mTLS status predicts survival benefit from adjuvant chemotherapy in gastric cancer patients treated with neoadjuvant therapy and surgery

Introduction

Gastric cancer (GC) ranks as the fifth most common malignancy and the fifth leading cause of cancer-related mortality worldwide, with nearly half of all cases occurring in Asia (1). Although radical gastrectomy with D2 lymphadenectomy is widely accepted as the standard surgical treatment for locally advanced gastric cancer (LAGC) (1), the prognosis for these patients remains generally poor (1,2).

Combining radical surgery with adjuvant chemotherapy (ACT) represents a key treatment modality for LAGC (3-6). Clinical trials such as ACTS-GC and CLASSIC have demonstrated the safety and efficacy of ACT (7-9), with the 5-year follow-up data from the JACCRO GC-07 study further confirming a survival benefit of ACT in stage III GC patients undergoing D2 dissection (10). In addition to ACT, neoadjuvant chemotherapy (NACT) has gained increasing interest due to its theoretical advantages—including tumor downstaging, eradication of micrometastases, improved treatment tolerance, and early assessment of chemosensitivity—thereby complementing postoperative treatment (11). The landmark MAGIC trial and the subsequent FLOT4 study demonstrated the efficacy of NACT in European populations (12-14). Moreover, studies such as RESOLVE, PRODIGY, and MATCH have highlighted the value of NACT in Asian patients (15-19).

Current guidelines from the National Comprehensive Cancer Network (NCCN) and the Chinese Society of Clinical Oncology (CSCO) recommend ACT for patients with esophageal or esophagogastric junction adenocarcinoma who achieve R0 resection after preoperative chemotherapy (20,21). Similarly, clinical trial protocols for NACT in GC often advocate subsequent ACT (22). However, real-world data indicate that only about half of the patients who undergo perioperative treatment actually initiate ACT, with completion rates (defined as receiving more than three cycles of ACT) ranging from 75% to 83% (14-18,23). These figures underscore the clinical challenges in delivering ACT after NACT and surgery, which may be attributed to factors such as malnutrition, immunosuppression, systemic inflammation, and organ dysfunction resulting from intensive NACT and extensive surgery (24,25).

Notably, several previous studies have reported that GC patients receiving preoperative chemotherapy and surgery derived limited survival benefit from ACT (26,27). Given the critical role of pathological staging and the postoperative tumor microenvironment in guiding ACT decisions, our study aims to evaluate the survival impact of ACT in LAGC patients—stratified by pathological stage and immune profile—who had received NACT followed by curative surgery. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-477/rc).

Methods

Patients

This study included 270 LAGC patients who received NACT followed by radical surgery at Zhongshan Hospital of Fudan University between January 1, 2010, and August 30, 2022. The inclusion criteria were as follows: (I) histologically confirmed gastric adenocarcinoma before initiating neoadjuvant treatment; (II) absence of distant metastasis or other unresectable factors at baseline; (III) completion of NACT followed by radical gastrectomy; and (IV) availability of clinicopathological information. The exclusion criteria were as follows: (I) patients who received neoadjuvant chemoradiation, neoadjuvant targeted therapy, or intraperitoneal chemotherapy; (II) patients who died within 1 month after the surgery; and (III) patients with other concurrent malignant tumors or other serious life-threatening diseases. According to the receipt of ACT, patients were divided into two groups: the ACT group and the no ACT group. Demographic characteristics, clinicopathological parameters, and survival information of each patient were collected retrospectively. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethical Committee of Zhongshan Hospital, Fudan University (No. B2023-229). Informed consent was obtained from all patients.

Treatment and follow-up

NACT regimens mainly included XELOX (oxaliplatin 130 mg/m² on day 1 of each cycle administered intravenously, oral capecitabine 1,000 mg/m² twice daily on days 1–14 of each 3-week cycle), DOS (docetaxel 75 mg/m² and oxaliplatin 130 mg/m² on day 1 of each cycle administered intravenously, oral S-1 a daily dose of 80, 100, or 120 mg on the basis of different body surface areas on days 1–14 of each 3-week cycle), SOX (oxaliplatin 130 mg/m² on day 1 of each cycle administered intravenously, oral S-1 a daily dose of 80, 100, or 120 mg based on different body surface areas on days 1–14 of each 3-week cycle) and FLOT [5-fluorouracil (5-FU) 2,600 mg/m² administered intravenously via peripherally inserted central catheter (PICC) for 24 hours on day 1, leucovorin 200 mg/m², oxaliplatin 85 mg/m², and docetaxel 50 mg/m² administered intravenously on day 1 of each 2-week cycle]. Radical gastrectomy with D2 lymph node dissection was scheduled 4–6 weeks after the last NACT. The indication for ACT was determined by the patient’s postoperative fitness and the pathological assessment of the resected specimen, incorporating the tumor regression grade (TRG), post-neoadjuvant therapy pathological tumor-node-metastasis (ypTNM) stage, and the presence of lymphovascular invasion (LVI) or perineural invasion (PNI). Patients in the ACT group received postoperative ACT for a duration of 3 to 6 months. The regimens used for ACT included the S-1 monotherapy regimen (S-1, 40–60 mg/m², administered orally twice daily on days 1–14 of each 3-week cycle), the XELOX regimen, the SOX regimen, the DOS regimen, the FLOT regimen, and the FOLFOX regimen (5-FU, 2,600 mg/m², administered intravenously via PICC over 24 hours on day 1, leucovorin 200 mg/m², and oxaliplatin 85 mg/m² administered intravenously on day 1 of each 2-week cycle).

Patients were followed up every 3 months for the first 2 years after surgery, every 6 months during the next 3 years, and yearly thereafter. Follow-up was completed on May 30th, 2024. The median duration of follow-up was 42.10 months [interquartile range (IQR), 25.90–69.20 months].

Pathological analysis

Paraffin-embedded surgical specimens were cut into serial sections with a thickness of 4 µm. Immunohistochemistry (IHC) was performed to detect the expression of CD19 (90176S, Cell Signaling Technology, Danvers, USA) and CD8 (D263403, Sangon Biotech, Shanghai, China). The staining procedure followed a standard IHC protocol, which included deparaffinization and rehydration of tissue sections, antigen retrieval, and blocking of endogenous peroxidase activity. Subsequently, sections were incubated with the corresponding primary antibodies at 4°C overnight. After washing, the sections were incubated with a horseradish peroxidase-conjugated secondary antibody (Servicebio, Wuhan, China). Color development was carried out using 3,3'-diaminobenzidine substrate (DAKO, Santa Clara, USA), followed by counterstaining with hematoxylin. The tumor bed was defined as the area that included the residual tumor, the tumor stroma, and the regression bed. Mature tertiary lymphoid structures (mTLSs) were identified based on morphological features observed on hematoxylin and eosin (H&E) stained slides, with additional confirmation through CD19 and CD8 IHC staining. H&E staining quantification was conducted on digitally scanned whole slide images (WSIs) employing Qupath software (v0.5.1).

Statistical analysis

Baseline body mass index (BMI) was defined as the BMI measured before the patient received the first cycle of NACT. Post-NACT BMI was defined as the BMI recorded after completing NACT but prior to undergoing curative gastrectomy. Although the majority of laboratory assessments were performed during a standardized follow-up at 1 month post-surgery, a minimal variation in sampling timing existed for a small subset of patients; all samples were drawn within the window of 2 weeks to 3 months after the curative gastrectomy. The calculation formula for the Systemic Immune-Inflammation Index (SII) is SII = platelet count (×10⁹/L) × neutrophil count (×10⁹/L) / lymphocyte count (×10⁹/L). The Prognostic Nutritional Index is calculated as serum albumin (g/L) + 5 × total peripheral blood lymphocyte count (×10⁹/L).

Clinicopathological features were reported as either the mean and range or as frequency and proportion, as appropriate. The differences in clinical and pathological characteristics between the two groups were analyzed using either the unpaired Student’s t-test or the chi-square test. Overall survival (OS) curves were generated using the Kaplan-Meier (KM) method, and differences in OS based on the receipt of ACT were compared using the log-rank test in the entire cohort. Subgroup analyses were also performed based on various pathological characteristics, including ypT stage, ypN stage, LVI, PNI, TRG, and infiltration of tertiary lymphoid structures. In each subgroup, the OS differences between the two groups were similarly evaluated using the KM method and the log-rank test.

The propensity score method was used to minimize the potential bias arising from confounding covariates using propensity score matching (PSM) in SPSS. The variables that showed statistically significant differences between the two groups in Table 1, including sex, age, LVI, and NACT cycles, were selected as matching factors. To maximize the number of participants, the Nearest Neighbor Matching algorithm was applied with a caliper value of 0.2, using a 1:2 matching ratio between the no ACT group and the ACT group. After completing the matching process, a subsequent analysis of differences between the two matched groups was conducted, and no statistically significant differences were found in any of the clinicopathological characteristics (Figure S1).

A multivariable Cox proportional hazards model was constructed by incorporating relevant prognostic factors, including lymph node metastasis, LVI, signet ring cell (SRC) component, and Lauren classification. To evaluate the predictive role of mTLS, an interaction term between ACT and mTLS status was introduced into the Cox model. The significance of the interaction was assessed using the likelihood ratio test. Based on the interaction test results, patients were stratified according to mTLS status. Within each subgroup, adjusted Cox models were built to calculate the hazard ratio (HR) and its 95% confidence interval (CI) for ACT. Forest plots were used to visually present the adjusted HRs for ACT across the subgroups.

All P values were two-sided, and a P value of less than 0.05 denoted statistical significance. The statistical analysis was performed with R (4.0.3), SPSS software (SPSS Statistics, version 29) and Prism (10.0.3).

Results

Differences in clinicopathological characteristics and prognosis between the ACT group and the no ACT group

The patient selection process was illustrated in Figure 1A. Among the 270 enrolled patients, 66 (24.4%) did not receive ACT, and 204 (75.6%) received ACT. The differences in clinical and pathological characteristics between the two groups are summarized in Table 1. The ACT group had a significantly higher proportion of patients who received more than 3 cycles of NACT and a higher frequency of LVI positivity compared to the no ACT group (P=0.03 and P=0.04, respectively).

Figure 1 Cohort selection and survival analysis by ACT status. Flowchart of cohort selection (A). KM survival curves of patients with or without ACT in the total cohort (B) and the PSM cohort (C). ACT, adjuvant chemotherapy; KM, Kaplan-Meier; PSM, propensity score matching.

After PSM, 64 patients remained in the ACT group and 116 in the no-ACT group. As shown in Table 2, no significant differences in clinicopathological characteristics were observed between the two groups after matching.

OS did not differ significantly between the ACT and no ACT groups, either in the overall cohort (P=0.36, Figure 1B) or in the PSM-matched cohort (P=0.46, Figure 1C).

Impact of ACT on survival in different pathological subgroups

In patients with relatively early postoperative pathological staging, such as ypT0–2 (Figure 2A, left) and ypN0 (Figure 2B, left), subgroup KM analysis revealed no significant effect of ACT on OS (P=0.79 and P=0.79, respectively). Among patients with postoperative pathological features indicating high risk of recurrence and metastasis, such as ypT3–4 (Figure 2A, right) and ypN+ (Figure 2B, right), OS did not differ significantly between the ACT and no ACT groups (P=0.50 and P=0.54, respectively).

Figure 2 KM survival curves of patients with different ypT stage (A) and ypN stage (B) stratified by the receipt of ACT. ACT, adjuvant chemotherapy; KM, Kaplan-Meier; ypN, post-neoadjuvant therapy pathological node; ypT, post-neoadjuvant therapy pathological tumor.

GC of different histological subtypes often exhibit distinct biological behaviors. We stratified the PSM cohort according to the Lauren classification into intestinal and diffuse/mixed subtypes and compared survival between ACT and no ACT subgroups. The KM analysis indicated no statistically significant difference in OS between the ACT and no ACT groups, either in the intestinal-type (Figure 3A, left, P=0.16) or the diffuse/mixed-type (Figure 3A, right, P=0.16) subgroups. LVI and PNI are two commonly used indicators of tumor invasiveness. In the PSM cohort, neither the LVI (+) nor LVI (−) groups derived a significant survival benefit from ACT (Figure 3B). This finding was consistently observed across subgroups stratified by PNI status (Figure 3C).

Figure 3 KM survival curves of patients with different histological subtypes (A), LVI status (B), and PNI status (C) stratified by the receipt of ACT. ACT, adjuvant chemotherapy; KM, Kaplan-Meier; LVI, lymphovascular invasion; PNI, perineural invasion.

In addition, we evaluated the impact of ACT on survival according to the extent of tumor regression after NACT. ACT did not confer a significant survival benefit to GC patients, regardless of TRG (Figure 4A-4D).

Figure 4 KM survival curves of patients with Becker-TRG 1a (pCR) (A), Becker-TRG 1b (MPR) (B), Becker-TRG 2 (C), and Becker-TRG 3 (D) stratified by the receipt of ACT. Becker-TRG 1a = no residual tumor cells; Becker-TRG 1b = residual viable tumor less than 10%; Becker-TRG 2 = residual viable tumor 10–50%; Becker-TRG 3 = residual viable tumor more than 50%. ACT, adjuvant chemotherapy; KM, Kaplan-Meier; MPR, major pathological response; pCR, pathological complete response; TRG, tumor regression grade.

Impact of ACT on survival in patients with or without mTLS

We collected H&E-stained slides of primary gastric lesions from patients in the PSM cohort and, based on slide availability, established an mTLS cohort (n=82). On H&E-stained sections, we observed mTLSs with prominent germinal centers. These structures were further validated by IHC staining, which revealed dense CD19⁺ cell clusters in the center of the identified mTLSs and diffuse CD8⁺ T cell infiltration in the surrounding areas (Figure 5A).

Figure 5 Impact of ACT on survival in patients with or without mTLS. Representative H&E-stained image of a typical mTLS and corresponding expression of CD19 and CD8 (A). KM survival curves comparing OS between the ACT and no ACT groups in the mTLS cohort (B). KM survival curves comparing OS between mTLS-positive and -negative patients in the mTLS cohort (C). KM survival curves comparing the ACT and no ACT groups among mTLS negative (D) and mTLS positive (E) patients. Forest plot showing adjusted hazard ratios for ACT, stratified by mTLS status (F). *, P<0.05; **, P<0.01. ACT, adjuvant chemotherapy; H&E, hematoxylin and eosin; KM, Kaplan-Meier; mTLS, mature tertiary lymphoid structures; OS, overall survival.

In the mTLS cohort, no significant differences were observed in clinicopathological characteristics (Table 3) or OS between the ACT and no ACT groups (P=0.40, Figure 5B). We also assessed the distribution of mTLSs across various pathological features and found that a better tumor regression was associated with a greater proportion of mTLS-positive cases (Figure S2A, P=0.03). However, no correlations were detected between mTLS presence and ypN stage or Lauren classification (Figure S2B,S2C). Furthermore, within the mTLS cohort, subgroup analyses stratified by TRG, ypN status, and LVI still revealed no statistically significant survival difference between ACT and no ACT groups (Figure S2D-S2F).

Based on the presence of mTLS in the tumor bed, patients in the mTLS cohort were classified into mTLS-positive and -negative groups. No significant difference in OS was observed between these two subgroups (P=0.38, Figure 5C). However, within the mTLS negative subgroup, patients who received ACT showed significantly better OS than those who did not (P=0.006, Figure 5D). Conversely, among mTLS positive patients, the no ACT group exhibited superior OS (P=0.02, Figure 5E).

We further evaluated the interaction between mTLS status and ACT using a likelihood ratio test, which indicated a statistically significant interaction effect (P<0.001, Figure 5F). In a stratified multivariable Cox regression analysis, ACT was associated with significantly worse OS in mTLS-positive patients (HR =3.127, 95% CI: 1.009–9.691, P=0.048), whereas it conferred a significant survival benefit in mTLS-negative patients (HR =0.02, 95% CI: 0.001–0.623, P=0.03) (Figure 5F).

Discussion

The primary goal of ACT is to eliminate residual micrometastases following surgical resection, thereby reducing the risk of recurrence and metastasis and improving the prognosis of patients with cancer (28). While the survival benefit of ACT for LAGC is well-established, perioperative chemotherapy, which combines NACT and ACT, has become the standard treatment for these patients (21,29). Nevertheless, current major clinical guidelines do not provide specific recommendations on which patients should receive ACT after NACT or which may safely forgo it (6,20,21,29). Since NACT itself aims to eliminate micrometastases and evaluate chemosensitivity, it remains unclear whether all patients completing NACT still require postoperative ACT, and if so, who would derive meaningful benefit. To date, no prospective studies have directly compared NACT plus ACT versus NACT alone.

In this retrospective study evaluating 180 LAGC patients who underwent NACT and R0 gastrectomy, we found that ACT did not significantly improve OS in the overall cohort. This observation is consistent with the 10-year follow-up results of the SAKK 43/99 trial, which reported no survival difference between patients receiving only preoperative or only postoperative docetaxel-based chemotherapy (30). Similarly, two studies based on the National Cancer Database (NCDB) indicated that patients in the United States who received preoperative chemotherapy and gastrectomy derived limited OS benefit from additional ACT (26,27). In contrast, another NCDB analysis suggested that ACT following neoadjuvant chemoradiation and surgery improved OS compared to observation alone in gastroesophageal adenocarcinoma. However, this study included patients with gastroesophageal adenocarcinoma undergoing neoadjuvant chemoradiation, and its generalizability to GC patients undergoing NACT remains uncertain. In addition, patients who underwent R1 resection were not excluded from this study (31).

In clinical practice, pathological features such as tumor infiltration depth, lymph node metastasis, histological subtype, LVI, and PNI often inform ACT decisions. TRG serves as a key efficacy endpoint of NACT, a strong prognostic marker, and a potential guide for postoperative treatment strategies. However, in our study, stratification based on these pathological characteristics did not reveal any subgroup that derived significant survival benefit from ACT. Previous studies have indicated that the SRC component percentage is associated with response to NACT and serves as an independent prognostic factor in GC (32-34). In our cohort, however, accurate assessment of SRC percentage was not feasible due to tumor regression after NACT. Therefore, our analysis was based on Lauren classification, which showed no significant OS benefit from ACT in either the intestinal or diffuse/mixed subtypes. We plan to incorporate baseline and post-treatment SRC data in future studies to further explore this aspect. A multicenter retrospective analysis by Lin et al. suggested that ACT after NACT and surgery may be particularly beneficial in patients with a lymph node ratio (LNR) ≥9% (35). Compared to that study, our cohort had a lower proportion of patients with pathologically positive lymph nodes (57.6% vs. 63.2–70%), which may partly explain the discrepancy in conclusions (31,35).

Notably, we found that the presence or absence of mTLS in the post-treatment tumor bed may help identify patients who are more or less likely to benefit from ACT. Although the tumor immune microenvironment represents a complex interplay of multiple factors, mTLS status in our cohort was only associated with TRG but not with other key prognostic features such as ypN stage or Lauren classification—none of which were predictive of ACT benefit. Further likelihood ratio tests and multivariable Cox regression confirmed that the effect of ACT on OS depended significantly on mTLS status, establishing mTLS as a strong predictive biomarker for ACT efficacy. The presence of mTLSs may identify a patient subgroup with a distinct immune microenvironment who not only fail to benefit from ACT but may potentially be adversely affected by it.

We observed that a subset of patients who did not receive ACT still exhibited favorable outcomes, prompting us to explore the characteristics of this group. This phenomenon is reminiscent of the established role of mismatch repair-deficient (dMMR)/microsatellite instability-high (MSI-H) status as a biomarker of favorable prognosis and response to immunotherapy in GC (36,37). Similar to how dMMR/MSI-H identifies tumors with a highly immunogenic phenotype that may not benefit from—or could even be harmed by—conventional chemotherapy (38), we propose that mTLS might serve as an analogous histological marker of a robust, pre-existing anti-tumor immune response. mTLSs are ectopic lymphoid aggregates that develop in non-lymphoid tissues such as tumors and play a crucial role in anti-tumor immune responses (39). The presence of mTLSs, characterized by germinal centers, has been confirmed to be associated with a favorable prognosis and treatment response in GC patients (40,41). In the context of NACT, which remodels the tumor microenvironment by inducing immunogenic cell death and altering immune cell infiltration (42,43), evidence from other cancers suggests that NACT can promote mTLS formation and maturation (44-47). However, the utility of mTLS for guiding post-neoadjuvant ACT has remained largely unexplored. To our knowledge, our study is the first to demonstrate that mTLS status following NACT can stratify survival benefit from subsequent ACT in GC. Both mTLS and dMMR/MSI-H may serve as biomarkers to identify a patient subgroup with pre-existing potent anti-tumor immunity. For these patients, additional cytotoxic treatment may be unnecessary and could potentially be detrimental. Given that mTLS assessment can be performed using routinely available H&E-stained slides without imposing significant burdens on clinical pathology workflows, incorporation of mTLS scoring could serve as a feasible and low-cost decision-support tool for de-escalating ACT in the post-NACT setting. This approach may facilitate more personalized treatment selection and help avoid unnecessary chemotherapy in mTLS positive patients. Furthermore, the current lack of evidence supporting adjuvant immune checkpoint inhibitors (ICIs) after GC surgery may be related to the removal of lymph nodes—key sites for anti-tumor immune responses—during standard D2 dissection. In this context, patients lacking mTLS may rely more on systemic chemotherapy to eliminate residual disease, whereas those with mTLS may already possess potent local and systemic immune surveillance, which could be compromised by chemotherapy-related immunosuppression but potentially enhanced by ICIs. In future clinical trials, mTLS status could be considered a stratification factor for deciding on ACT or exploring ICI-based regimens as an alternative to chemotherapy.

In some cases, inadequate tumor regression or postoperative recurrence may be attributed to chemoresistance. For such patients, alternative postoperative strategies—including immunotherapy or targeted therapy—should be considered. In our PSM cohort, 85.7% (36/42) of patients with pathological complete response (pCR) and major pathological response (MPR) had received ≥3 cycles of NACT. We believe that for these patients with a favorable pathological response, adequate preoperative chemotherapy may have already achieved the intended therapeutic effect. Moreover, the cumulative toxicity of chemotherapy, surgical stress, and postoperative complications often leads to deteriorated immune and nutritional status following NACT and gastrectomy. A considerable proportion of patients may be physically unable to tolerate ACT, and the survival benefit of such treatment in this setting remains uncertain. Based on our findings, evaluating post-NACT mTLS status may help identify patients who can safely forgo ACT, thereby improving postoperative quality of life and reducing treatment-related morbidity—particularly among elderly or frail individuals.

However, several limitations of this study should be considered. First, the relatively small sample size and limited follow-up duration may have limited the statistical power, and residual confounding factors might persist despite the use of PSM. Second, due to the retrospective design, detailed information regarding chemotherapy dose reductions, treatment delays, and actual dose intensity was not systematically collected or analyzed. Third, the relatively low proportion of ypN(+) patients in our cohort may have influenced the assessment of survival differences within this subgroup, potentially limiting the generalizability of our findings. Furthermore, more comprehensive profiling of the post-treatment tumor immune microenvironment was not feasible, which restricted our ability to identify additional biomarkers associated with ACT response. Future prospective studies with larger cohorts are warranted to validate these findings and to explore the underlying mechanisms in greater depth.

Conclusions

In summary, our study demonstrated that the presence of mTLS in the tumor bed following NACT serves as a key predictor for survival benefit from ACT in GC patients. Specifically, patients with mTLS did not derive significant survival improvement from ACT, whereas those without mTLS appeared to benefit from ACT. These findings suggest that mTLS status could help guide personalized treatment strategies after NACT and surgery. However, these results should be interpreted as preliminary evidence, and further well-controlled prospective studies are warranted to validate their clinical applicability.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-477/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-477/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-477/prf

Funding: This work was supported by Project of Science and Technology of Xiamen City (No. 3502Z20224015).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-477/coif). All authors report that this work was supported by Project of Science and Technology of Xiamen City (No. 3502Z20224015). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethical Committee of Zhongshan Hospital, Fudan University (No. B2023-229). Informed consent was obtained from all patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sundar R, Nakayama I, Markar SR, et al. Gastric cancer. Lancet 2025;405:2087-102. [Crossref] [PubMed]
2. Huang C, Liu H, Hu Y, et al. Laparoscopic vs Open Distal Gastrectomy for Locally Advanced Gastric Cancer: Five-Year Outcomes From the CLASS-01 Randomized Clinical Trial. JAMA Surg 2022;157:9-17. [Crossref] [PubMed]
3. Shitara K, Fleitas T, Kawakami H, et al. Pan-Asian adapted ESMO Clinical Practice Guidelines for the diagnosis, treatment and follow-up of patients with gastric cancer. ESMO Open 2024;9:102226. [Crossref] [PubMed]
4. Kim IH, Kang SJ, Choi W, et al. Korean Practice Guidelines for Gastric Cancer 2024: An Evidence-based, Multidisciplinary Approach (Update of 2022 Guideline). J Gastric Cancer 2025;25:5-114. [Crossref] [PubMed]
5. Wang FH, Zhang XT, Tang L, et al. The Chinese Society of Clinical Oncology (CSCO): Clinical guidelines for the diagnosis and treatment of gastric cancer, 2023. Cancer Commun (Lond) 2024;44:127-72. [Crossref] [PubMed]
6. Japanese Gastric Cancer Treatment Guidelines 2021 (6th edition). Gastric Cancer 2023;26:1-25.
7. Sakuramoto S, Sasako M, Yamaguchi T, et al. Adjuvant chemotherapy for gastric cancer with S-1, an oral fluoropyrimidine. N Engl J Med 2007;357:1810-20. [Crossref] [PubMed]
8. Sasako M, Sakuramoto S, Katai H, et al. Five-year outcomes of a randomized phase III trial comparing adjuvant chemotherapy with S-1 versus surgery alone in stage II or III gastric cancer. J Clin Oncol 2011;29:4387-93. [Crossref] [PubMed]
9. Noh SH, Park SR, Yang HK, et al. Adjuvant capecitabine plus oxaliplatin for gastric cancer after D2 gastrectomy (CLASSIC): 5-year follow-up of an open-label, randomised phase 3 trial. Lancet Oncol 2014;15:1389-96. [Crossref] [PubMed]
10. Ichikawa W, Yoshida K, Kodera Y, et al. Five-year overall survival with S-1 plus docetaxel as adjuvant treatment in curatively resected pStage III gastric cancer in JACCRO GC-07. J Clin Oncol 2023;41:4065. [Crossref]
11. De Vita F, Giuliani F, Galizia G, et al. Neo-adjuvant and adjuvant chemotherapy of gastric cancer. Ann Oncol 2007;18:vi120-3. [Crossref] [PubMed]
12. Cunningham D, Allum WH, Stenning SP, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006;355:11-20. [Crossref] [PubMed]
13. Al-Batran SE, Hofheinz RD, Pauligk C, et al. Histopathological regression after neoadjuvant docetaxel, oxaliplatin, fluorouracil, and leucovorin versus epirubicin, cisplatin, and fluorouracil or capecitabine in patients with resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4-AIO): results from the phase 2 part of a multicentre, open-label, randomised phase 2/3 trial. Lancet Oncol 2016;17:1697-708. [Crossref] [PubMed]
14. Al-Batran SE, Homann N, Pauligk C, et al. Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4): a randomised, phase 2/3 trial. Lancet 2019;393:1948-57. [Crossref] [PubMed]
15. Kang YK, Yook JH, Park YK, et al. PRODIGY: A Phase III Study of Neoadjuvant Docetaxel, Oxaliplatin, and S-1 Plus Surgery and Adjuvant S-1 Versus Surgery and Adjuvant S-1 for Resectable Advanced Gastric Cancer. J Clin Oncol 2021;39:2903-13. [Crossref] [PubMed]
16. Kang YK, Kim HD, Yook JH, et al. Neoadjuvant Docetaxel, Oxaliplatin, and S-1 Plus Surgery and Adjuvant S-1 for Resectable Advanced Gastric Cancer: Updated Overall Survival Outcomes From Phase III PRODIGY. J Clin Oncol 2024;42:2961-5. [Crossref] [PubMed]
17. Wang X, Lu C, Wei B, et al. Perioperative versus adjuvant S-1 plus oxaliplatin chemotherapy for stage II/III resectable gastric cancer (RESONANCE): a randomized, open-label, phase 3 trial. J Hematol Oncol 2024;17:17. [Crossref] [PubMed]
18. Zhang X, Liang H, Li Z, et al. Perioperative or postoperative adjuvant oxaliplatin with S-1 versus adjuvant oxaliplatin with capecitabine in patients with locally advanced gastric or gastro-oesophageal junction adenocarcinoma undergoing D2 gastrectomy (RESOLVE): final report of a randomised, open-label, phase 3 trial. Lancet Oncol 2025;26:312-9. [Crossref] [PubMed]
19. Jiang Z, Xie Y, Zhang W, et al. Perioperative chemotherapy with docetaxel plus oxaliplatin and S-1 (DOS) versus oxaliplatin plus S-1 (SOX) for the treatment of locally advanced gastric or gastro-esophageal junction adenocarcinoma (MATCH): an open-label, randomized, phase 2 clinical trial. Gastric Cancer 2024;27:571-9. [Crossref] [PubMed]
20. Ajani JA, D'Amico TA, Bentrem DJ, et al. Gastric Cancer, Version 2.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2022;20:167-92. [Crossref] [PubMed]
21. Wang FH, Zhang XT, Tang L, et al. The Chinese Society of Clinical Oncology (CSCO): Clinical guidelines for the diagnosis and treatment of gastric cancer, 2023. Cancer Commun (Lond) 2024;44:127-72. [Crossref] [PubMed]
22. Aggelis V, Cunningham D, Lordick F, et al. Peri-operative therapy for operable gastroesophageal adenocarcinoma: past, present and future. Ann Oncol 2018;29:1377-85. [Crossref] [PubMed]
23. Wang X, Li S, Sun Y, et al. The protocol of a prospective, multicenter, randomized, controlled phase III study evaluating different cycles of oxaliplatin combined with S-1 (SOX) as neoadjuvant chemotherapy for patients with locally advanced gastric cancer: RESONANCE-II trial. BMC Cancer 2021;21:20. [Crossref] [PubMed]
24. Seong YW, Han SJ, Jung W, et al. Perioperative change in neutrophil-to-lymphocyte ratio (NLR) is a prognostic factor in patients with completely resected primary pulmonary sarcomatoid carcinoma. J Thorac Dis 2019;11:819-26. [Crossref] [PubMed]
25. Herzberg H, Lifshitz K, Golan S, et al. Association between early change in neutrophil-to-lymphocyte ratio after radical cystectomy and treatment outcomes. BJU Int 2022;130:470-7. [Crossref] [PubMed]
26. Drake JA, Stiles ZE, Tsao MW, et al. Analysis of the Survival Impact of Postoperative Chemotherapy After Preoperative Chemotherapy and Resection for Gastric Cancer. Ann Surg Oncol 2021;28:1417-27. [Crossref] [PubMed]
27. Deng L, Groman A, Jiang C, et al. Association of Preoperative Chemosensitivity With Postoperative Survival in Patients With Resected Gastric Adenocarcinoma. JAMA Netw Open 2021;4:e2135340. [Crossref] [PubMed]
28. Kotani D, Oki E, Nakamura Y, et al. Molecular residual disease and efficacy of adjuvant chemotherapy in patients with colorectal cancer. Nat Med 2023;29:127-34. [Crossref] [PubMed]
29. Lordick F, Carneiro F, Cascinu S, et al. Gastric cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 2022;33:1005-20. [Crossref] [PubMed]
30. Fazio N, Biffi R, Maibach R, et al. Preoperative versus postoperative docetaxel-cisplatin-fluorouracil (TCF) chemotherapy in locally advanced resectable gastric carcinoma: 10-year follow-up of the SAKK 43/99 phase III trial. Ann Oncol 2016;27:668-73. [Crossref] [PubMed]
31. Mokdad AA, Yopp AC, Polanco PM, et al. Adjuvant Chemotherapy vs Postoperative Observation Following Preoperative Chemoradiotherapy and Resection in Gastroesophageal Cancer: A Propensity Score-Matched Analysis. JAMA Oncol 2018;4:31-8. [Crossref] [PubMed]
32. Roviello F, Marano L, Ambrosio MR, et al. Signet ring cell percentage in poorly cohesive gastric cancer patients: A potential novel predictor of survival. Eur J Surg Oncol 2022;48:561-9. [Crossref] [PubMed]
33. Marano L, Ambrosio MR, Resca L, et al. The Percentage of Signet Ring Cells Is Inversely Related to Aggressive Behavior and Poor Prognosis in Mixed-Type Gastric Cancer. Front Oncol 2022;12:897218. [Crossref] [PubMed]
34. Sorrenti S, Malerba S, Lori E, et al. Impact of Neoadjuvant Chemotherapy on Survival Outcomes in Gastric Signet-Ring Cell Carcinoma. Cancers (Basel) 2025;17:2400. [Crossref] [PubMed]
35. Lin JX, Tang YH, Lin GJ, et al. Association of Adjuvant Chemotherapy With Overall Survival Among Patients With Locally Advanced Gastric Cancer After Neoadjuvant Chemotherapy. JAMA Netw Open 2022;5:e225557. [Crossref] [PubMed]
36. Ratti M, Lampis A, Hahne JC, et al. Microsatellite instability in gastric cancer: molecular bases, clinical perspectives, and new treatment approaches. Cell Mol Life Sci 2018;75:4151-62. [Crossref] [PubMed]
37. Puliga E, Corso S, Pietrantonio F, et al. Microsatellite instability in Gastric Cancer: Between lights and shadows. Cancer Treat Rev 2021;95:102175. [Crossref] [PubMed]
38. Kim JW, Cho SY, Chae J, et al. Adjuvant Chemotherapy in Microsatellite Instability-High Gastric Cancer. Cancer Res Treat 2020;52:1178-87. [Crossref] [PubMed]
39. Sautès-Fridman C, Petitprez F, Calderaro J, et al. Tertiary lymphoid structures in the era of cancer immunotherapy. Nat Rev Cancer 2019;19:307-25. [Crossref] [PubMed]
40. Yin YX, Ling YH, Wei XL, et al. Impact of mature tertiary lymphoid structures on prognosis and therapeutic response of Epstein-Barr virus-associated gastric cancer patients. Front Immunol 2022;13:973085. [Crossref] [PubMed]
41. Li Z, Jiang Y, Li B, et al. Development and Validation of a Machine Learning Model for Detection and Classification of Tertiary Lymphoid Structures in Gastrointestinal Cancers. JAMA Netw Open 2023;6:e2252553. [Crossref] [PubMed]
42. Xing X, Shi J, Jia Y, et al. Effect of neoadjuvant chemotherapy on the immune microenvironment in gastric cancer as determined by multiplex immunofluorescence and T cell receptor repertoire analysis. J Immunother Cancer 2022;10:e003984. [Crossref] [PubMed]
43. Tang R, Xu J, Wang W, et al. Targeting neoadjuvant chemotherapy-induced metabolic reprogramming in pancreatic cancer promotes anti-tumor immunity and chemo-response. Cell Rep Med 2023;4:101234. [Crossref] [PubMed]
44. Sun X, Liu W, Sun L, et al. Maturation and abundance of tertiary lymphoid structures are associated with the efficacy of neoadjuvant chemoimmunotherapy in resectable non-small cell lung cancer. J Immunother Cancer 2022;10:e005531. [Crossref] [PubMed]
45. Lu Y, Zhao Q, Liao JY, et al. Complement Signals Determine Opposite Effects of B Cells in Chemotherapy-Induced Immunity. Cell 2020;180:1081-1097.e24. [Crossref] [PubMed]
46. Germain C, Gnjatic S, Tamzalit F, et al. Presence of B cells in tertiary lymphoid structures is associated with a protective immunity in patients with lung cancer. Am J Respir Crit Care Med 2014;189:832-44. [Crossref] [PubMed]
47. Kuwabara S, Tsuchikawa T, Nakamura T, et al. Prognostic relevance of tertiary lymphoid organs following neoadjuvant chemoradiotherapy in pancreatic ductal adenocarcinoma. Cancer Sci 2019;110:1853-62. [Crossref] [PubMed]