Advanced hepatocellular carcinoma (HCC) is the fifth commonest malignancy worldwide and one of the most challenging to treat. Most single agent chemotherapy drugs produce zero responses in HCC with the best response rate being 35%. While there has been progress in systemic chemotherapy for advanced HCC, the two major obstacles to more successful therapeutic outcome is the high frequency of overexpressed drug resistance genes in HCC that includes classical multidrug resistance genes such as p-glycoprotein, topoisomerase-II, glutathione-S-transferase, glyoxalase-2 related protein, heat shock proteins, p53 mutations, bcl-2 and bcl-xL. The other problem is that HCC largely arises on a background of liver cirrhosis and liver dysfunction, making it difficult to deliver some chemotherapy agents. In fact, cirrhosis with ascites, portal vein thrombosis and high bilirubin are associated with poor response to chemotherapy. Performance status is the strongest predictor to poor outcome to systemic chemotherapy.

The most well studied chemotherapy agent for HCC is the anthracycline, doxorubicin, with an total average response rate of 18% in 734 patients and a median survival of 4 months. A prospective Hong Kong clinical trial suggests a small survival benefit with doxorubicin over placebo (10.6 weeks versus 7.5 weeks). Doxorubicin is limited by its contraindication in patients with significant liver dysfunction. 5-fluorouracil (5-FU) is easier to deliver for HCC even on the background of compromised liver function. When combined with folinic acid, the response rate can be as high as 28% in a series of 25 HCC patients. The oral 5-FU prodrug, capecitabine is similarly safer in HCC patients with abnormal liver function and with one United States study reporting a 14% response rate which includes one radiologic complete response. Single agent gemcitabine may induce some responses in HCC patients with reports ranging from 0% in both a Harvard and Austrian study to 2.1% in a China series to 17.8% in a Taiwanese study.

Combination chemotherapy produces better results in HCC. Cisplatin (CDDP)-containing regimens such as epirubicin, 5-FU and CDDP produces a response rate of up to 25% a regimen comprising gemcitabine and CDDP produces a response rate of 21%. However, median survivals were still short. A National Cancer Centre clinical study is evaluating gemcitabine and CDDP in our local population. More recently, a French Phase II study reported 19% response rate and 48% stable disease to combination gemcitabine and oxaliplatin. One compelling Hong Kong study with CDDP, interferon-alpha (IFN- a ), doxorubicin, and infusional 5-FU (PIAF), a modification of an MD Anderson regimen, reported a response rate of 26% in advanced HCC and pathologic complete response in 44% of HCC patients who were able to undergo surgical resection following chemoimmunotherapy cytoreduction. The overall median survival was 8.9 months. It was also shown that normalization of tumour marker alphafetoprotein ( a FP) was predictive of pathologic complete remission in these HCC patients. This illustrates that radiological response evaluation may underestimate the true response. Toxicity was a problem in this intensive regimen with a 4% mortality and significant morbidity. Combinations of 5-FU with IFN- a have divergent results with one study reporting responses in 9 of 36 HCC patients. The role of single agent IFN- a against HCC may be related to its anti-angiogenesis mechanism, its ability to induce anti-tumour immunity or both, but clinical studies have again ranged from zero response rates to studies suggesting response rates better than doxorubicin with a survival benefit.

An oral anti-angiogenesis drug, thalidomide, has been tested in HCC with some modest responses (6%) and 50% a FP decrease in another 10 of 68 patients.

New combinations such as capecitabine and avastin (an antibody against vascular endothelial growth factor), the focus of an ongoing Asia-Pacific study, are promising, not least for its ability to produce low toxicity and potentially target this highly vascular tumour.

While there has been encouraging data in both molecular (targeted therapy) and immunotherapy (eg : vaccines) in liver cancer in vitro studies and animal models, these are not definitive standard of care in the bedside setting and should be, and are being assessed, in clinical trials.

Certainly, HCC is a tumour in need of much new therapeutic developments to make a significant clinical impact.

Dr Toh Han Chong

Senior Consultant

Department of Medical Oncology, National Cancer Centre, Singapore