Cancer Chemother Pharmacol (2014) 73:139–149 DOI 10.1007/s00280-013-2332-5
Original Article
Enhanced efficacy of postoperative adjuvant chemotherapy in advanced gastric cancer: results from a phase 3 randomized trial (AMC0101) Yoon‑Koo Kang · Jeong Hwan Yook · Heung‑Moon Chang · Min‑Hee Ryu · Changhoon Yoo · Dae Young Zang · Jae‑Lyun Lee · Tae Won Kim · Dae Hyun Yang · Se Jin Jang · Young Soo Park · Young Jack Lee · Hwoon‑Yong Jung · Jin‑Ho Kim · Byung Sik Kim
Received: 28 April 2013 / Accepted: 17 October 2013 / Published online: 27 October 2013 © Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose To improve the efficacy of adjuvant chemotherapy with mitomycin-C and fluoropyrimidine (Mf) in gastric cancer, we designed a new regimen (iceMFP) and investigated in a phase III study. Methods We randomly assigned 640 patients with resectable and macroscopically recognizable serosa-invading gastric cancer to Mf or iceMFP group during operation. The Mf consisted of intravenous mitomycin-C (20 mg/m2) at 3–6 weeks after surgery and oral doxifluridine (460– 600 mg/m2/day) starting 4 weeks after the administration of mitomycin-C and continuing for 3 months. The iceMFP consisted of intraoperative intraperitoneal cisplatin (100 mg), intravenous mitomycin-C (15 mg/m2) on This study was presented in part at the 44th ASCO annual meeting, May 30–June 3, 2008, in Chicago, USA, the 36th ESMO annual congress, September 23–27, 2011, in Stockholm, Sweden, and 2012 Gastrointestinal Cancers Symposium, January 19–21, 2012, in San Francisco, USA. Electronic supplementary material The online version of this article (doi:10.1007/s00280-013-2332-5) contains supplementary material, which is available to authorized users. Y.-K. Kang (*) · H.-M. Chang · M.-H. Ryu · C. Yoo · J.-L. Lee · T. W. Kim Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic‑ro 43‑gil, Songpa‑gu, Seoul 138‑736, Korea e-mail: [email protected]
postoperative day 1, followed by oral doxifluridine for 12 months, and six monthly intravenous cisplatin (60 mg/ m2). The primary endpoint was 3-year recurrence-free survival (RFS). Results A total of 521 patients (258 in Mf, 263 in iceMFP) were eligible for analysis after excluding patients with stage I disease, distant metastasis, or R1 resection. With a median follow-up of 3.5 years, the iceMFP group had a higher RFS (hazard ratio [HR] 0.70; 95 % confidence interval [CI] 0.54–0.90; p = 0.006; 3-year RFS 60 % vs. 50 %) and overall survival (HR 0.71; 95 % CI 0.53–0.95; p = 0.02; 3-year overall survival, 71 vs. 60 %) compared with the Mf group. This was confirmed at extension analysis after a median 6.6 years of follow-up. Both regimens were well tolerated with no differences in surgical complications. Conclusion The efficacy of adjuvant Mf was significantly improved by the additional therapeutic strategies of iceMFP. Considering negative results of AMC0201, these suggest that early initiation of chemotherapy and/or intraperitoneal cisplatin played a distinct role in the improved efficacy. D. H. Yang Department of Surgery, Hallym University Medical Center, Hallym University College of Medicine, Anyang, Korea S. J. Jang · Y. S. Park Department of Pathology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
J. H. Yook · B. S. Kim Department of Surgery, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
Y. J. Lee LSK Global Pharma Services, Seoul, Korea
D. Y. Zang Department of Internal Medicine, Hallym University Medical Center, Hallym University College of Medicine, Anyang, Korea
H.-Y. Jung · J.-H. Kim Department of Gastroenterology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
13
140
Keywords Mitomycin-C · Doxifluridine · Cisplatin · Intraperitoneal chemotherapy · Adjuvant chemotherapy · Advanced gastric cancer
Introduction After much debate, it is now globally agreed that adjuvant treatment improves survival of patients with advanced gastric cancer (AGC) undergoing curative surgery. However, there are still geographical differences in standard adjuvant treatments. The intergroup-0116 study of postoperative chemoradiation [1] and the MAGIC trial of perioperative chemotherapy [2] demonstrated the survival benefits in AGC over surgery alone and have become standard adjuvant therapy in the United States and Europe, respectively. In Asia, where D2 surgery is a standard procedure, postoperative chemotherapy with S-1 alone or combination of capecitabine and oxaliplatin has become standard adjuvant therapy on the basis of the results of the ACTS-GC [3] and the recent CLASSIC trials [4]. In 2000, when we designed this trial, adjuvant chemotherapy already was the community standard in Korea and Japan [5, 6]. Although there had been no pivotal study to prove the benefit of adjuvant chemotherapy, several metaanalyses had consistently suggested small but significant benefits of adjuvant chemotherapy [7, 8]. Therefore, we could not consider a clinical trial of adjuvant treatment without a treatment control arm. Instead, we tried to further improve the efficacy of adjuvant chemotherapy with the addition of therapeutic strategies available and suggested to be effective at that time. Combination regimens of mitomycin-C and 5-fluorouracil had been widely investigated and commonly used as adjuvant chemotherapy in Asia [5, 6, 9]. The efficacy of these regimens over surgery alone was subsequently confirmed in recent meta-analysis conducted by the GASTRIC group [10]. In a Spanish phase III study, a combination of mitomycin-C and 3 months of oral fluoropyrimidine (Mf) significantly prolonged survival of patients with resected stage III AGC compared with surgery alone [11]. Although this study was limited by small sample size, potential efficacy and feasibility of Mf regimen was well presented. We therefore adopted the Mf regimen as the control treatment for our adjuvant trials. We considered the following four strategies to enhance the efficacy of the Mf regimen: 1. to add cisplatin since it is active and synergistic with fluoropyrimidine in metastatic disease; 2. to prolong the duration of oral fluoropyrimidine; 3. to initiate the adjuvant chemotherapy immediately after surgery; 4. to add intraperitoneal chemotherapy to prevent the peritoneal recurrence which is common in gastric cancer, especially in cases with serosal involvement. In a companion study (AMC0201), we
13
Cancer Chemother Pharmacol (2014) 73:139–149
evaluated the efficacy of adding the first two strategies to the Mf regimen in patients with postoperative stage II–IV (without distant metastasis) gastric cancers [12]. The AMC0101 study was designed to evaluate the efficacy of adding all the above four strategies to the Mf regimen in patients with AGC macroscopically involving serosa. The result of primary analysis in this study was previously presented in part at the ASCO annual meeting, 2008 [13]. Here we report the final results of the AMC0101 study with inclusion of the results from 3-year extension analysis for the assessment of long-term outcome.
Materials and methods AMC 0101 was a randomized, open-label, phase III trial, conducted at two institutions in Korea. The protocol was approved by the institutional review board at each institution, and the study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the Good Clinical Practice guidelines, as defined by the International Conference on Harmonization. This study was registered in ClinicalTrials.gov (NCT00296322). All patients provided written informed consent before surgery. Patients Patients with histologically proven gastric adenocarcinoma were eligible if their tumors could be curatively resected with D2 dissection. During the surgery, macroscopic involvement of serosal layer was determined as disruption of serosal continuity by the inspection and palpation of the operating surgeon. If this was recognized, patients were entered to this study and randomized. Other inclusion criteria were age between 18 and 70 years, an Eastern Cooperative Oncology Group (ECOG) performance status of 0–2, and adequate hematologic, hepatic, and renal function. Patients who initially entered into this study but who had a pathologically positive resection margin (R1 resection), postoperative stage I or IV with distant metastasis were excluded. Study treatment During surgery, the eligible patients were randomly assigned in a 1:1 ratio to the control (Mf) or experimental (iceMFP) arms. Randomization was stratified by hospital and based on a permutation block method. Patients and investigators were not masked to treatment allocation. For patients assigned to the Mf arm, 20 mg/m2 intravenous mitomycin-C was given 3–6 weeks after surgery, and 4 weeks later, followed by 3 months of 460 (600 after interim analysis) mg/m2/day oral doxifluridine. For patients
Cancer Chemother Pharmacol (2014) 73:139–149
assigned to the iceMFP arm, 100 mg of cisplatin in 1 liter of normal saline was placed intraperitoneally for 2 h before closure of the abdomen, and 15 mg/m2 mitomycin-C was given intravenously on postoperative day 1. Beginning 4 weeks after surgery, 60 mg/m2 intravenous cisplatin was given monthly for 6 months and 460 (increased to 600 after interim analysis) mg/m2/day oral doxifluridine was administered for 12 months. In interim analysis to evaluate safety, performed in February 2004, a less than expected toxicity of doxifluridine prompted us to increase the dose of doxifluridine to 600 mg/m2/day. Assessment At baseline, patients underwent a history, physical examination, computed tomography (CT) scan of the abdomen and pelvis with contrast enhancement, chest radiography, and laboratory tests, including complete blood count, chemistry, and coagulation tests. After surgery, all resected specimens were staged at local pathology laboratories according to the American Joint Committee on Cancer (AJCC) 6th edition. Adverse events and laboratory profiles were assessed every month during treatment. After completion of chemotherapy, clinical assessments were performed every 3 months within 2 years from surgery, every 6 months between 2 and 5 years from surgery, and every 1 year thereafter. To detect recurrence, plain chest radiography and abdominopelvic CT scans were performed every 6 months for 5 years from surgery and every 1 year thereafter, and gastroscopy, every 1 year. If there was any sign or symptom indicating recurrence, investigations were then done to verify whether patients were recurrence-free. If findings on imaging studies were suggestive but not conclusive, serial follow-up studies were performed to verify the recurrence, and the date of the first recognition of findings suggestive of recurrence was defined as the date of recurrence. Adverse events were graded according to the National Cancer Institute’s Common Toxicity Criteria, version 2.0. Statistical analysis The primary endpoint was 3-year recurrence-free survival (RFS), and secondary endpoints were overall survival (OS), disease recurrence, and safety. This trial was designed for 527 patients to be randomly assigned to receive either Mf or iceMFP with a power of 80 % to detect a hazard ratio of 0.6574 for RFS (an improvement of 3-year RFS rate from 55 % for Mf to 67.5 % for iceMFP). Final analysis was planned after 192 events occurred (recurrences or deaths). Safety interim analysis was scheduled to review the toxicity profiles of study regimens after half of the planned number (260) of patients were enrolled. RFS was defined as the
141
time from randomization to documented disease recurrence or death, and OS was defined as the time from randomization to death from any cause. Patients were censored if they were recurrence-free or alive at the last follow-up. Kaplan– Meier curves were used to estimate survival and compared using the log-rank test on an intention-to-treat basis. The Cox regression model was used to estimate hazard ratios regarding the comparison of the efficacy between Mf and iceMFP arm in primary analysis and in prespecified subgroup analyses by potential prognostic factors. All tests were two sided, and a p value less than 0.05 was considered to indicate statistical significance. Confidence intervals (CI) were at the 95 % level. Descriptive analyses were used to represent adverse events. Statistical analyses were performed by a qualified biostatistician (YJL) using SAS version 9.1 (SAS Institute Inc, Cary, NC) and SPSS version 14.0 (SPSS, Chicago, IL). After 229 events occurred with a median follow-up period of 3.5 years, the database was initially frozen on March 20, 2008, and planned primary analysis was performed. Afterward, to verify the long-term results of this study, 3-year extension analysis was performed. The data cutoff for this extension analysis was April 11, 2011. Median follow-up at that time was 6.6 years (maximum, 114.1 months).
Results Between October 2001 and April 2007, 640 patients were randomly assigned to the Mf (318) or iceMFP (322) arms. As prespecified in the protocol, a total of 119 patients (60 in Mf and 59 in iceMFP) were excluded after randomization because of ineligibility. Therefore, a total of 521 patients (258 in Mf and 263 in iceMFP) were included in the analysis for efficacy (Fig. 1). The most common reason for study exclusion was inadequate pathologic stage (stage I or IV with distant metastasis, or microscopic involvement of the resection margin). Table 1 shows that baseline characteristics were well balanced between the two groups. Systemic chemotherapy was administered at a median time from surgery of 22 days in the Mf and 1 day in the iceMFP group. At the time of primary analysis, the median followup was 36.2 months in the Mf group and 38.2 months in the iceMFP group. During this period, 118 (46 %) and 94 (36 %) patients had recurrence, and 103 (40 %) and 82 (31 %) died in the Mf and iceMFP groups, respectively. Compared with the Mf group, the hazard ratio for recurrence in the iceMFP group was 0.70 (95 % CI 0.54–0.90; p = 0.006), suggesting a 30 % risk reduction for recurrence. The RFS rate at 3 years was 60 % (95 % CI 54–67) in the iceMFP group and 50 % (95 % CI 43–57) in the Mf group (Fig. 2a). Median RFS was not reached in the iceMFP
13
142
Cancer Chemother Pharmacol (2014) 73:139–149
Fig. 1 CONSORT diagram
group, whereas it was 34.5 months (95 % CI 24.2–63.8) in the Mf group. Compared with the Mf group, the hazard ratio for death was 0.71 (95 % CI 0.53–0.95; p = 0.02) in the iceMFP group. The 3-year OS rate was 71 % (95 % CI 65–77) in the iceMFP group and 60 % (95 % CI 53–66) in the Mf group. Median OS was estimated as 69.7 months (95 % CI 55.0-not assessable) in the iceMFP group and 54.3 months (95 % CI 37.6-not assessable) in the Mf group (Fig. 2b). Benefits regarding disease recurrence were observed in most subgroups in the iceMFP arm (Fig. 3), and there was no significant interaction between treatment groups and baseline characteristics. RFS and OS were not affected by whether patients received 460 mg/m2/day (278 patients) or 600 mg/m2/day (243) as the starting dose of doxifluridine (p = 0.23 and p = 0.13, respectively). Locoregional recurrences occurred with similar frequency in both groups (6 % in Mf vs. 4 % in iceMFP). The iceMFP group had decreased distant and peritoneal recurrence compared with the Mf group (supplementary data). However, the difference was statistically borderline for distant recurrence (32 vs. 40 %; p = 0.05) and not significant for peritoneal recurrence (17 vs. 23 %; p = 0.08). There was no treatment-related death in both arms. Analyses for adverse events were available for 509 patients
13
(249 in the Mf group vs. 260 in the iceMFP group), and 12 patients (9 vs. 3) were excluded from the safety evaluation population because of not receiving the assigned treatment or loss to follow-up. Adverse events that occurred in 10 % or more of patients are summarized in Table 2. Overall, patients who received iceMFP had more adverse events of any grade than those receiving Mf. Grade 3 or 4 neutropenia was more frequent in the iceMFP (35 %) than in the Mf group (12 %). However, there was no febrile neutropenia. Among non-hematologic toxicities, grade 3 or 4 anorexia, nausea, and vomiting were more common in the iceMFP group. There was no surgical mortality. We could not find clinically significant increase in surgical complications related to the intraperitoneal and intensive chemotherapy (14 % in Mf and 10 % in iceMFP; p = 0.23; supplementary data). The dose of doxifluridine was reduced in 84 patients (34 %) of the Mf group and 176 (70 %) of the iceMFP group. The main reason for dose reduction of doxifluridine was hematologic toxicity (70 vs. 171). The dose of cisplatin was reduced in 6 patients (2 %) of the iceMFP group. The chemotherapy cycle was delayed in 33 patients (13 %) in the Mf group and 88 (35 %) in the iceMFP group. The median relative dose intensity was at least 96 and 78 % for
Cancer Chemother Pharmacol (2014) 73:139–149
143
Table 1 Baseline patient characteristics Characteristic
Mf group, n = 258
iceMFP group, n = 263
Median age, years (min–max) Male gender ECOG performance status 0–1 2 Primary sitea Proximal Distal Multiple/diffuse Type of surgery Total gastrectomy Subtotal gastrectomy Lauren’s classification Intestinal type Diffuse type Mixed type Unknown Tumor stage T1 T2 T3 T4 Nodal stage N0 N1 N2 N3 AJCC stage 6th ed. II IIIA IIIB IV (M0) Dose of doxifluridine 460 mg/m2/day
56 (29–69) 176 (68 %)
53 (25–69) 175 (67 %)
247 (96 %) 11 (4 %)
248 (94 %) 15 (6 %)
26 (10 %) 189 (73 %) 43 (17 %)
25 (9 %) 191 (73 %) 47 (18 %)
141 (55 %) 117 (45 %)
132 (50 %) 131 (50 %)
76 (29 %) 154 (60 %) 25 (10 %) 3 (1 %)
85 (32 %) 147 (56 %) 26 (10 %) 5 (2 %)
1 (
Original Article
Enhanced efficacy of postoperative adjuvant chemotherapy in advanced gastric cancer: results from a phase 3 randomized trial (AMC0101) Yoon‑Koo Kang · Jeong Hwan Yook · Heung‑Moon Chang · Min‑Hee Ryu · Changhoon Yoo · Dae Young Zang · Jae‑Lyun Lee · Tae Won Kim · Dae Hyun Yang · Se Jin Jang · Young Soo Park · Young Jack Lee · Hwoon‑Yong Jung · Jin‑Ho Kim · Byung Sik Kim
Received: 28 April 2013 / Accepted: 17 October 2013 / Published online: 27 October 2013 © Springer-Verlag Berlin Heidelberg 2013
Abstract Purpose To improve the efficacy of adjuvant chemotherapy with mitomycin-C and fluoropyrimidine (Mf) in gastric cancer, we designed a new regimen (iceMFP) and investigated in a phase III study. Methods We randomly assigned 640 patients with resectable and macroscopically recognizable serosa-invading gastric cancer to Mf or iceMFP group during operation. The Mf consisted of intravenous mitomycin-C (20 mg/m2) at 3–6 weeks after surgery and oral doxifluridine (460– 600 mg/m2/day) starting 4 weeks after the administration of mitomycin-C and continuing for 3 months. The iceMFP consisted of intraoperative intraperitoneal cisplatin (100 mg), intravenous mitomycin-C (15 mg/m2) on This study was presented in part at the 44th ASCO annual meeting, May 30–June 3, 2008, in Chicago, USA, the 36th ESMO annual congress, September 23–27, 2011, in Stockholm, Sweden, and 2012 Gastrointestinal Cancers Symposium, January 19–21, 2012, in San Francisco, USA. Electronic supplementary material The online version of this article (doi:10.1007/s00280-013-2332-5) contains supplementary material, which is available to authorized users. Y.-K. Kang (*) · H.-M. Chang · M.-H. Ryu · C. Yoo · J.-L. Lee · T. W. Kim Department of Oncology, Asan Medical Center, University of Ulsan College of Medicine, 88, Olympic‑ro 43‑gil, Songpa‑gu, Seoul 138‑736, Korea e-mail: [email protected]
postoperative day 1, followed by oral doxifluridine for 12 months, and six monthly intravenous cisplatin (60 mg/ m2). The primary endpoint was 3-year recurrence-free survival (RFS). Results A total of 521 patients (258 in Mf, 263 in iceMFP) were eligible for analysis after excluding patients with stage I disease, distant metastasis, or R1 resection. With a median follow-up of 3.5 years, the iceMFP group had a higher RFS (hazard ratio [HR] 0.70; 95 % confidence interval [CI] 0.54–0.90; p = 0.006; 3-year RFS 60 % vs. 50 %) and overall survival (HR 0.71; 95 % CI 0.53–0.95; p = 0.02; 3-year overall survival, 71 vs. 60 %) compared with the Mf group. This was confirmed at extension analysis after a median 6.6 years of follow-up. Both regimens were well tolerated with no differences in surgical complications. Conclusion The efficacy of adjuvant Mf was significantly improved by the additional therapeutic strategies of iceMFP. Considering negative results of AMC0201, these suggest that early initiation of chemotherapy and/or intraperitoneal cisplatin played a distinct role in the improved efficacy. D. H. Yang Department of Surgery, Hallym University Medical Center, Hallym University College of Medicine, Anyang, Korea S. J. Jang · Y. S. Park Department of Pathology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
J. H. Yook · B. S. Kim Department of Surgery, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
Y. J. Lee LSK Global Pharma Services, Seoul, Korea
D. Y. Zang Department of Internal Medicine, Hallym University Medical Center, Hallym University College of Medicine, Anyang, Korea
H.-Y. Jung · J.-H. Kim Department of Gastroenterology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea
13
140
Keywords Mitomycin-C · Doxifluridine · Cisplatin · Intraperitoneal chemotherapy · Adjuvant chemotherapy · Advanced gastric cancer
Introduction After much debate, it is now globally agreed that adjuvant treatment improves survival of patients with advanced gastric cancer (AGC) undergoing curative surgery. However, there are still geographical differences in standard adjuvant treatments. The intergroup-0116 study of postoperative chemoradiation [1] and the MAGIC trial of perioperative chemotherapy [2] demonstrated the survival benefits in AGC over surgery alone and have become standard adjuvant therapy in the United States and Europe, respectively. In Asia, where D2 surgery is a standard procedure, postoperative chemotherapy with S-1 alone or combination of capecitabine and oxaliplatin has become standard adjuvant therapy on the basis of the results of the ACTS-GC [3] and the recent CLASSIC trials [4]. In 2000, when we designed this trial, adjuvant chemotherapy already was the community standard in Korea and Japan [5, 6]. Although there had been no pivotal study to prove the benefit of adjuvant chemotherapy, several metaanalyses had consistently suggested small but significant benefits of adjuvant chemotherapy [7, 8]. Therefore, we could not consider a clinical trial of adjuvant treatment without a treatment control arm. Instead, we tried to further improve the efficacy of adjuvant chemotherapy with the addition of therapeutic strategies available and suggested to be effective at that time. Combination regimens of mitomycin-C and 5-fluorouracil had been widely investigated and commonly used as adjuvant chemotherapy in Asia [5, 6, 9]. The efficacy of these regimens over surgery alone was subsequently confirmed in recent meta-analysis conducted by the GASTRIC group [10]. In a Spanish phase III study, a combination of mitomycin-C and 3 months of oral fluoropyrimidine (Mf) significantly prolonged survival of patients with resected stage III AGC compared with surgery alone [11]. Although this study was limited by small sample size, potential efficacy and feasibility of Mf regimen was well presented. We therefore adopted the Mf regimen as the control treatment for our adjuvant trials. We considered the following four strategies to enhance the efficacy of the Mf regimen: 1. to add cisplatin since it is active and synergistic with fluoropyrimidine in metastatic disease; 2. to prolong the duration of oral fluoropyrimidine; 3. to initiate the adjuvant chemotherapy immediately after surgery; 4. to add intraperitoneal chemotherapy to prevent the peritoneal recurrence which is common in gastric cancer, especially in cases with serosal involvement. In a companion study (AMC0201), we
13
Cancer Chemother Pharmacol (2014) 73:139–149
evaluated the efficacy of adding the first two strategies to the Mf regimen in patients with postoperative stage II–IV (without distant metastasis) gastric cancers [12]. The AMC0101 study was designed to evaluate the efficacy of adding all the above four strategies to the Mf regimen in patients with AGC macroscopically involving serosa. The result of primary analysis in this study was previously presented in part at the ASCO annual meeting, 2008 [13]. Here we report the final results of the AMC0101 study with inclusion of the results from 3-year extension analysis for the assessment of long-term outcome.
Materials and methods AMC 0101 was a randomized, open-label, phase III trial, conducted at two institutions in Korea. The protocol was approved by the institutional review board at each institution, and the study was conducted in accordance with the ethical principles of the Declaration of Helsinki and the Good Clinical Practice guidelines, as defined by the International Conference on Harmonization. This study was registered in ClinicalTrials.gov (NCT00296322). All patients provided written informed consent before surgery. Patients Patients with histologically proven gastric adenocarcinoma were eligible if their tumors could be curatively resected with D2 dissection. During the surgery, macroscopic involvement of serosal layer was determined as disruption of serosal continuity by the inspection and palpation of the operating surgeon. If this was recognized, patients were entered to this study and randomized. Other inclusion criteria were age between 18 and 70 years, an Eastern Cooperative Oncology Group (ECOG) performance status of 0–2, and adequate hematologic, hepatic, and renal function. Patients who initially entered into this study but who had a pathologically positive resection margin (R1 resection), postoperative stage I or IV with distant metastasis were excluded. Study treatment During surgery, the eligible patients were randomly assigned in a 1:1 ratio to the control (Mf) or experimental (iceMFP) arms. Randomization was stratified by hospital and based on a permutation block method. Patients and investigators were not masked to treatment allocation. For patients assigned to the Mf arm, 20 mg/m2 intravenous mitomycin-C was given 3–6 weeks after surgery, and 4 weeks later, followed by 3 months of 460 (600 after interim analysis) mg/m2/day oral doxifluridine. For patients
Cancer Chemother Pharmacol (2014) 73:139–149
assigned to the iceMFP arm, 100 mg of cisplatin in 1 liter of normal saline was placed intraperitoneally for 2 h before closure of the abdomen, and 15 mg/m2 mitomycin-C was given intravenously on postoperative day 1. Beginning 4 weeks after surgery, 60 mg/m2 intravenous cisplatin was given monthly for 6 months and 460 (increased to 600 after interim analysis) mg/m2/day oral doxifluridine was administered for 12 months. In interim analysis to evaluate safety, performed in February 2004, a less than expected toxicity of doxifluridine prompted us to increase the dose of doxifluridine to 600 mg/m2/day. Assessment At baseline, patients underwent a history, physical examination, computed tomography (CT) scan of the abdomen and pelvis with contrast enhancement, chest radiography, and laboratory tests, including complete blood count, chemistry, and coagulation tests. After surgery, all resected specimens were staged at local pathology laboratories according to the American Joint Committee on Cancer (AJCC) 6th edition. Adverse events and laboratory profiles were assessed every month during treatment. After completion of chemotherapy, clinical assessments were performed every 3 months within 2 years from surgery, every 6 months between 2 and 5 years from surgery, and every 1 year thereafter. To detect recurrence, plain chest radiography and abdominopelvic CT scans were performed every 6 months for 5 years from surgery and every 1 year thereafter, and gastroscopy, every 1 year. If there was any sign or symptom indicating recurrence, investigations were then done to verify whether patients were recurrence-free. If findings on imaging studies were suggestive but not conclusive, serial follow-up studies were performed to verify the recurrence, and the date of the first recognition of findings suggestive of recurrence was defined as the date of recurrence. Adverse events were graded according to the National Cancer Institute’s Common Toxicity Criteria, version 2.0. Statistical analysis The primary endpoint was 3-year recurrence-free survival (RFS), and secondary endpoints were overall survival (OS), disease recurrence, and safety. This trial was designed for 527 patients to be randomly assigned to receive either Mf or iceMFP with a power of 80 % to detect a hazard ratio of 0.6574 for RFS (an improvement of 3-year RFS rate from 55 % for Mf to 67.5 % for iceMFP). Final analysis was planned after 192 events occurred (recurrences or deaths). Safety interim analysis was scheduled to review the toxicity profiles of study regimens after half of the planned number (260) of patients were enrolled. RFS was defined as the
141
time from randomization to documented disease recurrence or death, and OS was defined as the time from randomization to death from any cause. Patients were censored if they were recurrence-free or alive at the last follow-up. Kaplan– Meier curves were used to estimate survival and compared using the log-rank test on an intention-to-treat basis. The Cox regression model was used to estimate hazard ratios regarding the comparison of the efficacy between Mf and iceMFP arm in primary analysis and in prespecified subgroup analyses by potential prognostic factors. All tests were two sided, and a p value less than 0.05 was considered to indicate statistical significance. Confidence intervals (CI) were at the 95 % level. Descriptive analyses were used to represent adverse events. Statistical analyses were performed by a qualified biostatistician (YJL) using SAS version 9.1 (SAS Institute Inc, Cary, NC) and SPSS version 14.0 (SPSS, Chicago, IL). After 229 events occurred with a median follow-up period of 3.5 years, the database was initially frozen on March 20, 2008, and planned primary analysis was performed. Afterward, to verify the long-term results of this study, 3-year extension analysis was performed. The data cutoff for this extension analysis was April 11, 2011. Median follow-up at that time was 6.6 years (maximum, 114.1 months).
Results Between October 2001 and April 2007, 640 patients were randomly assigned to the Mf (318) or iceMFP (322) arms. As prespecified in the protocol, a total of 119 patients (60 in Mf and 59 in iceMFP) were excluded after randomization because of ineligibility. Therefore, a total of 521 patients (258 in Mf and 263 in iceMFP) were included in the analysis for efficacy (Fig. 1). The most common reason for study exclusion was inadequate pathologic stage (stage I or IV with distant metastasis, or microscopic involvement of the resection margin). Table 1 shows that baseline characteristics were well balanced between the two groups. Systemic chemotherapy was administered at a median time from surgery of 22 days in the Mf and 1 day in the iceMFP group. At the time of primary analysis, the median followup was 36.2 months in the Mf group and 38.2 months in the iceMFP group. During this period, 118 (46 %) and 94 (36 %) patients had recurrence, and 103 (40 %) and 82 (31 %) died in the Mf and iceMFP groups, respectively. Compared with the Mf group, the hazard ratio for recurrence in the iceMFP group was 0.70 (95 % CI 0.54–0.90; p = 0.006), suggesting a 30 % risk reduction for recurrence. The RFS rate at 3 years was 60 % (95 % CI 54–67) in the iceMFP group and 50 % (95 % CI 43–57) in the Mf group (Fig. 2a). Median RFS was not reached in the iceMFP
13
142
Cancer Chemother Pharmacol (2014) 73:139–149
Fig. 1 CONSORT diagram
group, whereas it was 34.5 months (95 % CI 24.2–63.8) in the Mf group. Compared with the Mf group, the hazard ratio for death was 0.71 (95 % CI 0.53–0.95; p = 0.02) in the iceMFP group. The 3-year OS rate was 71 % (95 % CI 65–77) in the iceMFP group and 60 % (95 % CI 53–66) in the Mf group. Median OS was estimated as 69.7 months (95 % CI 55.0-not assessable) in the iceMFP group and 54.3 months (95 % CI 37.6-not assessable) in the Mf group (Fig. 2b). Benefits regarding disease recurrence were observed in most subgroups in the iceMFP arm (Fig. 3), and there was no significant interaction between treatment groups and baseline characteristics. RFS and OS were not affected by whether patients received 460 mg/m2/day (278 patients) or 600 mg/m2/day (243) as the starting dose of doxifluridine (p = 0.23 and p = 0.13, respectively). Locoregional recurrences occurred with similar frequency in both groups (6 % in Mf vs. 4 % in iceMFP). The iceMFP group had decreased distant and peritoneal recurrence compared with the Mf group (supplementary data). However, the difference was statistically borderline for distant recurrence (32 vs. 40 %; p = 0.05) and not significant for peritoneal recurrence (17 vs. 23 %; p = 0.08). There was no treatment-related death in both arms. Analyses for adverse events were available for 509 patients
13
(249 in the Mf group vs. 260 in the iceMFP group), and 12 patients (9 vs. 3) were excluded from the safety evaluation population because of not receiving the assigned treatment or loss to follow-up. Adverse events that occurred in 10 % or more of patients are summarized in Table 2. Overall, patients who received iceMFP had more adverse events of any grade than those receiving Mf. Grade 3 or 4 neutropenia was more frequent in the iceMFP (35 %) than in the Mf group (12 %). However, there was no febrile neutropenia. Among non-hematologic toxicities, grade 3 or 4 anorexia, nausea, and vomiting were more common in the iceMFP group. There was no surgical mortality. We could not find clinically significant increase in surgical complications related to the intraperitoneal and intensive chemotherapy (14 % in Mf and 10 % in iceMFP; p = 0.23; supplementary data). The dose of doxifluridine was reduced in 84 patients (34 %) of the Mf group and 176 (70 %) of the iceMFP group. The main reason for dose reduction of doxifluridine was hematologic toxicity (70 vs. 171). The dose of cisplatin was reduced in 6 patients (2 %) of the iceMFP group. The chemotherapy cycle was delayed in 33 patients (13 %) in the Mf group and 88 (35 %) in the iceMFP group. The median relative dose intensity was at least 96 and 78 % for
Cancer Chemother Pharmacol (2014) 73:139–149
143
Table 1 Baseline patient characteristics Characteristic
Mf group, n = 258
iceMFP group, n = 263
Median age, years (min–max) Male gender ECOG performance status 0–1 2 Primary sitea Proximal Distal Multiple/diffuse Type of surgery Total gastrectomy Subtotal gastrectomy Lauren’s classification Intestinal type Diffuse type Mixed type Unknown Tumor stage T1 T2 T3 T4 Nodal stage N0 N1 N2 N3 AJCC stage 6th ed. II IIIA IIIB IV (M0) Dose of doxifluridine 460 mg/m2/day
56 (29–69) 176 (68 %)
53 (25–69) 175 (67 %)
247 (96 %) 11 (4 %)
248 (94 %) 15 (6 %)
26 (10 %) 189 (73 %) 43 (17 %)
25 (9 %) 191 (73 %) 47 (18 %)
141 (55 %) 117 (45 %)
132 (50 %) 131 (50 %)
76 (29 %) 154 (60 %) 25 (10 %) 3 (1 %)
85 (32 %) 147 (56 %) 26 (10 %) 5 (2 %)
1 (
