Correlations of ¹⁸F-FDG PET/CT metabolic parameters and systemic immune-inflammation index with distant metastasis of small-size (T1) NSCLC

Introduction

Lung cancer is a leading cause of cancer mortality, with non-small cell lung cancer (NSCLC) comprising 85% of cases (1). Approximately one-third of patients are diagnosed at an advanced stage, missing the optimal surgical window (2). Although T1 NSCLC (tumors <3 cm) generally has a better prognosis due to lower metastasis likelihood (3,4), a notable proportion (24–27%) present with distant metastases (tumor-node-metastasis (TNM) stage IV) (5-7) and are associated with poor outcomes, presenting the contradictory feature of “small tumor size yet early metastasis”. This subgroup of patients harbors potential metastasis risk factors that cannot be identified solely by conventional staging criteria. Therefore, the original intention of this study was to identify reliable risk factors associated with distant metastasis in T1-stage NSCLC patients. Identifying risk factors for distant metastasis in small-size NSCLC is crucial for selecting patients who may benefit from earlier intervention. Translation of these findings into clinical practice addresses the critical unmet need for accurate risk stratification in T1-stage NSCLC, enabling reliable distinction between high- and low-metastatic-risk patients. This biomarker panel facilitates individualized management: high-risk patients may benefit from intensified upfront therapies (e.g., neoadjuvant chemotherapy, targeted therapy, or immunotherapy) to reduce metastasis, whereas low-risk patients are suited for minimally invasive strategies to avoid overtreatment. Additionally, it guides tailored follow-up, with close surveillance for high-risk cohorts and simplified monitoring for low-risk groups. Collectively, these results complement the TNM system and substantially improve the precision of clinical decision-making for early-stage NSCLC.

The nuclear metabolic imaging modality fluorine-18-fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography/computed tomography (PET/CT) is widely used for diagnosis, staging, and treatment monitoring in lung cancer due to its ability to provide both whole-body anatomical and metabolic data of the tumor lesion, providing comprehensive disease assessment. Compared to conventional imaging methods such as computed tomography (CT) and magnetic resonance imaging (MRI), it offers greater accuracy in evaluating tumor phenotypes (8-10). Its role in staging and assessing therapy response in lung cancer is well established (11,12). Local immune responses and cancer-associated inflammation are critical drivers of tumor progression, including in NSCLC (13). Among inflammatory markers, the neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), platelet-to-lymphocyte ratio (PLR), and systemic immune-inflammation index (SII) are recognized indicators of immune status and systemic inflammation (14,15). Recent studies have demonstrated that such inflammatory parameters are prognostic in several solid tumors, including NSCLC (16-18). Elevated tumor glycolysis and systemic inflammation synergize to drive metastasis. Several studies have examined links between ¹⁸F-FDG PET/CT parameters and the above inflammatory markers in tumor progression (19,20). However, the association with distant metastasis in small-size NSCLC remains unclear.

This study investigated the correlation between ¹⁸F-FDG PET/CT metabolic parameters and serum inflammatory markers in small-size NSCLC metastasis, aiming to clarify the metabolic-inflammatory interaction driving early dissemination. We aimed to distinguish high-risk patients prone to metastasis from the overall T1-stage population. This will provide a scientific basis for formulating individualized treatment strategies, avoiding undertreatment or overtreatment, and ultimately improving patients’ prognosis. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1701/rc).

Methods

Study population

This retrospective study enrolled 243 patients with NSCLC tumors smaller than 3 cm (T1), with (n=114) or without (n=129) distant metastasis, who underwent ¹⁸F-FDG PET/CT one to two weeks before tumor resection or biopsy between January 2019 and December 2020 at Shanghai Chest Hospital (Figure 1). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Shanghai Chest Hospital. Informed consent was provided by all the patients.

Figure 1 Study flowchart that shows the inclusion and exclusion criteria. A total of 243 patients were finally enrolled in our study. ¹⁸F-FDG, fluorine-18-fluorodeoxyglucose; NSCLC, non-small cell lung cancer; PET/CT, positron emission tomography/computed tomography.

The inclusion criteria were as follows: no prior anti-tumor treatment before the PET/CT scan; pathologically confirmed NSCLC; and solid nodule with maximum diameter <3 cm. The exclusion criteria were as follows: ground-glass density nodules; incomplete clinical data; active infection or inflammation; hematological disease; cardiovascular/cerebrovascular events within 1 month; and blood transfusion within 3 months. To focus on patients with stage T1 NSCLC with or without distant metastasis, individuals with N1 or N2 lymph node involvement identified on PET/CT were also excluded.

PET/CT scanning and parameter measurement

The PET/CT scanning process is described in our previously published article (21). ¹⁸F-FDG PET/CT was performed after fasting for at least 8 hours and 60 minutes after intravenous administration of ¹⁸F-FDG (5 MBq/kg, radiochemical purity 95%, provided by Shanghai Kexin Pharmaceutical, China). The blood glucose levels, measured just before tracer administration, were <8.0 mmol/L in all patients. The imaging was performed with a combined PET/CT device (Biograph mCT64; Siemens, Erlangen, Germany). Immediately after CT scan using a 64-slice helical CT (120 keV, 30–100 mA in Auto mA mode, 5.00 mm slice thickness), PET was performed with an acquisition time of 3 minutes per frame in 3-dimensional (3D) mode. CT data were used for attenuation correction, and images were iteratively reconstructed using a trueX + time-of-flight (TOF) algorithm (21 subsets, 3 iterations). Axial, sagittal, and coronal images were examined using Syngo software and then transferred in Digital Imaging and Communications in Medicine (DICOM) format to a Syngo workstation (Siemens). The regions of interest (ROIs) of primary lesions were manually defined. The manual delineation of the lesions was performed by Lihua Wang. All delineations were completed under the supervision of a senior radiologist with more than 20 years of clinical experience. Standardized uptake value peak (SUVpeak), standardized uptake value maximum (SUVmax), standardized uptake value mean (SUVmean), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) (TLG = SUVmean × MTV) of ROIs were recorded by two nuclear medicine physicians, who were blinded to patient characteristics. The discrepancies between physicians’ assessments were resolved by consensus through discussion, and patients with diffuse lung cancers with a hard-to-define ROI of the primary lesion were discussed and excluded. The determination of parameters such as SUVmax, SUVmean, MTV, and TLG can be found in our published article (21). The following parameters were assessed for each primary lung tumor:

• SUVpeak: determined by averaging pixel values over a 1.0 cc spherical kernel centered at the area of highest uptake, proposed as an alternative evaluation method;
• The coefficient of variation (COV) was calculated as standard deviation (SD)/SUVmean.

All parameters were measured on the PET images by using four SUV thresholds (1.0, 1.5, 2.0, 2.5), defined as the SUV values above which voxels within the 3D ROI were considered part of the metabolic volume.

The ROIs of primary lesions were manually defined, and SUVmax and diameters were recorded by two experienced nuclear medicine physicians (W.X. and C.C., approximately 20 and 15 years of experience in interpretation of PET/CT images, respectively) blinded to the study design. The differences between physicians’ assessments were resolved by consensus through discussion, and they reached an almost perfect agreement. Representative examples are shown in Figure 2.

Figure 2 ¹⁸F-FDG PET/CT images of different TNM stages of stage T1 NSCLC. (A) Representative clinical and imaging findings of a 66-year-old male patient pathological diagnosis of large cell carcinoma, ¹⁸F-FDG PET/CT identified a primary pulmonary lesion localized to the upper lobe of the right lung, with a maximum diameter of approximately 2.84 cm, with an ¹⁸F-FDG uptake (SUVmax) of 14.6. No lymph node metastasis and no distant metastasis. The combination of T1c, N0, and M0 yielded a final TNM staging of Stage IA3. (B) Representative clinical and imaging findings of a 56-year-old male patient pathological diagnosis of lung adenocarcinoma, ¹⁸F-FDG PET/CT identified a primary pulmonary lesion localized to the middle lobe of the right lung, with a maximum diameter of approximately 1.71 cm, with an ¹⁸F-FDG uptake (SUVmax) of 7.1. Concurrent metastatic involvement was confirmed in the right supraclavicular lymph nodes and mediastinal lymph nodes via either pathological verification. The combination of T1b, N3, and M0 yielded a final TNM staging of Stage IIIB. (C) Representative clinical and imaging findings of a 52-year-old female patient pathological diagnosis of lung adenocarcinoma, ¹⁸F-FDG PET/CT identified a primary pulmonary lesion localized to the upper lobe of the right lung, with a maximum axial diameter of approximately 1.91 cm, with an 18F-FDG uptake (SUVmax) of 17.6. Metastatic lesions were simultaneously observed in both the supraclavicular lymph nodes and the mediastinal lymph nodes. At the same time, multiple bone metastases were also found. The combination of T1b, N3, and M1c yielded a final TNM staging of Stage IVB. (D) Representative clinical and imaging findings of a 42-year-old female patient pathological diagnosis of lung squamous cell carcinoma, ¹⁸F-FDG PET/CT identified a primary pulmonary lesion localized to the left lower lobe of the lung, with a maximum axial diameter of approximately 2.44 cm, with an ¹⁸F-FDG uptake (SUVmax) of 16.4. The patient presented with multiple lymph node metastases, bone metastases and liver metastases. The combination of T1c, N3, and M1c yielded a final TNM staging of Stage IVB. ¹⁸F-FDG, fluorine-18-fluorodeoxyglucose; NSCLC, non-small cell lung cancer; PET/CT, positron emission tomography/computed tomography; SUVmax, maximum standardized uptake value; TNM, tumor-node-metastasis.

Serum inflammation factors examination

Complete blood count data were analyzed within one week before tumor resection or biopsy. Data included white cell, neutrophil, monocyte, platelet, and lymphocyte counts, and fibrinogen concentration. The calculation of NLR, MLR, and PLR indicators refers to the articles we have published (21). SII = platelet count × neutrophil count/lymphocyte count.

Determination of programmed death ligand 1 (PD-L1) and detection of gene mutations

All data were extracted from our hospital’s database. PD-L1 tumor proportion score (TPS) <1%, 1–49%, and ≥50% were considered to be PD-L1-negative, PD-L1-moderate, and PD-L1-strong, respectively. For the gene mutation assay, five routine gene mutations were tested in each patient, including epidermal growth factor receptor (EGFR) mutation, anaplastic lymphoma kinase (ALK) rearrangement, ROS proto-oncogene 1, receptor tyrosine kinase (ROS1) fusion, Kirsten rat sarcoma viral oncogene homolog (KRAS), and B-Raf proto-oncogene, serine/threonine kinase (BRAF) mutations. All determinations of PD-L1 expression status and gene mutations were performed using the submitted lung tumor specimens.

Statistical analysis

Statistical analyses were conducted using GraphPad Prism 6 (GraphPad Software, San Diego, CA, USA) and SPSS 19.0. Quantitative data were expressed as mean ± SD. Statistical differences between groups were evaluated using unpaired two-tailed t-tests or one-way analysis of variance (ANOVA). Risk factors for metastasis were identified using univariate and multivariate logistic regression models. Factors with P<0.05 in univariate analysis were included in the multivariate analysis. Categorical variables were expressed as numbers and percentages and compared using chi-square or Fisher’s exact tests, with odds ratios (ORs) and 95% confidence intervals (CIs). The area under the receiver operating characteristic (ROC) curve (AUC) was calculated, and the cutoff values, with the highest Youden index being used as the optimal cutoff threshold for predicting malignant. A P value <0.05 (*), <0.01 (**) or <0.001 (***) was considered statistically significant.

Results

Patient characteristics

The characteristics of the 243 patients with NSCLC enrolled in this study are presented in Table 1 and Table S1. Patients were randomly assigned to modeling and validation cohorts in a 3:1 ratio (182:61). There were no significant differences in patient characteristics between the two cohorts (P>0.05). Staging was conducted according to the 8th edition of the Lung Cancer Staging Manual (22). Nonsmokers were defined as individuals with a lifetime cigarette consumption of <100.

Correlation of ¹⁸F-FDG PET/CT parameters with metastasis of small-size NSCLC

¹⁸F-FDG PET/CT images of the 182 patients with T1 NSCLC in the modeling cohort were analyzed. SUVpeak, SUVmax, SUVmean, MTV, TLG, and COV values were recorded under different SUV thresholds (1.0, 1.5, 2.0, and 2.5). To evaluate the relationship between tumor glycolytic activity and malignant patterns, we assessed the correlation between these PET/CT parameters and metastasis status in the 182 patients. As shown in Figure 3, SUVmax, SUVmean (at all thresholds), MTV (at all thresholds), TLG (at all thresholds), and COV (at thresholds 1.5, 2.0, and 2.5) were significantly higher in the metastasis group than they were in the non-metastasis group (all P<0.05). However, COV at threshold 1.0 (COV1.0) showed no significant difference (P=0.65). ROC analysis (Figure 4) identified optimal cut-off values for predicting metastasis.

Figure 3 ¹⁸F-FDG PET/CT parameter values were higher in metastatic NSCLC than non-metastatic NSCLC in the modeling cohort. (A) The mean of SUVmax in the metastasis group were higher than those in the non-metastasis group (10.83±4.06 vs. 9.46±4.67, P<0.05). (B) The mean of SUVpeak in the metastasis group were higher than those in the non-metastasis group (7.27±3.33 vs. 5.86±3.22, P<0.05). (C) The mean of SUVmean1.0, SUVmean1.5, SUVmean2.0, and SUVmean2.5 in the metastasis group were higher than those in the non-metastasis group (4.10±1.28 vs. 3.48±1.27, 4.45±1.22 vs. 3.92±1.35, 4.84±1.22 vs. 4.37±1.38 and 5.22±1.29 vs. 4.76±1.47, all P<0.05). (D) The mean of MTV1.0, MTV 1.5, MTV 2.0, and MTV 2.5 in the metastasis group were higher than those in the non-metastasis group (9.38±3.40 vs. 6.28±3.08, 8.18±4.87 vs. 5.10±2.54, 7.08±3.64 vs. 4.17±2.19 and 6.22±2.33 vs. 3.53±1.90, all P<0.05). (E) The mean of TLG1.0, TLG1.5, TLG2.0, and TLG 2.5 in the metastasis group were higher than those in the no-metastasis group (42.89±23.10 vs. 24.14±10.98, 41.17±23.69 vs. 22.50±10.34, 38.91±29.49 vs. 20.76±9.71 and 37.15±29.58 vs. 19.43±9.47, all P<0.05). (F) The mean of COV1.0, COV1.5, COV2.0, and COV2.5 in the metastasis group were higher than those in the non-metastasis group (0.58±0.12 vs. 0.56±0.33, 0.47±0.14 vs. 0.41±0.11, 0.44±0.13 vs. 0.37±0.14 and 0.40±0.13 vs. 0.32±0.14, all were <0.05 except COV1.0). *, P<0.05. ¹⁸F-FDG, fluorine-18-fluorodeoxyglucose; COV, coefficient of variation; MTV, metabolic tumor volume; NSCLC, non-small cell lung cancer; PET/CT, positron emission tomography/computed tomography; SUVmax, maximum standardized uptake value; SUVmean, mean standardized uptake value; SUVpeak, peak standardized uptake value; TLG, total lesion glycolysis; TNM, tumor-node-metastasis.

Figure 4 ROC curve analysis for the optimal cut of value of ¹⁸F-FDG PET/CT parameters. (A) ROC curve analysis of ¹⁸F-FDG PET/CT parameter to predict metastasis in the modelling cohort. With the optimal cutoff values of 4.10 [AUC (95% CI) =0.63 (0.55–0.70); P<0.05], a high SUVmax could predict metastasis with sensitivity and specificity of 70.60% and 49.50%, respectively. (B) ROC curve analysis of ¹⁸F-FDG PET/CT parameter to predict metastasis in the modelling cohort. With the optimal cutoff values of 8.33 [AUC (95% CI) =0.60 (0.52–0.68); P<0.05], a high SUVpeak could predict metastasis with sensitivity and specificity of 84.70% and 37.10%, respectively. (C) ROC curve analysis of ¹⁸F-FDG PET/CT parameter to predict metastasis in the modelling cohort. With the optimal cutoff values of 3.03, 3.47, 4.10, and 4.50, respectively [AUC (95% CI) = 0.63 (0.54–0.71), 0.61 (0.53–0.67), 0.61 (0.53–0.69), 0.60 (0.52–0.68), respectively; P both <0.05], a high SUVmean1.0, SUVmean1.5, SUVmean2.0, SUVmean2.5, could predict metastasis with sensitivity and specificity of 81.20%, 82.40%, 71.80%, 70.60% and 41.20%, 42.30%, 50.50%, 50.50%, respectively. (D) ROC curve analysis of ¹⁸F-FDG PET/CT parameter to predict metastasis in the modelling cohort. With the optimal cutoff values of 5.38, 5.72, 4.28, and 4.51, respectively [AUC (95% CI) =0.63 (0.55–0.71), 0.92 (0.98–0.96), 0.67 (0.59–0.75), 0.67 (0.59–0.75), respectively; P both <0.05], a high MTV1.0, MTV1.5, MTV2.0, and MTV2.5, could predict metastasis with sensitivity and specificity of 69.40%, 60.00%, 63.50%, 57.60%, and 57.70%, 69.10%, 68.00%, 72.20%, respectively. (E) ROC curve analysis of ¹⁸F-FDG PET/CT parameter to predict metastasis in the modelling cohort. With the optimal cutoff values of 25.12, 24.68, 22.70, and 24.08, respectively [AUC (95% CI) =0.65 (0.57–0.73), 0.65 (0.57–0.73), 0.66 (0.58–0.74), 0.66 (0.58–0.74), respectively; P both <0.05], a high TLG1.0, TLG 1.5, TLG 2.0, and TLG2.5, could predict metastasis with sensitivity and specificity of 60.00%, 58.80%, 58.80%, 56.50%, and 70.10%, 71.10%, 74.20%, respectively. (F) ROC curve analysis of ¹⁸F-FDG PET/CT parameter to predict metastasis in the modelling cohort. With the optimal cutoff values of 0.40, 0.33, and 0.32, respectively [AUC (95% CI) =0.62 (0.54–0.70), 0.63 (0.55–0.71), 0.64 (0.56–0.72), respectively; P both <0.05], a high COV1.5, COV2.0, and COV2.5, could predict metastasis with sensitivity and specificity of 85.90%, 89.40%, 74.10%, and 36.10%, 34.00%, 50.50%, respectively. ¹⁸F-FDG, fluorine-18-fluorodeoxyglucose; AUC, area under the curve; CI, confidence interval; COV, coefficient of variation; MTV, metabolic tumor volume; PET/CT, positron emission tomography/computed tomography; ROC, receiver operating characteristic; SUVmax, maximum standardized uptake value; SUVmean, mean standardized uptake value; SUVpeak, peak standardized uptake value; TLG, total lesion glycolysis.

The optimal cutoff values for SUVpeak, SUVmax, SUVmean1.0, MTV1.0, and TLG1.0 for metastasis were >4.10, 8.33, 3.03, 5.38, and 25.12, respectively. The optimal cutoff values for SUVmean1.5, MTV1.5, TLG1.5, and COV1.5 were >3.47, 5.72, 24.68, and 0.40, respectively. The optimal cutoff values for SUVmean2.0, MTV2.0, TLG2.0, and COV2.0 were >4.10, 4.28, 22.70, and 0.33, respectively. The corresponding values for SUVmean2.5, MTV2.5, TLG2.5, and COV2.5 were >4.50, 4.51, 24.08, and 0.32, respectively.

We then grouped patients into categories based on the above-mentioned cutoff values, and subsequent univariate and multivariate logistic regression analyses revealed that a COV2.0 value greater than 0.33 was a significant independent risk factor on ¹⁸F-FDG PET/CT for metastasis in small-size NSCLC [hazard ratio (HR) =3.03, P=0.01, Table 2].

Correlation of serum inflammation factors with metastasis of small-size NSCLC

To investigate tumor-associated systemic inflammatory marker changes in small-size NSCLC patients, we analyzed pretreatment serum markers, including white cell, neutrophil, monocyte, platelet, and lymphocyte count, as well as fibrinogen concentration, NLR, MLR, PLR, and SII in the 182 patients with or without metastasis in the modeling cohort enrolled in our study.

As shown in Figure 5, the levels of serum white cells (P<0.001), neutrophils (P<0.001), platelets (P=0.03), fibrinogen (P<0.001), NLR (P<0.001), PLR (P=0.001), and SII (P<0.001) were significantly higher in metastatic patients than in non-metastatic patients. However, monocyte, and lymphocyte count, as well as MLR, did not differ significantly (Figure S1).

Figure 5 Serum inflammatory factors were higher in metastatic NSCLC than in non-metastatic NSCLC in the modeling cohort. (A) The mean of white cell count in the metastasis group was higher than that in the non-metastasis group [(7.63±3.22) vs. (6.35±1.64) ×10⁹, P<0.01]. (B) The mean of neutrophil count in the metastasis group was higher than that in the non-metastasis group [(5.35±2.07) vs. (4.10±1.36) ×10⁹, P<0.01]. (C) The mean of platelet count in the metastasis group was higher than that in the non-metastasis group [(240.61±68.44) vs. (218.93±60.17) ×10⁹, P<0.01]. (D) The mean of fibrinogen in the metastasis group was higher than that in the non-metastasis group (2.82±0.59 vs. 3.46±1.0 g/L, P=0.001). (E) The mean of NLR in the metastasis group was higher than that in the non-metastasis group (3.73±1.32 vs. 2.56±1.15, P<0.001). (F) The mean of PLR in the metastasis group was higher than that in the non-metastasis group (161.32±59.64 vs. 135.31±46.08, P=0.001). (G) The mean of SII in the metastasis group was higher than that in the non-metastasis group (914.64±233.35 vs. 553.84±277.68, P<0.001). *, P<0.05; **, P<0.01. NLR, neutrophil-to-lymphocyte ratio; NSCLC, non-small cell lung cancer; PLR, platelet-to-lymphocyte ratio; SII, systemic immune-inflammation index.

Further ROC analysis (Figure 6) revealed that the optimal cutoff values of white cell count, neutrophil count, platelet count, fibrinogen concentration, NLR, PLR, and SII for predicting metastasis were 6.85, 3.95, 219.50, 3.54, 2.35, 127.80, and 641.68, respectively.

Figure 6 ROC curve analysis of the optimal cutoff value of inflammation factors to predict metastasis in the modelling cohort. (A) ROC curve analysis of inflammation factors to predict metastasis in the modelling cohort. With the optimal cutoff values of 6.85, 3.95, 219.50, and 3.54 respectively [AUC (95% CI) =0.64 (0.56–0.72), 0.68 (0.60–0.75), 0.59 (0.51–0.68), 0.69 (0.61–0.77), respectively; P all <0.01], a high white cell count, neutrophil count, platelet count, and fibrinogen could predict metastasis with sensitivity and specificity of 57.70%, 75.30%, 60.00%, 49.40%, and 67.00%, 67.00%, 60.80%, and 90.70%, respectively. (B) ROC curve analysis of inflammation factors to predict metastasis in the modelling cohort. With the optimal cutoff values of 2.35, 127.80, and 641.68 respectively [AUC (95% CI) =0.68 (0.68–0.76), 0.64 (0.56–0.73), 0.71 (0.63–0.78), respectively; P all <0.01], a high NLR, PLR, and SII could predict metastasis with sensitivity and specificity of 72.90%, 71.80%, and 61.20%, and 54.60%, 56.70%, and 73.20%, respectively. AUC, area under the curve; CI, confidence interval; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; ROC, receiver operating characteristic; SII, systemic immune-inflammation index.

Patients were then grouped according to these cutoff values, and subsequent univariate and multivariate logistic regression analyses indicated that SII greater than 641.68 was an independent inflammation-related risk factor for metastasis of small-size NSCLC (HR =2.84, P=0.003, Table 3).

Correlation of clinicopathological features with metastasis of small-size NSCLC

We also analyzed the correlation of clinicopathological features with metastasis in the modeling cohort. The clinicopathological factors included age, gender, histological subtype, maximum tumor diameter, and smoking history. As a result, no significant differences in these factors were observed between the two groups (Table 4).

Correlation between ¹⁸F-FDG PET/CT parameters and serum inflammation factors in small-size NSCLC

To investigate the association between tumor glycolysis and systemic inflammation in small-size NSCLC metastasis, we grouped patients into low/high COV2.0 groups using a cutoff value of 0.33, and into low/high SII groups using a cutoff of 641.68 (the optimal values associated with tumor metastasis). Analysis showed that COV2.0 was positively associated with serum SII levels (OR =4.55; P=0.03, Table 5). These data indicate that pretreatment serum SII may also be a tumor-glycolysis-associated inflammatory marker in small-size NSCLC.

Co-high COV2.0/SII is associated with metastasis of small-size NSCLC

Since both COV2.0 and SII were independently associated with metastasis (as shown in Tables 2,3), we combined COV2.0 (>0.33) and SII (>641.68) to define a co-high COV2.0/SII group. We analyzed the association of COV2.0, SII, and co-high COV2.0/SII status with metastasis in the 182 patients with NSCLC in the modeling cohort. Univariate and multivariate analyses showed that co-high COV2.0/SII status was an independent risk factor for metastasis (HR =5.08, 95% CI: 2.63–9.81, P<0.001), even after adjusting for COV2.0 or SII alone (Table 6).

Furthermore, in the 61 patients with NSCLC in the validation cohort, which was independent of the modeling cohort, the probability of co-high COV2.0/SII status being associated with metastasis was 84.20% (16/19), significantly higher than in the non-metastasis group (15.80%, 3/19) (P=0.043). In contrast, the probability of co-low COV2.0/SII status being associated with metastasis was 17.90% (5/28), significantly lower than in the non-metastasis group (82.10%, 23/28) (P<0.001, Table 7).

PD-L1 expression and gene mutations in primary lung tumors in 243 patients with NSCLC

Among the 243 patients included in this study, 109 of 114 with distant metastasis and 59 of 129 without metastasis underwent PD-L1 measurement, totaling 168 cases. Statistical analysis of the correlation between PD-L1 levels and patient characteristics (such as age, gender, smoking history, or pathological subtype) was not significant (Table S2).

Among the 243 patients included in this study, 109 of 114 with distant metastasis and 80 of 129 without metastasis underwent gene mutation testing, including EGFR mutation, ALK rearrangement, ROS1 fusion, KRAS mutation, and BRAF mutation. The results showed that EGFR mutation was identified in 82 cases, with a mutation rate of 43.39% (82/189); ALK rearrangement was detected in 27 cases, accounting for 14.29% (27/189); ROS1 fusion was found in 27 cases, with a positive rate of 14.29% (27/189); KRAS mutation was observed in 16 cases, with a mutation rate of 8.47% (16/189); and BRAF mutation was confirmed in 5 cases, corresponding to a mutation rate of 2.65% (5/189). Statistical analysis of the correlation between gene mutations and patient characteristics (such as age, gender, smoking history, or pathological subtype) was not significant (Table S3).

We analyzed the association between gene mutations, COV2.0, and serum SII in 243 patients and found no statistical significance (Table S4). Furthermore, no significant association with metastasis risk was found in this group (Table S5).

Discussion

In this study, we examined two groups of small primary lung lesions: one without metastasis (T1N0M0) and one with multiple metastases (T1N3M0–T1N3M1). We aimed to identify a combined indicator based on PET/CT metabolic profiles and peripheral blood inflammation associated with multiple metastases in small lesions. This is the first investigation of the relationship between the metabolic index COV2.0 and the inflammatory index SII with tumor metastasis in small-size NSCLC. We found that pretherapeutic COV2.0 >0.33 and SII >641.68 (co-high) were closely associated with metastasis in this cohort. This finding may help improve understanding of lung cancer metastasis from the dual perspectives of tumor metabolism and host immune inflammation.

In theory, tumors at stage I behave less aggressively and may carry a lower risk of lymph node involvement. Reports have shown that the presence of mediastinal lymph node metastasis in patients with stage I NSCLC, as determined by CT, ranges from 6% to 21% (23). Pathological lymph node metastasis is detected in fewer than 10% of patients with clinical-stage I NSCLC (24). For N1 metastasis, surgery is the optimal treatment. For confirmed N2 metastasis, surgery alone is insufficient; multimodality therapy—including chemotherapy and/or radiotherapy—can be used, either alone or in combination with surgery (25). The presence of N3 lymph node metastasis or distant organ metastasis renders the disease inoperable and requires medical treatment. Clinically, we have observed that many patients with ¹⁸F-FDG PET/CT-identified small primary lesions also present with distant metastases.

Screening potential biomarkers of NSCLC metastasis may help to predict high-risk patients who are susceptible to metastasis. The study’s core aim was to examine the primary tumor’s metabolic features. Tissue biopsy or random sampling cannot fully reflect tumor heterogeneity. Therefore, the non-invasive assessment of tumor heterogeneity is valuable for predicting survival and guiding intensive therapy (26). Accordingly, the potential utility of tumor ¹⁸F-FDG heterogeneity as a surrogate imaging marker for tumor heterogeneity is receiving increasing interest.¹⁸F-FDG PET/CT is a well-established non-invasive imaging modality for evaluating lung cancer. In PET/CT imaging, the most commonly used features to assess tumor metabolic activity are SUVmax, SUVmean, MTV, and TLG (27). However, regional ¹⁸F-FDG accumulation can be significantly influenced by local pathophysiological processes such as hypoxia, angiogenesis, proliferation, and cell death. Thus, heterogeneity of ¹⁸F-FDG uptake may partially reflect these diverse components within the tumor tissue (26). Intratumoral histopathologic heterogeneity has been well correlated with increased tumor aggressiveness (28). The divergent ¹⁸F-FDG avidity of tumor components and cancer cell phenotypes—as well as local physiological conditions—may explain the observed association between histopathologic heterogeneity and the distribution of ¹⁸F-FDG uptake within tumors. Intratumoral metabolic heterogeneity assessed by ¹⁸F-FDG PET/CT has been shown to correlate with histopathologic heterogeneity in malignant tumors, including NSCLC. Studies have evaluated the utility of metabolic heterogeneity, as measured by ¹⁸F-FDG PET/CT, in primary NSCLC with clinically suspected N2-stage disease for predicting pathological mediastinal lymph node metastasis. Combining conventional imaging parameters, the COV of the primary tumor can enhance the accurate detection of N2 metastasis. The COV of the primary tumor may serve a complementary role to conventional imaging in providing nodal information prior to biopsy (25). This study selected the COV as the core indicator for its strengths in metabolic heterogeneity assessment and clinical practicality, aiming to identify a simple, reliable marker for high distant metastasis risk in T1-stage NSCLC. COV quantifies tumor metabolic distribution dispersion, directly reflecting intratumoral FDG uptake inhomogeneity—unlike traditional parameters (SUVmax, SUVmean, MTV, TLG) that focus only on overall metabolic intensity/volume, it specifically captures intratumoral differential characteristics. Notably, COV is simple and readily accessible: derived directly from SUV standard deviation and mean via routine ¹⁸F-FDG PET/CT (no complex post-processing, algorithms, software, or extra tests/radiation), ensuring high patient compliance. Calculated from objective SUV values, it is less affected by observer subjectivity (jointly verified by two radiologists), meets clinical reliability requirements with strong reproducibility, and is suitable for routine clinical application. Adopting a tiered SUVmax threshold system of 1.0, 1.5, 2.0, and 2.5 boasts prominent clinical and diagnostic advantages in PET/CT quantitative analysis: it abandons the limitations of a one-size-fits-all single cutoff, aligns with the continuous gradient of tissue glucose metabolism from normal/benign to malignant states, and enables precise gradated risk stratification of lesions instead of a simple binary normal/abnormal classification, thus facilitating targeted and individualized clinical management and follow-up strategy formulation. This tiered approach enhances diagnostic flexibility and accuracy by adapting to the metabolic characteristics of different lesion types and clinical scenarios, effectively reducing the rates of misdiagnosis and missed diagnosis; it also forms a reasonable buffer zone to accommodate the inherent 10–15% variability of SUVmax caused by equipment, scanning parameters, and individual patient factors, improving the practicality and comparability of clinical application. Moreover, it can accurately capture the dynamic changes of tumor glucose metabolism during treatment, serving as a more objective quantitative index for evaluating therapeutic efficacy, and further avoids medical over-treatment for low-risk benign lesions and delayed diagnosis and treatment for medium- and high-risk malignant lesions, which is more in line with the actual clinical diagnosis and treatment needs and provides a standardized quantitative basis for multicenter collaborative diagnosis and dynamic follow-up of lesions.

In this study, we investigated ¹⁸F-FDG heterogeneity as an independent marker for metastasis. We tested the value of the COV as a parameter of heterogeneous ¹⁸F-FDG uptake for determining metastasis risk. In small-size NSCLC patients, lung lesions with greater ¹⁸F-FDG heterogeneity, as indicated by a high COV2.0 (>0.33), were identified as a significant and independent risk factor for metastatic involvement. Although COV may not be the most sophisticated or precise parameter for assessing tissue heterogeneity, it is widely accessible and easily applicable in routine clinical practice without specialized software. Thus, COV may be routinely tested as an ¹⁸F-FDG PET/CT parameter to help evaluate metastasis risk in small-size NSCLC.

Oncogenic drivers, hypoxia, and metabolic alterations can trigger pro-inflammatory signals that promote an inflammatory tumor microenvironment (TME), contributing to tumor growth through immunological suppression (29,30). Beyond local inflammation, tumor-associated systemic inflammatory reactions also contribute to tumor progression (30). Mechanically, neutrophils can activate both endothelial and parenchymal cells to enhance circulating tumor cell adhesion, thereby promoting distant metastasis (31). Platelets shield circulating tumor cells from immune attack (32), induce epithelial-mesenchymal transition (EMT), and facilitate metastasis (33). Lymphocytes play an important role in immune surveillance and defense (34). Therefore, inflammation-related indexes such as the NLR, MLR, and PLR have been investigated in many studies associated with tumor progression, and have been shown to be crucial for malignant progression of many cancer types. We previously found that pretreatment serum NLR and MLR in thymic epithelial tumor (TET) patients are tumor-progression and tumor-glycolysis-related inflammatory markers. Other researchers have reported that inflammation-based factors are associated with aggressive tumor characteristics in various cancers (35). SII, which combines neutrophils, platelets, and lymphocytes (36,37), has been identified as a strong prognostic biomarker for predicting outcomes in operable NSCLC as well as associating with chemoradiation resistance of locally advanced NSCLC (38,39). In this study, although we did not observe any significant differences in serum monocyte or lymphocyte counts or MLR between patients with small-size NSCLC with and without metastasis, we found that NLR, PLR, and SII were all significantly higher in the metastasis group than in the non-metastasis group. We also demonstrated that pretreatment serum SII is a systemic inflammatory marker associated with tumor glycolytic activity. Here, we found SII associates with high COV in small-size NSCLC, and COV/SII co-high status is an independent risk factor for metastasis. These findings further underscore the importance of glycolysis-associated systemic inflammatory responses in lung cancer progression.

Glucose is one of the most important metabolic substrates for tumor cells, as most tumors exhibit increased ¹⁸F-FDG uptake when ¹⁸F-FDG PET scanning. The radical glucose uptake and its fermentation to lactate of cancer cells alters the metabolic status of immune cells in the TME. Inflammatory mediators produced by tumors or innate immune responses can suppress specific antitumor immune mechanisms. Further, EMT, which can be triggered by inflammatory cytokines in TME, has been well-validated as one critical mechanism for cancer metastasis by providing cells with migratory, invasive, and immune-evasive properties (40). The above information explains the positive correlation between co-high COV2.0/SII and metastasis observed in this study.

Our study also explored the potential correlations between PD-L1 expression, common gene mutations, and the metastasis, metabolic parameters and inflammatory markers of small-sized NSCLC. Among the 243 patients included in the study, 168 underwent PD-L1 testing and 189 underwent gene mutation testing. Statistical analysis revealed that there was no significant correlation between PD-L1 expression levels, gene mutations, and patient characteristics such as age, gender, smoking history, or pathological subtype. Furthermore, no significant association with the risk of metastasis was found in this group.

A deeper understanding of the interplay between glucose metabolism, inflammation, and immune suppression holds substantial promise for advancing integrated biomarker research in oncology. In this study, we identified three clinically valuable non-invasive biomarkers, PET/CT-derived COV2.0 (>0.33), serum SII (>641.68), and their co-high status—that enable precise risk stratification of distant metastasis in patients with early-stage small-size NSCLC (tumor size <3 cm, T1 stage). Our key finding confirmed that the co-high status of COV2.0 and SII represents an independent risk factor for metastasis in small-size NSCLC. This observation has direct clinical implications: for T1 NSCLC patients with co-high COV2.0 and SII, more aggressive therapeutic interventions combined with intensified follow-up surveillance are warranted to facilitate the early detection of occult metastasis and ultimately improve overall patient prognosis. Notably, the identified correlations between COV2.0 and SII, as well as between their co-high status and NSCLC metastatic potential, provide critical preliminary evidence to justify prospective investigations into the underlying pathophysiological and immunological mechanisms. Translating these findings into clinical practice, these readily measurable biomarkers address a long-standing unmet need in T1 NSCLC management—the accurate identification of high-risk patients—by enabling clinicians to reliably distinguish individuals at elevated metastatic risk from those at low risk. Beyond risk stratification, this biomarker panel offers a robust foundation for individualized patient management. High-risk patients may derive significant benefit from intensified upfront therapies, including neoadjuvant chemotherapy, targeted therapy, or immune checkpoint inhibitors, to mitigate metastatic potential. Conversely, low-risk patients are ideal candidates for less invasive treatment strategies, thereby minimizing overtreatment-related morbidity and healthcare burden. Furthermore, these biomarkers inform tailored follow-up protocols: close surveillance for high-risk cohorts to enable early metastasis detection, and streamlined monitoring for low-risk groups to reduce unnecessary healthcare utilization. Collectively, these insights complement traditional TNM staging systems and significantly enhance the precision of clinical decision-making in the management of early-stage NSCLC.

Our study has certain limitations. First, it was conducted retrospectively and relied on existing medical records and imaging data. As a single-center retrospective study, selection bias was inevitable. Multicenter, prospective collaborations are needed to confirm these results. Second, most patients in our cohort had adenocarcinoma, with few cases of squamous cell carcinoma. This imbalance may have affected the statistical comparison of metastasis frequency across histological subtypes.

Conclusions

Our study pioneers the identification of a synergistic “metabolic heterogeneity-systemic inflammation” axis driving distant metastasis in small-size NSCLC. Enhanced glycolysis, along with a corresponding systemic immune response, plays a critical role in NSCLC metastasis. COV2.0 (>0.33) and SII (>641.68) are independent risk factors, with their co-high status showing the strongest predictive power (HR =5.08). The combination of metabolic profiles from ¹⁸F-FDG PET/CT and serum inflammation factors may assist in evaluating metastasis risk in small-size NSCLC, thereby informing more rational treatment strategies for lung cancer.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1701/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1701/dss

Funding: This work was supported by the National Natural Science Foundation of China (Nos. 82372007, 82272044, 82202219, and 82302239), the Explorer Program of Shanghai (No. 23TS1400900), Natural Science Foundation of Shanghai (No. 24ZR1464100), and Science and Technology Commission of Shanghai Municipality Experimental Animal Research Fund (No. 24141900500).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1701/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 Institutional Review Board of Shanghai Chest Hospital. Informed consent was taken from all the patients.

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. Thai AA, Solomon BJ, Sequist LV, Gainor JF, Heist RS. Lung cancer. Lancet 2021;398:535-54. [Crossref] [PubMed]
2. Howlader N, Forjaz G, Mooradian MJ, Meza R, Kong CY, Cronin KA, Mariotto AB, Lowy DR, Feuer EJ. The Effect of Advances in Lung-Cancer Treatment on Population Mortality. N Engl J Med 2020;383:640-9. [Crossref] [PubMed]
3. Woodard GA, Li A, Boffa DJ. Role of adjuvant therapy in T1-2N0 resected non-small cell lung cancer. J Thorac Cardiovasc Surg 2022;163:1685-92. [Crossref] [PubMed]
4. Mei W, Yao W, Song Z, Jiao W, Zhu L, Huang Q, An C, Shi J, Yu G, Sun P, Zhang Y, Shen J, Xu C, Yang H, Wang Q, Zhu Z. Development and validation of prognostic nomogram for T(1-3)N(0)M(0) non-small cell lung cancer after curative resection. BMC Cancer 2023;23:715. [Crossref] [PubMed]
5. Kużdżał B, Moszczyński K, Żanowska K, Hauer J, Popovchenko S, Bryndza M, Warmus J, Trybalski Ł, Rudnicka L, Kocoń P. Correlation between 18-FDG standardized uptake value and tumor grade in patients with resectable non-small cell lung cancer. Transl Cancer Res 2023;12:3530-7. [Crossref] [PubMed]
6. Jung KJ, Lee KS, Kim H, Kwon OJ, Kim J, Shim YM, Kim TS. T1 lung cancer on CT: frequency of extrathoracic metastases. J Comput Assist Tomogr 2000;24:711-8. [Crossref] [PubMed]
7. Collon T, Ba O, Grivaux M, Dore P, Azarian R, Orion B, Boyer J, Raffy O, Jourdain B, Beraud A, Paillot N, Jouveshomme S, Mordacque C, Zureik M, Marsal L, Piquet J, Blanchon F. Le cancer bronchique primitif non à petites cellules: analyse des 419 cas de tumeurs T1 (≤ 3 cm) de l’étude KBP-2000-CPHG. Revue de Pneumologie Clinique 2004;60:333-43. [Crossref] [PubMed]
8. Kandathil A, Subramaniam RM. FDG PET/CT for Primary Staging of Lung Cancer and Mesothelioma. Semin Nucl Med 2022;52:650-61. [Crossref] [PubMed]
9. Expert Panel on Thoracic Imaging. de Groot PM, Chung JH, Ackman JB, Berry MF, Carter BW, Colletti PM, Hobbs SB, McComb BL, Movsas B, Tong BC, Walker CM, Yom SS, Kanne JP. ACR Appropriateness Criteria® Noninvasive Clinical Staging of Primary Lung Cancer. J Am Coll Radiol 2019;16:S184-S195.
10. Léger MA, Routy B, Juneau D. FDG PET/CT for Evaluation of Immunotherapy Response in Lung Cancer Patients. Semin Nucl Med 2022;52:707-19. [Crossref] [PubMed]
11. Eze C, Schmidt-Hegemann NS, Sawicki LM, Kirchner J, Roengvoraphoj O, Käsmann L, Mittlmeier LM, Kunz WG, Tufman A, Dinkel J, Ricke J, Belka C, Manapov F, Unterrainer M. PET/CT imaging for evaluation of multimodal treatment efficacy and toxicity in advanced NSCLC-current state and future directions. Eur J Nucl Med Mol Imaging 2021;48:3975-89. [Crossref] [PubMed]
12. Nestle U, Schimek-Jasch T, Kremp S, Schaefer-Schuler A, Mix M, Küsters A, et al. Imaging-based target volume reduction in chemoradiotherapy for locally advanced non-small-cell lung cancer (PET-Plan): a multicentre, open-label, randomised, controlled trial. Lancet Oncol 2020;21:581-92. [Crossref] [PubMed]
13. Diakos CI, Charles KA, McMillan DC, Clarke SJ. Cancer-related inflammation and treatment effectiveness. Lancet Oncol 2014;15:e493-503. [Crossref] [PubMed]
14. Liu J, Liu X, Li Y, Quan J, Wei S, An S, Yang R, Liu J. The association of neutrophil to lymphocyte ratio, mean platelet volume, and platelet distribution width with diabetic retinopathy and nephropathy: a meta-analysis. Biosci Rep 2018;38:BSR20180172. [Crossref] [PubMed]
15. Wu CC, Wu CH, Lee CH, Cheng CI. Association between neutrophil percentage-to-albumin ratio (NPAR), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR) and long-term mortality in community-dwelling adults with heart failure: evidence from US NHANES 2005-2016. BMC Cardiovasc Disord 2023;23:312. [Crossref] [PubMed]
16. Stone RL, Nick AM, McNeish IA, Balkwill F, Han HD, Bottsford-Miller J, et al. Paraneoplastic thrombocytosis in ovarian cancer. N Engl J Med 2012;366:610-8. [Crossref] [PubMed]
17. Bayraktaroglu M, Yildiz BP. Prognostic significance of neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio in non-small cell lung cancer. Medicine (Baltimore) 2023;102:e34180. [Crossref] [PubMed]
18. Gu XB, Tian T, Tian XJ, Zhang XJ. Prognostic significance of neutrophil-to-lymphocyte ratio in non-small cell lung cancer: a meta-analysis. Sci Rep 2015;5:12493. [Crossref] [PubMed]
19. Wang L, Ruan M, Yan H, Lei B, Sun X, Chang C, Liu L, Xie W. Pretreatment serum neutrophil-to-lymphocyte and monocyte-to-lymphocyte ratios: Two tumor-related systemic inflammatory markers in patients with thymic epithelial tumors. Cytokine 2020;133:155149. [Crossref] [PubMed]
20. Xu J, Li Y, Hu S, Lu L, Gao Z, Yuan H. The significant value of predicting prognosis in patients with colorectal cancer using (18)F-FDG PET metabolic parameters of primary tumors and hematological parameters. Ann Nucl Med 2019;33:32-8. [Crossref] [PubMed]
21. Wang L, Ruan M, Lei B, Yan H, Sun X, Chang C, Liu L, Xie W. The potential of (18)F-FDG PET/CT in predicting PDL1 expression status in pulmonary lesions of untreated stage IIIB-IV non-small-cell lung cancer. Lung Cancer 2020;150:44-52. [Crossref] [PubMed]
22. Lim W, Ridge CA, Nicholson AG, Mirsadraee S. The 8(th) lung cancer TNM classification and clinical staging system: review of the changes and clinical implications. Quant Imaging Med Surg 2018;8:709-18. [Crossref] [PubMed]
23. Wang J, Welch K, Wang L, Kong FM. Negative predictive value of positron emission tomography and computed tomography for stage T1-2N0 non-small-cell lung cancer: a meta-analysis. Clin Lung Cancer 2012;13:81-9. [Crossref] [PubMed]
24. Ding N, Mao Y, Gao S, Xue Q, Wang D, Zhao J, Gao Y, Huang J, Shao K, Feng F, Zhao Y, Yuan L. Predictors of lymph node metastasis and possible selective lymph node dissection in clinical stage IA non-small cell lung cancer. J Thorac Dis 2018;10:4061-8. [Crossref] [PubMed]
25. Pahk K, Chung JH, Yi E, Kim S, Lee SH. Metabolic tumor heterogeneity analysis by F-18 FDG PET/CT predicts mediastinal lymph node metastasis in non-small cell lung cancer patients with clinically suspected N2. Eur J Radiol 2018;106:145-9. [Crossref] [PubMed]
26. Pellegrino S, Fonti R, Hakkak Moghadam Torbati A, Bologna R, Morra R, Damiano V, Matano E, De Placido S, Del Vecchio S. Heterogeneity of Glycolytic Phenotype Determined by (18)F-FDG PET/CT Using Coefficient of Variation in Patients with Advanced Non-Small Cell Lung Cancer. Diagnostics (Basel) 2023;13:2448. [Crossref] [PubMed]
27. Vokes EE, Govindan R, Iscoe N, Hossain AM, San Antonio B, Chouaki N, Koczywas M, Senan S. The Impact of Staging by Positron-Emission Tomography on Overall Survival and Progression-Free Survival in Patients With Locally Advanced NSCLC. J Thorac Oncol 2018;13:1183-8. [Crossref] [PubMed]
28. Son SH, Kim DH, Hong CM, Kim CY, Jeong SY, Lee SW, Lee J, Ahn BC. Prognostic implication of intratumoral metabolic heterogeneity in invasive ductal carcinoma of the breast. BMC Cancer 2014;14:585. [Crossref] [PubMed]
29. Grivennikov SI, Greten FR, Karin M. Immunity, inflammation, and cancer. Cell 2010;140:883-99. [Crossref] [PubMed]
30. Nakamura K, Smyth MJ. Targeting cancer-related inflammation in the era of immunotherapy. Immunol Cell Biol 2017;95:325-32. [Crossref] [PubMed]
31. Liu J, Li S, Zhang S, Liu Y, Ma L, Zhu J, Xin Y, Wang Y, Yang C, Cheng Y. Systemic immune-inflammation index, neutrophil-to-lymphocyte ratio, platelet-to-lymphocyte ratio can predict clinical outcomes in patients with metastatic non-small-cell lung cancer treated with nivolumab. J Clin Lab Anal 2019;33:e22964. [Crossref] [PubMed]
32. Li N, Yu Z, Zhang X, Liu T, Sun YX, Wang RT, Yu KJ. Elevated mean platelet volume predicts poor prognosis in colorectal cancer. Sci Rep 2017;7:10261. [Crossref] [PubMed]
33. Stanger BZ, Kahn ML. Platelets and tumor cells: a new form of border control. Cancer Cell 2013;24:9-11. [Crossref] [PubMed]
34. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflammation. Nature 2008;454:436-44. [Crossref] [PubMed]
35. Deng Q, He B, Liu X, Yue J, Ying H, Pan Y, Sun H, Chen J, Wang F, Gao T, Zhang L, Wang S. Prognostic value of pre-operative inflammatory response biomarkers in gastric cancer patients and the construction of a predictive model. J Transl Med 2015;13:66. [Crossref] [PubMed]
36. Guo W, Cai S, Zhang F, Shao F, Zhang G, Zhou Y, Zhao L, Tan F, Gao S, He J. Systemic immune-inflammation index (SII) is useful to predict survival outcomes in patients with surgically resected non-small cell lung cancer. Thorac Cancer 2019;10:761-8. [Crossref] [PubMed]
37. Dirican N, Dirican A, Anar C, Atalay S, Ozturk O, Bircan A, Akkaya A, Cakir M. A New Inflammatory Prognostic Index, Based on C-reactive Protein, the Neutrophil to Lymphocyte Ratio and Serum Albumin is Useful for Predicting Prognosis in Non-Small Cell Lung Cancer Cases. Asian Pac J Cancer Prev 2016;17:5101-6. [PubMed]
38. Tong YS, Tan J, Zhou XL, Song YQ, Song YJ. Systemic immune-inflammation index predicting chemoradiation resistance and poor outcome in patients with stage III non-small cell lung cancer. J Transl Med 2017;15:221. [Crossref] [PubMed]
39. Gao Y, Zhang H, Li Y, Wang D, Ma Y, Chen Q. Preoperative increased systemic immune-inflammation index predicts poor prognosis in patients with operable non-small cell lung cancer. Clin Chim Acta 2018;484:272-7. [Crossref] [PubMed]
40. Li R, Ong SL, Tran LM, Jing Z, Liu B, Park SJ, Huang ZL, Walser TC, Heinrich EL, Lee G, Salehi-Rad R, Crosson WP, Pagano PC, Paul MK, Xu S, Herschman H, Krysan K, Dubinett S. Chronic IL-1β-induced inflammation regulates epithelial-to-mesenchymal transition memory phenotypes via epigenetic modifications in non-small cell lung cancer. Sci Rep 2020;10:377. Erratum in: Sci Rep 2020;10:4386. [Crossref] [PubMed]