# Integration of the GRIm Score with Pathologic Immune and Stromal Markers to Develop a Combined Prognostic Model in Gastric Cancer: A Retrospective Single-Center Study

## TL;DR

This study shows that combining the GRIm score with immune and stromal markers improves survival prediction in gastric cancer patients.

## Contribution

The novel contribution is integrating the GRIm score with tumor microenvironment markers to enhance prognostic stratification in gastric cancer.

## Key findings

- Low GRIm scores correlate with longer disease-free and overall survival in gastric cancer patients.
- Combining GRIm with tumor-stroma ratio (TSR) provides the best survival stratification.
- GRIm combined with PD-L1 and TIL density also improves prognostic discrimination.

## Abstract

Background and Objectives: The Gustave Roussy Immune (GRIm) score, reflecting systemic inflammation and nutritional status, has emerged as a simple and reproducible prognostic biomarker in various malignancies. However, its prognostic interaction with tumor microenvironmental factors remains unclear in gastric cancer. The primary aim of this study was to evaluate the prognostic value of the GRIm score in patients with resectable gastric adenocarcinoma, while the secondary aim was to determine whether integrating the GRIm score with tumor microenvironment–related pathological markers could improve prognostic stratification. Materials and Methods: This retrospective study analyzed 188 patients with resectable gastric adenocarcinoma treated at the Trakya University Faculty of Medicine between 2007 and 2018. GRIm scores were calculated from preoperative lactate dehydrogenase (LDH), albumin, and neutrophil-to-lymphocyte ratio (NLR) values. Pathologic parameters, including programmed death-ligand 1 (PD-L1) expression (combined positive score [CPS] ≥ 1 vs. <1), tumor–stroma ratio (TSR; stromal component ≥ 50% vs. <50%), and tumor-infiltrating lymphocyte (TIL) density (CD8+ ≥ 10% vs. <10%), were evaluated on surgical specimens. Survival outcomes were assessed using Kaplan–Meier and multivariate Cox analyses. Results: The study population had a mean age of 61.8 years and was predominantly male (72.3%). Patients with low GRIm scores had significantly longer disease-free survival (DFS; 24 vs. 12 months; p = 0.004) and overall survival (OS; 32 vs. 19 months; p = 0.006). In multivariate analysis, the GRIm score remained an independent predictor for both disease-free survival (p = 0.035) and overall survival (p = 0.044). Among combined models, the GRIm–TSR classification provided the most pronounced stratification (median DFS = 35 vs. 12 months; OS = 45 vs. 19 months; p = 0.014 and 0.001, respectively), retaining independent prognostic significance (hazard ratio [HR] = 1.23; p = 0.005). Integrating GRIm with PD-L1 and TIL density also improved prognostic discrimination. Conclusions: The GRIm score is a robust and cost-effective biomarker that independently predicts disease-free survival and overall survival in resectable gastric adenocarcinoma. Its combination with microenvironmental markers—PD-L1, TIL, and TSR—captures complementary biological dimensions of tumor aggressiveness, offering an integrative and clinically feasible framework for individualized risk assessment and postoperative management. Prospective multicenter validation is warranted.