The therapeutic options for hepatocellular carcinoma (HCC) have been so far rather inadequate. Sorafenib has shown an overall survival benefit and has become the new standard of care for advanced HCC. Nevertheless, in clinical practice, some patients are discontinuing this drug because of side effects, and misinterpretation of radiographic response may contribute to this. We highlight the importance of prolonged sorafenib administration, even at reduced dose, and of qualitative and careful radiographic evaluation. We observed two partial and two complete responses, one histologically confirmed, with progression-free survival ranging from 12 to 62 mo. Three of the responses were achieved following substantial dose reductions, and a gradual change in lesion density preceded or paralleled tumor shrinkage, as seen by computed tomography. This report supports the feasibility of dose adjustments to allow prolonged administration of sorafenib, and highlights the need for new imaging criteria for a more appropriate characterization of response in HCC.

Therapeutic options for advanced hepatocellular carcinoma (HCC) have been so far inadequate[1-5]. Recent progress in molecular biology has allowed identification of new therapeutic targets, including vascular endothelial growth factor (VEGF), which is overexpressed and related to progression-free survival (PFS) and overall survival (OS) in HCC[6,7]. Sorafenib, an oral multi-kinase inhibitor, blocks tumor cell proliferation and angiogenesis by targeting RAF, VEGF receptors, platelet-derived growth factor receptor β, c-KIT and FLT3 signaling pathways[8].

Efficacy and safety of sorafenib have been demonstrated by phase II/III trials, which have proven that 400 mg bid sorafenib significantly prolongs OS, reduces the risk of death by 31%, and increases the time to radiological progression[9-11]. Consequently, sorafenib has become the standard of care for patients with advanced HCC.

We present four cases of unresectable, systemic-treatment-naïve HCC, enrolled in phase II (HOPE) and III (SHARP) trials at the Department of Medical Oncology and Hematology of the Istituto Clinico Humanitas in Rozzano (Milan, Italy), who obtained long-lasting objective responses. All patients were diagnosed by histology and computed tomography (CT) or magnetic resonance imaging (MRI). Their peculiar responses and the personalized management of side effects provide suggestions to optimize the use of sorafenib in the management of HCC patients.

Sorafenib is the only effective systemic therapy for the treatment of HCC, but side effects lead to treatment discontinuation in some patients. Nowadays, HCC is treated by hepatologists or oncologists. The former may be less accustomed to the typical side effects of anticancer drugs, and the latter may not be keen to manage problems related to underlying liver cirrhosis. This report proves the importance of a multidisciplinary approach in the management of advanced HCC patients.

The described cases highlight how, in case of sorafenib-related side effects, reductions and pauses in the administered dose can allow long-term treatments. Efficacy of conventional cytotoxic agents is strictly related to the administered dose. With new targeted agents, length of treatment, rather than dose intensity, may be fundamental for tumor control. The winning strategy may lie in managing side effects, and tailoring the anticancer regimen to the characteristics of the patients, rather than discontinuing treatment at the appearance of signs of intolerance. Unfortunately, data on drug blood levels that are needed to achieve and maintain target inhibition are inadequate. The multi-target nature of sorafenib is one additional challenge, which means that various factors play a role in the activity of this agent[6-8].

In three of the four reported cases, objective responses were achieved following substantial dose reductions. In patient 2, partial response was observed at month 16 of the study (7 mo at full dose and 9 at half dose). In patient 3, sorafenib dose was reduced by 50% during the month 6 on treatment; an objective response was seen in month 8 of therapy, and a complete response was achieved after a total of 34 mo on study. In patient 4, sorafenib dose was reduced by 50% after just 10 d of therapy, and 20 d later, a partial response was observed. One month later, dose was further reduced to 25%, and complete response was documented after 24 mo on study. The lesson appears clear: the recommended sorafenib dose is 400 mg bid; when needed, dose reductions, by limiting side effects, offer a better quality of life and can allow long-term administration and achievement/maintenance of tumor control.

The radiological features of responding lesions are another issue worthy of attention. When assessing the efficacy of targeted therapies by imaging, a gradual change in tumor density and blood flow may be observed before tumor shrinkage. However, the uncommon radiological patterns can lead to late recognition of responses, or worse, to misleading evaluations. Indeed, in two of the reported cases (patients 3 and 4), lesions seemed unchanged and, at first sight, had been considered active. On the contrary, these lesions were responding and being substituted by residual scars (in patient 4, this was confirmed by liver biopsy). This issue deserves serious consideration and calls for new, more appropriate methods of appraisal. With targeted therapies, traditional methods of quantitative evaluation, such as WHO criteria or Response Evaluation Criteria in Solid Tumors, may not be optimal, and the need for qualitative standardized measurements becomes more pressing[9,12-14]. Although positron emission tomography is not reliable for evaluation of HCC, Hounsfield unit (HU) density scale for CT scan imaging, and signal patterns for MRI can be used to measure tumor necrosis. Combination of these methods with dimensional measurements allows more precise characterization of sorafenib responses in this disease[13,14]. In our experience, a lesion density reduction to 40 HU on CT scan can be considered indicative of tumor necrosis.

An additional mean of response evaluation may be provided by analysis of blood samples. At present, there is no agreement on specific biomarkers of response to sorafenib. We analyzed routine laboratory tests and did not identify any correlation in our patients, however, we did not search for alternative signals such as markers of inflammation and oxidative stress. The recommendations that derive from the reported cases are to tailor dose and schedule, to administer the drug until progressive disease is observed, and to evaluate critically dimensional and density changes in tumor lesions. Anticancer drugs have evolved in recent years, and so should the way physicians view tumor treatment strategies. Although this calls for more personalized treatment plans, agreement on standardized and more appropriate assessment techniques will allow more conscious decision making in the treatment of advanced HCC patients[13,14].