Incidência e fatores de risco para recidiva no sistema nervoso central em crianças e adolescentes com leucemia linfóide aguda: um estudoretrospectivo (março/2001 a agosto/2009) no Serviço de Hematologia do Hospital das Clínicas da UFMG

Abstract

OBJECTIVES: To describe the status of central nervous system (CNS) infiltration and the outcome of children with acute lymphoblastic leukemia (ALL) diagnosed at the Hematology Service, Hospital das Clínicas, Federal University of Minas Gerais (UFMG), Brazil. To classify the children with ALL according to the degree of CNS involvement and the occurrence, or not, of traumatic bleeding due to lumbar puncture (LP). To evaluate the incidence and identify the risk factors for CNS infiltration at diagnosis and at relapse. PATIENTS AND METHODS: Patients less than 18 years old, with a diagnosis of ALL between March 2001 and August 2009, treated according to the Grupo Cooperativo Brasileiro para Tratamento daLeucemia Infantil GBTLI LLA-99, were enrolled in this retrospective study. Clinical and laboratory data were obtained via chart review.RESULTS: The study population consisted of 199 patients. The follow up period ranged from two days to 8.7 years (median: 3.3 years). Boys comprised 57% of the studied population (114 out of 199). Medianage at diagnosis ranged between two months and 16.8 years (median: 5.59 years). Data on first cerebrospinal fluid results were obtained in 194 patients. Patient distribution with regard to CNS involvement at diagnosis was as follows: CNS1 (atraumatic LP; absence of blasts) - 61%; CNS2(atraumatic LP; leucocyte count less than 5/µl; presence of blasts) - 1%; CNS3 (atraumatic LP; leucocyte count equal or superior to 5/µl; presence of blasts) 0; negative TLP (traumatic LP; absence of blasts) - 37%; and positive TLP (traumatic LP; presence of blasts) - 1%. Data on place where first LP was performed were collected in 155 patients. In 43% of patients (67 out of 155), LP was performed in the hospital ward, without anesthesia, and, in 57% (88 out of 155), in the operating room, under deep sedation. A possible association between traumatic lumbar puncture and the following variables was evaluated: age, leucocyte count at diagnosis, place where LP was first performed, and platelet countbefore the procedure. The only significant association observed was that between age of the patient and traumatic lumbar puncture. Median age was 3.3 years for the patients who presented traumatic lumbarpuncture and 6.1 years, for those who did not (p=0.0004). The estimated probability of Overall Survival at 5 years for the whole group was 69.5% (±3.6%). The estimated probability of Event Free Survival (EFS) at5 years for the whole group was 58.8% (±4.0%). An estimated EFS of 76.2% (±5.4%) and of 46.6% (±5.3%) were observed in the low-risk and the high-risk (for relapse) groups, respectively (p=0.00002). CNS relapse occurred in 18 (9%) of the 199 patients, 11 of whom had isolated relapses. The cumulative incidence of CNS (isolated or combined) relapse for the whole group was 10.9% at 8 years. The cumulative incidence of CNS relapse was evaluated according to the immunophenotype (B versus T) at diagnosis, the leucocyte count at diagnosis, the interval between diagnosis of ALL and first LP, and thepresence or absence of trauma at first LP. There was a significant difference between the children with an absolute leucocyte count of more or of less than 50.000/mm3: The cumulative incidence of CNS relapse was much greater in those with hiperleucocytosis. The other comparisons between the groups did not show any significant difference as to the occurrence of CNS relapse. CONCLUSIONS: The cumulative incidence of (isolated or combined) CNS relapse in the present studywas superior to that of international studies. The EFS at five years was inferior to the results reported in the literature. Initial leucocyte count greater than or equal to 50.000/mm³, allocation to a high risk group,and delay (over 7 days) in first lumbar puncture were poor prognostic factors regarding EFS. A high incidence of traumatic lumbar puncture at diagnosis was also observed. Age was the only variable associated with trauma in this procedure. A leucocyte count at diagnosis of more than 50.000/mm³ was the only factor associated with a greater incidence of CNS relapse.