Surveillance of hepatocellular carcinoma: is only ultrasound enough?

See Article on 239

Early detection of hepatocellular carcinoma (HCC) is very important because the 5-year survival rate of early stage HCC (about 40% to 70%) is better than that of advanced stage HCC (less than 5%) [1]. In some countries, high-risk patients undergo a hepatocarcinoma surveillance program using ultrasound and serum alpha-fetoprotein (AFP) levels.

However, there are several drawbacks of surveillance program using ultrasonography. For high chance of receiving curative treatment, the stage of HCC should be low, such as very early (BCLC 0) or early (BCLC A) stage. However, detection of small lesions by ultrasound is not easy. According to a meta-analysis [2], the sensitivity of ultrasonography was 65% to 80%. Although accompanying with measurement of serum AFP, the sensitivity to detect early HCC was not increased [3]. Therefore, it seems not to be enough to perform ultrasonography as the only imaging method for surveillance of HCC. According to a retrospective study by Wong, et al. [4] mean size of detected HCC lesion was 3.2 cm and 61% of patients (158/257) had a tumor less than 3 cm of the largest dimension. Another study performed by Kim, et al. [5] shows that only one tumor was detected by ultrasonography only among 1,100 screenings for 407 patients, but 26 cases were detected by the surveillance by magnetic resonance imaging (MRI). They said that MRI with hepatospecific agent such as gadoxetic acid (Gd-EOB-DTPA) is more sensitive than ultrasonography only to detect early stage HCC in high-risk patients. Additionally, MRI also showed a significantly lower false-positive rate and higher positive predictive value than ultrasonography. Moreover, their result of ultrasonographic surveillance (28%) was very poorer than that of meta-analysis (63%) [3], and they explained that a kind of lead-time bias and inherent distortions of the liver parenchyma by advanced cirrhosis affected the result. Although it was not appeared in these papers, variable level of ultrasonographic quality is one of the most important issues of ultrasonographic surveillance for HCC. In Korea, a nationwide screening program for HCC has been performed from 2003 as a part of the National Cancer Screening Program (NCSP), and its quality assessment was also performed. According to the report about the quality control of screening liver ultrasonography [6], 21% of hospitals (143/685) and 32% (645/1,985) of private clinics did not passed the quality assessment. In addition, there are some patients’ factors to interfere an ultrasonographic exam, such as obesity, coarse echotexture, thickened adipose tissue of abdominal wall, increased waist circumference, uncooperative patient, a lot of bowel gas, and difficulty of position change due to limited patients’ movement [4]. Apart from these limitations, ultrasonography still has an unconquerable limitation: ultrasonography is operator-dependent. The quality of exam by experienced operator is different from that by novice, and the result can be also different.

To solve the problems of conventional imaging surveillance of HCC, we should consider introducing other objective imaging modalities such as CT or MRI. As mentioned above, the sensitivity of MRI to detect the curable HCC lesion which is able to be treated by locoregional treatments, such as resection, radiofrequency ablation, microwave ablation, and cryotherapy, was very significantly higher than that of ultrasonography. However, a major consideration is the cost-effectiveness of the surveillance program using MRI. Therefore, it would be desirable that MRI surveillance should be limited to those who will be able to be cured; for example, early stage HCC, compensated cirrhosis, and good performance status. Because the development risk of HCC is not same in the risk population, a tailored strategy using these imaging and serologic examinations should be considered. In addition, an effective quality assessment system for surveillance test including ultrasonography should be introduced and the surveillance program should be controlled to achieve the purpose of HCC surveillance.