Preoperative ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT)-based strategy for predicting negative mediastinal lymph node metastasis status in non-small cell lung cancer

Introduction

Accurate preoperative assessment of mediastinal lymph node (MLN) metastasis is essential for guiding surgical decision-making in resectable non-small cell lung cancer (NSCLC) (1,2). While ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) is the standard for MLN staging (3,4), extensive research has sought to refine that diagnostic accuracy by integrating clinical and radiological features and machine learning algorithms (5-7). However, these efforts primarily focus on maximizing overall detection of metastasis. In the context of the modern paradigm shift toward precision surgical de-escalation, a critical knowledge gap remains: the identification of a minimal-risk cohort.

Confidently defining patients with negative MLN metastasis status can provide potential candidate for minimally invasive selective lymph node dissection (LND). While systematic LND has been the standard practice for lung cancer surgery (8), recently some scholars have raised that selective removal of MLNs may have potential to reduce injury to mediastinal structures and preserve immunobiological functions (9). We and other teams previously utilized radiological features to screen for potential candidates for selective MLN dissection (10-13), yet there is a lack of comprehensive information of pre-operative PET/CT results in assisting the selection for patients with minimal risk of MLN metastasis. As preoperative ¹⁸F-FDG PET/CT is increasingly available in clinical practices, we hypothesize that metabolic and anatomic information provided by PET/CT could increase accuracy in predicting negative MLN metastasis.

In this study, we examined pre-operative ¹⁸F-FDG PET/CT findings and patterns of MLN metastasis in a large cohort of East-Asian stage I–III NSCLC patients who underwent systematic LND. By integrating radiological features including tumor size, presence of ground-glass opacity (GGO), and the metabolic activity of both the primary tumor and MLNs measured by PET/CT, we propose a preliminary novel strategy of identifying patient with minimal MLN metastasis risk. This approach may help identify candidates for selective LND or even no LND. We present this article in accordance with the STROBE reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0096/rc).

Methods

Patients and enrollment criteria

This retrospective cohort study included stage I–III NSCLC patients who underwent preoperative ¹⁸F-FDG PET/CT, curative surgical resection and complete systematic LND at the Department of Thoracic Surgery, Fudan University Shanghai Cancer Center (FUSCC), from January 2008 to March 2024. 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 Fudan University Shanghai Cancer Center (IRB2008223-9). Individual consent for this retrospective analysis was waived due to the minimal risk posed to patients.

Inclusion criteria for this study were pathologically confirmed invasive stage I–III NSCLC with systematic LND. Systematic LND was performed according to the American Joint Committee on Cancer (AJCC) 8th edition guidelines. Stations 2R/4R for right-sided, 5/6 for left-sided, and 7/8 were routinely dissected for all tumors. Station 3, 4L, 9 are fully dissected except special scenarios: station 4L for left-sided tumors, station 3 for right-sided tumors, and station 9 for upper- and middle-lobe tumors were not dissected in some cases, when metastasis risk was considered minimal. Main bronchus nodes, interlobar nodes, and lobar nodes were also routinely sampled.

Exclusion criteria included: history of neoadjuvant therapy; history of previous malignant disease, positive surgical margins, and synchronous lung cancer; no available chest high-resolution CT (HRCT) scans for review, or the interval between PET/CT and surgery is longer than 90 days; adenocarcinoma in situ (AIS), minimally invasive adenocarcinoma (MIA) and lepidic predominant adenocarcinoma (LPA), because previous results shown that these lesions are always free of lymph node involvement (14).

¹⁸F-FDG PET/CT imaging protocol

All patients underwent standardized ¹⁸F-FDG PET/CT in Fudan University Shanghai Cancer Center. After fasting, ¹⁸F-FDG was injected intravenously and whole-body imaging performed 60 min later using a Biograph 16 PET/CT scanner (Siemens, Knoxville, TN, USA). CT parameters were 120 kV, 150 mA, 0.33 s per rotation, 3.0 mm thickness. PET images were reconstructed with TrueX (2 iterations, 24 subsets, 2 mm full width at half maximum) to a 200×200 matrix, yielding 4.07×4.07×3.0 mm³ voxels. Standard uptake value (SUV) values were normalized to injected dose and body weight. Primary tumor and MLN SUVmax were extracted from preoperative reports and reviewed by nuclear medicine experts at our center.

High-resolution computer tomography protocol

Tumor size, consolidation-to-tumor ratio (CTR), and MLN enlargement were evaluated from the last preoperative HRCT [layer thickness 1.0 mm, lung window width 1,600 Hounsfield unit (HU), level −600 HU]. Tumor size was defined as the largest diameter of the primary lesion. CTR was calculated as the ratio of solid to total tumor diameter and classified into four categories: CTR =0 (pure GGO), 0< CTR <0.5 (GGO-predominant), 0.5≤ CTR <1 (solid-predominant), and 1 (solid). J.Y. and F.F. retrospectively assessed CTR categories on HRCT. N1/N2 lymph node enlargement status was defined by increased short-axis diameter <10 mm of corresponding lymph node, evaluated by radiologists prior to surgery and extracted from reports.

Clinical-pathological data

Clinical data was obtained from our patient electronic health record database, including sex, age, history of smoking, number of resected lymph nodes and station information of metastasized lymph nodes. Clinical T and N staging information was extracted from the last PET/CT report prior to surgical resection. Pathological diagnosis of each patient’s surgical specimens was obtained from the case report form of the pathologist’s diagnostic results at our institution. All tumor, node, metastasis (TNM) staging was according to the eighth edition of the International Association for the Study of Lung Cancer (IASLC) TNM classification.

Statistical analysis

Statistical analyses were conducted using R (v4.4.2) and Stata (v18.0). ROC curve was generated using pROC package in R. DCA was performed using the rmda package. Key diagnostic performance metrics were computed utilizing the caret package. All statistical significance was set to P value <0.05.

Results

Patient characteristics

A total of 2,414 patients were included in this cohort. A flow chart of patient selection process is shown in Figure S1. Detailed patient characteristics are listed in Table 1. The cohort had 959 females (35.9%) and 1,455 males (60.3%), with a median age of 64 years. Eighty-two (3.4%) patients had pure GGNs (CTR =0), 105 (4.4%) patients had GGO-dominant nodules (0< CTR <0.5), 329 (13.6%) patients had solid-dominant nodules (0.5≤ CTR <1), and 1899 (78.7%) patients had solid tumors (CTR=1). No GGO-dominant subsolid invasive nodules with CTR ≤0.5 occurred N2 metastasis (Figure 1A), and primary lung tumor size is positively associated with higher lymph node metastatic burden (Figure 1B), which is similar to prior international multicenter findings (15).

Figure 1 Tumor CTR (A), tumor size (B), tumor SUVmax (C) and MLN SUVmax (D) are associated with mediastinal lymph node metastasis status. CTR, consolidation-to-tumor ratio; MLN, mediastinal lymph node; N, node; SUVmax, maximum standard uptake value.

We examined PET/CT radiological features for predicting MLN metastasis. No N2 metastasis was observed when primary tumor SUVmax is below the conventional threshold of 2.5 (Figure 1C), while patients with nodal SUVmax <2.5 still had metastases (Figure 1D). An optimal MLN SUVmax cut-off of 3.9 was identified based on the Youden index, resulting in a maximum Youden of 0.357 and a corresponding accuracy of 0.733 (Figure S2). Alluvial plots (Figure 2) illustrated a concentration of larger tumors and higher SUVmax values in both primary lesions and MLNs in higher N stages. This trend was consistently observed in both lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LSCC) pathological subtypes (Figure 2B,2C).

Figure 2 Alluvial plot reveals relationship between tumor size, tumor SUVmax, MLN SUVmax and N staging in all participants (A), adenocarcinoma subtype (B) and squamous cell carcinoma subtype (C). CTR, consolidation-to-tumor ratio; LSCC, lung squamous cell carcinoma; LUAD, lung adenocarcinoma; MLN, mediastinal lymph node; N, node; SUVmax, maximum standard uptake value.

Logistic regression model identified clinical-radiological features associated with MLN metastasis

Univariate logistic regression identified clinical and radiological features predicting pN2 status (Table 2). A subsequent multivariate model using pre-operative covariates found that CT N2 enlargement, CTR group, and MLN SUVmax were significant independent predictors (P<0.001).

Notably, primary tumor SUVmax was significant in univariate analysis but lost significance in the multivariate model (P=0.32). This was not due to multicollinearity (variance inflation factor <1.20) (Table S1, Figure S3), indicating other factors provided stronger, independent predictive power.

The multivariate model demonstrated the best performance (area under the curve =0.7521), significantly superior to MLN SUVmax alone (DeLong’s test, P=0.0001) (Figure S4A). Decision curve analysis showed higher net benefit (10–40% thresholds) (Figure S4B), and internal validation (500 bootstraps) confirmed good model calibration, supporting the combined use of preoperative CT and PET/CT metrics (Figure S4C, Table 3). We conducted a post-hoc assessment of the model’s power. Our cohort includes 497 events. Given that our final multivariable model included 6 predictors, our events per variable (EPV) ratio of 82.8:1, providing robust protection against overfitting and ensuring sufficient statistical power.

Characteristics of occult MLN N2 metastasis

Occult MLN metastasis refers to the scenario where metastatic lymph nodes, missed by preoperative imaging, are confirmed postoperatively, resulting in pathological upstaging (16). We studied the patterns of occult MLN metastasis, defined by the disagreement of N2 status in preoperative PET/CT reports (Table 4). Of 451 pathologically confirmed pN2 cases, 358 cases were correctly identified, and 93 cases are considered as occult. Significant factors of high occult metastasis risks include age (P=0.03), female sex (P<0.001), non-smoking (P<0.001), LUAD subtype (P<0.001), smaller tumor size, tumor SUVmax and MLN SUVmax (all P<0.001). The association of occult MLN metastasis with smaller, less FDG-avid tumors reveals a key limitation of PET/CT-based staging.

Preliminary strategy for predicting negative MLN metastases of cT1 tumors based on PET/CT results

LND strategy of stage I lung cancer is a subject of intense controversies. Currently, American College of Surgeons Commission on Cancer (ACS CoC) recommend dissection of 1 hilar lymph node and 3 MLNs in cT1 tumors (17). Therefore, we focused on analysis of 1,573 cases of cT1a-cT1c tumor in our cohort. We previously described a 6-criterion CT-based strategy including tumor size, CTR and tumor segmental location (10) and validated the effectiveness and safety of implementing these strategies in prospective cohort (18,19). We suggest that incorporating PET/CT metrics could further simplify selection and expand suitability for selective LND.

Among 185 cases with pure GGOs or CTR <0.5, no MLN N2 metastasis was observed, consistent with their non-invasive nature. For 266 solid-predominant cT1 nodules (0.5≤ CTR <1), N2 rate was 4.89%, where 189 cases (71.0% of this group) where MLN SUVmax is less than the conventional and widely accepted threshold of 2.5 (20) showed no metastasis. Meanwhile, 1,120 solid tumors (CTR =1) had higher N2 involvement (18.5%), but in 130 cases (11.6%) where primary tumor SUVmax <2.5 and MLN SUVmax <2.5, no metastasis occurred (Figure 3).

Figure 3 Strategy for predicting negative mediastinal lymph node metastasis status of non-small cell lung cancer based on pre-operative PET/CT results. CT, computed tomography; CTR, consolidation-to-tumor ratio; IMLN, inferior mediastinal lymph node; LND, lymph node dissection; NSCLC, non-small cell lung cancer; PET, positron emission tomography; SLND, systemic lymph node dissection; SMLN, superior mediastinal lymph node; SUVmax, maximum standard uptake value.

Discussion

¹⁸F-FDG PET/CT is guideline recommended for lung cancer staging (3,4). A meta-analysis reported PET/CT have an overall sensitivity 0.77–0.81 and specificity 0.79–0.90 for mediastinal N2 nodal staging, influenced by scanner, tumor pathological subtype, FDG dosage, and study origin institutes (21). In this study, we utilized the largest single-center NSCLC cohort (n=2,414) with ¹⁸F-FDG PET/CT data on MLN status to our knowledge (22,23). Several research emphasized on improving N2 staging accuracy via PET/CT radiological features. Kameyama et al. proposed I/C-SUV ratio, which is ratio between ipsilateral hilar node SUV and contralateral hilar node SUV, achieving accuracy achieving accuracy of 71.3% when >1.34 with SUV >2.5 (24). Aokage et al developed a predictive formula including tumor diameter, CTR, and solid density (12). Uses of artificial intelligence, especially machine learning and computer vision, also show potential to improve MLN staging (25-27). While the existing literature has predominantly focused on maximizing the overall diagnostic precision for detecting metastasis, our study is specifically designed to accurately identify a patient cohort with a minimal risk of MLN metastasis.

Systematic LND or sampling is recommended by the National Comprehensive Cancer Network (NCCN) (3) and European Society for Medical Oncology guidelines (4) as standard intra-operative lymph node evaluation. However, recent findings suggest radical lymphadenectomy may not provide excessive benefits to all patients. The pivotal ACOSOG Z0030 trial on 1023 T1–T2 NSCLC patients receiving limited sampling or full dissection found no difference in disease-free survival or recurrence (28). Furthermore, a meta-analysis revealed postoperative complication rates were generally higher for systematic LND than sampling (29). These findings led to a search for minimally invasive, selective MLN dissection surgical practice for patients with no risk of metastases. JCOG1413 is a landmark ongoing randomized controlled study (RCT) on long-term benefits of lobe-specific nodal dissection (30).

Currently, strategies of stratifying patients with minimal risks of N2 metastasis suitable for surgical de-escalation is unclear, especially for clinical stage I NSCLC. Within our cohort, the aggregate 20.6% occult N2 metastasis rate underscores the continued necessity of systematic LND for high-risk individuals. This observation aligns with recent retrospective data in cT1b–c tumors suggesting that systematic LND remains warranted regardless of metabolic activity on PET/CT (31,32). However, this high baseline risk highlights the urgent clinical need for more granular stratification; by identifying a specific subset where this risk is significantly attenuated, our model provides a pragmatic framework to balance oncological safety with the goals of treatment de-escalation.

Our team previously conducted series of studies and proposed a selective dissection strategy based on preoperative CT findings including cT stage, CTR and segmental location (10). Development of this strategy has led to ECTOP-1009, a multi-center RCT in 302 peripheral T1N0 invasive NSCLC patients (18). No MLN metastasis was found in patients selected by our criteria, and selective LND resulted in a shorter surgery duration, fewer blood loss, shorter hospital stays and no lymphadenectomy-related postoperative complications (19). This comprehensive cohort of patients provides the much evidence necessary to challenge the routine use of PET/CT and systematic LND (33).

Based on previous promising studies on selective MLND and the current research on PET/CT metrics characteristics, we proposed a novel preliminary strategy for selective LND in stage I NSCLC using preoperative PET/CT metrics. First, for pure GGOs and GGO-dominant nodules where CTR <0.5, we recommend omitting unnecessary for PET/CT scans and systemic LND, to avoid over-diagnosis and over-surgery. Second, for solid-predominant subsolid nodules where 0.5≤ CTR <1, MLN SUVmax of PET/CT can distinguish nodal metastases risk using a conventional threshold of 2.5. For MLN SUVmax <2.5, no systemic LND is needed; for MLN SUVmax ≥2.5, based on prior results on tumor metastases patterns based on tumor location, a location-specific dissection strategy can be used: upper lobe tumor can spare inferior mediastinal lymph node (IMLN) dissection; lower lobe tumor can spare superior mediastinal lymph node (SMLN) dissection. As for solid nodules where CTR =1, only tumors with primary tumor SUVmax <2.5 and MLN SUVmax <2.5 may be considered for sparing of MLN dissection. Other cases should rigorously perform systemic LND or adhere to previous segment location-based strategies (10). Our proposed strategy is capable of stratifying subgroups with none N2 metastasis risks in our comprehensive cohort, which lays the groundwork for future prospective validations.

There are several limitations in this study. First, this single-center, retrospective analysis on East-Asian population inevitably carries the risk of selection bias. The accuracy of PET/CT for assessing MLN status may be influenced by variability in PET/CT scanner models, imaging practices, and interpretation criteria over the time course of this study, although these factors remained relatively constant within a single center. While rigorous cross-validation is performed within our internal cohort, external and prospective validation is needed. Multiple prospective trials are in progress (ECTOP-1003, NCT03216551; ECTOP-1009, NCT04527419; ECTOP-1018; NCT05970913) or preparation for future validation for the effectiveness of PET/CT in guiding selective MLN dissection. Second, the multivariate logistic regression model demonstrated only moderate overall discrimination. However, it is important to note that our primary objective was not to maximize aggregate statistical performance, but to develop a clinically transparent strategy with a high negative predictive power to ensure oncological safety during surgical de-escalation. While this model provides a pragmatic tool for immediate bedside use, we acknowledge that the future integration of multimodal data and superior algorithms may offer more granular risk stratification and superior predictive power. Third, although all patients underwent systematic LND, procedural differences between surgeons may introduced inconsistencies in nodal sampling, and long-term recurrence and survival are needed for comparative analysis.

Conclusions

In conclusion, we utilized pre-operative PET/CT radiological findings, including GGO proportion, primary tumor size and SUVmax of both tumor and MLN to predict the absence of MLN metastasis in NSCLC. Based on these findings, we propose a preliminary PET/CT-based predictive strategy for negative MLN status, which is warranted for future prospective validations.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0096/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0096/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0096/prf

Funding: The study was supported by National Key R&D Program of China (No. 2022YFA1103900), National Natural Science Foundation of China (No. 82430099), the Clinical Research Special Project of Shanghai Municipal Health Commission (No. 202440070), and the Medical Research Special Project for Shanghai Science and Technology Innovation Action Plan (No. 24Y12800400).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0096/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 Fudan University Shanghai Cancer Center (IRB2008223-9). Individual consent for this retrospective analysis was waived due to the minimal risk posed to 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. Park BJ, Kim TH, Shin S, et al. Recommended Change in the N Descriptor Proposed by the International Association for the Study of Lung Cancer: A Validation Study. J Thorac Oncol 2019;14:1962-9. [Crossref] [PubMed]
2. Mulita A, Valsamaki P, Bekou E, et al. Benefits from (18)F-FDG PET-CT-Based Radiotherapy Planning in Stage III Non-Small-Cell Lung Cancer: A Prospective Single-Center Study. Cancers (Basel) 2025;17:1969. [Crossref] [PubMed]
3. Riely GJ, Wood DE, Ettinger DS, et al. Non-Small Cell Lung Cancer, Version 4.2024, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2024;22:249-74. [Crossref] [PubMed]
4. Postmus PE, Kerr KM, Oudkerk M, et al. Early and locally advanced non-small-cell lung cancer (NSCLC): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2017;28:iv1-iv21. [Crossref] [PubMed]
5. Sathekge C, Maes J, Maes A, et al. FDG PET/CT for Staging Lung Carcinoma: An Update. Semin Nucl Med 2025;55:167-74. [Crossref] [PubMed]
6. Huang Q, Mo J, Deng X, et al. NHOC/NHOP as novel biomarkers for predicting lymph node metastasis in NSCLC using PET/CT radiomics and machine learning: a two-center retrospective study. Eur J Nucl Med Mol Imaging 2026;53:2387-402. [Crossref] [PubMed]
7. Rogasch JMM, Frost N, Bluemel S, et al. FDG-PET/CT for pretherapeutic lymph node staging in non-small cell lung cancer: A tailored approach to the ESTS/ESMO guideline workflow. Lung Cancer 2021;157:66-74. [Crossref] [PubMed]
8. Patel AJ, Bille A. Lymph node dissection in lung cancer surgery. Front Surg 2024;11:1389943. [Crossref] [PubMed]
9. Jiang C, Zhang Y, Fu F, et al. A Shift in Paradigm: Selective Lymph Node Dissection for Minimizing Oversurgery in Early Stage Lung Cancer. J Thorac Oncol 2024;19:25-35. [Crossref] [PubMed]
10. Zhang Y, Fu F, Wen Z, et al. Segment Location and Ground Glass Opacity Ratio Reliably Predict Node-Negative Status in Lung Cancer. Ann Thorac Surg 2020;109:1061-8. [Crossref] [PubMed]
11. Kamigaichi A, Aokage K, Ikeno T, et al. Long-term survival outcomes after lobe-specific nodal dissection in patients with early non-small-cell lung cancer. Eur J Cardiothorac Surg 2023;63:ezad016. [Crossref] [PubMed]
12. Aokage K, Suzuki K, Wakabayashi M, et al. Predicting pathological lymph node status in clinical stage IA peripheral lung adenocarcinoma. Eur J Cardiothorac Surg 2021;60:64-71. [Crossref] [PubMed]
13. Jiang C, Deng C, Deng P, et al. Optimizing Mediastinal Lymph Node Dissection by Integrating Tumour Location, Consolidation Tumour Ratio, and Size for Lung Cancer. Eur J Cardiothorac Surg 2026;68:ezag023. [Crossref] [PubMed]
14. Cheng X, Zheng D, Li Y, et al. Tumor histology predicts mediastinal nodal status and may be used to guide limited lymphadenectomy in patients with clinical stage I non-small cell lung cancer. J Thorac Cardiovasc Surg 2018;155:2648-2656.e2. [Crossref] [PubMed]
15. Zhang Y, Fu F, Chen H. Management of Ground-Glass Opacities in the Lung Cancer Spectrum. Ann Thorac Surg 2020;110:1796-804. [Crossref] [PubMed]
16. Beyaz F, Verhoeven RLJ, Schuurbiers OCJ, et al. Occult lymph node metastases in clinical N0/N1 NSCLC; A single center in-depth analysis. Lung Cancer 2020;150:186-94. [Crossref] [PubMed]
17. Khouzam MS, Wood DE, Vigneswaran W, et al. Impact of Federal Lung Cancer Screening Policy on the Incidence of Early-stage Lung Cancer. Ann Thorac Surg 2023;115:827-33. [Crossref] [PubMed]
18. Zhang Y, Deng C, Zheng Q, et al. Selective Mediastinal Lymph Node Dissection Strategy for Clinical T1N0 Invasive Lung Cancer: A Prospective, Multicenter, Clinical Trial. J Thorac Oncol 2023;18:931-9. [Crossref] [PubMed]
19. Zhang Y, Qian B, Song Q, et al. Phase III Study of Mediastinal Lymph Node Dissection for Ground Glass Opacity-Dominant Lung Adenocarcinoma. J Clin Oncol 2025;43:3081-9. [Crossref] [PubMed]
20. Hellwig D, Graeter TP, Ukena D, et al. 18F-FDG PET for mediastinal staging of lung cancer: which SUV threshold makes sense? J Nucl Med 2007;48:1761-6. [Crossref] [PubMed]
21. Schmidt-Hansen M, Baldwin DR, Zamora J. FDG-PET/CT imaging for mediastinal staging in patients with potentially resectable non-small cell lung cancer. JAMA 2015;313:1465-6. [Crossref] [PubMed]
22. Billè A, Okiror L, Skanjeti A, et al. Evaluation of integrated positron emission tomography and computed tomography accuracy in detecting lymph node metastasis in patients with adenocarcinoma vs squamous cell carcinoma. Eur J Cardiothorac Surg 2013;43:574-9. [Crossref] [PubMed]
23. Choi SY, Song MJ, Kwon BS, et al. Clinical characteristics of post-operative N2 nodal upstaging in non-small cell lung cancer: a retrospective cohort study. Transl Lung Cancer Res 2025;14:5218-29. [Crossref] [PubMed]
24. Kameyama K, Imai K, Ishiyama K, et al. New PET/CT criterion for predicting lymph node metastasis in resectable advanced (stage IB-III) lung cancer: The standard uptake values ratio of ipsilateral/contralateral hilar nodes. Thorac Cancer 2022;13:708-15. [Crossref] [PubMed]
25. Chen Z, Yang H, Qi M, et al. Enhancing (18)F-FDG PET image quality and lesion diagnostic performance across different body mass index using the deep progressive learning reconstruction algorithm. Cancer Imaging 2025;25:58. [Crossref] [PubMed]
26. Zhong Y, Cai C, Chen T, et al. PET/CT based cross-modal deep learning signature to predict occult nodal metastasis in lung cancer. Nat Commun 2023;14:7513. [Crossref] [PubMed]
27. Leivaditis V, Maniatopoulos AA, Lausberg H, et al. Artificial Intelligence in Thoracic Surgery: A Review Bridging Innovation and Clinical Practice for the Next Generation of Surgical Care. J Clin Med 2025;14:2729. [Crossref] [PubMed]
28. Darling GE, Allen MS, Decker PA, et al. Randomized trial of mediastinal lymph node sampling versus complete lymphadenectomy during pulmonary resection in the patient with N0 or N1 (less than hilar) non-small cell carcinoma: results of the American College of Surgery Oncology Group Z0030 Trial. J Thorac Cardiovasc Surg 2011;141:662-70. [Crossref] [PubMed]
29. Mokhles S, Macbeth F, Treasure T, et al. Systematic lymphadenectomy versus sampling of ipsilateral mediastinal lymph-nodes during lobectomy for non-small-cell lung cancer: a systematic review of randomized trials and a meta-analysis. Eur J Cardiothorac Surg 2017;51:1149-56. [Crossref] [PubMed]
30. Hishida T, Saji H, Watanabe SI, et al. A randomized Phase III trial of lobe-specific vs. systematic nodal dissection for clinical Stage I-II non-small cell lung cancer (JCOG1413). Jpn J Clin Oncol 2018;48:190-4. [Crossref] [PubMed]
31. Watanabe Y, Uchida S, Hattori A, et al. A Comparative Analysis of Nodal Spread in Solid-Predominant vs Pure-Solid Stage IA Hypometabolic Lung Adenocarcinoma. Ann Thorac Surg 2025; Epub ahead of print. [Crossref]
32. Wu Q, Lv Z, Lei T. Beyond SUVmax and Morphology: The Critical Role of Tumor Invasive Size in Hypometabolic Nodal Spread. Ann Thorac Surg 2025; Epub ahead of print. [Crossref]
33. Watanabe Y, Suzuki K. Role of Radiologic Morphology in Nodal Spread of Clinical Stage IA Hypometabolic Lung Adenocarcinoma. Ann Thorac Surg 2026; Epub ahead of print. [Crossref]