Effect of Lumbar Puncture Timing on Neurotoxicity with Blinatumomab Therapy in a Real-World Cohort of Acute Lymphoblastic Leukemia (ALL)

BACKGROUND:

Blinatumomab is a CD19/CD3 bispecific T-cell engager (BiTE) therapy which has revolutionized the treatment landscape of ALL. Cytokine release syndrome (CRS) and neurotoxicity are two major toxicities that can occur, with neurotoxicity being the most common cause for early termination of therapy. To date, the optimal timing of lumbar punctures (LP) with intrathecal chemotherapy (IT) during blinatumomab therapy and its effect on neurotoxicity remains unclear. Here we report our experience with blinatumomab related neurotoxicity.

METHODS:

We conducted a retrospective chart review of patients who were treated with blinatumomab between January 2016 to June 2024. Neurotoxicity was defined as any neurological event that was attributed to blinatumomab based on medical documentation or that led to interventions (e.g., stopping the drug and/or treatment with steroids). Time from start of blinatumomab to LP was categorized as follows. Day 0 was the start day of blinatumomab therapy. Window 1 = LP between Day -7 to day 14, Window 2: Day 0 to Day 28; Window 3 = Day 15 - Day 42 ; Window 4 = Day 29-.

RESULTS:

A total of 69 patients received blinatumomab during the study period with 179 corresponding cycles. Median number of cycles per patient was 2 (Range 1 to 8). There were 70 encounters for cycle 1, 50 for cycle 2, 28 for cycle 3, 4 for cycle 4, 6 for cycle 5, 3 for cycle 6 and 1 each for cycle 7 and 8. 28 patients received blinatumomab for relapsed/refractory disease (R/R), 25 for measurable residual disease (MRD+) disease and 16 for maintenance/consolidation (MRD-). Median blast % at the initiation of blinatumomab was 50% (Range 7-96%). 73 cycles were associated with at least 1 LP. IT was given with each cycle. 48 LPs were done within window 1; 47 in window 2 ; 22 in window 3 and 19 during window 4, 3 LPs were outside the 42 days cycle window.

19 patients experienced at least 1 neurotoxicity event during treatment with blinatumomab (19/69; 27%), with 23 corresponding neurotoxicity events (23/179; 12.8%). 19 events required permanent or temporary cessation of blinatumomab. 3 patients needed permanent discontinuation.17 events occurred in cycle 1 (24%), 5 in cycle 2 (10%) and 1 in cycle 3 (5%) (p 0.05). Median time to neurotoxicity was 3 days (Range 0 to18). 17 events (17/23 ; 73.9%) occurred with concurrent grade I CRS. All events resolved with appropriate interventions. Median blast % in the neurotoxicity cohort was 44.5% vs. 51% in the no neurotoxicity cohort. Incidence of neurotoxicity in R/R, MRD+ and MRD- cohorts was 16%, 14% and 18%, respectively.

Incidence of ICANS in patients with vs. without LP was 13% vs. 12 % for any cycle. During Cycle 1 there was no statistically significant association found between timing of LP and neurotoxicity, although incidence was numerically higher in window 1 or 2 vs. window 3 or 4 (26.9 vs. 18%).

CONCLUSIONS:

The neurotoxicity incidence in our cohort is comparable to previously reported data and predominantly seen during cycle 1 and cycle 2 of therapy. No difference was seen based on disease setting of blinatumomab administration or blast percentage at initiation. Administration of LP with IT concurrently with blinatumomab did not increase the risk for ICANS. Early (prior to day 14) versus late (after day 14) LP with IT did not significantly alter neurotoxicity rates. Confirmation of our findings in larger cohorts is warranted given the small sample size and retrospective nature of this study.