A Radiologic Review of the New TNM Classification for Lung Cancer

Cardiopulmonary Imaging Review Kligerman and bbott New TNM Classification for Lung Cancer Cardiopulmonary Imaging Review FOCUS ON: Seth Kligerman 1 Gerald bbott 2 Kligerman S, bbott G Keywords: International ssociation for the Study of Lung Cancer (ISLC), lung cancer, staging, TNM DOI:10.2214/JR.09.3354 Received July 21, 2009; accepted after revision September 1, 2009. Presented at the 2009 annual meeting of the merican Roentgen Ray Society, oston, M. Winner of the Silver Medal. 1 Department of Diagnostic Radiology, University of Maryland Medical Center, 22 S Greene St., Rm. N2W78, altimore, MD 21201. ddress correspondence to S. Kligerman (sethkligerman@hotmail.com). 2 Department of Radiology, Massachusetts General Hospital and Harvard Medical School, oston, M. CME This article is available for CME credit. See www.arrs.org for more information. JR 2010; 194:562 573 0361 803X/10/1943 562 merican Roentgen Ray Society Radiologic Review of the New TNM Classification for Lung Cancer OJECTIVE. In 2009, a new TNM staging system was published by the International Union gainst Cancer and the merican Joint Committee on Cancer. The new edition will encompass non small cell lung cancer, small cell lung cancer, and bronchopulmonary carcinoids. This article will review many important changes that have been made in the revised staging system. CONCLUSION. It is important that radiologists learn the new system and understand the reasons for the changes to provide more accurate clinical staging. I n the United States, lung cancer remains the most common cause of cancer-related death in both men and women. In 2008, it was estimated that lung cancer was the cause of 57% of cancer-related deaths, accounting for more deaths than breast, prostate, colon, rectal, and pancreatic cancers combined [1]. In 2009, the seventh edition of the TNM staging system for lung cancer was published by the International Union gainst Cancer and the merican Joint Committee on Cancer, based on proposals from the International Staging Project of the International ssociation for the Study of Lung Cancer (ISLC). In addition to non small cell lung cancer (NSCLC), the new classification system will be used to stage both small cell lung cancer (SCLC) and bronchopulmonary carcinoid tumors. Many important revisions have been made to the TNM classification of lung cancer, and it is important for the radiologist to learn of these changes to provide more accurate clinical staging. ISLC Population and Methodology etween 1990 and 2000, the data from 100,869 cases of newly diagnosed primary lung cancer were submitted to the Cancer Research and iostatistics (CR) office. The data originated from 46 different databases collected across 19 countries in North merica, sia, Europe, and ustralia. Of these 100,869 patients, 19,854 were excluded because of unknown histology, incomplete survival data, incomplete stage information, recurrent disease, or data collected outside of the study period [2]. Less common lung neoplasms, such as carcinoid tumors and sarcomas, were also excluded. Of the remaining 81,105 eligible cases, 67,725 (83.5%) were NSCLC and 13,290 (16.4%) were SCLC [3]. Of the 81,105 patients with primary lung cancer, 36% were treated with surgery alone, 11% with radiotherapy alone, 21% with chemotherapy alone, 23% with a combination of therapies, and 9% were treated supportively [4]. Two primary methods of lung cancer staging are available: clinical staging and pathologic staging. In clinical staging, information is provided by noninvasive or minimally invasive techniques, such as physical examination, radiologic examination, endoscopic ultrasound, bronchoscopy, mediastinoscopy, and thoracoscopy. In pathologic staging, information obtained from clinical staging is combined with findings from both the invasive surgical procedure and the pathologic evaluation of the excised tissue [5]. Clinical staging is important and can help to determine the next appropriate step in therapy, such as the decision to proceed with pathologic staging. However, direct comparison between the two systems is difficult because selection bias often leads to improved outcomes in those who obtain pathologic staging versus those who are only staged clinically [4]. It is important to remember that pathologic staging remains the reference standard because the overall level of agreement between the two systems only ranges from 35% to 55% [5]. Of the 67,725 cases 562 JR:194, March 2010

New TNM Classification for Lung Cancer of NSCLC in the ISLC study, clinical staging was available in 53,640 cases, whereas pathologic staging was available in 33,933 cases [4, 6]. Overlap of data occurred when patients had both clinical and pathologic staging data available. Using this large data set, survival statistics were calculated on the basis of the prognostic impact of various factors, including the T, N, and M designations as well as the final stage. djustments were made for cell type, sex, age, and the region where the data were collected. ll data were internally validated by origin and type of database. dditionally, external validation was obtained by testing the results against data collected from 1990 to 2002 in the National Cancer Institute s Surveillance, Epidemiology, and End Results (SEER) database. TLE 1: Seventh Edition of the TNM Classification of Lung Cancer Compared With the Sixth Edition Size Tumor Designation Prior System (Sixth Edition) New System (Seventh Edition) Five-Year Survival Rate (%) 2 cm T1 T1a a 77 d > 2 but 3 cm T1 T1b a 71 d > 3 but 5 cm T2 T2a a 58 d > 5 but 7 cm T2 T2b a 49 d > 7 cm T2 T3 a 35 d Pleural or pericardial invasion Visceral pleura T2 T2a b or T2b c N e Parietal pleura T3 T3 N e Mediastinal pleura T3 T3 N e Parietal pericardium T3 T3 N e Central airway invasion Tumor extending into mainstem bronchus > 2 cm from carina T2 T2a b or T2b c N e Tumor extending into mainstem bronchus 2 cm from carina T3 T3 N e Tumor extending to carina T4 T4 N e Lung atelectasis Tumor causing atelectasis of less than entire lung T2 T2a b or T2b c N e Tumor causing atelectasis of entire lung T3 T3 N e Soft tissue invasion Chest wall and superior sulcus T3 T3 N e Diaphragm T3 T3 N e Mediastinum T4 T4 N e Heart or great vessels T4 T4 N e Trachea T4 T4 N e Esophagus T4 T4 N e Osseous invasion Rib T3 T3 N e Vertebral body T4 T4 N e Nerve invasion Phrenic nerve T3 T3 N e Recurrent laryngeal nerve T4 T4 N e Note Cells in bold indicate a change in the designation from the sixth edition. N indicates not applicable. a T designation is listed for tumors completely surrounded by lung. Designation can increase depending on presence and extent of invasion. b T2a designation if tumor measures 5 cm in long-axis diameter. c T2b designation if tumor measures > 5 cm but 7 cm in long-axis diameter. d Survival based on patients staged pathologically with complete resection of tumor (R0) and no nodal or extranodal metastatic disease (N0M0). e Individual survival statistics not calculated due to limited information. s a group, 5-year survival rate in patients pathologically staged with a T3 and T4 designation (excluding those with tumors > 7 cm or satellite nodules), any R, any N, and M0 was 31% and 22%, respectively. f Survival based on patients staged pathologically with complete or incomplete resection of tumor (any R), any nodal disease (any N), and M0. g Survival based on patients staged pathologically with any tumor designation (any T) and M0. h Survival based on patients staged clinically with any T and any N. (Table 1 continues on next page) JR:194, March 2010 563

Kligerman and bbott TLE 1: Seventh Edition of the TNM Classification of Lung Cancer Compared With the Sixth Edition (continued) Satellite nodules T Designation The tumor (T) designation is determined by the size of the primary tumor as measured in the long-axis diameter, extent of invasion of the primary tumor, and presence or absence of satellite nodules (Table 1). Of 67,725 cases of NSCLC initially included in the ISLC database, 18,198 cases met the inclusion criteria for T designation analysis by having accurate clinical or pathologic staging and no metastatic disease. Using these cases, survival statistics were calculated based on the tumor size in patients who had undergone pathologic staging and surgical resection of the primary tumor. fter surgical resection, patients were classified as R0 if there was no residual disease, R1 if microscopic residual disease was present, or R2 if gross residual disease was present. In patients with nodules measuring 2 cm or less that were completely surrounded by lung or visceral pleura, in the absence of nodal or extranodal metastatic disease (N0M0) and with complete resection of tumor (R0), the 5-year survival was 77%. y comparison, patients with nodules measuring greater than 2 cm but less than or equal to 3 cm using the same criteria had a 5-year survival of 71% [7]. On the basis of these statistically significant findings, the T1 designation has been divided into T1a and Tumor Designation T1b (Fig. 1) using the cut points of 2 and 3 cm, respectively. Similar calculations were performed for completely resected lung masses greater than 3 cm in patients with pathologically staged N0M0 disease, and survival differences were Prior System (Sixth Edition) New System (Seventh Edition) Five-Year Survival Rate (%) Same lobe T4 T3 28 f Same lung, different lobe M1 T4 22 f Lymph node designation No lymphadenopathy N0 N0 56 g Ipsilateral, peripheral, or hilar interlobar zone involvement N1 N1 38 g Ipsilateral upper, aorticopulmonary, lower, or subcarinal zone involvement N2 N2 22 g Supraclavicular or contralateral upper, aorticopulmonary, lower, hilar interlobar, or peripheral zone involvement Metastatic disease designation N3 N3 6 g Contralateral lung metastases M1 M1a 3 h Pleural or pericardial dissemination T4 M1a 2 h Distant metastases M1 M1b 1 h Note Cells in bold indicate a change in the designation from the sixth edition. N indicates not applicable. a T designation is listed for tumors completely surrounded by lung. Designation can increase depending on presence and extent of invasion. b T2a designation if tumor measures 5 cm in long-axis diameter. c T2b designation if tumor measures > 5 cm but 7 cm in long-axis diameter. d Survival based on patients staged pathologically with complete resection of tumor (R0) and no nodal or extranodal metastatic disease (N0M0). e Individual survival statistics not calculated due to limited information. s a group, 5-year survival rate in patients pathologically staged with a T3 and T4 designation (excluding those with tumors > 7 cm or satellite nodules), any R, any N, and M0 was 31% and 22%, respectively. f Survival based on patients staged pathologically with complete or incomplete resection of tumor (any R), any nodal disease (any N), and M0. g Survival based on patients staged pathologically with any tumor designation (any T) and M0. h Survival based on patients staged clinically with any T and any N. optimized using cut points of 3, 5, and 7 cm. Tumors greater than 3 cm but less than or equal to 5 cm had a 5-year survival of 58%. Tumors greater than 5 cm but less than or equal to 7 cm had a 5-year survival of 49% [7]. On the basis of these findings, the T2 Fig. 1 Changes to T1 designation for non small cell lung carcinoma., xial CT image in 53-year-old man shows 1.6 1.5 cm adenocarcinoma in right upper lobe. ny tumor measuring 2 cm in greatest diameter completely surrounded by lung or visceral pleura is now classified as T1a., xial CT image in 67-year-old woman with 2.2 1.5 cm adenocarcinoma in right upper lobe. ny tumor measuring > 2 cm to 3 cm and completely surrounded by lung or visceral pleura is now classified as T1b. 564 JR:194, March 2010

New TNM Classification for Lung Cancer designation has been divided based on size: tumors greater than 3 cm but less than or equal to 5 cm are designated as T2a and tumors greater than 5 cm but less than or equal to 7 cm are designated as T2b (Fig. 2). For tumors greater than 7 cm, 5-year survival was 35%, which was similar to the 41% 5-year survival in pathologically staged T3N0M0 tumors [7]. This led to the reclassification of tumors greater than 7 cm as T3 (Fig. 3). In the previous edition of the TNM staging system, tumors were designated as T2 based not only on size but also on the extent of invasion. Tumors that invaded the visceral pleura and extended into the mainstem bronchus but were greater than 2 cm from the carina or caused atelectasis or postobstructive pneumonia without complete collapse of an entire lung were also classified as T2. However, alterations were necessary given the subdivision of the T2 designation on the basis of size in the revised staging system. Now any tumor less than or equal to 5 cm that meets any of the previously mentioned invasion criteria will be designated as T2a. Similarly, tumors greater than 5 cm but less than or equal to 7 cm with the same degree of invasion will be designated as T2b. ccurate pathologic distinction between degrees of visceral pleural invasion has been shown to be an important prognostic indicator of survival [8]. However, because the visceral pleura may have anywhere from four to six histologic layers, the exact definition of visceral pleural invasion has been debated. Therefore, the ISLC has defined visceral pleural invasion as invasion extending through the elastic layer of the visceral pleura to the surface of the visceral pleura and has recommended the use of elastin stains for determination of this feature [8]. Important changes have also been made to the T3 designation in the revised TNM classification. s stated previously, tumors greater than 7 cm in size are now designated as T3. dditionally, patients with satellite nodules in the same lobe, previously designated as T4, have been reclassified as T3 (Fig. 3). In patients with any pattern of nodal disease (any N), M0, and with or without complete resection (any R), 5-year survival was 28%, which is similar to the 31% for other T3 lesions using the same staging criteria [7]. Many of the T3 designations will remain unchanged in the new revisions, primarily because their incidence was not high enough for accurate statistical survival analysis. Tumors that invade the diaphragm, phrenic nerve, mediastinal pleura, or parietal pericardium (Fig. 4) are still classified as T3. Similarly, tumors that invade the parietal pleura or chest wall, including superior sulcus tumors or tumors that destroy ribs, are still classified as T3. Tumors in the mainstem bronchi less than 2 cm Fig. 2 Changes to T2 designation for non small cell lung carcinoma., xial CT scan in 72-year-old man shows 3.2 2.2 cm adenocarcinoma in right upper lobe. ny tumor > 3 cm but 5 cm completely surrounded by lung is now classified as T2a., xial chest CT in 77-year-old man shows 5.5 5.2 cm squamous cell carcinoma in left upper lobe. ny tumor > 5 cm and 7 cm completely surrounded by lung or visceral pleura is now designated as T2b. Fig. 3 Revisions to T3 designation for non small cell lung carcinoma., xial chest CT in 78-year-old woman shows 8.3 6.4 cm large cell carcinoma in right upper lobe. ny tumor measuring 7 cm is now designated at least as T3., Curved coronal multiplanar reformation CT image in 71-year-old man shows small 1.6-cm adenocarcinoma in left upper lobe nodule with adjacent 8.0-mm satellite nodule. Presence of satellite nodule in same lobe has been redesignated as T3 in new staging system. from the carina but not involving the carina or tumors of any size that cause atelectasis or postobstructive pneumonia of an entire lung (Fig. 4) will still be designated as T3. Tumors with a malignant nodule in a different lobe but in the same lung were classified JR:194, March 2010 565

Kligerman and bbott Fig. 4 T3 designation for non small cell lung carcinoma., xial CT image in 47-year-old woman shows 6.7 4.2 cm mass in lingula with invasion into chest wall (white arrow) and parietal pericardium (black arrow). Mass also invades into location of phrenic nerve (star), which was confirmed at pathology., Coronal chest CT image in 44-year-old woman shows 6.8 5.6 cm large cell carcinoma in right lower lobe with invasion of diaphragm but no hepatic invasion on pathologic evaluation. C, Coronal contrast-enhanced CT image in 66-year-old man shows complete atelectasis of left lung by 6.4 3.5 cm left hilar mass (white arrow), which is difficult to differentiate from surrounding atelectatic lung. Mass extends into mainstem bronchus (black arrow) 1.9 cm from carina but does not involve carina. Fig. 5 Revision to T4 designation for non small cell lung carcinoma. Coronal CT image in 64-yearold man shows 5.2 4.5 cm large cell carcinoma in right upper lobe with 1.4-cm metastatic nodule in right middle lobe (arrow). Satellite nodule in same lung but in different lobe from primary tumor was previously classified as metastatic disease but is now designated as T4. as M1 disease in the prior edition. However, the 5-year survival was 22% in patients pathologically staged with this pattern of disease, any pattern of nodal disease (any N), and with or without complete resection (any R). This is identical to the 22% 5-year survival in patients with other T4 tumors staged using similar pathologic criteria, leading to its reclassification as T4 [7] (Fig. 5 and Table 1). Other masses that meet the T4 designation include tumors that invade the mediastinum, carina, trachea, esophagus, great vessels, and heart (Fig. 6). Invasion of the recurrent laryngeal nerve or vertebral body also will lead to designation of a tumor as T4 (Fig. 7). Fig. 6 T4 designation for non small cell lung carcinoma., xial CT image in 51-year-old man shows 8.8 6.5 cm adenocarcinoma invading into mediastinum with invasion of right pulmonary artery, bronchus intermedius, and carina., xial CT image in 63-year-old woman shows 7.2 6.1 cm squamous cell carcinoma extending into mediastinum, with invasion of trachea (white arrow) and esophagus (black arrow). C, xial multiplanar reformation image 8 mm thick in 84-year-old woman shows large squamous cell carcinoma engulfing heavily calcified branches of left coronary artery (circle). Mass also invades and obstructs left inferior and superior pulmonary veins (arrow). ny mass that invades mediastinum, esophagus, trachea, carina, or great vessels is still designated as T4. ecause the incidence of invasion of these vital structures was relatively low in the study population, survival statistics could not be calculated, and therefore their designation remains unchanged. Numerous changes have been made to the T designation for NSCLC between the new and prior TNM classifications (Table 1). These changes were validated internally by the CR database and externally by the SEER database. dditionally, many recent publications have also validated these recommendations, adding further support to the new revisions [9 12]. C C 566 JR:194, March 2010

New TNM Classification for Lung Cancer N Designation The nodal (N) designation is determined by the presence or absence of metastatic involvement of lymph nodes throughout the thorax (Table 1). To determine if changes needed to be made to the N designation for NSCLC, 68,463 patients with M0 disease were evaluated. However, slight differences between the merican Thoracic Society (Mountain- Dressler) and Japanese nodal maps made statistical analysis difficult. To reconcile these differences, a new nodal chart was created that placed lymph nodes into seven specific zones: supraclavicular, upper, aorticopulmonary, subcarinal, lower, hilar interlobar, and peripheral [13] (Fig. 8). lthough the nomenclature has changed, the general concept remains the same. Patients without nodal metastatic disease are designated as N0. Patients with N1 disease are defined as having metastatic involvement of lymph nodes in the ipsilateral peripheral or hilar zones (Fig. 9). The N2 designation signifies metastatic extension to lymph nodes in the ipsilateral mediastinal (upper, aorticopulmonary, lower) or subcarinal lymph node zones (Fig. 10). The N3 nodal designation includes metastatic involvement of any nodes in the supraclavicular lymph node zone or nodes in contralateral mediastinal, hilar interlobar, or peripheral zones (Fig. 11). When survival statistics were evaluated based on these nodal zones, survival continually decreased in both clinically and pathologically staged patients as the nodal designation increased. For instance, in those clinically staged with any T designation (any T) and without extranodal metastatic disease (M0), the 5-year survival was 42%, 29%, 16%, and 7% for the N0, N1, N2, and N3 designations, respectively [3]. Survival differences were also calculated on the basis of the number of lymph node zones involved in any single nodal designation. For instance, in pathologically staged patients with any T and M0, those with nodal metastases to a single N1 zone had a median survival of 52 months whereas those with metastatic spread to nodes in multiple N1 zones had a median survival of only 31 months [3]. Similar decreases in survival were also seen in patients with multiple N2 nodal zone involvement (median survival 19 months) compared with those with disease in a single N2 nodal zone (median survival 35 months) [3]. Interestingly, these results showed improved survival in patients with a single N2 zone involved compared with those with multiple N1 zones involved. These findings, recently validated by an external study [14], raised the possibility of subdividing the N1 and N2 classifications into N1a (singlezone N1), N1b (multiple-zone N1), N2a (single-zone N2), and N2b (multiple-zone N2). However, when these new categories were analyzed in conjunction with each T stage category, an insufficient number of patients were present in each group for valid statistical analysis. lthough no changes will be made to the N designation in the seventh edition of the TNM classification, the above findings suggest that overall disease burden, in addition to the anatomic location of pathologically involved lymph nodes, contributes to survival. Fig. 7 T4 designation for non small cell lung carcinoma., Coronal CT image in 48-year-old woman with increasing hoarseness and left arm weakness shows large mass invading mediastinum with encasement of aorta, left subclavian artery, and left recurrent laryngeal nerve (black arrow). Mass also extends superiorly into soft tissue of neck with edema of subcutaneous tissues on left due to paralysis from brachial plexus invasion (white arrow)., xial CT image in 55-year-old man with adenocarcinoma at right apex invading vertebral body and compressing spinal cord. ny tumor that invades recurrent laryngeal nerve or vertebral body is designated as T4, which is unchanged from prior system. M Designation The M designation refers to the presence or absence of metastatic disease within or outside of the thorax (Table 1). For evaluation of the M designation, survival statistical analyses were performed on 5,592 clinically staged T4M0 and M1 patients. Clinically staged patients with pleural dissemination of disease, any degree of nodal metastases (any N), and no extranodal metastatic disease (M0) had 1-year and 5-year survival of 36% and 2%, respectively, similar to the 45% 1-year and 3% 5-year survival in those with contralateral lung nodules. This is significantly worse survival than the 53% 1-year and 15% 5-year survival in clinically staged patients with other T4, any N, and M0 lesions [15]. ased on these findings, dissemination of disease to the pleura or pericardium as well as metastatic nodules to the opposite lung will now be designated as M1a (Fig. 12). Metastatic disease outside of the thorax, with a significantly worse 1-year and 5-year survival of 22% and 1%, respectively, has been reclassified as M1b [15] (Fig. 13). lthough there was a small but significant survival difference between those with single and multiple sites of extrathoracic metastatic disease, not enough cases were available for analysis to support further subdivision of the M1b classification [15]. s with the T and N descriptors, external validation has confirmed the importance of the revisions to the M descriptor [11, 12]. When 12,901 patients with T4 or M1 disease were evaluated from the California Cancer Registry, the analyses confirmed that the revised M descriptors are better correlated with survival when compared with the prior edition of the TNM staging system [12]. The only discrepancy was the decreased survival in patients with dissemination of disease to the pericardium, a group with survival similar to those with distant metastatic disease. Small Cell Lung Cancer Unlike NSCLC, SCLC is seen almost exclusively in smokers and is known for its rapid JR:194, March 2010 567

Kligerman and bbott Nodal Zones Supraclavicular Upper orticopulmonary Lower Subcarinal Hilar Interlobar Peripheral Nodal Stations Supraclavicular Low cervical Sternal notch Upper paratracheal (left) Upper paratracheal (right) Prevascular Retrotracheal Lower paratracheal (right) Lower paratracheal (left) Subaortic (aorticopulmonary window) Paraaortic (ascending aorta or phrenic) Paraesophogeal (below carina) Pulmonary ligament Subcarinal Hilar Interlobar Lobar Segmental Subsegmental Fig. 8 International ssociation for the Study of Lung Cancer nodal zones in revised staging system for lung cancer., Comparison of seven new lymph node zones (colored rows) compared with lymph node stations in Mountain-Dressler classification used in prior system. E, Single coronal and three axial CT images through mediastinum show imaging-based map of location of seven new lymph node zones (colored circles) in revised classification. doubling time, high growth fraction, early development of metastatic disease, and initial sensitivity to chemotherapy and radiation [16 18]. Once thought to represent nearly a quarter of all cases of newly diagnosed lung cancer, the incidence of SCLC has steadily declined [19]. Despite its initial response to treatment, the long-term survival of patients with SCLC continues to be much worse than that of patients with NSCLC [18]. Of the 81,015 patients with primary lung cancers analyzed in the CR database, SCLC comprised 13,290 (16.4%) cases. Unlike NSCLC, the staging of SCLC had previously been subdivided into two main categories, limited disease and extensive disease. Limited disease was classified as disease limited to a single hemithorax, although local extension of the primary tumor and supraclavicular lymphadenopathy could be present if they could be included in the same radiation port as the primary D Fig. 9 N1 designation for non small cell lung carcinoma., Curved coronal multiplanar reformation image from CT scan in 55-year-old woman with 1.7-cm right upper lobe adenocarcinoma (T1a) and enlarged lymph node in right hilar zone (arrow)., Corresponding PET image shows 18 F-FDG uptake in both nodule and lymph node. Metastatic disease to lymph nodes in ipsilateral hilum is designated as N1, which is unchanged. C E 568 JR:194, March 2010

New TNM Classification for Lung Cancer Fig. 10 N2 designation for non small cell lung carcinoma. Curved axial multiplanar reformation CT image in 46-year-old woman shows 3.8 1.8 cm adenocarcinoma (T2a) in left upper lobe and metastatic involvement to both ipsilateral hilar (white arrow) and aorticopulmonary (black arrow) lymph node zones. Metastatic disease to ipsilateral mediastinal or subcarinal lymph nodes is still designated as N2. tumor [20, 21]. ll other disease was staged as extensive disease. Of the 13,290 cases of SCLC, 3,430 cases of clinically staged SCLC without distant metastatic disease and with full TNM data were available for statistical analysis. Except for the lack of statistical significance between those with N0 and N1 disease, analysis showed a significant decrease in survival as both the T and N designations increased [21]. Similarly, except for a very small subgroup of eight stage II patients that was a statistical outlier, survival decreased as the stage increased [21]. Given these findings, the ISLC has recommended that the new Fig. 11 N3 designation for non small cell lung carcinoma. Curved coronal multiplanar reformation image from CT scan in 62-year-old woman with 3.2 2.8 cm adenocarcinoma (T2a) in left upper lobe shows pathologic enlargement of lymph nodes in subcarinal zone (arrowhead), ipsilateral upper zone (white arrow), and supraclavicular lymph nodes (black arrow). Metastatic involvement of any lymph node in supraclavicular zone is designated as N3, which is unchanged from prior staging system. TNM staging system be applied to SCLC (Fig. 14). This recommendation is supported by external validation that has also shown survival being more accurately predicted using the revised TNM staging system [22]. Fig. 12 Changes to M1 designation for non small cell lung carcinoma., Coronal CT image in 51-year-old man with 3.4 3.2 cm adenocarcinoma (T2a) in right lower lobe and single 5-mm nodule in left lower lobe (arrow), which was pathologically similar to primary tumor and was thought to represent contralateral metastatic nodule., xial CT image in 37-year-old woman with large left lung mass invading mediastinum (T4). Complex pleural and pericardial effusions were malignant at cytology. Nodules in contralateral lung and malignant pleural or pericardial disease are classified as M1a in new staging system. ronchopulmonary Carcinoid Tumors ronchopulmonary carcinoid tumors are malignant neuroendocrine tumors that are divided into typical and atypical subtypes based on pathologic criteria. Carcinoid tumors exist within a spectrum of neuroendocrine tumors of the lung that range from low-grade typical carcinoid tumors to intermediate-grade atypical carcinoid tumors and high-grade small cell carcinomas and large cell carcinomas [23]. Using combined data from both the ISLC and SEER databases, survival analysis was performed on 1,829 pathologically staged patients with bronchopulmonary carcinoid tumors using the revised TNM staging criteria. s with NSCLC and SCLC, as the T, N, and M designations of bronchopulmonary carcinoid tumors independently increased, there was a statistically significant decrease in 5-year survival. Similarly, as the stage increased, 5-year survival decreased [8]. On the basis of these findings, it has been recommended that bronchopulmonary carcinoid tumors be staged using the newly revised TNM classification (Fig. 15). Prognostic Factors Of the prognostic factors evaluated, pathologic stage, sex, cell type, and age were all independently significant [24] (Table 2). These factors were also significant when clinically staged cases were evaluated in a separate analysis [2]. Earlier stage of disease at the time of diagnosis, female sex, and younger age were all favorable prognostic factors [24]. In terms of cell type, patients with bronchioloalveolar carcinoma, either pure or mixed, have the best survival across all calculations [24]. This survival benefit is well described in the literature [25 30]. Survival in patients with adenocarcinomas and squamous cell carcinomas was similar when calculated across age and sex. However, when survival was calculated after making adjustments for pathologic stage, cell type, sex, and age, there was a significant survival advantage in those diagnosed with squamous cell carcinomas compared with adenocarcinomas and large cell carcinomas [24]. similar survival advantage was also noted in patients with squamous cell carcinomas who were clinically staged with III disease [2]. lthough this would suggest some survival advantage in those with squamous cell carcinoma compared with these two cell types, this benefit is debated in the literature [31 35]. Smoking history was only available in 2,467 of the 9,173 pathologically staged patients. Current smokers had an increased incidence of squamous cell carcinoma as well as an overall worse prognosis when compared with never smokers and prior smokers [24]. These findings have been validated in multiple external studies [26, 36 38]. Prognosis between former smokers and never smokers was less well-defined and in most JR:194, March 2010 569

Kligerman and bbott Fig. 13 Changes to M1 designation for non small cell lung carcinoma., Fused PET/CT image in 49-year-old man shows 2.8-cm right lower lobe nodule (T1b) with intense 18 F-FDG uptake., Fused coronal image from PET/CT shows increased FDG uptake in right adrenal gland (arrow) and bone. Metastatic disease outside of thorax is now classified as M1b. Fig. 14 Classification of small cell lung cancer (SCLC) using revised TNM system. xial CT image in 59-year-old woman shows 6.2-cm SCLC extending into mediastinum and invading pulmonary artery without evidence of disease elsewhere (T4N0M0). SCLC should be staged using new TNM classification for lung cancer. Fig. 15 Classification of bronchopulmonary carcinoid tumors using revised TNM system. xial 10-mm-thick maximum-intensity-projection image in 38-year-old woman with atypical carcinoid tumor shows multiple nodules and masses confined to right lower lobe without disease elsewhere (T3N0M0). instances, there was no survival difference between these two subsets [24]. Performance status was available in 3,027 of the 9,173 pathologically staged patients with NSCLC. Decreased performance status was associated with an increased mortality and was shown to be an independently significant factor in determining survival [24]. similar benefit was shown in clinically staged patients [2]. Of the 13,290 cases of SCLC, prognostic factors were evaluated in 6,609 clinically staged patients. Using the prior system for the staging of SCLC, 2,870 patients had limited disease and 3,739 had extensive disease. s seen with NSCLC, increasing age, male sex, and decreased performance status are all associated with decreased survival [2]. Extensive disease was also associated with a worse survival than in those patients with limited disease [2]. dditional factors such as biologic markers, genetic markers, and functional imaging with PET have shown significant prognostic implications in the literature [2, 39, 40]. However, because the data in the CR database were collected from 1990 to 2000, this information was not available for statistical analysis. Changes to the Staging System In addition to revising the T, N, and M designators, significant changes to the final staging system were made to best correlate decreasing survival with increasing stage (Table 3). t the same time, changes were made to prevent overlap in the survival curves between the different stages. In the absence of metastatic disease, all T4 tumors with N0 or N1 nodes have been downstaged from III to III. T2b masses with N0M0 disease will be upstaged from I to II and T2a masses with N1M0 disease will be downstaged from II to II [3]. Treatment Implications y making adjustments to the T designation, M designation, and final stage based on both short- and long-term survival characteristics, the revised staging system for lung cancer is more accurately correlated with survival when compared with the prior staging system. Importantly, survival statistics from external studies have also confirmed the validity of these changes. Not only are these changes better correlated with survival, but they also more closely reflect trends in both definitive and palliative treatment. For instance, in the prior TNM revision, patients with tumors designated as T4 on the basis of invasion were at least staged as III, even those with N0M0 disease. lthough these patients were classically considered nonsurgical, successful resection with improved survival has been shown to be possible in select patients with direct invasion of the mediastinum, carina, pulmonary artery, vertebral body, superior vena cava, aorta, and left atrium [41 47]. These findings support the shift of locally invasive T4 tumors from stage III to III in the absence of N2 or N3 and M1a or M1b disease. Conversely, although no distinction was made between T4 tumors due to invasion and T4 tumors due to malignant pleural or pericardial disease in the prior revision, those tumors with malignant pleural or pericardial dissemination were clinically subcategorized as wet III. Those patients were deemed inappropriate candidates for definitive local therapy and thus were clinically regarded as having stage IV disease and treated only with systemic therapy [42, 45, 48]. Patients with satellite nodules in the same lobe and in the same lung but a different lobe, even in the absence of nodal metastases, would have been staged as III and IV, respectively. ccording to the prior system, this staging would have precluded these patients from 570 JR:194, March 2010

New TNM Classification for Lung Cancer TLE 2: Prognostic Factors of Survival in Pathologically Staged Patients With Non Small Cell Lung Carcinoma TNM stage Factor Median Survival (mo) Five-Year Survival Rate (%) I 95 66 I 75 56 II 44 43 II 29 35 III 19 23 Cell type ronchoalveolar carcinoma 83 61 denocarcinoma 45 44 Squamous cell carcinoma 44 43 Large cell carcinoma 34 41 denosquamous carcinoma 26 29 Sex Female 66 52 Male 40 41 ge (y) < 70 49 46 70 38 38 TLE 3: Revisions to Stage Groupings in the Seventh Edition of the TNM Classification for Lung Tumors Compared With the Sixth Edition Stage in Seventh Edition Stage in Sixth Edition N0 N1 N2 N3 T1a T1 I II III III T1b T1 I II III III T2a T2 I II (II) III III T2b T2 II II III III T3 (> 7 cm) T2 II (I) III (II) III III T3 (invasion) T3 II III III III T3 (satellite nodule, same lobe) T4 III (III) III (III) III (III) III T4 (invasion) T4 III (III) III (III) III III T4 (ipsilateral nodule, different lobe) M1 III (IV) III (IV) III (IV) III (IV) M1a (pleural or pericardial dissemination) T4 IV (III) IV (III) IV (III) IV (III) M1a (contralateral lung nodules) M1 IV IV IV IV M1b (distant metastatic disease) M1 IV IV IV IV Note Cells in bold indicate a change in the stage from the sixth edition. djacent stage in parentheses represents staging from the sixth edition. attempted surgical resection. However, many patients with this pattern of disease did undergo surgical resection and their outcomes were much improved compared with patients with distant metastatic disease [45, 49 51]. This surgical evidence further supports the shift in staging in the revised seventh edition of the TNM classification of lung cancer. lthough staging can help direct therapy, it is important to remember that prospective clinical trials should be used to define optimal patient treatment and not staging systems based on retrospective survival data. For SCLC, radiotherapy and chemotherapy remain the mainstays of treatment [18]. However, a more well-defined tumor and nodal classification system will allow more precise treatment. This precision will be of growing importance as increased radiotherapy doses or more aggressive combined therapies are explored for patients with more localized disease [21]. Tumor size and invasion also can help determine treatment in SCLC. The term limited disease makes no distinction between a small intraparenchymal SCLC without nodal disease and a large mass invading the mediastinum with metastatic disease to multiple ipsilateral lymph nodes. This is an important distinction because the treatments can differ. Studies have shown that even when using the prior TNM classification, patients with stage I and possibly those with stage II SCLC can benefit from surgery and neoadjuvant chemotherapy, whereas those with more extensive localized disease should receive radiation and chemotherapy alone [52 54]. Limitations Despite the extensive power of the study, there are some limitations to the revised staging system for lung cancer. The new system is based on retrospective data that were primarily collected from databases not specifically designed to study the TNM classification of lung cancer [55]. lthough tumor size was often included in the data, more precise information such as the exact site of tumor invasion and extent of lymph node involvement was often omitted. For this reason, much of the T3 and T4 classification remains unchanged, and no changes were made to the N classification despite significant survival differences in those with single versus multiple nodal zone involvement. dditionally, important tumor characteristics, such as the presence or absence of lymphangitic spread of tumor, were not evaluated. Prognostic information based on tumor biology and tumor genetics was not included in the study because most of these data were not widely available. Factors such as the reliability and accuracy of imaging in clinical staging and the prognostic impact of PET imaging were not addressed. Staging and treatment strategies were not uniform because the data were collected from 46 databases across 19 countries. Finally, although the new system is the best method to date to predict patient survival, all of the survival data comes from retrospective analysis and has not been tested prospectively. Given these limitations, a prospective database is currently being developed to assess the validity of each component JR:194, March 2010 571

Kligerman and bbott of the TNM classification as well as to assess multiple other factors that are important in lung cancer staging and prognosis [55]. Oncol 2008; 3:1384 1390 9. Ye C, Masterman JR, Huberman MS, et al. Subdivision of the T1 size descriptor for stage I non- cancer. J Thorac Oncol 2007; 2:1067 1077 22. Ignatius Ou SH, Zell J. The applicability of the proposed ISLC staging revisions to small cell Conclusion Many important changes have occurred with the adoption of the seventh edition of the TNM staging system for lung cancer. These changes are better correlated with survival characteristics and current trends in definitive and palliative therapy. ecause imaging plays such a crucial role in the evaluation of lung cancer, it is imperative that radiologists become well versed in the new revision of the TNM staging system to optimize patient care. References 1. Jemal, Siegel R, Ward E, et al. Cancer statistics, 2008. C Cancer J Clin 2008; 58:71 96 2. Sculier JP, Chansky K, Crowley JJ, Van Meerbeeck J, Goldstraw P. The impact of additional prognostic factors on survival and their relationship with the anatomical extent of disease expressed by the 6th edition of the TNM Classification of Malignant Tumors and the proposals for the 7th edition. J Thorac Oncol 2008; 3:457 466 3. Goldstraw P, Crowley J, Chansky K, et al. The ISLC Lung Cancer Staging Project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM Classification of Malignant Tumours. J Thorac Oncol 2007; 2:706 714 4. Groome P, olejack V, Crowley JJ, et al. The ISLC Lung Cancer Staging Project: validation of the proposals for revision of the T, N, and M descriptors and consequent stage groupings in the forthcoming (seventh) edition of the TNM Classification of Malignant Tumours. J Thorac Oncol 2007; 2:694 705 5. Lopez-Encuentra, Garcia-Lujan R, Rivas JJ, Rodriguez-Rodriguez J, Torres-Lanza J, Varela- Simo G. Comparison between clinical and pathologic staging in 2,994 cases of lung cancer. nn Thorac Surg 2005; 79:974 979; discussion, 979 6. Matin T, Goldberg M. Surgical staging of lung cancer. Oncology (Williston Park) 1999; 13:679 685; discussion, 685, 689, 693 7. Rami-Porta R, all D, Crowley J, et al. The ISLC Lung Cancer Staging Project: proposals for the revision of the T descriptors in the forthcoming (seventh) edition of the TNM classification for lung cancer. J Thorac Oncol 2007; 2:593 602 8. Travis WD, rambilla E, Rami-Porta R, et al. Visceral pleural invasion: pathologic criteria and use of elastic stains: proposal for the 7th edition of the TNM classification for lung cancer. J Thorac small cell lung cancer has prognostic value: a single institution experience. Chest 2009; 136:710 715 10. Rami-Porta R. New TNM classification for lung cancer [in Spanish]. rch ronconeumol 2009; 45:159 161 11. Kameyama K, Takahashi M, Ohata K, et al. Evaluation of the new TNM staging system proposed by the International ssociation for the Study of Lung Cancer at a single institution. J Thorac Cardiovasc Surg 2009; 137:1180 1184 12. Ou SH, Zell J. Validation study of the proposed ISLC staging revisions of the T4 and M nonsmall cell lung cancer descriptors using data from 23,583 patients in the California Cancer Registry. J Thorac Oncol 2008; 3:216 227 13. Rusch VW, samura H, Watanabe H, Giroux DJ, Rami-Porta R, Goldstraw P. The ISLC Lung Cancer Staging Project: a proposal for a new international lymph node map in the forthcoming seventh edition of the TNM classification for lung cancer. J Thorac Oncol 2009; 4:568 577 14. Lee JG, Lee CY, ae MK, et al. Validity of International ssociation for the Study of Lung Cancer proposals for the revision of N descriptors in lung cancer. J Thorac Oncol 2008; 3:1421 1426 15. Postmus PE, rambilla E, Chansky K, et al. The ISLC Lung Cancer Staging Project: proposals for revision of the M descriptors in the forthcoming (seventh) edition of the TNM classification of lung cancer. J Thorac Oncol 2007; 2:686 693 16. Johnson DH. Management of small cell lung cancer: current state of the art. Chest 1999; 116[6 suppl]:525s 530S 17. Elias D. Small cell lung cancer: state-of-the-art therapy in 1996. Chest 1997; 112;[4 suppl]:251s 258S 18. Sher T, Dy GK, djei. Small cell lung cancer. Mayo Clin Proc 2008; 83:355 367 19. Govindan R, Page N, Morgensztern D, et al. Changing epidemiology of small-cell lung cancer in the United States over the last 30 years: analysis of the surveillance, epidemiologic, and end results database. J Clin Oncol 2006; 24:4539 4544 20. Micke P, Faldum, Metz T, et al. Staging small cell lung cancer: Veterans dministration Lung Study Group versus International ssociation for the Study of Lung Cancer what limits limited disease? Lung Cancer 2002; 37:271 276 21. Shepherd F, Crowley J, Van Houtte P, et al. The International ssociation for the Study of Lung Cancer Lung Cancer Staging Project: proposals regarding the clinical staging of small cell lung cancer in the forthcoming (seventh) edition of the tumor, node, metastasis classification for lung lung cancer (SCLC) with comparison to the current UICC 6th TNM edition. J Thorac Oncol 2009; 4:300 310 23. Hage R, de la Riviere, Seldenrijk C, van den osch JM. Update in pulmonary carcinoid tumors: a review article. nn Surg Oncol 2003; 10:697 704 24. Chansky K, Sculier JP, Crowley JJ, Giroux D, Van Meerbeeck J, Goldstraw P. The International ssociation for the Study of Lung Cancer Staging Project: prognostic factors and pathologic TNM stage in surgically managed non-small cell lung cancer. J Thorac Oncol 2009; 4:792 801 25. reathnach OS, Kwiatkowski DJ, Finkelstein DM, et al. ronchioloalveolar carcinoma of the lung: recurrences and survival in patients with stage I disease. J Thorac Cardiovasc Surg 2001; 121:42 47 26. ryant, Cerfolio RJ. Differences in epidemiology, histology, and survival between cigarette smokers and never-smokers who develop nonsmall cell lung cancer. Chest 2007; 132:185 192 27. Liu YY, Chen YM, Huang MH, Perng RP. Prognosis and recurrent patterns in bronchioloalveolar carcinoma. Chest 2000; 118:940 947 28. Rena O, Papalia E, Ruffini E, et al. Stage I pure bronchioloalveolar carcinoma: recurrences, survival and comparison with adenocarcinoma of the lung. Eur J Cardiothorac Surg 2003; 23:409 414 29. Zell J, Ignatius Ou SH, Ziogas, nton-culver H. Validation of the proposed International ssociation for the Study of Lung Cancer non-small cell lung cancer staging system revisions for advanced bronchioloalveolar carcinoma using data from the California Cancer Registry. J Thorac Oncol 2007; 2:1078 1085 30. Campione, Ligabue T, Luzzi L, et al. Impact of size, histology, and gender on stage I non-small cell lung cancer. sian Cardiovasc Thorac nn 2004; 12:149 153 31. Janssen-Heijnen ML, Coebergh JW. The changing epidemiology of lung cancer in Europe. Lung Cancer 2003; 41:245 258 32. Makitaro R, Paakko P, Huhti E, loigu R, Kinnula VL. n epidemiological study of lung cancer: history and histological types in a general population in northern Finland. Eur Respir J 1999; 13:436 440 33. Devesa SS, ray F, Vizcaino P, Parkin DM. International lung cancer trends by histologic type: male:female differences diminishing and adenocarcinoma rates rising. Int J Cancer 2005; 117: 294 299 34. Mountain CF, Lukeman JM, Hammar SP, et al. Lung cancer classification: the relationship of dis- 572 JR:194, March 2010

New TNM Classification for Lung Cancer ease extent and cell type to survival in a clinical trials population. J Surg Oncol 1987; 35:147 156 35. Okamoto T, Maruyama R, Suemitsu R, et al. Prognostic value of the histological subtype in completely resected non-small cell lung cancer. Interact Cardiovasc Thorac Surg 2006; 5:362 366 36. Zhang Z, Xu F, Wang S, Li N, Wang C. Influence of smoking on histologic type and the efficacy of adjuvant chemotherapy in resected non-small cell lung cancer. Lung Cancer 2008; 60:434 440 37. Tsao S, Liu D, Lee JJ, Spitz M, Hong WK. Smoking affects treatment outcome in patients with advanced nonsmall cell lung cancer. Cancer 2006; 106:2428 2436 38. Toh CK, Wong EH, Lim WT, et al. The impact of smoking status on the behavior and survival outcome of patients with advanced non-small cell lung cancer: a retrospective analysis. Chest 2004; 126:1750 1756 39. Erasmus JJ, Rohren E, Swisher SG. Prognosis and reevaluation of lung cancer by positron emission tomography imaging. Proc m Thorac Soc 2009; 6:171 179 40. Molina JR, Yang P, Cassivi SD, Schild SE, djei. Non-small cell lung cancer: epidemiology, risk factors, treatment, and survivorship. Mayo Clin Proc 2008; 83:584 594 41. Garrido P, Gonzalez-Larriba JL, Insa, et al. Long-term survival associated with complete resection after induction chemotherapy in stage III (N2) and III (T4N0-1) non small-cell lung cancer patients: the Spanish Lung Cancer Group Trial 9901. J Clin Oncol 2007; 25:4736 4742 42. Jett JR, Schild SE, Keith RL, Kesler K. Treatment of non-small cell lung cancer, stage III: CCP evidence-based clinical practice guidelines (2nd edition). Chest 2007; 132[3 suppl]:266s 276S 43. Ohta M, Hirabayasi H, Shiono H, et al. Surgical resection for lung cancer with infiltration of the thoracic aorta. J Thorac Cardiovasc Surg 2005; 129:804 808 44. nraku M, Waddell TK, de Perrot M, et al. Induction chemoradiotherapy facilitates radical resection of T4 non-small cell lung cancer invading the spine. J Thorac Cardiovasc Surg 2009; 137:441 447 45. Gallo E, Donington JS. The role of surgery in the treatment of stage III non-small-cell lung cancer. Curr Oncol Rep 2007; 9:247 254 46. Stamatis G, Eberhardt W, Stuben G, ildat S, Dahler O, Hillejan L. Preoperative chemoradiotherapy and surgery for selected non-small cell lung cancer III subgroups: long-term results. nn Thorac Surg 1999; 68:1144 1149 47. Fukuse T, Wada H, Hitomi S. Extended operation for non-small cell lung cancer invading great vessels and left atrium. Eur J Cardiothorac Surg 1997; 11:664 669 48. ltundag O, Stewart DJ, Stevens DC, et al. The risk of distant metastases in patients with nonsmall cell lung cancer (NSCLC) with cytologically proven malignant pleural effusion, stage III: a retrospective analysis. (abstr) J Clin Oncol 2005; 23:9568 49. okage K, Ishii G, Nagai K, et al. Intrapulmonary metastasis in resected pathologic stage III nonsmall cell lung cancer: possible contribution of aerogenous metastasis to the favorable outcome. J Thorac Cardiovasc Surg 2007; 134:386 391 50. Roy MS, Donington JS. Management of locally advanced non small cell lung cancer from a surgical perspective. Curr Treat Options Oncol 2007; 8:1 14 51. attafarano RJ, Meyers F, Guthrie TJ, Cooper JD, Patterson G. Surgical resection of multifocal non-small cell lung cancer is associated with prolonged survival. nn Thorac Surg 2002; 74:988 993; discussion, 993 994 52. Waddell TK, Shepherd F. Should aggressive surgery ever be part of the management of small cell lung cancer? Thorac Surg Clin 2004; 14:271 281 53. Nakamura H, Kazuyuki S, Kawasaki N, Taguchi M, Kato H. History of limited resection for nonsmall cell lung cancer. nn Thorac Cardiovasc Surg 2005; 11:356 362 54. Coolen L, Van den Eeckhout, Deneffe G, Demedts M, Vansteenkiste J. Surgical treatment of small cell lung cancer. Eur J Cardiothorac Surg 1995; 9:59 64 55. Giroux DJ, Rami-Porta R, Chansky K, et al. The ISLC Lung Cancer Staging Project: data elements for the prospective project. J Thorac Oncol 2009; 4:679 683 FOR YOUR INFORMTION This article is available for CME credit. See www.arrs.org for more information. JR:194, March 2010 573