Circulating exosomal gastric cancer-associated long noncoding RNA1 as a noninvasive biomarker for predicting chemotherapy response and prognosis of advanced gastric cancer: A multi-cohort, multi-phase study

Background A previous validated study has identified the diagnostic value of circulating exosomal lncRNA-GC1 for detecting and monitoring gastric cancer. We aimed to further determine the predictive role of circulating exosomal lncRNA-GC1 for prognosis and chemotherapy response. Methods We retrospectively conducted a multi-phase analysis with four independent cohorts of 981 patients. A training cohort was used to generate the predictive model. One internal and two external cohorts were recruited as validation cohorts. Patients with stage II or III gastric cancer in the combined cohort were used to evaluate the predictive value of circulating exosomal lncRNA-GC1 for chemotherapy response. Findings In the training cohort, circulating exosomal lncRNA-GC1 was identified as an independent prognostic predictor for disease-free and overall survival. A prognostic risk stratification model based on circulating exosomal lncRNA-GC1 and AJCC stage revealed better predictive accuracy for disease-free and overall survival than the traditional AJCC stage system alone (C-index: DFS 0.701 vs 0.614; OS 0.720 vs 0.611, both P<0.05). And it has been further verified in the validation cohorts. In interaction analysis, for stage II and III GC, patients with low-level of circulating exosomal lncRNA-GC1 derived more survival benefit from adjuvant chemotherapy (P < 0.05); while those with high-level did not. Interpretation Measurement of circulating exosomal lncRNA-GC1 provides clinically important prognostic information and could complement the AJCC stage to optimize decision-making for selecting patients who could benefit more from fluorouracil-based chemotherapy after surgery. Funding The funders are listed in the Acknowledgement. Copyright (c) 2022 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/) eBioMedicine 103971 Published https://doi.org/10.1016/j. ebiom.2022.103971