Inr J. Radumm O,lCO/O~J’ Bml Ph.w, Vol. 24. pp. !i-23 Printed I” the U.S.A. All rights reserved.
( opynght
0360-3u16/9? $5.00 + .oo cc 1992 Pergamon Press Ltd.
??Clinical Original Contribution
COMBINED MODALITY THERAPY FOR STAGE III NON-SMALL CELL LUNG CARCINOMA: RESULTS OF TREATMENT AND PATTERNS OF FAILURE S.
REDDY,
E. KAPLAN,
M.D.,*
M. S. LEE, M.D.,*
P. BONOMI,
M.D.,+ MERYL GALE, M.D.,+ C. F. KITTLE,
M.D.+
Rush-Presbyterian-St.
M.D.,+
L. P. FABER,
S. G. TAYLOR, M.D.,*
AND F. R. HENDRICKSON,
IV, M.D.,+
W. WARREN,
M.D.,$
M.D.*
Luke’s Medical Center, Chicago, IL
Patients with Stage III non-small cell lung carcinoma continue to pose a therapeutic problem with dismal cure rates. In an effort to improve on these results, 129 patients with biopsy-proven clinical Stage III non-small cell lung carcinoma from November 1982 through November 1987, were entered into two consecutive Phase II studies at Rush-Presbyterian-St. Luke’s Medical Center. Treatment in the first study consisted of Cisplatin and S-Fluorouracil infusion with concomitant split course radiation; in the second Etoposide was added. Radiation and chemotherapy were given simultaneously on days one through five of each cycle in a preoperative fashion for four cycles in patients considered eligible for surgery and in a definitive fashion for six cycles in patients considered ineligible for surgery. Radiation was given in 2 Gy fractions for a planned preoperative dose of 40 Gy and a definitive dose of 60 Gy. Surgical resection was attempted four to five weeks later in patients treated preoperatively. Thus, 83 patients were treated preoperatively and 46 definitively. Eighty-three patients (64%) had IIIA disease and IIIB disease was found in the remainder of the patients. Sixty-two patients (75%) in the eligible for surgery group had a thoracotomy after the combined treatment with a resectability rate of 97% and an operative mortality rate of 5%. There were 17 patients (27%) with no evidence of residual cancer in the resected specimen. Three-year survival for the eligible for surgery group at 40% was significantly better than 19% observed in the ineligible for surgery group (p = 0.003). Seventy-six percent of the patients with no residual cancer in the resected specimen are recurrencefree at three years compared to 34% of the patients with gross residual. A total of 81 patients have failed after their treatment; 49 (59%) in the eligible for surgery group and 32 (70%) in the ineligible for surgery group. Of all the patients who failed, local failure alone and as a component occurred in 21 (26%) and 36 (44%) patients, respectively. Failure in distant sites alone was noted in 56% of the overall failures. Severe toxicity was unusual. There were three treatment related deaths (2%). Radiation esophagitis and pneumonitis were only mild to moderate seen in less than 10% of the patients. Survival rates and patterns of failure according to the stage of the disease, histology, treatment group and pathologic response will be presented in detail. Non-small cell lung carcinoma, Stage III, Combined modality therapy, Patterns of failure.
Carcinoma of the lung is the most commonly occurring cancer in the United States, with 157,000 new cases pre-
to local extension or patients’ medical condition. They are conventionally treated with external beam radiation alone with a low probability of long-term survival (10, 14). Thus, patients with Stage III lung carcinoma continue
dieted to be diagnosed in 1990 and 142,000 deaths attributable to it (21). Approximately seventy-five percent of the cases will have non-small cell histology, of whom more than a third present with locally advanced disease. Although these patients do not have proven metastatic disease at the time of diagnosis, a large proportion of them are not amenable to surgical resection either secondary
to pose a therapeutic problem and demonstrate a need for innovative methods of therapy to improve the existing dismal cure rates. Analysis of the first sites of failure in treated patients reveals both locoregional and distant sites to be at risk for failure (13, 22). Both Mluorouracil (5-FU) and Cisplatin are known radiation sensitizers. When used simultaneously, they
Presented at the 32nd Annual ASTRO Meeting in Miami, FL, October 1990. * Department of Therapeutic Radiology. + Department of Medical Oncology. * Department of Thoracic Surgery. Reprint requests to: Salitha Reddy, M.D., Rush-Presbyterian-
St. Luke’s Medical Center, 1653 West Congress Parkway, Chicago, IL 606 12. Acknowledgement-The authors wish to express their thanks to Norma G. Tirado for her secretarial assistance in the preparation of this manuscript. Accepted for publication 2 October 199 1.
INTRODUCTION
17
18
I. J. Radiation
Oncology
0 Biology 0 Physics
have synergistic cytotoxicity and have yielded high response rates as shown in the treatment of head and neck cancers (2,6, 12). These drugs have also been reported to give a 37% response rate in patients with localized nonsmall cell lung carcinoma (NSCLC) (1). Cisplatin-Etoposide regimen has produced a clinical response rate of 56% in previously untreated patients with NSCLC and has produced the highest one year survival rate in Stage IV patients treated in sequential Phase II trials conducted by Eastern Cooperative Oncology Group (ECOG) (7, 11). In addition, Etoposide (VP-16) and Cisplatin have been shown to be synergistic in vitro (19). In an effort to improve both locoregional control and survival rates in patients with Stage III NSCLC, a multimodality approach consisting of concomitant systemic chemotherapy and split-course radiation therapy was developed at Rush-Presbyterian-St. Luke’s Medical Center. Preliminary results of this approach have been published (I 5, 23). We have updated those results and will present the feasibility and toxicity of this modality, survival rates, and patterns of failure according to the treatment, stage, and response to treatment. METHODS
AND MATERIALS
Between November 1982 and November 1987, 129 patients with AJCC clinical Stage III MO, histologically confirmed NSCLC were entered into two consecutive Phase II studies. Pre-treatment evaluation included chest x-ray, CBC, blood chemistry, computed tomography of the chest, brain and upper abdomen, and a bone scan. Pleural effusions were evaluated cytologically and if they were found to be positive, the patient was eliminated from the study. In addition, patients with the ECOG PS worse than 3 and patients found to have distant metastatic disease were excluded from the study. Other eligibility requirements included leucocyte count greater than 4 X 103/ml, platelet counts greater than 100 X 103/ml, blood urea nitrogen value less than 25 mg/dl, serum creatinine value less than 1.5 mg/dl, and bilirubin level less than 1.5 mg/dl. Each patient was initially evaluated by a thoracic surgeon who classified the patient as being eligible for surgery or ineligible for surgery. Patients with potentially resectable tumors or those who could tolerate the resection were classified as belonging to the former group, whereas patients with technically unresectable tumors irrespective of their response to preoperative treatment or those who were medically inoperable were classified as the latter group. Tumors were classified as being technically unresectable when any one or more of the following features were identified; involvement of more than two tracheal rings; superior venacaval obstruction; right vocal cord paralysis; vertebral body erosion; contralateral mediastinal and hilar disease and infiltration into great vessels and mediastinal structures. The presence of cardiorespiratory insufficiency making the patients either unable to withstand the an-
Volume 24. Number
1. IW
aesthia or the pulmonary resection were considered to be medically inoperable. Patients in the eligible for surgery group were reevaluated four to five weeks after completing the preoperative treatment and if there was no progression of disease or medical contraindications, they underwent surgical resection. Prior to surgery, patients were restaged with a repeat bone scan and CAT scan of the head, chest, and upper abdomen. Repeat bronchoscopic examination was performed to evaluate planned resection margins. Surgery consisted of segmental resection, lobectomy, or pneumonectomy as determined by the thoracic surgeon. Patients who were found to have either inoperable or unresectable disease or refused surgery were recommended to complete the combined treatment. provided no progression of disease was noted. Patients in the ES group were to receive four cycles of concomitant chemotherapy and radiation and those belonging to the ineligible for surgery group were treated with planned six cycles of concomitant therapy. The first Phase II study consisted of delivering combination chemotherapy and split course radiation administered every other week for five days. On day 1, Cisplatin 60 mg/m2 IV was given over 1 hr, 5-FU 800 mg/m*/day as a 24 hr infusion for five days. In the second Phase II study, VP- 16 60 mg/m2/day IV over 1 hr on days 1-4 was added. The 5-FU was shortened by 24 hr and the interval between each cycle was increased to 21 days to minimize hematologic cytotoxicity. All three drugs were reduced by 25% for severe nausea and vomiting. If the symptoms persisted at the reduced dose, the dose was further reduced to 50% of the initial dose. 5-FU and VP16 doses were reduced by 25% if the leucocyte count was 3-4 X 10i/ml and by 50% if it was 2 < 3 X 103/ml. Treatment was completely withheld if the leucocyte count was lower than 2 X 103/ml or the platelet count was lower than 100 X 103/dl. Cisplatin dose was modified for renal toxicity. It was reduced by 25% if the blood creatinine level was > 1.5 < 2.0 mg/dl; by 50% if the level was >2.0 < 2.5 mg/dl. and completely withheld if the creatinine level was more than 2.5 mg/dl. Radiation therapy was given as 2 Gy fractions on days l-5 of each cycle which was repeated every 14 days in Study I and 21 days in Study Il. Patients in the eligible for surgery group were to receive a total dose of 40 Gy and those in the ineligible for surgery group were to receive 60 Gy. The initial 11 patients received two 1.16 Gy fractions daily, separated by at least 4 hours with an intent of delivering higher total doses and reducing the late toxicity. All the later patients received 2 Gy fractions as it was found that doses of 40-60 Gy with 2 Gy fractions were well-tolerated. Only megavoltage equipment was used. The treatment fields included the primary tumor and regional lymph nodes with a l-2 cm margin. Bilateral supraclavicular and mediastinal lymph nodes, subcarinal and ipsilateral hilar nodes were always treated. Contralateral hilar nodes were treated only when they were clin-
19
Lung carcinoma 0 S. REDDYel al.
ically involved or there was contralateral mediastinal disease. All doses were calculated at the midplane of the target volume and the minimal treatment distances to the isocenter were at least 80 ems. An anterior and posterior parallel opposed technique was used for the first 40 Gy. In patients receiving more than 40 Gy, the boost field used was designed to encompass the initial gross disease with a 1 cm margin, usually via opposed oblique or lateral fields, avoiding the spinal cord in its entirety, to a total dose of 60 Gy. All patients were followed up to a minimum of three years after treatment or until death. Survival and diseasefree survival rates were calculated using the Kaplan-Meir Method (8). The log rank method was used to test the significance of differences between survival rates according to the various prognostic factors.
RESULTS Pretreatment characteristics of patients and tumors according to the Study were as shown in Table 1. Sixtythree patients were entered into Study I consisting of two drugs and concomitant radiation therapy, whereas 66 patients were treated according to the Study II protocol with three drugs and radiation. Eighty-seven patients were male and 42 were female. Most of the patients were ambulatory (PS 0- 1) and were eligible for surgery. Ten and 17 patients in the ES and IES groups respectively had PS 2-3. Fortysix (38%) patients had lost >5% of their body weight before starting treatment of whom 26 patients belonged to the ES group. Squamous cell carcinoma was the most common histology seen in 57 patients (44%). Adenocarcinoma and large cell carcinoma were seen in 38 and 34 patients, Table 1. Characteristics of patients and tumors by study Study I (N = 63)
Sex Male Female Performance status 0 and 1 2 and 3 Weight loss O-S% >5% Histology Squamous cell Adenocarcinoma Large cell Eligibility for surgery Yes No Stage IIIA BIB
Study II (N = 66)
No.
%
No.
%
42 21
67 33
45 21
68 32
50 13
79 21
52 14
19 21
36 22
62 38
40 24
62 38
28 22 13
44 35 21
29 16 21
44 24 32
54 9
86 14
29 37
44 56
50 13
79 21
33 33
50 50
Table 2. Patient classification according to disease stage T Stage N Stage 0 1 2
1
2
3
4
Total
41
18
59
4
14
23
10
51
1 5
6 20
5 69
6 34
18
54% 3
80%
respectively. Sixty-four percent of the patients had AJCC Stage IIIA disease and the remaining patients had Stage BIB disease. IIIA disease was found in 83% of the ES group, but only in 30% of the IES group. Study II had more patients with Stage IIIB disease (50% vs. 21%) and more patients who were ineligible for surgery (56% vs. 14%) than Study I. T and N stage distribution was as shown in Table 2. One hundred and three patients (80%) had the local extension of the primary staged as being T3 or Tq. Mediastinal node involvement was seen in more than half of the patients. Histologic confirmation of nodal involvement was obtained in 35 patients. Superior sulcus tumors were seen in nine patients. Forty-three (33%) patients did not complete the planned course of treatment for various reasons. Tumor progression in 10; treatment related deaths in three; cardiac related deaths in six; worsening of pulmonary insufficiency in three; unknown in one; and patient refusal in 20. Twelve patients in the eligible for surgery group refused surgery after completion of preoperative treatment and seven of these received more concomitant therapy. Eight patients in the ineligible for surgery group refused to complete their planned six cycles of treatment. All of these patients are included in the final analysis of survival rates. Actuarial survival and recurrence-free survival rates for the entire group were identical as shown in Figure 1. The two- and three-year survival rates were 40% and 32%, respectively with a median survival of 535 days (17.6 months). Patients with IIIA disease had a survival rate of 40% at three years, which was significantly better than 22% seen in patients with Stage IIIB (Fig. 2). Although there were more patients with worse prognostic factors in Study II, the actuarial survival rates were identical for patients in both the studies (Fig. 3).
Fig. 1. Survival and recurrence free survival of all patients.
20
1. J. Radiation Oncology 0 Biology 0 Physics
Volume 24. Number I_ 1992 Table 3. Reasons
DAYS
for no surgery in eligible patients
No. = 21 (25%).
Fig. 2. Actuarial survival according to Stage.
There were 83 patients in the eligible for surgery group treated preoperatively. Of these 62 (75%) patients had a thoracotomy performed after the combined modality treatment. The reasons for not performing surgery were as shown in Table 3. All but two patients had resection of the tumor at surgery. The surgical procedures performed were pneumonectomy in 24; lobectomy in 29, and segmental resection in seven, six of whom had chest wall involvement. Thus, the resectability rate was 97%. All the resected specimens were assessed pathologically for response to the preoperative treatment. Seventeen of 62 patients (27%) had no evidence of residual cancer in the resected lung or regional nodes. Eleven patients had microscopic residual cancer and 34 had gross residual cancer. The three-year survival rate for the group of patients classified as eligible for surgery at 40% was significantly better than those who were ineligible for surgery p = 0.003 (Fig. 4). Patients in the ineligible for surgery group who received hyperfractionated radiation were analyzed separately and as they were found to have similar outcome to those receiving one fraction a day, all patients were analyzed together. The survival at three years for 62 patients who had surgery was 47% compared to 17% for the patients who did not have surgery (p = 0.0001) (Fig. 5). Patients with no residual cancer after evaluation of the resected specimen had a survival rate of 76% which was significantly better (a = 0.000 1) than those with residual tumor (Fig. 6). Patterns 0Jfailure Seventy-three percent of the patients who were to fail, did so by the first year of treatment and 93% of the patients failed by two years. Eighty-one patients have recurred with the sites of first failure being local only in 2 1 (26%): distant
only in 45 (56%); both distant and local in 15 (18%). Thus. local failure alone occurred in 26% of the patients who failed, and as a component of overall failure, occurred in 44% of the patients. Distant failure as a component occurred in 74% of all failures. Brain only as the first site of failure was seen in 10 patients or 12% of all failures. The pattern of first failure was analyzed according to several variables as shown in Table 4. There was a trend towards increased failure rates in patients with IIIB disease, patients with nodal involvement, patients who were ineligible for surgery and those who were non-responders. However, of the patients failing, the pattern of failure was similar among all the groups.
Severe toxicity of combined treatment was unusual (Table 5). Alopecia was seen in all patients receiving VP16. Nausea and vomiting were common side effects seen in 8 1% of the patients. However, only 13% of the patients had severe nausea and vomiting. Moderate esophagitis and pneumonitis occurred in 3% and 6% of the patients. respectively. Combined treatment related deaths were seen in three patients, all from sepsis during a period of leucopenia. Two of these patients who expired were nonambulatory (PS 2-3) and had lost more than 5%’of their usual body weight before starting treatment. Because of this lethal toxicity, the starting dose of VP-16 and 5-FU was reduced by 25% in patients who had lost more than 5% of their usual body weight or whose PS was ECOG 2 or 3. There were no deaths attributable to treatment after this change was made. Another six patients had suffered cardiac death and four of these patients were in the ineligible for surgery group. The operative mortality was 5% (3/62). All three patients who died had pneumonectomy. Two post-operative deaths were due to pulmonary emboli and one was caused by pulmonary infection in
iI&
pzor-:“”
0.4 0.2
*
I 0
I
0
250
540
760
rma vso
l500
9750
zmo
2260
2.500
2750
ot-; 0
254
: 750
: 1000
v--t-+,250
-2
1500
,750
2ooo
2250
2500
2750
DAYS
DAYS
Fig. 3. Actuarial survival according to study.
: 500
Fig. 4. Actuarial
survival according
to eligibility for surgery.
21
Lung carcinoma 0 S. REDDY uf al.
Fig. 5. Actuarial survival according to *surgery.
Fig. 6. Actuarial survival according to response.
the remaining lung. Major complications occurred in 14 patients and they were as shown in Table 6.
Surgical resection has been shown to be well-tolerated after preoperative radiation with 20’70five-year survival rates in locally advanced NSCLC ( 17,20). As locoregional failure is a predominant component of overall failure pattern in these patients, surgical resection, whenever possible, was performed as part of the protocol. The overall results in these two Phase II studies support the feasibility of the program with acceptable toxicity. There were three deaths related to the concomitant therapy, early in the course of the study. After appropriate changes were made, no further deaths secondary to treatment occurred. Cardiac deaths in six patients are of concern. Although this patient population may be at a high risk for cardio-vascular disease, a possible association of 5-FU and myocardial injury has been noted (16). The majority of the patients who failed to complete their planned treatment were in the eligible for surgery group and refused to have the operation after preoperative treatment. Progression of disease during the split-course of radiation was a theoretical concern in this study. However, only ten (8%) patients had their disease progression during the treatment, of whom only one patient had progression locally alone. This is lower than that observed in patients treated with continuous preoperative radiation therapy at our institution (where about 16% have progressed with
DISCUSSION
The majority of patients who have locally advanced NSCLC will succumb to their disease. Studies evaluating the patterns of failure in this group of patients have revealed failure in both local and distant sites to be a significant problem (13, 17, 22). Improvements on these results can be made only with treatment approaches that can not only improve locoregional control, but also be more effective against micrometastasis. The optimal sequence for chemotherapy and radiation in Stage III NSCLC has not been demonstrated. Using chemotherapy early in the course of treatment provides an opportunity to treat micrometastasis and to deliver the agents at full dose intensity. Since Cisplatin-Etoposide containing regimens have been demonstrated to be the most effective in the management of NSCLC patients, and with 5-FU demonstrated to have radiation sensitizing properties, we have initiated the Phase II studies incorporating all these rationale in the management of patients with Stage III NSCLC (2, 11). Because of the concomitant treatment, it was necessary to use split-course therapy to avoid any undue morbidity and mortality in this group of patients.
Table 4. Patterns of failure % of all failures
Variable Stage IIIA BIB Study I II Eligibility for surgery Yes No Nodal status NU N+ Path response PO P+
No.
%
failure
failure
Local only
Distant only
Both
48 33
58 12
25 27
63 45
12 27
41 40
65 61
27 2s
61 50
12 25
49 32
59 IO
24 28
65 41
10 31
31 50
53 72
3s 20
52 58
13 22
3 30
18 67
33 23
67 63
13
22
1. J. Radiation Oncology 0 Biology 0 Physics
more than half doing so locoregionally) ( 17), or that seen in patients treated with radiation alone (13). Since all patients were followed for a minimum of three years after treatment and 93% of the recurrences appeared within two years, we feel that the recurrence-free survival rates will flatten at the level of 32% seen at three years. This compares favorably with other recently published reports of combined modality treatments or the best subgroup of patients treated with hyperfractionated radiation (3, 5, 18). If indeed this treatment modality is superior to other modalities needs to be tested in a randomized study. In spite of having patients with worse prognostic factors in Study II, the survival rates were identical in both the studies suggesting increased efficacy of the treatment by the addition of VP-16. Although at variance with that reported by Curran et al. (4), the survival rate for IIIA patients was significantly better than that seen in IIIB patients, indicating the usefulness of the current AJCC staging system. The rationale for preoperative radiation is to increase the resectability of gross tumor, to sterilize the microscopic disease, and to reduce the number of cells capable of implantation at the time of surgery. Several studies have demonstrated resectability rates ranging from 50-72% with the usage of preoperative radiation, which is comparable to the 75% resectability rate observed in our group of patients having more locally advanced disease (20,24). The operative mortality rate of 5% was also minimal although resection was found to be difficult because of
Table 5. Toxicity of treatment Toxicity Nausea and vomiting Mild Moderate Severe Esophagitis Mild Moderate Pneumonitis Mild Moderate Alopecia Sepsis
% of patients
34 34 13 II 3 2 6 In all patients receiving VP- 16 2
Volume 24, Number I, 1992 Table 6. Surgical results
Operative mortality Resection Complications Fistula Tracheotomy Atelectasis Pneumonitis Stroke
.
No.
%
3/62 60162 14 4 2 5 2
5 97 22
1
marked fibrosis around the hila and fascial, and vascular planes were usually obliterated. There were four bronchial fistulae after resection, three of them occurring after pneumonectomy. Depending on the total dose of radiation delivered preoperatively, other authors have reported a range of 20% to 54% of the patients having no persistent tumor or only having microscopic disease (20,24, 25). Our results, documenting 27% of the patients having no demonstrable tumor in the resected specimen, especially considering that all the patients in our study had Stage III disease are encouraging. This was higher than 16% seen in similar stage patients treated at our institution with preoperative radiation only ( 17). This might be of importance especially with the results indicating that patients with no residual disease in the resected specimen have a significantly better survival than those with residual disease. If this is merely a reflection of the extent of the disease, it needs to be tested in a larger group of patients than we had. Most of the patients whose disease recurred did so within two years of their diagnosis as found by others (13). With local failure as a component of overall failures occurring in 44% of the patients and distant failure in 74% of the patients, we feel that future studies should continue to strive for better local and distant control rates. Patients in the eligible for surgery group have fared significantly better as to survival than those in the ineligible for surgery group. Although it is possible that selection factors other than surgical procedures could have accounted for the differences, we feel that surgery, if possible, should remain as a component in the overall management of patients with locally advanced NSCLC since local failure is a significant component of overall failures. Surgical resection has been reported to reduce the local recurrence (9).
REFERENCES Byfield, J. E.; Stanton, W.; Sharp, T. R.; Frankel, S. S. Koziol, J. Phase I-II study of 120-hour infused 5-FU and split course radiation therapy in localized non-small cell lung cancer. Cancer Treat. Rep. 67: 933-935; 1983. Byfield, J. E.; Calabro-Jones, P. M., Klisak, I.; Kulhanian, F. Pharmacologic requirements for obtaining sensitization of human tumor cells in vitro to combined 5-fluorouracil
or florafur and x-rays. Int. J. Radiat. Oncol. Biol. Phys. 8: 1923-1933; 1982. 3. Cox, J. D.; Azarnis, N.; Byhardt, R. W.; Shin, K. H.; Emami, B.; Pajak, T. F.: A randomized Phase I/II trial of hyperfractionated radiation therapy with total doses of 60.0 Gy to 79.2 Gy: possible survival benefit with 269.6 Gy in favorable patients with radiation therapy oncology group stage
Lung carcinoma 0 S. REDDY d al.
4.
5.
6.
I.
8.
9.
10.
II.
12.
13.
14.
III non-small cell lung carcinoma: Report of Radiation Therapy Oncology Group 83-l 1. J. Clint. Oncol. 8: 15431555; 1990. Curran, J. R.; Stafford, W. J. Lack of apparent difference in outcome between clinically staged IIIA and IIIB nonsmall cell lung cancer treated with radiation therapy. J. Clin. Oncol. 8: 409-4 15; 1990. Dillman, R. 0.; Seagren, S. L.; Propert, K. J.: Guerra, J.; Eaton, W. L.; Frei, E. F.; Perry, M. C.; Carey, R. W.; Frei III, E. F.; and Green, M. R. Protochemotherapy improves survival in regional non-small cell lung cancer. N. Engl. J. Med. 323: 14: 940-945; 1990. Douple, E. B.; Richmond, R. C. Platinum complexes as radiosensitizers of hypoxic mammalian cells. Br. J. Cancer 37: 98-102; 1978. Finkelstein, D. M.; Ettinger, D. S.; Ruckdeschel, J. C. Longterm survivors in metastatic non-small-cell lung cancer: an Eastern Cooperative Oncology Group study. J. Clin. Oncol. 4: 702-709; 1986. Kaplan, E. L.; Meier, P. Non-parametric estimation from incomplete observations. J. Am. Stat, Assoc. 53: 457-48 1; 1958. Katz, H. R. The effect of resection on local failure in irradiated non-oat cell carcinoma ofthe lung. Int. J. Rad. Oncol. Biol. Phys. 9: 1793-1805; 1983. Komaki, R.; Cox, J. D.; Hartz, A. J.; Byhardt, R. W.; PerezTamayo, C.; Clowry, L.; Choi, H.; Wilson, F.; Lopes da Conceicao, A.; Rangala, N. Characteristics of long-term survivors after treatment for inoperable carcinoma of the lung. Am. J. Clin. Oncol. 8: 362-370; 1985. Longeval, E.; Klastersky, J. Combination chemotherapy with cisplatin and etoposide in bronchogenic squamous cell carcinoma and adenocarcinoma. Cancer 50: 275 l-2756; 1982. Murthy, A. K.; Taylor, IV, S. G.; Showel, J.; Calderarelli, D. C.; Hutchinson Jr., J. C.; Holinger, L. D.; Witt, T. R.; Hoover, S. Treatment of advanced head and neck cancer with concomitant radiation and chemotherapy. lnt. J. Radiat. Oncol. Biol. Phys. 13: 01-07; 1987. Perez, C. A.; Pajak, T. F.; Rubin, P.; Simpson, J. R.; Mohiuddin, M.; Brady, L. W.; Perez-Tamayo, R.; Rotman, M. Long-term observations of the patterns of failure in patients with unresectable non-oat cell carcinoma of the lung treated with definitive radiotherapy. Report by the Radiation Therapy Oncology Group. Cancer. 59: 1874- 188 I ; 1987. Perez, C. A.; Stanley, K.; Rubin, P.; Kramer, S.; Brady, L.; Perez-Tamayo, R.; Brown, G. S.; Concannon, J.; Rotman,
15.
16. 17.
18.
19.
20.
21.
23
M.; Seydel, H. G. A prospective randomized study of various irradiation doses and fractionation schedules in the treatment of inoperable non-oat cell carcinoma of the lung. Cancer 45: 2744-2753; 1980. Pincus, M.; Reddy, S.; Lee, M. S.; Bonomi, P.; Taylor, S.; Rowland, K.; Faber, L. P.; Warren. W.; Kittle, C. F.; Hendrickson, F. R. Preoperative combined modality therapy for stage III M non-small cell lung carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 15: 189-195, 1988. Pottage, A.; Holt, S.; Ludgate, S.; Langlands, A. 0.; Fluorouracil cardio toxicity. Br. Med. J. 1: 547-548; 1978. Reddy, S.; Faber, L. P.; Baumann, L. A.; Lee, M. S.; Jensik, R. J.; Kittle, F. C.; Bonomi, P. D.; Taylor IV, S. G.; Hendrickson, F. R. Preoperative radiation therapy in regionally localized stage III non-small cell lung carcinoma: long-term results and patterns of failure. Int. J. Radiat. Oncol. Biol. Phys. 19: 287-292; 1990. Robinow, J. S.; Shaw, E. G.; Eagan, R. T.; Lee, R. E.: Creagan. E. T.; Frytak, S.; Richardson, R. L.; Jett, J. R.; Burton, J. K.; Ilstrup, D. M. Results of combination chemotherapy and thoracic radiation therapy for unresectable non-small cell carcinoma of the lung. lnt. J. Radiat. Oncol. Biol. Phys. 17: 1203-1210; 1989. Schabel. F. M. Jr.; Trader, M. W.; Laster, W. R. Cisdichiorodiammineplatinum, II. Combination of chemotherapy and cross-resistance studies with tumors of mice. Cancer Treat. Rep. 63: 1459-1473; 1979. Sherman, D. M.; Neptune, W.; Weichselbaum, R.: Order, S. E.; Piro, A. J. An aggressive approach to marginally resectable lung cancer. Cancer. 41: 2040-2045; 1978. Silverberg, E.; Boring, C. C.; Squires. T. S. Cancer statistics, Ca-A Cancer Journal for Clinicians. 40: 9-26; 1990.
22. Stanley, K.: Cox, J. D.; Patrovich. Z.; Paig, C. Patterns of failure in patients with inoperable carcinoma of the lung. Cancer. 47: 2725-2729; 198 I. 23. Taylor IV, S. G.; Trybula, M.; Bonomi, P. D.; Faber, L. P.; Lee, M. S.: Reddy, S.: Maffey, S. C.: Mathisen, D. J.; Jensik, R. J.; Kittle, C. F. Simultaneous cisplatin fluorouracil infusion and radiation followed by surgical resection in regionally localized stage III. non-small cell lung cancer. Ann. Thorac. Surg. 43: 87-91; 1987. 24. Warram, J. Preoperative irradiation of cancer of the lung: final report of a therapeutic trial. A collaborative study. Cancer 36: 914-925; 1975. 25. Widow, W. Preoperative irradiation of bronchial carcinoma. Cancer 28: 798-80 I; 197 I.
( opynght
0360-3u16/9? $5.00 + .oo cc 1992 Pergamon Press Ltd.
??Clinical Original Contribution
COMBINED MODALITY THERAPY FOR STAGE III NON-SMALL CELL LUNG CARCINOMA: RESULTS OF TREATMENT AND PATTERNS OF FAILURE S.
REDDY,
E. KAPLAN,
M.D.,*
M. S. LEE, M.D.,*
P. BONOMI,
M.D.,+ MERYL GALE, M.D.,+ C. F. KITTLE,
M.D.+
Rush-Presbyterian-St.
M.D.,+
L. P. FABER,
S. G. TAYLOR, M.D.,*
AND F. R. HENDRICKSON,
IV, M.D.,+
W. WARREN,
M.D.,$
M.D.*
Luke’s Medical Center, Chicago, IL
Patients with Stage III non-small cell lung carcinoma continue to pose a therapeutic problem with dismal cure rates. In an effort to improve on these results, 129 patients with biopsy-proven clinical Stage III non-small cell lung carcinoma from November 1982 through November 1987, were entered into two consecutive Phase II studies at Rush-Presbyterian-St. Luke’s Medical Center. Treatment in the first study consisted of Cisplatin and S-Fluorouracil infusion with concomitant split course radiation; in the second Etoposide was added. Radiation and chemotherapy were given simultaneously on days one through five of each cycle in a preoperative fashion for four cycles in patients considered eligible for surgery and in a definitive fashion for six cycles in patients considered ineligible for surgery. Radiation was given in 2 Gy fractions for a planned preoperative dose of 40 Gy and a definitive dose of 60 Gy. Surgical resection was attempted four to five weeks later in patients treated preoperatively. Thus, 83 patients were treated preoperatively and 46 definitively. Eighty-three patients (64%) had IIIA disease and IIIB disease was found in the remainder of the patients. Sixty-two patients (75%) in the eligible for surgery group had a thoracotomy after the combined treatment with a resectability rate of 97% and an operative mortality rate of 5%. There were 17 patients (27%) with no evidence of residual cancer in the resected specimen. Three-year survival for the eligible for surgery group at 40% was significantly better than 19% observed in the ineligible for surgery group (p = 0.003). Seventy-six percent of the patients with no residual cancer in the resected specimen are recurrencefree at three years compared to 34% of the patients with gross residual. A total of 81 patients have failed after their treatment; 49 (59%) in the eligible for surgery group and 32 (70%) in the ineligible for surgery group. Of all the patients who failed, local failure alone and as a component occurred in 21 (26%) and 36 (44%) patients, respectively. Failure in distant sites alone was noted in 56% of the overall failures. Severe toxicity was unusual. There were three treatment related deaths (2%). Radiation esophagitis and pneumonitis were only mild to moderate seen in less than 10% of the patients. Survival rates and patterns of failure according to the stage of the disease, histology, treatment group and pathologic response will be presented in detail. Non-small cell lung carcinoma, Stage III, Combined modality therapy, Patterns of failure.
Carcinoma of the lung is the most commonly occurring cancer in the United States, with 157,000 new cases pre-
to local extension or patients’ medical condition. They are conventionally treated with external beam radiation alone with a low probability of long-term survival (10, 14). Thus, patients with Stage III lung carcinoma continue
dieted to be diagnosed in 1990 and 142,000 deaths attributable to it (21). Approximately seventy-five percent of the cases will have non-small cell histology, of whom more than a third present with locally advanced disease. Although these patients do not have proven metastatic disease at the time of diagnosis, a large proportion of them are not amenable to surgical resection either secondary
to pose a therapeutic problem and demonstrate a need for innovative methods of therapy to improve the existing dismal cure rates. Analysis of the first sites of failure in treated patients reveals both locoregional and distant sites to be at risk for failure (13, 22). Both Mluorouracil (5-FU) and Cisplatin are known radiation sensitizers. When used simultaneously, they
Presented at the 32nd Annual ASTRO Meeting in Miami, FL, October 1990. * Department of Therapeutic Radiology. + Department of Medical Oncology. * Department of Thoracic Surgery. Reprint requests to: Salitha Reddy, M.D., Rush-Presbyterian-
St. Luke’s Medical Center, 1653 West Congress Parkway, Chicago, IL 606 12. Acknowledgement-The authors wish to express their thanks to Norma G. Tirado for her secretarial assistance in the preparation of this manuscript. Accepted for publication 2 October 199 1.
INTRODUCTION
17
18
I. J. Radiation
Oncology
0 Biology 0 Physics
have synergistic cytotoxicity and have yielded high response rates as shown in the treatment of head and neck cancers (2,6, 12). These drugs have also been reported to give a 37% response rate in patients with localized nonsmall cell lung carcinoma (NSCLC) (1). Cisplatin-Etoposide regimen has produced a clinical response rate of 56% in previously untreated patients with NSCLC and has produced the highest one year survival rate in Stage IV patients treated in sequential Phase II trials conducted by Eastern Cooperative Oncology Group (ECOG) (7, 11). In addition, Etoposide (VP-16) and Cisplatin have been shown to be synergistic in vitro (19). In an effort to improve both locoregional control and survival rates in patients with Stage III NSCLC, a multimodality approach consisting of concomitant systemic chemotherapy and split-course radiation therapy was developed at Rush-Presbyterian-St. Luke’s Medical Center. Preliminary results of this approach have been published (I 5, 23). We have updated those results and will present the feasibility and toxicity of this modality, survival rates, and patterns of failure according to the treatment, stage, and response to treatment. METHODS
AND MATERIALS
Between November 1982 and November 1987, 129 patients with AJCC clinical Stage III MO, histologically confirmed NSCLC were entered into two consecutive Phase II studies. Pre-treatment evaluation included chest x-ray, CBC, blood chemistry, computed tomography of the chest, brain and upper abdomen, and a bone scan. Pleural effusions were evaluated cytologically and if they were found to be positive, the patient was eliminated from the study. In addition, patients with the ECOG PS worse than 3 and patients found to have distant metastatic disease were excluded from the study. Other eligibility requirements included leucocyte count greater than 4 X 103/ml, platelet counts greater than 100 X 103/ml, blood urea nitrogen value less than 25 mg/dl, serum creatinine value less than 1.5 mg/dl, and bilirubin level less than 1.5 mg/dl. Each patient was initially evaluated by a thoracic surgeon who classified the patient as being eligible for surgery or ineligible for surgery. Patients with potentially resectable tumors or those who could tolerate the resection were classified as belonging to the former group, whereas patients with technically unresectable tumors irrespective of their response to preoperative treatment or those who were medically inoperable were classified as the latter group. Tumors were classified as being technically unresectable when any one or more of the following features were identified; involvement of more than two tracheal rings; superior venacaval obstruction; right vocal cord paralysis; vertebral body erosion; contralateral mediastinal and hilar disease and infiltration into great vessels and mediastinal structures. The presence of cardiorespiratory insufficiency making the patients either unable to withstand the an-
Volume 24. Number
1. IW
aesthia or the pulmonary resection were considered to be medically inoperable. Patients in the eligible for surgery group were reevaluated four to five weeks after completing the preoperative treatment and if there was no progression of disease or medical contraindications, they underwent surgical resection. Prior to surgery, patients were restaged with a repeat bone scan and CAT scan of the head, chest, and upper abdomen. Repeat bronchoscopic examination was performed to evaluate planned resection margins. Surgery consisted of segmental resection, lobectomy, or pneumonectomy as determined by the thoracic surgeon. Patients who were found to have either inoperable or unresectable disease or refused surgery were recommended to complete the combined treatment. provided no progression of disease was noted. Patients in the ES group were to receive four cycles of concomitant chemotherapy and radiation and those belonging to the ineligible for surgery group were treated with planned six cycles of concomitant therapy. The first Phase II study consisted of delivering combination chemotherapy and split course radiation administered every other week for five days. On day 1, Cisplatin 60 mg/m2 IV was given over 1 hr, 5-FU 800 mg/m*/day as a 24 hr infusion for five days. In the second Phase II study, VP- 16 60 mg/m2/day IV over 1 hr on days 1-4 was added. The 5-FU was shortened by 24 hr and the interval between each cycle was increased to 21 days to minimize hematologic cytotoxicity. All three drugs were reduced by 25% for severe nausea and vomiting. If the symptoms persisted at the reduced dose, the dose was further reduced to 50% of the initial dose. 5-FU and VP16 doses were reduced by 25% if the leucocyte count was 3-4 X 10i/ml and by 50% if it was 2 < 3 X 103/ml. Treatment was completely withheld if the leucocyte count was lower than 2 X 103/ml or the platelet count was lower than 100 X 103/dl. Cisplatin dose was modified for renal toxicity. It was reduced by 25% if the blood creatinine level was > 1.5 < 2.0 mg/dl; by 50% if the level was >2.0 < 2.5 mg/dl. and completely withheld if the creatinine level was more than 2.5 mg/dl. Radiation therapy was given as 2 Gy fractions on days l-5 of each cycle which was repeated every 14 days in Study I and 21 days in Study Il. Patients in the eligible for surgery group were to receive a total dose of 40 Gy and those in the ineligible for surgery group were to receive 60 Gy. The initial 11 patients received two 1.16 Gy fractions daily, separated by at least 4 hours with an intent of delivering higher total doses and reducing the late toxicity. All the later patients received 2 Gy fractions as it was found that doses of 40-60 Gy with 2 Gy fractions were well-tolerated. Only megavoltage equipment was used. The treatment fields included the primary tumor and regional lymph nodes with a l-2 cm margin. Bilateral supraclavicular and mediastinal lymph nodes, subcarinal and ipsilateral hilar nodes were always treated. Contralateral hilar nodes were treated only when they were clin-
19
Lung carcinoma 0 S. REDDYel al.
ically involved or there was contralateral mediastinal disease. All doses were calculated at the midplane of the target volume and the minimal treatment distances to the isocenter were at least 80 ems. An anterior and posterior parallel opposed technique was used for the first 40 Gy. In patients receiving more than 40 Gy, the boost field used was designed to encompass the initial gross disease with a 1 cm margin, usually via opposed oblique or lateral fields, avoiding the spinal cord in its entirety, to a total dose of 60 Gy. All patients were followed up to a minimum of three years after treatment or until death. Survival and diseasefree survival rates were calculated using the Kaplan-Meir Method (8). The log rank method was used to test the significance of differences between survival rates according to the various prognostic factors.
RESULTS Pretreatment characteristics of patients and tumors according to the Study were as shown in Table 1. Sixtythree patients were entered into Study I consisting of two drugs and concomitant radiation therapy, whereas 66 patients were treated according to the Study II protocol with three drugs and radiation. Eighty-seven patients were male and 42 were female. Most of the patients were ambulatory (PS 0- 1) and were eligible for surgery. Ten and 17 patients in the ES and IES groups respectively had PS 2-3. Fortysix (38%) patients had lost >5% of their body weight before starting treatment of whom 26 patients belonged to the ES group. Squamous cell carcinoma was the most common histology seen in 57 patients (44%). Adenocarcinoma and large cell carcinoma were seen in 38 and 34 patients, Table 1. Characteristics of patients and tumors by study Study I (N = 63)
Sex Male Female Performance status 0 and 1 2 and 3 Weight loss O-S% >5% Histology Squamous cell Adenocarcinoma Large cell Eligibility for surgery Yes No Stage IIIA BIB
Study II (N = 66)
No.
%
No.
%
42 21
67 33
45 21
68 32
50 13
79 21
52 14
19 21
36 22
62 38
40 24
62 38
28 22 13
44 35 21
29 16 21
44 24 32
54 9
86 14
29 37
44 56
50 13
79 21
33 33
50 50
Table 2. Patient classification according to disease stage T Stage N Stage 0 1 2
1
2
3
4
Total
41
18
59
4
14
23
10
51
1 5
6 20
5 69
6 34
18
54% 3
80%
respectively. Sixty-four percent of the patients had AJCC Stage IIIA disease and the remaining patients had Stage BIB disease. IIIA disease was found in 83% of the ES group, but only in 30% of the IES group. Study II had more patients with Stage IIIB disease (50% vs. 21%) and more patients who were ineligible for surgery (56% vs. 14%) than Study I. T and N stage distribution was as shown in Table 2. One hundred and three patients (80%) had the local extension of the primary staged as being T3 or Tq. Mediastinal node involvement was seen in more than half of the patients. Histologic confirmation of nodal involvement was obtained in 35 patients. Superior sulcus tumors were seen in nine patients. Forty-three (33%) patients did not complete the planned course of treatment for various reasons. Tumor progression in 10; treatment related deaths in three; cardiac related deaths in six; worsening of pulmonary insufficiency in three; unknown in one; and patient refusal in 20. Twelve patients in the eligible for surgery group refused surgery after completion of preoperative treatment and seven of these received more concomitant therapy. Eight patients in the ineligible for surgery group refused to complete their planned six cycles of treatment. All of these patients are included in the final analysis of survival rates. Actuarial survival and recurrence-free survival rates for the entire group were identical as shown in Figure 1. The two- and three-year survival rates were 40% and 32%, respectively with a median survival of 535 days (17.6 months). Patients with IIIA disease had a survival rate of 40% at three years, which was significantly better than 22% seen in patients with Stage IIIB (Fig. 2). Although there were more patients with worse prognostic factors in Study II, the actuarial survival rates were identical for patients in both the studies (Fig. 3).
Fig. 1. Survival and recurrence free survival of all patients.
20
1. J. Radiation Oncology 0 Biology 0 Physics
Volume 24. Number I_ 1992 Table 3. Reasons
DAYS
for no surgery in eligible patients
No. = 21 (25%).
Fig. 2. Actuarial survival according to Stage.
There were 83 patients in the eligible for surgery group treated preoperatively. Of these 62 (75%) patients had a thoracotomy performed after the combined modality treatment. The reasons for not performing surgery were as shown in Table 3. All but two patients had resection of the tumor at surgery. The surgical procedures performed were pneumonectomy in 24; lobectomy in 29, and segmental resection in seven, six of whom had chest wall involvement. Thus, the resectability rate was 97%. All the resected specimens were assessed pathologically for response to the preoperative treatment. Seventeen of 62 patients (27%) had no evidence of residual cancer in the resected lung or regional nodes. Eleven patients had microscopic residual cancer and 34 had gross residual cancer. The three-year survival rate for the group of patients classified as eligible for surgery at 40% was significantly better than those who were ineligible for surgery p = 0.003 (Fig. 4). Patients in the ineligible for surgery group who received hyperfractionated radiation were analyzed separately and as they were found to have similar outcome to those receiving one fraction a day, all patients were analyzed together. The survival at three years for 62 patients who had surgery was 47% compared to 17% for the patients who did not have surgery (p = 0.0001) (Fig. 5). Patients with no residual cancer after evaluation of the resected specimen had a survival rate of 76% which was significantly better (a = 0.000 1) than those with residual tumor (Fig. 6). Patterns 0Jfailure Seventy-three percent of the patients who were to fail, did so by the first year of treatment and 93% of the patients failed by two years. Eighty-one patients have recurred with the sites of first failure being local only in 2 1 (26%): distant
only in 45 (56%); both distant and local in 15 (18%). Thus. local failure alone occurred in 26% of the patients who failed, and as a component of overall failure, occurred in 44% of the patients. Distant failure as a component occurred in 74% of all failures. Brain only as the first site of failure was seen in 10 patients or 12% of all failures. The pattern of first failure was analyzed according to several variables as shown in Table 4. There was a trend towards increased failure rates in patients with IIIB disease, patients with nodal involvement, patients who were ineligible for surgery and those who were non-responders. However, of the patients failing, the pattern of failure was similar among all the groups.
Severe toxicity of combined treatment was unusual (Table 5). Alopecia was seen in all patients receiving VP16. Nausea and vomiting were common side effects seen in 8 1% of the patients. However, only 13% of the patients had severe nausea and vomiting. Moderate esophagitis and pneumonitis occurred in 3% and 6% of the patients. respectively. Combined treatment related deaths were seen in three patients, all from sepsis during a period of leucopenia. Two of these patients who expired were nonambulatory (PS 2-3) and had lost more than 5%’of their usual body weight before starting treatment. Because of this lethal toxicity, the starting dose of VP-16 and 5-FU was reduced by 25% in patients who had lost more than 5% of their usual body weight or whose PS was ECOG 2 or 3. There were no deaths attributable to treatment after this change was made. Another six patients had suffered cardiac death and four of these patients were in the ineligible for surgery group. The operative mortality was 5% (3/62). All three patients who died had pneumonectomy. Two post-operative deaths were due to pulmonary emboli and one was caused by pulmonary infection in
iI&
pzor-:“”
0.4 0.2
*
I 0
I
0
250
540
760
rma vso
l500
9750
zmo
2260
2.500
2750
ot-; 0
254
: 750
: 1000
v--t-+,250
-2
1500
,750
2ooo
2250
2500
2750
DAYS
DAYS
Fig. 3. Actuarial survival according to study.
: 500
Fig. 4. Actuarial
survival according
to eligibility for surgery.
21
Lung carcinoma 0 S. REDDY uf al.
Fig. 5. Actuarial survival according to *surgery.
Fig. 6. Actuarial survival according to response.
the remaining lung. Major complications occurred in 14 patients and they were as shown in Table 6.
Surgical resection has been shown to be well-tolerated after preoperative radiation with 20’70five-year survival rates in locally advanced NSCLC ( 17,20). As locoregional failure is a predominant component of overall failure pattern in these patients, surgical resection, whenever possible, was performed as part of the protocol. The overall results in these two Phase II studies support the feasibility of the program with acceptable toxicity. There were three deaths related to the concomitant therapy, early in the course of the study. After appropriate changes were made, no further deaths secondary to treatment occurred. Cardiac deaths in six patients are of concern. Although this patient population may be at a high risk for cardio-vascular disease, a possible association of 5-FU and myocardial injury has been noted (16). The majority of the patients who failed to complete their planned treatment were in the eligible for surgery group and refused to have the operation after preoperative treatment. Progression of disease during the split-course of radiation was a theoretical concern in this study. However, only ten (8%) patients had their disease progression during the treatment, of whom only one patient had progression locally alone. This is lower than that observed in patients treated with continuous preoperative radiation therapy at our institution (where about 16% have progressed with
DISCUSSION
The majority of patients who have locally advanced NSCLC will succumb to their disease. Studies evaluating the patterns of failure in this group of patients have revealed failure in both local and distant sites to be a significant problem (13, 17, 22). Improvements on these results can be made only with treatment approaches that can not only improve locoregional control, but also be more effective against micrometastasis. The optimal sequence for chemotherapy and radiation in Stage III NSCLC has not been demonstrated. Using chemotherapy early in the course of treatment provides an opportunity to treat micrometastasis and to deliver the agents at full dose intensity. Since Cisplatin-Etoposide containing regimens have been demonstrated to be the most effective in the management of NSCLC patients, and with 5-FU demonstrated to have radiation sensitizing properties, we have initiated the Phase II studies incorporating all these rationale in the management of patients with Stage III NSCLC (2, 11). Because of the concomitant treatment, it was necessary to use split-course therapy to avoid any undue morbidity and mortality in this group of patients.
Table 4. Patterns of failure % of all failures
Variable Stage IIIA BIB Study I II Eligibility for surgery Yes No Nodal status NU N+ Path response PO P+
No.
%
failure
failure
Local only
Distant only
Both
48 33
58 12
25 27
63 45
12 27
41 40
65 61
27 2s
61 50
12 25
49 32
59 IO
24 28
65 41
10 31
31 50
53 72
3s 20
52 58
13 22
3 30
18 67
33 23
67 63
13
22
1. J. Radiation Oncology 0 Biology 0 Physics
more than half doing so locoregionally) ( 17), or that seen in patients treated with radiation alone (13). Since all patients were followed for a minimum of three years after treatment and 93% of the recurrences appeared within two years, we feel that the recurrence-free survival rates will flatten at the level of 32% seen at three years. This compares favorably with other recently published reports of combined modality treatments or the best subgroup of patients treated with hyperfractionated radiation (3, 5, 18). If indeed this treatment modality is superior to other modalities needs to be tested in a randomized study. In spite of having patients with worse prognostic factors in Study II, the survival rates were identical in both the studies suggesting increased efficacy of the treatment by the addition of VP-16. Although at variance with that reported by Curran et al. (4), the survival rate for IIIA patients was significantly better than that seen in IIIB patients, indicating the usefulness of the current AJCC staging system. The rationale for preoperative radiation is to increase the resectability of gross tumor, to sterilize the microscopic disease, and to reduce the number of cells capable of implantation at the time of surgery. Several studies have demonstrated resectability rates ranging from 50-72% with the usage of preoperative radiation, which is comparable to the 75% resectability rate observed in our group of patients having more locally advanced disease (20,24). The operative mortality rate of 5% was also minimal although resection was found to be difficult because of
Table 5. Toxicity of treatment Toxicity Nausea and vomiting Mild Moderate Severe Esophagitis Mild Moderate Pneumonitis Mild Moderate Alopecia Sepsis
% of patients
34 34 13 II 3 2 6 In all patients receiving VP- 16 2
Volume 24, Number I, 1992 Table 6. Surgical results
Operative mortality Resection Complications Fistula Tracheotomy Atelectasis Pneumonitis Stroke
.
No.
%
3/62 60162 14 4 2 5 2
5 97 22
1
marked fibrosis around the hila and fascial, and vascular planes were usually obliterated. There were four bronchial fistulae after resection, three of them occurring after pneumonectomy. Depending on the total dose of radiation delivered preoperatively, other authors have reported a range of 20% to 54% of the patients having no persistent tumor or only having microscopic disease (20,24, 25). Our results, documenting 27% of the patients having no demonstrable tumor in the resected specimen, especially considering that all the patients in our study had Stage III disease are encouraging. This was higher than 16% seen in similar stage patients treated at our institution with preoperative radiation only ( 17). This might be of importance especially with the results indicating that patients with no residual disease in the resected specimen have a significantly better survival than those with residual disease. If this is merely a reflection of the extent of the disease, it needs to be tested in a larger group of patients than we had. Most of the patients whose disease recurred did so within two years of their diagnosis as found by others (13). With local failure as a component of overall failures occurring in 44% of the patients and distant failure in 74% of the patients, we feel that future studies should continue to strive for better local and distant control rates. Patients in the eligible for surgery group have fared significantly better as to survival than those in the ineligible for surgery group. Although it is possible that selection factors other than surgical procedures could have accounted for the differences, we feel that surgery, if possible, should remain as a component in the overall management of patients with locally advanced NSCLC since local failure is a significant component of overall failures. Surgical resection has been reported to reduce the local recurrence (9).
REFERENCES Byfield, J. E.; Stanton, W.; Sharp, T. R.; Frankel, S. S. Koziol, J. Phase I-II study of 120-hour infused 5-FU and split course radiation therapy in localized non-small cell lung cancer. Cancer Treat. Rep. 67: 933-935; 1983. Byfield, J. E.; Calabro-Jones, P. M., Klisak, I.; Kulhanian, F. Pharmacologic requirements for obtaining sensitization of human tumor cells in vitro to combined 5-fluorouracil
or florafur and x-rays. Int. J. Radiat. Oncol. Biol. Phys. 8: 1923-1933; 1982. 3. Cox, J. D.; Azarnis, N.; Byhardt, R. W.; Shin, K. H.; Emami, B.; Pajak, T. F.: A randomized Phase I/II trial of hyperfractionated radiation therapy with total doses of 60.0 Gy to 79.2 Gy: possible survival benefit with 269.6 Gy in favorable patients with radiation therapy oncology group stage
Lung carcinoma 0 S. REDDY d al.
4.
5.
6.
I.
8.
9.
10.
II.
12.
13.
14.
III non-small cell lung carcinoma: Report of Radiation Therapy Oncology Group 83-l 1. J. Clint. Oncol. 8: 15431555; 1990. Curran, J. R.; Stafford, W. J. Lack of apparent difference in outcome between clinically staged IIIA and IIIB nonsmall cell lung cancer treated with radiation therapy. J. Clin. Oncol. 8: 409-4 15; 1990. Dillman, R. 0.; Seagren, S. L.; Propert, K. J.: Guerra, J.; Eaton, W. L.; Frei, E. F.; Perry, M. C.; Carey, R. W.; Frei III, E. F.; and Green, M. R. Protochemotherapy improves survival in regional non-small cell lung cancer. N. Engl. J. Med. 323: 14: 940-945; 1990. Douple, E. B.; Richmond, R. C. Platinum complexes as radiosensitizers of hypoxic mammalian cells. Br. J. Cancer 37: 98-102; 1978. Finkelstein, D. M.; Ettinger, D. S.; Ruckdeschel, J. C. Longterm survivors in metastatic non-small-cell lung cancer: an Eastern Cooperative Oncology Group study. J. Clin. Oncol. 4: 702-709; 1986. Kaplan, E. L.; Meier, P. Non-parametric estimation from incomplete observations. J. Am. Stat, Assoc. 53: 457-48 1; 1958. Katz, H. R. The effect of resection on local failure in irradiated non-oat cell carcinoma ofthe lung. Int. J. Rad. Oncol. Biol. Phys. 9: 1793-1805; 1983. Komaki, R.; Cox, J. D.; Hartz, A. J.; Byhardt, R. W.; PerezTamayo, C.; Clowry, L.; Choi, H.; Wilson, F.; Lopes da Conceicao, A.; Rangala, N. Characteristics of long-term survivors after treatment for inoperable carcinoma of the lung. Am. J. Clin. Oncol. 8: 362-370; 1985. Longeval, E.; Klastersky, J. Combination chemotherapy with cisplatin and etoposide in bronchogenic squamous cell carcinoma and adenocarcinoma. Cancer 50: 275 l-2756; 1982. Murthy, A. K.; Taylor, IV, S. G.; Showel, J.; Calderarelli, D. C.; Hutchinson Jr., J. C.; Holinger, L. D.; Witt, T. R.; Hoover, S. Treatment of advanced head and neck cancer with concomitant radiation and chemotherapy. lnt. J. Radiat. Oncol. Biol. Phys. 13: 01-07; 1987. Perez, C. A.; Pajak, T. F.; Rubin, P.; Simpson, J. R.; Mohiuddin, M.; Brady, L. W.; Perez-Tamayo, R.; Rotman, M. Long-term observations of the patterns of failure in patients with unresectable non-oat cell carcinoma of the lung treated with definitive radiotherapy. Report by the Radiation Therapy Oncology Group. Cancer. 59: 1874- 188 I ; 1987. Perez, C. A.; Stanley, K.; Rubin, P.; Kramer, S.; Brady, L.; Perez-Tamayo, R.; Brown, G. S.; Concannon, J.; Rotman,
15.
16. 17.
18.
19.
20.
21.
23
M.; Seydel, H. G. A prospective randomized study of various irradiation doses and fractionation schedules in the treatment of inoperable non-oat cell carcinoma of the lung. Cancer 45: 2744-2753; 1980. Pincus, M.; Reddy, S.; Lee, M. S.; Bonomi, P.; Taylor, S.; Rowland, K.; Faber, L. P.; Warren. W.; Kittle, C. F.; Hendrickson, F. R. Preoperative combined modality therapy for stage III M non-small cell lung carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 15: 189-195, 1988. Pottage, A.; Holt, S.; Ludgate, S.; Langlands, A. 0.; Fluorouracil cardio toxicity. Br. Med. J. 1: 547-548; 1978. Reddy, S.; Faber, L. P.; Baumann, L. A.; Lee, M. S.; Jensik, R. J.; Kittle, F. C.; Bonomi, P. D.; Taylor IV, S. G.; Hendrickson, F. R. Preoperative radiation therapy in regionally localized stage III non-small cell lung carcinoma: long-term results and patterns of failure. Int. J. Radiat. Oncol. Biol. Phys. 19: 287-292; 1990. Robinow, J. S.; Shaw, E. G.; Eagan, R. T.; Lee, R. E.: Creagan. E. T.; Frytak, S.; Richardson, R. L.; Jett, J. R.; Burton, J. K.; Ilstrup, D. M. Results of combination chemotherapy and thoracic radiation therapy for unresectable non-small cell carcinoma of the lung. lnt. J. Radiat. Oncol. Biol. Phys. 17: 1203-1210; 1989. Schabel. F. M. Jr.; Trader, M. W.; Laster, W. R. Cisdichiorodiammineplatinum, II. Combination of chemotherapy and cross-resistance studies with tumors of mice. Cancer Treat. Rep. 63: 1459-1473; 1979. Sherman, D. M.; Neptune, W.; Weichselbaum, R.: Order, S. E.; Piro, A. J. An aggressive approach to marginally resectable lung cancer. Cancer. 41: 2040-2045; 1978. Silverberg, E.; Boring, C. C.; Squires. T. S. Cancer statistics, Ca-A Cancer Journal for Clinicians. 40: 9-26; 1990.
22. Stanley, K.: Cox, J. D.; Patrovich. Z.; Paig, C. Patterns of failure in patients with inoperable carcinoma of the lung. Cancer. 47: 2725-2729; 198 I. 23. Taylor IV, S. G.; Trybula, M.; Bonomi, P. D.; Faber, L. P.; Lee, M. S.: Reddy, S.: Maffey, S. C.: Mathisen, D. J.; Jensik, R. J.; Kittle, C. F. Simultaneous cisplatin fluorouracil infusion and radiation followed by surgical resection in regionally localized stage III. non-small cell lung cancer. Ann. Thorac. Surg. 43: 87-91; 1987. 24. Warram, J. Preoperative irradiation of cancer of the lung: final report of a therapeutic trial. A collaborative study. Cancer 36: 914-925; 1975. 25. Widow, W. Preoperative irradiation of bronchial carcinoma. Cancer 28: 798-80 I; 197 I.
