Thirteen RCTs (n=3,224) were included in the review. Sample sizes ranged from 26 to 624 patients. No details of randomisation methods and allocation concealment were reported in nine studies. Intention-to-treat analyses were reported in three studies; use of such analyses were not reported in seven studies and not performed in three studies. Drop-out rates (where reported) ranged from zero to 11.7%.
There were statistically significant benefits observed in overall survival for the group that received neoadjuvant chemotherapy compared to patients who received surgery alone (HR 0.84, 95% CI 0.77 to 0.92; 13 RCTs). Statistical heterogeneity was observed, but this was not statistically significant (I2=24%, p=0.20).
The results of the sensitivity analyses continued to show benefits in favour of neoadjuvant chemotherapy and surgery when the largest two trials were removed (HR 0.83, 95% CI 0.75 to 0.93; n=2,144) and on removal of four Chinese trials (HR 0.83, 95% CI 0.73 to 0.93; n=1,672).
Subgroup analysis of patients with Stage III NSCLC showed statistically significant benefits in survival for the group that received neoadjuvant chemotherapy (HR 0.84, 95% CI 0.75 to 0.95, I2=43%; eight RCTs, n=1,586)