Intensive chemotherapy in children with acute lymphoblastic leukemia. Interim analysis in a referral center in Colombia

Results: 119 patients were included. Death occurred in two cases during induction (1.67%) and in nine (7.7%) due to treatment, all of them caused by infection/sepsis and in complete remission. Six patients abandoned treatment (5%), while seven relapses occurred (5.9%). All patients experienced some type of side effect related to chemotherapy, the most frequent being febrile neutropenia (41.2%) and grade 3-4 infections (15.8%). Overall survival and event-free survival rates were 79.9% and 73.3%, respectively.


Introduction
Acute lymphoblastic leukemia (ALL) is the most common neoplasm in pediatric patients (1).Overall cure rates for childhood ALL have improved over the years and current survival rates vary from 75% to 85% in patients treated in high-income countries (2); with the continuous improvement of survival rates, the goal of current clinical protocols is to reduce adverse reactions related to treatment (3).In contrast, in low-income countries, the possibilities of cure are lower, probably due to disease status at diagnosis, treatment abandonment, high rates of relapse and death caused by toxicity or side effects related to treatment (2).
In Colombia, the Cancer Registry of Cali (RPCC by its acronym in Spanish) reported a rate of 41% of survival in children with leukemia between 1994 and 2003 (4); the Protocol of Public Health Surveillance on Childhood Cancer reported that high mortality in children with leukemia is given by deaths during the first year of treatment, possibly because of poor access to treatment, low intensity treatment and toxicity caused by it.It also mentions that the causes of such low survival rates have not been identified, so intervention on them has not been possible (5).
Better cure rates are affected by additional barriers; the evaluation in children with ALL by Suarez et al. in Bogotá (6) reports that delays due to non-medical reasons in treatment are common and predict treatment failure.The same study also reported mortality during induction of 7%, death in complete remission of 3% and abandonment rates of 9% (6).According to information provided by the National Institute of Health of Colombia, 455 new cases of ALL were confirmed in 2013, with an overall mortality related to treatment of 15.6% (7).
In a study conducted in Central America, specifically in lowincome countries, overall mortality related to treatment of 9.3% and mortality during induction of 5.5% were described (2).
Within the International BFM Study Group (Berlin-Frankfurt-Münster) (I-BFM-SG), the ALL strategy committee has developed several protocols with optimal clinical results over the last 20 years, most of them derived from the original BFM.Chemotherapy treatments proposed by the BFM for resource-limited countries have some changes aimed at local needs and conditions.In 2007, an amendment to the Intercontinental BFM 2002 Protocol was defined as more suitable for the diagnosis and treatment of children with ALL in Colombia (8).This protocol was adapted and applied for about five years at Fundación Hospital La Misericordia in Bogotá (Fundación HOMI), referral center for care of children with ALL in Colombia.
In 2012, the implementation of a modified version of the "Protocol for study and treatment of childhood lymphoblastic leukemia, ALL PINDA 2009, ALL IC BFM 2009" was decided.This approach to treatment is the same currently used in Argentina, Uruguay, Chile and Colombia, where the protocol is conducted by the National Cancer Institute.
In 2014, the BFM group published their experience with the BFM ALL IC 2002 treatment protocol, showing results from 15 countries of three continents, most of them with average incomes, as in the case of Colombia.The results of the implementation of this protocol show an improvement in the outcome of treatment for ALL, with event-free survival of 74% and overall survival of 82% (9).
There are no publications in Colombia that evaluate the results obtained with this type of treatment strategies, including survival and deleterious effects.This work aims at conducting an interim assessment of death during induction, treatment-related death (in remission), relapse and treatment abandonment, as well as at describing the most frequent adverse reactions observed with chemotherapy drugs, assessing overall and general event-free survival rates, and comparing -by risk group-previous outcomes with previous reports to assess the results of the implementation of this protocol at the institution.

Materials and methods
The description of a cohort of patients was performed.The inclusion criteria for the study were: age under 18, confirmed acute lymphoblastic leukemia diagnosis at Fundación HOMI between November 1, 2012 and December 31, 2014, treatment using the protocol BFM ALL IC 2009 (Table 1) in the same institution and continuity in treatment during the time of evaluation; those who started treatment and abandoned are also included until the time of abandonment.Exclusion criteria included being diagnosed in other institutions and being transferred to Fundación HOMI to continue treatment.Information was collected from clinical records.
This study was approved by the ethics committee of the institution prior to the review of the database, and the principle of privacy and confidentiality was preserved.Follow-up time was considered in months and was measured from the moment of diagnosis of the disease until the outcomes, which were defined as death during induction, death related to treatment, relapse, abandonment and transfer.The followup was done until September 30, 2015 and lasted between 9 and 34 months.The protocol was valid until the date of termination of the study.
Qualitative variables were presented as absolute and relative frequencies for the descriptive analysis.Similarly, a survival analysis was performed using Kaplan-Meier and Log Rank test for comparison of curves.High risk patients received prophylactic or therapeutic radiation accordingly to their condition.

Diagnosis
Diagnosis of ALL was confirmed with the presence 25% or more lymphoblasts in bone marrow.Flow cytometry based on EuroFlow panel criteria (10) was used for immunological classification of tumor cells, karyotype cytogenetic assessment was performed, and translocations t (12:21), t (4:11) and t (9:22) were identified with fluorescence in situ hybridization (FISH).The involvement of the central nervous system (CNS) was determined through identification of the cells using the cytospin method, for posterior classification according to the status: Status 1: No clinical evidence disease, including facial paralysis that may be attributable to leukemia; no images -computerized axial tomography scan (CT scan) or magnetic resonance imaging (MRI) taken by suspicion-with evidence of CNS abnormalities attributable to leukemia, normal fundus or cerebrospinal fluid with no blasts and no other evidence of CNS leukemia.Status 2: Blasts clearly identifiable in a CSF cytocentrifuge with cell count of <5/uL and CSF ratio of red blood cells (RBCs): leukocytes (LEU) ≤100:1; with this ratio between RBCs and LEU, a lumbar puncture is considered non-traumatic and CSF non-contaminated with blood.Lymphoblasts in a CSF cytocentrifuge and ratio GR: LEU> 100:1; with this ratio between erythrocytes and leukocytes, lumbar puncture is considered traumatic and CSF contaminated with blood or as a traumatic lumbar puncture -CSF contaminated with blood-associated with an initial leukocyte count of >50000/uL.Status 3: Abnormal mass in the brain and/or meninges detected through CT/MRI, cranial nerve palsies, regardless of origin; although CSF does not show blasts nor abnormal masses in the images, isolated compromise of the retina is evident but with CFS with no blasts nor masses in CT/MRI or a non-traumatic lumbar puncture with a cell count CSF>5/uL and most of the blasts in the cytocentrifuge.

Risk classification
Risk classification was established by the BFM group considering clinical and laboratory criteria evaluated in previous protocols; according to the characteristics described in Table 2, patients were classified as standard risk, intermediate risk and high risk.Source: Own elaboration based on the data obtained in the study protocol.

Outcomes
Treatment-related death, defined as death during induction or death in complete remission occurred; the first death was specified as death in the first 33 days of treatment and the second, as death after this period without clinical or paraclinical evidence of disease activity.Secondary outcomes were abandonment, defined as the interruption of treatment for four or more weeks without medical reason, transfer to another institution due to the change of treatment center by the insurer, and relapse in bone marrow caused by the reappearance of lymphoblasts ≥25% in bone marrow.
In the SNC, relapses were established by the appearance of cells >5/uL CSF and indisputable lymphoblasts identified in cytocentrifuge or intracerebral mass in CT/MRI without blasts in cerebrospinal fluid (CSF), peripheral blood (PB) or bone marrow (BM) -a biopsy may be necessary for diagnosis-.In the testicles, relapses were verified by a steady insensitive unilateral or bilateral increase of one or both testicles, with volume >2 and deviations measured with the Prader orchidometer -the diagnosis must be confirmed through biopsy-and combined by simultaneous compromise of two or more compartments or locations; BM relapse is considered compromised when >5% lymphoblasts.
Another outcome was the description of adverse reactions to chemotherapy drugs, which were evaluated based on the criteria of the National Cancer Institute (NCI CTC v2.0): incidence of infections, defined as those with identified pathogen and antibiotic treatment and sepsis; incidence of cardiac, liver toxicity and mucositis grade 3 and 4; requirement for admission to the intensive care unit and transfusion requirement, which includes transfusions of red blood cells, platelets, fresh frozen plasma and cryoprecipitate.

Statistical analysis
The event-free and overall survival were estimated according to the Kaplan Meier method.Event-free survival was defined as the period between the start of treatment and presentation of an event (whichever comes first: death, relapse, second malignancy, transfer or abandonment), for those who have not presented any, the event was the last control alive.Overall survival was defined as the time between the start of treatment and the last control alive, regardless of the condition of the disease.

Results
The sample includes 119 patients who meet the inclusion criteria in the specified period.Demographic characteristics are shown in Table 3.

Mortality
Of 119 patients, two (1.67%) died during induction, nine (7.7%) died in remission related to treatment due to infection/sepsis, seven of them were classified as high risk, two as intermediate risk, and there were no deaths in those with standard risk.Deaths that occurred during the treatment stages were: two (18%) in phase Ib, two (18%) in block HR1#1, two (18%) in block HR 3#1, one (10%) in HR3 block HR3#2 and two (18%) in protocol II phase A.

Abandonment and relapse
Six abandonments (5%) and seven relapses occurred (5.9%), the latter in two high-risk patients and five in intermediate-risk patients.Five relapses (71.4%) were isolated bone marrow, one (14.3%)was isolated CNS and one (14.3%)was isolated testicular.In two patients, relapses occurred after completing treatment -at 30 and 26 months after diagnosis-; one patient was considered high risk and the other intermediate risk.The remaining five patients -a high-risk patient and four-intermediate risk patients-relapsed within the first year of treatment.

Drug-related adverse reactions
A description of the adverse reactions related to chemotherapy drugs used in the treatment for each phase and risk group was made (Table 4 and 5).
All the patients had some adverse reactions related to chemotherapy drugs; the most frequent was febrile neutropenia during induction phase for all risks and in the consolidation phase for high risk patients.In the mM protocol, lower incidence of complications in general was found, as well as less febrile neutropenia, grade 3-4 infections and admission to the pediatric intensive care unit (PICU).The transfusion requirement was high in all groups of patients, especially in the induction phase.

Comparison with previous studies
After comparing the adverse reactions to chemotherapy drugs observed in this study (ALL IC-BFM 2009 protocol) with those reported in the Intercontinental Trial BFM ALL IC 2002 (9) (Table 6), the main findings were lower incidence of infections for all risks -demarcated as an identified infection pathogen requiring antibiotics IV or septic shock-and increased incidence of liver toxicity and mucositis in the Intercontinental Trial BFM ALL IC 2002.Source: Own elaboration based on the data obtained in the study.

Interim survival analysis in the sample
Overall survival at the moment of this evaluation was 79.9%, similar for patients with phenotype B and T. According to the risk, survival was 100% for standard risk, 85.4% for intermediate risk and 62.5% for high risk (Figure 1).

Figure 1 .
Figure 1.Probability of 2-years survival curves.Source: Own elaboration based on the data obtained in the study.
SR: standard risk; IR: intermediate risk; HR: high risk.Source: Own elaboration based on the data obtained in the study.

Table 2 .
Risk classification of BFM ALL IC 2009 Protocol.

Table 3 .
Patient characteristics and results of initial treatment according to the risk group in the total population.
*<5% blasts by morphology.Source: Own elaboration based on the data obtained in the study.

Table 4 .
Adverse reactions related to medication per treatment phase in protocol ALL IC BFM 2009.In IB protocol to cytarabine and in blocks to E. coli asparaginase.† Infections with clinical diagnosis, without identified pathogen and antibiotic treatment.
*Source: Own elaboration based on the data obtained in the study.

Table 5 .
Adverse reactions related to medications per treatment phase and risk group in protocol BFM ALL IC 2009.
SR: standard risk; IR: intermediate risk; HR: high risk; PICU: pediatric intensive care unit.* In IB protocol to cytarabine and in blocks E. coli asparaginase † Infections with clinical diagnosis, without identified pathogen and antibiotic treatment.Source: Own elaboration based on the data obtained in the study.

Table 6 .
Comparison of incidence of grade 3 and 4 non-hematologic complications.