Toward a rational management of very early hepatocellular carcinoma


  • Mario Strazzabosco M.D., Ph.D

    Corresponding author
    1. Department of Internal Medicine Yale University New Haven, CT
    2. Department of Surgery and Interdisciplinary Medicine University of Milan-Bicocca Milan, Italy
    • Department of Internal Medicine, Section of Digestive Diseases, Yale University School of Medicine, 333 Cedar Street, LMP 1080, 06520 New Haven, CT 06520===

    Search for more papers by this author
    • fax: 1-203-785-7273

  • Potential conflict of interest: Nothing to report.

See Article on Page 1426.


AFP, alpha-fetoprotein; BCLC, Barcelona Clinic Liver Cancer staging system; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; LT, liver transplantation; MELD, Model for End-Stage Liver Disease; PH, portal hypertension; RFA, radiofrequency ablation.

As the incidence of hepatocellular carcinoma (HCC) rises in western countries, including the United States, its natural history is also changing, as a result of more-widespread surveillance, improved imaging technology, and novel treatments.1,2 Currently, small HCC lesions are being diagnosed with greater frequency, but the management of these patients remains challenging. Several options are available, from liver transplantation (LT) to radiofrequency ablation (RFA) or surgical resection, with overlapping short-term results.

There is reasonable agreement that patients presenting with an HCC ≤2 cm and a history of decompensated cirrhosis, or a Model for End-Stage Liver Disease (MELD) score higher than 15, should be listed for transplant and the tumor should be ablated before liver function decreases further. Ablation remains the treatment of choice, if transplant cannot be offered.

On the other hand, the treatment of patients with a single lesion ≤2 cm in diameter, compensated cirrhosis, and very good functional status (very early HCCs or stage 0, according to the Barcelona Clinic Liver Cancer [BCLC] classification) may create a therapeutic dilemma.

According to the BCLC algorithm, which has been endorsed by the American and European liver study associations (the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver, respectively),3 resection of the lesion is recommended for patients without portal hypertension (PH) and with healthy bilirubin. Furthermore, several retrospective reports showed that short-term results of surgical resection and RFA for BCLC stage 0 HCC are similar,4,5 suggesting that BCLC stage 0 patients may be equally considered for hepatic resection or RFA.

Therapy with RFA, a procedure that is most effective when applied to singe-nodule HCC less than 3 cm in diameter, has increasingly gained acceptance as a first-line modality for very early HCC. Livraghi et al.6 reported on the outcomes of patients with a single HCC lesion ≤2 cm and well-compensated disease and noted a 68.5% survival rate at 5 years, with a high HCC recurrence rate at 5 years (80%).

The largest retrospective experience on the outcomes of surgical resection in very early HCC was reported on by Ikai et al. (2,320 patients), who showed a 3- and 5-year survival of 84% and 66%, respectively.7 Lee el al.8 reported a 3-year survival of 82.5% for tumors ≤2 cm and 67.7% for tumors between 2 and 3 cm. Overall survival for very early HCC was markedly affected, if microvascular invasion was found on histology (54.8% at 3 years).

This issue of HEPATOLOGY publishes the result of a combined effort from two large referral centers, based in New York and Milan,9 reporting on the outcomes and prognostic factors of surgical resection in patients with HCC <2 cm in diameter. In this large western series, survival at 5 years was 70% and was influenced by the presence of satellites and by platelet count, with the best results obtained in patients with more than 150,000 platelets (83% survival at 5 years). After 5 years of follow-up, HCC recurred in 68% of patients; prognostic factors for recurrence were the presence of satellites, cirrhosis, and performance of a nonanatomical resection. Contrary to larger hepatomas, in patients with very early HCC, recurrence peaked at 30 months after surgery, suggesting a major influence of the cirrhotic field.

In this study, platelet level is proposed as a surrogate index of PH. Although this is a useful rule of thumb, a key component in the evaluation of patients with very early HCC should remain the accurate determination of the degree of PH. An elevated hepatic venous pressure gradient is a significant risk factor for HCC, a predictor of postresection survival, and the most accurate prognostic predictor for patients with liver cirrhosis.10,11

Roayaie et al. also showed that at least one third of these early tumors present with microvascular invasion or satellites.9 On the other hand, approximately 5% of these lesions turn out to be either benign lesion or hepatocholangiocarcinomas.15 Thus, surgical resection has the advantage of confirming the diagnosis and providing important pathologic information that may affect the following treatment strategy (i.e., prevention of recurrence, indication, and timing for transplantation).

LT offers a curative treatment of both the HCC and the accompanying chronic liver disease and has become the treatment of choice for nonresectable BCLC stage A HCCs.3 However, the role of transplantation in the management of very early HCC must be framed in light of the ongoing shortage of available organs, the current basis for organ allocation that favors medical urgency, and the competing interests between transplant candidates with and without HCC.

The natural history and prognostic factors of patients with HCC are different from those of patients with cirrhosis alone, in which the MELD system efficiently predicts short-term mortality; as a result, the current allocation system still favors HCC over non-neoplastic indications.12,13 In the July issue of Hepatology, Toso et al.,14 using the Scientific Registry of Transplant Recipients database and a proportional hazard competitive risk model, proposed a novel score that should enable a fair allocation of liver grafts between oncologic and nononcologic transplant candidates. Because of the different scales between HCC score (based on age, diagnosis, MELD, size and number of nodules, and alpha-fetoprotein [AFP] levels) and MELD, a clever MELD equivalent was created and should enable a direct comparison of the drop-out risks of both types of transplant candidates. According to this model, an HCV-infected, well-compensated patient with cirrhosis and with very early HCC and normal AFP would be assigned a MELD-equivalent score of 7, consistent with the current practice of not listing very early HCC with conserved liver function.

Accordingly, transplantation for very early HCC should be considered when the HCC recurs (rescue transplantation) or if certain predictors are present at pathology (the principe transplantation).15 Using an intention-to-treat approach, Fuks at al.16 retrospectively compared resection, followed by salvage transplantation, versus first-line transplantation in a group of 330 potential transplant candidates with early HCC. The article, published in the February issue of Hepatology, shows that 5-year overall and disease-free survivals were similar in the two groups (approximately 70%). Among the patients in the resection group, 22 patients did not experience a recurrence (in most of these cases, patients had a single nodule of less than 2 cm and no satellites) and 20 were transplanted before recurrence because of liver deterioration. Of the remaining 90 patients, in 30, HCC recurred ouside the Milan criteria. Among the 60 cases in which recurrence was inside the Milan criteria, 21 could not be transplanted because of advanced age, comorbidities, or refusal. Pathologic analysis of the resected specimens contained important prognostic information. This study indicates that salvage transplantation allows remarkable survivals at 5 years with the use of a lower number of grafts. However, the risk of recurrence beyond the Milan criteria remains substantial, suggesting the use of tissue analysis as a selection criterion and that salvage transplantation should be restricted to those patients with favorable oncologic predictors. Still, it remains unclear whether patients with unfavorable oncologic indications should be offered immediate transplantation (de principe transplantation).

“Secondary prevention” of HCC is of extreme importance; close surveillance and immediate control of a recurrence could reduce the percentage of patients experiencing a multifocal HCC recurrence, which precludes transplantation. Key elements of this strategy include regular post-treatment dynamic imaging, along with regular serum AFP determination, in those with elevated AFP. In Fuks et al.'s study, most patients that recurred outside the Milan criteria did not have a regular follow-up.15 Clinical trials with medical agents able to prevent HCC recurrence are anxiously awaited.

Maintenance of liver function and treatment of the underlying cause of cirrhosis play a major role in this strategy. Patients with cirrhosis and with compensated disease have a remarkable survival, if progression can be prevented or delayed. For patients with alcohol-or HBV-induced cirrhosis clinical compensation can be achieved and maintained in the majority of patients, provided that alcohol consumption ceases and HBV treatment is instituted, if indicated. For hepatitis C virus (HCV)-infected patients with well-compensated cirrhosis and very early HCC, HCV therapy should be considered after treatment of the tumor, depending on individualized patient assessment, despite the lower sustained virologic response rate.

In summary, proper treatment of early HCC is complex; these patients should be managed in referral centers that are able to offer all of the above-mentioned procedures. Short-term results of surgical resection for BCLC stage 0 HCC are comparable to those achievable with LT. Prompt treatment of a very early HCC is justified by the ability to ensure disease control up front and then to embark on a program of close patient monitoring. Resection and RFA are both reasonable options and should be considered on an individual and center-specific basis. However, resection provides more-accurate histological diagnosis and staging. Patients treated for very early HCC must be monitored closely in a program of surveillance and strict follow-up, with prompt referral for further treatment or transplant in the event of recurrence. Every effort must be implemented to provide etiologic treatment and to meticulously manage the patient with underlying cirrhosis. Patients should be entered into clinical trials addressing the prevention of HCC recurrence Further studies are needed to compare outcomes after long-term follow-up (up to 10 years) with the various treatments as well as to determine the role of medical treatments in the secondary prevention of HCC after treatment.


This work was was funded by the National Institutes of Health (DK079005) and the Silvio O. Conte Digestive Diseases Research Core Centers (DK034989).