In urgent clinical scenarios with very sick patients, chemotherapy remains the core initial treatment. Targeted or immunotherapies can be added later once results for key biomarkers like MSI, HER2, PD-L1, and CLDN18.2 are available. This prioritizes immediate patient stability over a delayed precision approach.
The Matterhorn trial (durvalumab + triplet FLOT chemo) succeeded while KEYNOTE-585 (pembrolizumab + doublet chemo) failed. This stark contrast suggests the immunotherapy's success is contingent on a highly effective chemotherapy partner, making the more potent triplet regimen the superior backbone for perioperative treatment.
Retrospective data suggests that chemotherapy for localized MSI-high gastroesophageal cancer may be neutral or even deleterious. The concern is that it could blunt the patient's immune response, hindering the profound efficacy of checkpoint inhibitors in this specific subgroup. Experts are increasingly avoiding chemotherapy here.
Unlike some other cancers, early debulking surgery for multi-system metastatic gastric cancer is strongly discouraged. The Renaissance trial showed this approach can be deleterious. Surgery should only be considered for isolated sites of disease after a prolonged, sustained response of over one to two years.
While immunotherapy is approved for gastroesophageal cancer with a PD-L1 CPS score of 1 or higher, clinical data reveals the most significant and durable survival benefit is largely confined to patients with high expression (CPS ≥10) or MSI-high status. The benefit for patients with CPS scores from 1 to 9 is considered borderline.
The approved dose of trastuzumab deruxtecan (TDXD) for gastric cancer is 6.4 mg/kg, higher and more toxic than in other tumors. Clinicians must use a multi-agent antiemetic cocktail (including NK1 inhibitors and olanzapine) and consider a lower starting dose of 5.4 mg/kg for elderly or poor-performance patients.
In small Phase 2 trials, combining anti-PD-1 and anti-CTLA-4 therapies for 2-3 months yielded pathologic complete response rates of 60% in localized MSI-high gastroesophageal cancer. This stunning result, with high long-term survival, opens the possibility of non-operative management for these patients.
Unlike previous trastuzumab-IO combinations that harmed PD-L1 negative patients, the zanidatumab-IO regimen benefits HER2+ patients regardless of PD-L1 status. This is likely because zanidatumab is a more immunogenic molecule, potentiating the checkpoint inhibitor's effect even in a less inflamed tumor microenvironment.
Despite a tumor being HER2-positive, clinical trial data shows that adding adjuvant HER2-targeted therapy like trastuzumab after surgery provides no survival benefit. This is a crucial negative finding that helps clinicians avoid ineffective and potentially toxic therapy in the post-operative setting.