Abstract
Background: Based on the results of phase I/II studies, S-1 plus cisplatin (CDDP) and vinorelbine (VNR) plus CDDP are commonly used chemoradiotherapeutic regimens for the treatment of non-small cell lung cancer(NSCCLC) in Japan. However, there have been no studies that have compared S-1 and CDDP combined with thoracic radiotherapy (TRT) with VNR and CDDP combined with TRT. Patients and Methods: A total of 39 and 50 patients with stage III non-small cell lung cancer (NSCLC) were treated with S-1 and CDDP plus concurrent TRT, or with VNR and CDDP plus concurrent TRT, respectively, between 2002 and 2010. Results: In the S-1 plus CDDP plus TRT group, the median progression-free survival (PFS) and the median overall survival (OS) were 327 days and 1012 days, respectively. In the VNR plus CDDP plus TRT group, the median PFS and the median OS were 328 days and 905 days, respectively. The differences in the PFS and OS were not statistically significant. Grade 3 or more leukopenia and neutropenia were significantly more common in the VNR plus CDDP plus TRT group. Grade 3 or more thrombocytopenia, esophagitis and eruption tended to be more common in the S-1 plus CDDP plus TRT group. Conclusion: Due to the difference in the toxicity profiles of the two combinations, S-1 plus CDDP plus TRT or VNR plus CDDP plus TRT should be selected depending on each patient's baseline characteristics.
- S-1
- vinorelbine
- cisplatin
- non-small-cell lung cancer
- chemoradiotherapy
The standard treatment for stage III locally advanced non-small cell lung cancer (NSCLC) was combined modality of thoracic radiotherapy (TRT) and chemotherapy (1, 2). Phase III studies have since been conducted to assess the efficacy and toxicity of concurrent chemoradiotherapy (CRT) in comparison with the one of sequential chemoradiotherapy. In two studies, namely a Japanese Study (3) and the radiation therapy oncology group (RTOG) 9410 (4), which employed older, second-generation regimens, the survival period was reported to be significantly prolonged by concurrent CRT, although the toxicity was worse. Thus, the standard of treatment for stage III locally advanced lung cancer is currently recognized as concurrent CRT.
In patients with stage IIIB or IV NSCLC, it has been demonstrated that third-generation agents combined with cisplatin or carboplatin yield superior survival as compared to regimens containing second-generation agents (5-7). Based on these data, a phase III trial was conducted to compare third-generation chemotherapy with second-generation chemotherapy used in concurrent CRT regimens for patients with unresectable stage III NSCLC. The phase III study which compared mitomycin, vindesine plus cisplatin, irinotecan plus carboplatin and paclitaxel plus carboplatin used as concurrent CRT regimens for stage III NSCLC, showed that paclitaxel plus carboplatin was equally efficacious and exhibited a more favorable toxicity profile among the three arms [paclitaxel plus carboplatin arm: median progression-free survival (PFS) 9.5 months, median survival time (MST) 22.0 months] (8). On the other hand, because cisplatin-based chemotherapy is slightly superior to carboplatin-based chemotherapy in terms of the response rate and in prolonging survival of patients with advanced NSCLC (9), additional studies were conducted to assess benefits of the cisplatin-based chemoradiotherapy. A phase II study of S-1 and cisplatin with concurrent TRT for stage III NSCLC led to a median PFS of 12.0 months and MST of 33.1 months (10). In addition, a larger phase I study which recruited 18 patients and used vinorelbine plus cisplatin as concurrent CRT gave a median PFS of 15.6 months and an MST of 30.4 months (11).
Based on the findings of these clinical trials, S-1 plus cisplatin or vinorelbine plus cisplatin are now commonly used as CRT for stage III NSCLC in Japan. However, so far, there has been no clinical trial that has compared these CRT regimens. Therefore, we conducted this retrospective study to compare and clarify the efficacy and toxicity of S-1 plus cisplatin and vinorelbine plus cisplatin as concurrent CRT regimens.
Patients and Methods
Patient selection. We reviewed 39 and 50 consecutive patients with inoperable stage III NSCLC who were treated with S-1 and cisplatin plus concurrent TRT, and with vinorelbine and cisplatin plus concurrent TRT, respectively, at Shizuoka Cancer Center between July 2002 and December 2010. The TNM stage was classified using TNM stage version 6 (12). In terms of the T factor, T4 disease, or the presence of pulmonary metastasis in the same lobe, was considered ‘unresectable’. In terms of the N factor, clinically apparent or histologically/ cytologically proven multiple N2, bulky N2, N3 or both N1 positive and N2 positive disease were considered ‘unresectable’. In general, lymph nodes that were larger than 10 mm in the minor axis were considered to be metastatic. To confirm the presence of N2 disease, which was detected by chest computed tomography (CT) and smaller than 10 mm in the minor axis, 18F-fluoro-deoxy-glucose positron-emission tomography (FDG-PET) and/or mediastinoscopy was performed. Chest CT, abdominal CT, bone scintigram or FDG-PET, and brain magnetic resonance imaging (MRI)/CT were performed before CRT for all patients.
The inclusion criteria for CRT in our institution are generally as follows: age ≤75 years, performance status (PS) of 0-1, white blood cell count ≥3.0 × 103/mm3, neutrophil count ≥1.5 × 103/mm3, platelet count ≥1.0 × 105/mm3, serum creatinine ≤1.5 mg/dl, total bilirubin ≤1.5 mg/dl and transaminase level less than twice the upper limit of the normal value. The exclusion criteria were interstitial lung disease identified by a chest x-ray, malignant pleural or pericardial effusion, and serious complications, such as severe respiratory failure, active infectious diseases, serious heart diseases, and poorly controlled hypertension/diabetes mellitus. All patients gave informed consent before CRT.
Chemotherapy. S-1 plus cisplatin plus TRT. S-1 (40 mg/m2) was administered orally twice, daily on days 1-14, along with intravenous infusion of cisplatin (60 mg/m2) on day one. The treatment cycles were repeated every four weeks for a maximum of four cycles. The oral doses of S-1 for each patient were assigned based on their body surface area (BSA). The three doses of S-1 that were administered based on the BSA were: 40 mg, BSA ≤1.25 m2; 50 mg, 1.25 m2 <BSA≤1.50 m2; and 60 mg, BSA >1.50 m2. In general, if the entry eligibility criteria for CRT were not met, subsequent cycles of treatment were withheld until the noted abnormality had resolved. If there was no resolution of the abnormality after seven weeks from the first day of the cycle, chemotherapy was stopped. Generally, the doses of S-1 were reduced in the event of grade 4 hematological toxicity, or grade 3 or more non-hematological toxicity during the previous treatment cycle. For the subsequent courses, S-1 was reduced from 60, 50, or 40 mg twice daily to 50, 40, or 25 mg twice daily, respectively.
Vinorelbone plus ciplatin plus TRT. Vinorelbine (20 mg/m2), on days 1 and 8 and cisplatin (80 mg/m2) on day one were administered intravenously. The treatment cycles were repeated every four weeks for a maximum of four cycles. In general, if the entry eligibility criteria for the CRT were not met, subsequent cycles of treatment were withheld until the noted abnormality had resolved. If there was no resolution of the abnormality after seven weeks from the first day of the cycle, chemotherapy was stopped. Generally, the doses of vinorelbine and cisplatin were reduced in the event of grade 4 hematological toxicity, or grade 3 or more severe non-hematological toxicity during the previous treatment cycle. When the white blood cell, neutrophil or platelet counts were below 2000/mm3, 1000/mm3 or 75,000/mm3, respectively, or active when infection was present, the administration of vinorelbine on day eight was omitted.
Supportive care: All patients received prophylactic antiemetic therapy consisting of a 5-HT3 antagonist, metoclopramide, and dexamethasone. The use of granulocyte colony-stimulating factor during radiotherapy was not permitted.
Radiotherapy. TRT was concurrently started in the first cycle of chemotherapy. All the patients were required to undergo chest CT to facilitate treatment planning. The primary tumor (gross tumor volume; GTV primary) was delineated in the pulmonary windows, and the nodal involvement (GTV node) was delineated in the mediastinal windows. The clinical target volume (CTV) included the GTV primary, GTV node, ipsilateral hilum, and elective mediastinum, for which the lower border was 3.0 cm below the carina up to 40 Gy. After 40 Gy, CTV included the GTV primary and GTV node. The planning target volume (PTV) was the CTV plus a margin to ensure that the prescribed dose was actually delivered to the CTV. The prescribed dose was 60 Gy in 30 fractions (fr). The heterogeneity correction was used, and the dose was prescribed at an isocenter. The PTV was to cover 90% of the isodose surface. Portal verification was carried out for all treatment fields. It was ensured that the normal lung volume receiving more than 20 Gy (V20) was equal to or less than 35% of the total lung volume. The maximal dose to the spinal cord was not to exceed 45 Gy at any level.
Evaluation of efficacy and toxicity. The tumor response was evaluated in accordance with the response evaluation criteria in solid tumors (RECIST) ver. 1.0 (13). Acute adverse events were evaluated until 4 weeks after the last administration of chemotherapy/TRT, or until the patient's death, in accordance with the common terminology criteria for adverse events (CTCAE) ver. 3.0 (14). Late adverse events were scored according to the European Organization for Research and Treatment of Cancer/Radiation Therapy Oncology Group late radiation morbidity scoring scheme.
Statistical analysis. To analyze the progression-free (PFS) and overall survival (OS), survival curves were drawn using the Kaplan-Meier method. The PFS was calculated from the date of initiation of the CRT to the date of detection of disease progression or the date of death from any cause. The PFS was censored at the date of the last visit for those patients who were still alive without any documented disease progression. The OS was calculated from the date of initiation of the CRT to the date of death. The OS was censored at the date of the last visit for those patients whose deaths could not be confirmed. The PFS and OS were compared using the log-rank test according to the chemotherapeutic regimens (S-1 plus cisplatin plus TRT vs. vinorelbine plus cisplatin plus TRT). Cox proportional hazard models were used to calculate the hazard ratio (HR) and adjusted HR of the PFS and OS according to chemotherapeutic regimens. To evaluate the differences in the patients' characteristics, the chi-square test or Fisher's exact test were used, depending on the numbers of the patients. P-values <0.05 were considered to be statistically significant. All statistical analyses were performed by the application of JMP ver. 8.0 for Windows (SAS Institute Inc., Cary, NC, USA).
Results
Patients' characteristics. From July 2002 to December 2010, 39 patients received S-1 plus cisplatin plus TRT (planned dose: 60 Gy/30 fr) and 50 patients received vinorelbine plus cisplatin plus TRT (planned dose: 60 Gy/30 fr) at the Shizuoka Cancer Center. Table I shows the patients' characteristics. There were no statistically significant differences between the two groups in terms of gender, age, histology, performance status (PS), smoking history, presence of epidermal growth factor receptor (EGFR) mutations and stage. However, there was a trend for more male patients and a better PS in the S-1 plus cisplatin plus TRT group.
Toxicity, administered cycles of chemotherapy, and discontinuation of radiotherapy. Table II shows the toxicities in the two groups, respectively. With regard to grade 3 or higher toxicities, leukopenia and neutropenia were significantly more common in the vinorelbine-treated group than in the group treated with S-1 (leukopenia, p<0.0001; neutropenia, p<0.0001). On the other hand, grade 3 or more severe thrombocytopenia, esophagitis and eruption tended to be more common in the S-1-treated group than in the vinorelbine-treated group (thrombocytopenia, p=0.0825; esophagitis, p=0.0805; eruption, p=0.0805). There were no treatment-related deaths in either group.
The median number of cycles administered was 4 (range, 1-4) in the S-1-treated group, and 4 (range, 1-4) in the vinorelbine-treated group. There was no significant difference in the number of cycles administered between the two groups (Mann-Whitney test, p=0.321).
Discontinuation or omission of radiotherapy tended to be more common in the vinorelbine-treated group than in the S-1-treated group [9 patients (18%) in vinorelbine-treated group and 5 patients (13%) in the S-1-treated group, p=0.569]. The reasons were grade 3 febrile neutropenia in four patients, grade 4 leukopenia/neutropenia in four patients and grade 3 infection in one patient among the 9 patients in the vinorelbine-treated group. All nine patients recovered from toxicities and completed 60 Gy of TRT. The reasons for discontinuation or omission of radiotherapy were grade 2 pneumonitis in three patients and grade 3/4 thrombocytopenia in two patients among the five patients in the S-1-treated group. Although the two patients with grade 3/4 thrombocytopenia recovered and completed 60 Gy of TRT, the three patients with grade 2 pneumonitis discontinued TRT permanently.
Response to therapy and survival. Among the 39 patients in the S-1-treated group, 25, 12, 1 and 1 showed a complete response (CR)/ partial response (PR), stable disease (SD), progressive disease (PD) and were not evaluable (NE), respectively. The response rate was 64%, and the disease control rate was 95% (Table III). On the other hand, among the 50 patients in the vinorelbine-treated group, 38 and 12 patients showed a CR/ PR and SD, respectively. The response rate was 76% and the disease control rate was 100% (Table III). The differences in the response rate and disease control rate between the two groups were not statistically significant (response rate, p=0.321; disease control rate, p=0.438).
In the S-1-treated group, the median PFS and the median OS were 327 days and 1012 days, respectively (Figures 1 and 2). In the vinorelbine treated group, the median PFS and the median OS were 328 days and 905 days, respectively (Figures 1 and 2). There were no statistically significant differences in the PFS and OS between the two groups [PFS: log-rank p=0.528, HR 0.849 (95% confidence interval CI=0.511-1.42); OS: log-rank p=0.972, adjusted HR (age, gender, PS) 0.832 (95% CI=0.434-1.60)].
Failure site after chemoradiotherapy. There were 27 (69%) and 29 (58%) cases of disease relapse in the S-1-treated group and the vinorelbine-treated group, respectively. Among the 27 patients with recurrence in the S-1 plus cisplatin plus TRT group, infield relapses were observed in 11 patients (41%, 8 without and 3 with relapse outside of the radiation fields). Distant metastases were the first sites of failure in 19 patients (70%). Nineteen patients received second-line or further chemotherapy. On the other hand, among the 29 patients with recurrence in the vinorelbine plus cisplatin plus TRT group, infield relapses were observed in 18 patients (62%, 10 without and 8 with relapse outside of the radiation fields). Distant metastases were the first sites of the failure in 19 patients (66%). Nineteen patients received second line or further chemotherapy.
Discussion
As mentioned earlier, as was as in the national cancer comprehensive network (NCCN) guidelines, the standard treatment for unresectable stage III NSCLC is considered to be concurrent chemoradiotherapy (15). Furthermore, in the NCCN guidelines, the recommended chemotherapeutic regimens are etoposide plus cisplatin, vinblastine plus cisplatin or paclitaxel plus carboplatin. However, because third-generation agents combined with platinum agents yield superior survival as compared to regimens containing second-generation agents, and because cisplatin-based chemotherapy is slightly superior to carboplatin-based chemotherapy in terms of the response rate and in prolonging survival in patients with advanced NSCLC, cisplatin with S-1 or vinorelbine is commonly used as concurrent CRT for stage III NSCLC in Japan. However, there have been no previous studies that have compared these combinations with TRT in terms of their efficacy and toxicity. By clarifying the differences in the efficacy and toxicity of these regimens, this study helps physicians to decide which regimen to use for particular patients.
Although there was a trend toward a better response rate being observed in the vinorelbine-treated group, the response rate, disease control rate, PFS and OS were comparable between the two groups. In addition, the PFS and OS were longer than the ones observed in the previous phase II or III studies evaluating concurrent chemoradiotherapy using carboplatin plus paclitaxel or cisplatin plus etoposide as chemoradiotherapeutic regimens. Taking into account that the patients' backgrounds are usually better in clinical studies than in general clinical practice, the PFS and OS in this study are satisfactory, and indicate that both these combinations might be appropriate for use in concurrent CRT regimens.
With regard to toxicities, leukopenia and neutropenia were significantly more common in the vinorelbine-treated group than in the S-1-treated group. On the other hand, thrombocytopenia, esophagitis and eruption tended to occur more frequently in the S-1-treated group than in the vinorelbine-treated group. The reasons for discontinuation or omission of radiotherapy were related to these toxicities. Therefore, the patients with lower neutrophil/leukocyte counts at baseline should be treated using the S-1 combination, while those with a lower platelet count should be treated using the vinorelbine combination to control the local recurrence. Although there were no significant differences in the frequency and severity of pneumonitis, three patients with grade 2 pneumonitis discontinued TRT in the S-1-treated group.
There are a few limitations of this study that must be addressed. The severity of non-hematological toxicities, in particular, may have been underestimated in the present study because of its retrospective nature. However, patients were treated as inpatients during most of the treatment period, and the toxicity data were recorded in detail in the patients' medical records. The frequency and severity of both hematological and non-hematological toxicities were comparable to those of the previous prospective phase I or II studies. All patients were evaluated for their evaluable lesions approximately every two months by CT, MRI, bone scintigraphy or PET during the treatment period and every three to six months after treatment. However, the intervals between evaluations in the present study were not as accurate as those in a prospective study. Therefore, the comparison of the PFS between the two groups was less reliable.
In conclusion, using cisplatin plus S-1 or vinorelbine as concurrent CRT regimen is warranted, taking the present data and the data from previous studies using etoposide plus cisplatin, vinblastine plus cisplatin, and paclitaxel plus carboplatin as CRT regimens into account. In addition, because their toxicity profiles are different, the selection of these combination therapies should be based on the individual patient's baseline characteristics. However, a further randomized trial will be necessary to fully evaluate the usefulness of the current findings.
Footnotes
-
Conflict of Interest
None of the Authors have any financial or personal relationships that could influence their work.
- Received November 1, 2011.
- Revision received December 12, 2011.
- Accepted December 13, 2011.
- Copyright© 2012 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved