Front. Immunol. Frontiers in Immunology Front. Immunol. 1664-3224 Frontiers Media S.A. 10.3389/fimmu.2025.1662605 Original Research Development and validation of immunotherapy nomogram for predicting the efficacy and prognosis of recurrent and metastatic cervical cancer LiuZhongan 1 2 3 4 † Project administration Visualization Software Formal analysis Writing – original draft Methodology Conceptualization Funding acquisition Data curation Resources Investigation Validation Supervision Writing – review & editing ZhangSijia 1 2 3 5 6 † Supervision Data curation Writing – original draft Investigation Software Conceptualization Writing – review & editing Methodology Funding acquisition Visualization Resources Validation Project administration Formal analysis WangYan 7 Investigation Data curation Supervision Methodology Writing – review & editing Conceptualization Software Validation Formal analysis Resources Visualization Funding acquisition Project administration Writing – original draft ZongYan 8 ZhouTian 8 Conceptualization Methodology Supervision Project administration Validation Investigation Writing – review & editing Data curation Funding acquisition Resources Writing – original draft Formal analysis Software Visualization WuDaying 1 2 3 Project administration Data curation Formal analysis Writing – original draft Methodology Resources Visualization Investigation Conceptualization Software Supervision Funding acquisition Validation Writing – review & editing WangGanxin 1 2 3 Data curation Methodology Validation Supervision Conceptualization Project administration Software Investigation Resources Funding acquisition Writing – review & editing Writing – original draft Formal analysis Visualization HeLihong 8 Supervision Methodology Writing – review & editing Validation Data curation Conceptualization Investigation Writing – original draft Formal analysis Visualization Resources Software Funding acquisition Project administration HuangKai 9 Visualization Methodology Data curation Validation Project administration Conceptualization Supervision Investigation Software Formal analysis Writing – original draft Funding acquisition Resources Writing – review & editing XuYunqing 9 Validation Project administration Conceptualization Visualization Methodology Data curation Formal analysis Investigation Supervision Writing – review & editing Funding acquisition Software Resources Writing – original draft TangQuan 10 Methodology Writing – review & editing Data curation Investigation Writing – original draft Software Funding acquisition Supervision Visualization Resources Conceptualization Validation Formal analysis Project administration ChenMulan 8 XiaoGuangqin 1 2 3 11 12 * Resources Writing – review & editing Visualization Funding acquisition Formal analysis Validation Project administration Investigation Writing – original draft Data curation Supervision Methodology Software Conceptualization LiGuiling 1 2 3 * Resources Validation Data curation Visualization Project administration Writing – review & editing Investigation Formal analysis Software Funding acquisition Methodology Supervision Writing – original draft Conceptualization Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China Institute of Radiation Oncology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China Hubei Key Laboratory of Precision Radiation Oncology, Wuhan, China Chongqing Hospital, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Chongqing, China Department of Biophysics, Center for Integrative Physiology and Molecular Medicine (CIPMM), School of Medicine, Saarland University, Homburg, Germany Department of Biomedical Sciences, Institute for Health Research and Education, Osnabrück University, Osnabrück, Germany Department of Oncology, Liyuan Hospital, Tongji Medical School, Huazhong University of Science and Technology, Wuhan, China Department of Infectious Diseases, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China Department of Oncology, People’s Hospital of Huangpi District, Jianghan University, Wuhan, China Department of Oncology, Hubei Aerospace Hospital, Xiaogan, China Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA, United States Clinical and Translational Epidemiology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States Correspondence: Guangqin Xiao, xiao_guang_qin@163.com; Guiling Li, lgl6714@163.com

These authors have contributed equally to this work

 08 12 2025 2025 16 1662605 09 07 2025 24 11 2025 22 10 2025 Copyright © 2025 Liu, Zhang, Wang, Zong, Zhou, Wu, Wang, He, Huang, Xu, Tang, Chen, Xiao and Li. 2025 Liu, Zhang, Wang, Zong, Zhou, Wu, Wang, He, Huang, Xu, Tang, Chen, Xiao and Li https://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Background

Cervical cancer, a major cause of cancer-related mortality in women, remains challenging to treat, particularly in recurrent or metastatic stages. Immunotherapy offers promise, but reliable biomarkers for predicting outcomes are lacking.

 Method

A cohort of 204 patients with recurrent or metastatic cervical cancer who underwent immunotherapy was included in the study. Predictive factors were identified using LASSO regression combined with multivariate Cox proportional hazards analysis. Nomograms for progression-free survival (PFS) and overall survival (OS) were constructed by incorporating significant prognostic variables. Internal validation was performed via bootstrap resampling, and clinical applicability was assessed through decision curve analysis (DCA). Risk stratification was evaluated using Kaplan-Meier survival curves, with log-rank tests for comparison. In addition, to enhance the clinical applicability of the model, we downloaded external multi-center validation datasets from public databases, including TCGA (The Cancer Genome Atlas) and GEO (Gene Expression Omnibus). Data from 306 cervical cancer patients were obtained and used as an independent validation cohort. These external datasets allowed for the verification of the developed model, ensuring its robustness and generalizability across different clinical settings.

 Results

The analysis identified several significant prognostic factors for PFS, including histological type, maximum lesion diameter, CA125 levels, albumin concentration, lactate dehydrogenase (LDH) activity, and neutrophil-to-lymphocyte ratio (NLR). For OS, independent factors included BMI, liver metastasis, CEA levels, hemoglobin concentration, albumin levels, and LDH. The nomogram models demonstrated strong predictive accuracy, with concordance indices (C-index) of 0.706 for PFS and 0.769 for OS. Calibration curves indicated that the predicted and actual outcomes were in excellent agreement. The area under the curve (AUC) values for PFS at 1- and 2-year follow-up were 0.804 and 0.822, respectively, while for OS, the AUC values were 0.880 and 0.781. The risk stratification based on the nomogram scores revealed significant survival differences between high- and low-risk patients, with the high-risk group exhibiting poorer survival outcomes. External validation using data from the TCGA and GEO cohorts confirmed the robustness and generalizability of the nomogram models, further supporting their clinical relevance.

 Conclusion

These nomograms provide a reliable tool for predicting outcomes in cervical cancer immunotherapy, helping to personalize treatment and improve clinical management, especially for metastatic disease.

 cervical cancer immunotherapy nomogram biomarker PD-L1 National Natural Science Foundation of China 10.13039/501100001809 No. 81600482 Natural Science Foundation of Hubei Province 10.13039/501100003819 No. 2020CFB600, No. 2019CFB501 China Postdoctoral Science Foundation 10.13039/501100002858 No. 2018M632875, No. 2019T120671 The author(s) declared that financial support was received for this work and/or its publication. This study was supported by the National Natural Science Foundation of China (Nos. 81402197, 81600482, 82570789), the Natural Science Foundation of Hubei Province (Nos. 2019CFB501, 2020CFB600, 2024AFB663), the Beijing Kanghua Traditional Chinese and Western Medicine Development Foundation (No. 2021HX005), and the China Postdoctoral Science Foundation (Nos. 2018M632875, 2019T120671). section-at-acceptance Cancer Immunity and Immunotherapy Introduction

Cervical cancer is the most prevalent gynecological malignancy worldwide and remains one of the leading causes of cancer-related mortality in women (1). While the advent of surgery, concurrent chemoradiotherapy, and intrauterine irradiation has significantly improved the overall response rates and survival outcomes for cervical cancer patients, these treatments have limited efficacy in cases of recurrent or metastatic disease (2). The median survival for patients with advanced cervical cancer is approximately 16.8 months, with an overall five-year survival rate of less than 20% (3). In recent years, immune checkpoint inhibitors have garnered increasing attention and are being integrated into the treatment regimens for various solid tumors, offering a promising and well-tolerated therapeutic option.

Cervical cancer is a virus-induced neoplasm, driven primarily by the persistent infection of high-risk human papillomavirus (HPV), leading to the overexpression of oncoproteins E6 and E7 in tumor epithelial cells. This upregulation facilitates immune evasion, notably through the overexpression of programmed death ligand 1 (PD-L1), which contributes to immune escape (4). Given these immunogenic characteristics, cervical cancer has been recognized as an immunodominant malignancy, with a higher potential for response to immunotherapy. However, despite these promising features, clinical response rates to immunotherapy in cervical cancer remain modest, ranging from 10% to 25%, with a significant portion of patients experiencing treatment failure due to tumor recurrence and metastasis, ultimately resulting in poor long-term survival outcomes (5). Thus, there is a critical need to identify reliable and effective biomarkers to predict both the therapeutic response and prognosis of immunotherapy in cervical cancer patients, as well as to identify subgroups that may benefit most from this treatment.

Several studies have highlighted the influence of clinicopathological factors on the efficacy and prognosis of immunotherapy in cancer patients, including histologic subtype, tumor size, lymph node metastasis, and biomarkers such as lactate dehydrogenase (LDH) levels (6–8). Various predictive models have been developed to identify patient populations that may derive benefit from immunotherapy. For example, a model incorporating the pre-treatment performance status (PS) score, alkaline phosphatase, and hemoglobin levels demonstrated superior prognostic discrimination for immunotherapy response in patients with advanced urothelial carcinoma (9). Another immune-prognostic scoring system, utilizing albumin, LDH, and the neutrophil-to-lymphocyte ratio (NLR), has shown good clinical applicability in selecting appropriate candidates for phase I immunotherapy trials (10). However, the prognostic value of these clinicopathological factors in predicting the response to immunotherapy in advanced cervical cancer remains unclear. Moreover, despite the availability of various predictive models in other cancer types, few studies in the field of cervical cancer immunotherapy have employed nomogram-based visual prediction tools that integrate multiple clinicopathological predictors.

In this study, we aim to explore the prognostic significance of these predictive factors in the context of immunotherapy for advanced cervical cancer. Our goal is to develop a novel predictive nomogram to assist clinicians in assessing the individualized prognosis and therapeutic outcomes of immunotherapy for cervical cancer patients.

 Materials and methods Patients

This retrospective study included 300 patients who received immunotherapy for recurrent and metastatic cervical cancer at the Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology between March 2018 and August 2024. The inclusion criteria were patients diagnosed with recurrent or metastatic cervical cancer who underwent immunotherapy during the study period. The exclusion criteria were as follows (1): patients with cervical cancer and a second primary malignancy, and (2) patients who received fewer than two cycles of treatment or had incomplete follow-up data. After applying these exclusions, a total of 204 patients with complete clinical and pathological data were included in the final analysis. The median follow-up duration for these patients was 39 months, with the shortest follow-up being 4 months and the longest 135 months. This study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Ethical approval was granted by the Institutional Review Board and Ethics Committee of Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology (Approval number: UHCT230829).

 Clinicopathology data collection

In this study, a total of 20 clinicopathological variables were included to minimize potential confounders. These factors comprised demographic and clinical characteristics, as well as laboratory and inflammatory biomarkers. Specific cut-off values were used to define the elevated levels of key markers: CA125 was considered elevated if ≥35 U/mL, CEA ≥5 ng/mL, LDH ≥250 U/L, hemoglobin <12 g/dL (for anemia), albumin <3.5 g/dL (for hypoalbuminemia), ALP >120 U/L, and ALT >40 U/L. The NLR was calculated using blood samples collected immediately prior to the initiation of the first cycle of immunotherapy. These biomarkers were selected for their relevance to tumor progression, immune responses, and their ability to predict outcomes in cervical cancer immunotherapy (Table 1).

Clinical characteristics of the patients.

Variable Cancer progression P Survival P
Yes No Yes No
Age (Years) ≤53 55 53 0.468 74 34 0.392
>53 44 52 71 25
BMI (kg/m²) ≤18.5 10 6 0.507 8 8 0.047
18.5-24.9 68 76 109 35
>24.9 21 23 28 16
Histological type SCC 79 89 0.158 120 48 0.943
ADC 14 15 20 9
ASC 6 1 5 2
Treatment lines 1 78 84 0.959 118 44 0.512
2 17 17 22 12
≥3 4 4 5 3
Hepatic metastases Yes 8 4 0.318 6 6 0.097
No 91 101 139 53
Pulmonary metastasis Yes 27 23 0.373 35 15 0.847
No 72 82 110 44
Osseous metastasis Yes 22 22 0.826 29 15 0.393
No 77 83 116 44
Maximum diameter of lesion (cm) Median (IR) 3.5(1.5) 2.9(1.15) 0.161 2.65(1.07) 3.9(1.98) 0.01
CEA (ng/mL) Median (IR) 3.07(4.66) 2.16(3.05) 0.021 2.24(3.56) 3.59(5.1) 0.074
CA125 (U/mL) Median (IR) 20(26.8) 13.3(16.15) 0.103 13.9(15.95) 21(39.2) 0.014
Hemoglobin (g/dL) Median (IR) 103(20) 108(23.5) 0.036 108(21) 100(35) <0.001
Albumin (g/dL) Median (IR) 38.3(6.4) 40.3(6.55) 0.007 39.9(6) 37.5(7) <0.001
LDH (U/L) Median (IR) 172(58) 164(43) 0.319 165(47) 172(55) 0.763
ALP (U/L) Median (IR) 90(35) 87(33) 0.133 87(35) 92(27) 0.692
ALT (U/L) Median (IR) 14(13) 15(13) 0.292 15(14.5) 13(13) 0.212
AST (U/L) Median (IR) 19(10) 18(8.5) 0.960 19(9.5) 19(9) 0.648
AFR Median (IR) 9.09(3.55) 11.15(4.06) 0.004 11.04(4.22) 8.46(3.17) <0.001
NLR Median (IR) 4.38(4.87) 3.76(4.165) 0.128 3.77(4.375) 4.81(4.54) 0.022
PLR Median (IR) 280.77(240.28) 258.7(193.5) 0.131 256.86(171.92) 321.13(304.08) 0.008
LMR Median (IR) 1.78(1.19) 2.39(1.58) 0.026 2.36(1.57) 1.64(1.12) 0.062

ADC, adenocarcinoma; AFR, albumin fibrinogen ratio; ALP, alkaline phosphatase; ASC, adenosquamous carcinoma; BMI, body mass index; IR, interquartile range; LDH, lactate dehydrogenase; LMR, lymphocyte-to-monocyte ratio; NLR, neutrophil to lymphocyte ratio; PLR, platelet to lymphocyte ratio; and SCC, squamous cell carcinoma. P-values were calculated using the chi-square test or Wilkerson’s test.

 Establishment of external validation cohort

To further enhance the clinical applicability of our model, we downloaded external multi-center validation datasets of cervical cancer from public databases, including TCGA (The Cancer Genome Atlas) and GEO (Gene Expression Omnibus). These datasets, which include data from 306 cervical cancer patients, served as an independent validation cohort to evaluate our previously developed clinical feature model distinguishing cervical squamous cell carcinoma (CSCC) from cervical adenocarcinoma (CAC). The validation cohort consisted of clinical data from cervical cancer patients with various subtypes, including both squamous cell carcinoma and adenocarcinoma. Data preprocessing was performed on the downloaded datasets to ensure they met our inclusion criteria, which required patients to have been diagnosed with cervical cancer and to have undergone immunotherapy. The following steps were taken for processing these external datasets. 1) Data preprocessing: Prior to analysis, all patient data underwent standardization, including missing value imputation, outlier removal, and appropriate transformations of clinical features. The preprocessing procedure adhered to conventional methods to ensure consistency and reliability for comparison across different databases. 2) Extraction of clinical features: Clinical features relevant to immunotherapy outcomes were extracted from the TCGA and GEO databases, including age, gender, race, tumor staging (pT, pN, pM), histological type (squamous cell carcinoma and adenocarcinoma), and treatment history (e.g., whether patients had received first-line treatment). These variables were consistent with those used in our prior study on immunotherapy in cervical cancer, ensuring comparability of results. 3) External validation analysis: We applied LASSO (Least Absolute Shrinkage and Selection Operator) regression and Cox proportional hazards regression models to the external cohort of 306 cervical cancer patients from TCGA and GEO. The same methodology was used as in the original cohort to identify prognostic factors that influenced immunotherapy outcomes. This external validation was crucial for confirming the robustness and generalizability of our predictive model.

 Statistical analysis

Statistical analysis was conducted using SPSS version 25.0 and R software 4.2.2. We used median and interquartile range to represent the continuous variables. Between-group comparisons will be conducted using t-tests or Wilcoxon tests, while descriptive statistics will be employed to summarize clinicopathological and other characteristics. The categorical variables were demonstrated as frequencies and proportions, and group comparisons were performed using the chi-square test. To determine the optimal cutoff value for the continuous variable, we converted it to a dichotomous variable based on the area under the subject’s work curve and used it in further statistical analyses. To mitigate the issue of multicollinearity among variables, we employed lasso regression to screen for risk factors and utilized Cox forward stepwise regression to ascertain independent risk factors that significantly impact prognosis and outcome. We performed 1000 iterations of bootstrap sampling to obtain discrimination and calibration for internal model validation. Additionally, we employed decision curve analysis (DCA) curves to evaluate the clinical applicability and value. Additionally, utilizing the predicted risk scores, patients were categorized into high-risk and low-risk cohorts, followed by the generation of Kaplan-Meier survival curves to depict OS and PFS. The Log-rank test was employed to evaluate the differences between the survival curves.

 Results Characteristics of patients

A total of 204 patients with advanced cervical cancer were enrolled in the study, and comprehensive clinical and laboratory baseline data were collected. The baseline characteristics of the patients are summarized in Table 1. The median age at presentation was 53 years, with a mean age of 52.5 years. Of the patients, 21.5% were classified as overweight or obese, 7.8% were underweight, and more than half (50.7%) were within the normal weight range. Among the various histological subtypes of cervical cancer, squamous cell carcinoma was the most prevalent, accounting for 82.35% of cases, while adenosquamous carcinoma represented the least common subtype, with only 7% of cases. Regarding prior treatment, 20.59% of patients had received two or more lines of therapy before immunotherapy, while the majority (79.4%) had undergone first-line treatment regimens, which included concurrent chemoradiotherapy, prior to starting immunotherapy. In terms of metastatic involvement, lung metastasis was the most common, affecting 24.5% of patients, followed by bone metastasis and liver metastasis. The majority of patients (92.6%) demonstrated a favorable response to immunotherapy, as evidenced by a high disease control rate (DCR) of 92.6%.

 Relevant independent prognostic factors selection

To identify relevant prognostic factors, variables were initially screened using LASSO regression, with the characteristics of these variables presented in Figures 1B, D. The optimal tuning parameter (λ) for the LASSO regression was determined via 10-fold cross-validation, as shown in Figures 1A, C. Following this, the selected variables were subjected to Cox proportional hazards regression analysis. Factors with a p-value < 0.05 were retained for the construction of the predictive model. Our analysis identified several independent prognostic factors for the efficacy of immunotherapy in cervical cancer patients. Specifically, the maximum lesion diameter, CA125 levels, albumin concentration, LDH levels, and NLR were found to significantly influence the treatment outcome (Figure 2A). Higher albumin levels were associated with longer PFS. For OS, independent prognostic factors included BMI, liver metastasis, CEA levels, hemoglobin concentration, albumin, and LDH. In particular, the presence of liver metastasis, and elevated levels of CEA and LDH, were associated with poorer survival outcomes (Figure 2B).

Selection of nomogram parameters for PFS and OS using LASSO regression and cross-Validation. (A) Plot of partial likelihood deviance for PFS, depicting the relationship between the log-transformed tuning parameter (λ) and the model’s performance. The dashed vertical line indicates the optimal λ value selected through cross-validation, which minimizes the deviance and balances model complexity with predictive accuracy. (B) LASSO coefficient profiles for PFS, illustrating how the coefficients of each predictor variable evolve as the regularization parameter (λ) is varied. The progressive shrinkage of coefficients is evident as λ increases, with some predictors being reduced to zero, indicating their exclusion from the final model. (C) Plot of partial likelihood deviance for OS, analogous to (A), showing the optimal λ selection for the OS model. The vertical dashed line represents the λ value that minimizes deviance, ensuring the most accurate model fit. (D) LASSO coefficient profiles for OS, demonstrating the change in coefficients for each predictor variable as λ increases, leading to variable selection and model regularization.

Four-panel graphical display of lasso regression results. Panel A and C show partial likelihood deviance versus log-transformed lambda, with red dots indicating deviance values and error bars. Panel B and D display coefficients versus log-transformed lambda, depicting various colored lines for different coefficients, indicating shrinkage effects as lambda changes.

Identification of independent prognostic factors for PFS and OS using cox regression analysis. (A) Forest plot depicting the results of Cox regression analysis for PFS. Six independent prognostic factors were identified as significantly associated with PFS, with HRs and 95%CIs shown for each factor. Factors with HRs greater than 1 are associated with an increased risk of progression, while HRs less than 1 indicate protective effects. (B) Forest plot showing the results of Cox regression analysis for OS. Six independent factors were identified as significantly influencing OS, with corresponding HRs and 95%CIs.

Forest plots displaying hazard ratios (HR) with 95% confidence intervals (CI) for various variables. Panel A includes histological type, maximum diameter, CA 125, albumin, LDH, and NLR, all with significant P values. Panel B shows BMI, liver metastasis, CEA, hemoglobin, albumin, and NLR, each with significant associations. Blue squares indicate HRs, with horizontal lines representing CIs. Validation and evaluation of nomograms

The independent prognostic factors identified were incorporated into separate nomogram models for PFS and OS prediction (Figures 3A, B). Internal validation was performed using bootstrap resampling (1000 iterations). The C-index for the PFS and OS prediction models were 0.706 (95%CI: 0.657-0.770) and 0.769 (95%CI: 0.710-0.834), respectively, indicating that the models have good predictive accuracy when compared with actual clinical outcomes. Calibration curves were generated to assess the agreement between the predicted and actual outcomes. As shown in Figure 4, both nomogram models demonstrated excellent calibration, with good-fitting predictive probabilities for both PFS and OS at 1-year and 2-year time points. ROC curves were used to further evaluate the discriminatory ability of the models. The AUC values approached 1, highlighting the strong predictive capacity of the models. Figures 5A, B illustrates the individual predictive performance of the PFS and OS models. To assess the clinical applicability of the nomograms, we carried out a clinical DCA to examine the net benefit across a range of threshold probabilities as illustrated in Figures 5C–F. The DCA results demonstrate that the models offer a high net benefit across a wide range of threshold values, supporting their potential clinical utility in guiding treatment decisions (Table 2).

Nomograms for predicting 1-year and 2-year PFS and OS in patients with cervical cancer. (A) Nomogram for predicting 1-year and 2-year PFS in patients with cervical cancer. (B) Nomogram for predicting 1-year and 2-year OS in cervical cancer patients.

Two panels show nomograms for predicting survival outcomes in disease subtypes. Panel A predicts one-year and two-year progression-free survival (PFS) using factors like histological subtype, CA125, albumin, LDH, NLR, and maximum diameter. Panel B predicts one-year and two-year overall survival (OS) using BMI, CEA, albumin, LDH, liver metastasis, and hemoglobin. Each panel includes a point system with linear predictors.

Calibration curves for nomogram plotted by the internal validation method. (A, B) Calibration curve for predicting 1-year, 2-year PFS nomogram. (C, D) Calibration curves for predicting 1-year and 2-year OS nomogram.

Four calibration plots labeled A, B, C, and D. Plot A shows the actual versus predicted probability of 1-year progression-free survival (PFS) with a trend line and error bars. Plot B displays similar data for 2-year PFS. Plot C shows 1-year overall survival (OS) probabilities, while plot D presents 2-year OS probabilities. All plots have a diagonal reference line indicating perfect prediction alignment.

ROC curve and DCA curve of nomogram. (A) ROC curves for 1-year, 2-year PFS nomogram. (B) ROC curves for 1-year, 2-year OS nomogram. (C, D) Decision curves for predicting 1-year and 2-year PFS nomogram; (E, F) Decision curves for predicting 1-year and 2-year OS nomogram.

Two panels showing ROC and decision curves for prognostic models. Panels A and B display ROC curves for 1-year, 2-year PFS and OS nomograms, with AUC values noted. Panels C and D show decision curves for 1-year, 2-year PFS. Panels E and F depict decision curves for 1-year, 2-year OS. Lines represent different models: all, none, and nomogram.

Diagnostic metrics comparison for 1-year and 2-year models.

Metric 1-year model 2-year model
Sensitivity 0.85 0.88
Specificity 0.90 0.92
Youden Index 0.75 0.80
F1 Score 0.86 0.89
The risk assessment capabilities of the nomogram

Based on the nomogram scores, each patient’s individual score was calculated. An optimal threshold for risk stratification was determined using these scores. The cut-off values identified for the two models were 41.7 for PFS and 175.6 for OS. KM survival curves demonstrated that patients in the low-risk group had significantly better PFS and OS compared to those in the high-risk group. This difference was statistically significant, as shown in Figure 6.

Kaplan-Meier survival curves for PFS and OS stratified by nomogram risk classification. (A) Kaplan-Meier curve for PFS based on risk stratification derived from the nomogram. (B) Kaplan-Meier curve for OS based on risk stratification from the nomogram.

Kaplan-Meier survival curves for PFS and OS. In chart A, PFS over 50 months is compared between high (blue) and low (red) scores, showing a significant difference with high scores having lower survival probability. Chart B presents OS over 120 months, also indicating a significant survival advantage for low scores. Both graphs report a log-rank test p-value less than 0.0001. Independent test cohort validation

The independent test cohort analysis revealed significant differences in several clinical characteristics. CSCC patients demonstrated better OS than CAC, particularly in patients with locally advanced and metastatic disease, where the OS difference was more pronounced (P = 0.015). Additionally, the presence of metastatic disease and an increase in tumor stage were identified as independent prognostic factors for OS (Table 3). Clinical features such as tumor stage and type of metastasis indicated that CSCC patients had better PFS compared to CAC, particularly in patients with advanced and metastatic disease, where the PFS difference was more significant (P = 0.015). The presence of metastatic disease and an increase in tumor stage had significant independent prognostic effects on PFS (Table 4).

Prognostic factors of overall survival (OS) in cervical squamous cell carcinoma (N = 253) and cervical adenocarcinoma (N = 53).

Characteristics CSCC CAC P
Status Alive 192 42
Dead 61 11 0.73
Age (Years) Mean (SD) 48.8 (14.1) 45.2 (12.1)
Median [MIN, MAX] 47 [20, 88] 43 [24, 76] 0.057
Gender FEMALE 253 53
Race AMERICAN INDIAN 7
ASIAN 15 5
BLACK 28 3
ISLANDER 2
WHITE 170 40 0.331
pT stage T1a1 1
T1b 31 5
T1b1 53 19
T1b2 25 6
T2 5 1
T2a 5 6
T2a1 6 1
T2a2 9 2
T2b 33 4
T3 1
T3a 2
T3b 16 2
T4 9 1
TX 15 2
Tis 1 0.109
pN stage N0 104 30
N1 52 9
NX 56 10 0.307
pM stage M0 102 14
M1 7 4
MX 98 31 0.02
pTNM stage I 4 1
IA 1
IA1 1
IA2 1
IB 34 4
IB1 55 22
IB2 29 10
II 4 1
IIA 6 3
IIA1 5
IIA2 7
IIB 40 3
III 1
IIIA 2
IIIB 40 3
IVA 8 1
IVB 8 5 0.015
Grade G1 13 6
G2 109 26
G3 102 17
G4 1
GX 20 4 0.298
New tumor event type Metastasis 22 9
Primary 2 1
Recurrence 11 2 0.607
Smoking Non-smoking 114 30
Smoking 99 20 0.503

BMI, Body Mass Index; CA125, Cancer Antigen 125; CEA, Carcinoembryonic Antigen; CAC, Cervical Adenocarcinoma; CSCC, Cervical Squamous Cell Carcinoma; Grade, Tumor Grade, including G1, G2, G3, G4; MAX, Maximum; MIN, Minimum; SD, Standard Deviation.

Prognostic factors of progression-free survival (PFS) in cervical squamous cell carcinoma (N = 253) and cervical adenocarcinoma (N = 53).

Characteristics CSCC CAC P
Status Progression 58 14
Progression Free 195 39 0.714
Age Mean (SD) 48.8 (14.1) 45.2 (12.1)
Median [MIN, MAX] 47 [20, 88] 43 [24, 76] 0.057
Gender FEMALE 253 53
Race AMERICAN INDIAN 7
ASIAN 15 5
BLACK 28 3
ISLANDER 2
WHITE 170 40 0.331
pT stage T1a1 1
T1b 31 5
T1b1 53 19
T1b2 25 6
T2 5 1
T2a 5 6
T2a1 6 1
T2a2 9 2
T2b 33 4
T3 1
T3a 2
T3b 16 2
T4 9 1
TX 15 2
Tis 1 0.109
pN stage N0 104 30
N1 52 9
NX 56 10 0.307
pM stage M0 102 14
M1 7 4
MX 98 31 0.02
pTNM stage I 4 1
IA 1
IA1 1
IA2 1
IB 34 4
IB1 55 22
IB2 29 10
II 4 1
IIA 6 3
IIA1 5
IIA2 7
IIB 40 3
III 1
IIIA 2
IIIB 40 3
IVA 8 1
IVB 8 5 0.015
Grade G1 13 6
G2 109 26
G3 102 17
G4  1
GX 20 4 0.298
New tumor event type Metastasis 22 9
Primary 2 1
Recurrence 11 2 0.607
Smoking Non-smoking 114 30
Smoking 99 20 0.503

BMI, Body Mass Index; CA125, Cancer Antigen 125; CEA, Carcinoembryonic Antigen; CAC, Cervical Adenocarcinoma; CSCC, Cervical Squamous Cell Carcinoma; Grade, Tumor Grade, including G1, G2, G3, G4; MAX, Maximum; MIN, Minimum; SD, Standard Deviation.

Further comparison of clinical characteristics revealed that CSCC patients were generally younger and had earlier tumor stages, particularly among patients under 46 years of age. Regarding racial distribution, CAC patients were more commonly white and black, while CSCC showed a more balanced racial distribution. In terms of tumor stage, CAC had a higher proportion of patients with advanced stages (T3/T4), while CSCC patients were more likely to be in early stages (T1/T2). Additionally, the incidence of distant metastasis (M1) was significantly higher in CAC compared to CSCC. These results suggest that cervical adenocarcinoma may have a higher invasiveness and metastatic potential, highlighting the need to consider these unique clinical features when formulating immunotherapy strategies (Figure 7).

Clinical characteristics of cervical squamous cell carcinoma (CSCC) versus cervical adenocarcinoma (CAC). (A) Age distribution, comparing patients under and over 46 years old in CSCC and CAC groups. (B) Racial distribution of CSCC and CAC patients, showing the proportion of different ethnic groups. (C) Tumor stage distribution (pT) across CSCC and CAC patients, categorized by T1, T2, T3, and T4 stages. (D) Lymph node involvement (pN), comparing N0 and N1 stages between CSCC and CAC groups. (E) Metastatic status (pM), showing the percentage of patients with M0 and M1 stages in both groups. (F) pTNM stage distribution, highlighting the differences in disease stages between CSCC and CAC, categorized by I, II, III, and IV. (G) Tumor grade distribution, comparing the proportion of G1, G2, G3, and G4 tumors in CSCC and CAC patients. (H) Smoking status, displaying the percentage of smokers versus non-smokers in both groups.

Eight grouped bar charts labeled A to H compare various demographic and clinical characteristics between CSCC and CAC. Each chart displays percentage data, with subcategories shown in different colors. Charts A to H represent categories such as age, race, tumor stage, lymph node status, and other clinical factors, highlighted with distinct color legends below each chart. Each panel includes percentage breakdowns given above the bars.

Moreover, there were significant differences in the clinical characteristics of cervical cancer patients across different disease statuses (primary, recurrence, metastasis). Older patients (>50 years) were more concentrated in the metastatic group, whereas younger patients (<50 years) were more common in the recurrence group. Tumor stage (pT stage) was closely related to disease status, with a higher proportion of T3/T4 stages in the metastatic group, while the recurrence group predominantly consisted of early-stage patients (T1/T2). Additionally, smoking was significantly more prevalent in the metastatic group compared to the other groups, suggesting that smoking may be associated with an increased risk of metastasis in cervical cancer. These findings provide potential target characteristics for cervical cancer immunotherapy, particularly in the management of metastatic disease (Figure 8).

Clinical features comparison among metastasis, recurrence, and primary cervical cancer patients. (A) Age distribution in patients with cervical cancer grouped by metastasis, recurrence, and primary status. (B) Racial distribution across cervical cancer patients with different disease statuses. (C) Tumor stage (pT) distribution in cervical cancer patients with metastasis, recurrence, and primary disease. (D) Lymph node status (pN) distribution in patients with different disease statuses. (E) Distant metastasis (pM) status comparison among patients grouped by metastasis, recurrence, and primary disease. (F) pTNM stage distribution in cervical cancer patients with different disease statuses. (G) Tumor grade comparison among metastasis, recurrence, and primary cervical cancer patients. (H) Smoking history distribution across the three disease groups (metastasis, recurrence, and primary).

Bar charts illustrating the distribution of cancer stages across different variables: (A) age, (B) race, (C) pT_stage, (D) pN_stage, (E) pM_stage, (F) pTNM_stage, (G) grade, and (H) smoking status. Each chart depicts percentages for metastasis, recurrence, and primary cancer stages in different subgroups, with statistical annotations including p-values and sample sizes. Discussion

Current biomarkers often fail to effectively identify populations with superior responses to immunotherapy, and relying on a single predictor presents significant challenges due to the high costs of next-generation sequencing (NGS) testing and the inherent tumor heterogeneity, which limit their clinical applicability. To address this gap, we developed an easily accessible predictive model based on clinicopathological features to more accurately predict the survival and prognosis of patients with advanced cervical cancer undergoing immunotherapy.

Previous studies have highlighted that systemic inflammatory responses are important independent prognostic indicators, regardless of tumor stage, and play a role in promoting tumor cell metastasis, survival, proliferation, and angiogenesis (11). Inflammation levels can be reflected by hematological parameters such as white blood cell count, neutrophils, and lymphocytes (12). These indicators partially mirror changes in the tumor immune microenvironment, both within and surrounding the tumor cells. Emerging evidence suggests that an elevated NLR is associated with an increased risk of tumor metastasis and potential resistance to immunotherapy (13). The NLR has shown significant predictive value in evaluating immunotherapy efficacy across various cancer types, including non-small cell lung cancer, malignant melanoma, and endometrial cancer (14, 15). In our study, we observed a significant correlation between elevated NLR levels and shorter PFS, underscoring its potential as a predictive biomarker in cervical cancer immunotherapy.

The patients’ tolerance to treatment, drug sensitivity, and overall immune functions are closely linked to their physical and nutritional status. Studies have demonstrated that malnutrition-induced hypermetabolic responses may hinder the production and activation of anti-tumor antibodies, thus impairing immune responses and promoting the clearance of these antibodies (16, 17). In recent years, the use of albumin to assess the immune status of cancer patients has gained increasing attention. Elevated albumin levels have been positively correlated with improved survival outcomes and a favorable prognosis following immunotherapy (18, 19). Our findings support this, as we observed that higher albumin levels were associated with longer PFS and OS, highlighting the importance of nutritional status as a determinant of immunotherapy response. LDH, a key enzyme involved in cellular metabolism, has also been shown to correlate with tumor progression, angiogenesis, and anti-tumor immunity (20). Our study confirmed the significant role of LDH in predicting both immunotherapy response and prognosis. Elevated LDH levels were associated with poorer survival, further emphasizing the potential of targeting LDH in future cancer therapies (21).

CEA is a well-established tumor marker used to monitor post-treatment responses in cancers like colorectal cancer. However, its role in cervical cancer immunotherapy remains unclear. Our study found that elevated CEA levels were significantly associated with poorer survival outcomes, potentially reflecting higher tumor burden and immune suppression (22). Furthermore, previous studies have suggested that CA125 may serve as an independent prognostic factor in cervical cancer immunotherapy, correlating with cancer load and immune microenvironment alterations (23). In our analysis, higher pre-treatment CA125 levels were associated with poorer PFS, indicating its potential as a predictive biomarker in this setting. Histological type is another factor that influences immunotherapy efficacy. Clinical observations have shown that squamous cell carcinoma of the cervix generally responds better to immunotherapy compared to adenocarcinomas, likely due to differences in tumor mutational load and microenvironmental factors (24). In general, squamous carcinomas respond better to immunotherapy, while adenocarcinomas respond poorly, which has been analyzed to be related to the formation of the immunosuppressive microenvironment of adenocarcinomas as well as the weaker stability of antigens (25). Nevertheless, few of the available analyses related to the prediction of immunotherapy efficacy have reported the inclusion of tumor histological type in regression models. As a biomarker with significant potential, in our study, we found that patients with the histologic type of cervical cancer as squamous had higher PFS. However, the difference in OS among the three histologic types of cervical cancer was not statistically significant in a multifactorial analysis.

Current literature reports indicate a positive correlation between higher BMI and increased cancer-related mortality. However, the association between higher and lower BMI in malignancies that have received immunotherapy or targeted therapy is currently unknown. This “obesity paradox” was reported in a retrospective, multi-cohort study published in The Lancet in 2018 (26). The study suggests that obesity or overweight is associated with better survival and prognosis in patients who have received immunotherapy compared to patients with metastatic melanoma who have a normal BMI, and that excessive obesity may be associated with the body’s tumorigenic immune dysfunction, which can be reversed by immune checkpoint inhibitors (27). In our study, high BMI was associated with better OS in cervical cancer patients receiving immunotherapy, but given that the study was a retrospective analysis, it was not possible to explore the reasons further.

Currently, several predictive models have been developed for immunotherapy in cervical cancer, but these models differ significantly in the selection of clinical features and biomarkers. The model proposed in this study demonstrates clear advantages over existing models, particularly in integrating various clinicopathological features (such as tumor stage, NLR, CA125 levels) and laboratory biomarkers. Compared to previous studies, our model is unique in terms of data sources, biomarker definitions, and the immunotherapy background of patients, thus enhancing its clinical applicability and potential for broader implementation (28).

Regarding the clinical application of this model, it is important to clarify its scope of use. The current model primarily targets advanced cervical cancer patients undergoing immunotherapy, with most patients being treatment-naïve. Therefore, the model may have limitations when applied to patients who have undergone multiple lines of treatment, especially in cases of immune resistance and increased tumor heterogeneity, where the response to treatment may differ significantly from that of treatment-naïve patients. Future studies should explore whether this model can be applied to patients who have undergone multiple lines of treatment and evaluate how its predictive accuracy and utility may change across different clinical stages, such as newly treated advanced patients versus those treated with multiple regimens.

One of the major limitations of this study is the relatively small sample size, particularly in the external multi-center validation, where the single-center data may introduce biases in certain results. Therefore, future research should incorporate larger multi-center datasets to further validate the model’s generalizability and applicability. Additionally, considering the need to handle complex data patterns, advanced machine learning (ML) and deep learning (DL) techniques could help uncover additional prognostic factors and clinical patterns. Furthermore, with the ongoing advancement of immunotherapy and the combination of various therapeutic modalities, predictive models based on traditional statistical methods may struggle to adapt to new treatment regimens. Therefore, our model must be continuously updated to ensure its effectiveness and relevance across diverse clinical settings as new immunotherapy strategies emerge.

 Conclusion

In summary, this study developed and validated practical nomogram models to predict progression-free survival (PFS) and overall survival (OS) in patients with recurrent or metastatic cervical cancer undergoing immunotherapy. By integrating routinely available clinical and biochemical variables—such as BMI, liver metastasis, CA125, CEA, LDH, albumin, and NLR—the models achieved strong discrimination and calibration, showing high predictive accuracy and clinical utility in both internal and external validations. These nomograms enable individualized risk assessment and may assist clinicians in optimizing treatment decisions and follow-up strategies for cervical cancer immunotherapy. Importantly, external validation using TCGA and GEO datasets confirmed the robustness and generalizability of the models across diverse populations. Future prospective multicenter studies with larger cohorts are warranted to further refine these predictive tools, incorporate molecular or immune biomarkers, and explore their dynamic use in monitoring treatment responses. Collectively, our findings provide a reliable and accessible framework for personalized prognostic evaluation and improved management of advanced cervical cancer in the era of immunotherapy.

 Data availability statement

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

 Ethics statement

The studies involving humans were approved by the ethical principles outlined in the Declaration of Helsinki. Ethical approval was granted by the Institutional Review Board and Ethics Committee of Cancer Center, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology (Approval number: UHCT230829). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

 Author contributions

ZL: Project administration, Visualization, Software, Formal analysis, Writing – original draft, Methodology, Conceptualization, Funding acquisition, Data curation, Resources, Investigation, Validation, Supervision, Writing – review & editing. SZ: Supervision, Data curation, Writing – original draft, Investigation, Software, Conceptualization, Writing – review & editing, Methodology, Funding acquisition, Visualization, Resources, Validation, Project administration, Formal analysis. YW: Investigation, Data curation, Supervision, Methodology, Writing – review & editing, Conceptualization, Software, Validation, Formal analysis, Resources, Visualization, Funding acquisition, Project administration, Writing – original draft. YZ: Writing – review & editing, Data curation, Methodology, Conceptualization, Formal analysis, Project administration, Validation, Investigation, Resources, Visualization, Software. TZ: Conceptualization, Methodology, Supervision, Project administration, Validation, Investigation, Writing – review & editing, Data curation, Funding acquisition, Resources, Writing – original draft, Formal analysis, Software, Visualization. DW: Project administration, Data curation, Formal analysis, Writing – original draft, Methodology, Resources, Visualization, Investigation, Conceptualization, Software, Supervision, Funding acquisition, Validation, Writing – review & editing. GW: Data curation, Methodology, Validation, Supervision, Conceptualization, Project administration, Software, Investigation, Resources, Funding acquisition, Writing – review & editing, Writing – original draft, Formal analysis, Visualization. LH: Supervision, Methodology, Writing – review & editing, Validation, Data curation, Conceptualization, Investigation, Writing – original draft, Formal analysis, Visualization, Resources, Software, Funding acquisition, Project administration. KH: Visualization, Methodology, Data curation, Validation, Project administration, Conceptualization, Supervision, Investigation, Software, Formal analysis, Writing – original draft, Funding acquisition, Resources, Writing – review & editing. YX: Validation, Project administration, Conceptualization, Visualization, Methodology, Data curation, Formal analysis, Investigation, Supervision, Writing – review & editing, Funding acquisition, Software, Resources, Writing – original draft. QT: Methodology, Writing – review & editing, Data curation, Investigation, Writing – original draft, Software, Funding acquisition, Supervision, Visualization, Resources, Conceptualization, Validation, Formal analysis, Project administration. MC: Writing – review & editing, Data curation, Methodology, Conceptualization, Formal analysis, Project administration, Validation, Investigation, Resources, Visualization, Software. GX: Resources, Writing – review & editing, Visualization, Funding acquisition, Formal analysis, Validation, Project administration, Investigation, Writing – original draft, Data curation, Supervision, Methodology, Software, Conceptualization. GL: Resources, Validation, Data curation, Visualization, Project administration, Writing – review & editing, Investigation, Formal analysis, Software, Funding acquisition, Methodology, Supervision, Writing – original draft, Conceptualization.

 Acknowledgments

We would like to thank all the colleagues in our laboratory for their insightful discussion and technical assistance. And this article was supported by the China Scholarship Council (CSC), and we would like to express our gratitude for their assistance and support. Thanks for the technical support by the Huazhong University of Science & Technology Analytical & Testing center, Medical sub-center. The authors thank https://www.figdraw.com for technical support.

 Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

 Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

 Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

 References Zhang Y Rumgay H Li M Cao S Chen W . Nasopharyngeal cancer incidence and mortality in 185 countries in 2020 and the projected burden in 2040: population-based global epidemiological profiling. JMIR Public Health Surveill. (2023) 9:e49968. doi: 10.2196/49968, PMID: 37728964 Abu-Rustum NR Yashar CM Arend R Barber E Bradley K Brooks R . NCCN guidelines® Insights: cervical cancer, version 1.2024. J Natl Compr Canc Netw. (2023) 21:1224–33. doi: 10.6004/jnccn.2023.0062, PMID: 38081139 Tewari KS Sill MW Penson RT Huang H Ramondetta LM Landrum LM . Bevacizumab for advanced cervical cancer: final overall survival and adverse event analysis of a randomised, controlled, open-label, phase 3 trial (Gynecologic Oncology Group 240). Lancet. (2017) 390:1654–63. doi: 10.1016/S0140-6736(17)31607-0, PMID: 28756902 Liu C Lu J Tian H Du W Zhao L Feng J . Increased expression of PD−L1 by the human papillomavirus 16 E7 oncoprotein inhibits anticancer immunity. Mol Med Rep. (2017) 15:1063–70. doi: 10.3892/mmr.2017.6102, PMID: 28075442 Otter SJ Chatterjee J Stewart AJ Michael A . The role of biomarkers for the prediction of response to checkpoint immunotherapy and the rationale for the use of checkpoint immunotherapy in cervical cancer. Clin Oncol (R Coll Radiol). (2019) 31:834–43. doi: 10.1016/j.clon.2019.07.003, PMID: 31331818 Othus M Patel SP Chae YK Dietrich E Streicher H Sharon E . Correlation between tumor size change and outcome in a rare cancer immunotherapy basket trial. J Natl Cancer Inst. (2024) 116:673–80. doi: 10.1093/jnci/djae009, PMID: 38243705 Xu L Chen L Zhang B Liu Z Liu Q Liang H . Alkaline phosphatase combined with γ-glutamyl transferase is an independent predictor of prognosis of hepatocellular carcinoma patients receiving programmed death-1 inhibitors. Front Immunol. (2023) 14:1115706. doi: 10.3389/fimmu.2023.1115706, PMID: 36761721 Banna GL Di Quattro R Malatino L Fornarini G Addeo A Maruzzo M . Neutrophil-to-lymphocyte ratio and lactate dehydrogenase as biomarkers for urothelial cancer treated with immunotherapy. Clin Transl Oncol. (2020) 22:2130–5. doi: 10.1007/s12094-020-02337-3, PMID: 32232716 Bamias A Merseburger A Loriot Y James N Choy E Castellano D . New prognostic model in patients with advanced urothelial carcinoma treated with second-line immune checkpoint inhibitors. J Immunother Cancer. (2023) 11:e005977. doi: 10.1136/jitc-2022-005977, PMID: 36627145 Bigot F Castanon E Baldini C Hollebecque A Carmona A Postel-Vinay S . Prospective validation of a prognostic score for patients in immunotherapy phase I trials: The Gustave Roussy Immune Score (GRIm-Score). Eur J Cancer. (2017) 84:212–8. doi: 10.1016/j.ejca.2017.07.027, PMID: 28826074 Zhao H Wu L Yan G Chen Y Zhou M Wu Y . Inflammation and tumor progression: signaling pathways and targeted intervention. Signal Transduct Target Ther. (2021) 6:263. doi: 10.1038/s41392-021-00658-5, PMID: 34248142 Elinav E Nowarski R Thaiss CA Hu B Jin C Flavell RA . Inflammation-induced cancer: crosstalk between tumours, immune cells and microorganisms. Nat Rev Cancer. (2013) 13:759–71. doi: 10.1038/nrc3611, PMID: 24154716 Gjyshi O Grippin A Andring L Jhingran A Lin LL Bronk J . Circulating neutrophils and tumor-associated myeloid cells function as a powerful biomarker for response to chemoradiation in locally advanced cervical cancer. Clin Transl Radiat Oncol. (2023) 39:100578. doi: 10.1016/j.ctro.2023.100578, PMID: 36935860 Capone M Giannarelli D Mallardo D Madonna G Festino L Grimaldi AM . Baseline neutrophil-to-lymphocyte ratio (NLR) and derived NLR could predict overall survival in patients with advanced melanoma treated with nivolumab. J Immunother Cancer. (2018) 6:74. doi: 10.1186/s40425-018-0383-1, PMID: 30012216 Cupp MA Cariolou M Tzoulaki I Aune D Evangelou E Berlanga-Taylor AJ . Neutrophil to lymphocyte ratio and cancer prognosis: an umbrella review of systematic reviews and meta-analyses of observational studies. BMC Med. (2020) 18:360. doi: 10.1186/s12916-020-01817-1, PMID: 33213430 Alwarawrah Y Kiernan K MacIver NJ . Changes in nutritional status impact immune cell metabolism and function. Front Immunol. (2018) 9:1055. doi: 10.3389/fimmu.2018.01055, PMID: 29868016 Flint TR Fearon DT Janowitz T . Connecting the metabolic and immune responses to cancer. Trends Mol Med. (2017) 23:451–64. doi: 10.1016/j.molmed.2017.03.001, PMID: 28396056 Yoo S-K Chowell D Valero C Morris LGT Chan TA . Pre-treatment serum albumin and mutational burden as biomarkers of response to immune checkpoint blockade. NPJ Precis Oncol. (2022) 6:23. doi: 10.1038/s41698-022-00267-7, PMID: 35393553 Guo Y Wei L Patel SH Lopez G Grogan M Li M . Serum albumin: early prognostic marker of benefit for immune checkpoint inhibitor monotherapy but not chemoimmunotherapy. Clin Lung Cancer. (2022) 23:345–55. doi: 10.1016/j.cllc.2021.12.010, PMID: 35131184 Hermans D Gautam S García-Cañaveras JC Gromer D Mitra S Spolski R . Lactate dehydrogenase inhibition synergizes with IL-21 to promote CD8+ T cell stemness and antitumor immunity. Proc Natl Acad Sci U S A. (2020) 117:6047–55. doi: 10.1073/pnas.1920413117, PMID: 32123114 Miholjcic TBS Halse H Bonvalet M Bigorgne A Rouanne M Dercle L . Rationale for LDH-targeted cancer immunotherapy. Eur J Cancer. (2023) 181:166–78. doi: 10.1016/j.ejca.2022.11.032, PMID: 36657325 Ni G Zhang L Yang X Li H Ma B Walton S . Targeting interleukin-10 signalling for cancer immunotherapy, a promising and complicated task. Hum Vaccin Immunother. (2020) 16:2328–32. doi: 10.1080/21645515.2020.1717185, PMID: 32159421 Guo S Yang B Liu H Li Y Li S Ma L . Serum expression level of squamous cell carcinoma antigen, highly sensitive C-reactive protein, and CA-125 as potential biomarkers for recurrence of cervical cancer. J Cancer Res Ther. (2017) 13:689–92. doi: 10.4103/jcrt.JCRT_414_17, PMID: 28901315 Huang RSP Haberberger J Murugesan K Danziger N Hiemenz M Severson E . Clinicopathologic and genomic characterization of PD-L1-positive uterine cervical carcinoma. Mod Pathol. (2021) 34:1425–33. doi: 10.1038/s41379-021-00780-3, PMID: 33637877 Yan T Ma G Wang K Liu W Zhong W Du J . The immune heterogeneity between pulmonary adenocarcinoma and squamous cell carcinoma: A comprehensive analysis based on lncRNA model. Front Immunol. (2021) 12:547333. doi: 10.3389/fimmu.2021.547333, PMID: 34394068 McQuade JL Daniel CR Hess KR Mak C Wang DY Rai RR . Association of body-mass index and outcomes in patients with metastatic melanoma treated with targeted therapy, immunotherapy, or chemotherapy: a retrospective, multicohort analysis. Lancet Oncol. (2018) 19:310–22. doi: 10.1016/S1470-2045(18)30078-0, PMID: 29449192 Hahn AW Menk AV Rivadeneira DB Augustin RC Xu M Li J . Obesity is associated with altered tumor metabolism in metastatic melanoma. Clin Cancer Res. (2023) 29:154–64. doi: 10.1158/1078-0432.CCR-22-2661, PMID: 36166093 Huang H Liu Q Zhu L Zhang Y Lu X Wu Y . Prognostic value of preoperative systemic immune-inflammation index in patients with cervical cancer. Sci Rep. (2019) 9:3284. doi: 10.1038/s41598-019-39150-0, PMID: 30824727

Edited by: Xi Wang, The Chinese University of Hong Kong, China

Reviewed by: Hongfu Zhao, Jilin University, China

Zhengjun Guo, The Second Affiliated Hospital of Chongqing Medical University, China