Qualitative MR features to identify non-enhancing tumors within glioblastoma’s T2-FLAIR hyperintense lesions

Shota Yamamoto, Yoshiko Okita, Hideyuki Arita, Takahiro Sanada, Mio Sakai, Atsuko Arisawa, Naoki Kagawa, Eku Shimosegawa, Katsuyuki Nakanishi, Manabu Kinoshita, Haruhiko Kishima

Abstract

Purpose: To identify qualitative MRI features of non-(contrast)-enhancing tumor (nCET) in glioblastoma’s T2-FLAIR hyperintense lesion. Methods: Thirty-three histologically confirmed glioblastoma patients whose T1-, T2- and contrast-enhanced T1-weighted MRI and 11 C-methionine positron emission tomography (Met-PET) were available were included in this study. Met-PET was utilized as a surrogate for tumor burden. Imaging features for identifying nCET were searched by qualitative examination of 156 targets. A new scoring system to identify nCET was established and validated by two independent observers. Results: Three imaging features were found helpful for identifying nCET; “Bulky”, “Around the rim of contrast-enhancement (Around-rim),” and “High-intensity on T1WI and low-intensity on T2WI (HighT1LowT2)” resulting in an nCET score = 2 x bulky – 2 x Around-rim + HighT1LowT2. The nCET score’s classification performances of two independent observers measured by AUC were 0.78 and 0.80, with sensitivities and specificities using a threshold of four being 0.443 and 0.771, and 0.916 and 0.768, respectively. The weighted kappa coefficient for the nCET score was 0.946. Conclusion: The current investigation demonstrated that qualitative assessments of glioblastoma’s MRI might help identify nCET in T2/FLAIR high-intensity lesions. The novel nCET score is expected to aid in expanding treatment targets within the T2/FLAIR high-intensity lesions.

Introduction

Glioblastoma is a notoriously aggressive brain cancer. Traditionally, surgeons focus on removing the "contrast-enhancing" core—the part of the tumor that lights up on a standard MRI. However, tumor cells often migrate into the surrounding area, which appears as a bright "cloud" on T2-FLAIR MRI sequences. While removing this extra area (known as "supratotal resection") can significantly improve patient survival, it presents a clinical dilemma: how to distinguish between a dense cluster of migrating tumor cells (non-enhancing tumor, or nCET) and simple fluid/swelling (edema).

Advanced imaging like Met-PET can identify these cells, but it is expensive and not widely available. This study seeks a way to identify these "hidden" tumors using standard MRI features that a surgeon can assess qualitatively.

Methods

The researchers analyzed 33 patients with confirmed glioblastoma who had undergone both standard MRI and $^{11}$C-methionine PET (Met-PET). They used Met-PET as the "ground truth" for tumor density, as high methionine uptake is a reliable surrogate for high tumor cell concentration. They identified 156 specific targets within the T2-FLAIR regions and looked for qualitative MRI features that correlated with high Met-PET signals.

Through multiple linear regression, they distilled these into a three-feature scoring system:

  1. Bulky: A convex shape pushing out surrounding tissue or involvement of gray matter (viewed on T2WI).
  2. Around-rim: Targets located immediately outside the contrast-enhancing rim (viewed on Gd-T1WI). This was found to be a negative predictor of dense tumor.
  3. HighT1LowT2: A specific intensity profile where the target's T1 signal is higher and T2 signal is lower than the surrounding reference points.
Qualitative MRI features and the nCET scoring system application.

The nCET Score Formula: $$Score = (2 \times Bulky) - (2 \times Around\text{-}rim) + (1 \times HighT1LowT2)$$

Key Results

  • Predictive Power: The scoring system achieved an Area Under the Curve (AUC) of 0.78 to 0.80 when tested by independent observers.
  • Optimal Threshold: A score of $\ge 4$ was identified as the best threshold for identifying non-enhancing tumor. More than 50% of regions with a score of $\ge 4$ were confirmed as high-density tumor by Met-PET.
  • Reliability: The study found a weighted kappa coefficient of 0.946 for the nCET score, indicating near-perfect agreement between different observers.
  • Correlation: There was a significant positive correlation ($r = 0.52$ to $0.59$) between the nCET score and the actual tumor-to-normal (T/N) ratio measured by Met-PET.

Discussion and Conclusion

A notable finding was the "Around-rim" surprise: being right next to the enhancing core was actually a negative predictor for dense non-enhancing tumor. This suggests that the immediate perimeter of the main tumor mass might be more characterized by reactive swelling or diffuse infiltration rather than the dense "bulky" tumor clusters found further out.

While many researchers are pushing for AI and complex quantitative mapping, this study proves that "expert eyes" using a structured rubric can be highly effective. This makes the findings immediately applicable in clinical settings that lack advanced computational infrastructure. The novel nCET score is expected to aid in expanding treatment targets within T2/FLAIR high-intensity lesions, potentially improving surgical outcomes for glioblastoma patients.

Summary updated

BibTeX
@inproceedings{yamamoto2023qualitative,
  author = {Shota Yamamoto and Yoshiko Okita and Hideyuki Arita and Takahiro Sanada and Mio Sakai and Atsuko Arisawa and Naoki Kagawa and Eku Shimosegawa and Katsuyuki Nakanishi and Manabu Kinoshita and Haruhiko Kishima},
  title = {{Qualitative MR features to identify non-enhancing tumors within glioblastoma’s T2-FLAIR hyperintense lesions}},
  booktitle = {Research},
  year = {2023},
  doi = {10.21203/rs.3.rs-3289259/v1},
}
Figure