Transarterial (chemo)embolisation for liver metastases
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To study the beneficial and harmful effects of transarterial (chemo)embolisation compared with no intervention or placebo intervention in patients with liver metastases.
Background
Description of the condition
Primary liver tumours and liver metastases from colorectal carcinoma are the two most common malignant tumours to affect the liver (Lau 2000; Michel 2002). Primary liver tumours arise from malignant cells within the liver, and hepatocellular carcinoma represents the most common form of primary liver cancer (Lau 2000; Michel 2002). Metastatic liver disease is more common than primary liver cancer and develops when malignant cells migrate from other organs to the liver (Bilchik 2000; McCarter 2000). The liver is second only to the lymph nodes as the most common site for metastatic disease (Weiss 1986). More than half of the patients with metastatic liver disease will die from metastatic complications (Wood 1976; Markovic 1998). The most common primary sites for liver metastases are lung, breast, colon and rectum, and uterus. On pre‐operative imaging, liver metastases are found in 35% patients with colorectal cancer, and 8% to 30% of the remaining colorectal cancer patients will subsequently be found to have liver involvement. Almost half of patients dying from stomach, pancreas, or breast cancer are found to have liver metastases at autopsy while in patients with endometrial cancer it occurs in about 40% of patients (Hugh 1997). Colorectal carcinoma is third leading cancer in the United States and third in cancer‐related deaths. Approximately 142,570 new patients with large bowel cancer are diagnosed each year in the United States, of which 102,900 are colon and the remainder rectal cancers. Annually, approximately 51,370 Americans die of colorectal cancer, accounting for approximately 9% of all cancer deaths (Jemal 2010). Globally, age‐adjusted annual incidence rate for colorectal cancer is 178.3 per 100,000 (IARC 2008). The highest incidence is observed in North America (age adjusted 30.1 per 100,000), Australia and New Zealand (age adjusted 39.0), northern Europe (age adjusted 30.5), and western Europe (age adjusted 33.1). Lower incidences are observed in Africa (age adjusted 5.9) and Asia (age adjusted 12.9). Globally, age‐adjusted mortality for colorectal cancer is 8.2 per 100,000, and it is higher in the countries with higher incidence, and it is lower in the countries with lower incidence. In the United States, five‐year survival after the diagnosis of colorectal cancer is 66.6% (National Cancer Institute 2009). In all developed countries analysed together, the estimated survival is 55% (Parkin 2002) with the lowest survival reported for Eastern Europe (35% male and 36% female), while in developing countries analysed together it is 39% (Parkin 2002) with the lowest survival reported for Sub‐Saharan Africa (13% for male and 14% for female). Approximately 50% of colorectal cancer patients will develop recurrence within five years of initial diagnosis with the liver being the most common site for metastatic disease (Geoghegan 1999).
Globally, age‐standardised incidence and mortality rates for lung cancer are 23.0 and 19.4 per 100,000 people of both sexes, stomach 14.1 and 10.3, pancreas 3.9 and 3.7, breast 39.0 and 12.5 per 100,000 women, and corpus uteri 8.2 and 2.0 (IARC 2008). In United States five‐year survival after the diagnosis of lung cancer is 16.4%, stomach 26.7%, pancreas 5.7%, breast cancer 89.9%, and corpus uteri cancer 84.1% (National Cancer Institute 2009). In all developed countries analysed together estimated annual survival after the diagnosis of lung cancer is 13% in men and 20% in women; the estimated survival for stomach cancer is 35% in men and 31% in women; for breast cancer it is 75%; and for the cancer of corpus uteri it is 82% (Parkin 2002). In all developing countries analysed together, estimated survival after the diagnosis of lung cancer is 12% in men and women, estimated survival for stomach cancer is 21% in men and 20% in women, for breast cancer it is 57%, and for the cancer of corpus uteri it is 67% (Parkin 2002).
For many patients, the progressive involvement of the liver is the primary determinant of long‐term survival. Surgical resection is the only curative option for patients with malignant liver neoplasm with median survival times of 21 to 46 months or five‐year survival of 20% to 58% (McLoughlin 2006). However, only 20% of patients with hepatic tumours are candidates for resection as their metastases have spread too extensively in the liver (Bilchik 2000; Bipat 2007). Options for patients with unresectable liver metastases include chemotherapy delivered intra‐arterially (5‐fluorouracil), called 'regional chemotherapy', or systemic chemotherapy (5‐fluorouracil, irinotecan, oxaliplatin, leucovorin, capecitabine) or monoclonal antibodies (such as bevacizumab or cetuximab). Other methods include local tumour ablative techniques, such as transarterial (chemo)embolisation, percutaneous ethanol injection, microwave coagulation, laser‐induced thermotherapy, radiofrequency ablation, or cryosurgical ablation.
Description of the intervention
Chemoembolisation is defined as a selective administration of chemotherapy usually combined with embolisation of the vascular supply to the tumour (Vogl 2009). This treatment results in selective ischaemic and chemotherapeutic effects on liver metastases (Vogl 2007). Chemoembolisation is based on the concept that the blood supply to hepatic tumours originates predominantly from the hepatic artery (Breedis 1954; Vogl 2003). Therefore, embolisation of the hepatic artery can lead to selective necrosis of the liver tumour while it leaves normal parenchyma virtually unaffected (Jaeger 1996; Vogl 2003).
How the intervention might work
The selective administration of the drugs may prevent extensive liver parenchyma damages. Portal vein thrombosis, high grade liver dysfunction, and hepatorenal syndrome are common contraindications for TAE/TACE. In hepatocellular carcinoma TAE/TACE may reduce tumour growth, but randomised trials and meta‐analyses assessing survival have found no significant effect on mortality (Oliveri 2011).
Why it is important to do this review
In patients with liver metastases, local or regional treatment methods can provide local control, but it is uncertain what the long‐term outcomes of some of these therapies are. Systematic reviews may help to establish the effectiveness and the trade off between the benefit and harm associated with different non‐surgical ablation methods for the treatment of all forms of malignant liver tumours (primary and metastases). Reviews published so far focus mostly on primary liver tumours or colorectal cancer liver metastases and include studies up to April 2006 (Llovet 2003; Decadt 2004; ASERNIP‐S 2006; Lopez 2006; Sutherland 2006). The methods used in these reviews have lacked clarity on how the risk of systematic errors (bias) and the risks of random errors (play of chance) have been addressed (Oliveri 2011). Therefore, a systematic review dealing with all types of malignant liver metastases is warranted.
Objectives
To study the beneficial and harmful effects of transarterial (chemo)embolisation compared with no intervention or placebo intervention in patients with liver metastases.
Methods
Criteria for considering studies for this review
Types of studies
We will include all randomised clinical trials assessing beneficial and harmful effects of transarterial (chemo)embolisation, irrespective of publication status, language, or blinding. Quasi‐randomised and observational studies that will come up with the search, will be considered only for the report of data on harm.
Types of participants
Patients with liver metastases no matter the location of the primary tumour.
Types of interventions
Transarterial (chemo)embolisation compared with no intervention, or placebo intervention. Co‐interventions will be allowed if provided equally to the experimental and control groups of the individual randomised trial.
Types of outcome measures
Primary outcomes
1. Mortality at last follow‐up.
2. Time to mortality.
3. All adverse events and complications in separate and in total. The International Conference on Harmonisation (ICH) Guidelines (ICH‐GCP 1997) defines adverse events as serious and non‐serious. A serious fatal or nonfatal adverse event is any event that leads to death, is life‐threatening, requires in‐patient hospitalisation or prolongation of existing hospitalisation, results in persistent or significant disability, and any important medical event which may have jeopardised the patient or requires intervention to prevent it. All other adverse events will be considered non‐serious.
4. Quality of life.
Secondary outcomes
1. Failure or proportion of patients with recurrence.
2. Time to progression of liver metastasis.
3. Tumour response measures (complete response, partial response, stable disease, disease progression).
Search methods for identification of studies
Electronic searches
We will search the Cochrane Hepato‐Biliary Group Controlled Trials Register (Gluud 2011), the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, MEDLINE, EMBASE, Science Citation Index Expanded, LILACS, and CINAHL (Royle 2003) as well as the WHO international trial registry platform (WHO 2011).
One global search will be used for all non‐surgical ablation methods for primary malignant liver tumours and liver metastases. Preliminary search strategies with the expected time span of the searches are given in Appendix 1. The searches will be improved at the review stage, if necessary.
In addition, we will assess for inclusion all Food and Drug Administration (FDA) approvals and investigational device exemptions as found on the FDA web site (FDA 2011).
Searching other resources
We will search reference lists of reviews (such as Schwartz 2004 and Lopez 2006), Health Technology Assessment (HTA) reports (such as ASERNIP‐S 2006), all Cochrane reviews, and all trials included for relevant studies.
Data collection and analysis
Selection of studies
Two authors will independently evaluate titles and abstracts for ordering papers (RR and MB). Any differences in opinion will be resolved by discussion, or, if necessary, by consulting a third author (JK). For titles and abstracts that potentially fit our inclusion criteria, full papers will be ordered. These papers will be assessed by two independent authors (RR and MB), and differences in opinion will be resolved using the above mentioned procedure.
Data extraction and management
We will extract the relevant information on participant characteristics, interventions, study outcome measures, and data on the outcome measures for our review, as well as information on the design and methodology of the trials. Quality assessment of the trials, fulfilling the inclusion criteria, and data extraction from the retrieved for final evaluation trials will be done by one author (RR, MB, or RW) and checked by a second author (RR, MB, or RW).
Assessment of risk of bias in included studies
We will assess the methodological quality of the trials, and hence risk of bias, based on the domains described below (Schulz 1995; Moher 1998; Kjaergard 2001; Gluud 2008; Wood 2008). This assessment will be presented by trial and will be used to describe the results of each trial in relation to their reliability. Weak points will be stressed, and recommendations for further research will be presented with reference to the shortcomings of existing trials. Due to the limited number of existing trials, we expect that meta‐regression analyses using individual quality criteria will not be feasible.
Allocation sequence generation
‐ Low risk of bias: sequence generation was achieved using computer random number generation or a random number table. Drawing lots, tossing a coin, shuffling cards and throwing dice are adequate if performed by an independent adjudicator.
‐ Uncertain risk of bias: the trial is described as randomised, but the method of sequence generation was not specified.
‐ High risk of bias: the sequence generation method is not, or may not be, random. Quasi‐randomised studies, those using dates, names, or admittance numbers in order to allocate patients are inadequate and will be excluded for the assessment of benefits but not for harms.
Allocation concealment
‐ Low risk of bias: allocation was controlled by a central and independent randomisation unit, sequentially numbered, opaque and sealed envelopes or similar, so that intervention allocations could not have been foreseen in advance of, or during, enrolment.
‐ Uncertain risk of bias: the trial was described as randomised but the method used to conceal the allocation was not described, so that intervention allocations may have been foreseen in advance of, or during, enrolment.
‐ High risk of bias: if the allocation sequence was known to the investigators who assigned participants or if the study was quasi‐randomised. Quasi‐randomised studies will be excluded for the assessment of benefits but not for harms.
Blinding of participants, personnel, and outcome assessors
‐ Low risk of bias (blinding was performed adequately, or the outcome measurement is not likely to be influenced by lack of blinding).
‐ Uncertain risk of bias (there is insufficient information to assess whether the type of blinding used is likely to induce bias on the estimate of effect).
‐ High risk of bias (no blinding or incomplete blinding, and the outcome or the outcome measurement is likely to be influenced by lack of blinding).
Incomplete outcome data
‐ Low risk of bias (the underlying reasons for missingness are unlikely to make treatment effects departure from plausible values, or proper methods have been employed to handle missing data).
‐ Uncertain risk of bias (there is insufficient information to assess whether the missing data mechanism in combination with the method used to handle missing data is likely to induce bias on the estimate of effect).
‐ High risk of bias (the crude estimate of effects (eg, complete case estimate) will clearly be biased due to the underlying reasons for missingness, and the methods used to handle missing data are unsatisfactory).
Selective outcome reporting
‐ Low risk of bias: pre‐defined, or clinically relevant and reasonably expected outcomes are reported on.
‐ Uncertain risk of bias: not all pre‐defined, or clinically relevant and reasonably expected outcomes are reported on or are not reported fully, or it is unclear whether data on these outcomes were recorded or not.
‐ High risk of bias: one or more clinically relevant and reasonably expected outcomes were not reported on; data on these outcomes were likely to have been recorded.
Other sources of bias
‐ Low risk of other bias; the trial appears to be free of other components that could put it at risk of bias.
‐ Uncertain risk of other bias; the trial may or may not be free of other components that could put it at risk of bias.
‐ High risk of other bias; there are other factors in the trial that could put it at risk of bias, eg, for‐profit involvement, authors have conducted trials on the same topic etc.
Trials judged as having 'low risk of bias' in all of the above specified individual domains will be considered 'trials with low risk of bias'.
For an extra subgroup analysis we will consider a trial with a lower risk of bias if three or more domain items are met, including sequence generation and allocation concealment. The trials will be described as having 'high risk of bias' if none of the domain items are met, or one or two items were met not including sequence generation or allocation concealment.
Measures of treatment effect
For dichotomous variables, we plan to calculate the relative risk (RR) with 95% confidence interval. For continuous variables, we plan to calculate the standardised mean difference (SMD) (for outcomes such as quality of life when different scales could be used) with 95% confidence interval (CI). For outcomes such as hazard ratio for death, we plan to use the generic inverse variance method for the meta‐analysis.
Unit of analysis issues
The number of the randomised participants will be used to calculate estimates of intervention effects and confidence intervals. In cluster randomised trials, the unit of analysis will be the cluster. For cross‐over trials, we will include only data from the first intervention period (Higgins 2011).
Dealing with missing data
Data will be analysed using the intention‐to‐treat principle, that is, patients with missing data (in all treatment groups of a trial) will be considered as treatment failures, and all randomised patients will be included in the denominator.
Sensitivity analyses will be performed to assess how sensitive results are to reasonable changes in the assumptions that are made, ie, to perform a worst‐best case scenario, best‐worst case scenario, best‐best case scenario, and a worst‐worst case scenario. The potential impact of missing data on findings of the review will be discussed in the discussion section.
Assessment of heterogeneity
Heterogeneity will be assessed using the Chi2 and I2 methods. Any plausible possible causes of heterogeneity will be discussed.
Data synthesis
The authors will follow the instructions given in The Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) and The Cochrane Hepato‐Biliary Group Module (Gluud 2011).
The evidence synthesis will most likely be done in a narrative way, but where possible and meaningful, we will use meta‐analyses. We will calculate estimates using the random‐effects model (DerSimonian 1986) and the fixed‐effect model (Mantel 1959; Greenland 1985). We will present both results if there are discrepancies in the results. If not, we plan to report the random‐effects model (DerSimonian 1986), and we plan to measure the quantity of heterogeneity using I2 (Higgins 2011).
In principle, all data are suitable for meta‐analysis. Measures of effect will be calculated as relevant (hazard ratios, odds ratios, relative risks, risk differences, mean differences, and standardised mean differences). Where possible, hazard ratios will be calculated using methods described by Parmar and Tierney (Parmar 1998). Information (eg, hazard rates, P values, events, ratios, curve data, and information on follow‐up) will be extracted from the publication, and, if necessary, they will be entered into a Microsoft Office Excel 2003 spreadsheet to calculate hazard ratios (Tierney 2007). Where data are available for the same outcomes using similar methods, these will be meta‐analysed statistically. If data cannot be meta‐analysed statistically, for example, in case of extreme heterogeneity, results will be presented in a forest plot, without the estimate, in order to show the variance of the effects. Cross‐over trials will be included using results of the first period only (before cross‐over), as if they were parallel trials.
If a meta‐analysis is not possible, the results will be presented in a narrative way. This means that we will present text, tables, and figures to summarise the data and to allow the reader to judge the results based on the differences and similarities of the included studies and their risk of bias assessment. The trials will be grouped by intervention, patient characteristics and outcomes, and the most important characteristics of the included studies will be described, including a detailed review of the methodological shortcomings of the study.
If available data allow, we will use funnel plots to identify any possible small trial biases, such as publication bias. If present, we will discuss possible implications of our findings.
Where possible, we will examine apparently significant beneficial and harmful intervention effects with trial sequential analyses (CTU 2011) in order to evaluate if these apparent effects could be caused by random error (‘play of chance’) (Bangalore 2008; Wetterslev 2008; Brok 2009; Thorlund 2009).
We will create a 'Summary of findings' table including, where possible, survival, response, recurrence, and adverse events.
Subgroup analysis and investigation of heterogeneity
Subgroup analyses will be performed where possible, based on the following prognostic indicators: age, sex, tumour size, location of primary tumour, and use of any co‐interventions.
Sensitivity analysis
Outcomes after intervention, after 6 months or less, 6 to 12 months, and 1 year or more will be summarised separately.
