L‐asparaginase doses number as a prognostic factor in childhood acute lymphoblastic leukemia: A survival analysis study

Abstract Background The survival of children with acute lymphoblastic leukemia (ALL) has improved due to changes in the treatment and the disease diagnosis. A significant advance was the incorporation of asparaginase. However, hypersensitivity reactions are a common cause of early discontinuation of this drug. Aim The proposed study aims to evaluate early interruptions and the influence of the number of asparaginase doses effectively administered on the prognosis of patients with ALL. Methods and Results An observational cohort study was carried out, with retrospective data collection, in medical records. The prognostic variables indicated in the protocol applied were used, and the principal outcomes were 5 years event‐free survival (EFS) and 5 years of overall survival (OS) probability. Statistical analyzes were performed using SPPS 20.0 and R. In Cox's proportional hazards model for EFS and OS, variables of prognostic importance (n = 126 children) were: high‐risk group (HGR), by the protocol classification, and less than 10 doses of asparaginase. The increased risk of events and death in HGR, who did less than 10 doses, was 3.6 and 7 times, respectively. The study did not show statistical significance for the number of asparaginase doses in patients who were not at high risk. Conclusions We demonstrated that the early interruption of asparaginase treatment could negatively impact the prognosis of patients with ALL, especially HGR, reinforcing the need for careful diagnosis of reactions and the availability of alternative types of asparaginase.

the 1970s is considered fundamental for the best survival results. 2,4,5 Asparaginase is an enzyme that hydrolyzes asparagine, an essential amino acid that the leukemic blast cannot synthesize in aspartic acid and ammonia. The first type of asparaginase to be widely used in the treatment of ALL was Escherichia coli native asparaginase.
However, the high incidence of hypersensitivity reactions (average 30%), which end up causing early treatment interruption, caused others types of asparaginase to replace native asparaginase in many countries. 4,6 E. coli's PEG-asparaginase has a chemical conjugation with polyethylene glycol that reduces its immunogenicity and increases the half-life, thus reducing the number of doses to be administered, making it the first line of treatment in most developed countries. However, although it is less immunogenic than native asparaginase, 5% to 18% of people can develop antibodies against PEG-asparaginase. For this reason, from 2011, in Europe and the United States, the use of asparaginase from Erwinia chrysanthemi was approved, to guarantee the continuity of treatment of people who presented antibodies against PEG-asparaginase. [6][7][8][9][10] In Brazil, PEG-asparaginase was only registered in 2018. The native asparaginase of E. chrisantemi remains unregistered with the National Health Surveillance Agency (ANVISA). The unavailability of alternatives for continuing treatment with asparaginase makes many people stop their treatments early. Even in countries where E. chrisantemi asparaginase is used as a second line of treatment, this problem can occur due to its shortage or toxicity. 10,11 As asparaginase is a fundamental medication in the treatment of ALL, the need to use a lower number of doses than recommended in the protocols can directly impact patient survival. Gupta et al, 10 showed a risk ratio of 1.5 in the analysis of event-free survival (EFS) for the high-risk patients compared to those who had their treatment stopped early with those who took all doses.
The proposed study aims at the impact of early interruptions and the influence of the number of asparaginase doses effectively administered on the prognosis of patients with ALL treated at a pediatric teaching hospital located in Rio de Janeiro (Brazil) over 10 years. based protocols with high-dose Methotrexate (2 g/m 2 for non-T-ALL and 5 g/m 2 for T-ALL) and native Asparaginase intravenously. Nursling leukemia patients (<1 year) treated with a specific protocol (Interfant 99 and Interfant 06) were excluded. Two infants, close to   1 year of age, were treated with the ALL-BFM-IC and were included   in the study. During the study period, children older than 12 years were not treated at the institution.
The IPPMG-UFRJ (Instituto de Puericultura e Pediatria Martagão Gesteira da Universidade Federal do Rio Janeiro) is a quaternary pediatric teaching hospital with had an oncohematology service for more than 50 years, being one of the four reference sites in the state of Rio de Janeiro for treatment of childhood ALL. The study patients were identified through the high complexity procedure authorization lists (APAC) available at the institution.

| Data source and potential predictors
Data were collected in medical records by using a data collection form.  collected too the number of asparaginase doses effectively administered during the treatment to each patient.
The main outcomes were 5 years EFS and 5 years of Overall Survival probability. 12  Univariate and multivariable analysis were conducted using time-to-event techniques (Cox proportional hazard models), 13 which allowed the identification of the hazard ratio (HR) for the analysed variables.   Table 1.  In the HRG patients, the parsimonious predictive model obtained through the Cox proportional hazards multivariate analysis, the only identified prognostic factors was the use of less than 10 Lasparaginase doses (HR = 3.603 -CI: 1.317-9.861, p = .013).

| Predictors of overall survival
The KM OS analysis, shown in

| L-asparaginase doses number
The total number of L-asparaginase doses effectively received during the treatment by the HRG patients and the other risk groups were not significantly different (median 9 (IQR = 4) vs. 8 (IQR = 4) doses, p = .207). Nonetheless, the L-asparaginase dose number was the most important prognostic factor in HRG patients. This study did not indicate a prognostic significance of L-asparaginase dose number for the non-HRG patients.

| DISCUSSION
In the study population, there was a slight predominance of boys (56.3%), as well as more children from 1 to 5 years (60%). This slight difference in the incidence of ALL in boys (57.1%), as well as the greater number of cases in the age group from 1 to 5 years old (78%) was also observed by Stary et al, 12  asparaginase. 10 We confirmed in the present study that for high-risk patients, especially those who are unable to take at least 10 doses, continuity of treatment with Erwinia's asparaginase would be essential.
In Brazil, the country where the present study was carried out, the alternative to asparaginase derived from E. coli (native and PEGasparaginase) is not available. Therefore, once a hypersensitivity reaction is observed, treatment with asparaginase is definitely suspended.
Reactions related to asparaginase infusion are the leading cause of early treatment interruption since allergic hypersensitivity reactions are associated with the formation of neutralizing antibodies that make it impossible to continue treatment with the type of asparaginase used, being indicative of resistance to treatment. Pancreatitis may cause the early termination of asparaginase treatment without a possible change of type of asparaginase formulation. Thrombosis can lead to delays in treatment, as it requires temporary interruption of therapy, according to ALL-BFM-IC protocol. 6,12,15,16 However, not all infusion reactions to asparaginase require the interruption of treatment with asparaginase. Non-allergic hypersensitivity reaction and the hyperammonemia reaction can also occur during the asparaginase infusion. Recent works highlighted careful monitoring and correct classification of reactions for kind and grade so that a treatment interruption or exchange for an alternative type of asparaginase is performed unnecessarily. [17][18][19][20] It is imperative to avoid unnecessary early interruptions, and to guarantee alternatives to asparaginases derived from E. coli when necessary.

| CONCLUSION
High-risk and non-high-risk patients have been shown to have a different impact on global and EFS for the early discontinuation of asparaginase treatment. The first group needs to use at least 10 doses of E. coli native asparaginase. On the other hand, for the second, induction completeness is necessary, but when the reaction is after this phase of treatment, no decrease in survival was identified due to early interruption.
These findings reinforce the need for differential diagnosis of adverse reactions that do not prevent the asparaginase use and the availability of other formulations of this enzyme for high-risk LLA treatment, especially in developing countries.

ACKNOWLEDGMENTS
We would like to thank the IPPMG team, especially the hematologists, the pharmacy, the nursing staff of the hematology ward and the Aquário carioca (outpatient chemotherapy room), and the clinical analysis laboratory, for all their support in carrying out this study.

CONFLICT OF INTEREST
We declare the absence of conflicts of interest of a personal, commercial, academic, political, or financial nature and that the manuscript has not been submitted elsewhere nor previously published. project administration (supporting); writingreview and editing (lead).

ETHICS STATEMENT
The research project was submitted to and approved by the Research Ethics Committee of the Pediatrics Institute of the Federal University of Rio de Janeiro (2.787.465) Retrospective secondary data was collected without any interaction between researchers and children or parents. Personal information of the participants was kept blinded to investigators. Waiver of parental permission (Written informed consent from the participant's legal guardian) was requested and authorized by REC.

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