A needle in a haystack: Identifying biomarkers to personalize systemic therapy in patients with hepatocellular carcinoma


  • Henning Wege M.D., Ph.D.,

    Corresponding author
    1. Department of Internal Medicine, Gastroenterology and Hepatology University Medical Center Hamburg-Eppendorf Hamburg, Germany
    • Department of Internal Medicine, Gastroenterology and Hepatology, University Medical Center Hamburg-Eppendorf, Martinistraße 52, D-20246 Hamburg, Germany===

    Search for more papers by this author
    • fax: +40-40-7410-59092

  • Tim H. Brümmendorf M.D., Ph.D.,

    1. Department of Oncology and Hematology University Medical Center Aachen Aachen, Germany
    Search for more papers by this author
  • Carlo Gambacorti-Passerini M.D., Ph.D.

    1. Clinical Research Unit Section of Hematology University of Milano Bicocca San Gerardo Hospital, Monza, Italy
    Search for more papers by this author

  • Potential conflict of interest: Dr. Wege received grants from Bayer.

See Article on Page 1407.

Hepatocellular carcinoma (HCC), the most prevalent primary liver cancer, causes the third-highest mortality rate after lung and colon cancer worldwide. Although most cases occur in Asia, steadily rising incidence rates have been observed in Europe and North America during the last two decades. As a result, HCC constitutes a major health problem in the care and management of patients with liver cirrhosis. In patients with cirrhosis or chronic active hepatitis B, ultrasound surveillance is recommended every 6 months to increase the rate of early HCC detection. However, more than 70% of cases still present in intermediate or advanced stage worldwide, without curative treatment options.1 Advanced HCC (stage C according to the Barcelona Clinic Liver Cancer classification) is characterized by macrovascular invasion, extrahepatic spread (lymph node involvement or metastases), or cancer-related symptoms (Eastern Cooperative Oncology Group 1-2). In these patients, palliative treatment is possible, employing systemic therapy.

Genetic profiling has suggested that HCC progression is attributed to a number of altered signaling pathways as well as epigenetic mechanisms. Thus, as in other tumor entities, targeted agents are investigated as novel therapeutic options. Along this line, angiogenesis inhibition is a prime therapeutic target in solid tumors, especially in highly vascularized HCC. Since the successful completion of the SHARP study in 2007,2 the antiproliferative and angiostatic multi–tyrosin kinase inhibitor (TKI), sorafenib, has been approved as the first systemic agent for treatment of patients with unresectable or metastatic HCC and preserved liver function. Sorafenib primarily inhibits BRAF/vascular endothelial growth factor receptor (VEGFR)/platelet-derived growth factor receptor tyrosin kinases mediating cell proliferation and angiogenesis; it also blocks many additional kinases, given its lack of selectivity.3 The mechanism underlying the antitumor effect of sorafenib is complex, and even RAF-independent signaling has recently been described as a significant pathway of sorafenib-induced cell death.4 At present, sorafenib represents the only drug with statistically significant, but clinically modest, benefit in terms of improvement in overall survival (OS), time to progression (TTP), and disease control rate. Efficacy of sorafenib in HCC was demonstrated in two large randomized, controlled trials (RCT): SHARP and the Asia-Pacific study,2, 5 sorafenib significantly reduces the risk of death and prolongs median OS by approximately 3 months. Survival benefit is based on an extended TTP. Sorafenib does not induce tumor-size reduction, and radiologic response has to be confirmed by a decrease in viable tumor mass. Therefore, to identify novel drugs with activity in HCC in future and ongoing trials, modified Response Evaluation Criteria In Solid Tumors criteria have been proposed.6


FGF3/4, fibroblast growth factors 3 and 4; FGFR, FGF receptor; HCC, hepatocellular carcinoma; OS, overall survival; RCT, randomized, controlled trial; TKI, multi–tyrosin kinase inhibitor; TTP, time to progression; VEGFR, vascular endothelial growth factor receptor.

Sorafenib is associated with relevant toxicities, especially in patients with compromised liver function (stage Child-Pugh B). Patients with cirrhosis suffer from fatigue, and in such patients, sorafenib may cause a rapid deterioration in quality of life. Therefore, predictive (bio)markers to guide therapy with sorafenib are urgently needed. Many clinical parameters and serum markers as well as genomic signatures have been suggested; for example, development of hypertension and diarrhea, known side effects of sorafenib, under therapy seem to be associated with a favorable outcome.7, 8 Furthermore, in Child-Pugh B patients, baseline aspartate aminotransferase has been proposed to predict for patients more likely to benefit from sorafenib.9 To date, no molecular predictive biomarker for sorafenib response in HCC treatment has been validated and accepted in clinical practice. Efforts in such a direction were performed during the SHARP molecular biomarker program, but were of limited success.10 Furthermore in the same trial, 10 plasma biomarkers implicated in hepatocarcinogenesis were measured serially in 305 patients, and only high s-c-KIT or low hepatocyte growth factor showed trends toward enhanced survival. In addition, exploratory subset analyses of the Asia-Pacific trial also failed to identify markers associated with superior response to sorafenib.5

Working along this line of research, currently, Arao et al. have successfully identified a novel potential molecular biomarker: fibroblast growth factors 3 and 4 (FGF3/FGF4) amplification in a patient with partial response and long-term survival under sorafenib. Based on this finding and by employing copy number polymerase chain reaction, fluorescence in situ hybridization, and gene-expression analysis, the investigators extended their investigation and retrospectively identified FGF3/FGF4 amplification in 30% of sorafenib responders (3 of 1011). This frequency is significantly increased, compared with the overall reported rate of FGF3/FGF4 amplification, amounting to 2.4% in unselected patients with HCC. Therefore, the investigators conclude that FGF3/FGF4 amplification with consecutive up-regulation of gene expression seems to be associated with a substantially increased likelihood and degree of response to sorafenib. These findings were functionally confirmed by the screening of various cancer cell lines for FGF3/FGF4 up-regulation and by its correlation with in vitro response to sorafenib. In addition, enhanced proliferation in a cell clone expressing FGF4 was reversed by sorafenib in a nude mouse assay.

Although elegantly performed and scientifically sound, the study by Arao et al. is somewhat hampered by several limitations. First, a small cohort of patients is studied and retrospective analyses may lead to a selection bias. Second, an unbiased genome-wide evaluation was performed only in the index patient, and the finding in this patient (FGF3/FGF4 gene amplification) was confirmed in 3 of 10 other patients, but without genome-wide assessment in this population. Therefore, it remains unclear whether the proposed amplification is a driving genetic alteration rendering the affected cells particularly susceptible to sorafenib treatment and, if so, by which mechanism. Alternatively, the finding could represent a secondary bystander phenomenon. Based on clinical and molecular data, up-regulated signaling through FGF family receptors (FGFRs) has been identified as a promising therapeutic target in patients with HCC. Signaling through the FGFR family is associated with fibrogenesis and development of cirrhosis, the main risk factor for malignant transformation in chronic liver disease. FGF/FGFR-mediated signaling is highly conserved, and downstream pathways include RAS/RAF mitogen-associated protein kinase and phosphoinositide 3-kinase/protein kinase B cascade.12 Several alterations in FGFR signaling correlate with outcomes in patients with HCC.13 A recent clinical study demonstrated FGF8, FGF17, and FGF18 and their receptors, FGFR2, FGFR3, and FGFR4, to be up-regulated in HCC.14 Additional studies also suggested that FGF4, FGF5, FGF9, and FGF12 may be involved in HCC progression and metastasis.15 Furthermore, FGF up-regulation is one of the proposed escape mechanisms for HCC under sorafenib therapy. In fact, escape from sorafenib treatment by FGF activation was the rationale for an international RCT evaluating brivanib as a dual VEGFR/FGF inhibitor in sorafenib progressors. Along this line, it will be interesting to see whether altered signaling through FGF plays a role for resistance development under sorafenib in HCC (and other sorafenib-responsive malignancies).

Finally, it is important to note that the majority of evaluated sorafenib responders have an underlying viral etiology (11 of 13), and chronic hepatitis C is the most prevalent risk factor in Japan. However, in Europe and North America, other etiologies, in particular, alcoholic and nonalcoholic steatohepatitis have a growing effect. Therefore, transferability of the results to other regions of the world might be limited.

The study by Arao et al. exemplifies a valuable translational research approach (namely, from bedside to bench and back) and, as such, an important next step toward a personalized systemic treatment approach in HCC. A prospective evaluation of the suggested target in a large cohort of patients, including patients from Europe and North America, is clearly warranted. Sorafenib represents the standard treatment for patients with advanced HCC and preserved liver function (stage Child-Pugh A). FGF-signaling alterations have been identified as promising targets in patients with advanced HCC, and already, several novel agents targeting this receptor family, such as brivanib, dovitininb, and intedanib, are being investigated in clinical trials. These new agents will be evaluated in RCTs against sorafenib. This underscores the need for genome-wide sequencing, followed by the functional analysis of targets identified, to personalize molecular targeted therapy in patients with HCC. Development of personalized treatment algorithms has been identified as an urgent medical need and has been proposed as a short-term clinical aim by the investigators of the new EASL/EORTC Clinical Practice Guidelines for HCC.16