Sage Journals: Discover world-class research

Abstract

Background:

Introduced by the International Association for the Study of Lung Cancer (IASLC) in 2005, uncertain resection (Run) categorizes a new subclass of residual tumor. Despite several studies, the prognostic significance of Run in operable non-small cell lung cancer (NSCLC) remains unclear.

Objectives:

This study aimed to investigate the prognostic influence of Run in operable NSCLC, focusing on the impact of the four elements that comprise R descriptors on patient survival.

Design:

A systematic review and meta-analysis were conducted to synthesize data from relevant clinical studies.

Methods:

We developed search strategies to identify relevant clinical studies across databases including PubMed, Embase, Cochrane Library, Web of Science, CNKI, and Wanfang up to June 2024. Quantitative analysis was performed with Stata 15 to investigate the prognostic influence of Run, the extent of mediastinal lymph node removal, and the highest mediastinal lymph node involvement (HMLI). We also summarized the main findings from studies on pleural lavage cytology (PLC) and carcinoma in situ in operable NSCLC.

Results:

Compared to complete resection, Run-associated patients exhibited inferior 5-year overall survival (OS) and disease-free survival (DFS; risk ratio (RR) = 1.31, 95% confidence interval (CI) 1.19–1.44; RR = 1.43, 95% CI 1.28–1.60). Limited lymphadenectomy (L-LA) in cI stage showed similar survival benefit (OS, RR = 0.97, 95% CI 0.90–1.06; DFS, RR = 1.06, 95% CI 0.97–1.15), in contrast with systematic lymph node dissection (SLND). For pN2-III patients, HMLI indicated poorer OS (hazard ratio (HR) = 1.22, 95% CI 1.14–1.31) and DFS (HR = 1.25, 95% CI 1.14–1.36).

Conclusion:

IASLC’s residual tumor classification correlated with significant survival differences. Compared with R0, Run was associated with inferior 5-year OS and DFS. L-LA seemed to provide equivalent survival benefits, in contrast to SLND. For patients with low invasiveness in stage cI, L-LA could be considered as a preferred option. HMLI predicts poorer survival in pN2-III patients, and positive PLC significantly worsened long-term survival in operable NSCLC, particularly at early stage.

Keywords

highest mediastinal lymph node mediastinal lymph node dissection pleural lavage cytology prognosis uncertain resection

Introduction

Despite a decline in global smoking rates over several decades, lung cancer remains the most prevalent cancer in China as of 2022 and continues to be the leading cause of cancer-related deaths among both men and women.¹ Non-small cell lung cancer (NSCLC) accounts for 85% of all lung cancer cases, with approximately 25% diagnosed at an early stage.^2,3 Surgical resection is the recommended treatment for patients with curable stage.

The definition of complete resection is based on the residual tumor classification proposed by the Union for International Cancer Control (UICC), which includes three categories: R0, R1, and R2. R0 indicates no evidence of residual tumor, while R1 and R2 denote microscopic and macroscopic residual tumors, respectively.⁴ Due to limitations in the traditional R classification for predicting regional recurrence, the International Association for the Study of Lung Cancer (IASLC) proposed a modified R classification system, comprising complete resection, uncertain resection (Run), and incomplete resection.⁵ When determining postoperative treatment strategies for patients, the TNM classification and R classification serve as important references. However, in guidelines such as NCCN, Chinese Society of Clinical Oncology (CSCO), and European Society for Medical Oncology (ESMO), the presence of R1/R2 in the R classification affects treatment decisions. For example, in stage IIB NSCLC, adjuvant chemotherapy is recommended for R0 resection, while reoperation or concurrent chemoradiotherapy is suggested for R1/R2 resection. In contrast, there are no specific treatment recommendations for patients with Run resection. For a lung resection to be deemed complete, all the following criteria must be met: (a) microscopically confirmed negative resection margins; (b) systematic lymph node dissection (SLND) or lobe-specific systematic lymph node dissection (L-SND), including at least six groups of lymph nodes based on the primary tumor’s location in the lung lobe, with three groups from intrapulmonary and/or hilar stations and three from mediastinal stations, one of which must be subcarinal; (c) separately resected nodules or nodules at the margins of the major lung specimen should not exhibit tumor extension beyond the capsule; (d) the highest mediastinal lymph nodes removed must be negative.

Incomplete resection includes R1 (microscopic residual tumor) and R2 resection (macroscopic residual tumor). Any resection involving the following is considered as incomplete: (a) tumor involvement at the resection margins; (b) tumor extension beyond the capsule in separately resected nodules or at the margins of the major lung specimen; (c) known positive but unremoved lymph nodes; (d) positive cytology in pleural or pericardial effusion.

Uncertain resection must have tumor-free margins but may include one or more of the following situations: (1) intraoperative lymph node assessment fails to meet the standards of SLND or L-SND; (2) the highest mediastinal lymph node is invaded; (3) carcinoma in situ (CIS) is confirmed at the bronchial margin; (4) pleural lavage cytology (PLC) is positive.⁵ Professional guidelines have provided stratified pathways for R0 and R1/R2, but there is currently no consensus on the follow-up management of Run patients.

In this article, we first conducted a literature search and meta-analysis on the prognosis of applying the IASLC R staging system. We then performed a meta-analysis focusing on the elements included in Run, particularly different lymph node dissection methods and the involvement of the highest mediastinal lymph node. To our knowledge, this is the first article to conduct a meta-analysis of observational studies on residual tumor classification based on IASLC descriptors. Currently, SLND remains the recommended practice for all operable NSCLC patients, regardless of tumor stage or location.⁶ Although the value of accurate histopathological staging of nodal status with SLND is widely recognized, it typically requires longer operation times and may increase perioperative complications. Consequently, many surgeons prefer limited lymphadenectomy (L-LA), such as mediastinal lymph node sampling (MLNS) and L-SND, especially in early-stage NSCLC.⁷ This highlights the need to explore the prognosis for patients reclassified as having uncertain resection (Run). According to most professional guidelines, patients with stage II and III are recommended for adjuvant chemotherapy or targeted therapy, regardless of residual tumor status. Therefore, we summarized the main findings regarding different lymph node removal methods in clinical stage I NSCLC, aiming to provide evidence for developing follow-up schemes and treatment plans for patients classified as Run, particularly those in clinical stage IA disease. Highest mediastinal lymph node involvement (HMLI) is the second most common reason for migration from UICC R0 to IASLC Run.⁸ Thus, a meta-analysis was conducted to clarify the clinical evidence and its impact on patient survival, such as overall survival (OS) and disease-free survival (DFS), with the goal of informing the development of follow-up protocols and treatment strategies for patients classified as pN2-III NSCLC. Numerous studies have established the prognostic significance of PLC in early-stage NSCLC.^9–12 Consequently, this article includes a brief systematic review of PLC’s prognostic value, specifically focusing on implications for stage I patients.

Methods

Protocol and registration

This protocol was registered with Prospective Registry of Systematic Reviews (PROSPERO) under the number CRD42023443935 (available at: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42023443935). This systematic review and meta-analysis were conducted in alignment with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.¹³

Literature source and search strategy

This systematic review aggregated data from multiple databases, including PubMed, Cochrane Library databases, Web of Science, EMBASE, CNKI, and Wanfang databases, covering publications from their inception up to January 2024. The search utilized keywords and Medical Subject Headings terms, such as lung neoplasms, NSCLC, early-stage lung cancer, surgical procedures, operative, surgical resection, resection margins, uncertain resection, incomplete resection, suboptimal resection, lymph node excision, SLND, complete lymph node dissection, selective lymph node dissection, limited lymph node dissection, lymph node sampling, highest mediastinal lymph node, PLC, cytological analysis, prognosis, survival outcome, and mortality. The detailed search strategies of the databases are provided in Supplemental File 1.

In addition to electronic database searches, we conducted manual searches of references within systematic reviews, meta-analyses, and clinical trial registries, including the US Clinical Trials Registry, the World Health Organization International Clinical Trials Registry Platform, and the Chinese Clinical Trial Registry. We also reviewed conference proceedings from prominent organizations such as the American Society of Clinical Oncology, the ESMO, the American Association for Cancer Research, the World Conference on Lung Cancer, and the CSCO. The initial database searches were performed in January 2024 and updated in June 2024.

Study selection

Studies were included based on the following criteria: (1) patients diagnosed with primary NSCLC who underwent surgical treatment; (2) study designs including randomized controlled trials (RCTs) and observational studies; (3) study outcomes including OS or DFS; (4) studies with full accessibility in Chinese or English languages. Excluded were case reports, non-English/Chinese publications, and studies lacking full-text access. No study was excluded for flaws in study design or quality. For the purpose of this analysis, subgroup analyses that presented data by tumor stage were considered as separate studies. In cases of multiple publications or patients overlap, the most recent publication was chosen.

Quality assessment and data extraction

The methodological quality of each included RCT was assessed by two reviewers, J.Y. and X.L., using the Cochrane Collaboration’s Risk of Bias tool.¹⁴ The evaluation criteria included random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, completeness of outcome data, selective outcome reporting, and potential other biases. Randomized clinical trials were initially viewed as high-quality evidence but may be downgraded due to potential bias, imprecision, inconsistency, indirectness, and publication bias. The Newcastle-Ottawa Scale (NOS) was used by M.H. and M.Z. to evaluate study quality for observational studies, categorizing them into low, moderate, and high quality based on selection, comparability, and outcomes. Studies scoring 0–3, 4–6, 7–9 were categorized as low, moderate, and high quality, respectively.¹⁵ The ROBIS tool was used to evaluate the level of bias in the included meta-analysis, while the AMSTAR-2 scale was used to assess their methodological quality. In cases of disagreement, another investigator (T.Z.) was brought in to discuss in order to reach a consensus.

Data extraction included authors, publication year, patient demographics, tumor stage, study design, sample size, intervention details, and survival outcomes. Survival data were analyzed using risk ratio (RR) for dichotomous data and hazard ratio (HR) for time-to-event outcomes. When both univariate and multivariate results were reported by the authors, we opted to use the multivariate analysis findings. Survival rates were directly extracted or estimated from survival curves using Getdata Graph Digitizer.¹⁶

Statistical analysis

Survival discrepancies based on different groups were analyzed using RRs or HRs with 95% confidence intervals (CIs). Study heterogeneity was assessed using the I² statistic, with fixed-effect models applied for I² < 50% and random-effect models otherwise. Subgroup analysis was performed according to study design (RCTs vs cohort studies), lymph node removal method (SLND vs nSLND), tumor stage, and methods for controlling confounding factors. Publication bias was evaluated using funnel plots and quantitatively assessed through Egger’s and Begg’s tests. If necessary, the trim-and-fill method was used, considering a p-value <0.05 as statistically significant. Analysis was performed using Stata 15.0 (Stata Corporation, College Station, TX, USA), and graphical data were presented using GraphPad Prism 9.0 (GraphPad Software, San Diego, CA, USA).

Results

The validation of uncertain resection

Characteristics of the included studies

A total of 123 articles were identified, of which 13 studies involving 37,149 patients met the inclusion criteria.^8,17–28 The PRISMA flow diagram illustrates the search process (Figure 1). For this meta-analysis, we searched Web of Science (n = 36), PubMed (n = 40), Embase (n = 107), Cochrane Library (n = 15), CNKI (n = 2), and Wanfang (n = 1). After removing 78 duplicates, 123 records were screened. We excluded 81 records (77 irrelevant, 4 reviews). Full texts of 42 reports were assessed; 30 were excluded (20 not meeting criteria, 8 no full text, 2 duplicates). One study was added from references and three from a previous review. Ultimately, 13 studies were included in the analysis. Key characteristics of the included studies—including data sources, study periods, sample sizes, 5-year OS/DFS rates, HRs for OS/DFS, and NOS quality scores are summarized in Table 1. All studies were rated as high quality based on NOS criteria (Figure S1(A), Table S1). Three studies focused on operable stage III patients, six included surgical patients across stages I–III, one targeted IA cases, and one examined patients with clinical N0 but pathological pN2 disease. Five-year OS rates were extracted from all studies, with three derived from survival curves using GetData Graph Digitizer. Direct HRs for OS were reported in 11 studies.

Figure 1.

Flow diagram showing the search process of articles included in section “The validation of uncertain resection.”

Table 1.

Main characteristics of 13 eligible studies enrolled in section “The validation of uncertain resection.”.

Author	Period span	Country	Sample size	Stage	R0(N, %)	Run(N, %)	R1/2(N, %)	5-Year OS rate (%)			HR of OS		5-Year RFS rate (%)			HR of DFS
Author	Period span	Country	Sample size	Stage	R0(N, %)	Run(N, %)	R1/2(N, %)	R0	Run	R1/2	HR (Run vs R0)	HR (R1/2 vs R0)	R0	Run	R1/2	HR (Run vs R0)	HR (R1/2 vs R0)
Gagliasso 2017¹⁷	1998–2007	Italy	1277	I–IV	1003 (78.54)	185 (14.49)	89 (6.97)	58.80	37.30	15.70	1.35 (1.10–1.66)	2.30 (1.79–2.96)	NA	NA	NA	NA	NA
Edwards 2020¹⁸	1999–2010	World database	14712	I–III	6070 (41.26)	8185 (55.63)	457 (3.11)	75.78	71.55	34.21	NA	NA	NA	NA	NA	NA	NA
Osarogiagbon 2020⁸	2009–2019	USA	3361	I–IV	1119 (33.29)	2044 (60.82)	198 (5.89)	64.00	54.00	33.00	1.36 (1.19–1.56)	2.18 1.75–2.71)	NA	NA	NA	NA	NA
Yun 2021¹⁹	2004–2018	Korea	1039	III	432 (41.58)	212 (20.40)	395 (38.02)	54.70	45.80	36.20	1.06 (0.85–1.34)	1.48 (1.22–1.81)	NA	NA	NA	NA	NA
Lee 2022²⁰	2003–2019	Korea	910	III	302 (33.19)	329 (36.15)	279 (30.66)	60.00	52.90	48.40	1.18 (0.93–1.51)	1.51 (1.17–1.96)	54.90	46.20	31.90	1.20 (0.94–1.52)	1.50 (1.17–1.52)
Wang 2022²¹	1999–2018	China	2782	I–III	1897 (68.19)	885 (31.81)	NA	82.12^a	67.04^a	NA	1.30 (1.09–1.56)	NA	74.16^a	60.11^a	NA	1.18 (1.05–1.38 )	NA
Ren 2022²²	2009–2013	China	5293	I–III	3819 (72.15)	1371 (25.90)	103 (1.95)	71.00	46.00	34.00	1.41 (1.28–1.54)	1.74 (1.36–2.22)	63.54	40.63	26.04	1.52 (1.38–1.67)	1.71 (1.33–2.19)
Kadomatsu 2022²³	2014–2015	Japan	350	I–III	194 (55.43)	156 (44.57)	NA	79.38^a	75.00^a	NA	NA	NA	NA	NA	NA	NA	NA
Chen 2023²⁴	2013–2018	China	682	I–III	489 (71.70)	110 (16.13)	83 (12.17)	67.60	49.80	45.70	1.54 (1.02–2.33)	1.45 (1.00–2.11)	60.30	37.10	30.00	1.59 (1.09–2.31)	1.77 (1.27–0.47)
Wen 2023²⁵	2010–2019	China	5200	I–III	3228 (62.08)	1727 7(33.21)	245 (4.71)	80.13^a	69.32^a	37.16	1.40 (1.22–1.60)	NA	68.12	55.60	25.93	1.29 (1.15–1.43)	NA
Marziali 2023²⁶	2015–2020	Italy	68	cN0	37 (54.41)	31 (45.59)	NA	78.00	40.00	NA	2.00 (0.70–6.10)	NA	69.00	20.00	NA	3.20 (1.40–7.40)	NA
Liu 2024²⁷	2015–2018	China	468	III	249 9 (53.21)	219 (46.79)	NA	59.20	42.10	NA	1.45 (1.07–1.99)	NA	49.80	27.20	NA	1.26 (0.94–1.68)	NA
Si 2024²⁸	2014–2017	China	1007	IA	408 (40.52)	599 (59.48)	NA	94.30	92.40	NA	2.01 (1.16–3.47)	NA	91.80	88.30	NA	1.58 (1.02–2.46)	NA

The survival rate was extracted from the survival curves by Getdata Graph Digitizer.

DFS, disease-free survival; HR, hazard ratio; NA, not available; NOS, Newcastle-Ottawa Scale; OS, overall survival; RFS, relapse-free survival.

OS assessment based on IASLC R classification

Significant heterogeneity was detected among the 13 pooled studies (I² = 81.2%). A random-effects model revealed a pooled RR of 1.31 (95% CI 1.19–1.44) indicating superior 5-year OS rate in the R0 group compared to the Run group (Figure 2(a)). Eleven studies provided HR data comparing R-un and R0 resections, demonstrating worse OS for Run with no significant heterogeneity (HR = 1.34, 95% CI 1.27–1.42, p < 0.001, Figure S2(A)). Patients in the R1/2 group had a 1.98-fold higher mortality risk than the R0 group (RR = 1.98, 95% CI 1.59–2.47, Figure S2(C)).

Figure 2.

Forest plot of survival difference between Run and R0. (a) 5-Year OS difference. (b) 5-Year DFS difference.

DFS assessment based on IASLC R classification

Eight studies reported 5-year DFS rates comparing Run and R0 resections. Meta-analysis showed significantly lower DFS in the R-un group (RR = 1.43, 95% CI 1.28–1.60, Figure 2(b)). Pooled HRs from eight studies further confirmed worse DFS outcomes for Run (HR = 1.33, 95% CI 1.20–1.47, Figure S2(B)).

Subgroup and sensitivity analysis

Strong heterogeneity was observed among the 13 studies that were included in the present 5-year OS rate synthesis (I² = 81.2%). A subgroup analysis was conducted to explore the sources of heterogeneity in the OS analyses. Participants were divided into six predefined subgroups based on factors such as gender, age, data source, stage, percentage of patients with HMLI, and percentage of patients in N0. The results showed that there was significant heterogeneity in 5-year OS among the subgroups. Notably, studies from China showed more inferior OS in Run group compared to those from non-China (China: RR = 1.48, 95% CI 1.38–1.59; non-China: RR = 1.16, 95% CI 1.11–1.21), with no significant heterogeneity in either group (China, I² = 0%; non-China, I² = 26.7%). Despite the different data source, the trends in the results of the two groups were consistent (Figure S3(A)). Sensitivity analysis was conducted to assess whether the results were influenced by any single study. The studies from Edwards and Ren were identified as the main source of heterogeneity in overall studies^18,22 (Figure S3(B)). In Edwards’ study, patients reclassified as Run group had the highest 5-year OS rate, showing the least difference compared to those in complete resection group. In this study, 87% of the cases were clinically staged as I or II, and 8% as IIIA. According to most current studies, for early-stage patients, such as those with stage IA or ground-glass nodules, complete lymph node dissection does not significantly impact OS. Therefore, in section “Extent of removal for mediastinal lymph node stations,” we will discuss in detail the impact of different lymph node dissection methods on patient prognosis. As mentioned in other articles, the long-term prognosis of patients with HMLI is inferior to those with incomplete lymph node dissection.^8,19 In Ren’s study, the proportion of HMLI was 37.27%, the highest among the enrolled studies. Hence, in Ren’s study, the comparison between Run and R0 resection showed the greatest difference. After excluding these two studies, the pooled RR remained unchanged (RR = 1.43; 95% CI: 1.29–1.58). Thus, we retained the two studies in the final analysis and applied the Mantel-Haenszel method for the final synthesis. In the combined analysis of HR, no significant heterogeneity was observed (Figure S3(C)).

Publication bias

For the OS synthesis, the potential publication bias was evaluated using a funnel plot. Neither the visual inspection of the funnel plots nor Egger’s test revealed any significant publication bias, either in the 5-year OS synthesis or in the HR synthesis derived from OS (Figure S3(D) and (E); Egger’s test, p = 0.620; Begg’s test, p = 0.669).

Extent of removal for mediastinal lymph node stations

Characteristics of the included research

In this section, we reviewed the surgical procedures in each study, as the definition and execution of SLND varied across the included studies, as did the lymph node maps used. We established SLND standards based on the ESTS criteria, which require complete removal of all lymph nodes and surrounding fat tissue between the anatomical landmarks of the lung lobe where the tumor is located. This includes a minimum of six stations, with at least three stations must be from mediastinal lymph node zones, including station 7. We identified 1695 articles and ultimately selected 30 studies involving 11,766 patients for further analysis^29–58 (Figure 3). General characteristics of these studies are summarized in Table 2. These studies included 4 RCTs and 26 observational studies, with 2 studies derived from the registry databases. The data sources were from China, Japan, Korea, France, and Italy. Of these eligible studies, 16 compared long-term survival between SLND and MLNS, while 10 compared SLND with L-SND. The general characteristics of these studies are summarized in Table 2.

Figure 3.

Flow diagram showing the search process of articles included in section “Extent of removal for mediastinal lymph node stations.”

Table 2.

Main characteristics of 30 eligible studies enrolled in section “Extent of removal for mediastinal lymph node stations.”.

Author	Period span	Country	Stage	Study design	Sample size	SLND (N, %)	nSLND (N, %)	Control	5-Year OS rate (%)		HR of OS (reference: nSLND)	5-Year DFS rate (%)		HR of DFS (reference: nSLND)
Author	Period span	Country	Stage	Study design	Sample size	SLND (N, %)	nSLND (N, %)	Control	SLND	nSLND	HR of OS (reference: nSLND)	SLND	nSLND	HR of DFS (reference: nSLND)
Sugi 1998²⁹	1985–1992	Japan	cI	RCT	115	59	56	MLNS	81.40	83.90	NA	NA	NA	NA
Wu 2002³⁰	1989–1995	China	cI	RCT	156	58	98	MLNS	82.16	57.49	NA	NA	NA	NA
Aoki 2003³¹	1985–2001	Japan	cI	Os	49	22	27	nSLND	55.50	44.80	NA	NA	NA	NA
Lardinois 2005³²	1995–1996	Swiland	pI	Os	46	24	22	MLNS	61.28^a	45.23^a	NA	61.20^a	36.29^a	NA
Okada 2006³³	1997–2002	Japan	c and pI	Os	735	358	377	LSND	79.70	83.20	1.32 (0.93–1.86)	73.40	76.40	1.07 (0.80–1.45)
Ma 2008³⁴	1998–2002	China	cIA	Os	105	42	63	MLNS	80.40	61.20	NA	75.50	59.50	NA
Su 2008³⁵	1994–2003	China	cI	OS	319	139	180	MLNS	76.40	65.90	0.55 (0.35–0.86)	NA	NA	NA
Takizawa 2008³⁶	1994–2000	Japan	cI	Os	119	58	61	MLNS	78.00	76.20	NA	NA	NA	NA
Zhang 2008³⁷	1997–2003	China	IA	OS	98	54	44	MLNS	80.40	53.00	0.35 (1.15–9.01)	NA	NA	NA
Okami 2009³⁸	1999	Japan	cI	Os	367	127	240	nSLND	53.70	56.80	1.09 (0.75–1.60)	NA	NA	NA
Chen 2010³⁹	2000–2003	China	cI	Os	97	28	69	MLNS	84.36	68.54	0.48 (0.26–0.90)	NA	NA	NA
Hu 2010⁴⁰	1998–2004	China	cI	Os	73	22	51	MLNS	81.81	80.39	NA	NA	NA	NA
Chen 2012⁴¹	1998–2005	China	cI	Os	298	148	150	LSND	73.60	68.70	NA	NA	NA	NA
Jiang 2013⁴²	2005–2008	China	cI	Os	403	309	94	LSND	74.60	68.50	NA	74.40	69.10	NA
Ma 2013⁴³	2004–2008	China	cI	RCT	96	51	45	LSND	64.90	69.70	NA	60.80	66.00	NA
Zhang 2013⁴⁴	2006–2007	China	pI	RCT	76	37	39	MLNS	77.18	60.19	NA	NA	NA	NA
Jeon 2014⁴⁵	2000–2010	Korea	cI	Os	211	90	121	MLNS	90.60	81.76	0.32(0.11–0.89)	NA	NA	NA
Wu 2014⁴⁶	2003–2007	China	cI	Os	112	66	46	MLNS	77.50	69.60	NA	28.60	37.90	NA
Guerrera 2015⁴⁷	2004–2012	Italy	cI	Os	848	710	138	MLNS	NA	NA	0.58(0.42–0.81)	NA	NA	NA
Hishida 2016⁴⁸	2004	Japan	cI	Os	4652	3531	1121	LSND	79.20	83.70	NA	73.10	78.00	NA
Shen 2017⁴⁹	1998–2005	China	cI	Os	317	161	156	MLNS	73.90	62.80	0.71 (0.52–0.98)	NA	NA	NA
Wang 2019⁵⁰	1999–2014	China	pI	Os	468^b	234	234	MLNS	83.06	74.10	0.77 (0.53–0.99)	74.92	67.42	0.71 (0.51–0.98)
Guo 2021⁵¹	2010–2016	China	cIA	Os	191	104	87	nSLND	98.10	93.80	0.77 (0.53–0.99)	95.80	87.64	0.77 (0.53–0.99)
Hattori 2021⁵²	2008–2016	Japan	cI	Os	228^b	114	114	LSND	78.80	79.90	0.79 (0.53–1.19)	70.40	66.50	NA
Katsumata 2021⁵³	2003–2016	Japan	cI	Os	228^b	171	57	MLNS	74.00	82.40	NA	NA	NA	NA
Zhao 2021⁵⁴	2014–2017	China	cIA	Os	200^b	100	100	LSND	96.90	96.70	0.41 (0.11–1.49)	95.50	95.60	NA
Kamigaichi 2023⁵⁵	2006–2014	Japan	cI	Os	751^b	372	379	LSND	83.06^a	84.70^a	NA	75.99	76.70	NA
Sui 2023⁵⁶	2011–2016	China	cI	Os	210	72	138	LSND	98.60	97.20	1.04 (0.11–10.00)	93.00	95.70	1.07 (0.11–10.31)
Wo 2023⁵⁷	2008–2016	China	cI	Os	132^b	66	66	LSND	90.33^a	94.56^a	NA	85.60^a	89.75^a	NA
Woo 2023⁵⁸	2013–2019	Korea	cI	Os	66^b	33	33	MLNS	100.00	100.00	NA	97.13	94.98	NA

The survival rate was not directly in article but could be extracted from the survival curves by Getdata Graph Digitizer.

Propensity score analysis was used in study.

DFS, disease-free survival; HR, hazard ratio; NA, not available; NOS, Newcastle-Ottawa Scale; nSLND, nonsystematic lymph node dissection; Os, observational study; OS, overall survival; RCT, randomized controlled trial; SLND, systematic lymph node dissection.

Risk of bias and quality of studies

All observational studies were rated as moderate to good quality by NOS grading system (Figure S1(A), Table S1). We identified four RCTs involving 443 patients. Performance bias during lung resection was unavoidable. Two trials failed to perform an intention-to-treat analysis, and the methods for sequence generation and allocation concealment were additional sources of bias (Figure S1(B)).

Assessment of OS

Among the 30 enrolled studies, 29 studies provided 5-year OS rates, and 13 studies provided HRs. We extracted 5-year OS rates as primary endpoint directly from the full text of 26 studies and used Getdata software to convert survival curves into 5-year OS rates in 3 studies.^32,55,57 The combined analysis (Figure 4(a)) showed no significant difference in OS between SLND and MLNS (RR = 0.97, 95% CI 0.90–1.06, I² = 31.5%, Figure 4(a)). Given the completely different nature of these study designs, we separately report the meta-analysis results of observational studies and RCTs. The analysis results of observational studies show that SLND provides comparable survival benefits to MLNS (RR = 0.99, 95% CI 0.91–1.08, Figure 5(a)). The analysis of four RCT studies also shows similar results (RR = 0.77, 95% CI 0.50–1.18, Figure 5(b)). This result suggests that SLND is an alternative to MLNS in clinical stage I NSCLC.

Figure 4.

Forest plot of survival difference between SLND and nSLND in stage cI NSCLC patients. (a) 5-Year OS difference. (b) 5-Year DFS difference.

Figure 5.

Forest plot of 5-year OS difference between SLND and nSLND in subgroups of stage cI NSCLC patients. (a) 5-Year OS difference in observational studies. (b) 5-Year OS difference in RCT studies. (c) 5-Year OS difference in L-SND group. (d) 5-Year OS difference in MLNS group. (e) 5-Year OS difference in tumor ⩽2 cm. (f) 5-Year OS difference in studies with equilibrated baseline characteristics.

Subgroup analyses were conducted based on control arm (MLNS or L-SND) and tumor stage. Given the different definitions of stage I across TNM editions, we performed subgroup analyses based on tumor diameter. Eight studies reported 5-year OS rates for tumors ⩽2 cm, with no significant difference observed (RR = 1.20, 95% CI 0.88–1.64, Figure 5(c)). Interestingly, subgroup analyses based on surgical procedure showed that SLND was associated with higher 5-year OS compared to MLNS (RR = 0.71 95% CI 0.63–0.80, Figure 5(d)), while SLND appeared to offer worse survival outcomes than L-SND (RR = 1.13, 95% CI 1.01–1.26, Figure 5(e)).

We conducted a comprehensive assessment of baseline characteristics in the enrolled literature comparing SLND and nSLND. The key characteristics examined included age, gender, pathological type, tumor size, and surgical method. Of the studies reviewed, 11 exhibited baseline imbalances. Among these, propensity score matching was employed in seven observational studies to adjust for potential confounding factors.^{50,52–55,57} However, the remaining four studies did not employ propensity score matching.^42,45,51,58 Given this discrepancy, we excluded these four studies from further analysis. The reanalysis yielded a pooled RR of 0.99 (95% CI 0.91–1.08, Figure 5(f)).

Assessment of DFS

Among the enrolled studies, 15 provided DFS data (15 studies with 5-year DFS rates and 4 with HR). We extracted the 5-year DFS rates as the primary outcome directly from the full text of 13 studies. For the remaining two studies, we used Getdata software to transform the survival curves into 5-year DFS rates.^57,59 The pooled results of the 5-year DFS rates are shown in Figure 4(b). No significant heterogeneity was detected among these 15 clinical studies (I² = 12.3%). A fixed-effect model was applied for the DFS analysis. No significant difference was observed between SLND and nSLND with a pooled RR of 1.06 (95% CI 0.97–1.15, Figure 4(b)).

Publication bias

Risk of publication bias was assessed using a funnel plot in OS synthesis. The plots were asymmetrical on visual inspection, suggesting potential risks of publication biases (Figure S4(A)). The p value of Begg’s test was 0.358, of Egger’s test was 0.002. By trim-and-fill analyses, seven hypothesized studies were imputed. Meta-analysis incorporating these seven studies did not significantly change the results (RR, 0.96, 95% CI: 0.85–1.07, Figure S4(B)).

Highest mediastinal lymph node involvement

Characteristics of the included research

A total of 13 studies were identified, involving 6943 patients with operable pN2-stage NSCLC. These studies compared survival data that were compared between patients with and without HMLI metastasis^{8,18,22,26,59–67} (Figure 6, Table 3). All of the enrolled studies were observational and conducted in China, Korea, Japan, the USA, and Italy. With the exception of Marziali’s study, which included patients with clinical N0 status between 2015 and 2020, the remaining studies focused on patients with pN2 NSCLC. Regarding survival data, all of the 13 studies reported HR based on OS data and 6 studies provided data on DFS data. The quality assessment of the enrolled studies is shown in Figure S1(A) and Table S1.

Figure 6.

Flow diagram showing the search process of articles included in section “Highest mediastinal lymph node involvement.”

Table 3.

Main characteristics of 13 eligible studies enrolled in section “Highest mediastinal lymph node involvement.”.

Author	Period span	Country	Stage	Sample size	HMLI+(N, %)	HMLI−(N, %)	5-Year OS rate (%)		HR of OS	5-Year DFS rate (%)		HR of DFS
Author	Period span	Country	Stage	Sample size	HMLI+(N, %)	HMLI−(N, %)	HMLI+	HMLI−	HR of OS	HMLI+	HMLI−	HR of DFS
Sakao 2006⁶⁰	1996–2003	Japan	pN2	53	14 (26.42)	39 (73.58)	0.00	52.00	3.23 (1.15–9.01)	NA	NA	NA
Zheng 2011⁵⁹	1999–2005	China	pN2	549	246 (44.81)	303 (55.19)	13.00	29.00	1.58 (1.22–2.06)	NA	NA	NA
Nakagiri 2011⁶¹	1992–2007	Japan	pN2	117	73 (62.39)	44 (37.60)	34.70	52.60	1.75 (0.85–3.60)	21.10	41.20	NA
Park 2019⁶²	2000–2015	Korea	pN2	339	142 (41.89)	197 (58.11)	40.00	42.10	1.02 (0.75–1.37)	30.10	31.80	NA
Zhang 2019⁶³	2006–2016	China	pN2	288	132 (45.83)	156 (54.17)	NA	NA	1.01 (0.67–1.52)	NA	NA	1.12 (0.79–1.60)
Osarogiagbon 2020⁸	2009–2019	USA	pN2	196	119 (60.71)	77 (39.29)	NA	NA^a	1.08 (0.71–1.67)	NA	NA	NA
Edwards 2020¹⁸	1999–2010	World database	pN2	1800	738 (41.00)	1062 (59.00)	NA	NA^a	1.32 (1.15–1.52)	NA	NA	NA
Wang 2021⁶⁴	2004–2015	China	pN2	218	109 (50.00)	109 (50.00)	NA^a	NA^a	1.00 (0.70–1.43)	NA	NA	NA
Ren 2022²²	2009–2013	China	pN2	1177	511 (43.42)	666 (56.58)	NA^a	NA^a	1.23 (1.07–1.42)	NA^a	NA^a	1.21 (1.06–1.39)
Marziali 2023²⁶	2015–2020	Italy	cN0 →pN2	68	31 (45.59)	37 (54.41)	40.00	78.00	2.00 (0.70–6.10)	20.00	69.00	3.20 (1.40–7.40)
Guo 2023⁶⁵	2012–2016	China	pN2	1409	677 (48.05)	732 (51.95)	NA^a	NA^a	1.16 (1.00–1.35)	NA^a	NA^a	1.27 (1.10–1.47)
Liu 2023⁶⁶	2007–2016	Italy	pN2	261	128 (49.04)	133 (50.96)	NA^a	NA^a	1.31 (0.96–1.78)	69.00	20.00	1.39 (1.03–1.87)
Liu 2024²⁷	2015–2018	China	pN2	468	219 (46.79)	249 (53.21)	42.10	59.20	1.70 (1.19–2.42)	27.20	49.80	1.46 (1.06–2.02)

NA, the survival rate was not directly in article but could be extracted from the survival curves by Getdata Graph Digitizer.

DFS, disease-free survival; HR, hazard ratio; NA, not available; NOS, Newcastle-Ottawa Scale; OS, overall survival.

Assessment of OS and DFS

The meta-analysis results of the combined HR and 95% CI are presented in Figure 6. No significant heterogeneity was detected among these clinical studies included in the analysis (I² = 8.3%). Therefore, a fixed-effect model was applied. The results indicated that HMLI was an indicator of inferior prognosis, with a pooled HR of 1.22 (95% CI 1.14–1.31, Figure 7(a)). Additionally, five studies provided relevant HR data for the 5-year DFS rate of HMLI in pN2 NSCLC. The presence of HMLI was associated with shorter DFS (HR = 1.25, 95% CI 1.14–1.36, Figure 7(b)).

Figure 7.

Forest plot of survival difference between HMLI and nHMLI. (a) 5-Year OS difference. (b) 5-Year DFS difference.

Publication bias

The risk of publication bias was assessed using a funnel plot (Figure S4(C)). No evidence of significant publication bias was detected, as indicated by visual inspection of the funnel plots and formal statistical tests (Egger’s test, p = 0.331; Begg’s test, p = 0.428).

Prognostic significance of PLC

Malignant cells identified in pleural lavage fluid in patients without pleural effusion indicate micrometastasis, with the rate in operable NSCLC ranging from 3% to 13%. To date, four meta-analyses have reported on the prognostic significance of PLC, consolidating the conclusion that PLC+ is a strong predictor of survival in patients with lung cancer^9–12 (Table 4). We conducted a bias risk assessment using the ROBIS tool, with two articles rated as high risk and two articles rated as low risk. AMSTAR2 was used to evaluate the methodological quality, and all four articles were rated as moderate quality (Table S4). Despite the differences in bias risk and methodological quality among these four meta-analyses, the overall results consistently support the conclusion that positive PLC is a significant factor for poor prognosis, especially in stage I patients. Although there are limitations in bias risk and methodological quality, the consistency of the results from these four meta-analyses provides a certain degree of reliability to the conclusion.

Table 4.

Main result of four meta-analysis about PLC value in operable NSCLC.

Year	2008⁹		2010¹⁰	2012¹¹	2016¹²
Author	Li		Lim	Saso	Wang
Number of studies	15		31	17	28
Sample size	6391		8763	4450	20,714
Time of PLC	Before resection	After resection	NA	Before resection	Before resection	After resection
Positive rare (n, %)	421 (7.00%)	107 (6.10%)	511 (5.80%)	709 (7.14%)	1434 (7.30%)
HR of OS (PLC+ vs PLC−)	7.06 (5.04–9.90)	8.47 (6.12–11.73)	1.47 (1.29–1.67)	2.08 (1.71–2.52)	2.89 (2.48–3.37)	2.70 (1.90–3.83)
5 year-OS in stage I (PLC+ vs PLC−)	NA		53.60% vs 76.91%	4.20 (2.65–6.65)	3.29 (2.55–4.25)	4.85 (2.31–10.20)

CI, confidence interval; HR, hazard ratio; NA, not available; NSCLC, non-small cell lung cancer; OS, overall survival; PLC, pleural lavage cytology.

Current studies can be categorized into two groups based on the timing of intraoperative pulmonary lavage: before and after lung resection. In 2008, Li et al. conducted a meta-analysis of 15 studies involving 6391 patients to assess the prognostic value of PLC on OS. Their results indicated that positive PLC, either before or after lung cancer resection, adversely impacted OS, with HR of 7.06 (95% CI, 5.04–9.90) and 8.47 (95% CI, 6.12–11.73), respectively.⁹ In 2010, the International PLC Collaboration Group conducted a meta-analysis of 31 clinical studies from 22 medical centers, comprising individual data from 8763 patients, of whom 511 (5.8%) had positive PLC results. This analysis revealed that patients with positive PLC had lower OS rate, with the most significant difference observed in stage I patients.¹⁰ A 2012 meta-analysis focusing on the prognostic value of preoperative PLC results, including 4450 patients, found that positive preoperative PLC was significantly associated with an increased risk of local and distant cancer recurrence. Among these, nine studies presented survival data for stage I patients, indicating a more pronounced effect in this group.¹¹ An updated pooled analysis in 2016 with a total of 20,714 patients. Subgroup analysis of stage I patients in Wang et al.’s¹² study also showed that pre-operation, post-operation, and combined PLC were associated with adverse survival rate, with HR of 3.29 (95% CI 2.55–4.25), 4.85 (95% CI 2.31–10.20), and 3.16 (95% CI 2.53–3.94), respectively. These consistent results suggest that PLC has stronger prognostic significance in stage I patients compared to the overall population.

Nevertheless, PLC status has not yet been incorporated into the TNM classification. Mizuno et al.’s⁶⁷ study found that the prognosis in patients with pI NSCLC with PLC+ was comparable to that in patients with p-IIIA disease. Several reports have recommended that NSCLC be staged as pT3 for all patients with a PLC+ status, while others have proposed increasing the pT stage by 1.^10,68–70 Ogawa et al. conducted a study on the clinical benefits of adjuvant chemotherapy in 53 patients with positive PLC findings who underwent curative-intent surgery for lung adenocarcinoma. They found a significant difference in the 5-year DFS when comparing patients treated with surgery and adjuvant chemotherapy versus those treated with surgery alone.⁷¹ PLC+ represents a negative prognostic factor that is likely to influence the long-term outcomes but not the treatment strategy of care in patients with Stage pII-III and selected cases of Stage pIB, as adjuvant therapy is already recommended for these patients. For patients with PLC+ in stage pI, it is important to evaluate the necessity of adjuvant therapy.

As discussed above, for early-stage patients, particularly those in stage I, positive PLC had greater prognostic value and is closely related to postoperative malignant pleurisy and distant recurrence. However, in clinical practice, methods such as the volume of lavage fluid used, type of solution, and timing of lavage still require standardization. Whether adjuvant therapy should be recommended for stage I patients with positive PLC needs validation through multicenter prospective clinical trials.

Discussion

More than a decade ago, the IASLC proposed definitions for complete, incomplete, and uncertain resections in lung cancer. This definition was based on the previous R classification of the UICC and considered various aspects related to surgical treatment, including intraoperative nodal assessment, assessment methods, criteria for pathological absence of nodal disease, and the prognostic significance of PLC and CIS.⁵ Given the distinct prognosis, Run was introduced in the seventh edition of the TNM classification.⁴ Distinguishing between complete resection and Run is clinically important and provides additional information beyond the R descriptor alone. In recent years, several independent groups have published articles on the prognostic value of the IASLC resection definitions, using institutional, international, and population-based databases. However, the results of these studies have been inconsistent. Therefore, we conducted the system review and meta-analysis to synthesize all available evidence. As far as I know, this is the first meta-analysis on this topic.

In this meta-analysis, we included 13 observational studies with sample sizes ranging from 68 to 14,712. The initial combined analysis showed a RR of 1.31 for 5-year OS (95% CI 1.19–1.44) and 1.43 for 5-year DFS (95% CI 1.28–1.60). These results indicate that Run is associated with inferior outcomes compared to complete resection. Thus, the IASLC R classification better identifies risk than the previous classification. Given the survival deficit observed in the Run group, additional treatment protocols after surgery should be considered.²² Consistent with proposals for revising the classification of residual tumor, the OS of patients in the Run category consistently falls between that of the R0 and R1/2 groups. These differences in OS remained significant even after adjusting for potential confounders through multivariate analysis.⁷² Due to the independence of the four elements that comprise the R descriptors, significant heterogeneity exists when analyzed together. Although evidence for specific descriptors is generally limited and partially consistent, we discuss each element individually for its impact on the prognosis of early-stage NSCLC.

Our research aimed to examine the differences in long-term survival between SLND and L-LA in patients with clinical stage I NSCLC. We found that patients with stage cI NSCLC had similar 5-year OS rates when comparing SLND to L-LA (RR = 0.97, 95% CI 0.90–1.06). We included 31 studies involving 12,231 patients from 1985 to 2023, based on TNM editions 5–8. To minimize the impact of varying tumor diameter definitions in stage I, we specifically analyzed the survival data from stage IA2 patients (tumor diameter ⩽2 cm). The results showed no heterogeneity among the studies and no difference in 5-year survival between SLND and L-LA. Stratified analyses based on study design and statistical methods yielded consistent results across subgroups, indicating that SLND provides equivalent OS benefits compared to L-LA. Interestingly, subgroup analysis based on the method of L-LA showed that SLND provided equal OS benefits, while L-SND seemed to offer a higher 5-year OS rate. The ACOSOG Z0030 trial is currently the largest prospective multicenter randomized controlled study comparing SLND and sampling. There were no statistically significant differences between the two groups in terms of postoperative median OS, 5-year DFS, or local and distant recurrence. Since survival and recurrence data of stage cI were not included in the published result, this RCT study was not included in section “Extent of removal for mediastinal lymph node stations.” Stratification based on T1 showed no difference in long-term survival between MLNS and SLND. Based on our current meta-analysis results, it is speculated that including this study would not affect the overall conclusion.

The pooled results in section “Extent of removal for mediastinal lymph node stations” should be interpreted with caution. First, surgical techniques for SLND are not yet standardized. The extend of SLND performed is not based on objective or reproducible criteria but primarily depends on the surgeon’s preference and the patient’s condition. There is a lack of details regarding partial or complete removal of lymph nodes and the number of lymph nodes removed from different stations in the studies reviewed. Secondly, although authors have used the term “systematic” to define lymph node dissection, their definitions of SLND are inconsistent. Organizations such as the IASLC, NCCN, ACOSOG, and ESTS have each published different standards for SLND. For example, in some Japanese studies involving bilateral upper lobe lung tumors, removal of stations 8 and 9 is not mandatory. Given the potential bias introduced by these inconsistent definitions, we established, the following criteria for SLND to minimize the confounding: complete removal of all mediastinal lymph nodes and surrounding fat tissue between the anatomical landmarks of the lung lobe where the tumor is located, and SLND involving the removal of at least six lymph node stations, with at least three groups coming from mediastinal lymph nodes, including station 7. The standard based on the number of lymph node dissections was not adopted in this section of the study.^73,74 Even with a relatively consistent standard, some potential biases may still be inevitable. Lastly, certain high-risk factors for postoperative recurrence, such as spread through air spaces (STAS), have been identified. Several systematic reviews and meta-analyses have confirmed that tumor STAS is associated with shorter DFS and OS in stage I NSCLC.^75,76 Some surgeons suggest that when STAS is found in sublobar resection specimens, subsequent lobectomy should be performed in T1N0M0 adenocarcinoma patients. Despite these limitations, the results of the meta-analysis in section “Extent of removal for mediastinal lymph node stations” indicate that less rigorous nodal assessment was not associated with worse prognosis, particularly, in early-stage NSCLC or in more favorable histologic subtypes. Considering the lower postoperative complications and the effectiveness of tumor-draining lymph nodes for immunotherapy, L-LA may be prioritized in less invasive stage cIA patients.

The presence of HMLI is an important category of Run. The results of this meta-analysis showed that both OS and DFS were inferior in the metastasis group compared to the nHMLI group. For this population, more aggressive postoperative follow-up and adjuvant therapy may be considered. Several retrospective studies have shown significant survival benefits with postoperative radiotherapy (PORT) in pN2-III patients.^77,78 Nevertheless, recent RCTs, such as the Lung ART and PORT-C trials, showed that PORT did not improve long-term survival in pN2-III NSCLC.^22,79 The Lung ART study included 501 patients, with 203 being Run patients. However, subgroup analysis of different surgical resection statuses, including patients with HMLI, has not been published.

Under the current circumstances, sub-classification of lymph node metastasis in patients with pN2-III NSCLC probably has more clinical implications for prognosis, including multiple station metastases of mediastinal lymph nodes, HMLI, and distribution of metastatic nodes (skip N2 or non-skip N2).⁶¹ To my furthest knowledge, only two studies have reported survival data on PORT for HMLI in pN2 NSCLC. In Guo et al.’s retrospective study, patients with positive HMLI had worse prognosis in pN2 NSCLC. For these patients, PORT significantly improved both relapse-free survival (RFS; HR = 0.699, 0.59–0.83) and OS (HR = 0.79, 0.66–0.95).⁶⁵ Similar results were reported by Deng et al.⁸⁰ in a study of 140 patients in pN2-III. As suggested by this meta-analysis, HMLI is an independent factor for poorer OS and DFS, and further research is needed to clarify whether this population could benefit from PORT.

Bronchial margin involvement with CIS is a rare occurrence, with reported incidences ranging from 0.02%–2.4% in cohorts of 255–4493, totaling 138 cases (0.9%) out of 16,000 reported resections.⁸¹ In several validation studies mentioned above, the incidence of CIS was 0.09% (13/14,293), 0 (0/3361), 0 (0/910), and 0.03% (2/5299) separately, with no reported survival data.^8,18,20,22 Due to its low incidence, stratified analysis is not feasible in current large cohort or database studies. In a study of 1277 consecutive NSCLC patients, 1003 (78.5%) underwent complete resection, 185 (14.5%) had uncertain resection, and 89 (7.0%) had incomplete resection. Five patients showed in situ cancer at the bronchial margin, with a 5-year survival rate of 40.0%. The highest mediastinal lymph node positivity and less rigorous lymph node assessment had 5-year survival rate of 28.8% and 44.2%, respectively.¹⁷ Based on the available literature data, the presence of CIS does not appear to influence the prognosis of patients after surgical resection.

According to the standards established by IASLC in 2005, lobectomy is mandatory for the definition of complete resection. Since the 1990s, the pathological types of early lung cancer have gradually changed due to the increasing detection of ground-glass opacities (GGOs). Preoperative thin-layer CT can predict the malignancy and lymph node metastasis of GGOs, allowing for sublobar resection in cases of low-grade malignant nodules identified through imaging. This raises questions about the necessity of lobectomy. After Japan Clinical Oncology Group0201 (JCOG0201) established radiologically noninvasive criteria for lung adenocarcinoma, JCOG and the West Japan Oncology Group conducted a series of prospective imaging-guided studies. These studies proposed more surgical options for some peripheral lesions, such as segmentectomy and wedge resection.^82–86 Precise segmentectomy enables the simultaneous management of multiple lung nodules while maintaining the option for additional treatment for potential second primary cancer, respiratory diseases, or cerebrovascular diseases. This aspect was not addressed in the IASLC definitions of resection completeness. With evolving surgical techniques for lung cancer, a standardized approach to the extent and invasiveness of nodal resection may need revision, as suggested by Lee et al.⁸⁷

The presence of circulating tumor cells or tumor DNA (ctDNA) in the bloodstream indicates the existence of tiny residual lesions, and their persistence or recurrence implies a poor prognosis after resection. When the IASLC definition was published, signs of these minor residual diseases were not considered. A systematic review of 13 eligible studies demonstrated a poor prognosis in patients with postoperative molecular residual disease (MRD). Patients with positive MRD were most likely to benefit from adjuvant treatment in terms of RFS.⁸⁸ Similarly, the presence of circulating tumor cells (CTCs) at the baseline or postoperative period was associated with an increased risk of recurrence and death compared with the absence of CTC in lung cancer patients.⁸⁹ Cancer cells or their fragments in the blood could be considered as the cells in other body spaces or anatomic structures. They are in a fluid tissue, blood, in a well-defined anatomic system, the circulatory, but to date, they are not coded in the TNM classification. Rami-Porta⁹⁰ have suggested adding the letter B, for blood, to the TNM (TNMB) as a new component of the classification. That would certainly complement the TNM classification, but liquid biopsy is not universally available and perhaps it is too early to consider it in the TNM classification. It also requires more research to quantify the amount of cells or of circulating tumor DNA or RNA that make a difference in prognosis. However, the presence of tumor material in the blood after treatment should be considered as minimal residual disease, although more research is needed to see how to classify it and how to manage it.

Conclusion

IASLC has updated the definition of complete resection and reclassified the surgical resections into complete, uncertain, and incomplete resections. This updated classification has been validated in several studies, demonstrating its prognostic value. Our literature review and meta-analysis in section “Extent of removal for mediastinal lymph node stations” compared the prognosis of Run with R0, revealing that L-LA could be a viable option for some favorable early-stage NSCLC patients. Additionally, the presence of HMLI indicates poorer survival in pN2-III patients, while PLC is associated with a worse prognosis in operable NSCLC, particularly in stage I. To our knowledge, the proposed R-classification for the ninth edition of lung cancer stage classification maintains the same overall structure.⁷² We believe this review will offer a foundation for further refining the R-classification, despite the limitations of existing data. Moreover, with advancements in imaging technology, more precise preoperative staging, and the accumulation of liquid biopsy data, the revision of IASLC definitions should be more comprehensive. This includes incorporating the appropriate extent of surgical resection, STAS, and the presence of CTCs/ctDNA. The definition of resection completeness will continue to evolve in line with emerging evidence, aiding in practical use and encouraging further research.

Supplemental Material

sj-docx-1-tam-10.1177_17588359251344789 – Supplemental material for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis

Supplemental material, sj-docx-1-tam-10.1177_17588359251344789 for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis by Mingming Hu, Xiaomi Li, Maike Zheng, Jiaqi Yu, Baohua Lu, Ying Wang, Xiaoqing Cao, Chongyu Su, Yujie Dong, Xu Zhang and Tongmei Zhang in Therapeutic Advances in Medical Oncology

Supplemental Material

sj-docx-3-tam-10.1177_17588359251344789 – Supplemental material for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis

Supplemental material, sj-docx-3-tam-10.1177_17588359251344789 for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis by Mingming Hu, Xiaomi Li, Maike Zheng, Jiaqi Yu, Baohua Lu, Ying Wang, Xiaoqing Cao, Chongyu Su, Yujie Dong, Xu Zhang and Tongmei Zhang in Therapeutic Advances in Medical Oncology

Supplemental Material

sj-pdf-2-tam-10.1177_17588359251344789 – Supplemental material for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis

Supplemental material, sj-pdf-2-tam-10.1177_17588359251344789 for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis by Mingming Hu, Xiaomi Li, Maike Zheng, Jiaqi Yu, Baohua Lu, Ying Wang, Xiaoqing Cao, Chongyu Su, Yujie Dong, Xu Zhang and Tongmei Zhang in Therapeutic Advances in Medical Oncology

Footnotes

Acknowledgements

None.

Declarations

ORCID iD

Tongmei Zhang

Supplemental material

Supplemental material for this article is available online.

References

Han

Zheng

Zeng

, et al. Cancer incidence and mortality in China, 2022. J Natl Cancer Cent 2024; 4(1): 47–53.

Zhang

Nan

Chen

, et al. Circular RNA circSATB2 promotes progression of non-small cell lung cancer cells. Mol Cancer 2020; 19(1): 101.

Friedlaender

Addeo

Russo

, et al. Targeted therapies in early stage NSCLC: hype or hope?. Int J Mol Sci 2020; 21(17): 6329.

Wittekind

Brierley

Sobin

, editors. TNM Supplement: A Commentary on Uniform Use. 4th ed. Oxford: Wiley-Blackwell; 2012.

Rami-Porta

Wittekind

Goldstraw

Complete resection in lung cancer surgery: proposed definition. Lung Cancer 2005; 49(1): 25–33.

Lardinois

De Leyn

Van Schil

, et al. ESTS guidelines for intraoperative lymph node staging in non-small cell lung cancer. Eur J Cardiothorac Surg 2006; 30(5): 787–792.

Sun

Liu

, et al. Impact of histological components on selecting limited lymphadenectomy for lung adenocarcinoma ⩽2 cm. Lung Cancer 2020; 150: 36–43.

Osarogiagbon

Faris

Stevens

, et al. Beyond margin status: population-based validation of the proposed international association for the study of lung cancer residual tumor classification recategorization. J Thorac Oncol 2020; 15(3): 371–382.

Shi

Liang

, et al. Prognostic significance of pleural lavage cytology in patients with lung cancer: a meta-analysis. Lung Cancer 2008; 60(2): 183–192.

10.

Lim

Clough

Goldstraw

, et al. Impact of positive pleural lavage cytology on survival in patients having lung resection for non-small-cell lung cancer: an international individual patient data meta-analysis. J Thorac Cardiovasc Surg 2010; 139(6): 1441–1446.

11.

Saso

Rao

Ashrafian

, et al. Positive pre-resection pleural lavage cytology is associated with increased risk of lung cancer recurrence in patients undergoing surgical resection: a meta-analysis of 4450 patients. Thorax 2012; 67(6): 526–532.

12.

Wang

Ling

, et al. Prognostic value of pleural lavage cytology in patients with lung cancer resection: an updated meta-analysis. PLoS One 2016; 11(7): e157518.

13.

Page

Mckenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021; 372: n71.

14.

Higgins

JPT

Altman

Gotzsche

, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 2011; 343: d5928.

15.

Zaman

Subramaniam

Afroz

, et al. Circulating tumour DNA (ctDNA) as a predictor of clinical outcome in non-small cell lung cancer undergoing targeted therapies: a systematic review and meta-analysis. Cancers (Basel) 2023; 15(9): 2425.

16.

Wang

Yao

Chen

, et al. Stem cell therapy for Crohn’s disease: systematic review and meta-analysis of preclinical and clinical studies. Stem Cell Res Ther 2021; 12(1): 463.

17.

Gagliasso

Migliaretti

Ardissone

Assessing the prognostic impact of the International Association for the Study of Lung Cancer proposed definitions of complete, uncertain, and incomplete resection in non-small cell lung cancer surgery. Lung Cancer 2017; 111: 124–130.

18.

Edwards

Chansky

Van Schil

, et al. The IASLC lung cancer staging project: analysis of resection margin status and proposals for residual tumor descriptors for non-small cell lung cancer. J Thorac Oncol 2020; 15(3): 344–359.

19.

Yun

Lee

Choi

, et al. A validation study of the recommended change in residual tumor descriptors proposed by the international association for the study of lung cancer for patients with pN2 NSCLC. J Thorac Oncol 2021; 16(5): 817–826.

20.

Lee

Hong

, et al. Validation of the IASLC residual tumor classification in patients with stage III-N2 non-small cell lung cancer undergoing neoadjuvant chemoradiotherapy followed by surgery. Ann Surg 2023; 277(6): e1355–e1363.

21.

Wang

Cai

Yang

Investigation of the non-small cell lung cancer patients with uncertain resection: a population-based study. Lung Cancer 2022; 171: 47–55.

22.

Ren

She

Tang

, et al. Prognostic evaluation of the proposed residual tumor classification in a Chinese non-small cell lung cancer population. J Surg Oncol 2022; 125(6): 1061–1070.

23.

Kadomatsu

Nakamura

Ueno

, et al. Prognostic value of uncertain resection for overall survival in non-small cell lung cancer. Ann Thorac Surg 2022; 114(4): 1262–1268.

24.

Chen

Zhong

Wen

, et al. Prognostic significance of the proposed residual tumor classification in patients with NSCLC after sleeve lobectomy. JTO Clin Res Rep 2023; 4(10): 100574.

25.

Wen

Zhao

, et al. Residual tumor descriptors proposed by the International Association for the Study of Lung Cancer may not be applicable to stage I and ground-glass opacity-featured non-small cell lung cancer. Transl Lung Cancer Res 2023; 12(11): 2157–2168.

26.

Marziali

Frasca

Ambrogi

, et al. Prognostic significance of uncertain resection for metastasis in the highest mediastinal lymph node after surgery for clinical N0 non-small cell lung cancer. Front Surg 2023; 10: 1115696.

27.

Liu

Shi

Cao

, et al. Prognostic significance of the highest mediastinal lymph node involvement in patients with stage III–N2 non-small cell lung cancer. Ann Surg Oncol 2024; 31(8): 5028–5037.

28.

Zhao

, et al. Spread through air spaces in residual tumor classification for clinical IA lung adenocarcinoma. Ann Thor Surg 2024; 118: 825–833.

29.

Sugi

Nawata

Fujita

, et al. Systematic lymph node dissection for clinically diagnosed peripheral non-small-cell lung cancer less than 2 cm in diameter. World J Surg 1998; 22(3): 290–294, 294–295.

30.

Huang

Wang

, et al. A randomized trial of systematic nodal dissection in resectable non-small cell lung cancer. Lung Cancer 2002; 36(1): 1–6.

31.

Aoki

Tsuchida

Watanabe

, et al. Surgical strategy for clinical stage I non-small cell lung cancer in octogenarians. Eur J Cardiothorac Surg 2003; 23(4): 446–450.

32.

Lardinois

Suter

Hakki

, et al. Morbidity, survival, and site of recurrence after mediastinal lymph-node dissection versus systematic sampling after complete resection for non-small cell lung cancer. Ann Thorac Surg 2005; 80(1): 268–274, 274–275.

33.

Okada

Sakamoto

Yuki

, et al. Selective mediastinal lymphadenectomy for clinico-surgical stage I non-small cell lung cancer. Ann Thorac Surg 2006; 81(3): 1028–1032.

34.

Wang

, et al. Impact of different mediastinal lymphadenectomy on clinical-stage IA non-small cell lung cancer. Zhonghua Wai Ke Za Zhi (Chinese J Surg) 2008; 46(9): 670–673.

35.

Wang

Long

, et al. Mediastinal lymph node dissection affects survival in patients with stage I non-small cell lung cancer. Thorac Cardiovasc Surg 2008; 56(4): 226–230.

36.

Takizawa

Kondo

Matsuoka

, et al. Effect of mediastinal lymph nodes sampling in patients with clinical stage I non-small cell lung cancer. J Med Invest 2008; 55(1–2): 37–43.

37.

Zhang

XW.

Correlated analysis of prognostic factors in patients with stage Ⅰ a non-small cell lung cancer. Int J Respir 2008; 28(10): 599–601.

38.

Okami

Higashiyama

Asamura

, et al. Pulmonary resection in patients aged 80 years or over with clinical stage I non-small cell lung cancer: prognostic factors for overall survival and risk factors for postoperative complications. J Thorac Oncol 2009; 4(10): 1247–1253.

39.

Chen

Huang

Xie

, et al. Effect of extent of lymphadenectomy in stage I non-small cell lung cancer patients. Chin J Cancer Prev Treat 2010; 17(3): 213–214.

40.

Shen

Ding

, et al. Study on the prognosis of the T1a non-small cell lung cancer. Chin J Lung Cancer 2010; 13(3): 206–210.

41.

Chen

Mao

Song

, et al. [Retrospective study on lobe-specific lymph node dissection for patients with early-stage non-small cell lung cancer]. Zhongguo Fei Ai Za Zhi 2012; 15(9): 531–538.

42.

Jiang

Chen

, et al. Selective mediastinal lymphadenectomy without intraoperative frozen section examinations for clinical stage I non-small-cell lung cancer: retrospective study of 403 cases. World J Surg 2013; 37(2): 392–397.

43.

Zhang

, et al. Comparison of lobe-specific mediastinal lymphadenectomy versus systematic mediastinal lymphadenectomy for clinical stage T₁a N₀ M₀ non-small cell lung cancer. J Cancer Res Ther 2013; 9(Suppl. 2): S101–S105.

44.

Zhang

Mao

, et al. Comparison of complete and minimal mediastinal lymph node dissection for non-small cell lung cancer: results of a prospective randomized trial. Thorac Cancer 2013; 4(4): 416–421.

45.

Jeon

Moon

Kim

, et al. Extent of removal for mediastinal nodal stations for patients with clinical stage I non-small cell lung cancer: effect on outcome. Thorac Cardiovasc Surg 2014; 62(7): 599–604.

46.

Yan

, et al. Does an extended mediastinal lymphadenectomy improve outcome after R0 resection in lung cancer? Chin J Cancer Res 2014; 26(2): 183–191.

47.

Guerrera

Errico

Evangelista

, et al. Exploring stage I non-small-cell lung cancer: development of a prognostic model predicting 5-year survival after surgical resection. Eur J Cardiothorac Surg 2015; 47(6): 1037–1043.

48.

Hishida

Miyaoka

Yokoi

, et al. Lobe-specific nodal dissection for clinical stage I and II NSCLC: Japanese multi-institutional retrospective study using a propensity score analysis. J Thorac Oncol 2016; 11(9): 1529–1537.

49.

Shen-Tu

Mao

Pan

, et al. Lymph node dissection and survival in patients with early stage nonsmall cell lung cancer: a 10-year cohort study. Medicine (Baltimore) 2017; 96(43): e8356.

50.

Wang

Chen

, et al. Impact of different types of lymphadenectomy combined with different extents of tumor resection on survival outcomes of stage I non-small-cell lung cancer: a large-cohort real-world study. Front Oncol 2019; 9: 642.

51.

Guo

Z-Y

Ren

J-H

Y-Y

, et al. The significance of systematic lymph node dissection in surgery for early-stage non-small cell lung cancer patients aged ⩽40 years. J Thorac Dis 2021; 13(2): 1196–1204.

52.

Hattori

Matsunaga

Takamochi

, et al. Extent of mediastinal nodal dissection in stage I non-small cell lung cancer with a radiological pure-solid appearance. Eur J Surg Oncol 2021; 47(7): 1797–1804.

53.

Katsumata

Tane

Suzuki

, et al. Mediastinal lymph node dissection for the elderly with clinical stage I non-small cell lung cancer. Gen Thorac Cardiovasc Surg 2021; 69(12): 1560–1566.

54.

Zhao

Mao

, et al. Lobe-specific lymph node dissection in clinical stage IA solid-dominant non-small-cell lung cancer: a propensity score matching study. Clin Lung Cancer 2021; 22(2): e201–e210.

55.

Kamigaichi

Aokage

Ikeno

, 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(2): ezad016.

56.

Sui

Yang

Yin

, et al. The comparison of lobe-specific or systematic mediastinal lymph node dissection for early-stage lung adenocarcinoma with consolidation tumor ratio over 0.5. Clin Lung Cancer 2023; 24(1): 51–59.

57.

Chen

, et al. Lobe-specific lymph node dissection may be feasible for clinical N0 solid-predominant part-solid lung adenocarcinoma with solid component diameter ⩽2 cm. Clin Lung Cancer 2023; 24(5): 437–444.

58.

Woo

Cha

Lee

, et al. Impact of extended mediastinal lymph node dissection for stage I ground-glass opacity lesions. J Thorac Dis 2023; 15(11): 6029–6039.

59.

Zheng

Jiang

, et al. Define relative incomplete resection by highest mediastinal lymph node metastasis for non-small cell lung cancers: rationale based on prognosis analysis. Lung Cancer 2011; 72(3): 348–354.

60.

Sakao

Miyamoto

Yamazaki

, et al. Prognostic significance of metastasis to the highest mediastinal lymph node in nonsmall cell lung cancer. Ann Thorac Surg 2006; 81(1): 292–297.

61.

Nakagiri

Sawabata

Funaki

, et al. Validation of pN2 sub-classifications in patients with pathological stage IIIA N2 non-small cell lung cancer. Interact Cardiovasc Thorac Surg 2011; 12(5): 733–738.

62.

Park

Byun

Lee

, et al. Clinical implications of uncertain resection in scenarios of metastasis of the highest or most distant mediastinal lymph node station following surgical treatment of non-small-cell lung cancer. Lung Cancer 2019; 138: 1–5.

63.

Zhang

Wang

, et al. [Correlation between lymph node ratio and clinicopathological features and prognosis of IIIa-N2 non-small cell lung cancer]. Zhongguo Fei Ai Za Zhi 2019; 22(11): 702–708.

64.

Wang

Liu

, et al. Propensity-matched analysis of clinical relevance of the highest mediastinal lymph node metastasis. Ann Thorac Surg 2021; 111(1): 277–282.

65.

Guo

Zhang

, et al. The additional radiotherapy to adjuvant chemotherapy improves the prognosis of stage III-N2 with highest mediastinal lymph node metastasis in non-small cell lung cancer. J Cancer Res Clin Oncol 2023; 149(14): 13311–13321.

66.

Liu

Zhang

Lin

, et al. [The impact of the highest mediastinal lymph node metastasis on postoperative recurrence and survival in non-small cell lung cancer patients]. Zhonghua Wai Ke Za Zhi 2023; 61(12): 1086–1092.

67.

Mizuno

Isaka

Ono

, et al. Impact of positive pleural lavage cytology for each stage of non-small cell lung cancer patients. Ann Thorac Surg 2021; 111(5): 1696–1702.

68.

Kaneda

Yokoi

Ito

, et al. The value of pleural lavage cytology examined during surgery for primary lung cancer. Eur J Cardiothorac Surg 2012; 41(6): 1335–1341.

69.

Nakao

Hoshi

Ishikawa

, et al. Prognosis of non-small-cell lung cancer patients with positive pleural lavage cytology. Interact Cardiovasc Thorac Surg 2015; 20(6): 777–782.

70.

Tomizawa

Nishino

Sesumi

, et al. Prognostic impact of pleural lavage cytology in patients with primary lung cancer. Lung Cancer 2016; 102: 60–64.

71.

Ogawa

Uchino

Tanaka

, et al. Efficacy of adjuvant chemotherapy for lung adenocarcinoma patients with positive pleural lavage cytology findings. Interact Cardiovasc Thorac Surg 2015; 21(1): 34–39.

72.

Detterbeck

Ostrowski

Hoffmann

, et al. The international association for the study of lung cancer lung cancer staging project: proposals for revision of the classification of residual tumor after resection for the forthcoming (ninth) edition of the TNM classification of lung cancer. J Thorac Oncol 2024; 19(7): 1052–1072.

73.

Doddoli

Aragon

Barlesi

, et al. Does the extent of lymph node dissection influence outcome in patients with stage I non-small-cell lung cancer?. Eur J Cardiothorac Surg 2005; 27(4): 680–685.

74.

American Joint Committee on Cancer. AJCC cancer staging manual. 6th ed. New York, NY: Springer, 2002.

75.

Chen

Mao

Wen

, et al. Tumor spread through air spaces in non-small cell lung cancer: a systematic review and meta-analysis. Ann Thorac Surg 2019; 108(3): 945–954.

76.

Nurmamat

Mamatabdulla

, et al. [The effect of tumor spread through air spaces on the prognosis of patients with stage Ⅰ non-small cell lung cancer: a meta-analysis]. Zhonghua Yi Xue Za Zhi 2023; 103(45): 3683–3690.

77.

Robinson

Patel

Bradley

, et al. Postoperative radiotherapy for pathologic N2 non-small-cell lung cancer treated with adjuvant chemotherapy: a review of the National Cancer Data Base. J Clin Oncol 2015; 33(8): 870–876.

78.

Yun

Lee

Choi

, et al. The addition of radiotherapy to adjuvant chemotherapy has a combinatorial effect in pN2 non-small cell lung cancer only with extranodal invasion or multiple N2 metastasis. Lung Cancer 2021; 155: 94–102.

79.

Hui

Men

, et al. Effect of postoperative radiotherapy for patients with pIIIA-N2 non-small cell lung cancer after complete resection and adjuvant chemotherapy: the phase 3 PORT-C randomized clinical trial. JAMA Oncol 2021; 7(8): 1178–1185.

80.

Deng

Wang

Xiu

, et al. Uncertain resection of highest mediastinal lymph node positive among pN2 non-small cell lung cancer patients: survival analysis of postoperative radiotherapy and driver gene mutations. Jpn J Radiol 2023; 41(5): 551–560.

81.

Vallieres

Van Houtte

Travis

, et al. Carcinoma in situ at the bronchial resection margin: a review. J Thorac Oncol 2011; 6(10): 1617–1623.

82.

Asamura

Hishida

Suzuki

, et al. Radiographically determined noninvasive adenocarcinoma of the lung: survival outcomes of Japan Clinical Oncology Group 0201. J Thorac Cardiovasc Surg 2013; 146(1): 24–30.

83.

Hattori

Suzuki

Takamochi

, et al. Prognostic impact of a ground-glass opacity component in clinical stage IA non-small cell lung cancer. J Thorac Cardiovasc Surg 2021; 161(4): 1469–1480.

84.

Saji

Okada

Tsuboi

, et al. Segmentectomy versus lobectomy in small-sized peripheral non-small-cell lung cancer (JCOG0802/WJOG4607L): a multicentre, open-label, phase 3, randomised, controlled, non-inferiority trial. Lancet 2022; 399(10335): 1607–1617.

85.

Suzuki

Watanabe

Wakabayashi

, et al. A single-arm study of sublobar resection for ground-glass opacity dominant peripheral lung cancer. J Thorac Cardiovasc Surg 2022; 163(1): 289–301.e2.

86.

Aokage

Suzuki

Saji

, et al. Segmentectomy for ground-glass-dominant lung cancer with a tumour diameter of 3 cm or less including ground-glass opacity (JCOG1211): a multicentre, single-arm, confirmatory, phase 3 trial. Lancet Respir Med 2023; 11(6): 540–549.

87.

Lee

Hong

Cho

, et al. Reclassifying the international association for the study of lung cancer residual tumor classification according to the extent of nodal dissection for NSCLC: one size does not fit all. J Thorac Oncol 2022; 17(7): 890–899.

88.

Verze

Pluchino

Leonetti

, et al. Role of ctDNA for the detection of minimal residual disease in resected non-small cell lung cancer: a systematic review. Transl Lung Cancer Res 2022; 11(12): 2588–2600.

89.

Wankhede

Grover

Hofman

Circulating tumor cells as a predictive biomarker in resectable lung cancer: a systematic review and meta-analysis. Cancers (Basel) 2022; 14(24): 6112.

90.

Rami-Porta

The evolving concept of complete resection in lung cancer surgery. Cancers (Basel) 2021; 13(11): 2583.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB

0.01 MB

0.51 MB

The IASLC uncertain resection,general overview,current evidence,and future prospects: a systematic review and meta-analysis

Abstract

Background:

Objectives:

Design:

Methods:

Results:

Conclusion:

Keywords

Introduction

Methods

Protocol and registration

Literature source and search strategy

Study selection

Quality assessment and data extraction

Statistical analysis

Results

The validation of uncertain resection

Characteristics of the included studies

OS assessment based on IASLC R classification

DFS assessment based on IASLC R classification

Subgroup and sensitivity analysis

Publication bias

Extent of removal for mediastinal lymph node stations

Characteristics of the included research

Risk of bias and quality of studies

Assessment of OS

Assessment of DFS

Publication bias

Highest mediastinal lymph node involvement

Characteristics of the included research

Assessment of OS and DFS

Publication bias

Prognostic significance of PLC

Discussion

Conclusion

Supplemental Material

sj-docx-1-tam-10.1177_17588359251344789 – Supplemental material for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis

Supplemental Material

sj-docx-3-tam-10.1177_17588359251344789 – Supplemental material for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis

Supplemental Material

sj-pdf-2-tam-10.1177_17588359251344789 – Supplemental material for The IASLC uncertain resection, general overview, current evidence, and future prospects: a systematic review and meta-analysis

Footnotes

Acknowledgements

Declarations

ORCID iD

Supplemental material

References

Supplementary Material