- Email: [email protected]
Induction chemoradiation compared with induction radiation for lung cancer involving the superior sulcus
Induction Chemoradiation Compared With Induction Radiation for Lung Cancer Involving the Superior Sulcus Cameron D. Wright, MD, Matthew T. Menard, MD, John C. Wain, MD, Dean M. Donahue, MD, Hermes C. Grillo, MD, Thomas J. Lynch, MD, Noah C. Choi, MD, and Douglas J. Mathisen, MD General Thoracic Surgical Unit, Medical Oncology Unit, and Radiation Oncology Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts
Background. The usual approach of induction radiation therapy (RT) followed by resection of superior sulcus tumors results in many incomplete resections, a high local recurrence rate, and suboptimal survival. Induction chemoradiotherapy (CT/RT) has been shown to reduce local and distant recurrences and improve survival in stage III lung cancer. We investigated the role of induction CT/RT in superior sulcus patients. Methods. This was a single-institution, retrospective study. Results. From 1985 to 2000, 35 consecutive patients underwent induction treatment followed by resection of a superior sulcus tumor. All patients had mediastinoscopy first to exclude N2 disease, and all were N0 at final pathologic examination. Twenty patients had induction RT (mean, 39 Gy), and 15 had induction CT/RT (mean, 51 Gy) with concurrent cisplatin-based chemotherapy. There was no treatment mortality. Complete resection
was performed in 16 of 20 (80%) of the RT patients and in 14 of 15 (93%) of the CT/RT patients (p ⴝ 0.15). The pathologic response from the induction treatment was complete or near complete in 7 of 20 (35%) of the RT patients and in 13 of 15 (87%) of the CT/RT patients (p ⴝ 0.001). The median follow-up was 167 months in the RT patients and 51 months in the CT/RT patients. Two-year and 4-year survival was 49% and 49% (95% confidence interval, 26% to 71%) in the RT patients and 93% and 84% (95% confidence interval, 63% to 100%) in the CT/RT patients, respectively (p ⴝ 0.01). The local recurrence rate was 6 of 20 (30%) in the RT patients and 0 in the CT/RT patients (p ⴝ 0.02). Conclusions. Induction CT/RT for superior sulcus tumors appears to offer improved survival compared with induction RT alone. (Ann Thorac Surg 2002;73:1541– 4) © 2002 by The Society of Thoracic Surgeons
N
atomic area [2–5]. These high incomplete resection rates doom the patient to inadequate local control and lead to death, suggesting the need for better local control measures. Dartevelle and colleagues [7] recently introduced the anterior approach for select patients, which appears to enhance complete resection rates. Despite aggressive resection the most common site of first relapse is local [5]. Induction chemoradiotherapy (CT/RT) of stage III non– small-cell lung cancer has proven to be very effective at downstaging tumors and seemingly improving survival in a group of patients with a very high chance of both local and distant relapse [8, 9]. It would seem a priori that combined modality therapy would be most useful when tumor downstaging would facilitate a complete resection in the absence of N2 disease, a known marker for occult micrometastatic disease and subsequent death from distant metastases. We started treating patients with superior sulcus tumors with combined modality therapy based on our stage IIIA N2 results in 1991 [8] and have gradually changed our approach such that now we favor that approach. This period of practice pattern change in a single institution allows us to examine two approaches of induction treatment (RT and CT/RT) on outcome.
on–small-cell lung cancer involving structures at the thoracic inlet represents a small subset of potentially resectable lung cancer patients. Initially thought to be inoperable, the approach of low-dose (30 Gy) induction radiation therapy (RT) followed by resection was discovered by serendipity and promoted by Shaw and associates [1]. Numerous reports suggest that a 30% to 40% 5-year survival will be achieved with that approach [2–5]. A recent large series with multivariate analysis confirmed that the three most important prognostic factors were the ability to perform a complete resection and the T and N status of the tumor [5]. Although preoperative RT is favored by most surgeons, others have favored resection followed by RT, preferring to give full-dose RT that is not split course for radiobiologic reasons [6]. Most reported series document a relatively high incomplete resection rate (30% to 50%), highlighting the difficulty in performing a complete resection in this challenging an-
Accepted for publication Jan 23, 2002. Address reprint requests to Dr Wright, Blake 1570, Massachusetts General Hospital, Boston, MA 02114; e-mail: [email protected].
© 2002 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
0003-4975/02/$22.00 PII S0003-4975(02)03471-9
1542
WRIGHT ET AL INDUCTION THERAPY FOR SUPERIOR SULCUS TUMORS
Patients and Methods Patients with non–small-cell lung cancer invading the superior pulmonary sulcus were identified from the thoracic surgery database during a 15-year period (1985 to 2000). Only patients undergoing induction therapy followed by resection were reviewed. A retrospective chart review was performed, and the following data were recorded: patient age, sex, presence of pain pretreatment, induction treatment, adjuvant treatment, complete resection status, TNM status, length of stay, treatment-related complications, radiologic and pathologic response to induction treatment, disease status at last follow-up, and the sites of any relapse.
Patient Evaluation All patients had a chest computed tomographic scan, which included the liver and adrenal glands, and a diagnosis made by fine-needle aspiration. All had routine brain and bone scans to exclude occult metastatic disease. Since 1990, the majority of patients (70%) have had a magnetic resonance scan of the thoracic inlet area to delineate the anatomy. All patients had mediastinoscopy before starting induction treatment, and all were found to be negative for N2 nodal metastases.
Radiation Therapy The radiation target included the primary tumor, ipsilateral hilum, mid and upper mediastinum, and ipsilateral supraclavicular regions. Patients having induction RT had a preoperative mean dose of 39 Gy (range, 27 to 64 Gy) delivered in 2-Gy fractions, and the majority of patients had a postoperative boost dose (mean, 10 Gy; range, 0 to 27 Gy). Patients who had induction CT/RT had a preoperative mean dose of 51 Gy (range, 40 to 66 Gy). Four patients had hyperfractionated RT with 180 cGy given twice daily [8]. One CT/RT patient with an incomplete resection had postoperative RT.
Chemotherapy Four patients had two preoperative and one postoperative cycles of concurrent chemotherapy consisting of cisplatin 100 mg/m2 on day 1, 5-fluoruracil 30 mg/kg per day for 3 days starting day 1, and vinblastine 4 mg/m2 on day 1 [8]. Eleven patients had two preoperative cycles of concurrent chemotherapy consisting of cisplatin 50 mg/m2 on days 1, 8, 29, and 36 and etoposide 50 mg/m2 on days 1 to 5 and 29 to 33 [9]. Five patients had two additional cycles of the same chemotherapy postoperatively based on residual tumor in the resection specimen.
Statistical Analysis Survival was estimated by the product-limit method of Kaplan and Meier, and curves were compared by the log-rank test. Cox regression analysis was used to assess prognostic factors for their significance in predicting survival. Categorical variables were compared by the 2 test. Continuous variables were compared with the Student’s t test. All significance values were two-tailed.
Ann Thorac Surg 2002;73:1541– 4
Table 1. Patient Characteristics Induction Factor Year (average) Age (mean) T3 T4 Amount RT (Gy) Anterior approach Lobectomy Complete resection LOS
RT
CT/RT
p Value
1989 (1985–99) 54 (33– 86) 17 3 39 (30 – 63) 1/20 17/20 16/20 11 (6 –26)
1996 (1991–99) 58 (42– 68) 8 7 51 (40 – 66) 5/15 13/15 14/15 7.5 (4 –26)
⬍ 0.0001 0.31
CT/RT ⫽ chemoradiotherapy; tion therapy.
LOS ⫽ length of stay;
0.04 0.0002 0.03 0.89 0.15 0.07 RT ⫽ radia-
Results Thirty-five patients underwent treatment of a superior sulcus tumor with induction therapy followed by resection from 1985 to 2000. Eighteen patients were men and 17 were women. All patients had mediastinoscopy before treatment, and all were found to be free of mediastinal nodal metastases. Compared with CT/RT patients, those who had induction RT were treated on average 7 years earlier, were of the same age, had less RT, were less often resected by the anterior approach, had the same rate of lobectomy, trended toward more complete resections and a longer length of stay, and had less T4 tumors (Table 1). Side effects of induction RT were grade 2 esophagitis in 2 patients and fatigue and anorexia in 1 patient each. Side effects of induction CT/RT were grade 2 esophagitis in 4 patients and subclavian vein thrombosis in 1 patient. No patient suffered complications requiring hospitalization during induction therapy. There was no operative mortality and only five major postoperative complications (atrial fibrillation, 2; pneumonia, 1; prolonged air leak, 1; and cerebrospinal fluid leak, 1). Thirty patients underwent lobectomy, and 5 had a wedge resection. Twenty-nine patients had the classic extended posterolateral thoracotomy approach, and 6 had a combined anterior and posterior approach. The T4 tumors included 5 patients who underwent partial vertebrectomy and 5 patients who had subclavian artery resection. In the induction RT patients, 33% had a major (⬎ 50% reduction) radiologic response, 33% had a minor (⬎ 25% reduction) response, and 33% had no (⬍ 25% reduction) response. In the induction CT/RT patients, 70% had a major radiologic response and 30% had a minor response. All patients were N0 on the final pathology report. There was a marked difference in pathologic response based on the induction therapy favoring CT/RT (p ⫽ 0.001; Table 2). There was a marked survival advantage seen in the CT/RT patients over those with RT alone (p ⫽ 0.01; Fig 1). The 2-year and 4-year survival estimates in the CT/RT group were 93% (95% confidence interval, 81% to 100%) and 84% (95% confidence interval, 63% to 100%) compared with 49% (95% confidence interval, 26% to 71%)
Ann Thorac Surg 2002;73:1541– 4
1543
WRIGHT ET AL INDUCTION THERAPY FOR SUPERIOR SULCUS TUMORS
Table 2. Influence of Induction Therapy on Pathologic Response
Table 3. Influence of Induction Therapy on Site of First Recurrence
Induction
Pathologic response
RT
CT/RT
p Value
Complete Microscopic Tumor
2 5 13
10 3 2
0.001
CT/RT ⫽ chemoradiotherapy;
RT ⫽ radiation therapy.
and 49% (95% confidence interval, 26% to 71%) in the RT patients. The hazard ratio for RT was 5.6 (95% confidence interval, 1.2 to 25.2; p ⫽ 0.02). The median follow-up was 167 months in the RT patients and 51 months in the CT/RT patients. Six patients had a locoregional recurrence, and 10 had a distant recurrence noted during follow-up. Sites of distant recurrence included brain (n ⫽ 5), adrenal (n ⫽ 2), lung (n ⫽ 2), and bone (n ⫽ 1). There was no difference in distant recurrences between the two induction groups (Table 3). There was a reduction in local recurrence in the CT/RT induction patients (p ⫽ 0.02).
Comment The two different induction groups were fairly evenly matched for the three important prognostic factors for superior sulcus tumor patients (T and N status and complete resection) [5]. All patients were N0. The rate of complete resections was higher in the CT/RT group but not significantly (p ⫽ 0.15) so, probably owing to the small numbers of patients in the study. The number of complete resections in the CT/RT group is actually fairly high because more T4 tumors were resected. There were more T4 tumors in the CT/RT group (p ⫽ 0.04), probably
Fig 1. Kaplan–Meier survival curve for superior sulcus tumor patients treated by induction radiation therapy (RT) and induction chemoradiotherapy (CT/RT). (Cum ⫽ cumulative.)
Recurrence
RT
CT/RT
p Value
Local Distant
6/20 7/20
0/15 3/15
0.02 0.33
CT/RT ⫽ chemoradiotherapy;
RT ⫽ radiation therapy.
reflecting both greater confidence in the ability to perform a complete resection by an anterior approach and to sterilize the tumor with the induction therapy. Despite having more T4 tumors, the group that underwent CT/RT as induction therapy had a markedly improved survival. The morbidity was not appreciably greater in patients receiving CT/RT, and there was no mortality with either treatment. There was a trend for a reduction in length of stay in the group receiving CT/RT probably related to our use of pathways and reflects the average 7-year difference between the two groups [10]. Certainly the length of stay does not appear to be prolonged with the addition of chemotherapy to induction RT. Induction CT/RT was remarkably effective at sterilizing the tumor, with a 67% complete response rate and an 87% rate of near or complete response. It is likely that the improved survival seen was caused in large part by this very high sterilization rate. Two previously published reports have also suggested that CT/RT is the optimal induction strategy for superior sulcus tumor patients. Martinez-Monge and coworkers [11] in 1994 reported 18 patients treated with two different mitomycin, vindesine, and cisplatin regimens. There was a very high treatment mortality rate of 17%, a high pathologic complete response rate (71%), and a favorable 56% 4-year survival rate. Attar and associates [12] in 1998 briefly reported on 11 patients treated with carboplatin and paclitaxel. Treatment toxicity was not stated, but the 5-year survival was 72%. Rusch and colleagues [13] recently reported a large prospective multiinstitutional trial (Intergroup 0160) of induction CT/RT (based on Southwest Oncology Group 8805 [9]) for superior sulcus tumors. There were 111 eligible patients, and 102 had the planned induction treatment. There were 5 (5%) treatment-related deaths. Eighty-three patients underwent thoracotomy, and 76 (92%) had a complete resection. The pathologic response to induction CT/RT was complete or with only minimal microscopic residual disease in 65% of the patients. The 2-year and 4-year survival was 55%. The complete resection rate was high (90%), and the locoregional recurrence rate was low (23%). The authors concluded that the complete resection rate and survival appeared improved compared with historic experience. A randomized trial of therapy in this small subset of patients with non–small-cell lung cancer will probably never be accomplished because of the rarity of this tumor. Accordingly, careful scrutiny of the Intergroup trial and the other case series with comparison to historic experience with induction RT is necessary to determine whether an apparent improvement in care is real. Our series is particularly helpful in this regard inasmuch as
1544
WRIGHT ET AL INDUCTION THERAPY FOR SUPERIOR SULCUS TUMORS
there is a period of time during which the two induction programs overlapped, the patients were carefully staged both radiographically and by mediastinoscopy, and the surgical strategy was uniform. A fair comparison between the two induction programs might conclude that induction CT/RT (1) can be given with acceptably low morbidity, (2) trends toward a higher complete resection rate, (3) results in a very high pathologic response rate, (4) results in a reduced locoregional recurrence rate (the ultimate test of the completeness of resection), and (5) offers improved survival. There are several limitations of our study. The first is the retrospective nature of this consecutive case series in which treatment was not by protocol. The second is that there may have been case selection bias favoring better patients undergoing induction CT/RT. However, the three major prognostic factors were fairly evenly distributed between the two groups, and there were more T4 tumors in the CT/RT group. Third, there was probably also an element of time selection bias favoring the CT/RT patients who likely benefited from better staging (new spiral computed tomographic scan), improved surgical procedure owing to use of the anterior approach and the learning effect, and a general improvement of care that comes with the time. Nonetheless, many patients were treated simultaneously with both induction programs, and all patients were vigorously staged. Fourth, the median follow-up was longer in the RT patients, which allows more time for failure. Lastly, the sample size is small, and several comparisons (complete resection rate and length of stay) would likely have been significant if the study were larger (type II error). Induction CT/RT followed by resection for superior sulcus tumors appears superior to the historic approach of induction RT alone. Future trials should investigate the feasibility of dose escalation of radiation as medium-dose RT (50 Gy) appears well tolerated and local control is still not ideal [14]. Other chemotherapeutic agents that act as radiation sensitizers should be evaluated in these patients now that a baseline pathologic response rate has been established for the standard regimen of cisplatin and etoposide. Adjuvant postoperative chemotherapy and central nervous system prophylaxis need to be investigated as distant metastases now represent the most common site of failure.
Ann Thorac Surg 2002;73:1541– 4
References 1. Shaw RR, Paulson DL, Kee JL Jr. Treatment of the superior sulcus tumor by irradiation followed by resection. Ann Surg 1961;7:29– 40. 2. Hagen MP, Choi NC, Mathisen DJ, Wain JC, Wright CD, Grillo HC. Superior sulcus lung tumors: impact of local control on survival. J Thorac Cardiovasc Surg 1999;117: 1086–94. 3. Sartori F, Rea F, Calabro F, Mazzucco C, Bortolotti L, Tomio L. Carcinoma of the superior pulmonary sulcus: results of irradiation and radical resection. J Thorac Cardiovasc Surg 1992;104:679– 83. 4. Maggi G, Casadio C, Pischedda F, et al. Combined radiosurgical treatment in superior sulcus tumors. Ann Thorac Surg 1994;57:198 –202. 5. Rusch VW, Parekh KR, Leon L, et al. Factors determining outcome after surgical resection of T3 and T4 lung cancers of the superior sulcus. J Thorac Cardiovasc Surg 2000;119: 1147–53. 6. Cox JD. In: Shields TW, ed. General thoracic surgery, 3rd ed. Philadelphia: Lea & Febiger, 1989:1150–9. 7. Dartevelle PG, Chapelier AR, Macchiarini P, et al. Anterior transcervical-thoracic approach for radical resection of lung tumors invading the thoracic inlet. J Thoracic Cardiovasc Surg 1993;105:1025–34. 8. Choi NC, Carey RW, Daley W, et al. Potential impact on survival of improved tumor downstaging and resection rate by preoperative twice-daily radiation and concurrent chemotherapy in stage IIIA non-small cell lung cancer. J Clin Oncol 1997;15:712–22. 9. Albain KS, Rusch VW, Crowley JJ, et al. Concurrent cisplatin/etoposide plus chest radiotherapy followed by surgery for stages IIIA (N2) and IIIB non-small cell lung cancer: mature results of Southwest Oncology Group Phase II Study 8805. J Clin Oncol 1995;13:1880–92. 10. Wright CD, Wain JC, Grillo HC, Moncure AC, Macaluso SM, Mathisen DJ. Pulmonary lobectomy patient care pathway: a model to control cost and maintain quality. Ann Thorac Surg 1997;64:299 –302. 11. Martinez-Monge R, Herreros J, Aristu JJ, Aramendia JM, Azinovic I. Combined treatment in superior sulcus tumors. Am J Clin Oncol 1994;17:317–22. 12. Attar S, Krasna MJ, Sonett JR, et al. Superior sulcus (pancoast) tumor. experience with 105 patients. Ann Thorac Surg 1998;66:193– 8. 13. Rusch VW, Giroux DJ, Kraut MJ, et al. Induction chemoradiation and surgical resection for non-small cell carcinomas of the superior sulcus: initial results of Southwest Oncology Group Trial 9416 (Intergroup 0160). J Thorac Cardiovasc Surg 2001;121:472– 83. 14. Fuller DB, Chambers JS. Superior sulcus tumors: combined modality. Ann Thorac Surg 1994;57:1133–9.
Background. The usual approach of induction radiation therapy (RT) followed by resection of superior sulcus tumors results in many incomplete resections, a high local recurrence rate, and suboptimal survival. Induction chemoradiotherapy (CT/RT) has been shown to reduce local and distant recurrences and improve survival in stage III lung cancer. We investigated the role of induction CT/RT in superior sulcus patients. Methods. This was a single-institution, retrospective study. Results. From 1985 to 2000, 35 consecutive patients underwent induction treatment followed by resection of a superior sulcus tumor. All patients had mediastinoscopy first to exclude N2 disease, and all were N0 at final pathologic examination. Twenty patients had induction RT (mean, 39 Gy), and 15 had induction CT/RT (mean, 51 Gy) with concurrent cisplatin-based chemotherapy. There was no treatment mortality. Complete resection
was performed in 16 of 20 (80%) of the RT patients and in 14 of 15 (93%) of the CT/RT patients (p ⴝ 0.15). The pathologic response from the induction treatment was complete or near complete in 7 of 20 (35%) of the RT patients and in 13 of 15 (87%) of the CT/RT patients (p ⴝ 0.001). The median follow-up was 167 months in the RT patients and 51 months in the CT/RT patients. Two-year and 4-year survival was 49% and 49% (95% confidence interval, 26% to 71%) in the RT patients and 93% and 84% (95% confidence interval, 63% to 100%) in the CT/RT patients, respectively (p ⴝ 0.01). The local recurrence rate was 6 of 20 (30%) in the RT patients and 0 in the CT/RT patients (p ⴝ 0.02). Conclusions. Induction CT/RT for superior sulcus tumors appears to offer improved survival compared with induction RT alone. (Ann Thorac Surg 2002;73:1541– 4) © 2002 by The Society of Thoracic Surgeons
N
atomic area [2–5]. These high incomplete resection rates doom the patient to inadequate local control and lead to death, suggesting the need for better local control measures. Dartevelle and colleagues [7] recently introduced the anterior approach for select patients, which appears to enhance complete resection rates. Despite aggressive resection the most common site of first relapse is local [5]. Induction chemoradiotherapy (CT/RT) of stage III non– small-cell lung cancer has proven to be very effective at downstaging tumors and seemingly improving survival in a group of patients with a very high chance of both local and distant relapse [8, 9]. It would seem a priori that combined modality therapy would be most useful when tumor downstaging would facilitate a complete resection in the absence of N2 disease, a known marker for occult micrometastatic disease and subsequent death from distant metastases. We started treating patients with superior sulcus tumors with combined modality therapy based on our stage IIIA N2 results in 1991 [8] and have gradually changed our approach such that now we favor that approach. This period of practice pattern change in a single institution allows us to examine two approaches of induction treatment (RT and CT/RT) on outcome.
on–small-cell lung cancer involving structures at the thoracic inlet represents a small subset of potentially resectable lung cancer patients. Initially thought to be inoperable, the approach of low-dose (30 Gy) induction radiation therapy (RT) followed by resection was discovered by serendipity and promoted by Shaw and associates [1]. Numerous reports suggest that a 30% to 40% 5-year survival will be achieved with that approach [2–5]. A recent large series with multivariate analysis confirmed that the three most important prognostic factors were the ability to perform a complete resection and the T and N status of the tumor [5]. Although preoperative RT is favored by most surgeons, others have favored resection followed by RT, preferring to give full-dose RT that is not split course for radiobiologic reasons [6]. Most reported series document a relatively high incomplete resection rate (30% to 50%), highlighting the difficulty in performing a complete resection in this challenging an-
Accepted for publication Jan 23, 2002. Address reprint requests to Dr Wright, Blake 1570, Massachusetts General Hospital, Boston, MA 02114; e-mail: [email protected].
© 2002 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
0003-4975/02/$22.00 PII S0003-4975(02)03471-9
1542
WRIGHT ET AL INDUCTION THERAPY FOR SUPERIOR SULCUS TUMORS
Patients and Methods Patients with non–small-cell lung cancer invading the superior pulmonary sulcus were identified from the thoracic surgery database during a 15-year period (1985 to 2000). Only patients undergoing induction therapy followed by resection were reviewed. A retrospective chart review was performed, and the following data were recorded: patient age, sex, presence of pain pretreatment, induction treatment, adjuvant treatment, complete resection status, TNM status, length of stay, treatment-related complications, radiologic and pathologic response to induction treatment, disease status at last follow-up, and the sites of any relapse.
Patient Evaluation All patients had a chest computed tomographic scan, which included the liver and adrenal glands, and a diagnosis made by fine-needle aspiration. All had routine brain and bone scans to exclude occult metastatic disease. Since 1990, the majority of patients (70%) have had a magnetic resonance scan of the thoracic inlet area to delineate the anatomy. All patients had mediastinoscopy before starting induction treatment, and all were found to be negative for N2 nodal metastases.
Radiation Therapy The radiation target included the primary tumor, ipsilateral hilum, mid and upper mediastinum, and ipsilateral supraclavicular regions. Patients having induction RT had a preoperative mean dose of 39 Gy (range, 27 to 64 Gy) delivered in 2-Gy fractions, and the majority of patients had a postoperative boost dose (mean, 10 Gy; range, 0 to 27 Gy). Patients who had induction CT/RT had a preoperative mean dose of 51 Gy (range, 40 to 66 Gy). Four patients had hyperfractionated RT with 180 cGy given twice daily [8]. One CT/RT patient with an incomplete resection had postoperative RT.
Chemotherapy Four patients had two preoperative and one postoperative cycles of concurrent chemotherapy consisting of cisplatin 100 mg/m2 on day 1, 5-fluoruracil 30 mg/kg per day for 3 days starting day 1, and vinblastine 4 mg/m2 on day 1 [8]. Eleven patients had two preoperative cycles of concurrent chemotherapy consisting of cisplatin 50 mg/m2 on days 1, 8, 29, and 36 and etoposide 50 mg/m2 on days 1 to 5 and 29 to 33 [9]. Five patients had two additional cycles of the same chemotherapy postoperatively based on residual tumor in the resection specimen.
Statistical Analysis Survival was estimated by the product-limit method of Kaplan and Meier, and curves were compared by the log-rank test. Cox regression analysis was used to assess prognostic factors for their significance in predicting survival. Categorical variables were compared by the 2 test. Continuous variables were compared with the Student’s t test. All significance values were two-tailed.
Ann Thorac Surg 2002;73:1541– 4
Table 1. Patient Characteristics Induction Factor Year (average) Age (mean) T3 T4 Amount RT (Gy) Anterior approach Lobectomy Complete resection LOS
RT
CT/RT
p Value
1989 (1985–99) 54 (33– 86) 17 3 39 (30 – 63) 1/20 17/20 16/20 11 (6 –26)
1996 (1991–99) 58 (42– 68) 8 7 51 (40 – 66) 5/15 13/15 14/15 7.5 (4 –26)
⬍ 0.0001 0.31
CT/RT ⫽ chemoradiotherapy; tion therapy.
LOS ⫽ length of stay;
0.04 0.0002 0.03 0.89 0.15 0.07 RT ⫽ radia-
Results Thirty-five patients underwent treatment of a superior sulcus tumor with induction therapy followed by resection from 1985 to 2000. Eighteen patients were men and 17 were women. All patients had mediastinoscopy before treatment, and all were found to be free of mediastinal nodal metastases. Compared with CT/RT patients, those who had induction RT were treated on average 7 years earlier, were of the same age, had less RT, were less often resected by the anterior approach, had the same rate of lobectomy, trended toward more complete resections and a longer length of stay, and had less T4 tumors (Table 1). Side effects of induction RT were grade 2 esophagitis in 2 patients and fatigue and anorexia in 1 patient each. Side effects of induction CT/RT were grade 2 esophagitis in 4 patients and subclavian vein thrombosis in 1 patient. No patient suffered complications requiring hospitalization during induction therapy. There was no operative mortality and only five major postoperative complications (atrial fibrillation, 2; pneumonia, 1; prolonged air leak, 1; and cerebrospinal fluid leak, 1). Thirty patients underwent lobectomy, and 5 had a wedge resection. Twenty-nine patients had the classic extended posterolateral thoracotomy approach, and 6 had a combined anterior and posterior approach. The T4 tumors included 5 patients who underwent partial vertebrectomy and 5 patients who had subclavian artery resection. In the induction RT patients, 33% had a major (⬎ 50% reduction) radiologic response, 33% had a minor (⬎ 25% reduction) response, and 33% had no (⬍ 25% reduction) response. In the induction CT/RT patients, 70% had a major radiologic response and 30% had a minor response. All patients were N0 on the final pathology report. There was a marked difference in pathologic response based on the induction therapy favoring CT/RT (p ⫽ 0.001; Table 2). There was a marked survival advantage seen in the CT/RT patients over those with RT alone (p ⫽ 0.01; Fig 1). The 2-year and 4-year survival estimates in the CT/RT group were 93% (95% confidence interval, 81% to 100%) and 84% (95% confidence interval, 63% to 100%) compared with 49% (95% confidence interval, 26% to 71%)
Ann Thorac Surg 2002;73:1541– 4
1543
WRIGHT ET AL INDUCTION THERAPY FOR SUPERIOR SULCUS TUMORS
Table 2. Influence of Induction Therapy on Pathologic Response
Table 3. Influence of Induction Therapy on Site of First Recurrence
Induction
Pathologic response
RT
CT/RT
p Value
Complete Microscopic Tumor
2 5 13
10 3 2
0.001
CT/RT ⫽ chemoradiotherapy;
RT ⫽ radiation therapy.
and 49% (95% confidence interval, 26% to 71%) in the RT patients. The hazard ratio for RT was 5.6 (95% confidence interval, 1.2 to 25.2; p ⫽ 0.02). The median follow-up was 167 months in the RT patients and 51 months in the CT/RT patients. Six patients had a locoregional recurrence, and 10 had a distant recurrence noted during follow-up. Sites of distant recurrence included brain (n ⫽ 5), adrenal (n ⫽ 2), lung (n ⫽ 2), and bone (n ⫽ 1). There was no difference in distant recurrences between the two induction groups (Table 3). There was a reduction in local recurrence in the CT/RT induction patients (p ⫽ 0.02).
Comment The two different induction groups were fairly evenly matched for the three important prognostic factors for superior sulcus tumor patients (T and N status and complete resection) [5]. All patients were N0. The rate of complete resections was higher in the CT/RT group but not significantly (p ⫽ 0.15) so, probably owing to the small numbers of patients in the study. The number of complete resections in the CT/RT group is actually fairly high because more T4 tumors were resected. There were more T4 tumors in the CT/RT group (p ⫽ 0.04), probably
Fig 1. Kaplan–Meier survival curve for superior sulcus tumor patients treated by induction radiation therapy (RT) and induction chemoradiotherapy (CT/RT). (Cum ⫽ cumulative.)
Recurrence
RT
CT/RT
p Value
Local Distant
6/20 7/20
0/15 3/15
0.02 0.33
CT/RT ⫽ chemoradiotherapy;
RT ⫽ radiation therapy.
reflecting both greater confidence in the ability to perform a complete resection by an anterior approach and to sterilize the tumor with the induction therapy. Despite having more T4 tumors, the group that underwent CT/RT as induction therapy had a markedly improved survival. The morbidity was not appreciably greater in patients receiving CT/RT, and there was no mortality with either treatment. There was a trend for a reduction in length of stay in the group receiving CT/RT probably related to our use of pathways and reflects the average 7-year difference between the two groups [10]. Certainly the length of stay does not appear to be prolonged with the addition of chemotherapy to induction RT. Induction CT/RT was remarkably effective at sterilizing the tumor, with a 67% complete response rate and an 87% rate of near or complete response. It is likely that the improved survival seen was caused in large part by this very high sterilization rate. Two previously published reports have also suggested that CT/RT is the optimal induction strategy for superior sulcus tumor patients. Martinez-Monge and coworkers [11] in 1994 reported 18 patients treated with two different mitomycin, vindesine, and cisplatin regimens. There was a very high treatment mortality rate of 17%, a high pathologic complete response rate (71%), and a favorable 56% 4-year survival rate. Attar and associates [12] in 1998 briefly reported on 11 patients treated with carboplatin and paclitaxel. Treatment toxicity was not stated, but the 5-year survival was 72%. Rusch and colleagues [13] recently reported a large prospective multiinstitutional trial (Intergroup 0160) of induction CT/RT (based on Southwest Oncology Group 8805 [9]) for superior sulcus tumors. There were 111 eligible patients, and 102 had the planned induction treatment. There were 5 (5%) treatment-related deaths. Eighty-three patients underwent thoracotomy, and 76 (92%) had a complete resection. The pathologic response to induction CT/RT was complete or with only minimal microscopic residual disease in 65% of the patients. The 2-year and 4-year survival was 55%. The complete resection rate was high (90%), and the locoregional recurrence rate was low (23%). The authors concluded that the complete resection rate and survival appeared improved compared with historic experience. A randomized trial of therapy in this small subset of patients with non–small-cell lung cancer will probably never be accomplished because of the rarity of this tumor. Accordingly, careful scrutiny of the Intergroup trial and the other case series with comparison to historic experience with induction RT is necessary to determine whether an apparent improvement in care is real. Our series is particularly helpful in this regard inasmuch as
1544
WRIGHT ET AL INDUCTION THERAPY FOR SUPERIOR SULCUS TUMORS
there is a period of time during which the two induction programs overlapped, the patients were carefully staged both radiographically and by mediastinoscopy, and the surgical strategy was uniform. A fair comparison between the two induction programs might conclude that induction CT/RT (1) can be given with acceptably low morbidity, (2) trends toward a higher complete resection rate, (3) results in a very high pathologic response rate, (4) results in a reduced locoregional recurrence rate (the ultimate test of the completeness of resection), and (5) offers improved survival. There are several limitations of our study. The first is the retrospective nature of this consecutive case series in which treatment was not by protocol. The second is that there may have been case selection bias favoring better patients undergoing induction CT/RT. However, the three major prognostic factors were fairly evenly distributed between the two groups, and there were more T4 tumors in the CT/RT group. Third, there was probably also an element of time selection bias favoring the CT/RT patients who likely benefited from better staging (new spiral computed tomographic scan), improved surgical procedure owing to use of the anterior approach and the learning effect, and a general improvement of care that comes with the time. Nonetheless, many patients were treated simultaneously with both induction programs, and all patients were vigorously staged. Fourth, the median follow-up was longer in the RT patients, which allows more time for failure. Lastly, the sample size is small, and several comparisons (complete resection rate and length of stay) would likely have been significant if the study were larger (type II error). Induction CT/RT followed by resection for superior sulcus tumors appears superior to the historic approach of induction RT alone. Future trials should investigate the feasibility of dose escalation of radiation as medium-dose RT (50 Gy) appears well tolerated and local control is still not ideal [14]. Other chemotherapeutic agents that act as radiation sensitizers should be evaluated in these patients now that a baseline pathologic response rate has been established for the standard regimen of cisplatin and etoposide. Adjuvant postoperative chemotherapy and central nervous system prophylaxis need to be investigated as distant metastases now represent the most common site of failure.
Ann Thorac Surg 2002;73:1541– 4
References 1. Shaw RR, Paulson DL, Kee JL Jr. Treatment of the superior sulcus tumor by irradiation followed by resection. Ann Surg 1961;7:29– 40. 2. Hagen MP, Choi NC, Mathisen DJ, Wain JC, Wright CD, Grillo HC. Superior sulcus lung tumors: impact of local control on survival. J Thorac Cardiovasc Surg 1999;117: 1086–94. 3. Sartori F, Rea F, Calabro F, Mazzucco C, Bortolotti L, Tomio L. Carcinoma of the superior pulmonary sulcus: results of irradiation and radical resection. J Thorac Cardiovasc Surg 1992;104:679– 83. 4. Maggi G, Casadio C, Pischedda F, et al. Combined radiosurgical treatment in superior sulcus tumors. Ann Thorac Surg 1994;57:198 –202. 5. Rusch VW, Parekh KR, Leon L, et al. Factors determining outcome after surgical resection of T3 and T4 lung cancers of the superior sulcus. J Thorac Cardiovasc Surg 2000;119: 1147–53. 6. Cox JD. In: Shields TW, ed. General thoracic surgery, 3rd ed. Philadelphia: Lea & Febiger, 1989:1150–9. 7. Dartevelle PG, Chapelier AR, Macchiarini P, et al. Anterior transcervical-thoracic approach for radical resection of lung tumors invading the thoracic inlet. J Thoracic Cardiovasc Surg 1993;105:1025–34. 8. Choi NC, Carey RW, Daley W, et al. Potential impact on survival of improved tumor downstaging and resection rate by preoperative twice-daily radiation and concurrent chemotherapy in stage IIIA non-small cell lung cancer. J Clin Oncol 1997;15:712–22. 9. Albain KS, Rusch VW, Crowley JJ, et al. Concurrent cisplatin/etoposide plus chest radiotherapy followed by surgery for stages IIIA (N2) and IIIB non-small cell lung cancer: mature results of Southwest Oncology Group Phase II Study 8805. J Clin Oncol 1995;13:1880–92. 10. Wright CD, Wain JC, Grillo HC, Moncure AC, Macaluso SM, Mathisen DJ. Pulmonary lobectomy patient care pathway: a model to control cost and maintain quality. Ann Thorac Surg 1997;64:299 –302. 11. Martinez-Monge R, Herreros J, Aristu JJ, Aramendia JM, Azinovic I. Combined treatment in superior sulcus tumors. Am J Clin Oncol 1994;17:317–22. 12. Attar S, Krasna MJ, Sonett JR, et al. Superior sulcus (pancoast) tumor. experience with 105 patients. Ann Thorac Surg 1998;66:193– 8. 13. Rusch VW, Giroux DJ, Kraut MJ, et al. Induction chemoradiation and surgical resection for non-small cell carcinomas of the superior sulcus: initial results of Southwest Oncology Group Trial 9416 (Intergroup 0160). J Thorac Cardiovasc Surg 2001;121:472– 83. 14. Fuller DB, Chambers JS. Superior sulcus tumors: combined modality. Ann Thorac Surg 1994;57:1133–9.