Should a biopsy precede liver resection or transplantation for presumed hepatocellular carcinoma when the alfa fetoprotein is normal?

Abstract

Hepatocellular carcinoma (HCC) commonly occurs on a background of cirrhosis, and various clinical studies have shown the incidence of new tumors to be between 1% and 5% per year (1). Regular surveillance with liver ultrasound and measurement of the serum alfa fetoprotein (AFP) is now commonplace in the detection of tumors at an earlier stage in their natural history, when treatment is most effective. With more sensitive imaging techniques such as triphasic multislice spiral computed tomography (CT), nodules smaller than 1 cm can be detected. The nature of these nodules are often uncertain, and, at this stage, the AFP is not usually raised. Current guidelines (2) recommend that when a nodule is less than 2 cm, it should be biopsied because current imaging techniques are not sufficiently accurate to differentiate benign nodules from HCC. For nodules greater than 2 cm, imaging techniques should confidently establish the diagnosis without the need for biopsy. However, Durand et al. (3) performed biopsies on 137 patients under ultrasound before they underwent liver resection or transplantation for suspected HCC. Thirteen of 15 with a negative biopsy were found to have HCC after surgery, but the remaining 2 had benign nodules. Between March 1999 and August 2001, 16 patients underwent a targeted biopsy for presumed HCC in our unit. As per our protocol at the time (before the European Association for the Study of Liver Disease (EASL) consensus guidelines being published) (2), all patients had nodules diagnosed on a screening liver ultrasound and had a normal AFP. Furthermore, CT demonstrated the presence of typical hypervascular nodule(s), with enhancement in the arterial phase in all patients (Table 1). An experienced radiologist then performed targeted biopsy under CT guidance with an 18 gauge Trucut needle. Eleven of the 16 patients receiving biopsies had the diagnosis of HCC confirmed. We report the outcome of the other five patients whose initial diagnosis of HCC was refuted on the basis of a targeted biopsy of the nodule(s). The histology was negative in all five for HCC, demonstrating cirrhosis only. Patients were then followed up with 3 monthly AFP measurements and CT scans for 4 years. Three of the patients (2, 4, and 5 in Table 1) showed no change in nodule size. Patient 1 and 3 subsequently developed HCC. Patient 1 showed no change at 3 months, but at 6 months the AFP rose to 28 ng/mL, and the nodule increased from 4×4 cm to 5.5×5.5 cm. Repeat biopsy confirmed well-differentiated HCC. Patient 3’s nodule increased in size from 2.5×2.5 cm to 3×3 cm after 3 months, and a repeat biopsy confirmed well-differentiated HCC, although the AFP remained normal at 8 ng/mL. To date, none of the patients receiving biopsies have developed needle-track seeding.Table 1: Characteristics of patients with negative first targeted biopsy for hepatocellular carcinoma (HCC)Although two of the five patients were eventually shown to have HCC, three patients had the diagnosis refuted, thereby avoiding the risk of aggressive treatments such as resection or transplantation in a patient with no malignancy. This has to be balanced against the risk of needle-track seeding of the tumor, which occurred in 1.6% of patients in the study by Durand et al. Our results suggest that biopsy of a presumed HCC may be important in those patients who have focal liver nodules suggestive of HCC and normal AFP. However, patients with presumed HCC, but in whom the liver biopsy does not confirm this, should be closely followed up because of the potential for false-negatives. Debbie L. Shawcross Nickolai Naoumov Ioannis Pachiadakis Constantinos Mamais Roger Williams Rajiv Jalan Alice Gillams William Lees