Abstract Gastro-oesophageal cancer (GOC) ranks among the most prevalent cancers worldwide, its asymptomatic nature in early stages often delays diagnosis, contributing to a poor 5-year survival rate of 18%. Standard treatment typically involves chemotherapy administered in both the pre- and post-operative settings. While chemotherapy can benefit some patients, its overall efficacy is frequently limited by intrinsic resistance, subjecting others to substantial toxicity without considerable therapeutic benefit. Advances in targeted therapies in GOC have lagged behind those in other malignancies, largely due to pronounced intra- and inter-tumour heterogeneity. Nonetheless, the human epidermal growth factor receptor 2 (HER2), is over-expressed in 20% of cases. The landmark ToGA trial established trastuzumab, a HER2-targeting monoclonal antibody, combined with chemotherapy for advanced HER2-positive GOC. To further refine HER2 targeting, antibody-drug conjugates such as trastuzumab emtansine (T-DM1) and trastuzumab deruxtecan (T-DXd) have been developed, showing promising results in preclinical and clinical studies. Despite these advances, current methods for assessing treatment response remain inadequate for informing clinical decisions in real time. This highlights the urgent need for predictive and dynamic tools to evaluate therapeutic efficacy. To address this, we developed a biobank of patient-derived cancer organoids with varying HER2 expression levels to explore individualised responses to therapies. These findings highlighted anti-HER2 therapy, T-DXd, activated DNA damage response pathways, specifically ataxia-telangiectasia mutated (ATM) and Rad3-related (ATR) kinases. These results suggest that co-targeting DNA repair pathways could potentiate the effects of T-DXd. Further investigation is warranted to validate these insights and optimise therapeutic strategies for GOC.
Savage et al. (2026) studied this question.