The purpose was to determine the efficacy and toxicity of irinotecan and cisplatin with concurrent split-course thoracic radiotherapy (TRT) in locally advanced nonsmall-cell lung cancer.
The purpose was to determine the efficacy and toxicity of irinotecan and cisplatin with concurrent split-course thoracic radiotherapy (TRT) in locally advanced nonsmall-cell lung cancer.
Fifty patients fulfilling the following eligibility criteria were enrolled: chemotherapy-naive, good performance status (PS, 0-2), age <75, stage III, and adequate organ function. The patients received irinotecan 60 mg/m2 intravenously on Days 1, 8, and 15, and cisplatin 80 mg/m2 intravenously on Day 1 in the first group. The doses were reduced to 50 and 60 mg/m2, respectively, in the second group. Two cycles of chemotherapy were repeated every 4 weeks. Split-course thoracic radiotherapy of 2 Gy/day commenced on Day 2 of each chemotherapy cycle, with 28 and 32 Gy administered in the first and second cycles, respectively.
Fifty patients were eligible and 48 (16 in the first, 32 in the second group) patients were assessable for response, toxicity, and survival. The overall response was 83% (95% confidence interval [CI], 70%–93%). Grade 4 leukopenia, neutropenia, grade 3 or 4 diarrhea, pneumonitis, esophagitis, and fatigue occurred in 21%, 48%, 19%, 10%, and 19%, respectively. The median time to progression was 8.2 months. The median overall survival time and the 2- and 5-year survival rates were 20.1 months, 47.1%, and 17.1%, respectively. In subgroup analysis, grade 4 neutropenia, grade 3 or 4 diarrhea, the overall response, and the median survival times of the first/second groups were 63%/41%, 19%/19%, 75%/88%, and 13.1/33.4 months, respectively.
This combined modality of irinotecan and cisplatin with concurrent TRT is active and further investigations are warranted at the second group dose level. Cancer 2007. © 2007 American Cancer Society.
Lung cancer is the leading cause of cancer death worldwide. It is also the Japanese leading cause of death, with 57,000 deaths (18.3% of all cancer deaths) in 2003.1 Of all lung cancer cases, approximately 80% are nonsmall-cell lung cancer (NSCLC),2 and 25% to 40% of NSCLC patients have stage III, locally advanced disease.3 Thoracic radiotherapy (TRT) to the primary tumor and regional lymph nodes has been the traditional treatment for patients with locally advanced, unresectable stage III NSCLC.4 As the therapy provided 5-year survival rates of only 5% to 7%, with a median survival time of 6 to 11 months,5 a combined modality of chemotherapy and radiotherapy was attempted. Three meta-analyses of randomized trials demonstrated that combination chemoradiotherapy was superior to radiotherapy alone.6–8 Phase 3 trials have suggested that the concurrent administration of these 2 modalities improves long-term survival compared with sequential strategies.9–11
Irinotecan is a semisynthetic, water-soluble prodrug that is metabolized to the active metabolite SN-38, which inhibits the function of DNA topoisomerase I in cancer cells.12, 13 Clinical studies of irinotecan alone have shown a broad spectrum of antitumor activity against various human cancers including NSCLC,14 and preclinical studies demonstrated synergism and noncross-resistance between the combination of cisplatin with SN-38 and irinotecan.15, 16 Phase 2 and 3 trials of irinotecan/cisplatin therapy in Japan yielded a 44% to 52% response rate and 10.1 to 11.5 months of median survival for NSCLC,17, 18 and a preclinical study showed an enhancement effect with the use of irinotecan on tumor radiosensitivity.19 A previous Japanese trial of irinotecan/cisplatin with concurrent standard radiotherapy of 60 Gy in 30 fractions failed because of unacceptable toxicity.20 In our phase 1 trial,21 we incorporated split-course radiotherapy with a rest period between 2 chemotherapy cycles and demonstrated tolerability and a good response rate of 69.6% (95% confidence interval [CI], 47.1%–86.8%).
Based on these results, we conducted a phase 2 trial of irinotecan/cisplatin with concurrent split-course radiotherapy for locally advanced NSCLC. The main objective of the trial was to determine the efficacy and safety of this regimen.
The study protocol was reviewed and approved by the ethics committee of each institution and written informed consent was obtained from all study participants. This study is an independent collaborative (not-sponsored) group study. The participating institutions and investigators were: Second Department of Internal Medicine, Nagasaki University School of Medicine, Takashi Kasai, Shigeru Kawabata, Tetsuro Kanda, Takeshi Kitazaki, Shigeru Kohno, Yoshifumi Soejima, Hiroshi Soda, Nanae Tomonaga, Junji Tsurutani, Katsumi Nakatomi, Hirofumi Nakano, Reiji Nakano, Yoichi Nakamura, Satoru Fujino, and Hiroyuki Yamaguchi; Division of Respiratory Diseases, Department of Medicine, Kawasaki Medical School, Mikio Oka, Minoru Fukuda; Japanese Red-Cross Nagasaki Genbaku Hospital, Masaaki Fukuda, Tetsuya Iida; National Hospital Organization Nagasaki Medical Center, Akitoshi Kinoshita, Seiji Doi; Sasebo General Hospital, Seiji Nagashima; Nagasaki Municipal Hospital, Hiroshi Takatani; Nagasaki University School of Pharmaceutical Science, Kazuhiro Tsukamoto; Department of Radiology and Radiation Biology, Nagasaki University School of Medicine, Nobuyuki Hayashi; Saiseikai Futsukaichi Hospital, Toru Rikimaru; National Hospital Organization Kyusyu Medical Center, Masao Ichiki.
Patients with previously untreated, unresectable, and locally advanced stage III NSCLC were enrolled, but patients with malignant pleural effusion were excluded. Eligibility criteria included the following: a histologically confirmed diagnosis of NSCLC; age <75 years; Eastern Cooperative Oncology Group (ECOG) performance status (PS) ≤2; adequate bone marrow function (leukocyte count ≥4000/μL, platelet count ≥10 × 104/μL, and hemoglobin level ≥10.0 g/dL); serum bilirubin level ≤1.5 mg/dL; alanine and aspartate aminotransferase (ALT, AST) levels ≤2 times the upper limit of normal; serum creatinine level ≤1.5 mg/dL, and PaO2 ≥70 torr; and no medical problems severe enough to prevent compliance with the protocol.
Treatment commenced within 1 week of enrollment and 2 cycles of irinotecan/cisplatin therapy were repeated at 4-week intervals. Based on our phase 1 study,21 patients received 60 mg/m2 of irinotecan on Days 1, 8, and 15, and 80 mg/m2 of cisplatin on Day 1. After 16 patients were enrolled the chemotherapy dose was reduced to irinotecan 50 mg/m2 and cisplatin 60 mg/m2. Irinotecan was not administered on Day 8 or 15 in the cycle if the leukocyte count was <3000/μL, the platelet count was <10 × 104/μL, or the patient had diarrhea on those days. The next cycle commenced after the leukocyte and platelet counts reached at least 3000/μL and 10 × 104/μL, respectively. After completion of combined modality treatment, additional irinotecan/cisplatin therapy was optionally permitted.
Thoracic radiation was administered once daily with a split-schedule: 5 days/week with 2 Gy/day from Day 2 of each chemotherapy cycle, with a total of 28 and 32 Gy provided in the first and second cycles, respectively. There was a break in the split-course radiation of approximately 10 days. Based on a recent chest computed tomography (CT) scan, the radiation volumes and fields were individualized for each patient. The radiation fields encompassed the areas of all visible tumor and involved lymph nodes (≥1.0 cm) with a margin of 1.5 to 2 cm for each, as well as ipsilateral hilar, superior mediastinal, and subcarinal nodes. The area of the lung field included in the radiation field was not greater than half of the area of the ipsilateral lung. All radiation was delivered with megavoltage linear accelerators. Isodose curves and dose-volume histograms (DVHs) were calculated for the lungs and the esophagus. Patients received the intended radiotherapy with a total dose of 60 Gy. The maximum dose to any level of the spinal cord did not exceed 45 Gy. Local control was defined as no evidence of progressive disease within the planned radiation field. The treatment schema is shown in Figure 1.
Irinotecan treatment was omitted on Days 8 and 15 if the leukocyte count fell below 3000/μL, platelet count <10 × 104/μL, or any diarrhea occurred. Leukocytes ≥3000/μL and platelets ≥10 × 104/μL were mandatory to commence the second cycle of treatment, and if levels fell below these limits the second cycle was postponed until the counts recovered. Doses of irinotecan and cisplatin were reduced to 80% when dose-limiting toxicity (grade 4 neutropenia lasting 4 days or more, grade 4 thrombocytopenia, and grade 3 or greater nonhematologic toxicities except nausea and vomiting) occurred during the first treatment cycle.
Radiation was interrupted if grade 4 hematologic toxicity occurred during radiation, and restarted after recovery to grade 3 or less. If grade 3 or greater esophagitis occurred, it was interrupted and restarted after recovery to grade 2 or less. If esophagitis did not resolve it was discontinued. If PaO2 fell to 10 torr or a patient had a fever of 38 °C or higher, both radiotherapy and chemotherapy were interrupted and restarted as soon as possible after recovery.
Eligibility, assessability, and tumor responses were determined by external reviewers. Drug toxicity was graded according to the National Cancer Institute Common Toxicity Criteria, v. 2.22 Before the first cycle a blood cell count, urinalysis, and biochemistry tests were performed to assess renal and hepatic function and electrolytes. This monitoring was repeated during treatment, whereas other investigations were repeated, as necessary, to evaluate marker lesions. After the completion of treatment each disease was assessed and tumors restaged. Tumor response was classified according to the World Health Organization (WHO) criteria23: complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD).
The primary endpoint of this study was to estimate the objective response rate. The 2-stage accrual design described by Simon24 was used. Assuming an overall response rate of 50% for standard therapy, a target response rate of 70% was established. Alpha = 0.05, beta = 0.20, and the estimated required number of patients was more than 37. Considering unfitness, dropout, and discontinuation, the sample size of this study was determined to be 45. Overall survival was calculated by the Kaplan-Meier method.25 Subset-analysis was performed by Mann-Whitney U, log-rank, and Breslow-Gehan-Wilcoxon analysis. The statistical test was 2-sided and a probability value <0.05 was defined as statistically significant.
Fifty patients from 12 institutions were enrolled in this trial between September 1998 and August 2003. One patient was ineligible because of malignant effusion after enrollment. Another patient withdrew informed consent and transferred to a different hospital. Forty-eight patients received the planned treatment and were evaluated for toxicity, response, and survival. The baseline patient characteristics are shown in Table 1. Patients in PS 2 were eligible but were not enrolled in the study.
|Entry no. of patients||50|
|Eligible no. of patients||48|
|ECOG performance status|
|Squamous cell carcinoma||15||31|
|Large cell carcinoma||2||4|
A total of 133 cycles of irinotecan plus cisplatin were administered to the 48 patients: 1 cycle in 4 patients, 2 cycles in 19, 3 in 11, 4 in 12, and 5 in 2. Then the latter 25 (52%) patients followed the same consolidative regimen chemotherapy after chemoradiotherapy. Thirty (23%) irinotecan administrations were skipped on Day 8, 67 (50%) on Day 15, including 19 (14%) on both days. The major reasons for omission on Days 8 and 15 were leukopenia 60 cases, diarrhea 8 cases, patient refusal 4 cases, nausea and infection 3 cases each, reduction in PS, liver dysfunction, and leukopenia/thrombocytopenia 2 cases, pneumonitis, fever, ileus, leukopenia/diarrhea, leukopenia/liver dysfunction 1 case. The administration rate of irinotecan on Days 8 and 15 was 77% and 50%, respectively. The average treatment delay of the second cycle was 1.6 days per case. The dose intensity of irinotecan in the first 2 cycles was 27.5 (60/80 level, 27.9; 50/60 level, 27.4) mg/m2/week, which was 69.0% (60/80 level, 62.0%; 50/60 level, 73.1%) of the projected dose intensity. The dose intensity rate relative to the projected dose intensity of cisplatin was 96.4% (60/80 level, 96.7%; 50/60 level, 96.1%).
Of the first 16 patients, 5 (31%) could not complete the planned chemoradiotherapy because of refusal (3), prolonged neutropenia (1), and PD (1). The patients who refused experienced grade 3 fatigue, esophagitis, hiccoughs, or grade 4 neutropenia. This problem was discussed by the investigators and irinotecan and cisplatin doses were reduced to 50 and 60 mg/m2, respectively, which were nearly 80% of the initial dose, and 3 of 4 patients experienced a partial response at the same dose level in our phase 1 trial.21 After drug-dose reduction, compliance with the treatment including TRT became favorable and almost all patients completed the treatment, except for 2 patients who experienced pneumonitis and prolonged infection.
The toxicities during the treatment are listed in Table 2. All 48 patients were assessable for toxicity. Forty-one (85%) patients experienced grade 3 or 4 hematologic toxicity, and 23 (48%) had grade 4. The principal grade 3 or 4 hematologic toxicity was leukopenia and neutropenia in 39 (81%) patients, and the principal grade 4 toxicity was neutropenia in 23 (48%) patients. Febrile neutropenia occurred in 3 (6%) patients. The major nonhematologic toxicities were infection, gastrointestinal toxicity, and fatigue. One patient had grade 4 bloody diarrhea with ileus and multiple digestive ulcers during the second cycle and underwent a blood transfusion. Grade 3 radiation pneumonitis occurred in 5 (10%) patients. Grade 3 skin toxicity, mucositis, and alopecia were each observed in 1 (2%) patient. There was no treatment-related death.
|Adverse event||WHO Grade|
All 48 patients were assessed for response. Objective tumor response was observed in 40 patients, with an overall response rate of 83% (95% CI, 70%–93%). Stable disease was observed in 7 (15%) patients, and the remaining 1 (2%) patient had progressive disease. Responses at each stage were 100% (11 PR) in IIIA and 78% (29 PR, 7 SD, and 1 PD) in IIIB.
The median potential follow-up time was 61.7 (range, 29.8–89.0) months. Twelve patients survived and the other 36 patients died during the follow-up period. The progression-free survival of 48 patients is shown in Figure 2A. Median time to tumor progression was 8.2 (95% CI, 6.0–10.3) months, and the 1-, 2-, 3-, 4-, and 5-year progression-free survival rates were 40.4%, 22.7%, 17.3%, 17.3%, and 17.3%, respectively. The overall survival of 48 patients is shown in Figure 2B. Median survival time was 21.0 (95% CI: 16.8–25.2) months, and the 1-, 2-, 3-, 4-, and 5-year survival rates were 73.9%, 47.1%, 31.4%, 25.7%, and 17.1%, respectively.
Overall survival before and after drug reduction is shown in Figure 3A. The curves separated at about 9 months, and suggested a significant difference in favor of postdrug dose reduction (irinotecan 50 mg/m2 + cisplatin 60 mg/m2 + TRT) in both log-rank (P = .011) and Breslow-Gehan-Wilcoxon (P = .009) analyses. The median survival time, 1-, 2-, 3-, 4-, and 5-year survival rates before and after dose reduction were 13.1 (95% CI, 4.5–21.7) months, 55.0%, 27.5%, 6.9%, 6.9%, 6.9%, and 33.4 (95% CI, 16.3–50.4) months, 83.7%, 56.9%, 45.0%, 31.5%, 15.7%, respectively. The differences between the 2 chemotherapy dose levels are shown in Table 3. No differences were observed in baseline characteristics. After dose reduction, the dose intensity (DI) of cisplatin was reduced by almost 5 mg/m2/week (P < .001), whereas irinotecan was not reduced because of higher execution rates on Day 8 (which was not significant) and Day 15 (P = .032). Seven patients could not complete the planned treatment as stated in the treatment administration paragraph, so the differences of completion rates were significant (P = .022) between the 2 dose levels. The overall survival curves by irinotecan execution times on Days 8 and 15 are shown in Figure 3B,C, respectively. Both curves favored high execution rates and the differences were analyzed by 2 methods. The difference of curves on Day 15 was significant in both log-rank (P = .026) and Breslow-Gehan-Wilcoxon (P = .031) tests. The difference of curves on Day 8 was significant in the log-rank tests (P = .014) but was not significant in Breslow-Gehan-Wilcoxon (P = .074).
|Stage IIIA (%)||25||22||.810|
|PS 0 (%)||25||38||.392|
|DI of CP||27.9||27.4||.991|
|DI of DP||19.3||14.4||<.001|
|D8 rate (%)||72||83||.268|
|D15 rate (%)||22||48||.032|
|G3/4 neutropenia (%)||88||78||.438|
|G4 neutropenia (%)||63||41||.157|
|G3/4 diarrhea (%)||19||19||>.999|
|G3/4 pneumonitis (%)||6||13||.509|
|Mean CT cycles||2.4||3.0||.104|
Of the 45 evaluable patients in this investigation, 31 (69%) recurred. The first sites of disease progression were as follows: local only in 13 (42%) patients, local and distant in 4 (13%), and distant only in 14 (45%). In 6 (19%) patients, the initial site of recurrence was the brain. Of these, 3 recurred solely in the brain. Other recurrence sites included pleural effusion, bone, and lung (outside the TRT field) in 4 patients, and the liver, supraclavicular lymph node, ascites, and pericardial effusion in 1 patient. Treatment failure within 6 months is presented in Table 4. The significant failure factors were young age (≤64), female, not completing the treatment, dose 60/80, and irinotecan low execution on Day 15. These factors also impacted overall survival except for young age (data not shown).
|Factor||No. of recurrences (%)||P|
|Age, ≤64 y, ≥65 y||11 (38)||1 (6)||.033|
|Sex, female, male||5 (42)||7 (21)||.043|
|PS, 0, 1||4 (25)||8 (28)||.878|
|Stage, IIIA, IIIB||2 (20)||10 (27)||.593|
|Histology, sq, non-sq||2 (13)||10 (33)||.283|
|Completion, yes, no||7 (25)||5 (71)||.010|
|Dose, 60/80, 50/60||8 (62)||4 (13)||.001|
|Response, PR, SD+PD||10 (26)||2 (29)||.902|
|1, 2||2 (50)||5 (26)|
|3, 4||4 (36)||1 (8)|
|0, 1||1 (20)||5 (45)|
|0, 1||11 (48)||1 (9)|
Irinotecan/cisplatin therapy yielded 31.0% to 43.7% overall response and 46.5% to 59.2% 1-year survival in Japanese randomized phase 3 trials,18, 26 and 28.8% overall response and 37% 1-year survival in a phase 2 trial in the US27 for stage IIIB, IV NSCLC. Preclinical and clinical studies demonstrated the radiosensitizing activity of irinotecan.19, 28 Therefore, irinotecan/cisplatin plus TRT is considered 1 of the most active strategies for locally advanced NSCLC. Although the Fox Chase Cancer Center Group reported the tolerability of irinotecan/cisplatin with standard thoracic radiotherapy,29 a previous Japanese dose-finding trial of irinotecan/cisplatin with concurrent 60 Gy radiotherapy was not completed because of unacceptable toxicity.20 We incorporated split-course radiotherapy, which was applied in the trial by Furuse et al.,9 because of safety considerations for normal tissue, and combined full-dose irinotecan/cisplatin (60/80 mg/m2) chemotherapy and 60 Gy thoracic radiotherapy in the phase 1 trial.21 In the present phase 2 trial, however, we were confronted with severe toxicity of grade 3/4 neutropenia (88%), a low irinotecan execution rate on Day 15 (22%), a low completion rate (69%), and patients' refusal to continue the treatment in the first 60/80 dose level, and decided to reduce the dose of irinotecan/cisplatin to 50/60 mg/m2. After dose reduction the trial went smoothly and the dose of 50/60 mg/m2 was considered the recommended dose for this combined modality. The results of the present study, including a response rate of 83%, median survival time of 21.0 months, and 2-year survival time of 47.1%, are encouraging.
One problem of phase 2 trials is that most trials evaluate only the maximum tolerated dose defined by the phase 1 trial. The present trial could not continue at the maximum dose; however, dose reduction fortunately achieved good results. The completion rates of therapy significantly increased from 69% to 94%. Although a cautious estimation might be needed in subset analysis, the median survival time of 33.4 months at the 50/60 dose level of the present trial was promising. In contrast, the median survival time of the 60/80 level, at which we could not continue the trial, was only 13.1 months. The first possible reason for these differences between the 2 dose levels is that full-dose chemotherapy with concurrent radiotherapy may be too toxic for normal tissue when using third-generation new cytotoxic agents. We incorporated split-course radiotherapy to cope with this problem, but it was not sufficient. The second possible reason is that chemotherapy between chemotherapy cycles, as on Day 15, may play an important role in chemoradiotherapy, although there is a negative trial of concurrent chemoradiotherapy using a daily radiosensitizer.30 The irinotecan execution rate on Day 15 at the 50/60 level (48%) was significantly higher than that at the 60/80 level (22%) and, as a result, the dose intensities of irinotecan at both levels were very similar (27.9% in 60/80 vs 27.4% in 50/60). In an examination of early (within 6 months) treatment failure, higher irinotecan execution on Day 15 was associated with higher local and distant control (Table 4) and longer survival (Fig. 3C). On Day 8 of irinotecan treatment, there were no differences in treatment failure, but it might have affected survival after 1.4 years (Fig. 3B). The recommended doses of 40/60 mg/m2 of irinotecan/cisplatin chemoradiotherapy for small-cell lung cancer (SCLC)31, 32 and 50/60 mg/m2 defined in the present study for NSCLC are very similar.
The main toxicity of our irinotecan/cisplatin with radiotherapy was hematologic. Comparing grade 3 or higher toxicities in the present trial with those for early concurrent radiotherapy arms of trials by Furuse et al.,9 Fournel et al.,11 and Belani et al.,33 diarrhea (19%) was higher in our trial, thrombocytopenia was higher in the trial by Furuse et al. (53% vs 6%–16%), and nausea/vomiting (7% vs 19%–24%) and neutropenia (26% vs 77%–99%) were lower in the trial by Belani et al. These differences varied based on the nature of the chemotherapy regimen used (irinotecan/cisplatin vs mitomicin/vindesine/cisplatin vs cisplatin/etoposide vs paclitaxel/carboplatin). Interestingly, esophagitis was lower in the former 2 trials compared with the latter (3%–10% vs 28%–32%). Considered with the median survival of these trials (20.1, 16.5, 16.3, 16.3 months, respectively), it seems that split-course concurrent radiotherapy could reduce the risk of esophagitis without loss of activity compared with continuous use. Gandara et al.34 reported a phase 2 trial of consolidation docetaxel after concurrent chemoradiotherapy with cisplatin plus etoposide for stage IIIB NSCLC, and achieved a median survival time of 26 months. The report recognized the possible important role of consolidation chemotherapy, and the treatment became 1 of the key regimens in this field. Although our therapy did not plan consolidation like docetaxel, optional irinotecan plus cisplatin continuation after chemoradiotherapy, which was used in 25 patients (52%), might have a favorable influence. Recently, Jeremic et al.35 reported a phase 2 trial of chemoradiotherapy for stage III NSCLC. They used twice daily hyperfractionated radiotherapy to a total dose of 67.6 Gy concurrently with daily 25 mg/m2 carboplatin and 10 mg/m2 paclitaxel after 30 mg/m2 paclitaxel on Day 1, and achieved the best median survival time of 28 months; however, their treatment is complicated for practical treatment. Our therapy at the 50/60 level also encouraged survival potential and was suitable for practical use.
In conclusion, our multicenter phase 2 trial demonstrated the encouraging activity of irinotecan and cisplatin with concurrent radiotherapy for patients with stage III NSCLC. Further investigations are warranted at the 50 mg/m2 irinotecan and 60 mg/m2 cisplatin dose level.