CHICAGO – The neurotoxicity in adult patients with relapsed or refractory B-cell acute lymphocytic leukemia (B-ALL) treated with CD19 CAR T cells is a separate process from cytokine release syndrome (CRS) and needs to be treated separately, according to a new study presented at the American Society of Clinical Oncology annual meeting (abstract 3019).
CD19-specific CAR-modified T cells produce high, durable anti-tumor activity, but can be associated with treatment-related toxicities, including CRS and neurotoxicity.
Neurotoxicity is poorly understood and it hasn't been clear where to focus further research, said Bianca Santomasso, MD, of Memorial Sloan Kettering Cancer Center in New York, New York.
Dr Santomasso and colleagues had conducted a phase 1 study using autologous 19-CAR T cells in adult patients with relapsing/refractory B-ALL, with high response rates.
To gain a better understanding of CD19 CAR T-cell-associated neurotoxicity, they analyzed clinical and research parameters after stratifying patients by neurotoxicity grade.
At ASCO, she reported neurologic symptom presentation in 51 adult patients with relapsed/refractory B-ALL who were treated with CAR T cells following conditioning chemotherapy, along with cerebrospinal fluid (CSF) data and neuroimaging findings associated with neurotoxicity.
Of the 51 patients treated, 10 patients (20%) developed mild neurologic symptoms (grade 1 or 2) and 21 patients (41%) developed severe neurotoxicity (grade 3 or 4).
No grade 5 neurotoxicity or diffuse cerebral edema occurred and, in all but one case, neurologic symptoms fully resolved.
Fourteen patients (27.4%) developed severe CRS; 6 patients received tocilizumab alone, 13 patients tocilizumab plus steroids, 4 patients steroids alone, and 29 patients supportive care.
The cytokines IL-6, IL-8, IL-10, interferon gamma, and granulocyte-colony stimulating factor were elevated in CSF over serum at the time of neurotoxicity and correlated with CSF protein levels.
“We found no significant correlation between neurotoxicity grade and the CAR T-cell concentration in the CSF during neurotoxicity,” Dr Santomasso said. “Instead, CSF protein level was correlated with neurotoxicity grade.”
Neurotoxicity was associated with peak CAR T-cell expansion in the blood and peak serum levels of several cytokines associated with T-cell activation or proliferation, she said.
“CAR T cells traffic to the CSF of patients with all grades of neurotoxicity, including grade 0, and there is no significant correlation between CAR T cells in CSF at the time of acute neurotoxicity and grade of neurotoxicity,” Dr Santomasso reported. “This suggests that neurotoxicity is not directly mediated by CAR T cells, which cross into the spinal fluid.”
Some chemokines/cytokines are elevated in CSF relative to serum, suggesting CNS-specific production of these factors, she said.
“Clinicians treating these patients tend to lump CRS and neurotoxicity together. Now we have an increasing understanding that these adverse events are separated in time and possibly in underlying pathology,” said Dr Santomasso.
“Even when patients recover from CRS, they could still be at risk for neurotoxicity.”
She noted that the neurotoxicity symptoms that B-ALL patients develop in the setting of CAR T cells are manageable.
In a subset of patients with severe neurotoxicity, T2/FLAIR changes were observed, which resolved with steroids and neurologic symptom resolution.