Tumor immunophenotyping-derived signature identifies prognosis and neoadjuvant immunotherapeutic responsiveness in gastric cancer

Abstract Background The effectiveness of neoadjuvant immune checkpoint inhibition (ICI) therapy has been confirmed by clinical trials; however, patients that are suitable for receiving this therapy remain unspecified. Previous studies have demonstrated that the tumor microenvironment (TME) dominates immunotherapy; therefore, an effective classification strategy for the TME is needed. Methods Five publicly available datasets (n = 1,426) were used to identify the immunophenotypic features of the TME and to screen for core molecules. The training cohort (n = 506) was used to construct the immunophenotypic score (IPS) and six independent external centers (n = 638) were recruited as validation. Besides, 52 patients with GC receiving neoadjuvant anti-PD-1 therapy was enrolled to explore the value of the IPS in neoadjuvant ICI therapy. Moreover, the IPS-specific TME was profiled by multiplexed immunohistochemical staining and immunohistochemical staining. Results Five immunophenotype-related features (WARS, UBE2L6, GZMB, BATF2, and LAG-3) associated with prognosis and the immunotherapeutic response in GC were identified, forming the IPS. The data from seven medical centers (n = 1,144) indicated that the IPS is a robust and independent biomarker for GC and superior to the traditional TNM stage. Furthermore, IPSLow was defined as the immune-activated tumor that benefited from neoadjuvant anti-PD-1 therapy, while IPSHigh exhibited more immune unresponsive signals. Notably, patients with IPSLow and PD-L1 (CPS) ≥ 5 were observed to be the most favorable group for neoadjuvant anti-PD-1 treatment. Conclusions The IPS could serve as a valid quantitative tool for immunophenotyping to improve clinical outcomes, and it provides an effective reference for the implementation of neoadjuvant ICI therapy for patients with GC.