In, .I. Radialron OncologicBIO/. Phys Vol. Printed in the U.S.A. All rights reserved.
20, PP
945-951 Copyright
0360.3016/91 $3 00 + .oO 0 I99 Pergamon Press plc
I
??Original Contribution
RADIATION
JERRY
D. SLATER, CLIFTON
THERAPY FOLLOWING RESECTION OF NON-SMALL CELL BRONCHOGENIC CARCINOMA
M.D.,*
MOUNTAIN,
NANCY
A. ELLERBROEK,
M.D.,+
MARY
JANE
AND LESTER The University
M.D.,* OSWALD,
J. PETERS,
H. THOMAS B.S.,*
JACK
BARKLEY A. ROTH,
JR.,
M.D.,*
M.D.+
M.D.*
of Texas M. D. Anderson Cancer Center, Houston, Texas
Between 1970 and 1982, 102 patients received postoperative radiotherapy after attempted curative resection of bronchogenic carcinoma at The University of Texas M. D. Anderson Cancer Center. Surviving patients had a
minimum follow-up of 3 years. Eight patients had pathological Stage I disease, 29 Stage II, and 65 Stage III. The 5-year actuarial survivals for patients with stages I, II, and III disease were 83%, 55%, and 38%, respectively (p = -04). Corresponding values for patients with NO, Nl, and N2 disease were 74%, 56%, and 28% (p = .Ol). No significant differences in survival were seen based on T stage or tumor histology. Nine patients had gross residual disease following surgery, and 19 had microscopic residual disease. The 5-year actuarial survival was 78% for 12 patients without nodal disease who had known gross (4 patients) or microscopic (8 patients) residual tumor following attempted curative resection. The pathologic status of the hilar and mediastinal lymph nodes was the most significant factor affecting the frequency of metastatic relapse, with 19% of patients with NO, 33% of those with Nl, and 69% of those with N2 disease developing distant disease. The low overall rate of recurrence intrathoracically (16%) confirms that postoperative radiotherapy is effective in preventing local relapse even in patients with proven nodal involvement. The impact of adjuvant radiation therapy on survival cannot be determined from these data, and further data are needed, preferably from well designed prospective studies. Non-small cell lung cancer, Postoperative
radiation therapy.
radiotherapy on local disease control and survival in patients with lung cancer seen at The University of Texas M. D. Anderson Cancer Center.
INTRODUCTION Carcinoma of the lung is the most deadly malignancy and the leading cause of cancer death in the United States. The American Cancer Society estimates that there will be 157,000 new cases and 142,000 deaths from lung cancer in 1990 ( 19). Fewer than 25% of patients with cancer confined to the chest are considered surgical candidates, either because of tumor extent or poor medical condition. However, surgical resection offers the best chance of long-term survival and is the recommended treatment when possible. Despite apparent complete surgical resection, 5-year survival rates for patients with Stage II and III lung carcinoma have been reported to be 20-40% and O-25%, respectively (2,4,6, 8,9, 12-17,20, 2 1). In an effort to improve local control and survival, postoperative radiotherapy has been used. This has decreased the local recurrence rate, although its impact on long-term survival is more controversial. This review will present the results of postoperative
METHODS
AND
MATERIALS
Between 1970 and 1982, 102 patients who underwent attempted curative resection for non-small cell bronchogenie carcinoma were referred to the M. D. Anderson Cancer Center Department of Clinical Radiotherapy for consideration of postoperative irradiation. Patients who had radiation therapy at other institutions were excluded. No patient had another primary malignancy or evidence of distant metastases at the start of radiotherapy, and all living patients had a minimum follow-up of 3 years. Surgery consisted of wedge resection in 6 patients, lobectomy in 43, bilobectomy in 3, and pneumonectomy in 50. Squamous cell carcinoma was seen in 58 patients
Presented at the 30th Annual Scientific Meeting of the American Society for Therapeutic Radiology and Oncology in New Orleans, LA, October 9-14, 1988. * Dept. of Clinical Radiotherapy. + Dept. of Thoracic Surgery. Reprint requests to: Nancy A. Ellerbroek, M.D., Department
of Radiation Oncology, Loma Linda University Medical Center, 11234 Anderson St., Loma Linda, CA 92354. Supported by grant CA06294 awarded by the National Cancer Institute, United States Department of Health and Human Services. Accepted 945
for publication
9 November
1990.
I. J. Radiation Oncology 0 Biology 0 Physics
946
May 1, 1991, Volume 20, Number 5
and adenocarcinoma in 39; 5 patients had large cell, bronchioalveolar, or unclassified carcinoma. Tumors were pathologically staged according to the American Joint Committee on Cancer (AJC) classification (1); 8 patients had Stage I disease, 29 Stage II, and 65 Stage III. The relationships of tumor stage to histology are shown in Table 1. Twenty patients had NO, 35 N 1, and 47 N2 disease. During the same time period, 771 patients with an initial diagnosis of non-small cell lung cancer underwent curative resection. There were 655 patients whose AJC stage was known and who did not have small cell carcinoma or operative deaths; of those, 333 had Stage I disease, 109 Stage II, and 213 Stage III. There were 588 apparent complete resections, and 67 cases where microscopic or gross residual disease remained. Of 645 whose N Stage was known, 382 had NO disease, 148 Nl, and 115 N2. Fifteen patients received chemotherapy with cyclophosphamide, Adriamycin, and cisplatin as part of their primary treatment. Thirteen of those 15 had Stage III disease. In one patient cisplatin was omitted; another also had received DTIC and BCG. Chemotherapy followed surgery in all cases, and usually consisted of 1 to 3 cycles of CAP (median 2) and preceded radiotherapy in all but one patient. Seven patients had additional chemotherapy after radiotherapy. Radiotherapy was started approximately 4 weeks postoperatively in all but two patients unless the patient received chemotherapy, in which case it was started an average of 12 weeks after surgery (range 2-26 weeks). The indications for patient referral for radiotherapy changed during the time of this analysis. Prior to 1978, patients were not routinely referred for postoperative radiotherapy unless the surgeon believed there was increased risk of recurrent disease because of microscopic or gross residual tumor, extensive lymph node involvement, or evidence of extension of tumor outside the lymph nodes. From 1978 on, patients were also routinely referred for radiotherapy if they had positive hilar or mediastinal lymph nodes.
Table 1. Tumor
Squamous carcinoma n = 58
TIN0
1
Adenocarcinoma n = 39 Other n=5 Total n = Total number.
1
RESULTS Overall survival Actuarial survivals were calculated using the BerksonGage life table method (2). Patients who died of unknown causes were classified as having died of cancer. The overall
pathologic
stage and histology
Stage II (n = 29)
Stage I (n = 8) Histology
Overall, 28 patients had residual disease following surgery: 9 with gross residual disease and 19 with microscopic residual disease suspected because of positive surgical margins. All patients with NO disease who were referred for radiation therapy had suspected residual tumor, although tumor was confirmed in only 12 of 20 such cases upon review of pathologic and surgical reports. The other eight patients had large primary tumors and were considered at increased risk for local recurrence despite apparent complete removal of the tumors. Radiation therapy fields included the mediastinum, ipsilateral hilum, and supraclavicular lymph nodes in addition to areas of known microscopic or macroscopic disease. Eighty-four patients were treated with a continuous course of radiation, with the majority receiving 50 Gy in 25 fractions over 5 weeks. Treatment was given AP-PA initially, and a reduced-field boost, usually of 5-10 Gy, was given to 9 of 28 patients with postoperative residual disease, usually using opposed oblique fields designed to encompass the area of residual disease. Doses were prescribed to the midline or the isocenter without lung corrections. As CT planning was not in use during the study period, the treatment volume was verified with conventional simulation and portal films. In 18 patients a regimen of split-course radiation therapy, generally consisting of 30 Gy in 10 fractions, was planned, followed by a 3-week rest, at which time the patients were reevaluated for administration of an additional 30 Gy in 10 fractions. Of the 18 patients, 5 did not receive the second course of radiation either because they refused further treatment or because their conditions had deteriorated. Elective cranial irradiation was given to four patients with adenocarcinoma.
Stage III (n = 65)
TlNl
T2Nl
2
2
19
7
2
1
8
7
2
1
29
15
T2NO
4
3
T3NO
T3Nl
3
TIN2
T2N2
T3N2
2
17
4
3
17
1
TxN2
1
2 3
5
36
5
1
Non-small cell bronchogenic carcinoma 0 J. D. Survival
Survival
of All Patients
Fig. 1. Overall survival patterns of 102 patients treated with postoperative radiotherapy following attempted complete surgical resection.
2- and 5-year survival rates are 59% and 46%, respectively (Fig. 1). Only one patient is alive with disease.
Histology and survival The actuarial 2- and 5-year survivals (Fig. 2) for those with squamous cell carcinoma are 67% and 55%, respectively, and 50% and 38% for adenocarcinoma (p = .39).
Stage and survival The pathologic stage of disease affected survival as shown in Figure 3. At 5 years, the survivals were 83%, 55%, and 38% for patients with stages I, II, and III disease, respectively (p = .04). Extent of nodal involvement had the most significant impact on survival. Figure 4 shows the following 5-year survivals for patients with the stated disease stages who received postoperative irradiation: 74% for NO, 56% for Nl, and 28% for N2 patients (p = .O 1). There was no significant difference in survival for NO patients with known (78%) or suspected residual (71%).
Survival
947
SLATER et al.
Fig. 3. Survival of patients
of each patients
according
to disease stage.
Patients with Tl tumors had a survival rate of 68%, whereas 5-year survivals for those with T2 and T3 tumors were 43% and 5 I%, respectively. The increased survivals associated with T3 as compared with T2 tumors were probably caused by a higher incidence of mediastinal node involvement in our group of patients with T2 tumors. Of the patients with T3 NO disease, 3 with Pancoast tumors had a median survival of 68 months (range, 7 to 116), and 12 with other presentations had a median survival of 24 months (range, 2 to 124). One patient with Stage I, 2 with Stage II, and 12 with Stage III disease received chemotherapy as part of their primary treatment. When the Stage III disease patients who received chemotherapy were compared with those who did not, there was no detectable difference in actuarial survival (p = .90). Because of the small number of patients who received chemotherapy as part of other protocols and the heterogeneous group of patients involved, no evaluation of the effects of chemotherapy can be made. The 5 patients who did not complete the second part of their split course of treatment had a short median survival of 2 months; in 4 of the 5 the course was not completed because of progressive disease or complications.
by Histology Survival
Fig. 2. Survival type.
by Stage
according
to histologic
by Nodal Status
tumor Fig. 4. Survival of patients
according
to nodal status.
1. J. Radiation
948
Oncology
0 Biology 0 Physics
Putterns oftumor recurrence The cause of death was not known for 14 patients; thus they were excluded from analysis of disease recurrence sites. Of the remaining 88 patients, 49 have failed (Table 2) including 27 with squamous cell carcinoma (53%) and 20 with adenocarcinoma (59%). Two patients with pathologic Stage I (29%), 9 with Stage II (38%) and 38 with Stage III disease (67%) had recurrences.
May I, I99 I. Volume 20, Number 5
The local recurrence rates for Stages I, II, and III disease were 14%, 4%, and 2 l%, respectively. In 6 of 12 patients with Stage III disease who had local recurrence, it was their only site of failure. The local recurrence rate was 12% for patients with NO disease and 10% and 2 1% for those with N 1 and N2 disease, respectively. The local recurrence rate for patients with NO and known residual disease ( 10%) did not differ significantly from that for patients with a high suspicion of residual disease (17%).
Local tumor recurrence Fourteen patients ( 16%) had recurrences within the irradiated volume. The time to recurrence had a range of 3-40 months after completion of radiotherapy, with a mean of 13 months. Local recurrence as the only site of failure was seen in 7 patients and distant metastasis alone in 35 (Table 2). Seven patients had recurrences at both local and distant sites: three had local recurrences prior to developing distant metastases 1 to 5 months later, two developed distant metastases prior to local recurrence, and two others developed both simultaneously. It was difficult in this retrospective study to make a correlation between dose and local control. Only one of the patients who had a local recurrence had received less than 50 Gy. An analysis of the fields used and the locations of the recurrences yielded one case where a patient with an upper lobe primary had disease recur in an untreated ipsilateral supraclavicular region. One was possibly a second primary, occurring in the right upper lobe 40 months after treatment for a left lower lobe primary, and the other recurrences were in the treated area. There were 2 local recurrences in 8 patients with macroscopic residual disease postoperatively and 3 recurrences in 17 patients with microscopic residual disease. The overall local recurrence rate for patients with known gross or microscopic residual disease (20%) did not differ significantly from that for patients with NO disease and only suspected residual disease ( 17%). In 8 ( 16%) of 5 1 evaluable patients with squamous cell carcinoma, disease recurred locally; in 6 this was their only site of recurrence. Six patients with adenocarcinoma (18%) had local, recurrence; in one this was the only site of failure.
Distant diiseuse Distant metastases have developed in 42 (48%) of the 88 patients evaluable for patterns of recurrence, with a mean onset of 12 months following radiation therapy. In 35 (83%) of the patients who developed distant metastases, this was their only site of failure. The pathologic status of the hilar and mediastinal lymph nodes was the most significant factor affecting the metastatic rate, with 3/16 (19%) of patients with NO, lO/ 30 (33%) with Nl, and 29/42 (69%) with N2 disease developing distant metastases. The data revealed a trend toward more frequent distant metastases as the overall disease stage and T stage increase (Table 3). Distant metastases were seen in 56% of patients with adenocarcinoma and in 4 1% with squamous cell carcinoma. The most common initial sites of metastases were lung (outside the treatment field), and bone (Table 4). One of four patients with adenocarcinoma who received elective cranial irradiation eventually developed brain metastases.
Complications Twelve patients developed symptoms related to the radiation, with the majority of these symptoms being selflimited or causing minimal discomfort. Symptomatic pneumonitis was seen in two patients following radiation therapy, whereas four patients developed pericarditis symptoms that resolved with conservative management. Four patients died of complications of treatment: one patient died of pneumonia 6 months after irradiation but
Table 3. Incidence Table
2. Tumor histology and sites of recurrence
Histology Squamous cell n= 51 Adenocarcinoma n = 34 Other n=3 Total = 50/88 n = Total number.
Local
Distant
Local and distant
6
19
2
1
14
5
0
2
0
7
35
7
Stage
11
III Tl+ T2 T3 NO Nl N2
of distant
metastasis Distant
and stage of disease metastases*
l/7 9/24 32157 3/8 30/57 9/20 3/16 10/30 29142
(14%) (38%) (56%) (30%) (53%) (45%) (19%) (33%) (69%)
* Excludes 14 patients whose cause of death was not known. + Excludes one patient with TX primary stage.
Non-small cell bronchogenic carcinoma 0 J. D. SLATERetal.
949
Table 4. Tumor histology and first site of distant metastasis Tumor Sites Lung
Other or multiple
Total
Brain
Liver
Bone
Squamous cell Adenocarcinoma Other
5 3 1
3 1 0
5 3 1
5 5 0
3 7 0
21 19 2
Total
9
4
9
10
10
42
Histology
was noted to have “radiation fibrosis,” which may have contributed to his death; one died with Klebsiella pneumonia shortly after completion of irradiation; one developed a tracheoesophageal fistula after receiving 20 Gy associated with necrosis of tumor involving those structures; and one developed a hydro-pneumothorax early in the course of irradiation. Four patients developed subcutaneous fibrosis in the lower neck, which was associated with minimal limitations in mobility. Two others, both of whom were treated for Pancoast tumors, developed apparent brachial plexopathy with mild muscle wasting and paresthesias, one after receiving 60 Gy in 20 fractions over 50 days and another following 54 Gy in 26 fractions over 37 days. Recurrent tumor was a possible cause of the neurologic symptoms in the patient who had received 60 Gy.
DISCUSSION The goal of delivering postoperative radiation therapy is to improve control of the local tumor and, it is hoped, to increase the survival of these patients. Local recurrence, even of early-stage disease, can be a significant problem. After apparent complete surgical resection in patients with T 1-T2, NO-N 1 disease, local recurrence rates have been reported to be as high as 41%, with most studies indicating a lo-25% incidence (4, 8, 9, 20). The ability of postoperative radiotherapy to improve the local disease control rate has been shown (4, 20). In our group, 37 patients with T 1-T2, NO-N 1 disease received postoperative radiotherapy. Two (6%) of 3 1 evaluable patients had local recurrences whereas only 1 (4%) of 26 of these patients without evidence of residual disease did. Radiation therapy is not routinely given to patients with no lymph node involvement; however, when the surgical margins are involved or when there is gross residual disease, radiation can be used to decrease the likelihood of local recurrence. In our study, 20 patients without lymph node involvement received postoperative radiation therapy for residual disease (12 patients) or for suspected residual disease (8 patients) because of primary tumor size or extent. Despite these high-risk factors, local control was obtained in 88% of patients, a finding consistent with that for patients who had undergone complete resection (7, 8, 14, 20, 21).
Radiation therapy also decreases the local recurrence rate in patients with more advanced disease. The Lung Cancer Study Group, in a randomized trial of postoperative radiation therapy versus no further treatment for patients with Stage II or III squamous cell carcinoma, found a significant decrease in local recurrence rate in the patients treated with radiation (11, 12). Retrospective studies have also reported decreased local recurrence following postoperative irradiation. Choi et al. reported that 65% of all recurrences of Nl, N2, and T3 squamous carcinoma treated by surgery alone were regional; this figure decreased to 32% in those treated with postoperative radiation therapy (3). Kirsh and Sloan reported results for 20 patients with mediastinal metastases who underwent only surgical resection ( 10). Eight of 10 patients in whom the cause of death was known died with evidence of recurrent disease in the thorax. However, only 4 of 54 patients who received postoperative radiotherapy in whom the cause of death was known developed local recurrence. In our group of 65 patients with Stage III disease, the local recurrence rate with postoperative radiation therapy was 22%; it was the same for all patients who had Stage III disease with hilar or mediastinal node involvement. Excluding those Stage II and III disease patients who had known residual disease postoperatively, the local recurrence rate was 15%. Although the effectiveness of radiation therapy in improving control of local disease is well documented, its effect on improving survival is more controversial. Resection of Stages I and II non-small cell carcinomas of the lung without adjuvant treatment has resulted in 5year survival rates of 20-60%, with the higher survival for patients without lymph node involvement (6, 9, 13, 15, 18, 2 1). Van Houtte et al. conducted a randomized trial on the effects of postoperative irradiation in patients without evidence of lymph node metastases and found no increase in survival associated with radiation therapy (20). There are, however, subsets of patients with early-stage disease and poor prognostic factors (such as residual disease) who might benefit from postoperative radiation therapy to eradicate the remaining tumor. In our study group, 20 patients without nodal involvement were thought to be at high risk for recurrence because they were known to have residual disease postoperatively (I 2
950
1. J. Radiation Oncology 0 Biology 0 Physics
patients) or were suspected of having residual disease due to tumor size or extent (8 patients). Despite these findings, their 5-year survivals were 78% and 7 1% for known or suspected residual disease, respectively, following radiation therapy. The survival after surgery in patients with more advanced disease is poor. The American Joint Committee on Cancer Task Force on Lung reported on 36 1 patients with Stage III disease who underwent complete resection and had a 5-year survival of 16% which decreased to only 8% when mediastinal nodes were involved ( 13, 14). Studies have found that after resection, 5-year survivals range from 0% to 25% for patients with Stage III disease overall and are usually less than 10% for those with mediastinal node involvement (9, 13, 14, 17, 2 1). In our group of patients treated with postoperative radiotherapy, the 5year survivals were 36% and 26% for those with Stage III disease and mediastinal involvement, respectively. Retrospective studies have shown increased survival in some subsets of patients following postoperative radiotherapy. Choi et al. reported increased survivals for patients with N 1, N2, and T3 adenocarcinoma; 5-year survival rates after postoperative radiotherapy were 43%, compared with 8% after surgery alone (3). Kirsch and Sloan reported a 26% 5-year survival after postoperative radiotherapy, including 36% for those with squamous cell carcinoma and 13% for those with adenocarcinoma compared with no 5-year survivors among 20 patients who did not receive adjuvant radiotherapy (10). Green et al. found a 35% 5-year survival for patients with nodal metastases who received postoperative radiotherapy but only a 3% survival for those treated with surgery alone (7). The randomized trial of the Lung Cancer Study Group
May I. 199 1. Volume 20, Number 5
found that for patients with Stage II or III squamous cell carcinoma, there was no significant survival difference between those treated with and without adjuvant radiotherapy, although patients with N2 disease had a significantly reduced overall recurrence rate ( 11, 12). Although that study was a prospective, randomized trial, the results can be applied only to the squamous cell carcinoma patients studied, almost two-thirds of whom had Stage II disease. Among those assigned to receive postoperative radiation therapy only 74% received within 5% of the total dose prescribed. Because of this and other criticisms of that study, the question of the effect of postoperative radiation therapy on survival has not yet been answered in a satisfactory way (5, 12).
SUMMARY Surgical resection of lung cancer is associated with significant local and distant recurrence rates, even with apparent complete excision. Radiotherapy can decrease the local recurrences in those patients with minimal residual disease or with hilar or mediastinal lymph node metastases. In this series, postoperative radiotherapy was also beneficial for the 12 patients with disease-negative nodes who were at high risk for recurrent disease because of known residual tumor; these patients had a 90% local control rate and a 78% 5-year survival. The low overall recurrence rate of intrathoracic disease (16%) confirms that postoperative radiotherapy is effective in preventing local relapse even in patients with proven nodal involvement. The impact on survival of these patients cannot be determined from these data, and could only be determined with a well-designed prospective, randomized study.
REFERENCES 1. American Joint Committee for Cancer Staging and End Results Reporting, Task Force on the Lung, C. F. Mountain, Chairman. Cancer of the lung. Manual for staging of cancer. Chicago: AJCS; 1978:59-65. 2. Berkson, J.; Gage, R. P. Calculation of survival rates for cancer. Proc. Staff Meet. Mayo Clin. 25:270-286; 1950. 3. Choi, N. C.; Grille, H. C.; Gardiello, M.; Scannell, J. G.; Wilkins, E. W. Jr. Basis for new strategies in postoperative radiotherapy of bronchogenic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 6:31-35; 1980. 4. Chung, C. K.; Stryker, J. A.; O’Neill, M.; DeMuth, W. E. Jr. Evaluation of adjuvant postoperative radiotherapy for lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 8: 1877-1880; 1982. 5. Cox, J. D. Postoperative mediastinal radiation for cancer of the lung. N. Engl. J. Med. 316:1475; 1987. 6. Ferguson, M. K.; Little, A. G.; Golomb, H. M.; Hoffman, P. C.; DeMeester, T. R.; Beveridge, R.; Skinner, D. B. The role of adjuvant therapy after resection of Tl Nl MO and T2 N 1 MO non-small cell lung cancer. J. Thorac. Cardiovasc. Surg. 91:344-349; 1986.
7. Green, N.; Kurahara, S. S.; George, F. W. Postresection irradiation for primary lung cancer. Radiology 116:405407; 1975. 8. lascone, C.; DeMeester, T. R.; Albertucci, M.; Little, A. G.; Golomb, H. M. Local recurrence of resectable non-oat cell carcinoma of the lung: a warning against conservative treatment for NO and Nl disease. Cancer 57:471-476; 1986. 9. Immerman, S. C.; Vanecko, R. M.; Fry, W. A.; Head, L. R.; Shields, T. W. Site of recurrence in patients with stages I and II carcinoma of the lung resected for cure. Ann. Thorac. Surg. 32:23-27; 198 1. 10. Kirsh, M. M.; Sloan, H. Mediastinal metastases in bronchogenic carcinoma: influence of postoperative irradiation, cell type, and location. Ann. Thorac. Surg. 33:459-463; 1982. 11. The Lung Cancer Study Group. A randomized comparison of the effects of adjuvant therapy on resected stages II and III non-small cell carcinoma of the lung. Ann. Surg. 202: 335-341; 1985. 12. Lung Cancer Study Group. Effects of postoperative mediastinal radiation on completely resected stage II and stage
Non-small
13. 14.
15.
16. 17.
cell bronchogenic
III epidermoid cancer of the lung. N. Engl. J. Med. 315: 1377-1381; 1986. Mountain, C. F. Assessment of the role of surgery for control of lung cancer. Ann. Thorac. Surg. 24:365-373; 1977. Mountain, C. F.; McMurtrey, M. J.; Frazier, 0. H. Regional extension of lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 6:1013-1020; 1980. Mountain, C. F. Biologic, physiologic, and technical determinants in surgical therapy for lung cancer. In: Straus, M. J., ed. Lung cancer: clinical diagnosis and treatment. New York; Grune & Stratton; 1983:245-260. Naruke, T. Staging of N2 disease. Chest 89(Suppl.):338339; 1986. Pearson, F. G. Radical surgery for N2 disease. Chest 89(Suppl.):339-340; 1986.
carcinoma
0 J. D. SLATER et al.
951
18. Shields, T. W. Classification and prognosis of surgically treated patients with bronchial carcinoma: Analysis of VASOG studies. Int. J. Radiat. Oncol. Biol. Phys. 6: 102 l-1027; 1980. 19. Silverberg, E.; Lubera J. Cancer statistics, 1990. Cancer 40: 9-26; 1990. 20. Van Houtte, P. V.; Rocmans, P.; Smets, P.; Goffin, J.-C.; Lustman-Marechal, J.; Vanderhoeft, Pl; Henry, J. Postoperative radiation therapy in lung cancer: a controlled trial after resection ofcurative design. Int. J. Radiat. Oncol. Biol. Phys. 6:983-986; 1980. 2 1. Wright, J. L.; Coppin, C.; Mullen, B. J.; Pare, J. A. P.; Rutherford, T. F.; Ling, H.; Gerein A. N.; Miyagishima, R. T.; Hogg, J. C. Surgical treatment of lung cancer: promise and problems of early diagnosis. Can. J. Surg. 29:205-207; 1986.
20, PP
945-951 Copyright
0360.3016/91 $3 00 + .oO 0 I99 Pergamon Press plc
I
??Original Contribution
RADIATION
JERRY
D. SLATER, CLIFTON
THERAPY FOLLOWING RESECTION OF NON-SMALL CELL BRONCHOGENIC CARCINOMA
M.D.,*
MOUNTAIN,
NANCY
A. ELLERBROEK,
M.D.,+
MARY
JANE
AND LESTER The University
M.D.,* OSWALD,
J. PETERS,
H. THOMAS B.S.,*
JACK
BARKLEY A. ROTH,
JR.,
M.D.,*
M.D.+
M.D.*
of Texas M. D. Anderson Cancer Center, Houston, Texas
Between 1970 and 1982, 102 patients received postoperative radiotherapy after attempted curative resection of bronchogenic carcinoma at The University of Texas M. D. Anderson Cancer Center. Surviving patients had a
minimum follow-up of 3 years. Eight patients had pathological Stage I disease, 29 Stage II, and 65 Stage III. The 5-year actuarial survivals for patients with stages I, II, and III disease were 83%, 55%, and 38%, respectively (p = -04). Corresponding values for patients with NO, Nl, and N2 disease were 74%, 56%, and 28% (p = .Ol). No significant differences in survival were seen based on T stage or tumor histology. Nine patients had gross residual disease following surgery, and 19 had microscopic residual disease. The 5-year actuarial survival was 78% for 12 patients without nodal disease who had known gross (4 patients) or microscopic (8 patients) residual tumor following attempted curative resection. The pathologic status of the hilar and mediastinal lymph nodes was the most significant factor affecting the frequency of metastatic relapse, with 19% of patients with NO, 33% of those with Nl, and 69% of those with N2 disease developing distant disease. The low overall rate of recurrence intrathoracically (16%) confirms that postoperative radiotherapy is effective in preventing local relapse even in patients with proven nodal involvement. The impact of adjuvant radiation therapy on survival cannot be determined from these data, and further data are needed, preferably from well designed prospective studies. Non-small cell lung cancer, Postoperative
radiation therapy.
radiotherapy on local disease control and survival in patients with lung cancer seen at The University of Texas M. D. Anderson Cancer Center.
INTRODUCTION Carcinoma of the lung is the most deadly malignancy and the leading cause of cancer death in the United States. The American Cancer Society estimates that there will be 157,000 new cases and 142,000 deaths from lung cancer in 1990 ( 19). Fewer than 25% of patients with cancer confined to the chest are considered surgical candidates, either because of tumor extent or poor medical condition. However, surgical resection offers the best chance of long-term survival and is the recommended treatment when possible. Despite apparent complete surgical resection, 5-year survival rates for patients with Stage II and III lung carcinoma have been reported to be 20-40% and O-25%, respectively (2,4,6, 8,9, 12-17,20, 2 1). In an effort to improve local control and survival, postoperative radiotherapy has been used. This has decreased the local recurrence rate, although its impact on long-term survival is more controversial. This review will present the results of postoperative
METHODS
AND
MATERIALS
Between 1970 and 1982, 102 patients who underwent attempted curative resection for non-small cell bronchogenie carcinoma were referred to the M. D. Anderson Cancer Center Department of Clinical Radiotherapy for consideration of postoperative irradiation. Patients who had radiation therapy at other institutions were excluded. No patient had another primary malignancy or evidence of distant metastases at the start of radiotherapy, and all living patients had a minimum follow-up of 3 years. Surgery consisted of wedge resection in 6 patients, lobectomy in 43, bilobectomy in 3, and pneumonectomy in 50. Squamous cell carcinoma was seen in 58 patients
Presented at the 30th Annual Scientific Meeting of the American Society for Therapeutic Radiology and Oncology in New Orleans, LA, October 9-14, 1988. * Dept. of Clinical Radiotherapy. + Dept. of Thoracic Surgery. Reprint requests to: Nancy A. Ellerbroek, M.D., Department
of Radiation Oncology, Loma Linda University Medical Center, 11234 Anderson St., Loma Linda, CA 92354. Supported by grant CA06294 awarded by the National Cancer Institute, United States Department of Health and Human Services. Accepted 945
for publication
9 November
1990.
I. J. Radiation Oncology 0 Biology 0 Physics
946
May 1, 1991, Volume 20, Number 5
and adenocarcinoma in 39; 5 patients had large cell, bronchioalveolar, or unclassified carcinoma. Tumors were pathologically staged according to the American Joint Committee on Cancer (AJC) classification (1); 8 patients had Stage I disease, 29 Stage II, and 65 Stage III. The relationships of tumor stage to histology are shown in Table 1. Twenty patients had NO, 35 N 1, and 47 N2 disease. During the same time period, 771 patients with an initial diagnosis of non-small cell lung cancer underwent curative resection. There were 655 patients whose AJC stage was known and who did not have small cell carcinoma or operative deaths; of those, 333 had Stage I disease, 109 Stage II, and 213 Stage III. There were 588 apparent complete resections, and 67 cases where microscopic or gross residual disease remained. Of 645 whose N Stage was known, 382 had NO disease, 148 Nl, and 115 N2. Fifteen patients received chemotherapy with cyclophosphamide, Adriamycin, and cisplatin as part of their primary treatment. Thirteen of those 15 had Stage III disease. In one patient cisplatin was omitted; another also had received DTIC and BCG. Chemotherapy followed surgery in all cases, and usually consisted of 1 to 3 cycles of CAP (median 2) and preceded radiotherapy in all but one patient. Seven patients had additional chemotherapy after radiotherapy. Radiotherapy was started approximately 4 weeks postoperatively in all but two patients unless the patient received chemotherapy, in which case it was started an average of 12 weeks after surgery (range 2-26 weeks). The indications for patient referral for radiotherapy changed during the time of this analysis. Prior to 1978, patients were not routinely referred for postoperative radiotherapy unless the surgeon believed there was increased risk of recurrent disease because of microscopic or gross residual tumor, extensive lymph node involvement, or evidence of extension of tumor outside the lymph nodes. From 1978 on, patients were also routinely referred for radiotherapy if they had positive hilar or mediastinal lymph nodes.
Table 1. Tumor
Squamous carcinoma n = 58
TIN0
1
Adenocarcinoma n = 39 Other n=5 Total n = Total number.
1
RESULTS Overall survival Actuarial survivals were calculated using the BerksonGage life table method (2). Patients who died of unknown causes were classified as having died of cancer. The overall
pathologic
stage and histology
Stage II (n = 29)
Stage I (n = 8) Histology
Overall, 28 patients had residual disease following surgery: 9 with gross residual disease and 19 with microscopic residual disease suspected because of positive surgical margins. All patients with NO disease who were referred for radiation therapy had suspected residual tumor, although tumor was confirmed in only 12 of 20 such cases upon review of pathologic and surgical reports. The other eight patients had large primary tumors and were considered at increased risk for local recurrence despite apparent complete removal of the tumors. Radiation therapy fields included the mediastinum, ipsilateral hilum, and supraclavicular lymph nodes in addition to areas of known microscopic or macroscopic disease. Eighty-four patients were treated with a continuous course of radiation, with the majority receiving 50 Gy in 25 fractions over 5 weeks. Treatment was given AP-PA initially, and a reduced-field boost, usually of 5-10 Gy, was given to 9 of 28 patients with postoperative residual disease, usually using opposed oblique fields designed to encompass the area of residual disease. Doses were prescribed to the midline or the isocenter without lung corrections. As CT planning was not in use during the study period, the treatment volume was verified with conventional simulation and portal films. In 18 patients a regimen of split-course radiation therapy, generally consisting of 30 Gy in 10 fractions, was planned, followed by a 3-week rest, at which time the patients were reevaluated for administration of an additional 30 Gy in 10 fractions. Of the 18 patients, 5 did not receive the second course of radiation either because they refused further treatment or because their conditions had deteriorated. Elective cranial irradiation was given to four patients with adenocarcinoma.
Stage III (n = 65)
TlNl
T2Nl
2
2
19
7
2
1
8
7
2
1
29
15
T2NO
4
3
T3NO
T3Nl
3
TIN2
T2N2
T3N2
2
17
4
3
17
1
TxN2
1
2 3
5
36
5
1
Non-small cell bronchogenic carcinoma 0 J. D. Survival
Survival
of All Patients
Fig. 1. Overall survival patterns of 102 patients treated with postoperative radiotherapy following attempted complete surgical resection.
2- and 5-year survival rates are 59% and 46%, respectively (Fig. 1). Only one patient is alive with disease.
Histology and survival The actuarial 2- and 5-year survivals (Fig. 2) for those with squamous cell carcinoma are 67% and 55%, respectively, and 50% and 38% for adenocarcinoma (p = .39).
Stage and survival The pathologic stage of disease affected survival as shown in Figure 3. At 5 years, the survivals were 83%, 55%, and 38% for patients with stages I, II, and III disease, respectively (p = .04). Extent of nodal involvement had the most significant impact on survival. Figure 4 shows the following 5-year survivals for patients with the stated disease stages who received postoperative irradiation: 74% for NO, 56% for Nl, and 28% for N2 patients (p = .O 1). There was no significant difference in survival for NO patients with known (78%) or suspected residual (71%).
Survival
947
SLATER et al.
Fig. 3. Survival of patients
of each patients
according
to disease stage.
Patients with Tl tumors had a survival rate of 68%, whereas 5-year survivals for those with T2 and T3 tumors were 43% and 5 I%, respectively. The increased survivals associated with T3 as compared with T2 tumors were probably caused by a higher incidence of mediastinal node involvement in our group of patients with T2 tumors. Of the patients with T3 NO disease, 3 with Pancoast tumors had a median survival of 68 months (range, 7 to 116), and 12 with other presentations had a median survival of 24 months (range, 2 to 124). One patient with Stage I, 2 with Stage II, and 12 with Stage III disease received chemotherapy as part of their primary treatment. When the Stage III disease patients who received chemotherapy were compared with those who did not, there was no detectable difference in actuarial survival (p = .90). Because of the small number of patients who received chemotherapy as part of other protocols and the heterogeneous group of patients involved, no evaluation of the effects of chemotherapy can be made. The 5 patients who did not complete the second part of their split course of treatment had a short median survival of 2 months; in 4 of the 5 the course was not completed because of progressive disease or complications.
by Histology Survival
Fig. 2. Survival type.
by Stage
according
to histologic
by Nodal Status
tumor Fig. 4. Survival of patients
according
to nodal status.
1. J. Radiation
948
Oncology
0 Biology 0 Physics
Putterns oftumor recurrence The cause of death was not known for 14 patients; thus they were excluded from analysis of disease recurrence sites. Of the remaining 88 patients, 49 have failed (Table 2) including 27 with squamous cell carcinoma (53%) and 20 with adenocarcinoma (59%). Two patients with pathologic Stage I (29%), 9 with Stage II (38%) and 38 with Stage III disease (67%) had recurrences.
May I, I99 I. Volume 20, Number 5
The local recurrence rates for Stages I, II, and III disease were 14%, 4%, and 2 l%, respectively. In 6 of 12 patients with Stage III disease who had local recurrence, it was their only site of failure. The local recurrence rate was 12% for patients with NO disease and 10% and 2 1% for those with N 1 and N2 disease, respectively. The local recurrence rate for patients with NO and known residual disease ( 10%) did not differ significantly from that for patients with a high suspicion of residual disease (17%).
Local tumor recurrence Fourteen patients ( 16%) had recurrences within the irradiated volume. The time to recurrence had a range of 3-40 months after completion of radiotherapy, with a mean of 13 months. Local recurrence as the only site of failure was seen in 7 patients and distant metastasis alone in 35 (Table 2). Seven patients had recurrences at both local and distant sites: three had local recurrences prior to developing distant metastases 1 to 5 months later, two developed distant metastases prior to local recurrence, and two others developed both simultaneously. It was difficult in this retrospective study to make a correlation between dose and local control. Only one of the patients who had a local recurrence had received less than 50 Gy. An analysis of the fields used and the locations of the recurrences yielded one case where a patient with an upper lobe primary had disease recur in an untreated ipsilateral supraclavicular region. One was possibly a second primary, occurring in the right upper lobe 40 months after treatment for a left lower lobe primary, and the other recurrences were in the treated area. There were 2 local recurrences in 8 patients with macroscopic residual disease postoperatively and 3 recurrences in 17 patients with microscopic residual disease. The overall local recurrence rate for patients with known gross or microscopic residual disease (20%) did not differ significantly from that for patients with NO disease and only suspected residual disease ( 17%). In 8 ( 16%) of 5 1 evaluable patients with squamous cell carcinoma, disease recurred locally; in 6 this was their only site of recurrence. Six patients with adenocarcinoma (18%) had local, recurrence; in one this was the only site of failure.
Distant diiseuse Distant metastases have developed in 42 (48%) of the 88 patients evaluable for patterns of recurrence, with a mean onset of 12 months following radiation therapy. In 35 (83%) of the patients who developed distant metastases, this was their only site of failure. The pathologic status of the hilar and mediastinal lymph nodes was the most significant factor affecting the metastatic rate, with 3/16 (19%) of patients with NO, lO/ 30 (33%) with Nl, and 29/42 (69%) with N2 disease developing distant metastases. The data revealed a trend toward more frequent distant metastases as the overall disease stage and T stage increase (Table 3). Distant metastases were seen in 56% of patients with adenocarcinoma and in 4 1% with squamous cell carcinoma. The most common initial sites of metastases were lung (outside the treatment field), and bone (Table 4). One of four patients with adenocarcinoma who received elective cranial irradiation eventually developed brain metastases.
Complications Twelve patients developed symptoms related to the radiation, with the majority of these symptoms being selflimited or causing minimal discomfort. Symptomatic pneumonitis was seen in two patients following radiation therapy, whereas four patients developed pericarditis symptoms that resolved with conservative management. Four patients died of complications of treatment: one patient died of pneumonia 6 months after irradiation but
Table 3. Incidence Table
2. Tumor histology and sites of recurrence
Histology Squamous cell n= 51 Adenocarcinoma n = 34 Other n=3 Total = 50/88 n = Total number.
Local
Distant
Local and distant
6
19
2
1
14
5
0
2
0
7
35
7
Stage
11
III Tl+ T2 T3 NO Nl N2
of distant
metastasis Distant
and stage of disease metastases*
l/7 9/24 32157 3/8 30/57 9/20 3/16 10/30 29142
(14%) (38%) (56%) (30%) (53%) (45%) (19%) (33%) (69%)
* Excludes 14 patients whose cause of death was not known. + Excludes one patient with TX primary stage.
Non-small cell bronchogenic carcinoma 0 J. D. SLATERetal.
949
Table 4. Tumor histology and first site of distant metastasis Tumor Sites Lung
Other or multiple
Total
Brain
Liver
Bone
Squamous cell Adenocarcinoma Other
5 3 1
3 1 0
5 3 1
5 5 0
3 7 0
21 19 2
Total
9
4
9
10
10
42
Histology
was noted to have “radiation fibrosis,” which may have contributed to his death; one died with Klebsiella pneumonia shortly after completion of irradiation; one developed a tracheoesophageal fistula after receiving 20 Gy associated with necrosis of tumor involving those structures; and one developed a hydro-pneumothorax early in the course of irradiation. Four patients developed subcutaneous fibrosis in the lower neck, which was associated with minimal limitations in mobility. Two others, both of whom were treated for Pancoast tumors, developed apparent brachial plexopathy with mild muscle wasting and paresthesias, one after receiving 60 Gy in 20 fractions over 50 days and another following 54 Gy in 26 fractions over 37 days. Recurrent tumor was a possible cause of the neurologic symptoms in the patient who had received 60 Gy.
DISCUSSION The goal of delivering postoperative radiation therapy is to improve control of the local tumor and, it is hoped, to increase the survival of these patients. Local recurrence, even of early-stage disease, can be a significant problem. After apparent complete surgical resection in patients with T 1-T2, NO-N 1 disease, local recurrence rates have been reported to be as high as 41%, with most studies indicating a lo-25% incidence (4, 8, 9, 20). The ability of postoperative radiotherapy to improve the local disease control rate has been shown (4, 20). In our group, 37 patients with T 1-T2, NO-N 1 disease received postoperative radiotherapy. Two (6%) of 3 1 evaluable patients had local recurrences whereas only 1 (4%) of 26 of these patients without evidence of residual disease did. Radiation therapy is not routinely given to patients with no lymph node involvement; however, when the surgical margins are involved or when there is gross residual disease, radiation can be used to decrease the likelihood of local recurrence. In our study, 20 patients without lymph node involvement received postoperative radiation therapy for residual disease (12 patients) or for suspected residual disease (8 patients) because of primary tumor size or extent. Despite these high-risk factors, local control was obtained in 88% of patients, a finding consistent with that for patients who had undergone complete resection (7, 8, 14, 20, 21).
Radiation therapy also decreases the local recurrence rate in patients with more advanced disease. The Lung Cancer Study Group, in a randomized trial of postoperative radiation therapy versus no further treatment for patients with Stage II or III squamous cell carcinoma, found a significant decrease in local recurrence rate in the patients treated with radiation (11, 12). Retrospective studies have also reported decreased local recurrence following postoperative irradiation. Choi et al. reported that 65% of all recurrences of Nl, N2, and T3 squamous carcinoma treated by surgery alone were regional; this figure decreased to 32% in those treated with postoperative radiation therapy (3). Kirsh and Sloan reported results for 20 patients with mediastinal metastases who underwent only surgical resection ( 10). Eight of 10 patients in whom the cause of death was known died with evidence of recurrent disease in the thorax. However, only 4 of 54 patients who received postoperative radiotherapy in whom the cause of death was known developed local recurrence. In our group of 65 patients with Stage III disease, the local recurrence rate with postoperative radiation therapy was 22%; it was the same for all patients who had Stage III disease with hilar or mediastinal node involvement. Excluding those Stage II and III disease patients who had known residual disease postoperatively, the local recurrence rate was 15%. Although the effectiveness of radiation therapy in improving control of local disease is well documented, its effect on improving survival is more controversial. Resection of Stages I and II non-small cell carcinomas of the lung without adjuvant treatment has resulted in 5year survival rates of 20-60%, with the higher survival for patients without lymph node involvement (6, 9, 13, 15, 18, 2 1). Van Houtte et al. conducted a randomized trial on the effects of postoperative irradiation in patients without evidence of lymph node metastases and found no increase in survival associated with radiation therapy (20). There are, however, subsets of patients with early-stage disease and poor prognostic factors (such as residual disease) who might benefit from postoperative radiation therapy to eradicate the remaining tumor. In our study group, 20 patients without nodal involvement were thought to be at high risk for recurrence because they were known to have residual disease postoperatively (I 2
950
1. J. Radiation Oncology 0 Biology 0 Physics
patients) or were suspected of having residual disease due to tumor size or extent (8 patients). Despite these findings, their 5-year survivals were 78% and 7 1% for known or suspected residual disease, respectively, following radiation therapy. The survival after surgery in patients with more advanced disease is poor. The American Joint Committee on Cancer Task Force on Lung reported on 36 1 patients with Stage III disease who underwent complete resection and had a 5-year survival of 16% which decreased to only 8% when mediastinal nodes were involved ( 13, 14). Studies have found that after resection, 5-year survivals range from 0% to 25% for patients with Stage III disease overall and are usually less than 10% for those with mediastinal node involvement (9, 13, 14, 17, 2 1). In our group of patients treated with postoperative radiotherapy, the 5year survivals were 36% and 26% for those with Stage III disease and mediastinal involvement, respectively. Retrospective studies have shown increased survival in some subsets of patients following postoperative radiotherapy. Choi et al. reported increased survivals for patients with N 1, N2, and T3 adenocarcinoma; 5-year survival rates after postoperative radiotherapy were 43%, compared with 8% after surgery alone (3). Kirsch and Sloan reported a 26% 5-year survival after postoperative radiotherapy, including 36% for those with squamous cell carcinoma and 13% for those with adenocarcinoma compared with no 5-year survivors among 20 patients who did not receive adjuvant radiotherapy (10). Green et al. found a 35% 5-year survival for patients with nodal metastases who received postoperative radiotherapy but only a 3% survival for those treated with surgery alone (7). The randomized trial of the Lung Cancer Study Group
May I. 199 1. Volume 20, Number 5
found that for patients with Stage II or III squamous cell carcinoma, there was no significant survival difference between those treated with and without adjuvant radiotherapy, although patients with N2 disease had a significantly reduced overall recurrence rate ( 11, 12). Although that study was a prospective, randomized trial, the results can be applied only to the squamous cell carcinoma patients studied, almost two-thirds of whom had Stage II disease. Among those assigned to receive postoperative radiation therapy only 74% received within 5% of the total dose prescribed. Because of this and other criticisms of that study, the question of the effect of postoperative radiation therapy on survival has not yet been answered in a satisfactory way (5, 12).
SUMMARY Surgical resection of lung cancer is associated with significant local and distant recurrence rates, even with apparent complete excision. Radiotherapy can decrease the local recurrences in those patients with minimal residual disease or with hilar or mediastinal lymph node metastases. In this series, postoperative radiotherapy was also beneficial for the 12 patients with disease-negative nodes who were at high risk for recurrent disease because of known residual tumor; these patients had a 90% local control rate and a 78% 5-year survival. The low overall recurrence rate of intrathoracic disease (16%) confirms that postoperative radiotherapy is effective in preventing local relapse even in patients with proven nodal involvement. The impact on survival of these patients cannot be determined from these data, and could only be determined with a well-designed prospective, randomized study.
REFERENCES 1. American Joint Committee for Cancer Staging and End Results Reporting, Task Force on the Lung, C. F. Mountain, Chairman. Cancer of the lung. Manual for staging of cancer. Chicago: AJCS; 1978:59-65. 2. Berkson, J.; Gage, R. P. Calculation of survival rates for cancer. Proc. Staff Meet. Mayo Clin. 25:270-286; 1950. 3. Choi, N. C.; Grille, H. C.; Gardiello, M.; Scannell, J. G.; Wilkins, E. W. Jr. Basis for new strategies in postoperative radiotherapy of bronchogenic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 6:31-35; 1980. 4. Chung, C. K.; Stryker, J. A.; O’Neill, M.; DeMuth, W. E. Jr. Evaluation of adjuvant postoperative radiotherapy for lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 8: 1877-1880; 1982. 5. Cox, J. D. Postoperative mediastinal radiation for cancer of the lung. N. Engl. J. Med. 316:1475; 1987. 6. Ferguson, M. K.; Little, A. G.; Golomb, H. M.; Hoffman, P. C.; DeMeester, T. R.; Beveridge, R.; Skinner, D. B. The role of adjuvant therapy after resection of Tl Nl MO and T2 N 1 MO non-small cell lung cancer. J. Thorac. Cardiovasc. Surg. 91:344-349; 1986.
7. Green, N.; Kurahara, S. S.; George, F. W. Postresection irradiation for primary lung cancer. Radiology 116:405407; 1975. 8. lascone, C.; DeMeester, T. R.; Albertucci, M.; Little, A. G.; Golomb, H. M. Local recurrence of resectable non-oat cell carcinoma of the lung: a warning against conservative treatment for NO and Nl disease. Cancer 57:471-476; 1986. 9. Immerman, S. C.; Vanecko, R. M.; Fry, W. A.; Head, L. R.; Shields, T. W. Site of recurrence in patients with stages I and II carcinoma of the lung resected for cure. Ann. Thorac. Surg. 32:23-27; 198 1. 10. Kirsh, M. M.; Sloan, H. Mediastinal metastases in bronchogenic carcinoma: influence of postoperative irradiation, cell type, and location. Ann. Thorac. Surg. 33:459-463; 1982. 11. The Lung Cancer Study Group. A randomized comparison of the effects of adjuvant therapy on resected stages II and III non-small cell carcinoma of the lung. Ann. Surg. 202: 335-341; 1985. 12. Lung Cancer Study Group. Effects of postoperative mediastinal radiation on completely resected stage II and stage
Non-small
13. 14.
15.
16. 17.
cell bronchogenic
III epidermoid cancer of the lung. N. Engl. J. Med. 315: 1377-1381; 1986. Mountain, C. F. Assessment of the role of surgery for control of lung cancer. Ann. Thorac. Surg. 24:365-373; 1977. Mountain, C. F.; McMurtrey, M. J.; Frazier, 0. H. Regional extension of lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 6:1013-1020; 1980. Mountain, C. F. Biologic, physiologic, and technical determinants in surgical therapy for lung cancer. In: Straus, M. J., ed. Lung cancer: clinical diagnosis and treatment. New York; Grune & Stratton; 1983:245-260. Naruke, T. Staging of N2 disease. Chest 89(Suppl.):338339; 1986. Pearson, F. G. Radical surgery for N2 disease. Chest 89(Suppl.):339-340; 1986.
carcinoma
0 J. D. SLATER et al.
951
18. Shields, T. W. Classification and prognosis of surgically treated patients with bronchial carcinoma: Analysis of VASOG studies. Int. J. Radiat. Oncol. Biol. Phys. 6: 102 l-1027; 1980. 19. Silverberg, E.; Lubera J. Cancer statistics, 1990. Cancer 40: 9-26; 1990. 20. Van Houtte, P. V.; Rocmans, P.; Smets, P.; Goffin, J.-C.; Lustman-Marechal, J.; Vanderhoeft, Pl; Henry, J. Postoperative radiation therapy in lung cancer: a controlled trial after resection ofcurative design. Int. J. Radiat. Oncol. Biol. Phys. 6:983-986; 1980. 2 1. Wright, J. L.; Coppin, C.; Mullen, B. J.; Pare, J. A. P.; Rutherford, T. F.; Ling, H.; Gerein A. N.; Miyagishima, R. T.; Hogg, J. C. Surgical treatment of lung cancer: promise and problems of early diagnosis. Can. J. Surg. 29:205-207; 1986.
