Nonlinear association between standardized uptake value (SUV) index based on ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography/computed tomography (PET/CT) and the new International Association for the Study of Lung Cancer (IASLC) grading of lung invasive non-mucinous adenocarcinoma

Introduction

Lung cancer remains the leading cause of cancer-related mortality in many countries and ranks first in both incidence and death in China (1-3). Among its histological subtypes, lung adenocarcinoma is the most prevalent and is characterized by marked heterogeneity and generally poor prognosis (4). Invasive non-mucinous adenocarcinoma (INMA) constitutes approximately more than 90% of all lung adenocarcinoma cases (5). The previous histologic grading system, as defined in the 2015 World Health Organization (WHO) classification of thoracic tumors, categorized INMA into three differentiation levels based solely on the predominant histologic growth pattern: well-differentiated (lepidic predominant), moderately differentiated (acinar or papillary predominant), and poorly differentiated (solid or micropapillary predominant) (6). Although this system provides some prognostic value, its reliance on the dominant growth pattern alone, without accounting for minor high-grade components, limits its predictive accuracy (7). Multiple studies have shown that even a small proportion of high-grade patterns, such as micropapillary or solid components, is associated with poorer clinical outcomes (8,9). To address these limitations, the Pathology Committee of the International Association for the Study of Lung Cancer (IASLC) proposed a new grading system for lung INMA in 2020 (10). This system incorporates both the predominant histologic subtype and the proportion of high-grade patterns (10,11), classifying lung INMA into three grades: Grade 1 (well-differentiated), predominantly lepidic with no or less than 20% high-grade components; Grade 2 (moderately differentiated), predominantly acinar or papillary with no or less than 20% high-grade components; and Grade 3 (poorly differentiated), any histological pattern with 20% or more high-grade components. The high-grade components include solid, micropapillary, and complex glandular patterns (i.e., cribriform, fused glands, or single cells infiltrating desmoplastic stroma). Studies have demonstrated that compared with traditional grading system based solely on the dominant growth pattern, the IASLC grading significantly improves the accuracy of prognostic stratification (12). Moreover, determining tumor grading preoperatively is essential for guiding therapeutic decision-making and improving clinical outcomes in patients with lung INMA (13,14). The updated IASLC grading system has been officially adopted by the 2021 edition of the WHO classification of thoracic tumors (5). Therefore, early identification of IASLC grading system is critical for enabling personalized treatment strategies and improving prognosis of patients with lung INMA.

Currently, definitive grading of lung INMA can only be achieved through complete histological sampling via surgical resection. However, preoperative invasive procedures such as needle biopsy or bronchoscopy often yield limited tissue, making it difficult to comprehensively assess the full growth pattern of the tumor. Moreover, these procedures carry inherent risks of complications (15). As such, the development of preoperative non-invasive imaging methods to make up for the shortcomings of traditional invasive methods is crucial for refining diagnostic strategies and optimize surgical decision-making. ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) positron emission tomography/computed tomography (PET/CT) is a well-established imaging modality that combines anatomical and functional data, allowing for the visualization of tumor morphology via computed tomography (CT) and metabolic activity via PET. This dual-capability renders PET/CT uniquely advantageous and has led to its widespread application in the diagnosis, tumor, node, metastasis (TNM) stage, and treatment response assessment of lung cancer (16,17). In our previous work, we have summarized the current landscape of non-invasive imaging techniques used to predict the new IASLC grading of lung adenocarcinomas (18). Multiple studies have highlighted the potential of ¹⁸F-FDG PET/CT in preoperatively estimating IASLC grading of lung INMA; however, the relationship between conventional PET metabolic parameters and IASLC grading remains poorly defined (19,20).

In this study, we retrospectively analyzed the clinical and imaging data of patients with histologically confirmed lung INMA who underwent preoperative ¹⁸F-FDG PET/CT, with particular focus on their FDG uptake characteristics. Recognizing the limitations of maximum standardized uptake value (SUV_max), which can be influenced by variables such as blood glucose level, body weight, and scan parameters, we focused on the SUV index. Compared with SUV_max, SUV index offers the advantage of reducing inter-individual variability by normalizing FDG uptake relative to hepatic uptake, which is less affected by body composition and metabolic variations. The normalization enhances reproducibility and comparability, especially in retrospective and longitudinal studies (21). Furthermore, the lesion-to-liver FDG ratio has proven more diagnostically robust than absolute SUV values in discriminating certain diseases (22).

This study systematically evaluated the association between SUV index and IASLC grading, and aimed to establish a reliable foundation for developing non-invasive preoperative prediction models. Ultimately, such models could enhance personalized treatment strategies and improve prognosis in patients with lung adenocarcinoma. The overall workflow of this study is shown in Figure 1. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2364/rc).

Figure 1 Overall workflow of this study. ¹⁸F-FDG, ¹⁸F-fluorodeoxyglucose; 3D, three-dimensional; CI, confidence interval; CT, computed tomography; CTR, consolidation to tumor ratio; INMA, invasive non-mucinous adenocarcinoma; LLL, left lower lobe; LUL, left upper lobe; MTV, metabolic tumor volume; OR, odds ratio; PET, positron emission tomography; PET/CT, positron emission tomography/computed tomography; RLL, right lower lobe; RML, right middle lobe; RUL, right upper lobe; SUV, standardized uptake value; SUVmax, maximum standardized uptake value.

Methods

General data

We retrospectively collected the clinical and imaging data of 1,046 patients who underwent ¹⁸F-FDG PET/CT at the Department of Nuclear Medicine, The Third Affiliated Hospital of Soochow University, between January 2018 and June 2023 and were pathologically confirmed to have lung adenocarcinoma following surgical resection. The inclusion criteria were as follows: (I) histopathological confirmation of lung INMA; (II) completion of ¹⁸F-FDG PET/CT imaging within 30 days prior to surgery; (III) availability of complete pathological data for IASLC grading; and (IV) no prior history of other malignancies or cancer-related treatments. Patients were excluded if they met any of the following conditions: (I) poor image quality or lesions that could not be reliably assessed; (II) absence of breath-hold chest CT images acquired on the same scanner; (III) presence of more than two concurrent pulmonary lesions; or (IV) severe hepatic dysfunction, defined as serum alanine aminotransferase (ALT) or aspartate aminotransferase (AST) levels exceeding five times the upper limit of normal. For cases diagnosed before the release of the updated IASLC grading system, archived histopathological slides were retrieved and re-evaluated according to the current IASLC grading criteria. The re-assessment was performed by two experienced thoracic pathologists who were blinded to the imaging findings and clinical outcomes. Tumors were reclassified based on the predominant histological patterns and the proportion of high-grade components as defined by the new IASLC grading guidelines. In cases of discrepant grading, a consensus diagnosis was reached through joint review. After applying the above criteria, a total of 782 patients with lung INMA were included in the final analysis, comprising 324 males and 458 females, with a mean age of 64.1±9.1 years. Clinical data, including age, gender, smoking history, pathological findings, and TNM stage (according to the 9th edition of the Union for International Cancer Control classification) were collected. The patient selection flowchart is shown in Figure 2. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments and approved by the Institutional Ethics Committee of The Third Affiliated Hospital of Soochow University (approval No. [2018] K013). The requirement for informed consent was waived due to the retrospective nature of the study.

Figure 2 Flowchart of patient selection. ¹⁸F-FDG, ¹⁸F-fluorodeoxyglucose; INMA, invasive non-mucinous adenocarcinoma; PET/CT, positron emission tomography/computed tomography.

Image acquisition

All PET/CT examinations were performed using a Siemens Biograph mCT (64-slice) scanner (Siemens Healthineers, Erlangen, Germany). The imaging agent used was ¹⁸F-FDG, provided by Nanjing Jiangyuan Andico Positron Research & Development Co., Ltd. (Nanjing, China), with a radiochemical purity greater than 95%, and was intravenously administered at approximately 4.44 MBq/kg body weight. The scanning range extended from the skull to the mid-femur. In accordance with the European Association of Nuclear Medicine (EANM) guidelines version 2.0, patients fasted for 4–6 hours before the scan. On the day of the examination, height, weight, and blood glucose levels were recorded. Fasting blood glucose was required to be less than 11 mmol/L. PET/CT imaging was performed 55–75 minutes after intravenous injection of ¹⁸F-FDG. During the scan, patients were asked to lie in a supine position with arms raised above the head. A low-dose CT scan was performed first [tube voltage: 100 kV; tube current automatically adjusted using CareDose 4D (Siemens) based on body habitus and tissue density], followed by PET acquisition in three-dimensional (3D) mode at 2 minutes per bed position. After whole-body PET/CT imaging, a breath-hold chest CT was immediately acquired with thin-slice reconstruction at 1-mm thickness. All images were reconstructed using the TrueD workstation (Siemens Healthcare) with lung window settings [width: 1,200 Hounsfield units (HU); level: −600 HU] and mediastinal window settings (width: 350 HU; level: 40 HU). The TrueX + TOF (ultraHD-PET) method was used for image reconstruction (23). All PET/CT examinations were performed using consistent scanning and reconstruction parameters throughout the 2018–2023 study period.

Image analysis

PET/CT images were independently processed and analyzed by two experienced nuclear medicine physicians, each with more than 5 years of clinical experience. All tumor volume of interest (VOI) segmentations generated by the above two physicians were subsequently reviewed and corrected by a third physician with over 10 years of PET/CT diagnostic experience. PET and CT images were fused using ITK-SNAP (version 4.2, https://www.itksnap.org), and VOIs were manually delineated on the fused PET/CT images using 3D Slicer (version 4.11.2, http://www.slicer.org).

After image processing, PET metabolic features and CT quantitative features were extracted using the Pyradiomics module in Python (version 3.10.11), based on the two segmented VOIs. For each parameter, the average value from both VOIs was used in further analysis. CT morphological features were independently assessed by the two physicians in a blinded manner (without knowledge of the pathology results). In cases of discrepancy, a consensus was reached through discussion. PET metabolic features included SUV_max, metabolic tumor volume (MTV), total lesion glycolysis (TLG), and SUV index. MTV refers to the total volume of voxels (3D pixels) within a tumor that exhibit abnormal metabolic activity, typically defined by FDG uptake above a certain threshold. TLG is a composite metabolic parameter that integrates both MTV and the SUV_mean within that volume. It quantifies the total glycolytic activity of a metabolically active tumor lesion. SUV index is the ratio of tumor SUV_max to hepatic SUV_mean (SUV index = tumor SUV_max / hepatic SUV_mean). Hepatic SUV_mean was measured using an automatically generated region of interest (ROI) located 1 cm from the liver boundary in the right lobe by PET Liver Uptake Measurement module in 3D Slicer (24). CT quantitative features included maximum tumor diameter (D_max), lesion volume, mean CT value, and consolidation to tumor ratio (CTR). CTR was defined as the ratio of the maximum diameter of the solid (consolidation) component to the maximum overall tumor diameter on lung window images, measured on axial CT images. CT morphological features included lesion type (solid or subsolid), location (left upper lobe, left lower lobe, right upper lobe, right middle lobe, or right lower lobe), shape (round/quasi-round or irregular), margin (smooth or lobulated), spiculation, pleural indentation, vascular convergence, vacuolar sign, and air bronchogram as previously described in established radiological criteria for lung adenocarcinoma (4). To assess the reproducibility of VOI-based measurements, the intraclass correlation coefficient (ICC) was calculated for PET metabolic features and CT quantitative features. Parameters with ICC values greater than 0.75 were considered to have good agreement.

Statistical analysis

The 782 patients with lung INMA were divided into tertiles based on their SUV index values. Baseline characteristics were summarized by the group. Categorical variables were expressed as frequencies and percentages, normally distributed continuous variables as mean ± standard deviation, and non-normally distributed continuous variables as median and interquartile range. Differences among the three groups were assessed using the chi-square test for categorical variables, one-way analysis of variance (ANOVA) for normally distributed continuous variables, and the Kruskal-Wallis test for non-normally distributed continuous variables. Univariate logistic regression analysis was used to assess the association between different lesion features and the risk of Grade 3 lung INMA. Multivariate logistic regression analysis was then conducted to evaluate the independent and combined effects of the SUV index on the risk of Grade 3 lung INMA (a binary variable). Univariate and multivariable logistic regression analyses were performed with Grade 3 lung INMA as the outcome, using Grades 1–2 combined as the reference category. Three regression models were constructed: an unadjusted model (equivalent to the univariate analysis), a preliminarily adjusted model (Model I), and a fully adjusted model (Model II). Model I was adjusted for basic demographic variables selected a priori, including age, gender, and smoking history. Covariates were included in the Model II if they altered the estimated effect of the SUV index on the risk of Grade 3 lung INMA by more than 10% or were significantly associated with the outcome (P<0.1), as potential confounding factors. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. In the multivariate logistic regression analysis, to explore the relationship between the SUV index and Grade 3 lung INMA across different subgroups, stratified analyses were performed using the first tertile (Tertile 1) of the SUV index as the reference category. A generalized additive model (GAM) and smooth curve fitting were used to assess the relationship between SUV index and IASLC grading, and to determine whether there was a nonlinear relationship, whether there was a threshold or saturation effect, and whether a general linear regression model was appropriate. If a nonlinear relationship was detected, a segmented regression model was used to further explore the association. Interaction effects between subgroups were evaluated using stratified analysis. All statistical analyses were performed using R software (version 3.4.3; http://www.R-project.org). A P value <0.05 was considered statistically significant.

Results

Baseline characteristics

Among the 782 patients with lung INMA, 324 were male and 458 were female, with a mean age of 64.1±9.1 years. According to the IASLC grading, 58 cases (7.4%) were classified as Grade 1, 520 cases (66.5%) as Grade 2, and 204 cases (26.1%) as Grade 3. Based on the tertiles of the SUV index, the patients were divided into three groups: Tertile 1 (0.139–1.367, n=261), Tertile 2 (1.381–3.915, n=260), and Tertile 3 (3.919–12.841, n=261). Clinical and imaging characteristics across the three tertile groups are summarized in Table 1. Detailed variables across SUV index tertiles are provided in Table S1. No significant differences were observed among the groups with respect to age, lesion location, or shape (all P>0.05). In contrast, significant differences were found in gender, smoking history, TNM stage, lesion type, margin, spiculation, pleural indentation, vascular convergence, vacuolar sign, air bronchogram, mean CT value, D_max, CTR, lesion volume, MTV, and TLG (all P<0.05). The proportion of Grade 3 lung INMA increased significantly across SUV index tertiles, from 3.8% in Tertile 1 to 15.0% in Tertile 2 and 59.4% in Tertile 3 (P<0.05). Excellent interobserver agreement was observed for the PET and CT parameters extracted from manually segmented VOIs, with ICCs ranging from 0.921 to 0.996 (all P<0.001).

Univariate logistic regression analysis

The results of univariate logistic regression analysis of most variables for their association with IASLC grading (Grade 3 vs. Grades 1–2) are summarized in Table 2 (detailed variables are provided in Table S2). The results indicated that age, gender, smoking history, TNM stage, lesion type, margin, pleural indentation, vascular convergence, air bronchogram, mean CT value, D_max, CTR, MTV, TLG, and SUV index were all potential risk factors for Grade 3 lung INMA (OR range, 0.316–1,627.533; all P<0.1).

Multivariate logistic regression analysis of SUV index and IASLC grading

The results of both univariate and multivariate logistic regression analyses for continuous SUV index and SUV index tertiles are summarized in Table 3. The unadjusted model corresponded to univariate analysis; the preliminarily adjusted model (Model I) included covariates such as gender, age, and smoking history; the fully adjusted model (Model II) incorporated gender, age, smoking history, TNM stage, lesion type, margin, mean CT value, D_max, CTR, MTV, TLG, pleural indentation, vascular convergence, and air bronchogram. Across all models, unadjusted, Model I, and Model II, the risk of Grade 3 lung INMA increased significantly with a higher SUV index. The ORs were 1.960 (95% CI: 1.775–2.164, P<0.001), 1.964 (95% CI: 1.770–2.179, P<0.001), and 2.086 (95% CI: 1.728–2.517, P<0.001), respectively. Similarly, in the analysis using SUV index tertiles, a significant trend was observed in the increasing risk of Grade 3 lung INMA with higher tertile levels (all P<0.001, Table 3). In particular, when compared with Tertile 1, Tertile 3 showed a markedly increased risk of Grade 3 lung INMA, with ORs of 36.703 (unadjusted), 33.672 (Model I), and 9.545 (Model II), respectively (all P<0.001).

Curve fitting and threshold effect analysis

A GAM was applied to examine the relationship between the SUV index and Grade 3 lung INMA. After adjusting for confounding factors, including gender, age, smoking history, TNM stage, lesion type, margin, mean CT value, D_max, CTR, MTV, TLG, pleural indentation, vascular convergence, and air bronchogram (the same as the adjusted variables in Model II), a nonlinear association was observed between SUV index and IASLC grading of lung INMA. Specifically, the risk of Grade 3 lung INMA initially increased gradually with rising SUV index, then rose sharply, and finally plateaued, forming an approximate “S”-shaped curve (degrees of freedom: 2.488, P<0.001; Figure 3).

Figure 3 Relationship between SUV index and risk of Grade 3 lung INMA. The red solid line represents the fitted curve of the probability of Grade 3 lung INMA against the SUV index, whereas the blue dashed lines indicate the 95% CI (the small vertical lines just above the X-line represent the distribution of individual data points across the SUV index range). The model is adjusted for gender, age, smoking history, TNM stage, lesion type, margin, mean CT value, D_max, CTR, MTV, TLG, pleural indentation, vascular convergence, and air bronchogram. CI, confidence interval; CT, computed tomography; CTR, consolidation to tumor ratio; D_max, maximum tumor diameter; INMA, invasive non-mucinous adenocarcinoma; MTV, metabolic tumor volume; SUV, standardized uptake value; TLG, total lesion glycolysis; TNM, tumor, node, metastasis.

Further assessment using a segmented logistic regression model evaluated the presence of threshold or saturation effects (Table 4). The log-likelihood ratio test indicated a significant nonlinear relationship with two inflection points at SUV index values of 5.636 and 8.041 (P=0.001). When the SUV index was below 5.636, the risk of Grade 3 lung INMA increased modestly with a rising SUV index (OR =1.437; 95% CI: 1.204–1.715; P<0.001). Between 5.636 and 8.041, the risk increased markedly (OR =3.359; 95% CI: 1.730–6.521; P<0.001), demonstrating a classic threshold effect. When the SUV index exceeded 8.041, the association stabilized, with no significant further increase in risk (OR =1.020; 95% CI: 0.344–3.025; P=0.972).

When stratified by SUV index tertiles, the average probability of Grade 3 lung INMA increased progressively with rising SUV index levels. The mean probabilities of Grade 3 lung INMA for Tertile 1 to Tertile 3 were 3.8% (95% CI: 1.7–8.6%), 8.1% (95% CI: 4.4–14.5%), and 27.8% (95% CI: 17.1–41.8%), respectively (P<0.001). The highest average probability of Grade 3 lung INMA was observed in the Tertile 3 group (Figure 4). Representative cases are shown in Figure 5.

Figure 4 Relationship between SUV index tertiles and the risk of Grade 3 lung INMA. The black dashed line represents the fitted probability curve of Grade 3 lung INMA across SUV index tertile groups, with the red line indicating the 95% CI. Adjustments were made for gender, age, smoking history, TNM stage, lesion type, margin, mean CT value, D_max, CTR, MTV, TLG, pleural indentation, vascular convergence, and air bronchogram. CI, confidence interval; CT, computed tomography; CTR, consolidation to tumor ratio; D_max, maximum tumor diameter; INMA, invasive non-mucinous adenocarcinoma; MTV, metabolic tumor volume; SUV, standardized uptake value; TLG, total lesion glycolysis; TNM, tumor, node, metastasis.

Figure 5 Representative cases of lung INMA with different IASLC Grades. (A-C) A 74-year-old female with Grade 1 lung INMA: (A) CT lung window showing a 12-mm pure ground-glass nodule in the left lower lobe (arrow); (B,C) PET image and PET/CT fusion image of the lesion with an SUV index of 0.59. (D-F) A 50-year-old male with Grade 2 lung INMA: (D) CT lung window showing a 20-mm part-solid ground-glass nodule in the right upper lobe (arrow); (E,F) PET image and PET/CT fusion image with an SUV index of 4.46. (G-I) A 73-year-old male with Grade 3 lung INMA: (G) CT lung window showing a 23-mm solid nodule in the left upper lobe (arrow); (H,I) PET image and PET/CT fusion image with an SUV index of 7.60. (J-L) A 73-year-old male with Grade 3 lung INMA: (J) CT lung window showing a 34-mm solid mass in the right upper lobe (arrow); (K,L) PET image and PET/CT fusion image with an SUV index of 9.88. CT, computed tomography; IASLC, International Association for the Study of Lung Cancer; INMA, invasive non-mucinous adenocarcinoma; PET, positron emission tomography; SUV, standardized uptake value.

Interaction analysis

Stratified analysis was further conducted to explore whether interactions existed between the SUV index and various stratification factors on IASLC grading. The results demonstrated that lesion location, shape, volume, spiculation, and vacuolar sign did not significantly modify the association between SUV index and IASLC grading (all P>0.05) (Figure 6).

Figure 6 Stratified analysis of the association between SUV index and risk of Grade 3 Lung INMA. Adjusted for gender, age, smoking history, TNM stage, lesion type, margin, mean CT value, D_max, CTR, MTV, TLG, pleural indentation, vascular convergence, and air bronchogram. CI, confidence interval; CT, computed tomography; CTR, consolidation to tumor ratio; D_max, maximum tumor diameter; INMA, invasive non-mucinous adenocarcinoma; LLL, left lower lobe; LUL, left upper lobe; MTV, metabolic tumor volume; OR, odds ratio; RLL, right lower lobe; RML, right middle lobe; RUL, right upper lobe; SUV, standardized uptake value; TLG, total lesion glycolysis; TNM, tumor, node, metastasis.

Discussion

At present, ¹⁸F-FDG PET/CT has been widely applied in the diagnosis, staging, and therapeutic evaluation of lung cancer (16,17). Among various parameters, the SUV_max is a critical metric extensively used to quantify tumor glucose metabolism and has been employed to predict prognosis in non-small cell lung cancer (25). Studies by Kawaguchi et al. (20) and Jeon et al. (8) have both demonstrated that the SUV_max of IASLC Grade 3 lung INMA is significantly higher than that of Grades 1 and 2. Fujikawa et al. (14) have reported a significant increase of SUV_max with tumor grade and identified SUV_max as an independent predictor for Grade 3 lung INMA. However, SUV_max, as the most commonly used metabolic parameter in PET/CT, exhibits considerable inter-individual variability and is susceptible to confounding factors such as body weight, blood glucose levels, injected dose, uptake time, and scanner model. The SUV index, a more standardized form of SUV_max, is less influenced by individual patient variability and enhances consistency across multicenter studies. It has demonstrated promising predictive value in the diagnosis and assessment of various diseases (26,27). To investigate the association between SUV index and IASLC grading, this retrospective study analyzed data from lung INMA patients who underwent preoperative PET/CT scans.

Regarding clinical and CT features of lung INMA, our univariate logistic regression analysis revealed that gender, age, smoking history, pure solid lesions, mean CT value, D_max, CTR, lobulation sign, pleural indentation, vascular convergence, and air bronchogram were all independently associated with IASLC grading. These findings were consistent with previous reports. For instance, Volmonen et al. (19) observed a decrease in air bronchogram and an increase in CTR from Grade 1 to Grade 3 lung INMA. Liang et al. (28) similarly found that mean CT value, CTR, and tumor long axis increase with tumor grade. Kawaguchi et al. (20) and Fujikawa et al. (14) have reported that solid nodules are more common in Grade 3 lung INMA compared to Grades 1 and 2, in agreement with our results. As for the PET parameters, this study innovatively identified the SUV index as an independent factor associated with IASLC grading, highlighting its potential advantages in individualized metabolic evaluation. Additionally, TLG and MTV were also correlated with IASLC grading in univariate analyses. Yang et al. (29) similarly reported a close relationship between TLG and IASLC grading, aligning with our findings. However, their conclusion of no correlation between MTV and IASLC grading differs from ours. This study revealed that both MTV and TLG were associated with SUV index and also correlated with IASLC grading, suggesting that they may act as confounding factors between the two.

To control for confounding factors, this study further explored the association between the SUV index and IASLC grading through multivariate logistic regression analysis. After comprehensive adjustment for confounders, the SUV index was confirmed as an independent factor associated with IASLC grading. Specifically, the risk of Grade 3 lung INMA increased with a rising SUV index, independent of lesion location, shape, volume, spiculation, or vacuolar sign.

Further analyses using GAM and segmented regression revealed a nonlinear relationship between the SUV index and Grade 3 lung INMA risk, with a threshold effect. This suggests that the SUV index may provide a potential tool for non-invasive preoperative risk stratification and serve as an important predictor of tumor histological malignancy. Based on breakpoint analysis, we propose a hypothetical clinical decision framework. When the SUV index is <5.636, the tumor is likely in a moderately to highly differentiated stage, exhibiting relatively indolent metabolic characteristics with a lower Grade 3 risk. This suggests that these patients may have tumors with more favorable biological behavior, and future research could explore whether this applies to less invasive surgical approaches or more conservative follow-up strategies. Moreover, the threshold range of SUV index between 5.636 and 8.041 delineates a metabolic risk-alert zone that may directly inform several key aspects of clinical management. First, for lesions with SUV index falling within this interval, clinicians should consider prioritizing preoperative biopsy to confirm histological grade. Second, an SUV index in this range may support opting for more extensive resection (e.g., lobectomy over sublobar resection) even in early-stage disease, given the higher likelihood of Grade 3 lung INMA and its association with occult invasion or micrometastases (30,31). Third, the same metabolic alert zone should prompt systematic lymph node dissection rather than limited sampling, as Grade 3 lung INMA is more prone to nodal involvement (32). Finally, patients with SUV index in this critical range may benefit from intensified preoperative evaluation, including multidisciplinary tumor-board review, advanced imaging, and possibly neoadjuvant therapy trials in select settings. When the SUV index exceeds 8.041, the risk plateaus. Patients in this stage should be considered at extremely high risk, and more aggressive treatment and prognostic monitoring should be implemented to control their elevated metastasis risk.

Multiple studies have confirmed that high-grade growth patterns are closely associated with increased invasiveness and poorer prognosis (33). From a metabolic perspective, tumors exhibiting these patterns demonstrate markedly aberrant glucose metabolism, primarily due to the following mechanisms. First, FDG, a glucose analog, is taken up by tumor cells in proportion to the enhancement of the Warburg effect (34). Tumor cells with high-grade growth patterns display a pronounced Warburg effect characterized by elevated aerobic glycolysis (35). This phenomenon is tightly linked to the malignant phenotype of tumors and is molecularly driven by the overexpression of glucose transporter 1 (GLUT-1), which directly facilitates intracellular FDG uptake (36). The SUV index threshold observed in this study (5.636) may correspond to a critical level of GLUT-1 expression, which facilitates increased FDG uptake in Grade 3 lung INMA. The plateau observed beyond an SUV index of 8.041 may reflect biologic ceiling effects, including glycolytic inhibition related to tumor microenvironment acidosis, which has been shown to reduce glycolytic efficiency and limit further increases in FDG uptake (37). Additional factors, such as oxygen supply limitations and ATP production constraints, may also contribute to this plateau. Moreover, the elevated FDG uptake seen in high-grade patterns may partially reflect the biological link between tumor metabolic activity and circulating tumor cell dissemination, as metabolically active tumor subtypes often exhibit more aggressive invasive characteristics (38). IASLC studies have demonstrated that tumors containing 20% or more high-grade components behave similarly to tumors dominated by these patterns and are more invasive (10). Therefore, the significantly elevated SUV index in Grade 3 lung INMA essentially reflects the biological features of high-grade growth patterns, namely, enhanced proliferative capacity, pronounced metabolic reprogramming, and worse clinical prognosis. This association not only provides functional imaging evidence supporting the IASLC grading but also suggests that ¹⁸F-FDG PET/CT may hold considerable value in non-invasive preoperative assessment of tumor malignancy.

Despite methodological advances and a large sample size, several limitations remain in this study. (I) This was a single-center retrospective study; the potential enrichment of operable tumors and preferential inclusion of lesions more likely to undergo PET/CT may limit the generalizability of our findings and introduce selection bias across the broader clinical spectrum of lung INMA. (II) Although the overall sample size was substantial, cases of Grade 1 lung INMA were relatively scarce. Future studies should aim to increase the number of Grade 1 cases. (III) Although this study revealed a nonlinear relationship between the SUV index and IASLC grading, numerous confounding factors remain. Therefore, predicting IASLC grading still requires a comprehensive evaluation that integrates multiple clinical and imaging features. (IV) In recent years, artificial intelligence approaches, such as radiomics and deep learning, have emerged as promising tools for predicting IASLC grading (39). These methods offer a more comprehensive assessment of tumor heterogeneity and capture more complex nonlinear relationships compared to traditional statistical models. In future work, we plan to conduct studies incorporating radiomics and tumor microenvironment analyses.

Conclusions

This study demonstrated a nonlinear “S”-shaped association between the SUV index and IASLC grading, with the most pronounced risk increase observed within the 5.636–8.041 interval. This elucidation suggested that the SUV index might serve as a valuable metabolic parameter for non-invasive preoperative prognostic stratification. Future efforts combining additional imaging features and artificial intelligence techniques could further refine IASLC grading prediction models, ultimately providing more precise tools for personalized treatment and prognosis assessment in patients with lung INMA.

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-aw-2364/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2364/dss

Funding: This study was supported by Changzhou Clinical Medical Center (Nuclear Medicine) (No. CZZX202204); Top Talent of Changzhou “The 14th Five-Year Plan” High-Level Health Talents Training Project (Nos. 2022CZBJ037, 2024CZBJ008); Major Project of Changzhou Health Commission (No. ZD202405); and Postgraduate Research & Practice Innovation Program of Jiangsu Province (Nos. SJCX25_1794, SJCX25_1795).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-aw-2364/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.This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Institutional Ethics Committee of The Third Affiliated Hospital of Soochow University (approval No. [2018] K013), and the requirement for informed consent was waived due to the retrospective nature of the 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, Ferlay J, Siegel RL, Soerjomataram I, Jemal A. 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. Han B, Zheng R, Zeng H, Wang S, Sun K, Chen R, Li L, Wei W, He J. Cancer incidence and mortality in China, 2022. J Natl Cancer Cent 2024;4:47-53. [Crossref] [PubMed]
3. Siegel RL, Kratzer TB, Wagle NS, Sung H, Jemal A. Cancer statistics, 2026. CA Cancer J Clin 2026;76:e70043. [Crossref] [PubMed]
4. Travis WD, Brambilla E, Noguchi M, Nicholson AG, Geisinger KR, Yatabe Y, et al. International association for the study of lung cancer/american thoracic society/european respiratory society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol 2011;6:244-85. [Crossref] [PubMed]
5. Nicholson AG, Tsao MS, Beasley MB, Borczuk AC, Brambilla E, Cooper WA, Dacic S, Jain D, Kerr KM, Lantuejoul S, Noguchi M, Papotti M, Rekhtman N, Scagliotti G, van Schil P, Sholl L, Yatabe Y, Yoshida A, Travis WD. The 2021 WHO Classification of Lung Tumors: Impact of Advances Since 2015. J Thorac Oncol 2022;17:362-87. [Crossref] [PubMed]
6. Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JHM, Beasley MB, Chirieac LR, Dacic S, Duhig E, Flieder DB, Geisinger K, Hirsch FR, Ishikawa Y, Kerr KM, Noguchi M, Pelosi G, Powell CA, Tsao MS, Wistuba I, Panel WHO. The 2015 World Health Organization Classification of Lung Tumors: Impact of Genetic, Clinical and Radiologic Advances Since the 2004 Classification. J Thorac Oncol 2015;10:1243-60. [Crossref] [PubMed]
7. Wang W, Hu Z, Zhao J, Huang Y, Rao S, Yang J, Xiao S, Cao R, Ye L. Both the presence of a micropapillary component and the micropapillary predominant subtype predict poor prognosis after lung adenocarcinoma resection: a meta-analysis. J Cardiothorac Surg 2020;15:154. [Crossref] [PubMed]
8. Jeon HW, Kim YD, Sim SB, Moon MH. Significant difference in recurrence according to the proportion of high grade patterns in stage IA lung adenocarcinoma. Thorac Cancer 2021;12:1952-8. [Crossref] [PubMed]
9. Chang C, Sun X, Zhao W, Wang R, Qian X, Lei B, Wang L, Liu L, Ruan M, Xie W, Shen J. Minor components of micropapillary and solid subtypes in lung invasive adenocarcinoma (≤3 cm): PET/CT findings and correlations with lymph node metastasis. Radiol Med 2020;125:257-64. [Crossref] [PubMed]
10. Moreira AL, Ocampo PSS, Xia Y, Zhong H, Russell PA, Minami Y, et al. A Grading System for Invasive Pulmonary Adenocarcinoma: A Proposal From the International Association for the Study of Lung Cancer Pathology Committee. J Thorac Oncol 2020;15:1599-610. [Crossref] [PubMed]
11. Borczuk AC. Updates in grading and invasion assessment in lung adenocarcinoma. Mod Pathol 2022;35:28-35. [Crossref] [PubMed]
12. Yanagawa N, Sugai M, Shikanai S, Sugimoto R, Osakabe M, Uesugi N, Saito H, Maemondo M, Sugai T. The new IASLC grading system for invasive non-mucinous lung adenocarcinoma is a more useful indicator of patient survival compared with previous grading systems. J Surg Oncol 2023;127:174-82. [Crossref] [PubMed]
13. Hou L, Wang T, Chen D, She Y, Deng J, Yang M, Zhang Y, Zhao M, Zhong Y, Ma M, Zhao G, Chen Y, Xie D, Zhu Y, Chen Q, Wu C, Chen CMulti-omics Classifier for Pulmonary Nodules (MISSION) Collaborative Group. Prognostic and predictive value of the newly proposed grading system of invasive pulmonary adenocarcinoma in Chinese patients: a retrospective multicohort study. Mod Pathol 2022;35:749-56. [Crossref] [PubMed]
14. Fujikawa R, Muraoka Y, Kashima J, Yoshida Y, Ito K, Watanabe H, Kusumoto M, Watanabe SI, Yatabe Y. Clinicopathologic and Genotypic Features of Lung Adenocarcinoma Characterized by the International Association for the Study of Lung Cancer Grading System. J Thorac Oncol 2022;17:700-7. [Crossref] [PubMed]
15. Cataluña JJ, Perpiñá M, Greses JV, Calvo V, Padilla JD, París F. Cell type accuracy of bronchial biopsy specimens in primary lung cancer. Chest 1996;109:1199-203. [Crossref] [PubMed]
16. Groheux D, Quere G, Blanc E, Lemarignier C, Vercellino L, de Margerie-Mellon C, Merlet P, Querellou S. FDG PET-CT for solitary pulmonary nodule and lung cancer: Literature review. Diagn Interv Imaging 2016;97:1003-17. [Crossref] [PubMed]
17. Sibille L, Seifert R, Avramovic N, Vehren T, Spottiswoode B, Zuehlsdorff S, Schäfers M. (18)F-FDG PET/CT Uptake Classification in Lymphoma and Lung Cancer by Using Deep Convolutional Neural Networks. Radiology 2020;294:445-52. [Crossref] [PubMed]
18. Feng J, Shao X, Gao J, Ge X, Sun Y, Shi Y, Wang Y, Niu R. Application and progress of non-invasive imaging in predicting lung invasive non-mucinous adenocarcinoma under the new IASLC grading guidelines. Insights Imaging 2025;16:4. [Crossref] [PubMed]
19. Volmonen K, Sederholm A, Rönty M, Paajanen J, Knuuttila A, Jartti A. Association of CT findings with invasive subtypes and the new grading system of lung adenocarcinoma. Clin Radiol 2023;78:e251-9. [Crossref] [PubMed]
20. Kawaguchi M, Kato H, Hanamatsu Y, Suto T, Noda Y, Kaneko Y, Iwata H, Hyodo F, Miyazaki T, Matsuo M. Computed Tomography and (18)F-Fluorodeoxyglucose-Positron Emission Tomography/Computed Tomography Imaging Biomarkers of Lung Invasive Non-mucinous Adenocarcinoma: Prediction of Grade 3 Tumour Based on World Health Organization Grading System. Clin Oncol (R Coll Radiol) 2023;35:e601-10. [Crossref] [PubMed]
21. Rong N, Yuetao W, Xiaoliang S, Jianfeng W, Zhenxing J, Mei X, Yunmei S, Peiqi L, Xiaosong W, Xiaonan S. Correlation analysis between SUV index in 18F-FDG PET/CT imaging and invasiveness of early lung adenocarcinoma. Chin J Nucl Med Mol Imaging 2022;42:257-62.
22. Blake MA, Slattery JM, Kalra MK, Halpern EF, Fischman AJ, Mueller PR, Boland GW. Adrenal lesions: characterization with fused PET/CT image in patients with proved or suspected malignancy--initial experience. Radiology 2006;238:970-7. [Crossref] [PubMed]
23. Niu R, Shao X, Shao X, Jiang Z, Wang J, Wang Y. Establishment and verification of a prediction model based on clinical characteristics and positron emission tomography/computed tomography (PET/CT) parameters for distinguishing malignant from benign ground-glass nodules. Quant Imaging Med Surg 2021;11:1710-22. [Crossref] [PubMed]
24. Bauer C, Sun S, Sun W, Otis J, Wallace A, Smith BJ, Sunderland JJ, Graham MM, Sonka M, Buatti JM, Beichel RR. Automated measurement of uptake in cerebellum, liver, and aortic arch in full-body FDG PET/CT scans. Med Phys 2012;39:3112-23. [Crossref] [PubMed]
25. Paesmans M, Berghmans T, Dusart M, Garcia C, Hossein-Foucher C, Lafitte JJ, Mascaux C, Meert AP, Roelandts M, Scherpereel A, Terrones Munoz V, Sculier JP. European Lung Cancer Working Party, and on behalf of the IASLC Lung Cancer Staging Project. Primary tumor standardized uptake value measured on fluorodeoxyglucose positron emission tomography is of prognostic value for survival in non-small cell lung cancer: update of a systematic review and meta-analysis by the European Lung Cancer Working Party for the International Association for the Study of Lung Cancer Staging Project. J Thorac Oncol 2010;5:612-9. [Crossref] [PubMed]
26. Shi YM, Niu R, Shao XL, Zhang FF, Shao XN, Wang JF, Wang XS, Liu B, Yu WJ, Wang YT. Tumor-to-liver standard uptake ratio using fluorine-18 fluorodeoxyglucose positron emission tomography computed tomography effectively predict occult lymph node metastasis of non-small cell lung cancer patients. Nucl Med Commun 2020;41:459-68. [Crossref] [PubMed]
27. Carretta A, Bandiera A, Muriana P, Viscardi S, Ciriaco P, Gajate AMS, Arrigoni G, Lazzari C, Gregorc V, Negri G. Prognostic role of positron emission tomography and computed tomography parameters in stage I lung adenocarcinoma. Radiol Oncol 2020;54:278-84. [Crossref] [PubMed]
28. Liang M, Tang W, Tan F, Zeng H, Guo C, Feng F, Wu N. Preoperative prognostic prediction for stage I lung adenocarcinomas: Impact of the computed tomography features associated with the new histological grading system. Front Oncol 2023;13:1103269. [Crossref] [PubMed]
29. Yang H, Liu X, Wang L, Zhou W, Tian Y, Dong Y, Zhou K, Chen L, Wang M, Wu H. 18 F-FDG PET/CT characteristics of IASLC grade 3 invasive adenocarcinoma and the value of 18 F-FDG PET/CT for preoperative prediction: a new prognostication model. Nucl Med Commun 2024;45:338-46. [Crossref] [PubMed]
30. Hegedűs F, Zombori-Tóth N, Kiss S, Lantos T, Zombori T. Prognostic impact of the IASLC grading system of lung adenocarcinoma: a systematic review and meta-analysis. Histopathology 2024;85:51-61. [Crossref] [PubMed]
31. Wang K, Liu X, Ding Y, Sun S, Li J, Geng H, Xu M, Wang M, Li X, Sun D. A pretreatment prediction model of grade 3 tumors classed by the IASLC grading system in lung adenocarcinoma. BMC Pulm Med 2023;23:377. [Crossref] [PubMed]
32. Sun W, Su H, Liu J, Zhang L, Li M, Xie H, Xu L, Zhao S, She Y, Tang H, Wu C, Ke H, Chen CSurgical Thoracic Alliance of Rising Star Group. Impact of histological components on selecting limited lymphadenectomy for lung adenocarcinoma ≤2 cm. Lung Cancer 2020;150:36-43. [Crossref] [PubMed]
33. Qian F, Yang W, Wang R, Xu J, Wang S, Zhang Y, Jin B, Yu K, Han B. Prognostic significance and adjuvant chemotherapy survival benefits of a solid or micropapillary pattern in patients with resected stage IB lung adenocarcinoma. J Thorac Cardiovasc Surg 2018;155:1227-35.e2. [Crossref] [PubMed]
34. WARBURG O. On respiratory impairment in cancer cells. Science 1956;124:269-70.
35. Bu L, Tu N, Wang K, Zhou Y, Xie X, Han X, Lin H, Feng H. Relationship between (18)F-FDG PET/CT Semi-Quantitative Parameters and International Association for the Study of Lung Cancer, American Thoracic Society/European Respiratory Society Classification in Lung Adenocarcinomas. Korean J Radiol 2022;23:112-23. [Crossref] [PubMed]
36. de Geus-Oei LF, van Krieken JH, Aliredjo RP, Krabbe PF, Frielink C, Verhagen AF, Boerman OC, Oyen WJ. Biological correlates of FDG uptake in non-small cell lung cancer. Lung Cancer 2007;55:79-87. [Crossref] [PubMed]
37. Zhao J, Jin D, Huang M, Ji J, Xu X, Wang F, Zhou L, Bao B, Jiang F, Xu W, Lu X, Xiao M. Glycolysis in the tumor microenvironment: a driver of cancer progression and a promising therapeutic target. Front Cell Dev Biol 2024;12:1416472. [Crossref] [PubMed]
38. Nair VS, Keu KV, Luttgen MS, Kolatkar A, Vasanawala M, Kuschner W, Bethel K, Iagaru AH, Hoh C, Shrager JB, Loo BW Jr, Bazhenova L, Nieva J, Gambhir SS, Kuhn P. An observational study of circulating tumor cells and (18)F-FDG PET uptake in patients with treatment-naive non-small cell lung cancer. PLoS One 2013;8:e67733. [Crossref] [PubMed]
39. Zhong Y, Cai C, Chen T, Gui H, Chen C, Deng J, Yang M, Yu B, Song Y, Wang T, Chen Y, Shi H, Xie D, Chen C, She Y. PET/CT-based deep learning grading signature to optimize surgical decisions for clinical stage I invasive lung adenocarcinoma and biologic basis under its prediction: a multicenter study. Eur J Nucl Med Mol Imaging 2024;51:521-34. [Crossref] [PubMed]