Prognostic irrelevance of CD34 expression in childhood B-precursor acute lymphoblastic leukemia.

Sir, Phenotypic subclassification within B-precursor acute lymphoblastic leukemia (ALL) has been shown to have prognostic significance in children. Common ALL antigen (CALLA) expression in Bprecursor ALL has been associated in some, but not all, studies with a favorable outcome. According to the notion that acute leukemia is a clonal expression with maturation arrest, CD34 expression on leukemic cells might define particular characteristics of immaturity.1 Several studies in childhood B-lineage leukemia ALL have reported a strong association between CD34 expression and favorable prognosis,2-4 but other studies have not confirmed this association.5 In order to evaluate the prevalence of CD34 expression and to investigate possible correlations with the presenting clinical-biological features and treatment outcome, we analyzed 52 B-lineage ALL children consecutively admitted to our Service and treated with the AIEOP ALL-91 protocols. The leukemic population was considered CD34 positive if at least 10% of bone marrow blast cells reacted with the HPCA-2 monoclonal antibodies. All subjects expressed CD19. At diagnosis, there was no statistically significant difference in CD20-antigen expression rate between CD34+ and CD34– patients. Clinical and biological characteristics of the patients are reported in Table 1. The presence of CD34 on blast cells did not correlate with a higher probability of achieving complete response (CR) (p=0.24), nor did it affect the duration of disease free survival (DFS) (p=0.65) or overall survival (OS) (p=0.45) at 63 months of follow-up. There was no significant difference in relapse (REL) (p=0.62) or death rate (p=0.36) between CD34+ and CD34– patients. After a mean follow-up of 49 months, 17 CD34+ and 19 CD34– patients are in complete continuous remission (CCR) (p=0.38). Thus, in a series of 52 de novo B-ALL we could not confirm previous reports on the prognostic value of CD34. Differences in clinics and treatment could explain these contrasting results. For instance, in one study2 patients were enrolled in two successive total therapy studies with different follow-up duration. In another report,4 at diagnosis CD34– subjects had significantly more unfavorable prognostic factors, i.e. age > 10 years, WBC count >103109/L and more frequent CNS involvement. These factors could have been responsible for the worse outcome of the CD34– subjects. Finally, the population studied by Borowitz3 was randomized to receive two different consolidation regimens. In the study by Wanhaeke,5 as in ours, the outcome of CD34+ children with B-ALL was no better than the CD34– children. Some differences between our study and theirs should, however, be kept in mind: their cut-off for CD34 positivity was 30% and no subject with myeloid co-expression or bilineal leukemia was enrolled. Long term prognostic value of CD34 has also been analyzed in adult patients with de novo acute lymphoblastic or myeloid leukemia. In two studies on adult ALL,4,9 no statistical differences were seen in DFS or OS between CD34+ and CD34– cases, although CD34 expression was associated with major adverse prognostic factors. Similarly, a number of studies on adult AML blasts could not confirm the former results and failed to demonstrate an adverse effect of CD34 expression on the outcome of induction therapy, remission duration or event free survival.6-9