Introduction: With increased use of CAR-T for relapsed/refractory (R/R) large B-cell lymphoma (LBCL), CAR-T related complications including cytokine release syndrome (CRS) and ICANS pose a significant clinical challenge. While CAR-T mediated inflammation leading to endothelial activation and blood-brain barrier disruption may play a key role in ICANS, the exact mechanism remains unclear. Prognostic or predictive biomarkers for ICANS are not well established. Recent reports (Karschnia et al, Blood 2019) suggested an association between ICANS and inferior overall survival (OS). To better understand ICANS, we herein report a single-center analysis of LBCL patients treated with CD19-directed CAR-T, exploring the associated clinical features, predictive biomarkers, and its prognostic significance.
Methods: Patients (pts) with R/R LBCL treated with axicabtagene ciloleucel (Axi-cel) between 4/2018-5/2019 were identified. Data regarding patient and disease characteristics, treatment course, and clinical outcomes was collected (Table 1). Laboratory variables were collected at time of mononuclear cell harvest, day of initiation of lymphodepletive therapy, and day of CAR-T infusion (D0). CRS and ICANS were graded per the Lee and CTCAE v4.03 criteria, respectively.
Time to progression (TTP) and OS were estimated by the Kaplan-Meier method and groups compared with the logrank test. Cox Proportional Hazard Model was applied for prognostic modeling. Binary logistic regression was used for multivariable analysis of patient characteristics at D0 associated with ICANS.
Results: Forty-five pts with R/R LBCL (35 DLBCL, 7 TFL, 3 PMBCL) treated with Axi-cel were identified (Table 1). Twenty-five pts developed ICANS: n=7 with Grade (Gr) I-II, n=18 with Gr III-IV. Most common initial ICANS symptoms were dysgraphia, confusion, and somnolence; median time to ICANS was 5 days (range 3-11 days). Acute abnormalities were seen on brain MRI in 7 (28%) ICANS pts; EEG done in 10 ICANS pts showed diffuse slowing in all pts and focal slowing in 3 pts. All ICANS pts had preceding CRS, treated with tocilizumab (n=25, median 2.5 doses) and siltuximab (n=2). Twenty-three (92%) pts with ICANS required steroid therapy, with a median total dose equivalent to 221 mg of dexamethasone for a median duration of 12.5 days. Two pts exhibited protracted neurotoxicity manifested by short-term memory loss and profound weakness with immobility.
Twenty-two (49%, 95% CI 34-64%) pts achieved CR, 16 PR, and 5 PD. At time of analysis, n=18 had disease relapse/progression and n=35 were alive. Censoring pts that progressed, the median observation time for TTP was 9.3 months. Censoring pts who died, the median observation time for OS was 7.9 months. At 9 months OS (±1 SE) was 76.1%±7.4%, and 56.0%±8.0% were progression-free.
Logistic regression showed increasing fibrinogen level at D0 was associated with increasing risk of ICANS (p=0.003) and specifically, Gr III-IV ICANS [p<0.001, OR 3.02 for a case with a D0 fibrinogen of 533.5 mg/dL (3rd quartile of population distribution) vs. 382 mg/dL (1st quartile)]. Germinal center B-cell (GCB) subtype was also associated with risk of Gr III-IV ICANS (p=0.022, OR=11.9). Patient age, sex, race, disease stage at diagnosis, and D0 platelet count, WBC, LDH, CRP, and ferritin were not predictive of ICANS.
ICANS was not associated with TTP (p=0.66, HR=1.24, 95% CI 0.48-3.2) nor with OS (p=0.47, HR=1.67, 95% CI 0.42-6.7). Elevated baseline fibrinogen was associated with inferior TTP (p=0.04, HR=1.89, 95% CI 1.03-3.5) and shorter OS (p=0.07, HR=2.12, 95% CI 0.94-4.8); Neither duration nor total dose of steroid treatment significantly affected TTP or OS.
Conclusion: ICANS after Axi-cel for R/R LBCL was seen in approximately 50% of patients, 72% of which was severe (Gr III-IV). Contrary to prior reports, no association was seen between development of ICANS and TTP or OS. Elevated D0 fibrinogen and GCB subtype identified pts at higher risk for ICANS. D0 fibrinogen was also a prognostic marker for inferior TTP and OS.
The novel finding of baseline fibrinogen as a marker both predictive of ICANS and prognostic for treatment outcomes after CAR-T may be attributed to its role in inflammation and vascular injury, which warrants further investigation.
Hutnick:Kite/Gilead: Other: Yescarta Speakers Bureau, Speakers Bureau. Badros:Celgene Corporation: Consultancy; Amgen: Consultancy.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal