This phase III trial aimed to compare ARX788, a site-specific, construct-homogeneous antibody-drug conjugate, with lapatinib plus capecitabine in patients with human epidermal growth factor receptor 2 (HER2)-positive advanced breast cancer (ABC) who had progressed on one line of trastuzumab based regimen. Eligible patients were randomized (1:1) to receive ARX788 (1.5 mg/kg, IV, Q3W) or lapatinib plus capecitabine (LC: lapatinib 1250 mg QD; capecitabine 1000 mg/m2 BID, days 1-14, Q3W) and stratified by prior chemotherapy lines (0-1 versus >1) and visceral metastasis (yes versus no). The primary outcome was progression-free survival (PFS) assessed by a blinded independent central review (BICR). A total of 441 patients were randomly assigned to receive either ARX788 (n = 221) or LC (n = 220). The median PFS was 11.3 (95% confidence interval [CI], 8.4-13.8) months with ARX788 compared with 8.2 (95% CI, 6.9-8.7) months with LC, as per BICR (hazard ratio [HR] 0.64, p = 0.0006). Frequencies of treatment-related adverse events (TRAEs) of any grade were 98.6% and 99.1% for ARX788 and LC, respectively. Grade ≥3 TRAEs were 41.4% and 40.0%, respectively, the most common adverse events were blurred vision (12.3%), dry eye (9.1%), keratopathy (5.9%), and interstitial lung disease (ILD, 5.9%) with ARX788; hand-foot syndrome (18.1%) and hypokalemia (5.1%) with LC; all the hematological and gastrointestinal events of grade ≥3 with ARX788 were less than 3%. Six treatment-related deaths occurred, with three cases possibly related to ILD. ARX788 significantly improved PFS compared with LC in patients with HER2-positive ABC with a distinct toxicity profile, supporting it as a potential treatment option.