Introduction. In a modern context of improved management of acute lymphoblastic leukemia (ALL), central nervous system (CNS) involvement at diagnosis remains an obstacle towards long-term cure. We have previously reported that flow cytometry (FCM) is better than conventional cytology (CC) in demonstrating the presence of leukemic cells in patients' (pts) cerebrospinal fluid (CSF), especially in samples with low cell counts. In the framework of the national Campus ALL program aimed at improving the management of adult ALL patients enrolled in the GIMEMA protocols, we retrospectively evaluated the incidence of occult CNS positivity and its impact on outcome in 221 adult pts with newly diagnosed ALL from 11 centers.

Methods. Ninety-four patients (42%) were females and 127 (58%) males, with a median age of 44 years (range 17-80), a median white blood cell (WBC) count of 10.6x109/L (range 0.1-457). One hundred and seventy pts (77%) had B-lineage ALL. Cytogenetic/genetic data were available in 167 (75%) pts: 58 (35%) had a BCR/ABL rearrangement, 14 (8%) a complex karyotype and 11 (6%) a MLL rearrangement. Pts were treated according to the GIMEMA/NILG ALL protocols or with the Hyper-CVAD program. Ninety-eight pts underwent an allogeneic stem cell transplant (ASCT). Median follow up was 26.8 months (range 1-136.6). All CSF samples were evaluated both by CC and FCM. The presence of ≥10 clonally restricted or phenotypically abnormal events was regarded as a FCM positivity. Based on the results of CSF examination, three different categories were recognized: manifest CNS+ (CC+FCM+), occult CNS+ (CC-FCM+) and CNS- (CC-FCM-).

Results. Overall, 16 (7%) pts had manifest CNS+, 39 (17%) occult CNS+ and 166 (75%) were CNS-Median age, WBC count, B/T lineage, cytogenetic/genetic features did not differ significantly between the three categories. A complete remission (CR) was achieved in 178 (80%) pts, 9 (4%) died early in induction and 104/178 (58%) experienced a relapse. The frequency of CR rate did not vary significantly across the three identified categories. In univariate analysis, the CNS status correlated significantly with the incidence of relapse (p=.004) and with censor (<.00002). The four-year overall survival (OS) for manifest CNS+, occult CNS+ and CNS- pts was 6%, 17% and 46%, respectively (p=.0007) (Figure). No difference in terms of OS was observed between manifestand occult CNS+ pts (p=.30). In multivariate analysis a baseline status of manifestor occult CNS+ (RR: 1.98, 95%CI 1.47-2.43, p=0.00001) and age>45 years (RR:1.45, 95% CI 1.47-2.43, p=0.04)were independently associated with a lower OS.

Conclusions. Our large, multicenter CAMPUS ALL study showed that i) in ALL adult pts, FCM allows to detect occult CNS disease, even in conditions of low spinal fluid leukemic count; ii) the presence of occult CNS disease is associated with an unfavorable outcome. Further prospective studies on larger series are needed to confirm these data.


Del Principe:Gilead: Membership on an entity's Board of Directors or advisory committees. Fracchiolla:Gilead: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Pfizer Pharmaceuticals: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Merck: Honoraria, Speakers Bureau; Amgen: Membership on an entity's Board of Directors or advisory committees. Foà:NOVARTIS: Speakers Bureau; CELGENE: Other: ADVISORY BOARD, Speakers Bureau; AMGEN: Other: ADVISORY BOARD; JANSSEN: Other: ADVISORY BOARD, Speakers Bureau; CELTRION: Other: ADVISORY BOARD; GILEAD: Speakers Bureau; ABBVIE: Other: ADVISORY BOARD, Speakers Bureau; INCYTE: Other: ADVISORY BOARD; ROCHE: Other: ADVISORY BOARD, Speakers Bureau.

Author notes


Asterisk with author names denotes non-ASH members.