Four RCTs were included (2,518 participants, range 146 to 1,725). Three studies scored three points on the Jadad scale and one study scored two points. All trials reported withdrawals and drop-outs, but none reported use of double-blinding.
Pemetrexed plus platinum chemotherapy improved overall survival by 9% compared with other regimens (HR 0.91, 95% CI 0.83 to 1.00, 4 RCTs); the risk reduction was 13% in patients with non-squamous cell carcinoma (3 RCTs), although this result was largely driven by the one trial which used carboplatin. There were no statistically significant differences for progression-free survival (two RCTs) and overall response (3 RCTs).
Pemetrexed plus platinum chemotherapy resulted in less grade 3-4 neutropenia (OR 0.50, 95% CI 0.34 to 0.74; four RCTs) and leukopenia (OR 0.41, 95% CI 0.25 to 0.65; three RCTs), but more grade 3-4 nausea (OR 1.63, 95% CI 1.11 to 2.39; four RCTs). However, there were no significant differences for vomiting and diarrhoea. Further toxicity results were reported.
There was no evidence of statistically significant heterogeneity or publication bias, except for the grade 3-4 haematological toxicity outcomes, which were all subject to statistically significant heterogeneity.