Trastuzumab Deruxtecan (T-DXd) in HER2-Positive Advanced Gastric and Gastroesophageal Junction Cancer: Evidence, Sequencing, Patient Selection, and Clinical Decision-Making.

PURPOSE: To provide a clinically oriented review of trastuzumab deruxtecan (T-DXd) in HER2-positive advanced gastric or gastroesophageal junction adenocarcinoma, focusing on evidence, post-progression HER2 reassessment, second-line sequencing, patient selection, regional implementation, and toxicity management. METHODS: This narrative review integrates pivotal DESTINY-Gastric trials, HER2 reassessment studies, realworld cohorts, safety analyses, pathology guidance, and regulatory documents available through August 2026. RESULTS: DESTINY-Gastric04 established T-DXd as a preferred second-line treatment after progression on trastuzumab-containing therapy, improving median overall survival versus ramucirumab plus paclitaxel (14.7 vs 11.4 months; hazard ratio 0.70). HER2 expression is heterogeneous and dynamic; postprogression reassessment is therefore recommended whenever feasible using biopsy-specific scoring criteria. The 494 of 1,088 patients randomized in DESTINY-Gastric04 should not be interpreted as a 55% HER2-loss rate because screening attrition was multifactorial. T-DXd is preferred when HER2 positivity is retained, whereas ramucirumab plus paclitaxel remains appropriate when HER2 loss is documented. When HER2 status is unconfirmed, treatment should be individualized, recognizing regional regulatory differences. Older patients require careful toxicity assessment, particularly for interstitial lung disease (ILD)/pneumonitis. Grade 2 or higher ILD requires permanent discontinuation; grade 1 reinitiation must follow the applicable regional label. CONCLUSION: T-DXd is a preferred second-line option for post-progression HER2-positive advanced gastric or gastroesophageal junction cancer, but optimal use requires careful HER2 reassessment and regionappropriate safety management. Evidence after T-DXd failure, ctDNA-guided selection, and sequential CLDN18.2-directed therapy remains limited and requires prospective validation.