当前位置: 首页 > 期刊 > 《临床肿瘤学》 > 2005年第12期 > 正文
编号:11332070
Phase II Trial of High-Dose Conformal Radiation Therapy With Concurrent Hepatic Artery Floxuridine for Unresectable Intrahepatic Malignancie
http://www.100md.com 《临床肿瘤学》
     the Department of Radiation Oncology, Division of Hematology Oncology, and Department of Surgery, University of Michigan, Ann Arbor, MI

    ABSTRACT

    PURPOSE: A phase II trial was conducted to determine if high-dose radiation with concurrent hepatic arterial floxuridine would improve survival in patients with unresectable intrahepatic malignancies.

    PATIENTS AND METHODS: Three-dimensional conformal high-dose radiation therapy was delivered concurrently with hepatic arterial floxuridine in 128 patients. The radiation dose was based on a normal-tissue complication probability model and subjected the patient to an estimated maximum risk of radiation-induced liver disease of 10% to 15%. The study design provided more than 80% power to detect a two-fold increase in median survival compared with historical controls at a 5% significance level.

    RESULTS: The median radiation dose delivered was 60.75 Gy (1.5-Gy fractions bid). At a median follow-up time of 16 months (26 months in patients who were alive) the median survival was 15.8 months (95% CI, 12.6 to 18.3 months), significantly longer than in the historical control. The actuarial 3-year survival was 17%. The total dose was the only significant predictor of survival. Primary hepatobiliary tumors had a significantly greater tendency to remain confined to the liver than did colorectal cancer metastases. Overall toxicity was acceptable, with 27 patients (21%) and 11 patients (9%) developing grade 3 and 4 toxicity, respectively, and one treatment-related death.

    CONCLUSION: The results suggest that, compared with historical controls, high-dose focal liver irradiation with hepatic artery floxuridine prolongs survival in patients with unresectable chemotherapy-refractory metastatic colorectal cancer and primary hepatobiliary tumors. This provides a rationale for intensification of local therapy for unresectable hepatobiliary cancers and integration of this regimen with newer systemic therapy for patients with colorectal cancer.

    INTRODUCTION

    Although surgical resection is the most effective therapy for hepatocellular carcinoma (HCC) and bile duct cancer, most patients with hepatobiliary cancer and liver-confined metastases from colorectal cancer present with unresectable disease. A number of possible therapies have been evaluated for these patients, including liver transplantation (for HCC), chemoembolization, and various ablative techniques. These therapies have important limitations. Transplantation is possible only for tumors less than 5 cm1 so relatively few patients qualify. Historical series demonstrate that chemoembolization is of minimal efficacy for colorectal cancer metastatic to the liver.2 Although a recent meta-analysis in patients with HCC suggests that chemoembolization can improve survival compared with supportive care in selected asymptomatic patients without vascular invasion,3 some randomized trials have shown no difference between chemoembolization and best supportive care.4,5 Uncontrolled retrospective reviews of ablative techniques such as radiofrequency ablation suggest that small tumors can be controlled for long periods of time and possibly eradicated, but radiofrequency ablation cannot be applied if a tumor is larger than about 5 cm in diameter or abuts important blood vessels or bile ducts.6 Thus, many patients have unresectable disease that cannot be addressed by other local therapies, and these individuals have a median overall survival in the range of 6 to 7 months.2,7

    We have enrolled such patients onto a series of clinical trials of focal liver radiation therapy (RT) with concurrent hepatic arterial floxuridine. Traditionally, RT had played a minor role in the treatment of intrahepatic cancers because of the low tolerance of the liver when the whole organ is irradiated. We initially hypothesized that, just as the surgeon could resect parts of the liver, it would be possible to deliver tumoricidal doses of radiation to intrahepatic cancers if sufficient normal liver could be spared. Such an approach would depend both on the use of conformal techniques, to minimize the volume of liver irradiated, and on determining the maximum safe dose, as a function of the volume of normal liver irradiated. The first step was to develop a normal tissue complication probability (NTCP) model that quantitatively described the relationship between dose and volumes irradiated and the probability of developing radiation-induced liver disease (RILD).8 A phase I/II trial was then conducted to test prospectively the safety of the model parameters and to begin to develop efficacy data at the maximum-tolerated dose.9 The RT dose was individualized and was based on the volume of normal liver that could be spared. We prescribed the maximal possible dose that would subject the individual to an estimated risk not exceeding 10% to 15%. In that study, we established the maximum tolerable dose (as a function of volume) and reported preliminary survival outcomes that seemed to be better than those anticipated based on historical comparisons.

    We then wished to complete this experience by treating an additional cohort of patients in a phase II trial under the conditions previously established. In addition to the primary end point of survival, we wished to determine in this phase II trial the factors that predict response and survival.

    PATIENTS AND METHODS

    Patients

    Between April 1996 and April 2003, 128 patients with unresectable intrahepatic primary hepatobiliary cancers or liver metastases from colorectal cancer were accrued to two consecutive prospective clinical trials of focal liver RT and hepatic arterial floxuridine. Both trials were approved by the institutional review board of the University of Michigan. The first, a phase I/II trial, has been described in detail previously.9 The second study (UMCC 2001-008) was a phase II trial using identical eligibility criteria and treatment regimen at the established phase II level. Briefly, patients needed to have an estimated life expectancy of at least 12 weeks and an Eastern Cooperative Oncology Group performance status of 2 or less. All patients were required to have nearly normal liver (international normalized ratio < 1.2) and renal function (creatinine < 1.5), and adequate bone marrow reserve. Patients with colorectal cancer with extrahepatic disease were eligible if progression of intrahepatic disease was believed to be the greatest short-term threat to the patient’s life. In addition, the tumor(s) had to be treatable by radiation portals that excluded at least approximately 10% of the normal liver from irradiation. More precisely, Veff (defined as the normal liver volume, which if irradiated uniformly to the isocenter dose, would be associated with the same NTCP as the nonuniform dose distribution actually delivered10) had to be 0.90.

    RT

    Computed tomography or magnetic resonance imaging was used to delineate the gross tumor volume (GTV). The clinical target volume included the GTV plus a 1-cm margin within the liver. The planning target volume (PTV) included the clinical target volume plus a 0.5-cm uniform expansion for setup uncertainty, plus an additional 0.3 to 3 cm margin in the craniocaudal dimension to account for breathing motion. The latter was based on the magnitude of excursion of the diaphragm with breathing and was determined using fluoroscopy at the time of simulation. More recently, patients were treated using active breathing control.11 Briefly, the patient breathes through a system that monitors lung volume and cuts off air flow, with patient cooperation, at a predetermined phase of the breathing cycle. Acquisition of treatment planning computed tomography images as well as treatment delivery is carried out only during this predetermined phase. Three-dimensional conformal planning was used to ensure that the PTV was covered by the 95% isodose envelope while maximally sparing normal liver.

    Radiation therapy was delivered concurrently with hepatic arterial floxuridine in two 2-week blocks (1.5-Gy fractions, twice daily with an interfraction interval of at least 6 hours Monday through Friday, and once on Saturday) separated by a 2-week break. This break was required because most patients received floxuridine via a percutaneous brachial artery catheter (see Chemotherapy). The additional radiation after the two radiochemotherapy blocks was delivered without chemotherapy. The total dose was dependent on the volume of normal liver that could be spared from irradiation. Briefly, for each patient, Veff was calculated10 and was entered into our previously described8 NTCP model to estimate the dose that would subject the patient to a maximum risk of radiation-induced liver disease of 10% to 15%. The prescribed dose to the isocenter ranged from 40 to 90 Gy (median, 60.75 Gy). The dose to the duodenum and stomach was not to exceed 68 Gy in 1.5-Gy fractions, and the maximal allowable dose to the spinal cord was 37.5 Gy in 1.5-Gy fractions. If more that half of one kidney needed to receive 20 Gy or greater, then no more than 10% of the other kidney could receive greater than 18 Gy.

    An example of a typical beam arrangement and fields used is shown in Figure 1. The resultant dose distribution and dose-volume histogram is shown in Figure 2.

    Chemotherapy

    Concurrent continuous-infusion hepatic arterial floxuridine (0.2 mg/kg/d) was delivered as described, typically through a percutaneous hepatic arterial catheter placed through the brachial artery or with a hepatic artery catheter and pump placed at time of a previous laparotomy. Hepatic artery perfusion was verified with technetium-99m–macroaggregated albumin before floxuridine delivery.

    Evaluation

    After completion of treatment, patients were assessed by physical examination every 2 weeks for 2 months, then every 2 months for 6 months, every 6 months to 24 months, and annually thereafter. Liver function, renal function, and CBCs were assessed every visit. Abdominal/pelvic computed tomography or magnetic resonance imaging, chest x-ray, and carcinoembryonic antigen or alpha-fetoprotein level were obtained at 2 months, 6 months, and every follow-up visit thereafter. In patients with measurable disease, objective response was assessed using standard bidimensional criteria. Toxicity was scored using the National Cancer Institute Common Toxicity Criteria. RILD was defined as the development of elevated liver function tests and nonmalignant ascites, in the absence of progressive disease.

    Statistical Analysis

    All analyses were performed using SAS 9.1 (SAS Institute, Cary, NC) or R 2.0.1 (R Project for Statistical Computing, http://www.r-project.org/).

    GTV. GTV was highly skewed, and logarithmically transformed for analyses. The effect of demographic characteristics (age, sex, and race/ethnicity) and disease type (cholangiocarcinoma, HCC, or metastatic colorectal cancer) on GTV was described by means of a generalized linear model.

    RT dose. The effect of demographic variables, disease type, and GTV on prescribed RT dose was explored by means of a general linear model. Delivered, as opposed to prescribed RT dose was used as a predictor in models of response and survival.

    Objective response. Patients’ best responses were categorized as complete response, partial response, stable disease, or progressive disease. Because few patients had complete response or progressive disease, these categories were merged for analysis with partial response and stable disease, respectively. Patients without measurable disease could not be evaluated for response and were omitted from analyses where response was a predictor or outcome. The influence of demographic variables, disease type, GTV, and delivered RT dose on response was assessed by means of logistic regression, as follows. Response was first modeled as a function of GTV and RT dose, with and without disease type and demographic covariates. The statistical significance of disease type and demographic variables as a group was evaluated with a likelihood ratio test. If the test was statistically significant, disease type and demographic variables were retained in a model for response that included RT dose and GTV. Significant predictors in that model were retained for the final model.

    Survival and progression-free survival. Survival and progression-free survival (PF survival) were modeled using proportional hazards (Cox) regression. The same predictor selection strategy used for objective response was applied to survival and PF survival; objective response was studied as a predictor for survival and PF survival in a separate analysis. Median survival and CIs were estimated using the product-limit (Kaplan-Meier) method. The effect of RT dose on survival and PF survival was also characterized using a penalized smoothing spline model, with the df determined by Akaike’s information criterion.12,13 Penalized smoothing splines produce a smooth, semiparametric function of predictors (in this case, RT dose), the form of which is determined by the data. This smooth function is then used in the proportional hazards model. The risk function is the component of the hazard (also known as the conditional failure density) function that changes according to the predictors (in this case, RT dose); the risk function is plotted against RT dose to demonstrate its effect.

    Test of hypothesis against historical controls. UMCC 2001-008 was powered to test the hypothesis that survival could be increased compared with historical controls, based on an exponential accelerated failure time model. Two-sided likelihood ratio tests were performed against the null hypotheses that the median survival of patients with metastatic colorectal cancer who have progressive disease after fluorouracil (FU)/leucovorin (in earlier patients) and FU/leucovorin followed by irinotecan and/or oxaliplatin (in later patients) is 9 months, that the median survival of patients with cholangiocarcinoma is 9 months, and that the median survival of patients with hepatocellular cancer is 8 months. The justification for these values is provided in the Discussion. Tests were performed for all patients and for the individual disease groups.

    RESULTS

    Patients and Treatment Characteristics

    Table 1 lists the demographic and clinical characteristics of the patients by disease type. The median age was 60.3 years (range, 26 to 80 years). Age and sex were not related to disease type, but a significantly higher proportion of nonwhites presented with HCC than whites (63% v 24%). Disease type was the only statistically significant predictor of GTV among patients with measurable disease (P < .0012); HCCs were significantly larger than colorectal cancer metastases or cholangiocarcinomas (P < .0002). GTV was the only significant predictor of RT dose (P < .0001) and the relationship between GTV and RT dose did not vary significantly among disease types (Fig 3).

    Response and Survival

    Demographic variables, disease type, GTV, and RT dose were not statistically significant predictors of response (dichotomized as progressive disease plus stable disease v partial response plus complete response) in logistic regression models. The distribution of objective response by disease type is listed in Table 2.

    At the time of analysis, 109 of 128 patients (85%) had died. The actuarial 3-year overall survival was 17% and the median survival was 15.8 months (95% CI, 12.6 to 18.3 months) from the start of RT. Median survival in HCC, cholangiocarcinoma, and metastatic colorectal cancer was 15.2, 13.3, and 17.2 months, respectively.

    The primary goal of this study was to test the hypothesis that high-dose conformal RT combined with hepatic arterial floxuridine could improve the survival of patients with intrahepatic cancer who were deemed ineligible for resection or ablation. Indeed, we found that this treatment seemed to improve survival when tested against median survivals of 9 months (metastatic colorectal cancer), 9 months (cholangiocarcinoma), and 8 months (HCC; P < .0001). Disease-specific null hypotheses were also rejected for colorectal cancer (P < .014), cholangiocarcinoma (P < .0008), and HCC (P < .0001).

    Demographic variables, disease type, and GTV were not statistically significant predictors of survival, but RT dose was a statistically significant (P < .01) predictor of survival. The median survival of the 64 patients who received less than the median dose of 60.7 Gy was 15.2 months (95% CI, 9.5 to 16.4 months), whereas the median survival of the patients who received more than 60.7 Gy was 18.4 months (95% CI, 12.9 to 22.8 months). Patients who received doses 75 Gy (upper quartile) had a significantly higher overall survival than those receiving lower doses (23.9 v 14.9 months; P < .01; Fig 4). The relationship between RT dose delivered and survival was virtually identical in the three disease types. Figure 5A demonstrates the pattern of change in the risk as RT dose increases; there is little effect of dose below 60 Gy, and then a steady increase in survival is observed as RT dose increases to 90 Gy.

    PF Survival

    The median PF survival was 12.0 months (95% CI, 9.4 to 14.3 months). Demographic variables, disease type, GTV, or RT dose (as a linear function) delivered were not significant predictors of PF survival (P > .05). However, given that survival was significantly improved in patients receiving 75 Gy or more, a similar test was performed for PF survival, and, in this test, the median PF survival of patients receiving 75 Gy or more was improved compared with that of other patients (20.7 v 10.9 months; P < .05; Fig 6). Figure 5B displays an effect of RT dose on PF survival similar to that of RT dose on survival, but less pronounced.

    Pattern of Failure

    Primary hepatobiliary tumors had a greater tendency to remain confined to the liver than did metastatic colorectal cancer (P < .0001; Table 3). The 3-year freedom from extrahepatic progression rates were 48.5% for primary hepatic cancers and 15.1% for metastatic colorectal cancer (P < .003).

    Toxicity

    The toxicity associated with protocol therapy is summarized in Table 4. There were 39 patients (30%) who developed grade 1/2 toxicity, 38 patients (30%) with grade 3/4 toxicity, and one treatment-related death. The most common severe complications were upper GI ulceration and bleeding (5%), RILD (4%), and catheter-related problems (3%).

    DISCUSSION

    In this study, we found that high-dose conformal RT with concurrent hepatic arterial floxuridine may prolong the survival of patients with unresectable chemotherapy-refractory colorectal cancer metastatic to the liver and with unresectable primary hepatobiliary cancer. The total dose was the only significant factor associated with survival. Because there was only a weak inverse relationship between dose and tumor size, this suggests that increased RT dose permitted by the use of conformal therapy and a NTCP model played an important role in the observed effect on survival. In addition, we found that patients with colorectal cancer tended to experience progression distantly, whereas those with primary hepatobiliary cancer tended to experience progression locally. This suggests that future efforts should focus on intensifying local therapy for patients with unresectable hepatobiliary cancer and on combining this form of regional therapy with newer active systemic chemotherapy for patients with metastatic colorectal cancer.

    Our results regarding primary hepatobiliary cancers that could not be resected or treated with radiofrequency ablation therapy should be placed in the context of the overall modest recent progress in the development of new therapies for these diseases. An overview of medical treatments for unresectable HCC showed no survival benefit for systemic or hepatic artery chemotherapy.14 A review of randomized controlled trials15 examining different regimens of intravenous or hepatic arterial chemotherapy (with or without embolization of the hepatic artery), hormonal, and immunotherapy regimens found some evidence of benefit for tamoxifen and transcatheter arterial embolization but concluded that all other nonsurgical treatments were ineffective. Indeed, a large recent randomized controlled trial16 showed no benefit of tamoxifen over placebo. In a large retrospective analysis of 314 patients with HCC, Stuart et al17 reported an overall median survival of 10 months but only 2 to 4 months in those who were untreated or received systemic chemotherapy alone. Other chemotherapy trials using doxorubicin or gemcitabine have reported median survivals of less than 4 months and 5 to 8 months, respectively.18 Although there has been intermittent exploration of yttrium-90 microspheres in the treatment of HCC,19 the lack of prospective trials to assess efficacy systematically makes it difficult to define their role.

    For patients with cholangiocarcinoma, an extensive review of 65 clinical trials20 failed to document benefit of any nonsurgical therapy. Our own review of recent chemotherapy trials for patients with advanced cholangiocarcinoma suggests a median survival of approximately 9 months.21-35 Against this background, the median survival of 15.2 months in our patients with HCC and 13.3 months in patients with cholangiocarcinoma seems promising.

    The results reported here are in agreement with the more tentative outcomes reported by Dawson et al in 2000.9 The median survivals of 17.2 and 15.8 months, in all patients and in patients with metastatic colorectal cancer, respectively, are remarkably similar to the median survivals of 18 and 16 months reported previously. As before, RT dose continues to be a strong predictor of survival, and the only predictor on multivariate analysis.

    Evidence of benefit from high-dose RT in patients with unresectable intrahepatic malignancies has been provided by other groups as well. Mohiuddin et al36 reported higher rates and longer duration of palliation as well as improved survival in patients with liver metastases from colorectal cancer treated with a boost to the tumor (33 to 60 Gy), compared with those treated with whole-liver RT alone. Park et al37 conducted a RT dose-escalation study in patients with HCC. Patients were excluded if they had extrahepatic metastasis, liver cirrhosis of Child class C, tumors larger than two thirds of the liver, and an Eastern Cooperative Oncology Group performance status of greater than 3. The mean RT dose was 48.2 Gy in 1.8-Gy fractions. An objective response was observed in 106 (67.1%) of 158 patients. Similar to our findings, they reported that the total dose was the most significant factor associated with tumor response. The response rates in patients treated with doses less than 40, 40 to 50, and more than 50 Gy were 29.2%, 68.6%, and 77.1%, respectively. Overall survival rates were 41.8% and 19.9% at 1 and 2 years, respectively, and the median survival was 10 months. Others have reported similar results.38-40 However, our trial differs in its design; we use an NTCP model to maximize dose on an individual basis; this permits, in turn, the delivery of higher and potentially more effective doses of RT safely.

    Although our trial has focused on unresectable tumors that could not be treated by radiofrequency ablation, it is possible that high-dose RT may be appropriate for smaller tumors commonly treated at present with radiofrequency ablation. A study by Herfarth et al41 demonstrated that a single fraction of 14 to 26 Gy can control up to two thirds of small lesions (median volume, 10 mL) for longer than 18 months. These results suggest that the efficacy of high-dose RT may approach that of radiofrequency ablation for small liver lesions. Additional study of this approach is currently planned by the Radiation Therapy Oncology Group.

    Our analysis of the pattern of failure demonstrated an important difference between patients with metastatic colorectal cancer and those with primary hepatobiliary malignancies. The former tended to experience treatment failure outside of the liver more often than the latter. This observation is in agreement with surgical series reporting that approximately half of all treatment failures after resection of liver metastases are extrahepatic.42 Thus, whereas intensification of liver-directed therapy alone might improve the outcome of patients with primary hepatobiliary cancer, intensification alone seems unlikely to have as large an impact on patients with colorectal cancer metastases. We would hypothesize that greater gains could be realized for patients with metastatic colorectal cancer by combining high-dose RT with more effective systemic therapy.

    Although this trial has accrued patients with colorectal cancer who have experienced progression during chemotherapy, a limitation of this study is that the definition of chemotherapy refractoriness has changed with the sequential introduction of irinotecan, oxaliplatin, and more recently, cetuximab. However, evidence suggests that the survival of patients receiving second-line chemotherapy as well as those experiencing progression after second-line therapy has remained relatively constant during this time period. For instance, at the time this trial was initiated, the expected survival of patients refractory to FU receiving second-line chemotherapy was 9 to 12 months.43-45 More recently, it was reported that the median survival for chemotherapy-na?ve patients treated with folinic acid, fluorouracil, and irinotecan, or FOLFOX6 (folinic acid, fluorouracil, and oxaliplatin) was approximately 21 months.46 The fact that the median PF survival in this trial was approximately 9 months is consistent with the estimate that the median survival of patients receiving second-line chemotherapy at present is approximately 12 months. Likewise, the survival of patients experiencing progression after second-line therapy is approximately 6 months.43,45 Our study population consisted of a mixture of patients who had experienced progression after first-line chemotherapy or were chemotherapy refractory, and who had lesions that were unresectable and not eligible for local ablative techniques. In this context, the achieved median survival of 17.2 months in these patients is encouraging.

    The regimen described in this article contains elements, such as twice daily fractionation and hepatic artery infusion of floxuridine, which may not be easily transferred to many clinics. The relative contribution of these elements to the overall efficacy of the regimen is not clear. Twice-daily fractionation was incorporated into the protocol design to maximize the proportion of the total RT dose that would be delivered concurrently with floxuridine and to allow maximal opportunity for interaction between the two modalities. However, as newer delivery methods increase the safety of large-dose fractions, the practical appeal and potential theoretical advantages of shorter-duration courses and hypofractionation increase. In contrast, evidence suggests that in patients with metastatic colorectal cancer, hepatic arterial infusion of chemotherapy is associated with higher response rates47,48 and possibly survival,49,50 when compared with systemic chemotherapy. Because of a high first-pass liver extraction, levels of floxuridine in the systemic circulation are low, resulting in little radiosensitization of normal structures51; this likely improves the therapeutic ratio.

    The regimen was associated with a 9% probability of grade 4 toxicity and one treatment-related fatality. With a median potential follow-up of 26 months, we believe this represents a robust estimate of toxicity derived from a mature data set, which is not likely to change significantly. Although these results are not out of line with toxicity reported in other high-intensity regimens, we currently are investigating technical improvements in treatment delivery that might reduce these rates. Our initial treatment planning studies suggest that intensity-modulated RT, optimized using our NTCP model, can permit us to increase the tumor dose by up to 10 Gy while maintaining or reducing the dose to normal structures.52

    Authors' Disclosures of Potential Conflicts of Interest

    The authors indicated no potential conflicts of interest.

    NOTES

    Supported by Grants No. CA85684 and MO1RR00042 from the National Institutes of Health, Bethesda, MD.

    Authors' disclosures of potential conflicts of interest are found at the end of this article.

    REFERENCES

    Marsh JW, Geller DA, Finkelstein SD, et al: Role of liver transplantation for hepatobiliary malignant disorders. Lancet Oncol 5:480-488, 2004

    Alexander HR, Allegra CJ, Lawrence TS: Metastatic liver cancer, in Rosenberg SA (ed): Cancer Principles and Practice of Oncology (ed 6). Philadelphia PA, Lippincott Williams & Wilkins, 2001, pp 2690-2709

    Llovet JM, Burroughs A, Bruix J: Hepatocellular carcinoma. Lancet 362:1907-1917, 2003

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

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

    Abdalla EK, Vauthey JN, Ellis LM, et al: Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 239:818-827, 2004

    Venook AP: Treatment of hepatocellular carcinoma: Too many options. J Clin Oncol 12:1323-1334, 1994

    Dawson LA, Normolle D, Balter JM, et al: Analysis of radiation induced liver disease using the Lyman NTCP model. Int J Radiat Oncol Biol Phys 53:810-821, 2002

    Dawson LA, McGinn CJ, Normolle D, et al: Escalated focal liver radiation and hepatic artery floxuridine for unresectable liver malignancies. J Clin Oncol 18:2210-2218, 2000

    Kutcher GJ, Burman C: Calculation of complication probability factors for non-uniform normal tissue irradiation: The effective volume method. Int J Radiat Oncol Biol Phys 16:1623-1630, 1989

    Balter JM, Brock KK, Litzenberg DW, et al: Daily targeting of intrahepatic tumors for radiotherapy. Int J Radiat Oncol Biol Phys 52:266-271, 2002

    Lindstrom M: Penalized estimation of free-knot splines. J Comput Graph Stat 8:333-352, 1999

    Therneau T, Grambsch P: Modeling survival data: Extending the Cox model, in Statistics for Biology and Health. New York, NY, Springer-Verlag, 2000, pp 124-126

    Mathurin P, Rixe O, Carbonell N, et al: Review article: Overview of medical treatments in unresectable hepatocellular carcinoma–an impossible meta-analysis? Aliment Pharmacol Ther 12:111-126, 1998

    Simonetti RG, Liberati A, Angiolini C, et al: Treatment of hepatocellular carcinoma: A systematic review of randomized controlled trials. Ann Oncol 8:117-136, 1997

    Chow PK, Tai BC, Tan CK, et al: High-dose tamoxifen in the treatment of inoperable hepatocellular carcinoma: A multicenter randomized controlled trial. Hepatology 36:1221-1226, 2002

    Stuart KE, Anand AJ, Jenkins RL: Hepatocellular carcinoma in the United States: Prognostic features, treatment outcome, and survival. Cancer 77:2217-2222, 1996

    Fuchs CS, Clark JW, Ryan DP, et al: A phase II trial of gemcitabine in patients with advanced hepatocellular carcinoma. Cancer 94:3186-3191, 2002

    Salem R, Lewandowski R, Roberts C, et al: Use of yttrium-90 glass microspheres (TheraSphere) for the treatment of unresectable hepatocellular carcinoma in patients with portal vein thrombosis. J Vasc Interv Radiol 15:335-345, 2004

    Hejna M, Pruckmayer M, Raderer M: The role of chemotherapy and radiation in the management of biliary cancer: A review of the literature. Eur J Cancer 34:977-986, 1998

    Alberts SR, Al-Khatib H, Mahoney MR, et al: Gemcitabine, 5-fluorouracil, and leucovorin in advanced biliary tract and gallbladder carcinoma: A North Central Cancer Treatment Group phase II trial. Cancer 103:111-118, 2005

    Andre T, Tournigand C, Rosmorduc O, et al: Gemcitabine combined with oxaliplatin (GEMOX) in advanced biliary tract adenocarcinoma: A GERCOR study. Ann Oncol 15:1339-1343, 2004

    Ueno H, Okusaka T, Ikeda M, et al: Phase II study of S-1 in patients with advanced biliary tract cancer. Br J Cancer 91:1769-1774, 2004

    Raderer M, Hejna MH, Valencak JB, et al: Two consecutive phase II studies of 5-fluorouracil/leucovorin/mitomycin C and of gemcitabine in patients with advanced biliary cancer. Oncology 56:177-180, 1999

    Colleoni M, Di Bartolomeo M, Di Leo A, et al: Oral chemotherapy with doxifluridine and folinic acid in biliary tract cancer. Eur J Cancer 31A:2426-2427, 1995

    Choi CW, Choi IK, Seo JH, et al: Effects of 5-fluorouracil and leucovorin in the treatment of pancreatic-biliary tract adenocarcinomas. Am J Clin Oncol 23:425-428, 2000

    Patt YZ, Jones DV Jr, Hoque A, et al: Phase II trial of intravenous fluorouracil and subcutaneous interferon alfa-2b for biliary tract cancer. J Clin Oncol 14:2311-2315, 1996

    Patt YZ, Hassan MM, Aguayo A, et al: Oral capecitabine for the treatment of hepatocellular carcinoma, cholangiocarcinoma, and gallbladder carcinoma. Cancer 101:578-586, 2004

    Knox JJ, Hedley D, Oza A, et al: Gemcitabine concurrent with continuous infusional 5-fluorouracil in advanced biliary cancers: A review of the Princess Margaret Hospital experience. Ann Oncol 15:770-774, 2004

    Tsavaris N, Kosmas C, Gouveris P, et al: Weekly gemcitabine for the treatment of biliary tract and gallbladder cancer. Invest New Drugs 22:193-198, 2004

    Gebbia V, Giuliani F, Maiello E, et al: Treatment of inoperable and/or metastatic biliary tree carcinomas with single-agent gemcitabine or in combination with levofolinic acid and infusional fluorouracil: Results of a multicenter phase II study. J Clin Oncol 19:4089-4091, 2001

    Penz M, Kornek GV, Raderer M, et al: Phase II trial of two-weekly gemcitabine in patients with advanced biliary tract cancer. Ann Oncol 12:183-186, 2001

    Kuhn R, Hribaschek A, Eichelmann K, et al: Outpatient therapy with gemcitabine and docetaxel for gallbladder, biliary, and cholangio-carcinomas. Invest New Drugs 20:351-356, 2002

    Papakostas P, Kouroussis C, Androulakis N, et al: First-line chemotherapy with docetaxel for unresectable or metastatic carcinoma of the biliary tract: A multicentre phase II study. Eur J Cancer 37:1833-1838, 2001

    Kubicka S, Rudolph KL, Tietze MK, et al: Phase II study of systemic gemcitabine chemotherapy for advanced unresectable hepatobiliary carcinomas. Hepatogastroenterology 48:783-789, 2001

    Mohiuddin M, Chen E, Ahmad N: Combined liver radiation and chemotherapy for palliation of hepatic metastases from colorectal cancer. J Clin Oncol 14:722-728, 1996

    Park HC, Seong J, Han KH, et al: Dose-response relationship in local radiotherapy for hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 54:150-155, 2002

    Cheng JC, Chuang VP, Cheng SH, et al: Local radiotherapy with or without transcatheter arterial chemoembolization for patients with unresectable hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 47:435-442, 2000

    Li B, Yu J, Wang L, et al: Study of local three-dimensional conformal radiotherapy combined with transcatheter arterial chemoembolization for patients with stage III hepatocellular carcinoma. Am J Clin Oncol 26:e92-e99, 2003

    Yamada K, Izaki K, Sugimoto K, et al: Prospective trial of combined transcatheter arterial chemoembolization and three-dimensional conformal radiotherapy for portal vein tumor thrombus in patients with unresectable hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 57:113-119, 2003

    Herfarth KK, Debus J, Lohr F, et al: Stereotactic single-dose radiation therapy of liver tumors: Results of a phase I/II trial. J Clin Oncol 19:164-170, 2001

    Saltz LB, Ahmad SA, Vauthey J: Colorectal cancer: Management of advanced disease, in Kelsen DP, Daly JM, Kern SE, (eds): Gastrointestinal Oncology: Principles and Practice. Philadelphia, PA, Lippincott Williams & Wilkins, 2002, pp 825-852

    Cunningham D, Pyrhonen S, James RD, et al: Randomised trial of irinotecan plus supportive care versus supportive care alone after fluorouracil failure for patients with metastatic colorectal cancer. Lancet 352:1413-1418, 1998

    Rougier P, Van Cutsem E, Bajetta E, et al: Randomised trial of irinotecan versus fluorouracil by continuous infusion after fluorouracil failure in patients with metastatic colorectal cancer. Lancet 352:1407-1412, 1998

    Rougier P, Lepille D, Bennouna J, et al: Antitumour activity of three second-line treatment combinations in patients with metastatic colorectal cancer after optimal 5-FU regimen failure: A randomised, multicentre phase II study. Ann Oncol 13:1558-1567, 2002

    Tournigand C, Andre T, Achille E, et al: FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: A randomized GERCOR study. J Clin Oncol 22:229-237, 2004

    Harmantas A, Rotstein LE, Langer B: Regional versus systemic chemotherapy in the treatment of colorectal carcinoma metastatic to the liver: Is there a survival difference? Meta-analysis of the published literature. Cancer 78:1639-1645, 1996

    Reappraisal of hepatic arterial infusion in the treatment of nonresectable liver metastases from colorectal cancer: Meta-Analysis Group in Cancer. J Natl Cancer Inst 88:252-258, 1996

    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 341:2039-2048, 1999

    Kemeny NE, Niedzwiecki D, Hollis DR: Hepatic arterial infusion (HAI) versus systemic therapy for hepatic metastases from colorectal cancer: A CALGB randomized trial of efficacy, quality of life (QOL), cost effectiveness, and molecular markers. Proc Am Soc Clin Oncol 22:252, 2003 (abstr 1010)

    Robertson JM, Lawrence TS, Andrews JC, et al: Long term results of hepatic artery fluorodeoxyuridine and conformal radiation therapy for primary hepatobiliary cancers. Int J Radiat Oncol Biol Phys 37:325, 1997

    Thomas EE, Ten Haken RK, Chapet O, et al: The use of biological parameters (EUD and NTCP) in IMRT optimization and its impact on the maximum safe dose deliverable to intrahepatic tumors. Int J Radiat Oncol Biol Phys 60:S179, 2004 (abstr 82)(Edgar Ben-Josef, Daniel N)