Asparaginase is a cornerstone in the treatment of acute lymphoblastic leukemia (ALL), but it also causes a range of toxicities of which several are severe and have lifelong consequences.

Prolonged asparagine depletion (e.g. PEG-asparaginase q 2 weeks for half a year) seems to improve event-free-survival (EFS), but may also increase the risk of asparaginase related toxicities.


The NOPHO ALL2008 protocol (Toft, Eur J Haematol 2016) included a randomized trial for non-HR ALL children (age 1-17 years) comparing intermittent and continuous asparaginase treatment from day 92. The primary endpoint was non-inferiority with respect to EFS between the two arms with reduced toxicities as secondary endpoint. The study opened January 1st 2009 and closed March 1st 2016. End of follow-up was by December 31st 2016.

Before start of the randomization all children were treated with five doses of PEG-asparaginase 1000 IU/m2 i.m. at two weeks intervals from treatment day 30 (end of induction). From treatment day 92 (start of delayed intensification) patients received continuous asparaginase treatment (standard arm) for another 10 doses at 2w intervals or intermittent (experimental arm) for 3 doses at 6w intervals. Patients in the experimental arm did not receive asparaginase during dexamethasone containing delayed intensification.

Toxicities were captured by 1) case-report-forms, 2) quarterly prospective toxicity registration of pre-defined treatment toxicities, 3) a separate and detailed asparaginase registration, 4) deep phenotyping studies on selected toxicities, and 5) confirmation of toxicities and dates of onset by contact to the treating centers.


A total of 629 Nordic children were included in the study, with 312 and 317 randomized to the experimental and standard arm, respectively. Recruitment was less than anticipated primarily due to parents declining treatment reduction (Tulstrup, Ped Blood Cancer 2016) but also because of asparaginase associated toxicites before start of randomization.

No significant differences were observed regarding sex, age, immunophenotype or risk group between the two arms. Patients were followed from diagnosis with delayed entry at start of randomization until relapse, DCR1, SMN or end of follow-up, whichever came first. Median follow-up was 4.4 years (IQR: 2.9-6.2) with 5-year and 7-year EFS of 0.92 and 0.91, respectively.

Events (defined as relapses, DCR1 and SMN) were 19 and 22 in the experimental and standard arm, respectively. In a multiple Cox regression analysis adjusted for age, sex, WBC at diagnosis and risk group, and stratified by immunophenotype the hazard ratio for the experimental arm compared to the standard arm was 0.89 (95% CI: 0.48-1.66) (p=0.72).

Three year cumulative incidences in the whole randomized cohort for the various toxicities were 0.022 for allergy, 0.044 for pancreatitis, 0.059 for thrombosis, 0.060 for osteonecrosis (ON) and 0.031 for invasive fungal infection.

In an adjusted (sex, age, risk group) Cox regression, the hazard ratio for the experimental arm versus the standard arm was significantly reduced for pancreatitis (p=0.017), non-significantly reduced for allergy for IR patients (p=0.089) (no effect for SR patients), significantly reduced for thrombosis in boys (p=0.027), (no effect for girls). No significant differences were observed regarding ON and invasive fungal infections.


This study does not determine if extended asparaginase treatment improves EFS, but if such therapy is given, an intermittent rather than a continuous asparaginase treatment strategy results in almost identical EFS in non-HR patients with lower costs and a markedly reduced risk of serious toxicities.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract