AUTHORS
Anushareddy Muddasani, MD, Carolina D Schinke, MD2, Sharmilan Thanendrarajan, MD3, Maurizio Zangari, MD4, Frits van Rhee, MD, PhD5
AFFILIATIONS
1 University of Arkansas for Medical Sciences, Little rock, Arkansas, United States
2 University of Arkansas for Medical Sciences, Little rock, Arkansas, United States
3 University of Arkansas for Medical Sciences, Little rock, Arkansas, United States
4 University of Arkansas for Medical Sciences, Little rock, Arkansas, United States
5 University of Arkansas for Medical Sciences, Little rock, Arkansas, United States
ABSTRACT
BACKGROUND: Peak absolute lymphocyte count (ALC) after ciltacabtagene autoleucel (cilta-cel) has been proposed as a widely available surrogate for CAR T-cell expansion. Hosoya et al. showed that a peak ALC ≥3000 cells/µL predicts delayed non-ICANS neurotoxicity, particularly parkinsonism. In a single-center series, Forsberg et al. reported that a peak ALC >5000 cells/µL was associated with atypical non-ICANS neurologic events and inferior overall survival. Whether the lower threshold of 3000 cells/µL likewise identifies patients with lower overall survival or a shorter interval to subsequent therapy is not established. We evaluated the association between a peak post-infusion ALC above 3000 cells/µL and overall survival (OS), time to next treatment (TTNT), and delayed neurotoxicity in multiple myeloma.
METHODS: We conducted a retrospective propensity-matched cohort study using a federated electronic health record network (TriNetX) for adults with multiple myeloma who received cilta-cel. Patients were stratified by peak ALC within 30 days of infusion into high (>3000 cells/µL) and low (<3000 cells/µL) cohorts. Propensity score matching on demographics, comorbidities, prior transplant, prior myeloma therapies, and presence of extramedullary disease yielded 355 patients per cohort. Primary endpoints were OS and TTNT, defined as time from infusion to initiation of any subsequent myeloma therapy. The secondary endpoint was a composite of delayed non-ICANS neurotoxicity occurring beyond the acute post-infusion period, identified using diagnostic codes. Risk ratios (RR) with 95% confidence intervals (CI) were calculated for the neurotoxicity outcome. Kaplan-Meier estimates were generated for OS and TTNT, with between-group differences assessed by log-rank test and hazard ratios (HR) with 95% CIs estimated via Cox proportional hazards models. All analyses were performed within the TriNetX analytics platform. Estimates are presented for high versus low peak ALC.
RESULTS: Among 710 matched patients, 1-year OS was 90.1% in the high-ALC cohort and 93.4% in the low-ALC cohort, with 27 versus 16 deaths (HR 1.73, 95% CI 0.93-3.22, p=0.078). TTNT was similar between cohorts, with 87.8% versus 89.0% of patients free of subsequent therapy at 1 year (34 vs 26 events; HR 1.35, 95% CI 0.81-2.25, p=0.248). Neither median OS nor median TTNT was reached. Delayed non-ICANS neurotoxicity occurred in 15.2% of high-ALC patients compared with 7.9% of low-ALC patients (RR 1.93, 95% CI 1.25-2.97, p=0.0023).
CONCLUSIONS: In this propensity-matched real-world cohort, a high peak ALC within 30 days of cilta-cel infusion was associated with a nearly two-fold higher risk of delayed non-ICANS neurotoxicity, without a corresponding improvement in overall survival or time to next treatment. A numerically higher mortality in the high-ALC group did not reach statistical significance. These findings are concordant with prior work establishing peak ALC as an accessible surrogate for CAR T-cell expansion and a predictor of delayed neurotoxicity after cilta-cel, and extend it by showing that the same threshold did not identify a subgroup with more favorable survival or a longer interval to subsequent therapy. A high peak ALC may therefore be more useful for identifying patients at risk of delayed neurotoxicity than for predicting survival or interval to subsequent therapy.
DISCLOSURES
Anushareddy Muddasani, MD: No relevant conflicts of interest to declare.