BPG is committed to discovery and dissemination of knowledge
Retrospective Study Open Access
Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.
World J Gastrointest Oncol. Jun 15, 2026; 18(6): 117582
Published online Jun 15, 2026. doi: 10.4251/wjgo.v18.i6.117582
Prognostic significance of baseline C-reactive protein/albumin ratio for long-term outcomes in advanced gastric cancer under immunochemotherapy
Shuai Mei, Mei-Ling Ye, Qiu-Lin Hao, Geng-Chen Li, Zhi-Yuan Yao, Hai-Yan Wang, Department of Oncology, The Affiliated Hospital of Xuzhou Medical University, Xuzhou 221000, Jiangsu Province, China
ORCID number: Hai-Yan Wang (0009-0002-4508-9187).
Co-first authors: Shuai Mei and Mei-Ling Ye.
Co-corresponding authors: Zhi-Yuan Yao and Hai-Yan Wang.
Author contributions: Mei S, Ye ML, Yao ZY, and Wang HY contributed to the conceptualization, writing-review and editing of this manuscript, and the project administration and the supervision of this manuscript; Mei S and Wang HY were responsible for the methodology of this study; Mei S and Ye ML contributed to the formal analysis of this manuscript and the visualization of this article, responsible for the validation of this manuscript; Mei S, Ye ML, and Hao QL took part in the writing-original draft and investigation of this manuscript; Hao QL and Li GC took part in the data curation of this study; Li GC took part in the resources; Yao ZY and Wang HY were involved in the supervision of this study; Mei S and Ye ML contributed equally to the manuscript, they are co-first authors of this manuscript; Yao ZY and Wang HY contributed to this manuscript equally, they are co-corresponding authors of this study.
AI contribution statement: No entirety or any portion of the main text was AI-generated. All content was independently written by the authors. Only ChatGPT was used for language polishing and grammatical revision, not for full-text translation or manuscript writing. No AI tool participated in the study design, data analysis or interpretation of results. All figures and images in the manuscript are original experimental results of the authors, and no images were generated by AI.
Institutional review board statement: This research was carried out following the Declaration of Helsinki and received approval from the Ethics Committee at the Affiliated Hospital of Xuzhou Medical University (approval No. XYFY2025-KL385-01).
Informed consent statement: Given the retrospective design of this investigation, the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University granted us an exemption from obtaining written informed consent.
Conflict-of-interest statement: The authors declare that they have no conflict of interest.
Data sharing statement: Technical appendix, statistical code, and dataset available from the corresponding author at xzhaiyan68@163.com. Participants gave informed consent for data sharing.
Corresponding author: Hai-Yan Wang, MD, Doctor, Department of Oncology, The Affiliated Hospital of Xuzhou Medical University, No. 99 Huaihai West Road, Quanshan District, Xuzhou 221000, Jiangsu Province, China. xzhaiyan68@163.com
Received: December 11, 2025
Revised: January 5, 2026
Accepted: January 27, 2026
Published online: June 15, 2026
Processing time: 181 Days and 1.7 Hours

Abstract
BACKGROUND

Patients with advanced gastric cancer typically exhibit a dismal prognosis. Although combined immunotherapy and chemotherapy have been shown to improve survival in a subset of patients, substantial interindividual heterogeneity in therapeutic response persists, and reliable, easily obtainable prognostic biomarkers remain limited. The C-reactive protein/albumin ratio (CAR), a composite indicator of systemic inflammation and nutritional status, has demonstrated prognostic relevance across multiple malignancies. Nevertheless, its long-term prognostic significance in advanced gastric cancer patients receiving combined immunochemotherapy has not yet been clearly defined.

AIM

To assess baseline CAR’s prognostic value and develop a model for guiding personalized immunochemotherapy in advanced gastric cancer.

METHODS

Retrospectively, we collected clinical data from 200 advanced gastric cancer patients treated with first-line immunochemotherapy at the Affiliated Hospital of Xuzhou Medical University. The optimal CAR cutoff was determined by receiver operating characteristic curve (ROC) analysis, stratifying patients into high and low CAR groups. Survival was assessed using the Kaplan-Meier method and compared with the log-rank test. Cox regression evaluated CAR’s association with progression-free survival (PFS) and overall survival (OS). A prognostic nomogram was then constructed based on significant factors.

RESULTS

ROC curve analysis showed that the baseline CAR had an area under the curve of 0.80 [95% confidence intervals (CI): 0.76-0.85] for predicting OS, with an optimal cutoff value of 0.21. Based on this threshold, patients were stratified into a low CAR group (CAR < 0.21; n = 120) and a high CAR group (CAR ≥ 0.21; n = 80). The low CAR group exhibited a significantly longer median PFS of 12.3 months (95%CI: 9.5-21.0) compared with 7.2 months (95%CI: 6.2-9.8) in the high CAR group (P = 0.0023). Similarly, the median OS was markedly prolonged in the low CAR group (20.4 months; 95%CI: 18.5-25.1) relative to the high CAR group (14.3 months; 95%CI: 13.4-16.1; P < 0.001). The low CAR group also demonstrated superior objective response rate (ORR) and disease control rate (DCR), at 59.17% and 79.17%, respectively, compared with 45.00% and 66.25% in the high CAR group (P = 0.049; P = 0.041). Multivariate Cox regression analysis identified baseline CAR, programmed death-ligand 1 expression, mismatch repair status, tumor stage, and peritoneal metastasis as independent prognostic factors for both PFS and OS. A prognostic column-plot model integrating these variables exhibited robust discriminative ability and calibration in both the training and validation cohorts, with C-indices of 0.80 and 0.82 for PFS, and 0.78 and 0.73 for OS, respectively.

CONCLUSION

The baseline CAR index is an independent prognostic factor for long-term outcomes in advanced gastric cancer patients treated with combination immunotherapy and chemotherapy. Lower CAR levels correlate with improved PFS, OS, and higher ORR and DCR. The calibration curve model, incorporating CAR and clinical-pathological factors, demonstrates strong predictive performance, providing valuable insights for prognosis and personalized treatment.

Key Words: Gastric cancer; C-reactive protein/albumin ratio; Programmed death-1 inhibitor; Predictive model; Efficacy; Safety

Core Tip: The baseline C-reactive protein/albumin ratio (CAR) is a significant independent prognostic factor for advanced gastric cancer patients receiving combination immunotherapy and chemotherapy, with lower CAR levels associated with improved progression-free survival, overall survival, and higher objective response and disease control rates. By incorporating both inflammatory and nutritional status, the CAR index provides a more comprehensive evaluation of the immune microenvironment compared to traditional single inflammatory biomarkers. The CAR-integrated nomogram shows strong predictive accuracy, guiding personalized immunotherapy strategies. However, prospective, multi-center studies are needed to validate and refine its clinical applicability.



INTRODUCTION

Gastric cancer, one of the most prevalent malignancies of the digestive system globally, remains a leading cause of cancer-related mortality[1,2]. Its incidence and mortality rates are particularly high in East Asia[1,3]. Due to the fact that, most patients are diagnosed at locally advanced or metastatic stages, the limited number of candidates for curative surgery results in a generally poor prognosis for advanced gastric cancer[2,4]. Recent advancements in first-line treatment, combining immune checkpoint inhibitors with chemotherapy, have improved survival outcomes for certain patients[5-7]. However, significant interindividual variations in treatment response persist, underscoring the urgent need for accessible and reliable biomarkers to predict therapeutic efficacy and assess prognosis[8-10].

The C-reactive protein (CRP)/albumin (ALB) ratio (CAR) is a comprehensive biomarker that reflects systemic inflammatory responses and nutritional status[11-13]. As an acute-phase reactant, CRP indicates the severity of systemic inflammation, while serum ALB levels are associated with nutritional status, immune function, and chronic inflammation[11,14]. CAR integrates both inflammatory and nutritional signals, and has been shown to be a strong prognostic indicator in various solid tumors, including hepatocellular carcinoma, lung cancer, and colorectal cancer[13,15-18]. Emerging evidence suggests that elevated CAR levels are linked to tumor progression, an immune-suppressive microenvironment, and poor survival outcomes, positioning CAR as a promising biomarker for predicting the efficacy of immunotherapy[19-22].

Although existing studies have explored the prognostic value of CAR in gastric cancer, evidence regarding its predictive role in advanced gastric cancer patients receiving combination immunotherapy and chemotherapy remains limited[23,24]. On one hand, there is a lack of systematic studies evaluating the relationship between baseline CAR index and the efficacy of combination immunotherapy and chemotherapy, as well as long-term survival outcomes[23]. On the other hand, prognostic models based on CAR for this patient population have not been sufficiently reported[25]. Furthermore, with combination immunotherapy and chemotherapy increasingly becoming a first-line treatment for advanced gastric cancer, traditional prognostic indicators based on tumor burden or histological features no longer comprehensively reflect the heterogeneity of immune-related responses[5,26]. In contrast, simple clinical indices that simultaneously reflect systemic inflammation and immune status may hold greater clinical potential[27]. Therefore, assessing the prognostic value of CAR in the context of combination immunotherapy and chemotherapy is of significant importance.

Against this background, the aim of this study is to evaluate the long-term prognostic value of baseline CAR index in advanced gastric cancer patients undergoing combination immunotherapy and chemotherapy. Additionally, we aim to develop a nomogram-based predictive model centered on CAR to provide a more reliable and accessible reference for prognosis assessment and personalized treatment strategies in advanced gastric cancer.

MATERIALS AND METHODS
Study design and patients

This study is a single-center, retrospective analysis aimed at evaluating the efficacy and prognosis of sintilimab combined with chemotherapy in human epidermal growth factor receptor 2-negative (HER-2-negative), unresectable locally advanced or metastatic gastric cancer and gastroesophageal junction adenocarcinoma. The study was conducted at the Affiliated Hospital of Xuzhou Medical University, spanning from January 2021 to December 2024. The inclusion criteria were as follows: (1) Histologically or cytologically confirmed unresectable or metastatic gastric cancer or gastroesophageal junction adenocarcinoma; (2) Age ≥ 18 years; (3) HER-2 negative status; (4) An Eastern Cooperative Oncology Group (ECOG) performance status of 0-1; (5) Presence of measurable lesions according to response evaluation criteria in solid tumors version (RECIST v1.1) criteria; (6) Receipt of at least two cycles of sintilimab combined with chemotherapy; and (7) Availability of complete follow-up data. The exclusion criteria included: (1) Patients who had previously received systemic therapy; (2) Those with severe dysfunction of the heart, liver, kidneys, or other vital organs; (3) Hemoglobin < 80 g/L or platelets < 75 × 109/L; (4) Presence of autoimmune diseases; (5) Known allergy to the study drug; (6) Patients with other advanced malignancies; (7) Patients with evidence of active infection at baseline; (8) Patients with active inflammatory diseases or chronic inflammatory conditions; and (9) Patients with known liver cirrhosis or clinically significant chronic liver disease. The study was approved by the Ethics Committee of the Affiliated Hospital of Xuzhou Medical University (Ethical approval number: No. XYFY2025-KL385-01). Given the retrospective nature of the study, the Ethics Committee waived the requirement for informed consent, and the use of data complied with relevant legal and regulatory requirements.

Definition

The CAR index is calculated using the following formula: CAR = CRP concentration (mg/L)/serum ALB concentration (g/L). Both CRP and ALB concentrations are baseline hematological parameters, collected within one week prior to the initiation of combination immunotherapy and chemotherapy. CRP levels were measured using chemiluminescent immunoassay, a method known for its high sensitivity and accuracy in quantifying CRP levels in serum. ALB concentration was determined using an automated biochemical analyzer, ensuring stable and standardized results in accordance with clinical laboratory testing requirements.

In this study, overall survival (OS) in advanced gastric cancer patients receiving combination immunotherapy and chemotherapy was the primary endpoint. A receiver operating characteristic (ROC) curve was constructed based on the baseline CAR values obtained prior to treatment. The area under the curve (AUC) was 0.80 [95% confidence intervals (CI): 0.76-0.85], and the optimal cutoff value was determined to be 0.21 by maximizing the product of sensitivity (75.20%) and specificity (84.25%).

Data collection

Data were collected through the electronic medical record system, which included patients’ baseline demographic information (e.g., gender, age, body mass index, ECOG performance status) as well as baseline hematological parameters obtained within one week prior to initiation of combination immunotherapy and chemotherapy [e.g., neutrophil and lymphocyte counts, CRP, ALB levels, carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9)]. Additionally, tumor characteristics were recorded, including tumor location, tumor differentiation, metastatic sites, programmed death-ligand 1 (PD-L1) expression, mismatch repair (MMR) status, and Epstein-Barr virus (EBV) status. All data were verified and entered by trained professionals to ensure the reliability and accuracy of the study.

Treatment protocol

All enrolled patients received sintilimab combined with chemotherapy. Sintilimab was administered as a 200 mg intravenous infusion every 3 weeks. The chemotherapy regimen was selected based on the patient’s condition and included one of the following two options: Capecitabine was administered orally at a dose of 1000 mg/m2 twice daily after meals for 14 consecutive days (Days 1-14), with oxaliplatin given as a 130 mg/m2 intravenous infusion on Day 1. This regimen was repeated every 3 weeks for a total of 6 cycles. Alternatively, Tegafur-Uracil was administered orally at a dose based on body surface area (BSA): 40 mg for BSA < 1.25 m2, 50 mg for BSA ≥ 1.25 m2 but < 1.5 m2, and 60 mg for BSA ≥ 1.5 m2, twice daily after meals for Days 1-14, with oxaliplatin administered as a 130 mg/m2 intravenous infusion on Day 1. This regimen was also repeated every 3 weeks for 6 cycles. For patients without disease progression, maintenance therapy consisted of sintilimab combined with either Tegafur-Uracil or capecitabine, continued until disease progression, patient death, or intolerable adverse events. All patients were followed up until death or the data cutoff date (December 31, 2024).

Efficacy evaluation

The efficacy of this study was assessed according to the RECIST v1.1, focusing on both recent treatment outcomes and survival endpoints. Recent treatment outcomes were classified into four categories: Complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD). CR was defined as the complete disappearance of all measurable lesions, sustained for at least 4 weeks; PR was defined as a reduction of at least 30% in the maximum diameter of at least one measurable lesion, sustained for at least 4 weeks; SD referred to cases where tumor size did not meet the criteria for PR or PD, sustained for at least 4 weeks; and PD was defined as an increase of more than 20% in the size of the tumor or the appearance of new lesions. Based on these criteria, the objective response rate (ORR) and disease control rate (DCR) were calculated, with ORR representing the proportion of patients with CR or PR, and DCR representing the proportion of patients with CR, PR, or SD.

Survival outcomes included progression-free survival (PFS) and OS. PFS was defined as the time from the initiation of treatment until the first occurrence of disease progression or death, used to assess the effectiveness of disease control post-treatment. OS was defined as the time from the initiation of treatment until death from any cause. All efficacy assessments were conducted through regular imaging examinations, evaluated by independent radiologists, with follow-up imaging performed at weeks 6 and 12 post-treatment, and every three months thereafter, until disease progression or patient death. The final follow-up occurred in December 2024, with a median overall follow-up duration of 20 months.

Statistical analysis

All statistical analyses were performed using R software (version 4.5.1). Continuous variables were described using median values, and intergroup comparisons were conducted using the Wilcoxon rank-sum test. Categorical variables were compared using the χ2 test. Survival curves were plotted using the Kaplan-Meier method and compared with the Log-rank test. Cox proportional hazards models were used to identify independent prognostic factors affecting PFS and OS, with hazard ratios (HR) and 95%CI reported. To assess the collinearity among the biological upstream variables of CRP and ALB, the variance inflation factor (VIF) was calculated. All VIF values were found to be below 5, indicating that there was no significant collinearity among the selected variables. A nomogram model was constructed based on the results of multivariate analysis, and its predictive performance was evaluated using Harrell’s C-index, ROC curve analysis, AUC, and bootstrap calibration (1000 iterations). All statistical tests were two-sided, and a P value < 0.05 was considered statistically significant.

RESULTS
Patient characteristics

This study is a single-center retrospective analysis that included 200 advanced gastric cancer patients who received first-line combination immunotherapy and chemotherapy at the Affiliated Hospital of Xuzhou Medical University. Based on the optimal cutoff value derived from the ROC curve, patients were stratified into two groups: A low CAR index group (n = 120) and a high CAR index group (n = 80). Table 1 summarizes the baseline characteristics of the patients, including general demographic information (e.g., gender, age, body mass index, ECOG performance status) and hematological parameters obtained within one week prior to the initiation of immunochemotherapy (e.g., neutrophil and lymphocyte counts, CRP concentration, ALB concentration, CEA, CA199). Additionally, tumor characteristics were collected, including tumor location, tumor staging, tumor differentiation, metastatic sites, PD-L1 expression status, MMR status, and EBV infection status. Significant differences between the two groups were observed in ALB concentration, CRP levels, and CAR index (P < 0.001).

Table 1 Baseline characteristics stratified by C-reactive protein to albumin ratio (< 0.21 vs ≥ 0.21), mean ± SD/n (%).
Variables
Total (n = 200)
High CAR (n = 80)
Low CAR (n = 120)
Statistic
P value
Neutrophil3.91 ± 2.324.05 ± 2.443.81 ± 2.24t = 0.720.475
Lymphocyte1.38 ± 0.591.36 ± 0.671.39 ± 0.53t = -0.370.713
Albumin39.24 ± 5.7336.59 ± 6.2941.01 ± 4.55t = -5.42< 0.001
CRP17.33 ± 27.0939.74 ± 31.522.38 ± 2.28t = 10.58< 0.001
CAR0.49 ± 0.791.13 ± 0.940.06 ± 0.06t = 10.14< 0.001
Sexχ2 = 1.420.233
Female75 (37.50)26 (32.50)49 (40.83)
Male125 (62.50)54 (67.50)71 (59.17)
Ageχ2 = 0.420.515
< 60 years78 (39.00)29 (36.25)49 (40.83)
≥ 60 years122 (61.00)51 (63.75)71 (59.17)
BMIχ2 = 1.000.318
< 25166 (83.00)69 (86.25)97 (80.83)
≥ 2534 (17.00)11 (13.75)23 (19.17)
CEAχ2 = 2.540.111
< 379 (39.50)37 (46.25)42 (35.00)
≥ 3121 (60.50)43 (53.75)78 (65.00)
CA199χ2 = 1.030.309
< 37126 (63.00)47 (58.75)79 (65.83)
≥ 3774 (37.00)33 (41.25)41 (34.17)
ECOG PSχ2 = 2.080.149
072 (36.00)24 (30.00)48 (40.00)
1128 (64.00)56 (70.00)72 (60.00)
Stagingχ2 = 0.480.487
III91 (45.50)34 (42.50)57 (47.50)
IV109 (54.50)46 (57.50)63 (52.50)
Differentiationχ2 = 2.080.149
Poorly/undifferentiated180 (90.00)75 (93.75)105 (87.50)
Medium/high20 (10.00)5 (6.25)15 (12.50)
Siteχ2 = 0.000.949
Gastric and esophageal binding58 (29.00)23 (28.75)35 (29.17)
Stomach142 (71.00)57 (71.25)85 (70.83)
Peritoneal metastasisχ2 = 2.840.092
No157 (78.50)58 (72.50)99 (82.50)
Yes43 (21.50)22 (27.50)21 (17.50)
Liver metastasisχ2 = 3.540.060
No156 (78.00)57 (71.25)99 (82.50)
Yes44 (22.00)23 (28.75)21 (17.50)
EBV statusχ2 = 1.660.198
No-infect187 (93.50)77 (96.25)110 (91.67)
Infect13 (6.50)3 (3.75)10 (8.33)
PD-L1 expressionχ2 = 2.290.130
CPS < 587 (43.50)40 (50.00)47 (39.17)
CPS ≥ 5113 (56.50)40 (50.00)73 (60.83)
MMR statusχ2 = 2.240.135
pMMR168 (84.00)71 (88.75)97 (80.83)
dMMR32 (16.00)9 (11.25)23 (19.17)
Tumor response

Table 2 presents the treatment response data for the two groups of patients. In the low CAR index group, 7 patients achieved CR (5.83%), 64 patients achieved PR (53.33%), 24 patients had SD (20.00%), and 25 patients experienced PD (20.83%). In the high CAR index group, 4 patients achieved CR (5.00%), 32 patients achieved PR (40.00%), 17 patients had SD (21.25%), and 27 patients experienced PD (33.75%). The ORR in the low CAR index group was modestly but significantly higher than that in the high CAR index group (59.17% vs 45.00%, P = 0.049), and the DCR was also modestly but significantly higher (79.17% vs 66.25%, P = 0.041).

Table 2 Tumor responses stratified by C-reactive protein to albumin ratio (< 0.21 vs ≥ 0.21), n (%).
Variables
Total (n = 200)
High (n = 80)
Low (n = 120)
Statistic
P value
CR
No189 (94.50)76 (95.00)113 (94.17)
Yes11 (5.50)4 (5.00)7 (5.83)
PR
No104 (52.00)48 (60.00)56 (46.67)
Yes96 (48.00)32 (40.00)64 (53.33)
SD
No159 (79.50)63 (78.75)96 (80.00)
Yes41 (20.50)17 (21.25)24 (20.00)
PD
No148 (74.00)53 (66.25)95 (79.17)
Yes52 (26.00)27 (33.75)25 (20.83)
ORRχ2 = 3.870.049
No93 (46.50)44 (55.00)49 (40.83)
Yes107 (53.50)36 (45.00)71 (59.17)
DCRχ2 = 4.160.041
No52 (26.00)27 (33.75)25 (20.83)
Yes148 (74.00)53 (66.25)95 (79.17)
PFS and OS

Figure 1A shows that the median PFS in the low CAR index group was 12.3 months (95%CI: 9.5-21.0), significantly longer than the 7.2 months (95%CI: 6.2-9.8) observed in the high CAR index group (P = 0.0023). Regarding OS, the median OS in the low CAR index group was 20.4 months (95%CI: 18.5-25.1), significantly longer than the 14.3 months (95%CI: 13.4-16.1) in the high CAR index group (P < 0.0001) (Figure 1B).

Figure 1
Figure 1 Effects of different C-reactive protein/albumin ratio index on the long-term prognosis of gastric cancer patients. A: Kaplan-Meier plot of progression-free survival in the C-reactive protein/albumin ratio (CAR) index < 0.21 and CAR index ≥ 0.21 groups; B: Kaplan-Meier plot of overall survival in the CAR index < 0.21 and CAR index ≥ 0.21 groups. CAR: C-reactive protein/albumin ratio; PFS: Progression-free survival; OS: Overall survival; CI: Confidence interval.
Univariate and multifactorial analyses of PFS

Univariate Cox regression analysis for PFS revealed that the CAR index, MMR status, PD-L1 expression, tumor stage, tumor differentiation, and peritoneal metastasis were significantly associated with PFS. Further multivariate Cox regression analysis identified the following independent prognostic factors for PFS: Low CAR index (HR = 0.67, 95%CI: 0.46-0.97, P = 0.034), high PD-L1 expression (HR = 0.42, 95%CI: 0.28-0.62, P < 0.001), deficient MMR (dMMR) (HR = 0.34, 95%CI: 0.18-0.65, P = 0.001), advanced tumor stage (HR = 1.67, 95%CI: 1.10-2.52, P = 0.015), and peritoneal metastasis (HR = 1.95, 95%CI: 1.31-2.88, P < 0.001) (Table 3).

Table 3 Univariate and multivariate analyses of prognostic factors for progression-free survival.
VariablesUnivariate
Multivariate
P value
HR (95%CI)
P value
HR (95%CI)
PD-L1 expression (CPS < 5 vs CPS ≥ 5)< 0.0010.34 (0.24-0.50)< 0.0010.42 (0.28-0.62)
MMR status (pMMR vs dMMR)< 0.0010.25 (0.13-0.48)0.0010.34 (0.18-0.65)
CAR (high vs low)0.0030.57 (0.40-0.82)0.0340.67 (0.46-0.97)
Sex (male vs female)0.9081.02 (0.70-1.49)
Age (≥ 60 vs < 60), years0.1141.36 (0.93-2.00)
BMI (< 25 vs ≥ 25), kg/m20.8131.06 (0.66-1.70)
CEA (< 3 vs ≥ 3), ng/mL0.8531.04 (0.72-1.50)
CA199 (< 37 vs ≥ 37), ng/mL0.3651.19 (0.82-1.73)
ECOG (0 vs 1)0.8211.04 (0.72-1.52)
Staging (III vs IV) < 0.0012.55 (1.74-3.75)0.0151.67 (1.10-2.52)
Differentiation (poorly/undifferentiated vs medium/high)0.0410.45 (0.21-0.97)0.2350.62 (0.29-1.36)
Site (gastric and esophageal binding vs stomach)0.7720.94 (0.63-1.40)
Peritoneal metastasis (no vs yes)< 0.0012.08 (1.44-3.00)< 0.0011.95 (1.31-2.88)
Liver metastasis (no vs yes)0.0721.47 (0.97-2.25)
EBV status (infect vs no-infect)0.0990.47 (0.19-1.15)
Neutrophil0.2801.04 (0.97-1.11)
Lymphocyte0.9711.01 (0.75-1.36)
Univariate and multifactorial analyses of OS

Univariate Cox regression analysis for OS showed that the CAR index, MMR status, PD-L1 expression, tumor stage and peritoneal metastasis were significantly associated with OS. Further multivariate Cox regression analysis identified the following independent prognostic factors for OS: Low CAR index (HR = 0.51, 95%CI: 0.35-0.73, P < 0.001), high PD-L1 expression (HR = 0.45, 95%CI: 0.30-0.66, P < 0.001), dMMR (HR = 0.34, 95%CI: 0.18-0.64, P = 0.001), advanced tumor stage (HR = 1.70, 95%CI: 1.12-2.58, P = 0.012), and peritoneal metastasis (HR = 1.73, 95%CI: 1.17-2.54, P = 0.005) (Table 4).

Table 4 Univariate and multivariate analyses of prognostic factors for overall survival.
VariablesUnivariate
Multivariate
P value
HR (95%CI)
P value
HR (95%CI)
PD-L1 expression (CPS < 5 vs CPS ≥ 5)< 0.0010.34 (0.23-0.49)< 0.0010.45 (0.30-0.66)
MMR status (pMMR vs dMMR)< 0.0010.22 (0.12-0.42)0.0010.34 (0.18-0.64)
CAR (high vs low)< 0.0010.45 (0.31-0.64)< 0.0010.51 (0.35-0.73)
Sex (male vs female)0.5950.91 (0.63-1.31)
Age (≥ 60 vs < 60), years0.7021.07 (0.74-1.55)
BMI (< 25 vs ≥ 25), kg/m20.4080.82 (0.51-1.31)
CEA (< 3 vs ≥ 3), ng/mL0.6441.09 (0.76-1.57)
CA199 (< 37 vs ≥ 37), ng/mL0.5441.12 (0.78-1.62)
ECOG (0 vs 1)0.3261.21 (0.83-1.76)
Staging (III vs IV)< 0.0012.69 (1.83-3.94)0.0121.70 (1.12-2.58)
Differentiation (poorly/undifferentiated vs medium/high)0.1310.59 (0.30-1.17)
Site (gastric and esophageal binding vs stomach)0.8231.05 (0.70-1.57)
Peritoneal metastasis (no vs yes)< 0.0012.02 (1.41-2.91)0.0051.73 (1.17-2.54)
Liver metastasis (no vs yes)0.0641.50 (0.98-2.29)
EBV status (infect vs no-infect)0.1960.60 (0.28-1.30)
Neutrophil0.1781.05 (0.98-1.12)
Lymphocyte0.3680.87 (0.65-1.17)
Adverse events

Adverse events (AEs) were assessed according to the Common Terminology Criteria for Adverse Events (CTCAE 5.0) by the National Cancer Institute, with grading from 1 to 4. Common adverse events included leukopenia, anemia, neutropenia, nausea, and fever (Table 5). All patients received standard supportive care during treatment, including infection control, antiemetics, and blood transfusions. Although no patient permanently discontinued treatment due to adverse events, a proportion of patients experienced grade ≥ 3 adverse events, with some patients requiring temporary treatment delays or dose adjustments according to protocol specified management guidelines to ensure safety and tolerability. The incidence of adverse events at each grade did not differ significantly between the two groups (P > 0.05).

Table 5 Comparison of adverse events stratified by C-reactive protein to albumin ratio (< 0.21 vs ≥ 0.21) in advanced gastric cancer, n (%).
Variables
Total (n = 200)
High (n = 80)
Low (n = 120)
χ2 value
P value
Leukopenia79 (39.50)34 (42.50)45 (37.50)0.500.479
Anemia83 (41.50)32 (40.00)51 (42.50)0.120.725
Neutropenia80 (40.00)32 (40.00)48 (40.00)0.001.000
Thrombocytopenia72 (36.00)33 (41.25)39 (32.50)1.600.207
Nausea61 (30.50)26 (32.50)35 (29.17)0.250.616
Pyrexia51 (25.50)23 (28.75)28 (23.33)0.740.389
Elevated ALT37 (18.50)18 (22.50)19 (15.83)1.410.234
Elevated AST41 (20.50)17 (21.25)24 (20.00)0.050.830
Hypothyroidism37 (18.50)15 (18.75)22 (18.33)0.010.941
Diarrhea39 (19.50)19 (23.75)20 (16.67)1.530.215
Hypertension26 (13.00)13 (16.25)13 (10.83)1.250.264
Pneumonitis22 (11.00)10 (12.50)12 (10.00)0.310.580
Proteinuria23 (11.50)10 (12.50)13 (10.83)0.130.717
Hyperbilirubinemi18 (9.00)7 (8.75)11 (9.17)0.010.920
Fatigue26 (13.00)12 (15.00)14 (11.67)0.470.492
≥ 3 grades: Thrombocytopenia35 (17.50)16 (20.00)19 (15.83)0.580.447
≥ 3 grades: Neutropenia25 (12.50)12 (15.00)13 (10.83)0.760.383
≥ 3 grades: Leukopenia20 (10.00)7 (8.75)13 (10.83)0.230.630
≥ 3 grades: Anemia15 (7.50)5 (6.25)10 (8.33)0.300.584
≥ 3 grades: Diarrhea6 (3.00)3 (3.75)3 (2.50)0.010.933
≥ 3 grades: Nausea9 (4.50)4 (5.00)5 (4.17)0.001.000
≥ 3 grades: Elevated ALT7 (3.50)3 (3.75)4 (3.33)0.001.000
≥ 3 grades: Elevated AST6 (3.00)2 (2.50)4 (3.33)0.001.000
≥ 3 grades: Hypothyroidism5 (2.50)2 (2.50)3 (2.50)0.001.000
≥ 3 grades: Hypertension4 (2.00)2 (2.50)2 (1.67)0.001.000
≥ 3 grades: Pneumonitis3 (1.50)2 (2.50)1 (0.83)0.130.722
≥ 3 grades: Hyperbilirubinemia4 (2.00)2 (2.50)2 (1.67)0.001.000
Nomogram model construction

Through multivariate Cox regression analysis, we identified variables significantly associated with patient survival (OS or PFS) (P < 0.05) and calculated the regression coefficients (β values) for each variable. Using the “rms” package in R software (version 4.5.1), the regression coefficients were converted into corresponding points, with each variable’s score being proportional to its risk effect. The higher the risk effect, the higher the score. By summing the scores of all variables, we obtained the total score for each patient. Specifically, independent prognostic factors for PFS and OS included CAR index, MMR status, PD-L1 expression, tumor stage, and peritoneal metastasis. These factors were used to construct prognostic models for predicting PFS at 9, 12, and 15 months (Figure 2). Similarly, based on the independent prognostic factors for OS, we constructed prognostic models to predict survival at 12, 15, and 18 months (Figure 3).

Figure 2
Figure 2 Graph depicting the prognostic model for predicting 9-, 12-, and 15-month progression-free survival. PD-L1: Programmed death-ligand 1; CAR: C-reactive protein/albumin ratio; CPS: Combined positive score; MMR: Mismatch repair; pMMR: Proficient mismatch repair; dMMR: Deficient mismatch repair.
Figure 3
Figure 3 Graph depicting the prognostic model for predicting 12-, 15-, and 18-month overall survival. PD-L1: Programmed death-ligand 1; CAR: C-reactive protein/albumin ratio; CPS: Combined positive score; MMR: Mismatch repair; pMMR: Proficient mismatch repair; dMMR: Deficient mismatch repair.
Model validation

To evaluate the predictive performance of the nomogram model, all patients were randomly divided into a training set (140 cases) and a validation set (60 cases) in a 7:3 ratio. The baseline characteristics of the two groups are presented in Table 6. The nomogram model was initially constructed using the training set, followed by internal validation to assess its efficacy. Internal validation was performed using the Bootstrap resampling method (1000 iterations), and the model’s performance was evaluated using the C-index, ROC curve, and calibration curve.

Table 6 Comparison of features between the training and validation sets, mean ± SD/n (%).
Variables
Total (n = 200)
Test (n = 60)
Train (n = 140)
Statistic
P value
Neutrophil3.91 ± 2.324.21 ± 3.063.78 ± 1.92t = 1.000.320
Lymphocyte1.38 ± 0.591.40 ± 0.691.37 ± 0.54t = 0.350.724
PD-L1 expressionχ2 = 0.120.732
CPS < 587 (43.50)25 (41.67)62 (44.29)
CPS ≥ 5113 (56.50)35 (58.33)78 (55.71)
MMR statusχ2 = 1.200.274
pMMR168 (84.00)53 (88.33)115 (82.14)
dMMR32 (16.00)7 (11.67)25 (17.86)
CARχ2 = 0.001.000
High80 (40.00)24 (40.00)56 (40.00)
Low120 (60.00)36 (60.00)84 (60.00)
Sexχ2 = 0.630.426
Female75 (37.50)20 (33.33)55 (39.29)
Male125 (62.50)40 (66.67)85 (60.71)
Ageχ2 = 0.580.448
< 60 years78 (39.00)21 (35.00)57 (40.71)
≥ 60 years122 (61.00)39 (65.00)83 (59.29)
BMIχ2 = 1.730.189
< 25166 (83.00)53 (88.33)113 (80.71)
≥ 2534 (17.00)7 (11.67)27 (19.29)
CEAχ2 = 0.010.925
< 379 (39.50)24 (40.00)55 (39.29)
≥ 3121 (60.50)36 (60.00)85 (60.71)
CA199χ2 = 2.760.097
< 37126 (63.00)43 (71.67)83 (59.29)
≥ 3774 (37.00)17 (28.33)57 (40.71)
ECOG PSχ2 = 0.700.403
072 (36.00)19 (31.67)53 (37.86)
1128 (64.00)41 (68.33)87 (62.14)
Stagingχ2 = 0.160.687
III91 (45.50)26 (43.33)65 (46.43)
IV109 (54.50)34 (56.67)75 (53.57)
Differentiationχ2 = 0.260.607
Poorly/undifferentiated180 (90.00)55 (91.67)125 (89.29)
Medium/high20 (10.00)5 (8.33)15 (10.71)
Siteχ2 = 0.040.838
Gastric and esophageal binding58 (29.00)18 (30.00)40 (28.57)
Stomach142 (71.00)42 (70.00)100 (71.43)
Peritoneal metastasisχ2 = 0.170.679
No157 (78.50)46 (76.67)111 (79.29)
Yes43 (21.50)14 (23.33)29 (20.71)
Liver metastasisχ2 = 0.200.655
No156 (78.00)48 (80.00)108 (77.14)
Yes44 (22.00)12 (20.00)32 (22.86)
EBV statusχ2 = 0.060.802
No-infect187 (93.50)57 (95.00)130 (92.86)
infect13 (6.50)3 (5.00)10 (7.14)

For PFS prediction, the C-index was 0.80 (95%CI: 0.74-0.87) in the training set and 0.82 (95%CI: 0.79-0.86) in the validation set, demonstrating good discriminative ability. ROC curve analysis further confirmed the model’s predictive accuracy, with AUC values in the training set of 0.85 (95%CI: 0.79-0.92) at 9 months, 0.87 (95%CI: 0.81-0.94) at 12 months, and 0.86 (95%CI: 0.78-0.94) at 15 months. For the validation set, the AUC values were 0.77 (95%CI: 0.64-0.90) at 9 months, 0.76 (95%CI: 0.61-0.91) at 12 months, and 0.77 (95%CI: 0.58-0.97) at 15 months (Figure 4A and B).

Figure 4
Figure 4 Graph illustrating the receiver operating characteristic curves for a prognostic model. A: Receiver operating characteristic (ROC) curves for the training set 9-, 12-, and 15-month progression-free survival (PFS); B: ROC curves for the validation set 9-, 12-, and 15-month PFS; C: ROC curves for the training set 12-, 15-, and 18-month overall survival (OS); D: ROC curves for the validation set 12-, 15-, and 18-month OS. AUC: Area under the curve.

For OS prediction, the C-index was 0.78 (95%CI: 0.74-0.83) in the training set and 0.73 (95%CI: 0.68-0.79) in the validation set. ROC curve analysis showed AUC values of 0.80 (95%CI: 0.71-0.89) at 12 months, 0.82 (95%CI: 0.74-0.90) at 15 months, and 0.83 (95%CI: 0.75-0.92) at 18 months in the training set. In the validation set, the AUC values were 0.75 (95%CI: 0.60-0.90) at 12 months, 0.80 (95%CI: 0.67-0.93) at 15 months, and 0.78 (95%CI: 0.65-0.92) at 18 months (Figure 4C and D).

The calibration curve demonstrated excellent agreement between the predicted survival probabilities and the actual observed outcomes, further validating the high predictive accuracy of the nomogram model in clinical application (Figure 5). These results suggest that this model possesses strong prognostic predictive ability and can be effectively applied for prognostic assessment in clinical patients.

Figure 5
Figure 5 Graph illustrating the calibration plots for a prognostic model. A: Calibration plots for the training set 9-, 12-, and 15-month progression-free survival (PFS); B: Calibration plots for the validation set 9-, 12-, and 15-month PFS; C: Calibration plots for the training set 12-, 15-, and 18-month overall survival (OS); D: Calibration plots for the validation set 12-, 15-, and 18-month OS.
DISCUSSION

This study evaluated the prognostic value of the baseline CAR index in predicting the long-term outcomes of advanced gastric cancer patients receiving combination immunotherapy and chemotherapy. The results showed that patients with a low CAR index had better PFS, OS, DCR, and ORR compared to those with a high CAR index. Further Cox regression analysis confirmed that the CAR index was significantly associated with both PFS and OS, and that it interacted with other independent prognostic factors, such as PD-L1 expression and MMR protein expression, to build a prognostic model with strong clinical predictive performance. These findings suggest that the baseline CAR index can serve as an independent predictor for long-term prognosis in advanced gastric cancer patients undergoing combination immunotherapy and chemotherapy.

The immune microenvironment in advanced gastric cancer is typically characterized by immune evasion and systemic inflammation[8,28]. Low ALB levels often indicate malnutrition and impaired immune function, while elevated CRP levels reflect widespread inflammatory responses[14]. A high CAR index typically reflects an enhanced immunosuppressive state, which may be associated with poor responses to immunotherapy[20,21]. Previous studies have demonstrated that elevated CAR is closely related to immune evasion mechanisms, such as the upregulation of PD-L1 on tumor cells, which inhibits T-cell-mediated attack and reduces the efficacy of immunotherapy. In contrast, a low CAR index may reflect a more active immune response, potentially improving treatment outcomes[19].

Additionally, a high CAR index is often accompanied by systemic inflammation. By activating pro-inflammatory signaling pathways, such as nuclear factor kappa-B and Janus kinase/signal transducer and activator of transcription, it promotes the release of immunosuppressive factors like interleukin-6 and tumor necrosis factor-alpha, which not only support tumor cell growth but also inhibit immune cell function, further diminishing the effectiveness of immunotherapy[28,29]. Elevated CAR levels, which correlate with decreased ALB levels, suggest an exacerbation of immune tolerance, posing an additional challenge to immunotherapy. Therefore, the CAR index significantly impacts the immune microenvironment, activates pro-inflammatory pathways, and facilitates immune evasion mechanisms, making it an important biomarker for predicting the response to combination immunotherapy and chemotherapy[30].

Compared to previous studies, the CAR index offers significant advantages as a predictive marker for immunotherapy. Traditional inflammatory biomarkers, such as CRP or ALB, generally reflect only a single aspect of the immune microenvironment and fail to comprehensively reveal the complex interactions between immune evasion, immune tolerance, and systemic inflammation[11,13]. The CAR index, by integrating CRP and ALB levels, provides a more precise reflection of the immune status and systemic inflammatory response in cancer patients. It reveals the dual effects of immune suppression and inflammation, offering more sensitive and specific prognostic information[15,17]. Furthermore, the CAR index can help identify the efficacy differences between various immunotherapeutic approaches (e.g., immune checkpoint inhibitors and cell therapies), thereby providing critical insights for selecting individualized treatment strategies in clinical practice[20,22]. Compared to individual biomarkers, the CAR index offers a more comprehensive assessment of the immune microenvironment, making it an effective biomarker for predicting immunotherapy response and possessing higher clinical applicability.

In clinical practice, using this predictive model, clinicians can rapidly assess the immune microenvironment of patients based on their baseline CAR index and predict their likely response to immunotherapy. Specifically, a low CAR index is generally associated with a better immune response and higher treatment efficacy, while a high CAR index indicates poorer treatment responses and prognosis. Therefore, for patients with a high CAR index, clinicians may consider more aggressive supportive therapies (e.g., infection management, ALB supplementation) and combination treatment strategies (e.g., chemotherapy or targeted therapy) to optimize treatment outcomes and improve patient quality of life.

However, it is important to clarify that the CAR index, as examined in this study, remains primarily a prognostic indicator rather than a strictly predictive biomarker of immunotherapy benefit. Unlike biomarkers that directly reflect tumor intrinsic sensitivity to immune checkpoint inhibition, such as PD-L1 or MMR status, CAR captures systemic host factors including inflammation and nutritional status that are associated with overall disease trajectory and survival rather than specific therapeutic mechanisms[31]. Consequently, while CAR stratifies patients by overall risk and may flag those at risk for poorer outcomes, its use as a definitive determinant of treatment selection should be approached cautiously until prospective validation is achieved[31].

Furthermore, the dual nature of CAR as a marker of both inflammation and nutrition suggests potential utility beyond prognosis[32]. Elevated CAR reflects a state of systemic inflammation and relative nutritional depletion, both of which are potentially modifiable clinical features. This raises the possibility that CAR could be leveraged in treatment stratification and preintervention optimization, for example by identifying patients who may benefit from nutritional support, anti-inflammatory interventions, or tailored supportive care prior to or alongside immunotherapy[32]. Such strategies might not only improve general patient status but also potentially enhance immunotherapy effectiveness by mitigating inflammation related immune suppression. Future studies should therefore explore whether interventions targeting inflammation and nutritional status before or during treatment can modulate CAR levels and translate into improved clinical outcomes[31].

Although this study provides clinically significant results, several limitations remain. First, as a single-center retrospective analysis, there may be selection bias, and the findings need further validation in multi-center prospective cohorts. Second, the CAR index can be influenced by factors such as infections, inflammation, and nutritional interventions, which may affect its stability in certain contexts[11,14]. Additionally, this study did not perform external validation, and the CAR cutoff was determined in a data driven manner, which may introduce bias and overfitting, thereby limiting the generalizability of the predictive model. Future research should consider integrating immune microenvironment-related molecular biomarkers to explore more precise and comprehensive predictive models.

In conclusion, this study demonstrates that the baseline CAR index is an independent prognostic indicator for advanced gastric cancer patients undergoing combination immunotherapy and chemotherapy. It can aid in identifying patients with poorer prognoses. The nomogram model based on CAR demonstrates strong predictive performance and provides valuable reference for the development of individualized treatment strategies in clinical settings.

CONCLUSION

This study confirms that the CAR index is an independent prognostic factor for advanced gastric cancer patients undergoing combination immunotherapy and chemotherapy, effectively identifying subgroups with different survival risks. The nomogram model, integrating CAR with clinical characteristics, demonstrates good discriminative ability and calibration performance, providing a simple tool for individualized prognostic assessment. As a composite indicator reflecting inflammation and nutritional status, CAR shows potential in predicting the outcomes of combination immunotherapy and chemotherapy. However, its clinical applicability still requires further validation through prospective studies.

References
1.  Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021;71:209-249.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 76817]  [Cited by in RCA: 69825]  [Article Influence: 13965.0]  [Reference Citation Analysis (51)]
2.  Arnold M, Abnet CC, Neale RE, Vignat J, Giovannucci EL, McGlynn KA, Bray F. Global Burden of 5 Major Types of Gastrointestinal Cancer. Gastroenterology. 2020;159:335-349.e15.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 1700]  [Cited by in RCA: 1509]  [Article Influence: 251.5]  [Reference Citation Analysis (7)]
3.  Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68:394-424.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 53448]  [Cited by in RCA: 56132]  [Article Influence: 7016.5]  [Reference Citation Analysis (27)]
4.  Van Cutsem E, Sagaert X, Topal B, Haustermans K, Prenen H. Gastric cancer. Lancet. 2016;388:2654-2664.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 1498]  [Cited by in RCA: 1484]  [Article Influence: 148.4]  [Reference Citation Analysis (3)]
5.  Fuchs CS, Doi T, Jang RW, Muro K, Satoh T, Machado M, Sun W, Jalal SI, Shah MA, Metges JP, Garrido M, Golan T, Mandala M, Wainberg ZA, Catenacci DV, Ohtsu A, Shitara K, Geva R, Bleeker J, Ko AH, Ku G, Philip P, Enzinger PC, Bang YJ, Levitan D, Wang J, Rosales M, Dalal RP, Yoon HH. Safety and Efficacy of Pembrolizumab Monotherapy in Patients With Previously Treated Advanced Gastric and Gastroesophageal Junction Cancer: Phase 2 Clinical KEYNOTE-059 Trial. JAMA Oncol. 2018;4:e180013.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 1522]  [Cited by in RCA: 1477]  [Article Influence: 184.6]  [Reference Citation Analysis (6)]
6.  Janjigian YY, Ajani JA, Moehler M, Shen L, Garrido M, Gallardo C, Wyrwicz L, Yamaguchi K, Cleary JM, Elimova E, Karamouzis M, Bruges R, Skoczylas T, Bragagnoli A, Liu T, Tehfe M, Zander T, Kowalyszyn R, Pazo-Cid R, Schenker M, Feeny K, Wang R, Lei M, Chen C, Nathani R, Shitara K. First-Line Nivolumab Plus Chemotherapy for Advanced Gastric, Gastroesophageal Junction, and Esophageal Adenocarcinoma: 3-Year Follow-Up of the Phase III CheckMate 649 Trial. J Clin Oncol. 2024;42:2012-2020.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 204]  [Cited by in RCA: 178]  [Article Influence: 89.0]  [Reference Citation Analysis (4)]
7.  Yamaguchi K, Minashi K, Sakai D, Nishina T, Omuro Y, Tsuda M, Iwagami S, Kawakami H, Esaki T, Sugimoto N, Oshima T, Kato K, Amagai K, Hosaka H, Komine K, Yasui H, Negoro Y, Ishido K, Tsushima T, Han S, Shiratori S, Takami T, Shitara K. Phase IIb study of pembrolizumab combined with S-1 + oxaliplatin or S-1 + cisplatin as first-line chemotherapy for gastric cancer. Cancer Sci. 2022;113:2814-2827.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 22]  [Cited by in RCA: 22]  [Article Influence: 5.5]  [Reference Citation Analysis (0)]
8.  Cristescu R, Mogg R, Ayers M, Albright A, Murphy E, Yearley J, Sher X, Liu XQ, Lu H, Nebozhyn M, Zhang C, Lunceford JK, Joe A, Cheng J, Webber AL, Ibrahim N, Plimack ER, Ott PA, Seiwert TY, Ribas A, McClanahan TK, Tomassini JE, Loboda A, Kaufman D. Pan-tumor genomic biomarkers for PD-1 checkpoint blockade-based immunotherapy. Science. 2018;362:eaar3593.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 1035]  [Cited by in RCA: 1793]  [Article Influence: 224.1]  [Reference Citation Analysis (2)]
9.  Topalian SL, Taube JM, Anders RA, Pardoll DM. Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy. Nat Rev Cancer. 2016;16:275-287.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 2344]  [Cited by in RCA: 2195]  [Article Influence: 219.5]  [Reference Citation Analysis (3)]
10.  Yao ZY, Ma X, Cui YZ, Liu J, Han ZX, Song J. Impact of triglyceride-glucose index on the long-term prognosis of advanced gastric cancer patients receiving immunotherapy combined with chemotherapy. World J Gastroenterol. 2025;31:102249.  [PubMed]  [DOI]  [Full Text]
11.  Kinoshita A, Onoda H, Imai N, Iwaku A, Oishi M, Tanaka K, Fushiya N, Koike K, Nishino H, Matsushima M. The C-reactive protein/albumin ratio, a novel inflammation-based prognostic score, predicts outcomes in patients with hepatocellular carcinoma. Ann Surg Oncol. 2015;22:803-810.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 354]  [Cited by in RCA: 345]  [Article Influence: 31.4]  [Reference Citation Analysis (0)]
12.  McMillan DC. The systemic inflammation-based Glasgow Prognostic Score: a decade of experience in patients with cancer. Cancer Treat Rev. 2013;39:534-540.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 1148]  [Cited by in RCA: 1075]  [Article Influence: 82.7]  [Reference Citation Analysis (4)]
13.  Nakamura T, Gaston CL, Reddy K, Iwata S, Nishio J. Inflammatory Biomarkers in Cancer. Mediators Inflamm. 2016;2016:7282797.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 1]  [Cited by in RCA: 3]  [Article Influence: 0.3]  [Reference Citation Analysis (0)]
14.  Roxburgh CS, McMillan DC. Role of systemic inflammatory response in predicting survival in patients with primary operable cancer. Future Oncol. 2010;6:149-163.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 761]  [Cited by in RCA: 741]  [Article Influence: 46.3]  [Reference Citation Analysis (3)]
15.  Guthrie GJ, Roxburgh CS, Farhan-Alanie OM, Horgan PG, McMillan DC. Comparison of the prognostic value of longitudinal measurements of systemic inflammation in patients undergoing curative resection of colorectal cancer. Br J Cancer. 2013;109:24-28.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 119]  [Cited by in RCA: 114]  [Article Influence: 8.8]  [Reference Citation Analysis (4)]
16.  Pan Y, Lou Y, Wang L. Prognostic value of C-reactive protein to albumin ratio in metastatic colorectal cancer: A systematic review and meta-analysis. Medicine (Baltimore). 2021;100:e27783.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 5]  [Cited by in RCA: 15]  [Article Influence: 3.0]  [Reference Citation Analysis (0)]
17.  Utsumi M, Inagaki M, Kitada K, Tokunaga N, Kondo M, Yunoki K, Sakurai Y, Hamano R, Miyasou H, Tsunemitsu Y, Otsuka S. Preoperative C-reactive protein-to-albumin ratio as a prognostic factor in biliary tract cancer: A systematic review and meta-analysis. Medicine (Baltimore). 2023;102:e33656.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 7]  [Cited by in RCA: 10]  [Article Influence: 3.3]  [Reference Citation Analysis (0)]
18.  Yamauchi Y, Safi S, Muley T, Warth A, Herth FJF, Dienemann H, Hoffmann H, Eichhorn ME. C-reactive protein-albumin ratio is an independent prognostic predictor of tumor recurrence in stage IIIA-N2 lung adenocarcinoma patients. Lung Cancer. 2017;114:62-67.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 12]  [Cited by in RCA: 31]  [Article Influence: 3.4]  [Reference Citation Analysis (0)]
19.  Anandavadivelan P, Lagergren P. Cachexia in patients with oesophageal cancer. Nat Rev Clin Oncol. 2016;13:185-198.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 115]  [Cited by in RCA: 202]  [Article Influence: 18.4]  [Reference Citation Analysis (0)]
20.  Fang Q, Yu J, Li W, Luo J, Deng Q, Chen B, He Y, Zhang J, Zhou C. Prognostic value of inflammatory and nutritional indexes among advanced NSCLC patients receiving PD-1 inhibitor therapy. Clin Exp Pharmacol Physiol. 2023;50:178-190.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 1]  [Cited by in RCA: 33]  [Article Influence: 11.0]  [Reference Citation Analysis (0)]
21.  Wu J, Tan W, Chen L, Huang Z, Mai S. Clinicopathologic and prognostic significance of C-reactive protein/albumin ratio in patients with solid tumors: an updated systemic review and meta-analysis. Oncotarget. 2018;9:13934-13947.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 21]  [Cited by in RCA: 31]  [Article Influence: 3.9]  [Reference Citation Analysis (0)]
22.  Zhou W, Zhang GL. C-reactive protein to albumin ratio predicts the outcome in renal cell carcinoma: A meta-analysis. PLoS One. 2019;14:e0224266.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 10]  [Cited by in RCA: 20]  [Article Influence: 2.9]  [Reference Citation Analysis (0)]
23.  Alkurt EG, Durak D, Turhan VB, Sahiner IT. Effect of C-Reactive Protein-to-Albumin Ratio on Prognosis in Gastric Cancer Patients. Cureus. 2022;14:e23972.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in RCA: 8]  [Reference Citation Analysis (0)]
24.  Sun Y, Yan F, Yu X, Sun Z, Zhou G. Retrospective study on the prognostic prediction of inflammatory markers and the C-reactive protein/albumin ratio in first-line immunotherapy for advanced HER2 negative gastric cancer patients. Transl Cancer Res. 2025;14:2043-2053.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in RCA: 5]  [Reference Citation Analysis (0)]
25.  Iasonos A, Schrag D, Raj GV, Panageas KS. How to build and interpret a nomogram for cancer prognosis. J Clin Oncol. 2008;26:1364-1370.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 2571]  [Cited by in RCA: 2479]  [Article Influence: 137.7]  [Reference Citation Analysis (5)]
26.  Huynh J, Patel K, Gong J, Cho M, Malla M, Parikh A, Klempner S. Immunotherapy in Gastroesophageal Cancers: Current Evidence and Ongoing Trials. Curr Treat Options Oncol. 2021;22:100.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in Crossref: 20]  [Cited by in RCA: 17]  [Article Influence: 3.4]  [Reference Citation Analysis (0)]
27.  Yao Z, Li G, Pan D, Pei Z, Fang Y, Liu H, Han Z. Roles and functions of tumor-infiltrating lymphocytes and tertiary lymphoid structures in gastric cancer progression. Front Immunol. 2025;16:1595070.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 4]  [Cited by in RCA: 10]  [Article Influence: 10.0]  [Reference Citation Analysis (1)]
28.  Le DT, Durham JN, Smith KN, Wang H, Bartlett BR, Aulakh LK, Lu S, Kemberling H, Wilt C, Luber BS, Wong F, Azad NS, Rucki AA, Laheru D, Donehower R, Zaheer A, Fisher GA, Crocenzi TS, Lee JJ, Greten TF, Duffy AG, Ciombor KK, Eyring AD, Lam BH, Joe A, Kang SP, Holdhoff M, Danilova L, Cope L, Meyer C, Zhou S, Goldberg RM, Armstrong DK, Bever KM, Fader AN, Taube J, Housseau F, Spetzler D, Xiao N, Pardoll DM, Papadopoulos N, Kinzler KW, Eshleman JR, Vogelstein B, Anders RA, Diaz LA Jr. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. 2017;357:409-413.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 5580]  [Cited by in RCA: 5231]  [Article Influence: 581.2]  [Reference Citation Analysis (11)]
29.  Kim ES, Kim SY, Moon A. C-Reactive Protein Signaling Pathways in Tumor Progression. Biomol Ther (Seoul). 2023;31:473-483.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in RCA: 41]  [Reference Citation Analysis (22)]
30.  Xu HJ, Ma Y, Deng F, Ju WB, Sun XY, Wang H. The prognostic value of C-reactive protein/albumin ratio in human malignancies: an updated meta-analysis. Onco Targets Ther. 2017;10:3059-3070.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 39]  [Cited by in RCA: 73]  [Article Influence: 8.1]  [Reference Citation Analysis (0)]
31.  Dai M, Wu W. Prognostic role of C-reactive protein to albumin ratio in cancer patients treated with immune checkpoint inhibitors: a meta-analysis. Front Oncol. 2023;13:1148786.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Cited by in RCA: 14]  [Reference Citation Analysis (0)]
32.  Li N, Tian GW, Wang Y, Zhang H, Wang ZH, Li G. Prognostic Role of the Pretreatment C-Reactive Protein/Albumin Ratio in Solid Cancers: A Meta-Analysis. Sci Rep. 2017;7:41298.  [RCA]  [PubMed]  [DOI]  [Full Text]  [Full Text (PDF)]  [Cited by in Crossref: 35]  [Cited by in RCA: 36]  [Article Influence: 4.0]  [Reference Citation Analysis (0)]
Footnotes

Peer review: Externally peer reviewed.

Peer-review model: Single blind

Specialty type: Oncology

Country of origin: China

Peer-review report’s classification

Scientific quality: Grade B

Novelty: Grade B

Creativity or innovation: Grade C

Scientific significance: Grade B

P-Reviewer: Zhang HL, PhD, Researcher, Malaysia S-Editor: Fan M L-Editor: A P-Editor: Zhao YQ

Write to the Help Desk