- Email: [email protected]
Postoperative radiotherapy after pneumonectomy: Impact of modern treatment facilities
Ini I Radmrwn Oncr~lo~y Bwl Phy, Vol. 27. pp. 525-529 Prmted I” the U S.A. All nghts reserved.
Copyright
0360.3016/93 $6.00 + .OO 0 I993 Pergamon Press Ltd.
??Clinical Original Contribution
POSTOPERATIVE RADIOTHERAPY AFTER PNEUMONECTOMY: IMPACT OF MODERN TREATMENT FACILITIES PATRICIA
PHLIPS,
M.D.,’
PIERRE
AND PAUL ‘Department
ROCMANS, VAN
M.D.,2
HOUTTE,
PATRICK
M.D.,
VANDERHOEFT,
M.D.2
PH.D.’
of Radiation Oncology, Institut Jules Bordet; and *Department of Thoracic Surgery. Hopital Erasme, Brussels, Belgium
Purpose: The present study was undertaken to see how modern treatment facilities, computed tomography (CT)based treatment planning and linear accelerator, have modified the results of postoperative irradiation after a pneumonectomy for lung cancer. Methods and Materials: Between 1970-1985, 103 patients were treated in our department after a pneumonectomy: 50 patients with a TlT2NO tumor and 53 patients with a T3, Nl or N2 tumor. Three groups were considered: 27 patients had only surgical resection, 51 patients were irradiated postoperatively with a Co60 source, and 25 patients were treated using those modern facilities. Results: The S-year survival varies from 4% to 31% according to the tumor extent but also to the radiation technique. Patients treated with a Co60 source had a dismal 5-year survival rate (8%) whereas patients treated with the modern facilities had a S-year survival rate of 30% similar to the 31% of the control surgical group including less advanced tumors. Conclusion: Linear accelerator and computed tomography-based treatment planning improved the accuracy of postoperative thoracic irradiation and allow to deliver high doses to the mediastinum even after a pneumonectomy. Lung cancer, Postoperative
radiotherapy, CT scan.
therapy: CT based treatment planning, individually tailored blocks, lung factor correction for tissue inhomogeneity, linear accelerator are helpful tools to better select and perform the radiation procedure. The present work was designed to see if those developments had modified the outcome of patients treated with postoperative radiation after a pneumonectomy.
INTRODUCIION Thoracic postoperative radiation for lung cancers still remains a controversial issue but it represents a challenge for the radiotherapist: a dose able to control microscopic residual disease must be delivered without exceeding the tolerance of many normal and vital organs such as the lung, the heart, or the spinal cord (13). A small benefit in terms of survival or local control may easily be outweighed by an increase in radiation-induced late damage. The poor quality of the radiation program used in several studies run through the 60’s and 70’s was often advocated to explain the negative results (2, 8, 10). Indeed, in our own trial, postoperative radiation did not improve the survival of patients with a complete resection of a lung cancer without lymph node involvement. Furthermore, the patients irradiated after pneumonectomy had quite a dismal survival ( 16). This was believed to be related to the radiation technique and to an increase in radiationinduced late damage. During the last two decades, several technical developments were introduced in the daily practice of radio-
METHODS
AND
MATERIALS
From 1970 to 1985,490 consecutive patients with lung cancer were seen in the department of radiation oncology before or after surgery. All the patients included in the present analysis had to fulfill the following two conditions: (a) a complete resection of a lung tumor without known distant metastases and (b) a pneumonectomy. Excluded were patients with a lobectomy, a bilobectomy, or an incomplete resection, those treated with chemotherapy and radiation or by preoperative radiation. The present series included 103 pneumonectomized patients. In fact, during that period our treatment policy was as follows:
Accepted for publication
Reprint requests to: Paul Van Houtte, M.D., Department of Radiation Oncology, Institut Jules Bordet, rue HCger Bordet 1, 1000 Brussels, Belgium. 525
15 April 1993.
I. J. Radiation
526
Oncology
0 Biology 0 Physics
Volume 27, Number 3, 1993
Table 1. Radiation technique Treatment Machine
Table 3. Histology
sq.
planning
Number fields
Dose talc.
CT scan
Lung factor correction
Dose GY
3 3 3
Yes Yes Yes
No (Yes)* Yes
No Yes Yes
60 60 56
Co60 BetaCo Linac
* CT scan performed during treatment.
After a complete resection without lymph node involvement (pN0) ( 1970- 1975), patients were included in a randomized trial, which compared a control surgical group with postoperative radiotherapy ( 16). After a complete resection without lymph node involvement (pN0) ( 1976- 1978), patients received routine postoperative radiotherapy. This approach was disregarded after an update analysis of the prior randomized trial. After a complete resection of a tumor extending beyond the lung (pT3) or with positive lymph nodes (pN 1 or pN2), routine postoperative radiotherapy was used. All patients were staged after surgery according to the 1979 AJC classification (1). This series included 50 Tl T2NO tumors, 16 T3N0, 16 N 1, and 2 1 N2. All major histological variants were treated: 76 squamous cell carcinomas, 12 adenocarcinomas, 9 large cell and 6 small cell carcinomas.
Radiation technique During that period, the radiation technique varied according to the technical developments. Three different approaches were used:
Co60 technique
(Co60): a dose of 60 Gy was delivered to the mediastinum with a three fields set-up using a Co60 source. Dose calculations were made using a computer but without lung factor correction and CT scan. This technique was used for patients treated within the randomized trial (16). Betatron-Co60 technique (betaco): a similar dose of 60 Gy was delivered using an anterior field from a 35 MeV betatron and two posterior oblique fields with wedge from a Co60 unit. Dose calculations were made using a SIDOS-U treatment planning taking into ac-
Table 2. Treatment
Contr R* Co60 R* BetaCo Co60 Linac
TlT2(T3) TlT2(T3) TIT2(T3) T3,Nl,N2 T3,Nl,N2
* Randomized
trial.
Contr. Co60 Linac
R
20 38 18 76
and treatment
Adeno 2 6 4 12
Large
SCCI
3 3 3 9
2 4
27 51 25 103
6
count lung factor correction. A CT scan for dosimetry was made during the course of the treatment whenever it was feasible. The technique has already been published ( 14). 3. The linac technique (linac): a dose of 56 Gy was delivered to the mediastinum using a three or four fields set-up from a 18 MeV linear accelerator. A CT scan for dosimetry was made before treatment. Treatment planning was based on this CT and lung factor correction was applied. The main characteristics of the different radiation techniques are summarized in Table I. During the whole period, the target volume included the mediastinum from the sternal notch or 2 cm above the mediastinoscopy scar down to 5 cm below the carina. So, according to the treatment and the tumor extent, five different groups may be considered: the two arms of the randomized trials including only patients with a tumor without lymph node involvement (control, Co60R), a group of patients with the same tumor extent but treated with the betatron-Co60 technique (betaco), and two series of patients presenting more advanced disease (T3, N 1, or N2) treated with the two different radiation techniques (Co60 and linac) (Table 2). For practical reasons, all patients treated with postoperative radiation before 1980 were pooled together regardless of the radiation technique or the tumor extent (Co60 group) and were compared to those treated after 1980 using all modern facilities (linac group).
RESULTS The histological distribution is quite similar within the different groups with a higher proportion of squamous cell carcinoma (Table 3). The pTNM tumor distribution among the different radiation techniques reflects our treatment policy over the years. All the patients within the control group (surgery alone) had a tumor without lymph node involvement (pT lT2NOMO). In contrast, the linac group only included patients with a tumor extending
categories
Tumor extent
No. patients
NO NO NO
27 16 14 Co60 2125
1970-1975 1970-1975 1976-1978 1970-1975 1980-1985
Table 4. Staging and treatment
Control Co60 Linac
T 1,T2NO
T3NO
NI
N2
23 27
4 4 8 16
6 10 16
14 7 21
50
27 51 25 103
Postoperative radiotherapy after pneumonectomy 0 P.
527
PHLIPS et nl.
”
U
r
__
‘b
40
a
*- t-+- -t---I t 8%
I
1
0
10
20
30
40
50
60
70
MONTHS -~-f-m Co60 51 pts
--*--Control
Fig. 1. Postop. RT and pneumonectomy:
No 27
pts
~cf-
Linac
influence of radiation technique. All known acute and late complications after radiation were reported but some patients died at home and precise information is often lacking. Ten complications were observed among the 76 irradiated patients compared to 5 out of 27 without radiation: respiratory distress (six patients), cardiac problems (three patients) and fistula (six patients). The linac technique led to a decrease in reported late effects: one patient died of respiratory insufficiency while six patients from the Co60 group died of respiratory and cardiac insufficiency or of a fistula. Furthermore, there was no single incidence of myelopathy (Table 5).
beyond the lung (pT3) or with positive lymph nodes (pN 1 or pN2) (Table 4). The Co60 group included 27 patients with a Tl or T2 No tumor and 24 with more advanced disease (T3, Nl or N2). Long-term survival was significantly decreased after postoperative radiation using a Co60 source with a 5-year survival rate of 8% compared to 3 1% for the control group (Fig. 1). No difference was observed between the two different radiation techniques: either three fields of a Co60 source or one anterior field of a 35 MeV beta&on and two posterior oblique fields from a Co60 source. In contrast, for the group of patients treated with the linac technique, the 5-year survival rate reached 30%. Although this comparison may be hazardous due to the progress made over the years, the survival of the T3, Nl, or N2 patients treated with all the modern radiation facilities (linac group) was similar to the surgical control group including only completely resected tumors without lymph node involvement. In the present series, the main cause of failure remains distant metastases. Locoregional recurrences are rare regardless of the radiation technique (Fig. 2).
DISCUSSION This study presents all the well known limits of any retrospective approach: patients selection, progress in imaging procedure and treatment technique, better knowledge of the disease itself etc. Nevertheless, during the study period, patients selection remained very consistent in the choice of a possible course of postoperative radiation. In a prior review, about 80% of the patients operated received the planned course of adjuvant treatment in our cancer
Percent 60% 50% 40% 30% 20% 10% 0% ! Local
Both 0
Linac
m
Fig. 2. Postop radiotherapy
N+ 25 pts
Control
Distant R
~
Co60
for lung CA: pattern
of failure.
I. J.
528
Radiation
Table 5. Postop radioth.
Number
of pts.
Respiratory
insuf.
Oncology
0 Biology
0
Physics
complications
cont.
Co60
Linac
27 1 (1)
51 4
25 1 (1)
(3) 3
Cardiac
toxicity
(2) 4
Fistula
2 (2)
(1)
5 Total Complic. ( ) Lethal compl.
center
(3) IS%> 11%
( 15). The low number
with the linac technique
9
of patients
only reflects
1 (1) 4% 4%
(6) 180/o 12%
treated
our changes
after
1980
in treat-
ment philosophy after the publication of our randomized trial: postoperative radiotherapy was only performed in presence of a T3, Nl, or N2 tumor. Furthermore, all patients were treated by the same surgical and radiotherapeutic teams. So, the improvement in survival that we have observed among patients treated with a course of postoperative irradiation may be related to the introduction of linear accelerator, CT-scan-based treatment planning and a more accurate staging procedure. Indeed, in our randomized trial for patients without lymph node involvement, the type of surgical resection did dramatically influence the survival in the irradiated group; after a pneumonectomy, a dismal survival was observed but in presence of a lobectomy, the results were similar for the control and irradiated groups ( 16). Such a difference between lobectomy and pneumonectomy was not observed among patients presenting with locally advanced disease (T3, Nl or N2 tumor) treated with the more modern technique. Several factors may have contributed to this difference: the small dose reduction, the use of lung correction factor for tissue inhomogeneity, the use of CT-scan-based treatment planning to select the radiation procedure or the introduction of a linear accelerator. In this study, we did not observe any survival difference between the two groups of patients treated at least partially with a Co60 source (Co60 and BetaCo group). The latter takes into account lung factor correction for tissue inhomogeneity. This factor should lead to a greater dose modification after a lobectomy than after a pneumonectomy: in our own experience, this difference ranged between 3 to 15% (14).
Volume
27. Number
3, 1993
This difference may explain some of the late damage induced to structure lying within the mediastinum, great vessels, and esophagus; this is certainly not the case for lung tissue as the doses delivered already exceed its tolerance. The major contribution of CT-scan-based treatment planning has been well-documented both for inoperable lung cancer and for postoperative radiation (6, 7, 11, 14). In our prior study, the target volume dose was overestimated by 5% when comparing treatment planning based on orthogonal X rays to the one done with a CT scan ( 14). Furthermore, possible mediastinal or lung shifts were not easily detected with orthogonal X rays, which was not the case with CT scan. A main limiting factor in the design of a course of postoperative radiation is certainly the tolerance of the remaining lung parenchyma. The dose required to control microscopic disease will necessarily exceed the lung tolerance. So, limiting the amount of normal lung tissue included in the treatment fields represents a major requirement to avoid life threalening late damage. Today, a CT scan for treatment planning is performed after surgery in treatment position before the beginning of the radiation. This will allow to design for each patient the best field set-up taking into account an abnormal anatomy such as a mediastinal shift and limiting as much as possible the volume of normal lung irradiated. This is one possible explanation for the difference observed between the different groups. The next improvement will certainly be due to the introduction of a true 3-dimensional (3-D) treatment planning instead of using a single or a few planes: then, the 3-D isodose displays and dosevolume-histograms may become routinely available (5). Nevertheless, a control group without postoperative radiation should be randomly selected to assess that postoperative radiation can be safely carried out with high doses even after a pneumonectomy. Our preliminary results with an adequate technique are in good agreement with some of the data published (3, 4. 12). Postoperative radiation, especially in the presence of positive lymph nodes, remains a controversial issue for lung cancers due partially to the lack of well-designed randomized trial including a large number of patients as well as an adequate staging procedure and well-designed radiation programs (9). Those are the reasons why the GETCB and the EORTC have launched such a study which requires the use of CT scan-based treatment planning and linear accelerator to deliver the prescribed 60 Gy to the mediastinum.
REFERENCES American Joint Committee on Cancer. Task force on lung. Staging of lung cancer. Chicago: American Joint Committee on Cancer; 1979. Bangma, P. J. Postoperative radiotherapy. In: Deeley, T. J., ed. Carcinoma ofthe bronchus: Modern radiotherapy. New York: Appleton-Century-Crofts; 1972: 163- 170. Choi, N.; Grille, H.; Gardiello, M.: Scannell, J. G.; Wilkins,
E. W. Basis for new strategies in postoperative radiotherapy of bronchogenic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 6:31-35;1980. 4. Emami, B.; Kim, T.; Roper, C.; Simpson, J.; Pilepich, M.; Hederman, M. Postoperative radiation therapy in the management of lung cancer. Radiology 164:251-253:1987. 5. Emami, B.; Purdy, J. A.; Manolis, J.; Barest, G.; Cheng, E.;
Postoperative radiotherapy after pneumonectomy 0 P.
6.
7.
8.
9.
10.
Coia, L.; Doppke, K.; Galvin, J.; LoSasso, T.; Matthews, J.; Munzenrider, J.; Shank, B. Three-dimensional treatment planning for lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 21:217-227;1991. Goittein, M.; Wittenberg, J.; Mendcondo, M.; Doucette, J.; Friedberg, C.; Ferrucci, J.; Gunderson, L.; Linggood, R.; Shipley, W. U.; Fineberg, H. V. The value of CT scanning in radiation treatment planning: A prospective study. Int. J. Radiat. Oncol. Biol. Phys. 5: 1787- 1798; 1979. Hobday, P.; Hodson, N. J.; Husband, J.; Parker, R. P.; Macdonald, J. S. Computed tomography applied to radiotherapy treatment planning: Techniques and results. Radiology 133:477-482; 1979. Israel, L.; Bonadonna, G.; Sylvester, R.; members of the EORTC Lung Cancer Group. Controlled study with adjuvant radiotherapy, chemotherapy, immunotherapy and chemoimmunotherapy in operable squamous cell carcinoma of the lung. In: Muggia, F. M., Rozencweig, M., eds. Lung cancer: Progress in therapeutic research. New York: Raven Press; 1979:443-452. Lung Cancer Study Group. Effects of postoperative mediastinal radiation on completely resected stage II and stage III epidermoid cancer of the lung. New Engl. J. Med. 3 15: 1377-1381;1986. Patterson, R.; Russell, M. H. Clinical trial in malignant disease IV lung cancer. Clin. Radiol. 13:141-144;1962.
PHLIPS et al.
529
11. Perez, C. A.; Stanley, K.; Grundy, G.; Hanson, W.; Rubin, P.; Kramer, S.; Brady, L. W.; Marks, J. E.; Perez-Tamayo, R.; Brown, J.; Concannon, J.; Rotman, M. Impact of irradiation technique and tumor extent in tumor control and survival of patients with unresectable non-oat cell carcinoma of the lung. Report by the Radiation Therapy Oncology Group. Cancer 50: 109 l- 1099; 1982. 12. Slater, J. D.; Ellerbroek, N. A.; Barkley, Jr., H.; Mountain, C.; Oswald, M. J.; Roth, J. A.; Peters, L. J. Radiation therapy following resection of non-small cell bronchogenic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 20:945-951;1991. 13. Van Houtte, P. Postoperative radiotherapy for lung cancer. Lung Cancer 7:57-64; 199 1. 14. Van Houtte, P.; Piron, A.; Lustman-MarCchal, J.; Osteaux, M.; Henry, J. Computed axial tomography (CAT) contribution for dosimetry and treatment evaluation in lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 6:995-1000;1980. 15. Van Houtte, P.; Rocmans, P.; Nguyen, T. H.; LustmanMarechal, J.; Vanderhoeft, P.; Henry, J. Irradiation postoperatoire dans le cancer pulmonaire. J. Eur. Radiother. I: 17-25;1981. 16. Van Houtte, P.; Rocmans, P.; Smets, P.; Goffin, J. C.; Lustman-Marechal, J.; Vanderhoeft, P.; Henry, J. Postoperative radiation therapy in lung cancer: a controlled trial after resection of curative design. Int. J. Radiat. Oncol. Biol. Phys. 6:983-986; 1980.
Copyright
0360.3016/93 $6.00 + .OO 0 I993 Pergamon Press Ltd.
??Clinical Original Contribution
POSTOPERATIVE RADIOTHERAPY AFTER PNEUMONECTOMY: IMPACT OF MODERN TREATMENT FACILITIES PATRICIA
PHLIPS,
M.D.,’
PIERRE
AND PAUL ‘Department
ROCMANS, VAN
M.D.,2
HOUTTE,
PATRICK
M.D.,
VANDERHOEFT,
M.D.2
PH.D.’
of Radiation Oncology, Institut Jules Bordet; and *Department of Thoracic Surgery. Hopital Erasme, Brussels, Belgium
Purpose: The present study was undertaken to see how modern treatment facilities, computed tomography (CT)based treatment planning and linear accelerator, have modified the results of postoperative irradiation after a pneumonectomy for lung cancer. Methods and Materials: Between 1970-1985, 103 patients were treated in our department after a pneumonectomy: 50 patients with a TlT2NO tumor and 53 patients with a T3, Nl or N2 tumor. Three groups were considered: 27 patients had only surgical resection, 51 patients were irradiated postoperatively with a Co60 source, and 25 patients were treated using those modern facilities. Results: The S-year survival varies from 4% to 31% according to the tumor extent but also to the radiation technique. Patients treated with a Co60 source had a dismal 5-year survival rate (8%) whereas patients treated with the modern facilities had a S-year survival rate of 30% similar to the 31% of the control surgical group including less advanced tumors. Conclusion: Linear accelerator and computed tomography-based treatment planning improved the accuracy of postoperative thoracic irradiation and allow to deliver high doses to the mediastinum even after a pneumonectomy. Lung cancer, Postoperative
radiotherapy, CT scan.
therapy: CT based treatment planning, individually tailored blocks, lung factor correction for tissue inhomogeneity, linear accelerator are helpful tools to better select and perform the radiation procedure. The present work was designed to see if those developments had modified the outcome of patients treated with postoperative radiation after a pneumonectomy.
INTRODUCIION Thoracic postoperative radiation for lung cancers still remains a controversial issue but it represents a challenge for the radiotherapist: a dose able to control microscopic residual disease must be delivered without exceeding the tolerance of many normal and vital organs such as the lung, the heart, or the spinal cord (13). A small benefit in terms of survival or local control may easily be outweighed by an increase in radiation-induced late damage. The poor quality of the radiation program used in several studies run through the 60’s and 70’s was often advocated to explain the negative results (2, 8, 10). Indeed, in our own trial, postoperative radiation did not improve the survival of patients with a complete resection of a lung cancer without lymph node involvement. Furthermore, the patients irradiated after pneumonectomy had quite a dismal survival ( 16). This was believed to be related to the radiation technique and to an increase in radiationinduced late damage. During the last two decades, several technical developments were introduced in the daily practice of radio-
METHODS
AND
MATERIALS
From 1970 to 1985,490 consecutive patients with lung cancer were seen in the department of radiation oncology before or after surgery. All the patients included in the present analysis had to fulfill the following two conditions: (a) a complete resection of a lung tumor without known distant metastases and (b) a pneumonectomy. Excluded were patients with a lobectomy, a bilobectomy, or an incomplete resection, those treated with chemotherapy and radiation or by preoperative radiation. The present series included 103 pneumonectomized patients. In fact, during that period our treatment policy was as follows:
Accepted for publication
Reprint requests to: Paul Van Houtte, M.D., Department of Radiation Oncology, Institut Jules Bordet, rue HCger Bordet 1, 1000 Brussels, Belgium. 525
15 April 1993.
I. J. Radiation
526
Oncology
0 Biology 0 Physics
Volume 27, Number 3, 1993
Table 1. Radiation technique Treatment Machine
Table 3. Histology
sq.
planning
Number fields
Dose talc.
CT scan
Lung factor correction
Dose GY
3 3 3
Yes Yes Yes
No (Yes)* Yes
No Yes Yes
60 60 56
Co60 BetaCo Linac
* CT scan performed during treatment.
After a complete resection without lymph node involvement (pN0) ( 1970- 1975), patients were included in a randomized trial, which compared a control surgical group with postoperative radiotherapy ( 16). After a complete resection without lymph node involvement (pN0) ( 1976- 1978), patients received routine postoperative radiotherapy. This approach was disregarded after an update analysis of the prior randomized trial. After a complete resection of a tumor extending beyond the lung (pT3) or with positive lymph nodes (pN 1 or pN2), routine postoperative radiotherapy was used. All patients were staged after surgery according to the 1979 AJC classification (1). This series included 50 Tl T2NO tumors, 16 T3N0, 16 N 1, and 2 1 N2. All major histological variants were treated: 76 squamous cell carcinomas, 12 adenocarcinomas, 9 large cell and 6 small cell carcinomas.
Radiation technique During that period, the radiation technique varied according to the technical developments. Three different approaches were used:
Co60 technique
(Co60): a dose of 60 Gy was delivered to the mediastinum with a three fields set-up using a Co60 source. Dose calculations were made using a computer but without lung factor correction and CT scan. This technique was used for patients treated within the randomized trial (16). Betatron-Co60 technique (betaco): a similar dose of 60 Gy was delivered using an anterior field from a 35 MeV betatron and two posterior oblique fields with wedge from a Co60 unit. Dose calculations were made using a SIDOS-U treatment planning taking into ac-
Table 2. Treatment
Contr R* Co60 R* BetaCo Co60 Linac
TlT2(T3) TlT2(T3) TIT2(T3) T3,Nl,N2 T3,Nl,N2
* Randomized
trial.
Contr. Co60 Linac
R
20 38 18 76
and treatment
Adeno 2 6 4 12
Large
SCCI
3 3 3 9
2 4
27 51 25 103
6
count lung factor correction. A CT scan for dosimetry was made during the course of the treatment whenever it was feasible. The technique has already been published ( 14). 3. The linac technique (linac): a dose of 56 Gy was delivered to the mediastinum using a three or four fields set-up from a 18 MeV linear accelerator. A CT scan for dosimetry was made before treatment. Treatment planning was based on this CT and lung factor correction was applied. The main characteristics of the different radiation techniques are summarized in Table I. During the whole period, the target volume included the mediastinum from the sternal notch or 2 cm above the mediastinoscopy scar down to 5 cm below the carina. So, according to the treatment and the tumor extent, five different groups may be considered: the two arms of the randomized trials including only patients with a tumor without lymph node involvement (control, Co60R), a group of patients with the same tumor extent but treated with the betatron-Co60 technique (betaco), and two series of patients presenting more advanced disease (T3, N 1, or N2) treated with the two different radiation techniques (Co60 and linac) (Table 2). For practical reasons, all patients treated with postoperative radiation before 1980 were pooled together regardless of the radiation technique or the tumor extent (Co60 group) and were compared to those treated after 1980 using all modern facilities (linac group).
RESULTS The histological distribution is quite similar within the different groups with a higher proportion of squamous cell carcinoma (Table 3). The pTNM tumor distribution among the different radiation techniques reflects our treatment policy over the years. All the patients within the control group (surgery alone) had a tumor without lymph node involvement (pT lT2NOMO). In contrast, the linac group only included patients with a tumor extending
categories
Tumor extent
No. patients
NO NO NO
27 16 14 Co60 2125
1970-1975 1970-1975 1976-1978 1970-1975 1980-1985
Table 4. Staging and treatment
Control Co60 Linac
T 1,T2NO
T3NO
NI
N2
23 27
4 4 8 16
6 10 16
14 7 21
50
27 51 25 103
Postoperative radiotherapy after pneumonectomy 0 P.
527
PHLIPS et nl.
”
U
r
__
‘b
40
a
*- t-+- -t---I t 8%
I
1
0
10
20
30
40
50
60
70
MONTHS -~-f-m Co60 51 pts
--*--Control
Fig. 1. Postop. RT and pneumonectomy:
No 27
pts
~cf-
Linac
influence of radiation technique. All known acute and late complications after radiation were reported but some patients died at home and precise information is often lacking. Ten complications were observed among the 76 irradiated patients compared to 5 out of 27 without radiation: respiratory distress (six patients), cardiac problems (three patients) and fistula (six patients). The linac technique led to a decrease in reported late effects: one patient died of respiratory insufficiency while six patients from the Co60 group died of respiratory and cardiac insufficiency or of a fistula. Furthermore, there was no single incidence of myelopathy (Table 5).
beyond the lung (pT3) or with positive lymph nodes (pN 1 or pN2) (Table 4). The Co60 group included 27 patients with a Tl or T2 No tumor and 24 with more advanced disease (T3, Nl or N2). Long-term survival was significantly decreased after postoperative radiation using a Co60 source with a 5-year survival rate of 8% compared to 3 1% for the control group (Fig. 1). No difference was observed between the two different radiation techniques: either three fields of a Co60 source or one anterior field of a 35 MeV beta&on and two posterior oblique fields from a Co60 source. In contrast, for the group of patients treated with the linac technique, the 5-year survival rate reached 30%. Although this comparison may be hazardous due to the progress made over the years, the survival of the T3, Nl, or N2 patients treated with all the modern radiation facilities (linac group) was similar to the surgical control group including only completely resected tumors without lymph node involvement. In the present series, the main cause of failure remains distant metastases. Locoregional recurrences are rare regardless of the radiation technique (Fig. 2).
DISCUSSION This study presents all the well known limits of any retrospective approach: patients selection, progress in imaging procedure and treatment technique, better knowledge of the disease itself etc. Nevertheless, during the study period, patients selection remained very consistent in the choice of a possible course of postoperative radiation. In a prior review, about 80% of the patients operated received the planned course of adjuvant treatment in our cancer
Percent 60% 50% 40% 30% 20% 10% 0% ! Local
Both 0
Linac
m
Fig. 2. Postop radiotherapy
N+ 25 pts
Control
Distant R
~
Co60
for lung CA: pattern
of failure.
I. J.
528
Radiation
Table 5. Postop radioth.
Number
of pts.
Respiratory
insuf.
Oncology
0 Biology
0
Physics
complications
cont.
Co60
Linac
27 1 (1)
51 4
25 1 (1)
(3) 3
Cardiac
toxicity
(2) 4
Fistula
2 (2)
(1)
5 Total Complic. ( ) Lethal compl.
center
(3) IS%> 11%
( 15). The low number
with the linac technique
9
of patients
only reflects
1 (1) 4% 4%
(6) 180/o 12%
treated
our changes
after
1980
in treat-
ment philosophy after the publication of our randomized trial: postoperative radiotherapy was only performed in presence of a T3, Nl, or N2 tumor. Furthermore, all patients were treated by the same surgical and radiotherapeutic teams. So, the improvement in survival that we have observed among patients treated with a course of postoperative irradiation may be related to the introduction of linear accelerator, CT-scan-based treatment planning and a more accurate staging procedure. Indeed, in our randomized trial for patients without lymph node involvement, the type of surgical resection did dramatically influence the survival in the irradiated group; after a pneumonectomy, a dismal survival was observed but in presence of a lobectomy, the results were similar for the control and irradiated groups ( 16). Such a difference between lobectomy and pneumonectomy was not observed among patients presenting with locally advanced disease (T3, Nl or N2 tumor) treated with the more modern technique. Several factors may have contributed to this difference: the small dose reduction, the use of lung correction factor for tissue inhomogeneity, the use of CT-scan-based treatment planning to select the radiation procedure or the introduction of a linear accelerator. In this study, we did not observe any survival difference between the two groups of patients treated at least partially with a Co60 source (Co60 and BetaCo group). The latter takes into account lung factor correction for tissue inhomogeneity. This factor should lead to a greater dose modification after a lobectomy than after a pneumonectomy: in our own experience, this difference ranged between 3 to 15% (14).
Volume
27. Number
3, 1993
This difference may explain some of the late damage induced to structure lying within the mediastinum, great vessels, and esophagus; this is certainly not the case for lung tissue as the doses delivered already exceed its tolerance. The major contribution of CT-scan-based treatment planning has been well-documented both for inoperable lung cancer and for postoperative radiation (6, 7, 11, 14). In our prior study, the target volume dose was overestimated by 5% when comparing treatment planning based on orthogonal X rays to the one done with a CT scan ( 14). Furthermore, possible mediastinal or lung shifts were not easily detected with orthogonal X rays, which was not the case with CT scan. A main limiting factor in the design of a course of postoperative radiation is certainly the tolerance of the remaining lung parenchyma. The dose required to control microscopic disease will necessarily exceed the lung tolerance. So, limiting the amount of normal lung tissue included in the treatment fields represents a major requirement to avoid life threalening late damage. Today, a CT scan for treatment planning is performed after surgery in treatment position before the beginning of the radiation. This will allow to design for each patient the best field set-up taking into account an abnormal anatomy such as a mediastinal shift and limiting as much as possible the volume of normal lung irradiated. This is one possible explanation for the difference observed between the different groups. The next improvement will certainly be due to the introduction of a true 3-dimensional (3-D) treatment planning instead of using a single or a few planes: then, the 3-D isodose displays and dosevolume-histograms may become routinely available (5). Nevertheless, a control group without postoperative radiation should be randomly selected to assess that postoperative radiation can be safely carried out with high doses even after a pneumonectomy. Our preliminary results with an adequate technique are in good agreement with some of the data published (3, 4. 12). Postoperative radiation, especially in the presence of positive lymph nodes, remains a controversial issue for lung cancers due partially to the lack of well-designed randomized trial including a large number of patients as well as an adequate staging procedure and well-designed radiation programs (9). Those are the reasons why the GETCB and the EORTC have launched such a study which requires the use of CT scan-based treatment planning and linear accelerator to deliver the prescribed 60 Gy to the mediastinum.
REFERENCES American Joint Committee on Cancer. Task force on lung. Staging of lung cancer. Chicago: American Joint Committee on Cancer; 1979. Bangma, P. J. Postoperative radiotherapy. In: Deeley, T. J., ed. Carcinoma ofthe bronchus: Modern radiotherapy. New York: Appleton-Century-Crofts; 1972: 163- 170. Choi, N.; Grille, H.; Gardiello, M.: Scannell, J. G.; Wilkins,
E. W. Basis for new strategies in postoperative radiotherapy of bronchogenic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 6:31-35;1980. 4. Emami, B.; Kim, T.; Roper, C.; Simpson, J.; Pilepich, M.; Hederman, M. Postoperative radiation therapy in the management of lung cancer. Radiology 164:251-253:1987. 5. Emami, B.; Purdy, J. A.; Manolis, J.; Barest, G.; Cheng, E.;
Postoperative radiotherapy after pneumonectomy 0 P.
6.
7.
8.
9.
10.
Coia, L.; Doppke, K.; Galvin, J.; LoSasso, T.; Matthews, J.; Munzenrider, J.; Shank, B. Three-dimensional treatment planning for lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 21:217-227;1991. Goittein, M.; Wittenberg, J.; Mendcondo, M.; Doucette, J.; Friedberg, C.; Ferrucci, J.; Gunderson, L.; Linggood, R.; Shipley, W. U.; Fineberg, H. V. The value of CT scanning in radiation treatment planning: A prospective study. Int. J. Radiat. Oncol. Biol. Phys. 5: 1787- 1798; 1979. Hobday, P.; Hodson, N. J.; Husband, J.; Parker, R. P.; Macdonald, J. S. Computed tomography applied to radiotherapy treatment planning: Techniques and results. Radiology 133:477-482; 1979. Israel, L.; Bonadonna, G.; Sylvester, R.; members of the EORTC Lung Cancer Group. Controlled study with adjuvant radiotherapy, chemotherapy, immunotherapy and chemoimmunotherapy in operable squamous cell carcinoma of the lung. In: Muggia, F. M., Rozencweig, M., eds. Lung cancer: Progress in therapeutic research. New York: Raven Press; 1979:443-452. Lung Cancer Study Group. Effects of postoperative mediastinal radiation on completely resected stage II and stage III epidermoid cancer of the lung. New Engl. J. Med. 3 15: 1377-1381;1986. Patterson, R.; Russell, M. H. Clinical trial in malignant disease IV lung cancer. Clin. Radiol. 13:141-144;1962.
PHLIPS et al.
529
11. Perez, C. A.; Stanley, K.; Grundy, G.; Hanson, W.; Rubin, P.; Kramer, S.; Brady, L. W.; Marks, J. E.; Perez-Tamayo, R.; Brown, J.; Concannon, J.; Rotman, M. Impact of irradiation technique and tumor extent in tumor control and survival of patients with unresectable non-oat cell carcinoma of the lung. Report by the Radiation Therapy Oncology Group. Cancer 50: 109 l- 1099; 1982. 12. Slater, J. D.; Ellerbroek, N. A.; Barkley, Jr., H.; Mountain, C.; Oswald, M. J.; Roth, J. A.; Peters, L. J. Radiation therapy following resection of non-small cell bronchogenic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 20:945-951;1991. 13. Van Houtte, P. Postoperative radiotherapy for lung cancer. Lung Cancer 7:57-64; 199 1. 14. Van Houtte, P.; Piron, A.; Lustman-MarCchal, J.; Osteaux, M.; Henry, J. Computed axial tomography (CAT) contribution for dosimetry and treatment evaluation in lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 6:995-1000;1980. 15. Van Houtte, P.; Rocmans, P.; Nguyen, T. H.; LustmanMarechal, J.; Vanderhoeft, P.; Henry, J. Irradiation postoperatoire dans le cancer pulmonaire. J. Eur. Radiother. I: 17-25;1981. 16. Van Houtte, P.; Rocmans, P.; Smets, P.; Goffin, J. C.; Lustman-Marechal, J.; Vanderhoeft, P.; Henry, J. Postoperative radiation therapy in lung cancer: a controlled trial after resection of curative design. Int. J. Radiat. Oncol. Biol. Phys. 6:983-986; 1980.