ABSTRACT
Lung adenocarcinoma exhibits diverse pathological features, some of which for example, histological grading and tumor spread through air spaces (STAS), can be used for prognosis. Extra‐alveolar invasion (EX) is a unique pathological factor focused on depth of invasion, and is defined as tumor infiltration into periarterial, peribronchiolar, or interlobular septal stroma. We recently suggested EX as a prognostic indicator, but its significance in sublobar resection remains unclear. We investigated the prognostic impact of EX in 116 patients with pT1mi–pT2a non‐mucinous lung adenocarcinoma treated by segmentectomy or wedge resection. EX was evaluated using elastic fiber staining. Among 10 recurrent cases, 8 were EX‐positive, and 2 were EX‐negative and STAS‐positive. The EX(−)/STAS(+) cases had local recurrence, while the EX(+) cases exhibited metastatic recurrence. The pT1abc EX(+) group showed significantly shorter recurrence‐free survival (RFS) than the pT1abc EX(−) group, but no significant differences were observed when compared with the pT2 group. EX(+ )/STAS(−) and EX(+ )/STAS(+) groups showed significantly shorter RFS than EX(−)/STAS(−) group, but EX(−)/STAS(+) group did not. In conclusion, EX was associated with poor prognosis after sublobar resection of lung adenocarcinoma. Combined evaluation of EX and STAS may improve postoperative risk stratification.
Keywords: extra‐alveolar invasion, lung adenocarcinoma, prognosis, spread through air spaces
Extra‐alveolar invasion (EX), which is defined as tumor infiltration into periarterial, peribronchiolar, or interlobular septal stroma, was associated with poor prognosis in sublobar‐resected lung adenocarcinoma cases. The EX(−)/STAS(+) cases had local recurrence, while the EX(+) cases exhibited metastatic recurrence. No recurrence was observed in EX(−)/STAS(−) cases. Combined evaluation of EX and STAS may improve postoperative risk stratification and identify patients who would benefit from additional therapy or closer surveillance.
Abbreviations
- CT
computed tomography
- EVG
Elastica van Gieson
- EX
extra‐alveolar invasion
- RFS
recurrence‐free survival
- STAS
spread through air spaces
1. Introduction
In recent years, less extensive surgeries for lung cancer, such as segmentectomy, have been introduced to preserve the respiratory function and reduce invasiveness, although lobectomy has long been the standard surgical procedure. A large‐scale clinical trial, JCOG0802/WJOG4607L demonstrated that segmentectomy achieved superior overall survival compared to lobectomy in patients with peripheral‐type non‐small cell lung cancer measuring ≤ 2 cm [1]. Segmentectomy is becoming established as a standard treatment option. However, risk stratification for local recurrence and distant metastasis after sublobar resection is significantly challenging.
In lung adenocarcinoma, high‐grade histological patterns, histological grading systems, and tumor spread through air spaces (STAS) have been identified as pathological prognostic factors [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]. Evaluation of these growth patterns is based on subjective morphological assessment, raising the issues related to reproducibility and inter‐pathologist variability. Furthermore, unlike gastrointestinal tumors, the TNM classification for lung cancer [13] does not assess the depth of invasion in the lung histological structure, except for the pleura and main bronchi. Considering this background, we recently reported a study that focused on lung adenocarcinoma invasion into the extra‐alveolar stroma by dividing the lung parenchyma into alveolar and extra‐alveolar areas [14]. Extra‐alveolar invasion (EX) is defined as tumor infiltration into extra‐alveolar structures such as periarterial, peribronchial/peribronchiolar, or interlobular septal areas. EX was an independent poor prognostic factor for recurrence‐free survival (RFS) in non‐mucinous lung adenocarcinoma. Notably, the pT1/EX‐positive cases showed significantly shorter RFS than the pT1/EX‐negative cases, while there was no statistically significant difference in RFS between pT1/EX‐positive cases and pT2 cases. Furthermore, postoperative recurrence was not observed in EX‐negative cases, regardless of the STAS status. These results suggested that EX may serve as a pathological factor for identifying tumors with a high risk of recurrence in early lung adenocarcinomas.
These findings on EX were obtained from lung cancer cases that underwent lobectomy or more extensive procedures. The significance of EX in sublobar resection cases is yet to be verified. Even in small tumors, prognosis after sublobar resection for invasive lung adenocarcinoma is not uniformly favorable, and the pathological features that affect curability in sublobar resection are unknown. While STAS is correlated with local recurrence around resection margins in sublobar resection cases [10], the extent to which EX influences prognosis in these patients remains unclear. Furthermore, the relationship between EX and STAS is poorly understood. If EX can be established as a new pathological risk stratification factor for sublobar resection cases, it can contribute to individualized postoperative follow‐up strategies.
The purpose of this study was to clarify the prognostic significance of EX status in sublobar resection.
2. Materials and Methods
2.1. Case Selection and Clinicopathological Evaluation
The study was approved by the Ethics Committee of The University of Tokyo (approval number: 10461‐15), and informed consent was obtained from all participants through the opt‐out method. We retrospectively reviewed 185 consecutive cases that underwent sublobar resection (segmentectomy or wedge resection) for lung adenocarcinoma at The University of Tokyo Hospital (Tokyo, Japan) between January 2015 and December 2021. Among these, 52 cases with non‐invasive (adenocarcinoma in situ), 4 cases with mucinous adenocarcinoma, and 10 cases with multiple invasive lung tumors or intrapulmonary metastatic tumors, 2 cases with tumor invasion at the resection margin, and 1 case with severe pulmonary fibrosis were excluded. Thus, data from 116 cases that underwent sublobar resection for pT1mi to pT2a non‐mucinous lung adenocarcinoma were analyzed.
Clinical and prognostic data were retrospectively collected from medical records. Postoperative follow‐up procedures, including physical examination and chest computed tomography (CT), were performed every 6 months. Fluorodeoxyglucose‐positron emission tomography and brain magnetic resonance imaging were performed when recurrence was suspected based on CT findings or clinical symptoms.
Glass slides of all cases were reviewed by two pulmonary pathologists (M. Suzuki and A.S‐U.), and the pathological stages were re‐evaluated according to the Union for International Cancer Control TNM classification ninth edition [13]. All resected tissues were fixed in 10% or 20% formalin, embedded in paraffin, and stained with hematoxylin and eosin. All tumor tissues were also stained with Elastica van Gieson (EVG). Histological classification and assessment were performed in accordance with the fifth edition of the WHO classification of lung tumors [15].
2.2. Evaluation of EX and Pleural Invasion
We differentiated the alveolar and extra‐alveolar regions by identifying the elastic layer between these with EVG‐stained sections, and evaluated EX using the same assessment criteria as in our previous investigation [14]. EX was further classified into periarterial, peribronchial/peribronchiolar, and interlobular septal types of invasion. Each invasion was defined as follows:
Periarterial invasion (Figure 1A,B): Tumor cells infiltrated into the fibrous stroma surrounding the pulmonary artery beyond the elastic layer of the alveolar margin with or without direct invasion into the arterial wall but not reaching the arterial lumen. Direct invasion into the vascular lumen and tumor embolism were evaluated as vascular invasion, which is a different factor from periarterial invasion.
Figure 1.
Examples of extra‐alveolar invasion. (A, B) Examples of periarterial invasion. Tumor cells infiltrate into the periarterial stroma across the elastic layer of the alveolar margin (blue arrowhead). (C, D) Examples of peribronchiolar invasion. Tumor cells infiltrate into the bronchiolar wall across the elastic layer of the alveolar margin (blue arrowhead). (E) An example of interlobular septal invasion. Tumor cells infiltrate into the interlobular septum across the elastic layer of the alveolar margin (blue arrowhead). (F) Tumor cells infiltrate beyond the internal elastic layer (blue arrowhead) but do not cross the external elastic layer (blue arrow) beneath the visceral pleura. Such a type of invasion is classified as equivalent to interlobular septal invasion. Staining method: Elastica van Gieson staining. Magnification: x50 (A, B); x100 (C–F).
Peribronchial/peribronchiolar invasion (Figure 1C,D): Tumor cells infiltrated into the bronchial and/or bronchiolar wall beyond the elastic layer of the alveolar margin. Intraepithelial extension of tumor cells along the airway (intrabronchial or intrabronchiolar extension) was excluded.
Interlobular septal invasion (Figure 1E,F): Tumor cells infiltrated into the interlobular septum beyond the elastic layer of the alveolar margin. Infiltration into the fibrous stroma around the pulmonary veins was also included in the interlobular septal invasion. The stroma between the internal and external elastic layers beneath the visceral pleura is a structure continuous with the interlobular septa; therefore, tumor infiltration into the stroma was included in interlobular septal invasion (Figure 1F).
The minimum tumor volume criterion for determining EX was defined as one or more clusters of ≥ 5 cancer cells infiltrating beyond the elastic layer of the alveolar margin. EX(−) was defined as the absence of any of the above EX factors, and EX(+) was defined as the presence of one or more of the above EX factors. Areas where the alveolar boundaries were difficult to identify due to extensive collapse were excluded from the EX assessment. Pleural invasion was defined as infiltration of the visceral pleura beyond the external elastic layer by cancer cells.
2.3. Comparative Prognostic Analysis of Lobectomy and Sublobar Resection Cases
To investigate differences in prognosis of EX‐positive cases between the two different surgical procedures, we compared RFS of EX‐positive cases in this cohort with that of 85 EX‐positive cases in pT1a to pT2a stages, which were obtained from the lobectomy cohort used in our previous study [14].
2.4. Statistical Analysis
Fisher's exact test or Pearson's chi‐square test was used to evaluate differences between qualitative variables. The Kruskal–Wallis test was used to compare continuous variables in patient groups. RFS was calculated using the Kaplan–Meier method and compared using the log‐rank test. RFS was defined as the time from the date of surgery to the date of recurrence, death due to any cause, or last follow‐up. Patients who were alive when the last follow‐up was done were censored for that time point. Since this cohort had only 1 case of lung cancer death, overall survival analysis was not performed. For multiple comparisons of survival curves (three or more curves), Holm‐Bonferroni correction was used to adjust the p values in the log‐rank test. The Cox proportional hazards model was used to perform univariate and multivariable analyses. We estimated the cumulative incidence of lung cancer recurrence with death from other causes considered as a competing event, and compared groups using Gray's test. Although an adequate number of cases with recurrence events could not be obtained, multivariable Cox regression analyses were performed by combining EX with each clinicopathological factor individually to evaluate the robustness of its prognostic impact. Smoking index was dichotomized at 600 (≥ 600 vs. < 600), and invasive size was categorized using a cutoff of 11 mm (≥ 11 mm vs. < 11 mm) for statistical analyses. In all statistical tests, p values < 0.05 were considered statistically significant. Statistical analysis was conducted using EZR version 1.36 (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria).
3. Results
3.1. Clinicopathological Characteristics
The clinicopathological characteristics of 116 lung adenocarcinomas are summarized in Table 1. The surgical margin was negative for all cases. RFS for all pT stages is shown in Supplementary Figure 1. Median follow‐up period was 1485 days (range, 22–3177 days). To clarify the prognostic impact of EX on early‐stage lung adenocarcinomas, we analyzed the present cohort by grouping it into the pT1mi, pT1abc EX(−), pT1abc EX(+), and pT2 (Table 1). Lung cancer recurrence was observed in 10 cases, and death from other causes was observed in 4 cases. Of the 4 cases of death from other causes, 3 were pT1mi, and the other was pT1a and EX‐negative. None of the pT1mi cases showed EX or recurrence of lung cancer, so these were excluded from subsequent prognostic analyses. One case, which presented a 5 mm invasive tumor with a predominant acinar component, was included in pT1a because it lacked lepidic growth and did not meet the criteria for minimally invasive adenocarcinoma. The tumor had infiltrated beyond the internal elastic layer beneath the visceral pleura (Figure 1F), corresponding to interlobular septal invasion, and metastasized to the lung and skin postoperatively. Additionally, three cases that were STAS‐positive and with invasions < 5 mm in size were classified as pT1a, and all were EX‐negative. All pT2 cases were EX‐positive.
Table 1.
Clinicopathological characteristics of the conventional lung adenocarcinomas.
| Variable | Total | pT1mi | pT1abc EX(−) | pT1abc EX(+) | pT2 | p value |
|---|---|---|---|---|---|---|
| Number of patients | 116 | 53 | 43 | 11 | 9 | |
| Surgical procedure | 0.707 | |||||
| Wedge resection | 69 | 30 | 25 | 7 | 7 | |
| Segmentectomy | 47 | 23 | 18 | 4 | 2 | |
| Age (y) | 0.977 | |||||
| Range | 41–91 | 41–91 | 49–88 | 51–84 | 42–86 | |
| Median | 72 | 71 | 74 | 73 | 74 | |
| Sex | 0.753 | |||||
| Female | 56 | 27 | 18 | 6 | 5 | |
| Male | 60 | 26 | 25 | 5 | 4 | |
| Smoking | 0.798 | |||||
| Never | 46 | 19 | 18 | 4 | 5 | |
| Ever | 70 | 34 | 25 | 7 | 4 | |
| Smoking index | 0.776 | |||||
| Range | 0–3000 | 0–3000 | 0–2250 | 0–3000 | 0–920 | |
| Median | 200 | 200 | 160 | 480 | 0 | |
| Size of invasion (mm) | 0.0023 | |||||
| Range | 1–25 | 1–4 | 2–22 | 5–20 | 6–25 | |
| Median | 6 | 3 | 7 | 11 | 16 | |
| Predominant histological pattern | < 0.0001 | |||||
| Lepidic | 74 | 53 | 21 | 0 | 0 | |
| Papillary | 13 | 0 | 11 | 2 | 0 | |
| Acinar | 21 | 0 | 8 | 7 | 6 | |
| Solid | 7 | 0 | 2 | 2 | 3 | |
| Micropapillary | 1 | 0 | 1 | 0 | 0 | |
| Histological grade | < 0.0001 | |||||
| G1 | 21 | — | 21 | 0 | 0 | |
| G2 | 29 | — | 19 | 7 | 3 | |
| G3 | 13 | — | 3 | 4 | 6 | |
| Spread through air spaces (STAS) | < 0.0001 | |||||
| STAS (−) | 88 | 53 | 27 | 5 | 3 | |
| STAS (+) | 28 | 0 | 16 | 6 | 6 | |
| Lymphovascular involvement (LVI) | < 0.0001 | |||||
| LVI (−) | 104 | 53 | 43 | 7 | 1 | |
| LVI (+) | 12 | 0 | 0 | 4 | 8 | |
| Extra‐alveolar invasion | < 0.0001 | |||||
| Periarterial | 11 | 0 | 0 | 4 | 7 | |
| Peribronchial/peribronchiolar | 6 | 0 | 0 | 3 | 3 | |
| Interlobular septal | 12 | 0 | 0 | 8 | 4 | |
| Pleural invasion (pl) | < 0.0001 | |||||
| pl (−) | 107 | 53 | 43 | 11 | 0 | |
| pl (+) | 9 | 0 | 0 | 0 | 9 | |
| pT | < 0.0001 | |||||
| pT1mi | 53 | 53 | 0 | 0 | 0 | |
| pT1a | 39 | 0 | 33 | 6 | 0 | |
| pT1b | 14 | 0 | 9 | 5 | 0 | |
| pT1c | 1 | 0 | 1 | 0 | 0 | |
| pT2a | 9 | 0 | 0 | 0 | 9 | |
| pN | 0.325 | |||||
| pN0 | 47 | 23 | 18 | 4 | 2 | |
| pN1 | 1 | 0 | 0 | 0 | 1 | |
| pNX | 68 | 30 | 25 | 7 | 6 | |
| pStage | < 0.0001 | |||||
| IA | 107 | 53 | 43 | 11 | 0 | |
| IB | 8 | 0 | 0 | 0 | 8 | |
| II | 1 | 0 | 0 | 0 | 1 | |
| III or IV | 0 | 0 | 0 | 0 | 0 |
Abbreviation: EX, extra‐alveolar invasion.
Correlations between EX and other pathological factors are shown in Supplementary Table 1. EX was correlated with histological grade, pleural invasion, lymphatic invasion, and vascular invasion. There was no significant correlation between EX and STAS.
The histological characteristics of the 10 recurrent cases are summarized in Figure 2. All recurrent cases showed either EX or STAS or both; 8 cases (80%) were positive for EX, and 8 cases (80%) were positive for STAS, 2 cases were EX‐positive and STAS‐negative, and 2 cases were EX‐negative and STAS‐positive. Some recurrent cases were also positive for other histological factors such as pleural invasion and/or lymphovascular invasion. However, 6 (60%) cases were negative for pleural invasion, and 4 (40%) cases were negative for lymphovascular invasion. One case (case 1 in Figure 2) was positive only for EX. The sites for recurrence in EX‐positive cases were lymph node metastases in 2 cases, intrapulmonary metastases in 2 cases, extrapulmonary metastases in 2 cases, pleural dissemination in 1 case, and local recurrence in 2 cases. Both cases with EX‐negative/STAS‐positive showed local recurrence.
Figure 2.
Pathological characteristics of recurrent cases. The pathological factors in each column represent those of the primary lesions. The EX columns are positive if any of the three EX factors are positive. Aci, acinar; EX, extra‐alveolar invasion; F, female; LVI, lymphovascular involvement; M, male; Mipap, micropapillary; S, segmentectomy; Sol, solid; STAS, spread through air spaces; Pap, papillary; W, wedge resection.
3.2. Prognostic Analyses
All the three EX factors (periarterial, peribronchial/peribronchiolar, and interlobular septal invasion) were associated with significantly shorter RFS (Figure 3A–C). The EX(+) group had a significantly lower recurrence‐free rate than the EX(−) group (Figure 3D). Univariate analyses of the Cox proportional hazards model showed that five pathological factors, including EX, were significantly associated with shorter RFS (Table 2). To evaluate the robustness of the prognostic impact of EX, multivariable Cox regression models were constructed by combining EX with one clinicopathological factor at a time (Table 3). Although statistical significance diminished in several models, EX generally retained a high hazard ratio, and the magnitude of the association remained relatively stable, suggesting that EX may be associated with prognosis. In the cumulative incidence analysis, all pathological factors, including EX, were significantly associated with a higher rate of lung cancer recurrence (Figure 4A–E).
Figure 3.
Recurrence‐free survival (RFS) analysis for lung adenocarcinoma patients with extra‐alveolar invasion (EX). (A) RFS for periarterial invasion. (B) RFS for peribronchial/peribronchiolar invasion. (C) RFS for interlobular septal invasion. (D) RFS for EX.
Table 2.
Univariate analysis of recurrence‐free survival.
| Factor | HR (95% CI) | p value |
|---|---|---|
| Age | 1.089 (1.006–1.179) | 0.03564 |
| Sex (male) | 1.819 (0.4782–6.92) | 0.3801 |
| Smoking index (≧ 600) | 2.798 (0.8515–9.194) | 0.09006 |
| Invasive size (≧ 11 mm) | 5.311 (1.404–20.09) | 0.01388 |
| Histological grade | ||
| Grade 1 | Referrence | |
| Grade 2 | 1.978 (0.2054–19.05) | 0.5551 |
| Grade 3 | 11.680 (1.4310–95.32) | 0.02178 |
| Pleural invasion | 3.209 (0.9369–10.99) | 0.06341 |
| Lymphovascular invasion | 6.579 (2.000–21.64) | 0.001932 |
| Spread through air spaces | 4.738 (1.198–18.73) | 0.02657 |
| Extra‐alveolar invasion | 7.951 (2.073–30.49) | 0.0025 |
Abbreviations: CI, confidence interval; HR2, hazard ratio.
Table 3.
Multivariable analyses evaluating the robustness of extra‑alveolar invasion (EX).
| Model | Covariates in the model | HR for EX (95% CI) | p value for EX |
|---|---|---|---|
| 1 | EX + Age | 9.342 (2.180–40.030) | 0.002616 |
| 2 | EX + Sex (male) | 8.138 (2.1400–30.950) | 0.002095 |
| 3 | EX + Smoking index (≧ 600) | 7.823 (2.004–30.540) | 0.003074 |
| 4 | EX + Invasive size (≧ 11 mm) | 5.063 (1.1440–22.41) | 0.03259 |
| 5 | EX + Histological grade | 4.022 (0.75320–21.48) | 0.1034 |
| 6 | EX + Pleural invasion | 10.4200 (2.1570–50.380) | 0.003543 |
| 7 | EX + Lymphovascular invasion | 4.965 (0.8068–30.550) | 0.08391 |
| 8 | EX + Spread through air spaces | 5.678 (1.3790–23.38) | 0.01619 |
Abbreviations: CI, confidence interval; HR, hazard ratio.
Figure 4.
Cumulative incidence (CI) analysis of lung cancer recurrence with pathological factors. (A) CI of lung cancer recurrence with histological grade. (B) CI of lung cancer recurrence with pleural invasion. (C) CI of lung cancer recurrence with lymphovascular invasion. (D) CI of lung cancer recurrence with spread through air spaces (STAS). (E) CI of lung cancer recurrence with extra‐alveolar invasion (EX).
Three‐year RFS rates were 97.6% in the pT1abc EX(−), 62.3% in the pT1abc EX(+), and 77.8% in the pT2 groups, and 5‐year RFS rates were 92.7% in the pT1abc EX(−), 62.3% in the pT1abc EX(+), and 46.7% in the pT2 groups (Figure 5A). The pT1abc EX(+) group showed significantly shorter RFS than the pT1abc EX(−) group. There were no significant differences between the pT1abc EX (+) and pT2 groups. In the cumulative incidence analysis, the pT1abc EX(+) group had a significantly higher recurrence rate than the pT1abc EX(−) group, while the statistically significant difference in recurrence rate was not observed between the pT1abc EX(+) group and the pT2 group (Figure 5B).
Figure 5.
Prognostic analysis using combined pathological factors. Recurrence‐free survival (A) and cumulative incidence of lung cancer recurrence (B) with EX and pathological T stage. Recurrence‐free survival (C) and cumulative incidence of lung cancer recurrence (D) with EX and STAS. EX: extra‐alveolar invasion, STAS: spread through air spaces.
To examine the combined prognostic impact of EX and STAS, prognostic analysis was performed by grouping the cohort into the EX(−)/STAS(−) (n = 27), EX(−)/STAS(+) (n = 16), EX(+ )/STAS(−) (n = 8), and EX(+ )/STAS(+) (n = 12). EX(+ )/STAS(−) and EX(+ )/STAS(+) groups showed significantly shorter RFS than EX(−)/STAS(−) group (Figure 5C), but EX(−)/STAS(+) group did not. The EX(+ )/STAS(+) group showed a shorter RFS than the EX(+ )/STAS(−) group, although the difference was not statistically significant. In the cumulative incidence analysis, the EX(+ )/STAS(+) group showed a significantly higher recurrence rate than the EX(−)/STAS(−) group. No significant differences in recurrence rate were observed among EX(−)/STAS(+ ), EX(+ )/STAS(−) and EX(−)/STAS(−) groups (Figure 5D).
We compared the RFS of 20 EX‐positive cases that underwent sublobar resection and 85 EX‐positive cases that underwent lobectomy. The EX‐positive sublobar resection group showed shorter RFS than the EX‐positive lobectomy group, but the difference was not statistically significant (Supplementary Figure 2A). A comparative analysis limited to cases ranging from pT1a to pT1c also revealed similar results (Supplementary Figure 2B).
4. Discussion
EX is a unique pathological factor that is focused on the histological structures of lungs and the depth of invasion of lung cancers, unlike previously reported prognostic factors, which are defined by cancer‐specific histological patterns in the lungs, such as STAS or high‐grade patterns (solid, micropapillary, complex glandular pattern, etc.). Fujita et al. first reported on the prognostic significance of EX in 2009 [16]. They examined 157 cases of completely resected lung adenocarcinomas measuring ≤ 30 mm in whole tumor size. They demonstrated that lung adenocarcinoma cases without EX, regardless of destruction of the alveolar framework, had a 5‐year disease‐specific survival rate of 100%. Additionally, the 5‐year disease‐specific survival rate was 90.1% in the group which was positive for only one extra‐alveolar involvement factor (e.g., bronchial, perivascular, interlobular septal, or pleural invasion), and 66.7% in the group which was positive for two or more factors. We recently performed a prognostic analysis of 178 cases of lung adenocarcinoma that underwent complete resection (lobectomy or more), demonstrating that EX is an independent poor prognostic factor for RFS [14]. Regardless of STAS status or histological grade, there was no lymph node metastasis or postoperative recurrence observed in the EX‐negative group [14]. This study focused on sublobar resection cases. Because the number of events was limited, we were unable to perform a sufficient multivariable analysis to conclusively determine whether EX is an independent prognostic factor. However, the EX‐positive group had a significantly poorer prognosis than the EX‐negative group. EX showed a relatively high hazard ratio in the univariate analysis and demonstrated considerable robustness across the multivariable models that combined EX with individual clinicopathological factors, indicating that EX may be associated with poor prognosis. Furthermore, most recurrent cases in this study were EX‐positive. Notably, however, 2 of the 10 recurrent cases were EX‐negative and STAS‐positive, which differs from findings in lobectomy cases.
STAS is a distinctive progression pattern of lung cancer. It is also a risk factor for recurrence after sublobar resection [10, 17, 18]. In this study, STAS was identified to be a significant poor prognostic factor in the univariate analysis on RFS. In the subgroup analysis that combined EX and STAS, EX(+ )/STAS(−) group had a significantly shorter RFS than the EX(−)/STAS(−) group, but EX(−)/STAS(+) group did not have a short RFS. The previous lobectomy cases study revealed no recurrence in the EX‐negative group regardless of STAS status [14]. These findings suggested that EX had a greater impact on prognosis than STAS regardless of the surgical procedure. In this study, metastatic recurrence in the lungs, other organs, or lymph nodes was observed only in the EX‐positive group; therefore, EX appears to be associated with metastatic recurrence. In contrast, no metastatic recurrence was observed in EX(−)/STAS(+) group in both lobectomy and sublobar resection cohorts, while two cases with EX(−)/STAS(+) showed local recurrence, suggesting that STAS may be associated with local recurrence rather than metastatic recurrence. Since no recurrence was observed in EX(−)/STAS(−) cases, these two factors may be important for sublobar resection cases. These findings suggest that EX and STAS could be beneficial for selecting additional treatments, such as lobectomy or adjuvant therapy, for sublobar resection cases.
Extra‐alveolar stroma of bronchovascular bundles and interlobular septum contains abundant lymphatic vessels and small arteries/veins [19]. Tumor infiltration into the stroma increases the chances for lymphovascular invasion. This lung histological feature likely contributes to a high risk of metastatic recurrence in the EX‐positive group highlighting the importance of the depth of invasion in lung cancer. The present study demonstrated that EX was associated with lymphatic and vascular invasion, which appears to support this notion. Recently, Filipello et al. proposed a classification system dividing the invasive forms of lung adenocarcinoma into true invasion and alveolar filling growth [20]. True invasion includes stromal invasion into the extra‐alveolar region. The authors also stated that elastic fiber staining was useful for distinguishing adenocarcinoma in situ with iatrogenic collapse from invasive adenocarcinoma [20]. Elastic fiber staining is also useful for evaluating pleural and vascular invasion [21], and enables clear and easy identification of EX. We agree to the previous findings on identifying true invasion to the extra alveolar stroma and strongly advocate the use of elastic fiber staining for the assessment of tumor invasion in lung cancer pathology.
In this cohort, we identified one case of metastatic recurrence in a small lung cancer measuring 5 mm in invasive diameter. The tumor infiltrated the stroma between the internal and external elastic layers beneath the visceral pleura, and was classified as being equivalent to interlobular septal invasion. The International Association for the Study of Lung Cancer defines visceral pleural invasion as infiltration beyond the visceral pleural elastic layer [22]. In Japan, tumor infiltration beyond the external elastic layer is adopted as the criterion for visceral pleural invasion [23]. Some reports indicate that tumor invasion beyond the external elastic layer is significantly associated with a poor prognosis [24, 25]. However, the impact of tumor invasion beyond the internal elastic layer on prognosis varies across the studies. In a study by Hamasaki et al., 26 lung cancer cases with internal elastic layer invasion had shorter overall survival than 73 cases without such invasion, but the difference was not statistically significant [24]. In contrast, Yeh et al. showed minimal differences in recurrence‐free and overall survival between 483 cases with internal elastic layer invasion and 502 cases without such invasion [25]. The results of both the above studies suggest no impact of stroma invasion between internal and external elastic laminae; however, these studies did not assess EX status at other sites than the pleura, suggesting that cases without internal elastic layer invasion may be associated with EX at peribronchial/periarterial/interlobular spaces. To clarify the prognostic impact of invasion into the pleural stroma, a comparison of prognosis between the group with such invasion and the EX‐negative group is necessary.
Further verification is needed, but if EX is indeed a prognostic factor for lung adenocarcinoma treated with sublobar resection, the pathological assessment of EX may influence clinical decisions regarding additional treatment. Although no statistically significant difference was observed, EX‐positive cases undergoing sublobar resection showed shorter RFS than those undergoing lobectomy. Because this cohort included many pNX cases, the association between EX and lymph node metastasis was unclear; however, in a previous study of lobectomy cases, lymph node metastasis was observed only in EX‐positive cases [14]. Additional treatment with lobectomy and lymph node dissection may improve the prognosis for EX‐positive cases who underwent sublobar resection. Additionally, all cases with postoperative metastatic recurrence were EX‐positive, suggesting that EX may serve as an indicator for adjuvant chemotherapy. Further study is needed to determine whether EX can serve as a criterion for the indication of additional treatment.
This study has several limitations. First, it was a single‐center study, and the cohort included relatively few EX‐positive cases. There were no EX‐positive cases with pT1c or histological grade 1. There were also a few cases with lung cancer‐related death, preventing an analysis of overall survival. Because the number of cases with lung cancer recurrence was insufficient, we could not perform a sufficiently adjusted multivariable analysis to conclusively determine whether EX is an independent prognostic factor. It remains unclear whether EX is merely a surrogate marker for known pathological prognostic factors, such as tumor size, pleural invasion, and histological grade, or it is a stronger prognostic factor than these factors. To confirm whether EX is truly a poor prognostic factor, a study involving a larger cohort, preferably from multiple institutions, would be necessary. Second, the sublobar resection cohort included many pNX cases. It is possible that the tumor stage may have been underestimated due to the omission of lymph node sampling, which is common in sublobar resection cases. Third, this study did not include distance from the resection margin. The cohort contained some cases in which retrospective assessment of margin distance was difficult, preventing examination of the relationship between margin distance and prognosis. In STAS‐positive sublobar resection cases, it has been reported that securing a margin distance exceeding the tumor diameter did not improve prognosis [18]. Future studies should clarify the relationship between EX and margin distance. Fourth, the reproducibility of the EX evaluation has not been examined. Similar to the previous lobectomy case study [14], in this study, two pathologists evaluated EX; however, training may be necessary for EX assessment. It is necessary to verify the reproducibility of this evaluation before incorporating the pathological assessment of EX into routine practice. The degree of variation among pathologists in EX evaluation requires further investigation.
In conclusion, the EX(+) group showed significantly shorter RFS compared to the EX(−) group in sublobar resection cases of non‐mucinous lung adenocarcinoma. Metastases to the lung, other organs, or lymph nodes were observed only in EX(+) group, and EX can be a risk factor for metastatic recurrence. No recurrence was observed in EX(−)/STAS(−) cases. Combined assessment of EX and STAS may improve postoperative risk stratification and help identify patients who could benefit from additional treatment or intensive follow‐up in sublobar resection cases. Further investigation in larger cohorts is necessary to verify the impact of EX on prognosis and the reproducibility of its assessment.
Author Contributions
Masaki Suzuki and Aya Shinozaki‐Ushiku conceived and designed the study and analyzed and interpreted the data. Junichi Takada, Koji Aze, Shinji Yuhara, Yue Cong, Mitsuaki Kawashima, and Masaaki Sato enrolled patients. Aya Shinozaki‐Ushiku critically revised the manuscript for important intellectual content. Tetsuo Ushiku gave final approval of the study.
Conflicts of Interest
Tetsuo Ushiku, the Editor‐in‐Chief of Pathology International, is also a co‐author of this article. To avoid any potential bias, he was not involved in any editorial decisions regarding the acceptance of this manuscript. The remaining authors declare no conflicts of interest.
Supporting information
Figure S1: Recurrence‐free survival for each pathological T stage.
Figure S2: Recurrence‐free survival (RFS) of the EX‐positive sublobar resection and lobectomy groups. (A) RFS of EX‐positive cases ranging from pT1a to pT2a. (B) RFS of EX‐positive cases limited to pT1a‐c.
Table S1: Correlations between EX and other pathological factors.
Acknowledgments
We thank Dr. Ryo Sadachi, Clinical Research Promotion Center, The University of Tokyo Hospital, for providing valuable support with the statistical analyses. We would like to thank Editage (www.editage.jp) for English language editing.
Portions of the methodological description resemble our previously published work because EX was evaluated using identical criteria in both studies. These overlapping sections have been appropriately cited to ensure transparency. This work was supported, in part, by the Japan Society for the Promotion of Science [25K10261].
Data Availability Statement
The datasets generated and/or analyzed during the current study are not publicly available due to patient confidentiality and ethical restrictions.
References
- 1. Saji H., Okada M., Tsuboi M., 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 399 (2022): 1607–1617. [DOI] [PubMed] [Google Scholar]
- 2. Yoshizawa A., Motoi N., Riely G. J., et al., “Impact of Proposed IASLC/ATS/ERS Classification of Lung Adenocarcinoma: Prognostic Subgroups and Implications for further Revision of Staging Based on Analysis of 514 Stage I Cases,” Modern Pathology 24 (2011): 653–664. [DOI] [PubMed] [Google Scholar]
- 3. Warth A., Muley T., Meister M., et al., “The Novel Histologic International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society Classification System of Lung Adenocarcinoma is a Stage‐Independent Predictor of Survival,” Journal of Clinical Oncology 30 (2012): 1438–1446. [DOI] [PubMed] [Google Scholar]
- 4. Tsuta K., Kawago M., Inoue E., et al., “The Utility of the Proposed IASLC/ATS/ERS Lung Adenocarcinoma Subtypes for Disease Prognosis and Correlation of Driver Gene Alterations,” Lung Cancer 81 (2013): 371–376. [DOI] [PubMed] [Google Scholar]
- 5. Kamiya K., Hayashi Y., Douguchi J., et al., “Histopathological Features and Prognostic Significance of the Micropapillary Pattern in Lung Adenocarcinoma,” Modern Pathology 21 (2008): 992–1001. [DOI] [PubMed] [Google Scholar]
- 6. Yanagawa N., Shiono S., Abiko M., Katahira M., Osakabe M., and Ogata S., “The Clinical Impact of Solid and Micropapillary Patterns in Resected Lung Adenocarcinoma,” Journal of Thoracic Oncology 11 (2016): 1976–1983. [DOI] [PubMed] [Google Scholar]
- 7. Kadota K., Yeh Y.‐C., Sima C. S., et al., “The Cribriform Pattern Identifies a Subset of Acinar Predominant Tumors With Poor Prognosis in Patients With Stage I Lung Adenocarcinoma: A Conceptual Proposal to Classify Cribriform Predominant Tumors as a Distinct Histologic Subtype,” Modern Pathology 27 (2014): 690–700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Moreira A. L., Joubert P., Downey R. J., and Rekhtman N., “Cribriform and Fused Glands Are Patterns of High‐Grade Pulmonary Adenocarcinoma,” Human Pathology 45 (2014): 213–220. [DOI] [PubMed] [Google Scholar]
- 9. Moreira A. L., Ocampo P. S. S., Xia Y., et al., “A Grading System for Invasive Pulmonary Adenocarcinoma: A Proposal From the International Association for the Study of Lung Cancer Pathology Committee,” Journal of Thoracic Oncology 15 (2020): 1599–1610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kadota K., Nitadori J., Sima C. S., et al., “Tumor Spread Through Air Spaces is an Important Pattern of Invasion and Impacts the Frequency and Location of Recurrences After Limited Resection for Small Stage I Lung Adenocarcinomas,” Journal of Thoracic Oncology 10 (2015): 806–814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Mino‐Kenudson M., “Significance of Tumor Spread Through Air Spaces (STAS) in Lung Cancer From the Pathologist Perspective,” Translational Lung Cancer Research 9 (2020): 847–859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Yang Y., Xie X., Wang Y., et al., “A Systematic Review and Meta‐Analysis of the Influence of STAS on the Long‐Term Prognosis of Stage I Lung Adenocarcinoma,” Translational Cancer Research 10 (2021): 2428–2436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Brierley J. D., Giuliani M., O'Sullivan B., Rous B., and Van Eycken L., TNM Classification of Malignant Tumours, 9th ed. (Wiley‐Blackwell, 2025). [Google Scholar]
- 14. Suzuki M., Shinozaki‐Ushiku A., Yuhara S., Nagano M., Sato M., and Ushiku T., “The Prognostic Impact of Extra‐Alveolar Invasion in Lung Adenocarcinoma,” Lung Cancer 205 (2025): 108612. [DOI] [PubMed] [Google Scholar]
- 15. WHO Classification of Tumors Editorial Board , Thoracic Tumors, 5th ed. (International Agency for Research on Cancer, 2021). [Google Scholar]
- 16. Fujita A., Kameda Y., and Goya T., “Clinicopathology of Stromal Invasion in Lung Adenocarcinoma,” Pathology International 59 (2009): 1–6. [DOI] [PubMed] [Google Scholar]
- 17. Masai K., Sakurai H., Sukeda A., et al., “Prognostic Impact of Margin Distance and Tumor Spread Through Air Spaces in Limited Resection for Primary Lung Cancer,” Journal of Thoracic Oncology 12 (2017): 1788–1797. [DOI] [PubMed] [Google Scholar]
- 18. Eguchi T., Kameda K., Lu S., et al., “Lobectomy is Associated With Better Outcomes Than Sublobar Resection in Spread Through Air Spaces (STAS)‐Positive T1 Lung Adenocarcinoma: A Propensity Score‐Matched Analysis,” Journal of Thoracic Oncology 14 (2019): 87–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Weber E., Sozio F., Borghini A., Sestini P., and Renzoni E., “Pulmonary Lymphatic Vessel Morphology: A Review,” Annals of Anatomy ‐ Anatomischer Anzeiger 218 (2018): 110–117. [DOI] [PubMed] [Google Scholar]
- 20. Filipello F., Blaauwgeers H., Lissenberg‐Witte B., et al., “Stereologic Consequences of Iatrogenic Collapse: The Morphology of Adenocarcinoma In Situ Overlaps With Invasive Patterns. Proposal for a Necessary Modified Classification of Pulmonary Adenocarcinomas,” Lung Cancer 197 (2024): 107987. [DOI] [PubMed] [Google Scholar]
- 21. Thunnissen E., Motoi N., Minami Y., et al., “Elastin in Pulmonary Pathology: Relevance in Tumours With a Lepidic or Papillary Appearance. A Comprehensive Understanding From a Morphological Viewpoint,” Histopathology 80 (2022): 457–467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Asamura H., “International Association for the Study of Lung Cancer.” Staging Manual in Thoracic Oncology, 3rd ed. (Editorial Rx Press, 2024). [Google Scholar]
- 23. The Japan Lung Cancer Society , General Rule for Clinical and Pathological Record of Lung Cancer, 9th ed. (Kanehara Publishing Co, 2025). [Google Scholar]
- 24. Hamasaki M., Kato F., Koga K., et al., “Invasion of the Inner and Outer Layers of the Visceral Pleura in pT1 Size Lung Adenocarcinoma Measuring ≤ 3 cm: Correlation With Malignant Aggressiveness and Prognosis,” Virchows Archiv 461 (2012): 513–519. [DOI] [PubMed] [Google Scholar]
- 25. Yeh Y.‐C., Chiang C.‐H., Hsu P.‐K., et al., “Tumor Cell Invasion of the External Elastic Lamina Designates Visceral Pleural Invasion and Predicts Poorer Patient Outcomes in Pulmonary Nonmucinous Invasive Adenocarcinoma,” Journal of Thoracic Oncology 20 (2025): 1791–1800. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Recurrence‐free survival for each pathological T stage.
Figure S2: Recurrence‐free survival (RFS) of the EX‐positive sublobar resection and lobectomy groups. (A) RFS of EX‐positive cases ranging from pT1a to pT2a. (B) RFS of EX‐positive cases limited to pT1a‐c.
Table S1: Correlations between EX and other pathological factors.
Data Availability Statement
The datasets generated and/or analyzed during the current study are not publicly available due to patient confidentiality and ethical restrictions.