Anakinra Prophylaxis for Recurrent Severe ICANS Associated With Tarlatamab in Small Cell Lung Cancer: Case Report Authors: David W. Shia, Suwicha Limvorasak, Karen L. Reckamp, Kamya Sankar Published: 2026 ABSTRACT Tarlatamab is a bispecific T-cell engager that has demonstrated promising efficacy in treating patients with relapsed small cell lung cancer (SCLC). The incidence of immune-mediated adverse events, including cytokine release syndrome and immune effector cell-associated neurotoxicity syndrome (ICANS), has been reported at relatively low frequencies, with severe events occurring in a minority of patients. Consequently, real-world experience in managing severe ICANS associated with tarlatamab remains limited. We present a case of relapsed SCLC treated with tarlatamab that was complicated by recurrent episodes of severe ICANS. Subsequent administrations incorporated prophylactic anakinra, an interleukin-1 receptor antagonist, to mitigate neurotoxicity, which was well-tolerated and enabled the continuation of therapy, eliciting a significant clinical and radiographic response, including intracranial response. This case warrants further study of interleukin-1 blockade as a potential strategy for ICANS prophylaxis in patients with SCLC who may be at a higher risk for severe ICANS when treated with bispecific antibody therapy. Introduction Small cell lung cancer is an aggressive malignancy that is responsive to chemotherapy but commonly recurs. Tarlatamab is the first T-cell engager antibody-based therapy to gain FDA approval for solid tumors1 specifically for use upon progression following platinum-based combination chemotherapy with or without immune checkpoint inhibitor therapy for patients with extensive-stage small cell lung cancer (ES-SCLC).2 Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) are unique immune-mediated adverse effects associated with T-cell engaging therapies. ICANS with tarlatamab has been reported in approximately 10% of patients, with grade3 ≥ 2 events occurring in 1.3% and recurrent events occurring in 1.6%. Severe ICANS (grade3 ≥ 3) is rare, with no cases observed at the 10 mg dose in the phase II trial, which utilized prophylactic dexamethasone (8 mg, IV) with the first two doses.1 Furthermore, a single case of grade 5 ICANS with death attributed to concurrent CRS was reported in the randomized phase III trial.2 Thus, experience with the management of severe ICANS resulting from tarlatamab remains limited. Importantly, recent reports suggest a far higher rate of ICANS, ranging from 23.3 to 40.9%, with tarlatamab treatment in the real-world setting compared to that observed in trials.3,4 While IL-6 inhibitors such as tocilizumab are effective in the management of CRS, the role of cytokine-targeted therapies in treating ICANS among patients receiving CD3-engaging BiTE therapies remains poorly defined. Anakinra, an IL-1 receptor antagonist, has been studied as a prophylactic strategy in a single-arm phase II study of patients with relapsed/refractory large B-cell lymphoma receiving anti-CD19 CAR-T treatment.5 Despite the relatively low sample size of this study, the rates of all-grade and grade3 ≥ 3 ICANS were 22% and 9%, respectively. These rates compare favorably to prior clinical trial and real-world data with axicabtagene ciloleucel, which reported all-grade and grade3 ≥ 3 neurotoxicity rates of 56–64% and 28–38%, respectively.5 A recent retrospective analysis of rates of CRS and ICANS before and after a change in institutional toxicity management to incorporate anakinra further found that while rates of CRS and ICANS were similar, the duration of CRS and ICANS was significantly shorter with the protocol that incorporated anakinra use. Furthermore, this study found no difference in complete response rate between groups, suggesting that anakinra did not negatively impact overall efficacy.6 Here, we report a case of ES-SCLC treated with tarlatamab complicated by recurrent severe ICANS with rapid onset within 24 hours of administration requiring dose interruption, in which anakinra was utilized as a prophylaxis strategy. Case Presentation A 56-year-old Asian male with a 40 pack-year smoking history presented with cough and urinary retention. He was diagnosed with ES-SCLC, arising from a right lower lobe primary lung mass, with diffuse lymphadenopathy, scattered soft tissue metastases, and intracranial and spinal cord involvement. He received palliative spine and whole-brain radiation (Figure 1), followed by carboplatin, etoposide, and atezolizumab, achieving a significant reduction in thoracic disease and marked improvement in intracranial metastases (Figure 3A and B). He transitioned to maintenance atezolizumab and lurbinectedin but developed intractable low back pain after two cycles. MRI demonstrated diffuse spinal cord and cauda equina enhancement consistent with leptomeningeal disease with concurrent progression of extensive intracranial metastases involving innumerable lesions in the posterior fossa, midbrain, and bilateral cerebral hemispheres. PET-CT and MRI demonstrated systemic disease progression (Figure 2A and 3C).Figure 1Treatment timelineFigure 2Radiographic systemic response to tarlatamab (CT chest, PET/CT); (A) Representative maximal intensity projections of FDG-PET/CT at relapse after first-line therapy (left) and after initial 2 doses of tarlatamab and 2 additional doses of rechallenge (right), (B) Representative sections of CT chest at relapse after first-line therapy (left) and after initial 2 doses of tarlatamab (right).Figure 3Magnetic resonance characterization of central nervous system response to whole brain radiation with carboplatin + etoposide + atezolizumab, subsequent recurrence and response to 2L tarlatamab; representative sections of MRI brain and cervical and lumbar spine at (A) diagnosis prior to 1L treatment showing innumerable intracranial lesions, (B) after whole brain radiation and 2 cycles of carboplatin + etoposide + atezolizumab showing partial response, (C) at relapse prior to tarlatamab treatment showing marked intracranial disease progression, (D) after 2 cycles of tarlatamab demonstrating near complete resolution of previously innumerable intracranial lesions. All images of the brain are contrast-enhanced T2 FLAIR. The images of the cervical and lumbar spine are contrast-enhanced T1, with the exception of lumbar spine in part D, which is contrast-enhanced T1 FLAIR. Treatment timeline Radiographic systemic response to tarlatamab (CT chest, PET/CT); (A) Representative maximal intensity projections of FDG-PET/CT at relapse after first-line therapy (left) and after initial 2 doses of tarlatamab and 2 additional doses of rechallenge (right), (B) Representative sections of CT chest at relapse after first-line therapy (left) and after initial 2 doses of tarlatamab (right). Magnetic resonance characterization of central nervous system response to whole brain radiation with carboplatin + etoposide + atezolizumab, subsequent recurrence and response to 2L tarlatamab; representative sections of MRI brain and cervical and lumbar spine at (A) diagnosis prior to 1L treatment showing innumerable intracranial lesions, (B) after whole brain radiation and 2 cycles of carboplatin + etoposide + atezolizumab showing partial response, (C) at relapse prior to tarlatamab treatment showing marked intracranial disease progression, (D) after 2 cycles of tarlatamab demonstrating near complete resolution of previously innumerable intracranial lesions. All images of the brain are contrast-enhanced T2 FLAIR. The images of the cervical and lumbar spine are contrast-enhanced T1, with the exception of lumbar spine in part D, which is contrast-enhanced T1 FLAIR. He initiated second-line therapy with tarlatamab (1 mg, IV, cycle 1 day 1) with dexamethasone (8 mg, IV) premedication administered in the outpatient setting with a scheduled hospital admission for observation. He developed somnolence and confusion 16 hours after completion of the initial dose. Brain imaging was unremarkable, and electroencephalogram demonstrated diffuse slowing. His immune effector cell encephalopathy (ICE) score declined from 9 at baseline to 1, consistent with grade 3 ICANS. There was no evidence of concurrent CRS. Dexamethasone (10 mg, IV) was administered every 6 hours per institutional protocol, resulting in marked improvement in mentation within 24 hours. Nine days after the initial dose, he recovered to his baseline mental status with significant improvement in low back pain, no longer requiring analgesics. After receiving the day 8 dose of tarlatamab (10 mg, IV) with dexamethasone (8 mg, IV) premedication, he again developed neurotoxicity, becoming non-verbal 26 hours after the infusion. His ICE score declined from 10 to 0, consistent with recurrent grade 3 ICANS. Dexamethasone (10 mg, IV, q6h) was reinitiated, with improvement in mental status within 24 hours. Interval CT imaging five weeks after the priming dose demonstrated a marked reduction in the size of the primary lung mass and associated lymphadenopathy (Figure 2B). The patient reported improvement in fatigue, increased ambulation, and a decrease in the size of palpable soft tissue nodules. Given his clinical and radiographic response, re-challenge with tarlatamab with repeated step-up dosing (1 mg, IV; cycle 1 day 15) was offered after discussion of the risk of recurrent high-grade ICANS. In addition to pre-medication with dexamethasone (8 mg, IV), prophylactic anakinra (100 mg, subcutaneous) was administered 18 hours after tarlatamab infusion. He tolerated this dose without any evidence of recurrent ICANS. Prophylactic anakinra was again administered with a subsequent 10 mg dose of tarlatamab, without evidence of recurrent neurotoxicity or other common tarlatamab-related adverse events, such as dysgeusia. All doses thereafter were administered without anakinra or dexamethasone premedication, and no further ICANS events occurred. Restaging imaging after two cycles of tarlatamab demonstrated a significant response at all disease sites, including resolution of cervical, thoracic, and lumbar cord nodular enhancement and near-complete resolution of the intracranial disease (Figure 2A and 3D). He remains on therapy with continued clinical improvement. Discussion While severe ICANS rates were low in the trials that evaluated tarlatamab for the treatment of SCLC,1,2 the incidence is likely to be higher in a real-world setting. Real-world data suggest that a higher intracranial disease burden, particularly untreated, active, or high-volume brain metastases, may correlate with an increased risk and severity of ICANS. The recurrence of grade3 ≥ 3 ICANS warrants consideration of the permanent discontinuation of therapy. However, rapid responses to the initial dosing of tarlatamab, including intracranial disease, have been observed,7 which may reduce the risk of recurrent ICANS due to a lower tumor burden at the time of subsequent dosing.3 Additionally, while concurrent radiotherapy with tarlatamab appears safe,8 the relationship between prior whole-brain radiation and the risk of ICANS will require further delineation. In this case, the patient experienced recurrent grade 3 ICANS after the first and second doses of tarlatamab with a concurrent rapid disease response. As tarlatamab moves from the third-line to the second-line setting, broader clinical use will allow for a more accurate estimation of the incidence of severe ICANS. While the pathogenesis of ICANS remains incompletely understood, the activation of endothelial cells and disruption of the blood-brain barrier, in part by inflammatory monocyte-derived cytokines such as TNF-α, IL-6, and IL-1, are putative mechanisms. Preclinical evidence suggests a more central role of IL-1 over IL-6 in the pathogenesis of both CRS and ICANS, as IL-1 production precedes that of IL-6 by up to 24 hours and administration of anakinra, but not tocilizumab, prevents the occurrence of a rapid-onset neurologic syndrome in leukemic mouse models treated with anti-CD19 CAR-T cells.9 Differential ability to cross the blood-brain barrier may also account for differences in efficacy.10 Given the patient’s excellent initial response and marked clinical improvement, re-challenge was attempted with the use of a single prophylactic dose of anakinra. Anakinra is generally well tolerated, though common adverse effects include injection site reactions and an increased risk of infection, especially in combination with TNF inhibitors such as etanercept. The timing of anakinra initiation was based on detectable increases in IL-1RA and IL-6 levels within 24 hours of CAR-T administration in ZUMA-1.11 An early report of cytokine dynamics in patients treated with tarlatamab seems to support this.12 No further neurotoxicity occurred, and anakinra was subsequently de-escalated. While the improved tolerance cannot be fully attributed to anakinra, since the patient already had a reduced disease burden at the time of re-challenge, this case nonetheless highlights a potential role for IL-1 blockade in tarlatamab-associated ICANS prophylaxis. Recent retrospective data have also suggested a role for tocilizumab as prophylaxis against CRS and ICANS in patients treated with tarlatamab.4 Comparatively, anakinra offers a more favorable pharmacokinetic profile with a shorter half-life and may allow for a shorter duration of immunosuppression, but the absence of endpoint clinical data precludes a definitive conclusion at this time. Further studies are needed to define risk factors for high-grade ICANS with tarlatamab and to identify patient subgroups who may benefit from prophylactic interventions. Conclusion This is a case of relapsed ES-SCLC treated with tarlatamab, complicated by recurrent severe ICANS upon therapy initiation. A rapid response to initial treatment, including a marked intracranial response, merited consideration of re-challenge. Anakinra was used prophylactically in subsequent doses, was well tolerated, and did not result in any adverse effects. Further investigation is needed to define the risk of severe ICANS in real-world practice and to evaluate the potential role of IL-1 blockade in safely optimizing tarlatamab therapy in patients who are at risk of developing severe ICANS. CRediT Authorship Contribution Statement David W. Shia: Data curation, Visualization, Writing – original draft, Writing – review & editing. Suwicha Limvorasak: Visualization, Writing – review & editing. Karen L. Reckamp: Supervision, Writing – review & editing. Kamya Sankar: Conceptualization, Visualization, Data curation, Supervision, Writing – review & editing. Disclosure K.L.R has research funding to the institution: BioNTech; Genentech; Blueprint; Daiichi Sankyo; Elevation Oncology; Johnson & Johnson and serves as a consultant with honoraria to self: AstraZeneca, Genentech, GlaxoSmithKline, Immunity Bio, Johnson & Johnson, Lilly, Merus, Novartis, Nuvation, Pfizer. K.S. has research funding to their institution from Genentech, Black Diamond Therapeutics, Eli Lilly, Harpoon Therapeutics, AstraZeneca, Amgen, and Merck. All the remaining authors no conflict of interest.