CNS Involvement in Adult Patients with B-Cell Acute Lymphoblastic Leukemia

Introduction: Central nervous system (CNS) involvement in patients with B-acute lymphoblastic leukemia (ALL) occurs in <10% of patients at the time of diagnosis and approximately 5% of patients at the time of relapse. Given continued advances in both diagnostic tools and therapeutic agents, we aim to describe the incidence of CNS involvement at diagnosis and risk of CNS relapse in a modern cohort of adult patients with ALL. Methods: We performed an IRB-approved retrospective analysis of all patients with newly diagnosed B-ALL treated at our institution between October 2011 and July 2020. Cytogenetic risk was categorized using established criteria (PMID 34657128, 33895809) and the threshold for complex cytogenetics was ³5 chromosomal abnormalities. CNS relapse was confirmed on cerebrospinal fluid cytology, corrected using the Steinherz/Bleyer algorithm as appropriate. Minimal residual disease (MRD) was assessed using flow cytometry, +/- polymerase chain reaction and next generation sequencing. All tests were 2-sided and p-values less than 0.05 were considered statistically significant. Analyses were conducted using SAS 9.4, S-plus 8.2, and SPSS v26. Results: Of 462 patients included in this analysis, 194 (42%) were female, 265 (57%) were white/Caucasian, 32 (7%) were black, and 19 (4%) were Asian. One hundred nineteen patients (26%) were Hispanic. The median age at diagnosis was 49 years (interquartile range (IQR) 34-65). Median lactate dehydrogenase (LDH) at diagnosis was 995 U/L (IQR 588-2124 U/L). One hundred eighty-six patients (40%) had BCR-ABL rearrangements, 37 (8%) had rearrangements in MLL, while 52 (11%) had Philadelphia-like ALL, of whom 31 (7%) had alterations in CRLF and 21 (5%) had alterations in JAK2. One hundred eighteen patients (26%) had high risk molecular alterations, 138 patients (30%) were low hypodiploid/near triploid, and complex cytogenetics were seen in 19 patients (4%). The most common initial therapy regimens included hyperfractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone (hyper-CVAD), +/- tyrosine kinase inhibitor and/or anti-CD20 monoclonal antibody (307 patients, 64%). One hundred and four patients (22%) received mini-hyper-CVD (PMID 37187201), with or without inotuzumab ozogamicin, blinatumomab, or anti-CD20 monoclonal antibody. Four hundred and one patients achieved a complete response to first-line therapy (87%), and 155 patients (32%) were MRD negative (MRD-). At a median follow-up of 45 months (IQR 13-82 months), 238 patients (52%) were still alive. Thirty-two patients (7%) had CNS involvement at the time of diagnosis. Twenty-six patients (6%) developed new CNS disease at a median of 23 months following diagnosis (IQR 13-41 months), 12 of whom (46%) had concurrent marrow relapse. Median CNS-relapse free survival was 71 months (95% confidence interval 48-94 months). Median survival after identification of CNS disease in the relapsed setting was 6 months (IQR 3-16 months). On univariate analysis CNS involvement at diagnosis was associated with increased circulating peripheral blasts (median 66% vs. 37%, p = 0.02), with a trend towards older age at diagnosis (median 57 vs. 49 years, p = 0.09). CNS relapse was associated with MRD positivity after first-line therapy (12/134 (9%) MRD+ patients compared to 14/328 (4%) MRD-/unknown patients, p = 0.05), elevated LDH at diagnosis (median 9115 vs. 3157, p < 0.001) and increased peripheral blasts at diagnosis (68% vs. 37%, p = 0.01). Molecular risk classification, complex cytogenetics, presence of BCR-ABL or MLL rearrangement, and first-line therapy regimen were not associated with risk of CNS relapse. Conclusions We found that MRD- response was associated with a decreased risk of CNS relapse, and increased peripheral blasts were associated with increased risk of CNS involvement at diagnosis and subsequent relapse. Further research is needed to characterize risk factors for CNS involvement in adults with ALL to better influence therapeutic strategies.