Real world experience of anakinra for the treatment of immune effector cell-associated neurotoxicity syndrome in large B-cell lymphoma Free

Background Anakinra, an IL-1 receptor antagonist approved for rheumatologic conditions, has been increasingly used off-label as a corticosteroid-sparing agent for refractory cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) following chimeric antigen receptor T-cell therapy (CART). Preclinical models have demonstrated that anakinra may be effective for CRS and ICANS without compromising CAR T-cell function, and small retrospective studies support its safety in this setting. While anakinra has been shown to be effective in a fraction of cases of corticosteroid-refractory ICANS, real-world data remain limited, and the identification of clinical and laboratory factors to guide patient selection and optimal treatment initiation is warranted.

Methods This is a retrospective single center analysis of patients with large B-cell lymphoma (LBCL) consecutively treated with anakinra for ICANS after standard of care anti-CD19 CART between February 2020 and November 2024. Responders were defined by ICANS resolution after initiation of anakinra. Baseline characteristics were collected pre-lymphodepletion. Serum inflammatory and tumor burden biomarkers were collected during the first 30 days after CART infusion. CRS and ICANS graded per ASTCT criteria; lymphoma response per Lugano 2014 criteria. Chi-square test or Fisher's exact test and Wilcoxon rank sum test were used for categorical and continuous variables, respectively. The area under the curve (AUC) for ferritin, lactate dehydrogenase (LDH), and C-reactive protein (CRP) was calculated over the first 30 days after CART and compared between groups using the Wilcoxon rank sum test.

Results Forty-two patients were included: 31 (74%) were responders and 11 (26%) were non-responders. The median age was 64.5 years (range 19-78) and 54.8% (n=23) were male. Most (66.7%, n=28) had stage IV disease, median International Prognostic Index score was 3 (range 1-5), median prior therapy lines were 2 (range 1-6), and 73.8% (n=31) had refractory disease. Thirty-six (85.7%) patients received axicabtagene ciloleucel, 4 (9.5%) lisocabtagene maraleucel and 2 (4.8%) tisagenlecleucel. At anakinra initiation, median ICANS grade (G) was 3; 42.6% had concomitant CRS of any grade and 16.7% had G3-4 CRS. Thirty-nine (92.9%) patients received corticosteroids prior to anakinra, with a median cumulative dexamethasone equivalent dose of 40.0 mg (range 0-1041.9 mg); 95.2% received additional corticosteroids after initiation of anakinra, with a median cumulative dexamethasone equivalent of 585.0 mg (range 0-2432.1 mg).

Compared to responders (n=31), non-responders (n=11) had significantly greater corticosteroid exposure (median cumulative dexamethasone equivalent dose, 1040 mg vs 398 mg, p < 0.001), lower baseline hemoglobin (median 8.5 vs 10.2 g/dL, p = 0.010), higher baseline CRP (median 73.0 vs 11.9 mg/L, p = 0.015), higher baseline serum ferritin (median 1236 vs 399 ng/mL, p = 0.040), and higher baseline serum LDH (median 367.0 vs 218.5 U/L, p = 0.0165). During hospitalization, non-responders had significantly higher peak ICANS severity, with 100% experiencing G4 ICANS versus 25.8% among responders (p < 0.001). Non-responders had a higher rate of G3-4 CRS during the same timeframe (72.7% vs 25.8%, p = 0.006). During the first 30 days after CART infusion, non-responders had a significantly higher AUC for CRP (median 1165.6 vs 530.2 mg × day/L, p = 0.011) and a numerically higher AUC for ferritin (126323.0 vs 43845.5 ng × day/mL, p = 0.0553), but not LDH (9644.0 vs 7164.5 U × day/L, p = 0.3451). When comparing 179 LBCL patients who received CART before 05/02/20 (date of introduction of anakinra for CRS/ICANS management at our institution) to the 325 who did afterwards, CRS/ICANS related mortality declined from 8.57% (6 of 70 deaths) to 4.14% (7 of 169 deaths).

Conclusion This study represents the largest real-world experience of patients with LBCL treated with anakinra for corticosteroid-refractory ICANS after CART. The use of anakinra was associated with a high response rate, responses being more likely in patients with low inflammatory biomarkers, and with a decrease in proportion of deaths attributable to CRS/ICANS. Larger prospective studies are needed to better optimize patient selection and use of anakinra in patients with CART-associated toxicities, and to potentially facilitate formal approval of this agent, currently used off-label.