Efficacy and safety of sintilimab combined with nab-paclitaxel plus S-1 for neoadjuvant treatment of locally advanced gastric cancer

Abstract

BACKGROUND

Gastric cancer is a leading global cause of cancer mortality, with poor survival in locally advanced stages. While immune checkpoint inhibitors (ICIs) like sintilimab have improved outcomes in advanced disease, their role as neoadjuvant therapy remains understudied. This study investigates sintilimab combined with nab-paclitaxel/S-1 as preoperative treatment for locally advanced gastric cancer (LAGC), addressing an unmet need for effective neoadjuvant strategies.

AIM

To explore the efficacy and safety of combination treatment with sintilimab and nab-paclitaxel plus S-1 as neoadjuvant therapy for LAGC.

METHODS

Clinical data from 82 patients diagnosed with LAGC, who underwent preoperative treatment and surgery between April 2020 and December 2022, were included. Patients were divided into 2 groups according to treatment regimen: ICI (sintilimab + nab-paclitaxel + S-1; and non-ICI (nab-paclitaxel + S-1). Imaging and pathological efficacy, intra- and postoperative conditions, molecular subtypes, short-term survival outcomes, and safety were compared between the 2 groups.

RESULTS

Imaging evaluation of therapeutic efficacy revealed that the inclusion of ICI yielded a significantly higher complete response rate (13.2% vs 0.0%; P = 0.048), and objective response rate (69.8% vs 31.0%, P = 0.001) compared with non-ICI treatment. Pathological evaluation revealed that the ICI group exhibited a significantly higher pathological complete response rate (13.2% vs 0.0%; P = 0.048) and major pathological response rate (35.8% vs 13.8%; P = 0.041) than those in the non-ICI group. The two-year disease-free survival rate in the ICI group was greater than that in the non-ICI group (83.0% vs 55.2%; P = 0.043). The use of ICI did not increase the incidence of adverse reactions (47.2% vs 41.4%; P = 0.614) or perioperative adverse events (18.9% vs 13.8%; P = 0.761).

CONCLUSION

The combination of sintilimab with nab-paclitaxel + S-1 for neoadjuvant treatment of LAGC improved efficacy in patients without increasing adverse drug reactions and perioperative adverse events, suggesting that this treatment regimen is safe and feasible.

Key Words: Gastric cancer; Neoadjuvant therapy; Immune checkpoint inhibitor; Efficacy; Safety

Core Tip: This study investigated sintilimab (an immune checkpoint inhibitor) plus nab-paclitaxel and S-1 as neoadjuvant therapy for locally advanced gastric cancer. The combination significantly improved pathological complete response and major pathological response rates, while also demonstrating superior 2-year disease-free survival compared to chemotherapy alone, without additional safety concerns. These findings support its potential as an effective preoperative treatment strategy for locally advanced gastric cancer.

INTRODUCTION

Gastric cancer is the fifth most common malignant tumor and a major cause of cancer-related death worldwide[1]. Surgery remains the mainstay of early gastric cancer treatment; however, in the early stage(s) of gastric cancer, symptoms are not always obvious, and many patients are already in the middle or late stage when diagnosed[2]. Furthermore, pure surgical treatment has poor efficacy for patients diagnosed with locally advanced gastric cancer (LAGC) and often needs to be combined with postoperative adjuvant therapy to improve prognosis[3]. However, the five-year survival rate of patients with LAGC remains low, and there is an urgent need for new treatment modes and means to improve patient prognosis. Multiple studies[4,5] have shown that, compared with simple surgery followed by adjuvant therapy, neoadjuvant therapy can downstage the tumor, increase the R0 resection rate, and ultimately improve prognosis and prolong survival in patients with LAGC. Immune checkpoint inhibitors (ICIs) have emerged as a new antitumor therapy in recent years, demonstrating promising clinical efficacy against melanoma, lung cancer, gastric cancer, and other cancers[6]. Studies, such as CheckMate-649[7], ATTRACTION-4[8], and KEYNOTE-059[9], have demonstrated that ICI combined with chemotherapy significantly improve the prognosis of advanced gastric cancer patients with manageable toxic reactions. Recently, exploratory studies have been conducted on the perioperative treatment of LAGC using ICI combined with chemotherapy.

André et al[10] reported that neoadjuvant treatment of DNA deficient mismatch repair (dMMR)/microsatellite instability-high LAGC with nivolumab in combination with ipilimumab resulted in a high pathological complete response (pCR) rate without unexpected toxicity. Sun et al[11] reported a nationwide multicenter retrospective study confirming the short-term efficacy and safety of neoadjuvant treatment of LAGC with ICIs in combination with chemotherapy. Cui et al[12] also reported that neoadjuvant treatment of LAGC with ICI in combination with chemotherapy neoadjuvant treatment for LAGC resulted in better pathological outcomes without an increase in associated drug toxicities. A meta-analysis concluded that ICI in combination with chemotherapy was safe and effective in patients with LAGC and may be a more desirable neoadjuvant treatment strategy[13]. However, these studies involved a large number of ICI drug classes and chemotherapy regimens and did not perform subgroup analyses of different ICI classes or chemotherapy regimens. Sintilimab - an ICI - has demonstrated good efficacy and tolerability in patients with advanced gastric cancer[14,15], and the ORIENT-16[16] study established the efficacy of sintilimab in the treatment of advanced gastric cancer. The efficacy and safety of sintilimab for perioperative treatment of LAGC, however, have not been reported in real-world clinical settings. As such, the present study aimed to investigate the efficacy and safety of sintilimab in combination with nab-paclitaxel + S-1 chemotherapy for the preoperative treatment of LAGC.

MATERIALS AND METHODS

Patients

This retrospective cohort study evaluated 82 consecutive patients with LAGC treated at Zhangzhou Affiliated Hospital of Fujian Medical University from April 2020 to December 2022. All patients received neoadjuvant therapy followed by radical gastrectomy after completing 2-4 treatment cycles. The patient screening flowchart is shown in Figure 1. The inclusion criteria were pathological diagnosis of gastric adenocarcinoma and clinically staged as stage III-IVa. The tumor-node-metastasis (TNM) staging is carried out according to the 8^th edition of the American Joint Committee on Cancer/TNM staging system[17]. Individuals with a history of other malignant tumors and those who had received treatment with targeted drugs or other antitumor medications were excluded.

Figure 1 Flow diagram of the patient selection process. ICI: Immune checkpoint inhibitor.

Perioperative chemotherapy and immunotherapy

The dose of nab-paclitaxel was 260 mg/m², administered by intravenous infusion once every 3 weeks. On day 1, S-1 was administered twice daily for 14 days. A single dose of a 5-fluorouracil derivative (S-1) was calculated based on patient body surface area (BSA), as follows: BSA < 1.25 m², 40 mg per time; BSA ≥ 1.25 m² and < 1.5 m², 50 mg per time; and BSA ≥ 1.5 m², 60 mg per time. There was a seven-day interval, with 1 treatment cycle lasting 21 days. Sintilimab was administered as a fixed 200 mg dose in 1-hour IV infusions. Treatment-related adverse events were managed through protocol-defined interventions: Dose reduction (25% decrement to 150 mg), treatment interruption (median 7 days), or supportive care. These measures enabled all affected patients to resume therapy and complete subsequent cycles without further complications. Chemotherapy-related events were recorded in accordance with the Common Terminology Criteria for Adverse Events system[18].

Surgery

Both treatment groups underwent standardized radical gastrectomy following completion of 2-4 therapy cycles. All surgical procedures were performed by the same surgical team, with the specific operative approach (total vs subtotal gastrectomy) determined by tumor characteristics (location and size) and patient anatomy. When clinically indicated, en bloc resection of adjacent involved organs was performed to achieve R0 resection. In accordance with the Sixth Edition of the Japanese Gastric Cancer Treatment Guidelines[19], D2 lymph node dissection was performed. Postoperative complications were surgery-related complications that occurred within 30 days after surgery and their severity were assessed according to the Clavien-Dindo grading system[20,21].

Endpoints

The main study endpoint was tumor regression grading (TRG) [pCR and major pathological response (MPR)], and the secondary study endpoints included objective response rate (ORR), disease-free survival (DFS), overall survival (OS), TNM staging, R0 resection, drug-related adverse reactions, and perioperative treatment.

Pathological response

TRG was performed according to the Becker criteria[22,23]. pCR = (number of patients with TRG grade 1a)/total number of patients × 100. The pCR refers to the complete regression of both the primary tumor and regional lymph nodes. MPR = (number of patients with TRG1a grade + number of patients with TRG1b grade)/total number of patients × 100.

All patients underwent abdominal-enhanced computed tomography or magnetic resonance imaging of the target lesions before and after regular regimen treatment, and the imaging efficacy of the target lesions was assessed according to the Response Evaluation Criteria in Solid Tumors version 1.1[24]. The ORR was calculated as: ORR = [number of patients with complete response (CR) + number of patients with partial response (PR)]/total number of patients × 100. The disease control rate (DCR) was calculated as: DCR = (number of patients with CR + number of patients with PR + number of patients with stable disease)/total number of patients × 100. All abdominal imaging data were reviewed and evaluated by 2 physicians.

Follow-up

Patient follow-up was conducted through outpatient visits or telephone interviews. OS was measured from the date of treatment initiation to death from any cause, with surviving patients censored at their last documented follow-up. DFS was calculated from the date of surgery to the first occurrence of either tumor recurrence (local or distant), disease progression, or death from any cause; patients without events were censored at their final follow-up. The cutoff date for all survival analyses was July 20, 2024.

Statistical analysis

Statistical analysis was performed using SPSS version 25.0 (IBM Corp., Armonk, NY, United States). Categorical data are expressed as frequency (percentage), and comparisons between groups were performed using the χ²d test or Fisher’s exact test. Continuous variables conforming to normal distribution are expressed as mean with standard deviation, and comparisons between groups were performed using the t-test. Non-continuous variables are expressed as median (range), and comparisons between groups were performed using the Mann-Whitney U test. Survival analysis was performed using Kaplan-Meier curve analysis. Single-multifactorial analysis was performed by plain logistic regression analysis. Differences with P < 0.05 were considered to be statistically significant.

RESULTS

Clinical characteristics

Eighty-two patients with progressive LAGC, who fulfilled the inclusion criteria were divided into 2 groups according to preoperative treatment plan: ICI [sintilimab + nab-paclitaxel + S-1 (n = 53)]; and non-ICI [nab-paclitaxel + S-1 (n = 29)]. There were no significant differences between the 2 groups in terms of age, sex, Eastern Cooperative Oncology Group score, TNM clinical stage, lesion site, pathological differentiation type, nerve infiltration, choroidal infiltration, or tumor volume (P > 0.05). Detailed information is provided in Table 1.

Table 1 Baseline characteristics in group immune checkpoint inhibitor and group non-immune checkpoint inhibitor, n (%).

Variable	Group ICI (n = 53)	Group non-ICI (n = 29)	P value
Age (years)
≥ 65	31 (58.5)	12 (41.4)	0.138
< 65	22 (41.5)	17 (58.6)	-
Gender
Male	39 (73.6)	21 (72.4)	0.909
Female	14 (26.4)	8 (27.6)	-
ECOG
0-1	47 (88.7)	27 (93.1)	0.798
2	6 (13.3)	2 (6.9)	-
C-TNM
III	39 (73.6)	23 (79.3)	0.564
IVa	14 (26.4)	6 (20.7)	-
Tumor location
Upper	30 (56.6)	22 (75.9)	0.174
Middle	13 (24.5)	3 (10.3)	-
Lower	10 (18.9)	4 (13.8)	-
Differentiation degree
Poorly	22 (41.5)	16 (55.2)	0.236
Moderately-highly	31 (58.5)	13 (44.8)	-
Vascular tumor embolus	29 (54.7)	21 (72.4)	0.116
Nerve infiltration	29 (54.7)	22 (75.9)	0.059
Tumor size
> 5 cm	11 (20.8)	11 (37.9)	0.093
≤ 5 cm	42 (79.2)	18 (62.1)	-

ICI: Immune checkpoint inhibitor; ECOG: Eastern Cooperative Oncology Group; C-TNM: Tumor-node-metastasis.

Imaging response

Evaluating treatment efficacy for solid tumors according to Response Evaluation Criteria in Solid Tumors version 1.1, responses in the ICI group were distributed as follows: CR, n = 7 (13.2%); PR, n = 30 (56.6%); stable disease, n = 1 (28.3%); and progressive disease (PD), n = 1 (1.9%). The ORR was 62.3% and the DCR was 98.1%. Treatment responses in the non-ICI group were distributed as follows: CR, n = 0 (0%); PR, n = 10 (34.5%); stable disease, n = 18 (62.1%); and PD, n = 1 (3.4%). The ORR was 34.5% and the DCR was 96.6%. The CR (13.2% vs 0.0%; P = 0.048) and PR (56.6% vs 31.0%; P = 0.037) rates were significantly higher in the ICI group than those in the non-ICI group (69.8% vs 31.0%; P = 0.001). Further details of treatment response rates are summarized in Table 2.

Table 2 Radiological response in group immune checkpoint inhibitor and group non-immune checkpoint inhibitor, n (%).

Variable	Group A (n = 53)	Group B (n = 29)	P value
CR	7 (13.2)	0 (0.0)	0.048
PR	30 (56.6)	9 (31.0)	0.037
SD	15 (28.3)	18 (62.1)	0.004
PD	1 (1.9)	2 (6.7)	0.549
ORR	37 (69.8)	9 (31.0)	0.001
DCR	52 (98.1)	27 (93.1)	0.284

CR: Complete response; PR: Partial response; SD: Stable disease; PD: Progressive disease; ORR: Objective response rate; DCR: Disease control rate.

Pathological response

The postoperative pathological response was assessed according to the Becker criteria, with 7 (13.2%) patients in the ICI group achieving pCR (TRG1a), whereas pCR did not occur in the non-ICI group, and the difference between the groups was statistically significant (13.2% vs 0.0%; P = 0.048). The MPR in the ICI group (35.8% vs 13.8%; P = 0.041) and the y pathological tumor stage 0 rate (15.1% vs 0.0%; P = 0.046) were significantly higher than those in the non-ICI group, and the difference between the groups was statistically significant. The y pathological node stage 0 rate in the ICI group was higher than that in the non-ICI group (47.2% vs 34.5%; P = 0.267), although the difference was not statistically significant. Unifactorial analysis revealed that factors affecting MPR included programmed cell death/programmed death-ligand 1 (PD-L1) expression and ICI use. Multifactorial analysis revealed that positive PD-L1 expression (odds ratio = 5.317; 95% confidence interval: 1.107-25.530; P = 0.037) and ICI use (odds ratio = 3.448; 95% confidence interval: 1.014-11.728); P = 0.047) were independent protective factors against MPR. Detailed information is provided in Table 3 and Supplementary Table 1.

Table 3 Pathological evaluation of response in group immune checkpoint inhibitor and group non-immune checkpoint inhibitor, n (%).

Variable	Group ICI (n = 53)	Group non-ICI (n = 29)	P value
pCR	7 (13.2)	0 (0.0)	0.048
MPR	19 (35.8)	4 (13.8)	0.041
ypT0	8 (15.1)	0 (0.0)	0.046
ypN0	25 (47.2)	10 (34.5)	0.267
TRG1a	7 (13.2)	0 (0.0)	0.030
TRG1b	12 (22.6)	4 (13.8)	-
TRG2	23 (43.4)	14 (48.3)	-
TRG3	11 (20.8)	11 (39.7)	-

ICI: Immune checkpoint inhibitor; pCR: Pathologic complete response; MPR: Major pathologic response; ypT0: y pathological tumor stage 0; ypN0: y pathological node stage 0; TRG: Tumor regression grading.

Comparison of postoperative conditions

In this study, the differences between the 2 groups in the number of hospitalization days, extubation time, and resumption of diet were not statistically significant. The R0 resection rate was 94.3% in the ICI group and 93.1% in the non-ICI group. Three (5.7%) patients in the ICI group underwent R1 resection and 2 (6.9%) in the non-ICI group underwent R1 resection. Four (7.5%) patients in the ICI group underwent combined splenic pancreatic body-caudal resection and 1 (1.9%) partial resection of the left lobe of the liver, whereas 2 (6.9%) pancreatic body-caudal resections were performed in the non-ICI group; the difference between the groups was not statistically significant. Ten (18.9%) complications occurred in the ICI group: Gastric emptying disorder (n = 1); pulmonary infection(s) (n = 4); abdominal infection(s) (n = 3); anastomotic leakage (n = 1), and intussusception (n = 1). Four (13.8%) patients in the non-ICI group experienced complications: Pulmonary infection(s) (n = 2); abdominal infection (n = 1); and anastomotic fistula (n = 1). There was 1 (1.9%) grade I complication in the ICI group, 7 (13.2%) grade II, 2 (3.8%) grade III, and 3 (10.3%) grade II complications. One (3.4%) grade III complication occurred in the non-ICI group, with no grade IV or V complications in either group, and difference between the groups were not statistically significant. All patients who experienced complications recovered and were discharged from hospital after symptomatic treatment. No surgery-related deaths occurred in either group (Supplementary Tables 2 and 3).

Follow-up results

The median follow-up for all patients was 26 months (range: 18-50 months). Survival analysis revealed that the two-year DFS was higher in the ICI group than that in the non-ICI group (83.0% vs 55.2%; P = 0.043; Figure 2) whereas there was no statistical difference in the two-year OS between the ICI and non-ICI groups (81.1% vs 82.8%; P = 0.816; Figure 2).

Figure 2 Survival curves. A: Disease free survival for group immune checkpoint inhibitor (ICI) and group non-ICI (P = 0.043); B: Overall survival for group ICI and group non-ICI (P = 0.816). DFS: Disease-free survival; OS: Overall survival; ICI: Immune checkpoint inhibitor.

Immunohistochemistry

Detection of the molecular phenotypes of patients in the ICI group revealed that 40 (75.5%) patients had a combined positive score (CPS) ≥ 1 and 13 (24.5%) had a CPS < 1, and the difference in TRG grading comparison between the two groups of CPS ≥ 1 and CPS < 1 was statistically significant. Thirty (56.6%) patients had a CPS ≥ 5, and 23 (43.4%) had a CPS < 5. The relationship between the expression status of human epidermal growth factor receptor 2 (HER-2), MMR, and Epstein-Barr virus (EBV) in relation to pCR and MPR was examined: 4 patients with dMMR, of which 2 (50.0%) achieved pCR and 4 (100.0%) achieved MPR; and 3 patients with positive EBV expression, of whom 1 (33.3%) achieved pCR and 2 (66.6%) achieved MPR. There were a total of 7 patients with positive HER-2 expression, of whom 1 (14.3%) achieved pCR and 2 (28.6%) achieved MPR. Detailed information is provided in Supplementary Table 4 and Figure 3.

Figure 3 Immunohistochemistry. A and B: Expression of combined positive score in group immune checkpoint inhibitor (ICI); C: Expression of mismatch repair in group ICI; D: Expression of Epstein-Barr virus in group ICI; E: Expression of human epidermal growth factor receptor 2 in group ICI. CPS: Combined positive score; TRG: Tumor regression grading; pMMR: DNA proficient mismatch repair; dMMR: DNA deficient mismatch repair; EBV: Epstein-Barr virus; HER-2: Human epidermal growth factor receptor 2.

Safety profile

During preoperative treatment, both groups experienced different levels of drug toxicities, with 12 (22.6%) grade III or higher drug toxicities in the ICI group and 6 (17.2%) in the non-ICI group. The differences in toxic side-effects were, however, not statistically significant. The most common hematological adverse reactions were leukopenia, neutropenia, and anemia, while non-hematological adverse reactions included elevated transaminase levels and gastrointestinal symptoms. Five (9.4%) patients with abnormal thyroid function and 1 (1.96%) with hypoadrenocorticism in the ICI group experienced recovery after symptomatic treatment with clinical intervention or suspension of treatment. No treatment-related deaths occurred in either group (Table 4).

Table 4 Perioperative treatment-related adverse events in group immune checkpoint inhibitor and group non-immune checkpoint inhibitor, n (%).

Variable	Group ICI (n = 53)						Group non-ICI (n = 29)
	0	I	II	III	V	Total	0	I	II	III	V	Total	P value
Leucopenia	38	9	2	2	2	15 (28.3)	22	3	2	1	1	7 (24.1)	0.684
Neutropenia	35	10	5	2	1	18 (34.0)	21	3	3	2	0	8 (27.6)	0.553
Anemia	40	9	3	1	0	13 (24.5)	24	2	3	0	0	5 (17.2)	0.446
Thrombocytopenia	51	1	1	0	0	2 (3.8)	29	0	0	0	0	0 (0.0)	0.537
Serum AST/ALT increase	38	10	2	2	1	15 (28.3)	21	6	2	0	0	8 (27.6)	0.945
Elevated bilirubin index	49	3	1	0	0	4 (7.5)	27	2	0	0	0	2 (6.9)	1.000
Electrolyte disturbance	46	6	0	1	0	7 (13.2)	27	2	0	0	0	2 (6.9)	0.614
Serum creatinine increase	50	3	0	0	0	3 (5.7)	29	0	0	0	0	0 (0.0)	0.490
Hyper- or hypothyroidism	46	3	2	0	0	5 (9.4)	29	0	0	0	0	0 (0.0)	0.221
Myocardial enzyme increase	49	4	0	0	0	4 (7.5)	29	0	0	0	0	0 (0.0)	0.327
Pneumonia	53	0	0	0	0	0 (0.0)	29	0	0	0	0	0 (0.0)	NA
Adrenocortical insufficiency	52	0	0	1	0	1 (1.9)	29	0	0	0	0	0 (0.0)	1.000
Nausea/vomiting	40	9	1	1	2	13 (24.5)	23	2	1	1	2	6 (20.7)	0.694
Grade 0	-	-	-	-	-	28 (52.8)	-	-	-	-	-	17 (58.6)	0.967
Grade I	-	-	-	-	-	8 (15.1)	-	-	-	-	-	4 (13.8)	-
Grade II	-	-	-	-	-	5 (9.4)	-	-	-	-	-	3 (10.3)	-
Grade III	-	-	-	-	-	6 (11.3)	-	-	-	-	-	2 (6.9)	-
Grade IV	-	-	-	-	-	6 (11.3)	-	-	-	-	-	3 (10.3)	-
Overall rate	-	-	-	-	-	25 (47.2)	-	-	-	-	-	12 (41.4)	0.614
Any grade	-	-	-	-	-	41 (77.4)	-	-	-	-	-	23 (82.8)	0.564

ICI: Immune checkpoint inhibitor; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase.

DISCUSSION

The present study investigated the efficacy and safety of sintilimab combined with albumin-bound paclitaxel and tigio (i.e., S-1) for the preoperative treatment of patients diagnosed with LAGC. The results demonstrated that sintilimab combination therapy significantly enhanced treatment efficacy compared to chemotherapy alone (P < 0.05), while maintaining comparable safety profiles in terms of drug-related adverse reactions and perioperative complications. These findings support the safety and clinical feasibility of sintilimab in this therapeutic setting.

For some patients with resectable LAGC, due to the high likelihood of experiencing micrometastatic lesions before surgery, deeper tumor infiltration, and even invasion of neighboring organs, the scope of resection for surgical treatment is larger, with more intraoperative and postoperative complications[25]. Recurrence or metastasis occurs in most patients after surgery, which results in a shorter OS period and poor prognosis. Neoadjuvant therapy can compensate for the limitations of surgery, reduce tumor stage, and eliminate micrometastatic lesions, thus reducing the risks for recurrence, rapidly improving tumor-related symptoms, and determining whether the tumor is sensitive to chemotherapy to increase the likelihood of radical resection, improve R0 resection rates, reduce postoperative complications and mortality, and does not increase perioperative risk[26,27]. Several large studies[4,5,26] have demonstrated that preoperative chemotherapy in combination with surgical treatment not only improves surgical cure rates but also improves patient survival compared with surgical treatment alone. However, the population that benefits from neoadjuvant chemotherapy alone is limited, and there is an urgent need for new therapeutic approaches to increase the number of individuals who would from neoadjuvant therapy.

In recent years, the use of ICIs has brought hope to patients with advanced gastric cancer. ICIs, represented by anti-programmed death 1/PD-L1 antibodies, have significantly improved the prognosis of patients with advanced gastric cancer and other cancers, and 2 large phase III studies, CheckMate-649[7] and ORIENT-16[16], have solidified the position of ICIs in combination with chemotherapy for the first-line treatment of advanced gastric cancer. As such, the use of ICI combined with chemotherapy for the preoperative treatment of LAGC may be a new therapeutic approach. Several retrospective studies[28-34] have also confirmed that the combination of ICI and chemotherapy has shown good efficacy, with satisfactory pCR and survival outcomes. Moreover, laparoscopic surgery after neoadjuvant immunotherapy combined with chemotherapy is safe and feasible. In this study, the CR (13.2% vs 0.0%; P = 0.048), PR (56.6% vs 31.0%; P = 0.037), and ORR in the ICI group were found to be significantly higher than those in the non-ICI group (69.8% vs 31.0%; P = 0.001). In terms of pathology, the pCR was 13.2% (7/53) and MPR was 35.8% (19/53) in the ICI group, and 0.0% (0/29) and 13.8% (4/29), respectively, in the non-ICI group. The pCR and MPR in the ICI group were significantly higher than those in the non-ICI group, and the use of ICI significantly improved the pathological response of the tumor, which greatly improved the pCR and MPR (P < 0.05). Second, the y pathological tumor stage 0 (18.2% vs 0%) and y pathological node stage 0 rates in the ICI group were greater than those in the non-ICI group, suggesting that the use of ICI in preoperative treatment is more advantageous in terms of local tumor downstaging effects. Lin et al[32] concluded that the use of ICIs in the perioperative period was an independent protective factor for MPR and lymph node staging according to multifactorial analysis. In the present study, PD-L1 positivity and ICI use were independent protective factors for MPR, which is similar to the above findings because previous studies have reported that preoperative chemotherapy may lead to a variety of adverse effects, and the use of ICI may also produce immunotherapy-related adverse effects (irAEs), which may increase postoperative complications. However, a retrospective analysis of gastric cancer resection reported that preoperative chemotherapy followed by gastric cancer resection did not significantly increase the incidence of perioperative adverse events in patients compared with the surgical regimen alone[26]. Su et al[30] concluded that the use of ICIs in preoperative therapy does not increase the associated perioperative risks. Sun et al[11] found that ICI combined with chemotherapy for preoperative treatment of LAGC, compared to chemotherapy alone, showed no statistically significant difference in the incidence of surgical complications between the two groups (17.44% vs 16.15%). In the present study, there were no intraoperative complications or patients who experienced surgical death in the 2 groups, and total complication rates was comparable (18.9% vs 13.8%). Moreover, there were no grade IV-V complications, and differences in the complications at all levels were not statistically significant, which was similar to the above findings. Moreover, patients who experienced postoperative complications were discharged from hospital after the appropriate active treatments, and there were no perioperative deaths. Second, there were no significant differences in the length of surgery, postoperative hospitalization, postoperative diet recovery, postoperative extubation, or combined organ resection between the 2 groups. Therefore, compared with chemotherapy alone, sintilimab combined with chemotherapy as a preoperative treatment option for LAGC did not affect postoperative recovery or increase the occurrence of postoperative complications.

Huang et al[34] reported a 1-year OS of 92.8% and a DFS of 88.9% for sintilimab combined with oxaliplatin + S-1, and Jiang et al[35] reported a 1-year OS and DFS of 94.1% and 90.3%, respectively, for sintilimab combined with XELOX (oxaliplatin + capecitabine). In contrast, in this study, the two-year DFS in the ICI group was significantly higher than that in the non-ICI group(83.0% vs 55.2%; P = 0.043). The ICI group exhibited significantly better 2-year DFS rates than the non-ICI group (83.0% vs 55.2%, P = 0.043), although the 2-year OS rates showed no statistically significant difference between groups (81.1% vs 82.8%, P = 0.816). This finding is consistent with early reports from the FLOT4-AIO[36] and CheckMate-816[37] trials. However, due to the relatively small sample size, shorter follow-up duration, and the evaluation of immunotherapy efficacy in an unselected population in this study, further validation is required. Expanding the sample size or conducting stratified analyses based on PD-L1 expression levels will help clarify which patient subgroups truly benefit from immunotherapy.

There are several defined biomarkers regarding gastric cancer, including PD-L1 expression, MMR status, EBV positivity, and HER-2 expression status. PD-L1 is a promising therapeutic and prognostic target for gastric cancer, and the CPS has progressively developed as a predictive marker for immunotherapy response in gastric cancer and is frequently used as a stratification marker in other studies that can guide patients in the selection of immunotherapy[38,39]. Kim et al[40] achieved better PR with ICI in patients with PD-L1 (+) metastatic gastric cancer compared with PD-L1 (-). The KEYNOTE-059, KEYNOTE-061, KEYNOTE-062 clinical trials achieved better long-term survival can be obtained by using ICI in patients with CPS ≥ 10[41]. However, higher pCR and MPR were observed in preoperative treatment with ICI in combination with chemotherapy in patients with CPS ≥ 5 as reported by Huang et al[34]. The present study concluded that patients with CPS ≥ 1 were more likely to experience a better pathological outcome. dMMR status can lead to DNA mismatches, followed by a rapid in vivo expansion of specific T cells, which can make patients with dMMR more sensitive to ICIs[42]. There were 4 patients with dMMR in this study, of whom 2 achieved pCR and all achieved MPR. This indicates that patients with dMMR may be able to achieve a better pathological response by receiving ICI in combination with chemotherapy in the perioperative period. It has been shown that EBV-positive patients with gastric cancer may be more suitable candidates for ICI treatment[43]. In contrast, in this study, only 3 patients were EBV-positive, and 1 achieved pCR with an MPR rate of 66.6%, which may indicate that EBV-positive patients with gastric cancer are more suitable for immunotherapy. Due to the small number of patients in this study, more clinical studies are needed to further confirm this.

Adverse reactions of different grades occurred in both groups of patients, with the total incidence of hematological toxic reactions comparable with that of non-hematological toxic reactions, which were normalized by symptomatic clinical interventions or suspension of treatment. Results of several studies have shown that increasing ICI with preoperative treatment did not significantly increase the incidence of adverse reactions[13,32,34]. In this study, the incidence of adverse reactions was slightly higher in the ICI group than in the non-ICI group (47.2% vs 41.4%; P = 0.614), and for drug toxicities of grade III or higher, there were 12 cases (22.6%) in the ICI group and 6 (17.2%) in the non-ICI group, although the difference was not statistically significant (P = 0.564). The observed phenomenon may be attributed to the distinct pharmacological ICIs. While the incidence of neutropenia (34.0% vs 28.3%) and elevated transaminases (28.3% vs 24.5%) was slightly higher in the ICI group compared to the non-ICI group, these adverse events were typically transient. All cases were effectively managed with therapeutic interventions, such as corticosteroid supplementation, dose adjustments, and supportive measures (e.g., granulocyte colony-stimulating factor for neutropenia and hepatoprotective agents for transaminase elevation). Importantly, none of the patients required permanent treatment discontinuation due to these adverse effects. The most common hematotoxic reactions included leukopenia (n = 11, 36.6%), neutropenia (n = 9, 30.0%), elevation of transaminase levels (n = 7, 23.3%), and reduction in hemoglobin level (n = 15, 36.7%). Nausea, vomiting, and diarrhea were the common non-hematotoxic reactions during the course of sintilimab in combination with XELOX as the preoperative treatment regimen[33]. Similarly, the most frequent hematotoxic reactions among patients in the ICI group in this study included: Neutropenia (n = 18, 34.0%); leukopenia (n = 15, 28.3%); reduction in hemoglobin (n = 13, 24.5%); and elevation of transaminase levels (n = 15, 28.3%). The majority of irAEs are mild to moderate and resolve on their own after discontinuation of drug without the need for specific treatment. However, serious adverse reactions such as immune-associated pneumonia, myocarditis, neurotoxicity, and lethal diarrhea, sometime occur. Previous studies have shown that endocrine organs and the skin are the favored sites for irAEs, followed by the gastrointestinal tract and lungs, and most of the reactions are generally mild to moderate[44]. Is the present study, 6 (11.3%) irAEs occurred in the ICI group, of which 5 occurred in patients with abnormal thyroid function, and 1 was grade 3 hypoadrenocorticism. Patients who experienced irAEs in this study successfully completed treatment after the symptoms were appropriately addressed. The use of ICIs has changed the pattern of tumor treatment; however, the occurrence of irAEs should not be underestimated, and the management of irAEs in clinical practice needs to be further strengthened.

The present study had some limitations, the first of which was its single-center, retrospective design. However, the chemotherapy regimens used in this study were consistent, which reduced the bias caused by differences in treatment outcomes due to different chemotherapy regimens, and increase the credibility of the conclusions drawn from this study. Nevertheless, they still need to be further confirmed in prospective larger-scale clinical studies. Second, the follow-up period was relatively short, and the OS data have not been completed in full, and require validation through further follow-up.

CONCLUSION

In conclusion, ICI combined with chemotherapy was an effective preoperative treatment option for the management of LAGC and did not increase the incidence of adverse reactions or perioperative adverse events. However, this conclusion needs to be further confirmed in larger-scale, prospective clinical studies.