Liao 2010.

Methods	Study design: parallel randomised clinical trial Study duration: November 2005 to November 2007 Duration of follow‐up: date of TACE therapy until the last follow‐up visit (or death) up to November 2009 (median 12 months; range 2–38 months)
Participants	Country: China Setting: inpatient, 1 centre People with liver function Child‐Pugh class A or B; ECOG PS 0–2; no distant metastasis; no obvious myelosuppression or renal damage Number: treatment group 24; control group 24 Mean age: treatment group 62 years; control group 63 years Sex (M/F): treatment group 15/9; control group 14/10 Tumour size: treatment group 3.5–11 cm; control group 3.8–11.5 cm Child‐Pugh class (A/B): treatment group 17/7; control group 17/7 Stage (III/IV): treatment group 18/6; control group 16/8
Interventions	Control group: 3–5 courses of TACE with 1‐month interval. TACE comprised hepatic arterial infusion chemotherapy and hepatic artery embolisation. 5‐fluorouracil 1000–1250 mg; cisplatin 70–90 mg; adriamycin 50–60 mg; peripheral embolisation with iodine oil emulsion and central embolisation with gelfoam Treatment group: 3–5 courses of TACE with 1‐month interval; 3‐DCRT delivered 1–2 weeks after the last course of TACE if liver function tests were normal. Radiotherapy: Varian23EX linear accelerator used to deliver 6 MV high‐energy X‐ray radiotherapy. Sum of radiation doses received by each participants ranged from 40 Gy to 66 Gy (median 55 Gy) with 2–2.5 Gy/day, radiation administered in 20–33 sessions
Outcomes	According to RECIST, treatment efficacy divided into CR, PR, SD, and PD Response rate was the sum of CR+PR 1‐ and 3‐year survival rates Toxic effects Death rate and cause
Notes	Source of funding: Project of Quzhou Science and Technology Bureau No information on the compliance and ITT/PP analysis Attempted to contact the first author, without success
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "randomization was performed by drawing lots."
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not reported for all outcomes
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported for all outcomes
Incomplete outcome data (attrition bias) All outcomes	High risk	There were postrandomisation dropouts and all relevant outcomes were reported for only 45 participants (i.e. 23 in treatment and 22 in control group).
Selective reporting (reporting bias)	High risk	1 predefined outcome (serious adverse events) not reported. Protocol not available
Other bias	Low risk	No other potential source of bias identified, baseline characteristics comparable