Four studies (229 patients) were included in the review including three RCTs and one retrospective study. Sample sizes ranged from 19 to 114 patients. The Jadad scores for the three RCTs were 3 points; the retrospective study was not assigned a score for methodological quality.
There were statistically significant benefits of gemcitabine-based chemoradiotherapy with higher survival rates at 12-months follow up (RR 1.54, 95% CI 1.05 to 2.26; Ι²=0%) compared with 5-fluorouracil-based chemoradiotherapy. There were no differences between the chemoradiotherapy regimens at six-months or 24 months of follow-up.
Sensitivity analyses showed a similar pattern of results at 12-months follow-up with higher survival rates observed with gemcitabine-based chemoradiotherapy compared with 5-fluorouracil-based chemoradiotherapy when the trial including an additional chemotherapeutic agent was excluded from the analysis (RR 1.51, 95% CI 1.00 to 2.29; Ι²=0%; three studies, 167 patients) and when the smallest trial was excluded from the analysis (RR 1.59, 95% CI 1.06 to 2.37; Ι²=0%; three studies, 210 patients).
In the sensitivity analysis of the three RCTs, there were no significant differences between the treatment groups at six-months or 12-months follow-up, but the survival rate at 24 months was higher for the gemcitabine-based chemoradiotherapy (RR 5.49, 95% CI 1.01 to 29.87, three RCTs; Ι²=0%; 115 patients).
There were significantly more adverse events observed with gemcitabine-based chemoradiotherapy compared to 5-fluorouracil-based chemoradiotherapy with increased leucocytopenia (RR 5.02, 95% CI 1.86 to 13.54; three studies), thrombocytopenia (RR 6.10, 95% CI 1.69 to 21.99; three studies) and gastro-intestinal bleeding (RR 4.26, 95% CI 1.25 to 14.48). There were no significant differences between the treatment groups in anaemia.
The funnel plots for survival at six months, 12 months and 24 months showed some asymmetry, which suggested the possibility of publication bias.