Covered Versus Uncovered Self-conformable Metal Stent for Palliation of Primary Malignant Extra-hepatic Biliary Strictures: a Randomized Multicenter Study
Covered Versus Uncovered Self-conformable Metal Stent for Palliation of Primary Malignant Extra-hepatic Biliary Strictures: a Randomized Multicenter Study
The purpose of this study is to compare the duration of stent patency of a covered vs. an uncovered biliary self-expandable metal stents (SEMS) placed to relieve biliary obstruction in patients with inoperable extrahepatic malignant biliary obstruction.
Cancer of the pancreas, gallbladder, or bile ducts is the most common cause of malignant obstruction of the biliary tree. Patients who have unresectable tumors have a dismal prognosis in terms of survival and quality of life. In these cases 5-year survival is less than 2% and palliation, such as the establishment of a biliary drainage, is the only treatment available. Two types of stents are routinely used: plastic stents (PS) and self-expandable metal stents (SEMS). The first generation SEMS are uncovered and recurrent obstruction, most frequently caused by tumor ingrowth through the metal mesh, is seen in 16-46%. Recently, covered SEMS have been introduced to prevent tumor ingrowth. Covered SEMS are associated with stent occlusion in 14% of patients. As can be expected, the most frequent cause of stent obstruction in these patients is sludge formation. Stent migration, and cholecystitis and pancreatitis caused by obstruction of the cystic duct and pancreatic duct, respectively, have been suggested to occur more frequently with covered SEMS. To date, however, one randomized trial and three comparative studies compared covered with uncovered SEMS, have found only a non statistically significant trend towards more frequent occurrence of these complications.
From these initial studies comparing uncovered to covered SEMS, it suggested that stent patency may be longer with covered SEMS. However, supporting evidence for the superior efficacy of covered SEMS is lacking. In addition, the issue of safety of covered SEMS, as well as the real world effectiveness of the self conformable SEMS, warrant further investigation.
In this study, the Investigators will include patients with symptoms (jaundice, cholangitis) due to malignant extrahepatic biliary tree obstruction (pancreatic cancer, cholangiocarcinoma, gallbladder cancer, or metastatic lymphadenopathy) who are not candidates for surgical cure either because the tumor is inoperable or because of the patient's poor medical condition due to comorbidities and/or advanced age.
Patients with extrahepatic malignancy in whom a diagnostic work up is still ongoing to establish the possibility of performing a curative approach will not be immediately enrolled. Patients who have been previously treated with a plastic stent will be eligible if the plastic stent was placed within the 4 weeks prior to enrolment in this study.
The purpose of this study is to compare the duration of stent patency of a covered vs. an uncovered biliary self-expandable metal stents (SEMS) placed to relieve biliary obstruction in patients with inoperable extrahepatic malignant biliary obstruction.
1. Primary Aim: To compare the duration of stent patency of a covered vs. an uncovered biliary SEMS placed to relieve biliary obstruction in patients with inoperable extrahepatic malignant biliary obstruction.
Secondary Aims:
In patients with inoperable extrahepatic malignant biliary obstruction managed with SEMS:
1.1 Primary endpoints
It will be defined as an occurrence of the following items:
1.2 Secondary endpoints
Patients with extrahepatic malignancy in whom a diagnostic work up is still ongoing to establish the possibility of performing a curative approach will not be immediately enrolled. Patients who have been previously treated with a plastic stent will be eligible if the plastic stent was placed within the 4 weeks prior to enrolment in this study.
Materials Fully covered SEMS: Niti-S Biliary ComVi Stent; Uncovered SEMS: Niti-S (D type) stent Sample size calculation: The primary end point of the study is stent occlusion. The number of patients in each group required to demonstrate a statistically significant difference in SEMS patency with an 80% power is 70 for a 22% difference,121 for a 17% difference, and 248 for a 12% difference in the obstruction rate between the two groups. Estimated sample size is 121 for a 17% difference, and 70 for a 22% difference.
With a lower (75%) power, 63 patients per treatment group are required to detect a difference of 22%, 108 for a 17% difference and 222 for a 12% difference. This computation is based on data on obstruction percentage reported in literature.
The target enrolment for this study will be 70 patients per study arm. Considering time to occlusion analysis, a total of 140 patients will detect a treatment difference at a two sided 0.05 significance level, with 80% power, if the true hazard ratio is at least 1.76.
Treatment of data Data storage, management, and analysis will be centralized. An electronic database will be constructed to collect the data. The program will be distributed to all participating centers and the data will be entered at the time of the encounters with the subjects such as at the time of endoscopy for stent placement, follow-up visit, or follow-up telephone call. Randomization assignment (stent type) will be coded. Standard operating procedures for regularly backing up the data will be employed at each facility and centrally.
Every 6 months the compact disk (CD) with the study site data will be sent to the coordinating center where the data manger will download the data and merge it with the previously collected study data.
Security measures will be adopted before the mailing of the CD in order to avoid any possible disclosure of the privacy: the data will be encrypted and transformed in numbers.
The data manager will remove information regarding the type of SEMS used prior to sending the data to the statistician. Therefore, data analysis will be performed by a statistician who is blinded to the type of stent.
Data sheet
Clinical data (Form A):
patient demographics (gender, age)
medical history related to diagnosis and history
concomitant medications and treatments
endoscopic and/or MRI and/or CT-scan exam for confirmation and location of stent
stenosis (the endoscopic examination could be done immediately before the stent placement procedure)
liver function tests
1 week, 1 month, 3 and 6 months after placement (Form B):
liver function tests
concomitant medications and treatment
confirmation of stent position via supine X-ray
Complications 1. week, 1 month, 3 and 6 months after placement (Form E): HRQL questionnaires Analysis Descriptive statistics, including graphical displays, will be used to summarize all study variables. The unit of analysis will be the patient. For continuous variables, means, medians, standard deviations, percentiles, ranges, box plots and histograms will be generated. For categorical variables, frequencies and proportions will be generated. The investigators will examine all variables to determine if parametric distributional assumptions (e.g. normality for the continuous variables) are valid.
Differences between continuous variables will be determined by parametric tests, or, when appropriate by non-parametric tests. Differing frequencies of variables at different times within each group (dysphagia score, body weight, etc) will be compared with tests for related samples.
To address the primary aim, differences in duration of stent patency, the Kaplan-Meier method will be used to estimate stent patency in each group and the log-rank test will be used for an unadjusted comparison between groups. Then a Cox proportional hazard model will be constructed to compare time to stent occlusion adjusted for important potential confounders. Stent patency will be calculated in days and will represent the interval between the time of stent insertion and the time of its replacement or the death of the patient with concomitant cholangitis.
To address the secondary aims, relationships between complication rates and stent type will be examined by the chi-square or the exact Fisher tests. Logistic regression will be used to compare stent complication rates adjusted for important potential confounders. Health-related quality of life (HRQL) will be evaluated by a paired t-test to determine the impact of stent placement (i.e. compare baseline HRQL and month 3 HRQL) by Student's t -test to compare the differences in HRQL at baseline and 3 months between study groups. Linear regression models will be constructed to assess HQRL while adjusting for factors other than stent type. Total direct costs for each study group will be compared and cost effectiveness modelled.
For all analyses the Statistical Package for Social Sciences software (SPSS, Inc. for Windows will be used.
Adverse events are defined as any undesirable experience occurring to a subject during a clinical trial. All adverse events reported spontaneously by the subject or observed by the investigator or his staff will be recorded.
A serious adverse event is any untoward medical occurrence or effect that at any level results in death:
Inclusion Criteria:
Exclusion Criteria: