Dear Editor,
We sincerely thank Dr. Sun and Zhang for their interest in our study entitled “Non-contrast MRI for Detection of Late Recurrent Hepatocellular Carcinoma After Curative Treatment: Comparison to CT”, and for their constructive comments [
1,
2]. In our study, we demonstrated that non-contrast MRI (NC-MRI) offers significantly higher sensitivity and accuracy than contrast-enhanced CT (CECT) in detecting recurrent hepatocellular carcinoma (HCC) in patients who had remained recurrence-free for over two years following curative treatment. While our findings highlight the potential clinical value of NC-MRI in this setting, we agree with the authors that several important considerations are necessary for real-world implementation.
NC-MRI has several advantages over CECT, including the absence of ionizing radiation and the avoidance of contrast agent administration. Owing to these strengths and its favorable diagnostic performance, NC-MRI has been explored not only for detecting HCC recurrence, as in our study, but also for primary HCC surveillance in at-risk populations, with promising results reported in recent studies [
3,
4]. However, NC-MRI also has several drawbacks. The cost of NC-MRI tends to be higher than that of CECT, although the exact expenses may vary by country. Additionally, NC-MRI generally requires longer examination times compared to CECT. Nevertheless, the NC-MRI protocol used in our study was substantially shorter than that of standard contrast-enhanced liver MRI protocols, with an average scan time of 9.5 minutes. To support broader clinical adoption of NC-MRI for detecting recurrent HCC, further cost-effectiveness analyses are warranted. For this purpose, reliable data will be essential to determine whether the diagnostic advantages of NC-MRI translate into meaningful economic benefits over CECT in routine clinical practice.
Beyond cost and scan time, image quality and interpretability are also critical factors for the clinical adoption of NC-MRI. As NC-MRI protocols typically include key sequences essential for HCC detection, such as T2-weighted imaging and diffusion-weighted imaging (DWI), acquisition of high-quality images is paramount [
5]. In particular, DWI plays a central role in the diagnostic performance of NCMRI, but it is also highly susceptible to artifacts, underscoring the importance of optimal image acquisition. Additionally, given the inherent subjectivity in lesion detection and interpretation on both NC-MRI and CECT, inter-reader agreement represents another important consideration. In our study, image analyses were conducted independently at each participating center by board-certified radiologists with over a decade of experience in liver imaging, applying standardized criteria to minimize variability in interpretation across sites. However, our study did not include centralized image review; thus, we were unable to assess inter-reader agreement for NC-MRI and CECT in detecting recurrent HCC. To better evaluate the consistency and reliability of image interpretation in clinical practice, further studies incorporating centralized image assessment by multiple independent readers are needed.
In addition to imaging modalities, serum tumor markers such as alpha-fetoprotein (AFP) and Protein Induced by Vitamin K Absence or Antagonist-II (PIVKA-II) can aid in the detection of recurrent HCC following curative treatment [
6]. In our study, among the 22 patients with recurrent HCC, three exhibited elevated PIVKA-II levels, and one had elevated levels of both AFP and PIVKA-II. The remaining 18 patients demonstrated normal levels of these markers. When incorporating elevated tumor marker levels alongside imaging findings, the recalculated sensitivity for detecting recurrent HCC was 45.5% (10/22) for CECT and 86.4% (19/22) for NC-MRI, with NC-MRI maintaining significantly higher sensitivity than CECT (
P=0.004). These results suggest that NC-MRI provides superior sensitivity for recurrence detection, irrespective of tumor marker status.
In the context of surveillance, the occurrence of interval cancers, defined as recurrent lesions missed on prior imaging but detected between scheduled follow-up examinations, serves as an important metric for evaluating the efficacy of imaging-based surveillance strategies [
7]. In our study, all participants adhered to scheduled follow-up imaging protocols with both CECT and NC-MRI, and none underwent unscheduled interim imaging. Moreover, as this study employed a single-arm, intra-individual, head-to-head comparison design, with all participants undergoing both imaging modalities, we were unable to assess the incidence of interval cancers attributable to each modality independently.
Another important factor to consider is the individual risk of recurrence, as it may influence the cost-effectiveness of each imaging modality to detect recurrent HCC. In our study, we included participants who had remained free of recurrence for more than two years following curative treatment, with the aim of detecting late recurrence, presumably from de novo secondary primary HCC in the residual liver. During this period, the overall recurrence risk is relatively low compared to the first two years post-treatment, and the likelihood of aggressive patterns such as vascular invasion, disseminated micrometastases, or extrahepatic spread is also reduced. Consequently, our study did not evaluate the diagnostic performance of NC-MRI in detecting these more aggressive forms of recurrence. To address this limitation, future studies should include a broader spectrum of recurrence risk profiles and a larger patient population to comprehensively assess the diagnostic utility of NC-MRI in recurrent HCC.
In conclusion, our study provides preliminary evidence supporting the use of NC-MRI for detecting recurrent HCC, offering the advantage of avoiding contrast agent administration and radiation exposure. Nonetheless, further research involving participants with varying levels of recurrence risk, along with centralized image review by multiple readers, is necessary to thoroughly evaluate the diagnostic performance of NC-MRI. Additionally, cost-effectiveness analyses and individualized risk stratification will be essential to inform the routine clinical adoption of NC-MRI as a surveillance tool following curative treatment for HCC.
FOOTNOTES
-
Conflicts of Interest
The author has no conflicts to disclose.
Abbreviations
contrast-enhanced computed tomography
diffusion-weighted imaging
non-contrast magnetic resonance imaging
Protein Induced by Vitamin K Absence or antagonist-II
REFERENCES
- 1. Sun L, Zhang X. Revisiting the role of non-contrast MRI in recurrent HCC: Bridging diagnostic performance and realworld applicability: Letter to the editor on “Non-contrast MRI for detection of late recurrent hepatocellular carcinoma after curative treatment: A prospective multicenter comparison to contrastenhanced CT”. Clin Mol Hepatol 2026;32:e172-e174.
- 2. Kim DW, Chang W, Kim SY, Lim YS, Choi J, Cho J, et al. Noncontrast magnetic resonance imaging for detection of late recurrent hepatocellular carcinoma after curative treatment: a prospective multicenter comparison to contrast-enhanced computed tomography. Clin Mol Hepatol 2025;31:1285-1297.
- 3. Kim DH, Yoon JH, Choi MH, Lee CH, Kang TW, Kim HA, et al. Comparison of non-contrast abbreviated MRI and ultrasound as surveillance modalities for HCC. J Hepatol 2024;81:461-470.
- 4. Rhee H, Kim MJ, Kim DY, An C, Kang W, Han K, et al. Noncontrast magnetic resonance imaging vs ultrasonography for hepatocellular carcinoma surveillance: A Randomized, singlecenter trial. Gastroenterology 2025;168:1170-1177.e12.
- 5. Lee DH. Recent advances and issues in imaging modalities for hepatocellular carcinoma surveillance. J Liver Cancer 2025;25:31-40.
- 6. Lee J, Joo I, Lee DH, Jeon SK, Lee JM. Clinical outcomes of patients with a high alpha-fetoprotein level but without evident recurrence on CT or MRI in surveillance after curative-intent treatment for hepatocellular carcinoma. Abdom Radiol (NY) 2021;46:597-606.
- 7. Tzartzeva K, Obi J, Rich NE, Parikh ND, Marrero JA, Yopp A, et al. Surveillance imaging and alpha fetoprotein for early detection of hepatocellular carcinoma in patients with cirrhosis: A meta-analysis. Gastroenterology 2018;154:1706-1718.e1.
Citations
Citations to this article as recorded by
