Prognostic Significance of Computed Tomography in Resected N2 Lung Cancer Irene J. Cybulsky, MD, Louis A. Lanza, MD, M. Bernadette Ryan, MD, Joe B. Putnam, Jr, MD, Marion M. McMurtrey, MD, and Jack A. Roth, MD Department of Thoracic Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas

We reviewed 124 patients from 1982 to 1988 who had a resected primary non-small cell lung cancer metastatic to mediastinal (N2) lymph nodes and a preoperative assessment of the mediastinum with computed tomography of the chest. Sixty-three patients studied had computed tomographic evidence of mediastinal lymph node enlargement. In these patients the survival at 5 years was only 6.6%, compared with the 5-year survival of 13.5% in 61 patients in whom the mediastinum was normal. Plain

chest roentgenography with evidence of mediastinal adenopathy did not predict a poorer outcome. In addition, patients with tumors located in the left upper lobe were found to have an improved survival. These patients had a 5-year survival of 20.8%. Tumor histology, central location of the tumor, extranodal extension, and type of resection did not result in a significant survival difference. (Ann Thorac Surg 1992;54:533-7)

R

Material and Methods

esectability and outcome in patients with non-small cell lung cancer frequently centers on the presence and extent of nodal involvement. Patients who have metastases in mediastinal (N2) lymph nodes have a poorer prognosis. Nevertheless, selected patients with stage IIIa disease have in some series achieved 5-year survival rates greater than 30% [l-31. A computed tomographic (CT) scan of the chest allows better definition of the primary lesion and more accurate assessment of mediastinal lymphadenopathy [4-61. Correlation of nodal size with lymph node involvement by malignancy, however, is variable [7]. In the absence of mediastinal lymphadenopathy it is accepted practice to proceed to thoracotomy and resection [8]. If the CT scan shows resectable ipsilateral lymphadenopathy, some surgeons will proceed with pulmonary resection and complete mediastinal node dissection [l]. Evidence on the chest roentgenogram of mediastinal lymphadenopathy is associated with a poor prognosis after resection [9]. Patients with chest roentgenographic evidence of mediastinal lymph node involvement ("clinical N2") had only an 8% 3-year actuarial survival [9]. The prognostic significance of visibly enlarged lymph nodes on computed tomograms of patients with documented mediastinal lymph node metastases is unknown. We reviewed patients who underwent pulmonary resection for primary non-small cell lung cancer with malignant involvement of their mediastinal nodes who were staged with a preoperative CT scan to determine if mediastinal lymphadenopathy detected by CT scan is an adverse prognostic sign. Accepted for publication Feb 13, 1992 Address reprint requests to Dr Roth, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 109, Houston, TX 77030.

0 1992 by The Society of Thoracic Surgeons

A review was performed of all patients between Jan 1, 1982, and Dec 31, 1988, who underwent resection for primary lung cancer with involved (N2) mediastinal lymph nodes and who had a preoperative CT scan of the chest. Omitted were 22 patients who did not have a preoperative CT scan, 6 with small cell carcinoma, 7 with synchronous bilateral primary lung cancers, 4 who had received preoperative chemotherapy, and 4 who were found to have unresectable disease at thoracotomy. One hundred twenty-four patients comprise this study group. During this period all patients, including those with bulky lymphadenopathy, without evidence of metastatic disease outside of the chest, supraclavicular or axillary adenopathy, or metastatic lymph node involvement in the contralateral chest (N3 disease) whose tumor appeared potentially resectable and who could tolerate pulmonary resection were advised to undergo thoracotomy. The chest roentgenogram (CXR) and CT scan reports were reviewed for indication of enlargement of mediastinal lymph nodes. We used the radiology report as documented on the chart to more closely parallel situations commonly arising in daily practice. Sixty-four (52%) of the CT scans were performed at The University of Texas, M.D. Anderson Cancer Center. All scans were reviewed by staff radiologists. If the radiologist indicated that no notable mediastinal adenopathy was present this was recorded as a negative CT scan for mediastinal involvement. Reports that indicated the presence of adenopathy in which malignant involvement was questioned or believed to be likely were considered positive CT scans. Our radiologists follow the generally accepted criteria of considering lymph nodes to be normal in size when they are less than 1 cm in largest diameter [ 7 ] . A complete mediastinal node dissection was done in 118 patients (95%).Six patients had lymph node sampling 0003-4975/92/$5.00

534

CYBULSKY ET AL CT SCAN IN N2 DISEASE

Ann Thorac Surg 1992;54:533-7

Table 1. Distribution of Lung Tumors by Location" Site

CT Positive (63 cases)

CT Negative (61 cases)

28

17 3 9

Right upper lobe Right middle lobe Right lower lobe Left upp,r lobe Left lower lobe a

2 13 16 4

16 15

One patient had a right main bronchial tumor (CT negative).

CT

=

computed tomographic scan.

of one or more lymph node groups. The pathology reports were reviewed, and the number of nodes involved with cancer were documented for each nodal group. In 39 patients, a pathological analysis of individual lymph node groups was not performed. Postoperative adjuvant radiation therapy and chemotherapy were given to some patients entered into randomized clinical trials. Seventyfive patients received adjuvant mediastinal radiotherapy and 33 received adjuvant postoperative chemotherapy. One patient received preoperative radiation therapy. Doses of postoperative mediastinal radiation were usually 45 to 55 Gy. Chemotherapeutic agents used were predominantly cyclophosphamide, cis-platinum, and either doxorubicin )hydrochloride (Adriamycin; Adria Laboratories, Dublin, OH) or etoposide. Actuarial survival curves were calculated by the method of Kaplari and Meier using the SAS computer program, and statistical comparisons were made using the log-rank test and the Wilcoxon rank sum test. All p values were two-sided. Operative mortality (30-day and death during the same hospitalization as operation) was included in the survival analysis. In comparing patients with postoperative adjuvant treatment, only patients surviving beyond 30 days were included.

Results One hundred twenty-four patients were studied. Their mean age was 60 years (range, 38 to 80 years). There were 87 men and 37 women. More cancers were found in the upper lobes, with 45 (36%)in the right upper lobe (Table 1). Fifty percent of tumors were central. Fifty-eight patients (47%) had adenocarcinoma, and 41 patients (33%) had squamous carcinoma. The histology in the remaining patients included bronchoalveolar, 5; large cell, 7; poorly differentiated, 10; and adenosquamous, 3. The vast majority of patients had clinical T2 tumors (83%).Seventeen patients underwent staging with mediastinoscopy before resection. ldediastinoscopy was done primarily to exclude contralaterd N3 disease, and in no patient was this group of lymph nodes found to be involved. In 2 patients, biopsies oi ipsilateral lymph nodes were positive for carcinoma. Sixty-three patients (51%) had enlarged mediastinal lymph nodes by preoperative CT scan. In this group were 21 patients whose mediastinal abnormalities were appre-

ciated on CXR (the finding of an abnormal hilum is not included in this group). Correlation between the site of mediastinal lymph node metastasis and the site of enlarged nodes on the CT scan could not be assessed in 5 patients. In 8 patients (14%) the specified location of enlarged nodes did not correspond to pathologically involved nodes. In addition, occasionally two node groups were deemed to be enlarged, but only one contained metastatic disease. Table 2 demonstrates the different distribution of histology in patients with positive and negative CT scans. Fifty-seven percent of the negative CT scans were in patients with adenocarcinoma. The sensitivity of the CT scan was higher in patients with squamous cell type (68%), compared with patients with adenocarcinoma (40%). The sensitivity of the CT scan also varied with location of the primary tumor, with the highest rate in the right upper lobe (63%)and the lowest in the left lower lobe (26%). Three patients (2%), all from abroad, were lost to follow-up after 1, 3, and 7 months. The overall survival was 19.1% at 3 years and 10.0% at 5 years (Fig 1).There were nine 30-day postoperative or in-hospital deaths for an overall mortality rate of 7.2%. Four deaths occurred in patients after right pneumonectomy. The mortality of patients undergoing right pneumonectomy was 12.9% (4/31); it was 5.3% (5/93) for other procedures. Nonfatal morbidity occurred in 22 patients: supraventricular arrhythmia in 8, pneumonia in 4, prolonged air leak in 3, bronchopleural fistula in 2, chylothorax in 2, and others in 3. There was a statistically significant survival difference between patients with positive and negative CT scans (Fig 2). The survival rate at 3 and 5 years was 25.0% and 13.5% for the negative CT scan group and 13.2% and 6.6% for the positive CT scan group, respectively. When the patients with an abnormal mediastinum on CXR were excluded, the increased survival in the CT positive group remained: 3-year survival was 12.7% for positive CT patients and 5-year survival was 5.1% for negative CT patients. An improved survival was found in patients with left upper lobe lesions, with 3-year and 5-year survival rates of 38.7% and 20.3%, respectively (Fig 3). Fifty percent of patients with left upper lobe tumors had a positive CT scan. There was no difference in survival between patients

Table 2. Distribution by Histoloxy Histology

CT Positive (63 cases)

CT Negative (61 cases)

27 (43%)

14 (23%) 35 (57%) 3 (5%) 4 (7%) 5 (8%) 0

Squamous cell Adenocarcinoma Poorly differentiated Large cell Bronchoalveolar

Adenosquamous CT

=

computed tomographic scan.

23 (36%) 7 (11%) 3 (5%) 0 3 (5%)

535

CYBULSKY ET AL CT SCAN IN N2 DISEASE

Ann Thorac Surg 1992;54:53%7

0

TOTAL FAIL 107

1 .0

TOTAL 45 23 0 31 x 19

0 124

D

0.9

09

0.8

08

0.7

0. I

i

*

FAIL 39 LOG= RUL 22 LOC = RLL 24 LOC = LUL 18 LOC= LLL

0.6

4

z 2

0.5

c

0.5

K

0.4

0.4 0.3

n.

1

0.3 0.20.1.

1

0.0 0

10

20

30

50

40

60

70

80

90

MONTHS

with squamous cell carcinoma and those with adenocarcinoma ( p = 0.430) or in those with central versus peripheral lesions ( p = 0.836). Patients who underwent pneumonectomy had a trend toward worse survival ( p =

TOTAL FAIL 63 57 CT(+) A 61 50 CT(-)

0

0.9 0.8

0.7

3

0.6-

-1

4

6

' I-

0.5-

re

zn 0.40.3 0.2 0.1 -

0

10

20

30

40

50

I

10

20

30

40

50

60

70

80

90

MONTHS

Fig I. Actuarial survival curve for all patients.

1 .0

0

--

60

70

80

90

MONTHS

Fig 2 . Actuarial survival curves of patients with positive and negative computed tomographic (CT) scans of the chest.

Fig 3 . Actuarial survival by tumor locution. (LLL = left lower lobe; L O C = location; LUL = left upper lobe; RLL = right lower lobe; RUL = right upper lobe.)

0.087); however, this could be attributed to a higher operative mortality. The patients with involvement of only a single lymph node did not have an improvement in survival ( p = 0.201). No adverse impact on survival was observed in 51 patients who had microscopically positive resection margins or extranodal extension of tumor ( p = 0.973). There was a trend toward a decreased survival for patients with extranodal extension (33 patients) compared with patients without this finding and negative resection margins (73 patients) ( p = 0.070). A trend toward survival advantage was seen in patients who received postoperative mediastinal irradiation ( p = 0.155). There was no survival difference in patients who received adjuvant chemotherapy ( p = 0.460). Thus, these factors did not play a significant role in the survival difference attributable to the presence or absence of lymphadenopathy on the preoperative CT scan. Thirteen patients are alive at the present time, 4 with recurrent disease. Excluding those lost to follow-up (3), postoperative deaths (9), late deaths of unknown cause (ll),and late deaths not related to lung cancer (lo), the remaining 78 patients died secondary to recurrent disease. In 7, the site of recurrence was not documented. Forty-four patients had metastases to remote organs: brain, bone, liver, and adrenal gland. Twenty-three patients, as well as the 4 living patients, had recurrent disease within the chest or supraclavicular or axillary lymph nodes.

536

CYBULSKY ET AL CT SCAN IN N2 DISEASE

Commlent Mediastinal lymph nodes that exceed 1cm in diameter are more likely to harbor metastatic disease in patients with primary lung cancer than lymph nodes with a diameter of less than 1 cm [6, 71. The sensitivity in identifying mediastinal involvement in our study of 124 patients with involvecl mediastinal (N2) lymph nodes was only 51%. In studies [4,6,1&12] assessing the accuracy of CT scanning the sensitivity at this level varies between 57% [lo] and 95% [ll]. When a larger size criterion is used the specificity rises with a corresponding drop in sensitivity [12, 131. The techniques used in performing a CT scan such as the diameter of the cuts and vascular resolution influence the sensitivity and specificity [14]. Evidence of lymph node involvement must be confirmed histologically. When a mediastinal node dissection at thoracotomy was used to verify the findings on CT scan, a higher falsely negative CT scan rate was found [15]. Granulomatous disease and obstructive pneumonitis account for falsely positive CT scan results. A negative CT scan is considered sufficient to proceed with thoracotomy in patients who are surgical candidates [14, 151. The patients in whom the CT scan was falsely negative most commonly had a histological diagnosis of adenocarcinoma. In fact, 60% of patients with adenocarcinoma involving mediastinal (N2) nodes had a normal mediastinum on the CT scan, as opposed to only 35% of patients with squamous cell carcinoma. This trend was noted in an earlier study at this institution [16]. It was also found that adenocarcinoma could metastasize directly to mediastinal (N2) nodes, skipping hilar (Nl) nodes. Mediastinoscopy is both a sensitive and specific procedure for establishing mediastinal nodal involvement [5]. It may be used to exclude patients with advanced disease from surgical consideration and simultaneously provide a tissue diagnosis. The test, however, does involve an operative procedure and has an associated low rate of potentially serious complications. Computed tomographic scanning has been suggested as an alternative to mediastinoscopy, in the hope of eliminating an invasive, time-consuming, and more expensive procedure [6, 171. At the present time a CT scan is obtained in almost every patient with an abnormal CXR that suggests pulmonary malignancy. In addition to assessing the mediastinum, this technique improves definition of the primary lesion and excludes additional lesions. In 51% of patients in this series, the more advanced stage of the disease was completely unsuspected. The outcome in patients with a negative CT scan was better than in patients with a positive scan; however, the 5-year survival in this subgroup was only 13.5%. This falls short of the survival rates of 30% achieved by Martini and Flehinger [I] and 20% by Watanabe and associates [18]. It also falls below the 5-year survival in Pearson and colleagues’ [3] series of patients with preoperative falsely negative mediastinoscopy, who achieved an overall 5-year survival rate of 24%. Variability in outcome perhaps is due to different patient selection, size of primary tumor, location of tumor, extent of lymphadenopathy,

Ann Thorac Surg 1992;54:533-7

and the extent to which exclusion of N3 disease is vursued. Patients with a positive CT scan had a 5-year survival of 6.6%, which is similar to the 9% 5-year survival of patients who had a positive mediastinoscopy preoperatively 131. Evidence of mediastinal enlargement on CXR did not predict as poor an outcome in this series as described previously [9], perhaps because once the extent of adenopathy was better defined with the CT scan of the chest, only the more favorable patients were selected for resection. Some series report improved survival in patients with a “complete” or “curative” resection [l,3, 181. A uniform definition of these terms is lacking in the literature. Extranodal extension of tumor is an indication of incomplete resection to some [19], but not others [20]. Patients in this series with microscopically negative margins and without extranodal extension did not demonstrate a significant survival advantage. The number of lymph nodes involved with metastatic disease did not influence outcome, although this finding may have been influenced by incomplete pathologic breakdown of mediastinal lymph node groups in 31.5% of the patients. Some authors have found a survival disadvantage when multiple sites of lymph node metastases are present [21, 221, whereas others have not [2]. In Naruke and associates’ [2] series 11 of 35 patients who achieved 5-year survival had multiple lymph node groups involved. The type of operation performed did not have an impact on survival. It was not possible to determine if patients undergoing pneumonectomy had larger tumors, but the outcome in patients with central tumors was not different from that in patients with peripheral ones. A right pneumonectomy carries a high mortality [23-261, a pattern that has not changed in the last two decades. Patients at highest risk are those with T3 lesions or N2 disease, or patients whose postoperative forced expiratory volume in 1 second would be less than 33% of predicted [23]. Thirty-two patients had tumor located in the left upper lobe. Their outcome was significantly better, with 3-year and 5-year survival rates of 38.7% and 20.3%, respectively. Left upper lobe lesions most frequently metastasize to stations 5 and 6 node groups (ATS classification [27]). In a study of 35 patients with isolated metastases to these nodes, Patterson and colleagues [20] found a 5-year survival of 28%. Therefore, an increased representation of these patients in a surgical series of N2 patients will favorably affect survival. Postoperative adjuvant irradiation did not significantly prolong survival. This variable is difficult to assess in this study, as the indications for adjuvant radiotherapy were not standardized. This result is similar to the lack of survival benefit for postoperative irradiation shown in a prospective study by the Lung Cancer Study Group [28]. Patients with stage IIIa lung cancer involving mediastinal lymph nodes have a poor prognosis even with potentially resectable disease. The CT scan may underestimate mediastinal involvement, especially in patients with adenocarcinoma. We believe the data from this and other

Ann Thorac Surg 1992:54:53>7

studies support the following approach. If the CT scan is positive for mediastinal adenopathy a mediastinoscopy should be performed. This can confirm N2 disease and in some patients will demonstrate N3 disease. Patients who would be at high risk for surgical resection, such as those requiring a right pneumonectomy, should be considered for nonsurgical therapy. If N2 disease is confirmed histologically and appears resectable, we proceed with thoracotomy in patients with a CT scan negative for mediastinal adenopathy. The best prognosis in patients with mediastinal (N2) lymph node involvement is in those with left upper lobe lesions. In these patients, unless the lymphadenopathy is causing recurrent laryngeal nerve involvement, or appears extremely bulky, resection should be attempted. This study has demonstrated that when there is no evidence of adenopathy demonstrated by CT scanning, the outcome after resection and mediastinal node dissection is better than when enlarged lymph nodes are visualized. Patients in this group are considered for resection when the potential for long-term survival exceeds the estimated operative mortality. When possible, all patients with N2 disease are entered into clinical trials evaluating adjuvant therapy. We thank Giri Natarajan for her help with data management and Carol Torrence for assistance with manuscript preparation.

References 1. Martini N, Flehinger BJ. The role of surgery in N2 lung cancer. Surg Clin North Am 1987;67103749. 2. Naruke T, Goya T, Tsuchiya R, Suemasu K. The importance of surgery to non-small cell carcinoma of lung with mediastinal lymph node metastasis. Ann Thorac Surg 1988;46: 603-10. 3. Pearson FG, DeLarue NC, Ilves R, Todd TRJ, Cooper JD. Significance of positive superior mediastinal nodes identified at mediastinoscopy in patients with resectable cancer of the lung. J Thorac Cardiovasc Surg 1982;83:1-11. 4. McKenna RJ, Libshitz HI, Mountain CE, McMurtrey MJ. Roentgenographic evaluation of mediastinal nodes for preoperative assessment in lung cancer. Chest 1985;88:206-10. 5. Patterson GA, Ginsberg RJ, Poon PY, et al. A prospective evaluation of magnetic resonance imaging, computed tomography, and mediastinoscopy in the preoperative assessment of mediastinal node status in bronchogenic carcinoma. J Thorac Cardiovasc Surg 1987;94:679-84. 6. Gross BH, Glazer GM, Orringer MB, Spizarny DL, Flint A. Bronchogenic carcinoma metastatic to normal-sized lymph nodes: frequency and significance. Radiology 1988;166:714. 7. Aronchick JM. CT of mediastinal lymph nodes in patients with non-small cell lung carcinoma. Radio1 Clin North Am 1990;28:57%81. 8. Spiro SG, Goldstraw P. The staging of lung cancer. Thorax 1984;39:401-7. 9. Martini N, Flehinger BJ, Zaman MB, Beattie EJ Jr. Results of resection in non-oat cell carcinoma of the lung with mediastinal lymph node metastases. Ann Surg 1983;198:386-97.

CYBULSKY ET AL CT SCAN IN N2 DISEASE

537

10. Rhoads AC, Thomas JH, Hermreck AS, Pierce GE. Comparative studies of computerized tomography and mediastinoscopy for the staging of bronchogenic carcinoma. Am J Surg 1986;152:587-91. 11. Glazer GM, Orringer MB, Gross BH, Quint LE. The mediastinum in non-small cell lung cancer: CT-surgical correlation. AJR 1984;142:1101-5. 12. Staples CA, Miller RR, Evans KG, Nelems B. Mediastinal nodes in bronchogenic carcinoma: comparison between CT and mediastinoscopy. Radiology 1988;167:367-72. 13. Ratto GB, Frola C, Cantoni S, Motta G. Improving clinical efficacy of computed tomographic scan in the preoperative assessment of patients with non-small cell lung cancer. J Thorac Cardiovasc Surg 1990;99:41&25. 14. Ferguson MK, MacMahon H, Little AG, Golomb HM, Hoffman PC, Skinner DB. Regional accuracy of computed tomography of the mediastinum in staging of lung cancer. J Thorac Cardiovasc Surg 1986;91:49%504. 15. Daly BDT Jr, Faling LJ, Pugatch RD, et al. Computed tomography: an effective technique for mediastinal staging in lung cancer. J Thorac Cardiovasc Surg 1984;88:48&94. 16. Libshitz HI, McKenna RJ Jr, Mountain CF. Patterns of mediastinal metastases in bronchogenic carcinoma. Chest 1986;90:22932. 17. Osborne DR, Korobkin M, Ravin CE, et al. Comparison of plain radiography, conventional tomography, and computed tomography in detecting intrathoracic lymph node metastases from lung carcinoma. Radiology 1982;152:15741. 18. Watanabe Y, Shimizu J, Oda M, et al. Aggressive surgical intervention in N2 non-small cell cancer of the lung. Ann Thorac Surg 1991;51:253-61. 19. Mountain CF. Expanded possibilities for surgical treatment of lung cancer. Survival in stage IIIa disease. Chest 1990;97: 1045-51. 20. Patterson GA, Piazza D, Pearson FG, et al. Significance of metastatic disease in subaortic lymph nodes. Ann Thorac Surg 1987;43:155-9. 21. Thomas PA, Piantadosi S, Mountain CF. Should subcarinal lymph nodes be routinely examined in patients with nonsmall cell lung cancer? J Thorac Cardiovasc Surg 1988;95: 883-7. 22. Maggi G, Casadio C, Mancuso M, Oliaro A, Cianci R, Ruffini E. Resection and radical lymphadenectomy for lung cancer: prognostic significance of lymphatic metastases. Int Surg 1990;75:17. 23. Putnam JB Jr, Lammermeir DE, Colon R, McMurtrey MJ, Ali MK, Roth JA. Predicted pulmonary function and survival after pneumonectomy for primary lung carcinoma. Ann Thorac Surg 1990;49:909-15. 24. Ginsberg RJ, Hill LD, Eagan RT, et al. Modern thirty-day operative mortality for surgical resections in lung cancer. J Thorac Cardiovasc Surg 1983;86:65M. 25. Wahi R, McMurtrey MJ, DeCaro LF, et al. Determinants of perioperative morbidity and mortality after pneumonectomy. Ann Thorac Surg 1989;48:33-7. 26. Thompson-Evans EW. Resection for bronchial carcinoma in the elderly. Thorax 1973;28:864. 27. Mountain CF. A new international staging system for lung cancer. Chest 1986;89:225S-33S. 28. Lung Cancer Study Group. Effects of postoperative mediastinal radiation on completely resected stage I1 and stage I11 epidermoid cancer of the lung. N Engl J Med 1986;315: 1377-81.

Prognostic significance of computed tomography in resected N2 lung cancer.

We reviewed 124 patients from 1982 to 1988 who had a resected primary non-small cell lung cancer metastatic to mediastinal (N2) lymph nodes and a preo...
749KB Sizes 0 Downloads 0 Views