Page 99 - Read Online

Page 2 of 16 Brunsing et al. Hepatoma Res 2020;6:59 I http://dx.doi.org/10.20517/2394-5079.2020.50

INTRODUCTION
Imaging-based surveillance for hepatocellular carcinoma (HCC) aims to detect early-stage, potentially
curable tumors in asymptomatic high-risk patients to prolong life. First introduced about four decades
ago, it is now an established part of routine clinical care for patients with chronic hepatitis B or cirrhosis
in many countries across the globe. A randomized controlled trial of over 18,000 people with active or
chronic hepatitis B showed that semi-annual screening with a combination of ultrasound (US) and serum
[1]
[2,3]
alpha fetoprotein reduced HCC-related mortality by 37% . Based on the above findings, other studies ,
cost and availability considerations, US is recommended by most national and international hepatology
societies for HCC surveillance [4-10] . Since surveillance US does not permit a definitive diagnosis of HCC,
positive surveillance US exams prompt additional diagnostic tests, usually a contrast-enhanced multiphase
computed tomography (CT) or magnetic resonance imaging (MRI). Patients with negative US exams
return for routine surveillance US examinations, usually at six-month intervals.

Despite universal recommendation for use of US in HCC surveillance, the efficacy of this modality is
disappointing. US has low sensitivity for HCC [11,12] , in particular for patients with early-stage tumors [12-14] ,
ascites, cirrhosis or obesity [15-17] . Meta-analyses indicate that the sensitivity of surveillance US to detect
small (e.g., ≤ 2 cm) HCCs in patients with cirrhosis is less than 50%, i.e., more than half of patients
with potentially curable cancers are missed and may progress to advanced, incurable disease before
diagnosis [14,18,19] . Delayed diagnosis defeats the purpose of surveillance, which aims to detect patients with
[20]
[21]
very early- or early-stage HCC , allowing for curative therapies . The failure to detect early disease
[22]
contributes to HCC-related mortality .
A more sensitive surveillance test might improve outcomes in patients at risk for HCC. Compared to
US, both CT and MRI have superior reported diagnostic sensitivity to identify patients with HCC [16,19] ,
[15]
including those with early-stage tumors , however they also pose challenges as surveillance tools. CT
requires injection of iodinated intravenous contrast agents, which can cause allergic reactions and possibly
nephrotoxicity, potentially limiting the use of this modality in certain populations. In addition, CT exposes
patients to ionizing radiation, an important consideration in younger or middle-aged adults with well-
compensated cirrhosis. Conventional MRI provides higher sensitivity than CT [16,19] , but also requires
administration of intravenous contrast material; moreover, long exam duration, interpretation complexity,
and high cost hinder its suitability for surveillance.

Motivated to provide higher sensitivity than US while avoiding the limitations of CT and conventional
MRI, investigators have developed abbreviated MRI (AMRI) protocols that rely on a small number of
select sequences specifically tailored for HCC detection [12,23-36] . The rationale is that reduced scanner time
decreases costs and complexity, while improving patient comfort, without significantly compromising
HCC detection. AMRI also simplifies workflow and possibly interpretation, while utilizing fewer
resources. Recent studies suggest that AMRI might be a high-sensitivity and feasible alternative to US for
HCC surveillance, and a recent Markov model-based cost-utility analysis suggested AMRI-based HCC-
[37]
surveillance may be the most cost-effective strategy .
The purpose of this article is to review emerging concepts on AMRI-based HCC surveillance, including
technical aspects, diagnostic performance, current gaps in knowledge, and future directions.

AMRI: APPROACHES
Three general AMRI approaches have been developed: non-contrast AMRI, dynamic contrast-enhanced
AMRI, and hepatobiliary phase contrast-enhanced (HBP) AMRI. All can be completed in approximately
10 min or less of scanner time, considerably less than a complete or conventional MRI exam of the liver,