Lymph Node Metastases in Surgically Resected Solitary Ground-Glass Opacities: A Two-Center Retrospective Cohort Study and Pooled Literature Analysis

Evaluating the necessity of lymph node sampling in lung adenocarcinoma with ground glass opacities

BACKGROUND

Ground glass opacity is observed frequently in the early stages of lung adenocarcinoma and is associated with a favorable prognosis and a low incidence of lymph node metastasis. However, the necessity of lymph node sampling in these patients is questionable, although current guidelines still recommend it.

METHODS

Radiologic and clinical data were retrospectively collected and analyzed for 2,298 patients with lung cancer who underwent surgical resection for lesions ≤15 mm during 2022. Based on the consolidation tumor ratios, patients were categorized into 4 groups (pure ground glass opacity, ground glass opacity-predominant, solid-predominant, and pure solid). The incidence of lymph node metastasis in each group was examined.

RESULTS

A total of 2,298 patients with a median age of 54.0 years were enrolled in this study. Tumors were categorized into 4 types: 1,427 (62.1%) were pure ground glass opacity, which constituted the majority, while 421 (18.3%) were ground glass opacity-predominant, 330 (14.4%) were solid-predominant, and the remaining 120 (5.2%) were pure solid. Significant positive correlations were revealed between the consolidation tumor ratio group and pathologic grade (P < .001, ρ = 0.307), T stage (P < .001, ρ = 0.270), and N stage (P < .001, ρ = 0.105). Among the included cases, only 7 cases with metastasis were in the pure solid group. Within this group, 113 cases (94.2%) were N0, 5 cases (4.2%) were N1, and 2 cases (1.7%) were N2.

CONCLUSION

Lymph node metastasis exclusively occurred in the pure solid group, suggesting that for nodules <15 mm, lymph node sampling may be crucial for pure solid nodules but less so for those containing ground glass opacities.

Deep learning model based on primary tumor to predict lymph node status in clinical stage IA lung adenocarcinoma: a multicenter study

Objective

To develop a deep learning model to predict lymph node (LN) status in clinical stage IA lung adenocarcinoma patients.

Methods

This diagnostic study included 1,009 patients with pathologically confirmed clinical stage T1N0M0 lung adenocarcinoma from two independent datasets (699 from Cancer Hospital of Chinese Academy of Medical Sciences and 310 from PLA General Hospital) between January 2005 and December 2019. The Cancer Hospital dataset was randomly split into a training cohort (559 patients) and a validation cohort (140 patients) to train and tune a deep learning model based on a deep residual network (ResNet). The PLA Hospital dataset was used as a testing cohort to evaluate the generalization ability of the model. Thoracic radiologists manually segmented tumors and interpreted high-resolution computed tomography (HRCT) features for the model. The predictive performance was assessed by area under the curves (AUCs), accuracy, precision, recall, and F1 score. Subgroup analysis was performed to evaluate the potential bias of the study population.

Results

A total of 1,009 patients were included in this study; 409 (40.5%) were male and 600 (59.5%) were female. The median age was 57.0 years (inter-quartile range, IQR: 50.0-64.0). The deep learning model achieved AUCs of 0.906 (95% CI: 0.873-0.938) and 0.893 (95% CI: 0.857-0.930) for predicting pN0 disease in the testing cohort and a non-pure ground glass nodule (non-pGGN) testing cohort, respectively. No significant difference was detected between the testing cohort and the non-pGGN testing cohort (P = 0.622). The precisions of this model for predicting pN0 disease were 0.979 (95% CI: 0.963-0.995) and 0.983 (95% CI: 0.967-0.998) in the testing cohort and the non-pGGN testing cohort, respectively. The deep learning model achieved AUCs of 0.848 (95% CI: 0.798-0.898) and 0.831 (95% CI: 0.776-0.887) for predicting pN2 disease in the testing cohort and the non-pGGN testing cohort, respectively. No significant difference was detected between the testing cohort and the non-pGGN testing cohort (P = 0.657). The recalls of this model for predicting pN2 disease were 0.903 (95% CI: 0.870-0.936) and 0.931 (95% CI: 0.901-0.961) in the testing cohort and the non-pGGN testing cohort, respectively.

Conclusions

The superior performance of the deep learning model will help to target the extension of lymph node dissection and reduce the ineffective lymph node dissection in early-stage lung adenocarcinoma patients.

Robotic Versus Thoracoscopic Sub-lobar Resection for Octogenarians with Clinical Stage IA Non-small Cell Lung Cancer: A Propensity Score-Matched Real-World Study

BACKGROUND

Minimally invasive sub-lobectomy is sufficient in treating small early-stage non-small cell lung cancer (NSCLC). However, comparison of the feasibility and oncologic efficacy between robot-assisted thoracoscopic surgery (RATS) and video-assisted thoracoscopic surgery (VATS) in performing sub-lobectomy for early-stage NSCLC patients age 80 years or older is scarce.

METHODS

Octogenarians with clinical stage IA NSCLC (tumor size, ≤ 2 cm) undergoing minimally invasive wedge resection or segmentectomy at Shanghai Chest Hospital from 2011 to 2020 were retrospectively reviewed from a prospectively maintained database. Propensity score-matching (PSM) with a RATS versus VATS ratio of 1:4 was performed. Perioperative and long-term outcomes were analyzed.

RESULTS

The study identified 594 patients (48 RATS and 546 VATS patients), and PSM resulted in 45 cases in the RATS group and 180 cases in the VATS group. The RATS patients experienced less intraoperative bleeding (60 mL [interquartile range (IQR), 50-100 mL] vs. 80 mL [IQR, 50-100 mL]; P = 0.027) and a shorter postoperative hospital stay (4 days [IQR, 3-5 days] vs. 5 days [IQR, 4-6 days]; P = 0.041) than the VATS patients. The two surgical approaches were comparable concerning other perioperative outcomes and postoperative complications (20.00% vs. 26.11%; P = 0.396). Additionally, during a median follow-up period of 66 months, RATS and VATS achieved comparable 5-year overall survival (90.48% vs. 87.93%; P = 0.891), recurrence-free survival (83.37% vs. 83.18%; P = 0.782), and cumulative incidence of death. Further subgroup comparison also demonstrated comparable long-term outcomes between the two approaches. Finally, multivariate Cox analysis indicated that the surgical approach was not independently correlated with long-term outcomes.

CONCLUSIONS

The RATS approach shortened the postoperative hospital stay, reduced intraoperative bleeding by a statistically notable but clinically insignificant amount, and achieved long-term outcomes comparable with VATS in performing sub-lobectomy for octogenarians with early-stage small NSCLC.

Distinct impacts of radiological appearance on lymph node metastasis and prognosis based on solid size in clinical T1 non-small cell lung cancer

BACKGROUND

Solid nodules (SN) had more aggressive features and a poorer prognosis than part-solid nodules (PSN). This study aimed to evaluate the specific impacts of nodule radiological appearance (SN vs. PSN) on lymph node metastasis and prognosis based on solid size in cT1 non-small cell lung cancer (NSCLC).

METHODS

Patients with cT1 NSCLC who underwent anatomical resection between 2010 and 2019 were retrospectively screened. Univariable and multivariable logistic regression analyses were adopted to evaluate the associations between nodule radiological appearance and lymph node metastasis. The log-rank test and Cox regression analyses were applied for prognostic evaluation. The cumulative recurrence risk was evaluated by the competing risk model.

RESULTS

There were 958 and 665 NSCLC patients with PSN and SN. Compared to the PSN group, the SN arm had a higher overall lymph node metastasis rate (21.7% vs. 2.7%, P < 0.001), including nodal metastasis at N1 stations (17.7% vs. 2.1%), N2 stations (14.0% vs. 1.6%), and skip nodal metastasis (3.9% vs. 0.6%). However, for cT1a NSCLC, no significant difference existed between SN and PSN (0 vs. 0.4%, P = 1). In addition, the impacts of nodule radiological appearance on lymph node metastasis varied between nodal stations. Solid NSCLC had an inferior prognosis than part-solid patients (5-year disease-free survival: 79.3% vs. 96.2%, P < 0.001). The survival inferiority only existed for cT1b and cT1c NSCLC, but not for cT1a. Strikingly, even for patients with nodal involvement, SN still had a poorer disease-free survival (P = 0.048) and a higher cumulative incidence of recurrence (P < 0.001) than PSN. Specifically, SN had a higher recurrence risk than PSN at each site. Nevertheless, the distribution of recurrences between SN and PSN was similar, except that N2 lymph node recurrences were more frequent in solid NSCLC (28.21% vs. 7.69%, P = 0.041).

CONCLUSION

SN had higher risks of lymph node metastasis and poorer prognosis than PSN for cT1b and cT1c NSCLC, but not for cT1a. SN exhibited a greater proportion of N2 lymph node recurrence than PSN. SN and PSN needed distinct strategies for nodal evaluation and postoperative follow-up.