Regional hypometabolism on brain FDG-PET one month after receiving chimeric antigen receptor T-cell therapy may be associated with icans

Abstract Introduction: Chimeric antigen receptor (CAR) T-cell therapy is an established cellular therapy approved for patients with large B cell lymphoma (LBCL), multiple myeloma, indolent non-Hodgkin lymphoma (NHL) and B-acute lymphoblastic leukemia (B-ALL). Immune effector cell-associated neurotoxicity syndrome (ICANS) is a potentially life-threatening toxicity which manifests with neurocognitive symptoms, including seizures and rarely cerebral oedema. While MRI brain is recommended in patients developing severe ICANS, it rarely demonstrates ICANS-defining changes. We hypothesised that brain FDG-PET imaging may reveal patterns of metabolism associated with ICANS. Method: We retrospectively analysed CAR T-cell therapy recipients at the Alfred Hospital between 2019 and June 2025. Patients with a baseline and one-month post infusion brain FDG-PET, as well as a baseline MRI brain were included. Whole-body PET images were cropped to include the region between the skull and mandible using visual inspection and FSLROI in FSL. MRI T1 brain images were extracted with the MNI152 1mm voxel template. All FDG-PET images were co-registered with baseline MRI brain images and subsequently with the MNI152 template. Cerebellum was used as reference region for Standardised Uptake Value Ratio (SUVR) generation. Analysis of mean FDG intensity in the region of interest (frontal, parietal, temporal, occipital lobes, cerebellum, whole brain and basal ganglia) were calculated. Data analysis with descriptive statistics and the Wilcoxon signed ranked test were undertaken in R. Result: A total of 48 patients had complete radiological data. Median age was 72 years (range 32-83), 39.6% were female (n=19) and the majority (72.9%) were diagnosed with large B cell lymphoma (LBCL). The remaining were diagnosed with multiple myeloma (20.8%) and other non-Hodgkin lymphomas (6.2%). There was a median of two lines of treatment prior to CAR-T (range 1-8). Median percentage of SUVR changes at one month compared to baseline FDG PET were an increase of 2.29% for the frontal lobe (IQR -2.81 to 9.87), 3.05% for parietal lobe (IQR -4.67 to 10.19), 0.74% for temporal lobe (IQR -5.37 to 3.75), 0.42 for occipital lobe (IQR -8.23 to 10.33), 2.87% for whole brain (IQR -2.47 to 9.01) and 6.71% for basal ganglia (IQR 1.071 to 14.74; p-value <0.01). Unadjusted Wilcoxon ranked-sum comparisons by ICANS status revealed a statistically significant reduction of median SUVR in the temporal lobe at one month relative to baseline in patients who experienced ICANS (median -4.45%; p-value = 0.01); and a trend for frontal hypometabolism (median -1.31%; p value = 0.07). Conclusion: Day-28 FDG-PET demonstrated temporal hypometabolism in CAR T-cell therapy recipients who developed ICANS. Future studies should assess FDG-PET brain imaging at the time of ICANS and explore whether metabolic changes can serve as diagnostic or prognostic biomarkers of ICANS.