In his review, Dr. Venook correctly argues that, in the majority of pa;tients, hepatocellular carcinoma results from underlying liver disease; the most common culprit is cirrhosis, which, in turn, is frequently related to hepatitis B and/or hepatitis C exposure and alcohol(Drug information on alcohol) abuse. Given that patient outcomes are determined by the interplay between tumor growth and adequate hepatic reserve, and that most patients with hepatocellular carcinoma eventually die of liver failure, Dr. Venook argues that there is a good rationale for locoregional tumor control of hepatocellular carcinoma. Locoregional therapies may include hepatic intraarterial (HIA) chemotherapy, transarterial chemoembolization, Lipiodol chemo-embolization, radiation therapy (conformal external radiation therapy or intraarterially delivered radiation), or ablative procedures. These therapies are less aggressive than conventional resectional therapies, such as cryosurg-ery, percutaneous ethanol injection, radiofrequency ablation, and other intratumoral therapies.
Dr. Venook discusses several published HIA chemotherapy protocols, including EAP (etoposide, Adriamycin, and Platinol) and FLAP (floxuridine, leucovorin, Adriamycin, and Platinol), as well as the systemically administered chemotherapy regimen PIAF (cisplatin, interferon-alfa, Adriamycin, and fluorouracil(Drug information on fluorouracil)). Since HIA chemotherapy with FLAP requires pump implantation in most cases, and has been associated with increased toxicity in hepatitis-infected patients, this approach has to be reserved for selected individuals. Indeed, as Dr. Venook reports, in the University of California, San Francisco, (UCSF) experience, < 10% of all hepatocellular carcinoma patients were candidates for surgical placement of regional chemotherapy infusion devices.
With regard to transarterial chemoembolization, Dr. Venook discusses the pharmacologic principles underlying this approach, the various materials that can be used for the procedure, and the problems associated with documenting tumor response. The role of transarterial chemoembolization in patients awaiting orthotopic liver transplantation is also discussed. In this setting, chemoembolization is employed to diminish systemic tumor cell release at the time of the native liver explant and to prevent tumor progression during the wait for a donor liver. This approach has not been tested in a randomized trial.
The use of Lipiodol, an agent attractive to proponents of chemoembolization because of its unique ability to carry lipophilic drugs and its capacity to be radioactively labeled, has been subjected to a randomized trial. That study had to be closed early because no survival benefit seemed likely, and because acute hepatic failure frequently developed in the treated group. Conversely, a study of HIA chemotherapy with iodine-131 in Lipiodol, administered within 6 weeks following curative resection of hepatocellular carcinoma, demonstrated a significant increase in disease-free survival in the treated group.
Dr. Venook rightfully concludes that, as a consequence of coexistent liver disease in most patients, the ability to treat hepatocellular carcinoma regionally is limited. Therefore, newer agents and strategies are required, suggesting that all patients be treated in the context of research protocols. It is unclear why Dr. Venook states that the coexistence of liver dysfunction with liver tumor(s) makes regional treatment approaches more appealing. However, the suggestion that all patients be included in research protocols is very wise.
M. D. Anderson Approach to Hepatocellular Cancer
At M. D. Anderson Cancer Center, we have taken a multidisciplinary approach to the treatment of patients with hepatocellular carcinoma. Our recent review of patients referred to M. D. Anderson indicated that 23.9% of the risk for hepatocellular carcinoma could be attributed to hepatitis C infection; approximately 10% to hepatitis B infection, and 41% to daily alcohol consumption; no clear cause could be identified in 25% of cases. In addition, 50% to 60% of all hepatocellular carcinoma patients at M. D. Anderson had associated liver cirrhosis that could be diagnosed clinically.
The presence of cirrhosis poses a major challenge to the treatment of hepatocellular carcinoma. The associated thrombocytopenia and neutropenia exclude the use of myelosuppressive therapy. Ascites third space fluid volume expansion, along with intravascular hypovolemia, may prohibit the use of agents that require intravascular volume expansion, such as cisplatin(Drug information on cisplatin). Hyperbilirubinemia may prevent the use of anthracyclines or require dose adjustments. Portal vein occlusion by clot or tumor, which is commonly present, may rule out effective hepatic artery chemoembolization in most cases. Patients who have severe encephalopathy and liver failure are more likely to die from liver failure than from hepatocellular carcinoma. Thus, treatment of hepatocellular carcinoma must be tailored to the individual patients needs and must take into consideration residual functioning liver reserve.
Small hepatocellular carcinoma lesions may be managed by such interventions as resection, radiofrequency ablation, cryoablation, and alcohol ablation. However, these interventions are limited by the size, number, and location of the tumors and by the underlying liver reserve. Most importantly, none of these interventions targets micrometastatic disease or additional primary lesions present within or outside of the liver. Although orthotopic liver transplantation may be curative in patients with cirrhotic livers that contain very small tumors, the need for antirejection drugs following transplantation may promote tumor growth and accelerate tumor recurrence.
Limited Role for Liver-Directed Therapies
Therefore, at M. D. Anderson, we have focused our investigations on systemic therapies. We reserve liver-directed therapy for patients with disease limited to the liver who may become candidates for tumor resection, radiofrequency ablation, cryoablation, or orthotopic liver transplantation after a response to systemic or the addition of HIA chemotherapy. When opting for liver-directed therapy, we prefer HIA chemotherapy over transarterial chemoembolization for the following reasons: (1) Hepatic intraarterial chemotherapy is feasible despite compromise of portal blood flow, and it can be accomplished with implantable infusion pumps, while transarterial chemoembolization always requires arteriography. (2) Transarterial chemo-embolization is contraindicated in the presence of severe portal vein obstruction and may accelerate liver decompensation. (3) Repeated transarterial chemo-embolization is commonly associated with occlusion of the hepatic artery, preventing further access to that artery and precluding further intra-arterial therapy.
When used, HIA chemotherapy is delivered most commonly via an implantable infusion pump. Placement of such a pump secures patency of the hepatic artery and allows for long-term liver-directed therapy. As mentioned above, the FLAP regimen was associated with increased toxicity among hepatitis C virus and/or hepatitis B virusreactive patients. However, universal use of prophylactic filgrastim(Drug information on filgrastim)
(Neupogen) and use of lower chemo-therapy doses improved treatment tolerance and eliminated the survival difference between hepatitis-reactive and -nonreactive patients, resulting in an overall median survival of 15 months.[Y. Patt, unpublished data, 2000]
Systemic Therapies for Different Clinical Scenarios
When choosing systemic therapies, we have designed protocols appropriate for several different clinical scenarios. Thus, in patients who have hepatocellular carcinoma without liver cirrhosis or with Childs class A cirrhosis, treatment is begun with investigational agents that may have myelosuppressive properties, administered as part of phase II trials.
Our current phase II trial is investigating a topoisomerase I inhibitor, DX-8951f (Daiichi Pharmaceuticals, Japan) as front-line therapy. Patients who do not respond to this therapy will be treated as part of a planned phase III multi-institutional trial designed to assess the role of recombinant interferon-alfa-2b (Intron A) in advanced hepatocellular carcinoma. The combination of systemic PIAF (Platinol, interferon-alfa-2b, Adriamycin, and fluorouracil), reported by our group to be associated with histologic complete responses, will be compared to the same three chemotherapeutic agents without interferon (the PAF regimen).
Responders to PIAF or PAF whose cancer becomes resectable or amenable to ablative therapy, such as radiofrequency ablation, cryoablation, alcohol ablation, or even orthotopic liver transplantation, will undergo the procedure, and will receive 3 to 6 more months of post-surgical adjuvant treatment with the same regimen that induced the response. In patients whose disease responds to initial treatment or responds but is not surgically ablatable, and whose tumor is still confined to the liver, HIA with PIAF will be initiated.
Thus, at M. D. Anderson, regional HIA is not used as front-line therapy. Rather, use of this approach is reserved for selected patients.
Patients with Childs class B liver cirrhosis are not candidates for systemic or hepatic arterial PIAF chemobiotherapy or for other myelosuppressive treatment. These patients require tolerable therapies. We have completed a pilot study of capecitabine(Drug information on capecitabine) (Xeloda) in such patients. In this study, capecitabine treatment was tolerated even by patients with significant liver cirrhosis and thrombocytopenia. It was associated with a 15% partial response rate and a median survival of 15 months.[Y. Patt, data submitted for publication, 2000]
Another investigational agent, thalidomide(Drug information on thalidomide) (Thalomid) is undergoing phase II trials at our center. Initial analysis of the 12 patients treated with thalidomide demonstrated one partial response and disease stability in four patients. Major toxicities included drowsiness and skin reactions.[Y. Patt, data submitted for publication, 2000]
Yet another therapy that can be tolerated by patients with Childs class B liver cirrhosis is a continuous intravenous infusion of fluorouracil together with subcutaneous interferon-alfa-2b C. Resection, radiofrequency ablation, or orthotopic liver transplantation is offered to patients with some evidence of response to such therapy or at least long-term disease stability. However, patients whose disease progressed during therapy are not considered candidates for any of those surgical interventions, and, obviously, orthotopic transplantation should not be attempted.
The presence of Childs class C liver cirrhosis with refractory encephalopathy indicates that the patient is not a candidate for any anticancer therapy and should be treated palliatively and referred to hospice care. Patients with Childs class C liver cirrhosis who have very small tumors (less than 3 to 5 cm) who are placed on the waiting list for an orthotopic liver transplant may be candidates for a tolerable systemic therapy with such agents as capecitabine, fluorouracil, or thalidomide while awaiting transplantation. Obviously, prior to orthotopic liver transplantation, such patients must be reassessed to establish that they are still candidates for this procedure.
We believe that efforts and resources should be devoted to identifying agents that can be tolerated by hepatocellular cancer patients with liver cirrhosis. Although regional approaches are of interest, they are not appropriate for most hepatocellular carcinoma patients as front-line treatment, but rather, should be considered as part of a multidisciplinary approach aimed at making the tumor resectable or ablatable. These therapies should be subjected to clinical trials instead of being used in clinical practice.