Although ipsilateral intrapulmonary metastasis (PM), Evaluation of TMN Classification for Lung Carcinoma With Ipsilateral Intrapulmonary Metastasis

Transcription

1 Evaluation of TMN Classification for Lung Carcinoma With Ipsilateral Intrapulmonary Metastasis Morihito Okada, MD, Noriaki Tsubota, MD, Masahiro Yoshimura, MD, Yoshifumi Miyamoto, MD, and Reiko Nakai, MD Department of Thoracic Surgery, Hyogo Medical Center for Adults, Hyogo, Japan Background. Staging for lung cancer based on the TNM classification is an important predictive factor for prognosis. Recently, lung cancer with ipsilateral intrapulmonary metastasis (PM) was reclassified according to the revision of the TNM classification. To evaluate the prognostic importance of the new staging system for PM, we analyzed the postoperative survival of patients with non small cell lung carcinoma. Methods. Of 1,002 consecutive patients who underwent operation for primary lung cancer between June 1984 and December 1996, we reviewed the medical record of 889 patients who underwent complete resection for non small cell lung cancer. Results. We considered 89 patients (10.0%) to have synchronous ipsilateral PM. After reclassification to the former staging system revised in 1992, 5 patients were classified as stage I, 29 as stage IIIA, 48 as stage IIIB, and 7 as stage IV. In the new staging system revised in 1997, 48 patients were recategorized as stage IIIB, and 41 as stage IV. The 5-year survival of patients without PM (49.5%) was significantly better than that of patients with PM in primary-tumor lobe (29.6%, p 0.002) or in nonprimary-tumor ipsilateral lobe (23.4%, p ). Although the survival of patients with stage IV cancer without PM was significantly worse than that of patients with the new (1997) stage IV cancer with PM (p 0.02), it was similar to that of patients with the former (1992) stage IV cancer with PM. The survival of PM patients with N0 or N1 disease was significantly better than that of PM patients with N2 or N3 disease (p 0.001). Furthermore, in patients with the new (1997) stage IIIB cancer, the survival of N0 disease was better than that of N2 disease (p 0.007). Conclusions. Inasmuch as the prognosis of non small cell carcinoma in patients with PM strongly correlated with N factor rather than PM factor, N factor should be reflected in a staging designation. We therefore consider the new TNM classification for PM reclassified in 1997 to be less acceptable for surgical-pathologic staging than the revision in (Ann Thorac Surg 1999;68:326 31) 1999 by The Society of Thoracic Surgeons Although ipsilateral intrapulmonary metastasis (PM), which is one of the most common findings in non small cell cancer, was previously thought to influence adversely the prognosis as metastasis spread hematogenously to other organs classified as M1 [1, 2], it may have less impact on prognosis than extrapulmonary metastasis if PM is totally removed [3, 4]. In 1992, the Union Internationale Contre le Cancer and American Joint Committee on Cancer revised the TNM classification for lung cancer with PM by upgrading T factor as a local progression instead of M factor. Briefly, T factor was upgraded by a single unit if PM was located at the primary lobe, or was determined as T4 if PM occurred in the other lobes of the ipsilateral lung [5, 6]. However, the 1997 recommendations for staging have included new rules, namely that satellite tumors within the primarytumor lobe of the lung should be classified T4 and that Accepted for publication Feb 3, Address reprint requests to Dr Tsubota, Department of Thoracic Surgery, Hyogo Medical Center for Adults, Kitaohji-cho 13-70, Akashi City 673, Hyogo, Japan; n-tsubo@sanynet.ne.jp. ipsilateral PM in a distant, that is nonprimary-tumor, lobe of the lung should be classified M1 [7] (Table 1). In this study, our goal is to see whether the new staging system might be adequate through surgical-pathologic findings and postoperative prognoses of patients with synchronously detected and resected ipsilateral PM lesions. Material and Methods The medical records of all patients who underwent operation for primary lung cancer at our institution between June 1, 1984, and December 31, 1996, were reviewed. Among 1,002 consecutive patients, a total of 889 underwent complete resection for non small cell lung cancer. Intrapulmonary metastasis was defined as an independent tumor having the same histopathologic features as the primary one. The following criteria, based on and modified from those of Martini and Melamed [8], were used for designation of multiple primary lung cancers: (1) the tumors were anatomically separate (different lobes or lungs); and (2) the tumors were histolog by The Society of Thoracic Surgeons /99/$20.00 Published by Elsevier Science Inc PII S (99)

2 Ann Thorac Surg OKADA ET AL 1999;68: LUNG CANCER WITH INTRAPULMONARY METASTASIS 327 Table 1. Changes of the TNM Classification for Non Small Cell Lung Cancer Regarding Ipsilateral Intrapulmonary Metastasis Lesion Location pm1 M1 T factor upgraded T4 by a single unit pm2 M1 T4 M1 pm1 intrapulmonary metastasis in the lobe in which the primary lesion was located; pm2 intrapulmonary metastasis in the ipsilateral lobe in which the primary lesion was not located. ically different. If the tumors were histologically the same, they must have no systemic metastases or mediastinal spread. In cases in which all lesions were squamous cell carcinomas, a classification was accepted if a separate origin from an area of carcinoma in situ could be identified for each [9]. The tumors that met these criteria fully were excluded as multiple primary lesions. All cancers were staged using the criteria of the International Staging System for Lung Cancer [1, 2, 5 7], and staging was based on data obtained from imaging, bronchoscopy, invasive diagnostic techniques, operative findings, and pathologic examination. Careful intraoperative staging at the time of surgical resection was performed by dissecting intrapulmonary, hilar, and mediastinal nodes. The TNM subsets are used as follows: T1 N0 M0 or T2 N0 M0, stage I; T1 N1 M0 or T2 N1 M0, stage II; T3 N0 M0, T3 N1 M0, T3 N2 M0, T1 N2 M0, T2 N2 M0, or T3 N2 M0, stage IIIA; T4 N1 M0, T4 N2 M0, T4 N3 M0, T1 N3 M0, T2 N3 M0, or T3 N3 M0, stage IIIB; T1 N1 M1, T1 N2 M1, T1 N3 M1, T2 N1 M1, T2 N2 M1, T2 N3 M1, T3 N1 M1, T3 N2 M1, T3 N3 M1, T4 N1 M1, T4 N2 M1, or T4 N3 M1, stage IV. The changes of the TMN classification for PM are shown in Table 1. According to the TNM system revised in 1992 [5, 6], PM in the ipsilateral lung was alternatively included in T factor, instead of M1 in the TNM classification produced in 1986 [1, 2]. If PM was located in the primary lobe, T factor was upgraded by a single unit, and if PM was done in the other lobes of the ipsilateral lung, it was determined as T4. Furthermore, the latest revision in 1997 proposed that satellite tumors within the primary-tumor lobe should be classified T4 and PM in nonprimary-tumor lobes should be classified M1 [7]. On the other hand, PM in the contralateral lung, which was considered as distant metastasis, was included in M1 disease. In the ipsilateral lung, including the lobe in which the primary lesion was located, the presence of PM was first checked macroscopically by close palpation during operation. Postoperatively, surgeons, radiologists, and pathologists together checked satellite nodules macroscopically and by palpation in the fixed specimen sliced 1 cm in thickness. Suspicious nodules were evaluated microscopically in detail. Histologic typing was done according to the World Health Organization classification [10]. All cases of bronchioloalveolar carcinoma were excluded just because of the question of multicentricity of those primary cancers. Generally, the histopathology of adenomatous hyperplasia and atypical adenomatous hyperplasia is similar to that of PM of adenocarcinoma. We therefore paid attention to excluding adenomatous hyperplasia and atypical adenomatous hyperplasia. All patients underwent a surgical resection of the main and metastatic pulmonary lesions with a dissection of hilar and mediastinal lymph nodes. In patients with stage IV cancer, the distant metastatic lesions were treated with surgical resection or radiotherapy. All patients had a chest roentgenogram, a history, and a physical examination made at the follow-up visit, which took place at least twice a year. All deaths including the operative death were presumed to be associated with lung cancer, and the analysis included all types of death. The final date for evaluation was December 31, Overall follow-up ranged from 12 to 163 months, with a median of 69 months. Cumulative survival rates were calculated by the Kaplan-Meier estimation using the date of pulmonary resection as the starting point and the date of death or last follow-up as the end point, and differences in survival were determined by log-rank analysis. A multivariate analysis of independent prognostic factors was performed with Cox s proportional hazards regression model. Results Of 889 patients, 89 (10.0%) were histopathologically demonstrated to have ipsilateral and synchronous PM. The age of these patients ranged from 42 to 78 years, with an average of years. Fifty-five were men and 34 were women. Pathologic examination of the resected specimens showed adenocarcinoma in 71 (79.8%), squamous cell carcinoma in 17 (19.2%), and adenosquamous carcinoma in 1 (1.1%). Fifty-one patients (57.4%) had PM in the lobe in which the primary lesion was located (pm1), and 38 (42.6%) in an ipsilateral different lobe (pm2). Forty-two of 51 pm1 lesions (82.4%) were diagnosed incidentally by examination of the resected specimen and the other 9 (17.6%) were discovered during operation. On the other hand, 4 of 38 pm2 lesions (10.5%) were detected preoperatively, 22 (57.9%) were discovered during operation, and 12 (31.6%) were diagnosed postoperatively. Pneumonectomy including sleeve procedure was performed in 7 patients, lobectomy including sleeve technique in 69, and segmentectomy including partial resection in 13. Regarding N classification, 32 patients had N0 disease; 17, N1 disease; 33, N2 disease; and 7, N3 disease. Concerning staging of 800 patients without PM (pm0), 400 were classified as stage I, 109 as stage II, 196 as stage IIIA, 72 as stage IIIB, and 23 as stage IV (Table 2). Reclassification of patients with PM is shown in Table 2. In the 1986 staging system, all the patients with PM belonged to stage IV. After reclassification of the patients with PM to the 1992 staging system, 5 patients were classified as stage I, 29 as stage IIIA, 48 as stage IIIB, and 7 as stage IV. In the 1997 staging system, 48 patients were recategorized as stage IIIB, and 41 as stage IV. The 5-year survival of patients with pm0 (49.5%) was significantly better than that of patients with pm1 (29.6%,

3 328 OKADA ET AL Ann Thorac Surg LUNG CANCER WITH INTRAPULMONARY METASTASIS 1999;68: Table 2. Revision of the TNM Classification for Non Small Cell Lung Cancer Regarding Ipsilateral Intrapulmonary Metastasis (n 889) Stage PM ( ) (n 800) 1986 PM ( ) (n 89) I II IIIA IIIB IV PM ipsilateral intrapulmonary metastasis. Contralateral intrapulmonary metastasis was included in M1 and was not included in PM. p 0.002) or pm2 (23.4%, p ). There was no significant difference in survival between patients with pm1 and those with pm2 ( p 0.79) (Fig 1). The 5-year survival of PM patients with adenocarcinoma (26.8%) was significantly better than that of PM patients with squamous cell carcinoma (16.8%, p 0.02). There was no significant difference in survival between PM patients with well-differentiated adenocarcinoma and those with more poorly differentiated adenocarcinoma ( p 0.93) (Fig 2). In the 1997 staging system, the survival at 5 years of patients with stage IIIB and stage IV cancer with PM was 31.4% and 21.7%, respectively, although no significant difference was observed between the two groups ( p 0.31) (Fig 3). The 5-year survival of patients with stage IV cancer without PM (10.8%) was significantly worse than that of patients with the 1986 stage IV cancer with PM (26.5%, p ) or those with 1997 stage IV cancer with PM (21.7%, p 0.02). However, there was no significant difference in survival between patients with stage IV cancer without PM and those with 1992 stage IV cancer with PM ( p 0.58) (Fig 4), supporting the finding Fig 2. Actuarial survival curves of 88 patients with intrapulmonary metastasis grouped by pathologic types, except 1 patient with adenosquamous carcinoma. The survival of patients with intrapulmonary metastasis with adenocarcinoma was significantly better than that of patients with intrapulmonary metastasis with squamous cell carcinoma (p 0.02). that the survivals of the two groups were almost the same. The 5-year survival of PM patients with N0, N1, N2, and N3 disease was 45.4%, 44.3%, 6.6%, and 0%, respectively. The survival of PM patients with N0 or N1 disease was significantly better than that of PM patients with N2 or N3 disease ( p 0.001) (Fig 5). In patients with 1997 stage IIIB cancer with PM, the 5-year survival of N2 disease (0%) was significantly worse than that of N0 disease (67.1%, p 0.007) but was not significantly different from that of N1 disease (61.4%, p 0.08) (Fig 6). The multivariate analysis revealed that the prognosis of patients with non small cell carcinoma strongly correlated with N factor (hazard ratio, 3.267; 95% confidence interval, to 4.002; p ) rather than PM factor (hazard ratio, 1.425; 95% confidence interval, to 1.897; p ). Fig 1. Actuarial survival curves of 889 patients grouped by presence or absence of intrapulmonary metastasis, and its location. The survival of patients without intrapulmonary metastasis (pm0) was significantly better than that of patients with intrapulmonary metastasis in primary-tumor lobe (pm1, p 0.002) or in nonprimarytumor ipsilateral lobe (pm2, p ). Fig 3. Actuarial survival curves of 89 patients with intrapulmonary metastasis grouped by the 1997 staging system. No significant difference in survival was observed between patients with stage IIIB cancer and those with stage IV cancer (p 0.31).

4 Ann Thorac Surg OKADA ET AL 1999;68: LUNG CANCER WITH INTRAPULMONARY METASTASIS 329 Fig 4. Actuarial survival curves of patients with stage IV cancer grouped by presence or absence of intrapulmonary metastasis, and by various revisions of staging system. The survival of patients with stage IV cancer without intrapulmonary metastasis was significantly worse than that of patients with 1986 stage IV cancer with intrapulmonary metastasis (p ) or those with 1997 stage IV cancer with intrapulmonary metastasis (p 0.02), although there was no significant difference in survival between patients with stage IV cancer without intrapulmonary metastasis and those with 1992 stage IV cancer with intrapulmonary metastasis (p 0.58). Comment Fig 5. Actuarial survival curves of 89 patients with intrapulmonary metastasis grouped by pathologic N factor. The survival of patients with N0 or N1 disease was significantly better than that of patients with N2 or N3 disease (p 0.001). Fig 6. Actuarial survival curves of 46 patients with intrapulmonary metastasis who were assigned stage IIIB cancer on the 1997 staging system, grouped by pathologic N factor, except for 2 patients with N3 disease. The survival of patients with N0 disease was better than that of patients with N2 disease (p 0.007). Concerning staging of lung cancer, which serves as a guide to treatment and an indicator of prognosis, we have often been in situations that would create confusion. One such situation is the presence of plural pulmonary tumors. Although some researchers [8 10] have proposed various criteria to define multiple primary lung cancer, it is still difficult to distinguish synchronous multiple primary cancer from PM when the histologic type is identical. As a result, there is a fair chance for a contamination of multiple primary lung cancers existing in patients with PM. Deslauriers and colleagues [11] defined satellite nodules in resected lung carcinoma specimens as well-circumscribed carcinomatous foci adjacent to but clearly separated from the main tumor by normal lung parenchyma. Our examinations for satellite nodules by palpation and on slices of lung are effective for detecting PM, and may be a more extensive evaluation than most pathologists could routinely perform on lung cancer specimens. Miller [12] and Barsky and associates [13] raised the possibility that adenocarcinoma with bronchioloalveolar features of PM might represent multifocal bronchioloalveolar carcinoma rather than PM. The reason for an excellent survival of these cases raises questions about the biologic behavior of bronchioloalveolar carcinoma as compared with other types of adenocarcinoma. Therefore, we excluded all cases of bronchioloalveolar carcinoma from this study. Nakajima and coworkers [14] stated in their detailed histopathologic study that 30% of PM lesions might be adenomatous or atypical adenomatous hyperplasias, and patients with such lesions also exhibited a long postoperative survival. It is hard to distinguish adenomatous hyperplasia or atypical adenomatous hyperplasia from PM when the main lesion was well-differentiated adenocarcinoma. Indeed, atypical adenomatous hyperplasia has been regarded as a precancerous lesion, and its morphology is very similar to that of the peripheral part of well-differentiated adenocarcinoma. We demonstrated no significant differences in survival between PM patients with well-differentiated adenocarcinoma and those with a more poorly differentiated one, supporting little chance for a contamination of adenomatous hyperplasia or atypical adenomatous hyperplasia. The TNM classification of patients with ipsilateral PM is difficult, because of the small number of patients identified for end result studies. Several reports that led to the current rules for classifying patients with PM may provide insight into the complexity of developing internationally acceptable rules for classification. Deslauriers and associates [11], in the first published study of the influence of satellite nodules on prognosis, showed that

5 330 OKADA ET AL Ann Thorac Surg LUNG CANCER WITH INTRAPULMONARY METASTASIS 1999;68: satellite nodules, which they considered a manifestation of local disease, had a deleterious effect on prognosis and recommended that they be classified as T3. Actually, patients with non small cell carcinoma with PM have a better prognosis than those with other types of distant metastasis if a surgical resection is possible [3, 4]. These data suggested that PM might be an intrathoracic disease if no distant metastasis or invasion to the neighboring structures was observed, supporting that the incidence of PM in the ipsilateral lung was alternatively included in the T factor instead of M factor. On the other hand, Naruke and colleagues [15] supported a new rule in 1997 that the satellite tumor within the primary-tumor lobe should be classified T4 and ipsilateral PM in nonprimarytumor lobe should be classified M1. As a result of these studies, the American Joint Committee on Cancer and the Union Internationale Contre le Cancer revised the TNM classification of lung cancer with ipsilateral PM in 1992 [5, 6] and in 1997 [7]. In our study, we evaluated the prognosis of patients with non small cell carcinoma with ipsilateral PM to investigate the appropriateness of the TNM classification. The prognosis of our patients with PM was comparable with that of subgroups in the same stage revised in 1992 rather than in the same stage revised in 1997 or before Analysis showed that the prognosis of patients with non small cell carcinoma strongly correlated with N factor as compared with PM factor, supporting that N factor rather than PM factor should be reflected in a designation of the TNM staging system. In patients with PM who underwent resection, the ones with N0 or N1 disease did better at 5 years than those with N2 disease, confirming that nodal status is the best predictor of survival in resected lung cancer. According to the 1997 staging system, no significant difference was observed between the survival of patients with stage IIIB cancer with PM and those with stage IV cancer with PM. That was the reason that N factor would be utterly ignored in the 1997 staging system. Based on surgical-pathologic staging and the postoperative survival data, we consider the new classification for PM to be less acceptable than that revised in Naruke and coworkers [15] revealed that 5-year survival for all patients were stage I, 68.5%; stage II, 46.9%; stage IIIA, 26.1%; stage IIIB, 9.0%; and stage IV, 11.2%, and that 5-year survival rates for patients with pm1 and pm2 were 17.8% and 8.3%, respectively. Although the survival of all their patients grouped by stage was similar to that of our series (stage I, 67.2%; stage II, 49.4%; stage IIIA, 30.4%; stage IIIB, 16.7%; and stage IV, 10.8%), the survival of their patients with PM was much worse compared with that of ours (pm1, 29.6% and pm2, 23.4%) which was comparable with that of the studies by Shimizu and associates [16], Yoshino and coworkers [17] and Fujisawa and colleagues [18]. This significant difference in survival may be because of closer attention to detect PM lesion during and after the operation, which would affect stage bias. Although we discovered ipsilateral PM lesions in 10.0% of total patients, Naruke and associates [15] only found them in 5.9%. Checking for the presence of PM lesions by palpation during operation and in the fixed specimen sliced thinly after operation might provide the difference. Because these repeated modifications of the TNM staging system are certain to be a source of confusion to many, we must give attention to collecting significantly more data to confirm any classification rule. We could regard non small cell carcinoma with ipsilateral PM as representing rather locally advanced disease than systematic disease. With some certainty we can therefore recommend that patients with satellite lesions should be considered for surgical resection unless otherwise contraindicated. References 1. Hermanek P. TNM classification of malignant tumours, 4th fully rev ed. Berlin: Springer-Verlag, 1987: Mountain CF. The new international system for staging lung cancer. Chest 1986;89:225S 33S. 3. Naruke T, Goya T, Tuchiya R, Suemasu K. Prognosis and survival in resected lung carcinoma based on the new international staging system. J Thorac Cardiovasc Surg 1988; 96: Kunitoh H, Eguchi K, Yamada K, et al. Intrapulmonary sublesions detected before surgery in patients with lung cancer. Cancer 1992;70: American Joint Committee on Cancer. Manual for staging of cancer, 4th ed. Philadelphia: Lippincott, Mountain CF, Libshitz HI, Hermes KE. Lung cancer. A handbook for staging and imaging. Houston, TX: Charles P. Young, 1992: Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111: Martini N, Melamed MR. Multiple primary lung cancers. J Thorac Cardiovasc Surg 1975;70: Okada M, Tsubota N, Yoshimura M, Miyamoto Y. Surgical approach for multiple primary lung carcinomas. J Thorac Cardiovasc Surg 1998;115: World Health Organization histologic typing of lung tumours, 2nd ed. Am J Clin Pathol 1982;77: Deslauriers J, Brisson J, Cartier R, et al. Carcinoma of the lung. Evaluation of satellite nodules as a factor influencing prognosis after resection. J Thorac Cardiovasc Surg 1989;97: Miller RR. Bronchioloalveolar cell adenomas. Am J Surg Pathol 1990;14: Barsky SH, Grossman DA, Ho J, Holmes EC. The multifocality of bronchioloalveolar cell carcinoma: evidence and implications of a multiclonal origin. Mod Pathol 1994;7: Nakajima J, Furuse A, Oka T, Kohno T, Ohtsuka T. Excellent survival in a subgroup of patients with intrapulmonary metastasis of lung cancer. Ann Thorac Surg 1996;61: Naruke T, Tuchiya R, Kondo H, Asamura H, Nakayama H. Implications of staging in lung cancer. Chest 1997;112:242S 8S. 16. Shimizu N, Ando A, Date H, Teramoto S. Prognosis of undetected intrapulmonary metastasis in resected lung cancer. Cancer 1993;71: Yoshino I, Nakanishi R, Osaki T, et al. Postoperative prognosis in patients with non small cell cancer with synchronous ipsilateral intrapulmonary metastasis. Ann Thorac Surg 1997;64: Fujisawa T, Yamaguchi Y, Satoh Y, Hiroshima K, Ohwada H. Blood and lymphatic vessel invasion as prognostic factors for patients with primary resected nonsmall cell carcinoma of the lung with intrapulmonary metastasis. Cancer 1995;76: