LocaTIAN, locaTIAN, locaTIAN Free

In this issue of Blood, Kaulen et al¹ provide, to our knowledge, the first comprehensive description of tumor inflammation–associated neurotoxicity (TIAN) in the setting of chimeric antigen receptor (CAR) T-cell therapy for central nervous system (CNS) lymphoma. With new therapies come new toxicities, and TIAN was recently defined by consensus to capture a pseudoprogression syndrome that has been observed during CAR T-cell therapy for tumors of the brain and spinal cord.² The hypothesized mechanism is inflammation around the tumor mass triggered by T-cell antitumor activity, similar to what has been observed with immune checkpoint inhibitors in solid tumors. Unique to the brain, even small changes in lesion volume can be dangerous, especially if there is mass effect on critical structures or obstruction of cerebrospinal fluid flow.

Treatment of primary or secondary CNS lymphoma with CD19-directed CAR T cells can be effective, with complete response rates of ∼50% reported in real-life cohorts.^3,4 Patients with CNS disease were excluded from the pivotal CD19-directed CAR T-cell clinical trials because of concerns about focal CNS inflammation related to tumor location. This concern so far has not been borne out in the literature, which suggests a similar frequency of neurotoxicity after CD19-directed CAR T-cell treatment in patients with or without active CNS involvement. However, the main framework previously available for reporting these neurotoxicities has been immune effector cell–associated neurotoxicity syndrome (ICANS).⁵ ICANS is thought to be due to off-tumor inflammatory-mediated global cerebral dysfunction and typically develops in the setting of cytokine release syndrome. The new work by Kaulen et al suggests that this attribution will have to be reexamined and that concerns regarding complications due to focal peritumoral inflammation are indeed warranted.

In this single-institution experience, TIAN developed in 10 of 60 patients treated with CD19-targeted CAR T cells for primary (n = 20) or secondary CNS lymphoma (n = 40). Seven of these patients had been previously reported to have ICANS but were reclassified as TIAN based on imaging and symptom review. TIAN was diagnosed when focal neurologic symptoms such as new limb weakness correlated with tumor location. Encouragingly, patients with TIAN had better disease responses than patients without TIAN, suggesting that TIAN is a direct consequence of effective on-tumor, on-target T-cell activity. Histopathology findings in a patient with TIAN who died from pneumonia on day 29 after CAR T-cell infusion showed complete CNS lymphoma remission with dense immune cell infiltration, including CAR T cells. The concept of TIAN as a CNS-restricted focal inflammatory syndrome is also supported by correlative findings: patients with TIAN had low rates of concurrent cytokine release syndrome and no associated increases in systemic biomarkers of inflammation. Notably, CNS lymphoma lesion volume was predictive of TIAN and may serve as a future clinically useful TIAN prognosticator. This is in contrast with ICANS, which is strongly associated with cytokine release syndrome and elevated systemic inflammatory cytokines, does not correlate with CNS lesion size, and does not portend favorable disease outcomes. But note, patients can have TIAN and ICANS simultaneously: 1 patient with TIAN had concurrent aphasia and confusion that could not be explained by the location of the CNS lymphoma lesions. Both ICANS and TIAN start in the first week after CAR T-cell infusion (day 3.5 for TIAN and day 4 for ICANS), so careful neurologic phenotyping with imaging and bedside examination is crucial for distinguishing the 2 entities.

The TIAN consensus statement² distinguishes 2 subtypes of TIAN: symptoms related to lesion mass effect causing increased intracranial pressure, hydrocephalus, and/or brain herniation (type I); and localized dysfunction of signaling networks connected to the inflammatory focus (type II). Although distinction of these entities is not always obvious, particular vigilance is warranted in patients who are at risk of type I TIAN given the location of their tumor(s). This is illustrated by the fact that all high-grade TIAN was characterized as type I and the only death from TIAN occurred in a patient with a large infratentorial lesion who died from TIAN-associated mass effect. CAR T-cell treatment of all CNS tumors, including lymphoma, should trigger TIAN surveillance to ensure early recognition and appropriate management. TIAN was managed similarly to ICANS with steroids, interleukin-1 blockade, osmotic agents, and cerebrospinal fluid diversion in severe cases. Focal radiotherapy was also used, which would represent a unique strategy for TIAN.

The original TIAN criteria were designed to apply to all CNS tumors, but this study must be applied with caution to disease settings other than CNS lymphoma. First, the TIAN criteria were modified to exclude patients who only had headache and fever. This likely increased the specificity of the findings but hampers comparison with other studies that use the unmodified criteria. Second, the magnetic resonance imaging findings associated with TIAN (increased size of contrast enhancement with concurrent decrease of diffusion restriction) may not apply to other tumor types. Distinguishing TIAN from tumor progression remains a difficult task, with radiographic and examination findings often equivocal in the acute setting. The timing of symptoms can be helpful here. With the benefit of a retrospective analysis, Kaulen et al found that neurologic worsening due to progressive disease was noted a median of 22 days after CAR T-cell infusion, whereas TIAN was much earlier.

Clinical studies going forward should distinguish TIAN from ICANS not only because of differences in management but also because TIAN may be an indicator of effective tumor killing by CAR T cells. Although ICANS occurs independent of tumor location, location is fundamental to TIAN.

Conflict-of-interest disclosure: The authors declare no competing financial interests.