CT-based body composition and inflammatory nutritional biomarker nomogram for predicting early postoperative recurrence of non-small cell lung cancer: a multicenter study

Introduction

According to recent data released by the International Agency for Research on Cancer (IARC), lung cancer remains the leading malignancy worldwide in both incidence and mortality (1). Non-small cell lung cancer (NSCLC) accounts for 80–85% of all lung cancer cases. Currently, radical resection is a common treatment option for patients with early-stage and some cases of locally advanced NSCLC. However, Stage I lung adenocarcinoma recurs in 20–50% of cases post-resection (2), while resectable stage II or higher disease shows a 5-year event-free survival of 20% (3) and >50% of recurrences within two years (4), termed early recurrence (ER). Although the Tumor-Node-Metastasis (TNM) staging system is an essential tool for prognostic assessment and therapeutic guidance, considerable heterogeneity in outcomes is observed even among patients within the same stage, indicating the need to develop a more rational and broadly applicable prognostic model to effectively stratify high-risk patients and tailor individualized treatment strategies.

Increasing evidence underscores the critical role of inflammation in tumorigenesis, contributing not only to tumor initiation, progression, and metastasis but also to the remodeling of the tumor microenvironment. The interplay between tumor cells and the surrounding stroma, coupled with the influence of inflammatory cells on the tumor milieu, constitutes a central mechanism driving cancer progression (5). Furthermore, patients’ nutritional status and body composition are integral to evaluating immune competence in cancer patients. Sarcopenia has been significantly associated with early postoperative complications and reduced long-term survival across multiple malignancies (6-8), while decreased visceral adipose tissue correlates positively with heightened mortality risk in diverse cancer populations (9). Additionally, prior studies have demonstrated that systemic inflammatory responses are linked to cancer-associated weight and muscle loss, with consistent associations observed between alterations in body composition and systemic inflammation (10,11).

Computed tomography (CT), as a routine modality for diagnosis, staging, and follow-up of lung cancer, provides cross-sectional imaging of skeletal muscle and adipose tissue, enabling precise and detailed assessment of body composition without additional financial burden to patients (12,13). Although extensive research has explored the impact of body composition, inflammation, and nutritional indicators on cancer outcomes (14-16), their interrelationships and prognostic significance in early postoperative outcomes of NSCLC patients remain unclear. Therefore, this study aims to develop a body composition and inflammatory nutritional biomarker (BCINB) score by integrating CT-based body composition parameters with preoperative inflammatory and nutritional serum biomarkers to improve risk stratification for ER in NSCLC patients. Furthermore, we constructed and validated a prognostic nomogram incorporating the BCINB score and key clinicopathological variables, which provides a refined prognostic tool that may facilitate personalized therapeutic decision-making. We present this article in accordance with the TRIPOD reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1211/rc).

Methods

General information

Patients with stage I–III NSCLC (according to the AJCC 9th edition) who underwent radical surgery at Jiangxi Cancer Hospital & Institution, Jiangxi Clinical Research Center for Cancer, The Second Affiliated Hospital of Nanchang Medical College and The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University from August 2018 to March 2021 were retrospectively reviewed. Inclusion criteria comprised: (I) pathological diagnosis of primary NSCLC; (II) R0 resection with standard lymphadenectomy; and (III) complete blood tests and qualifying chest CT within two weeks pre-surgery. Exclusion criteria were: (I) histological types other than adenocarcinoma or squamous cell carcinoma; (II) prior anti-tumor treatments before surgery; (III) previous malignancies; (IV) preoperative systemic infections, autoimmune diseases, or hematologic disorders; and (V) unconfirmed recurrence/metastasis or follow-up under two years. After exclusions, 251 patients from Jiangxi Cancer Hospital formed the training cohort, and 104 patients from The Second Affiliated Hospital of Nanchang University served as the external test cohort (Figure S1, I). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Jiangxi Cancer Hospital & Institution (Nanchang, China, 2024ky217) and individual consent for this retrospective analysis was waived. The other institution was informed and approved this study.

Clinical pathological data and definition of variables

Demographic and clinical data included age, sex, body mass index (BMI), smoking history, and pathological TNM (pTNM) stage. All staging information (T, N, and overall stage) analyzed in this study was derived from pathological reports obtained after radical resection and standard lymphadenectomy. Preoperative laboratory tests—performed within 30 hours before surgery—covered carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9), blood cell counts (white, lymphocyte, monocyte, neutrophil), bilirubin, albumin, globulin, gamma-glutamyl transferase (GGT), aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphatase (ALP) levels. Surgical procedures were classified per International Lung Cancer Surgery Guidelines (17). The choice of surgical procedure was guided by tumor location, size, and patient-specific factors such as pulmonary function, consistent with current standard clinical practice for resectable NSCLC. Adjuvant chemotherapy was primarily given to patients with pTNM stage II–III disease in accordance with National Comprehensive Cancer Network (NCCN) Guidelines (18). Additionally, a small subset of high-risk stage IB patients, exhibiting features like lymphovascular invasion, visceral pleura invasion or poor differentiation, received adjuvant therapy based on multidisciplinary team decisions (Table S1). Pathological features from postoperative reports included tumor location, T and N stages, histology, differentiation, and presence of pleural, nerve, or vascular invasion.

Composite indices were calculated as follows: Albumin to alkaline phosphatase ratio (AAPR) = albumin/ALP; albumin to globulin ratio (AGR) = albumin/globulin; albumin-bilirubin index (ALBI) = log10(bilirubin) × 0.66 − albumin × 0.085; derived neutrophil-lymphocyte ratio (dNLR) = neutrophils/(white cells − neutrophils); platelet-lymphocyte ratio (PLR) = platelets/lymphocytes; prognostic nutritional index (PNI) = albumin + 5 × lymphocytes; systemic inflammation index (SII) = platelets × neutrophils/lymphocytes; lymphocyte-albumin (LA) = lymphocytes × albumin; advanced lung cancer inflammation index (ALI) = BMI × albumin/NLR; lymphocyte to monocyte ratio (LMR) = lymphocytes/monocytes; monocyte to albumin ratio (MAR) = monocytes/albumin; and systemic inflammatory response index (SIRI) = (monocytes × neutrophils)/lymphocytes. Continuous variable cutoffs were determined by maximum rank selection (19). ALBI grades were defined as Grade 1 (≤−2.60), Grade 2 (>−2.60 to ≤−1.39), and Grade 3 (≥−1.39).

CT scanning and parameter measurement

All NSCLC patients underwent non-contrast chest CT scans within two weeks before diagnosis, covering from the thoracic inlet to bilateral adrenal glands with scanner settings detailed in Table S2. CT images were retrieved from the Picture Archiving and Communication System and analyzed using RadiAntViewer software. Instead of abdominal CT, single axial slices at the first lumbar vertebra (L1) level were used to assess whole-body composition, as previously validated (20). A senior radiologist identified the L1 plane, and cross-sectional DICOM images were obtained. Semi-automated measurements were performed with sliceomatic 5.0 software (Figure S1, II).

Tissue Hounsfield unit (HU) thresholds were defined as (21,22): subcutaneous adipose tissue (SAT) −190 to −30 HU, skeletal muscle (SM) −29 to +150 HU, and intermuscular adipose tissue (IMAT) −150 to −30 HU. L1 skeletal muscle (L1-SM) included abdominal wall, diaphragm, intercostal, paravertebral, and psoas muscles (23). Areas were normalized by height squared (m²) to calculate indices: subcutaneous adipose tissue index (SATI), skeletal muscle index (SMI), and intermuscular adipose tissue index (IMATI). Average HU values per region were determined to assess tissue density: subcutaneous adipose tissue density (SATD), skeletal muscle density (SMD), and intermuscular adipose tissue density (IMATD). The ratio of IMATD to SMD yielded Intramuscular Adipose Content (IMAC), reflecting intramuscular fat content (21). Body composition was independently assessed in a double-blind manner by two radiologists with 9 and 19 years’ experience. Inter-observer consistency was evaluated on 30 randomly selected cases, with final values averaged. Gender-specific cutoff values were established due to sex differences (24) (see Appendix 1).

Follow-up

Postoperative follow-up for NSCLC patients was scheduled every three months during the first two years, including physical exams, serum tumor markers, chest and abdominal CT, and PET-CT scans (25). ER was defined as local recurrence or distant metastasis occurring within two years post-surgery. Local recurrence referred to confirmed tumor relapse at ipsilateral bronchial stump or regional lymph nodes (26), while distant metastasis involved confirmed spread to contralateral lung or distant organs. If both occurred simultaneously, classification favored distant metastasis (27). Tumor recurrence and metastasis were diagnosed radiologically by chest CT or PET-CT, and confirmed pathologically via bronchoscopy and biopsy. Recurrence-free survival (RFS) was the endpoint, measured from surgery date to first confirmed recurrence or metastasis via imaging or pathology.

Construction of the BCINB score

In the training cohort, univariate Cox regression analysis was conducted to identify prognostic BCINB with P values <0.05. These variables were then evaluated using the least absolute shrinkage and selection operator (LASSO)-Cox regression to determine their prognostic significance. The BCINB score was calculated based on features with non-zero coefficients. The algorithm was iterated 1,000 times to ensure analytical precision, and 10-fold cross-validation via the cv.glmnet function was applied to reduce overfitting. Non-parametric tests were used to compare BCINB scores among subgroups with different clinical and pathological characteristics to explore associations with adverse outcomes.

Constructing and validating the nomogram

Based on the independent predictors identified through multivariable Cox regression analysis, we developed a nomogram. To enhance model robustness and minimize overfitting, Bootstrap resampling with 1,000 iterations was applied. The nomogram’s generalizability was then evaluated using an external test dataset. Calibration curves assessed the model’s accuracy, while performance and clinical utility were evaluated through the concordance index (C-index), time-dependent receiver operating characteristic (ROC) curves, and decision curve analysis (DCA), compared against the BCINB score, pathological models, and the conventional pTNM staging system.

Nomogram risk scores were calculated for all patients, and using the training cohort median as the cutoff, patients were stratified into high- and low-risk groups. Subgroup analyses across different datasets and adjuvant chemotherapy status post-surgery were performed to validate the model’s robustness.

Statistical analysis

Continuous variables are presented as mean ± standard deviation. Inter-group comparisons were conducted using either the t-test or the Mann-Whitney U test. Categorical variables are reported as frequencies (percentages), and comparisons between groups were performed using the chi-square test. The intraclass correlation coefficient (ICC) was calculated to assess inter-observer variability. Kaplan-Meier curves were generated to illustrate RFS, and differences in RFS were analyzed using the log-rank test. All statistical analyses were performed using R Studio (version 4.1.2) with the following R packages: “glmnet”, “rms”, “survival”, “timeROC”, “pec”, “dcurves”, and “survivalROC”. P values were two-tailed, with statistical significance considered at P<0.05.

Results

Baseline characteristics of included patients

A total of 355 patients participated in this study, comprising 137 females (39%) and 218 males (61%). Among these participants, 244 patients (69%) were younger than 65 years, while 111 patients (31%) were 65 years or older. Based on the AJCC 9th Edition TNM staging system, 152 patients (43%) were classified as stage I, 100 patients (28%) as stage II, and 103 patients (29%) as stage III. The median follow-up period was 33 months, with an interquartile range of 24 to 41 months. During the follow-up, we documented 73 instances (21%) of ER; of these, 52 cases (21%) were reported from Jiangxi Cancer Hospital and 21 cases (20%) from The Second Affiliated Hospital of Nanchang University. Jiangxi Cancer Hospital served as the training cohort, while The Second Affiliated Hospital of Nanchang University was designated as the external test cohort. Detailed demographic and clinicopathological characteristics are presented in Table 1.

Interobserver reliability assessment

Results of the ICC analysis for body composition parameters revealed the following: SAT (ICC =0.994; 95% CI: 0.988–0.997), SM (ICC =0.958; 95% CI: 0.914–0.980), IMAT (ICC =0.949; 95% CI: 0.895–0.975), SATD (ICC =0.984; 95% CI: 0.967–0.992), SMD (ICC =0.906; 95% CI: 0.813–0.954), and IMATD (ICC =0.943; 95% CI: 0.884–0.972). These findings indicate a high level of measurement reproducibility and observer agreement between the two radiologists.

Construction and validation of the BCINB score

The correlation between 26 body composition parameters and inflammatory nutritional biomarkers is depicted in Figure 1A. Univariate Cox regression analysis identified 16 variables with P values less than 0.05 (Figure 1B). These variables were subsequently included in a LASSO-Cox regression analysis (Figure S2), which revealed eight non-zero coefficient features associated with prognosis in NSCLC: SII, IMAC, SIRI, IMAT, SATI, AGR, SMD, and SMI (Figure 1C). The BCINB score was constructed using the following formula: BCINB score = −0.243659 + SII * 0.21661141 + IMAC * 0.11574079 + SIRI * 0.21661141 + IMAT * 0.03888766 − SATI * 0.18791388 − AGR * 0.21661141 − SMD * 0.48366848 − SMI * 0.18791388.

Figure 1 Construction of the BCINB score. (A) Correlation heatmap of BCINB. (B) Univariate Cox regression analysis for recurrence-free survival. (C) Feature weight histogram generated by LASSO-Cox regression. (D) Comparative analysis of BCINB score distributions across different clinicopathological characteristics. Wilcoxon test was used for comparisons between two groups. AGR, albumin to globulin ratio; BCINB, body composition and inflammatory nutritional biomarker; cTNM, clinical Tumor-Node-Metastasis; IMAT, intermuscular adipose tissue; LASSO, least absolute shrinkage and selection operator; SATI, subcutaneous adipose tissue index; SII, systemic inflammation index; SIRI, systemic inflammatory response index; SM, skeletal muscle; SMD, skeletal muscle density.

We evaluated the differences in BCINB scores among various subgroups of clinical and pathological characteristics (Figure 1D). Patients characterized by male gender, age over 65 years, tumor size greater than 3 cm, and pTNM stage III exhibited significantly higher BCINB scores. However, no notable differences in BCINB scores were observed within the subgroups categorized by histological type and degree of differentiation.

Construction and evaluation of a nomogram model

In the multivariate Cox regression analysis, we adjusted for BCINB score and other clinicopathological characteristics, incorporating correlations determined to be statistically significant with RFS from univariate analyses (Table 2). Our results demonstrate that the BCINB score is an independent prognostic factor for patients with NSCLC (HR: 13.853, 95% CI: 6.393–30.018). Based on multivariate Cox regression analysis, we developed and validated a prognostic nomogram to predict ER in patients with NSCLC. This model incorporates the BCINB score, N stage, pleural invasion, nerve invasion, and vascular invasion (Figure 2A). Calibration plots for 1- and 2-year DFS showed strong agreement between nomogram predictions and observed probabilities in both the training and test cohorts (Figure 2B,2C). The C-index for the training cohort was 0.822 (95% CI: 0.780–0.864), and for the test cohort, it was 0.806 (95% CI: 0.736–0.869), both higher than those of the BCINB score, pathological model, and pTNM staging system (Table 3). In both the training and test cohorts, we analyzed the survival ROC curves for 1- and 2-year RFS rates. The 1- and 2-year RFS rates in the training cohort were 0.816 and 0.856 (Figure 3A,3B), respectively, while the test cohort showed rates of 0.825 and 0.871 (Figure 3C,3D), all of which exceeded those observed in the pathological model and pTNM staging system. Additionally, DCA highlights the clinical utility of the nomogram, demonstrating a wider range of threshold probabilities for predicting 1- and 2-year RFS rates in both the training and test cohorts, when compared to other staging systems (Figure 3E,3F).

Figure 2 Predictive nomogram incorporating BCINB score for 1- and 2-year RFS after radical resection of NSCLC and calibration curves. (A) Nomogram integrating the BCINB score and clinicopathological variables to predict 1- and 2-year DFS probabilities after radical resection of NSCLC. For example, a patient presenting with pleural invasion and vascular invasion, classified as N2 stage, has a BCINB score of 0.015 and a total nomogram score of 365 points, corresponding to an 82.9% probability of recurrence within 1 year and a 99.5% probability within 2 years. Calibration curves for 1- and 2-year RFS in the (B) training cohort and (C) external test cohort. *, P-value <0.05; **, P-value <0.01; ***, P-value <0.001. BCINB, body composition and inflammatory nutritional biomarker; DFS, disease-free survival; N, node; NSCLC, non-small cell lung cancer.

Figure 3 ROC curves of the established nomogram and staging systems in the training cohort for predicting 1-year (A) and 2-year (B) recurrence-free survival, along with the corresponding clinical decision curve analysis (E). ROC curves of multiple models in the external validation cohort for predicting 1-year (C) and 2-year (D) recurrence-free survival, and the associated clinical decision curve analysis (F). AUC, area under the curve; BCINB, body composition and inflammatory nutritional biomarker; cTNM, clinical Tumor-Node-Metastasis; ROC, receiver operating characteristic.

Prognostic stratification of the nomogram model

The linear predicted values derived from the nomogram model were defined as risk scores (Figure 4A). Using an optimal cutoff value of −0.013, determined from the median of the training cohort, we categorized all patients into low-risk and high-risk groups. Kaplan-Meier survival curves indicated that high-risk scores were significantly associated with ER in both the training and test cohorts (all P<0.001, Figure 4B,4C). These findings remained consistent in the postoperative adjuvant chemotherapy subgroup, with no statistically significant confounding factors (Figure 4D,4E).

Figure 4 Prognostic analysis. (A) Visualization of differences between high- and low-risk groups based on the median cutoff (−0.013) of the nomogram risk score, illustrating relapse time, relapse status, and clinicopathological factors. Kaplan-Meier DFS curves for patients in the (B) training cohort and (C) testing cohort. Kaplan-Meier DFS curves stratified by (D) no adjuvant chemotherapy and (E) adjuvant chemotherapy subgroups. BCINB, body composition and inflammatory nutritional biomarker; DFS, disease-free survival; N, node.

Discussion

The progression of malignancies is a dynamic process driven by the interplay between the host inflammatory microenvironment and metabolic dysregulation (28,29). Chronic inflammation promotes tumor immune evasion through activation of NF-κB and STAT3 signaling pathways (30-32), whereas metabolic abnormalities, such as skeletal muscle wasting and ectopic fat deposition, exacerbate systemic inflammatory responses (33). This bidirectional interaction along the “inflammation-metabolism axis” constitutes a central pathogenic mechanism in tumor progression. In this study, we innovatively integrated circulating inflammatory-nutritional biomarkers with quantitative body composition parameters derived from CT imaging to develop the BCINB prognostic scoring system. This approach aims to overcome the limitations of single-factor assessment by multidimensionally reflecting the interaction between host status and tumor biology. Our results demonstrate that the BCINB score is an independent predictor of ER following surgery in NSCLC patients (HR: 13.853, 95% CI: 6.393–30.018). Subgroup analyses revealed higher BCINB scores in males, patients older than 65 years, those with tumor diameter >3 cm, and stage III, indicating enhanced discriminatory capacity of the scoring system in high-risk populations.

The BCINB score comprises three circulating inflammatory-nutritional biomarkers (SII, SIRI, AGR) and five quantitative body composition parameters (SMI, SMD, SATI, IMAT, IMAC), whose prognostic value has been validated across multiple tumor types. SMI and SMD are key indicators for assessing skeletal muscle loss and muscle fatty infiltration; reductions in these parameters signify sarcopenia and myosteatosis, both strongly associated with poor prognosis. Recent evidence indicates that when SMD increased by 5 and 10 HU, the risk of death decreased by 8% and 16% respectively (34). Chaunzwa et al. (35) similarly reported a negative correlation between skeletal muscle mass decline and immunotherapy outcomes in NSCLC. Conversely, the “obesity paradox” has been widely observed, whereby moderate obesity may confer survival benefits (36-38). Alifano et al. (39) found that overweight and obese NSCLC patients exhibited improved postoperative survival, and Liu et al. (40) demonstrated that higher BMI correlates with prolonged progression-free survival in lung cancer patients receiving immunotherapy. However, recent studies highlight the dual effects of fat distribution, noting that BMI alone fails to capture body composition nuances relevant to tumor prognosis (41,42). Our study stratifies adipose tissue into SAT and IMAT to more precisely explore their prognostic implications. SAT, serving as an energy reservoir, may enhance antitumor immunity by secreting protective adipokines such as adiponectin, thereby improving insulin resistance (43,44). Lee et al. (45) demonstrated that greater SAT volume favorably impacts progression-free survival in NSCLC patients. In contrast, elevated IMAT and IMAC are closely linked to mitochondrial dysfunction in muscle (46). Moreover, metabolically active IMAT exhibits visceral fat-like inflammatory profiles by secreting cytokines such as IL-6 and TNF-α, fostering a local pro-tumorigenic microenvironment and perpetuating a vicious cycle with mitochondrial impairment (47,48). Consistently, in our scoring system, higher SMI, SMD, and SATI exert protective effects and positively correlate with prognosis, whereas increased IMAT and IMAC levels are associated with significantly RFS.

Regarding circulating inflammatory-nutritional biomarkers, the SII, which incorporates peripheral lymphocyte, platelet, and neutrophil counts, provides a comprehensive assessment of the host immune response and inflammatory status. Huai et al. (49) reported that low SII levels are associated with improved disease-free survival in NSCLC patients receiving neoadjuvant chemotherapy. In addition, Studies indicate that higher pre-treatment SIRI predicts poorer survival in patients undergoing chemoradiotherapy for advanced NSCLC (50), and SIRI is also an independent predictor of early progression in advanced NSCLC patients treated with PD-1 inhibitors (51). Additionally, the AGR is a crucial indicator of hepatic synthetic function and nutritional status. Compared to other malnutrition assessment tools, AGR has been shown to effectively stratify prognosis in cancer cachexia patients (52). Our findings align with previous studies, showing that decreased SII and SIRI levels, alongside increased AGR, are associated with more favorable clinical outcomes. Collectively, the BCINB score is closely linked to malnutrition, active inflammatory response, metabolic dysregulation, and tumor progression. The test cohort using an external dataset further supports the BCINB score’s robust predictive accuracy and generalizability for postoperative ER in NSCLC patients.

While the TNM staging system remains the gold standard for prognostic evaluation and treatment stratification in NSCLC, significant outcome heterogeneity persists among same-stage patients. This has driven the development of multiple predictive models in recent years to enhance postoperative recurrence risk assessment. Saw et al. (53) developed a postoperative recurrence prediction model based on clinicopathological characteristics—including adenocarcinoma subtype, grade, lymphovascular invasion, stage, and smoking status—to identify high-risk stage I EGFR-positive NSCLC patients, achieving a C-index of 0.706. Tu et al. (54) constructed a nomogram using data from 4,238 patients in the SEER database to facilitate individualized cancer-specific survival assessment for stage IB NSCLC patients without lymphovascular invasion and visceral pleural invasion. This model incorporated seven prognostic factors: sex, age, grade, histological subtype, tumor size, surgical type, and number of resected lymph nodes, yielding robust predictive accuracy (C-indexes of 0.755 and 0.726 in training and test cohorts, respectively). Wang et al. (55) developed a prognostic model based on clinicopathological features that outperformed TNM staging, with C-indexes of 0.669 vs. 0.580, 0.676 vs. 0.582, and 0.733 vs. 0.572 in the training, internal validation, and external test cohorts, respectively. Soomin et al. (56) evaluated the prognostic value of the Nobel and Underwood score for overall survival in NSCLC patients. The NUn score, a logistic regression model incorporating C-reactive protein, serum albumin, and white blood cell count, reflects systemic inflammation and nutritional status. When combined with age, American Society of Anesthesiologists Physical Status, TNM stage, and pleural invasion, this integrated predictive model exhibited superior performance compared to the TNM staging system alone (C-index of 0.832 vs. 0.72). Compared to these studies, our model incorporates not only conventional clinicopathological features but also body composition and inflammation-nutrition related parameters. The results demonstrate that, relative to clinical models, the integrated model significantly enhances the accuracy of predicting postoperative ER in NSCLC patients, with C-index of 0.822 and 0.806 in the training and test cohorts, respectively, and provides greater clinical net benefit.

In clinical practice, we obtained body composition parameters using CT imaging. Although both CT and MRI serve as baseline modalities for diagnosis and staging and can provide additional information without increasing examination costs, CT has become the preferred imaging modality for pulmonary lesions owing to its rapid acquisition, cost-effectiveness, and quantitative HU measurements that correlate with tissue pathophysiology (57), thus holding promising clinical translational potential. Furthermore, we selected the L1 vertebral level from chest CT scans for body composition assessment rather than the gold-standard L3 level. This approach reduces patient financial burden and improves clinical workflow efficiency. Liu et al. (20) demonstrated a strong correlation between body composition parameters obtained at L1 and L3 levels. Additionally, some studies have used the T12 level to measure SMI, SMD, and IMAT to predict in-hospital mortality among elderly inpatients (58), yielding results consistent with those derived from L1 measurements.

The limitations of this study include its retrospective design, which may introduce selection bias. Due to the limited sample size and event rate (73 cases of ER among 355 patients), coupled with the increased estimation variance caused by the integration of multi-dimensional variables and their potential interactions, the confidence intervals are widened. In subsequent studies, we plan to increase the sample size by expanding multi-center cohorts, with a particular focus on increasing the number of events (target ≥150 events), so as to improve the accuracy of estimation. Additionally, assessing body composition using CT images at a single axial slice at the L1 level may reflect regional specificity, such as the absence of visceral adipose tissue parameters, resulting in a relatively incomplete evaluation of whole-body composition. In subsequent studies, we plan to develop a segmentation model for the automatic extraction of body composition parameters from CT images. This automation can reduce measurement bias and reliance on professional radiologists, making it particularly suitable for low- and medium-volume institutions with limited resources and personnel. In addition, we aim to develop a user-friendly, web-based BCINB score calculator and integrate this automatic body composition extraction technology to facilitate clinical application. Furthermore, we used only preoperative baseline body composition as predictive indicators, lacking assessment of postoperative changes and their impact on long-term prognosis. Finally, there is no established consensus on cutoff values for body composition parameters, necessitating further validation in larger, diverse populations.

Conclusions

The BCINB score integrates comprehensive inflammatory and nutritional body composition parameters, offering a novel perspective for ER assessment in NSCLC patients. The nomogram based on BCINB facilitates more precise individualized prognostic evaluations for these patients.

Acknowledgments

The authors express their thanks to the patients and their families, and all researchers and medical staff members.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1211/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1211/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1211/prf

Funding: This study was supported by the Science and Technology Planning Project of Jiangxi Provincial Health Commission (Project No. 202510509 to Y.Z.) and the Science and Technology Planning Project of Jiangxi Provincial Health Commission (Project No. 202110081 to F.Z.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1211/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Jiangxi Cancer Hospital & Institution (Nanchang, China, 2024ky217) and individual consent for this retrospective analysis was waived. The other institution was informed and approved this study.

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

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Wang Y, Ding Y, Liu X, et al. Preoperative CT-based radiomics combined with tumour spread through air spaces can accurately predict early recurrence of stage I lung adenocarcinoma: a multicentre retrospective cohort study. Cancer Imaging 2023;23:83. [Crossref] [PubMed]
3. Donington J, Hu X, Zhang S, et al. Real-World Neoadjuvant Treatment Patterns and Outcomes in Resected Non-Small-Cell Lung Cancer. Clin Lung Cancer 2024;25:440-8. [Crossref] [PubMed]
4. Ettinger DS, Wood DE, Aisner DL, et al. NCCN Guidelines Insights: Non-Small Cell Lung Cancer, Version 2.2021. J Natl Compr Canc Netw 2021;19:254-66. [Crossref] [PubMed]
5. Greten FR, Grivennikov SI. Inflammation and Cancer: Triggers, Mechanisms, and Consequences. Immunity 2019;51:27-41. [Crossref] [PubMed]
6. Okugawa Y, Kitajima T, Yamamoto A, et al. Clinical Relevance of Myopenia and Myosteatosis in Colorectal Cancer. J Clin Med 2022;11:2617. [Crossref] [PubMed]
7. Findlay M, White K, Stapleton N, et al. Is sarcopenia a predictor of prognosis for patients undergoing radiotherapy for head and neck cancer? A meta-analysis. Clin Nutr 2021;40:1711-8. [Crossref] [PubMed]
8. Zhang XM, Dou QL, Zeng Y, et al. Sarcopenia as a predictor of mortality in women with breast cancer: a meta-analysis and systematic review. BMC Cancer 2020;20:172. [Crossref] [PubMed]
9. Li L, Li W, Xu D, et al. Association Between Visceral Fat Area and Cancer Prognosis: A Population-Based Multicenter Prospective Study. Am J Clin Nutr 2023;118:507-17. [Crossref] [PubMed]
10. Aziz MH, van Dongen JC, Saida L, et al. High Systemic Immune Inflammation Index Is Associated With Low Skeletal Muscle Quantity in Resectable Pancreatic Ductal Adenocarcinoma. Front Oncol 2022;12:827755. [Crossref] [PubMed]
11. Guo H, Feng S, Li Z, et al. Prognostic Value of Body Composition and Systemic Inflammatory Markers in Patients with Locally Advanced Cervical Cancer Following Chemoradiotherapy. J Inflamm Res 2023;16:5145-56. [Crossref] [PubMed]
12. Tolonen A, Pakarinen T, Sassi A, et al. Methodology, clinical applications, and future directions of body composition analysis using computed tomography (CT) images: A review. Eur J Radiol 2021;145:109943. [Crossref] [PubMed]
13. Bates DDB, Pickhardt PJ. CT-Derived Body Composition Assessment as a Prognostic Tool in Oncologic Patients: From Opportunistic Research to Artificial Intelligence-Based Clinical Implementation. AJR Am J Roentgenol 2022;219:671-80. [Crossref] [PubMed]
14. Zhang XW, Ge YZ, Song MM, et al. Prognostic Power of Nutrition-Inflammation Indicators in Patients With Breast Cancer. Clin Breast Cancer 2023;23:e312-21. [Crossref] [PubMed]
15. Teke K, Avci IE, Cinar NB, et al. Immune-inflammatory-nutritional status predicts oncologic outcomes after radical cystectomy for urothelial carcinoma of bladder. Actas Urol Esp (Engl Ed) 2023;47:430-40. [Crossref] [PubMed]
16. McSorley ST, Black DH, Horgan PG, et al. The relationship between tumour stage, systemic inflammation, body composition and survival in patients with colorectal cancer. Clin Nutr 2018;37:1279-85. [Crossref] [PubMed]
17. Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus limited resection for T1 N0 non-small cell lung cancer. Lung Cancer Study Group. Ann Thorac Surg 1995;60:615-22; discussion 622-3. [Crossref] [PubMed]
18. National Comprehensive Cancer Network. (2025). NCCN Guidelines for Non-Small Cell Lung Cancer (Version 7.2025). Available online: https://www.nccn.org/guidelines
19. Zhu J, Wang D, Liu C, et al. Development and validation of a new prognostic immune-inflammatory-nutritional score for predicting outcomes after curative resection for intrahepatic cholangiocarcinoma: A multicenter study. Front Immunol 2023;14:1165510. [Crossref] [PubMed]
20. Liu S, Han X, Li J, et al. Feasibility of using chest computed tomography (CT) imaging at the first lumbar vertebra (L1) level to assess skeletal muscle mass: a retrospective study. PeerJ 2023;11:e16652. [Crossref] [PubMed]
21. Shi S, Zhao YX, Fan JL, et al. Development and External Validation of a Nomogram Including Body Composition Parameters for Predicting Early Recurrence of Hepatocellular Carcinoma After Hepatectomy. Acad Radiol 2023;30:2940-53. [Crossref] [PubMed]
22. Recio-Boiles A, Galeas JN, Goldwasser B, et al. Enhancing evaluation of sarcopenia in patients with non-small cell lung cancer (NSCLC) by assessing skeletal muscle index (SMI) at the first lumbar (L1) level on routine chest computed tomography (CT). Support Care Cancer 2018;26:2353-9. [Crossref] [PubMed]
23. Kang DO, Park SY, Choi BG, et al. Prognostic Impact of Low Skeletal Muscle Mass on Major Adverse Cardiovascular Events in Coronary Artery Disease: A Propensity Score-Matched Analysis of a Single Center All-Comer Cohort. J Clin Med 2019;8:712. [Crossref] [PubMed]
24. Benmassaoud A, Roccarina D, Arico FM, et al. Sex is a major effect modifier between body composition and mortality in patients with cirrhosis assessed for liver transplantation. Liver Int 2023;43:160-9. [Crossref] [PubMed]
25. Ettinger DS, Wood DE, Aisner DL, et al. Non-Small Cell Lung Cancer, Version 3.2022, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2022;20:497-530. [Crossref] [PubMed]
26. Lou F, Huang J, Sima CS, et al. Patterns of recurrence and second primary lung cancer in early-stage lung cancer survivors followed with routine computed tomography surveillance. J Thorac Cardiovasc Surg 2013;145:75-81; discussion 81-2. [Crossref] [PubMed]
27. Jeong WG, Choi H, Chae KJ, et al. Prognosis and recurrence patterns in patients with early stage lung cancer: a multi-state model approach. Transl Lung Cancer Res 2022;11:1279-91. [Crossref] [PubMed]
28. Iyengar NM, Gucalp A, Dannenberg AJ, et al. Obesity and Cancer Mechanisms: Tumor Microenvironment and Inflammation. J Clin Oncol 2016;34:4270-6. [Crossref] [PubMed]
29. Balkwill F, Mantovani A. Inflammation and cancer: back to Virchow? Lancet 2001;357:539-45. [Crossref] [PubMed]
30. Cao Y, Yi Y, Han C, et al. NF-κB signaling pathway in tumor microenvironment. Front Immunol 2024;15:1476030. [Crossref] [PubMed]
31. Surai PF, Surai A, Earle-Payne K. Silymarin and Inflammation: Food for Thoughts. Antioxidants (Basel) 2024;13:98. [Crossref] [PubMed]
32. Taniguchi K, Karin M. NF-κB, inflammation, immunity and cancer: coming of age. Nat Rev Immunol 2018;18:309-24. [Crossref] [PubMed]
33. Fearon K, Strasser F, Anker SD, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol 2011;12:489-95. [Crossref] [PubMed]
34. Sjøblom B, Grønberg BH, Wentzel-Larsen T, et al. Skeletal muscle radiodensity is prognostic for survival in patients with advanced non-small cell lung cancer. Clin Nutr 2016;35:1386-93. [Crossref] [PubMed]
35. Chaunzwa TL, Qian JM, Li Q, et al. Body Composition in Advanced Non-Small Cell Lung Cancer Treated With Immunotherapy. JAMA Oncol 2024;10:773-83. [Crossref] [PubMed]
36. Zhang X, Rui M, Lin C, et al. The association between body mass index and efficacy of pembrolizumab as second-line therapy in patients with recurrent/metastatic head and neck squamous cell carcinoma. Cancer Med 2023;12:2702-12. [Crossref] [PubMed]
37. Abdel-Rahman O, Ghosh S, Morrish D. Impact of baseline body mass index on the outcomes of patients with neuroendocrine neoplasms. J Endocrinol Invest 2022;45:1683-8. [Crossref] [PubMed]
38. Li Y, Li C, Wu G, et al. The obesity paradox in patients with colorectal cancer: a systematic review and meta-analysis. Nutr Rev 2022;80:1755-68. [Crossref] [PubMed]
39. Alifano M, Daffré E, Iannelli A, et al. The Reality of Lung Cancer Paradox: The Impact of Body Mass Index on Long-Term Survival of Resected Lung Cancer. A French Nationwide Analysis from the Epithor Database. Cancers (Basel) 2021;13:4574. [Crossref] [PubMed]
40. Liu Z, Diao Y, Li X. Body mass index and serum markers associated with progression-free survival in lung cancer patients treated with immune checkpoint inhibitors. BMC Cancer 2022;22:824. [Crossref] [PubMed]
41. Wang J, Dong P, Qu Y, et al. Association of computed tomography-based body composition with survival in metastatic renal cancer patient received immunotherapy: a multicenter, retrospective study. Eur Radiol 2023;33:3232-42. [Crossref] [PubMed]
42. Prado CM, Cushen SJ, Orsso CE, et al. Sarcopenia and cachexia in the era of obesity: clinical and nutritional impact. Proc Nutr Soc 2016;75:188-98. [Crossref] [PubMed]
43. Inoue O, Goten C, Hashimuko D, et al. Single-cell transcriptomics identifies adipose tissue CD271(+) progenitors for enhanced angiogenesis in limb ischemia. Cell Rep Med 2023;4:101337. [Crossref] [PubMed]
44. Tran TT, Kahn CR. Transplantation of adipose tissue and stem cells: role in metabolism and disease. Nat Rev Endocrinol 2010;6:195-213. [Crossref] [PubMed]
45. Lee JW, Lee HS, Na JO, et al. Effect of adipose tissue volume on prognosis in patients with non-small cell lung cancer. Clin Imaging 2018;50:308-13. [Crossref] [PubMed]
46. Xu Z, You W, Chen W, et al. Single-cell RNA sequencing and lipidomics reveal cell and lipid dynamics of fat infiltration in skeletal muscle. J Cachexia Sarcopenia Muscle 2021;12:109-29. [Crossref] [PubMed]
47. Di Ludovico A, La Bella S, Ciarelli F, et al. Skeletal muscle as a pro- and anti-inflammatory tissue: insights from children to adults and ultrasound findings. J Ultrasound 2024;27:769-79. [Crossref] [PubMed]
48. Balan AI, Halațiu VB, Scridon A. Oxidative Stress, Inflammation, and Mitochondrial Dysfunction: A Link between Obesity and Atrial Fibrillation. Antioxidants (Basel) 2024;13:117. [Crossref] [PubMed]
49. Huai Q, Luo C, Song P, et al. Peripheral blood inflammatory biomarkers dynamics reflect treatment response and predict prognosis in non-small cell lung cancer patients with neoadjuvant immunotherapy. Cancer Sci 2023;114:4484-98. [Crossref] [PubMed]
50. Yang H, Wang K, Li B, et al. The Prognostic Role of Blood Inflammatory Biomarkers and EGFR Mutation Status in Stage IIIA/N2 Non-Small Cell Lung Cancer Patients Treated With Trimodality Therapy. Front Oncol 2021;11:707041. [Crossref] [PubMed]
51. Tang C, Zhang M, Jia H, et al. The systemic inflammation response index (SIRI) predicts survival in advanced non-small cell lung cancer patients undergoing immunotherapy and the construction of a nomogram model. Front Immunol 2024;15:1516737. [Crossref] [PubMed]
52. Xie HL, Zhang Q, Ruan GT, et al. Evaluation and Validation of the Prognostic Value of Serum Albumin to Globulin Ratio in Patients With Cancer Cachexia: Results From a Large Multicenter Collaboration. Front Oncol 2021;11:707705. [Crossref] [PubMed]
53. Saw SPL, Zhou S, Chen J, et al. Association of Clinicopathologic and Molecular Tumor Features With Recurrence in Resected Early-Stage Epidermal Growth Factor Receptor-Positive Non-Small Cell Lung Cancer. JAMA Netw Open 2021;4:e2131892. [Crossref] [PubMed]
54. Tu Z, Li C, Tian T, et al. A risk classification system predicting the cancer-specific survival for postoperative stage IB non-small-cell lung cancer patients without lymphovascular and visceral pleural invasion. Lung Cancer 2021;161:114-21. [Crossref] [PubMed]
55. Wang ZH, Deng L. Establishment and Validation of a Predictive Nomogram for Postoperative Survival of Stage I Non-Small Cell Lung Cancer. Int J Gen Med 2022;15:7287-98. [Crossref] [PubMed]
56. An S, Eo W, Kim DH, et al. Prognostic Value of the Noble and Underwood Score in Patients with Non-Small Cell Lung Cancer Undergoing Surgical Resection. J Cancer 2024;15:6185-95. [Crossref] [PubMed]
57. Charette N, Vandeputte C, Ameye L, et al. Prognostic value of adipose tissue and muscle mass in advanced colorectal cancer: a post hoc analysis of two non-randomized phase II trials. BMC Cancer 2019;19:134. [Crossref] [PubMed]
58. Shen Y, Luo L, Fu H, et al. Chest computed tomography-derived muscle mass and quality indicators, in-hospital outcomes, and costs in older inpatients. J Cachexia Sarcopenia Muscle 2022;13:966-75. [Crossref] [PubMed]