Adjuvant Chemotherapy After Radical Surgery for Non-Small-Cell Lung Cancer: A Randomized Study By A. Niiranen, S. Niitamo-Korhonen, M. Kouri, A. Assendelft, K. Mattson, and S. Pyrh6nen Purpose: The aims of this study were to assess the effect of adjuvant chemotherapy on overall survival, disease-free survival, and relapse pattern, as well as its toxicity in patients who underwent radical surgery for non-small-cell lung cancer (NSCLC). Patients and Methods: One hundred ten patients with T1-3N0 (World Health Organization [WHO] 1981) NSCLC underwent radical surgery during the period of 1982 through 1987. After surgery, the patients were randomized to receive adjuvant chemotherapy (n = 54) (cyclophosphamide 400 mg/m 2, doxorubicin 40 mg/m2, and 2 [CAP] for six cycles) or no active cisplatin 40 mg/m treatment (n = 56). Results: After 10 years from the start of the study, 61% of patients were alive in the chemotherapy group and 48%were alive in the control group (P = .050). Seventeen
patients (31%) in the CAP group and 27 patients (48%) in the control group had a recurrence during the follow-up period (P = .01). The 5-year survival rate was 67% in the chemotherapy group and was 56% in the control group (P = .050). The patients in the chemotherapy group who completed the planned treatment had a slightly better 5-year survival than those whose chemotherapy was discontinued (72.5% v 50.3%; P = .15). Chemotherapyrelated gastrointestinal toxicity grade 3 to 4 (WHO) occurred in 63% and was the main reason why patients refused further planned therapy. Conclusion: Our results suggest that patients with NSCLC at pathologic stage I who have undergone radical surgery benefit from adjuvant chemotherapy. J Clin Oncol 10:1927-1932. o 1992 by American Society of Clinical Oncology.
HERE IS GENERAL agreement that radical sur-
(World Health Organization [WHO] 1981) disease, who after radical surgery were randomized to receive adju-
T
gery is the therapy of choice for patients with stage I or II non-small-cell lung cancer (NSCLC).1,2 Provided that occult mediastinal lymph node involvement can be
vant chemotherapy or no adjuvant treatment. The objectives of the study were to compare overall survival, DFS,
excluded by careful preoperative staging, patients with
and the patterns of recurrence, and to determine the
T1, NO lesions, irrespective of histologic subtype, have
toxicity of the adjuvant chemotherapy.
an 80% chance of being alive 5 years after surgery. 3 ,4 In patients with T2, NO lesions, 5-year survival rates of 50% to 60% can be expected after curative resection, 4 and in most institutions these patients do not receive adjuvant therapy. 5 Patients with nonepidermoid carcinoma, however, have a much lower probability of survival than do
patients with epidermoid carcinoma. 6 However, the outcome of surgical treatment alone is jeopardized by the presence of occult local, regional, or distant disease that cannot be demonstrated at the time of diagnosis and surgery. Surgical failure in these instances takes the form of either local relapses or distant metastases.
Therefore, there is a rationale for adjuvant therapies both local and systemic. The use of radiotherapy as an adjuvant to surgery is based on its potential for control of local and regional occult residual disease. When either given preoperatively7 ,8 or postoperatively, 9 -11 irradiation has not improved survival in randomized stud78 ies. .
PATIENTS AND METHODS Patients During 1982 to 1987, 110 patients underwent radical surgery for bronchogenic NSCLC at the Helsinki University Central Hospital. The tumor-node-metastasis (TNM) classification of the American Joint Committee for Cancer Staging and End Results Reporting (AJC) of 1981 was used for pathologic staging. Only patients with a definite diagnosis of NSCLC by histologic examination and at pathologic stage T1-3NO (according to mediastinal dissection and sampling by surgery) MO were accepted for the study. A performance status of 60% or more according to the Karnofsky performance scale was required. Patients with a previous history of cancer (except basal cell skin cancer or carcinoma in situ of the uterine cervix) were excluded, as were patients older than 70 years of age. Cases were recorded according to tumor histology, TNM classification, and performance status before surgery. Four to 6 weeks after radical surgery, patients who fulfilled the entry criteria were assigned randomly to the adjuvant chemotherapy arm or the control arm.
During the past 10 years, there has been a resurgence of interest in preoperative or postoperative chemother-
apy in patients with NSCLC. The benefit of postoperative adjuvant chemotherapy has been apparent in patients with locally more advanced, although still resectable disease, and chemotherapy is effective in prolonging disease-free survival (DFS) time.
12
,13 Here
we report a study on patients with NSCLC, T1-T3NOMO
From the Departmentsof Radiotherapy and Oncology, and Pulmonary Medicine, Helsinki University CentralHospital, and the Department of PulmonaryDiseases,North Karelia CentralHospital,Finland. Submitted February24, 1992; acceptedJuly 30, 1992. Address reprintrequests to A. Niiranen, MD, Pdivdtie 5 A 2, 02210 Espoo, Finland. © 1992 by American Society of ClinicalOncology. 0732-183X/92/1012-0015$3.00/0
Journal of Clinical Oncology, Vol 10, No 12 (December), 1992: pp 1927-1932
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1927
1928
NIIRANEN ET AL
Treatment
Table 1. Clinical Characteristics of 110 Patients 2
Chemotherapy consisted of cyclophosphamide 400 mg/m , doxorubicin 40 mg/m 2, and cisplatin 40 mg/m 2 (CAP), which was repeated every 4 weeks for a total of six cycles. Evaluation All patients underwent general preoperative evaluation that included medical and physical examination, blood chemistry, radiologic examination (recent chest radiographs, conventional tomography or computed tomographic [CT] scan of primary tumor, hilum, and mediastinum), bronchoscopy, lung function tests, radionuclide bone scan, and liver scan or ultrasound scan of the upper abdomen. Brain scan was done only if clinically indicated. Patients were evaluated monthly for 6 months after surgery, then at 3-month intervals for 24 months, and subsequently at 6-month intervals until death or until 5 years had elapsed since surgery. Toxicity of chemotherapy was scored according to WHO criteria. The patient's performance status was recorded before surgery, at the time of randomization, and at each follow-up visit. StatisticalMethods Survival was calculated from the date of surgery. Actuarial survival was determined by the product-limit Kaplan-Meier method. Comparison of survival curves was made by the Mantel-Haenszel (log-rank) test. For comparison of intergroup differences, the x2 test was used. P values of less than .05 were considered statistically significant. RESULTS Balanceof Randomization Of 110 patients, 54 were randomized to the CAP arm and 56 to the control arm; there were 44 men in the CAP arm, 46 men in the control arm, and 10 women in each arm. For detailed patient characteristics, see Table 1. The two groups were well balanced in regard to stage of disease. Forty-nine patients in the CAP arm and 50 in the control arm had stage I disease. No one in the CAP arm and two patients in the control arm had stage II; five in the CAP arm and four in the control arm had stage III disease. Epidermoid carcinoma was the most frequent histologic diagnosis in 55% (62 of 110), followed by large-cell anaplastic carcinoma in 19%, adenocarcinoma in 18%, and bronchioalveolar carcinoma in 7%. The type of surgery for each patient is listed in Table 1; 56% (62 of 110) underwent lobectomy, 44% (48 of 110) required more extensive resection, 30% (33 of 110) underwent pneumonectomy, and 10% (11 of 110) had bilobectomy. Only 4% (four of 110) underwent surgery with sleeve resection. Compliance Of the 54 patients who were randomized into the CAP arm, seven received no chemotherapy (four refused, two had surgical complications that contraindicated chemo-
No. of patients Age, years Median Range Sex, male/female Karnofsky performance status (%) 100 90 80 < 80
CAP
Control
54
56
57 41-69 44/10
59 35-70 46/10
15 17 15 7
11 24 10 11
49 0 5
50 2 4
19 31 4
15 37 4
29
33
7 11
14 9
4
3
2 35 6 11
2 27 5 22
Stage, pathologic II III Tumor category 1 2 3 Cell type Epidermoid carcinoma (n = 62) Large-cell carcinoma (n = 21) Adenocarcinoma (n = 20) Bronchioalveolar carcinoma (n = 7) Extent of surgery Sleeve resection Lobectomy Bilobectomy Pneumonectomy
therapy, and one had an acute myocardial infarction before initiation of chemotherapy). Sixteen patients in the CAP group refused chemotherapy at some phase of their treatment mainly because of chemotherapyinduced nausea and vomiting; 31 of the 54 patients received their planned treatment and 38 of 54 received at least three cycles of chemotherapy. DFS and Relapse Pattern Seventeen patients (31%) in the CAP group and 27 patients in the control group (48%) had a recurrence during the follow-up (P = .01). Median disease-free periods in relapsed patients were 17 months (range, 2.5 to 54) in the CAP arm and 16 months (range, 2 to 71) for controls (Fig 1). No relapse was recorded in 36 patients (67%) in the CAP group and 28 (50%) in the control group. There were no data in regard to relapse for one patient in each group. No significant difference was observed in relapse pattern between the two groups. In both groups, local relapses were observed in 35%, and distant relapses were observed in 65%.
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1929
POSTOPERATIVE CHEMOTHERAPY IN NSCLC
C, 0.
C
4--
4-,
19. CL ci, a
Years Fig 1. DFS curves for patients by treatment 0, controls (n = 56). P = .01, log-rank test.
Survival Figure 2 shows the overall surviva I curves. The 5-year survival rate that was calculated fro m surgery was 67% in the CAP arm and 56% in the conttrol arm (P = .050). If deaths caused by cardiac disease were excluded, the difference was statistically more clea r (P = .019). At the time of this analysis (10 years fro m the start of the study), 61% were alive in the CAP g roup and 48%awere alive in the control group. All random included in the survival analyses regmized patients where
their planned treatment was completed or discontinued. Patients with T1 disease survived significantly longer than the others (P = .004), and for patients with T2 disease, survival time was significantly longer among those who received chemotherapy than among the controls (P = .002). CAP patients who completed their planned treatment survived longer than those whose CAP therapy was discontinued, but this difference was not statistically significant. Patients who underwent pneumonectomy had a poorer outcome than those with smaller operations (P = .002), a difference also not explained by tumor stage. The same trend appeared among patients at T1, T2, or T3 who underwent pneumonectomy. In all, 68% of patients (22 of 32) who underwent pneumonectomy had T2 disease, 18% (six of 32) had T1 disease, and 15% (five of 32) had T3 disease. These patients had a significantly worse prognosis than did other patients (P = .001). When survival was analyzed separately in the two main surgical groups, there was no difference in survival between the two treatment arms among pneumonectomy patients; 73% of these patients died during the follow-up period (seven of 10 in the CAP group and 16 of 22 in the control group). When the patients with smaller operations (lobectomy, bilobectomy, or resection) were analyzed separately, the difference in survival between the CAP group and the control group was rendered statistically nonsignificant. The 5-year survival rate of patients with smaller operations was 73.5% in the CAP group and 63.7% in the control group (P = .24). There was no significant difference in 5-year survival rt*
1
hptwppn
ntiritr
with
pnLdprmnld
rr-innmI
dl,_
nocarcinoma, and large-cell anaplastic carcinoma (63%, 50%, and 63%, respectively). The seven patients with bronchioalveolar carcinoma proved to have a better prognosis than the rest (85% alive after 5 years). There was no significant difference in survival by sex, age, or performance status. Fifteen of the 21 deaths in the CAP group were associated with carcinoma, five were caused by acute myocardial infarction or cardiac failure, and one was caused by pulmonary embolism. Of the 29 deaths in the control group, 26 were caused by carcinoma. One was caused by acute myocardial infarction, and two were caused by cardiac failure. The cardiac deaths occurred 1, 5, 6, 7, and 8 years after chemotherapy in the CAP group, and these intervals were 1, 5, and 7 years for the
oC
ci C-, C. U)
ci aU
Scontrols
Years Fig 2. Survival curves for patients by treeatment group. 0, CAP (n = 54); 0, controls, (n = 56). P = .05, log-n ank test.
Among patients who had undergone pneumonectomy, 19 died from cancer, seven in the CAP group and
12 in the control group. In the CAP group, among
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1930
NIIRANEN ET AL
patients with smaller operations, eight patients died from carcinoma, but five died as a result of some cardiopulmonary event. In the control group, the same figures were 12 and one, respectively. Toxicity Hematologic toxicity was not a problem and was observed in only 12 patients. Three patients had grade 3 leukopenia (WHO), five had grade 2 leukopenia, and two had grade 1 leukopenia. Thrombopenia was observed in four patients (grade 1). Nonhematologic side effects occurred in 38 patients who received CAP (Table 2). Nausea and vomiting were the most disturbing side effects. Twenty-four (63%) of the 38 patients with nonhematologic side effects experienced grade 3 to 4 (WHO) nausea and vomiting, and six patients had grade 2 nausea and vomiting. Twelve patients (22%) in the CAP arm refused chemotherapy because of nausea and vomiting. Six patients discontinued their chemotherapy for some other reason; three patients suffered grade 3 renal toxicity and three had grade 3 cardiac toxicity. Nausea did not correlate with weight loss during chemotherapy. To the contrary, there was slightly more weight loss among patients in the control group than in the CAP group: during the first 6 months, 20% in the control group, but only 12% in the CAP group, experienced a loss of 5% or more weight. DISCUSSION Surgical treatment is curative in a small proportion of patients with NSCLC. It is well known that a considerable proportion of NSCLC patients with apparently the best prognostic factors (small primary tumor, no lymph node involvement, complete resection) develop disseminating and fatal disease shortly after surgery because of micrometastases that were undetectable at surgery. Thus, there is a rationale for systemic adjuvant therapy. However, the results of adjuvant studies are controversial. In most studies, patients have had clinical stage III 9 14 15 disease. , ,
Table 2. Nonhematologic Toxicity (N = 38) WHO Grade 1
Type of
2
3
4
Toxicity
No.
%
No.
%
No.
%
No.
%
Nausea/vomiting Cardiac Renal Neurologic
8 0 0 0
21 0 0 0
6 1 3 2
16 3 8 5
13 2 2 1
34 5 5 3
11 3 0 0
29 8 0 0
Two different randomized trials by the Lung Cancer Study Group showed adjuvant CAP chemotherapy to be effective in the prolongation of the DFS of stage III patients. 16 Ferguson et a117 and Newman et al18 reported that in patients with T1-T2N1 disease resection in combination with adjuvant radiation therapy and chemotherapy improved the median survival. Controversial results by Martini et a119 in patients with T1N1 disease showed no improvement in survival with the use of adjuvant treatment. In the present study of patients with stage T1-T3NO disease, control of micrometastases was attempted with CAP. In the present study, patient selection was more stringent than in previous adjuvant studies because all patients with lymph node involvement at pathologic staging were excluded. However, we had two patients with stage II and nine patients with stage III disease in our study because of our criteria for exclusion. The CAP group and the control group were well balanced with regard to well-known prognostic factors. However, during the analysis a slight imbalance appeared in the extent of surgery between groups; more patients in the control arm were treated with pneumonectomy. Because these patients also had worse prognoses than did patients treated with less radical surgery, one cannot totally exclude the possibility that the extent of surgery could be a confounding factor in the comparison of outcomes. Contrary to other studies, women did not have a better prognosis than men. There is general agreement that recurrence, both regional and distant, is more common in patients with N1 than those with NO disease. 19 This observation is also supported by autopsy studies, which demonstrate that up to 20% to 35% of patients who died within 30 days of a potentially curative resection in fact had persistent disease, both local and distant.2 0,21 In the present study, 39% of all patients relapsed despite the fact that they had NO disease at the time of surgery. However, patients in the CAP arm had significantly fewer relapses than those in the control arm. Therefore, it seems plausible that adjuvant chemotherapy prevented the development of micrometastases possibly present at surgery. No significant differences in relapse pattern could be detected between the two treatment groups. It is noteworthy that contrary to other studies, 22 brain metastases were observed more often (56%) among patients with epidermoid carcinoma than among patients with other cell types. Patients with relapse of large-cell carcinoma were all in the control arm. The 5-year survival rate in stage I lung cancer treated by curative pulmonary resection is approximately 30% to
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1931
POSTOPERATIVE CHEMOTHERAPY IN NSCLC
50%; prognosis for the more advanced stages is worse. 23 In the present study, this low figure held only for the control group, in which the overall 5-year survival rate was 56%. The survival rate was 67% in the CAP group. Because more patients with pneumonectomy were randomized onto the control group, survival was analyzed separately among patients who had undergone smaller operations. There was no difference in survival rate among the pneumonectomy patients between the chemotherapy group and the control group; 72.7% in both groups died in 5 years. Those patients who had undergone lobectomy or bilobectomy had a slightly better prognosis and showed a slightly longer survival rate in the CAP group (73.5% v 63.7%). The situation was even better among those 31 patients in the CAP group who received their planned six cycles of chemotherapy; their 5-year survival rate was better than that of patients in the CAP group who did not receive their planned treatment. Belcher and Anderson 24 pointed out that the results of lobectomy for bronchial carcinoma were at least as good as, and often better than, the results of pneumonectomy. However, these better results were accounted for by the fact that the more restricted operation is only feasible for the more restricted lesions. In the present study there were equal numbers of carcinoma-related deaths in both groups (35% to 41%). When cardiac deaths were compared among patients undergoing pneumonectomy, three deaths (10%) in the control group were noted to have been caused by cardiac-respiratory diseases, whereas in the CAP group, only one patient died from cardiac disease. In contrast, in the CAP group, patients with smaller operations
included five deaths caused by cardiac-respiratory diseases (38%). Thus a statistically significant greater amount of patients in the CAP group died from some cause other than carcinoma (P = .019). Those cardiac deaths usually occurred more than 5 years after chemotherapy. Thus it is difficult to conclude whether, in this study, CAP therapy could have had some influence on cardiac deaths. Patients have died from acute myocardial infarction, and no specific investigations have been performed concerning a possible doxorubicin effect on cardiac muscle. The main problem throughout the chemotherapy was CAP-induced nausea and vomiting that was intractable with the antiemetic drugs in use at the time of the study. Of the patients randomized into the CAP group, 30% refused chemotherapy because of nausea. It also was difficult to motivate some patients to initiate such emetogenic treatment after they were informed that their surgery had been radical and no disease remained. The other side effects of CAP were not remarkable. With the more effective antiemetic drugs of the present day, nausea and vomiting would not prove to be so great a problem. Our results show that adjuvant chemotherapy like that used in the present study deserves further research in radically resected NSCLC. Furthermore, we conclude that patients in such studies should be stratified in groups according to extent of surgery before randomization to avoid the possible confounding effect of surgery that we faced in our trial. To improve compliance in such studies, modern and effective antiemetic treatment must be used.
REFERENCES 1. Mountain CF: The new international staging system for lung cancer. Chest 89:2258-2338, 1986 2. Gralla RJ, Kris MG: Management of non-small cell lung cancer. Recent Adv Clin Oncol 2:167-180, 1986 3. Martini N, Kris MG, Gralla RJ, et al: The effects of preoperative chemotherapy on the resectability of non-small cell lung carcinoma with mediastinal lymph node metastases (N2 MO). Ann Thorac Surg 45:370-379, 1988 4. Williams DE, Pairollera PC, David CS, et al: Survival of patients surgically treated for stage I lung cancer. J Thorac Cardiovasc Surg 82:70-76, 1981 5. Holmes EC: Surgical results and surgical adjuvant therapy for lung cancer. Am J Surg 143:691-693, 1982 6. Mountain CF: Expanded possibilities for surgical treatment of lung cancer. Survival in stage IIIa disease. Chest 97:1045-1051, 1990 7. Shields TW: Preoperative radiation therapy in the treatment of bronchial carcinoma. Cancer 30:1388-1394, 1972 8. Committee for Radiation Therapy Studies at the National
Cancer Institute: Preoperative irradiation of cancer of the lung: Final report of a therapeutic trial. A collaborative study. Cancer 36:914-925, 1975 9. Van Houtte P, Rocmans P, Smets P, et al: Postoperative radiation therapy in lung cancer: A controlled trial after resection of curative design. Int J Radiat Oncol Biol Phys 6:983-986, 1980 10. Choi NC: Role of postoperative radiation therapy in lung cancer with either metastases to regional lymph nodes (N1 or unforeseen N2) or direct invasion beyond visceral pleura (T3), in Choi NC, Grillo HC (eds): Thoracic Oncology. New York, NY, Raven Press, 1983, pp 129-146 11. Choi NC, Grillo HC, Gardiello M, et al: Basis for new strategies in postoperative radiotherapy of bronchogenic carcinoma. Int J Radiat Oncol Biol Phys 6:31-35, 1980 12. Holmes EC: Surgical adjuvant therapy of non-small cell lung cancer. J Surg Oncol 1:26-33, 1989 (suppl) 13. Klastersky J, Feld R, Kleisbauer JP, et al: Treatment of N2 non-small cell lung cancer (NSCLC). Eur J Cancer 26:798-801, 1990
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1932
NIIRANEN ET AL
14. Einhorn LH: Neoadjuvant therapy of stage III non-small cell lung cancer. Ann Thorac Surg 46:362-365, 1988 15. Astudillo J, Conill C: Role of postoperative radiation therapy in stage IIIa non small cell lung cancer. Ann Thorac Surg 50:618-623, 1990 16. Holmes EC: Surgical adjuvant chemotherapy in non-smallcell lung cancer. Semin Oncol 15:255-260, 1988 17. Ferguson MK, Little AG, Golomb HM, et al: The role of adjuvant therapy after resection of T1 N1 MO and T2 N1 MO non-small cell lung cancer. J Thorac Cardiovasc Surg 91:344-349, 1986 18. Newman SB, DeMeester TR, Golomb HM, et al: Treatment of modified stage II (T1 N1 MO, T2, N1 MO) non-small cell bronchogenic carcinoma. A combined modality approach. J Thorac Cardiovasc Surg 8:180-185, 1983 19. Martini N, Flehinger BJ, Nagasaki F, et al: Prognostic
significance of N1 disease in carcinoma of the lung. J Thorac Cardiovasc Surg 86:646-653, 1983 20. Matthews MJ, Kanhouwa S, Pickren J, et al: Frequency of residual and metastatic tumor in patients undergoing curative surgical resection for lung cancer. Cancer Chemother Rep 4:63-67, 1973 21. Naruke T, Suemasu K, Ishikawa S: Lymph node mapping and curability at various levels of metastasis in resected lung cancer. J Thorac Cardiovasc Surg 76:832-839, 1978 22. Figlin RA, Piantadosi S, Feld R: Intracranial recurrence of carcinoma after complete surgical resection of stage I, II and III non-small cell lung cancer. N Engl J Med 318:1300-1305, 1988 23. Lad T: Adjuvant therapy of stage II non-small cell lung cancer, in Bitran JD (ed): Lung-Cancer: A Comprehensive Treatise. Orlando, FL, Grune & Stratton, 1988, pp 173-182 24. Belcher JR, Anderson R: Surgical treatment of carcinoma of the bronchus. Br Med J 1:948-954, 1965
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
patients (31%) in the CAP group and 27 patients (48%) in the control group had a recurrence during the follow-up period (P = .01). The 5-year survival rate was 67% in the chemotherapy group and was 56% in the control group (P = .050). The patients in the chemotherapy group who completed the planned treatment had a slightly better 5-year survival than those whose chemotherapy was discontinued (72.5% v 50.3%; P = .15). Chemotherapyrelated gastrointestinal toxicity grade 3 to 4 (WHO) occurred in 63% and was the main reason why patients refused further planned therapy. Conclusion: Our results suggest that patients with NSCLC at pathologic stage I who have undergone radical surgery benefit from adjuvant chemotherapy. J Clin Oncol 10:1927-1932. o 1992 by American Society of Clinical Oncology.
HERE IS GENERAL agreement that radical sur-
(World Health Organization [WHO] 1981) disease, who after radical surgery were randomized to receive adju-
T
gery is the therapy of choice for patients with stage I or II non-small-cell lung cancer (NSCLC).1,2 Provided that occult mediastinal lymph node involvement can be
vant chemotherapy or no adjuvant treatment. The objectives of the study were to compare overall survival, DFS,
excluded by careful preoperative staging, patients with
and the patterns of recurrence, and to determine the
T1, NO lesions, irrespective of histologic subtype, have
toxicity of the adjuvant chemotherapy.
an 80% chance of being alive 5 years after surgery. 3 ,4 In patients with T2, NO lesions, 5-year survival rates of 50% to 60% can be expected after curative resection, 4 and in most institutions these patients do not receive adjuvant therapy. 5 Patients with nonepidermoid carcinoma, however, have a much lower probability of survival than do
patients with epidermoid carcinoma. 6 However, the outcome of surgical treatment alone is jeopardized by the presence of occult local, regional, or distant disease that cannot be demonstrated at the time of diagnosis and surgery. Surgical failure in these instances takes the form of either local relapses or distant metastases.
Therefore, there is a rationale for adjuvant therapies both local and systemic. The use of radiotherapy as an adjuvant to surgery is based on its potential for control of local and regional occult residual disease. When either given preoperatively7 ,8 or postoperatively, 9 -11 irradiation has not improved survival in randomized stud78 ies. .
PATIENTS AND METHODS Patients During 1982 to 1987, 110 patients underwent radical surgery for bronchogenic NSCLC at the Helsinki University Central Hospital. The tumor-node-metastasis (TNM) classification of the American Joint Committee for Cancer Staging and End Results Reporting (AJC) of 1981 was used for pathologic staging. Only patients with a definite diagnosis of NSCLC by histologic examination and at pathologic stage T1-3NO (according to mediastinal dissection and sampling by surgery) MO were accepted for the study. A performance status of 60% or more according to the Karnofsky performance scale was required. Patients with a previous history of cancer (except basal cell skin cancer or carcinoma in situ of the uterine cervix) were excluded, as were patients older than 70 years of age. Cases were recorded according to tumor histology, TNM classification, and performance status before surgery. Four to 6 weeks after radical surgery, patients who fulfilled the entry criteria were assigned randomly to the adjuvant chemotherapy arm or the control arm.
During the past 10 years, there has been a resurgence of interest in preoperative or postoperative chemother-
apy in patients with NSCLC. The benefit of postoperative adjuvant chemotherapy has been apparent in patients with locally more advanced, although still resectable disease, and chemotherapy is effective in prolonging disease-free survival (DFS) time.
12
,13 Here
we report a study on patients with NSCLC, T1-T3NOMO
From the Departmentsof Radiotherapy and Oncology, and Pulmonary Medicine, Helsinki University CentralHospital, and the Department of PulmonaryDiseases,North Karelia CentralHospital,Finland. Submitted February24, 1992; acceptedJuly 30, 1992. Address reprintrequests to A. Niiranen, MD, Pdivdtie 5 A 2, 02210 Espoo, Finland. © 1992 by American Society of ClinicalOncology. 0732-183X/92/1012-0015$3.00/0
Journal of Clinical Oncology, Vol 10, No 12 (December), 1992: pp 1927-1932
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1927
1928
NIIRANEN ET AL
Treatment
Table 1. Clinical Characteristics of 110 Patients 2
Chemotherapy consisted of cyclophosphamide 400 mg/m , doxorubicin 40 mg/m 2, and cisplatin 40 mg/m 2 (CAP), which was repeated every 4 weeks for a total of six cycles. Evaluation All patients underwent general preoperative evaluation that included medical and physical examination, blood chemistry, radiologic examination (recent chest radiographs, conventional tomography or computed tomographic [CT] scan of primary tumor, hilum, and mediastinum), bronchoscopy, lung function tests, radionuclide bone scan, and liver scan or ultrasound scan of the upper abdomen. Brain scan was done only if clinically indicated. Patients were evaluated monthly for 6 months after surgery, then at 3-month intervals for 24 months, and subsequently at 6-month intervals until death or until 5 years had elapsed since surgery. Toxicity of chemotherapy was scored according to WHO criteria. The patient's performance status was recorded before surgery, at the time of randomization, and at each follow-up visit. StatisticalMethods Survival was calculated from the date of surgery. Actuarial survival was determined by the product-limit Kaplan-Meier method. Comparison of survival curves was made by the Mantel-Haenszel (log-rank) test. For comparison of intergroup differences, the x2 test was used. P values of less than .05 were considered statistically significant. RESULTS Balanceof Randomization Of 110 patients, 54 were randomized to the CAP arm and 56 to the control arm; there were 44 men in the CAP arm, 46 men in the control arm, and 10 women in each arm. For detailed patient characteristics, see Table 1. The two groups were well balanced in regard to stage of disease. Forty-nine patients in the CAP arm and 50 in the control arm had stage I disease. No one in the CAP arm and two patients in the control arm had stage II; five in the CAP arm and four in the control arm had stage III disease. Epidermoid carcinoma was the most frequent histologic diagnosis in 55% (62 of 110), followed by large-cell anaplastic carcinoma in 19%, adenocarcinoma in 18%, and bronchioalveolar carcinoma in 7%. The type of surgery for each patient is listed in Table 1; 56% (62 of 110) underwent lobectomy, 44% (48 of 110) required more extensive resection, 30% (33 of 110) underwent pneumonectomy, and 10% (11 of 110) had bilobectomy. Only 4% (four of 110) underwent surgery with sleeve resection. Compliance Of the 54 patients who were randomized into the CAP arm, seven received no chemotherapy (four refused, two had surgical complications that contraindicated chemo-
No. of patients Age, years Median Range Sex, male/female Karnofsky performance status (%) 100 90 80 < 80
CAP
Control
54
56
57 41-69 44/10
59 35-70 46/10
15 17 15 7
11 24 10 11
49 0 5
50 2 4
19 31 4
15 37 4
29
33
7 11
14 9
4
3
2 35 6 11
2 27 5 22
Stage, pathologic II III Tumor category 1 2 3 Cell type Epidermoid carcinoma (n = 62) Large-cell carcinoma (n = 21) Adenocarcinoma (n = 20) Bronchioalveolar carcinoma (n = 7) Extent of surgery Sleeve resection Lobectomy Bilobectomy Pneumonectomy
therapy, and one had an acute myocardial infarction before initiation of chemotherapy). Sixteen patients in the CAP group refused chemotherapy at some phase of their treatment mainly because of chemotherapyinduced nausea and vomiting; 31 of the 54 patients received their planned treatment and 38 of 54 received at least three cycles of chemotherapy. DFS and Relapse Pattern Seventeen patients (31%) in the CAP group and 27 patients in the control group (48%) had a recurrence during the follow-up (P = .01). Median disease-free periods in relapsed patients were 17 months (range, 2.5 to 54) in the CAP arm and 16 months (range, 2 to 71) for controls (Fig 1). No relapse was recorded in 36 patients (67%) in the CAP group and 28 (50%) in the control group. There were no data in regard to relapse for one patient in each group. No significant difference was observed in relapse pattern between the two groups. In both groups, local relapses were observed in 35%, and distant relapses were observed in 65%.
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1929
POSTOPERATIVE CHEMOTHERAPY IN NSCLC
C, 0.
C
4--
4-,
19. CL ci, a
Years Fig 1. DFS curves for patients by treatment 0, controls (n = 56). P = .01, log-rank test.
Survival Figure 2 shows the overall surviva I curves. The 5-year survival rate that was calculated fro m surgery was 67% in the CAP arm and 56% in the conttrol arm (P = .050). If deaths caused by cardiac disease were excluded, the difference was statistically more clea r (P = .019). At the time of this analysis (10 years fro m the start of the study), 61% were alive in the CAP g roup and 48%awere alive in the control group. All random included in the survival analyses regmized patients where
their planned treatment was completed or discontinued. Patients with T1 disease survived significantly longer than the others (P = .004), and for patients with T2 disease, survival time was significantly longer among those who received chemotherapy than among the controls (P = .002). CAP patients who completed their planned treatment survived longer than those whose CAP therapy was discontinued, but this difference was not statistically significant. Patients who underwent pneumonectomy had a poorer outcome than those with smaller operations (P = .002), a difference also not explained by tumor stage. The same trend appeared among patients at T1, T2, or T3 who underwent pneumonectomy. In all, 68% of patients (22 of 32) who underwent pneumonectomy had T2 disease, 18% (six of 32) had T1 disease, and 15% (five of 32) had T3 disease. These patients had a significantly worse prognosis than did other patients (P = .001). When survival was analyzed separately in the two main surgical groups, there was no difference in survival between the two treatment arms among pneumonectomy patients; 73% of these patients died during the follow-up period (seven of 10 in the CAP group and 16 of 22 in the control group). When the patients with smaller operations (lobectomy, bilobectomy, or resection) were analyzed separately, the difference in survival between the CAP group and the control group was rendered statistically nonsignificant. The 5-year survival rate of patients with smaller operations was 73.5% in the CAP group and 63.7% in the control group (P = .24). There was no significant difference in 5-year survival rt*
1
hptwppn
ntiritr
with
pnLdprmnld
rr-innmI
dl,_
nocarcinoma, and large-cell anaplastic carcinoma (63%, 50%, and 63%, respectively). The seven patients with bronchioalveolar carcinoma proved to have a better prognosis than the rest (85% alive after 5 years). There was no significant difference in survival by sex, age, or performance status. Fifteen of the 21 deaths in the CAP group were associated with carcinoma, five were caused by acute myocardial infarction or cardiac failure, and one was caused by pulmonary embolism. Of the 29 deaths in the control group, 26 were caused by carcinoma. One was caused by acute myocardial infarction, and two were caused by cardiac failure. The cardiac deaths occurred 1, 5, 6, 7, and 8 years after chemotherapy in the CAP group, and these intervals were 1, 5, and 7 years for the
oC
ci C-, C. U)
ci aU
Scontrols
Years Fig 2. Survival curves for patients by treeatment group. 0, CAP (n = 54); 0, controls, (n = 56). P = .05, log-n ank test.
Among patients who had undergone pneumonectomy, 19 died from cancer, seven in the CAP group and
12 in the control group. In the CAP group, among
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1930
NIIRANEN ET AL
patients with smaller operations, eight patients died from carcinoma, but five died as a result of some cardiopulmonary event. In the control group, the same figures were 12 and one, respectively. Toxicity Hematologic toxicity was not a problem and was observed in only 12 patients. Three patients had grade 3 leukopenia (WHO), five had grade 2 leukopenia, and two had grade 1 leukopenia. Thrombopenia was observed in four patients (grade 1). Nonhematologic side effects occurred in 38 patients who received CAP (Table 2). Nausea and vomiting were the most disturbing side effects. Twenty-four (63%) of the 38 patients with nonhematologic side effects experienced grade 3 to 4 (WHO) nausea and vomiting, and six patients had grade 2 nausea and vomiting. Twelve patients (22%) in the CAP arm refused chemotherapy because of nausea and vomiting. Six patients discontinued their chemotherapy for some other reason; three patients suffered grade 3 renal toxicity and three had grade 3 cardiac toxicity. Nausea did not correlate with weight loss during chemotherapy. To the contrary, there was slightly more weight loss among patients in the control group than in the CAP group: during the first 6 months, 20% in the control group, but only 12% in the CAP group, experienced a loss of 5% or more weight. DISCUSSION Surgical treatment is curative in a small proportion of patients with NSCLC. It is well known that a considerable proportion of NSCLC patients with apparently the best prognostic factors (small primary tumor, no lymph node involvement, complete resection) develop disseminating and fatal disease shortly after surgery because of micrometastases that were undetectable at surgery. Thus, there is a rationale for systemic adjuvant therapy. However, the results of adjuvant studies are controversial. In most studies, patients have had clinical stage III 9 14 15 disease. , ,
Table 2. Nonhematologic Toxicity (N = 38) WHO Grade 1
Type of
2
3
4
Toxicity
No.
%
No.
%
No.
%
No.
%
Nausea/vomiting Cardiac Renal Neurologic
8 0 0 0
21 0 0 0
6 1 3 2
16 3 8 5
13 2 2 1
34 5 5 3
11 3 0 0
29 8 0 0
Two different randomized trials by the Lung Cancer Study Group showed adjuvant CAP chemotherapy to be effective in the prolongation of the DFS of stage III patients. 16 Ferguson et a117 and Newman et al18 reported that in patients with T1-T2N1 disease resection in combination with adjuvant radiation therapy and chemotherapy improved the median survival. Controversial results by Martini et a119 in patients with T1N1 disease showed no improvement in survival with the use of adjuvant treatment. In the present study of patients with stage T1-T3NO disease, control of micrometastases was attempted with CAP. In the present study, patient selection was more stringent than in previous adjuvant studies because all patients with lymph node involvement at pathologic staging were excluded. However, we had two patients with stage II and nine patients with stage III disease in our study because of our criteria for exclusion. The CAP group and the control group were well balanced with regard to well-known prognostic factors. However, during the analysis a slight imbalance appeared in the extent of surgery between groups; more patients in the control arm were treated with pneumonectomy. Because these patients also had worse prognoses than did patients treated with less radical surgery, one cannot totally exclude the possibility that the extent of surgery could be a confounding factor in the comparison of outcomes. Contrary to other studies, women did not have a better prognosis than men. There is general agreement that recurrence, both regional and distant, is more common in patients with N1 than those with NO disease. 19 This observation is also supported by autopsy studies, which demonstrate that up to 20% to 35% of patients who died within 30 days of a potentially curative resection in fact had persistent disease, both local and distant.2 0,21 In the present study, 39% of all patients relapsed despite the fact that they had NO disease at the time of surgery. However, patients in the CAP arm had significantly fewer relapses than those in the control arm. Therefore, it seems plausible that adjuvant chemotherapy prevented the development of micrometastases possibly present at surgery. No significant differences in relapse pattern could be detected between the two treatment groups. It is noteworthy that contrary to other studies, 22 brain metastases were observed more often (56%) among patients with epidermoid carcinoma than among patients with other cell types. Patients with relapse of large-cell carcinoma were all in the control arm. The 5-year survival rate in stage I lung cancer treated by curative pulmonary resection is approximately 30% to
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1931
POSTOPERATIVE CHEMOTHERAPY IN NSCLC
50%; prognosis for the more advanced stages is worse. 23 In the present study, this low figure held only for the control group, in which the overall 5-year survival rate was 56%. The survival rate was 67% in the CAP group. Because more patients with pneumonectomy were randomized onto the control group, survival was analyzed separately among patients who had undergone smaller operations. There was no difference in survival rate among the pneumonectomy patients between the chemotherapy group and the control group; 72.7% in both groups died in 5 years. Those patients who had undergone lobectomy or bilobectomy had a slightly better prognosis and showed a slightly longer survival rate in the CAP group (73.5% v 63.7%). The situation was even better among those 31 patients in the CAP group who received their planned six cycles of chemotherapy; their 5-year survival rate was better than that of patients in the CAP group who did not receive their planned treatment. Belcher and Anderson 24 pointed out that the results of lobectomy for bronchial carcinoma were at least as good as, and often better than, the results of pneumonectomy. However, these better results were accounted for by the fact that the more restricted operation is only feasible for the more restricted lesions. In the present study there were equal numbers of carcinoma-related deaths in both groups (35% to 41%). When cardiac deaths were compared among patients undergoing pneumonectomy, three deaths (10%) in the control group were noted to have been caused by cardiac-respiratory diseases, whereas in the CAP group, only one patient died from cardiac disease. In contrast, in the CAP group, patients with smaller operations
included five deaths caused by cardiac-respiratory diseases (38%). Thus a statistically significant greater amount of patients in the CAP group died from some cause other than carcinoma (P = .019). Those cardiac deaths usually occurred more than 5 years after chemotherapy. Thus it is difficult to conclude whether, in this study, CAP therapy could have had some influence on cardiac deaths. Patients have died from acute myocardial infarction, and no specific investigations have been performed concerning a possible doxorubicin effect on cardiac muscle. The main problem throughout the chemotherapy was CAP-induced nausea and vomiting that was intractable with the antiemetic drugs in use at the time of the study. Of the patients randomized into the CAP group, 30% refused chemotherapy because of nausea. It also was difficult to motivate some patients to initiate such emetogenic treatment after they were informed that their surgery had been radical and no disease remained. The other side effects of CAP were not remarkable. With the more effective antiemetic drugs of the present day, nausea and vomiting would not prove to be so great a problem. Our results show that adjuvant chemotherapy like that used in the present study deserves further research in radically resected NSCLC. Furthermore, we conclude that patients in such studies should be stratified in groups according to extent of surgery before randomization to avoid the possible confounding effect of surgery that we faced in our trial. To improve compliance in such studies, modern and effective antiemetic treatment must be used.
REFERENCES 1. Mountain CF: The new international staging system for lung cancer. Chest 89:2258-2338, 1986 2. Gralla RJ, Kris MG: Management of non-small cell lung cancer. Recent Adv Clin Oncol 2:167-180, 1986 3. Martini N, Kris MG, Gralla RJ, et al: The effects of preoperative chemotherapy on the resectability of non-small cell lung carcinoma with mediastinal lymph node metastases (N2 MO). Ann Thorac Surg 45:370-379, 1988 4. Williams DE, Pairollera PC, David CS, et al: Survival of patients surgically treated for stage I lung cancer. J Thorac Cardiovasc Surg 82:70-76, 1981 5. Holmes EC: Surgical results and surgical adjuvant therapy for lung cancer. Am J Surg 143:691-693, 1982 6. Mountain CF: Expanded possibilities for surgical treatment of lung cancer. Survival in stage IIIa disease. Chest 97:1045-1051, 1990 7. Shields TW: Preoperative radiation therapy in the treatment of bronchial carcinoma. Cancer 30:1388-1394, 1972 8. Committee for Radiation Therapy Studies at the National
Cancer Institute: Preoperative irradiation of cancer of the lung: Final report of a therapeutic trial. A collaborative study. Cancer 36:914-925, 1975 9. Van Houtte P, Rocmans P, Smets P, et al: Postoperative radiation therapy in lung cancer: A controlled trial after resection of curative design. Int J Radiat Oncol Biol Phys 6:983-986, 1980 10. Choi NC: Role of postoperative radiation therapy in lung cancer with either metastases to regional lymph nodes (N1 or unforeseen N2) or direct invasion beyond visceral pleura (T3), in Choi NC, Grillo HC (eds): Thoracic Oncology. New York, NY, Raven Press, 1983, pp 129-146 11. Choi NC, Grillo HC, Gardiello M, et al: Basis for new strategies in postoperative radiotherapy of bronchogenic carcinoma. Int J Radiat Oncol Biol Phys 6:31-35, 1980 12. Holmes EC: Surgical adjuvant therapy of non-small cell lung cancer. J Surg Oncol 1:26-33, 1989 (suppl) 13. Klastersky J, Feld R, Kleisbauer JP, et al: Treatment of N2 non-small cell lung cancer (NSCLC). Eur J Cancer 26:798-801, 1990
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
1932
NIIRANEN ET AL
14. Einhorn LH: Neoadjuvant therapy of stage III non-small cell lung cancer. Ann Thorac Surg 46:362-365, 1988 15. Astudillo J, Conill C: Role of postoperative radiation therapy in stage IIIa non small cell lung cancer. Ann Thorac Surg 50:618-623, 1990 16. Holmes EC: Surgical adjuvant chemotherapy in non-smallcell lung cancer. Semin Oncol 15:255-260, 1988 17. Ferguson MK, Little AG, Golomb HM, et al: The role of adjuvant therapy after resection of T1 N1 MO and T2 N1 MO non-small cell lung cancer. J Thorac Cardiovasc Surg 91:344-349, 1986 18. Newman SB, DeMeester TR, Golomb HM, et al: Treatment of modified stage II (T1 N1 MO, T2, N1 MO) non-small cell bronchogenic carcinoma. A combined modality approach. J Thorac Cardiovasc Surg 8:180-185, 1983 19. Martini N, Flehinger BJ, Nagasaki F, et al: Prognostic
significance of N1 disease in carcinoma of the lung. J Thorac Cardiovasc Surg 86:646-653, 1983 20. Matthews MJ, Kanhouwa S, Pickren J, et al: Frequency of residual and metastatic tumor in patients undergoing curative surgical resection for lung cancer. Cancer Chemother Rep 4:63-67, 1973 21. Naruke T, Suemasu K, Ishikawa S: Lymph node mapping and curability at various levels of metastasis in resected lung cancer. J Thorac Cardiovasc Surg 76:832-839, 1978 22. Figlin RA, Piantadosi S, Feld R: Intracranial recurrence of carcinoma after complete surgical resection of stage I, II and III non-small cell lung cancer. N Engl J Med 318:1300-1305, 1988 23. Lad T: Adjuvant therapy of stage II non-small cell lung cancer, in Bitran JD (ed): Lung-Cancer: A Comprehensive Treatise. Orlando, FL, Grune & Stratton, 1988, pp 173-182 24. Belcher JR, Anderson R: Surgical treatment of carcinoma of the bronchus. Br Med J 1:948-954, 1965
Downloaded from ascopubs.org by University of Newcastle Upon Tyne on April 7, 2019 from 128.240.208.034 Copyright © 2019 American Society of Clinical Oncology. All rights reserved.
