Role of abbreviated MRI protocol for screening of HCC in HCV related cirrhotic patients prior to direct-acting antiviral treatment

Abstract

Background: Chronic HCV infection is a global health problem causing progressive hepatic fibrosis, cirrhosis, and ultimately hepatocellular carcinoma (HCC). Recent advantage of direct-acting antiviral drugs (DAA) with a high sustained virologic response (SVR) reduces overall HCV-related morbidity and mortality, yet recent studies report a high recurrence rate of HCC after DAA; this calls availability of a reliable screening method to properly exclude HCC before DAA treatment. The primary objective of our cohort study was to assess the feasibility of an abbreviated MRI protocol as a screening tool for the detection of hepatic focal lesions/early HCC in patients with HCV-related liver cirrhosis. The study included 41 patients with HCV-related cirrhosis candidates to DAA therapy. All patients underwent routine screening for HCC by combined abdominal ultrasound and serum alfa-fetoprotein. An abbreviated MRI protocol (Abr-MR) including combined T2-weighted image and diffusion-weighted imaging (DWI) followed by dynamic contrast-enhanced MRI (CE-MRI) was performed for all subjects, assessing for presence and characterization of focal lesions. Results: For all included 41patients, no elevation of the alpha-fetoprotein was shown. Ultrasound detected a single focal lesion in one patient. Abbreviated MR demonstrated 15 focal lesions; for detected lesions, 1 lesion shows "shine though" and the rest showing moderate and high degrees of restriction. CE-MRI characterized lesions as 1 lesion = LIRADS-1, 3 lesions = LR-M, and 11 lesions = LR-5.The standard screening using combined ultrasound and alpha-fetoprotein had sensitivity, specificity, PPV, and NPV of 6.6 % (95% CI = 0.0034-0.29), 100% (95% CI = 0.89-1.000), 100% (95% CI = 0.051-1.000), and 68.8% (95% CI = 0.54-0.80). Abr-MR protocol showed sensitivity, specificity, PPV, and NPV of 100% for all parameters (95% CI = 0.79-1.000, 0.89-1.000, 0.79-1.000, and 0.89-1.000 respectively). Conclusion: In our study, we demonstrate the superiority of a proposed cost-effective Abr-MR protocol in the detection of hepatic focal lesions and small-sized HCC compared to routine screening using alpha-fetoprotein and ultrasound in HCV-related liver cirrhosis.

Meta-Analysis of the Accuracy of Abbreviated Magnetic Resonance Imaging for Hepatocellular Carcinoma Surveillance: Non-Contrast versus Hepatobiliary Phase-Abbreviated Magnetic Resonance Imaging

“…In search of the bibliographies, we found one additional eligible article. Finally, a total of 10 eligible articles reported the diagnostic performance of AMRI in HCC surveillance [11][12][13][26][27][28][29][30][31][32]. The study characteristics of 10 studies are shown in Table 1.…”

“…Two studies were prospective by design [11,31], and five studies exclusively included patients with cirrhosis [13,27,28,30,31]. The most common underlying liver disease was hepatitis C in four studies [12,26,31,32] and hepatitis B in four studies [11,13,29,30]. Three studies used only 1.5-T MRI scanners [13,27,31].…”

“…The most common underlying liver disease was hepatitis C in four studies [12,26,31,32] and hepatitis B in four studies [11,13,29,30]. Three studies used only 1.5-T MRI scanners [13,27,31]. Of the 10 included studies, three used HBP-AMRI protocols [12,26,29] and six used NC-AMRI protocols [11,13,27,28,30,31], with one study using both [32].…”

“…Of the 10 included studies, three used HBP-AMRI protocols [12,26,29] and six used NC-AMRI protocols [11,13,27,28,30,31], with one study using both [32]. Two studies only used pathology as the reference standard for the diagnosis of HCC [27,28], two only used imaging [30,31], and six used both pathological diagnosis and imaging as the reference standard [11][12][13]26,29,32]. Two studies performed multiple surveillance rounds [12,13], with surveillance intervals ranging between 5 and 8.8 months.…”

“…The results of the study qualities of the 10 included studies are shown in Figure S1. Of the 10 included studies, five had a high risk of bias in at least one of the four domains [12,26,[29][30][31]. In the patient-selection domain, three studies had an unclear risk of bias because they were unclear about whether patients were consecutively or randomly enrolled or not [27,28,31].…”