Prognostic Significance of Predominant Histologic Pattern and Nuclear Grade in Resected Adenocarcinoma of the Lung

Transcription

1 ORIGINAL ARTICLE Prognostic Significance of Predominant Histologic Pattern and Nuclear Grade in Resected Adenocarcinoma of the Lung Potential Parameters for a Grading System Jan H. von der Thüsen, MD, PhD,* Yuen S. Tham, BSc,* Holly Pattenden, BSc, Alexandra Rice, FRCPath,* Michael Dusmet, MD, Eric Lim, FRCS (C-Th), MD, and Andrew G. Nicholson, FRCPath, DM* Introduction: Currently, no agreed histologic grading system exists for lung adenocarcinomas (ADCs). With a recently updated consensus classification, the aim of this study was to assess potential prognostic factors identifiable on routine histology, which might be used as grading parameters. Methods: A retrospective study of resected pulmonary ADCs (n = 238) in patients with stage IA to IIIB disease was carried out in which various histopathological parameters were correlated with survival data. The relationship between these factors and patient survivability was analyzed using Cox proportional hazards regression. Results: Mitotic rate was found to be a highly significant prognostic marker (p = 0.008), as was overall nuclear grade (p < 0.001). ADC subtyping was also found to be potentially important, as lepidic predominant (hazard ratio 0.99, p = 0.023) and solid predominant (hazard ratio 1.01, p = 0.003) subtypes were found to be independent (to age and tumor, node, metastasis category) prognostic predictors. Vessel invasion within tumor approached significance as a negative prognostic factor (p = 0.067). Conclusions: This study showed not only that histologic subtype and mitotic rate are important prognostic factors in lung ADCs, but also that other criteria described previously may not be useful in our specific patient population. Key Words: Lung adenocarcinoma, Staging, Grading, Histologic parameters, Prognostic factor. (J Thorac Oncol. 2013;8:37 44) Lung cancer is a global disease with consistently one of the highest annual rates of cancer incidence and mortality, 1 with adenocarcinoma (ADC) currently the most common type Departments of *Histopathology and Thoracic Surgery, Royal Brompton and Harefield NHS Foundation Trust, London, United Kingdom; and Department of Pathology, Medisch Centrum Haaglanden, Den Haag, The Netherlands. Disclosure: The authors declare no conflict of interest. Address for correspondence: Jan H. von der Thüsen, MD, PhD, Department of Pathology, Medisch Centrum Haaglanden, Lijnbaan 32, 2512 VA Den Haag, The Netherlands. j.von.der.thusen@mchaaglanden.nl Copyright 2012 by the International Association for the Study of Lung Cancer ISSN: /12/ and increasing in incidence. 2 Despite recent advances in surgery and oncology, 3 lung ADC is still associated with a poor prognosis and survival, and is currently responsible for 40% of worldwide lung cancer cases, with an annual incidence of 1.5 million patients, and accounting for approximately 12% of global cancer burden. 4 Hence, pulmonary ADC is an important disease that will prove to be an increasingly major health economic challenge. Morphologically, ADCs are diverse with respect to many characteristics within and peripheral to the tumor. In particular, four histologic patterns (lepidic previously termed bronchioloalveolar, acinar, papillary and solid) were recommended as individual subtypes in the World Health Organization 2004 classification, 5 although this proved of little clinical value as most were classified as a fifth mixed category. 6 Subsequently, a multidisciplinary consensus proposal for reclassification was undertaken, adding a fifth pattern, micropapillary, 7 and recommending that ADCs be subtyped by predominant histologic pattern. 8 Evidence in stage I tumors has additionally suggested that this may have value in a grading system. 9,10 Furthermore, nuclear characteristics have also been shown to have prognostic significance, along with several other parameters such as lymphatic and blood vessel invasion, necrosis, tumor budding, aerobic spread, and mucinous morphology With no established histologic or cytological grading system currently in place for lung ADC, the purpose of this study was to analyze readily assessable morphological characteristics to determine which might be best used to construct a grading system. MATERIALS AND METHODS Sample Size This study falls under Research Ethics Committee (REC) reference no. 10/H0504/29 of the National Research Ethics Service (Southampton & South West Hampshire REC[B]). A retrospective study was conducted of 283 patients in whom a histologically proven ADC had been resected between May 1999 and August 2007 at the Royal Brompton Hospital in London. Of these, 45 patients were excluded because of a lack of minimum patient information. Of the remaining 238 patients, two had background lung fibrosis, one had received Journal of Thoracic Oncology Volume 8, Number 1, January

2 von der Thüsen et al. Journal of Thoracic Oncology Volume 8, Number 1, January 2013 radical radiotherapy, and one had been given neoadjuvant therapy. In addition, the slides of 11 cases were not available for review at the time of study. These 15 patients were also excluded. For the remaining 223 cases, standard hematoxylin and eosin (H&E) and Miller s elastin van Gieson stain (EvG) stained tumor slides were identified and retrieved from our archives, and reviewed and graded by two independent histopathologists. The number of slides retrieved per patient ranged from two to 13 (average: 6). Scoring Histologic classification of the tumors was based on International Association for the Study of Lung Cancer (IASLC)/American Thoracic Society (ATS)/ European Respiratory Society (ERS) criteria. 8 Each tumor was reviewed and histologically subtyped, recording the percentages in 5% increments relative to each histologic component. Histologic subtypes included lepidic, papillary, acinar, solid, and micropapillary (Fig.1) patterns. We further classified the tumors according to the predominant pattern present. The predominant pattern is defined as the pattern with the largest percentage, not necessarily 50% or greater. Nuclear grade was scored based on nuclear diameter (grade 1 3) and mitotic count (grade 1 3) in the most poorly differentiated component of the tumor as previously described by Kadota et al. 18 In short, nuclear grade was determined by the nuclear diameter of the tumor cell relative to small reticular lymphocyte nuclei. The mitotic count was measured by the number of mitoses per 10 high power fields (40 objective). The overall nuclear grade was determined by combining both nuclear and mitotic grades (Table 1). The presence of mucin was classified as (1) intracellular mucin, (2) extracellular mucin, (3) signet ring cell structure, (4) colloid carcinoma, and (5) mixed type (which represented a combination of these patterns) (Fig.1K N). Each mucin characteristic was recorded and the proportion recorded in percentages (5% increments) relative to the mucinous part of the tumor. In cases where tumor had invaded the lymphatics and/ or the blood vessels, these were identified and recorded as vascular invasion, with a differentiation between intratumoral and intraparenchymal vessel invasion. Other histologic factors such as aerobic spread and necrosis, which have previously been reported as potential prognostic factors, 9,10 were also investigated and recorded in binary form. Clinical data regarding date of diagnosis, date of death, and date of last follow-up were obtained from an existing patient database. We compared patient data with the various tumor characteristics, to investigate for correlation. Statistical Analysis Actuarial survival was calculated using Kaplan Meier methods, and comparisons were performed using the log-rank test. Cox proportional hazards regression was used to ascertain the covariates associated with survival while adjusting for age and tumor stage. Statistical analyses were performed using R (R core development team, Vienna, Austria) and Stata 9.2 (StataCorp, College Station, TX) software. RESULTS Clinical Characteristics As mentioned before, 15 of the 238 patients were excluded after initial review, leaving 223 patients for analysis (Table 2). Of these, there was virtually equal spread of men and women, with 109 male patients (51.1%) and 114 female patients (48.9%). The majority of this study population were white (154). There were 34 current smokers, 88 exsmokers, and 21 who had never smoked. In 80 cases, smoking history was unknown. The follow-up period ranged from 0.01 years (4 days) to 10.2 years (123 months), with a median of 3.5 years. The mean age of our patients was 64.2 years at date of diagnosis, with a range of 29 to 82 years, and a median of Patient age was a significant factor in determining patient survivability (p = 0.020). There were 138 right-sided and 84 leftsided tumors. Tumor Node Metastasis Stage The stage of the disease was a highly significant predictor of survival (p < 0.001). The mean maximum tumor diameter was 37 mm (range, mm), and according to tumor node metastasis staging (7th edition), T stage ranged from T1A to T3. The highest proportion of patients had T1B tumors (40.0%). Thirty-three patients had N1 disease and 35 were N2. As expected, T and N stage were both found to be highly significant prognostic factors with a p value of and p value less than (corrected for age), which serves to validate the data set. As such strong associations were found, all other parameters were corrected for age in the T and N stage. Histopathological Scoring After histologically reviewing each tumor, we found that the majority of patients (48%) had acinar predominant tumors (Table 2), which is in keeping with previous studies. This was followed by lepidic predominant (24.2%), solid predominant (17.9%), papillary predominant (5.8%), and micropapillary predominant (4.0%) subtypes. Each predominant subtype had an equal proportion of male and female patients. Regression analysis demonstrated that of the five histologic subtypes, lepidic (p = 0.023) and solid (p = 0.03) predominant subtypes had significant positive and negative prognostic implications, respectively (Table 3, Fig. 2). The presence of a micropapillary predominant pattern was not prognostically significant but this group had the highest proportion of nodal disease (66.7% being either N1 or N2). With regard to nuclear grading, nuclear diameter per se did not prove to be a useful marker for patient survival (p = 0.11). The mitotic rate, however, was found to be highly significant (p = 0.008) (Table 3, Fig. 3). With respect to mitotic rate, lepidic and papillary predominant subtypes had the highest proportion of the lowest rate (grade 1) with 79.6% and 69.2%, respectively. Solid predominant subtypes, however, had the highest proportion of tumors with highest mitotic rate (75% with grade 3) (Table 4). The overall nuclear grade combined the nuclear and mitotic rate scores and was divided into low (in aggregate 2 4) and high (aggregate 5 6) risk categories (Table 1), and this resulted in highly significant risk 38 Copyright 2012 by the International Association for the Study of Lung Cancer

3 Journal of Thoracic Oncology Volume 8, Number 1, January 2013 Histologic Grading of Pulmonary Adenocarcinoma FIGURE 1. Patterns of histologic subtypes and mucin distribution. A, Lepidic: proliferation of neoplastic cells along the preexisting alveolar wall. B, High-power (40 ) view of lepidic type adenocarcinoma (ADC), low nuclear grade. C, High-power (40 ) view of lepidic type ADC, high nuclear grade. D, Papillary: papillary structures with fibrovascular cores lined by malignant cuboidal to columnar tumor cells. Unlike in lepidic carcinoma, basic alveolar structure is altered. E, High-power (40 ) view of papillary type ADC, low nuclear grade. F, High-power (40 ) view of papillary type ADC, high nuclear grade. G, Equivocal pattern: architectural features of both lepidic and papillary ADC are present. H, Acinar: oval-to-round-shaped malignant glands with cuboidal or columnar cells that invade the stroma. I, Solid: solid, invasive sheets of tumor cells. J, Micropapillary: tufts of small clusters of glandular cells, lacking stromal cores. K, Intracellular mucin: cytomorphological features of mucin production (tall columnar cells). L, Extracellular mucin: mucin pooling in surrounding alveolar spaces. M, Signet ring cells: atypical round tumor cells with large amounts of intracytoplasmic mucin and peripheral nuclei. N, Colloid carcinoma: abundant pools of extracellular mucin that distend airspaces, containing clusters of mucin-secreting tumor cells. Copyright 2012 by the International Association for the Study of Lung Cancer 39

4 von der Thüsen et al. Journal of Thoracic Oncology Volume 8, Number 1, January 2013 stratification, with high-risk patients having a dramatically poorer prognosis than low-risk patients ( p < 0.001, Fig. 4). This proved to generate more divergent curves than mitotic rate alone (Fig. 3), and may therefore, prove to be a more useable prognostic parameter. TABLE 1. Nuclear and Mitotic Grade Risk Stratification Nuclear Features Category Grade Definition Nuclear diameter Mitotic count Overall grade (ND+MC) Small 1 ND 3 small lymphocytes Intermediate 2 3 small lymphocytes<nd 5 small lymphocytes Large 3 ND>5 small lymphocytes Low mitosis/10hpf Intermediate mitoses/10hpf High 3 5 mitoses/10hpf Low risk 2 4 High risk 5 6 Grade 1: Nuclear diameter and mitotic rate: Nuclear diameter is < 2 3 timessrl nuclei. Low mitotic rate of 0 1 mitosis/10 HPF. Grade 2: Nuclear diameter between 3 5 times SRL nuclei. Intermediate mitotic rate of 2 4 mitoses/10hpf. Grade 3: Nuclear diameter >5 times SRL nuclei. High mitotic rate of >5 mitoses/10hpf. Tumor cell undergoing mitosis is observed in the center of the field. SRL, small reticular lymphocyte; HPF, high-power field; ND, nuclear diameter; MC, mitotic count. Vessel invasion within tumor was close to being a statistically significant negative prognostic factor in our study (p = 0.067). There was also a difference in the rate of vessel invasion between tumor types, as 80% of solid predominant tumors have vessel invasion within tumor, whereas lepidic predominant only has 16.7% of vessel invasion in tumor (Table 5). There was minimal vessel invasion in the surrounding lung parenchyma, which only manifested in acinar, solid, and micropapillary subtype patterns, and vessel invasion in the surrounding lung was not a significant prognostic factor (p = 0.70). Only 13 tumors (5.8%) of the study population contained signet ring cells. As regression analysis showed that the presence of mucin did not alter patient survival (p < 0.98), the significance of signet ring morphology in patient survival was not tested. Necrosis was most prominent in solid predominant tumors, as 80% of the tumors contained a degree of necrosis. However, necrosis was not found to be a significant prognostic factor (p = 0.16). Aerobic spread (p = 0.16) was also not significant. DISCUSSION This study shows that the predominant patterns and mitotic rate in resected ADCs are pathologic factors that show prognostic significance, being easily recordable parameters that could potentially be used to construct a grading system for use alongside staging. TABLE 2. Patient Data Related to Predominant Subtype Lepidic Papillary Acinar Solid MP Number (%) 54 (24.2) 13 (5.8) 107 (48.0) 40 (17.9) 9 (4.0) Pure subtype (%) 8 (14.8) 1 (7.7) 9 (8.4) 3 (7.5) 2 (22.2) Sex (M/F) 29/25 7/6 52/55 21/19 5/4 Current Smoking Ex Never N/S Race W/O 35/4 10/2 71/5 32/7 5/3 N/S Follow-up in years (average) (3.86) (5.95) (3.58) (3.51) (3.90) N/S A B A Tumor stage 2B A B Unknown N0 46 (85.1) 9 (69.2) 72 (67.3) 25 (62.5) 3 (33.3) N1 1 (2.2) 3 (23.1) 20 (18.7) 5 (12.5) 2 (22.0) Node stage (%) N2 5 (9.0) 1 (8.0) 15 (14.0) 10 (25.0) 4 (44.0) Nx 2 (3.7) N1+2 6 (11.1) 4 (30.8) 35 (32.7) 15 (37.5) 6 (66.7) MP, micropapillary; M/F, male/female; W/O, white/others; N/S, not stated; FUY, follow-up in years. 40 Copyright 2012 by the International Association for the Study of Lung Cancer

5 Journal of Thoracic Oncology Volume 8, Number 1, January 2013 Histologic Grading of Pulmonary Adenocarcinoma With regard to the importance of predominant histologic pattern, much of the published literature has shown prognostic significance only in stage I patients. 9,10 Our study population comprised patients with a range of T stages, from T1A to T3, with clinical stage ranging from IA to IIIB, confirming that a predominant histologic pattern has prognostic TABLE 3. Impact of Potential Prognostic Factors on Patient Survival in Adenocarcinomas of the Lung by Univariate Analysis Variable Hazard Ratio p Histologic subtype Lepidic Papillary Acinar Solid Micropapillary Additional pathological factors Mucin Necrosis Vessel invasion: Tumor Lung 0.70 Nuclear grade: Nuclear diameter Mitotic rate Nuclear diameter + Mitotic rate 1.47 <0.001 Scar size Aerobic spread relevance in more advanced operable disease in a predominantly British population. This is inkeeping with a recent study from Australia, showing similar results in stage I III disease, 19 indicating that there is consistent prognostic significance in different global populations. However, our study did not show prognostic significance for a predominant micropapillary pattern, viewed to be a more aggressive subtype of ADC, 8 although this may be because of the small numbers as their survival data are consistently worse in our cohort (Fig. 2). Furthermore, there was a high rate of vascular and nodal involvement in those patients with micropapillary-predominant ADC, inkeeping with the published literature on early-stage disease 20 22, and so our data are supportive of this pattern being representative of higher-grade disease, if not being prognostically significant herein. This may be because of the relatively late stage of presentation in comparison with other studies, which could be presumed to negate some of the effect of early metastasis in the micropapillary pattern. Mitotic rate was the only other pathological factor that was significant in predicting patient survival. As mitotic rate of a tumor governs the rate of tumor proliferation, 23 higher mitotic rates are likely to have a faster and larger increase in volume and mass per unit time and are hence likely associated with higher aggression and lower prognosis. Higher rates of mitosis also increase the chances of mitotic errors and mutations and likely result in more poorly differentiated tumors. Using the mitotic rate as a prognostic marker has the advantage of easy introduction into routine histopathologic practice, as it does not require specialist skills or equipment. Also, scoring the number of mitotic Fs has previously been shown to have an adequate interobserver agreement in other carcinoma types, such as the grading of breast carcinoma. 24 FIGURE 2. Kaplan Meier survival plot of predominant histologic adenocarcinoma subtypes (*p < 0.05). Copyright 2012 by the International Association for the Study of Lung Cancer 41

6 von der Thüsen et al. Journal of Thoracic Oncology Volume 8, Number 1, January 2013 FIGURE 3. Kaplan Meier survival plot of different mitotic rate categories (1=0 1 mitoses/10 HPF; 2=2 4 mitoses/10 HPF; 3= 5 mitoses/10 HPF). HPF, high-power field. TABLE 4. Nuclear Characteristics for Predominant Adenocarcinoma Subtypes Grade Lepidic Papillary Acinar Solid MP Nuclear size (%) 1 17 (31.5) 4 (30.8) 4 (3.7) 0 (0) 1 (11.1) 2 33 (61.1) 7 (53.8) 49 (45.8) 9 (22.5) 4 (44.4) 3 4 (7.4) 2 (15.4) 54 (50.5) 31 (77.5) 4 (44.4) Mitotic rate (%) 1 43 (79.6) 9 (69.2) 46 (43.0) 4 (10) 4 (44.4) 2 5 (9.3) 3 (23.1) 21 (19.6) 6 (15) 2 (22.2) 3 6 (11.1) 1 (7.7) 40 (37.4) 30 (75) 3 (33.3) MP, micropapillary. TABLE 5. Histologic Parameters Related to Predominant Subtype Lepidic Papillary Acinar Solid MP Vascular invasion in tumor (%) 9 (16.7) 8 (62.0) 77 (72.0) 32 (80) 6 (66.7) Vascular invasion in lung (%) (4.7) 7 (17.5) 1 (11.1) Aerobic spread (%) 18 (33.3) 3 (23.1) 19 (17.8) 8 (20.0) 5 (55.6) Scar size(average) 0 25 (9.64) 0 98 (22.4) 0 85 (26.2) (26.2) 0 58 (10.1) Papillary subtype present (%) 17 (31.5) 13 (100) 30 (28.0) 1 (2.5) 2 (22.2) Necrosis (%) 6 (11.1) 3 (23.1) 56 (52.3) 33 (82.5) 5 (55.6) Cribriform pattern (%) (13.1) 4 (10) 0 Signet ring structure (%) 5 (9.3) 0 5 (4.7) 3 (7.5) 0 MP, micropapillary. Nuclear diameter on its own was not found to be a significant part of a nuclear grading system in this study, which contradicts studies which did find a significant correlation for cytonuclear structure, 25,26 but concurs with a study by Kadota et al. 18 in which the authors graded all aspects of tumor nuclear characteristics, including nuclear diameter, but found only mitotic count to provide useful prognostic information. Nonetheless, combining nuclear diameter and mitotic rate into 42 Copyright 2012 by the International Association for the Study of Lung Cancer

7 Journal of Thoracic Oncology Volume 8, Number 1, January 2013 Histologic Grading of Pulmonary Adenocarcinoma FIGURE 4. Kaplan Meier survival plot of combined nuclear and mitotic rate score. a single scoring system with classification into high- and lowrisk categories, would seem to yield a highly significant and divergent prognostic parameter. Of the remaining histologic parameters that were scored, only tumor vessel invasion approached prognostic significance (p < 0.067), and this has been previously described as an independent prognostic factor, 10,17 including a recent publication in this journal. 27 Most published data, however, are on stage I patients and so the later presentation in our cohort may, in part, explain the lack of prognostic significance in this study. We did find that compared with 66.7% of micropapillary predominant tumors, only 16.7% of lepidic predominant tumors had tumor vessel invasion, suggesting that the prognostic significance seen may be, in part, because of the propensity for vascular spread, and that micropapillary tumors are more likely to metastasize because of reduced cell adhesion and vascular invasion. 22 Vessel invasion in the lung adjacent to the tumor was not prognostically significant, possibly because of small sample size (n = 13), but it is worth noting that this only occurred in acinar, solid, and micropapillary predominant subtypes. None of the other histologic factors (presence of mucin, aerobic spread, nuclear grade, and necrosis) carried prognostic significance. Subtypes, such as signet ring variant of ADC, have been reported as having a poor prognosis, 23,28,29 but our numbers were probably too small to assess significance of this particular cytological feature. There are several limitations to this study which should be mentioned. First, in the process of selecting our population study, a number of patients were excluded because of the lack of follow-up. Furthermore, data were retrospective, although all histologic materials were collected in a uniform fashion in one institution, using the minimum data set of the Royal College of Pathologists with regard to representative tumor blocks. As both factors may be a source of bias, verification of the findings in a prospective study is desirable. For example, despite adhering to the minimum data set, the bias, thus introduced, precludes quantification of the total area of necrosis. Although blocks will include areas of necrosis, these would, by the nature of macroscopic sampling, preferentially be from viable tissue to allow for adequate microscopic assessment, with complete blocking of tumors only undertaken for small tumors. This would considerably skew the results against the microscopic quantification of necrosis in larger tumors. We feel it would, therefore, be inappropriate to assess this in any greater detail than present or absent. It is also worth recognizing that histologic grading of ADCs remains a relatively subjective exercise for which, even in a subspecialized setting of thoracic pathologists, there may be considerable interobserver variability. Thus, the morphologic distinction of papillary carcinoma pattern from acinar, lepidic, and micropapillary patterns is anecdotally difficult, and this may have led to a difference between other published data and ours. Thunnissen et al. 30 have investigated this matter further in a recent interobserver study of the reproducibility of histopathological subtypes and invasion in pulmonary ADC. In this study, interobserver variation was at its greatest when distinguishing lepidic from papillary/acinar patterns, in the context of how pathologists report lepidic pattern (i.e., constituted crosscutting of lepidic [pseudo-papillary/acinar] versus true invasion). An example of such a case is provided in Figure 1G. We feel that this probably explains the anomalous nature of our results in relation to the prognosis of papillary Copyright 2012 by the International Association for the Study of Lung Cancer 43

8 von der Thüsen et al. Journal of Thoracic Oncology Volume 8, Number 1, January 2013 ADCs, and we intend to investigate this further in a separate follow-up study in due course. In conclusion, our study shows that the predominant histologic pattern and mitotic rate are significant prognostic factors in resected lung ADCs, with a trend toward significance for vascular invasion. On the basis of these results, one could suggest establishing a risk stratification system based on histologic subtype and nuclear grade, as previously alluded to by Kadota et al. 18 ACKNOWLEDGMENTS This project was funded and supported by the National Institute for Health Research Respiratory Disease Biomedical Research Unit, at the Royal Brompton, and Harefield NHS Foundation Trust, and Imperial College London. REFERENCES 1. Jemal A, Siegel R, Ward E, et al. Cancer statistics, CA Cancer J Clin 2008;58: Alberg AJ, Ford JG, Samet JM; American College of Chest Physicians. Epidemiology of lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007;132(3 Suppl):29S 55S. 3. Goldstraw P, Ball D, Jett JR, et al. Non-small-cell lung cancer. Lancet 2011;378: American Cancer Society. Global Cancer Facts & Figures (2nd Ed). Atlanta, GA:American Cancer Society Travis WD, Brambilla E, Müller-Hermelink HK, et al. Pathology and Genetics: Tumors of the Lung, Pleura, Thymus and Heart. Lyon, France:IARC, Motoi N, Szoke J, Riely GJ, et al. Lung adenocarcinoma: modification of the 2004 WHO mixed subtype to include the major histologic subtype suggests correlations between papillary and micropapillary adenocarcinoma subtypes, EGFR mutations and gene expression analysis. Am J Surg Pathol 2008;32: Miyoshi T, Satoh Y, Okumura S, et al. Early-stage lung adenocarcinomas with a micropapillary pattern, a distinct pathologic marker for a significantly poor prognosis. Am J Surg Pathol 2003;27: Travis WD, Brambilla E, Noguchi M, 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 2011;6: Sica G, Yoshizawa A, Sima CS, et al. A grading system of lung adenocarcinomas based on histologic pattern is predictive of disease recurrence in stage I tumors. Am J Surg Pathol 2010;34: Yoshizawa A, Motoi N, Riely GJ, 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 2011;24: Nakazato Y, Minami Y, Kobayashi H, et al. Nuclear grading of primary pulmonary adenocarcinomas: correlation between nuclear size and prognosis. Cancer 2010;116: Petersen I, Kotb WF, Friedrich KH, Schlüns K, Böcking A, Dietel M. Core classification of lung cancer: correlating nuclear size and mitoses with ploidy and clinicopathological parameters. Lung Cancer 2009;65: Pechet TT, Carr SR, Collins JE, Cohn HE, Farber JL. Arterial invasion predicts early mortality in stage I non-small cell lung cancer. Ann Thorac Surg 2004;78: Miyoshi K, Moriyama S, Kunitomo T, Nawa S. Prognostic impact of intratumoral vessel invasion in completely resected pathologic stage I non-small cell lung cancer. J Thorac Cardiovasc Surg 2009;137: Funai K, Sugimura H, Morita T, Shundo Y, Shimizu K, Shiiya N. Lymphatic vessel invasion is a significant prognostic indicator in stage IA lung adenocarcinoma. Ann Surg Oncol 2011;18: Poleri C, Morero JL, Nieva B, et al. Risk of recurrence in patients with surgically resected stage I non-small cell lung carcinoma: histopathologic and immunohistochemical analysis. Chest 2003;123: Yamaguchi Y, Ishii G, Kojima M, et al. Histopathologic features of the tumor budding in adenocarcinoma of the lung: tumor budding as an index to predict the potential aggressiveness. J Thorac Oncol 2010;5: Kadota K, Suzuki K, Colovos C, et al. A nuclear grading system is a strong predictor of survival in epitheloid diffuse malignant pleural mesothelioma. Mod Pathol 2012;25: Russell PA, Wainer Z, Wright GM, Daniels M, Conron M, Williams RA. 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 2011;6: Maeda R, Isowa N, Onuma H, et al. Lung adenocarcinomas with micropapillary components. Gen Thorac Cardiovasc Surg 2009;57: Miyoshi T, Satoh Y, Okumura S, et al. Early-stage lung adenocarcinomas with a micropapillary pattern, a distinct pathologic marker for a significantly poor prognosis. Am J Surg Pathol. 2003;27: Miyoshi T, Shirakusa T, Ishikawa Y, et al. Possible mechanism of metastasis in lung adenocarcinomas with a micropapillary pattern. Pathol Int 2005;55: Tsuta K, Ishii G, Yoh K, et al. Primary lung carcinoma with signet-ring cell carcinoma components: clinicopathological analysis of 39 cases. Am J Surg Pathol 2004;28: Lehr HA, Hansen DA, Kussick S, et al. Assessment of proliferative activity in breast cancer: MIB-1 immunohistochemistry versus mitotic figure count. Hum Pathol 1999;30: Kobayashi Y, Yokose T, Kawamura K, et al. Cytologic factors associated with prognosis in patients with peripheral adenocarcinoma of the lung measuring 3 cm or less in greatest dimension. Cancer 2005;105: Minami Y, Matsuno Y, Iijima T, et al. Prognostication of small-sized primary pulmonary adenocarcinomas by histopathological and karyometric analysis. Lung Cancer 2005;48: Shimada Y, Saji H, Yoshida K, et al. Pathological vascular invasion and tumor differentiation predict cancer recurrence in stage IA non-smallcell lung cancer after complete surgical resection. J Thorac Oncol 2012;7: Castro CY, Moran CA, Flieder DG, Suster S. Primary signet ring cell adenocarcinomas of the lung: a clinicopathological study of 15 cases. Histopathology 2001;39: Ou SH, Ziogas A, Zell JA. Primary signet-ring carcinoma (SRC) of the lung: a population-based epidemiologic study of 262 cases with comparison to adenocarcinoma of the lung. J Thorac Oncol 2010;5: Thunnissen E, Beasley MB, Borczuk AC, et al. Reproducibility of histopathological subtypes and invasion in pulmonary adenocarcinoma. An international interobserver study. Mod Pathol July 20, doi: / modpathol [Epub ahead of print]. 44 Copyright 2012 by the International Association for the Study of Lung Cancer