Wang L, Chen MZ, Liu L, Jiang ZN, Zhang SM, Zhang MS, Zhang XX, Liu RQ, Wang DS. Novel inflammatory-nutritional prognostic index for advanced gastric cancer patients undergoing gastrectomy and prophylactic hyperthermic intraperitoneal chemotherapy. World J Gastrointest Surg 2025; 17(5): 102201 [DOI: 10.4240/wjgs.v17.i5.102201]
Corresponding Author of This Article
Dong-Sheng Wang, MD, Professor, Department of Gastrointestinal Surgery, The Affiliated Hospital of Qingdao University, No. 1677 Wutaishan Road, Huangdao District, Qingdao 266555, Shandong Province, China. wangdongsheng@qdu.edu.cn
Research Domain of This Article
Gastroenterology & Hepatology
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
World J Gastrointest Surg. May 27, 2025; 17(5): 102201 Published online May 27, 2025. doi: 10.4240/wjgs.v17.i5.102201
Novel inflammatory-nutritional prognostic index for advanced gastric cancer patients undergoing gastrectomy and prophylactic hyperthermic intraperitoneal chemotherapy
Co-corresponding authors: Rui-Qing Liu and Dong-Sheng Wang.
Author contributions: All authors read and approved the final manuscript; Wang DS, Liu RQ, Wang L, Chen MZ, Zhang MS, Zhang XX contributed to the study’s conception and design; Wang L, Liu L, and Jiang ZN were responsible for patient screening and data collection. Wang L, Chen MZ, and Zhang SM analyzed the data and made the figures and tables. Wang L, Chen MZ, and Liu RQ completed the first draft of the article. Wang DS, Liu RQ, Wang L, and Chen MZ participated in revising the manuscript before submission and during the formal revision. Wang L and Chen MZ contributed equally as co-first authors; During the creation and publication of this article, Liu RQ and Wang DS, as co-corresponding authors, made significant contributions to each key aspect. Wang DS took the lead in designing the research plan, ensuring its scientificity, feasibility, and innovation. Meanwhile, Wang DS was responsible for coordinating various resources to guarantee the smooth progress of the research work. Liu RQ was deeply involved in the experimental process, including specific tasks such as data collection and chart making. During the writing of the paper, Liu RQ systematically sorted out and analyzed the research results and wrote key chapters, making the expressions more accurate, fluent, and the logic more rigorous.
Institutional review board statement: The study was reviewed and approved by the institutional Ethics Committee of Affiliated Hospital of Qingdao University (No. QYFY WZLL28242).
Informed consent statement: The requirement of informed consent for enrolled patients was waived by the Institutional Ethics Committee because of the retrospective study design.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Data sharing statement: The datasets during the current study available from the corresponding author on reasonable request at wangdongsheng@qdu.edu.cn.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Dong-Sheng Wang, MD, Professor, Department of Gastrointestinal Surgery, The Affiliated Hospital of Qingdao University, No. 1677 Wutaishan Road, Huangdao District, Qingdao 266555, Shandong Province, China. wangdongsheng@qdu.edu.cn
Received: October 11, 2024 Revised: February 13, 2025 Accepted: March 21, 2025 Published online: May 27, 2025 Processing time: 223 Days and 13.3 Hours
Abstract
BACKGROUND
Prophylactic hyperthermic intraperitoneal chemotherapy (HIPEC) is one of the methods to prevent peritoneal metastasis of advanced gastric cancer (AGC). However, the prognosis of gastric cancer patients who receive this treatment are different.
AIM
To investigate whether inflammation and nutritional indicators affect the prognosis of AGC patients undergoing gastrectomy and prophylactic HIPEC, and to develop a novel inflammatory nutritional prognostic index (INPI). Additionally, we aimed to construct a nomogram model to visually predict the prognosis of these patients and provide more accurate guidance for clinical decision-making.
METHODS
Clinical data from 181 Locally AGC patients who underwent gastrectomy and prophylactic HIPEC treatment at The Affiliated Hospital of Qingdao University were retrospectively collected. Multicollinearity analysis and least absolute shrinkage and selection operator (LASSO) Cox regression were utilized to construct the INPI. Survival analyses were performed using the Kaplan-Meier method and log-rank test. Both univariate and multivariate Cox proportional hazards regression models were used to analyze independent prognostic factors, and a prognostic nomogram was generated. And the model was validated using the bootstrap method.
RESULTS
Clinical data from 181 locally AGC patients who underwent gastrectomy and prophylactic HIPEC treatment at The Affiliated Hospital of Qingdao University were retrospectively collected. Multicollinearity analysis and LASSO Cox regression were utilized to construct the INPI. Survival analyses were performed using the Kaplan-Meier method and log-rank test. Both univariate and multivariate Cox proportional hazards regression models were applied to analyze independent prognostic factors, and a prognostic nomogram was generated. And the model was validated using the bootstrap method.
CONCLUSION
Inflammation and nutrition indicators are associated with the prognosis of AGC patients undergoing gastrectomy and prophylactic HIPEC. The nomogram based on the INPI and clinical features supports personalized treatment strategies improving prognosis for AGC patients undergoing gastrectomy and prophylactic HIPEC.
Core Tip: This study introduces a novel inflammatory nutritional prognostic index (INPI) for advanced gastric cancer (AGC) patients undergoing gastrectomy and prophylactic hyperthermic intraperitoneal chemotherapy. Analyzing data from 181 patients, we established the INPI to stratify patients into three distinct risk groups, revealing significant prognostic differences. Our prognostic nomogram, integrating INPI with clinical features, demonstrated superior predictive performance compared to traditional tumor-node-metastasis staging, offering a valuable tool for personalizing treatment strategies and improving patient outcomes in AGC management.
Citation: Wang L, Chen MZ, Liu L, Jiang ZN, Zhang SM, Zhang MS, Zhang XX, Liu RQ, Wang DS. Novel inflammatory-nutritional prognostic index for advanced gastric cancer patients undergoing gastrectomy and prophylactic hyperthermic intraperitoneal chemotherapy. World J Gastrointest Surg 2025; 17(5): 102201
Gastric cancer (GC) is one of the most common gastrointestinal tumors and ranks fifth in incidence among all cancers and fourth in mortality[1]. The poor prognosis of GC patients is closely associated with local recurrence of the tumor; T4-stage GC patients have a high recurrence rate of up to 43.8%, with the peritoneum being one of the most common sites for GC recurrence and metastasis[2].
Early prevention and treatment of peritoneal metastasis of GC, including eradication of free cancer cells in the peritoneal cavity, is crucial for improving patient prognosis. Prophylactic hyperthermic intraperitoneal chemotherapy (HIPEC), which was originally proposed by Spratt et al[3], has shown unique efficacy in the treatment of peritoneal carcinomatosis resulting from various cancers[4-6]. Prophylactic HIPEC refers to postoperative treatment administered to GC patients with locally advanced disease but without visible peritoneal metastases. It has been proven to be an effective treatment for preventing postoperative peritoneal recurrence, extending the prognosis of advanced GC (AGC) patients[7-10]. However, its efficacy remains controversial, the study by Lee et al[6] evaluated the efficacy of prophylactic HIPEC after gastrectomy in patients with clinical T4 GC. The overall survival (OS) and disease-free survival (DFS) in the prophylactic HIPEC group were superior to those of patients who underwent gastrectomy alone (P = 0.035 and P = 0.017, respectively). However, the study by Fan et al[11] indicated that although prophylactic HIPEC did not increase postoperative complications in patients with locally AGC, it did not reduce the risk of peritoneal recurrence. The factors affecting the outcomes of AGC patients who undergo gastrectomy and prophylactic HIPEC are currently unknown and are the focus of our research.
Inflammation and nutrient availability play important roles in the occurrence and progression of cancer[12], and gastrectomy and prophylactic HIPEC are typically performed for patients with locally AGC, specifically those with cT4a disease. Studies have shown that patients with more advanced-stage tumors often have greater inflammation and poorer nutritional status[13]. We do not know whether preoperative inflammatory and nutritional status affects the efficacy of prophylactic HIPEC treatment. Many studies have demonstrated a close association between inflammatory and nutritional indicators and the prognosis of cancer patients, which can be used to assess and predict the prognosis of different cancers[14-17]. Therefore, we designed a retrospective study with the aim of establishing a novel inflammatory-nutritional prognostic index (INPI) based on inflammation and nutritional indices in combination with clinical features to predict the prognosis of AGC patients who underwent gastrectomy and prophylactic HIPEC treatment.
MATERIALS AND METHODS
Patients
In this study, we retrospectively collected the clinical data of local AGC patients who underwent gastrectomy followed by prophylactic HIPEC treatment at The Affiliated Hospital of Qingdao University from January 2016 to December 2019. This study received approval from the hospital's ethics committee (No. QYFY WZLL28242) Anonymize the data to maximize the protection of patient privacy. During the research process, regularly report the implementation of patient privacy protection measures to the Ethics Review Committee and accept the supervision and guidance of the Ethics Review Committee. The inclusion criteria were as follows: (1) Had diagnosed GC with a clinical stage of T4a, underwent curative gastrectomy and prophylactic HIPEC treatment, and underwent adjuvant chemotherapy; (2) Were aged between 18 and 80 years; and (3) Had complete clinical and pathological data, routine preoperative examination data, and postoperative follow-up data. The exclusion criteria for patients were as follows: (1) Had a previous history of other malignant tumors; (2) Used immunosuppressive agents such as steroids within the 3 months prior to surgery; (3) Had concomitant acute or chronic infections, rheumatic autoimmune diseases, hematological disorders, or other blood system diseases; and (4) Had an incurable tumor or the presence of distant metastases.
Data collection
The collected clinical data included the following: (1) General information - name, sex, age, neoadjuvant status, American Society of Anesthesiologists classification, and nutritional score; (2) Clinicopathological information - tumor site, pathological tumor type, pT stage, pN stage, histological grade, and lymphovascular invasion; (3) Surgical information - extent of gastric resection, reconstruction method, and heat perfusion frequency; and (4) Inflammation and nutrition-related indicators, including body mass index (BMI), neutrophil count, platelet count, lymphocyte count, albumin level, monocyte count, hemoglobin level, and globulin level.
The calculations of the relevant indices were as follows: Neutrophil-to-lymphocyte ratio (neutrophil count/lymphocyte count), platelet-to-lymphocyte ratio (platelet count/lymphocyte count), lymphocyte-to-monocyte ratio (lymphocyte count/monocyte count), prognostic nutritional index (PNI) (albumin value + 5 × lymphocyte count), systemic immune-inflammatory index (SII) (neutrophil count × platelet count/lymphocyte count), and platelet-to-albumin ratio (PAR) (platelet count/albumin).
Follow-up
Patients were followed up through outpatient visits and telephone interviews. For the first 3 years, follow-up was performed every 3 to 6 months, every 6 months for 3 to 5 years, and then every year after 5 years. The follow-up included blood biochemistry assessments, tumor marker analysis, abdominal computed tomography and electronic gastroscopy, and the follow-up period continued until March 2023. The time from surgery to death from any cause was defined as OS.
Statistical analysis
Continuous variables are presented as the mean ± SD or median and interquartile range (IQR). Categorical variables are presented as frequencies (percentages) and were compared using the χ2 test or Fisher's exact test, as appropriate. We used multicollinearity analysis for inflammation and nutrition-related indicators, initially excluding variables with a variance inflation factor greater than 10. Subsequently, a least absolute shrinkage and selection operator (LASSO) Cox regression model was used to select the optimal prognostic features from among all inflammation- and nutrition-related indicators for the construction of a new INPI. Survival analyses were performed using the Kaplan-Meier method and log-rank test, and both univariate and multivariate Cox proportional hazards regression models were used to analyze independent prognostic factors. Finally, the results of the multivariate analysis were used to generate a prognostic nomogram, and its predictive performance was validated using the bootstrap method (with 1000 repetitions and tenfold cross-validation). Model performance was assessed using the concordance index (C-index) and calibration curves. SPSS version 26.0 and R language version 4.2.3 was employed for the statistical analysis, and a P value < 0.05 was considered indicative of a statistically significant difference. The cutoff values for relevant features were determined using X-tile version 3.6.1.
RESULTS
Clinical characteristics
The clinical baseline data of the patients are shown in Table 1. A total of 181 patients were included, including 110 males (60.8%) and 71 females (39.2%). The median age was 64 years (IQR: 57-69). Among the patients, 26 had stage IIB disease (14.4%), 20 had stage IIIA disease (11%), 44 had stage IIIB disease (24.3%), and 91 had stage IIIC disease (50.3%). Tables 1 and 2 also list the baseline information for more than 10 inflammation and nutritional indicators. During the follow-up period, 114 out of 181 patients (62.9%) recurred or metastasis, and 102 (56.4%) of those patients died. The OS rates at 1, 2, and 3 years were 96.1%, 75.1%, and 53.6%, respectively (Table 3). The median OS was 41.3 months (95%CI: 35.7-47.0 months).
Table 1 Relationship between the death within 3 years and clinicopathological characteristics of 181 advanced gastric cancer patients.
Patients’ characteristics
No death (n = 97)
Death (n = 84)
χ2 value
P value
Sex
2.376
0.123
Male
64
46
Female
33
38
Age at diagnosis, years
< 0.001
< 60
55
9
41.651
≥ 60
42
75
NRS 2002
32.887
< 0.001
< 5
44
6
≥ 5
53
78
ASA score
1.960
0.581
1
1
0
2
53
44
3
42
40
4
1
0
pT stage
38.567
< 0.001
pT3
53
9
pT4
44
75
pN stage
68.524
< 0.001
N1
64
9
N2
24
26
N3
9
49
pTNM stage
38.140
< 0.001
II
40
2
III
57
82
Pathological tumor type
0.521
0.470
Adenocarcinoma
73
67
Mucinous adenocarcinoma or signet-ring cell carcinoma
24
17
Tumor site
Cardia/fundus
1
2
1.991
0.370
Body/angulus
28
31
Antrum/pylorus
68
51
Histological grade
0.062
0.804
G2
70
62
G3
27
22
Lymphovascular invasion
39.308
< 0.001
Yes
39
72
No
58
12
Surgical procedure
2.080
0.149
Open surgery
24
29
Laparoscopic surgery
73
55
Reconstruction method
0.283
0.868
Roux-en-Y
65
59
Billroth II + Braun
30
23
Billroth I
2
2
Extent of gastric resection
0.451
0.502
Distal
66
61
Total
31
23
Heat perfusion frequency
2.897
0.235
1
9
8
2
60
42
3
28
34
BMI (kg/m2)
30.885
< 0.001
< 22.1
8
37
≥ 22.1
89
47
Table 2 Relationship between the death within 3 years and inflammation and nutrition-related indicators of 181 advanced gastric cancer patients.
Patients’ characteristics
No death (n = 97)
Death (n = 84)
χ2 value
P value
Hemoglobin (g/L)
14.660
< 0.001
< 119
37
56
≥ 119
60
28
Albumin (g/L)
16.099
< 0.001
< 34.38
19
40
≥ 34.38
78
44
Neutrophil count (× 109/L)
13.977
< 0.001
< 5.2
83
59
≥ 5.2
8
25
Lymphocyte count (× 109/L)
5.177
0.023
< 1.42
27
37
≥ 1.42
70
47
Monocyte count (× 109/L)
5.389
0.020
> 0.43
65
42
≤ 0.43
32
42
Platelet count (× 109/L)
0.305
0.581
< 324
13
9
≥ 324
84
75
NLR
14.625
< 0.001
< 3.11
82
50
≥ 3.11
15
34
PLR
< 159.3
55
29
8.902
0.003
≥ 159.3
42
55
LMR
5.288
0.021
< 3.34
30
40
≥ 3.34
67
44
SII
19.731
< 0.001
< 801.43
56
21
≥ 801.43
41
63
PNI
17.113
< 0.001
< 43.42
23
45
≥ 43.42
74
39
PAR
21.146
< 0.001
< 9.09
84
47
≥ 9.09
13
37
AGR
5.070
0.024
< 1.12
10
19
≥ 1.12
87
65
Globulin (g/L)
0.009
0.926
< 30.02
34
30
≥ 30.02
63
54
Table 3 The cumulative recurrence rates and overall survival rates of 181 advanced gastric cancer patients undergoing gastrectomy and prophylactic hyperthermic intraperitoneal chemotherapy.
Rates
1 year
2 years
3 years
Median OS
Overall survival rates
96.1%
75.1%
53.6%
41.3 months
Construction of the INPI
The process of constructing the new INPI model is illustrated in Figure 1. Based on the results of the multicollinearity analysis, three highly correlated features, globulin, albumin, and platelet count, were excluded. Subsequently, LASSO-Cox regression was used to further select features, and seven features, namely, BMI, neutrophil count, SII, PNI, PAR, albumin-to-globulin ratio (AGR), and hemoglobin, that corresponded to the optimal value λ = 0.041 were selected from the initial 10 indicators for model construction (Figure 2). Optimal cutoff values for inflammation- and nutrition-related indicators were obtained using X-tile software, and the patients were categorized into high and low groups. Survival curves were generated using Kaplan-Meier analysis, and log-rank tests revealed statistically significant differences between the high- and low-risk groups (Figure 3A-G).
Figure 2 Construction of the inflammatory-nutritional prognostic index using the least absolute shrinkage and selection operator Cox regression model.
A: The least absolute shrinkage and selection operator coefficient curves of 12 inflammatory and nutritional indicators, showing the changes in the regression coefficients of each indicator in the model. As the regularization parameter λ increases, some coefficients are gradually shrunk to zero, thus achieving variable selection and enhancing the sparsity of the model; B: Ten-fold cross-validation was used for parameter tuning and selection of the least absolute shrinkage and selection operator model. The vertical "dotted line a" represents λmin, that is, the number of independent variables in the model corresponding to the minimum error. It indicates that under the λ =0.041, the cross-validation error of the model reaches the minimum. The vertical "dotted line b" represents λ1se, which is the number of independent variables in the model corresponding to the position of one standard error from the minimum error.
Figure 3 Survival curves via Kaplan-Meier analysis of 7 inflammation and nutrition indicators and inflammatory-nutritional prognostic index groups.
A: Survival curves of systemic immune-inflammatory index (≥ 801.43 vs < 801.43); B: Survival curves of albumin-to-globulin ratio (AGR) (≥ 1.12 vs < 1.12); C: Survival curves of body mass index (≥ 22.1 vs < 22.1); D: Survival curves of AGR hemoglobin (≥ 119 vs < 119); E: Survival curves of neutrophils (≥ 5.2 vs < 5.2); F: Survival curves of platelet-to-albumin ratio (≥ 9.09 vs < 9.09); G: Survival curves of prognostic nutritional index (≥ 43.42 vs < 43.42); H: Survival curves of inflammatory-nutritional prognostic index (low risk vs medium risk vs high risk). SII: Systemic immune-inflammatory index; AGR: Albumin-to-globulin ratio; PAR: Platelet-to-albumin ratio; INPI: Inflammatory nutritional prognostic index.
The relevant risk factors were defined as follows: Low BMI (BMI < 22.1 kg/m²), high neutrophil count (neutrophil ≥ 5.2 × 109/L), high SII (SII ≥ 801.43), low PNI (PNI < 43.42), high PAR (PAR ≥ 9.09), low AGR (AGR < 1.12), and low hemoglobin (hemoglobin < 119 g/L). The presence of any of the above risk factors was assigned a score of 1, and the total score constituted the new INPI. Using X-tile software, INPI scores were stratified into risk groups, with INPI scores ranging from 0-2 indicating the low-risk group (n = 101, 55.8%), INPI scores ranging from 3-4 indicating the intermediate-risk group (n = 60, 33.1%), and INPI scores ranging from 5-7 indicating the high-risk group (n = 20, 11.1%). There were significant differences in prognosis among the low-risk, intermediate-risk, and high-risk patients reclassified by the INPI (log rank P < 0.001; Figure 3H).
Prognostic nomogram for GC patients undergoing radical gastrectomy with prophylactic HIPEC
The results of univariate and multivariate analyses for baseline data and INPI scores are shown in Table 4. The study indicated a significant association between low BMI, high neutrophil count, high SII, low PNI, high PAR, low AGR, low hemoglobin, and shorter OS [hazard ratio (HR) = 3.84, 95%CI: 2.56-5.88; HR = 3.33, 95%CI: 2.15-5.15; HR = 2.57, 95%CI: 1.67-3.96; HR = 1.75, 95%CI: 2.56-3.84; HR = 3.01, 95%CI: 2.00-4.54; HR = 1.23, 95%CI: 1.96-3.22; HR = 1.58, 95%CI: 2.43-3.57]. Variables identified as potential prognostic factors (P < 0.05) in the univariate analysis were included in the multivariate analysis. The results of multivariate analysis indicated that the INPI was an independent prognostic factor for OS (HR = 3.70, 95%CI: 2.29-5.98; HR = 9.81, 95%CI: 4.97-19.37, P < 0.001). Additionally, age at diagnosis (year) (P < 0.001), Nutritional Risk Screening 2002 (NRS 2002) score (P = 0.037), pT stage (P = 0.028), pN stage (P < 0.001), and lymphovascular invasion (P = 0.01) were found to be independent prognostic factors for OS.
Table 4 Univariate and multivariate analyses of baseline characteristics for overall survival in 181 advanced gastric cancer patients.
Patients’ characteristics
Univariate analysis
Multivariate analysis
HR (95%CI)
P value
HR (95%CI)
P value
Sex (male vs female)
0.78 (0.52-1.15)
0.209
0.94 (0.63-1.41)
0.780
Age at diagnosis, years (> 60 vs ≤ 60)
6.98 (3.92-12.4)
< 0.001
3.66 (1.97-6.80)
< 0.001
INPI groups
Medium risk vs low risk
3.23 (2.08-5.00)
< 0.001
3.70 (2.29-5.98)
< 0.001
High risk vs low risk
15.0 (8.17-27.57)
< 0.001
9.81 (4.97-19.37)
< 0.001
NRS 2002 (> 5 vs ≤ 5)
4.18 (2.40-7.29)
< 0.001
1.88 (1.03-3.40)
0.037
ASA score
2 vs 1
0.76 (0.10-5.56)
0.794
3 vs 1
0.86 (0.11-6.29)
0.888
4 vs 1
1.35 (0.08-21.75)
0.830
pT stage (pT4 vs pT3)
4.46 (2.68-7.42)
< 0.001
2.06 (1.08-3.94)
0.028
pN stage
N2 vs N1
3.70 (2.09-6.53)
< 0.001
3.77 (1.83-7.78)
< 0.001
N3 vs N1
9.83 (5.70-16.96)
< 0.001
6.69 (3.30-13.56)
< 0.001
pTNM stage (IV vs III)
6.93 (3.34-14.40)
< 0.001
1.49 (0.5-4.54)
0.477
Pathological tumor type
Mucinous adenocarcinoma or signet-ring cell carcinoma vs adenocarcinoma
1.07 (0.67-1.72)
0.758
Tumor site
Body/angulus vs cardia/fundus
1.09 (0.26-4.55)
0.901
Antrum/pylorus vs cardia/fundus
0.92 (0.22-3.78)
0.916
Histological grade (G3 vs G1/G2)
12.8 (0.80-2.04)
0.309
Lymphovascular invasion (yes vs no)
3.37 (2.13-5.32)
< 0.001
1.91 (1.17-3.13)
0.010
Surgical procedure (open surgery vs laparoscopic surgery)
1.25 (0.83-1.90)
0.279
Reconstruction method
Billroth II + Braun vs Roux-en-Y
1.068 (0.69-1.64)
0.763
Billroth I vs Roux-en-Y
0.894 (0.21-3.64)
0.876
Heat perfusion frequency
2 times vs 1 time
0.79 (0.40-1.55)
0.498
3 times vs 1 time
0.86 (0.58-2.34)
0.658
Extent of gastric resection (distal vs total)
0.98 (0.65-1.49)
0.946
The prognostic nomogram based on the multivariate Cox analysis results is presented in Figure 4A. This analysis included all the independent prognostic factors for OS, including the INPI score, age at diagnosis, and NRS 2002 score. A higher total score indicated a worse clinical prognosis for AGC patients who underwent curative surgery with prophylactic HIPEC. The nomogram showed good performance, with a C-index of 0.838. Internal validation was performed using bootstraps with 1000 resamples to validate the reliability of the model, resulting in a C-index of 0.833. The calibration plots of the nomogram (method = ‘boot’, B = 1000) for predicting 3-year OS showed good performance with the ideal model, and the high concordance ensured accurate estimation of patient survival (Figure 4B).
Figure 4 A novel prognostic nomogram based on inflammatory-nutritional prognostic index for advanced gastric cancer patients undergoing curative surgery with prophylactic hyperthermic intraperitoneal chemotherapy.
A: The nomogram for predicting 2, 3 and 4 year survival probability in advanced gastric cancer patients undergoing curative surgery with prophylactic hyperthermic intraperitoneal chemotherapy; B: Calibration plots of the nomogram for 3-year survival probability using bootstraps with 1000 resample. NRS: Nutritional Risk Screening; INPI: Inflammatory nutritional prognostic index; OS: Overall survival.
Additionally, we constructed a nomogram using age at diagnosis, NRS-2002 score, pT stage, pN stage, and lymphovascular invasion and validated the model's reliability. The C-index was 0.86, and the adjusted C-index was 0.85, which was superior to the tumor-node-metastasis stage alone (C-index 0.812).
DISCUSSION
Through correlation analysis and survival analysis, we confirmed that inflammatory and nutritional indicators are closely associated with the prognosis of AGC patients undergoing gastrectomy and prophylactic HIPEC. In this study, we successfully constructed a novel INPI by combining inflammation and nutrition-related indicators to stratify patient risk effectively. The nomogram, constructed based on the INPI and clinicopathological features, was effective in predicting the survival of patients who underwent radical gastrectomy with prophylactic HIPEC for GC.
In this study, the overall 3-year survival rate was only 53.6%. GC patients are often diagnosed at advanced stages and have high rates of recurrence and mortality; peritoneal recurrence is one of the most common patterns of GC recurrence[18,19]. As an early intervention measure, prophylactic HIPEC is aimed at preventing postoperative peritoneal recurrence in patients with locally AGC who do not have obvious peritoneal metastases. However, the efficacy of prophylactic HIPEC remains controversial. The advantage of HIPEC lies in its ability to directly deliver a large amount of antineoplastic drugs into the peritoneal cavity, reduce their systemic toxicity, and utilize the synergistic effects of hyperthermia to enhance antitumor efficacy in multiple ways[20,21]. A study by Reutovich et al[22] showed that the incidence of metachronous peritoneal metastasis (MPM) was significantly lower in the prophylactic HIPEC treatment group than in the surgery-only group (12.8% vs 27.6%, P < 0.001), while the difference in the rate of complications was not statistically significant (P < 0.254). Prophylactic HIPEC has shown promising efficacy in preventing synchronous and MPM in Asian populations, but there is a lack of conclusive evidence proving its applicability in Western populations[8]. A study by Coccolini et al[23] demonstrated that prophylactic HIPEC combined with neoadjuvant chemotherapy can increase DFS and OS in AGC patients without peritoneal carcinomatosis, but there is no definitive evidence proving its efficacy. Therefore, our research focused primarily on identifying the factors that influence the prognosis of patients receiving prophylactic HIPEC. Accurate risk assessment and the development of personalized treatment strategies are of paramount importance.
This study demonstrated that the INPI, which was developed based on preoperative BMI, neutrophil count, the SII, the PNI, the PAR, the AGR, and hemoglobin, is an independent prognostic indicator for AGC patients undergoing prophylactic HIPEC treatment after curative surgery. Although globulin, albumin, and platelet count were excluded during the construction of INPI through multicollinearity analysis, they also play important roles in the occurrence and development of cancer. Platelets play a multifaceted role in tumor progression. Platelets can promote angiogenesis in tumor tissues by releasing transforming growth factor. They also help circulating tumor cells evade host immune surveillance, thereby facilitating tumor cell invasion and metastasis[24]. Albumin, as an important indicator of nutritional status, is involved in maintaining colloid osmotic pressure and transporting nutrients and drugs. Additionally, albumin can inhibit the production of pro-inflammatory factors such as tumor necrosis factor-α, thus exerting an anti-inflammatory effect[25,26]. The AGR composed of albumin and globulin, plays a significant role in tumor prediction. The study by Mao et al[27] indicated that AGR is an independent prognostic factor for OS (HR: 0.578, 95%CI: 0.373-0.897, P = 0.015), and it can serve as a prognostic biomarker for the OS of GC patients. Many studies have confirmed that inflammatory and nutritional indicators are closely related to the prognosis of GC patients. Except for AGR, the other risk factors included in the INPI are consistent with previous research findings. A large cohort study has shown that among patients who undergo radical GC surgery, those who are overweight or have mild to moderate obesity before surgery have better OS and CSS compared to patients with normal body weight[28]. The study by Wang et al[29] demonstrated that PAR is an independent predictor of CSS and DFS in patients with stage II-III GC.
The study by Inoue et al[30] included 447 patients who underwent radical gastrectomy. The results showed that the 5-year OS of patients with a high preoperative SII was significantly lower than that of the low SII group (P < 0.001), and the peritoneal recurrence rate in the high SII group was also higher than that in the low SII group (P = 0.028). The study by Sun et al[31] used the PNI to predict the prognosis of GC patients treated with immune checkpoint inhibitors. A total of 146 patients were included in the study. In all patients, the PFS and OS of the low PNI group were shorter (HR = 1.913, P = 0.013 and HR = 2.332, P = 0.001). Some meta-analysis studies have also shown that the OS rate of GC patients with preoperative anemia is poor (HR = 1.33, 95%CI: 1.21-1.45), and the DFS rate is significantly lower (HR = 1.62, 95%CI: 1.13-2.32)[32]. Survival analysis indicated that the INPI can be used to stratify patients into three risk groups, and patients in the high risk group have a poor prognosis. In clinical practice, the prognostic risk of patients can be evaluated based on their preoperative test indicators. For patients with a high INPI score, efforts can be made to improve their nutritional and inflammatory status. In terms of postoperative chemotherapy, the chemotherapy cycle can be extended or drugs can be administered at a high dose. Additionally, postoperative follow up for high risk patients should be strengthened, and close monitoring for tumor recurrence should be carried out. Notably, due to the presence of multicollinearity among novel inflammation and nutrition-related indicators, for enhanced model stability, we conducted collinearity analysis during INPI construction, selected features and identified the most crucial predictors associated with survival time through LASSO Cox regression.
Our study revealed that GC patients with high neutrophil counts have a poorer prognosis (Figure 3E), which also supports the interesting phenomenon of "self-contradiction" of neutrophils in tumor development[33]. The tumor microenvironment (TME) is closely related to tumor initiation and progression. It is composed of various factors, with inflammatory cells playing a crucial coordinating role. Inflammation is an indispensable element in tumor progression, promoting proliferation, survival, and migration[34,35]. Tumor-associated neutrophils within the TME play dual roles in cancer initiation and development. Neutrophils play an essential role in early-stage tumor defense by reducing cancer-related inflammation and stimulating T-cell activation and proliferation, among other mechanisms. Conversely, they can promote tumor growth and metastasis by reshaping TME[36]. Although research on the relationship between neutrophils and GC prognosis is limited, Feng et al's study[37] indicated that a high absolute neutrophil count is associated with poor prognosis in GC patients, which is consistent with our findings. The impact of inflammation and nutrition on tumors is intricately connected at the microscopic level, and the INPI may concretize this relationship.
We further combined the INPI with other independent clinical features to construct a prognostic nomogram for AGC patients, which demonstrated good prognostic performance. Since pathological information is unavailable preoperatively, the nomogram includes the clinical features of age at diagnosis and NRS-2002 score. The relationship between age and prognosis in patients with GC remains controversial[38,39]. Our study indicated that age > 60 years is an independent risk factor for GC prognosis. Regarding the NRS-2002 score, previous research[40] has used a cutoff value of 3 points to distinguish patients' nutritional status. In our study, we resegmented the scoring process, and the results showed that the NRS-2002 score has clinical value in predicting the prognosis of GC patients. Nomograms constructed based on the INPI and those constructed by combining postoperative pathology with clinical features both demonstrated good predictive performance, with similar adjusted C-indices (0.83 vs 0.85), indicating comparable predictive capabilities.
The strengths of this study lie in the inclusion of a substantial number of postoperative pathologically confirmed cT4-stage GC patients. Standard D2 gastrectomy was performed by skilled and experienced gastrointestinal surgeons, with the center conducting nearly 1000 GC surgeries annually to ensure procedural consistency. Additionally, this was a study focusing on AGC that considered the intrinsic characteristics of tumors. A graphical representation was constructed by integrating clinical features, the immune-inflammatory microenvironment, and patient nutritional status, providing a visual tool for intuitive prognosis prediction. Importantly, the parameters of the INPI are routine and readily available in clinical practice, rendering it a cost-effective and valuable tool for prognosis grading, optimization of treatment decision, and postoperative follow-up strategy guidance. Our study explored the use of prophylactic HIPEC as an early intervention for peritoneal metastasis in GC patients, representing a relatively novel research direction with practical implications for improving the prognosis of AGC patients.
However, our study has several limitations. This was a single-center retrospective analysis with a small sample size, which limits the generalizability of our research findings. A single-center study may introduce selection bias due to the unique patient population and treatment methods specific to our institution. Regarding the limited sample size, it may lead to model overfitting and reduce the precision of our estimates. We are currently preparing a multi-center validation study. By collaborating with other top cancer centers and standardizing the data collection process among these centers, we aim to address the issue of insufficient sample size and enhance the generalizability of our research findings. Additionally, the complex biological characteristics of cancer, such as genetic factors, socioeconomic influences, and lifestyle factors affecting the prognosis of patients with GC, could not be fully encompassed in our study.
CONCLUSION
As an effective and cost-efficient scoring system, the INPI has promising clinical value for predicting postoperative survival in AGC patients who undergo D2 gastrectomy with HIPEC. The prognostic performance of the INPI-based prognostic nomogram makes it an effective tool for devising personalized treatment strategies.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Gastroenterology and hepatology
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade B
Novelty: Grade B
Creativity or Innovation: Grade A
Scientific Significance: Grade B
P-Reviewer: Jiao Y S-Editor: Li L L-Editor: A P-Editor: Zhao YQ
van Driel WJ, Koole SN, Sikorska K, Schagen van Leeuwen JH, Schreuder HWR, Hermans RHM, de Hingh IHJT, van der Velden J, Arts HJ, Massuger LFAG, Aalbers AGJ, Verwaal VJ, Kieffer JM, Van de Vijver KK, van Tinteren H, Aaronson NK, Sonke GS. Hyperthermic Intraperitoneal Chemotherapy in Ovarian Cancer.N Engl J Med. 2018;378:230-240.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 844][Cited by in RCA: 959][Article Influence: 137.0][Reference Citation Analysis (0)]
Di Giorgio A, Gerardi C, Abatini C, Melotti G, Bonavina L, Torri V, Santullo F, Garattini S, De Luca M, Rulli E, Rulli E, Pacelli F; GOETH Investigators. Prophylactic surgery plus hyperthermic intraperitoneal chemotherapy (HIPEC CO2) versus standard surgery for gastric carcinoma at high risk of peritoneal carcinomatosis: short and long-term outcomes (GOETH STUDY)-a collaborative randomized controlled trial by ACOI, FONDAZIONE AIOM, SIC, SICE, and SICO.Trials. 2022;23:969.
[RCA] [PubMed] [DOI] [Full Text][Reference Citation Analysis (0)]
Muscaritoli M, Lucia S, Farcomeni A, Lorusso V, Saracino V, Barone C, Plastino F, Gori S, Magarotto R, Carteni G, Chiurazzi B, Pavese I, Marchetti L, Zagonel V, Bergo E, Tonini G, Imperatori M, Iacono C, Maiorana L, Pinto C, Rubino D, Cavanna L, Di Cicilia R, Gamucci T, Quadrini S, Palazzo S, Minardi S, Merlano M, Colucci G, Marchetti P; PreMiO Study Group. Prevalence of malnutrition in patients at first medical oncology visit: the PreMiO study.Oncotarget. 2017;8:79884-79896.
[RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)][Cited by in Crossref: 209][Cited by in RCA: 257][Article Influence: 32.1][Reference Citation Analysis (0)]
Fujino S, Myoshi N, Saso K, Sasaki M, Ishikawa S, Takahashi Y, Yasui M, Ohue M, Hata T, Matsuda C, Mizushima T, Mori M, Doki Y. The inflammation-nutrition score supports the prognostic prediction of the TNM stage for colorectal cancer patients after curative resection.Surg Today. 2020;50:163-170.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 13][Cited by in RCA: 13][Article Influence: 2.6][Reference Citation Analysis (0)]
Kim YJ, Hiratsuka Y, Suh SY, Won SH, Jung EH, Kang B, Lee SW, Ahn HY, Suh KJ, Kim JW, Kim SH, Kim JW, Lee KW, Kim JH, Lee JS. Performance of mid-upper arm circumference and other prognostic indices based on inflammation and nutrition in oncology outpatients: a tertiary cancer center study.Ann Palliat Med. 2022;11:3171-3180.
[RCA] [PubMed] [DOI] [Full Text][Reference Citation Analysis (0)]
Xie H, Ruan G, Wei L, Zhang H, Zhang Q, Ge Y, Lin S, Song M, Zhang X, Liu X, Zhang X, Li X, Zhang K, Yang M, Tang M, Deng L, Shi H. A novel inflammation-nutrition biomarker score for predicting prognosis of patients with cancer: results from a multicenter study.BMC Cancer. 2022;22:1311.
[RCA] [PubMed] [DOI] [Full Text][Reference Citation Analysis (0)]
Li JH, Zhang SW, Liu J, Shao MZ, Chen L. Review of clinical investigation on recurrence of gastric cancer following curative resection.Chin Med J (Engl). 2012;125:1479-1495.
[PubMed] [DOI] [Full Text]
