Review question:
The review is going to resolve conflicting evidence: whether the therapeutic effects of transarterial chemoembolization (TACE) for colorectal liver metastases.

Sources to be searched:
Medline; PubMed; Embase; Cochrane Library; Biological Abstracts; BA; Science Citation Index (SCI). Chinese data: cnki; CBM; 《医学中央雑誌》. NRR; CCT; CDC; DH; HC; FDA.

Types of study to be included
RCT study or parallel-group or cross-over design, irrespective of blinding, publication status, or language. For cross-over trials, only data from the first phase of the trial will be used. Quasi-randomised trials will be excluded.

Participants
Patients with unresectable colorectal liver metastases will be included. The presence of colorectal cancer should have been defined by histology, or cytology, or by clinical diagnosis(typical images indicating colorectal cancer on ultrasound (US), computed tomography (CT), or angiography), i.e. Patients at risk of colorectal cancer but without actual cancer will not be considered.

Interventions
Transarterial chemoembolization (TACE), (any type, dose, and schedule) versus placebo, no intervention, supportive therapy, or other treatments. TACE (any type, dose, and schedule) plus other intervention versus the same other intervention. TACE will be defined as the primary intervention if used in patients not suitable for surgery, local ablation of the tumours. TACE will be defined as adjuvant or neo-adjuvant intervention if used in patients in which other interventions, aimed to cure the primary tumours, were used: surgical ablation (resection, OLT), local ablation (by percutaneous ethanol or acid acetic injection or radiofrequency, etc), of the tumour(s).

Primary outcome
1. Number of patients dying (primary outcome measure). When the number is not reported by the authors we will contact them for further information.
2. Objective response, according to the World Health Organization (WHO 1981) criteria (See Appendix 1). When the authors have used different definition criteria, we will report them.

Key secondary outcomes
1. Number of patients who had colorectal cancer recurrence develop. Recurrence is defined as diagnosis of new colorectal cancer in patients in whom no signs of colorectal cancer were detected by using US and/or CT scan after the treatment had been applied. The diagnosis of new colorectal cancers will be made according to the diagnostic criteria described above under 'Types of participants'.
2. Number and type of adverse events, defined as any untoward medical occurrence in a patient in any of the two intervention groups. This occurrence should not necessarily have a causal relationship with the treatment, but did, however, result in a dose reduction or discontinuation of treatment. We have defined serious adverse events according to the ICH guidelines (ICH-GCP 1997) as any event that leads to death, is life-threatening, requires in-patient hospitalisation or prolongation of existing hospitalisation, results in persistent or significant disability or congenital anomaly/birth defect, and any important medical event which may have jeopardise the patient or requires intervention to prevent it. All other adverse events will be considered non-serious.
3. Quality of life: improvement of performance status, disappearance or reduction of pain, or whatever the investigators defined as quality of life.
4. Health economics: length of hospitalisation, cost for procedure, or whatever the investigators defined costs as in relation to any benefit

Strategy for data synthesis
Meta-analysis