Download Prognosis of early central lung cancer is better than T1 staged group

Prognosis and staging of superficial endobronchial lung cancer: the impact of invasion depth,
tumor diameter, and co-existent pneumonitis or atelectesis.
1
Chang CHEN, M.D., 2Hui ZHENG, M.D., 1Wen GAO, M.D., 3Ying ZHOU, M.D., 4 Sen JIANG, M.D.
and 1Hon Chi SUEN*, M.D.
1
Department of Thoracic Surgery, 2 Pathology, 3 Respiratory Medicine, and 4 Radiology, Tongji
University-Affiliated Shanghai Pulmonary Hospital, Shanghai 200433, China
*Dr. Hon Chi Suen is an Honored Professor at Tongji University-Affiliated Shanghai Pulmonary
Hospital.
Corresponding author:
Chang CHEN, M.D.
Zhengmin RD. 507, Shanghai 200433, China
TEL.: +86-21-65115006, ext. 2074
FAX: +86-21-65111298
E-mail address: [email protected]
1
Abbreviations
SELC
Superficial endobronchial lung cancer
EHLC
Early hilar lung cancer
2
Abstract
BACKGROUND Few reports exist discussing the surgical-pathological characteristics of superficial
endobronchial lung cancer (SELC), which has been defined as cancer growth limited to within the
bronchial wall. Its prognosis and corresponding TNM staging have not been fully clarified. As well, little
is known as to whether such T status is impacted by the existence of associated atelectasis or pneumonia
(which might be controversial, indicating either T1 or T2), and circumstantial invasion depth.
METHODS Between 1988 and 2007, 81 out of 8817 surgical patients met SELC criteria, in that there
was no detectable invasion beyond the bronchial wall. A retrospective review was performed, and
follow-up information was collected. RESULTS The overall five-year survival rate for all 81 patients
was 85.6%; for N0M0 (n=67), N1M0 (n=7) and N2M0 (n=7) patients, it was 89.3%, 75.0% and 60.0%,
respectively. Intraluminal tumor size measured 0.4-3.0 cm; obstructive atelectasis or pneumonia was
noted in 14 patients. The presence of tumor-associated obstructive atelectesis or pneumonia did not have a
significant impact upon prognosis (P=0.96), nor was the tumor’s greatest diameter (P=0.70). Histology
showed carcinoma in situ (level one) in 13 cases; invasion of the submucosal layer (level two) in 12,
involvement of the muscular layer (level three) in 20, invasion into the space between the muscular layer
and cartilage (level four) in 21, and bronchial cartilage infiltration in 15 (level five). In cases without
lymph node metastases, five year survival was 100% for the first three levels; 84.0% and 61.3% for the
fourth and fifth levels, respectively. CONCLUSIONS Relative to TNM-based prognostic data,
superficial endobronchial lung cancer exhibits much enhanced five-year survival rates, and therefore
should be placed at the forefront among tumors in the T1 class, regardless of tumor size or the presence of
secondary obstructive atelectasis or pneumonia. Lymph node metastasis is associated with a worse
3
prognosis. Survival is negatively impacted by tumor infiltration depth into the bronchial wall.
Key Words
Lung cancer, Prognosis, Staging, Surgery, Superficial endobronchial lung cancer, Early hilar lung cancer
4
Introduction
The term superficial endobronchial lung cancer (SELC) refers to an unusual group of lung cancers, all of
which are confined to the bronchial lumen both macroscopically and microscopically [1]. This term most
commonly is adopted by pulmonary pathologists during their gross description of tumors; whereas in
clinical practice, the tumor-related special clinical-pathological characteristics of SELC, including its
unusually good prognosis and corresponding staging significance, often is neglected. In the 1997 revised
International Staging System for lung Cancer, this type of lung cancer “of any size” was grouped into
classification category T1 [1]; no updated material has yet been available from the IASLC Staging Project.
However, such T classification may become clouded under certain circumstances. For instance, it has not
been clarified what occurs, prognostically, when these tumors are “associated with atelectasis or
obstructive pneumonitis that extends to the hilar region but does not involve the entire lung” [1]. Both of
these radiological findings suggest a T2 classification, rather than T1.
As well, little has been discussed about the surgical-pathological characteristics of SELC, including the
influence of tumor diameter, lymph node metastases, and bronchial circumstantial invasion depth upon
prognosis. Since this tumor grows longitudinally, superficially, and intraluminally along the bronchi, it
commonly is assumed to be an early stage lung cancer within limited extension, probably having a very
low incidence of nodal or remote involvement [2], and subsequently being associated with excellent long
term survival. However, the issues just described are not fully covered by the TNM staging system;
therefore further study is warranted.
5
There also has been some controversy over SELC’s precise T position, given the surprisingly excellent
prognosis attributed to early hilar lung cancer (EHLC) [3-5]. EHLCs initially were proposed by the
Japanese Lung Cancer Society to be those tumors, growing in major bronchi, which are confined to the
bronchial wall and which lack any lymph node or distant metastases. Consequently, EHLC appears to be
a subgroup of SELC, consisting of those SELC that lack any extra-bronchial invasion or metastases.
Strikingly, EHLC exhibits more than a 90% five-year survival rate [3-6], an outcome that is much better
than that attributed to ‘early stage’ lung cancer referring to stages Ia-Ib (T1N0M0 and T2N0M0) [7].
Therefore, the word ‘early’ may be discrepant in these two circumstances. In current cancer research
literature, the concepts of EHLC and SELC seem best communicated using TNM nomenclature.
In this study, we performed a retrospective review drawing from nineteen years of surgical experience at a
single institution, intending to address several key issues discussed above. Additionally, the
surgical-pathological characteristics of this uncommon lung cancer, SELC, were examined.
Patients and Methods
From April 1988 to June 2007, a total of 8817 operations were performed for non-small-cell lung cancer
in the Department of Thoracic Surgery at Shanghai Pulmonary Hospital, Shanghai, China. Among these
patients, 904 patients were staged pathologically as T1, utilizing the 1997 revised international TNM
system, including both peripheral and hilar types. Eighty-one patients met the criteria of no invasion
6
beyond the bronchial wall, either macroscopically or microscopically. We retrieved their detailed
information from the pathological data base. These 81 patients’ demographic data and their pre- and
post-operative medical information were extracted from case records as well.
All 81 patients had undergone a thorough preoperative evaluation assessing organ function, and all had
had their tumor staged. Remote metastases had been excluded by means of bone scan, computed
tomography or magnetic resonance imaging of both the brain and abdomen. Pre-operative fiberoptic
bronchoscopy had been performed routinely for the purposes of establishing a preoperative diagnosis,
assessing the extent of excision required, and predicting postoperative residual lung function. As a result,
intrabronchial neoplasm was noted in all patients, and biopsies had preoperatively confirmed a
histological diagnosis in 65 cases. Tumor-associated obstructive atelectasis and/or pneumonia were noted
in 14 patients, based upon both clinical presentation at admission and specimen pathological description.
These findings were added to the data base for statistical analysis.
Pathological information had been collected on tumor location, the tumor’s greatest diameter, the status of
lymph node metastases, and the histological diagnosis. We restaged all cancers, based upon the 1997
revised international TNM classification system. The pathologist in our research team re-examined and
recorded circumstantial tumor invasion depth in every case. The histological stratification method we
used was slightly modified from that introduced by Dr. Konaka in 1999 [12]. It was comprised of five
perpendicular levels: 1) carcinoma in situ; 2) no invasion of the superficial muscular layer; 3) invasion
into the muscular layer; 4) invasion beyond the muscular layer but not reaching cartilage; and 5)
7
bronchial cartilage involvement.
Follow-up information was obtained by contacting patients or their relatives either by phone or mail
inquiry. Overall and TMN-specific survival rates were calculated. The tumor’s greatest diameter, and the
co-existence of tumor-associated atelectasis or pneumonia were tested as possible prognosis-influencing
risk factors. Overall five-year survival rates and group survival comparisons were calculated by means of
the Kaplan-Meier method and Cox Regression. Inter-group differences associated with P values < 0.05
were considered to be statistically significant.
Results
There were 80 men and 1 woman with an average age of 59.8±8.0 years (range 35-75 years). Most of
these patients initially had presented with non-specific symptoms like cough, sputum production,
hemoptysis, wheezing or chest pain. Three patients initially were referred to a pulmonary specialist
because of perceived patchy abnormalities on plain chest films during yearly occupational physical
examinations. Eleven were nonsmokers and 70 patients were smokers. Excluding these non-smokers, the
average smoking index (cigarettes per day × years) was 1100.
Among these 81 patients, 29 were noted to have masses in the right upper lobe, 12 in the right lower lobe,
1 at the right main bronchus, 4 in the right middle lobe, 28 in the left upper lobe, and 7 in the left lower
lobe. Most patients underwent standard pulmonary resection and lymph node sampling. Surgical
8
procedures included lobectomy (n=48), sleeve lobectomy or lobectomy with bronchoplasty (n=19), and
pneumonectomy (n=14) depending upon the extent of disease. Complete resection was performed in 76
patients; in the remaining 5 individuals, the bronchial resection margins remained positive for residual
tumor infiltration, largely because these patients were deemed to be unable to tolerate more aggressive
resection and reconstruction; these 5 patients received postoperative salvage radiation to the cancer
-retained stumps. Adjuvant chemotherapies were administered to 34 patients; the regimens of which
comprised of MVP (Mitomycin 6mg/m2, d1 + Vindesine 3mg/m2, d1 + Cisplatin 25mg/m2, d1, d2), NP
(Vinorelbine 25mg/m2, d1, d8 + Cisplatin 50mg/m2, d1, d2 or Carboplatin, AUC 5, d1) and TP
(paclitaxel 150mg/m2, d1 + Cisplatin 50mg/m2, d1, d2 or Carboplatin, AUC 5, d1).
Final pathology reports indicate that 67 cases were free of lymph node metastases (N0), 7 had regional
nodal involvement (N1), and 7 had mediastinal node metastases (N2). All but 3 cases were squamous cell
carcinoma. These three cases included adeno-squamous carcinoma in one patient; and synchronous
ipsilateral intrabronchial squamous lung cancer, combined with a peripheral nodule of primary
adenocarcinoma in an adjacent lobe, in two. Lobectomies were performed for the central endobronchial
squamous cell cancers, and wedge resections for nodular peripheral adenocarcinoma. Postoperative
pathological examination suggested an N2 status for the one adeno-squamous carcinoma patient and N0
for the other two.
At the time of follow up, 16 patients were noted to have died. Five of these patients had died of
respiratory failure 48, 69, 84, 178, and 182 months postoperatively. Among the remaining 11 deceased
9
patients, two patients had N2 disease and one patient N1, and eight patients initially had been noted to be
free of lymph nodal involvement. During follow up, two patients developed a second squamous cell
carcinoma, one each at 104 and 168 months postoperatively. The five- and ten- year survivals for all 81
patients were 85.6% and 70.5% respectively. Five-year survival rates for N0M0, N1M0 and N2M0
patients were 89.3%, 75.0% and 60.0%, respectively. Figure 1 depicts the survival curves for each degree
of nodal involvement. In the following survival analysis, we only compared N0 cases, in aim to exclude
the impact of lymph metastasis upon survival.
Macroscopically, intraluminal tumor size measured from 0.4 to 3.0 centimeters, averaging 1.4 ±0.1 cm.
By COX regression analysis, tumor size was not an independent risk factor for cancer death (P=0.70).
Obstructive atelectasis or pneumonia had been noted in 11 N0 patients, which was compatible in case
records and with gross pathologic findings. Overall five-year survival for this group was 100%. The
five-year survival rate was 87.8% for the remaining 56 patients. COX regression analysis demonstrated
that prognosis was not impacted by the presence of obstructive atelectesis or pneumonia (P=0.96).
Therefore, in these SELC cases, pneumonitis or atelectasis related to endobronchial obstruction exerted
no apparent influence upon survival.
According to the histological stratification methods used (the details of which already have been
described in ‘Patients and Methods’), there were 13, 12, 20, 21, and 15 patients distributed between levels
1 (airway epithelium) and 5 (bronchial cartilage) respectively. Table 1 presents the surgical-pathological
characteristics based upon invasion depth stratifications. Five-year survival was 100% among all patients
10
assigned to the inner three levels; and was 84.0% and 61.3% for levels four and five, respectively. Figure
2 depicts the survival curves corresponding to each involved circumstantial level. Survival between the
five levels was significantly different (P=0.02), indicating that circumstantial invasion depth has impact
upon survival.
Discussion
Superficial endobronchial lung cancer (SELC) has been documented rarely [1, 2]. It has been defined as
lung cancer that is limited to within the bronchial lumen without microscopically-apparent cartilaginous
ring involvement [1]. Up to date, almost all available data have pertained to its subtype early hilar lung
cancer (EHLC), which accounts for approximately 1.4-3.5% of all primary lung cancers [4, 5, 8]. The
major discrepancy between these two categories, EHLC and SELC, has been the involvement of lymph
node metastasis [3-5]. The clinical incidence of SELC is similarly low as EHLC, in our series only around
0.9% (81/8817) of all operative cases. As well, there have been few descriptions of its
surgical-pathological characteristics, its prognostic relevance, and its relevance related to staging. In the
current study, we calculated long-term survival rates in patients with SELC, paying particular attention to
the corresponding staging position of the SELC, and the influence of co-existent atelectasis or
pneumonitis, tumor size, and tumor circumstantial invasion depth.
Albeit only estimated in a few studies, the prognosis of SELC largely has been examined from the
perspective of EHLC [4-5, 8-10], which truly only represents a subgroup of SELC patients, specifically
11
referring to those endobronchially-confined lung cancers without lymph node or distant metastases.
EHLC has been associated with excellent long-term survival, and was suggested by Terzi et al. to
represent “a subgroup of IA with a more favorable prognosis” [4]. Similar data were generated by
Watanabe and associates, who reported 100% and 96.9% 5-year and 10-year survival rates in 43 EHLC
patients, respectively [9]. In Nakamura’s series, 46 EHLC patients exhibited an 89.3% postoperative
five-year survival rate [3]. Our data are consistent with these earlier reports, five-year survival being
89.6% among our N0M0 SELC patients. However, whereas the earlier data only suggest that EHLC
enjoys favorable survival which is the N0 subset of SELCs, no prior data have been available which
allows one to assess SELC survival relative to T status or in N1 or N2 disease (stages II-III).
Taking the whole SELC group into consideration, our data suggest that SELC warrants being considered
among the most favorable presentations, within the T1 subset of patients. The overall five-year survival in
the present series was 85.6%, and it was 89.3%、75.0% and 60.0% for N0M0, N1M0 and N2M0 patients,
respectively. These rates appear superior to the 65-80% 5-year survival rates observed among all
T1-2N0M0 cases and hilar type lung cancers, and the even worse outcomes observed in TxN1-2M0
patients [1, 7, 11]. When dealing with N0M0 SELC (which is the same as EHLC), Terzi and colleagues
suggested that this was a particular subset of ‘T1’ cancers that includes Tis (in situ), and both
microinvasive and invasive cancer limited to within the bronchial wall [4]. Adding our data to the mix, it
is reasonable to deduce that SELC occupies a special T1 position versus stages I to III. To date, we cannot
explain why SELC patients in stages II-III (N1- N2) also appear to have a relatively favorable prognosis.
Only 7 N1 and 7 N2 cases were included in our series, thereby limiting whatever conclusions we can
12
draw. Further research is warranted to assess the prognosis in this apparently small and misunderstood
subset of SELC patients.
Contrary to common sense, the existence of N1 and N2 cases suggests that SELC is not always early
stage malignancy. Though the neoplasm usually is grossly limited or even impalpable during operation [9,
12], we found that lymph node metastases occurred more commonly than expected. In the present series,
17.3% of the SELC patients had metastatic spread (14/81). Moreover, as Figure 1 demonstrates, nodal
metastases appear to have a significant impact upon prognosis. It therefore seems inappropriate to insist
on calling SELC an ‘early’ type of central lung cancer.
Again, in this study, we found that tumor size, and tumor-associated pneumonia or atelectesis had no
influence upon survival among SELC patients. A statistically-significant relationship between tumor
diameter and depth of intrabronchial invasion previously was described by Konaka et al. [13] In his paper,
seventy roentgenologically-occult cancers were included and 82.9% (29/35) of the lesions with maximum
diameters >10 mm were observed to be outside the cartilaginous wall. In our series, most SELC tumors
had maximum diameters which had surpassed 10 mm, which is not that surprising given that a greater
percentage of our patients were symptomatic at presentation. Due to the time length of cases distribution,
present data were not sufficiently supportive of the role of adjuvant chemo- or radio-therapies on patients’
survivals.
Intraluminal growth of SELC commonly leads to bronchial blockage and subsequent atelectasis or
13
pneumonia [2, 4, 5]. Its relationship with prognosis has not been enough documented previously. Terzi
and colleagues noticed that all their seven EHLC patients with tumor-determined atelectasis survived
beyond six years [4]. Ishda et al. [5] noted that six of his 8 patients with atelectesis remained alive for the
duration of follow up, the longest more than 12 years post-operatively. Two patients died, one of heart
failure and one of unknown cause at about one and a half years. Our study confirms that most patients
with atelectasis have good five year survival rate, as opposed to what might be predicted, given that
secondary atelectasis and pneumonia generally are considered to be T2 factors [1]. We speculate that
obstructive atelectasis secondary to endobronchial tumor growth differs from the obstruction caused by
extrabronchial tumor invasion and compression, since these two different conditions might be associated
with different levels of tendency towards metastasis. Therefore, once a patient meets the inclusion
criterion of primary tumor limited within bronchial cartilage, the appropriate label is T1, irrespective of
tumor associated radiological signs.
In the current context, we used the histological stratification method, dividing SELC into five levels, and
a distinct survival trend was observed. This indicates that tumor invasion depth is a crucial impact
variable in survival analysis. In clinical practice, it has been very important to collect serial tumor
sections for meticulous assessment of tumor invasion depth across the bronchial wall. Armed with the
information gleaned from this study, it appears that this information could have relevance with respect to
SELC patient outcomes.
References
14
1．Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;
111:1710-1717.
2．Han NJ, Song KS, Lee KH, Seo JB, Lee JS, Lim TH, et al. Superficial endobronchial lung cancer:
radiologic-pathologic correlation. Korean J Radiol 2002; 3: 229-234.
3．Nakamura H, Kawasaki N, Hagiwara M, Ogata A, Saito M, Konaka C, et al. Early hilar lung
cancer--risk for multiple lung cancers and clinical outcome. Lung Cancer. 2001; 33:51-7.
4． Terzi A, Pelosi G. Falezza G, Lonardoni A, Pasini F, Calabro F. Early Hilar Lung Cancer-clinical
aspect and long term survival. Identification of a subgroup of stage Ia patients with a more favorable
prognosis. Lung Cancer 2000; 27:119-124.
5． Ishida T, Inoue T, Sugio K, Inoue K, Inuzuka S, Tateishi M, et al. Early squamous lung cancer and
longer survival rates. Respiration. 1993; 60:359-65.
6．Mathur PN, Edell E, Sutedja T, Vergnon JM, American College of Chest Physicians. Treatment of
early stage non-small cell lung cancer. Chest 2003; 123:176s-180s.
7． Chang MY, Sugarbaker DJ. Surgery for early stage non-small cell lung cancer. Seminars Surg
Oncology 2003; 21:74-84.
8．Watanabe Y, Shimizu J, Oda M, Iwa T, Takashima T, Kamimura R, et al. Early Hilar Lung Cancer: Its
Clinical Aspect. Journal of Surgical Oncology 1991; 48:75-76.
9． Watanabe Y, Murakami S, Oda M, Ohta Y, Watanabe S, Nozaki Z, et al. Surgical management of
early stage central (hilar) and peripheral non-small cell lung carcinoma. Cancer 2000; 89:2438-2444.
10．
Shimizu J, Watanabe Y, Oda M, Ohta Y, Tsunezuka Y, Itoh Y, et al. Evaluation of sleeve
segmentectomy for early hilar lung cancer. Int Surg 2002; 87:53-9.
15
11． Naruke T, Goya T, Tsuchiya R, Suemasu K. Prognosis and survival in resected lung carcinoma
based on the new international staging system. Thorac Cardiovasc Surg 1988; 96:440-447.
12． Miyazu Y, Miyazawa T, Kurimoto N, Iwamoto Y, Kanoh K, Kohno N. Endobronchial
ultrasonography in the assessment of centrally located early-stage lung cancer before photodynamic
therapy. Am J Respir Crit Care Med 2002; 165: 832-837.
13． Konaka C, Hirano T, Kato H, Furuse K, Takada M, Saito Y, et al. Comparison of endoscopic
features of early-stage squamous cell lung cancer and histological findings. British J Cancer 1999;
80:1435-1439.
16
Tables
Table 1. Surgical pathological characteristics of 81 SELCs based on invasion depth stratifications.
Level 1
Level 2
Level 3
Level 4
Level 5
epithelial
submucosal
muscular
extramuscle
cartilage
layer
layer
layer
layer
N0
13
12
17
18
7
N1
0
0
3
2
2
N2
0
0
0
1
6
1
1
4
3
2
Average tumor dimension in N0 (cm) 1.2±0.6
1.5±0.6
1.4±0.4
1.4±0.5
1.6±0.7
Five-year survival (%) in N0 cases
100
100
84.0
61.3
Limited within (by definition)
Cases number
Atelectesis / pneumonia in N0
100
Figures
Fig. 1. Survival curves after resection of endobronchial superficial lung cancers according to N status.
17
Fig. 2. Survival curves according to each circumstantial tumor invasion level in N0 SELC patients.
No Conflict of Interests existed.
18