Abstract Objective To compare the efficacy and safety of sacituzumab tirumotecan (sac-TMT) with docetaxel in patients with locally advanced or metastatic epidermal growth factor receptor ( EGFR) -mutated non-small cell lung cancer (NSCLC) after previous treatment failure with EGFR-tyrosine kinase inhibitors and platinum based chemotherapy. Design Multicentre, open label, randomised controlled trial. Setting 48 centres in China, 1 September 2023 to 31 December 2024. Participants 137 adults (aged 18-75 years) with EGFR -mutated advanced or metastatic NSCLC after previous treatment failure with EGFR-tyrosine kinase inhibitors and platinum based chemotherapy. Intervention Patients were randomly assigned (2:1) to receive sac-TMT (5 mg/kg) on days 1 and 15 of each four week cycle, or docetaxel (75 mg/m 2 ) on day 1 of each three week cycle. Patients in the docetaxel group were permitted to crossover to sac-TMT treatment on disease progression. Main outcome measures The primary endpoint was objective response rate as assessed by a blinded independent review committee (BIRC). The secondary endpoints included objective response rate assessed by the investigator; disease control rate, progression-free survival, time to response, and duration of response assessed by BIRC and the investigator; overall survival; and safety. Results 137 patients were randomised to receive sac-TMT (n=91) or docetaxel (n=46). Median follow-up was 12.2 months at the data cut-off for efficacy (31 December 2024). BIRC assessed objective response rate was significantly higher in the sac-TMT group (45% (41/91)) v docetaxel (16% (7/45)), with a difference of 29% (95% confidence interval (CI) 15% to 43%; one sided P<0.001). Median progression-free survival was longer with sac-TMT than with docetaxel assessed by BIRC (6.9 v 2.8 months; hazard ratio 0.30, 95% CI 0.20 to 0.46; one sided P<0.001) and the investigator (7.9 v 2.8 months; hazard ratio 0.23, 0.15 to 0.36; one sided P<0.001). The 12 month overall survival rate was 73% with sac-TMT and 54% with docetaxel (hazard ratio 0.49, 0.27 to 0.88; one sided P=0.007). After adjustment for crossover using the rank-preserving structural failure time model, sac-TMT also showed improved overall survival (hazard ratio 0.36, 0.20 to 0.66). Grade ≥3 treatment related adverse events were less frequent with sac-TMT than with docetaxel (56% v 72%), with no new safety signals identified. Conclusions Sac-TMT showed statistically significant and clinically meaningful improvements in objective response rate, progression-free survival, and overall survival compared with docetaxel, with a manageable safety profile in patients with EGFR -mutated locally advanced or metastatic NSCLC. Trial registration ClinicalTrials.gov NCT05631262 .