Integrating inflammation-based scores into gastric cancer prognosis

Abstract

The conventional tumor-node-metastasis (TNM) staging system is crucial for predicting the prognosis of gastric cancer. However, TNM staging fails to delineate tumor biology and host immune responses. Zhou et al have addressed this in a noteworthy study, they analyzed a cohort of 1071 patients with stage I-III gastric cancer who had undergone radical gastrectomy. The systemic inflammation response index (SIRI) and platelet-to-lymphocyte ratio (PLR) score were derived from the SIRI and PLR. Multivariate analysis identified the SIRI-PLR score as an independent predictor of survival. This was significant along with factors such as age, tumor size, and TNM stage. Importantly, they created a nomogram that integrated these variables, demonstrating superior predictive accuracy compared to the TNM staging system. The findings of this single-center cohort study require prospective multicenter validation to confirm their reproducibility and generalizability. Furthermore, inflammatory markers are dynamic and serial measurements can offer a more refined assessment of patient prognoses and guide treatment responses. These findings should be combined with genomic, immunological, and radiomics data. This approach promises precise and personalized risk stratification, thereby improving the prediction of prognosis in patients with gastric cancer.

Key Words: Gastric cancer; Radical gastrectomy; Systemic inflammation response index; Platelet-to-lymphocyte ratio; Prognosis; Nomogram

Core Tip: The conventional tumor-node-metastasis (TNM) staging system for gastric cancer has limitations as it often fails to capture tumor biology and host immune responses. Zhou et al developed a novel composite biomarker using the inflammatory markers, systemic inflammation response index and platelet-to-lymphocyte ratio. This score has been identified as an independent predictor of survival. A nomogram integrating this score with the TNM stage and other factors demonstrated superior predictive accuracy compared with the TNM system alone. Although promising, these findings require multicenter validation. Future studies should integrate these biomarkers with genomic data for more precise and personalized risk stratification of patients with gastric cancer.

INTRODUCTION

Gastric cancer (GC) remains a major health challenge worldwide, with nearly one million new cases annually contributing to more than 650000 deaths[1]. Clinical and histological evaluation play vital roles in the prognostic assessment of GC[2]. The tumor-node-metastasis (TNM) staging system serves as the gold standard in clinical practice for determining treatment strategies and outcome prediction for GC[3,4]. This anatomically based classification allows for objective and reproducible risk stratification by assessing tumor size and depth, the presence and extent of lymph node metastasis, and the presence of distant metastasis. However, clinicians routinely encounter patients who exhibit markedly different clinical courses, despite being assigned to the same TNM stage. For instance, some patients with stage II or stage III GC experience rapid postoperative recurrence, whereas others achieve curative outcomes and long-term survival[5,6]. This fundamental dilemma observed in clinical settings highlights a critical limitation of the TNM system: Its focus on the anatomical extent of the disease overlooks other essential prognostic factors such as the biological aggressiveness of the tumor itself and the host’s physiological response to cancer[7].

INTERPRETING SYSTEMIC INFLAMMATION RESPONSE INDEX AND PLATELET-LYMPHOCYTE RATIO IN GC

The article by Zhou et al[8] entitled “Clinical significance of systemic inflammation response index and platelet-lymphocyte ratio in patients with stage I-III gastric cancer” presents a practical and noteworthy approach to a key clinical issue. They retrospectively examined two easily accessible inflammation-based markers, the systemic inflammation response index (SIRI) and platelet-to-lymphocyte ratio (PLR), in 1071 patients. Rather than assessing each marker separately, the authors proposed a composite score, SIRI-PLR, based on receiver operating characteristic-derived cutoffs (SIRI: 0.55; PLR: 140.35). However, specific details, such as the area under the curve and the rationale for cutoff selection, were not disclosed. Verifying whether these cutoff values retain prognostic significance across various subgroups, such as different TNM stages or age cohorts, is also important. For instance, whether this threshold can effectively stratify risk in patients with stage I disease remains to be investigated.

Patients were classified into three groups: Score 0 (both markers low), score 1 (one marker elevated), and score 2 (both markers elevated), to better capture systemic inflammation and immunosuppression. Multivariate analysis showed that the SIRI-PLR score was independently associated with overall survival, regardless of age, tumor size, or TNM stage (score 1: Hazard ratio = 1.444; score 2: Hazard ratio = 1.759). Importantly, they constructed a prognostic nomogram incorporating the SIRI-PLR score along with clinical variables. This model achieved a C-index of 0.757, outperforming TNM-based prediction alone.

The strength of this study was its clinical feasibility. The SIRI-PLR score relies solely on routine blood tests and does not require expensive diagnostics, making it widely applicable, particularly in low-resource settings. As a simple yet informative tool, it may assist in refining postoperative risk stratification and guiding decisions such as adjuvant chemotherapy eligibility.

FROM PROGNOSIS TO PREDICTION: THE POTENTIAL OF SIRI-PLR

Although Zhou et al[8] demonstrated the clinical utility of SIRI-PLR as a simple composite marker, several key issues must be carefully considered.

First, compared with other commonly used inflammation-based scores, such as the neutrophil-to-lymphocyte ratio, systemic immune-inflammation index, and Glasgow Prognostic Score, the SIRI-PLR score may offer improved prognostic discrimination by integrating both immune suppression and inflammatory burden. Further comparative studies are needed to evaluate its predictive accuracy (e.g., the C-index), ease of implementation, and responsiveness to dynamic changes over time.

Second, the biological plausibility of composite markers must be addressed. Previous studies have suggested that individual inflammatory markers such as SIRI and PLR are associated with the prognosis of GC[9,10]. This hypothesis is supported by the complex roles of various blood cell components in cancer progression. Neutrophils promote tumor angiogenesis by releasing factors, such as vascular endothelial growth factors, and facilitating tissue invasion through the production of matrix metalloproteinases[11,12]. Platelets protect circulating tumor cells from shear stress to facilitate their survival, promote tumor cell arrest at the endothelium, and promote extravasation into distant tissues[13]. In contrast, lymphocytes, particularly cytotoxic T cells, play a central role in intertumoral immunity[14]. Therefore, an increase in neutrophils and platelets coupled with a decrease in lymphocytes can be interpreted as a state in which tumor-promoting mechanisms are activated, while antitumor immune defenses fail. The potential strength of the SIRI-PLR score employed by Zhou et al[8] may be due to its ability to simultaneously assess multiple immune and inflammatory pathways as well as to integrate complementary information provided by individual markers. The SIRI-PLR score provides a more comprehensive and sensitive reflection of the host immunological balance between antitumor immunity and tumor-promoting inflammation.

Third, the inherent limitations of the study design and the implications for future research must be considered. As this was a single-center retrospective analysis, caution should be exercised when interpreting the generalizability of the findings. To ensure the robustness of the results, prospective validation in multicenter cohorts including patients with diverse racial and healthcare backgrounds is essential. In particular, whether the cutoff values used in this study can be directly applied to Western populations that differ in dietary habits and genetic backgrounds remains a critical question that requires further investigation[15].

Fourth, the retrospective nature of this study makes it difficult to completely exclude unmeasured confounding factors. Systemic inflammation can be influenced not only by cancer itself but also by comorbidities such as autoimmune diseases, diabetes, and obesity[16-18]. Given the retrospective nature of Zhou et al’s study[8], completely eliminating the impact of unmeasured confounding variables is challenging. Future studies should implement methods such as multivariate adjustment or propensity score matching, and conduct sensitivity analyses to validate the independent prognostic value of the SIRI-PLR score in patients without major comorbidities.

Another key issue is the effect of treatment. Most patients likely received adjuvant chemotherapy, which can affect both inflammation and survival through factors such as the regimen type, completion rate, and myelosuppression. Whether the SIRI-PLR score reflects treatment-independent biological risk remains unclear. To address this, subgroup analyses comparing the prognostic performance between patients who received chemotherapy and those who did not are essential. Furthermore, longitudinal monitoring of the SIRI-PLR score during chemotherapy could enable real-time tracking of treatment response and early relapse, potentially enhancing its role beyond prognostication and establishing it as a dynamic predictive tool. Murphy et al[16] demonstrated that platelets can internalize and sequester extracellular DNA, including tumor-derived and fetal DNA, thereby potentially serving as novel reservoirs of clinically valuable genetic material. Thus, platelet- and inflammation-based scores, such as SIRI-PLR, may also be associated with circulating tumor-derived DNA. Clarifying this distinction is essential, especially in the context of therapy.

The SIRI-PLR score, which is derived from routine blood counts, may also serve as a predictive marker of immune checkpoint inhibitors’ (ICIs) efficacy. Further research is warranted to explore its association with tumor microenvironment indicators such as tumor-infiltrating lymphocytes, programmed cell death ligand 1 (PD-L1) expression, and neutrophil infiltration density[17]. Establishing these links could help elucidate the biological mechanisms through which the SIRI-PLR score predicts immunotherapy outcomes. Elevated neutrophil counts in GC are associated with increased infiltration of tumor-activated neutrophils that express immunosuppressive molecules, such as PD-L1, which suppress T-cell immunity and correlate with disease progression and poor survival[18]. Similarly, increased myeloid-derived suppressor cells in both peripheral blood and tumor tissues are linked to advanced disease stages and reduced survival, as these cells suppress T cell proliferation and effector function[19].

Zhou et al’s findings[8] could inform the use of ICIs in adjuvant settings. For example, even with PD-L1 positivity, patients with high SIRI-PLR scores may benefit from a combination of ICIs with chemotherapy or novel immunotherapies. Establishing the predictive value of this score for GC immunotherapy is an urgent research priority. In addition, inflammation is dynamic. Although this study only assessed the SIRI-PLR score preoperatively, its reproducibility remains an important consideration. Blood counts, particularly lymphocyte levels, may fluctuate depending on the specimen collection time or laboratory instrumentation. Therefore, future studies should investigate the repeatability of SIRI-PLR measurements using short-interval testing to ensure clinical reliability. Fluctuations during treatment can provide real-time insights into therapeutic response or early relapse, even before radiological detection[16]. Moving from a one-time snapshot to longitudinal monitoring can significantly enhance clinical utility.

CONCLUSION

A large-scale cohort study by Zhou et al[8] demonstrated that the SIRI-PLR score, derived from routine blood tests, is a simple and cost-effective prognostic tool that complements the TNM staging system. The accessibility of routine blood tests makes the SIRI-PLR score suitable for widespread clinical use, particularly in resource-limited settings. Beyond prognostication, this biomarker holds promise for predicting the therapeutic response to ICIs and for real-time monitoring of immune status during treatment. Prospective multicenter validation is necessary to establish its clinical utility. In the future, host-derived immune-inflammatory markers, such as SIRI-PLR, may be integrated with tumor genomics, liquid biopsy, and radiomic data to enable personalized care for patients with GC. The study by Zhou et al[8] represents an important step toward precision oncology.