|
|
||||||||
Department of Thoracic Surgery and Pathology, Japanese Red Cross Nagaoka Hospital, Niigata, Japan
Accepted for publication June 16, 2009.
* Address correspondence to Dr Koike, Department of Thoracic Surgery, Japanese Red Cross Nagaoka Hospital, 2-297-1 Sensyu, Nagaoka, Niigata, 940-2085 Japan (Email: t-koike{at}kj8.so-net.ne.jp).
| Abstract |
|---|
|
|
|---|
Methods: We designed a prospective phase II study evaluating limited resection for noninvasive BAC diagnosed by intraoperative pathologic examination. From 1999 to 2007, limited resection was the procedure in 46 patients (16 men and 30 women; median age, 69 years; range, 49 to 83) who were diagnosed intraoperatively as having noninvasive BAC. The first end point was the predictive value of the intraoperative pathologic examination for noninvasive BAC diagnosis. The second end point was overall survival, disease-free survival, and cancer-specific survival, calculated using the Kaplan-Meier method.
Results: We performed wedge resections for 44 patients and segmentectomy for 2 patients. Permanent pathologic examination revealed 3 patients had primary lung adenocarcinomas other than noninvasive BAC. The predictive value of intraoperative pathologic examination for noninvasive BAC diagnosis was 94%. During a median 51-month follow-up, there were only 2 cancer unrelated deaths. The 5-year overall survival rate and the disease-free survival rate were 93%, and the 5-year cancer-specific survival rate was 100%.
Conclusions: The results of our prospective phase II study indicate that limited resection, mainly by wedge resection, is a potentially curative surgical procedure and may be an acceptable alternative to lobectomy for patients with noninvasive BAC. Furthermore, an intraoperative pathologic diagnosis of noninvasive BAC is strongly predictive and allows for an intraoperative decision to perform a limited resection in these patients.
| Introduction |
|---|
|
|
|---|
Owing to the advent of refined chest computed tomography (CT) images with high resolution and CT screening programs, smaller and fainter lung cancers are being discovered [5]. Bronchioloalveolar carcinoma (BAC) is a subtype of lung adenocarcinoma that replaces alveolar epithelial cells as it grows [6]. Noguchi and associates [7] examined the presence of in situ noninvasive peripheral adenocarcinomas and defined BAC as a form of adenocarcinoma with a pure bronchioloalveolar growth pattern and no evidence of invasion in the revised histologic classification of lung and pleural tumors in 1999 [8]. Many groups have since suggested in situ lung carcinomas such as BAC are potential indications for limited surgical resection [9–11].
BAC presents with ground-glass opacity (GGO) on high-resolution CT (HRCT), and CT nodule consistency closely reflects the proportion of BAC [12–14]. Radiologic findings of GGO, indicating a BAC component, reportedly represent a positive prognostic factor, and GGO findings on HRCT correlate well with Noguchi classifications [15]. Characterizing and quantifying CT findings is generally subjective, however, and based on visual estimation by individuals, leading to high potential for marked interobserver discrepancies. In addition, lesions with GGO on HRCT were not always noninvasive BAC [16, 17]; therefore, it is difficult to preoperatively diagnose noninvasive BAC and plan for limited surgical resection based solely on CT findings.
We designed this prospective phase II study for intentional limited resection for noninvasive BAC diagnosed by intraoperative pathologic examination. We evaluated the diagnostic capability of intraoperative pathologic examination and report the long-term results and feasibility of limited resection for noninvasive BAC.
| Patients and Methods |
|---|
|
|
|---|
In April 1999, we began enrolling patients as candidates for this study who satisfied the following criteria: (1) suspected NSCLC on CT or suspected noninvasive BAC by preoperative pathologic findings, (2) solitary, peripheral nodule, (3) cT1 or 2 N0 M0, (4) no history of previous treatment for lung cancer, (5) general medical condition and respiratory function adequate for lobectomy, and (6) diagnosed as having noninvasive BAC by intraoperative pathologic examination.
From April 1999 to September 2007, intraoperative pathologic examinations were performed for noninvasive BAC diagnosis on 118 patients, and 46 satisfied these criteria, were enrolled in this study, and underwent limited resection. We performed wedge resections by palpating for the tumor and were careful to remove the entire tumor. We chose segmentectomy when the tumor was deep and too close to the hilum to remove by wedge resection.
Our pathologists examined the frozen-section specimen immediately. The lung specimen was sliced at the largest tumor diameter. In tumors accompanied by a pleural indentation or a solid portion, slices were placed to pass through such areas, and the tumor was sliced with care to ascertain the relation between the tumor and the surgical margin. After macroscopic examination, the portions most likely to contain invasive cancer were embedded, stained with hematoxylin and eosin (H-E), and examined microscopically.
When we identified cancer cells as increasing and replacing the alveolar epithelial cells without infiltrating pleural or peribronchiolar tissues, the tumor was diagnosed as noninvasive BAC. Also diagnosed as noninvasive BAC were tumors that were accompanied by alveolar collapse but no disruption of the alveolar wall lattice structure consisting of elastic fibers or no microscopic foci of solid growth filling alveolar lumen. Our pathologists identified disruption of the alveolar wall lattice structure by using the difference in brightness between elastic and collagen fibers with H-E staining. We did not consider mucinous BAC as an indication for limited resection because mucinous BAC tends to spread and form satellite tumors or pneumonic consolidation and thus has a worse prognosis than nonmucinous BAC [18]. Thus, we only performed limited resection for patients with nonmucinous BAC.
If the tumor was diagnosed as noninvasive BAC and no cancer cells were identified at the surgical margin, limited resection was completed without systematic lymph node dissection or intraoperative pathologic lymph node examination. If the surgical margin was not sufficient, an additional margin was resected and pathologic examination was performed again, by the same method. If the tumor was diagnosed as a primary lung carcinoma other than noninvasive BAC, lobectomy with systematic lymph node dissection or lymph node sampling was performed.
The characteristics of the 46 patients (16 men, 30 women) undergoing limited resection are reported in Table 1. The median age was 69 years (range, 49 to 83 years). The median tumor size measured by preoperative radiologic imaging was 1.6 cm (range, 0.8 to 3.0 cm). The preoperative CT findings showed 13 patients (28%) had tumors larger than 2 cm in diameter and 26 (57%) presented less than 50% GGO contents. Preoperative diagnosis was by CT-guided needle biopsy in 12 patients (26%), but 34 (74%) did not have a diagnosis of lung carcinomas.
|
As the first end point, we established the predictive value of intraoperative frozen-section examination for noninvasive BAC diagnosis. It was defined as the probability that a patient diagnosed as having noninvasive BAC by intraoperative pathologic examination would also be diagnosed as having noninvasive BAC by permanent pathologic examination. We considered that the predictive value of intraoperative frozen-section examination for noninvasive BAC diagnosis to be especially important because this value would be associated with the number of patients undergoing limited resection for primary lung adenocarcinomas other than noninvasive BAC. The second end point was overall survival, disease-free survival, and cancer-specific survival calculated using the Kaplan-Meier method.
| Results |
|---|
|
|
|---|
By permanent pathologic examination, 3 patients were diagnosed as having primary lung adenocarcinomas other than noninvasive BAC and were considered to not have had an indication for intentional limited resection. Therefore, the predictive value of intraoperative frozen-section examination for noninvasive BAC diagnosis was 94%. The tumors in these 3 patients were smaller than 2 cm in diameter, and the tumors in 2 of these patients had GGO contents of more than 50% on CT. We recommended that these 3 patients undergo completion lobectomy, but they refused and were followed up as outpatients without adjuvant chemotherapy or radiotherapy.
The rest of the patients had adenocarcinomas with a pure bronchioloalveolar growth pattern, with neither active fibroblastic proliferation nor evidence of pleural, vascular, or lymphatic involvement, according to the permanent pathologic examination. The tumors in all 46 patients were at pathologic stage I.
The median postoperative hospital stay was 10 days (range, 3 to 17 days). Postoperative complications included pneumothorax, gastric ulcer, neurogenic bladder, and a small wound dehiscence without infection, but none of the complications occurred that are often experienced after lobectomy, such as prolonged air leakage, pneumonia, chronic respiratory failure, and supraventricular tachycardia. There were no deaths associated with this operation.
The overall survival curve and the disease-free survival curve for the 46 patients undergoing limited resection are shown in Figure 1. During the median follow-up of 51 months, 2 deaths unrelated to cancer occurred. Accordingly, the overall survival curve and the disease-free survival curve for the 46 patients were the same. The 5-year overall survival rate and the disease-free survival rate were 93%. The 5-year cancer-specific survival rate was 100%. The 5-year overall survival rate and the disease-free survival rate for the 44 patients who underwent wedge resection were 93% (data not shown).
|
| Comment |
|---|
|
|
|---|
BAC is a subtype of lung adenocarcinoma that grows and replaces alveolar epithelial cells [6]. Noguchi and associates [7] classified lung adenocarcinomas into 6 categories on the basis of histologic characteristics and defined localized BAC (type A) and localized BAC with foci of collapse (type B) as pathologically and biologically in situ noninvasive peripheral carcinomas because no cases of type A and B showed lymph node metastasis and the 5-year survival rate for patients with these tumors was 100% [7]. In the revised histologic classification of lung and pleural tumors in 1999, BAC was clearly defined as a form of adenocarcinoma with a pure bronchioloalveolar growth pattern and no evidence of stromal, vascular, or pleural invasion [8]. Because these noninvasive BACs showed no metastasis or micrometastasis to mediastinal, hilum, or regional lymph nodes [20, 21], it was suggested that an in situ lung carcinoma such as noninvasive BAC is a potential indication for limited resection [9–11].
Several reports on prospective studies of limited resection for noninvasive BAC with intraoperative pathologic examination are summarized in Table 2 [15, 22–24]. Mainly wedge resection was performed in these studies and our study, and no cancer-related deaths or recurrences were seen during follow-up in all studies, including our study. Consequently, limited resection, mainly by wedge resection, was a potentially curative surgical procedure for noninvasive BAC.
|
In trials that have been conducted of limited resection for small lung nodules with abundant GGO contents on CT [25, 26], the postoperative prognosis was good because most resected tumors were noninvasive BAC or atypical adenomatous hyperplasia. However, among 70 tumors no larger than 1.5 cm seen as pure GGO, 8 (11.4%) were diagnosed as having primary lung adenocarcinomas other than noninvasive BAC (pure GGO: consisting of only homogeneous translucent density) [26]. On the other hand, in a study on the correlation between HRCT and pathologic findings, 8 tumors (11.3%) were diagnosed as Noguchi type A or B among 71 solid-density tumors no larger than 2 cm (solid-density: areas of the tumor opacity on mediastinal window images were more than half of those on lung window images) [27].
Therefore, if a decision for limited resection was based solely on CT findings, incorrect limited resection for primary lung adenocarcinomas other than noninvasive BAC could occur for tumors found to have a large proportion of GGO on CT, and formal lobectomy could occur for noninvasive BAC tumors found to have a small proportion of GGO or not so small in diameter on CT. Furthermore, it was suggested that to diagnose noninvasive BAC and determine the indications for limited resection based solely on preoperative CT findings would be difficult.
We thus determined the indication for limited resection based on the diagnosis made by intraoperative pathologic examination. Certainly, some discrepancies occurred between the results of intraoperative frozen-section diagnosis and the final pathologic diagnosis. We revised 3 of 46 patients to a diagnosis of primary lung adenocarcinoma other than noninvasive BAC after permanent pathologic examination. We considered that the predictive value of intraoperative frozen-section examination for noninvasive BAC diagnosis was especially important, because this value would be associated with the number of patients undergoing incorrect limited resection for primary lung adenocarcinomas other than noninvasive BAC.
In some prospective studies of limited resection for noninvasive BAC with intraoperative pathologic examination, the predictive value of intraoperative frozen-section examination for noninvasive BAC diagnosis was quite high at 93% to 100%, and BAC with no invasive features was almost precisely diagnosed by frozen-section (Table 2). Yoshida and associates [24] reported that with customized stapling cartridges, negative-pressure specimen preparation and inflation, stereoscopic microscopy, Victoria blue-van Gieson staining, and a skilled pathologist, frozen-section classification of Noguchi subtype remained highly accurate.
In addition, because we performed intraoperative pathologic examinations both for patients with adenocarcinomas suspected to be noninvasive BAC and all patients suspected to have NSCLC regardless of GGO content and tumor diameter on CT, it is reasonable to perform limited resections for not so small noninvasive BAC and noninvasive BAC with minor GGO contents on CT. In fact in our study, 25 of 43 patients (58%) diagnosed with noninvasive BAC by permanent pathologic examination presented with GGO contents smaller than 50% on CT and 13 (30%) were larger than 2 cm in diameter on CT.
On the other hand, it was difficult to predict the 3 patients revised to having a primary lung adenocarcinoma other than noninvasive BAC after permanent pathologic examination by preoperative radiologic findings because all 3 had tumors smaller than 2 cm in diameter and 2 had tumors with GGO contents more than 50% on CT. Furthermore, because our result of the predictive value of intraoperative frozen-section examination for noninvasive BAC diagnosis was quite high (94%) and not inferior to the results in other similar reports (Table 2), the criteria as candidates for our procedure did not affect the accuracy of the intraoperative pathologic examination for noninvasive BAC diagnosis.
If patients are diagnosed as having primary lung adenocarcinomas other than noninvasive BAC by permanent pathologic examination, limited resection is not considered an adequate surgical treatment. In fact, there has been a report that local recurrence occurred in a patient who had a tumor with a GGO ratio of 50% but exhibited vessel infiltrations on permanent pathologic examination after wedge resection [28]. Therefore, we need to recommend that such patients undergo a completion lobectomy.
This study has limitations because the patient number and the follow-up period were insufficient to reach a definitive conclusion. Further study and long-term follow-up will be necessary to determine whether limited resection can serve as a curative procedure for noninvasive BAC. Furthermore, the predictive value of intraoperative pathologic examination for noninvasive BAC diagnosis was not 100%, and some patients underwent limited resection for primary lung adenocarcinomas other than noninvasive BAC. We are going to attempt to further improve the predictive value by increasing our experience in this procedure.
In conclusion, the results of our prospective phase II study indicate that limited resection, mainly by wedge resection, is a potentially curative surgical procedure and may be an acceptable alternative to lobectomy for noninvasive BAC patients. Furthermore, an intraoperative pathologic diagnosis of noninvasive BAC is strongly predictive and allows for the intraoperative decision to perform a limited resection in these patients.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
P. E. Van Schil, H. Asamura, V. W. Rusch, T. Mitsudomi, M. Tsuboi, E. Brambilla, and W. D. Travis Surgical implications of the new IASLC/ATS/ERS adenocarcinoma classification Eur. Respir. J., February 1, 2012; 39(2): 478 - 486. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Schuchert, A. Kilic, A. Pennathur, K. S. Nason, D. O. Wilson, J. D. Luketich, and R. J. Landreneau Oncologic Outcomes After Surgical Resection of Subcentimeter Non-Small Cell Lung Cancer Ann. Thorac. Surg., June 1, 2011; 91(6): 1681 - 1688. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Iwasaki, M. Ohta, Y. Shintani, and N. Ikeda Successful intentional lobectomy for lung cancer after treating contralateral diaphragmatic eventration Interact CardioVasc Thorac Surg, June 1, 2010; 10(6): 1049 - 1050. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Sculier, T. Berghmans, and A. P. Meert Update in Lung Cancer and Mesothelioma 2009 Am. J. Respir. Crit. Care Med., April 15, 2010; 181(8): 773 - 781. [Full Text] [PDF] |
||||
![]() |
M. Okada Invited Commentary Ann. Thorac. Surg., October 1, 2009; 88(4): 1111 - 1111. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |