Hepatic arterial infusion using pirarubicin combined with systemic chemotherapy: a phase II study in patients with nonresectable liver metastases from colorectal cancer

D. Fallik1, M. Ychou2, J. Jacob3, P. Colin4, J. F. Seitz5, J. Baulieux6, A. Adenis7, J. Y. Douillard8, P. Couzigou9, R. Mahjoubi1, M. Ducreux1, M. Mahjoubi10 and P. Rougier1,11,+

1 Institut Gustave Roussy, Villejuif; 2 Institut Val d’Aurelle, Montpellier; 3 Centre François Baclesse, Caen; 4 Clinique Courlancy, Reims; 5 Centre Paoli Calmette, Marseilles; 6 Hopital Croix Rousse, Lyon; 7 Centre Oscar Lambret, Lille; 8 Centre René Gauducheau, Saint Herblain; 9 CHR Bordeaux, Pessac; 10 Laboratoire Aventis France, Paris; 11 Hopital Ambroise Paré, Boulogne, France

Received 5 November 2002; revised 16 December 2002; accepted 27 January 2003


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Background:

A prospective phase II study was performed to determine the feasibility, efficacy and safety of arterial hepatic infusion (HAI) using pirarubicin combined with intravenous chemotherapy.

Patients and methods:

From December 1991 to April 1994, 75 patients with unresectable colorectal metastases confined to the liver were included in this multicenter study to receive intra-arterial hepatic pirarubicin and a systemic monthly regimen of 5-fluorouracil (5-FU) and folinic acid. Sixty-four patients were analyzed in the intention-to-treat analysis and 61 in the per-protocol analysis.

Results:

Tolerance of this regimen was rather good; however, functional catheter problems were observed in 29 patients (45%) resulting in failure of HAI in 21 cases (33%) after a median of three cycles; vomiting grade 3 was present in 12.5% of patients, neutropenia grade 4 in 23% and alopecia grade 3 in 19%. The overall response rate was 31.9% in intention-to-treat analysis, and 39.3% in per-protocol analysis. Extrahepatic progression was reported in only 21.7% of patients. Time to hepatic progression and extra-hepatic progression was 8.3 and 15 months, respectively, in intention-to-treat analysis, and 11 and 18 months, respectively, in per-protocol analysis. Median survival was 19 and 20 months in intention-to-treat analysis and per-protocol, respectively.

Conclusions:

In our study, the combination of intra-arterial pirarubicin and intravenous chemotherapy demonstrated some efficacy and good tolerance in the treatment of isolated colorectal liver metastases. This treatment seems to prevent extra-hepatic spread and prolong survival time. The results of this study have to be confirmed by new trials using more active systemic chemotherapy.

Key words: colorectal cancer, hepatic arterial chemotherapy, liver metastases, pirarubicin


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Colorectal carcinoma is one of the most common cancers in France, with 33 500 new patients diagnosed every year—the third most abundant cancer after broncho-pulmonary cancer and breast cancer. For patients who develop advanced disease, liver metastases occur in 60% of cases. Hepatic resection is the best treatment when hepatic metastases are isolated, but this is only feasible in 10–20% of cases; for these patients, the 5-year survival rate is ~20–30% [13].

When metastases are unresectable but confined to the liver, arterial hepatic infusion (HAI) can be a valid therapeutic option. Used in these conditions, several studies have shown that HAI chemotherapy was able to increase response rates [4, 5]. The rationale for HAI has been extensively reported, and results from the liver’s dual blood supply and the clearance by the liver of many drugs. Briefly, it was demonstrated that liver metastases larger than 1mm are irrigated by the hepatic artery, whereas normal tissue liver is supplied by the portal vein, and that the hepatic clearance of many drugs before extra-hepatic passage allows one to increase, by a factor of 10–100, the drug’s exposure in the tumoral site and to reduce greatly the systemic side-effects [46].

Arterial hepatic infusion efficacy has been demonstrated in randomized trials using arterial infusion of 5'-deoxy-5-fluorouridine (FUDR), using an implanted pump (Infusaid 400; Infusaid Inc., Norwood, MA) and results in a significant increase of both the tumoral response rates in all studies [712] and the overall survival compared with symptomatic treatment or intravenous 5-fluorouracil (5-FU) bolus chemotherapy in two studies [12, 14]. The benefits of hepatic intra-arterial chemotherapy (HAIC) have been well illustrated in meta-analysis [13], and suggest an increase in median survival of 16 versus 12 months in favor of HAI, so encouraging the continued use of HAIC and the use new drugs, despite hepatic toxicity and systemic recurrences [13]. In contrast, a significant difference in terms of quality of life was not shown between HAIC and i.v. treatment [14, 15]. Finally, a medical economic study showed that the balance of cost/efficiency seemed to be acceptable for HAIC treatment [16].

Pirarubicin (THP)–doxorubicin is an anthracycline obtained by hemisynthesis of daunorubicin or doxorubicin. It is rapidly incorporated into tumor cells and shows antitumor activity by inhibiting nucleic acids synthesis, followed by cell death due to cessation of the cell cycle at the G2 phase. Like other anthracyclines, i.v. administration of pirarubicin demonstrates good clinical responses in breast cancer, cervical cancers and Hodgkin’s lymphoma, but not in colorectal cancer. Furthermore, HAIC with doxorubicin has not increased its therapeutic index and has shown limited advantages over systemic administration, with no reduction in systemic toxicities and a modest decrease in peripheral plasma levels. Nevertheless, THP–doxorubicin is active in HAIC on CRC liver metastases usually unresponsive to doxorubicin. In rabbits with Vx2 tumor implanted in the liver, the anti-tumor effect of THP upon HAI administration was better than that upon i.v. injection [17]. So the activity of THP was stronger than other anthracyclines because of its high clearance and high extraction during the first hepatic passage (80%). Toxicities caused by IA administration of pirarubicin, such cardiotoxicity, loss of hair and gastrointestinal adverse reactions, are low, which has been confirmed by experimental and clinical studies [18]. The risk of observing cardiac insufficiency is only 1% with a cumulative dose of 700 mg/m2 (against 20% with doxorubicin). The limiting toxicity is neutropenia and the dose advised using the systematic route is 50 mg/m2. Clinical studies in humans show that the tolerated dose during hepatic intra-arterial administration is greater than the tolerated dose when administered i.v. In this phase II study, in colorectal cancer patients with metastasis confined to the liver, THP (IA) was initiated at 60 mg/m2 with a 10 mg/m2 increment until grade 2 hematotoxicity. The maximum tolerated dose was 85 mg/m2 (range 60–120); 19 patients receiving a dose of HAI pirarubicin 60 mg/m2 of every 3 weeks showed good results, with a partial response (PR) observed in 33% of cases, with no hepatic or cardiac toxicity, but with severe neutropenia as the limiting toxicity [19].

These results demonstrate that HAI pirarubicin produced high locoregional concentrations with faster cellular uptake and reduced systemic exposure. It can also achieve responses in metastatic liver disease of colorectal origin, which led our group to set-up a multicentric phase II study testing the efficacy and tolerance of the combination of HAI pirarubicin with systemic 5-FU and folinic acid.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Eligibility criteria were as follows: histologically proven colorectal carcinoma; primary tumor totally resected; unresectable measurable or evaluable; hepatic metastases without evidence of extrahepatic disease; hepatic involvement <75%; age ≥18 years and ≤75 years; life expectancy >3 months; performance status of <3 [according to World Health Organization (WHO) criteria]; no prior chemotherapy; implanted catheter with subcutaneous access system allowing hepatic intra-arterial perfusion. Patients must have baseline normal functions: neutrophils >2 x 109/l; platelets >100 x 109/l; serum creatinine level <135 µmol/l; serum bilirubin level <35 µmol/l. Written informed consent was obtained from all patients. All patients had bidimensionally measurable or assessable disease documented by computed tomography (CT) scans of the abdomen. Computed tomography scans of the chest were performed if clinically indicated. Patients were excluded if they had any of the following criteria: previous myocardial infarction <6 months; previous angina pectoris; clinical cardiac insufficiency; chemotherapy or chemo-radiation within 6 months prior to study entry; malignancy or history of malignancy other than in situ cervix uteri carcinoma or localized basal/spinocellular cell skin cancer, non-controlled infection, pregnancy and women breastfeeding infants.

Treatment plan
Chemotherapy cycles were planned every 4 weeks. At day 1 of each cycle, HAI pirarubicin was administered at the dose of 70 mg/m2 during a 1 h perfusion through a subcutaneous access connected to a surgically placed arterial catheter. A decrease of 10 mg/m2 was indicated at the following cycle in case of WHO grade 4 neutropenia (neutrophils <0.5 x 109/l) and/or thrombocytopenia (platelets count <25 x 109/l) at day 15 of the previous cycle, and then eventually a second decrease of 10 mg/m2 if a grade 4 toxicity was observed again.

Intravenous chemotherapy was administered from days 1 to 5 of every cycle, using 5-FU at a dose of 400 mg/m2/day for 1-h infusion; this dose was decreased by 50 mg/m2 in the following cycle in cases of WHO grade 4 neutropenia and/or thrombocytopenia (platelets count <25 x 109/l), and/or grade 4 stomatitis or diarrhea at day 15 of the precedent cycle, a subsequent decrease of 50 mg/m2 was allowed. In case of persistence of grade 4 toxicity after two reductions, the treatment was stopped.

Folinic acid was administered by i.v. bolus at a dose of 20 mg/m2/day just before i.v. perfusion of 5-FU from day 1 to day 5, without dose adaptation. No other antitumoral treatment was authorized during the study (chemotherapy, radiotherapy). Symptomatic medications were authorized but avoided during the first cycle. Corticoids and non-steroidal anti-inflammatory drug (NSAID) were not allowed because of gastro-intestinal toxicity.

Follow-up
At inclusion and every 4 weeks physical examination, performance status and vital signs evaluations were collected. A biological work-up including ionogram, lactate dehydrogenase (LDH), {gamma}-glutamyl-transpeptidase, serum bilirubin, alkaline phosphatase, aspartate and alanine aminotransferase, serum creatinine, glycemia, prothrombine time and carcinoembryogenic antigen were performed. A complete blood count was made every week. An electrocardiogram was taken before each cycle.

Treatment evaluation
Efficacy evaluation was done using an abdominal CT scan at inclusion and then every three cycles. Chest X-ray and/or thoracic CT scan were carried out if there was any suspicion of lung metastases.

Tumor response was evaluated according to WHO criteria: a complete response (CR) was defined as the total resolution of all measurable sites of disease for a minimum of 4 weeks. A PR was defined as a ≥50% decrease in the sum of the products of the perpendicular dimensions of all measurable lesions for a minimum of 4 weeks without the appearance of new lesions; minor response corresponded to a decrease between 25% and 50%; stable disease (NC) was defined as no change or no increase or decrease >25% and no new lesions over 12 weeks, and progression (PD) was defined as an increase ≥25% or the development of new lesions.

Assessment criteria, sample size and statistical methods
The primary end point was response rate according to WHO criteria. The secondary end points were survival, time to hepatic progression, time to extra-hepatic progression and toxicity, which were analyzed according to WHO criteria.

Since the primary end point was response rate according to WHO criteria, 14 eligible and evaluable patients had to be included in succession. Indeed, if no response had been observed then the probability that treatment had an antitumoral activity >20% would have been <5%. It has been estimated that if at least one response was observed, 50 evaluable patients should be included to get a precise evaluation of the response rate.

Two analyses were planned: an intention-to-treat analysis including all eligible patients enrolled onto the protocol, and a per-protocol analysis, which considered only patients who received at least one complete HAI cycle. In the per-protocol analyses, all the responses were reviewed by an independent external panel of expert radiologists.

Survival was calculate from the date of randomization to the date of death. Survival and time to progression curves were estimated using the Kaplan–Meier method [20]. Cox’s proportional hazard modeling was used for the determination of prognosis factors for survival [21].

For each patient, the maximum grade for each type of toxicity was recorded. Frequency tables were analyzed to determine toxicity patterns.

Ethical considerations
The protocol was approved by the French Groupe de Reflexion sur l’Ethique Biomedicale de Bicêtre (GREBB).


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Patient characteristics
Seventy-five patients from December 1991 to April 1994 were included in this phase II trial. Patient characteristics, pathological data and pretreatment values of hematological and biological parameters are shown in Table 1. There were 48 men and 27 women; median age was 61 years (range 53–69). The median time from diagnosis to inclusion was 253 ± 435 days. Metastases were synchronous in 47% and metachronous in 53% of patients; the primary tumor was colon cancer for 71% of patients, rectal cancer in 35% and four patients presented synchronous primitive lesions to the colon and to the rectum. The median time between diagnosis of hepatic metastases and inclusion was 87 ± 147 days. The percentage of radiological hepatic invasion was <25% for 38% of patients, between 25% and 50% for 49% of patients, and >50% for 13% of patients. The maximum size of the largest liver metastases was >5 cm in 56% of patients.


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics at inclusion
 
Six patients were considered ineligible for the efficacy analysis because of an earlier catheter dysfunction and no HAI. Five others patients were considered inevaluable for the efficacy analysis for the following reasons: early death or major toxicity with treatment arrest in two cases after only one cycle; three patients had no radiological evaluation at 3 months. Sixty-nine patients were eligible and considered for intention-to-treat analysis; 64 were evaluable and included in the per-protocol analysis. Toxicity was assessed for 73 patients, since one patient did not receive pirarubicin, and no data were available for one patient. Four hundred and thirty-two cycles of treatment were administrated and 415 cycles completed (THP, 5-FU and AF). The median number of cycles received per patient was six (range 1–17).

Response rate
In the intention-to-treat analysis, one CR was observed (1.5%) and a 21 PRs (30.4%), the overall response rate was 31.9%. Twenty-six patients (37.7%) had NC and 21 patients (30.4%) only experienced PD.

Of the 64 patients considered for per-protocol analysis, a PR was observed in 24 patients (39.3%), 17 patients (27.9%) had stable disease and 20 patients (32.8%) had PD.

Time to hepatic progression
Median time to hepatic progression was 8.3 months [95% confidence interval (CI) 5.6–14.3] in intention-to-treat analysis and 11 months (95% CI 7.0–16.3) in per-protocol analysis (Figure 1).



View larger version (9K):
[in this window]
[in a new window]
 
Figure 1. Time to hepatic progression (intention-to-treat analysis)

 
Time to extra-hepatic progression
Extra-hepatic progression was observed in 15 of 69 (21.7%) patients as follows: lung (n = 8), lymph nodes (n = 1), bone (n = 2), adrenal glands (n = 2) and peritoneal carcinosis (n = 1). Median time to extra-hepatic progression was 15 months and 18 months in intention-to-treat and per-protocol analysis, respectively (Figure 2).



View larger version (7K):
[in this window]
[in a new window]
 
Figure 2. Time to extra-hepatic progression (intention-to-treat analysis).

 
Survival
The median follow-up period was 21 months; the 6 months survival rate was, in intention-to-treat analysis, 95% (standard deviation 3%), 75% (5%) at 1 year, 52% (6%) at 18 months and 33% (6%) at 2 years. Median survival was 19 months (95% CI 15–22 months) (Figure 3). In per-protocol analysis, median survival was 20 months (95% CI 15–24 months), the 1-year survival rate was 78% (5%) and the 2-year survival rate was 36% (6%).



View larger version (9K):
[in this window]
[in a new window]
 
Figure 3. Kaplan–Meier survival curve. Median survival 19 months.

 
According to univariate analysis, several prognosis factors were identified associated with overall survival: elevated serum carcinoembryonic antigen (CEA) (> normal; P = 0.044); elevation of the alkaline phosphatases (<170 U/l or ≥170 U/l; P <0.001); neutrophil count (<5000 versus ≥5000 mm3; P = 0.009); the liver invasion at CT scan (<25% versus 25–50% versus >50%; P <0.001); hepatic involvement assessed surgically (<25% versus 25–50% versus >50%; P = 0.009); and the performance status (0–1 versus 2; P = 0.016). On the other hand, according to multivariate analysis, we have identified prognosis factors associated independently with overall survival: elevation of alkaline phosphatases (P = 0.025); elevated serum CEA (P = 0.0299); liver invasion at CT scan (P <0.001); and performance status (0–1 versus 2; P = 0.081).

The results of the uni- and multivariate analysis showed no significant interaction between serum LDH and overall survival.

Toxicity
A functioning problem of the catheter appeared in 29 patients; intra-arterial hepatic catheter was unusable in 21 patients: five just after implantation and 16 after three cycles of chemotherapy. These complications included hepatic artery thrombosis, misperfusion and/or thrombosis of the catheter and catheter leakage. The average number of cycles without functioning problem was five. One patient discontinued therapy because of major infectious complications (cholangitis) and another patient because of abdominal pain. One patient died of respiratory insufficiency after one complete cycle.

Clinical toxicity was minimal; non-hematological toxicity was mainly nausea and vomiting grade 3 reported during 15 cycles in eight patients; alopecia grade 3 in 19% of the patients; and grade 3 diarrhea in 10% of patients. Table 2 shows complete results concerning clinical toxicity. No cardiac toxicity was reported during the study.


View this table:
[in this window]
[in a new window]
 
Table 2. Clinical toxicity
 
Hematological toxicities are listed in Table 3. Neutropenia grade 4 was reported for 27 cycles in 17 patients (23%), and anemia for six cycles in five patients (7%). Thrombocytopenia was rare (<1% of cycles).


View this table:
[in this window]
[in a new window]
 
Table 3. Hematological toxicity
 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
In the last 20 years, regional chemotherapy represented a valid therapeutic option for patients with unresectable liver metastases. However, HAI seemed less interesting since the development of more effective systemic protocols with new systemic agents, such as oxaliplatin or irinotecan, effective in colorectal cancer. Nevertheless, HAIC is still relevant in adjuvant situation after the resection of liver metastases, as reported by Memorial Sloan-Kettering Cancer Center [22], and its combination with active systemic chemotherapy may be an interesting approach in selected cases.

The objective of our study was to determine the feasibility, efficacy and safety of HAI using pirarubicin in combination with systemic chemotherapy using a 5-FU–folinic acid protocol in a palliative situation. The response rate in the present study (34.4%) does not appear to be different from those previously observed using intrahepatic FUDR. Most of the phase II studies have reported a response rate of about 40%, whichever agent is used [5]; however, it is interesting to underline the fact that this response rate may be obtained with a simplified HAI method using a subcutaneous access and a 1 h monthly intra-arterial administration in ambulatory patients. The median survival was 19 months in the present study which is equivalent to those reported by many studies [713]. The prognosis factors found in this study were performance status, hepatic replacement by metastases >50% and CEA elevation have already been reported [23].

The role of HAI has been widely discussed because of its limitations. Extra-hepatic progression is one of the principal factors of failure. In an early experience using intrahepatic FUDR, 50–70% extra-hepatic metastases were observed after 1 year of treatment [13]. However, this progression was not typical because it was often delayed and occurred in unusual sites, such as skin, brain, bones and suprarenal glands. In our study there was a relatively low rate of extra-hepatic evolution (23%), which suggests the possibility that the association of HAI with systemic chemotherapy may have provided some additional short-term therapeutic benefit and could be recommended to prevent occurrence of extra-hepatic metastases. A few prospective studies have demonstrated the benefit of HAI combined with intravenous chemotherapy. Lorenz et al. [24] revealed the superiority of HAIC with FUDR combined with 5-FU i.v. and folinic acid i.v. versus HAIC with FUDR alone, in terms of median time to progression (9.2 versus 5.9 months) and median survival (18.7 versus 12.7 months). In a randomized study by Safi et al. [25], 21 patients received treatment by FUDR (HAI i.v.) and 23 by FUDR (HAI); extra-hepatic progression was less frequent with the combined treatment (33% versus 61%) without significant difference in terms of hepatic response or overall survival. In a phase II study, combination of intravenous chemotherapy with 5-FU–folinic acid and HAIC with FUDR, a 62% objective response rate, a time to progression of 9 months and a median survival of 18 months were reported [26]. In a study conducted in the UK, the association of systemic folinic acid to an intra-arterial continuous infusion of 5-FU gave interesting results [2729], which has not been confirmed in a randomized trial.

Intra-arterial monotherapy with mitomycin C (MMC) gave only a 20% overall response rate (ORR) [30]; however, Cantore et al. [3133] reported the efficiency of the combination of 5-FU plus MMC plus epirubicin (HAI) associated with systemic 5-FU and folinic acid with an ORR of 50% and a median survival of 18 months.

Although the efficiency seems similar with intra-arterial pirarubicin, treatment with FUDR is regarded at present more critically due to its severe side effects. In our study, pirarubicin gave rise to no biliary toxicity, no cardiac toxicity and very low digestive toxicity. Biliary toxicity depends on the used protocol and has been observed mainly with the use of continuous HAI using FUDR; its frequency increases with the dose and duration of drip. In the French randomized trial, the risk of hepatitis was 35% and the risk of biliary sclerosis 25% after 1 year of HAIC [12]. No such risks were reported with the use of monthly HAI pirarubicin, apart from one patient who suffered from septic cholangitis.

Because the usefulness of HAIC therapy in controlling liver metastases in unresectable disease has previously been demonstrated, the search for the optimal intra-hepatic or systemic agent to use in combination with HAIC is the focus of current studies.

The Medical Research Council has conducted a prospective study comparing LV5FU2 (HAIC) with doses amounted to LV5FU2 (i.v.) with negative results, but with a very high rate of technical failure (ECCO 11).

The high rate of technical failure and catheter thrombosis in multicenter HAI studies is a major problem, and in this study five patients never received their local treatment for this reason. In 16 other patients, HAI has been interrupted for catheter dysfunction and an overall proportion of 39% of patients had functioning problems with their catheter. This underlines the necessity of using HAI in specialized centers.

The development of the combined treatment, using the most active products, could be a promising treatment allowing progression and widening indications to patients with extra-hepatic metastases of small volume. Finally, new modes of 5-FU administration, such as systemic chronomodulated chemotherapy [34].


    Conclusions
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
Hepatic intra-arterial chemotherapy is an effective method for the treatment of selected patients with colorectal cancer metastases. The results of this phase II study combining HAIC pirarubicin and systemic 5-FU–folinic acid have demonstrated some efficacy and good tolerance in the treatment of isolated colorectal liver metastases. The results of our trial seems to confirm that there is a place for HAIC using THP–doxorubicin combined with systemic chemotherapy. Additional studies should explore combinations of other systemic agents. When our study was developed, a 5 day bolus treatment regimen seemed to be optimal. Today, a conventional i.v. de Gramont regimen using continuous 5-FU infusion over 48 h is widely regarded as being more effective [35]. The results of this study consolidate the idea that it is necessary to realize trials comparing the most effective systemic chemotherapy to the same chemotherapy plus IA THP.


    Acknowledgements
 
We would like to thank M. Grandjean (Hopital Louis Mourier, Colombes) and M. Rivoire (Center Léon Bérard, Lyon) for participation in this trial, Mrs Rahouda Mahjoubi for assistance and QUANTA Medical for collection and analysis of the data.


    Footnotes
 
+ Correspondence to: Professor P. Rougier, Hôpital Ambroise Paré, 92100 Boulogne, France. Tel: +33-149-095325; Fax: +33-149-095327; E-mail: philippe.rougier{at}apr.ap-hop-paris.fr Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 Conclusions
 References
 
1. Wagner JS, Adson MA, Van Heerden JA et al. The natural history of hepatic metastases from colorectal carcinoma. A comparison with resective treatment. Ann Surg 1984; 199: 502–508.[ISI][Medline]

2. Colonna M, Grosclaude P, Faivre J et al. Cancer registry data based estimation of regional cancer incidence: application to breast and colorectal cancer in French administrative regions. J Epidemiol Community Health 1999; 53: 558–564.[Abstract]

3. Rougier P, Milan C, Lazorthes F et al. Prospective study of prognostic factors in patients with unresected hepatic metastases from colorectal cancer. Br J Surg 1995; 82: 1397–1400.[ISI][Medline]

4. Kemeny N. Review of regional therapy of liver metastases in colorectal cancer. Semin Oncol 1992; 19: 155–162.[ISI]

5. Rougier P, Lasser P, Elias D. Bases rationnelles et résultats de la chimiothérapie intra-artérielle. Ann Chir 1988; 42: 639–646.[ISI][Medline]

6. Taylor I, Bennett R, Sherriff S. The blood supply of colorectal liver metastases. Br J Cancer 1978; 38: 749–756.[ISI][Medline]

7. Kemeny MM, Goldberg D, Beatty JD et al. Results of a prospective randomized trial of continuous regional chemotherapy and hepatic resection as treatment of hepatic metastases from colorectal primaries. Cancer 1986; 57: 492–498.[ISI][Medline]

8. Chang AE, Schneider PD, Sugarbaker PH et al. A prospective randomized trial of regional versus systemic continuous 5-fluorodeoxyuridine chemotherapy in the treatment of colorectal liver metastases. Ann Surg 1987; 206: 685–693.[ISI][Medline]

9. Kemeny N, Daly J, Reichman B et al. Intrahepatic or systemic infusion of fluorodeoxyuridine in patients with liver metastases from colorectal carcinoma. Ann Intern Med 1987; 107: 459–465.[Medline]

10. Hohn DC, Stagg RJ, Friedman MA et al. A randomized trial of continuous intravenous versus hepatic intra-arterial floxuridine in patients with colorectal cancer metastatic to the liver: the Northern California Oncology Group trial. J Clin Oncol 1989; 7: 1646–1654.[Abstract]

11. Martin JK Jr, O’Connell MJ, Wieand HS et al. Intra-arterial floxuridine versus systemic fluorouracil for hepatic metastases from colorectal cancer. A randomized trial. Arch Surg 1990; 125: 1022–1027.[Abstract]

12. Rougier P, Laplanche A, Huguier M et al. Hepatic arterial infusion of floxuridine in patients with liver metastases from colorectal carcinoma: long-term results of a prospective randomized trial. J Clin Oncol 1992; 10: 1112–1118.[Abstract]

13. Meta-Analysis Group in Cancer. Reappraisal of hepatic arterial infusion in the treatment of non resectable liver metastases from colorectal cancer. J Natl Cancer Inst 1996; 88: 252–258.[Abstract/Free Full Text]

14. Allen-Mersh TG, Earlam S, Forfy C et al. Quality of life and survival with continuous hepatic-artery floxiuridine infusion for colorectal liver metastases. Lancet 1994; 344: 1255–1260.[ISI][Medline]

15. Earlam S, Glover C, Davies M et al. Effect of regional and systemic fluorinated pyrimidine chemotherapy on quality of life in colorectal liver metastasis patients. J Clin Oncol 1997; 15: 2022–2029.[Abstract]

16. Durand-Zaleski I, Roche B, Buyse M et al. Economic implications of hepatic arterial infusion chemotherapy in treatment of nonresectable colorectal liver metastases. Meta-Analysis Group in Cancer. J Natl Cancer Inst 1997; 89: 790–795.[Abstract/Free Full Text]

17. Munck JN, Riggi M, Rougier P et al. Pharmacokinetic and pharmacodynamic advantages of Pirarubicin over adriamycin after intra-arterial hepatic administration in the rabbit VX2 tumor model. Cancer Res 1993; 53: 1550–1554.[Abstract]

18. Munck JN, Rougier P, Chabot GG et al. Phase I and pharmacological study of intra-arterial hepatic administration of Pirarubicin in patients with advanced hepatic metastases. Eur J Cancer 1994; 30A: 289–294.

19. Rougier P, Munck JN, Elias D et al. Intra-arterial hepatic chemotherapy with Pirarubicin. Preclinical and clinical studies. Am J Clin Oncol 1990; 13: S1–S4.[ISI]

20. Kaplan E, Meier P. Nonparametric estimation from incomplete observation. J Am Stat Assoc 1958; 53: 457–481.[ISI]

21. Cox DR. Regression models and life tables. J R Stat Soc B 1972; 34: 187–220.[ISI]

22. Kemeny N, Huang Y, Cohen AM et al. Hepatic arterial infusion of chemotherapy after resection of hepatic metastases from colorectal cancer. N Engl J Med 1999; 341: 2039–2048.[Abstract/Free Full Text]

23. Elias D, Ducreux M, Rougier P et al. Chimiothérapie intra-artérielle hépatique. Expérience de 200 cas. Gastroenterol Clin Biol 1994; 18: 975–982.[ISI][Medline]

24. Lorenz M, Müller HH for the German Cooperative Group on Liver Metastases. Randomized, multicenter trial of fluorouracil plus leucovorin administrated either via hepatic arterial or intravenous infusion versus fluorodeoxyuridine administrated via hepatic arterial infusion in patients with nonresectable liver metastases from colorectal carcinoma. J Clin Oncol 2000; 18: 243–254.[Abstract/Free Full Text]

25. Safi F, Bittner R, Roscher R et al. Regional chemotherapy for hepatic metastases of colorectal carcinoma (continuous intra-arterial versus continuous intra-arterial/intravanous therapy). Result of a controlled clinical trial. Cancer 1989; 64: 379–387.[ISI][Medline]

26. O’Connell MJ, Nagorney DM, Bernath AM et al. Sequential intrahepatic fluorodeoxyuridine and systemic fluorouracil plus leucovorin for the treatment of metastatic colorectal cancer confined to the liver. J Clin Oncol 1998; 16: 2528–2533.[Abstract]

27. Howell JD, McArdle CS, Kerr DJ et al. A phase II study of regional 2-weekly 5-fluorouracil infusion with intravenous folinic acid in the treatment of colorectal liver metastases. Br J Cancer 1997; 76: 1390–1393.[ISI][Medline]

28. Howell JD, Warren HW, Anderson JH et al. Intra-arterial 5-fluorouracil and intravenous folinic acid in the treatment of liver metastases from colorectal cancer. Eur J Surg 1999; 165: 652–658.[CrossRef][ISI][Medline]

29. Warren HW, Anderson JH, O’Gorman P et al. A phase II study of regional 5-fluorouracil infusion with intravenous folinic acid for colorectal liver metastases. Br J Cancer 1994; 70: 677–680.[ISI][Medline]

30. Makela J, Kantola R, Tikkakoski T et al. Superselective intra-arterial chemotherapy with mitomycin C in hepatic metastases from colorectal cancer. J Surg Oncol 1997; 65: 127–131.[CrossRef][ISI][Medline]

31. Cantore M, Aitini E, Rabbi C et al. Combined intra-arterial locoregional and systemic treatment of nonresectable hepatic metastases of colorectal carcinoma. G Chir 1997; 18: 235–239.[Medline]

32. Kemeny N, Cohen A, Seiter K et al. Randomized trial of hepatic arterial floxuridine, mitomycin, and carmustine versus floxuridine alone in previously treated patients with liver metastases from colorectal cancer. J Clin Oncol 1993; 11: 330–335.[Abstract]

33. Link KH, Pillasch J, Formentini A et al. Downstaging by regional chemotherapy of non-resectable isolated colorectal liver metastases. Eur J Surg Oncol 1999; 25: 381–388.[CrossRef][ISI][Medline]

34. Focan C, Levi F, Kreutz F et al. Continuous delivery of venous 5-fluorouracil and arterial 5-fluorodeoxyuridine for hepatic metastases from colorectal cancer: feasibility and tolerance in a randomized phase II trial comparing flat versus chronomodulated infusion. Anticancer Drugs 1999; 10: 385–392.[ISI][Medline]

35. de Gramont A, Bosset JF, Milan C et al. Randomized trial comparing monthly low-dose leucovorin and fluorouracil bolus with bimonthly high-dose leucovorin and fluorouracil bolus plus continuous infusion for advanced colorectal cancer: a French Intergroup study. J Clin Oncol 1997; 15: 808–815.[Abstract]