SWI Signal Intensity Distinguishes Treatment-induced Tissue Damage from Recurrent Tumor in Post-treatment High-grade Glioma

Introduction: Conventional MRI methods are limited in their ability to assess the effectiveness of anti-glioma treatments. New radiographic methods are needed to distinguish effective tumor response from ineffective treatment effects that occur secondary to tissue damage in the tumor microenvironment. We used immunohistochemistry (IHC) to characterize MRI-localized biopsies from post-treatment recurrent high-grade glioma (HGG) patients and correlated these findings with SWI signal intensity to examine its ability to accurately distinguish recurrent tumor from treatment-induced tissue damage.

Methods: 102 post-treatment recurrent HGG MRI-localized biopsies (from 51 patients) were classified as either predominantly tumor, treatment effect (treatment-induced tissue damage), or infiltrated brain tissue using IHC-derived k-means clustering. Biopsies were stained via IHC for an endothelial cell marker (CD34) and measured using a percent area index to quantify its vascular profile. The tissue state classification of each MRI-localized biopsy was correlated with its corresponding radiographic data.

Results: Biopsies classified as predominantly treatment effect had significantly less vasculature than biopsies classified as predominantly infiltrated brain tissue (p = 0.0065) or recurrent tumor (p = 2.6 x 10-5). Multiple logistic regression analysis found no significant relationship between IHC-derived tissue state classification and T1-CE or FLAIR, but did reveal SWI signal intensity as a significant predictor of tissue state classification (p = 0.0049). Specifically, MRI-localized biopsies with treatment effect predominance had significantly higher SWI signal intensity than infiltrated brain tissue (p = 0.021) and recurrent tumor (p = 0.0019).

Conclusion : The SWI signal intensities of MRI-localized biopsies in post-treatment recurrent HGG patients were correlated with tissue state classifications. SWI and IHC data independently suggest that biopsies predominantly composed of treatment effect have disrupted vascular integrity, as evidenced by low CD34+ IHC staining and high SWI signal intensity relative to infiltrated brain and recurrent tumor samples. This imaging technique provides an attractive non-invasive method for clinically monitoring response to anti-glioma treatments.