Radio-Pathomic Cell Density Maps Outperform Advanced MRI for Distinguishing True Glioblastoma Progression from Pseudoprogression

For distinguishing true progression from pseudoprogression in glioblastoma, radio-pathomic cell density (RPM Cell) maps achieved an AUC of 0.827, significantly outperforming advanced diffusion and perfusion metrics (AUCs <0.687, p=0.006).

• The retrospective, single-site study analyzed 171 MR imaging sessions from 130 glioblastoma patients (139 true progression, 32 pseudoprogression).
• RPM Cell maps are generated from conventional imaging data (e.g., T1, FLAIR, ADC), requiring no additional contrast agents or scan time.
• The study's primary limitation is the lack of external validation across institutions or prospective testing, as it used only the UCSD-PTGBM public dataset.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.