Molecular profiling-selected treatment in metastatic gastric and oesophageal cancer: Real-life clinical experience.

Abstract

e15537 Background: We evaluated real-life clinical experience with molecular profiling (MP)-guided therapy in metastatic gastric and oesophageal cancer in Israel Methods: This multicenter retrospective cohort study included patients with metastatic gastric or oesophageal cancer who were treated in the participating institutions and underwent MP (Caris Molecular Intelligence). Treatment was considered as having benefit if the ratio between the longest progression-free survival (PFS) post MP and the last PFS pre MP was ≥1.3. The null hypothesis was that ≤15% of patients gain such benefit. Results: The analysis included 46 patients (61% males; median age, 58 years; 57% with poorly-differentiated tumours). At least one actionable (i.e., predictive of response to a specific therapy) biomarker was identified per patient. Immunohistochemistry was performed on all samples and identified 1-8 (median: 3) biomarkers per patient (most commonly: low TS, high TOPO1, high TOP2A). Twenty-eight patients received therapy post MP (1-4 lines). In the 1st-line post MP, 5 patients (18%) achieved partial response and 5 (18%) stable disease; the median (range) PFS was 4.3 (0.4-38.5) months. Twenty-four patients were evaluable for PFS ratio analysis; in 7 (29.2%), the ratio was ≥1.3. In one-sided exact binomial test vs. the null hypothesis, P = 0.019; therefore, the null hypothesis was rejected. Conclusions: Our findings demonstrated that implementing MP is feasible and that MP could provide clinical benefit for a considerable proportion (~30%) of patients with metastatic gastric or oesophageal cancer.