Background. Recurrence of ALL in CNS in adults is considered a poor prognostic feature but few studies have analyzed this issue. The objective of this study was to analyze the frequency, prognosis and predictive factors of CNS involvement and recurrence in adult patients with ALL treated with 4 PETHEMA protocols not including cranial irradiation for CNS prophylaxis.

Methods. From June 1989 to December 2003, 467 adult patients (≥ 15-years-old) diagnosed with ALL were treated with one of the four consecutive protocols of PETHEMA group: ALL-89 (standard and high risk, n=108), ALL-93 (high risk, n=222), ALL-96 (standard risk, n=84), and ALL-97 (Burkitt’s leukemia, n=53). CNS prophylaxis consisted of intrathecal (IT) injection of methotrexate (12 mg), cytarabine (30 mg) and hydrocortisone (20 mg), for 12–14 courses, together with high-dose systemic methotrexate and cytarabine during the early intensification phase. Cranial or craniospinal irradiation was not used in any case.

Results. The median (SD) age was 33 (16) years and 272 (58%) were males. ALL type according to the FAB classification was: L1 28%, L2 61%, L3 11%. Immunological subtypes were: early-pre-B 15%, common 45%, pre-B 5%, mature B 11% and T 24%. CNS involvement at diagnosis was observed in 18 (3.9%) patients. Predictive factors for CNS involvement at diagnosis were: L3/mature B ALL (p<0.0001) and testicular involvement (p=0.006). Overall, complete remission (CR) was achieved in 381 (81%) of the patients, of whom 159 (42%) relapsed: 137 (36%) in bone marrow (BM) and 22 (5.8%) in CNS (14 isolated and 8 combined CNS and BM). The median (range) CR duration prior to CNS recurrence was 1.06 yr (95%CI0.11–2.01) for isolated CNS relapse, 0.6 yr (95%CI o.30–0.89) for combined relapse and 0.93 (95%CI 0.78–1.07) for BM relapse (p=0.76). No correlation was found between initial CNS involvement and CNS relapse. An initial LDH value > 1,000 U/L was the only factor associated with higher risk of CNS relapse (p<0.001). Treatment of CNS relapse consisted of systemic and IT therapy in the 22 cases (cranial irradiation was added to one) and CR was attained in 7 (32%) out of 22 of these patients. Stem cell transplantation was performed in 4 patients and 3 patients developed a second CNS recurrence. The median overall survival (OS) after recurrence was 0.7 yr for isolated CNS relapse, 0.13 yr for combined relapse and 0.41 yr for BM relapse (p=0.11).

Conclusions. The frequency of CNS relapse in adult ALL patients receiving IT and systemic therapy for CNS prophylaxis is similar to that observed in protocols including cranial irradiation. An initial LDH value > 1,000 U/L was the only factor associated with higher risk of CNS relapse. Adult patients with CNS recurrence have a poor prognosis, although it is not different from that observed in BM relapses.

Author notes

Corresponding author

Sign in via your Institution