Introduction: This study aims to determine the potential value of MR-PWI and MR-DWI to differentiate immune therapy-induced inflammatory response from recurrent glioblastoma tumour growth. Both can present as contrast-enhancing lesions on conventional magnetic resonance imaging (MRI).
Methods: Patients with recurrent glioblastoma who could obtain a total or near-total resection were treated with dendritic cell immune therapy according to the HGG-IMMUNO-2003 trial. A retrospective analysis of 32 follow-up MRI examinations (mean follow-up time 21 months) in eight patients was performed for this pilot study. For the statistical analysis, the 32 examinations were divided into three groups: 0-obtained in patients that remained stable during the follow-up period, 1a-obtained in progressive-tumour patients at time points before definite progression and 1b-obtained in patients at or after progression.
Results: Maximum lesional rCBV ratios were highest in group 1b (Student t test, 9.25 ± 2.68; p < 0.001) and were higher in group 1a (4.87 ± 1.61, p < 0.001) compared to group 0 (1.22 ± 0.47). The minimum apparent diffusion coefficients (ADCs) in the contrast-enhancing regions were lower in group 1a (0.62 ± 0.06 × 10(-3) mm(2)/s) than in group 0 (1.03 ± 0.43 × 10(-3) mm(2)/s, p = 0.01) and higher in group 1b (0.76 ± 0.08) compared to 1a (p = 0.02). The minimum ADCs in the FLAIR-hyperintense region were lower in group 1a (0.62 ± 0.06, p = 0.02) compared to group 0 (0.76 ± 0.16) but not significantly different in group 1b (0.68 ± 0.07) from groups 0 and 1a (p = 0.33, p = 0.10). The mean ADCs of the FLAIR-hyperintense region and the mean ADCs of the contrast-enhancing lesion were not significantly different.
Conclusion: The maximum lesional rCBV ratios and minimum ADC values in the contrast-enhancing area are potential radiological markers to differentiate between immune therapy-induced inflammatory response and recurrent glioblastoma tumour growth in glioblastoma patients treated with immune therapy.