- Email: [email protected]
Current Readings: Pathology, Prognosis, and Lung Cancer
CURRENT READINGS
Current Readings: Pathology, Prognosis, and Lung Cancer James H. Suh, MD The 2011 International Association for the Study of Lung Cancer/American Thoracic Society/ European Respiratory Society international multidisciplinary classification of lung adenocarcinoma introduced the new categories of adenocarcinoma in situ, minimally invasive adenocarcinoma, and invasive mucinous adenocarcinoma, and replaced the category of mixed subtype adenocarcinoma with lepidic, acinar, papillary, micropapillary, and solid predominant adenocarcinoma. The aim of this manuscript is to evaluate whether the new classification can be applied successfully in determining prognosis of surgically resected patients. Six consecutive clinicopathologic studies using the new classification that were published between spring 2011 and fall 2012 were reviewed. Overall, they demonstrated excellent outcome for adenocarcinoma in situ and minimally invasive adenocarcinoma; intermediate outcome for lepidic, acinar, and papillary predominant adenocarcinoma; and poor outcome for solid and micropapillary predominant adenocarcinoma and invasive mucinous adenocarcinoma. As the new classification remains a proposal at this time, it is hoped that thoracic surgeons will play a leading role in its worldwide dissemination for clinical care and research. Semin Thoracic Surg 25:14-21 © 2013 Elsevier Inc. All rights reserved. Keywords: lung adenocarcinoma, pathology, prognosis, thoracic surgery In February 2011, Travis et al1 published a new proposal, the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society (IASLC/ATS/ERS) international multidisciplinary classification of lung adenocarcinoma. For the first time, the process included representatives from all medical specialties that participate in the diagnosis and management of lung cancer patients, not just pathologists. This proposal is intended to replace relevant sections within the 2004 World Health Organization (WHO) classification.2 In particular, it eliminates the categories of bronchioloalveolar carcinoma (BAC) and mixed subtype adenocarcinoma. BAC has been used in very different contexts by clinicians and pathologists, and mixed subtype adenocarcinoma has evolved into a wastebasket diagnosis applied to patients with widely disparate outcomes. Therefore, the primary objective of the new classification is to achieve precise correlations between predominant histologic
Department of Pathology, NYU Langone Medical Center, New York, New York. Dr. Suh reports receiving lecture and consulting fees from Pfizer. Address reprint requests to James H. Suh, MD, Department of Pathology, NYU Langone Medical Center, Tisch HW-467, 560 First Avenue, New York, NY 10016. E-mail: [email protected]
14
subtypes of lung adenocarcinoma and survival in surgically resected patients. Since early 2011, there have been 6 retrospective clinicopathologic studies using the new classification, including almost 2000 operated patients from 4 continents during the past 2 decades.3-8 This review will summarize the key points from each article, examine their findings regarding pathology and prognosis, and discuss potential implications for thoracic surgeons. 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 Yoshizawa A, Motoi N, Riely, GJ, et al. Mod Pathol 24:653-664, 2011 Yoshizawa et al3 conducted the first major study on the new classification (Table 1) to evaluate its utility in determining prognosis and implications for staging. In a retrospective review, the authors selected 514 stage I cases at the Memorial Sloan-Kettering Cancer Center (1995-2005) using UICC/AJCC seventh edition criteria, the majority of whom underwent lobectomy (83%) and mediastinal lymph node dissection 1043-0679/$-see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1053/j.semtcvs.2013.01.003
PATHOLOGY, PROGNOSIS, AND LUNG CANCER Table 1. IASLC/ATS/ERS international multidisciplinary classification of lung adenocarcinoma Preinvasive lesions Atypical adenomatous hyperplasia Adenocarcinoma in situ (ⱕ3 cm, formerly BAC) Nonmucinous Mucinous Mixed mucinous nonmucinous Minimally invasive adenocarcinoma (ⱕ3 cm lepidic predominant tumor with ⱕ5-mm invasion) Nonmucinous Mucinous Mixed mucinous nonmucinous Invasive adenocarcinoma Lepidic predominant (formerly nonmucinous BAC pattern, with ⬎5 mm invasion) Acinar predominant Papillary predominant Micropapillary predominant Solid predominant with mucin production Variants of invasive adenocarcinoma Invasive mucinous adenocarcinoma (formerly mucinous BAC) Colloid Fetal (low and high grade) Enteric Reprinted with permission Travis et al.1
and did not receive adjuvant therapy (94%). The median age was 69 years, 63% were female, and the median tumor size was 2.0 cm. An average of 8.6 slides (range: 1-31) per tumor were examined by 1-3 pathologists, depending on difficulty. Mucin stains were evaluated, if needed. Standard pathologic pa-
rameters such as 2004 WHO grade, necrosis, lymphovascular invasion, and visceral pleural invasion were included. In addition, invasive tumor size was calculated, either by ruler or by multiplication of percent invasion and total tumor size. Finally, overall and disease-free survivals were recorded. The median follow-up period was 48 months. Fiftynine percent had no evidence of disease, whereas 14% of patients died of disease, and 21% died of other causes. According to the 2004 WHO classification, 95% were classified as mixed subtype adenocarcinomas. In contrast, application of the new classification identified 3 prognostic subgroups: (a) 0.2% (of total cases) adenocarcinoma in situ (AIS) and 1.4% minimally invasive adenocarcinoma (MIA) had excellent outcomes (100% 5-year disease-free survivals); (b) 5.6% lepidic predominant adenocarcinoma (LPA; 90%), 45.1% acinar predominant adenocarcinoma (ACA; 84%), and 27.8% papillary predominant adenocarcinoma (PPA; 83%) had intermediate outcomes; and (c) 2.5% invasive mucinous adenocarcinoma (76%), 1.8% colloid predominant adenocarcinoma (71%), 13.0% solid predominant adenocarcinoma (SPA; 70%), and 2.3% micropapillary predominant adenocarcinoma (MPA; 67%) had the worst outcomes (Table 2, Fig. 1). Multivariate analysis also identified male gender, invasive tumor size, and necrosis as poor prognostic factors. Compared with total tumor size, invasive tumor size would decrease T-factor scoring in 14% of patients, mostly among MIA, LPA, and invasive mucinous adenocarcinoma, and correlate better with disease-free survival. The findings in this study validate the prognostic impact of the new classification and may be helpful in identifying candidates for adjuvant therapy.
Table 2. Five-year disease-free survival by IASLC subtype and prognostic subgroup IASLC/ATS/ERS classification subtypes
Number (%)
Low grade Adenocarcinoma in situ 1 (0.2%) Minimally invasive adenocarcinoma, nonmucinous 7 (1%) Minimally invasive adenocarcinoma, mixed 1 (0.2%) mucinous and nonmucinous Intermediate grade Lepidic predominant 29 (6%) Acinar predominant 232 (45%) Papillary predominant 143 (28%) High grade Micropapillary predominant 12 (2%) Solid predominant 67 (13%) Colloid predominant 9 (2%) Invasive mucinous adenocarcinoma, mixed mucinous/nonmucinous 13 (3%)
Disease-free survival at 5 years 100% 100% 100%
90% 84% 83% 67% 70% 71% 76%
Reprinted with permission Yoshizawa et al.3
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
15
PATHOLOGY, PROGNOSIS, AND LUNG CANCER
Figure 1. Five-year disease-free survival by IASLC subtype (A) and prognostic subgroup (B). Reprinted with permission Yoshizawa et al.3
DOES LUNG ADENOCARCINOMA SUBTYPE PREDICT PATIENT SURVIVAL? A CLINICOPATHOLOGIC STUDY BASED ON THE NEW IASLC/ATS/ERS INTERNATIONAL MULTIDISCIPLINARY LUNG ADENOCARCINOMA CLASSIFICATION Russell PA, Wainer Z, Wright GM, et al. J Thorac Oncol 6:1426-1504, 2011 Russell et al4 included all patients with lung adenocarcinoma who underwent resection with curative intent. They identified 210 stage I, II, and III patients at the St. Vincent’s Hospital of the University of Melbourne, Australia (1996-2009), using UICC/AJCC seventh edition criteria. Lobectomy was performed in 79% and mediastinal lymph node dissection was performed in 86% of cases in Russell et al. None received neoadjuvant chemotherapy. The median age was 67 years, 46% were female, and the median tumor size was 2.9 cm. An average of 5 slides (range: 1-20) per tumor were reviewed by 2 pathologists. Elastic Van Gieson stains were evaluated in all cases with a lepidic component to assess for invasion, and periodic acid-Schiff stains were evaluated in all cases with a solid component to assess for mucin. The presence or absence of lymphovascular invasion, perineural invasion, visceral pleural invasion, and lymph node metastases was recorded. Prognostic correlations between IASLC diagnosis and 5-year overall survival were explored. The median follow-up period was 49 months. As in the first study, 93% were classified as mixed subtype adenocarcinomas in the 2004 WHO classification. Using the new classification, the authors identified 4 prog16
nostic subgroups: (a) 0.5% AIS (100% 5-year survival), 3.3% MIA (100%), and 4.8% LPA (86%) had the best outcomes (overall 93%); (b) 12.4% PPA (71%) and 40.0% ACA (68%) had intermediate-good outcomes; (c) 4.8% invasive mucinous adenocarcinoma (51%) and 4.3% colloid predominant adenocarcinoma (51%) had intermediate-poor outcomes; and (d) 23.3% SPA (39%) and 6.7% MPA (38%) had the worst outcomes. They discovered that the correlations between IASLC diagnosis and prognosis persisted even after controlling for stage. Therefore, the findings in this study suggest that the new classification can be useful in determining prognosis of stage II and IIIA lung adenocarcinoma patients as well. ADENOCARCINOMAS WITH PROMINENT LEPIDIC SPREAD: RETROSPECTIVE REVIEW APPLYING NEW CLASSIFICATION OF THE ATS Xu L, Tavora F, Battafarano R, et al. Am J Surg Pathol 36:273-282, 2012 In contrast to the first 2 studies, Xu et al5 focused primarily on AIS, MIA, and LPA. They selected 87 mostly early-stage patients at the University of Maryland Medical Center (2007-2010) with surgically resected lung adenocarcinomas. Of the selected cases in Xu’s study, 30 had previous pathologic diagnoses of BAC or mixed subtype adenocarcinoma with BAC features and were included for further analysis. The type of surgical procedure, method of mediastinal lymph node assessment, and administration of adjuvant therapy were
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
PATHOLOGY, PROGNOSIS, AND LUNG CANCER not recorded. The median age was 69 years, 39% were female, and the mean tumor size was 2.3 cm. An unknown number of slides per tumor were reviewed by 2 pathologists. Elastic Van Gieson stains were evaluated, if already available. The presence or absence of lymph node metastases was recorded. Invasive tumor size was measured for all cases. Correlations between IASLC diagnosis and short-term outcome were explored. The mean follow-up periods ranged from 17 to 28 months. Eight of 87 tumors were reclassified from 2004 WHO mixed subtype adenocarcinoma to MIA. All 3 patients with AIS (3.4% of total cases) had no evidence of disease (28-month mean follow-up), and 7 of 8 patients with MIA (9.2%) also had no evidence of disease (22-month mean follow-up). In contrast, only 9 of 19 patients with LPA (21.8%) had no evidence of disease (17-22-month mean follow-up). The presence of a stromal desmoplastic response in areas of invasion conferred an increased likelihood of local recurrence, lymph node metastasis, and distant spread. Outcomes for other subtypes of adenocarcinoma were not reported. Finally, 2 of 8 cases of MIA were originally scored as T2 and T3 based on total tumor size, whereas invasive tumor size was significantly smaller. Although this study is limited by the focus on AIS, MIA, and LPA, it provides further evidence that both AIS and MIA should have an excellent prognosis. THE NOVEL HISTOLOGIC IASLC/ATS/ERS CLASSIFICATION SYSTEM OF LUNG ADENOCARCINOMA IS A STAGEINDEPENDENT PREDICTOR OF SURVIVAL Warth A, Muley T, Meister M, et al. J Clin Oncol 30: 1438-1446, 2012 Warth et al6 used the new classification to study patients with all stages of surgically resected adenocarcinoma— even stage IV. Warth and colleagues identified 500 stage I–IV patients at the University Hospital Heidelberg, Germany (2002-2008), using UICC/AJCC seventh edition criteria. The type of surgical procedure and method of mediastinal lymph node assessment were not recorded. None received neoadjuvant therapy, but 26% received adjuvant chemotherapy and 17% received adjuvant radiation therapy. The median age was 62 years and 40% were female. The median tumor size was not recorded. An average of 4.6 slides (range: 1-14) per tumor were reviewed by 2 pathologists. Mucin stains were evaluated for all cases. Other pathologic parameters such as lymphovascular invasion, visceral pleural invasion, and lymph
node status were not included. Correlations between IASLC diagnosis and 5-year overall, disease-specific, and disease-free survivals were explored. The median follow-up period was not reported. Ninety-three percent were classified as mixed subtype adenocarcinomas in the 2004 WHO classification. They did not identify any cases of AIS or MIA. However, application of the new classification produced 3 prognostic subgroups: (a) 8.2% LPA (78.5 months) had the best outcome; (b) 41.4% ACA (67.3 months) had an intermediate outcome; and (c) 36.6% SPA (58.1 months), 4.6% PPA (48.9 months), and 6.6% MPA (44.9 months) had the worst outcomes (overall 57.2 months). Multivariate analysis showed that IASLC diagnosis was a stageand therapy-independent predictor of survival. Invasive tumor size correlated better with outcome than total tumor size. Finally, they discovered that patients with SPAs received significant benefit from adjuvant chemoradiation therapy. The findings in this study support the utility of the new classification in all stages of resected lung adenocarcinoma patients. HISTOLOGIC PATTERNS AND MOLECULAR CHARACTERISTICS OF LUNG ADENOCARCINOMA ASSOCIATED WITH CLINICAL OUTCOME Solis LM, Behrens C, Raso MG, et al. Cancer 118: 2889-2899, 2012 Solis et al7 applied concepts derived from the new classification as well as adjunct studies to examine associations between pathology and prognosis. Solis et al identified 240 stage I–IV patients at the University of Texas M.D. Anderson Cancer Center (1997-2005) with surgically resected solitary lung nodules and first diagnosis of primary lung adenocarcinoma who did not receive neoadjuvant therapy. The type of surgical procedure and method of mediastinal lymph node assessment were not recorded. Thirty percent received adjuvant therapy. The median age was 67 years and 57% were female. The median tumor size was not recorded. One slide per centimeter of tumor was examined by 3 pathologists. In addition to morphology, immunohistochemistry for TTF-1 and a panel of 5 biomarkers called the FILM index was performed on tissue microarrays. The presence or absence of necrosis and lymphovascular invasion was recorded. Correlations between histologic subtype, immunohistochemistry, and 5-year overall and disease-free survivals were explored.
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
17
PATHOLOGY, PROGNOSIS, AND LUNG CANCER
Figure 2. Five-year recurrence-free survival of nonsolid tumors by TTF-1 expression (A) and FILM index score (B). Reprinted with permission Solis et al.7
The median follow-up period was 51 months. Eighty percent were mixed subtype adenocarcinomas according to the 2004 WHO classification. No cases of AIS were identified. Using concepts from the new classification, they identified 4 mutually exclusive histologic groups and 2 prognostic subgroups: (a) 38.3% (of total cases) tumors containing any amount of solid pattern (hazard ratio ⫽ 1.722) had a worse outcome than all nonsolid tumors, including; (b) 14.2% tumors containing any papillary pattern but no solid pattern; (c) 30.0% tumors containing lepidic and acinar patterns but no solid or papillary patterns; and (d) 17.5% pure acinar pattern tumors. The difference in outcome between patients with group a vs patients with groups b, c, and d tumors was seen at all stages. In nonsolid tumors, they found that increased TTF-1 expression and low FILM index score were correlated with better outcome, also regardless of stage (Fig. 2). The findings in this study suggest that the new classification can be used in alternative ways to determine prognosis in all stages of resected lung adenocarcinoma patients and that immunohistochemical studies can provide additional prognostic information beyond morphology in many cases. PROGNOSTIC SIGNIFICANCE OF THE IASLC/ATS/ERS CLASSIFICATION IN CHINESE PATIENTS—A SINGLEINSTITUTION RETROSPECTIVE STUDY OF 292 LUNG ADENOCARCINOMA CASES Gu J, Lu C, Guo J. J Surg Oncol (in press) In the last study, Gu et al8 used the new classification to study an ethnic population that has distinct clinicopathologic and molecular characteristics of lung adenocarcinoma. Gu and colleagues identified 292 stage I, II, and III patients at the Fudan University Hospital in Shanghai, 18
China (2005-2008), using UICC/AJCC seventh edition criteria. Lobectomy was performed in 95% and mediastinal lymph node dissection was performed in 100% of cases. None received neoadjuvant chemotherapy. The median age was 59 years and 52% were female. The median tumor size was not recorded. An average of 3.7 slides (range: 2-12) were reviewed by 2 pathologists. Additional parameters such as tumor location, tumor size, visceral pleural invasion, and results of immunohistochemical stains were recorded. Correlations between IASLC diagnosis and 5-year overall and disease-free survivals were explored. The median follow-up period was 44 months. Forty percent experienced recurrence or metastasis and 23% died. Ninety-one percent were mixed subtype adenocarcinomas in the 2004 WHO classification. Using the new classification, Gu et al identified 3 prognostic subgroups: (a) 0.3% AIS (100% 5-year disease-free survival) and 4.8% MIA (100%) had excellent outcomes; (b) 10.6% LPA (72%), 5.5% less common variants (62%), 12.3% PPA (56%), and 38.4% ACA (54%) had intermediate outcomes; and (c) 17.8% SPA (46%) and 10.3% MPA (26%) had the worst outcomes. In addition, multivariate analysis identified higher stage and absence of EGFR mutation as poor prognostic factors. The latter finding is of interest owing to the increased frequency of EGFR mutations in East Asian patients. Regardless of reported ethnic differences in the molecular genetics of lung adenocarcinoma, the findings in this study demonstrate that the predominantly morphologic approach of the new classification can be used to determine prognosis in Chinese lung adenocarcinoma patients as well. COMMENTARY The 1995 Noguchi classification identified 2 subtypes of lung adenocarcinoma with 100% 5-year survival after complete resection, which corresponded
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
PATHOLOGY, PROGNOSIS, AND LUNG CANCER
Figure 3. Adenocarcinoma in situ, nonmucinous. (A) The tumor is well-circumscribed and consists entirely of lepidic (formerly nonmucinous bronchioloalveolar carcinoma) growth (low-power). (B) The tumor cells line intact alveolar septae without stromal, vascular, or pleural invasion (high-power). Reprinted with permission Yoshizawa et al.3 (Color version of figure is available online.)
mostly to BAC in the 1999/2004 WHO classifications, a term that was first introduced by Liebow in 1960.2,9-11 However, by the early 21st century, BAC referred to both solitary lung nodules as well as widely metastatic disease. Furthermore, both the Noguchi classification (60% type C) and 2004 WHO classification (⬎90% mixed subtype) contained large categories of patients who exhibited sharply divergent outcome, creating the need for a new classification.3 The new classification divides tumors into 3 categories—preinvasive (AIS), minimally invasive (MIA), and overtly invasive adenocarcinomas (Figs. 3 and 4). Overtly invasive adenocarcinomas are classified as LPA, ACA, PPA, MPA, and SPA using comprehensive histologic subtyping. This morphologic tool requires extensive sampling of each tumor to determine the predominant subtype, which can be ⬍50%. In particular, complete sampling is necessary to exclude areas of invasion before a diagnosis of AIS can be made, which is problematic during intraoperative consultation due to inherent time constraints.1 Since the publication of the new classification in February 2011, there have been 6 retrospective studies, including 1843 patients of all stages who underwent surgery from 1995 to 2010 in the United States, Germany, China, and Australia3-8. In the 3 studies for which data are available, lobectomy was performed in 79%-95%, sublobar resection was performed in 0%-16%, and mediastinal lymph node dissection was performed in 86%-100% of cases.3,4,8 Depending on the study, 6%-30% underwent adjuvant therapy. The median follow-up periods ranged from 17 to 51 months, with most having at least 40 months of follow-up. Whereas 80%-95% of the tumors were originally diagnosed as mixed subtype adenocarcinomas, tumors were reclassified accord-
ing to the new classification in most cases. It appears that adequate sampling of tumors was performed in all studies—in one department, there was a change in policy to require submission of the entire tumor, whereas another chose to process a whole cross-section of tumor at its largest diameter.4,6 Therefore, these studies represent a large international cohort of patients who received the standard of care in both surgical and oncological management, allowing a detailed analysis of the relationship between pathology and prognosis using the new classification. Of the 1603 cases that used the exact terminology proposed by the new classification, there were 6 AIS (0.4%) and 36 MIA (2.2%). Although few in number, all patients diagnosed with either AIS or MIA demonstrated excellent outcome. Three of 6 AIS had 100% 5-year survival, and the other 3 cases had no evidence of disease at short-term follow-up. Similarly, 28 of 36 MIA had 100% 5-year survival, and 7 of 8 remaining cases had no evidence of disease at short-term follow-up. Therefore, future prospective studies could be constructed to determine whether surveillance is possible for these patients. Both AIS and MIA often have a ground-glass appearance on HRCT, are usually ⬍3 cm in size, and tend to occur in peripheral locations. If a ground-glass nodule begins to grow rapidly or develop a solid component, then resection should be considered. The appropriate extent of surgery remains controversial at this time, as lobectomy, segmentectomy, and wedge resection can be performed using a video-assisted thoracoscopic surgery or robotic approach. The ongoing CALGB trial, which randomizes patients with tumors ⬍2 cm, will provide further insight. Next, preoperative image-guided placement of a clip can
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
19
PATHOLOGY, PROGNOSIS, AND LUNG CANCER
Figure 4. Minimally invasive adenocarcinoma, nonmucinous. (A) The tumor consists mostly of lepidic growth with a central area of invasion that measures ⬍5 mm in greatest linear dimension (low-power). (B) There is lepidic growth on the left and acinar invasion on the right (medium-power). (C) Small angulated glands invade through a desmoplastic stroma, but there is no lymphovascular/pleural invasion or tumor necrosis (highpower). Reprinted with permission Travis et al.1 (Color version of figure is available online.)
help locate a subtle lesion at the time of surgery. As discussed earlier, the role of frozen section in this setting is ill-defined. Finally, as AIS and MIA have a low propensity for metastasis, a mediastinal lymph node dissection may not be necessary. Results of future randomized prospective trials could help with this decision.1,12 Nevertheless, the vast majority (97.4%) of lung adenocarcinomas in these studies were overtly invasive adenocarcinomas. Approximately 8% were LPA, 40% were ACA, 14% were PPA, 6% were MPA, and 22% were SPA. The remaining 10% consisted of less common variants such as invasive mucinous adenocarcinoma (formerly mucinous BAC), colloid predominant adenocarcinoma, and enteric adenocarcinoma. Remarkably, all 3 studies that applied the new classification strictly and measured 5-year survival showed similar findings regarding prognosis, despite the inclusion of patients with stage II and III disease in 2 of them.3,4,8 LPA was associated with good out20
come (72%-90%), ACA (54%-84%) and PPA (56%83%) with intermediate outcome, and MPA (26%67%) and SPA (39%-70%) with poor outcome. In addition to separating invasive adenocarcinomas into prognostically relevant categories, the new classification can help distinguish whether multiple tumors represent separate primaries or metastases in many cases. Therefore, dependingon functional capacity, some patients with multiple lung nodules have surgical options, including wedge resection for peripheral less invasive tumors, segmentectomy or lobectomy for central invasive tumors, and other combinations such as multiple wide wedge resections, lobectomy with wide wedge resection(s), bilobectomy, and pneumonectomy.1,12 The new classification also impacts staging. As areas of lepidic growth appear morphologically identical to AIS, it may be possible to subtract such areas before measurement of tumor size. For example, Yoshizawa et al3 found that using invasive tumor size decreased T-
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
PATHOLOGY, PROGNOSIS, AND LUNG CANCER factor scoring in 14% of patients. Similarly, Xu et al5 found that 2 of 8 MIAs had originally been scored as T2 or T3 by total tumor size. Based on this and similar data, introduction of Tis and Tmi categories could lead to more accurate staging in the future. In addition, 3 studies suggest that pathologic diagnosis in the new classification might be a stage-independent predictor of outcome.4,6,8 Although it is encouraging that prognostic correlations persist in patients presenting at all stages of disease, it is likely that some of the differences in survival are due to varying risks of developing lymph node involvement and distant metastases. Regardless of potential impacts on staging, it appears that the new classification identifies candidates for adjuvant therapy, which is critical because up to 40% of stage I patients will recur or die within 5 years.3 In particular, Warth et al6 discovered that patients with SPA received significant benefit from adjuvant chemoradiation therapy. Finally, it is important to discuss limitations of the new classification. Comprehensive histologic subtyping is a complex method and still relies on the ability of pathologists to distinguish between in situ and invasive growth, which can be difficult in the absence of desmoplastic stromal reaction.5 Furthermore, critics of the new classification point out that ⬍5% of tumors in most series are AIS or MIA, whereas the vast majority confers an intermediate prognosis, which does not represent much improvement from previous classifications. For these patients, immunohistochemical and molecular studies may be useful in further refining prognosis. Solis et al7 found that increased TTF-1 expression was correlated with favorable outcome, and Gu et al8 reported that absence of EGFR mutation was
1. 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 6:244-285, 2011 2. Travis WD, Brambilla E, Muller-Hermelink HK, et al: Pathology and Genetics: Tumours of the Lung, Pleura, Thymus and Heart, Vol. 1. Lyon, IARC, 2004 3. 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 24:653-664, 2011 4. Russell PA, Wainer Z, Wright GM, et al: 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
5.
6.
7.
8.
9.
an independent poor prognostic factor, which are of interest owing to the link between TTF-1 positivity and EGFR mutation in nonmucinous adenocarcinomas.13 In contrast, invasive mucinous adenocarcinoma is often negative for TTF-1, harbors KRAS mutation, and seems to carry a worse prognosis.1,3 Finally, the signet ring cell subtype was removed from the new classification, but recent reports identified this pattern in tumors with EML4-ALK translocation.3,14,15 Further studies are needed to determine the exact relationships between predominant subtype, cytologic appearance, and mutation status. In summary, although there are no absolute recommendations for surgical management in the new classification, it introduces several concepts that are relevant to the practice of thoracic surgeons, particularly due to the increased frequency of small lung nodules detected on imaging. The role of video-assisted thoracoscopic surgery, sublobar resection for early-stage disease, extent of lymph node dissection, and utility of intraoperative consultation are just a few key issues that were raised by the multidisciplinary process of creating the new classification.1,12 Although most of these require further study, it appears that the correlation of rapid and inexpensive morphology-based diagnosis and prognosis of surgically resected lung adenocarcinoma patients may prove to be extremely beneficial in identifying candidates for adjuvant therapy. Therefore, it is clear that thoracic surgeons are in an ideal position to facilitate the global adoption of the new classification to advance patient care and research, especially because no consensus exists among pathologists on how to use it.
classification. J Thorac Oncol 6:1496-1504, 2011 Xu L, Tavora F, Battafarano R, et al: Adenocarcinomas with prominent lepidic spread: Retrospective review applying new classification of the American Thoracic Society. Am J Surg Pathol 36:273-282, 2012 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 stageindependent predictor of survival. J Clin Oncol 30:1438-1446, 2012 Solis LM, Behrens C, Raso MG, et al: Histologic patterns and molecular characteristics of lung adenocarcinoma associated with clinical outcome. Cancer 118:2889-2899, 2012 Gu J, Lu C, Guo J, 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 (in press) Noguchi M, Morikawa A, Kawasaki M, et al: Small adenocarcinoma of the lung. Histologic
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
10.
11. 12.
13.
14.
15.
characteristics and prognosis. Cancer 75: 2844-2852, 1995 Travis WD, Colby TV, Corrin B, et al: Histologic Typing of Lung and Pleural Tumours, 3rd ed. Berlin, Springer, 1999 Liebow AA: Bronchiolo-alveolar carcinoma. Adv Intern Med 10:329-358, 1960 Van Schil PE, Asamura H, Rusch VW, et al: Surgical implications of the new IASLC/ATS/ ERS adenocarcinoma classification. Eur Respir J 39:478-486, 2012 Yatabe Y, Kosaka T, Takahashi T, et al: EGFR mutation is specific for terminal respiratory unit type adenocarcinoma. Am J Surg Pathol 29:633-639, 2005 Shaw AT, Yeap BY, Mino-Kenudson M, et al: Clinical features and outcome of patients with non-small cell lung cancer who harbor EML4ALK. J Clin Oncol 27:4247-4253, 2009 Yoshida A, Tsuta K, Nakamura H, et al: Comprehensive histologic analysis of ALK-rearranged lung carcinomas. Am J Surg Pathol 35: 1226-1234, 2011
21
Current Readings: Pathology, Prognosis, and Lung Cancer James H. Suh, MD The 2011 International Association for the Study of Lung Cancer/American Thoracic Society/ European Respiratory Society international multidisciplinary classification of lung adenocarcinoma introduced the new categories of adenocarcinoma in situ, minimally invasive adenocarcinoma, and invasive mucinous adenocarcinoma, and replaced the category of mixed subtype adenocarcinoma with lepidic, acinar, papillary, micropapillary, and solid predominant adenocarcinoma. The aim of this manuscript is to evaluate whether the new classification can be applied successfully in determining prognosis of surgically resected patients. Six consecutive clinicopathologic studies using the new classification that were published between spring 2011 and fall 2012 were reviewed. Overall, they demonstrated excellent outcome for adenocarcinoma in situ and minimally invasive adenocarcinoma; intermediate outcome for lepidic, acinar, and papillary predominant adenocarcinoma; and poor outcome for solid and micropapillary predominant adenocarcinoma and invasive mucinous adenocarcinoma. As the new classification remains a proposal at this time, it is hoped that thoracic surgeons will play a leading role in its worldwide dissemination for clinical care and research. Semin Thoracic Surg 25:14-21 © 2013 Elsevier Inc. All rights reserved. Keywords: lung adenocarcinoma, pathology, prognosis, thoracic surgery In February 2011, Travis et al1 published a new proposal, the International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society (IASLC/ATS/ERS) international multidisciplinary classification of lung adenocarcinoma. For the first time, the process included representatives from all medical specialties that participate in the diagnosis and management of lung cancer patients, not just pathologists. This proposal is intended to replace relevant sections within the 2004 World Health Organization (WHO) classification.2 In particular, it eliminates the categories of bronchioloalveolar carcinoma (BAC) and mixed subtype adenocarcinoma. BAC has been used in very different contexts by clinicians and pathologists, and mixed subtype adenocarcinoma has evolved into a wastebasket diagnosis applied to patients with widely disparate outcomes. Therefore, the primary objective of the new classification is to achieve precise correlations between predominant histologic
Department of Pathology, NYU Langone Medical Center, New York, New York. Dr. Suh reports receiving lecture and consulting fees from Pfizer. Address reprint requests to James H. Suh, MD, Department of Pathology, NYU Langone Medical Center, Tisch HW-467, 560 First Avenue, New York, NY 10016. E-mail: [email protected]
14
subtypes of lung adenocarcinoma and survival in surgically resected patients. Since early 2011, there have been 6 retrospective clinicopathologic studies using the new classification, including almost 2000 operated patients from 4 continents during the past 2 decades.3-8 This review will summarize the key points from each article, examine their findings regarding pathology and prognosis, and discuss potential implications for thoracic surgeons. 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 Yoshizawa A, Motoi N, Riely, GJ, et al. Mod Pathol 24:653-664, 2011 Yoshizawa et al3 conducted the first major study on the new classification (Table 1) to evaluate its utility in determining prognosis and implications for staging. In a retrospective review, the authors selected 514 stage I cases at the Memorial Sloan-Kettering Cancer Center (1995-2005) using UICC/AJCC seventh edition criteria, the majority of whom underwent lobectomy (83%) and mediastinal lymph node dissection 1043-0679/$-see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1053/j.semtcvs.2013.01.003
PATHOLOGY, PROGNOSIS, AND LUNG CANCER Table 1. IASLC/ATS/ERS international multidisciplinary classification of lung adenocarcinoma Preinvasive lesions Atypical adenomatous hyperplasia Adenocarcinoma in situ (ⱕ3 cm, formerly BAC) Nonmucinous Mucinous Mixed mucinous nonmucinous Minimally invasive adenocarcinoma (ⱕ3 cm lepidic predominant tumor with ⱕ5-mm invasion) Nonmucinous Mucinous Mixed mucinous nonmucinous Invasive adenocarcinoma Lepidic predominant (formerly nonmucinous BAC pattern, with ⬎5 mm invasion) Acinar predominant Papillary predominant Micropapillary predominant Solid predominant with mucin production Variants of invasive adenocarcinoma Invasive mucinous adenocarcinoma (formerly mucinous BAC) Colloid Fetal (low and high grade) Enteric Reprinted with permission Travis et al.1
and did not receive adjuvant therapy (94%). The median age was 69 years, 63% were female, and the median tumor size was 2.0 cm. An average of 8.6 slides (range: 1-31) per tumor were examined by 1-3 pathologists, depending on difficulty. Mucin stains were evaluated, if needed. Standard pathologic pa-
rameters such as 2004 WHO grade, necrosis, lymphovascular invasion, and visceral pleural invasion were included. In addition, invasive tumor size was calculated, either by ruler or by multiplication of percent invasion and total tumor size. Finally, overall and disease-free survivals were recorded. The median follow-up period was 48 months. Fiftynine percent had no evidence of disease, whereas 14% of patients died of disease, and 21% died of other causes. According to the 2004 WHO classification, 95% were classified as mixed subtype adenocarcinomas. In contrast, application of the new classification identified 3 prognostic subgroups: (a) 0.2% (of total cases) adenocarcinoma in situ (AIS) and 1.4% minimally invasive adenocarcinoma (MIA) had excellent outcomes (100% 5-year disease-free survivals); (b) 5.6% lepidic predominant adenocarcinoma (LPA; 90%), 45.1% acinar predominant adenocarcinoma (ACA; 84%), and 27.8% papillary predominant adenocarcinoma (PPA; 83%) had intermediate outcomes; and (c) 2.5% invasive mucinous adenocarcinoma (76%), 1.8% colloid predominant adenocarcinoma (71%), 13.0% solid predominant adenocarcinoma (SPA; 70%), and 2.3% micropapillary predominant adenocarcinoma (MPA; 67%) had the worst outcomes (Table 2, Fig. 1). Multivariate analysis also identified male gender, invasive tumor size, and necrosis as poor prognostic factors. Compared with total tumor size, invasive tumor size would decrease T-factor scoring in 14% of patients, mostly among MIA, LPA, and invasive mucinous adenocarcinoma, and correlate better with disease-free survival. The findings in this study validate the prognostic impact of the new classification and may be helpful in identifying candidates for adjuvant therapy.
Table 2. Five-year disease-free survival by IASLC subtype and prognostic subgroup IASLC/ATS/ERS classification subtypes
Number (%)
Low grade Adenocarcinoma in situ 1 (0.2%) Minimally invasive adenocarcinoma, nonmucinous 7 (1%) Minimally invasive adenocarcinoma, mixed 1 (0.2%) mucinous and nonmucinous Intermediate grade Lepidic predominant 29 (6%) Acinar predominant 232 (45%) Papillary predominant 143 (28%) High grade Micropapillary predominant 12 (2%) Solid predominant 67 (13%) Colloid predominant 9 (2%) Invasive mucinous adenocarcinoma, mixed mucinous/nonmucinous 13 (3%)
Disease-free survival at 5 years 100% 100% 100%
90% 84% 83% 67% 70% 71% 76%
Reprinted with permission Yoshizawa et al.3
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
15
PATHOLOGY, PROGNOSIS, AND LUNG CANCER
Figure 1. Five-year disease-free survival by IASLC subtype (A) and prognostic subgroup (B). Reprinted with permission Yoshizawa et al.3
DOES LUNG ADENOCARCINOMA SUBTYPE PREDICT PATIENT SURVIVAL? A CLINICOPATHOLOGIC STUDY BASED ON THE NEW IASLC/ATS/ERS INTERNATIONAL MULTIDISCIPLINARY LUNG ADENOCARCINOMA CLASSIFICATION Russell PA, Wainer Z, Wright GM, et al. J Thorac Oncol 6:1426-1504, 2011 Russell et al4 included all patients with lung adenocarcinoma who underwent resection with curative intent. They identified 210 stage I, II, and III patients at the St. Vincent’s Hospital of the University of Melbourne, Australia (1996-2009), using UICC/AJCC seventh edition criteria. Lobectomy was performed in 79% and mediastinal lymph node dissection was performed in 86% of cases in Russell et al. None received neoadjuvant chemotherapy. The median age was 67 years, 46% were female, and the median tumor size was 2.9 cm. An average of 5 slides (range: 1-20) per tumor were reviewed by 2 pathologists. Elastic Van Gieson stains were evaluated in all cases with a lepidic component to assess for invasion, and periodic acid-Schiff stains were evaluated in all cases with a solid component to assess for mucin. The presence or absence of lymphovascular invasion, perineural invasion, visceral pleural invasion, and lymph node metastases was recorded. Prognostic correlations between IASLC diagnosis and 5-year overall survival were explored. The median follow-up period was 49 months. As in the first study, 93% were classified as mixed subtype adenocarcinomas in the 2004 WHO classification. Using the new classification, the authors identified 4 prog16
nostic subgroups: (a) 0.5% AIS (100% 5-year survival), 3.3% MIA (100%), and 4.8% LPA (86%) had the best outcomes (overall 93%); (b) 12.4% PPA (71%) and 40.0% ACA (68%) had intermediate-good outcomes; (c) 4.8% invasive mucinous adenocarcinoma (51%) and 4.3% colloid predominant adenocarcinoma (51%) had intermediate-poor outcomes; and (d) 23.3% SPA (39%) and 6.7% MPA (38%) had the worst outcomes. They discovered that the correlations between IASLC diagnosis and prognosis persisted even after controlling for stage. Therefore, the findings in this study suggest that the new classification can be useful in determining prognosis of stage II and IIIA lung adenocarcinoma patients as well. ADENOCARCINOMAS WITH PROMINENT LEPIDIC SPREAD: RETROSPECTIVE REVIEW APPLYING NEW CLASSIFICATION OF THE ATS Xu L, Tavora F, Battafarano R, et al. Am J Surg Pathol 36:273-282, 2012 In contrast to the first 2 studies, Xu et al5 focused primarily on AIS, MIA, and LPA. They selected 87 mostly early-stage patients at the University of Maryland Medical Center (2007-2010) with surgically resected lung adenocarcinomas. Of the selected cases in Xu’s study, 30 had previous pathologic diagnoses of BAC or mixed subtype adenocarcinoma with BAC features and were included for further analysis. The type of surgical procedure, method of mediastinal lymph node assessment, and administration of adjuvant therapy were
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
PATHOLOGY, PROGNOSIS, AND LUNG CANCER not recorded. The median age was 69 years, 39% were female, and the mean tumor size was 2.3 cm. An unknown number of slides per tumor were reviewed by 2 pathologists. Elastic Van Gieson stains were evaluated, if already available. The presence or absence of lymph node metastases was recorded. Invasive tumor size was measured for all cases. Correlations between IASLC diagnosis and short-term outcome were explored. The mean follow-up periods ranged from 17 to 28 months. Eight of 87 tumors were reclassified from 2004 WHO mixed subtype adenocarcinoma to MIA. All 3 patients with AIS (3.4% of total cases) had no evidence of disease (28-month mean follow-up), and 7 of 8 patients with MIA (9.2%) also had no evidence of disease (22-month mean follow-up). In contrast, only 9 of 19 patients with LPA (21.8%) had no evidence of disease (17-22-month mean follow-up). The presence of a stromal desmoplastic response in areas of invasion conferred an increased likelihood of local recurrence, lymph node metastasis, and distant spread. Outcomes for other subtypes of adenocarcinoma were not reported. Finally, 2 of 8 cases of MIA were originally scored as T2 and T3 based on total tumor size, whereas invasive tumor size was significantly smaller. Although this study is limited by the focus on AIS, MIA, and LPA, it provides further evidence that both AIS and MIA should have an excellent prognosis. THE NOVEL HISTOLOGIC IASLC/ATS/ERS CLASSIFICATION SYSTEM OF LUNG ADENOCARCINOMA IS A STAGEINDEPENDENT PREDICTOR OF SURVIVAL Warth A, Muley T, Meister M, et al. J Clin Oncol 30: 1438-1446, 2012 Warth et al6 used the new classification to study patients with all stages of surgically resected adenocarcinoma— even stage IV. Warth and colleagues identified 500 stage I–IV patients at the University Hospital Heidelberg, Germany (2002-2008), using UICC/AJCC seventh edition criteria. The type of surgical procedure and method of mediastinal lymph node assessment were not recorded. None received neoadjuvant therapy, but 26% received adjuvant chemotherapy and 17% received adjuvant radiation therapy. The median age was 62 years and 40% were female. The median tumor size was not recorded. An average of 4.6 slides (range: 1-14) per tumor were reviewed by 2 pathologists. Mucin stains were evaluated for all cases. Other pathologic parameters such as lymphovascular invasion, visceral pleural invasion, and lymph
node status were not included. Correlations between IASLC diagnosis and 5-year overall, disease-specific, and disease-free survivals were explored. The median follow-up period was not reported. Ninety-three percent were classified as mixed subtype adenocarcinomas in the 2004 WHO classification. They did not identify any cases of AIS or MIA. However, application of the new classification produced 3 prognostic subgroups: (a) 8.2% LPA (78.5 months) had the best outcome; (b) 41.4% ACA (67.3 months) had an intermediate outcome; and (c) 36.6% SPA (58.1 months), 4.6% PPA (48.9 months), and 6.6% MPA (44.9 months) had the worst outcomes (overall 57.2 months). Multivariate analysis showed that IASLC diagnosis was a stageand therapy-independent predictor of survival. Invasive tumor size correlated better with outcome than total tumor size. Finally, they discovered that patients with SPAs received significant benefit from adjuvant chemoradiation therapy. The findings in this study support the utility of the new classification in all stages of resected lung adenocarcinoma patients. HISTOLOGIC PATTERNS AND MOLECULAR CHARACTERISTICS OF LUNG ADENOCARCINOMA ASSOCIATED WITH CLINICAL OUTCOME Solis LM, Behrens C, Raso MG, et al. Cancer 118: 2889-2899, 2012 Solis et al7 applied concepts derived from the new classification as well as adjunct studies to examine associations between pathology and prognosis. Solis et al identified 240 stage I–IV patients at the University of Texas M.D. Anderson Cancer Center (1997-2005) with surgically resected solitary lung nodules and first diagnosis of primary lung adenocarcinoma who did not receive neoadjuvant therapy. The type of surgical procedure and method of mediastinal lymph node assessment were not recorded. Thirty percent received adjuvant therapy. The median age was 67 years and 57% were female. The median tumor size was not recorded. One slide per centimeter of tumor was examined by 3 pathologists. In addition to morphology, immunohistochemistry for TTF-1 and a panel of 5 biomarkers called the FILM index was performed on tissue microarrays. The presence or absence of necrosis and lymphovascular invasion was recorded. Correlations between histologic subtype, immunohistochemistry, and 5-year overall and disease-free survivals were explored.
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
17
PATHOLOGY, PROGNOSIS, AND LUNG CANCER
Figure 2. Five-year recurrence-free survival of nonsolid tumors by TTF-1 expression (A) and FILM index score (B). Reprinted with permission Solis et al.7
The median follow-up period was 51 months. Eighty percent were mixed subtype adenocarcinomas according to the 2004 WHO classification. No cases of AIS were identified. Using concepts from the new classification, they identified 4 mutually exclusive histologic groups and 2 prognostic subgroups: (a) 38.3% (of total cases) tumors containing any amount of solid pattern (hazard ratio ⫽ 1.722) had a worse outcome than all nonsolid tumors, including; (b) 14.2% tumors containing any papillary pattern but no solid pattern; (c) 30.0% tumors containing lepidic and acinar patterns but no solid or papillary patterns; and (d) 17.5% pure acinar pattern tumors. The difference in outcome between patients with group a vs patients with groups b, c, and d tumors was seen at all stages. In nonsolid tumors, they found that increased TTF-1 expression and low FILM index score were correlated with better outcome, also regardless of stage (Fig. 2). The findings in this study suggest that the new classification can be used in alternative ways to determine prognosis in all stages of resected lung adenocarcinoma patients and that immunohistochemical studies can provide additional prognostic information beyond morphology in many cases. PROGNOSTIC SIGNIFICANCE OF THE IASLC/ATS/ERS CLASSIFICATION IN CHINESE PATIENTS—A SINGLEINSTITUTION RETROSPECTIVE STUDY OF 292 LUNG ADENOCARCINOMA CASES Gu J, Lu C, Guo J. J Surg Oncol (in press) In the last study, Gu et al8 used the new classification to study an ethnic population that has distinct clinicopathologic and molecular characteristics of lung adenocarcinoma. Gu and colleagues identified 292 stage I, II, and III patients at the Fudan University Hospital in Shanghai, 18
China (2005-2008), using UICC/AJCC seventh edition criteria. Lobectomy was performed in 95% and mediastinal lymph node dissection was performed in 100% of cases. None received neoadjuvant chemotherapy. The median age was 59 years and 52% were female. The median tumor size was not recorded. An average of 3.7 slides (range: 2-12) were reviewed by 2 pathologists. Additional parameters such as tumor location, tumor size, visceral pleural invasion, and results of immunohistochemical stains were recorded. Correlations between IASLC diagnosis and 5-year overall and disease-free survivals were explored. The median follow-up period was 44 months. Forty percent experienced recurrence or metastasis and 23% died. Ninety-one percent were mixed subtype adenocarcinomas in the 2004 WHO classification. Using the new classification, Gu et al identified 3 prognostic subgroups: (a) 0.3% AIS (100% 5-year disease-free survival) and 4.8% MIA (100%) had excellent outcomes; (b) 10.6% LPA (72%), 5.5% less common variants (62%), 12.3% PPA (56%), and 38.4% ACA (54%) had intermediate outcomes; and (c) 17.8% SPA (46%) and 10.3% MPA (26%) had the worst outcomes. In addition, multivariate analysis identified higher stage and absence of EGFR mutation as poor prognostic factors. The latter finding is of interest owing to the increased frequency of EGFR mutations in East Asian patients. Regardless of reported ethnic differences in the molecular genetics of lung adenocarcinoma, the findings in this study demonstrate that the predominantly morphologic approach of the new classification can be used to determine prognosis in Chinese lung adenocarcinoma patients as well. COMMENTARY The 1995 Noguchi classification identified 2 subtypes of lung adenocarcinoma with 100% 5-year survival after complete resection, which corresponded
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
PATHOLOGY, PROGNOSIS, AND LUNG CANCER
Figure 3. Adenocarcinoma in situ, nonmucinous. (A) The tumor is well-circumscribed and consists entirely of lepidic (formerly nonmucinous bronchioloalveolar carcinoma) growth (low-power). (B) The tumor cells line intact alveolar septae without stromal, vascular, or pleural invasion (high-power). Reprinted with permission Yoshizawa et al.3 (Color version of figure is available online.)
mostly to BAC in the 1999/2004 WHO classifications, a term that was first introduced by Liebow in 1960.2,9-11 However, by the early 21st century, BAC referred to both solitary lung nodules as well as widely metastatic disease. Furthermore, both the Noguchi classification (60% type C) and 2004 WHO classification (⬎90% mixed subtype) contained large categories of patients who exhibited sharply divergent outcome, creating the need for a new classification.3 The new classification divides tumors into 3 categories—preinvasive (AIS), minimally invasive (MIA), and overtly invasive adenocarcinomas (Figs. 3 and 4). Overtly invasive adenocarcinomas are classified as LPA, ACA, PPA, MPA, and SPA using comprehensive histologic subtyping. This morphologic tool requires extensive sampling of each tumor to determine the predominant subtype, which can be ⬍50%. In particular, complete sampling is necessary to exclude areas of invasion before a diagnosis of AIS can be made, which is problematic during intraoperative consultation due to inherent time constraints.1 Since the publication of the new classification in February 2011, there have been 6 retrospective studies, including 1843 patients of all stages who underwent surgery from 1995 to 2010 in the United States, Germany, China, and Australia3-8. In the 3 studies for which data are available, lobectomy was performed in 79%-95%, sublobar resection was performed in 0%-16%, and mediastinal lymph node dissection was performed in 86%-100% of cases.3,4,8 Depending on the study, 6%-30% underwent adjuvant therapy. The median follow-up periods ranged from 17 to 51 months, with most having at least 40 months of follow-up. Whereas 80%-95% of the tumors were originally diagnosed as mixed subtype adenocarcinomas, tumors were reclassified accord-
ing to the new classification in most cases. It appears that adequate sampling of tumors was performed in all studies—in one department, there was a change in policy to require submission of the entire tumor, whereas another chose to process a whole cross-section of tumor at its largest diameter.4,6 Therefore, these studies represent a large international cohort of patients who received the standard of care in both surgical and oncological management, allowing a detailed analysis of the relationship between pathology and prognosis using the new classification. Of the 1603 cases that used the exact terminology proposed by the new classification, there were 6 AIS (0.4%) and 36 MIA (2.2%). Although few in number, all patients diagnosed with either AIS or MIA demonstrated excellent outcome. Three of 6 AIS had 100% 5-year survival, and the other 3 cases had no evidence of disease at short-term follow-up. Similarly, 28 of 36 MIA had 100% 5-year survival, and 7 of 8 remaining cases had no evidence of disease at short-term follow-up. Therefore, future prospective studies could be constructed to determine whether surveillance is possible for these patients. Both AIS and MIA often have a ground-glass appearance on HRCT, are usually ⬍3 cm in size, and tend to occur in peripheral locations. If a ground-glass nodule begins to grow rapidly or develop a solid component, then resection should be considered. The appropriate extent of surgery remains controversial at this time, as lobectomy, segmentectomy, and wedge resection can be performed using a video-assisted thoracoscopic surgery or robotic approach. The ongoing CALGB trial, which randomizes patients with tumors ⬍2 cm, will provide further insight. Next, preoperative image-guided placement of a clip can
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
19
PATHOLOGY, PROGNOSIS, AND LUNG CANCER
Figure 4. Minimally invasive adenocarcinoma, nonmucinous. (A) The tumor consists mostly of lepidic growth with a central area of invasion that measures ⬍5 mm in greatest linear dimension (low-power). (B) There is lepidic growth on the left and acinar invasion on the right (medium-power). (C) Small angulated glands invade through a desmoplastic stroma, but there is no lymphovascular/pleural invasion or tumor necrosis (highpower). Reprinted with permission Travis et al.1 (Color version of figure is available online.)
help locate a subtle lesion at the time of surgery. As discussed earlier, the role of frozen section in this setting is ill-defined. Finally, as AIS and MIA have a low propensity for metastasis, a mediastinal lymph node dissection may not be necessary. Results of future randomized prospective trials could help with this decision.1,12 Nevertheless, the vast majority (97.4%) of lung adenocarcinomas in these studies were overtly invasive adenocarcinomas. Approximately 8% were LPA, 40% were ACA, 14% were PPA, 6% were MPA, and 22% were SPA. The remaining 10% consisted of less common variants such as invasive mucinous adenocarcinoma (formerly mucinous BAC), colloid predominant adenocarcinoma, and enteric adenocarcinoma. Remarkably, all 3 studies that applied the new classification strictly and measured 5-year survival showed similar findings regarding prognosis, despite the inclusion of patients with stage II and III disease in 2 of them.3,4,8 LPA was associated with good out20
come (72%-90%), ACA (54%-84%) and PPA (56%83%) with intermediate outcome, and MPA (26%67%) and SPA (39%-70%) with poor outcome. In addition to separating invasive adenocarcinomas into prognostically relevant categories, the new classification can help distinguish whether multiple tumors represent separate primaries or metastases in many cases. Therefore, dependingon functional capacity, some patients with multiple lung nodules have surgical options, including wedge resection for peripheral less invasive tumors, segmentectomy or lobectomy for central invasive tumors, and other combinations such as multiple wide wedge resections, lobectomy with wide wedge resection(s), bilobectomy, and pneumonectomy.1,12 The new classification also impacts staging. As areas of lepidic growth appear morphologically identical to AIS, it may be possible to subtract such areas before measurement of tumor size. For example, Yoshizawa et al3 found that using invasive tumor size decreased T-
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
PATHOLOGY, PROGNOSIS, AND LUNG CANCER factor scoring in 14% of patients. Similarly, Xu et al5 found that 2 of 8 MIAs had originally been scored as T2 or T3 by total tumor size. Based on this and similar data, introduction of Tis and Tmi categories could lead to more accurate staging in the future. In addition, 3 studies suggest that pathologic diagnosis in the new classification might be a stage-independent predictor of outcome.4,6,8 Although it is encouraging that prognostic correlations persist in patients presenting at all stages of disease, it is likely that some of the differences in survival are due to varying risks of developing lymph node involvement and distant metastases. Regardless of potential impacts on staging, it appears that the new classification identifies candidates for adjuvant therapy, which is critical because up to 40% of stage I patients will recur or die within 5 years.3 In particular, Warth et al6 discovered that patients with SPA received significant benefit from adjuvant chemoradiation therapy. Finally, it is important to discuss limitations of the new classification. Comprehensive histologic subtyping is a complex method and still relies on the ability of pathologists to distinguish between in situ and invasive growth, which can be difficult in the absence of desmoplastic stromal reaction.5 Furthermore, critics of the new classification point out that ⬍5% of tumors in most series are AIS or MIA, whereas the vast majority confers an intermediate prognosis, which does not represent much improvement from previous classifications. For these patients, immunohistochemical and molecular studies may be useful in further refining prognosis. Solis et al7 found that increased TTF-1 expression was correlated with favorable outcome, and Gu et al8 reported that absence of EGFR mutation was
1. 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 6:244-285, 2011 2. Travis WD, Brambilla E, Muller-Hermelink HK, et al: Pathology and Genetics: Tumours of the Lung, Pleura, Thymus and Heart, Vol. 1. Lyon, IARC, 2004 3. 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 24:653-664, 2011 4. Russell PA, Wainer Z, Wright GM, et al: 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
5.
6.
7.
8.
9.
an independent poor prognostic factor, which are of interest owing to the link between TTF-1 positivity and EGFR mutation in nonmucinous adenocarcinomas.13 In contrast, invasive mucinous adenocarcinoma is often negative for TTF-1, harbors KRAS mutation, and seems to carry a worse prognosis.1,3 Finally, the signet ring cell subtype was removed from the new classification, but recent reports identified this pattern in tumors with EML4-ALK translocation.3,14,15 Further studies are needed to determine the exact relationships between predominant subtype, cytologic appearance, and mutation status. In summary, although there are no absolute recommendations for surgical management in the new classification, it introduces several concepts that are relevant to the practice of thoracic surgeons, particularly due to the increased frequency of small lung nodules detected on imaging. The role of video-assisted thoracoscopic surgery, sublobar resection for early-stage disease, extent of lymph node dissection, and utility of intraoperative consultation are just a few key issues that were raised by the multidisciplinary process of creating the new classification.1,12 Although most of these require further study, it appears that the correlation of rapid and inexpensive morphology-based diagnosis and prognosis of surgically resected lung adenocarcinoma patients may prove to be extremely beneficial in identifying candidates for adjuvant therapy. Therefore, it is clear that thoracic surgeons are in an ideal position to facilitate the global adoption of the new classification to advance patient care and research, especially because no consensus exists among pathologists on how to use it.
classification. J Thorac Oncol 6:1496-1504, 2011 Xu L, Tavora F, Battafarano R, et al: Adenocarcinomas with prominent lepidic spread: Retrospective review applying new classification of the American Thoracic Society. Am J Surg Pathol 36:273-282, 2012 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 stageindependent predictor of survival. J Clin Oncol 30:1438-1446, 2012 Solis LM, Behrens C, Raso MG, et al: Histologic patterns and molecular characteristics of lung adenocarcinoma associated with clinical outcome. Cancer 118:2889-2899, 2012 Gu J, Lu C, Guo J, 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 (in press) Noguchi M, Morikawa A, Kawasaki M, et al: Small adenocarcinoma of the lung. Histologic
Seminars in Thoracic and Cardiovascular Surgery ● Volume 25, Number 1
10.
11. 12.
13.
14.
15.
characteristics and prognosis. Cancer 75: 2844-2852, 1995 Travis WD, Colby TV, Corrin B, et al: Histologic Typing of Lung and Pleural Tumours, 3rd ed. Berlin, Springer, 1999 Liebow AA: Bronchiolo-alveolar carcinoma. Adv Intern Med 10:329-358, 1960 Van Schil PE, Asamura H, Rusch VW, et al: Surgical implications of the new IASLC/ATS/ ERS adenocarcinoma classification. Eur Respir J 39:478-486, 2012 Yatabe Y, Kosaka T, Takahashi T, et al: EGFR mutation is specific for terminal respiratory unit type adenocarcinoma. Am J Surg Pathol 29:633-639, 2005 Shaw AT, Yeap BY, Mino-Kenudson M, et al: Clinical features and outcome of patients with non-small cell lung cancer who harbor EML4ALK. J Clin Oncol 27:4247-4253, 2009 Yoshida A, Tsuta K, Nakamura H, et al: Comprehensive histologic analysis of ALK-rearranged lung carcinomas. Am J Surg Pathol 35: 1226-1234, 2011
21