Mediastinal lymph node metastasis pattern of subsolid non-small cell lung cancer Authors: Xiao Sun, Yanlu Xin, Yanjun Jia, Hao Wang, Xiangyu Cui, Jun Chen Published: 2026 ABSTRACT BackgroundLymph node (LN) dissection is important in lung cancer surgery. LN metastasis was analyzed in each stage of non-small cell lung cancer (NSCLC) to inform the resection strategy, especially for subsolid nodules. This study aimed to explore LN metastasis distribution features of NSCLC, with a focus on subsolid nodules.MethodsPatients with NSCLC treated from January 2014 to December 2018 (n=6,273) were retrospectively grouped according to primary tumor locations and clinical stages. The incidence and distribution of LN metastasis were analyzed.ResultsClinical and pathological characteristics varied with different primary lesion locations. LN metastasis occurred in 1,356 patients (21.62%). Lobe-specific distribution was observed on mediastinal LN metastasis. However, LN metastasis in each region exceeded 5%, which warrants careful attention during surgical dissection. Subgroup analysis indicated more regular lobe-specific distribution in cT1N0 lesions, but not in advanced stages. LN metastasis was negligible (<1%) in subsolid nodules (≤3 cm) with radiographic consolidation-to-tumor ratio (CTR) ≤0.5, but significant in the CTR >0.5 group (2.34–6.60%) in each mediastinal region. For upper lobe tumors, upper mediastinum was invaded frequently (right, 4.34%; left 5.70%). In right middle lobe, the mediastinal (5.21%) or sub-carinal (6.06%) region was involved. Tumor invasion in lower lobes included the upper (left, 3.38%; right, 2.49%), sub-carinal (right, 5.88%; left, 4.00%) and lower (right, 2.34%; left, 2.33%) regions. Prognosis was significantly better in the CTR ≤0.5 group compared to the CTR >0.5 group (P<0.0001).ConclusionsSubsolid tumors with CTR ≤0.5 demonstrate an extremely low risk of mediastinal LN metastasis, which may help guide individualized lymphadenectomy strategies. For tumors with CTR >0.5, metastatic risk is significantly higher, suggesting more comprehensive LN evaluation may be warranted. Highlight box Key findings • Mediastinal lymph node (LN) metastasis in non-small cell lung cancer (NSCLC) shows a lobe-specific distribution pattern. • Subsolid nodules (≤3 cm) with consolidation-to-tumor ratio (CTR) ≤0.5 have a very low incidence of mediastinal LN metastasis. • Subsolid nodules with CTR >0.5 demonstrate a measurable risk of mediastinal LN involvement across multiple mediastinal regions. What is known and what is new? • LN dissection is essential for accurate staging and prognosis in NSCLC. • This study provides evidence that CTR stratification in subsolid nodules is associated with different patterns and risks of LN metastasis. What is the implication, and what should change now? • Lobe-specific LN dissection or sampling may be adequate for subsolid tumors with CTR ≤0.5. • Systematic mediastinal LN dissection should be considered for subsolid tumors with CTR >0.5, particularly in lower lobe lesions. Introduction The incidence and mortality rates of lung cancer are the highest of all malignant diseases. Invasion of the mediastinal lymph node (LN) significantly worsens the prognosis of lung cancer patients (1). LN dissection (LND) is an important part of lung cancer surgery, since it facilitates accurate staging. In addition, the prognosis of patients can be improved to some extent (2). The guidelines of the European Society of Thoracic Surgeons propose five strategies for lymphadenectomy; systematic LND is the most common approach for non-small cell lung cancer (NSCLC) surgery (3). However, excessive LND may increase trauma and complications. Accordingly, thoracic surgeons have focused on the strategy of mediastinal LND for decades. Evidence indicates that subsolid pulmonary nodules are a special subtype of NSCLC. The prognosis for patients with this subtype is good, with recurrence and metastasis rarely occurring (4). A series of studies have focused on the surgical strategy of subsolid nodules (5-7). Many surgeons have proposed that lobe-specific LND or biopsy is sufficient for the same prognosis with less trauma and reduced complications for lung nodules with a diameter <2 cm (8). However, research data remain inconclusive. To further inform the strategy of LND for subsolid nodules, we retrospectively statistically analyzed the incidence and distribution of mediastinal LN metastasis (LNM) in patients with NSCLC following surgery. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0514/rc). Methods Study subjects This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of The Affiliated Hospital of Qingdao University, Shandong, China (approval No. QYFYKY 2018-10-11-2). Informed consent for the inclusion in this study was obtained from the patients. Patients with NSCLC (n=8,210) who underwent surgical treatment at Affiliated Hospital of Qingdao University from January 2014 to December 2018 were initially included in this study. All patients with operable primary NSCLC underwent lung surgery combined with systematic mediastinal LND or sampling. The exclusion criteria were (I) use of neoadjuvant therapy or tumor transformation therapy, (II) other primary cancers, (III) multiple nodules in different lobes and difficulty in identifying the primary tumor location, and (IV) stage IIIB or more. Finally, 6,273 patients were included (the flow chart is shown in Figure S1). The patients were divided into five groups according to the primary tumor locations: right upper lobe (RUL, n=1,861), right middle lobe (RML, n=549), right lower lobe (RLL, n=1,222), left upper lobe (LUL, n=1,522), and left lower lobe (LLL, n=1,119). The inclusion and exclusion process used in this study is shown in Figure 1. All the cases were staged according to the 8th edition lung cancer staging classification of the American Joint Committee on Cancer. Pathological stage (p-stage) was assessed according to postoperative pathological result, while c-stage was assessed with computed tomography (CT) or positron emission tomography/computed tomography (PET-CT). Nodule size was measured using the longest diameter. Proportion of involved lymph nodes in different zones for subsolid nodules (≤3 cm) with CTR >0.5. CTR, consolidation-to-tumor ratio; LML, left middle lobe; LN, lymph node; LUL, left upper lobe; RLL, right left lobe; RML, right middle lobe; RUL, right upper lobe. LN analyses The grouping of mediastinal and intrapulmonary LNs was based on the LN map provided by International Association for the Study of Lung Cancer (9). To permit more intuitive study, the stations were stratified into upper mediastinum (stations 2R, 3 and 4R for right side; stations 4L, 5 and 6 for left side), lower mediastinum (stations 7, 8 and 9), and intrapulmonary (stations 10, 11 and 12). Stations 2L and 3 were rarely dissected and station 13 was just dissected in segmentectomy; these stations were excluded. The incidence of metastasis for each LN station (%) was defined as the number of patients with metastasis in that station divided by the number of patients who underwent LND at that station. Skip metastasis was defined as mediastinal LNM without intrapulmonary LN involvement. Statistical analyses Statistical analyses were performed using IBM SPSS 22.0 (IBM Corporation, Armonk, NY, USA). Pearson Chi-squared test was used to analyze the differences between groups for categorical variables. One-way analysis of variance and Kruskal-Wallis test were used to compare groups of continuous, normally distributed variables and abnormally distributed variables, respectively. Microsoft EXCEL 2016 (Microsoft Corporation, Redmond, WA, USA) was used for data procession. GraphPad Prism 9.2 software (GraphPad, San Diego, CA, USA) was used to draw statistical figures and Kaplan-Meier plot survival curves. P<0.05 was considered significantly different. For pairwise comparisons among five lung lobe subgroups, the Bonferroni method was applied to correct the significance level for multiple testing. The adjusted threshold was set at P<0.005, while a P value <0.05 was considered statistically significant for other analyses. Results Baseline information Among the total of 6,273 patients meeting the inclusion and exclusion criteria, 3,201 were male (mean age 62.19 years). The patients included 1,988 (43.22%) with a history of smoking and 1,358 (30.12%) with a history of alcohol consumption. The common pathological type was lung adenocarcinoma (n=4,518). There were no significant differences concerning male gender, age, height, tumor size and personal history for each lobe. The comparative clinicopathological characteristics among different groups are summarized in Table 1. Table 1Distribution of clinicopathologic characteristics stratified by tumor location (n=6,273)CharacteristicsTotal (N=6,273)RUL (n=1,861)RML (n=549)RLL (n=1,222)LUL (n=1,522)LLL (n=1,119)P valueBasic information Male3,201 (51.65)938 (50.40)274 (49.91)595 (48.73)815 (53.58)579 (51.741)0.119 Age (years)62.19±9.1762.09±9.0161.54±9.4562.12±9.1262.38±9.2262.52±9.270.360Tumor size (cm)2.86±1.882.64±1.732.99±1.912.95±1.942.82±1.793.23±2.080.006Histology<0.001 LUAD4,518 (72.89)1,438 (77.86)372 (69.27)850 (72.34)1,109 (72.48)749 (67.54) LSCC1,083 (17.47)266 (14.40)119 (22.16)219 (18.64)288 (18.82)251 (22.63) Others597 (9.63)143 (7.74)46 (8.57)106 (9.02)133 (8.69)109 (9.83)Lymph node metastasis0.003 N04,917 (78.38)1,535 (82.48)404 (73.59)955 (78.15)1,186 (77.92)837 (74.80) N1387 (6.17)79 (4.25)43 (7.83)81 (6.63)93 (6.11)91 (8.13) N2969 (15.45)247 (13.27)102 (18.58)186 (15.22)243 (15.97)191 (17.07)p-TNM stage<0.001 I3,409 (55.00)1,161 (62.86)320 (59.59)667 (56.77)796 (52.03)465 (41.93) II1,419 (22.89)340 (18.41)123 (22.91)272 (23.15)371 (24.25)313 (28.22) III1,370 (22.10)346 (18.73)94 (17.50)236 (20.09)363 (23.73)331 (29.85)Anatomical type<0.001 Central1,829 (29.51)400 (21.66)211 (39.29)259 (22.04)456 (29.80)365 (32.91) Peripheral4,507 (72.72)1,447 (78.34)326 (60.71)916 (77.96)1,074 (70.20)744 (67.09)Data are presented as mean ± standard deviation or n (%). LLL, left lower lobe; LSCC, lung squamous cell carcinoma; LUAD, lung adenocarcinoma; LUL, left upper lobe; N, lymph node; p, pathological; RLL, right left lobe; RML, right middle lobe; RUL, right upper lobe; TNM, tumor-node-metastasis. Distribution of clinicopathologic characteristics stratified by tumor location (n=6,273) Data are presented as mean ± standard deviation or n (%). LLL, left lower lobe; LSCC, lung squamous cell carcinoma; LUAD, lung adenocarcinoma; LUL, left upper lobe; N, lymph node; p, pathological; RLL, right left lobe; RML, right middle lobe; RUL, right upper lobe; TNM, tumor-node-metastasis. The clinical and pathological characteristics of lesions varied in each lobe. Tumors occurred most commonly in the RUL (n=1,847) and exhibited the highest incidence of adenocarcinoma (77.86%), most peripheral lesions (78.34%), most stage-I tumors (62.86%), lowest incidence of LNM (17.5%), and smallest tumor size (2.64 cm). For the RML, the central anatomical type was most frequent (41.58%). LLL tumors were largest (3.23 cm), with the fewest stage-I tumors (41.55%) and more instances of squamous cell carcinoma (22.43%). Other factors compared among different lobes included gene mutation, imaging characteristics, LN skipping metastasis, pathological characteristics, and differentiation. Most of these factors were statistically significantly different (Table S1). Incidence and distribution of LNM in each lobe LNM occurred in 1,356 patients (21.62%). Of these, 387 cases (6.17%) were N1, 624 (9.95%) were both N1 and N2 positive, and 345 (5.50%) were N2 only (N2 skip metastases). The incidence of LNM in each lobe was: RUL, 17.52%; RML, 26.41%; RLL, 21.85%; LUL, 22.08%; and LLL, 25.20% (Table 1). The distribution of mediastinal LNM is characterized by lobe specificity. For the upper lobe lesions, the upper mediastinal LNs were more likely to be invaded (RUL: 6.32% for station 2, 6.77% for station 3, 11.44% for station 4R. LUL: 9.38% for station 4L, 12.85% for station 5, and 9.23% for station 6). For RML, the highest incidence was observed in station 7 (13.21%), followed by station 3 (11.82%) and station 4R (11.41%). In lower lobe tumors, LNs in the lower mediastinal and sub-carinal regions were more susceptible. For the RLL group, station 7 was the most common site (12.34%), followed by station 4R (7.26%). For the LLL group, station 7 was the most common site of metastasis (12.05%). 4L were also commonly involved (8.93%), which was more frequent than stations 8 and 9. The incidence of LNM for station 7 was significant in each group. For the intrapulmonary stations (stations 10–12), there is no significant difference in the distribution of LNs among different lobes. The details of lobe-specific mediastinal LNM to each station are provided in Table 2. Table 2Distribution of LNM according to primary tumor locationsLN stationsRUL (n=1,861)RML (n=549)RLL (n=1,222)LUL (n=1,522)LLL (n=1,119)Total (%) (n=6,273)Station 299/1,566 (6.32)30/433 (6.93)43/887 (4.85)––5.96Station 322/325 (6.77)13/110 (11.82)9/202 (4.46)––6.91Station 4206/1,801 (11.44)59/517 (11.41)76/1,047 (7.26)53/565 (9.38)35/392 (8.93)9.93Station 5–––176/1,370 (12.85)50/909 (5.50)9.92Station 6–––68/737 (9.23)17/469 (3.62)7.05Station 754/1,763 (3.06)70/530 (13.21)148/1,199 (12.34)56/1,414 (3.96)128/1,062 (12.05)7.64Station 88/336 (2.38)4/97 (4.12)12/321 (3.74)8/395 (2.03)26/347 (7.49)3.88Station 94/639 (0.63)8/171 (4.68)24/575 (4.17)14/960 (1.46)50/788 (6.35)3.19Station 10127/1,680 (7.56)52/462 (11.26)97/1,012 (9.58)127/1,404 (9.05)131/1,034 (12.67)9.55Station 11115/1,600 (7.19)67/506 (13.24)142/1,109 (12.80)106/1,304 (8.13)142/1,010 (14.06)10.35Station 1280/996 (8.03)40/269 (14.87)65/662 (9.82)100/862 (11.60)73/605 (12.06)10.55LNM distribution was exhibited as number of LNM/number of dissected lymph nodes (proportion). LLL, left lower lobe; LN, lymph node; LNM, lymph node metastasis; LUL, left upper lobe; RLL, right left lobe; RML, right middle lobe; RUL, right upper lobe. Distribution of LNM according to primary tumor locations LNM distribution was exhibited as number of LNM/number of dissected lymph nodes (proportion). LLL, left lower lobe; LN, lymph node; LNM, lymph node metastasis; LUL, left upper lobe; RLL, right left lobe; RML, right middle lobe; RUL, right upper lobe. Subgroup analysis according to different clinical staging Generally, the surgical strategy is determined based on preoperative clinical staging. Accordingly, we further grouped the cases according to different clinical stages and analyzed the incidence of LNM in various lobes. This subgroup analysis revealed that cT1N0 lesions were always associated with lobe-specific LNM, especially for subsolid nodules. For advanced staging tumors (cT1N1–2 and T2–4N0–2), regardless of the primary lesion location, the LNM rate in each mediastinal region was sufficiently high (>5%) to be taken seriously (Tables S2,S3). Metastasis pattern of subsolid nodules (≤3 cm) with different radiographic consolidation-to-tumor ratio (CTR) Multiple research results have indicated that subsolid nodules with various CTRs significantly influence prognosis. A total of 2,159 patients with subsolid pulmonary nodules (≤3 cm in diameter) were enrolled in this study and stratified into two groups according to CTR: CTR ≤0.5 and CTR >0.5 (Table 3). Among them, 1,032 cases were in the CTR >0.5 group and 1,127 cases in the CTR ≤0.5 group. In the CTR >0.5 group, 829 patients underwent lobectomy, while the remaining patients received sublobar resection. There were 10 cases of isolated N1 LNM and 60 cases of N2 LNM. In the CTR ≤0.5 group, 577 patients underwent lobectomy, and the rest were treated with sublobar resection, with only 2 cases of N1 metastasis and 7 cases of N2 metastasis. All patients with LNM received targeted therapy or chemotherapy after surgery. Mediastinal LN metastases were rare (<1%) in the CTR ≤0.5 group (Figure S2). The incidence rate >5% in the CTR >0.5 group was significant in each mediastinal region (Figure 1). Table 3LNM distribution of subsolid nodules (≤3 cm) with different CTR (n=2,159)Clinical stageUpper mediastinal (right/left side)Sub-carinalLower mediastinalIntrapulmonaryStation 2R/4LStation 3/5Station 4R/6Total (%)Station 7Station 8Station 9Total (%)Station 10Station 11Station 12Total (%)Right side RUL CTR ≤0.51/313 (0.32)0/67 (0)2/356 (0.56)0.410/350 (0)0/56 (0)0/113 (0)01/338 (0.30)1/334 (0.30)2/218 (0.92)0.45 CTR >0.57/254 (2.76)3/52 (3.85)16/293 (5.46)4.344/298 (1.34)0/41 (0)0/103 (0)010/256 (3.91)9/243 (3.70)6/156 (3.85)3.82 RML CTR ≤0.50/60 (0)0/16 (0)1/67 (1.49)0.700/68 (0)0/14 (0)0/22 (0)00/64 (0)1/70 (1.43)0/29 (0)0.61 CTR >0.53/83 (3.61)2/13 (15.38)5/96 (5.21)5.216/99 (6.06)0/14 (0)0/32 (0)03/76 (3.95)5/89 (5.62)2/49 (4.08)4.67 RLL CTR ≤0.50/133 (0)0/26 (0)0/150 (0)01/184 (0.54)0/51 (0)0/87 (1.15)02/162 (1.23)0/166 (0)0/109 (0)0.46 CTR >0.55/161 (3.11)0/36 (0)8/188 (4.26)3.3811/187 (5.88)1/44 (2.27)2/85 (2.35)2.336/155 (3.87)12/173 (6.94)7/96 (7.29)5.90Left side LUL CTR ≤0.50/12 (0)2/230 (0.87)0/146 (0)0.520/228 (0)0/51 (0)0/144 (0)00/243 (0)0/209 (0)1/144 (0.69)0.17 CTR >0.51/74 (1.35)15/222 (6.76)8/125 (6.40)5.702/231 (0.87)0/67 (0)1/161 (0.62)0.4412/209 (5.74)10/199 (5.03)10/148 (6.76)5.76 LLL CTR ≤0.50/11 (0)0/115 (0)0/63 (0)01/131 (0.76)0/39 (0)1/106 (0.94)0.690/129 (0)0/129 (0)0/86 (0)0 CTR >0.50/23 (0)4/143 (2.80)2/75 (2.67)2.497/175 (4.00)1/47 (2.13)3/124 (2.42)2.3411/168 (6.55)12/165 (7.27)5/91 (5.49)6.60LNM distribution was exhibited as number of LNM/number of dissected lymph nodes (proportion). CTR, consolidation-to-tumor ratio; LLL, left lower lobe; LNM, lymph node metastasis; LUL, left upper lobe; RLL, right left lobe; RML, right middle lobe; RUL, right upper lobe. LNM distribution of subsolid nodules (≤3 cm) with different CTR (n=2,159) LNM distribution was exhibited as number of LNM/number of dissected lymph nodes (proportion). CTR, consolidation-to-tumor ratio; LLL, left lower lobe; LNM, lymph node metastasis; LUL, left upper lobe; RLL, right left lobe; RML, right middle lobe; RUL, right upper lobe. Upper lobe tumors conformed to the lobe-specific distribution pattern that upper mediastinal region (2R, 3, 4R for RUL and 4L, 5, 6 for LUL). These tumors were invaded frequently (4.34% for RUL and 5.70% for LUL). In the RML, either the upper mediastinal or sub-carinal region was involved (5.21% and 6.06% respectively). When tumors occurred in the lower lobes, each mediastinal region could be involved, especially the upper (3.38% for RLL and 2.49% for LLL) and sub-carinal (5.88% for RLL and 4.00% for LLL) regions. Stations 8 and 9 were occasionally invaded (2.34% for RLL and 2.33% for LLL) (Table 3). Interestingly, for lesions in various lobes, upper mediastinal LNs were more susceptible to involvement (Figure 1). For subsolid tumors with CTR ≤0.5, regional (lobe-specific) LND or sampling was sufficient. However, systemic mediastinal LN resection could be more reliable for subsolid nodules with CTR >0.5, especially lower lobe lesions. Survival analyses of subsolid nodules Subsolid tumors with CTR ≤0.5 had a significantly better overall survival (OS) and disease-free survival (DFS) than the CTR >0.5 group. The respective 10-year OS was 94.62% vs. 84.88% (P<0.0001, Figure 2A). The respective 10-year DFS was 94.06% vs. 73.32% (P<0.0001, Figure 2B). Comparison of OS (A) and DFS (B) between subsolid (≤3 cm) with different CTR. CTR, consolidation-to-tumor ratio; DFS, disease-free survival; OS, overall survival. Discussion Significantly different tumor sizes (P<0.001) and LNM patterns (P=0.006) were observed among the five lobes. RUL had the smallest tumor size and lowest LNM rate. The incidence of LNM was higher in RML than in other locations. LLL tumor sizes were largest. Upper lobe tumors malignancies were relative lower than middle- and lower-lobe tumor malignancies. Furthermore, cT1N0 lung tumors were observed associated with lobe-specific LNM, especially for subsolid nodules. For advanced staging tumors, the LNM rate in each mediastinal region was sufficiently high (>5%) to be taken seriously. The findings highlight the need for increased vigilance on LND for advanced staging tumors, especially for patients with middle- and lower-lobe involvement. The suitable extent of LND remains contentious. The traditional view holds that systematic mediastinal LND is the standard surgical approach for lung cancer (10-12). Lobe-specific LND was introduced in 1998 (12). This innovation shortened surgery duration, while providing similar long-term outcomes as selective LND in stage-I NSCLC. One study presented evidence of the preferential involvement of upper lobe tumors in the upper mediastinum, while lower lobe tumors always migrated to the lower mediastinum (13). Another study analyzed LNM and patient prognosis, and showed that lobe-specific LND is sufficient for early-stage NSCLC (14). A retrospective analysis of the incidence and distribution of 4,511 NSCLC patients revealed the frequent involvement of each mediastinal LN zone. This prompted the authors to propose that systemic LND was still necessary (15,16). Other authors advised that for upper lobe tumors, sub-carinal dissection was unnecessary, while for lower lobe tumors, extensive dissection is required (3). Recent evidence suggests that excessive dissection of mediastinal LNs may reduce the effectiveness of tumor immunotherapy (17). The present findings highlight the need to discuss the extent of LND based on different clinical stages and imaging features. For cT1N0 lung tumors, lobe-specific LND is recommended, especially for subsolid nodules. Systemic LND is more suitable for advanced stages patients, since the rate of involvement in each mediastinal LN region was sufficiently high (>5%). Different from previous studies, the present findings reveal the LNM of subsolid nodules for the first time. Subsolid pulmonary nodules are defined as pulmonary nodules with a diameter ≤3 cm and ground glass density on CT imaging (18). Multiple studies suggest that subsolid pulmonary nodules are a special clinical subtype with a generally good long-term prognosis and less recurrence and metastasis (19,20). However, there is a lack of clear guidelines for the diagnosis and treatment of this type of nodule, as well as it’s characteristics of LNM. In recent years, multiple renowned clinical studies and guidelines have addressed the treatment strategy for subsolid pulmonary nodules (21-23). CTR of subsolid nodules is considered to be related to prognosis and LNM. In the present study (24,25), subsolid nodules with CTR ≤0.5 displayed negligible LNM (<1%). For these nodules, lobe-specific LND or sampling is sufficient. For CTR >0.5 nodules, the incidence of LNM in each mediastinal region should not be ignored (2.34–6.60%). Notably, we observed that the lower lobe nodules (CTR >0.5) were more likely to invade upper mediastinal LNs (left, 3.38%; right, 2.49%). Prognosis was significantly better in the CTR ≤0.5 group compared to the CTR >0.5 group (P<0.0001) (Figure 2). These findings may provide important evidence for the surgical diagnosis and treatment of subsolid lung nodules. Additionally, we also observed some lobe-specific characteristics in hilar and medial LNM. RUL showed the lowest LNM rate, while the rates for LLL and RML were relatively higher. Lower lobe tumors exhibited a high probability of involvement of interlobar fissure. However, for upper lobe tumors, the involvement rate of station 10 and 11 LNs were far less than station 12 (Table 2). Further stratification of the data demonstrated that with elevated stages the incidence of RML tumor to station 10 was increased and higher than that of intrapulmonary LNs. In addition, for advanced-stage LLL patients, the involvement of LNs of hilar and interlobar-fissures stations increased significantly (Tables S2,S3). There were several limitations in our study. As a retrospective study, some data are incomplete, such as driver gene mutation. However, in this investigation, the missing data were not considered to be important and did not impact our main results. We will collect and supplement the data as far as possible in our next study. Additionally, LND in some cases did not involve standard systematic nodal dissection or systematic sampling. The unqualified samples have been excluded as far as possible. Thirdly, there was no systematic follow-up. Although the metastasis pattern of mediastinal LNM theoretically guides lymphadenectomy, the optimal method of LND should be chosen according to patients’ short- and long-term outcomes. Our follow-up results will be reported in a subsequent paper. Conclusions Given the extremely low metastatic risk in subsolid tumors with CTR ≤0.5, our findings may help inform individualized lymphadenectomy decisions. For those with CTR >0.5, especially lower lobe lesions, more comprehensive LN evaluation appears advisable. Supplementary The article’s supplementary files as Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-0514/prf