Abstract 159P
Background
Within the tumor microenvironment (TME), the role of B cells in influencing the prognosis and response to therapy of patients with gastric cancer (GC) remains poorly characterized. We investigated the predictive and prognostic value of B cells including an assessment of their spatial organization within the TME in one of the largest multi-cohort studies to date.
Methods
Using CD20 immunohistochemistry, we studied B cell abundance in resection specimens from more than 977 patients with resectable GC across three cohorts. The relationship between CD20 abundance, clinicopathological characteristics and overall survival (OS) was analyzed. Furthermore, digital spatial profiling of 1063 regions of interest from 15 patients was performed to characterize the distribution of B cells within different spatial regions.
Results
CD20 abundance was significantly higher in diffuse-type GC compared to intestinal-type GC (p=0.000012). Among all the included patients from the Korean CLASSIC trial (n=549), those with CD20-low diffuse-type GC (n=56) had the shortest OS compared to all other patients (median OS 49 months vs 62 months, HR:1.9, 95%CI: 1.2-3, p=0.003). The results were similar in Japanese patients (n=215) with CD20-low diffuse-type GC (n=17) compared to all other patients (median OS 49 vs 67 months, HR: 2.2, 95%CI: 1.2-4, p=0.011). Specifically, within the treatment arm of the CLASSIC trial, the difference between the CD20-low diffuse type on adjuvant chemotherapy vs other patients on adjuvant chemotherapy was no longer significant (median OS 62 vs 63 months, HR: 1.8, 95%CI: 0.88-3.5, p=0.108). Spatial analysis showed that there were significantly more B cells within the tumor cell regions compared to the intratumoral stroma and the adjacent normal regions. This was in particularly true for diffuse-type GC, while B cells were found to be more spread in different regions in intestinaltype GC.
Conclusions
B cell abundance, particularly in diffuse-type GC, is associated with poor prognosis and may identify a subgroup of GC patients who particularly benefit from adjuvant chemotherapy. These findings highlight the importance of B cells in GC and may help to improve risk stratification and inform treatment decisions.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
National Medical Research Council (NMRC); Clinician Scientist Development Unit (CSDU) NUS Medicine Student Research Mentorship Program.
Disclosure
T. Oshima: Financial Interests, Personal, Invited Speaker: Taiho, Astellas; Financial Interests, Personal, Writing Engagement: Astellas; Financial Interests, Personal, Other, research grant: Taiho, Kyowa Kirin, Chugai, Nihon Kayaku. H.I. Grabsch: Financial Interests, Personal, Other, Honoraria: MSD, NL. R. Sundar: Financial Interests, Personal, Advisory Board, Advisory Board, Invited Speaker: Bristol Myers Squibb, MSD; Financial Interests, Personal, Advisory Board: Merck, Bayer, Novartis, GSK, Pierre Fabre, Tavotek, AstraZeneca, Daiichi Sankyo, Beigene; Financial Interests, Personal, Advisory Board, Advisory Board, Invited Speaker, Travel: Eisai; Financial Interests, Personal, Advisory Board, Advisory Board, Invited Speaker, Travel for conferences. Funding declared is over several years (<10,000 Euro per year): Taiho; Financial Interests, Personal, Invited Speaker: Eli Lilly, BMS, Roche, Taiho, AstraZeneca, DKSH, Daiichi Sankyo, Beigene, Astellas; Financial Interests, Personal, Advisory Board, Travel for conference and Advisory Board, funding declared is over several years. All other authors have declared no conflicts of interest.