[ Special Features ]

The Eighth Edition Lung Cancer Stage

Classification
Frank C. Detterbeck, MD, FCCP; Daniel J. Boffa, MD; Anthony W. Kim, MD, FCCP; and Lynn T. Tanoue, MD, FCCP

Stage classification provides a nomenclature about the anatomic extent of a cancer; a

consistent language provides the ability to communicate about a specific patient and about
cohorts of patients in clinical studies. This paper summarizes the eighth edition of lung
cancer stage classification, which is the worldwide standard as of January 1, 2017. This
revision is based on a large global database, a sophisticated analysis, extensive internal
validation as well as multiple assessments confirming generalizability. Practicing clinicians
must be familiar with the stage classification system when managing contemporary patients
with lung cancer. CHEST 2017; 151(1):193-203

KEY WORDS: lung cancer; non-small cell lung cancer; prognosis; stage classification

Classification of tumor stage is a cornerstone this must evolve. To meet the needs of
of providing care for patients with cancer. stability and consistency while allowing for
The fundamental purpose of stage progress, formal periodic revisions are
classification is to provide a nomenclature undertaken. The Union Internationale
about the anatomic extent of disease that is Contre le Cancer (UICC) and American
used consistently around the world. This Joint Committee on Cancer (AJCC) serve as
enables reliable communication about a the official bodies that define, periodically
particular patient, provides an understanding review and refine the stage classification
of the extent of disease among patients in a systems; although separate, these
clinical trial, and thus enhances the ability of organizations work together to achieve
clinicians to make judgments about how well global consistency. In January 2017, the
particular management strategies and eighth edition of the stage classification takes
associated results apply to a new patient. effect around the world, although
implementation is delayed in the United
Although it is critical that stage classification
States to ensure that the cancer care
represents a stable, consistently used
community has the necessary infrastructure
nomenclature, periodic revisions are needed.
in place. This paper summarizes the eighth
As technology changes and the ability to
edition AJCC/UICC stage classification for
define nuances regarding tumor extent
lung cancer.
progresses, the nomenclature that describes

ABBREVIATIONS: AJCC = American Joint Committee on Cancer; CORRESPONDENCE TO: Frank C. Detterbeck, MD, FCCP, Yale
IASLC = International Association for the Study of Lung Cancer; GG/ University School of Medicine, Department of Surgery, Section of
L = ground glass/lepidic; NSCLC = non-small cell lung cancer; SPFC = Thoracic Surgery, PO Box 208062, New Haven, CT 06520; e-mail:
Staging and Prognostic Factors Committee; UICC = Union Inter- frank.detterbeck@yale.edu
nationale Contre le Cancer Copyright Ó 2016 American College of Chest Physicians. Published by
AFFILIATIONS: From the Department of Surgery (Drs Detterback, Elsevier Inc. All rights reserved.
Boffa, and Kim), Section of Thoracic Surgery, and the Department of DOI: http://dx.doi.org/10.1016/j.chest.2016.10.010
Internal Medicine (Dr Tanoue), Section of Pulmonary, Critical Care
and Sleep Medicine, Yale University School of Medicine, New Haven,
CT.

journal.publications.chestnet.org 193

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017

 Patients and Methods Staging and Prognostic Factors Committee (SPFC). The SPFC for lung
was divided into multiple subcommittees which developed proposals
Basic Concepts
that were refined by the entire committee according to a formal process.6
The description of the anatomic extent of a tumor consists of three
components: T for extent of the primary tumor, N for involvement For the eighth edition the IASLC SPFC assembled a new global
of lymph nodes, and M for distant metastases. Each T, N, and M database of 94,708 patients receiving a diagnosis between 1999 and
component is divided into several categories (eg, T1, T2). Various 2010 from 35 sources and 16 countries. Most (85%) of the patients
characteristics, known as descriptors, define what is included within underwent surgery (
other treatments) and came from Europe
a T, N, or M category. Specific combinations of T, N, and M (49%) and Asia (44%). An extensive statistical analysis of the
categories are grouped together into stage groups. database was conducted by the Cancer Research and Biostatistics
group according to a set of guiding principles.6 Outcomes within the
A prefix further specifies the context of the stage classification (Table 1). database varied by region and type of source data; therefore,
Clinical stage (c) is determined by all information available before a proposals for stage classification were made on the basis of the
surgical resection, including symptoms, physical signs, imaging, presence (or absence) of differences in prognosis between
procedures, and biopsies. Pathologic stage (p) is defined by the results (heterogeneity) and within (homogeneity) categories and stage
of a surgical resection (or, rarely, an aborted surgical resection) groupings that were consistent across multiple comparisons (clinical,
together with all clinical staging information. Thus c and p stage apply pathologic, R0, R-any, N0, N-any, within a geographic region,
to the composite stage group or TNM designation; application of the c histologic type, database type, etc.). An understanding of the nature
or p designation to individual T, N, or M components is confusing and and limitations of the available data influenced how heavily
discouraged.1-4 Resected tumors are further classified by the extent of particular analyses were weighed; this was supplemented by clinical
resection (Table 2). A “certainty factor” (C) can be used to reflect the and historic considerations in arriving at the final classification
extent of testing involved in defining the stage (eg, a simple history proposals.
and physical [C1], imaging and invasive biopsies [C2], surgical biopsy
[C3], or surgical resection [C4]).5
It is essential that a classification system is broadly applicable; therefore
Process, Analysis, and Validation the SPFC conducted analyses that demonstrated geographic, historic,
The International Association for the Study of Lung Cancer (IASLC), methodologic, spectrum, and follow-up transportability.6 External
being the largest multidisciplinary global organization involved in lung validation was demonstrated using the US-based National Cancer
cancer, began developing infrastructure to inform the AJCC/UICC Database. Further external validation within local or regional
stage classification revisions in 1996. This initiative led to the revisions databases is encouraged; to be useful, this should evaluate
of the seventh and now also the eighth editions of the lung cancer discriminatory ability (not prognostic prediction) and be conducted
stage classification. IASLC appoints an international multispecialty in a scientifically robust manner.6

Results confirmed previous size cutpoints and suggested further

cutpoints in 1-cm increments. Non-size T descriptors
T Component
were examined using multivariate Cox regression
The T component analysis was on the basis of 10,230 analysis that adjusted for age, sex, histologic type, and
c-stage and 22,257 p-stage tumors with sufficient geographic region, again in multiple cohorts. Tumors
detailed information.7 The impact of size was analyzed with one vs more than one positive descriptor within a T
using a running log rank statistic (initially in a p-stage category were considered, but this was not incorporated
N0 M0 R0 non-small cell lung cancer [NSCLC] cohort, into the classification because of inconsistent
but then substantiated in multiple others).7 This differences.6,7

The T component is divided into five T categories that

TABLE 1 ] Types of Staging Assessments are defined by various T descriptors, as summarized in
Prefix Name Definition
Table 3.7 Size plays a prominent role in defining the T
c Clinical Before initiation of any
treatment, using any and
all information available
TABLE 2 ] Residual Tumor After Treatment
(eg, including Symbol Name Definition
mediastinoscopy)
R0 No residual No identifiable tumor
p Pathologic After resection, made on the remaining, negative
basis of pathologic surgical margins
assessment
R1 Microscopic Microscopically positive
y Restaging After part or all of the treatment residual margins but no visible
has been given tumor remaining
r Recurrence Stage at time of a recurrence R2 Gross Gross (visible or palpable)
a Autopsy Stage as determined by autopsy residual tumor remaining

194 Special Features [ 151#1 CHEST JANUARY 2017 ]

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017
 TABLE 3 ] Definitions for T, N, and M Descriptors

category. In addition, the T category is determined by (different from the seventh edition classification).
invasion into adjacent central/mediastinal or peripheral Involvement of a T structure by tumor that is extending
structures. Finally, when an additional tumor nodule is from a nodal metastasis (eg, left recurrent nerve
present, the location of this relative to the primary involvement by an aortopulmonary window node
tumor determines the T category. metastasis) is not counted as T involvement.
Invasion of a main bronchus is classified as T2a Involvement of hilar fat is classified as T2a and
regardless of the distance from the carina; similarly, involvement of mediastinal fat as T4. The mediastinal
atelectasis extending to the hilum is designated as T2a, pleura has been omitted as a T descriptor; the results
regardless of whether it involves a lobe or an entire lung were inconsistent, and specific (isolated) mediastinal
(different from the seventh edition classification). pleural involvement was rare. Involvement of the
Involvement of the diaphragm is classified as T4 parietal pericardium is classified as T3 (this means that

journal.publications.chestnet.org 195

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017

 the fat overlying the pericardium should probably not be categories was demonstrated in separate comparisons
counted as T4). Involvement of the visceral pericardium within geographic regions (Asia, North/South America,
is designated as T4. A Pancoast tumor is classified as T4 Europe, and Australia).12
if there is clear involvement of C8 or higher nerve roots,
The four N categories remain the same in the eighth
cords of the brachial plexus, subclavian vessels, vertebral
edition as in the seventh edition (Table 3).12 The
bodies, lamina, or spinal canal. A tumor is classified as
category is determined by the location of involved
T3 if it involves thoracic nerve roots only (ie, T1 or T2
nodes. Figure 1 and e-Table 1 provide a description and
nerve roots).
diagram of the node map.13 Direct extension of the
When multiple T descriptors are applicable to a tumor, primary tumor into an adjacent node is counted as
the highest T category should be chosen. In other words, nodal involvement.
a small tumor with a higher T category by invasion
The SPFC considered further subdivisions that included
should be classified by the invasion (eg, a 1.5-cm tumor
the number of involved node stations (Table 4). These
with visceral pleural involvement would be T2a), and a
analyses showed differences between p-stage tumors
large tumor with a lower degree of invasion should be
with single vs multiple N1 or N2 station involvement,
classified according to the size (eg, a 5.5-cm tumor
but no difference between multiple N1 stations and a
involving the main bronchus would be classified as T3).
single N2 skip metastasis (no N1 involvement). This
How size should be measured is specifically addressed.8 subgrouping was not included in the stage classification,
The maximum dimension of the solid component (on however, primarily because it could not be assessed in
imaging, c-stage) or the invasive component (on c-stage tumors.
microscopy, p stage) is used to assign the T category;
The AJCC, UICC, and IASLC recommend that at least
however, the maximum dimension of the ground glass
six nodes are removed during surgical resection, three
or lepidic component should also be recorded. Further
from N1 and three from N2 stations (ie, a representative
details of how this should be measured on imaging were
node from each station) for accurate staging.12 There are
also addressed by the SPFC subcommittee. Slice
differences of opinion whether N0 status should be
thickness, window settings, degree of inspiration, and
recognized if more limited sampling has occurred and is
scanner parameters can affect the observed size8; in
negative; some would classify this as pN0, whereas
addition, there is significant inter- and intraobserver
others suggest the designation pN0(un) to show that
variability in size measurement with smaller lesions.9,10
there is a degree of uncertainty.
There are several special situations. A superficial
spreading tumor in the central airways is classified as M Component
T1a, regardless of location. Carcinoma in situ is The M component analysis included 1,059 nonsurgically
classified as Tis; note that this now applies to both managed NSCLC M1 tumors.14 This cohort was drawn
squamous carcinoma and adenocarcinoma.8 Minimally from a detailed “electronic data capture” portion of the
invasive adenocarcinoma is classified as T1a(mi). A database; most other submitted nonsurgically managed
minimally invasive adenocarcinoma has an invasive M1 tumors lacked sufficient detail. The M categories and
component of # 5 mm and a lepidic (noninvasive) descriptors are summarized in Table 3. Pleural/
component of # 3 cm.11 (Note that these diagnoses can pericardial nodules, pleural/pericardial effusion, and
are only be made in resected tumors.) contralateral/bilateral pulmonary nodules are classified
as M1a. M1b denotes tumors with a single distant
N Component (extrathoracic) metastasis. There was no consistent
The N component analysis was made on the basis of difference with respect to the site of metastasis among
38,910 c-stage and 31,426 p-stage tumors with sufficient tumors with a single distant metastatic focus. M1c
detailed information. The discrimination of the N includes tumors with multiple metastases, either
categories was first demonstrated in c-stage T-any M0 multiple metastases in a single organ or multiple
NSCLC cases, then confirmed in each T category and in metastases in multiple organs.
p-stage cases (ie, T-any M0 R-any and T-any M0 R0).
Patients with sufficient detail to be evaluable for the N Stage Groups
component analysis were largely contributed from For the stage group analysis, 17,477 c-stage tumors
Japan. However, geographic applicability of the N (16,595 T-any N-any M0 and 882 T-any N-any M1)

196 Special Features [ 151#1 CHEST JANUARY 2017 ]

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017
 Figure 1 – The International Association for the Study of Lung Cancer node map for lung cancer. With permission from Rusch et al.13

and 31,936 p-stage tumors (all T-any N-any M0) were period of case entry. After extensive testing in
available. Candidate stage grouping schemes were multiple subgroups, adjusted Cox regression
developed beginning with M0 tumors and best stage, analyses, validation set analyses, and considerations
using a recursive partitioning and amalgamation regarding practicality and clinical relevance, the
algorithm made on the basis of survival in a training stage grouping shown in Table 5 and Figures 2, 3, and 4
set, stratified by type of data submission and time was selected.

journal.publications.chestnet.org 197

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017

 TABLE 4 ] N Subclassification TABLE 5 ] Lung Cancer Stage Grouping (Eighth Edition)
Category Subclass Description T/M Label N0 N1 N2 N3
T1 T1a ≤1 IA1 IIB IIIA IIIB
Nx Regional lymph nodes cannot be
T1b >1-2 IA2 IIB IIIA IIIB
assessed
T1c >2-3 IA3 IIB IIIA IIIB
N0 No regional lymph node T2 T2a Cent, Yisc Pl IB IIB IIIA IIIB
involvement T2a >3-4 IB IIB IIIA IIIB
N1 N1a Single-station N1 involvement T2b >4-5 IIA IIB IIIA IIIB
T3 T3 >5-7 IIB IIIA IIIB IIIC
N1b Multiple-station N1 involvement T3 Inv IIB IIIA IIIB IIIC
N2 N2a1 Single-station N2 without N1 T3 Satell IIB IIIA IIIB IIIC
involvement (skip) T4 T4 >7 IIIA IIIA IIIB IIIC
N2a2 Single-station N2 with N1 T4 Inv IIIA IIIA IIIB IIIC
involvement T4 Ipsi Nod IIIA IIIA IIIB IIIC
M1 M1a Contr Nod IVA IVA IVA IVA
N2b Multiple-station N2 involvement M1a Pl Dissem IVA IVA IVA IVA
N3 N3 lymph node involvement M1b Single IVA IVA IVA IVA
M1c Multi IVB IVB IVB IVB
See Table 3 text and legend for expansion of abbreviations.

Stage I involves T1/T2a N0 M0 tumors; stage II involves

either T2b/T3 N0 M0 tumors or T1/T2 N1 M0 tumors. regardless of T or N classification. Stage IVB involves all
Stage III is now divided into three subgroups. Stage IIIA M1c tumors.
includes T4 N0 M0 and T3/4 N1 M0 tumors as well as
T1/T2 N2 M0 tumors. Stage IIIB tumors are either Multiple Pulmonary Sites of Lung Cancer
T3/T4 N2 M0 or T1/T2 N3 M0. Stage IIIC involves Patients with multiple pulmonary sites of lung cancer
T3/T4 N3 M0 tumors. Stage IV is divided into two are seen with increasing frequency. There has been
subgroups. Stage IVA includes all M1a and M1b tumors, significant variability in how TNM classification has

Figure 2 – Graphic illustration of stages 0, I, and II.

198 Special Features [ 151#1 CHEST JANUARY 2017 ]

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017
 Figure 3 – Graphic illustration of stage III.

been applied to these tumors.15 An SPFC subcommittee single “typical” lung cancers according to the stage and
proposed definitions and a schema for stage histologic type.18 Note that most second primary lung
classification of such tumors.16-19 Four patterns of cancers have the same histotype, and that there is
disease are distinguished (Table 6); the clinical substantial variability in biomarker patterns (ie, either
presentation, pathologic correlates, and biologic different in clearly related metastases or the same in
behavior of these suggest specific applications of TNM clearly different tumors). This means histologic type of
classification rules. The subcommittee developed a series biomarker patterns alone are not entirely reliable to
of criteria to define these four patterns of disease classify two tumors as separate primaries or related
(summarized in e-Tables 2-5).16 tumors; classification should take into account all
available information or involve a comprehensive
First, patients can present with second primary lung histologic assessment.18 Second primary lung cancers
cancers. The demographic characteristics, outcomes, and should be designated with a T, N, and M category for
recurrence patterns for each tumor are similar to that of each tumor.

Figure 4 – Graphic illustration of stage IV.

journal.publications.chestnet.org 199

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017

 TABLE 6 ] Schematic Summary of Patterns of Disease and TNM Classification of Patients With Lung Cancer With
Multiple Pulmonary Sites of Involvement
Second Primary Multifocal Pneumonic-Type Separate
Lung Cancer GG/L Nodules of Adenocarcinoma Tumor Nodule
Imaging Two or more distinct Multiple ground glass Patchy areas of ground Typical lung cancer
Features masses with imaging or part-solid nodules glass and consolidation (eg, solid, spiculated)
characteristic of lung with separate solid
cancer (eg, spiculated) nodule
Pathologic Different histotype or Adenocarcinomas with Same histology Distinct masses with
Features different morphology prominent lepidic throughout (most often the same morphology
by comprehensive component (typically invasive mucinous by comprehensive
histologic assessment varying degrees of adenocarcinoma) histologic assessment
AIS, MIA, LPA)
TNM Classi- Separate cTNM and T based on highest T T based on size or T3 if Location of separate
fication pTNM for each cancer lesion with (#/m) in single lobe, T4 or nodule relative to
indicating multiplicity; M1a if in different ipsi- primary site
single N and M or contralateral lobes; determines if T3, T4 or
single N and M M1a; single N and M
Conceptual Unrelated tumors Separate tumors, albeit Single tumor, diffuse Single tumor, with
View with similarities pulmonary involvement intrapulmonary
metastasis
AIS ¼ adenocarcinoma in situ; GG/L ¼ ground glass/lepidic; LPA ¼ lepidic predominant adenocarcinoma; MIA ¼ minimally invasive adenocarcinoma.
Reprinted with permission from Detterbeck et al.16

Second, some patients with a solid primary lung cancer in one lobe, T4 if involving multiple same-side lobes,
have one or more separate solid tumor nodule(s) of the and M1a if involving both lungs—with a single N and M
same histologic type (referred to as “intrapulmonary category for all areas of involvement.
metastasis” in the pathology community). The behavior
Discussion
of these tumors is similar to that of a similar solitary
Definition of stage classification of lung cancer has
tumor; outcomes are slightly inferior and affected by
undergone a transformative change with the
how they are treated.17 These tumors should be classified
engagement of the IASLC SPFC. The size of the
according to the location of the separate nodule relative
database, the sophistication of the analysis, and the
to the index tumor—T3 for a same-lobe, T4 for a same-
extent of internal and external validation are
side (different lobe), and M1a for an other-side
unprecedented among solid tumors. A debt is owed by
location—with a single N and M category.
the world to the many contributors who committed the
A third pattern of disease involves patients presenting time to provide the worldwide data that make this
with multiple lung cancer nodules with prominent possible; nevertheless, there are surely aspects of the
ground glass or lepidic (GG/L) features. This group has stage classification that can be improved. A different
different demographic characteristics, excellent type of engagement is now needed: investigators are
outcomes, and infrequent recurrences outside the lung encouraged to test the system and expose areas needing
parenchyma.19 These GG/L tumors should be further refinement. To be useful, an analysis must be
designated by the T category of the highest T lesion, the scientifically rigorous and robust. Proposed metrics for
number or “m” in parentheses (#/m) to indicate the such analyses have been outlined.6
multiplicity, and a collective N and M category for all.
The development of lung cancer stage classification rests
A comprehensive histologic assessment of each GG/L
on an unprecedented scientific foundation; nevertheless,
tumor nodule is not required.
limitations exist. Although the database is large and has
A fourth pattern of disease involves a form of lung global representation, it is still essentially a convenience
cancer that is radiologically similar to pneumonia (so- sample of available data. Regions other than Asia and
called “pneumonic-type” of lung cancer). Extrathoracic Europe are underrepresented. Nonsurgically managed
and nodal involvement is infrequent, but prognosis is patients are underrepresented in the IASLC database;
distinctly worse than for patients with multiple GG/L however, internal validation demonstrated geographic,
nodules.19 Diffuse pneumonic-type lung cancers are spectrum, and methodologic transportability.
designated by size (or T3 if size cannot be determined) if Furthermore, the eighth edition stage classification has

200 Special Features [ 151#1 CHEST JANUARY 2017 ]

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017
 been externally validated against the US-based National 1999 through 2010 database than the IASLC 1990
Cancer Database (publication underway) which consists through 1999 database (approximately 30% better but
largely of nonsurgically managed patients; therefore, this variable by stage).6 Outcomes have presumably
external validation demonstrates excellent improved further in the current period, but this is not
discriminatory validity of the eighth edition stage defined.
classification in the United States and in nonsurgical
cohorts. There is a strong need for a prognostic prediction model.
To be clinically useful, this should be current (applicable
It is important to understand the relationship between
to patients managed today), specific for an individual
stage classification and prognosis. Differences in
patient, and be accurate and validated. Each of these
prognosis that were consistent across multiple
aspects is inherently problematic. We must use data
subgroups and adjusted multivariate regression analyses
from the past (with known outcomes), yet apply it to the
were used prominently (but not exclusively) to decide
future in a rapidly changing field. It must be flexible to
how to classify tumors in groups that are sufficiently
accommodate new prognostic factors, but robust model
internally homogeneous but also distinct from one
development requires sufficient follow-up and a large
another. Additionally, although there is no question that
database that includes all potential factors. Validation
the anatomic extent of disease has an impact on
also requires follow-up and sufficiently large cohorts—
prognosis, prognosis is also affected by a multitude of
how can this be accomplished in a manner that is
other patient-related (eg, performance status, age,
personalized for countless individual patients around the
comorbidities, competing causes of death), tumor-
world? The SPFC is working to address these challenges.
related (eg, histologic subtype, grade, PET intensity,
In the meantime, we must recognize that a clinician’s
genomics), environment-related (eg, access to care,
ability to integrate complex information for an
quality of care), and treatment-related factors (eg, which
individual patient is what we have and is probably
treatment is chosen, treatment response). The variability
reasonably accurate. We should use the information
within the IASLC database and the need for careful
regarding the prognostic impact of the anatomic disease
adjusted analyses underscores this. Finally, stage
burden (ie, the outcomes noted for the stage groups), but
classification is a nomenclature and therefore must
we must account for the changing environment and
inherently remain relatively static and universally
additional factors affecting prognosis in making a
applicable, whereas prognosis is fluid, constantly
prediction for a particular patient.
changing, and specific to an individual patient, clinical
setting, and point in time.20
Stage classification is not a treatment guideline.
There is a strong temptation to focus on the outcomes of Treatment recommendations stem from the data we
the patients in the IASLC database (Table 7). These have regarding outcomes of patients managed according
outcomes represent an average of patients from various to a specific treatment strategy. Whatever name we put
parts of the world, diagnosed between 1999 and 2010 on tumors (ie, a seventh edition or an eighth edition
and treated in many different ways. How applicable this name) does not alter the data we have regarding specific
is to patients diagnosed today in a specific locale and treatment outcomes. The stage classification
treated in a specific way is highly questionable. This is nomenclature is used simply as a tool for
underscored by the variability in outcomes by regions communication among clinical trials and clinical
and type of source data in the IASLC database.6 guidelines, which is what one has to look to for
Furthermore, outcomes are substantially better in the treatment recommendations.

TABLE 7 ] 5-Year Survival (%)

Type IA1 IA2 IA3 IB IIA IIB IIIA IIIB IIIC IVA IVB
Clinical 92 83 77 68 60 53 36 26 13 10 0
Pathologic 90 85 80 73 65 56 41 24 12 - -
Average overall survival in the International Association for the Study of Lung Cancer global database of patients receiving a diagnosis between 1999
and 2010. Data from Goldstraw et al.21

journal.publications.chestnet.org 201

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017

 It is transiently of interest to consider what has changed Acknowledgments
between the seventh and eighth edition stage Financial/nonfinancial disclosures: None declared.
classifications. In short, the T categories have been Additional information: The e-Tables can be found in the
broken down further by size (in 1-cm increments up to Supplemental Materials section of the online article.
5 cm). Tumors that are > 5 to 7 cm are now T3 and T4
if > 7 cm. Central tumors involving a main bronchus or
References
1. Detterbeck F, Boffa DJ, Tanoue L, Wilson L. Details, difficulties and
causing obstructive atelectasis are all classified as T2a dilemmas regarding the new lung cancer staging system. Chest.
regardless of the distance to the carina or if the lung is 2010;137(5):1172-1180.

partially or completely atelectatic. Tumors involving the 2. UICC. TNM Classification of Malignant Tumours. 8th ed. Hoboken,
NJ: Wiley-Blackwell; 2016.
diaphragms are classified as T4. There are no changes in
3. AJCC. AJCC Cancer Staging Manual. 8th ed. New York: Springer;
the N categories. The M category now distinguishes 2016.
tumors with a solitary distant metastasis from multiple 4. Rami-Porta R. International Association for the Study of Lung
metastases. However, the relevance of comparing the Cancer (IASLC). IASLC Staging Manual in Thoracic Oncology. North
Fort Meyers, FL: Editorial Rx Press; 2016.
seventh and eighth editions will greatly diminish after a 5. Detterbeck F, Tanoue L, Boffa DJ. The new lung cancer staging
brief transition period. Besides, it is more complicated to system. Chest. 2009;136(1):260-271.
focus on the changes than to simply learn the new 6. Detterbeck F, Chansky K, Groome P, et al. The IASLC Lung Cancer
Staging Project: methodology and validation used in the development
classification system, especially when considering the of proposals for revision of the stage classification of non-small cell
stage groups. Therefore we have focused primarily on lung cancer in the forthcoming (8th) edition of the TNM Classification
of Lung Cancer. J Thor Oncol. 2016;11(9):1433-1446.
explanations that will facilitate implementation of the
7. Rami-Porta R, Bolejack V, Crowley J, et al. The IASLC Lung Cancer
new system. Staging Project: proposals for the revisions of the T descriptors in
the forthcoming eighth edition of the TNM Classification for Lung
Although the AJCC and UICC are almost completely Cancer. J Thorac Oncol. 2015;10(7):990-1003.
aligned regarding stage classification and definitions, 8. Travis D, Asamura H, Bankier AA, et al. The IASLC Lung Cancer
Staging Project: proposals for coding T categories for subsolid
there is a slight discrepancy of terms with the eighth nodules and assessment of tumor size in part solid tumors in the
edition. The AJCC eighth edition uses the singular forthcoming eighth edition of the TNM Classification of Lung
Cancer. J Thorac Oncol. 2016;11(8):1204-1223.
term “prognostic stage groups” to describe the
9. Kim H, Park CM, Woo S, et al. Pure and part-solid pulmonary
grouping system, regardless of whether if it is solely ground-glass nodules: measurement variability of volume and mass
anatomic or also includes nonanatomic factors in the in nodules with a solid portion less than or equal to 5 mm.
Radiology. 2013;269(2):585-593.
classification. The UICC uses the term “stage groups”
10. Nietert PJ, Ravenel JG, Leue WM, et al. Imprecision in automated
to refer to stage classification based strictly on volume measurements of pulmonary nodules and its effect on the
anatomic factors and “prognostic stage groups” for a level of uncertainty in volume doubling time estimation. Chest.
2009;135(6):1580-1587.
separate classification that includes nonanatomic
11. Travis W, Brambilla E, Burke A, Marx A, Nicholson A. WHO
factors. This is a minor point for lung cancer, because Classification of Tumours of the Lung, Pleura, Thymus and Heart.
there is no prognostic grouping incorporating 4th ed. Lyon, France: International Agency for Research on Cancer
(IARC); 2015.
nonanatomic factors.
12. Asamura H, Chansky K, Crowley J, et al. The IASLC Lung Cancer
Staging project: proposals for the revision of the N descriptors in the
forthcoming eighth edition of the TNM Classification for Lung
Conclusion Cancer. J Thorac Oncol. 2015;10(12):1675-1684.
The eighth edition of TNM classification of lung cancer 13. Rusch V, Asamura H, Watanabe H, et al. The IASLC Lung Cancer
Staging Project: a proposal for a new international lymph node map
is the worldwide standard as of January 1, 2017. An in the forthcoming 7th edition of the TNM classification for lung
extensive and multifaceted analysis served as the cancer. J Thorac Oncol. 2009;4(5):568-577.
foundation for this revision. The T component is 14. Eberhardt W, Mitchell J, Crowley J, et al. The IASLC Lung Cancer
Staging Project: proposals for the revision of the M descriptors in the
subdivided by primary tumor size in 1-cm increments as forthcoming (8th) edition of the TNM Classification of Lung Cancer.
well as other descriptors of invasion into adjacent J Thorac Oncol. 2015;10(11):1515-1522.
structures. The N component is determined by the 15. Fonseca A, Detterbeck FC. How many names for a rose: inconsistent
classification of multiple foci of lung cancer due to ambiguous rules.
location of involved lymph nodes. The M component is Lung Cancer. 2014;85(1):7-11.
subdivided into intrathoracic dissemination, a single 16. Detterbeck F, Nicholson F, Franklin W, et al. The IASLC
extrathoracic metastasis, and multiple metastases. These Lung Cancer Staging Project:summary of proposal revisions of the
classification of lung cancers with multiple pulmonary sites of
are coalesced into stage groups. It is essential that those involvement in the forthcoming eighth edition of the TNM
caring for these patients are familiar with this system Classification. J Thorac Oncol. 2016;11(5):639-650.
17. Detterbeck F, Bolejack V, Arenberg D, et al. The IASLC Lung Cancer
because it provides a universal language to describe the Staging Project: background data and proposals for the classification
anatomic extent of disease. of lung cancer with separate tumor nodules in the forthcoming

202 Special Features [ 151#1 CHEST JANUARY 2017 ]

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017
 eighth edition of the TNM Classification for Lung Cancer. J Thorac pneumonic-type of involvement in the forthcoming eight
Oncol. 2016;11(5):681-692. edition of the TNM classification. J Thorac Oncol. 2016;11(5):
18. Detterbeck F, Arenberg D, Asamura H, et al. The IASLC Lung Cancer 666-680.
Staging Project: background data and proposed criteria to distinguish 20. Detterbeck FC. Stage classification and prediction of prognosis: the
separate primary lung cancers from metastatic foci in patients with difference between accountants and speculators. J Thorac Oncol.
two lung tumors in the rothcoming eight edition of the TNM 2013;8(7):820-822.
Classification for Lung Cancer. J Thorac Oncol. 2016;11(5):651-665. 21. Goldstraw P, Chansky K, Crowley J, et al. The IASLC Lung
19. Detterbeck F, Arenberg D, Asamura H, et al. The IASLC Lung Cancer Staging Project: Proposals for Revision of the TNM
Cancer Staging Project: background data and proposals for the Stage Groupings in the Forthcoming (Eighth) Edition of the
application of TNM Staging rules to lung cancer presenting as TNM Classification for Lung Cancer. J Thorac Oncol. 2016;11(1):
multiple nodules with ground glass or lepidic features or a 39-51.

journal.publications.chestnet.org 203

Downloaded From: http://journal.publications.chestnet.org/ by Jean-Paul D Odemont on 01/15/2017