Differentiating primary central nervous system lymphoma from glioblastoma by time-dependent diffusion using...

Area of Science:

• Radiology
• Neuro-oncology
• Medical Imaging

Background:

• Differentiating primary central nervous system lymphomas (PCNSLs) from glioblastomas (GBMs) is crucial for treatment planning.
• Apparent diffusion coefficient (ADC) measurements are used for this differentiation, but their effectiveness can be influenced by diffusion time settings.

Purpose of the Study:

• To evaluate the impact of effective diffusion time on ADC-based differentiation between PCNSLs and GBMs.
• To investigate the utility of time-dependent diffusion MRI parameters for distinguishing these brain tumors.

Main Methods:

• Retrospective analysis of diffusion-weighted imaging (DWI) data from 21 PCNSL and 66 GBM patients.
• Acquisition of DWI sequences at two effective diffusion times (7.1 ms and 44.5 ms).
• Calculation of ADC, ADC changes (cADC), and relative ADC changes (rcADC) in enhancing and peritumoral regions; comparison of diagnostic performance using AUC.

Main Results:

• ADC values (ADC7.1ms and ADC44.5ms) were significantly lower in PCNSLs compared to GBMs in enhancing regions.
• cADC and rcADC values were significantly higher in PCNSLs than GBMs in enhancing regions (p < 0.01 for all).
• Mean rcADC demonstrated the highest diagnostic performance (AUC = 0.920) for differentiating PCNSLs from GBMs in enhancing regions.

Conclusions:

• Effective diffusion time significantly impacts ADC-based differentiation between PCNSLs and GBMs.
• Time-dependent diffusion MRI parameters, especially rcADC, offer valuable insights for distinguishing these intracranial lesions.