3505 Background: Antibody-drug conjugate (ADC) targeting HER2 represents standard therapy for HER2-positive gastric cancer and breast cancer. This study aimed to assess trastuzumab rezetecan (an HER2-targeted ADC) vs SOC in patients (pts) with chemotherapy-refractory, HER2-positive, RAS and RAF wild-type advanced CRC. Methods: In this randomized, open-label, multicenter, phase 3 trial, pts with HER2-positive, RAS and RAF wild-type, advanced CRC who have progressed after standard 2nd-line therapy were randomized 2:1 to receive trastuzumab rezetecan (4.8 mg/kg, d1, iv, q3w) or SOC (TAS-102 35 mg/m 2 , bid, d1-5 and d8-12, fruquintinib 5 mg, qd, d1-21, or regorafenib 160 mg, qd, d1-21; po, q4w). Randomization was stratified by HER2 status (IHC 3+ vs IHC 2+/ISH+) and ECOG PS (0 vs 1). Primary endpoint was PFS per RECIST v1.1 by independent review committee (IRC). Key secondary endpoint was OS. Results: As of data cutoff (Oct 31, 2025), 130 pts were randomly assigned to trastuzumab rezetecan (N=86) or SOC (N=44). 70.8% of pts had HER2 IHC 3+, and 57.7% had an ECOG PS of 1. Median follow-up was 9.6 mo (95% CI 7.4-10.7). PFS per IRC was significantly increased with trastuzumab rezetecan vs SOC (median 5.5 mo 95% CI 4.8-6.7 vs 2.8 mo 95% CI 2.2-4.2; HR 0.33 95% CI 0.21-0.53; 1-sided p<0.0001). Although median OS were immature, risk of death with trastuzumab rezetecan was reduced by 23% compared with SOC (HR 0.77 95% CI 0.35-1.73). PFS per investigator, ORR and DoR were also better with trastuzumab rezetecan vs SOC (Table 1). Grade ≥3 TRAEs occurred in 48.8% with trastuzumab rezetecan and 50.0% with SOC. No patients discontinued treatment due to TRAEs. Treatment-related death was reported in two pts (2.3%) with trastuzumab rezetecan (septic shock and myelosuppression). Conclusions: Compared with SOC, trastuzumab rezetecan achieved prolonged PFS and improved ORR in chemotherapy-refractory HER2-positive advanced CRC, with a tolerable safety profile. Clinical trial information: NCT06199973 . Efficacy. Trastuzumab rezetecan (N=86) SOC (N=44) Treatment effect (95% CI) 1-sided p value IRC-assessed PFS, mo 5.5 (4.8-6.7) 2.8 (2.2-4.2) HR, 0.33 (0.21 0.53) <0.0001 * Investigator-assessed PFS, mo 5.6 (5.0-6.8) 2.8 (1.5-4.1) HR, 0.31 (0.20-0.50) <0.0001 * OS, mo NR (14.6-NR) NR (12.5-NR) HR, 0.77 (0.35-1.73) 0.2651 * IRC-assessed ORR 35 (40.7; 30.2-51.8) 2 (4.5; 0.6-15.5) RD, 36.2 (24.1-48.2) <0.0001 † Investigator-assessed ORR 28 (32.6; 22.8-43.5) 2 (4.5; 0.6-15.5) RD, 28.0 (16.4-39.7) 0.0002 † IRC-assessed DoR, mo 4.4 (3.3-NR) 3.4 (2.7-NR) - - Investigator-assessed DoR, mo 5.8 (4.2-NR) 1.4 (NR-NR) - - Data are median (95% CI), or n (%; 95% CI), unless otherwise stated. * Stratified log-rank test. † Stratified Cochran-Mantel-Haenszel test. NR, not reached; RD, rate difference.
Li et al. (Wed,) studied this question.