Synopsis

Keywords: Liver, Cancer

Survival rates for hepatocellular carcinoma (HCC) are poor, and improved methods for early detection are required. One such method may be the detection of iron sparing in HCC compared to the liver tissue using R₂* mapping. We optimised sequences for the assessment of whole liver 3D R₂* mapping, and developed a HCC phantom comprising small spheres with a range of R₂* values to assess them. Agreement of the phantom R2* with reference values and between vendors, as well as the ability to detect small differences in R₂* indicated the sequences have strong potential to detect iron sparing.

Introduction

Survival rates from hepatocellular carcinoma (HCC), the most common primary liver cancer, are poor¹ despite curative treatments being available². Surveillance in at-risk individuals using ultrasound is recommended^2–4, but ultrasound has only 23%⁵-47%⁶ sensitivity for early HCC detection. There is a need for a better surveillance tool. Several studies have shown iron is lower in HCC compared to surrounding liver tissue^7–9. Since R₂* can be used to estimate iron, this may allow early detection of such iron sparing in HCCs. However, multi-echo gradient echo (ME-GRE) sequences for R₂* mapping have mainly been optimised to detect iron overload. We optimised 3D ME-GRE R₂* mapping sequences for detection of iron sparing across two vendors, and built a ‘HCC phantom’ to represent R₂* of iron in tumours and liver parenchyma to assess the sequences met requirements for detecting iron sparing.

Methods

Requirements for detecting HCC iron sparing were defined to detect iron of 0-3 mg/g dry weight (dw)¹⁰, with localised differences in iron of 0.2 mg/g dw and higher¹⁰ in regions of 1 cm or more². 3D ME-GRE sequences Siemens and Philips scanners were optimised at 3 T. Parameters are provided in Table 1.
HCC Phantom The HCC phantom was developed, using solutions of MnCl₂ and NiCl₂ across 17 concentrations to modulate R₂* and R₁¹¹, with the Alam¹² and Hankins¹³ equations used for conversion between R₂* and iron concentration. 10, 15 and 20 mm diameter polypropylene spheres were filled with solutions with R₂* to represent HCC iron concentrations. These spheres were attached to three plates within a hollow acrylic sphere (160 mm diameter) flood filled with a concentration to represent liver iron, see Figure 1A-C. The HCC phantom was scanned on both Philips and Siemens scanners using optimised 3D ME-GRE sequences.
Reference R₂* values were obtained by scanning tubes (25 mm diameter, 115 mm height) filled with each concentration, placed in a water-filled tub (210 mm diameter, 160 mm height), using a 2D ME-GRE sequence on the Philips scanner (see Table 1), robust against artifacts due to its thin slice thickness and repetitions to increase signal-to-noise ratio. R₁ was measured using a 5(3)3 MOLLI scheme.
Analysis T₂* and R₂* maps were generated using the MAGO method¹⁴. A 10/7.5/5 mm diameter circular ROI was placed on each 20/15/10 mm sphere and a 10 mm ROI on the background. A 15 mm diameter circular ROI was placed on each tube. Median R₂* was calculated for each ROI. Bland-Altman methods were used to compare R₂* and equivalent iron of the HCC phantom spheres with reference values from the tubes, and between vendors. To assess the ability to detect differences in iron of 0.2 mg/g dw, the R₂* of the 20 mm spheres were compared with adjacent background R₂*.

Results

The tubes provided reference R₂* values with equivalent iron concentration 0.4 mg/g dw – 2.0 mg/g dw (Figure 1B), whilst the HCC phantom background had equivalent iron concentration of 0.9 mg/g dw. R₁ of the tubes ranged from 11 to 20 s^-1, whilst the background was 14 s^-1. R₂* maps of the HCC phantom from each scanner are given in Figure 1D-E. Some spheres were excluded due to filling errors or could not be quantified, particularly on the Siemens data which appeared smoother. Bland-Altman plots show good agreement of R₂* between vendors for the 20 mm spheres, whilst for the smaller spheres there were discrepancies between vendors at higher R₂* values (Figure 2).
Bland-Altman plots comparing R₂* of the spheres with the reference tubes are given in Figure 3A. For the 20/15/10 mm spheres, the R₂* bias from the reference was 2/-9/-12 s^-1 on the Philips scanner and 1/-16/-10 s^-1 on the Siemens scanner. For all sphere sizes, the limits of agreement were larger for Siemens than Philips. Bland-Altman plots comparing equivalent iron of the spheres with the reference tubes are given in Figure 3B. The 20 mm spheres were similar to equivalent iron of ±0.2 mg/g or less from the reference, whilst the 15 and 10 mm spheres had some differences greater than ±0.2 mg/g dw from the reference.
ROIs placed on the 20 mm spheres to compare concentrations 3, 4 and 5 to adjacent background showed no overlap in the interquartile ranges (Figure 4) and thus potential to detect localised differences in-vivo.

Discussion

The HCC phantom showed that the 3D R₂* mapping sequences could be used to measure iron within the required range, with differences in iron of 0.2 mg/g dw detectable. The size of the sphere impacted the results: smaller 10 mm spheres had R₂* further from the reference values or were not visible on the R₂* maps, particularly on Siemens scanners, where the data appear smoothed compared to Philips despite an attempt to match sequence parameters, indicating that the sequences may not be suitable for detecting tumours of 10 mm or less.

Conclusion

The HCC phantom allowed assessment of the sequences to study HCC for detecting iron sparing. These optimised R₂* sequences are now being used in trials in patients with HCCs to assess the utility of the sequence as a surveillance tool.

Acknowledgements

EP was supported by an Industrial Fellowship from the Royal Commission for the Exhibition of 1851. This project was performed as part of the DeLIVER Consortium, funded by CRUK.

References

1. Yang JD, Hainaut P, Gores GJ, Amadou A, Plymoth A, Roberts LR. A global view of hepatocellular carcinoma: trends, risk, prevention and management. Nat Rev Gastroenterol Hepatol. 2019;16(10):589-604.
2. Galle PR, Forner A, Llovet JM, et al. EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J Hepatol. 2018;69(1):182-236.
3. Marrero JA, Kulik LM, Sirlin CB, et al. Diagnosis, Staging, and Management of Hepatocellular Carcinoma: 2018 Practice Guidance by the American Association for the Study of Liver Diseases. Hepatology. 2018;68(2):723-750.
4. Kokudo N, Hasegawa K, Akahane M, et al. Evidence-based Clinical Practice Guidelines for Hepatocellular Carcinoma: The Japan Society of Hepatology 2013 update (3rd JSH-HCC Guidelines). Hepatol Res. 2015;45(2):123-127.
5. Park HJ,Seo N, Kim SY. Current Landscape and Future Perspectives of Abbreviated MRI for Hepatocellular Carcinoma Surveillance. Korean J Radiol. 2022 Jun;23(6):598-614.
6. Tzartzeva K, Obi J, Rich NE, et al. Surveillance Imaging and Alpha Fetoprotein for Early Detection of Hepatocellular Carcinoma in Patients With Cirrhosis: A Meta-analysis. Gastroenterology. 2018;154(6):1706.
7. Ebara M, Fukuda H, Hatano R, et al. Metal contents in the liver of patients with chronic liver disease caused by hepatitis C virus: Reference to hepatocellular carcinoma. Oncology. 2003;65(4):323-330.
8. Tashiro H, Kawamoto T, Okubo T, Koide O. Variation in the Distribution of Trace Elements in Hepatoma. Biol Trace Elem Res. 2003;49.
9. Honda H, Kaneko K, Kanazawa Y, et al. MR imaging of hepatocellular carcinomas: Effect of Cu and Fe contents on signal intensity. Abdom Imaging. 1997;22(1):60-66.
10. Ebara M, Fukuda H, Kojima Y, et al. Small hepatocellular carcinoma: Relationship of signal intensity to histopathologic findings and metal content of the tumor and surrounding hepatic parenchyma. Radiology. 1999;210(1):81-88.
11. Stupic KF, Ainslie M, Boss MA, et al. A standard system phantom for magnetic resonance imaging. Magn Reson Med. 2021;86(3):1194-1211.
12. Alam MH, Auger D, McGill LA, et al. Comparison of 3 T and 1.5 T for T2∗magnetic resonance of tissue iron. J Cardiovasc Magn Reson. 2016;18(1).
13. Hankins JS, McCarville MB, Loeffler RB, et al. R2* magnetic resonance imaging of the liver in patients with iron overload. Blood. 2009;113(20):4853-4855.
14. Triay Bagur A, Hutton C, Irving B, Gyngell ML, Robson MD, Brady M. Magnitude-intrinsic water–fat ambiguity can be resolved with multipeak fat modeling and a multipoint search method. Magn Reson Med. 2019;82(1):460-475.