Background Despite remarkable benefit has been provided by immune checkpoint inhibitors in gastric cancer (GC), predictions of treatment response and prognosis remain unsatisfying, making identifying biomarkers desirable. We aim to develop and validate a CT imaging biomarker to predict the immunotherapy response in patients with GC and investigate the associated immune infiltration patterns.

Methods This retrospective study includes 294 GC patients who received anti-PD-1/PD-L1 immunotherapy from three independent medical centers between January 2017 to April 2022. A radiomics score (RS) was developed from the intratumoral and peritumoral features on pretreatment CT images to predict immunotherapy related progression-free survival (irPFS). The performance of the RS was evaluated by the area under the time-dependent receiver operating characteristics curve (AUC). Multivariable Cox regression analysis was performed to construct predictive nomogram of irPFS. C-index were used to determine the performance of nomogram. Bulk RNA sequencing of tumors in The Cancer Genome Atlas was used to investigate the RS-associated immune infiltration patterns.

Results Overall, 89 of 294 patients (median age, 57 years [IQR, 48–66 years]; 171 men) had objective response to immunotherapy. The RS included 13 CT features yielded 12-month AUCs of 0.787, 0.810 and 0.785 in training, internal validation, and external validation 1 cohorts, respectively, and 24-month AUC of 0.805 in external validation 2 cohort. Patients with low RS had longer irPFS in each cohort (P < 0.05). Multivariable Cox regression analyses showed RS is the independent prognostic factor of irPFS. Nomogram that integrated RS and clinical characteristic showed improved performance in predicting irPFS with C-index of 0.687–0.778 in training and validation cohorts. The CT imaging biomarker was associated with M1 macrophages cells infiltration.

Conclusions Noninvasive CT imaging biomarker can effectively predict immunotherapy outcomes in GC patients and is associated with innate immune system.

Acknowledgements We acknowledge the TCGA database for providing their platforms and contributors for uploading their meaningful datasets.

Ethics Approval Ethical approval was obtained from the institutional review board of the three participating centers, and informed consent was waived for this retrospective analysis.