I read with interest the article regarding the role of imaging techniques in the treatment of hepatocellular carcinoma (HCC) by Bruix et al. published in HEPATOLOGY.1 In this special article, the investigators made an updated, comprehensive review on the pivotal role of imaging techniques in the management of HCC. This article is very helpful for clinical hepatologists to tackle problems that are frequently encountered when facing various stages of HCC. However, several points may be misleading and need clarification.
As we know, serum alpha-fetoprotein (AFP) is widely used in the Asia-Pacific region to diagnose and follow-up HCC.2 This tumor marker is neglected by the investigators simply because of inadequate sensitivity. On the other hand, immunohistochemical staining for glypican 3, heat shock protein 70, glutamine synthase, and clathrin heavy chain, with doubtful value, and unfamiliar to most hepatologists, are believed by the investigators to be able to reinforce the diagnosis of HCC. This policy appears to be impractical. As stated in the article, there is risk of overestimation of the necrosis extent by imaging studies. Some patients classified as complete response have residual disease at the time of explant, if resected or transplanted. If patients have an elevated serum AFP before treatment, serum AFP is not likely to return to normal if residual cancers persist. By the supplement of serum AFP, the risk of overestimation could definitely be further reduced. Regarding the treatment interval of chemoembolization, the investigators suggested two treatment sessions at baseline, then repeated every 6 months. Follow-up monitoring beyond 1 month of therapy by computed tomography or magnetic resonance imaging was also suggested. If viable tumors are noted by imaging studies, another session of chemoembolization would be delayed until at the interval of 6 months. This would definitely elicit anxiety for HCC patients with large or multiple tumors.
Enlarged lymph nodes could evoke confusion in the staging of HCC. Indeed, the presence of slightly enlarged lymph nodes can be observed in cirrhosis of any etiology. Neither image can clearly recognize or exclude the presence of cancers. Hence, the investigators suggested that only lymph nodes greater than 2 cm could be declared as malignant involvement. This appears to be arbitrary, rather than evidence based. In my personal experience, malignant involvement of lymph nodes could be as small as 1 cm. In contrast, an individual with HCC and portal vein involvement and an abdominal lymph node up to 3 cm without cancer cells has also been encountered. Thus, assessment of malignant involvement cannot be simply based on the size of the lymph node. More studies are required to clarify this issue.