This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Haller, D. G. Search for Related Content PubMed PubMed Citation Articles by Haller, D. G. Social Bookmarking                            What's this?

Waiting for the Definitive Trial of Hepatic Arterial Chemotherapy for Colorectal Cancer...

University of Pennsylvania Cancer Center , Philadelphia, PA

LIVER-DIRECTED THERAPIES for colorectal cancer are firmly based in the natural history of the disease, in which the liver is the dominant site of metastases in the majority of patients and apparently the sole site in some. For this latter group of patients, surgical resection of metastases represents a standard of care, based on multiple case series demonstrating disease-free and overall survival rates exceeding 25%. Without surgery, treatment with systemic chemotherapy alone would likely result in only a tiny fraction of 5-year survivors, none of them free of disease. Given the data, it is not surprising that a randomized clinical trial comparing surgical resection with curative intent with systemic chemotherapy has neither been designed nor desired.

For other liver-directed therapies, in which reproducible, prolonged, disease-free survival rates have not been demonstrated, clinical trials are required to establish efficacy and utility. In this issue of the Journal of Clinical Oncology, Lorenz and Müller¹ report the results of a trial from the German Cooperative Group on Liver Metastases, one of a long series of studies designed to establish the role of hepatic arterial chemotherapy compared with intravenous systemic chemotherapy. The concepts that lead to consideration of hepatic arterial therapy are not new: aside from the site-specific rationale for regional therapy, there are also pharmacologic reasons for administering fluoropyrimidines directly into the blood supply of the liver. Floxuridine, in particular, meets the criteria of being active in colorectal cancer and having a high hepatic clearance, thereby allowing for a strong regional delivery advantage over systemic therapy. This latter advantage is considerably less for fluorouracil.

It has been 35 years since Sullivan and Zurac² reported on the use of prolonged ambulatory hepatic infusion of floxuridine in patients with hepatic metastases from colorectal cancer. In the initial series, 24 of 39 assessable patients achieved an objective response by the criteria established by the investigators, more than expected with the same drug given systemically. Subsequent to this report, the Central Oncology Group reported the results of the first randomized controlled trial of hepatic arterial infusion with fluorouracil compared with the same drug given systemically.³ Although the response rate with intra-arterial infusion was slightly higher in this trial, there was no survival benefit for the hepatic infusion. The small number of patients, the short duration of hepatic arterial infusion, and the use of fluorouracil rather than floxuridine confounded the results of the study. Further randomized trials of hepatic arterial infusion therapy were performed after this trial, the results of which were reviewed in a meta-analysis published in 1996.⁴ As in the first randomized trial, these studies were also plagued by methodologic problems, including potentially inadequate control arms, small numbers of patients, and technical problems or surgical complications associated with catheter placement. In addition, many studies, by design or by default, allowed for cross-over from the hepatic infusion arm to systemic chemotherapy and vice versa. The combined analysis showed a 41% response rate for hepatic infusion therapy, which is significantly higher than that seen with systemic chemotherapy. There was no survival advantage for the hepatic arterial approach, when intravenous chemotherapy was the control arm, although each individual trial may have been statistically underpowered to detect small but clinically relevant differences. Subset analyses in three trials that allowed cross-over suggested a survival benefit in patients initially allocated to hepatic arterial infusion therapy. However, this observation is subject to selection bias because patients who could be treated with hepatic arterial infusion after failure of systemic therapy were likely to have had metastases confined to the liver, as well as to have met other clinical criteria associated with longer survival.

From these initial studies, it was clear that a number of technical and methodologic complications associated with trials of hepatic arterial infusion therapy needed resolution before a definitive study of the efficacy of the approach could be performed. Early trials showed that improvements in preoperative staging, catheter placement, and postoperative management were mandatory. Inadequacies in any of these inevitably led to poor outcome, not only for the individual patient but also for analyses of study results. In many of these trials, significant numbers of patients failed to receive treatment because of surgical complications or catheter misplacement, which led to a potential underestimate of the benefits of hepatic infusion therapy when statistical analyses were based on intention to treat. In addition, the choice and doses of the infusion chemotherapy may not have been optimal in many trials. Both reducing the hepatic toxicity of floxuridine by adding dexamethasone to the infusion mixture and increasing the cytotoxicity of fluoropyrimidines with leucovorin have been explored in phase II trials.⁵ Hepatic arterial infusions with a combination of floxuridine, leucovorin, and dexamethasone have therefore been considered to be standard by many institutions for patients with hepatic metastases from colorectal cancer.

While refinements in hepatic arterial therapy have been made, changes in systemic chemotherapy for colorectal cancer have also occurred, which has made the choice of an appropriate control arm something of a moving target for randomized trials. In the first studies, systemic infusions of either floxuridine or bolus fluorouracil were considered to be adequate controls. In more recent years, clinical experience and meta-analytic approaches have supported the concept that biochemically modulated fluorouracil or continuous infusion of fluorouracil is likely to be more effective than bolus fluorouracil alone in metastatic colorectal cancer, by objective response if not always by overall survival. Use of these systemic approaches in clinical practice and as standard control arms in recent clinical trials has made it imperative that they also be used as the control arm of any recently designed trial comparing hepatic arterial infusion therapy to systemic therapy.

In this issue of the Journal of Clinical Oncology, the results of a large, randomized trial are presented in which patients were treated with systemic fluorouracil and leucovorin or with hepatic arterial infusions of either fluorouracil and leucovorin or floxuridine. Patients were randomized before surgery so that the two groups of patients assigned to the hepatic arterial infusion arms had more extensive staging, including laparotomy. The hepatic arterial infusion floxuridine therapy was relatively standard, although leucovorin and dexamethasone were not included in the regimen. The 5-day fluorouracil and leucovorin regimens were initially the same in dose and schedule, whether given intrahepatically or systemically. Earlier phase I/II trials of these treatment arms were not well documented, and the systemic arm proved intolerable, requiring dose reduction. Many of the same problems that plagued earlier trials were also present in this study. Overall, of the 168 patients who were entered, only slightly more than two thirds received the assigned treatment. Eighteen patients assigned to the hepatic arterial infusion arms were found to have extrahepatic disease, and many of them were given the systemic chemotherapy treatment. Overall, 86 patients received some form of cross-over therapy, with more patients on the systemic arm not receiving such treatment. Mortality was considerably higher in the group who underwent surgical exploration, thereby further decreasing the likelihood of detecting a survival advantage in the group assigned to the hepatic infusion arms. These apparent flaws in study execution are typical of hepatic arterial infusion trials and have little to do with the intent or quality of the investigators. Indeed, the consistency of these problems across such studies suggests that they are inherent to the technique and are reflective of what is also intrinsic to complex clinical trials as opposed to more easily controlled laboratory investigations.

The investigators of this trial wished to find an end point that could serve as a surrogate for efficacy of the hepatic arterial approach, uncomplicated by the cross-over issues that made earlier trials unassessable. Therefore, time to progression on the assigned treatment was selected as the primary statistical end point, with overall survival and response as secondary end points. As in early trials, the response rate for hepatic arterial therapy was higher than that for systemic therapy (45% for hepatic arterial fluorouracil and leucovorin, 43% for hepatic arterial floxuridine, and 20% for systemic fluorouracil and leucovorin). Median survival was not significantly different among the three arms, with a range of 12.7 months for the hepatic arterial infusion of floxuridine to 18.7 months for hepatic arterial infusion of fluorouracil and leucovorin. The systemic arm showed an impressive median survival of 17.6 months. For the primary end point, time to progression, there were no significant differences among the three treatments, although there was a trend favoring the hepatic arterial infusion of fluorouracil and leucovorin. One of the most interesting aspects of this trial was the differential risk of the development of extrahepatic metastases among the treatment arms, with the highest rate in the patients receiving hepatic arterial floxuridine and the lowest in the groups receiving the hepatic arterial or systemic fluorouracil and leucovorin. These data rather nicely support both the positive and negative aspects of hepatic-directed chemotherapy. The high clearance of floxuridine results in very low systemic levels of drug, likely resulting in a higher rate of development of extrahepatic metastases. The lower clearance of fluorouracil, with higher systemic drug levels, seems to result in improved control of extrahepatic disease as well as increased systemic toxicity. This latter effect was well documented in this trial, in that the toxicity profile of fluorouracil and leucovorin was similar, whether the drugs were given intrahepatically or systemically.

The implications of these findings are numerous. It does not seem that giving fluorouracil via a liver-directed pathway is significantly better than systemic therapy with the same drug. This observation has also been made in the analyses of adjuvant portal vein infusion studies for resected colorectal cancer, in which median survival is modestly prolonged without an obvious or consistent impact on hepatic recurrence. Additionally, if one is exploiting the high hepatic clearance of floxuridine, then some additional systemic chemotherapy would be advisable to reduce the progression of extrahepatic disease. This could be accomplished by the addition of a systemic drug to hepatic arterial floxuridine. Systemic fluorouracil (with or without leucovorin), irinotecan, and oxaliplatin are all legitimate options, and each is already being explored in conjunction with hepatic arterial infusion in phase I/II trials.

On the basis of the results of all of the trials performed to date and the findings in the current study, it is easy to agree with the conclusion of Lorenz and Müller that hepatic arterial therapy does not represent a standard treatment option for patients with hepatic metastases from colorectal cancer. Further studies are indeed warranted, to establish, as a proof of principle, that hepatic arterial therapy leads to improved survival compared with systemic therapy. In the clinic, this has been difficult to prove. A current trial being conducted in the Unites States (Cancer and Leukemia Group B 8481) has been designed to definitively answer this question. In this trial, patients are randomly assigned to hepatic arterial floxuridine, leucovorin, and dexamethasone or to systemic therapy with bolus fluorouracil and leucovorin. The systemic arm is administered using the 5-day technique developed at the Mayo Clinic; it has been widely used in other trials in advanced colorectal cancer as a standard control arm. To avoid the methodologic problems found in earlier trials, no cross-over is permitted, so those patients with systemic failure will be discouraged from having hepatic arterial therapies as salvage treatment. Companion studies in this trial will also address medical resource utilization, quality of life, and molecular markers of prognosis and response. The statistical design of the study requires 340 patients to detect a 50% improvement in median survival for the hepatic infusion arm. The statistical design is appropriate, but recent improvements in systemic chemotherapy and the realities of patient care raise issues as to whether the study design remains germane. With the introduction of irinotecan and oxaliplatin into clinical practice, response rates of 50% are now being obtained in large-scale trials of combination chemotherapy trials in previously untreated patients with colorectal cancer. These rates are equivalent to those seen in prior studies of hepatic arterial therapy. Although the current Cancer and Leukemia Group B trial is correctly designed to prove whether hepatic-directed floxuridine is superior to bolus fluorouracil and leucovorin, the control arm may no longer be considered standard practice, by virtue of the introduction of infusion fluorouracil and oral fluoropyrimidine for their lower toxicity and by the advent of combination chemotherapy programs. Although problematic for clinical investigators who must design studies based on best practice at the time a trial is designed, changes in systemic chemotherapy of colorectal cancer is a real phenomena that must be addressed. If one were to design a new trial today, an appropriate design might include a control arm of combination chemotherapy with fluorouracil and irinotecan or oxaliplatin and a treatment arm of intrahepatic floxuridine, leucovorin, and dexamethasone given in conjunction with the best systemic therapy that has yet been found to be tolerable in ongoing phase I/II trials. If one is optimistic about new advances in colorectal cancer, even this trial design might seem outdated by the time of its activation, much less its completion.

It may well be that the definitive determination of the role of hepatic directed chemotherapy for colorectal cancer will never be made, based on the problems that have beset previous attempts and the natural history of colorectal cancer, which is still predominantly one of systemic metastases. Perhaps the real role for hepatic arterial chemotherapy will be in the subset of patients who have already been selected for hepatic-directed therapies, such as liver resection. The very process by which patients are selected for these procedures also makes them ideal candidates for additional, intensive liver-directed therapy, since the liver remains a common site of relapse after hepatic resection. With increased radiographic surveillance of patients, and with improved surgical techniques, the absolute number of patients undergoing liver resection and other ablative techniques is likely to rise, presenting oncologists with the task of reducing recurrences after such potentially curative therapy. There are already preliminary data from two trials suggesting a benefit for hepatic arterial infusion therapy, with or without systemic chemotherapy, in patients who have undergone liver resection for colorectal cancer metastases. Perhaps it is in the adjuvant setting that the true efficacy and utility of hepatic-directed chemotherapy will be determined.

REFERENCES

1. Lorenz M, Müller H-H for the German Cooperative Group on Liver Metastases: Randomized, multicenter trial of fluorouracil plus leucovorin administered either via hepatic arterial or intravenous infusion versus fluorodeoxyuridine administered via hepatic arterial infusion in patients with nonresectable liver metastases from colorectal carcinoma. J Clin Oncol 18:243-254, 1999[Abstract/Free Full Text]

2. Sullivan RD, Zurac WZ: Chemotherapy for liver cancer by protracted ambulatory infusion. JAMA 194:93-98, 1965

3. Grage TB, Vassilopoulos PP, Shingleton WW, et al: Results of a prospective randomized study of hepatic artery infusion with 5-fluorouracil versus intravenous 5-fluorouracil in patients with hepatic metastases from colorectal cancer: A Central Oncology Group study. Surgery 86:550-555, 1979[Medline]

4. Meta-Analysis Group in Cancer: Reappraisal of hepatic arterial infusion in the treatment of nonresectable liver metastases from colon cancer. J Natl Cancer Inst 88:252-257, 1996[Abstract/Free Full Text]

5. Kemeny N, Conti JA, Cohen A, et al: Phase II study of hepatic arterial floxuridine, leucovorin, and dexamethasone for unresectable liver metastases from colorectal carcinoma. J Clin Oncol 12:2288-2295, 1994[Abstract/Free Full Text]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				M. S. Copur, M. Capadano, J. Lynch, T. Goertzen, T. McCowan, R. Brand, and M. Tempero Alternating Hepatic Arterial Infusion and Systemic Chemotherapy for Liver Metastases From Colorectal Cancer: A Phase II Trial Using Intermittent Percutaneous Hepatic Arterial Access J. Clin. Oncol., May 1, 2001; 19(9): 2404 - 2412. [Abstract] [Full Text] [PDF]

				D. J. Kerr and R. J. M. C. S. McArdle Hepatic Arterial Chemotherapy: Still on Trial J. Clin. Oncol., April 1, 2000; 18(8): 1806 - 1807. [Full Text] [PDF]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Haller, D. G. Search for Related Content PubMed PubMed Citation Articles by Haller, D. G. Social Bookmarking                            What's this?