.

2 |. METHODS

2.1 |. Patients

We conducted a retrospective chart review on all patients at our institution who received at least one dose of IV PEG following the FDA approval of ERW on November 18, 2011. This included all patients treated as an outpatient through the Pediatric Oncology Clinic or as an inpatient including both the general Pediatric Oncology inpatient ward and pediatric intensive care unit (ICU). Patients for whom asparaginase therapy was discontinued due to nonallergic, noninfusion reactions (e.g., pancreatitis, coagulopathy) were excluded. Chart review of all clinically apparent reactions was conducted and coded according to the Common Terminology Criteria for Adverse Events (CTCAE), version 4.03. This study was approved by the institutional review board of the Johns Hopkins School of Medicine. The primary end point was substitution to ERW from PEG. Secondary end points included grade of acute adverse events (aAEs), disposition following aAEs, SAA testing, and drug cost.

2.2 |. Premedication for asparaginase

Beginning on February 10, 2016, all patients receiving any asparaginase product were premedicated 30 minutes prior to infusion with diphenhydramine (1 mg/kg, maximum 50 mg/dose, IV or orally) and either IV or oral tablet famotidine (1 mg/kg/dose, maximum 20 mg) or PO liquid ranitidine(2 mg/kg/dose, maximum 150 mg). Additionally, for patients being rechallenged with PEG after a previous aAE, hydrocortisone (1 mg/kg, maximum 100 mg/dose) was included as part of the premedication. For acute reactions, IV diphenhydramine (1 mg/kg, maximum 50 mg/dose) was readministered. For refractory acute reactions or hemodynamic instability, IV hydrocortisone (1 mg/kg, maximum 100 mg/dose) was administered. Epinephrine was administered in cases of hemodynamic instability or airway compromise according to standard practices. These orders were incorporated into all chemotherapy ordersets in the electronic medical record that include an asparaginase product.

2.3 |. Therapeutic drug monitoring

At least 0.2 mL of either serum or plasma was submitted, chilled for SAA testing (Granger Genetics, previously AI BioTech, Richmond, Virginia) performed seven days (± 3 days for ease of scheduling, range, 4–10 days) after PEG administration. SAA testing was also performed two days (or three days for Friday administration) after ERW for any patients receiving their first dose of ERW. These time points were selected based upon pharmacokinetic and pharmacodynamic data indicating comparable SAA levels for the respective products and to allow for timely decision-making.^6–8,11,27

2.4 |. Statistical analysis

Statistical analysis was performed within the R environment. Changes in rates of drug substitution and clinically apparent reactions after implementation of the universal premedication policy were tested for statistical significance using a bootstrap resampling algorithm with 1 × 10⁶ resamplings. Other standard statistical tests (Student t, χ², etc.) were performed as indicated.

2.5 |. Cost modeling

For all patients who were substituted ERW for PEG prior to the implementation of the premedication policy, drug cost savings were calculated by directly converting the administered ERW doses to equivalent PEG doses and calculating the difference in drug cost according to prices as of June 2018.

3 |. RESULTS

3.1 |. Patient characteristics

From November 2011 through April 2018, 177 patients were treated with at least one dose of PEG. Of these patients, 109 patients were treated prior to the implementation of the universal premedication policy, and 55 were following implementation. An additional 13 patients received PEG both without and with premedication. Being at risk for reaction and substitution under both schemes, they were therefore included in both groups. This yielded a total of 122 patients treated without premedication and 68 patients treated with premedication. Patients were characterized according to sex, age at diagnosis, ethnicity, and disease (Table 1). Univariate analysis demonstrated that the post-policy cohort had a significantly greater age at diagnosis (11.3 years vs 8.0 years, P = 0.0006) due to an increase in the upper age limit of patients treated within the Johns Hopkins Children’s Center during the course of the period studied from age 18 to 25 years. The post-policy cohort also tended to have more males (70.6% vs 63.9%); however, this did not reach statistical significance (P = 0.42). Other parameters were not clinically or statistically significantly different between the two cohorts.

TABLE 1.

Patient characteristics

	All patients (n= 177)a		Pre-policy (n= 122)		Post-policy (n= 68)
Age at diagnosis (mean)b	9.1 y (4 m–24.9 y)		8.0 y (4 m–22.7 y)		11.3 y (10 m–24.9 y)		P = 0.0006
Sex (male)	117	(66.1%)c	78	(63.9%)	48	(70.6%)	P = 0.42
Ethnicity							P = 0.65
White	117	(66.1%)	79	(64.8%)	46	(67.7%)
Black/AA	31	(17.5%)	21	(17.2%)	12	(17.7%)
Hispanic	18	(10.2%)	13	(10.7%)	7	(10.3%)
Asian	10	(5.7%)	8	(6.6%)	3	(4.4%)
Diagnosis							P = 0.34d
B lineage	136	(76.8%)	94	(77.1%)	53	(77.9%)
Pre-B ALL	130	(73.5%)	90	(73.8%)	51	(75.0%)
Pre-B LLy	6	(3.4)	4	(3.3%)	2	(2.9%)
T lineage	33	(18.6%)	21	(17.2%)	14	(20.6%)
T ALL	23	(13.0%)	17	(13.9%)	7	(10.3%)
T LLy	10	(5.7%)	4	(3.3%)	7	(10.3%)
Other	8	(4.5%)	7	(5.7%)	1	(1.5%)
AML	1	(0.6%)	1	(0.8%)	0	(0.0%)
MPAL	3	(1.7%)	2	(1.6%)	1	(1.5%)
CML	3	(1.7%)	3	(2.5%)	0	(0.0%)
Undifferentiated	1	(0.6%)	1	(0.8%)	0	(0.0%)

AA, African American; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; CML, chronic myelogenous leukemia in lymphoid blast crisis; LLy, lymphoblastic lymphoma; MPAL, mixed phenotype acute leukemia.

^aThirteen patients were treated in both the pre- and post-policy periods and were, therefore, at risk for substitution under both and included in each cohort.

^bThe upper age limit of patients treated within the Children’s Center was increased to 25.99 years during the period studied.

^cPercentages may not total 100% due to rounding.

^dChi-squared according to major lineage (B, T, or other).

3.2 |. Drug substitution rate

Pre-policy, 21 of 122 patients had a clinically significant reaction to PEG that prompted substitution with ERW (17.2% of patients), compared with 5 of 68 patients post-policy (Figure 1). This substitution rate of 7.4% versus 17.2% is a 57.3% reduction in the rate of substitution to ERW, with 10.1 patients premedicated to avoid one substitution (RR 0.427, 95% CI, 0.27–0.69, P = 0.028). Univariate analysis demonstrated that male sex was significantly associated with substitution to ERW from PEG, but no correlation was found with any other demographic factors (Table 2). Of the five patients with ERW substitution post-policy, three were for severe breakthrough aAEs that maintained excellent levels of SAA (“infusion reactions), one for a severe breakthrough aAE associated with inadequate SAA levels presumably from neutralizing antibodies (true “hypersensitivity reaction”), and one for inadequate SAA levels without clinical manifestation (“silent inactivation”). Patients treated both pre- and post-policy were included in the analysis, as they were “at risk” for conversion under both treatment regimens; however, reanalysis excluding these patients demonstrated a substitution rate of 7.3% versus 17.4%, which is a 58.1% reduction (RR, 0.419; 95% CI, 0.25–0.71, P = 0.036), with a number needed to treat of 9.8—effectively unchanged from the analysis of the full cohort. This further supports the view that prior treatment is unrelated to reaction and represents infusional reactions instead of the development of true antibody-mediated hypersensitivity.

FIGURE 1.

Premedication reduces the risk of drug substitution and the rate of clinical reactions. The rate of substitution from PEG to ERW (left) and rate of clinically significant reactions (right) were compared between patients treated pre-policy and post-policy. Significance was tested by bootstrap resampling using 1 × 106 resamplings. Error bars are standard errors of the average

TABLE 2.

Relationship of demographics to risk of drug substitution

	Un-substituted (n= 151)		Substituted (n= 26)
Age at diagnosis (mean)	9.0 y (8 m–24.9 y)		9.4 y (4 m–23.3 y)		P = 0.81
Sex (male)	95	(62.9%)a	22	(84.6%)	P = 0.042
Ethnicity					P = 0.94
White	99	(65.6%)	18	(69.2%)
Black/AA	28	(18.5%)	3	(11.5%)
Hispanic	14	(9.3%)	4	(15.4%)
Asian	9	(6.0%)	1	(3.9%)
Diagnosis					P = 0.50b
B lineage	118	(78.2%)	18	(69.2%)
Pre-B ALL	112	(74.2%)	18	(69.2%)
Pre-B LLy	6	(4.0%)	0	(0.0%)
T lineage	26	(17.2%)	7	(26.9%)
T ALL	18	(11.9%)	5	(19.2%)
T LLy	8	(5.3%)	2	(7.7%)
Other	7	(4.6%)	1	(3.9%)
AML	1	(0.7%)	0	(0.0%)
MPAL	3	(2.0%)	0	(0.0%)
CML	3	(2.0%)	0	(0.0%)
Undifferentiated	0	(0.0%)	1	(3.9%)

AA, African American; ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; CML, chronic myelogenous leukemia in lymphoid blast crisis; LLy, lymphoblastic lymphoma; MPAL, mixed phenotype acute leukemia.

^aPercentages may not total 100% due to rounding.

^bChi-squared according to major lineage (B, T, or other).

3.3 |. Rate of AEs

The rate of aAEs pre-policy compared with post-policy was 17.2% versus 5.9%, a 65.8% reduction in adverse events (Figure 1), with 8.8 patients premedicated to prevent one adverse reaction (RR, 0.342; 95% CI, 0.2–0.58, P = 0.017). Proportion of grade 4 aAEs also decreased, with 15% of reactions pre-policy requiring an ICU admission for cardiovascular support, and no grade 4 aAEs or ICU admissions postpolicy (Figure 2).

FIGURE 2.

Premedication reduces the severity of clinically apparent reactions. Clinically significant reactions were graded according to CTCAE 4.03. Graphed are the percentages of all patients treated with PEG pre-policy and post-policy who experienced each grade of reaction. Rates are charted according to each organ system examined and for all organ systems in aggregate. No grade 5 reactions were experienced during the study period. No grade 4 reactions were experienced after implementation of the premedication policy

3.4 |. SAA levels

Universal SAA testing was feasible, with 86.7% compliance with this testing postpolicy. We confirmed prior studies that demonstrated superior SAA with PEG (mean, 0.90 IU/mL; range, 0.23–3.00 IU/mL) compared with ERW (mean, 0.15 IU/mL; range, 0.0–0.45 IU/mL, P = 2 × 10⁻⁸; Figure 3A), although it is important to note that the difference in the timing of SAA testing relative to half-life makes direct comparison difficult (PEG: 7 days after administration, with half-life of approximately five days; ERW: 48–72 hours after administration, with half-life of approximately 6–16 hours). Among PEG doses, there was no difference in the SAA levels after doses given to patients for whom PEG was subsequently substituted with ERW due to infusion reactions (mean, 0.74 IU/mL; range, 0.55–0.99 IU/mL) compared with SAA levels after doses given to patients who never required substitution with ERW (mean, 0.91 IU/mL, range, 0.23–3.00 IU/mL, P = 0.26; Figure 3B). Universal SAA monitoring confirmed that drug inactivation after IV PEG is a rare event, with undetectable SAA levels (consistent with neutralizing antibodies) seen in only 2 of 68 patients (2.9%). Among ERW doses, there was no difference in the SAA levels after doses given to patients who had previously inactivated IV PEG (based on undetectable SAA levels) compared with the SAA levels after doses given to patients without evidence of IV PEG inactivation (Figure 3C). Applying a PEG inactivation rate of 2.9% to the pre-policy cohort, we estimate that only 4 of the 21 drug substitutions would have been associated with true drug inactivation. This indicates that using clinical reaction alone to monitor for drug inactivation has a false-positive rate of 83%.

FIGURE 3.

TDM of asparaginase-based therapy. SAA testing was performed for post-policy patients one week after PEG administration or two to three days after ERW administration. A, Comparison of SAA levels for all completed doses of PEG with ERW doses. B, Comparison of SAA levels following PEG achieved by patients who completed therapy without drug substitution (“Never reacting”) with levels achieved by patients who would ultimately be substituted to ERW for all PEG doses prior to the final, offending dose (“Pre-reacting doses”). C, Comparison of SAA levels following ERW administration achieved by patients without evidence of prior PEG inactivation with those with negative PEG SAA testing

3.5 |. Drug cost analysis

Although the cost per vial of PEG ($17 453 for 3750 IU) is higher than the cost per vial of ERW ($4696 for 10 000 IU as of June 2018), on a single-dose basis, the cost of PEG ($11 636/m² per dose) and ERW ($11 740/m² per dose) is comparable. However, as ERW requires a minimum of six doses for equivalence to a single dose of PEG, the cost per cycle of ERW ($70 442/m² per cycle) is higher than that of PEG (a difference of $58 806/m² per dose of PEG substituted with ERW).

The average body surface area of our patient population was 1.07 m² (0.56–2.15 m²). Pre-policy, an average of 2.0 doses (range, 1–5 doses) of PEG were substituted with ERW for each patient who required drug substitution. Thus, the average drug cost savings for a patient who avoids substitution is approximately $125 779 ($58 806/m² × 1.07 m² × 2.0 doses). Premedication yields an absolute risk reduction for drug substitution of 0.098 (0.074 down from 0.172). Thus, the estimated drug cost savings per patient premedicated is $12 402 ($125 779 × 0.098). These results do not include the savings realized through fewer admissions due to reactions, fewer interventions for life-threatening reactions including critical care, and fewer subsequent clinic visits for multiple ERW administrations.

4 |. DISCUSSION

Asparaginase is a critical component of modern therapy for ALL, and maximizing its potency has been demonstrated to improve outcomes. IV PEG is the near-universal formulation used in ALL therapy, as it leads to increased drug exposure, and is more cost effective and convenient than ERW, which ideally should be reserved for patients who develop hypersensitivity-associated neutralizing antibodies that inactivate asparaginase activity after IV PEG administration.

The availability of the SAA assay makes it possible to distinguish reactions that produce neutralizing antibodies from other “infusion reactions,” which can be clinically severe but still maintain sufficient asparagine depletion. The SAA assay also allows detection of “silent inactivation.” Thus, SAA testing makes it unnecessary to avoid the use of premedication for fear that it may “mask” hypersensitivity reactions. Rather, if premedication can prevent or mitigate the severity of these distressing and sometime serious symptoms, then perhaps this should become standard practice.

This retrospective, single-institution study of premedication and SAA testing for all patients receiving PEG demonstrates that it is a feasible supportive care intervention that results in a significant decrease in substitutions to ERW, and maintenance of adequate serum SAA in most patients, even those with aAEs during the administration of the drug. This policy also reduced the rate and severity of adverse effects associated with the infusion, which can be clinically quite severe.

Implementation of universal SAA testing demonstrated very low rates of silent inactivation with IV PEG. This study also confirmed previous reports of the higher SAA levels achieved with IV PEG compared with ERW, which lead to decreased drug exposure (as measured by AUC) despite more frequent administrations. Furthermore, the low rates of inactivation in general suggest that clinically apparent reactions markedly overestimate the presence of inactivating antibodies, and that historically, using clinical response without TDM, approximately four of five substituted patients were unnecessarily switched to potentially less active therapy.

From the perspective of optimizing high-value healthcare, substantial cost savings were achieved with the implementation of this policy, with approximately $12 400 saved per patient premedicated. This is based solely on the decreased cost of one dose PEG compared with a six-dose course of ERW and does not incorporate the additional financial benefits of the decreased number of clinic visits when comparing PEG to ERW, ICU stays, and interventions for adverse events. This saving easily compensates for the additional cost of the premedication drugs and SAA for each patient.

We recognize that our study has a number of limitations. First, this is a retrospective review of consecutively treated patients before and after a change in clinical practice, and not a prospective, randomized trial. Although it is reasonable to consider this when evaluating the strength of evidence favoring universal premedication and SAA testing, the effect size is large and, with the exception of a slight shift toward older patients in the post-policy group, the characteristics of the comparator groups are similar. Thus, it is exceedingly unlikely that the observed differences are the result of biases introduced by the study design. Second, because we did not routinely measure the SAA levels of patients receiving IV PEG or ERW prior to the policy change, the sample sizes for the comparisons of potency of IV PEG versus ERW are limited to the patients treated after the policy was adopted. Nonetheless, the differences are statistically significant and consistent with prior published studies of both IV PEG and ERW, so we can be reasonably certain that the pattern of greater SAA after PEG than after ERW dosing would have been similarly demonstrable in the cohort of patients treated prior to the adoption of the policy. In addition, although direct comparison of SAA levels 7 days after PEG to 48 to 72 hours after ERW is not ideal given their different pharmacokinetic profiles, it is clear from multiple prior studies (further supported by our data) that the area under the curve for SAA after a single dose of PEG is greater than that for six doses of ERW. Thus, using drug exposure as a measure of potency, PEG is more potent than ERW.

In conclusion, a policy of universal premedication with real-time, TDM for asparaginase-based therapy is feasible and has demonstrated significant clinical benefits. It leads to fewer patients having PEG substituted with ERW, and subsequently higher serum asparaginase activities. It is associated with fewer and less severe clinically significant aAEs. It offers significant cost savings from both drug costs and less healthcare system utilization. Furthermore, although this study was conducted in an academic center, the commercial availability of rapid turnaround TDM allows this policy to be applied in almost any clinical setting. For these reasons, we propose universal premedication with TDM for IV PEG be considered standard of care. We advocate for its inclusion in the supportive care guidelines for cooperative group trials in pediatric ALL.

Abbreviations:

aAE

acute adverse event

AE

adverse event

ALL

acute lymphoblastic leukemia

CTCAE

Common Terminology Criteria for Adverse Events

ERW

L-asparaginase Erwinia chrysanthemi

IM

intramuscular

IV

intravenous

PEG

pegylated L-asparaginase

SAA

serum asparaginase activity

TDM

therapeutic drug monitoring

DATA AVAILABILITY

The data that support the findings of this study are available from the corresponding author upon reasonable request.