A great mitigator of non-ICANS neurotoxicities? Free

In this issue of Blood, Blumenberg et al¹ discuss the benefit and theoretical mechanism of using cyclophosphamide to mitigate potentially fatal non-ICANS (immune effector cell–associated neurotoxicity syndrome) neurotoxicities (NINTs), which can be difficult to otherwise manage and are associated with debilitating symptoms for patients. They review their experiences of using cyclophosphamide to treat 13 patients with relapsed and refractory multiple myeloma who received ciltacabtagene autoleucel, a B-cell maturation antigen (BCMA)–targeted chimeric antigen receptor (CAR) T-cell therapy and subsequently developed steroid-refractory NINTs.

BCMA-targeted CAR T-cell therapies have transformed the treatment paradigm for relapsed and refractory multiple myeloma, improving patient outcomes and overall survival.^2-4 Akin to other CAR T-cell therapies, this treatment has been associated with cytokine release syndrome (CRS) and ICANS. However, distinct from other CAR therapies, BCMA-targeted CAR T-cell therapy has also been associated with a neurotoxicity syndrome called NINT. Although the mechanism underpinning this toxicity has yet to be fully elucidated, NINTs are hypothesized to develop secondary to high CAR T-cell expansion.^5,6 Patients with NINTs can present with movement disorders (such as parkinsonism-like symptoms that can include tremor, bradykinesia, rigidity, gait changes, and masked facies potentially due to localized on-target, off-tumor toxicity in the basal ganglia),^7,8 cognitive symptoms, personality changes, cranial nerve palsies, peripheral neuropathies, and even Guillain-Barré syndrome. Although steroids can be used for the treatment of NINTs, there are a subset of patients who can continue to have symptoms that are refractory to steroid treatment, and this report highlights the potential benefit of using cyclophosphamide to ablate CAR T cells and mitigate NINTs in this patient population.

In this cohort, there were a total of 42 patients treated with commercially available ciltacabtagene autoleucel for refractory and relapsed multiple myeloma, and 13 patients (30.9%) developed NINTs. Of these 13 patients, all developed CRS and 3 had ICANS, and both CRS and ICANS completely resolved with tocilizumab or corticosteroids. Then, after a period of toxicity freedom, and notably, after 5 of 13 patients had achieved a partial or complete myeloma response, patients developed NINT symptoms. These clinical symptoms consisted of movement disorders with parkinsonism features such as bradykinesia (n = 4), rigidity (n = 6), gait changes (n = 5), abnormal posture (n = 1), and tremor (n = 6), neurocognitive changes that included aphasia (n = 3) and confusion (n = 4), personality changes (n = 3), with flat affect (n = 4) and reduced facial expression (n = 2), cranial nerve palsies (n = 4), peripheral neuropathies (n = 2), and Guillain-Barré syndrome (n = 1) requiring temporary mechanical ventilation. Patients were initially treated with corticosteroids, but this only led to partial symptom resolution in some patients (of note, patients with isolated cranial nerve palsies were more likely to respond to treatment with corticosteroids, with or without IV immunoglobulin). Accordingly, 6 patients were then given cyclophosphamide (dosage 2000 mg/m², aside from 1 patient who received 1500 mg/m²) at a median of 19 days after the onset of symptoms. Of 6 patients, 4 had durable resolution of their symptoms after cyclophosphamide administration, with complete resolution of symptoms in 2 patients within 68 and 144 days of cyclophosphamide treatment, respectively, and 2 patients experienced ongoing clinical improvement. This potential for a delayed resolution of NINT symptoms after cyclophosphamide treatment is an important prognostic pearl for clinicians. Furthermore, patients who received cyclophosphamide earlier after symptom onset (range, 5-21 days vs range, 20-272 days) responded better, suggesting that earlier intervention with cyclophosphamide may be more beneficial for symptomatic management. This will need to be validated in prospective studies.

Interestingly, most patients who developed NINTs had elevated peak absolute lymphocyte counts, and all 5 tested patients had elevated CAR T cells before cyclophosphamide treatment. These findings are indicative of CAR T-cell persistence after immunosuppressive treatment and are suggestive of the potential role of CAR T-cell expansion in the development of NINTs. Accordingly, the absolute lymphocyte counts decreased in all but 1 patient who received cyclophosphamide, illustrating that cyclophosphamide effectively ablated the CAR T cells in these patients. Importantly, after the ablation of these CAR T cells, the initial complete responses found in 5 patients were sustained (median follow-up of 284 days). Patients were at high risk for severe infections during this time, which underscores the need for close infection preventative and treatment measures while patients are undergoing BCMA CAR T-cell therapy.

Even though this study had only a small sample size and is retrospective in nature, it provides critical insight to a potential treatment regimen for NINTs. BCMA-targeted CAR T-cell therapies are increasingly being used for the treatment of refractory and relapsed multiple myeloma, and as the field enhances CAR T-cell treatments to further increase CAR T-cell expansion and persistence to improve clinical efficacy, it is possible that more patients will develop steroid-refractory NINTs, particularly if CAR T-cell expansion is the driving mechanism for the development of NINTs. It will therefore be imperative to identify anti-inflammatory treatments, such as cyclophosphamide, to safely manage and mitigate these neurotoxicities, as they can be profoundly debilitating, impair quality of life, and even be fatal, if untreated. Future larger cohort and multi-institutional studies will be needed to prospectively assess whether cyclophosphamide can be effectively used to treat NINTs and to obtain clinical, laboratory, and imaging correlatives before, during, and after treatment to improve our understanding of the mechanisms underlying NINT development. Thus, the findings in this report will help inform future prospective studies on BCMA-targeted CAR T-cell therapy and its associated neurotoxicities, educate clinicians on potential anti-inflammatory treatment strategies, and will help optimize the safe delivery of BCMA-targeted CAR T-cell therapy in this vulnerable patient population.

Conflict-of-interest disclosure: The author declares no competing financial interests.