Regional Strategies for Managing Hepatocellular Carcinoma

Hepatocellular carcinoma is a major public health problem worldwide, although at present it remains a relatively uncommon cancer in the United States. As pointed out by Dr. Venook in his elegant review of the topic, most hepatocellular carcinomas progress locoregionally. Hepatic failure is the most common mode of death for patients with this disease. For this reason, regional management strategies would appear to be attractive. Dr. Venook is to be commended for an accurate review of the literature regarding this issue. Unfortunately, that literature suffers from many limitations.

Limitations of Existing Clinical Trials

Numerous regional treatment strategies for hepatocellular carcinoma have been tried, but none has been adequately investigated to the point where it can be considered a clearly established part of routine standard care. The “clinical trials” reported are often small series from which few meaningful conclusions can be drawn. Many reports are retrospective reviews of institutional experiences. These are fraught with all of the serious limitations of retrospective analyses, including recall bias, selection bias, and other methodologic flaws.

Few, if any, trials have truly addressed the question of whether or not a particular regional therapy is superior to another regional approach, a less invasive systemic approach, or even supportive care. Rather, these studies have focused on presenting justifications for the therapeutic modalities that have been selected for use.

An all too familiar conclusion of many of these small, uncontrolled trials is that “the results appear to be superior to historical controls.” Aside from the obvious limitations of such a comparison, one must be careful to ensure that the appropriate historical controls have been selected. Most trials require adequate performance status and organ function for entry. Often, historical controls have been subjected to no such limitations. Thus, the very fact that the patient was selected for a phase II trial may put that patient into a superior prognostic category.

As pointed out by Dr. Venook, patients with hepatocellular carcinoma represent a diverse population of individuals with multiple related comorbidities. Patients with disease secondary to cirrhosis from hepatitis B vs hepatitis C vs other etiologies clearly have different long-term prognoses. In fact, in such patients, it is the underlying hepatitis and resultant cirrhosis that may be the primary life-threatening illness, rather than the hepatocellular carcinoma itself. In these cases, short of palliation of symptoms due to tumor bulk, there would appear to be little role for regional therapeutic strategies. Clearly, there is a need for appropriate control of the variables that these comorbidities bring, with larger studies and appropriate stratifications needed to sort out these issues.

Hepatic Intraarterial Chemotherapy

Hepatic intraarterial (HIA) chemotherapy has been widely explored in metastatic colorectal cancer, and large-scale, randomized studies are in progress in that disease. It is important to reemphasize, as pointed out by Dr. Venook, that hepatocellular carcinomas differ markedly from hepatic colorectal metastases, with respect to underlying biology and underlying associated comorbidities. Therefore, direct extrapolation of data from colorectal cancer to hepatocellular carcinoma is probably inappropriate.

When we evaluate the justification for HIA in hepatocellular carcinoma, the evidence to date consists primarily of small phase II trials, the largest of which contained 31 patients.[1] These trials reported extensive toxicity, and overall survival did not differ markedly from that which would be expected from historical reports. In fact, the one randomized, prospective investigation of intrahepatic chemotherapy in hepatocellular cancer, conducted during the post-resection adjuvant phase of treatment, showed that patients who received intrahepatic plus systemic therapy had a worse outcome than those treated with surgery alone.[2]

In the absence of additional data, therefore, this approach clearly should not be considered a standard treatment.

Chemoembolization

Chemoembolization is another popular treatment approach for which a critical review of the data is warranted. Again, the theoretical rationale for this therapeutic modality is strong, but the clinical evidence to support it is somewhat lacking.

The results with Lipiodol appear to be of particular concern. Two randomized, phase III studies, described by Dr. Venook, showed no apparent survival benefit. Both studies found a high percentage of patients with early signs of liver failure following Lipiodol chemoembolization.[3,4]

An additional retrospective review showed that routine planned administration of this technique was potentially inferior to the application of Lipiodol embolizations at the time of sympto-matic progression. Given the negative data, it is hard to be enthusiastic about the routine use of Lipiodol in the management of patients with hepatocellular cancer.

Assessing the ‘State of the Art’

What, then, can we conclude about the state of the art of regional management strategies of hepatocellular carcinoma? In general, we can say that we, the scientific community, have failed in our responsibility to critically evaluate the relative efficacy of these treatments. Individual institutions have perpetuated their preferred approach, often under the rubric of a phase II protocol, or without a specific protocol but then reviewed retrospectively. Data suggesting a lack of efficacy have failed to dramatically alter the choice of treatments that have been pursued.

When is the right time for an embolization? Should it be performed with concurrent chemotherapy or by itself? What agents should be used for embolization? We don’t know the answers to these questions because the head-to-head comparative trials that might provide those answers have not been done yet.

Clearly, if ever there were a need for randomized, multicenter trials, it is in this area. As newer locoregional modalities, such as alcohol ablation, cryosurgery, and radiofrequency ablation, enter the clinical arena, these, too, will require prospective, randomized evaluation. These newer modalities clearly have potential risks and increased costs. Only through the proper conduct of adequately powered, randomized, controlled clinical trials will we ever find out whether or not these interventions re truly in our patients’ best interests.

References:

1. Patt YZ, Charnsangavej C, Yoffe B, et al: Hepatic arterial infusion of floxuridine, leucovorin, doxorubicin, and cisplatin for hepatocellular carcinoma: Effects of hepatitis B and C viral infection on drug toxicity and patient survival. J Clin Oncol 12:1204-1211, 1994.

2. Lai ECS, Lo C-M, Fran S-T, et al: Postoperative adjuvant chemotherapy after curative resection of hepatocellular carcinoma. Arch Surg 133:183-188, 1998.

3. Groupe d’Etude et de Traitement du Carcinome Hepatocellulaire: A comparison of Lipiodol chemoembolization and conservative treatment for unresectable hepatocellular carcinoma. N Engl J Med 332:1256-1261, 1995.

4. Pelletier G, Ducreux M, Gay F, et al: Treatment of unresectable hepatocellular carcinoma with Lipiodol chemoembolization: A multicenter randomized trial. J Hepatol 29:129-134, 1998.