Ultrasound assessment of pleural invasion via visceral pleura interruption and arch distance-to-tumour diameter ratio in a rabbit model of subpleural lung cancer

Introduction

Lung cancer is the leading cause of cancer death (1,2), and pleural invasion is involved for approximately 30% of them (3), which is an adverse factor for treatment and prognosis (4-6). Surgical resection with lymph node dissection is the recommended standard treatment for early-stage non-small cell lung cancer (NSCLC). If the tumour invades the parietal pleura or chest wall (PL3), extrapleural or en bloc resection is performed with a greater extent of resection than that needed for tumours in other categories, and patients in such cases have a worse prognosis (7).

PL1 (tumour invades beyond the elastic layer of the visceral pleura) and PL2 (tumour invades the visceral pleura surface) tumours were identified as T2 descriptors indicating a poorer prognosis than PL0 (tumour invading beneath the elastic layer) tumours (8). With the introduction of the elastic tissue staining technique, the 5-year disease-free survival (DFS) and overall survival (OS) rates for patients with PL1 tumours (60.2%, 74.4%) were proven to be significantly greater than those for patients with PL2 tumours (28.8%, 50.0%) (9). Among T3 tumours, PL3 tumours (parietal pleural or chest wall invasion, <5 cm) yielded worse patient survival rates than did T3 tumours between 5 cm and 7 cm, as patient survival rates with PL3 tumours were similar to those with T4 tumours (>7 cm) (10). Therefore, precise preoperative evaluation of pleural invasion is highly important for clinical decision-making.

However, pathological pleural invasion is difficult to determine clearly during clinical staging. As computed tomography (CT) is a noninvasive imaging modality used for the preoperative staging of lung cancer, several findings on CT images have been reported for the prediction of visceral pleural invasion (VPI) (11-13). The arch distance-to-maximum tumour diameter ratio, with a sensitivity and specificity of 89.7% and 96.0%, respectively, for chest wall invasion was superior to the other conventional CT criteria, such as a tumour with more than 3 cm of contact with the pleura, an obtuse angle, or associated pleural thickening (11), but insufficient to differentiate PL1 from PL2 tumours (14,15). In another study, an irregular tumour-pleura interface margin on magnetic resonance image (MRI) was found in 10 of the 12 PL2 or PL3 tumours evaluated, and a smooth margin was found in 20 of the 21 PL0 or PL1 tumours evaluated (16), suggesting that the integrity of the visceral pleura was the key to distinguishing PL1 tumours from PL2 tumours, although this feature was limited by respiratory motion artefacts, allergies to contrast media, and patient claustrophobia.

Fortunately, arch distance-to-tumour diameter ratio and interruption of the visceral pleura could be detected simultaneously via ultrasound (US) images; this approach is potentially advantageous for pleural invasion detection because of its superior resolution and real-time imaging ability. Hence, we constructed a subpleural VX2 lung cancer model in rabbits under US guidance and investigated the feasibility of assessing the arch distance-to-tumour diameter ratio combined with visceral pleura interruption for evaluating pleural invasion in rabbits and provided the potential indicators for subsequent clinical trials in patients for lung cancer treatment and prognosis. We present this article in accordance with the STARD and ARRIVE reporting checklists (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0076/rc).

Methods

Modified protocol of developing a subpleural VX2 lung cancer in rabbits

Thirty New Zealand rabbits weighing 2.5kg to 3.0 kg, regardless of gender, were included in this study. They were purchased from Guangdong Mingzhu Biotechnology Co., Ltd. (SCSK 2022-0061) and were fed with standard laboratory diet and tap water ad libitum. Experiments were performed under a project license (No. HTSW221023) granted by the Animal Care Committee of Guangzhou Huateng Biomedical Technology Co., Ltd., in compliance with institutional guidelines for the care and use of animals. VX2 tumour tissue was donated by Yingjia Li’s team of Southern Hospital, Southern Medical University.

We modified a previous method for developing a lung cancer model to construct an animal model of subpleural lung cancer that is more prone to invade the pleura (17). The tumour implantation procedure started with intravenous injection of pentobarbital sodium (3%, 0.8 mL/kg). After skin preparation and routine disinfection of the target area of the chest, 0.4 mL of VX2 tumour tissue in suspension was loaded into a syringe with an 18-gauge needle. All US imaging of the model rabbits was performed using a Resona 7T (Mindray, China) with a 7-MHz linear array probe, unless otherwise specified. First, the needle was used to puncture the pulmonary parenchyma approximately 0.3–0.5 cm below the pleura at an angle of approximately 60° with respect to the chest wall under real-time US guidance; then, the needle was advanced another 0.3–0.5 cm at an angle of approximately 15° (Figure 1A). After the needle was in the proper location, the suspended tumour tissue was injected into the rabbit and the needle was quickly withdrawn; then, the pinhole was immediately covered with gauze and pressure was applied to prevent pneumothorax and bleeding. Penicillin (400,000 units per session) was intramuscularly injected for 3 days to prevent infection.

Figure 1 Schematic diagram of ultrasound-based assessment of pleural invasion. (A) Schematic of US-guided pleural region puncture: two needles are targeted at the region at 60° and 15° angles, respectively. (B) US-anatomical correspondence of pleural lesions: labeling parietal pleura, pleural cavity, visceral pleura, and correspondence between pleural lesions at different anatomical levels (PL0–PL3) and US-visible lesions (uPL1–uPL3). (C) US quantitative measurement of pleural tumours: left shows schematic diagram, right displays US image. Annotated measurements include Dper, Dpar, and Adist. Actual measured values are 0.98, 0.88, and 0.92 cm, respectively. Adist, arch distance; Dpar, diameter parallel to the chest wall; Dper, diameter perpendicular to the chest wall; US, ultrasound.

US scanning was performed on the animals starting from the 10th day post-tumour implantation to monitor VX2 tumour growth. The tumours were allowed to grow for 15–30 days until they reached an approximate diameter of 5–20 mm, after which the formal evaluation was initiated.

US evaluation

Rabbits with lung cancer determined to have lesions abutting the chest wall were selected as candidates in this study. US was performed by a single sonographer with 5 years of experience. Two experienced lung US sonographers who were blinded to the pathological findings evaluated the integrity of the visceral and parietal pleura. If results were inconsistent, the sonographers discussed until an agreement was reached.

US parameters, including acoustic gain, depth, and focus, were optimized for each tumour. The rabbits were anaesthetized with pentobarbital sodium and placed in a lateral recumbent position. The tumours were scanned by US intercostally, and dynamic videos were stored and analysed off-line. The diameters of the tumour parallel (Dpar) and perpendicular (Dper) to the chest wall on the largest section were measured, as was the length of the interface between the primary tumour and neighbouring structures (arch distance). All the lengths are expressed in centimetres. Tumours that were not clearly visible were not assessed. The degree of pleural invasion by lung cancer tumours was determined according to the US criteria (Figure 1B): uPL1, which indicates that the visceral pleura was intact; uPL2, which indicates that the visceral pleura was interrupted by the tumour and that the parietal pleura was smooth; and uPL3, which indicates that the visceral and parietal pleura were both interrupted by the tumour. The arch distance-to-tumour diameter ratio was calculated as A/D (Figure 1C).

Pathologic evaluation

After US examination, the rabbits were sacrificed, and the tumours were harvested for elastic tissue staining and hematoxylin and eosin staining. The degree of pleural invasion was reviewed by a pathologist who was blinded to the experiment based on the Union Internationale Contre le Cancer (UICC) TNM staging system (10). In this system, stage PL0 refers to either a tumour that is within the subpleural lung parenchyma or superficially invading the connective tissue of the visceral pleura beneath the elastic layer; stage PL1 refers to a tumour invading beyond the elastic layer of the visceral pleura; stage PL2 refers to a tumour invading the visceral pleura surface; and stage PL3 refers to a tumour invading any component of the parietal pleura or chest wall. Stages PL0 and PL1 were combined and denoted as stage PL0+1, indicating that it was difficult to distinguish tumours at stage PL0 from those at stage PL1 by US (Figure 1B).

Statistical analysis

Data were tested for normality. The sample size of 27 tumours was determined based on feasibility for this exploratory study rather than formal power calculation. Data for continuous variables are presented as the mean ± standard deviation and as frequencies with percentages for categorical variables. The characteristics of tumours with different degrees of pleural invasion were compared using one-way ANOVA and the Dunn multiple comparisons test. ROC curves were used to determine the cut-off values that yielded the highest combined sensitivity and specificity according to distinct clinical requirements. The areas under the ROC curves (AUCs) were compared following the protocol described by Hanley and McNeil. The consistency between the US assessment results and pathological results was analysed using linear weighted Kappa statistical analysis. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of outcomes based on US image evaluation were calculated. A P value <0.05 was considered to indicate statistical significance. The kappa coefficient was calculated as follows: <0.20, slight; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, substantial; and >0.81, excellent. The statistical data were analysed using SPSS version 26.0 (SPSS, Inc., Chicago, IL, USA).

Results

Tumour characteristics

To determine the efficacy of evaluating US images to assess pleural invasion, we first aimed to construct the rabbit model of subpleural lung cancer that is more prone to invade the pleura. Of the 30 rabbits, 2 had extensive pleural metastasis, and 1 died from an anaesthesia accident; these rabbits were excluded. Thus, a total of 27 tumours were included; the tumour growth time was 20.9±3.8 days; 6 tumours were not clearly visible on US images (due to small size or unfavorable location); these were pathologically confirmed as PL0/1 and classified as PL0+1 for analysis, but were excluded from subsequent analyses involving US-based quantitative measurements (arch distance-to-perpendicular tumour diameter ratio, A/Dper). The mean Dper, Dpar and arch distance (Adist) of the other 21 tumours were 0.81±0.21, 1.04±0.34, and 1.11±0.24 cm, respectively. The tumours were divided into groups PL0+1 (n=13), PL2 (n=8), or PL3 (n=6) according to their pathological status, and uPL1 (n=14), uPL2 (n=7), or uPL3 (n=6) according to their US data. The tumour sizes and tumour presentation on US images for each group are shown in Table 1 and Figure 2. The Dper of the PL3 group was smaller than that of the PL0+1 and PL2 groups (P<0.05), and there was no significant difference in the Dper in the other groups. There was no difference in the Dpar and Adist among staging groups. A/Dper and arch distance-to-Dpar ratio (A/Dpar) values of the PL3 tumours were greater than those of the PL0+1 and PL2 tumours (P<0.05), and there was no significant difference between the PL0+1 and PL2 tumours. The rabbit model of subpleural lung cancer with pleural invasion was successful for this study.

Figure 2 Ultrasound-pathology correlation in pleural lesion assessment. (A-D) Evolution of ultrasound imaging in pleural lesions. (A) Normal pleural line (arrows); (B) pleural thickening with localized bulging (black triangle: parietal pleura; hollow triangle: visceral pleura); (C,D) US appearance of pleural tumour (black/red triangle: parietal pleura; hollow triangle: visceral pleura), showing morphological changes with tumour protrusion into the pleural cavity. (E-H) Histopathological images of pleural lesions. (E) Tumour cellnests in pleura (×400, scale bar 50 µm); (F) Tumour tumour cells infiltrating pleuralstroma (×400, scale bar 50 µm); (G) tumour tissue on pleural surface (×400, scalebar 20 µm); (H) tumor tissue invading the parietal pleura (×200, scalebar 20 µm). US, ultrasound.

Arch distance-to-tumour diameter ratio in evaluating pleural invasion

To determine the cut-off values that yielded the highest combined sensitivity and specificity with respect to evaluating pleural invasion, conventional ROC curves were generated to analyse the A/Dper and A/Dpar (Figure 3). The optimal cut-off value, AUC and diagnostic efficiency of A/Dper and A/Dpar are shown in Table 2. Although A/Dper >1.55 and A/Dpar >1.12 were equivalent in distinguishing PL3 from VPI tumours, the former was more advantageous for distinguishing PL3 from PL2 tumours, with excellent consistency with pathology and a kappa value of 0.857. A/Dper >1.15 and A/Dpar >1.06 were insufficient to distinguish PL1 from PL2 tumours, with kappa values of 0.324 and 0.348, respectively. The finding that A/Dper >1.55 has higher discriminatory power for distinguishing PL3 from PL2 tumours provides a reference for subsequent clinical treatment decisions.

Figure 3 Three-dimensional ROC curves of US parameters A/Dper and A/Dpar for assessing pleural invasion. (A) PL1 vs. PL2; (B) PL2 vs. PL3; (C) VPI vs. PL3. The layered surfaces show the comparison of diagnostic performance at different cut-off values. A/Dpar, arch distance-to-parallel tumour diameter ratio; A/Dper, arch distance-to-perpendicular tumour diameter ratio; ROC, receiver operating characteristic; US, ultrasound; VPI, visceral pleural invasion.

Use of the pleura interruption for evaluating pleural invasion

To investigate the diagnostic efficacy of pleural interruption in the assessment of pleural invasion, US examination was performed at multiple angles, and the ribs were avoided. An evaluation of the continuity of the pleura on US images revealed a smooth margin in the visceral pleura in 12 of the 13 (92.3%) PL0+1 tumours and an interrupted margin in 12 of the 14 (85.7%) PL2 or PL3 tumours (P<0.001) and a smooth margin in the parietal pleura in 19 of the 21 (90.5%) PL0+1 or PL2 tumours and an interrupted margin in 4 of the 6 (66.7%) PL3 tumours (P=0.003). The sensitivity, specificity, PPV, NPV and accuracy of visceral pleura interruption in distinguishing PL0+1 from PL2 and PL3 tumours and parietal pleura interruption in distinguishing PL3 from PL0+1 and PL2 tumours are shown in Table 3. Assessment of PL3 tumours by parietal pleura interruption resulted in a 33.3% false-positive rate. The interruption of the visceral pleura detected on US images demonstrated the efficacy of this approach for separating PL2 tumours from PL1 tumours.

Visceral pleura interruption and the arch distance-to-tumour diameter ratio in evaluating pleural invasion

Next, the diagnostic efficacy of assessing visceral pleura interruption combined with the arch distance-to-tumour diameter ratio in different sequences for assessing pleural invasion was compared. The frequency of visceral pleura interruption followed by an A/Dper >1.55 offered better diagnostic efficacy than the sequence of reverse order, with higher sensitivity, specificity, PPV, NPV and accuracy, as shown in Table 4. The linear weighted kappa value between the pathological and US results was excellent at 0.821.

Based on the optimal sequence of “first observing whether the visceral pleura is interrupted, followed by calculating the A/Dper ratio in cases with interruption”, this study proposes a clinical decision-making flowchart: initially, US examination is performed to determine the integrity of the visceral pleura. If the visceral pleura is continuous without interruption, target-related lesions are preliminarily excluded; if the visceral pleura is clearly interrupted, the pleural space distance (A) and pleural vertical diameter (Dper) are further measured to calculate the A/Dper ratio. Finally, comprehensive lesion assessment is completed by integrating the ratio results and clinical data, providing a reference for the formulation of subsequent diagnostic strategies or treatment plans (Figure 4).

Figure 4 Clinical decision-making flowchart for ultrasonic evaluation of pleural lesions (visceral pleural interruption + A/Dper ratio sequence). A/Dper, arch distance-to-perpendicular tumour diameter ratio.

Discussion

Pleural invasion is an independent factor for poor prognosis in patients with NSCLC and is used to adjust the T stage, which is necessary for treatment and prognosis (18). Therefore, determining the depth of pleural invasion is highly important. This study is the first to evaluate the combination of visceral pleural interruption and the arch distance-to-tumour diameter ratio on US images for assessing pleural invasion in a rabbit model of subpleural lung cancer. Our findings demonstrate that PL0+1 tumours could be identified by a smooth visceral pleura, and PL3 tumours were subsequently distinguished by an A/Dper value >1.55, achieving excellent consistency between pathological and US results.

In order to investigate the role of US in evaluating the degree of pleural invasion of lung cancer, we developed the model in rabbits, because it was large enough and convenient for US observation (19,20). We modified the modelling technique by adjusting the puncture needle angle to ensure the tumour tissue adhered closely to the visceral pleura (Figure 1A). This simple and efficient method successfully induced varying degrees of pleural invasion in 21 out of 27 tumours.

In subsequent analyses, we innovatively applied the arch distance-to-tumour diameter ratio—originally used in CT imaging—to the evaluation of US images. A/Dper >1.55 and A/Dpar >1.12 maximized AUC of the ROC curves with values of 0.933 and 0.767, respectively, for distinguishing PL3 tumours from VPI tumours. Notably, using an A/Dper cut-off value >1.55 to differentiate PL3 from PL2 tumours yielded an AUC as high as 0.979, with excellent agreement between pathological and US assessments. This indicates that the diagnostic performance of US image evaluation is comparable to that of CT. The differences in cut-off values between US and CT images may be attributed to the different imaging planes: CT typically measures the largest cross-section in the axial plane, whereas US measurements are determined based on tumour-specific imaging angles. However, similar to CT, the ability of the arch distance-to-tumour diameter ratio measured on US images to differentiate between PL1 and PL2 pleural invasion remained limited.

According to pathological criteria, pleural interruption is a direct indicator of the depth of lung cancer invasion. Preoperative parietal pleura interruption on surgeon-performed US images was used for assessment of chest wall invasion in a clinical trial reaching a sensitivity and specificity of 90.9% and 85.7%, respectively (21). In contrast, the sensitivity and specificity in our study were 66.7% and 90.5%, respectively. Two PL3 (33.3%) tumours localized in the parietal pleura were judged as PL2 tumours in our study, whereas most tumours of this type invaded the chest wall in other studies. These findings collectively suggest that parietal pleural interruption alone may lack sufficient sensitivity for early detection of parietal pleural invasion.

Furthermore, whether the tumour invades the visceral pleural surface is a key pathological factor distinguishing PL2 from PL1 pleural invasion. A 7-MHz linear array probe was used to assess the continuity of the visceral pleura in our study. Among the 13 PL0+1 tumours, 12 showed no visceral pleural interruption, whereas 12 out of 14 PL2 or PL3 tumours exhibited interruption. The sensitivity, specificity, and accuracy achieved with US evaluation were 85.7%, 92.3%, and 88.9%, respectively, which were comparable to the values for analysis with MRI. Finally, we proposed a two-step diagnostic strategy: first, identifying PL0+1 tumours based on the absence of visceral pleural interruption, and then, for tumours with interruption, determining whether A/Dper was >1.55 to distinguish PL3 from PL2 tumours. This combined approach showed excellent agreement with pathological findings.

Several limitations should be acknowledged. First, the animal sample size was relatively small (PL0+1: n=13, PL2: n=8, PL3: n=6), which may have introduced bias. Additionally, we only included tumours that abutted the chest wall and were visible on US, potentially introducing selection bias. This limits generalizability to central or deeply located tumours not contacting the chest wall. Second, differentiation between PL0 and PL1 was not explored, as it is challenging to determine whether tumours have penetrated the elastic fiber layer on US images. Photoacoustic imaging with near-infrared contrast agents (700–900 nm) has been reported for in vivo elastin imaging (22,23) and may offer a potential solution for distinguishing PL0 from PL1 in the future. Third, model reproducibility across different operators was not formally tested. All tumour inoculations were performed by a single senior researcher with experience in animal model establishment. This may limit the general applicability of the model. Fourth, the thinner chest walls of rabbits compared to human lung cancer patients may result in lower imaging resolution in clinical practice. Before clinical translation, the same US protocol should be piloted in more human-like models (e.g., porcine models or cadaveric human lung specimens) to assess resolution drop-off. Fifth, due to experimental constraints and cost considerations for high-frequency, high-resolution CT in live animals, we did not collect concurrent CT data for direct comparison. Future studies should include head-to-head comparisons between US and CT. Sixth, the robustness of A/Dper measurement to variations in scanning angle was not systematically evaluated. In clinical practice, when tumours are not perfectly perpendicular to the probe, measurement variability may increase. Standardization of the scanning protocol to obtain the maximum arch height view is essential. Seventh, although two sonographers achieved substantial agreement in assessing pleural continuity (kappa =0.856), inter- and intra-observer reproducibility for A/Dper measurement were not formally evaluated. Eighth, a 7-MHz linear high-frequency probe was used to obtain high-resolution pleural images. Lower-frequency probes were tested for deeper penetration but resulted in blurring of the pleura, making integrity assessment difficult, although tumour size measurement was not significantly affected. Thus, combining a high-frequency probe for pleural integrity assessment with a low-frequency probe for tumour size measurement may hold better potential in future clinical studies.

Future directions and sample size recommendations

Based on our findings, we recommend that future human pilot studies enroll at least 50–60 patients with suspected peripheral NSCLC abutting the chest wall, using a standardized US scanning protocol, and prospectively record the time required for A/Dper measurement to assess clinical workflow integration. For the primary endpoint of diagnostic accuracy (using surgical pathology as the gold standard), longitudinal follow-up is not required. However, subsequent larger confirmatory trials would require at least 2 years of follow-up to evaluate local recurrence or disease-free survival.

Conclusions

In conclusion, we presented a novel strategy for preoperative US assessment of tumours invading the pleura in a rabbit model of lung cancer using visceral pleura interruption combined with the A/Dper. However, additional studies with larger sample sizes and multiple imaging modalities are needed to verify the clinical utility of US for predicting the prognosis of NSCLC patients.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD and ARRIVE reporting checklists. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0076/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0076/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0076/prf

Funding: This work was supported by the National Key Research and Development Program “Diagnosis and Treatment Equipment and Biomedical Materials” (No. 2023YFC2411700), Guangdong Provincial Key Laboratory Basic and Applied Basic Research Fund for Enterprise Collaboration (No. 2420210404000003), Major Clinical Research Project of Guangzhou Medical University Research Enhancement Program (No. GMUCR2025-02019), Guangzhou Health and Wellness Science and Technology Project (No. 20231A011085), National Science and Technology Innovation 2030 Major Project, Youth Talent Cultivation Program (No. 2023ZD0517100), and National Natural Science Foundation of China for Young Scientists Program (No. 82400007).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-1-0076/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. Experiments were performed under a project license (No. HTSW221023) granted by the Animal Care Committee of Guangzhou Huateng Biomedical Technology Co., Ltd., in compliance with institutional guidelines for the care and use of animals.

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. Siegel RL, Miller KD, Wagle NS, et al. Cancer statistics, 2023. CA Cancer J Clin 2023;73:17-48. [Crossref] [PubMed]
2. Qi J, Li M, Wang L, et al. National and subnational trends in cancer burden in China, 2005-20: an analysis of national mortality surveillance data. Lancet Public Health 2023;8:e943-55. [Crossref] [PubMed]
3. Jiang L, Liang W, Shen J, et al. The impact of visceral pleural invasion in node-negative non-small cell lung cancer: a systematic review and meta-analysis. Chest 2015;148:903-11. [Crossref] [PubMed]
4. Yang Z, Li X, Bai J, et al. Prognostic Factors for Survival of Stage IB Non-small Cell Lung Cancer Patients: A 10-Year Follow-Up Retrospective Study. Ann Surg Oncol 2023;30:7481-91. [Crossref] [PubMed]
5. Asamura H, Nishimura KK, Giroux DJ, et al. IASLC Lung Cancer Staging Project: The New Database to Inform Revisions in the Ninth Edition of the TNM Classification of Lung Cancer. J Thorac Oncol 2023;18:564-75.
6. Lula Lukadi J, Mariolo AV, Ozgur EG, et al. Upstaged from cT1a-c to pT2a lung cancer, related to visceral pleural invasion patients, after segmentectomy: is it an indication to complete resection to lobectomy? Interdiscip Cardiovasc Thorac Surg 2023;37:ivad102. [Crossref] [PubMed]
7. Zywiciel JF, Verm RA, Raad W, et al. En bloc chest wall resection in locally advanced cT3N2 (stage IIIB) lung cancer involving the chest wall: Revisiting guidelines. JTCVS Open 2024;18:221-31. [Crossref] [PubMed]
8. Liu J, Wang Y, Zhou X, et al. Updated perspectives on visceral pleural invasion in non-small cell lung cancer: A propensity score-matched analysis of the SEER database. Curr Probl Cancer 2025;56:101205. [Crossref] [PubMed]
9. Liang RB, Li P, Li BT, et al. Modification of Pathologic T Classification for Non-small Cell Lung Cancer With Visceral Pleural Invasion: Data From 1,055 Cases of Cancers ≤ 3 cm. Chest 2021;160:754-64. [Crossref] [PubMed]
10. Zhang W, Wang Z, Huang L, et al. Influence of pleural invasion on survival in pathologic T3-4N0M0 non-small cell lung cancer: a propensity score matching study based on the Surveillance, Epidemiology, and End Results database. Transl Lung Cancer Res 2024;13:3214-23. [Crossref] [PubMed]
11. Huang R, Zhao C, Yang J, et al. Nomogram based on radiomics and CT features for predicting visceral pleural invasion of invasive adenocarcinoma ≤ 2 cm: A multicenter study. Eur J Radiol 2025;190:112227. [Crossref] [PubMed]
12. Imai K, Minamiya Y, Ishiyama K, et al. Use of CT to evaluate pleural invasion in non-small cell lung cancer: measurement of the ratio of the interface between tumor and neighboring structures to maximum tumor diameter. Radiology 2013;267:619-26. [Crossref] [PubMed]
13. Wang Y, Lyu D, Deng X, et al. CT predictors of visceral pleural invasion in subsolid nodular pulmonary adenocarcinoma: differences between direct and indirect tumor-pleura contact. Cancer Imaging 2026;26:64. [Crossref] [PubMed]
14. Kim H, Goo JM, Kim YT, et al. CT-defined Visceral Pleural Invasion in T1 Lung Adenocarcinoma: Lack of Relationship to Disease-Free Survival. Radiology 2019;292:741-9. [Crossref] [PubMed]
15. Nishino M. Using CT to Evaluate Visceral Pleural Invasion: Caution Is Advised. Radiology 2019;292:750-1. [Crossref] [PubMed]
16. Zhang Y, Kwon W, Lee HY, et al. Imaging Assessment of Visceral Pleural Surface Invasion by Lung Cancer: Comparison of CT and Contrast-Enhanced Radial T1-Weighted Gradient Echo 3-Tesla MRI. Korean J Radiol 2021;22:829-39. [Crossref] [PubMed]
17. Xing J, He W, Ding YW, et al. Correlation between Contrast-Enhanced Ultrasound and Microvessel Density via CD31 and CD34 in a rabbit VX2 lung peripheral tumor model. Med Ultrason 2018;1:37-42. [Crossref] [PubMed]
18. Rami-Porta R, Bolejack V, Crowley J, et al. The IASLC Lung Cancer Staging Project: Proposals for the Revisions of the T Descriptors in the Forthcoming Eighth Edition of the TNM Classification for Lung Cancer. J Thorac Oncol 2015;10:990-1003.
19. Tanju S, Erus S, Selçukbiricik F, et al. Level of pleural invasion effects on prognosis in lung cancer. Tumori 2019;105:155-60. [Crossref] [PubMed]
20. On KC, Rho J, Yoon HY, et al. Tumor-Targeting Glycol Chitosan Nanoparticles for Image-Guided Surgery of Rabbit Orthotopic VX2 Lung Cancer. Pharmaceutics 2020;12:621. [Crossref] [PubMed]
21. Tahiri M, Khereba M, Thiffault V, et al. Preoperative assessment of chest wall invasion in non-small cell lung cancer using surgeon-performed ultrasound. Ann Thorac Surg 2014;98:984-9. [Crossref] [PubMed]
22. Fu Q, Zhu R, Song J, et al. Photoacoustic Imaging: Contrast Agents and Their Biomedical Applications. Adv Mater 2019;31:e1805875. [Crossref] [PubMed]
23. Su D, Teoh CL, Park SJ, et al. Seeing Elastin: A Near-Infrared Zwitterionic Fluorescent Probe for In Vivo Elastin Imaging. Chem 2018;4:1128-38.