Fat fraction quantification with MRI estimates tumor proliferation of hepatocellular carcinoma

Mengqi Huang, Fan Zhang, Zhen Li, Yan Luo, Jiali Li, Zixiong Wang, Liya Ma, Gen Chen, Xuemei HuFrontiers in Oncology 2024 · DOI

Abstract

We infer that intratumor fat can be identified in HCC and fat fraction quantification using quantitative multi-echo Dixon can distinguish low proliferative HCCs.

Introduction

Hepatocellular carcinoma (HCC) remains a leading cause of cancer-related mortality worldwide. Clinical management decisions—ranging from surgical resection and liver transplantation to ablation—depend heavily on the tumor's aggressiveness. Key pathological markers for aggressiveness include the Ki67 index, which indicates cell proliferation rates, and the presence of microvascular invasion (MVI). Currently, these markers are typically assessed postoperatively through histopathological analysis. There is a critical need for non-invasive, preoperative "virtual biopsy" techniques to estimate these factors and guide treatment planning. This study investigates the potential of quantitative MRI, specifically fat fraction quantification, to serve as a surrogate marker for tumor proliferation.

Methods and Materials

The study involved a retrospective analysis of 66 patients with histopathologically confirmed HCC who were treated at Tongji Hospital between August 2021 and March 2023.

Patient selection flowchart showing the exclusion criteria and final cohort of 66 patients.

The core imaging protocol utilized a quantitative multi-echo Dixon MRI sequence. This technique allows for the precise calculation of the Proton Density Fat Fraction (PDFF) and R2* mapping. Radiologists performed Region of Interest (ROI) analysis, manually delineating areas on the MRI slices where intratumoral fat was most prominent to calculate the tumor fat fraction (tFF).

Visualization of the liver tumor and corresponding MRI mappings: T1WI, Fat fraction mapping, and R2* mapping used for ROI analysis.

These MRI-derived measurements were correlated with pathological findings, specifically the Ki67 proliferation index and MVI status.

Results

The study revealed several key insights into the relationship between tumor fat content and biological behavior:

  • Tumor vs. Liver Fat Content: The median fat fraction within the tumor (tFF) was significantly higher than in the surrounding liver tissue (5.24% vs. 3.51%).
  • Correlation with Proliferation: A significant negative correlation was found between tFF and the Ki67 index (r = -0.306, P = 0.012). This indicates that as tumor cells divide more rapidly (higher Ki67), the detectable fat fraction decreases.
  • Identifying Low-Proliferation Nodules: The fat fraction was particularly effective at distinguishing "low-proliferation" HCCs (Ki67 ≤ 30%). High-proliferation nodules exhibited a median tFF of 2.10%, whereas low-proliferation nodules had a median tFF of 4.90%.
  • Diagnostic Performance: Using a tFF threshold of 3.39%, the researchers achieved an Area Under the Curve (AUC) of 0.747 for identifying low-proliferation HCC, with a sensitivity of 77.8% and a specificity of 69.2%.
  • MVI and R2*: Interestingly, tFF did not significantly differ between tumors with and without microvascular invasion, suggesting that metabolic pathways for proliferation and invasion are distinct. Additionally, R2* values were generally lower in the tumor than in the background liver.

Discussion

The findings suggest that intratumoral fat serves as a non-invasive indicator of tumor growth rates. The observed loss of fat in highly aggressive tumors may be linked to metabolic reprogramming, where rapidly dividing cells consume lipid stores for energy. While this study provides a promising "virtual biopsy" tool for identifying low-proliferation HCCs, it is limited by its single-center, retrospective design and relatively small sample size. Future research should focus on multi-center validation and exploring volumetric or automated radiomic analyses to better capture tumor heterogeneity.

Summary updated

Abstract

We infer that intratumor fat can be identified in HCC and fat fraction quantification using quantitative multi-echo Dixon can distinguish low proliferative HCCs.
BibTeX
@inproceedings{huang2024fat,
  author = {Mengqi Huang and Fan Zhang and Zhen Li and Yan Luo and Jiali Li and Zixiong Wang and Liya Ma and Gen Chen and Xuemei Hu and Xuemei Hu},
  title = {{Fat fraction quantification with MRI estimates tumor proliferation of hepatocellular carcinoma}},
  booktitle = {Frontiers in Oncology},
  year = {2024},
  doi = {10.3389/fonc.2024.1367907},
}
Figure