Nineteen RCTs (n = 1,554) were included in the review. Jadad scores ranged from 1 to 5, with five trials scoring 3 or higher. Sample size ranged from 20 to 250, 11 RCTs included 60 or fewer patients.
Glutathione (seven RCTs): two of six RCTs found statistically significantly lower levels of neurotoxicity in the interventions groups (p = 0.0001 and p = 0.01). The direction of effect was the same in the studies showing non-significant results. One RCT found significantly higher chemotherapy completion rates in the intervention group (58 per cent versus 39 per cent, p = 0.04), with a significantly lower rate of non-completion due to nephrotoxicity (p = 0.012). There were no significant differences in survival or response rate, although one subgroup analysis in one RCT showed an increased complete response rate (p = 0.014).
Melatonin (four RCTs): three RCTs reported statistically significant improved survival either at one year (two RCTs, p < 0.05 and p < 0.001) or five years (one RCT, p< 0.001) in the intervention groups. The final study reported a significantly higher rate of disease stabilisation in the melatonin group (p< 0.05).
Mixed interventions (two RCTs): one RCT evaluated vitamins C and E with beta carotene and the other assessed vitamins C and E with selenium. No significant differences between the groups were observed.
Vitamin E (one RCT): there was a significantly lower rate of neurotoxicity in the supplement group (31 per cent versus 86 per cent, p< 0.01). There was no significant difference in tumour response between the groups.
Ellagic acid (one RCT): there was a significantly lower rate of neutropenia in the supplement group (33 per cent versus 75 per cent, p < 0.05). There was no significant difference in complete response rate or two-year survival rate between the groups.
Vitamin A (two RCTs): one RCT found a higher rate of complete response in the supplement group (38 per cent versus 15 per cent, P < 0.02) and a higher projected 43-month survival rate (93 per cent versus 30 per cent, p < 0.02) and these outcomes also showed significant differences for a postmenopausal subgroup; the second RCT found no significant differences in efficacy outcomes, but a lower rate of grade 2 or higher toxicities in the supplement group (4 per cent versus 23 per cent, p = 0.002).
No significant differences between the groups were observed for N-acetylcysteine (one RCT) or vitamin C (one RCT).