Small hepatocellular carcinoma: MRI findings for predicting tumor growth rates

Acad Radiol. 2014 Nov;21(11):1455-64. doi: 10.1016/j.acra.2014.06.011.

Abstract

Rationale and objectives: Current clinical practice favors imaging rather than biopsy to diagnose hepatocellular carcinoma (HCC). There is a need to better understand tumor biology and aggressiveness of HCC. Our goal is to investigate magnetic resonance imaging (MRI) features of HCC that are associated with faster growth rates (GRs).

Materials and methods: After approval from institutional review board, a retrospective evaluation was performed of pre-liver transplant patients. Fifty-two patients who developed a >2 cm HCC on serial imaging were included in the study group, with a total of 60 HCCs seen. Precursor foci were identified on serial MRIs before the specific diagnostic features of >2 cm HCC could be made, and GRs and MRI features, including signal on T1- and T2-weighted images (WI), the presence of intralesional steatosis on chemical shift imaging, and enhancement pattern were analyzed. GRs were correlated with imaging features.

Results: The average GR of precursor lesions to >2 cm HCC was determined to be 0.23 cm/mo (standard deviation [SD], 0.32), with a doubling time of 5.26 months (SD, 5.44). The presence of increased signal intensity (SI) on T2-WI was associated with significantly higher growth (P = .0002), whereas increased intensity on T1-WI at the initial study was associated with a significantly lower GR (P = .0162). Furthermore, lesions with hypervascular enhancement with washout pattern had significantly higher GR (P = .0164). There is no evidence of differences in GRs seen in lesions with steatosis.

Conclusions: Small precursor lesions with increased SI on T2-WI and a washout pattern of enhancement are associated with faster GRs, which may suggest more aggressive tumor biology. These features may be helpful in patient management and surveillance for HCC.

Keywords: Liver neoplasms; MR; diagnosis; hepatocellular carcinoma; liver cancer; prognosis prediction.

MeSH terms

  • Algorithms*
  • Carcinoma, Hepatocellular / pathology*
  • Female
  • Humans
  • Image Enhancement / methods
  • Image Interpretation, Computer-Assisted / methods*
  • Liver Neoplasms / pathology*
  • Magnetic Resonance Imaging / methods*
  • Male
  • Middle Aged
  • Neoplasm Staging
  • Prognosis
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Tumor Burden