Nonpredominant lepidic pattern correlates with better outcome in invasive lung adenocarcinoma

Lung Cancer 90 (2015) 568–574

Contents lists available at ScienceDirect

Lung Cancer journal homepage: www.elsevier.com/locate/lungcan

Nonpredominant lepidic pattern correlates with better outcome in invasive lung adenocarcinoma Johanna M. Mäkinen a,b,∗ , Kirsi Laitakari c , Shirley Johnson c , Riitta Mäkitaro c , Risto Bloigu d , Elisa Lappi-Blanco a,b,1 , Riitta Kaarteenaho c,e,f,1 a

Department of Pathology, Center for Cancer Research and Translational Medicine, University of Oulu, POB 5000, 90014 Oulu, Finland Medical Research Center Oulu, Oulu University Hospital and University of Oulu, POB 20, 90029 Oulu, Finland c Department of Internal Medicine, Respiratory Research Unit, Medical Research Center Oulu, Oulu University Hospital and University of Oulu, POB 20, 90029 Oulu, Finland d Medical Informatics and Statistics Research Group, University of Oulu, POB 5000, 90014 Oulu, Finland e Unit of Medicine and Clinical Research, Pulmonary Division, University of Eastern Finland, POB 1627, 70211 Kuopio, Finland f Center of Medicine and Clinical Research, Division of Respiratory Medicine, Kuopio University Hospital, POB 100, 70029 Kuopio, Finland b

a r t i c l e

i n f o

Article history: Received 16 June 2015 Received in revised form 17 September 2015 Accepted 11 October 2015 Keywords: Adenocarcinoma Histology Lepidic Lung cancer Prognosis Subtype

a b s t r a c t Objectives: Histologic heterogeneity is a typical feature of pulmonary adenocarcinoma. This study aimed to deconstruct the intratumoral growth pattern composition and to examine the prognostic relevance of the current lung adenocarcinoma classification in a series of Finnish lung cancer patients. Materials and methods: A cohort of 112 patients with surgically operated stage I–IV lung adenocarcinoma was retrospectively reviewed. Histologic subtyping was performed according to the classification system established by the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society (IASLC/ATS/ERS). Systematically collected clinical information including survival data was correlated with the subtype status. In addition, emphasis was placed on the nonpredominant histologic patterns, gender and smoking history. Results: The most common subtype was acinar predominant adenocarcinoma with 56 cases (50%). Most tumors were composed of a mixture of two or more growth patterns, and single pattern tumors were rare (9.8%). Micropapillary predominant adenocarcinoma and solid predominant adenocarcinoma were the subtypes with the lowest disease-specific survival rates (5-year DSS 21.4% and 30.4%; shared mean DSS 46.3 months, p = 0.040). A nonpredominant lepidic component was observed in 24 (21.4%) tumors, and its presence predicted a better outcome (mean DSS 127.4 months vs. 55.7 months, p = 0.001). This association was confirmed by multivariate analysis (p = 0.004). Solid pattern and solid, papillary, micropapillary and cribriform predominant histology associated with smoking (p < 0.001), while mucinous pattern was more common in nonsmokers (p < 0.001) and in women (p = 0.050). Conclusions: Micropapillary and solid predominant adenocarcinomas showed significantly lower survival rate than other major subtypes, yet the prognostic value of the current lung adenocarcinoma classification is not limited only to the predominant growth patterns. The more favorable outcome associated with the nonpredominant lepidic pattern further emphasizes the importance of histologic subtyping and assessment of tumor heterogeneity in the diagnostics of lung adenocarcinoma. © 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

∗ Corresponding author at: Department of Pathology, Center for Cancer Research and Translational Medicine, University of Oulu, POB 5000, 90014 Oulu, Finland. Fax: +358 8 315 2177. E-mail address: johanna.makinen@oulu.fi (J.M. Mäkinen). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.lungcan.2015.10.014 0169-5002/© 2015 Elsevier Ireland Ltd. All rights reserved.

Lung cancer remains the leading cause of global cancer incidence and mortality [1]. In most countries, adenocarcinoma has become the most predominant histologic subtype accounting for approximately a half of all lung cancers [2]. Primary lung adenocarcinoma is known to be a very heterogeneous tumor in many aspects, and both its histopathology and the course of the disease

J.M. Mäkinen et al. / Lung Cancer 90 (2015) 568–574

are diverse. Acknowledging this, a proposal for a multidisciplinary lung adenocarcinoma classification was established in 2011 [3]. To date, the validity of the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society (IASLC/ATS/ERS) classification has been tested in multiple independent studies beginning from the early reports by Yoshizawa et al. [4] and Russell et al. [5], and most recently, the classification was adopted by the World Health Organization (WHO) [6]. In 2011, Yoshizawa et al. reported that the IASLC/ATS/ERS classification identified histologic subsets with significant prognostic differences in a series of stage I lung adenocarcinomas [4]. Subsequently, Russell et al. reported that subtyping according to the IASLC/ATS/ERS displayed a significant correlation with the 5-year survival in stage I–III lung adenocarcinomas, and in fact, the classification was the strongest predictor of patient survival, independent of the TNM stage [5]. In both studies, patients with adenocarcinoma in situ and microinvasive adenocarcinoma showed excellent 5-year survival, followed by lepidic predominant adenocarcinoma. Acinar predominant and papillary predominant adenocarcinomas shared an intermediate prognosis, while micropapillary predominant and solid predominant adenocarcinomas were associated with highgrade clinical behavior and a dismal prognosis. In 2012, the studies by Thunnissen et al. [7] and Warth et al. [8] validated the reproducibility of the IASLC/ATS/ERS classification, and they have been followed by several reports further validating its prognostic value [9–12]. Histologic subtyping has been proposed as being a potential screening method with which to identify the lung cancer patients at high risk of recurrent disease [11]. In these pioneering studies, micropapillary predominant and solid predominant adenocarcinomas are repeatedly identified as highgrade tumors. Currently, also the nonpredominant micropapillary and solid patterns are recognized as markers of unfavorable prognosis [9,13–15]. The aim of the present study was to investigate the correlation between the IASLC/ATS/ERS classification and patient outcome and clinical characteristics in a series of patients with surgically operated lung adenocarcinoma, assessing both the predominant and nonpredominant histologic patterns.

2. Materials and methods 2.1. Patients A retrospective analysis was conducted on 112 lung adenocarcinoma patients who underwent a complete surgical resection with curative intent in the Oulu University Hospital between January 1998 and December 2007. None of the patients had received preoperative chemotherapy or radiotherapy. The clinical and histologic data were re-evaluated by two clinicians and two pulmonary pathologists. Clinical information including age, sex, smoking history and follow-up data was collected systematically from the medical records using a specific formula planned for the study. A nonsmoker was defined as a person whose lifetime tobacco consumption equaled fewer than 100 cigarettes. Other relevant data, such as the location and the size of tumor, were traced from the original pathology reports (Table 1). Ethical approval for the study was obtained from the Ethical Committee of the Northern Ostrobothnia Hospital District in Oulu (statement 2/2008), and from the National Supervisory Authority for Welfare and Health (former National Authority of Medicolegal Affairs, Reg. no. 863/04/047/08). For the retrospective histologic material, informed consent permission was given by the National Supervisory Authority for Welfare and Health, the national licensing authority.

569

2.2. Histopathologic evaluation Surgical specimens had been routinely formalin fixed, paraffin embedded, sectioned, and stained with hematoxylin and eosin. Additionally, at least one representative tumor slide was stained with Alcian Blue PAS to evaluate the presence of intracytoplasmic mucin. One pathologist (JMM) reviewed all available histologic material from 112 cases, and all available tumor slides from each case were independently re-evaluated by two pathologists (JMM and ELB) according to the IASLC/ATS/ERS classification [3,16], blinded to clinical data. An average of 5 tumor slides (ranging from 1 to 16 slides) was processed per case. Five major growth patterns were defined: lepidic, acinar (including acinar and/or cribriform pattern), papillary, micropapillary and solid (Fig. 1A–F). Patterns were assessed semiquantitatively in 5% increments, and a predominant pattern was designated for each tumor. All tumors were confirmed invasive (no microinvasive adenocarcinoma or adenocarcinoma in situ included), and reclassified as either lepidic, acinar, papillary, micropapillary or solid predominant adenocarcinoma. Tumors with a single growth pattern were referred to as pure tumors. In addition, tumors exhibiting mucinous pattern were divided into invasive mucinous adenocarcinoma, mixed mucinous and nonmucinous adenocarcinoma, and colloid adenocarcinoma (Fig. 1G–H). 2.3. Pathological staging All tumors were restaged according to the TNM classification of malignant tumors (Union for International Cancer Control/American Joint Committee on Cancer, 7th edition) [17]. 2.4. Statistical analyses Statistical analyses were performed by IBM SPSS Statistics for Windows, Version 21.0 (IBM Corp. Released 2012. IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp.). Interobserver agreement over the histologic growth patterns was verified using Cohen’s kappa statistics. Associations between clinicopathologic and histologic parameters were analyzed using 2 test and Fisher’s exact test (categorical variables), and one-way analysis of variance. Survival analyses were performed using the Kaplan–Meier method with log-rank test and the Cox proportional hazard regression model. All tests were two sided, and p-values less than 0.05 were considered statistically significant. 3. Results 3.1. Patient characteristics Out of the 112 patients, 71 were men (63.4%) and 41 were women (36.6%). Their age ranged from 41 to 82 years, with a median of 66.0 years. The majority, 88 patients (78.6%) were current or ex-smokers, while 20 patients (17.9%) were nonsmokers. No smoking data was available from four patients (3.6%). Nineteen patients underwent pneumonectomy (17%), nine bilobectomy (8%), 73 lobectomy (65.2%), and 11 sublobar resection (segmentectomy or wedge resection, 9.8%) (Table 1). The postoperative 30-day mortality was 0.9% (1/112 patients). 3.2. Distribution of growth patterns According to the IASLC/ATS/ERS classification, the most frequent subtype in our study was acinar predominant adenocarcinoma (50%), followed by solid predominant adenocarcinoma (22.3%), invasive mucinous adenocarcinoma (8.9%), papillary predominant

570

J.M. Mäkinen et al. / Lung Cancer 90 (2015) 568–574

Table 1 Clinicopathological characteristics and 5-year disease specific survival (DSS) rate in lung adenocarcinoma. Univariate analysis.

All patients Gender Male Female Age (years) <65 ≥65 Smoking status Nonsmoker Former or current smoker No smoking data available Surgical procedure Sublobar resection Lobectomy Bilobectomy Pneumonectomy Tumor laterality Right Left pT T1 T2 T3 T4 pN N0 N1 N2 No LN dissection Pathological stage (TNM7) IA IB IIA IIB IIIA IIIB IV

n

(%)

5-year DSS rate (%)

112

100

41.9

71 41

63.4 36.6

40.1 45.3

51 61

45.5 54.5

44.1 40

20 88 4

17.9 78.6 3.6

47.4 40.9 33.3

11 73 9 19

9.8 65.2 8 17

40 47 25 31.6

73 39

65.2 34.8

38.5 48.4

38 55 15 4

33.9 49.1 13.4 3.6

70.1 31.1 13.3 25

64 21 12 15

57.1 18.8 10.7 13.4

56.8 28.6 10.6 21.4

32 26 20 12 17 0 5

28.6 23.2 17.9 10.7 15.2 0 4.5

77.2 40.1 35 16.7 13.7 – 0

p 0.425

0.534

0.889

0.253

0.528

<0.001

0.001

<0.001

adenocarcinoma (7.1%) and micropapillary predominant adenocarcinoma (6.3%) (Table 2). In addition, there were four cases of mixed invasive mucinous and nonmucinous adenocarcinoma (3.6%), and one case of lepidic predominant and colloid adenocarcinoma (0.9% each). Among the acinar predominant tumors we identified eight cribriform variants (7.1% of all adenocarcinomas). Most adenocarcinomas expressed histologic heterogeneity i.e. they were composed of a mixture of two or more growth patterns: One hundred and one tumors (90.2%) exhibited at least two, 50 tumors (44.6%) presented at least three, 23 tumors (20.5%) displayed at least four, and two tumors (1.8%) showed all five distinct growth patterns. The number of observed growth patterns correlated positively with the number of evaluated tumor slides (p = 0.043), but did not show significant correlation to tumor size. However, solid predominant adenocarcinomas displayed less heterogeneity than other adenocarcinomas, especially the micropapillary predominant tumors (p = 0.033). With respect to the 11 (9.8%) adenocarcinomas presenting with a single growth pattern, three were purely solid and eight were purely acinar (with

acinar and/or cribriform pattern). No cases of pure lepidic, papillary or micropapillary adenocarcinomas were identified. The vast majority, 108 tumors (96.4%), displayed acinar growth, and in 30 tumors (26.8%) the acinar component showed cribriform features. As a nonpredominant (minor) pattern, acinar pattern was observed in 39 tumors (34.8%). A nonpredominant micropapillary pattern was present in 46 tumors (41.4%), solid in 43 tumors (38.4%), and papillary in 27 tumors (24.1%). A nonpredominant lepidic tumor component was detected in 24 tumors (21.4%). Its frequency was highest in mucinous adenocarcinomas (69.2%), while it was only rarely observed in solid predominant tumors (8%, p < 0.001). The median percentage of the lepidic pattern was 5%, ranging from 5% to 45%. The interobserver kappa value for the predominant growth pattern was 0.85 (p < 0.001), indicating a very good interobserver agreement. For the nonpredominant patterns, the kappa values were the following: papillary 0.34 (p = 0.002), micropapillary 0.71 (p < 0.001), acinar/cribriform 0.71 (p < 0.001), solid 0.77 (p < 0.001),

Table 2 Adenocarcinoma subtypes according to IASLC/ATS/ERS: relationship with age, gender and smoking status and 5-year disease specific survival (DSS) rate. Subtype

n (%)

Lepidic predominant 1 (0.9) 48 (42.9) Acinar predominant 8 (7.1) Cribriform variant 8 (7.1) Papillary predominant Micropapillary predominant 7 (6.3) 25 (22.3) Solid predominant 15 (13.4) Mucinous variants

Mean age, years ± SD 64 ± 0 65.6 ± 8.2 64.5 ± 9.6 66.5 ± 12 64.4 ± 9.3 65 ± 8.3 64.6 ± 10.1

Gender M:F

Smoker:nonsmoker

5-Year DSS rate (%)

1:0 30:18 5:3 7:1 5:2 17:8 6:9

1:0 37:8 7:0 8:0 6:1 24:1 5:10

– 44.4 50 37.5 21.4 30.4 53.3

J.M. Mäkinen et al. / Lung Cancer 90 (2015) 568–574

571

Fig. 1. Histologic growth patterns of lung adenocarcinoma and mucinous adenocarcinoma variants (hematoxylin and eosin, original magnification ×100). (A) Nonmucinous lepidic, (B) acinar, (C) cribriform, (D) papillary, (E) micropapillary, and (F) solid pattern. (G) Invasive mucinous adenocarcinoma with lepidic and acinar patterns, and (H) colloid adenocarcinoma.

and lepidic 0.86 (p < 0.001), the interobserver agreement ranging from fair to very good.

3.3. Survival analysis

(20.5%) died of other causes. Of the remaining 21 patients (18.8%) alive at the end of the study, two (1.8%) had recurrent disease and 19 (17%) had no evidence of disease. The last follow-up date was 31.12.2012, and the median follow-up time 32.5 months (range 0–172 months).

In Kaplan–Meier analysis, the disease-specific 5-year survival (DSS) rate for the entire cohort was 41.9%, while the 5-year overall survival (OS) rate was 34.8%. During the follow-up, a total of 68 patients (60.7%) died of lung cancer and a further 23 patients

3.3.1. Predominant patterns For the solid predominant subtype, the 5-year DSS rate was 30.4%, and micropapillary predominant adenocarcinoma was asso-

572

J.M. Mäkinen et al. / Lung Cancer 90 (2015) 568–574

Fig. 2. Kaplan–Meier survival curves for disease-specific survival (DSS) according to (A) the IASLC/ATS/ERS classification, (B) the presence of a nonpredominant lepidic tumor component, (C) the three prognostic groups highlighting the impact of a nonpredominant lepidic pattern on the predominant growth patterns, and (D) the stage (pTNM).

ciated with a lower 5-year DSS rate (21.4%) than any other major subtype. The 5-year DSS rate for patients with papillary predominant adenocarcinoma was 37.5%, and for patients with acinar predominant tumors the rate was 45.3%. In terms of survival, the cribriform variants did not differ significantly from other acinar predominant adenocarcinomas. The mucinous adenocarcinoma variants shared the highest 5-year DSS rate, 53.3%. Our only patient with a lepidic predominant adenocarcinoma was alive with no signs of disease after five postoperative years, but died of histologically confirmed pulmonary squamous cell carcinoma three years later. Interestingly, single pattern tumors showed lower 5-year DSS rates than the more heterogeneous tumors (18.2% vs. 44.8%; p = 0.046). Fig. 2A shows the survival difference between solid and micropapillary predominant adenocarcinomas and other subtypes according to the IASLC/ATS/ERS classification (28.7% vs. 46.7% 5year DSS; p = 0.064, mean DSS 46.3 vs. 82.3 months; p = 0.040).

3.3.2. Nonpredominant patterns The presence of the nonpredominant lepidic pattern correlated with better outcome. The 5-year DSS rate for patients with a nonpredominant lepidic tumor component was 70.4% in comparison to 32.5% for patients without this component (p = 0.003; Fig. 2B), and the difference over mean DSS was 127.4 months vs. 55.7 months (p = 0.001). Both the mucinous and nonmucinous lepidic patterns were associated with a more favorable outcome (75% and 67.7%, respectively), the survival rate was higher when the lepidic component grew extensive (83.3% 5-year DSS; p = 0.011), and in multivariate analysis, nonpredominant lepidic pattern was a stage-

independent prognostic factor (HR 3.226; 95% CI, 1.468–7.090; p = 0.004). Based on the different combinations of predominant growth patterns (including both nonmucinous and mucinous histology) and a minor lepidic pattern, adenocarcinomas could be divided into three prognostic groups. The group with the most favorable prognosis included lepidic predominant tumors, and acinar or papillary predominant tumors with a nonpredominant lepidic pattern. The intermediate group included acinar or papillary predominant tumors lacking a lepidic pattern and micropapillary or solid predominant tumors with a nonpredominant lepidic pattern. Finally, the group with the worst prognosis consisted of micropapillary and solid predominant tumors lacking a lepidic component. The 5-year DSS rates for these groups were 76.2%, 37.5% and 22.5%, respectively (p = 0.002; Fig. 2C), and the mean DSS was 135.1, 64.0 and 40.0 months (p < 0.001). In multivariate analysis, this subclassification was a stage-independent predictor of survival (HR 6.560; 95% CI, 2.449–17.571; p < 0.001). The nonpredominant papillary, micropapillary and solid patterns did not possess any statistically significant prognostic value, and the simultaneous presence of a nonpredominant micropapillary and solid component or the progressive increase in their proportion did not potentiate their negative effect on survival.

3.3.3. Stage Stage (pTNM) was a strong predictor of survival (p < 0.001). The 5-year DSS rates for stage IA, IB, IIA, IIB, IIIA and IV patients were 77.2%, 40.1%, 35%, 16.7%, 13.7% and 0%, respectively (Table 1 and

J.M. Mäkinen et al. / Lung Cancer 90 (2015) 568–574

Fig. 2D). The prognostic impact of tumor stage (pT) and nodal status (pN) is presented in Table 1. 3.4. Smoking and gender Smoking history, gender and age did not exert any direct influence on survival in univariate analyses. However, smoking and female gender correlated with the subtype status (Table 2). The majority, 88.7% of men and 61% of women were smokers; 75% of nonsmokers in the study were female (p < 0.001). Smoking was strongly associated with papillary predominant (100% smokers), solid predominant (96%), and micropapillary predominant subtypes (87.5%), and the cribriform variant of acinar predominant adenocarcinoma (87.5%; p = 0.001). The solid predominant adenocarcinoma was the most common subtype found in smokers (27.3%), and the solid pattern in itself correlated with smoking. Among smokers, 68.2% of tumors displayed a solid component, while in the nonsmoking group its prevalence was only 20% (p < 0.001), and only one case of solid predominant adenocarcinoma was detected. Mucinous adenocarcinomas were associated with a nonsmoking history. Every second tumor in nonsmokers was designated as a mucinous variant with the most common subtype being invasive mucinous adenocarcinoma (35%). In smokers, only 5.7% of tumors exhibited mucinous pattern (p < 0.001). At the same time, 60% of mucinous adenocarcinomas were observed in women (p = 0.050). 4. Discussion In this study, we have shown a significant correlation between disease-specific survival (DSS) and the predominant histologic subtypes based on the IASLC/ATS/ERS classification in a cohort of 112 surgically operated lung adenocarcinomas. Moreover, we have demonstrated that the presence of a nonpredominant lepidic component correlates with a better prognosis. Furthermore, we found that mucinous pattern was associated with female gender and nonsmoking history, whereas solid pattern along with solid, papillary, micropapillary and cribriform predominant histology was more frequent in smokers. The presence of a nonpredominant lepidic tumor component correlated with better outcome independent of stage. As far as we are aware, this finding of the positive effect of a nonpredominant lepidic pattern has not been previously reported, although the excellent prognosis of lepidic predominant adenocarcinoma [5,9,18], and its correlation with epidermal growth factor receptor (EGFR) mutation status [12,19,20] has been documented. The nuclear expression of maspin, a tumor suppressor protein, has been observed in association with lepidic growth pattern, whereas the combined nuclear and cytoplasmic expression profile has been linked to invasion [21]. Araki et al. proposed that the low incidence of lymphatic vessel invasion in lepidic predominant adenocarcinoma could be the key factor for its favorable prognosis [18]. More recently, the presence of tumor spread through air spaces (STAS) has been shown to be exceedingly rare in lepidic predominant adenocarcinoma [22]. It is possible that the noninvasive nature of the nonpredominant lepidic component may alter lung cancer behavior. Furthermore, our finding of the rarity of the lepidic component in solid predominant adenocarcinomas may point to differences in the pathogenesis of these patterns. In line with previous reports [4,5,10], micropapillary and solid predominant subtypes predicted a more aggressive course of disease and poor prognosis. Patients with papillary predominant adenocarcinoma approached the low survival rate of these high-

573

grade subtypes, especially the solid predominant adenocarcinoma. While papillary tumors have usually been associated with an intermediate prognosis [4,5], our observation is consistent with the data published by Warth et al. who reported papillary predominant adenocarcinoma having survival rates similar to micropapillary and solid predominant adenocarcinomas [10]. The association between the risk of recurrence and a nonpredominant micropapillary or solid tumor component has been demonstrated in several recent studies [9,13–15]. In the current study, nonpredominant micropapillary and solid patterns did not display any significant prognostic relevance, but these observations may have been affected by the relatively small number of patients studied. The most common subtype was acinar predominant adenocarcinoma with a prevalence of 50%. The percentage of cribriform variants of acinar predominant adenocarcinomas (7.1% of all adenocarcinomas) corresponds to the figures reported earlier [23,24]. In our material, no survival difference could be observed between acinar predominant adenocarcinoma and its cribriform variant. The IASLC/ATS/ERS classification recommends to include cribriform predominant tumors under the acinar predominant subtype [3], and this practice is supported by our findings. However, there is growing evidence that cribriform growth pattern associates with other high-grade tumor components and metastatic potential, and should therefore be regarded as a high-grade variant of the acinar subtype [23–25]. Invasive mucinous adenocarcinoma has been considered as a subtype with an intermediate or even a poor prognosis [4,5,11]. In the present study, mucinous adenocarcinoma variants were associated with a surprisingly favorable outcome. The number of mucinous tumors in our study is low, but this finding could support a semiquantitative growth pattern analysis of mucinous tumors. Mucinous histology was associated with female gender and nonsmoking history. At the same time, the solid predominant subtype was observed more often in smokers, and solid pattern correlated with smoking history. The other subtypes associated with smoking were papillary and micropapillary predominant adenocarcinoma and the cribriform variant of acinar predominant adenocarcinoma. A recent study reported a higher frequency of lepidic features in never-smokers [26], and others have shown a connection between smoking and poorly differentiated adenocarcinoma components, namely solid [27] and cribriform [23,28] adenocarcinoma. Our results are in line with the recent studies indicating that the pathogenesis of lung cancer in smokers and nonsmokers may be different [29]. Male gender and smoking history have been associated with poorer survival in lung cancer, but we could not detect any significant survival differences attributable to either gender or smoking habits [4,12,27]. However, for example, our patients differ from the Japanese adenocarcinoma patients on the basis of higher prevalence of smoking and higher male–female ratio. Furthermore, EGFR mutations are likely to play a more important role in Asian than in Western countries. We conclude that the prognostic value of the current lung adenocarcinoma classification is not limited only to the predominant growth patterns. The more favorable clinical outcome associated with the nonpredominant lepidic pattern further emphasizes the importance of conducting a semiquantitative analysis of the histologic patterns in the diagnostics of lung adenocarcinoma. Our observations may yield novel prognostic biomarkers, and should be further validated in future studies.

Conflict of interest The authors declare no conflicts of interest.

574

J.M. Mäkinen et al. / Lung Cancer 90 (2015) 568–574

Sources of funding This work has been supported by the Foundation of the Finnish Anti-Tuberculosis Association, the Jalmari and Rauha Ahokas Foundation, the National Graduate School of Clinical Investigation, the Northern Finland Cancer Foundation, and a state subsidy of the Oulu University Hospital. Acknowledgement The authors wish to thank Dr. Ewen MacDonald for providing assistance with the language. References [1] A. Jemal, F. Bray, M.M. Center, J. Ferlay, E. Ward, D. Forman, Global cancer statistics, CA Cancer J. Clin. 61 (2011) 69–90. [2] S.S. Devesa, F. Bray, A.P. Vizcaino, D.M. Parkin, International lung cancer trends by histologic type: male:female differences diminishing and adenocarcinoma rates rising, Int. J. Cancer 117 (2005) 294–299. [3] W.D. Travis, E. Brambilla, M. Noguchi, et al., International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society international multidisciplinary classification of lung adenocarcinoma, J. Thorac. Oncol. 6 (2011) 244–285. [4] A. Yoshizawa, N. Motoi, G.J. Riely, 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, Mod. Pathol. 24 (2011) 653–664. [5] P.A. Russell, Z. Wainer, G.M. Wright, M. Daniels, M. Conron, R.A. Williams, Does lung adenocarcinoma subtype predict patient survival?: A clinicopathologic study based on the new International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society international multidisciplinary lung adenocarcinoma classification, J. Thorac. Oncol. 6 (2011) 1496–1504. [6] W.D. Travis, E. Brambilla, A.P. Burke, A. Marx, A.G. Nicholson, WHO Classification of Tumours of the Lung, Pleura, Thymus and Heart, 4th ed., IARC, Lyon, 2015. [7] E. Thunnissen, M.B. Beasley, A.C. Borczuk, et al., Reproducibility of histopathological subtypes and invasion in pulmonary adenocarcinoma. An international interobserver study, Mod. Pathol. 25 (2012) 1574–1583. [8] A. Warth, A. Stenzinger, A.C. von Brunneck, et al., Interobserver variability in the application of the novel IASLC/ATS/ERS classification for pulmonary adenocarcinomas, Eur. Respir J. 40 (2012) 1221–1227. [9] L.M. Solis, C. Behrens, M.G. Raso, et al., Histologic patterns and molecular characteristics of lung adenocarcinoma associated with clinical outcome, Cancer 118 (2012) 2889–2899. [10] A. Warth, T. Muley, M. Meister, 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, J. Clin. Oncol. 30 (2012) 1438–1446. [11] J. Gu, C. Lu, J. Guo, et al., Prognostic significance of the IASLC/ATS/ERS classification in Chinese patients—a single institution retrospective study of 292 lung adenocarcinoma, J. Surg. Oncol. 107 (2013) 474–480. [12] A. Yoshizawa, S. Sumiyoshi, M. Sonobe, et al., Validation of the IASLC/ATS/ERS lung adenocarcinoma classification for prognosis and association with EGFR and KRAS gene mutations: analysis of 440 Japanese patients, J. Thorac. Oncol. 8 (2013) 52–61.

[13] S. Sumiyoshi, A. Yoshizawa, M. Sonobe, et al., Pulmonary adenocarcinomas with micropapillary component significantly correlate with recurrence, but can be well controlled with EGFR tyrosine kinase inhibitors in the early stages, Lung Cancer 81 (2013) 53–59. [14] J. Nitadori, A.J. Bograd, K. Kadota, et al., Impact of micropapillary histologic subtype in selecting limited resection vs lobectomy for lung adenocarcinoma of 2 cm or smaller, J. Natl. Cancer Inst. 105 (2013) 1212–1220. [15] M.J. Cha, H.Y. Lee, K.S. Lee, et al., Micropapillary and solid subtypes of invasive lung adenocarcinoma: clinical predictors of histopathology and outcome, J. Thorac. Cardiovasc. Surg. 147 (2014) 921–928, e2. [16] W.D. Travis, E. Brambilla, M. Noguchi, et al., Diagnosis of lung adenocarcinoma in resected specimens: implications of the 2011 International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society classification, Arch. Pathol. Lab Med. 137 (2013) 685–705. [17] V.W. Rusch, H.D. Appleman, E. Blackstone, Lung, et al., AJCC Cancer Staging Handbook from the AJCC Cancer Staging Manual, in: S.B. Edge, D.R. Byrd, C.C. Compton, A.G. Fritz, F.L. Greene, A. Trotti (Eds.), 7th ed., Springer, 2010, pp. 299–323. [18] K. Araki, Y. Kidokoro, K. Hosoya, et al., Excellent prognosis of lepidic-predominant lung adenocarcinoma: low incidence of lymphatic vessel invasion as a key factor, Anticancer Res. 34 (2014) 3153–3156. [19] N. Rekhtman, D.C. Ang, G.J. Riely, M. Ladanyi, A.L. Moreira, KRAS mutations are associated with solid growth pattern and tumor-infiltrating leukocytes in lung adenocarcinoma, Mod. Pathol. 26 (2013) 1307–1319. [20] C. Villa, P.T. Cagle, M. Johnson, et al., Correlation of EGFR mutation status with predominant histologic subtype of adenocarcinoma according to the new lung adenocarcinoma classification of the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society, Arch. Pathol. Lab Med. 138 (2014) 1353–1357. [21] F. Lonardo, H. Guan, S. Dzinic, S. Sheng, Maspin expression patterns differ in the invasive versus lepidic growth pattern of pulmonary adenocarcinoma, Histopathology 65 (2014) 757–763. [22] A. Warth, T. Muley, C.A. Kossakowski, et al., Prognostic impact of intra-alveolar tumor spread in pulmonary adenocarcinoma, Am. J. Surg. Pathol. 39 (2015) 793–801. [23] K. Kadota, Y.C. Yeh, C.S. Sima, 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, Mod. Pathol. 27 (2014) 690–700. [24] A.L. Moreira, P. Joubert, R.J. Downey, N. Rekhtman, Cribriform and fused glands are patterns of high-grade pulmonary adenocarcinoma, Hum. Pathol. 45 (2014) 213–220. [25] L. Xu, F. Tavora, A. Burke, Histologic features associated with metastatic potential in invasive adenocarcinomas of the lung, Am. J. Surg. Pathol. 37 (2013) 1100–1108. [26] J. Mazieres, I. Rouquette, B. Lepage, et al., Specificities of lung adenocarcinoma in women who have never smoked, J. Thorac. Oncol. 8 (2013) 923–929. [27] R. Maeda, G. Ishii, J. Yoshida, T. Hishida, M. Nishimura, K. Nagai, Influence of cigarette smoking on histological subtypes of stage I lung adenocarcinoma, J. Thorac. Oncol. 6 (2011) 743–750. [28] A.C. Mackinnon Jr, A. Luevano, L.C. de Araujo, N. Rao, M. Le, S. Suster, Cribriform adenocarcinoma of the lung: clinicopathologic, immunohistochemical, and molecular analysis of 15 cases of a distinctive morphologic subtype of lung adenocarcinoma, Mod. Pathol. 27 (2014) 1063–1072. [29] M. Schiavon, G. Marulli, N. Nannini, et al., COPD-related adenocarcinoma presents low aggressiveness morphological and molecular features compared to smoker tumours, Lung Cancer 86 (2014) 311–317.