You are here
Semiquantitative Assessment of Tumor Spread through Air Spaces (STAS) in Early-Stage Lung Adenocarcinomas
Journal of Thoracic Oncology, July 2017, Volume 12, Issue 7, Pages 1046-1051
Commentary by Nir Peled
Air-born spread of cancer cells considered being safe, however there is no proven evidence associated with that assumption. Seeding ability of cancer cells as well mesenchymal to epithelial transition may play a role in this process. Although the numerous limitations of this study, it provides an early evidence for the importance of air-born spread of cancer cells and their ability to survive without a trivial seeding capacity.
Tumor spread through air spaces (STAS) has recently been reported as a form of tumor invasion having an unfavorable prognosis, but the significance of a small amount of STAS is not known. The aim of this study was to perform a semiquantitative assessment of STAS.
Small (≤2 cm) stage I lung adenocarcinomas surgically resected at our institution between 2003 and 2009 were assessed semiquantitatively in the most prominent area as no STAS, low STAS (1–4 single cells or clusters of STAS), or high STAS (≥5 single cells or clusters of STAS) by using a 20× objective and a 10× ocular lens. A statistical analysis was performed to determine the impact of clinicopathologic parameters on STAS and to clarify the relationship between STAS and patient survival.
STAS was assessed as no STAS in 109 of 208 cases (52.4%), as low STAS in 38 cases (18.3%), and as high STAS in 61 cases (29.3%). There were statistically significant associations between higher STAS and solid predominant invasive adenocarcinoma ( p < 0.001), pleural invasion ( p < 0.001), lymphatic invasion ( p < 0.001), vascular invasion ( p < 0.001), and tumor size of 10 mm or more ( p = 0.037). There was a significant association between increasing STAS and shorter recurrence-free survival (RFS) in univariate analysis (no STAS, 154.2 months; low STAS, 147.6 months; and high STAS, 115.6 months). In a multivariate Cox proportional hazards model, only STAS ( p = 0.015) remained a significant predictor of RFS.
We found that one-third of resected small adenocarcinomas had high STAS. Higher STAS was predictive of worse RFS.
Keywords: Lung adenocarcinoma, Tumor spread through air spaces, Invasion, Recurrence.
Lung cancer is a leading cause of cancer mortality in both the United States and Japan. 1 2 Surgery is the mainstay of treatment for small lung adenocarcinoma. However, approximately 20% of patients with lung adenocarcinoma 2 cm or smaller without pleural invasion or metastasis experience recurrence of the disease. 3
Since the publication of the multidisciplinary lung adenocarcinoma classification (International Association for the Study of Lung Cancer, American Thoracic Society, and European Respiratory Society) in 2011, 4 tumor morphology has come to be recognized as a prognostic factor as important as molecular characteristics. Patients with a lepidic predominant pattern have the most favorable prognosis, those with an acinar or papillary pattern have the second most favorable prognosis, and those with a solid or micropapillary pattern have the worst prognosis. 5 6 Kadota et al. 7 reported that tumor spread through air spaces (STAS), floating tumor cells, and small tumor clusters in the normal alveolar space adjacent to the edge of tumor were significant risk factors for recurrence of small lung adenocarcinoma in patients who undergo limited resection. The 2015 WHO classification 8 introduced STAS as a new route of invasion and included it as an exclusion criterion for minimally invasive adenocarcinoma. Warth et al. 9 reported that STAS significantly reduced recurrence-free survival (RFS), overall survival (OS), and disease-free survival in patients with resected adenocarcinomas of any stage and that this trend was the same in patients with widespread STAS and those with limited STAS. However, the significance of quantitative assessment of STAS has not been reported. The aim of this study was to investigate the relation between the STAS and clinical, pathologic, and prognostic factors as well as patients’ survival.
Materials and Methods
The study was approved by the institutional review board of Toranomon Hospital. A total of 208 patients with small (≤2 cm) stage I lung adenocarcinoma (minimally invasive or invasive) and no lymph node metastasis were identified as having undergone surgical resection including both lobectomy and limited resection between 2003 and 2009 at our institution. Patients with variants of adenocarcinoma (invasive mucinous, fetal, or colloid adenocarcinoma), multiple nodules, or other components (such as squamous or neuroendocrine differentiation) were excluded. Information on patients’ demographics, the surgical procedures performed, and postoperative follow-up was retrieved from the medical records. RFS was measured from the date of resection to development of clinical/radiographic recurrence of disease. Recurrences were categorized as locoregional or distant. 7 10 Patients who were alive without documented clinical or radiographic recurrence were censored on the date of last follow-up. OS was measured from the date of resection to death from any cause.
All archived hematoxylin and eosin–stained tumor slides (a median of three per patient, range 1–8) were reviewed by two of the authors (H. U. and T. F.). Each case was evaluated for histologic pattern on the basis of the WHO 2015 classification 8 ; necrosis; and presence or absence of pleural invasion, lymphatic invasion, and vascular invasion. The final stage of the tumor was determined in accordance with the American Joint Committee on Cancer classification, seventh edition. 11
Examination of STAS
The definition of STAS used in this study was based on the original report by Kadota et al. 7 First, we examined all tumor edges and selected one to three fields with abundant STAS. STAS was then assessed semiquantitatively in the most prominent area by using a BX53 microscope with a 20× objective and a 10× ocular lens (Olympus, Tokyo, Japan). Patients with micropapillary or solid nest–predominant STAS according to the original report by Kadota et al. 7 were classified as having no STAS, low STAS (1–4 clusters of micropapillary or solid nest–predominant STAS), or high STAS (≥5 clusters of STAS) as shown in Figure 1 . Patient with single cell–predominant STAS were classified as having no STAS, low STAS (1–4 tumor cells), or high STAS (≥5 tumor cells).
A univariate analysis was performed by the Clinical Study Support Company (Nagoya, Japan) with the chi-square test to determine the impact of clinicopathologic parameters on STAS. This analysis was conducted with SAS software, version 9.4 (SAS Institute Inc., Cary, NC). The RFS and OS analyses were performed by one of the authors (H. U.) using the Kaplan-Meier estimator and the log-rank test with SPSS software, version 23 (IBM Corp., Armonk, NY). Univariate and multivariate survival analysis was performed by the Clinical Study Support Company with a Cox proportional hazards regression model to adjust for potential confounders of the association between clinicopathologic characteristics and RFS/OS independently with SAS software, version 9.4. Spearman's test was performed by the Clinical Study Support Company to clarify the relationship between STAS and RFS/OS, also using SAS software, version 9.4. All p values were two sided and considered to be statistically significant at less than 0.05.
STAS and Patient Characteristics
The mean age of the study cohort (n = 208) was 66 years (range 32–87), and 51% were men. Of the 208 patients, 163 (78%) underwent lobectomy with ND2a-1 lymphadenectomy for patients younger than 80 years and ND1b lymphadenectomy for patients 80 years or older by video-assisted thoracic surgery, and 45 (22%) underwent limited resection with or without lymphadenectomy by video-assisted thoracic surgery. STAS was assessed as no STAS in 109 cases (52.4%), low STAS in 38 cases (18.3%), and high STAS in 61 cases (29.3%) ( Table 1 ). Of the 208 patients, 180 (87%) were pathologic stage T1aN0M0 and the remaining 28 (13%) were pathologic stage T2aN0M0.
n = 109
n = 38
n = 61
|Sex, M/F, n (%)||51 (47)/58 (53)||22 (58)/16 (42)||33 (54)/28 (46)||0.421|
|Age, y, n (%)|
|<65||42 (39)||21 (55)||30 (49)||0.143|
|≥65||67 (61)||17 (45)||31 (51)|
|Smoking, n (%)|
|Never||60 (55)||16 (42)||23 (38)||0.072|
|Former/current||49 (45)||22 (58)||38 (62)|
|Operation, n (%)|
|Lobectomy||75 (69)||30 (79)||58 (95)||<0.001|
|Limited resection||34 (31)||8 (21)||3 (5)|
|Tumor size, n (%)|
|<10 mm||13 (12)||0 (0)||3 (5)||0.037|
|≧10 mm||96 (88)||38 (100)||58 (95)|
|WHO 2015 classification, n (%)|
|MIA||39 (36)||0 (0)||0 (0)||<0.001|
|Lepidic (IA)||17 (16)||6 (16)||2 (3)|
|Acinar||29 (27)||21 (55)||31 (51)|
|Papillary||22 (20)||8 (21)||18 (30)|
|Solid||2 (2)||3 (8)||10 (16)|
|Presence of necrosis, n (%)||6 (6)||6 (16)||6 (10)||0.141|
|Presence of pleural invasion, n (%)||5 (5)||9 (24)||14 (23)||<0.001|
|Presence of lymphatic invasion, n (%)||26 (24)||21 (55)||43 (70)||<0.001|
|Presence of vascular invasion, n (%)||11 (10)||10 (26)||31 (51)||<0.001|
There was a statistically significant association between higher STAS and solid predominant invasive adenocarcinoma ( p < 0.001), pleural invasion ( p < 0.001), lymphatic invasion ( p < 0.001), vascular invasion ( p < 0.001), and tumor size of 10 mm or more ( p = 0.037) by the chi-square test (see Table 1 ). When the WHO 2015 classification was used, high STAS was found to be present in only two (8%) of 25 lepidic dominant invasive adenocarcinomas but in 10 (67%) of 15 solid predominant invasive adenocarcinomas.
The frequency of lymphatic invasion increased with increasing STAS (26 of 109 cases [24%] with no STAS, 21 of 38 [55%] with low STAS, and 43 of 61 [70%] with high STAS). A similar trend was seen for pleural and vascular invasion. STAS was present in only three of 16 tumors with a diameter less than 10 mm (19%) but in 96 of 192 tumors with a diameter of 10 mm or more (50%). Patients with higher STAS underwent lobectomy ( p < 0.001); three of 61 patients with high STAS (5%) underwent limited resection (see Table 1 ).
Relationship between STAS and RFS
Recurrence and disease progression occurred in 17 cases during follow-up. The incidence of recurrence was 2% (two of 109) for no STAS, 5% (two of 38) for low STAS, and 21% (13 of 61) for high STAS. There was a significant association between increasing STAS and shorter RFS (154.2 months with no STAS [95% confidence interval (CI): 149.6–158.8], 147.6 months with low STAS [95% CI: 137.5–157.7], and 115.6 months with high STAS [95% CI: 101.2–130.1]; p < 0.001) ( Fig. 2 ). Higher STAS ( p = 0.001), pleural invasion ( p < 0.001), and lymphatic invasion ( p = 0.014) were significantly unfavorable prognostic factors for RFS in the univariate Cox proportional hazards model ( Supplementary Table 1 ). However, only STAS remained a significant ( p = 0.015) predictor of RFS in the multivariate Cox proportional hazards model ( Table 2 ).
|Category||Adjusted HR||95% CI||p Value|
|Low vs. no||1.651||0.228–11.968||0.015|
|High vs. no||7.347||1.535–35.174|
|High vs. low||4.450||0.963–20.558|
|Presence of lymphatic invasion|
|Presence vs. absence||1.493||0.41–5.444||0.544|
|Presence of vascular invasion|
|Presence vs. absence||0.745||0.231–2.395||0.621|
|WHO 2015 classification|
|MIA vs. acinar||0.000||0.640|
|Lepidic (IA) vs. acinar||0.716||0.080–6.439|
|Papillary vs. acinar||0.792||0.243–2.584|
|Solid vs. acinar||2.680||0.654–10.987|
In the patients who underwent lobectomy (n = 163), higher STAS was associated with shorter RFS (140.4 months with no STAS [95% CI: 134.8–146.0], 150.4 months with low STAS [95% CI: 141.3–159.5], and 116.3 months with high STAS [95% CI: 101.5–131.0]; p < 0.001) ( Supplementary Fig. 1 A ). In patients who underwent limited resection (n = 45), there was no statistically significant relationship between STAS and RFS (152.9 with months no STAS [95% CI: 144.2–161.7], 117.5 months with low STAS [95% CI: 86.0–148.9], and 85.8 months with high STAS [95% CI: 28.1–143.6]; p = 0.084) ( Supplementary Fig. 1 B ). The data for type of operation and recurrence are shown in Supplementary Table 2 . Two patients with STAS who underwent limited resection experienced distant recurrence.
Relationship between STAS and OS
There were 20 deaths from any cause during follow-up. Nine of the 20 deaths occurred in patients with recurrence of lung cancer, and eight of these deaths were directly cancer related (the cause of death in the remaining patient was aspiration pneumonia). Five of the 11 patients without recurrence of lung cancer died for reasons related to another type of cancer; the cause was unknown in two patients, and the cause of death in one patient each was pneumonia, interstitial pneumonia, acute subdural hematoma, and road traffic accident. There was a significant association between increasing STAS and shorter OS (151.0 months with no STAS [95% CI: 145.4–156.5], 135.7 months with low STAS [95% CI: 120.6–150.9], and 127.3 months with high STAS [95% CI: 115.8–138.8]; p = 0.020) (see Fig. 2 ).
The results for overall survival in the univariate Cox proportional hazards model are shown in Supplementary Table 3 . A solid predominant pattern according to the WHO 2015 classification ( p < 0.001), higher STAS ( p = 0.034), and presence of necrosis ( p = 0.002) were statistically significant unfavorable prognostic factors for OS. The multivariate Cox proportional hazards model included age, type of operation, the WHO 2015 classification, higher STAS, and presence of necrosis, and presence of necrosis ( p = 0.020), identified age ( p = 0.021), and type of operation ( p = 0.023) as significant factors for unfavorable prognosis ( Supplementary Table 4 ).
Higher STAS was seen more often in patients with solid predominant invasive adenocarcinoma, pleural and lymphovascular invasion, and tumor size of 10 mm or more. Univariate and multivariate Cox proportional hazards regression analysis in this study showed that higher STAS was a significant unfavorable prognostic factor for RFS in patients with small lung adenocarcinoma. We found that about 20% of patients with high STAS developed a recurrence.
Clinicopathologic parameters related to STAS have been reported in three lung adenocarcinoma cohorts. In a report of 411 resected small (≤2 cm) stage I lung adenocarcinomas by Kadota et al., 7 lymphovascular invasion as well as micropapillary and solid predominant patterns were associated with presence of STAS. Micropapillary predominant, high-stage, node-positive, and metastasized adenocarcinomas were significantly associated with the presence of STAS in 569 resected pulmonary adenocarcinomas of any stage reported by Warth et al. 9 Finally, male sex, cigarette smoking, solid nodules, stage IB disease, and lymphovascular and pleural invasion were related to the presence of STAS in 318 stage I adenocarcinomas reported by Shiono and Yanagawa. 12 Our results for clinicopathologic parameters classified according to no, low, or high STAS are consistent with the findings of these three studies.
An early report indicated that aerogenous tumor spread is an unfavorable prognostic factor in patients with mucinous or nonmucinous adenocarcinoma, 13 and a subsequent study reported that clusters of floating cancer cells were a significant prognostic factor in patients with metastatic lung cancer of colorectal origin, regardless of whether they had undergone lobectomy or limited resection. 14 15 Onozato et al. 16 also pointed out that a large collection of tumor cells (a “tumor island”) in the alveolar space increased the risk for recurrence in patients with stage I lung adenocarcinoma. Three-dimensional reconstruction analysis showed that the tumor island was connected to the main tumor. Therefore, a tumor island was a type of aerogenous tumor spread, although whether tumor islands should be included in STAS is still debatable. 17 Finally, Kadota et al. 7 described tumor cells that spread within air spaces in the lung parenchyma beyond the edge of the main tumor and coined the term STAS . In the report, the presence of STAS was associated with a risk for recurrence in patients with small adenocarcinoma who had undergone limited resection. The original definition did not include the number of clusters of STAS. Warth et al. 9 subsequently defined STAS as a small solid nest containing five or more tumor cells and reported an association between STAS and significantly reduced OS and disease-free survival in a series of patients with resected stage I to IV pulmonary adenocarcinoma. Although there were slight differences about the definition of STAS, STAS was prognostic factor. 18 Our findings are consistent with those of the aforementioned studies.
In our study, STAS was a significant prognostic factor. Two patients with STAS who underwent limited resection experienced recurrence, which was distant in both cases. In the report by Kadota et al, 7 presence of STAS increased the risk for recurrence in patients who underwent limited resection but not in those who underwent lobectomy. They speculated that STAS may remain undetected in the alveolar space beyond the surgical margin and cause locoregional recurrence. However, in the cohort of patients with completely resected stage I to IV pulmonary adenocarcinoma reported by Warth et al, 9 STAS was still associated with decreased disease-free survival and OS. Our results are consistent with those in the report by Warth et al. 9 The exact reason for the association between STAS and prognosis is unknown, but tumor activity at the invasive front may be important.
This study has several limitations. First, it was retrospective in nature and conducted in a single center. Second, the molecular alterations were unknown. STAS has been reported to be associated with lower rates of EGFR mutation but higher rates of BRAF mutation. 9 12 Nevertheless, further studies of the molecular correlates of STAS are needed.
In summary, higher STAS based on semiquantitative assessment was seen more often in patients with resected solid predominant small adenocarcinoma, pleural and lymphovascular invasion, and a larger tumor size and was associated with worse RFS.
The authors did not receive any funding in the form of grants to conduct this study. We thank Tetsumi Toyoda (Clinical Study Support Company, Nagoya, Japan) for assistance with the statistical analysis.
- 1R.L. Siegel, K.D. Miller, A. Jemal. Cancer statistics, 2015. CA Cancer J Clin. 2015;65 :5-29
- 2Center for Cancer Control and Information Services NCC. Cancer statistics in Japan., . Accessed April 26, 2014.http://ganjoho.jp/reg-stat/
- 3R. Rami-Porta, D. Ball, J. Crowley. The IASLC Lung Cancer Staging Project: proposals for the revision of the T descriptors in the forthcoming (seventh) edition of the TNM classification for lung cancer. J Thorac Oncol. 2007;2 :593-602
- 4W.D. Travis, E. Brambilla, M. Noguchi. International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol. 2011;6 :244-285
- 5A. Yoshizawa, N. Motoi, G.J. Riely. 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. Mod Pathol. 2011;24 :653-664
- 6A. Warth, T. Muley, M. Meister. 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. J Clin Oncol. 2012;30 :1438-1446
- 7K. Kadota, J. Nitadori, C.S. Sima. 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. J Thorac Oncol. 2015;10 :806-814
- 8W. Travis, E. Brambilla, A. Burke, A. Marx, A. Nicholson. WHO Classification of Tumours of the Lung, Pleura, Thymus and Heart. World Health Organization Classification of Tumours, vol. 7
- 9A. Warth, T. Muley, C.A. Kossakowski. Prognostic impact of intra-alveolar tumor spread in pulmonary adenocarcinoma. Am J Surg Pathol. 2015;39 :793-801
- 10J. Nitadori, A.J. Bograd, K. Kadota. Impact of micropapillary histologic subtype in selecting limited resection vs lobectomy for lung adenocarcinoma of 2cm or smaller. J Natl Cancer Inst. 2013;105 :1212-1220
- 11S.B. Edge, D.R. Byrd, C.C. Compton, A.G. Fritz, F.L. Greene, A. Trotti. AJCC Cancer Staging Manual. 7th ed. 2010 Springer: New York, NY
- 12S. Shiono, N. Yanagawa. Spread through air spaces is a predictive factor of recurrence and a prognostic factor in stage I lung adenocarcinoma. Interact Cardiovasc Thorac Surg. 2016;23 :567-572
- 13Bronchioloalveolar carcinomas. Cell types, patterns of growth, and prognostic correlates. Cancer. 1986;57 :1555-1564
- 14S. Shiono, G. Ishii, K. Nagai. Predictive factors for local recurrence of resected colorectal lung metastases. Ann Thorac Surg. 2005;80 :1040-1045
- 15S. Shiono, G. Ishii, K. Nagai. Histopathologic prognostic factors in resected colorectal lung metastases. Ann Thorac Surg. 2005;79 :278-282
- 16M.L. Onozato, A.E. Kovach, B.Y. Yeap. Tumor islands in resected early-stage lung adenocarcinomas are associated with unique clinicopathologic and molecular characteristics and worse prognosis. Am J Surg Pathol. 2013;37 :287-294
- 17V. Morales-Oyarvide, M. Mino-Kenudson. Tumor islands and spread through air spaces: distinct patterns of invasion in lung adenocarcinoma. Pathol Int. 2016;66 :1-7
- 18A. Warth, M.B. Beasley, M. Mino-Kenudson. Breaking new ground: the evolving concept of spread through air spaces (STAS). J Thorac Oncol. 2017;12 :176-178
© 2017 International Association for the Study of Lung Cancer. Published by Elsevier Inc.