- © 2008 by American Society of Clinical Oncology
Phase III Study, V-15-32, of Gefitinib Versus Docetaxel in Previously Treated Japanese Patients With Non–Small-Cell Lung Cancer
- Riichiroh Maruyama,
- Yutaka Nishiwaki,
- Tomohide Tamura,
- Nobuyuki Yamamoto,
- Masahiro Tsuboi,
- Kazuhiko Nakagawa,
- Tetsu Shinkai,
- Shunichi Negoro,
- Fumio Imamura,
- Kenji Eguchi,
- Koji Takeda,
- Akira Inoue,
- Keisuke Tomii,
- Masao Harada,
- Noriyuki Masuda,
- Haiyi Jiang,
- Yohji Itoh,
- Yukito Ichinose,
- Nagahiro Saijo and
- Masahiro Fukuoka
- From the National Kyushu Cancer Center, Fukuoka; National Cancer Center Hospital East, Chiba; National Cancer Center Hospital; Tokyo Medical University Hospital, Tokyo; Shizuoka Cancer Center, Shizuoka; Kinki University School of Medicine; Osaka Medical Center for Cancer and Cardiovascular Diseases; Osaka City General Hospital; AstraZeneca KK, Osaka; Shikoku Cancer Center, Ehime; Hyogo Medical Center for Adults; Kobe City General Hospital, Hyogo; Tokai University Hospital, Kanagawa; Tohoku University Hospital, Miyagi; Hokkaido Cancer Center, Hokkaido; and Kitasato University School of Medicine, Kanagawa, Japan
- Corresponding author: Yukito Ichinose, MD, Department of Thoracic Oncology, National Kyushu Cancer Center, 3-1-1 Notame Minami-ku, Fukuoka, 811-1395, Japan; e-mail: yichinos{at}nk-cc.go.jp
Abstract
Purpose This phase III study (V-15-32) compared gefitinib (250 mg/d) with docetaxel (60 mg/m2) in patients (N = 489) with advanced/metastatic non–small-cell lung cancer (NSCLC) who had failed one or two chemotherapy regimens.
Methods The primary objective was to compare overall survival to demonstrate noninferiority for gefitinib relative to docetaxel. An unadjusted Cox regression model was used for the primary analysis.
Results Noninferiority in overall survival was not achieved (hazard ratio [HR], 1.12; 95.24% CI, 0.89 to 1.40) according to the predefined criterion (upper CI limit for HR ≤ 1.25); however, no significant difference in overall survival (P = .330) was apparent between treatments. Poststudy, 36% of gefitinib-treated patients received subsequent docetaxel, and 53% of docetaxel-treated patients received subsequent gefitinib. Gefitinib significantly improved objective response rate and quality of life versus docetaxel; progression-free survival, disease control rates, and symptom improvement were similar for the two treatments. Grades 3 to 4 adverse events occurred in 40.6% (gefitinib) and 81.6% (docetaxel) of patients. Incidence of interstitial lung disease was 5.7% (gefitinib) and 2.9% (docetaxel). Four deaths occurred due to adverse events in the gefitinib arm (three deaths as a result of interstitial lung disease, judged to be treatment related; one as a result of pneumonia, not treatment related), and none occurred in the docetaxel arm.
Conclusion Noninferiority in overall survival between gefitinib and docetaxel was not demonstrated according to predefined criteria; however, there was no statistically significant difference in overall survival. Secondary end points showed similar or superior efficacy for gefitinib compared with docetaxel. Gefitinib remains an effective treatment option for previously treated Japanese patients with NSCLC.
INTRODUCTION
In Japan, patients with advanced non–small-cell lung cancer (NSCLC) who fail first-line platinum-based therapy often receive second-line docetaxel.1,2 However, docetaxel has been associated with significant levels of toxicity, especially grades 3 to 4 neutropenia (40% to 67% and 63% to 73% for docetaxel 75 mg/m2 and 60 mg/m2, respectively).1-4 In North America and in European countries, docetaxel,3,4 pemetrexed,2 and erlotinib5 are approved second-line treatments for NSCLC.3,6
In phase II trials (IDEAL 1 and 2), the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor gefitinib (Iressa; AstraZeneca, London, United Kingdom) 250 mg/d showed response rates of 12% to 18% and median survival of 7.0 to 7.6 months in patients who had pretreated advanced NSCLC.7,8 A subset of Japanese patients in IDEAL 1 demonstrated a higher response rate (27.5%) and longer median survival (13.8 months) compared with the overall population.9 A phase III study (Iressa Survival Evaluation in Lung Cancer) in patients who had previously treated refractory NSCLC showed that gefitinib was associated with a nonsignificant trend toward improved overall survival versus placebo.10 Preplanned subgroup analyses demonstrated a statistically significant increase in survival for gefitinib compared with placebo in patients of Asian origin (hazard ratio [HR], 0.66; 95% CI, 0.48 to 0.91; P = .010; median survival, 9.5 v 5.5 months) and in never-smokers (HR, 0.67; 95% CI, 0.49 to 0.92; P = .012; median survival, 8.9 v 6.1 months).10,11
Reported here is the first phase III study to compare the effects of targeted therapy (gefitinib) with chemotherapy (docetaxel) on overall survival in Japanese patients with advanced/metastatic (stages IIIB to IV) or recurrent NSCLC who failed one or two chemotherapy regimens.
METHODS
Study Design
This multicenter, randomized, open-label, postmarketing clinical study (V-15-32) compared gefitinib with docetaxel in Japanese patients who had pretreated, locally advanced/metastatic (stages IIIB to IV) or recurrent NSCLC. Patients were randomly assigned by using stratification factors of sex (female v male), performance status (PS; 0 to 1 v 2), histology (adenocarcinoma v others), and study site.
The primary end point was overall survival, and the study aimed to show noninferiority of gefitinib versus docetaxel. Secondary end points were progression-free survival (PFS), time to treatment failure, objective response rate (ORR), disease control rate (DCR), quality of life (QoL), disease-related symptoms, safety, and tolerability.
A late protocol amendment included exploratory end points, such as EGFR gene copy number, protein expression, and mutation status of tumor tissue.
Patients
Patients age 20 years or older were eligible if they had the following: histologically or cytologically confirmed NSCLC (stages IIIB to IV) not amenable to curative surgery or radiotherapy, or postoperative recurrent NSCLC; failure of prior treatment with one or two chemotherapy regimens (≥ 1 platinum-based regimen); life expectancy of 3 months or greater; WHO PS 0 to 2; and measurable disease by Response Evaluation Criteria in Solid Tumors (RECIST). To improve recruitment, the protocol was amended approximately 6 months after study initiation to allow patients without measurable lesions to participate. This was not expected to greatly impact the primary end point.
Treatment
Gefitinib 250 mg/d was administered orally; docetaxel was administered every 3 weeks as a 1-hour intravenous infusion of 60 mg/m2 (ie, the approved dose in Japan). Patients received treatment until disease progression, intolerable toxicity, or discontinuation for another reason. Poststudy treatment was at physician and patient discretion; a switch to other study treatment was prohibited unless requested by the patient.
Assessments
Overall survival was assessed from date of random assignment to date of death as a result of any cause, or data were censored at the last date the patient was known to be alive. Tumor response by RECIST was performed at baseline, every 4 weeks for the first 24 weeks, and every 8 weeks thereafter. Complete response (CR) or partial response (PR) was confirmed on the basis of two consecutive examinations that were at least 28 days apart. Investigator assessment of best overall tumor response was used for the primary analysis; sensitivity analyses were performed with independent response evaluation committee assessment. PFS was defined as the time from random assignment to the earliest occurrence of disease progression or death from any cause; patients who had not progressed or died at data cutoff were censored at last tumor assessment. QoL was assessed with the FACT-L questionnaire at baseline and every 4 weeks during study treatment until week 12. The FACT-L total score and trial outcome index (TOI; sum of FACT-L physical well-being + functional well-being + additional concerns subscales) were calculated. Disease-related symptoms were assessed weekly with the FACT-L lung cancer subscale (LCS). Improvement was defined as an increase from baseline of at least six points for FACT-L or TOI, or an increase of at least two points for LCS, on two visits that were at least 28 days apart. Adverse events (AEs) were monitored and graded according to the National Cancer Institute Common Toxicity Criteria (NCI-CTC; version 2.0). Routine laboratory assessments were performed. EGFR gene copy number was determined by fluorescent in situ hybridization (FISH).12 EGFR mutations were assessed by direct sequencing of exon 18 to 21 of chromosome 7. EGFR protein expression was measured by immunohistochemistry with the DAKO EGFR pharmaDxTM kit (DAKO, Glostrup, Denmark).10
Statistical Analysis
The primary overall survival analysis was conducted in the intent-to-treat (ITT) population by estimating the HR and two-sided 95.24% CI for gefitinib versus docetaxel, derived from a Cox regression model without covariates (significance level adjusted because of interim analysis). Noninferiority was to be concluded if the upper CI limit was ≤ 1.25. Superiority was concluded if the upper CI limit was less than 1. A total of 296 death events were required for 90% power to demonstrate noninferiority, with the assumption that gefitinib had better overall survival than docetaxel (median survival, 14 v 12 months4), and the study plan was to recruit 484 patients.
Robustness of the primary conclusion was assessed by supportive analyses in the per-protocol population and by using a Cox regression model with covariate adjustment for sex (male v female), PS (0 or 1 v 2), tumor type (adenocarcinoma v other), smoking history (ever v never), number of prior chemotherapy regimens (1 v 2), age at random assignment (< 65 years v ≥ 65 years), time from diagnosis to random assignment (< 6 v 6 to 12 v > 12 months), and best response to prior chemotherapy (CR/PR v stable disease [SD] v progressive disease not assessable/unknown).
Preplanned subgroup analyses were performed on the basis of these covariates. Subgroups were first assessed for evidence of randomized treatment effect by subgroup interactions, to ensure that outcomes between subgroups were likely to be different; then, the subgroups for which evidence existed were examined further.
For PFS, the HR and its 95% CI for gefitinib versus docetaxel were calculated for the population that was assessable for response (defined as patients with ≥ 1 measurable lesion at baseline by RECIST) by using a Cox regression model without covariates. Supportive analyses were performed in the ITT population by using a model adjusted for covariates. Overall survival and PFS were summarized with Kaplan-Meier methods.
The ORR (proportion of CR + PR) and the DCR (proportion of CR + PR + SD ≥ 12 weeks) were estimated in the assessable-for-response population and were compared between treatments by generating an odds ratio and a 95% CI from a logistic regression model that included covariates.
The exploratory analysis of biomarker subgroups was performed with similar methods to the overall and clinical subgroup analyses when possible.
RESULTS
Patients
From September 2003 to January 2006, 490 patients were randomly assigned from 50 institutes. In the ITT population, 245 patients were randomly assigned to gefitinib, and 244 patients were randomly assigned to docetaxel; one patient was excluded because of a Good Clinical Practice violation (Fig 1). Treatment groups were generally well balanced for baseline demographics (Table 1), except for some small imbalances in smoking history (7% fewer never-smokers and 10% more ex-smokers in the gefitinib arm). The overall population was representative of an advanced, pretreated NSCLC population in a clinical trial setting in Japan. The median (range) duration of treatment for gefitinib was 58.5 (4 to 742) days and, for docetaxel, was 3 (1 to 12) cycles.
Poststudy, 36% of gefitinib-treated patients received subsequent docetaxel, and 40% received no other therapy except for gefitinib; 53% of docetaxel-treated patients received subsequent gefitinib, and 26% received no other therapy except for docetaxel.
Survival
At data cutoff for overall survival (October 31, 2006), overall mortality was 62.6%, and median follow-up was 21 months. Noninferiority in overall survival was not achieved (HR, 1.12; 95.24% CI, 0.89 to 1.40) according to the predefined criterion (upper CI limit for HR ≤ 1.25). However, no statistically significant difference in overall survival was apparent (P = .330; Fig 2A).
A supportive Cox analysis, which took into account imbalances in known prognostic factors, showed an HR of 1.01 (95% CI, 0.80 to 1.27; P = .914), which suggested that a demography imbalance that favored docetaxel may have had some impact on the primary, unadjusted, overall survival result.
The median survival and the 1-year survival rates were 11.5 months and 47.8%, respectively, for gefitinib and were 14.0 months and 53.7%, respectively, for docetaxel.
PFS
There was no significant difference between treatments in PFS in the unadjusted analysis (HR, 0.90; 95% CI, 0.72 to 1.12; P = .335); median PFS was 2.0 months with both treatments (Fig 2B). Similar PFS results were obtained from supportive Cox regression analysis adjusted for covariates (HR, 0.81; 95% CI, 0.65 to 1.02; P = .077).
Tumor Response
For ORR, gefitinib was statistically superior to docetaxel (22.5% v 12.8%; odds ratio, 2.14; 95% CI, 1.21 to 3.78; P = .009; Table 2). Gefitinib was similar to docetaxel in terms of DCR (34.0% v 33.2%; odds ratio, 1.08; 95% CI, 0.69 to 1.68; P = .735). The primary ORR results that were based on investigator judgment were generally consistent with those obtained from independent response evaluation committee assessment.
Symptom Improvement and QoL
Gefitinib showed statistically significant benefits compared with docetaxel in QoL improvement rates (FACT-L: 23.4% v 13.9%; P = .023; TOI: 20.5% v 8.7%; P = .002; Table 2), but there were no significant differences between treatments in LCS improvement rates (22.7% v 20.4%; P = .562).
Subgroup Analyses
Survival outcomes were generally consistent across subgroups, with the exception of best response to prior chemotherapy (treatment by subgroup interaction test P = .017). For patients with best response to prior chemotherapy of progressive disease, overall survival was numerically longer on gefitinib than on docetaxel, whereas patients with a best response of SD had significantly longer survival on docetaxel than on gefitinib (HR, 1.58; 95% CI, 1.09 to 2.27; P = .015; Fig 3A). However, the result was not supported by the PFS (Fig 3B) or ORR results in this subgroup, which favored gefitinib.
Safety
Gefitinib was associated with fewer dose interruptions or delays than docetaxel (26% v 52%, respectively). There were no clinically relevant differences in the frequencies of serious AEs or discontinuations of study treatment as a result of AEs between treatment groups (Table 3). Fewer NCI-CTC grades 3 to 4 AEs occurred with gefitinib compared with docetaxel (40.6% v 81.6%). There were four deaths as a result of AEs in the gefitinib arm (three as a result of interstitial lung disease that was considered by the investigator to be treatment related; one as a result of pneumonia that was not considered treatment-related), and none in the docetaxel arm.
The most common AEs with gefitinib were rash/acne (76.2%) and diarrhea (51.6%), and the most common AEs with docetaxel were neutropenia (79.5%) and alopecia (59.4%; Table 4). There was a higher incidence of grades 3 to 4 neutropenia with docetaxel (73.6%) compared with gefitinib (8.2%). Interstitial lung disease events occurred in 5.7% (n = 14) and 2.9% (n = 7) of patients who received gefitinib and docetaxel, respectively (Table 3).
Biomarkers
Of the 74 EGFR biomarker samples provided, 53 to 60 were assessable (depending on biomarker). Because of the late protocol amendment, these samples were from long-term survivors who were recruited early or from patients who were recruited later in the study. Compared with the overall study population, this subgroup was over-representative of some stratification factors on both treatment arms: good PS, females, never-smokers, greater than 12 months from diagnosis to random assignment, and best response to prior chemotherapy of CR/PR. There were insufficient events to allow meaningful evaluation of overall survival in relation to biomarker status, and the PFS and ORR data should be interpreted with caution.
Thirty-one (54.4%) of 57 patients had EGFR mutation–positive tumors, and 42 (70.0%) of 60 had EGFR FISH–positive tumors. There was a high degree of overlap between EGFR mutation and clinical characteristics (eg, high frequency in females, in those with adenocarcinoma, and in never-smokers). EGFR mutation–positive patients appeared to have better PFS than EGFR mutation–negative patients on both treatments (gefitinib-positive v gefitinib-negative HR, 0.33; 95% CI, 0.11 to 0.97; 17 events; docetaxel HR, 0.15; 95% CI, 0.04 to 0.57; 15 events). In addition, EGFR FISH–positive patients appeared to have better PFS than EGFR FISH–negative patients on both treatments (gefitinib-positive v gefitinib-negative HR, 0.75; 95% CI, 0.28 to 1.98; 18 events; docetaxel HR, 0.45; 95% CI, 0.14 to 1.41; 16 events). There were no clear PFS differences between gefitinib and docetaxel in any biomarker subgroups, although the number of events was small and the CIs for the HRs were wide. PFS could not be assessed for EGFR protein expression because of the small number of events in the expression-negative group. For EGFR mutation–positive patients, the ORR was 67% (six of 9 patients) with gefitinib administration and 46% (five of 11 patients) with docetaxel administration. For EGFR FISH–positive patients, the ORR was 46% (five of 11) with gefitinib administration and 33% (six of 18) with docetaxel administration. For EGFR expression–positive patients, the ORR was 36% (five of 14) with gefitinib administration and 31% (four of 13) with docetaxel administration. There were no responses among EGFR mutation–negative, or EGFR FISH–negative, patients, and there was one response (13%) of eight EGFR expression–negative patients who received docetaxel.
DISCUSSION
V-15-32 is the first phase III study to compare gefitinib versus docetaxel in previously treated Japanese patients who have advanced NSCLC. Both gefitinib and docetaxel demonstrated efficacy and tolerability, and findings were consistent with previous experience for both agents in Japan.
Although noninferiority in overall survival for gefitinib versus docetaxel was not proven, there was no statistically significant difference between the two treatments. The original statistical assumption was that gefitinib would have 20% longer survival than docetaxel; hence, the relatively small sample size for a noninferiority study. However, since the study was initiated, data from postmarketing experience in Japan (the SIGN study13) and substantial switching to the alternative study treatment on progression in V-15-32 indicated that it would be more likely that gefitinib and docetaxel had similar overall survival. With the assumption of equal survival, the chance (power) of showing noninferiority with this study size is reduced to 48%. The median survival with gefitinib 250 mg/d in our study was consistent with previous experience in Japan (11.5 v 13.8 months for Japanese subset of IDEAL 1).9 Docetaxel demonstrated a longer median survival in V-15-32 (14.0 months) compared with previous Japanese studies (7.8 to 9.4 months).1,4,14
In line with increasingly available therapy for NSCLC since the trial was designed and with standard practice in Japan, a large proportion of patients received additional anticancer therapy after discontinuation of the randomly assigned study treatment. Crossover was greater than initially expected, and differences in the number and types of patients who received these poststudy treatments complicated interpretation of survival results. A greater proportion of patients who received docetaxel received poststudy therapy compared with those who received gefitinib. Imbalances in the use of gefitinib after chemotherapy have been reported recently in a phase III study of Japanese patients with lung cancer who were treated with docetaxel and have been cited as a possible explanation for the prolonged median survival seen with docetaxel.15 INTEREST (Iressa NSCLC Trial Evaluating Response and Survival against Taxotere), a worldwide phase III trial that is comparing gefitinib with docetaxel in pretreated patients who have advanced NSCLC recently demonstrated that gefitinib had statistically noninferior survival to docetaxel.16 In contrast to V-15-32, INTEREST was larger (1,466 patients) and had subsequent therapies that were well-balanced between treatment arms.
Secondary end points, largely unaffected in this study by subsequent therapy, provided further evidence of the clinical efficacy of both gefitinib and docetaxel in Japanese patients. PFS was similar with gefitinib and docetaxel, and ORR was statistically significantly improved with gefitinib. The ORR in V-15-32 with gefitinib (22.5% v 12.8% with docetaxel) was consistent with a subset analysis from IDEAL 1 in Japanese patients (27.5%).3,8,9
A number of patient subgroups (including females, patients with adenocarcinoma, and never-smokers) have been reported previously to experience improved clinical benefit with gefitinib.2,4,7,8,10 Subgroup analyses in this study should be interpreted with caution, as the primary objective was not met, some subgroups were small, and there were imbalances in poststudy treatments. In between-treatment comparisons, no statistically significant overall survival benefit was found for gefitinib compared with docetaxel in any subgroup. However, when post hoc, within-treatment comparisons were performed, females, never-smokers, and patients with adenocarcinoma (and also patients with poor PS and > 12 months since diagnosis) had significantly longer survival than their opposite subgroups on both gefitinib and docetaxel (P < .001 for females v males, adenocarcinoma v others, and never-smokers v ever-smokers on both treatments). It appears that the subgroups typically associated with a gefitinib benefit were seen but that they also did well on docetaxel. However, the rate of subsequent gefitinib prescription in the docetaxel arm was high in these subgroups (eg, approximately two-thirds of docetaxel never-smokers and females had gefitinib as their first poststudy treatment); for PFS and ORR, which are largely unaffected by subsequent treatment, the benefit in these subgroups remained for gefitinib but not for docetaxel, which suggested that poststudy treatments are confounding the interpretation of overall survival in the subgroups.
AEs in our study were consistent with those previously observed, and the most commonly reported AEs were rash/acne and diarrhea for gefitinib and neutropenia for docetaxel. Docetaxel demonstrated a typically high incidence of neutropenia (79.5%) and febrile neutropenia (7.1%) compared with gefitinib (9.8% and 1.6%, respectively). These neutropenia levels that accompanied docetaxel treatment are consistent with previously reported studies in Japanese patients (95.4%1 and 81.5%4). The incidence of interstitial lung disease reported in this study with gefitinib (5.7%) is consistent with that reported in the Japanese postmarketing study (5.8%).17
Although the patient numbers were too small for firm conclusions, the biomarker data from this study suggest that EGFR mutation–positive or EGFR FISH–positive patients have a greater response to both gefitinib and docetaxel compared with EGFR mutation–or FISH–negative patients. The gefitinib data are consistent with several previous reports.18 The docetaxel data provide potential new information about EGFR biomarkers and chemotherapy; this has not been consistently seen before, because there are only a few small studies in the literature, and they have conflicting results.19 Hence, it is difficult to say conclusively that EGFR mutation or EGFR FISH–positivity predict for docetaxel as well as gefitinib benefit.
Although the study did not prove noninferior survival for gefitinib compared with docetaxel in this patient population, the clinical efficacy and tolerability of gefitinib 250 mg/d in Japanese patients who had NSCLC, reported here, is consistent with the clinical experience reported to date, and gefitinib remains an effective treatment option for previously treated Japanese patients who have locally advanced/metastatic NSCLC.
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Employment or Leadership Position: Haiyi Jiang, AstraZeneca (C); Yohji Itoh, AstraZeneca (C) Consultant or Advisory Role: None Stock Ownership: Haiyi Jiang, AstraZeneca; Yohji Itoh, AstraZeneca; Nagahiro Saijo, Takeda Honoraria: Yutaka Nishiwaki, AstraZeneca; Tomohide Tamura, AstraZeneca; Nobuyuki Yamamoto, AstraZeneca, Novartis; Masahiro Tsuboi, AstraZeneca, Bristol-Myers Squibb, Taiho, Sanofi-Aventis, Eli Lilly; Kazuhiko Nakagawa, AstraZeneca, Sanofi-Aventis; Tetsu Shinkai, AstraZeneca; Shunichi Negoro, AstraZeneca, Sanofi-Aventis; Kenji Eguchi, AstraZeneca, Chugai; Noriyuki Masuda, AstraZeneca; Yukito Ichinose, AstraZeneca; Nagahiro Saijo, AstraZeneca, Sanofi-Aventis, Novartis, Taiho, Chugai, Eli Lilly; Masahiro Fukuoka, AstraZeneca, Chugai, Eizai, Eli Lilly Research Funding: None Expert Testimony: None Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Yutaka Nishiwaki, Shunichi Negoro, Nagahiro Saijo, Masahiro Fukuoka
Administrative support: Haiyi Jiang, Yohji Itoh
Provision of study materials or patients: Riichiroh Maruyama, Yutaka Nishiwaki, Tomohide Tamura, Nobuyuki Yamamoto, Masahiro Tsuboi, Kazuhiko Nakagawa, Tetsu Shinkai, Shunichi Negoro, Fumio Imamura, Kenji Eguchi, Koji Takeda, Akira Inoue, Keisuke Tomii, Masao Harada, Noriyuki Masuda, Yukito Ichinose
Collection and assembly of data: Riichiroh Maruyama, Yutaka Nishiwaki, Tomohide Tamura, Nobuyuki Yamamoto, Masahiro Tsuboi, Kazuhiko Nakagawa, Tetsu Shinkai, Shunichi Negoro, Fumio Imamura, Kenji Eguchi, Koji Takeda, Akira Inoue, Keisuke Tomii, Masao Harada, Noriyuki Masuda, Yukito Ichinose
Data analysis and interpretation: Yutaka Nishiwaki, Shunichi Negoro, Haiyi Jiang, Yohji Itoh, Nagahiro Saijo, Masahiro Fukuoka
Manuscript writing: Riichiroh Maruyama, Haiyi Jiang, Yohji Itoh
Final approval of manuscript: Yukito Ichinose, Nagahiro Saijo, Masahiro Fukuoka
Appendix
Time to treatment failure results.
Time to treatment failure (TTF) was assessed from date of random assignment to the earliest date of either diagnosis of progressive disease or treatment discontinuation, or to death as a result of any cause (in the assessable-for-response [AFR] population). The hazard ratio (HR) and its 95% CI for gefitinib versus docetaxel were calculated for the AFR population by using a Cox regression model without covariates. Data were censored at the last date of treatment administration for patients who were lost to follow-up or who were still receiving treatment at the final analysis.
Gefitinib showed statistically significant benefits compared with docetaxel in TTF (HR, 0.63; 95% CI, 0.51 to 0.77; P < .001; Appendix Fig A1). Supportive Cox regression analysis for TTF showed similar results to the primary analysis (HR, 0.54; 95% CI, 0.44 to 0.68; P < .001). Another supportive analysis that used the ITT population also showed similar results (HR, 0.61; 95% CI, 0.51 to 0.74; P < .001). Differences in administration and timings of both treatments and some subjective factors influenced the decision to continue treatment; therefore, caution is required in the interpretation of TTF results.
Additional inclusion criteria and the main exclusion criteria.
Additional inclusion criteria included the following: WBC count of 4.0 to 12.0 × 109 cells/L; neutrophil count ≥ 2.0 × 109 cells/L; platelet count ≥ 100 × 109 cells/L; serum bilirubin ≤ 1.5 mg/dL; ALT or AST ≤ 2.5 × upper limit of the reference range; serum creatinine ≤ 1.5 mg/dL; and arterial oxygen tension ≥ 70 torr.
Patients were excluded if they had any of the following: received their last chemotherapy (including intrapleural administration) or radiotherapy within 4 weeks before enrollment; received prior treatment with a docetaxel-containing regimen or any anti-EGFR therapy; an allergy or suspected allergy to gefitinib or docetaxel; other coexisting malignancies diagnosed within the last 5 years, with exceptions; any unresolved chronic toxicity greater than NCI-CTC grade 2 from previous anticancer therapy; any evidence of severe or uncontrolled systemic diseases, as judged by the investigator; current status of pregnancy or breastfeeding; treatment with a nonapproved or investigational drug within 30 days before enrollment; intracerebral metastases; significant malabsorption syndrome; past history of, or concurrent, interstitial lung disease, idiopathic pulmonary fibrosis, or pneumoconiosis, or radiation pneumonia or drug-induced pneumonia that required corticosteroids; fever with suspected infection; or treatment with systemic corticosteroids for ≥ 4 weeks.
Gefitinib discontinuations and docetaxel premedication and dose reductions.
Temporary discontinuations of gefitinib for up to 14 consecutive days were allowed if a patient experienced either grade 3 diarrhea or a grade 3 to 4 nonhematologic toxicity until the event improved to grade 1 to 2. If a patient experienced grade 4 diarrhea, gefitinib was discontinued permanently. If a patient experienced dyspnea, fever, or cough, or if there was any change in these symptoms, a computed tomography or high-resolution computed tomography scan was performed for diagnosis of interstitial pneumonia. If diagnosed, gefitinib was stopped immediately, and steroid therapy administered. Gefitinib treatment was only resumed when the development of interstitial lung disease had been excluded.
Premedication against shock symptoms, anaphylactic reaction, and edema by docetaxel was allowed at the investigators’ discretion, according to Japanese prescribing guidelines. The docetaxel dose could be reduced to 50 mg/m2 if a patient experienced any of the following toxicities: WBC count less than 1.0 × 109 cells/L; grades 3 to 4 neutropenia associated with a fever of ≥ 38°C; platelet count less than 20 × 109 cells/L; or any other grade 3 nonhematologic toxicity (excluding anorexia). No re-escalation of the docetaxel dose was allowed. If further dose reduction was required, the patient was discontinued from the study.
Preplanned interim analysis.
When approximately half of the target number of events (155) were reached, an independent data monitoring committee conducted a preplanned interim analysis of overall survival and recommended that the study should continue as planned. AstraZeneca remained blind to the interim analysis results. To account for the interim analysis and to maintain the overall type I error rate at 5%, the significance level for the final analysis was set at 4.76% (two-sided), and corresponding 95.24% confidence intervals were based on the α-spending function of an O'Brien-Fleming–type error (O'Brien PC, Fleming TR: Biometrics 35:549-556, 1979).
Additional overall survival and progression-free survival analyses.
Supportive analysis of overall survival by using the per-protocol data showed similar results to primary analysis in the ITT population (HR, 1.12; 95% CI, 0.89 to 1.14; P = .310).
Supportive analysis of progression-free survival with the ITT population also showed similar results to the primary analysis in the AFR population (HR, 0.89; 95% CI, 0.73 to 1.09; P = .248).
Investigators and institutions.
M. Harada (Hokkaido Cancer Center); Y. Fujita (Dohoku National Hospital); H. Nakagawa (Hirosaki National Hospital); A. Inoue (Tohoku University Hospital); S. Sugawara (Sendai Kousei Hospital); T. Tsukamoto (Yamagata Prefectural Central Hospital); Y. Nishiwaki (National Cancer Center Hospital East); T. Tamura (National Cancer Center Hospital); T. Horai (Japanese Foundation for Cancer Research Cancer Institute Hospital); M. Tsuboi (Tokyo Medical University Hospital); M. Shibuya (Tokyo Metropolitan Komagome Hospital); K. Yoshimura (Toranomon Hospital); A. Tamura (Tokyo Hospital); R. Ieki (Metropolitan Bokutoh Hospital); K. Watanabe (Yokohama Municipal Citizen's Hospital); K. Noda (Kanagawa Cancer Center); K. Eguchi (Tokai University Hospital); N. Masuda (Kitasato University School of Medicine); T. Ogura (Kanagawa Cardiovascular and Respiratory Center); A. Yokoyama (Niigata Cancer Center); T. Kasahara (Kanazawa University Hospital); N. Yamamoto (Shizuoka Cancer Center); K. Yasuda (Iwata City Hospital); T. Mitsudomi and T. Hida (Aichi Cancer Center Hospital); H. Saito (Aichi Hospital Aichi Cancer Center); H. Saka (Nagoya Medical Center); H. Ueshima (Rinku General Medical Center Municipal Izumisano Hospital); K. Nakagawa (Kinki University School of Medicine); M. Kawahara (Kinki-chuo Chest Medical Center); K. Takeda and H. Tada (Osaka City General Hospital); S. Kudo (Osaka City University Hospital); K. Matsui (Osaka Prefectural Medical Center for Respiratory and Allergic Disease); F. Imamura (Osaka Medical Center for Cancer and Cardiovascular Disease); K. Komuta (Osaka Police Hospital); K. Tomii (Kobe City General Hospital); S. Negoro (Hyogo Cancer Center); N. Katakami (Institute of Biomedical Research and Innovation); K. Kiura (Okayama University Medical and Dental School Hospital); T. Maeda (Sanyo National Hospital); T. Shinkai (Shikoku Cancer Center); Y. Ichinose (Kyushu Cancer Center); K. Takayama (Kyusyu University Hospital); H. Souda (Nagasaki University Hospital); A. Kinoshita (Nagasaki Medical Center); M. Matsumoto (Kumamoto University Hospital); and H. Yokoyama (Beppu Medical Center).
Additional acknowledgments.
We thank Alison Armour, MD, Medical Science Director, AstraZeneca; and Claire Watkins, MSc, Global Statistician, AstraZeneca, for critical review and input to the manuscript and assistance with data analysis and interpretation.
Time to treatment failure in the assessable-for-response population. HR, hazard ratio.
Study flow. (*) Allocated to the docetaxel group. (†) The safety analysis, conducted according to treatment received, was performed on this population. ITT, intent to treat; GCP, Good Clinical Practice; DCO, data cutoff date for overall survival (October 31, 2006); QoL, quality of life; LCS, Lung Cancer Subscale.
(A) Overall survival in the intent-to-treat population; (B) Progression-free survival (PFS) in the assessable-for-response population. HR, hazard ratio.
Forest plots of (A) overall survival and (B) progression-free survival that compare treatment groups within clinically relevant subgroups. HR, hazard ratio; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; NE, not assessable.
Baseline Patient Characteristics in Intent-to-Treat Population
Response Rates and Improvement Rates
Summary of Adverse Event Data in the Assessable-for-Safety Population
Most Common Adverse Events
Acknowledgments
We thank all the patients and investigators who participated in the V-15-32 study. We also thank Ann Gordon, PhD, from Complete Medical Communications, who provided medical writing support funded by AstraZeneca.
Footnotes
-
Supported by AstraZeneca.
-
Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
-
Clinical trial information can be found for the following: NCT00252707.
- Received November 5, 2007.
- Accepted March 14, 2008.
