Advertisement

The IASLC Lung Cancer Staging Project: Proposals for the Revisions of the T Descriptors in the Forthcoming Eighth Edition of the TNM Classification for Lung Cancer

      Introduction:

      An international database was collected to inform the 8th edition of the anatomic classification of lung cancer. The present analyses concern its primary tumor (T) component.

      Methods:

      From 1999 to 2010, 77,156 evaluable patients, 70,967 with non–small-cell lung cancer, were collected; and 33,115 had either a clinical or a pathological classification, known tumor size, sufficient T information, and no metastases. Survival was measured from date of diagnosis or surgery for clinically and pathologically staged tumors. Tumor-size cutpoints were evaluated by the running log-rank statistics. T descriptors were evaluated in a multivariate Cox regression analysis adjusted for age, gender, histological type, and geographic region.

      Results:

      The 3-cm cutpoint significantly separates T1 from T2. From 1 to 5 cm, each centimeter separates tumors of significantly different prognosis. Prognosis of tumors greater than 5 cm but less than or equal to 7 cm is equivalent to T3, and that of those greater than 7 cm to T4. Bronchial involvement less than 2 cm from carina, but without involving it, and total atelectasis/pneumonitis have a T2 prognosis. Involvement of the diaphragm has a T4 prognosis. Invasion of the mediastinal pleura is a descriptor seldom used.

      Conclusions:

      Recommended changes are as follows: to subclassify T1 into T1a (⩽1 cm), T1b (>1 to ⩽2 cm), and T1c (>2 to ⩽3 cm); to subclassify T2 into T2a (>3 to ⩽4 cm) and T2b (>4 to ⩽5 cm); to reclassify tumors greater than 5 to less than or equal to 7 cm as T3; to reclassify tumors greater than 7 cm as T4; to group involvement of main bronchus as T2 regardless of distance from carina; to group partial and total atelectasis/pneumonitis as T2; to reclassify diaphragm invasion as T4; and to delete mediastinal pleura invasion as a T descriptor.

      Key Words

      The 7th edition of the tumor, node, and metastasis (TNM) classification of lung cancer published in 2009 was based on the most thorough data-based revision ever done to date.
      • Goldstraw P
      • Sobin L
      • Gospodarowocz M
      • Wittekind C
      • Edge SB
      • Byrd DR
      • Compton CC
      • Fritz AG
      • Greene FL
      • Trotti III, A
      A retrospective international database including 81,495 evaluable patients collected from 1990 to 2000 by the International Association for the Study of Lung Cancer (IASLC) and analyzed by Cancer Research And Biostatistics (CRAB) was used for the revision.
      • Goldstraw P
      • Crowley JJ
      The International Association for the Study of Lung Cancer international staging project on lung cancer.
      The revision consisted of changes in the T descriptors that emphasized the prognostic impact of tumor size and redefined the classification of additional tumor nodules and malignant pleural effusion, the subclassification of M1, the validation of the classification for bronchopulmonary carcinoid tumors, and the rearrangement of stage grouping, whereas the N descriptors remained the same. Despite the magnitude of the database not all descriptors could be validated.
      • Rami-Porta R
      • Goldstraw P
      Strength and weakness of the new TNM classification for lung cancer.
      The limitations of the retrospective database prompted the IASLC to launch a call for the collection of new data.
      • Giroux DJ
      • Rami-Porta R
      • Chansky K
      • et al.
      International Association for the Study of Lung Cancer International Staging Committee. The IASLC Lung Cancer Staging Project: data elements for the prospective project.
      The call resulted in a new database of 77,156 evaluable patients diagnosed with lung cancer from 1999 to 2010.
      • Rami-Porta R
      • Bolejack V
      • Giroux DJ
      • et al.
      International Association for the Study of Lung Cancer Staging and Prognostic Factors Committee, Advisory Board Members and Participating Institutions. The IASLC Lung Cancer Staging Project: the new database to inform the 8th edition of the TNM classification of lung cancer.
      This new database is being used now to inform the 8th edition of the TNM classification of lung cancer due to be published in 2016.
      This article presents the results of the analyses of the new IASLC database performed by the members of the Primary Tumor (T) Subcommittee of the IASLC Staging and Prognostic Factors Committee and the statisticians of CRAB concerning the T component of the TNM classification and its descriptors. The analyses were conducted to achieve predefined objectives: to further assess the prognostic impact of tumor size; to assess the prognostic power of each descriptor defining the different T categories; and to study new conditions not included in the present T descriptors, such as differences between parietal pleura and rib invasion.
      • Giroux DJ
      • Rami-Porta R
      • Chansky K
      • et al.
      International Association for the Study of Lung Cancer International Staging Committee. The IASLC Lung Cancer Staging Project: data elements for the prospective project.

      PATIENTS AND METHODS

      Population

      The total number of patients diagnosed with lung cancer between 1999 and 2010 submitted to CRAB was 94,708. After exclusions, 77,156 (70,967 with non–small-cell lung cancer [NSCLC] and 6189 with small-cell lung cancer) remained for analysis.
      • Rami-Porta R
      • Bolejack V
      • Giroux DJ
      • et al.
      International Association for the Study of Lung Cancer Staging and Prognostic Factors Committee, Advisory Board Members and Participating Institutions. The IASLC Lung Cancer Staging Project: the new database to inform the 8th edition of the TNM classification of lung cancer.
      In the NSCLC group, 33,115 patients met the T descriptors subcommittee’s initial analytic requirements of M0 NSCLC, a complete set of either clinical (c) TNM or pathological (p) TNM, known tumor size, and sufficiently detailed T descriptors to support the assigned T category. There was sufficient clinical T descriptor information for 13,012 patients, including 12,449 who were eventually operated, distributed as follows: 10,084 (81.0%) cN0, 907 (7.3%) cN1, 1327 (10.7%) cN2, and 131 (1.1%) cN3. As for the analysis of the pathologic T, the population excluded those who had induction treatment and consisted of 30,018 patients with complete pTN and M0 tumors (9915 of these also provided complete cTN categories; Table 1). Their distribution according to the pN component is 22,257 (74.2%) pN0, 3465 (11.5%) pN1, 4157 (13.9%) pN2, and 139 (0.5%) pN3. Asia was the geographic region that contributed most to the IASLC database: 10,294 (79%) patients with clinically staged tumors and 23,838 (79%) with pathologically staged ones came from Japan, South Korea, and People’s Republic of China (Supplementary Table 1, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). Adenocarcinoma was the most common cell type, with 64% of tumors both clinically and pathologically staged. Squamous cell carcinoma followed with 25% of clinically staged tumors and 27% of pathologically staged tumors (Supplementary Table 2, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). From the 30,018 patients with surgically resected and pathologically staged tumors, 28,150 (94%) were completely resected (Supplementary Table 3, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). To assess the completeness of resection, the information given by the data providers was considered. When the specific residual tumor (R) status was unknown, the case was grouped in the “any R” category.
      TABLE 1Number of M0 Non–Small-Cell Lung Cancer Cases Passing Initial Screening
      Criteria for T descriptor analysis: cases must have known tumor size, at least one T descriptor supporting the assigned T category, and no T descriptors suggesting a higher T category.
      N0Any N
      TotalT1T2T3T4TotalT1T2T3T4
      Clinically staged
       Total30, 10217, 4309498235781740, 26319, 18214, 39443802307
       Analyzed10, 2306436292671914913, 012710042391305368
      Clinically staged, surgically managed
       Total29, 15317,2489200217852736, 69718, 80713, 2533664973
       Analyzed10,0846416287368211312, 449702240491167113
      Clinically staged, nonsurgically managed
       Total949182298179290356637511417161334
       Analyzed1462053373656378190138157
      Pathologically staged
       Total26, 72212, 85710, 510278057536, 83014, 95415, 97347561147
       Analyzed22, 25711, 5598411210817930, 01813, 36812, 6283620402
      a Criteria for T descriptor analysis: cases must have known tumor size, at least one T descriptor supporting the assigned T category, and no T descriptors suggesting a higher T category.

      Statistical Analysis

      Survival was measured from the date of diagnosis for clinically staged patients and date of surgery for pathologically staged patients. Overall survival was assessed using the Kaplan–Meier method. Prognostic groups were assessed using Cox proportional hazards regression analysis.
      • Cox DR
      Regression models and life tables (with discussion).
      All survival and regression analyses were performed using SAS version 9.2.
      Tumor-size cutpoints were evaluated using a running log-rank statistics produced by each hypothetical cutpoint in the pN0M0R0 data set graphed against tumor size.
      • Crowley J
      • LeBlanc M
      • Jacobson J
      • Salmon S
      Some exploratory tools for survival analysis. Lecture Notes on Statistics.
      This was performed both to confirm the 7th edition T category cutpoints defined by size (T1a, b; T2a, b; and T3) and to identify possible additional size increments that could be useful. For evaluating possible new size cutpoints, the tumor size that coincided with the highest log-rank statistics, rounded to the nearest 1 cm, was chosen as the optimal cutpoint. The chosen cutpoint was then tested in the context of the 7th edition cutpoints in a multivariate Cox regression to assess their additional prognostic significance. In addition, the cutpoints were assessed in the N0M0 population regardless of resection completeness, in the M0R0 population regardless of nodal status, and in the clinically staged N0M0 population. R software was used to generate the log-rank statistics.
      • R Core Team
      T descriptors were evaluated individually among the population of cases where the given descriptor was evaluated along with at least one other descriptor within the given T category. To isolate the effects of each descriptor individually, cases with more than one positive descriptor within a T category were considered separately from those with only one positive descriptor. Descriptors identified for potential reclassification, on the basis of their respective survival outcome compared with other descriptors in the same or adjacent category, were then evaluated in a multivariate Cox regression analysis that adjusted for age, gender, histology, and geographic region. Specific comparisons were made to compare the survival of patients with a given descriptor against other cases within its category as defined by the 7th edition and against those in the proposed category. If a given descriptor was significantly different from others in the same 7th edition category, and similar to those in an adjacent category, it was considered to be evidence in support of the proposed change.

      Decisions on Recommendations

      The objective-based preliminary analyses of the new IASLC database were presented at the IASLC Staging and Prognostic Factors Committee General Meeting that took place in Sydney, Australia, on October 25 and 26, 2013. Most committee members attended this meeting. Further analyses on certain descriptors, such as visceral pleural invasion, and univariate and multivariate analyses, were suggested. Once these were completed, a core group of the committee membership directly involved with data analyses met with the IASLC statisticians at CRAB in Seattle on October 31, 2014. During this meeting the final recommendations were agreed upon discussion of the new results.

      RESULTS

      Tumor Size

      The running log-rank statistics showed that the 3-cm cutpoint is still valid to separate T1 from T2 tumors so classified exclusively according to tumor size, both in the pathological and clinical staging settings, but also when the population of patients with T2 tumors includes tumors classified by othet T2 descriptors other than size (Supplementary Fig. 1A and B, Supplemental Digital Content 2, http://links.lww.com/JTO/A835). These supplementary figures just show the 3-cm cutpoint, the best cutpoint for all sizes over all T categories.
      Figure thumbnail gr1
      FIGURE 1A, Survival of pathologically staged T1–T2 N0R0 tumors according to size only, at 1-cm intervals. B, Survival of clinically staged T1–T2 N0 tumors according to size only, at 1-cm intervals.
      When survival was analyzed by 1-cm increments in tumor size (⩽1 cm, >1 to 2 cm, >2 to 3 cm, >3 to 4 cm, >4 to 5 cm, >5 to 6 cm, >6 to 7 cm, and >7 cm), a progressive degradation of survival was observed for each 1-cm cutpoint. This was found not only in the population of patients with pT1-2 N0M0 and R0 tumors (Fig. 1A) but also in those with nodal involvement (Supplementary Fig. 2A, Supplemental Digital Content 2, http://links.lww.com/JTO/A835) and incomplete resections (Supplementary Fig. 2B, Supplemental Digital Content 2, http://links.lww.com/JTO/A835) and in those with clinically staged tumors with and without nodal involvement (Fig. 1B and Supplementary Fig. 2C, Supplemental Digital Content 2, http://links.lww.com/JTO/A835).
      Figure thumbnail gr2
      FIGURE 2A, Survival according to 7th edition and proposed T categories for pathologically staged T1–T4 N0M0R0 tumors. B, Survival according to 7th edition and proposed T categories for clinically staged T1–T4 N0M0 tumors.
      Univariate and multivariate analyses to further study the significance of pathological tumor size controlled for age, gender, cell type, and geographical region showed that survival was statistically significant for all tumor-size cutpoints. The 6-cm cutpoint did not add additional prognostic information after the other cutpoints were considered in a stepwise selection process (Table 2 and Supplementary Table 4, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). The same result was found in the univariate and multivariate analyses of clinical tumor size (Supplementary Tables 5 and 6, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). Further analyses to evaluate the new 1-cm cutpoints for pT1 tumors showed that they distinguish between risk groups (Table 3 and Supplementary Table 7, Supplemental Digital Content 1, http://links.lww.com/JTO/A834).
      TABLE 2Results of Univariate Analyses of Survival of Pathologically Staged T1-T3 N0M0R0 Cases According to Tumor Size and T2 and T3 Descriptors
      Survival from Surgery
      Variablen/N (%)HR (95% CI)P value
      Other histology vs. adeno7064/21, 122 (33)2.19 (2.07, 2.32)<0.001
      Squamous vs. other5237/21, 122 (25)1.96 (1.85, 2.07)<0.001
      Age ≥ 60 vs.<6016, 070/21, 014 (76)2.29 (2.11, 2.49)<0.001
      Male vs. female12, 457/20, 995 (59)1.86 (1.75, 1.98)<0.001
      Americas vs. Asia1873/21, 123 (9)1.79 (1.64, 1.97)<0.001
      Europe/Australia vs. Asia2361/21, 123 (11)2.61 (2.43, 2.80)<0.001
      Size >2 vs.≤2 cm12, 970/21, 123 (61)1.50 (1.39, 1.62)<0.001
      Size >3 vs.>2–3 cm7163/21, 123 (34)1.59 (1.47, 1.70)<0.001
      Size >5 vs.>3–5 cm1925/21, 123 (9)1.45 (1.31, 1.59)<0.001
      Size >7 vs.>5–7 cm606/21, 123 (3)1.45 (1.26, 1.67)<0.001
      Size >1 vs.≤1 cm19, 623/21, 122 (93)2.68 (2.28, 3.14)<0.001
      Size >4 vs.≤4 cm3669/21, 122 (17)2.43 (2.28, 2.58)<0.001
      Size >6 vs.≤6 cm1041/21, 122 (5)2.79 (2.55, 3.06)<0.001
      Multiple pT2 descriptors vs. other pT2, pT31817/9952 (18)1.17 (1.07, 1.27)<0.001
      pT3 vs. pT1–21882/21, 122 (9)2.63 (2.44, 2.83)<0.001
      pT2 main bronchus >2 cm vs. all others67/19, 013 (0)1.53 (0.98, 2.37)0.059
      pT3 main bronchus <2 cm vs. all others24/19, 013 (0)1.82 (0.91, 3.64)0.091
      pT2 atelectasis vs. all others161/11, 869 (1)1.98 (1.51, 2.61)<0.001
      pT3 atelectasis vs. all others8/11, 869 (0)3.06 (0.76, 12.24)0.114
      pT2 visceral pleura PL1 vs. PL02690/15, 685 (17)1.74 (1.60, 1.89)<0.001
      pT2 visceral pleura PL2 vs. PL0813/15, 685 (5)2.23 (1.97, 2.54)<0.001
      pT2 3–5 cm size only vs. pT1, pT2 ≤ 3 cm3320/21, 123 (16)1.79 (1.66, 1.93)<0.001
      pT2 3–5 cm plus other descriptor vs. pT1, pT2 ≤ 3 cm1362/21, 123 (6)2.22 (2.01, 2.46)<0.001
      pT2 5–7 cm size only vs. pT1, pT2 ≤ 3 cm586/21, 123 (3)2.59 (2.25, 2.99)<0.001
      pT2 5–7 cm plus other descriptor vs. pT1, pT2 ≤3 cm450/21, 123 (2)2.85 (2.46, 3.31)<0.001
      pT3 Single descriptor vs. pT1, pT2 ≤ 3 cm1556/21, 123 (7)3.20 (2.94, 3.49)<0.001
      pT3 Multiple pT3 descriptors vs. pT1, pT2 ≤ 3 cm326/21, 123 (2)4.27 (3.66, 4.99)<0.001
      The p value from Wald χ2 test in Cox Regression.
      HR, hazard ratio; 95% CI, 95% confidence interval; n, number with descriptor; N, number evaluated; %, percent with descriptor.
      TABLE 3Multivariate Survival Analyses of Proposed 1-cm Cutpoints in Pathologically Staged T1 Tumors
      Survival from Surgery
      Variablen/N (%)HR (95% CI)P Value
      Age ≥60 vs.<6012, 554/16, 644 (75)2.06 (1.87, 2.28)<0.001
      Americas vs. Asia1559/16, 644 (9)2.24 (2.01, 2.50)<0.001
      Europe/Australia vs. Asia1647/16, 644 (10)2.58 (2.36, 2.83)<0.001
      Male vs. female9371/16, 644 (56)1.70 (1.57, 1.83)<0.001
      Other histology vs. adeno4759/16, 644 (29)1.47 (1.31, 1.65)<0.001
      Squamous vs. other3473/16, 644 (21)0.98 (0.87, 1.10)0.685
      T1a >1–2 vs.<1 cm5462/16, 644 (33)1.45 (1.21, 1.74)<0.001
      T1b >2–3 vs.<1 cm4230/16, 644 (25)1.82 (1.52, 2.18)<0.001
      T2a <3 vs.<1 cm5611/16, 644 (34)2.43 (2.04, 2.90)<0.001
      Each size increment distinguishes between risk groups. A comparison of T2a less than 3 cm (T2a by descriptors other than size) versus larger T1 cases (T1b > 2–3 cm, not shown in table) indicates that T2a cases are appropriately in a higher risk category (p < 0.001).p value from Wald χ2 test in Cox regression.
      HR, hazard ratio; 95% CI, 95% confidence interval; n, number with descriptor; N, number evaluated, %=percent with descriptor.
      A comparison of T2a less than 3 cm (T2a by descriptor other than size, i.e., visceral pleura invasion) versus cases of similar size (T1b 2–3 cm) indicated that T2a cases are appropriately in a higher risk category (p < 0.001; Table 3). The same is true for clinically staged T1 tumors (Supplementary Table 7, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). However, pathologically and clinically staged tumors greater than 5 cm but equal to or less than 7 cm aligned better with a T3 prognosis than with a T2b (Table 4 and Supplementary Table 8, Supplemental Digital Content 1), and tumors classified as T3 by size greater than 7 cm had a survival similar to that of T4 tumors (Supplementary Tables 9–12, Supplemental Digital Content 1, http://links.lww.com/JTO/A834).
      TABLE 4Survival Comparisons of Pathologically Staged T2-T4 Tumors >4–5 cm, >5–7 cm, and >7 cm in Greatest Dimension
      Survival from Surgery
      Variablen/N (%)HR (95% CI)P Value
      UnivariateOther histology vs. adeno4357/10, 028 (43)1.61 (1.50, 1.73)<0.001
      Squamous vs. other3318/10, 028 (33)1.45 (1.35, 1.56)<0.001
      Age ≥60 vs.<607934/9987 (79)1.94 (1.76, 2.15)<0.001
      Male vs. female6599/9967 (66)1.53 (1.41, 1.65)<0.001
      Americas vs. Asia762/10, 028 (8)1.24 (1.09, 1.42)0.001
      Europe/Australia vs. Asia1439/10, 028 (14)1.90 (1.74, 2.07)<0.001
      Proposed T2b 4–5 cm vs. all others1480/10, 028 (15)1.10 (1.00, 1.21)0.046
      Proposed T3 5–7 cm vs. all others1417/10, 028 (14)1.48 (1.35, 1.62)<0.001
      Other T3 (excluding >7 cm) vs. all others828/10, 028 (8)1.30 (1.16, 1.46)<0.001
      Proposed T4 (including T3 > 7 cm) vs. all others761/10, 028 (8)2.14 (1.92, 2.38)<0.001
      MultivariateOther histology vs. adeno4312/9940 (43)1.28 (1.14, 1.43)<0.001
      Squamous vs. other3281/9940 (33)0.92 (0.82, 1.03)0.165
      Age ≥60 vs.<607891/9940 (79)1.95 (1.76, 2.16)<0.001
      Male vs. female6581/9940 (66)1.46 (1.35, 1.59)<0.001
      Americas vs. Asia761/9940 (8)1.45 (1.27, 1.66)<0.001
      Europe/Australia vs. Asia1428/9940 (14)1.82 (1.66, 1.99)<0.001
      Proposed T2b 4–5 cm vs. T2 3–4 cm1467/9940 (15)1.27 (1.15, 1.41)<0.001
      Proposed T3 5–7 cm vs. T2 3–4 cm1409/9940 (14)1.59 (1.44, 1.76)<0.001
      Other T3 (excluding >7 cm) vs. T2 3–4 cm821/9940 (8)1.62 (1.43, 1.83)<0.001
      Proposed T4 (Including T3>7 cm) vs. T2 3–4 cm757/9940 (8)2.24 (2.00, 2.52)<0.001
      Specific comparisons not shown in table: when survival of tumors greater than 5 to 7 cm is compared with that of tumors greater than 4 to 5 cm, the p value is 0.0002, indicating survival is significantly different for these groups. When survival of T3 tumors (excluding those >7 cm) is compared with that of tumors greater than 5 to 7 cm, the p value is 0.821, indicating survival is similar between these groups; p value from Wald χ2 test in Cox regression.
      HR, hazard ratio; 95% CI, 95% confidence interval; n, number with descriptor; N, number evaluated; %, percent with descriptor.

      Involvement of the Main Bronchus

      Involvement of the main bronchus 2 cm or more from the carina is well aligned as a T2 descriptor in all studied populations with pathologically staged tumors (N0M0R0, any N and any R) and in the populations with clinically staged tumors (N0 and any N). There are no statistically significant differences among survival of this T2 descriptor when it is compared with that of other T2 descriptors (Supplementary Fig. 3AE, Supplemental Digital Content 2, http://links.lww.com/JTO/A835). On the other hand, involvement of the main bronchus less than 2 cm from the carina, without invasion of the carina, a present T3 descriptor, has better prognosis than other T3 descriptors in all studied populations (Supplementary Fig. 4AE, Supplemental Digital Content 2, http://links.lww.com/JTO/A835). When the prognosis of T2 and T3 so defined by involvement of the main bronchus are compared, their prognosis is similar, and that of T3 main bronchus is better than prognosis of other T3 descriptors (Supplementary Fig. 5AE, Supplemental Digital Content 2, http://links.lww.com/JTO/A835). Multivariate analyses showed that, in pathologically and clinically staged tumors, involvement of main bronchus, regardless of distance to carina, does not seem to increase risk after adjusting for tumor size. This result supports the idea that T2 main bronchus is similar to other T2 cases, and that T3 main bronchus does not show significant increased risk over T2 (Table 5 and Supplementary Table 13, Supplemental Digital Content 1, http://links.lww.com/JTO/A834).
      TABLE 5Multivariate Survival Analyses of Pathologically Staged pT2–3 Tumors Based on Their Endobronchial Location
      Multivariate Results VariableSurvival from Surgery
      n/N (%)HR (95% CI)P Value
      Other histology vs. adenocarcinoma3725/8807 (42)1.42 (1.26, 1.60)<0.001
      Squamous vs. other2868/8807 (33)0.88 (0.78, 1.00)0.045
      Age ≥ 60 vs.<607031/8807 (80)1.96 (1.76, 2.20)<0.001
      Male vs. female5807/8807 (66)1.45 (1.33, 1.58)<0.001
      Americas vs. Asia234/8807 (3)1.74 (1.39, 2.18)<0.001
      Europe vs. Asia1031/8807 (12)1.98 (1.78, 2.21)<0.001
      Size >2 vs.≤ 2 cm7640/8807 (87)1.28 (1.09, 1.50)0.002
      Size >3 vs.2 to ≤3 cm6230/8807 (71)1.09 (0.97, 1.22)0.133
      Size >5 vs.3 to ≤5 cm1571/8807 (18)1.33 (1.20, 1.48)<0.001
      Size >7 vs.5 to ≤7 cm467/8807 (5)0.99 (0.83, 1.19)0.953
      pT2 main bronchus >2 cm from carina vs. pT2 without invasion67/8807 (1)1.08 (0.69, 1.69)0.725
      pT3 main bronchus <2 cm from carina vs. pT2 without invasion24/8807 (0)1.03 (0.51, 2.06)0.937
      pT3 other than main bronchus vs. pT2, pT3 with invasion of main bronchus1304/8807 (15)1.56 (1.39, 1.76)<0.001
      p value from Wald χ2 test in Cox regression.
      HR, hazard ratio; 95% CI, 95% confidence interval; n, number with descriptor; N, number evaluated; %, percent with descriptor.

      Atelectasis/Pneumonitis

      Partial atelectasis/pneumonitis is well aligned with other T2 descriptors both in the pathological and in the clinical settings. Five-year survival for those patients with T2 tumors so defined by partial atelectasis/pneumonitis, only, and N0M0R0, for those with other T2 descriptors, and for those with T2 tumors by size only were 72%, 70%, and 70%, respectively. Similar survival rates for the three groups of T2 tumors were found in patients with any N and any R tumors and in the population of patients with cT2 N0 and any N tumors. Total atelectasis/pneumonitis, a T3 descriptor, showed better prognosis than other T3 tumors with different descriptors, but there were seven cases only in the pathological setting, which precluded further analyses (Supplementary Table 14, Supplemental Digital Content 1, http://links.lww.com/JTO/A834).

      Visceral Pleural Invasion

      Visceral pleural invasion is well positioned as a T2 descriptor and confers a worse prognosis even after adjusting for the current tumor size cutpoints (Supplementary Table 15, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). The extent of the visceral pleura invasion as currently defined (PL0, tumor within the subpleural lung parenchyma or invades superficially into the pleural connective tissue beneath the elastic layer; PL1, tumor invades beyond the elastic layer of the visceral pleura; and PL2, tumor invades to the visceral pleura surface)
      • Travis WD
      • Brambilla E
      • Rami-Porta R
      • et al.
      International Staging Committee. Visceral pleural invasion: pathologic criteria and use of elastic stains: proposal for the 7th edition of the TNM classification for lung cancer.
      appropriately distinguishes between risk groups, and although the prognosis of PL1 and PL2 is worse than that of PL0, there are also significant differences between PL1 and PL2, the latter having a worse prognosis (Supplementary Table 16, Supple mental Digital Content 1, http://links.lww.com/JTO/A834). The increased risk associated with visceral pleura invasion is also found in the clinical staging setting (Supplementary Table 17, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). Further analyses in pathologically and clinically staged tumors show that pathologically staged tumor of greater than 3–4 cm with visceral pleura invasion has similar prognosis to that of those greater than 4–5 cm; and that tumors of greater than 4–5 cm with visceral pleura invasion have similar prognosis to that of those greater than 5–7 cm (Table 6). However, these differences are not so clear in the clinically staged patients (Supplementary Table 18, Supplemental Digital Content 1, http://links.lww.com/JTO/A834).
      TABLE 6Univariate and Multivariate Survival Analyses for Pathologically Staged T1, T2, and T3 Tumors Based on the Status of the Visceral Pleura for the Purpose of Assessing Upstaging Based on Visceral Pleura Invasion (Proposed Size Cutpoints for the 8th Edition)
      Survival from Surgery
      Variablen/N (%)HR (95% CI)P value
      UnivariateOther histology vs. adeno7020/21, 007 (33)2.20 (2.08, 2.33)<0.001
      Age ≥ 6015, 970/20, 899 (76)2.28 (2.10, 2.47)<0.001
      Male12, 380/20, 880 (59)1.87 (1.76, 1.99)<0.001
      Americas1871/21, 007 (9)1.55 (1.41, 1.69)<0.001
      Europe/Australia2358/21, 007 (11)2.47 (2.30, 2.65)<0.001
      T1b vs. T1a4652/20, 372 (23)0.47 (0.44, 0.51)<0.001
      T1c vs. T1a4247/20, 372 (21)0.81 (0.75, 0.87)<0.001
      T2a VPI, 0.1–1 cm vs. T1a25/20, 372 (0)0.17 (0.02, 1.20)0.075
      T2a VPI, 1–2 cm vs. T1a692/20, 372 (3)0.81 (0.68, 0.96)0.016
      T2a VPI, 2–3 cm vs. T1a1273/20, 372 (6)1.07 (0.96, 1.20)0.218
      T2a size only vs. T1a2405/20, 372 (12)1.22 (1.12, 1.33)<0.001
      T2a VPI, 3–4 cm vs. T1a1147/20, 372 (6)1.58 (1.42, 1.76)<0.001
      T2b size only vs. T1a1047/20, 372 (5)1.65 (1.48, 1.84)<0.001
      T2b VPI, 4–5 cm vs. T1a433/20, 372 (2)1.93 (1.64, 2.25)<0.001
      T3 vs. T1a2245/20, 372 (11)2.36 (2.19, 2.54)<0.001
      MultivariateOther histology vs. adeno6558/20, 163 (33)1.36 (1.28, 1.45)<0.001
      Age ≥ 6015, 395/20, 163 (76)2.03 (1.87, 2.22)<0.001
      Male11, 862/20, 163 (59)1.64 (1.54, 1.76)<0.001
      Americas1806/20, 163 (9)2.03 (1.85, 2.24)<0.001
      Europe/Australia2174/20, 163 (11)2.25 (2.09, 2.44)<0.001
      T1b vs. T1a4603/20, 163 (23)0.91 (0.80, 1.04)0.173
      T1c vs. T1a4213/20, 163 (21)1.28 (1.12, 1.45)<0.001
      T2a VPI, 0.1–1 cm vs. T1a25/20, 163 (0)0.26 (0.04, 1.87)0.182
      T2a VPI, 1–2 cm vs. T1a690/20, 163 (3)1.42 (1.16, 1.73)<0.001
      T2a VPI, 2–3 cm vs. T1a1262/20, 163 (6)1.80 (1.54, 2.10)<0.001
      T2a size only vs. T1a2378/20, 163 (12)1.58 (1.38, 1.81)<0.001
      T2a VPI, 3–4 cm vs. T1a1131/20, 163 (6)2.18 (1.88, 2.53)<0.001
      T2b size only vs. T1a1035/20, 163 (5)1.92 (1.65, 2.24)<0.001
      T2b VPI, 4–5 cm vs. T1a432/20, 163 (2)2.54 (2.10, 3.07)<0.001
      T3 vs. T1a2230/20, 163 (11)2.66 (2.34, 3.03)<0.001
      Specific comparisons not shown in table: T2a greater than 2 to 3 cm VPI versus T2a size only, p = 0.0661; T2a appropriate as proposed. T2a greater than 3 to 4 cm VPI versus T2a size only, p < 0.0001; support for upstaging to T2b. T2a greater than 3 to 4 cm VPI versus T2b size only, p = 0.0945; support for upstaging to T2b. T2b greater than 4 to 5 cm VPI versus T2b size only, p = 0.0036; support for upstaging to T3.T2b greater than 4 to 5 cm VPI versus T3, p = 0.5761; support for upstaging to T3.T2a 1 to 2 cm VPI versus T2a size only, p = 0.2399; T2a appropriate as proposed.p value from Wald χ2 test in Cox regression.
      HR, hazard ratio, 95% CI, 95% confidence interval.

      Diaphragm

      When survival of patients with tumors involving the diaphragm, a present T3 descriptor, is compared with survival of patients with other T3 tumors defined by other descriptors, it has worse prognosis, both in the pathological and clinical settings. This is confirmed by the multivariate analyses. Patients with pT3 tumors by diaphragm involvement have worse prognosis than those with pT3 tumors by other descriptors (p = 0.004) and even those with pT4 tumors (p = 0.02; Supplementary Table 19, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). Clinically staged tumors with diaphragm involvement have similar prognosis to those clinically classified as T4 (p = 0.09) and to those clinically classified as T3 (p = 0.121). (Supplementary Table 20, Supplemental Digital Content 1, http://links.lww.com/JTO/A834) The hazard ratio, however, is in the same direction as the pathological analyses.

      Other T3 and T4 Descriptors

      T3 descriptors parietal pericardium, mediastinal pleura, chest wall invasion, including Pancoast tumors and parietal pleural invasion, and additional tumor nodules in the same lobe of the primary tumor did not differ in prognosis compared with other T3 tumors. In the subgroup of patients with tumors invading the chest wall, there were no differences in survival between those with parietal pleural invasion (163 patients with tumors classified as pT3 by parietal pleura invasion N0M0R0, with a 56% 5-year survival rate) and those with more extensive chest wall involvement (405 patients with tumor classified as pT3 by chest wall invasion N0M0R0, with a 52% 5-year survival rate). Similar survivals were found for the pathologically staged tumors with any N and any R and in the clinically staged tumors with N0 and any N. Tumors clinically staged T3 by mediastinal pleura involvement tended to have better prognosis than other T3 tumors, but there were only 20 tumors so classified in the present database (data not shown). However, at pathological staging, mediastinal pleura invasion seems to have worse prognosis than other pT3 tumors (Supplementary Table 9, Supplemental Digital Content 1, http://links.lww.com/JTO/A834).
      A thorough analysis of the individual T4 descriptors was not possible because of the small number of patients in each group. However, from the survival graphs that could be developed, involvement of the mediastinum and of the vertebral body seems to be well aligned within the T4 descriptors; invasion of great vessels tended to have a slightly better prognosis than the other T4 tumors; and invasion of the heart tended to have the worst (data not shown). In the selected population of patients with completely resected T4 tumors with no nodal metastases, survival is very similar to that of patients with completely resected T3N0 tumors.
      Additional tumor nodules in a different ipsilateral lobe seem to have a slightly worse prognosis than other T4 descriptors, but the limited number of patients precludes meaningful analyses. Although having more than one additional tumor nodule in the same side seems to have worse prognosis than having only one, there are too few patients in each group to draw solid conclusions (data not shown).

      Single versus Multiple Descriptors for a T Category

      Multiple analyses were conducted to compare tumors in a given T category defined by either single or multiple T descriptors. Results were inconclusive, and sometimes they differed in the clinical and pathological settings. As an example, Supplementary Tables 21 and 22 (Supplemental Digital Content 1, http://links.lww.com/JTO/A834) show that, although in the pathological setting current pT2a tumors (>3 to 5 cm) with multiple T2 descriptors might be upstaged to pT2b, this could not be reproduced in the clinical setting. Therefore, no meaningful conclusions could be drawn from these analyses.

      Rearrangement of Descriptors

      On the basis of the results described above, the following rearrangement of descriptors was done for exploratory analyses: T1 tumors were subdivided into three groups at 1 cm cutpoints (⩽1 cm; >1 and ⩽2 cm; and >2 and ⩽3 cm); T2 tumors were subdivided into two subgroups (>3 and ⩽4 cm; >4 and ⩽5 cm); T2 tumors greater than 5 and less than or equal to 7 cm were reclassified as T3; T3 tumors greater than 7 cm were reclassified as T4; T2 and T3 tumors so classified by endobronchial location were combined as T2; and invasion of the diaphragm was reclassified as T4. When the survival of patients with tumors classified with the new descriptors was studied, survival curves separated nicely with no crossing over or superposition. This occurred in the three populations with pathologically staged tumors (N0M0R0, any N, and any R) and in those with clinically staged tumors (N0M0 and any NM0). (Fig. 2A and B and Supplementary Fig. 6AC, Supplemental Digital Content 2, http://links.lww.com/JTO/A835.) In addition, all survival comparisons were statistically significant, including the differences between T3 and T4, which are not significant in the current TNM classification (Tables 7 and 8).
      TABLE 7Survival Comparisons of Pathologically Staged Tumors According to the T Categories of the 7th Edition and to the Proposed T Categories for the 8th Edition
      7th Edition CategoriesProposed Categories
      ContrastEstimateLower LimitUpper LimitP ValueContrastEstimateLower LimitUpper LimitP Value
      T1a vs. T1b1.35851.23531.4940<0.0001T1a vs. T1b1.48991.23401.7988<0.0001
      T1b vs. T2a1.42921.31621.5520<0.0001T1b vs. T1c1.27671.15681.4090<0.0001
      T2a vs. T2b1.25201.11911.4007<0.0001T1c vs. T2a1.36471.25191.4878<0.0001
      T2b vs. T31.44861.28071.6384<0.0001T2a vs. T2b1.22181.10221.35430.0001
      T3 vs. T41.00450.76071.32640.9747T2b vs. T31.28951.15531.4392<0.0001
      T3 vs. T41.29971.14581.4742<0.0001
      TABLE 8Survival Comparisons of Clinically Staged Tumors According to the T Categories of the 7th Edition and to the Proposed T Categories for the 8th Edition
      7th Edition CategoriesProposed Categories
      ContrastEstimateLower LimitUpper LimitP valueContrastEstimateLower LimitUpper LimitP value
      T1a vs. T1b1.55341.38441.7430<0.0001T1a vs. T1b1.83801.42742.3668<0.0001
      T1b vs. T2a1.35181.21261.5070<0.0001T1b vs. T1c1.41651.25801.5949<0.0001
      T2a vs. T2b1.44651.22021.7149<0.0001T1c vs. T2a1.29671.15431.4567<0.0001
      T2b vs. T31.28041.06131.54490.0098T2a vs. T2b1.20381.03091.40560.0190
      T3 vs. T40.88510.67261.16480.3836T2b vs. T31.30311.09961.54430.0022
      T3 vs. T41.45421.22211.7305<0.0001

      DISCUSSION

      The analyses of tumor size in the new IASLC database provided solid ground not only to further subclassify tumors 3 cm or less in size (present T1 category) and those greater than 3 cm (present T2 category), but also to distribute tumor size as a descriptor of all T categories. The survival analyses according to 1-cm cutpoints showed that from 1 to 5 cm every centimeter counts, and that larger tumors are best aligned with either T3 (tumor size of more than 5 to 7 cm) or T4 (tumor size of more than 7 cm). This finding further confirms the common intuition that the larger the tumor, the worse the prognosis. This study confirms that, although the 3-cm cutpoint still remains a landmark to separate T1 from T2 tumors, there can be tumors less than 2 cm in greatest dimension with significantly different prognosis, and that the 5-cm cutpoint remains a useful one to separate tumors of different prognosis. The fact that tumors of 1 cm or less in greatest dimension are significantly different from larger ones is important in the light of the results of screening programs. In screening programs, 60% to 70% of detected lung cancers are in stage I,
      • The National Lung Screening Trial Research Team
      Reduced lung-cancer mortality with low-dose computed tomography screening.
      ,
      • Horeweg N
      • van der Aalst CM
      • Thunnissen E
      • et al.
      Characteristics of lung cancers detected by computer tomography screening in the randomized NELSON trial.
      and 56% are 1 cm or less in size.
      • Henschke CI
      • McCauley DI
      • Yankelevitz DF
      • et al.
      Early Lung Cancer Action Project: overall design and findings from baseline screening.
      These small tumors could constitute a particular group worthy of further studies regarding growth rate, tumor density (solid, part-solid, or pure ground glass opacity), intensity of uptake in positron emission tomography, type of resection, alternative nonsurgical therapies, molecular characterization, and genetic signatures. The cutpoints found in this study work well in each studied population in the clinical and pathological staging settings, represent a logical degradation of survival as tumor size increases, will be relevant for tumor stratification in future clinical trials, will allow the refinement of prognosis, and can be easily applied in clinical practice, while keeping compatibility with the size descriptors of the 7th edition.
      The analyses of this database could not address how to measure the size of part-solid adenocarcinomas with a lepidic component. A Subcommittee within the IASLC Staging and Prognostic Factors Committee was created to address this issue and give recommendations in a white paper that is being prepared at the time of this writing. In the meantime, the Union for International Cancer Control recommendation is to measure the invasive component of the tumor to define its T category.
      • Wittekind Ch
      • Compton CC
      • Brierley J
      • Sobin LH
      There is evidence that this measurement better predicts prognosis than the overall tumor size in lepidic predominant tumors.
      • 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.
      • Tsutani Y
      • Miyata Y
      • Nakayama H
      • et al.
      Prognostic significance of using solid versus whole tumor size on high-resolution computed tomography for predicting pathologic malignant grade of tumors in clinical stage IA lung adenocarcinoma: a multicenter study.
      • Maeyashiki T
      • Suzuki K
      • Hattori A
      • Matsunaga T
      • Takamochi K
      • Oh S
      The size of consolidation on thin-section computed tomography is a better predictor of survival than the maximum tumour dimension in resectable lung cancer.
      In the 7th edition of the TNM classification, involvement of the main bronchus is classified as T2 if it is located 2 cm or more from the carina and as T3 if it is less than 2 cm from the carina but without its invasion. This descriptor could not be studied reliably for the 7th edition of the TNM classification, but data from the new IASLC database revealed that endobronchial tumors either less than 2 or greater than 2 cm from the carina have the same prognosis in the clinical and pathological staging. This simplifies classification as this T3 descriptor can be merged with the T2 descriptor to form a single T2 descriptor in the 8th edition of the classification.
      One of the consequences of endobronchial tumor involvement is atelectasis and pneumonitis. In the 7th edition of the TNM classification, atelectasis/pneumonitis is a T2 descriptor, if it does not involve the whole lung, and a T3 descriptor, if it involves the whole lung. The present analysis has found that partial atelectasis/pneumonitis is well aligned with other T2 descriptors, but that total atelectasis/pneumonitis has better survival than other T3 descriptors. Although the patients with total atelectasis/pneumonitis are few and it is difficult to draw data-based conclusion, it seems logical to group total atelectasis/pneumonitis in the T2 category as it surely follows the same prognostic pattern seen for endobronchial location. Atelectasis/pneumonitis by itself has no prognostic value; it is what it represents, i.e., endobronchial growth, what is prognostic. While discussing this issue at the Staging and Prognostic Factors Committee meeting, the question of eliminating partial and total atelectasis/pneumonitis from the list of T descriptors arose. Reasons for the elimination were that there are few tumors classified with this descriptor and that the pathologists examining the resected specimen do not see the atelectasis because the lungs are deflated, and then, this descriptor cannot be used in pathological staging. It is true that there are few patients with completely resected tumors and no nodal involvement classified as T2-partial atelectasis/pneumonitis (95 patients) or T3-total atelectasis/pneumonitis (5 patients). However, the numbers rise when we consider tumors with nodal involvement (156 patients with pT2 any N, 100 with pT2N0 any R, and 25 with cT3 any N). For patients with no surgical option who will not need any further explorations, atelectasis/pneumonitis seen on chest radiography or computed tomography may be the only way to assign a T category to the tumor, avoiding the need of positron emission tomography or bronchoscopy, if these are not essential for treatment. It is also important to have in mind that pathological classification does not only derive from the pathological study performed by the pathologist, but from the information collected before resection. The letter of the second general rule of the TNM classification explicitly states that pathological classification “is based on evidence acquired before treatment, supplemented or modified by additional evidence acquired from surgery and from pathological examination.”
      • Edge SB
      • Byrd DR
      • Compton CC
      • Fritz AG
      • Greene FL
      • Trotti III, A
      ,
      • Goldstraw P
      ,
      • Sobin L
      • Gospodarowocz M
      • Wittekind C
      Therefore, a pT category based on the partial or total atelectasis/pneumonitis observed at clinical staging can be assigned to a tumor even if the pathologist cannot recognize it.
      Because the publication of the recommended definition of visceral pleura invasion in 2008,
      • Travis WD
      • Brambilla E
      • Rami-Porta R
      • et al.
      International Staging Committee. Visceral pleural invasion: pathologic criteria and use of elastic stains: proposal for the 7th edition of the TNM classification for lung cancer.
      many groups have reviewed their experiences and published their results. Yoshida et al.,
      • Yoshida J
      • Nagai K
      • Asamura H
      • et al.
      Japanese Joint Committee for Lung Cancer Registration. Visceral pleura invasion impact on non-small cell lung cancer patient survival: its implications for the forthcoming TNM staging based on a large-scale nation-wide database.
      reporting on an experience of 9758 patients who had undergone anatomical resection and of whom 2350 had visceral pleural invasion, concluded that, for tumors 7 cm or less in greatest dimension, the T category should be upstaged to the next T category if visceral pleura invasion (VPI) was present. Analyzing smaller series, others have found that T2a tumors with VPI had worse prognosis than other T2a tumors so classified by tumor size only and recommended to upstage them to the T2b category.
      • Kudo Y
      • Saji H
      • Shimada Y
      • et al.
      Impact of visceral pleural invasion on the survival of patients with non-small cell lung cancer.
      • Fibla JJ
      • Cassivi SD
      • Brunelli A
      • et al.
      Re-evaluation of the prognostic value of visceral pleura invasion in Stage IB non-small cell lung cancer using the prospective multicenter ACOSOG Z0030 trial data set.
      • Kawase A
      • Yoshida J
      • Miyaoka E
      • et al.
      Japanese Joint Committee of Lung Cancer Registry. Visceral pleural invasion classification in non-small-cell lung cancer in the 7th edition of the tumor, node, metastasis classification for lung cancer: validation analysis based on a large-scale nationwide database.
      Shim et al.
      • Shim HS
      • Park IK
      • Lee CY
      • Chung KY
      Prognostic significance of visceral pleural invasion in the forthcoming (seventh) edition of TNM classification for lung cancer.
      found that pathological T2b tumors by tumor size greater than 5 cm but not greater than 7 cm with VPI should be upstaged to T3. Maeda et al.
      • Maeda R
      • Yoshida J
      • Ishii G
      • Hishida T
      • Nishimura M
      • Nagai K
      Risk factors for tumor recurrence in patients with early-stage (stage I and II) non-small cell lung cancer: patient selection criteria for adjuvant chemotherapy according to the seventh edition TNM classification.
      identified VPI as a risk factor for recurrence in completely resected stage I and II tumors, whereas Nitadori et al.
      • Nitadori J
      • Colovos C
      • Kadota K
      • et al.
      Visceral pleural invasion does not affect recurrence or overall survival among patients with lung adenocarcinoma ⩽ 2 cm: a proposal to reclassify T1 lung adenocarcinoma.
      did not find VPI to be a risk factor for increased recurrence or reduced overall survival in adenocarcinomas 2 cm or less in greatest dimension and proposed to combine tumors of 2 cm or less with or without VPI with those of more than 2 but 3 cm or less without VPI into a new stage IA. The present analyses have shown that the presence of VPI confers worse prognosis even when adjusting by tumor size, and that there is no need to modify the present definition of visceral pleura invasion and its different categories (PL0, PL1, and PL2) as this study confirms that visceral pleura involvement is well positioned in the T2 category. The two categories of visceral pleura invasion (PL1 and PL2) are justified as there are statistically significant differences between them. Detailed analyses of VPI in the pathological staging setting show that tumors greater than 3 to 4 cm with VPI have a similar prognosis to those greater than 4 to 5 cm; and that those greater than 4 to 5 cm with VPI had a similar prognosis to those greater than 5 to 7 cm. These findings could be used to upstage tumors with VPI to the next tumor size category, but they could not be replicated clearly in the clinical staging setting, indicating that the clinical assessment of VPI is unreliable. VPI mainly is a pathological descriptor. At clinical staging, it can be assumed by the proximity of the tumor to the lung surface and by its retraction, and although it could be pathologically confirmed (by tru-cut biopsy, by thoracoscopic biopsy, or by wedge resection of the tumor mass), this confirmation is not often done, as it is unnecessary to plan therapy. The same is true for the distinction of PL1 and PL2. Even if this differentiation could be used to upstage tumors in case of PL2, this is a pathological finding that can be known at clinical staging only in exceptional cases. Therefore, the IASLC Staging and Prognostic Factors Committee members, although recognizing the prognostic value of VPI and of its different categories, decided not to modify its present position as a T2 descriptor, and not to use it to upstage tumors. However, they do recognize its value in the construction of postoperative prognostic groups, a new project of the IASLC Staging and Prognostic Factors Committee. VPI is an important prognostic factor that has been consolidated in the analyses of the present database. Therefore, the recommendation to search it with elastic stains when it is not evident or is inconclusive with hematoxylin and eosin stains is valid and should be emphasized.
      • Travis WD
      • Brambilla E
      • Rami-Porta R
      • et al.
      International Staging Committee. Visceral pleural invasion: pathologic criteria and use of elastic stains: proposal for the 7th edition of the TNM classification for lung cancer.
      Invasion of the diaphragm is a different issue. Reported series, usually with small number of patients, have shown that invasion of the diaphragm has a bad prognosis. Five-year survival for completely resected tumors ranges from 0% to 30%.
      • Padilla J
      • Calvo V
      • Morcillo A
      • García Zarza A
      • Blasco E
      • Pastor J
      [Resection of bronchogenic carcinoma invading the diaphragm].
      • Rocco G
      • Rendina EA
      • Meroni A
      • et al.
      Prognostic factors after surgical treatment of lung cancer invading the diaphragm.
      • Yokoi K
      • Tsuchiya R
      • Mori T
      • et al.
      Results of surgical treatment of lung cancer involving the diaphragm.
      • Galetta D
      • Borri A
      • Casiraghi M
      • et al.
      Outcome and prognostic factors of resected non-small-cell lung cancer invading the diaphragm.
      Nodal involvement, full-depth invasion, and primary repair, when compared with prosthetic repair, have been found to adversely affect prognosis.
      • Rocco G
      • Rendina EA
      • Meroni A
      • et al.
      Prognostic factors after surgical treatment of lung cancer invading the diaphragm.
      • Yokoi K
      • Tsuchiya R
      • Mori T
      • et al.
      Results of surgical treatment of lung cancer involving the diaphragm.
      • Galetta D
      • Borri A
      • Casiraghi M
      • et al.
      Outcome and prognostic factors of resected non-small-cell lung cancer invading the diaphragm.
      In this study, it has been confirmed that invasion of the diaphragm has a worse prognosis than that assigned to other T3 descriptors, but that prognosis is similar to that of T4 and even worse for those clinically staged. Present data thus support the upstaging of this descriptor to the T4 category.
      Mediastinal pleura invasion is seldom used as a descriptor. It is difficult to determine clinically, and this may account for its tendency to have a better prognosis than other T3 descriptors at clinical staging. Contrary to parietal pleura invasion, mediastinal pleural invasion is not associated with pain by itself. It can be assumed if the tumor is in contact with the mediastinum, but when there are more signs of mediastinal invasion, the tumor has already gone beyond the mediastinal pleura and invaded mediastinal tissue, a T4 descriptor. This may explain its worse prognosis when compared with other pT3 descriptors. (Supplementary Table 9, Supplemental Digital Content 1, http://links.lww.com/JTO/A834). At pathological staging, it is exceptional to find mediastinal pleura invasion with no more invasion into the mediastinal tissue. Therefore, the IASLC Staging and Prognostic Factors Committee members favor the elimination of mediastinal pleura invasion as a T descriptor.
      The analyses of the new IASLC database share some of the limitations of the first retrospective database used for the 7th edition of the TNM classification. The most important limitation is that many contributing databases were not designed to study the TNM classification and therefore lacked the detailed information needed for the specific study of each descriptor. Although tumor size was regularly recorded in all databases, the other specific descriptors to define the T categories were not recorded, and this lack of information prevented the validation of many T3 and T4 descriptors. This is probably the reason why the analyses on the impact on prognosis of tumors classified by multiple descriptors, when compared with those classified by one descriptor, only, could not lead to solid conclusions. However, there was enough information to study endobronchial location, atelectasis/pneumonitis, visceral pleura invasion, and diaphragm invasion to the extent of allowing recommendations for changes in the 8th edition and improving the capacity to separate groups of tumors with significantly different prognosis based on their anatomical extent. When these changes are incorporated as new descriptors and tested for survival, the resulting survival graphs are better separated, and the differences in prognosis are more significant than those observed in the 7th edition of the TNM classification. An important result of this rearrangement is that survival for T3 and T4 tumors is now different, whereas it was not in the 7th edition. (Fig. 2A and B; Supplementary Fig. 6AC, Supplemental Digital Content 2, http://links.lww.com/JTO/A835; and Tables 7 and 8). It is important to note that there were insufficient patients treated by radiotherapy and chemotherapy alone in the database to determine the generalizability of the new recommendations across nonsurgical treatment modalities, as the prognostic impact of the T descriptors may differ depending on the therapy applied.
      • Ball D
      • Mitchell A
      • Giroux D
      • Rami-Porta RIASLC Staging Committee and Participating Institutions
      Effect of tumor size on prognosis in patients treated with radical radiotherapy or chemoradiotherapy for non-small cell lung cancer. An analysis of the staging project database of the International Association for the Study of Lung Cancer.
      In addition, this database had a predominance of Asian patients, as opposed to the previous database used to inform the 7th edition that had a predominance of European cases. However, the multivariate analysis performed in this occasion was adjusted for geographical region to compensate for this geographical unbalance. The present database has no information on the epidermal growth factor receptor mutation status of the registered patients with adenocarcinoma. Therefore, its prognostic impact could not be assessed in the present analyses.
      As it was the case after the 7th edition was published, many specialists managing lung cancer patients used the changes in the classification to modify therapy. If the proposed IASLC recommendations are eventually introduced in the 8th edition of the TNM classification, they should not be interpreted as basis for changing treatment. They imply a taxonomic refinement and not new indications of already established treatment protocols that should ideally be derived from clinical trials.
      • Boffa DJ
      • Detterbeck FC
      • Smith EJ
      • et al.
      Should the 7th edition of the lung cancer stage classification system change treatment algorithms in non-small cell lung cancer?.
      ,
      • Boffa DJ
      • Greene FL
      Reacting to changes in staging designations in the 7th edition of the AJCC staging manual.
      So, for the T component, upstaging invasion of the diaphragm or tumors greater than 7 cm from T3 to T4 does not imply that these tumors should not be resected if they are amenable to complete resection.
      In conclusion, based on the results of the analyses of the new IASLC database, the IASLC Staging and Prognostic Factors Committee recommends the following changes in the T component for the 8th edition of the TNM classification of lung cancer:
      • 1.
        The subclassification of T1 into
        • T1a:
          tumor 1 cm or less in greatest dimension,
        • T1b:
          tumor more than 1 cm but not more than 2 cm in greatest dimension, and
        • T1c:
          tumor more than 2 cm but not more than 3 cm in greatest dimension;
      • 2.
        The subclassification of T2 into
        • T2a:
          tumor more than 3 cm but not more than 4 cm in greatest dimension and
        • T2b:
          tumor more than 4 cm but not more than 5 cm in greatest dimension;
      • 3.
        The reclassification of tumors more than 5 cm but not more than 7 cm in greatest dimension as T3;
      • 4.
        The reclassification of tumors more than 7 cm in greatest dimension as T4;
      • 5.
        The grouping of the involvement of the main bronchus as a T2 descriptor, regardless of distance from the carina, but without invasion of the carina;
      • 6.
        The grouping of partial and total atelectasis or pneumonitis as a T2 descriptor;
      • 7.
        The reclassification of diaphragm invasion as T4;
      • 8.
        To delete mediastinal pleura invasion as a T descriptor.
      The proposed changes to the T component reduce in part the arbitrary nature with which some descriptors had been assigned in the past to a certain T category, maintain the compatibility with previous classifications, and improve the prognostic discrimination of the different T categories. Therefore, they should be implemented in the new edition of the TNM classification of lung cancer.

      APPENDIX IASLC Staging and Prognostic Factors Committee

      Peter Goldstraw, Past Chair, Royal Brompton Hospital and Imperial College, London, United Kingdom; Ramón Rami-Porta, Chair, Hospital Universitari Mutua Terrassa, Terrassa, Spain; Hisao Asamura, Chair Elect, Keio University, Tokyo, Japan; David Ball, Peter MacCallum Cancer Centre, Melbourne, Australia; David Beer, University of Michigan, Ann Arbor, MI; Ricardo Beyruti, University of Sao Paulo, Brazil; Vanessa Bolejack, Kari Chansky, and John Crowley, Cancer Research And Biostatistics, Seattle, WA; Frank Detterbeck, Yale University, New Haven, CT; Wilfried Ernst Erich Eberhardt, West German Cancer Centre, University Hospital, Ruhrlandklinik, University Duisburg-Essen, Essen, Germany; John Edwards, Northern General Hospital, Sheffield, United Kingdom; Françoise Galateau-Sallé, Centre Hospitalier Universitaire, Caen, France; Dorothy Giroux, Cancer Research And Biostatistics, Seattle, WA; Fergus Gleeson, Churchill Hospital, Oxford, United Kingdom; Patti Groome, Queen’s Cancer Research Institute, Kingston, Ontario, Canada; James Huang, Memorial Sloan-Kettering Cancer Center, New York, NY; Catherine Kennedy, University of Sydney, Sydney, Australia; Jhingook Kim, Samsung Medical Center, Seoul, Korea; Young Tae Kim, Seoul National University, Seoul, South Korea; Laura Kingsbury, Cancer Research And Biostatistics, Seattle, WA; Haruhiko Kondo, Kyorin University Hospital, Tokyo, Japan; Mark Krasnik, Gentofte Hospital, Copenhagen, Denmark; Kaoru Kubota, Nippon Medical School Hospital, Tokyo, Japan; Antoon Lerut, University Hospitals, Leuven, Belgium; Gustavo Lyons, British Hospital, Buenos Aires, Argentina; Mirella Marino, Regina Elena National Cancer Institute, Rome, Italy; Edith M. Marom, MD Anderson Cancer Center, Houston, TX; Jan van Meerbeeck, Antwerp University Hospital, Edegem (Antwerp), Belgium; Alan Mitchell, Cancer Research And Biostatistics, Seattle, WA; Takashi Nakano, Hyogo College of Medicine, Hyogo, Japan; Andrew G. Nicholson, Royal Brompton and Harefield NHS Foundation Trust and Imperial College, London, United Kingdom; Anna Nowak, University of Western Australia, Perth, Australia; Michael Peake, Glenfield Hospital, Leicester, United Kingdom; Thomas Rice, Cleveland Clinic, Cleveland, OH; Kenneth Rosenzweig, Mount Sinai Hospital, New York, NY; Enrico Ruffini, University of Torino, Torino, Italy; Valerie Rusch, Memorial Sloan-Kettering Cancer Center, New York, NY; Nagahiro Saijo, National Cancer Center Hospital East, Chiba, Japan; Paul Van Schil, Antwerp University Hospital, Edegem (Antwerp), Belgium; Jean-Paul Sculier, Institut Jules Bordet, Brussels, Belgium; Lynn Shemanski, Cancer Research And Biostatistics, Seattle, WA; Kelly Stratton, Cancer Research And Biostatistics, Seattle, WA; Kenji Suzuki, Juntendo University, Tokyo, Japan; Yuji Tachimori, National Cancer Center, Tokyo, Japan; Charles F. Thomas Jr, Mayo Clinic, Rochester, MN; William Travis, Memorial Sloan-Kettering Cancer Center, New York, NY; Ming S. Tsao, The Princess Margaret Cancer Centre, Toronto, Ontario, Canada; Andrew Turrisi, Sinai Grace Hospital, Detroit, MI; Johan Vansteenkiste, University Hospitals, Leuven, Belgium; Hirokazu Watanabe, National Cancer Center Hospital, Tokyo, Japan; and Yi-Long Wu, Guangdong General Hospital, Guangzhou, People’s Republic of China.

      Advisory Board of the IASLC Mesothelioma Domain

      Paul Baas, The Netherlands Cancer Institute, Amsterdam, The Netherlands; Jeremy Erasmus, MD Anderson Cancer Center, Houston, TX; Seiki Hasegawa, Hyogo College of Medicine, Hyogo, Japan; Kouki Inai, Hiroshima University Postgraduate School, Hiroshima, Japan; Kemp Kernstine, City of Hope, Duarte, CA; Hedy Kindler, The University of Chicago Medical Center, Chicago, IL; Lee Krug, Memorial Sloan-Kettering Cancer Center, New York, NY; Kristiaan Nackaerts, University Hospitals, Leuven, Belgium; Harvey Pass, New York University, NY; and David Rice, MD Anderson Cancer Center, Houston, TX.

      Advisory Board of the IASLC Thymic Malignancies Domain

      Conrad Falkson, Queen’s University, Ontario, Canada; Pier Luigi Filosso, University of Torino, Italy; Giuseppe Giaccone, Georgetown University, Washington, DC; Kazuya Kondo, University of Tokushima, Tokushima, Japan; Marco Lucchi, University of Pisa, Pisa, Italy; and Meinoshin Okumura, Osaka University, Osaka, Japan.

      Advisory Board of the IASLC Esophageal Cancer Domain

      Eugene Blackstone, Cleveland Clinic, OH.

      Participating Institutions in the New IASLC Lung Cancer Staging Project

      F. Abad Cavaco and E. Ansótegui Barrera, Hospital La Fe, Valencia, Spain; J. Abal Arca and I. Parente Lamelas, Complejo Hospitalario de Ourense, Ourense, Spain; A. Arnau Obrer and R. Guijarro Jorge, Hospital General Universitario de Valencia, Valencia, Spain; D. Ball, Peter MacCallum Cancer Centre, Melbourne, Australia; G. K. Bascom, Good Samaritan Hospital, Kearney, NE; A. I. Blanco Orozco and M. A. González Castro, Hospital Virgen del Rocío, Sevilla, Spain; M. G. Blum, Penrose Cancer Center, Colorado Springs, CO; D. Chimondeguy, Hospital Universitario Austral, Argentina; V. Cvijanovic, Military Medical Academy, Belgrade, Serbia; S. Defranchi, Hospital Universitario-Fundacion Favaloro, Buenos Aires, Argentina; B. de Olaiz Navarro, Hospital de Getafe, Getafe, Spain; I. Escobar Campuzano and I. Macía Vidueira, Hospital de Bellvitge, L’Hospitalet de Llobregat, Spain; E. Fernández Araujo and F. Andreo García, Hospital Universitari Germans Trias i Pujol, Badalona, Spain; K. M. Fong, Prince Charles Hospital, Brisbane, Australia; G. Francisco Corral and S. Cerezo González, Hospital La Mancha Centro, Ciudad Real, Spain; J. Freixinet Gilart, Hospital Universitario “Dr. Negrín,” Las Palmas de Gran Canaria, Spain; L. García Arangüena, Hospital Sierrallana, Torrelavega, Spain; S. García Barajas, Hospital Infanta Cristina, Badajoz, Spain; P. Girard, L’Institut Mutualiste Montsouris, Paris, France; T. Goksel, Turkish Thoracic Society, Turkey; M. T. González Budiño, Hospital General Universitario de Oviedo, Oviedo, Spain; G. González Casaurrán, Hospital Gregorio Marañón, Madrid, Spain; J. A. Gullón Blanco, Hospital San Agustín, Avilés, Spain; J. Hernández Hernández, Hospital de Ávila, Avila, Spain; H. Hernández Rodríguez, Hospital Universitario de Tenerife, Santa Cruz de Tenerife, Spain; J. Herrero Collantes, Hospital Universitario Nuestra Señora de la Candelaria, Santa Cruz de Tenerife, Spain; M. Iglesias Heras, Hospital de Ávila, Ávila, Spain; J. M. Izquierdo Elena, Hospital Nuestra Señora de Aránzazu, Donostia, Spain; E. Jakobsen, Danish Lung Cancer Registry, Denmark; S. Kostas, Athens School of Medicine, Athens, Greece; P. León Atance and A. Núñez Ares, Complejo Hospitalario de Albacete, Albacete, Spain; M. Liao, Shanghai Lung Tumor Clinical Medical Center, Shanghai, China; M. Losanovscky, Clinica y Maternidad Suizo Argentina, Buenos Aires, Argentina; G. Lyons, Hospital Britanico de Buenos Aires, Buenos Aires, Argentina; R. Magaroles and L. De Esteban Júlvez, Hospital Joan XXIII, Tarragona. Spain; M. Mariñán Gorospe, Hospital de San Pedro de Logroño, Logroño, Spain; B. McCaughan and C. Kennedy, University of Sydney, Sydney, Australia; R. Melchor Íñiguez, Fundación Jiménez Díaz, Madrid, Spain; L. Miravet Sorribes, Hospital La Plana, Castellón, Spain; S. Naranjo Gozalo and C. Álvarez de Arriba, Hospital Universitario Marqués de Valdecilla, Santander, Spain; M. Núñez Delgado, Hospital de Meixoeiro, Vigo, Spain; J. Padilla Alarcón and J. C. Peñalver Cuesta, Instituto Valenciano de Oncología, Valencia, Spain; J. S. Park, Samsung Medical Center, Seoul, South Korea; H. Pass, New York University Langone Medical Center and Cancer Center, New York, NY; M. J. Pavón Fernández, Hospital “Severo Ochoa,” Leganés, Spain; M. Rosenberg, Alexander Fleming Institute and Hospital de Rehabilitación Respiratoria, Buenos Aires, Argentina; E. Ruffini, University of Torino, Torino, Italy; V. Rusch, Memorial Sloan-Kettering Cancer Center, New York, NY; J. Sánchez de Cos Escuín, Hospital de Cáceres, Cáceres, Spain; A. Saura Vinuesa, Hospital de Sagunto, Sagunto, Spain; M. Serra Mitjans, Hospital Universitari Mutua Terrassa, Terrassa, Spain; T. E. Strand, Cancer Registry of Norway, Norway; D. Subotic, Clinical Centre of Serbia, Belgrade, Serbia; S. Swisher, MD Anderson Cancer Center, Houston, TX; R. Terra, University of Sao Paulo Medical Center, Sao Paulo, Brazil; C. Thomas, Mayo Clinic Rochester, Rochester, MN; K. Tournoy, University Hospital Ghent, Belgium; P. Van Schil, Antwerp University Hospital, Edegem (Antwerp), Belgium; M. Velasquez, Fundacion Clinica Valle del Lili, Cali, Colombia; Y. L. Wu, Guangdong General Hospital, Guangzhou, China; and K. Yokoi, Japanese Joint Committee for Lung Cancer Registry, Osaka, Japan.

      Supplementary Material

      Supplementary Material

      REFERENCES

        • Goldstraw P
        Staging Manual in Thoracic Oncology. Editorial Rx Press, Orange Park, FL2009: 57-65
        • Sobin L
        • Gospodarowocz M
        • Wittekind C
        TNM Classification of Malignant Tumours. 7th Ed. Wiley-Blackwell, Oxford, England2009: 138-146
        • Edge SB
        • Byrd DR
        • Compton CC
        • Fritz AG
        • Greene FL
        • Trotti III, A
        Cancer Staging Manual. 7th Ed. Springer, New York, NY2010: 253-270
        • Goldstraw P
        • Crowley JJ
        The International Association for the Study of Lung Cancer international staging project on lung cancer.
        J Thorac Oncol. 2006; 1: 281-286
        • Rami-Porta R
        • Goldstraw P
        Strength and weakness of the new TNM classification for lung cancer.
        Eur Respir J. 2010; 36: 237-239
        • Giroux DJ
        • Rami-Porta R
        • Chansky K
        • et al.
        International Association for the Study of Lung Cancer International Staging Committee. The IASLC Lung Cancer Staging Project: data elements for the prospective project.
        J Thorac Oncol. 2009; 4: 679-683
        • Rami-Porta R
        • Bolejack V
        • Giroux DJ
        • et al.
        International Association for the Study of Lung Cancer Staging and Prognostic Factors Committee, Advisory Board Members and Participating Institutions. The IASLC Lung Cancer Staging Project: the new database to inform the 8th edition of the TNM classification of lung cancer.
        J Thorac Oncol. 2014; 9: 1618-1624
        • Cox DR
        Regression models and life tables (with discussion).
        J R Stat Soc. 1972; B34: 187-220
        • Crowley J
        • LeBlanc M
        • Jacobson J
        • Salmon S
        Some exploratory tools for survival analysis. Lecture Notes on Statistics.
        in: Proceedings of the First Seattle Symposium in Biostatistics: Survival Analysis. Springer, New York1997: 199-299
        • R Core Team
        R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria2013 (Accessed March 18, 2015)
        • Travis WD
        • Brambilla E
        • Rami-Porta R
        • et al.
        International Staging Committee. Visceral pleural invasion: pathologic criteria and use of elastic stains: proposal for the 7th edition of the TNM classification for lung cancer.
        J Thorac Oncol. 2008; 3: 1384-1390
        • The National Lung Screening Trial Research Team
        Reduced lung-cancer mortality with low-dose computed tomography screening.
        New Engl J Med. 2011; 365: 395-409
        • Horeweg N
        • van der Aalst CM
        • Thunnissen E
        • et al.
        Characteristics of lung cancers detected by computer tomography screening in the randomized NELSON trial.
        Am J Respir Crit Care Med. 2013; 187: 848-854
        • Henschke CI
        • McCauley DI
        • Yankelevitz DF
        • et al.
        Early Lung Cancer Action Project: overall design and findings from baseline screening.
        Lancet. 1999; 354: 99-105
        • Wittekind Ch
        • Compton CC
        • Brierley J
        • Sobin LH
        TNM Supplement. A Commentary on Uniform Use. 4th Ed. Wiley-Blackwell, Oxford, England2012: 6
        • Yoshizawa A
        • Motoi N
        • Riely GJ
        • et al.
        Impact of proposed IASLC/ATS/ERS classification of lung adenocarcinoma: prognostic subgroups and implications for further revision of staging based on analysis of 514 stage I cases.
        Mod Pathol. 2011; 24: 653-664
        • Tsutani Y
        • Miyata Y
        • Nakayama H
        • et al.
        Prognostic significance of using solid versus whole tumor size on high-resolution computed tomography for predicting pathologic malignant grade of tumors in clinical stage IA lung adenocarcinoma: a multicenter study.
        J Thorac Cardiovasc Surg. 2012; 143: 607-612
        • Maeyashiki T
        • Suzuki K
        • Hattori A
        • Matsunaga T
        • Takamochi K
        • Oh S
        The size of consolidation on thin-section computed tomography is a better predictor of survival than the maximum tumour dimension in resectable lung cancer.
        Eur J Cardiothorac Surg. 2013; 43: 915-918
        • Goldstraw P
        Staging Manual in Thoracic Oncology. Editorial Rx Press, Orange Park, FL2009: 32-33
        • Sobin L
        • Gospodarowocz M
        • Wittekind C
        TNM Classification of Malignant Tumours. 7th Ed. Wiley-Blackwell, Oxford, England2009: 7-9
        • Yoshida J
        • Nagai K
        • Asamura H
        • et al.
        Japanese Joint Committee for Lung Cancer Registration. Visceral pleura invasion impact on non-small cell lung cancer patient survival: its implications for the forthcoming TNM staging based on a large-scale nation-wide database.
        J Thorac Oncol. 2009; 4: 959-963
        • Kudo Y
        • Saji H
        • Shimada Y
        • et al.
        Impact of visceral pleural invasion on the survival of patients with non-small cell lung cancer.
        Lung Cancer. 2012; 78: 153-160
        • Fibla JJ
        • Cassivi SD
        • Brunelli A
        • et al.
        Re-evaluation of the prognostic value of visceral pleura invasion in Stage IB non-small cell lung cancer using the prospective multicenter ACOSOG Z0030 trial data set.
        Lung Cancer. 2012; 78: 259-262
        • Kawase A
        • Yoshida J
        • Miyaoka E
        • et al.
        Japanese Joint Committee of Lung Cancer Registry. Visceral pleural invasion classification in non-small-cell lung cancer in the 7th edition of the tumor, node, metastasis classification for lung cancer: validation analysis based on a large-scale nationwide database.
        J Thorac Oncol. 2013; 8: 606-611
        • Shim HS
        • Park IK
        • Lee CY
        • Chung KY
        Prognostic significance of visceral pleural invasion in the forthcoming (seventh) edition of TNM classification for lung cancer.
        Lung Cancer. 2009; 65: 161-165
        • Maeda R
        • Yoshida J
        • Ishii G
        • Hishida T
        • Nishimura M
        • Nagai K
        Risk factors for tumor recurrence in patients with early-stage (stage I and II) non-small cell lung cancer: patient selection criteria for adjuvant chemotherapy according to the seventh edition TNM classification.
        Chest. 2011; 140: 1494-1502
        • Nitadori J
        • Colovos C
        • Kadota K
        • et al.
        Visceral pleural invasion does not affect recurrence or overall survival among patients with lung adenocarcinoma ⩽ 2 cm: a proposal to reclassify T1 lung adenocarcinoma.
        Chest. 2013; 144: 1622-1631
        • Padilla J
        • Calvo V
        • Morcillo A
        • García Zarza A
        • Blasco E
        • Pastor J
        [Resection of bronchogenic carcinoma invading the diaphragm].
        Arch Bronconeumol. 1999; 35: 297-298
        • Rocco G
        • Rendina EA
        • Meroni A
        • et al.
        Prognostic factors after surgical treatment of lung cancer invading the diaphragm.
        Ann Thorac Surg. 1999; 68: 2065-2068
        • Yokoi K
        • Tsuchiya R
        • Mori T
        • et al.
        Results of surgical treatment of lung cancer involving the diaphragm.
        J Thorac Cardiovasc Surg. 2000; 120: 799-805
        • Galetta D
        • Borri A
        • Casiraghi M
        • et al.
        Outcome and prognostic factors of resected non-small-cell lung cancer invading the diaphragm.
        Interact Cardiovasc Thorac Surg. 2014; 19 (discussion 636): 6-632
        • Ball D
        • Mitchell A
        • Giroux D
        • Rami-Porta RIASLC Staging Committee and Participating Institutions
        Effect of tumor size on prognosis in patients treated with radical radiotherapy or chemoradiotherapy for non-small cell lung cancer. An analysis of the staging project database of the International Association for the Study of Lung Cancer.
        J Thorac Oncol. 2013; 8: 315-321
        • Boffa DJ
        • Detterbeck FC
        • Smith EJ
        • et al.
        Should the 7th edition of the lung cancer stage classification system change treatment algorithms in non-small cell lung cancer?.
        J Thorac Oncol. 2010; 5: 1779-1783
        • Boffa DJ
        • Greene FL
        Reacting to changes in staging designations in the 7th edition of the AJCC staging manual.
        Ann Surg Oncol. 2011; 18: 1-3